android_kernel_xiaomi_sm8350/drivers/net/wireless/ipw2200.c

11858 lines
319 KiB
C
Raw Normal View History

2005-03-25 13:31:53 -05:00
/******************************************************************************
Copyright(c) 2003 - 2006 Intel Corporation. All rights reserved.
2005-03-25 13:31:53 -05:00
802.11 status code portion of this file from ethereal-0.10.6:
Copyright 2000, Axis Communications AB
Ethereal - Network traffic analyzer
By Gerald Combs <gerald@ethereal.com>
Copyright 1998 Gerald Combs
This program is free software; you can redistribute it and/or modify it
under the terms of version 2 of the GNU General Public License as
2005-03-25 13:31:53 -05:00
published by the Free Software Foundation.
This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
2005-03-25 13:31:53 -05:00
more details.
2005-03-25 13:31:53 -05:00
You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc., 59
2005-03-25 13:31:53 -05:00
Temple Place - Suite 330, Boston, MA 02111-1307, USA.
2005-03-25 13:31:53 -05:00
The full GNU General Public License is included in this distribution in the
file called LICENSE.
2005-03-25 13:31:53 -05:00
Contact Information:
James P. Ketrenos <ipw2100-admin@linux.intel.com>
Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
******************************************************************************/
#include "ipw2200.h"
#include <linux/version.h>
2005-03-25 13:31:53 -05:00
#ifndef KBUILD_EXTMOD
#define VK "k"
#else
#define VK
#endif
#ifdef CONFIG_IPW2200_DEBUG
#define VD "d"
#else
#define VD
#endif
#ifdef CONFIG_IPW2200_MONITOR
#define VM "m"
#else
#define VM
#endif
#ifdef CONFIG_IPW2200_PROMISCUOUS
#define VP "p"
#else
#define VP
#endif
#ifdef CONFIG_IEEE80211_RADIOTAP
#define VR "r"
#else
#define VR
#endif
#ifdef CONFIG_IPW2200_QOS
#define VQ "q"
#else
#define VQ
#endif
#define IPW2200_VERSION "1.1.2" VK VD VM VP VR VQ
2005-03-25 13:31:53 -05:00
#define DRV_DESCRIPTION "Intel(R) PRO/Wireless 2200/2915 Network Driver"
#define DRV_COPYRIGHT "Copyright(c) 2003-2006 Intel Corporation"
2005-03-25 13:31:53 -05:00
#define DRV_VERSION IPW2200_VERSION
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#define ETH_P_80211_STATS (ETH_P_80211_RAW + 1)
2005-03-25 13:31:53 -05:00
MODULE_DESCRIPTION(DRV_DESCRIPTION);
MODULE_VERSION(DRV_VERSION);
MODULE_AUTHOR(DRV_COPYRIGHT);
MODULE_LICENSE("GPL");
static int cmdlog = 0;
#ifdef CONFIG_IPW2200_DEBUG
2005-03-25 13:31:53 -05:00
static int debug = 0;
#endif
2005-03-25 13:31:53 -05:00
static int channel = 0;
static int mode = 0;
static u32 ipw_debug_level;
static int associate = 1;
static int auto_create = 1;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
static int led = 0;
2005-03-25 13:31:53 -05:00
static int disable = 0;
static int bt_coexist = 0;
static int hwcrypto = 0;
static int roaming = 1;
2005-03-25 13:31:53 -05:00
static const char ipw_modes[] = {
'a', 'b', 'g', '?'
};
static int antenna = CFG_SYS_ANTENNA_BOTH;
2005-03-25 13:31:53 -05:00
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
static int rtap_iface = 0; /* def: 0 -- do not create rtap interface */
#endif
#ifdef CONFIG_IPW2200_QOS
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static int qos_enable = 0;
static int qos_burst_enable = 0;
static int qos_no_ack_mask = 0;
static int burst_duration_CCK = 0;
static int burst_duration_OFDM = 0;
static struct ieee80211_qos_parameters def_qos_parameters_OFDM = {
{QOS_TX0_CW_MIN_OFDM, QOS_TX1_CW_MIN_OFDM, QOS_TX2_CW_MIN_OFDM,
QOS_TX3_CW_MIN_OFDM},
{QOS_TX0_CW_MAX_OFDM, QOS_TX1_CW_MAX_OFDM, QOS_TX2_CW_MAX_OFDM,
QOS_TX3_CW_MAX_OFDM},
{QOS_TX0_AIFS, QOS_TX1_AIFS, QOS_TX2_AIFS, QOS_TX3_AIFS},
{QOS_TX0_ACM, QOS_TX1_ACM, QOS_TX2_ACM, QOS_TX3_ACM},
{QOS_TX0_TXOP_LIMIT_OFDM, QOS_TX1_TXOP_LIMIT_OFDM,
QOS_TX2_TXOP_LIMIT_OFDM, QOS_TX3_TXOP_LIMIT_OFDM}
};
static struct ieee80211_qos_parameters def_qos_parameters_CCK = {
{QOS_TX0_CW_MIN_CCK, QOS_TX1_CW_MIN_CCK, QOS_TX2_CW_MIN_CCK,
QOS_TX3_CW_MIN_CCK},
{QOS_TX0_CW_MAX_CCK, QOS_TX1_CW_MAX_CCK, QOS_TX2_CW_MAX_CCK,
QOS_TX3_CW_MAX_CCK},
{QOS_TX0_AIFS, QOS_TX1_AIFS, QOS_TX2_AIFS, QOS_TX3_AIFS},
{QOS_TX0_ACM, QOS_TX1_ACM, QOS_TX2_ACM, QOS_TX3_ACM},
{QOS_TX0_TXOP_LIMIT_CCK, QOS_TX1_TXOP_LIMIT_CCK, QOS_TX2_TXOP_LIMIT_CCK,
QOS_TX3_TXOP_LIMIT_CCK}
};
static struct ieee80211_qos_parameters def_parameters_OFDM = {
{DEF_TX0_CW_MIN_OFDM, DEF_TX1_CW_MIN_OFDM, DEF_TX2_CW_MIN_OFDM,
DEF_TX3_CW_MIN_OFDM},
{DEF_TX0_CW_MAX_OFDM, DEF_TX1_CW_MAX_OFDM, DEF_TX2_CW_MAX_OFDM,
DEF_TX3_CW_MAX_OFDM},
{DEF_TX0_AIFS, DEF_TX1_AIFS, DEF_TX2_AIFS, DEF_TX3_AIFS},
{DEF_TX0_ACM, DEF_TX1_ACM, DEF_TX2_ACM, DEF_TX3_ACM},
{DEF_TX0_TXOP_LIMIT_OFDM, DEF_TX1_TXOP_LIMIT_OFDM,
DEF_TX2_TXOP_LIMIT_OFDM, DEF_TX3_TXOP_LIMIT_OFDM}
};
static struct ieee80211_qos_parameters def_parameters_CCK = {
{DEF_TX0_CW_MIN_CCK, DEF_TX1_CW_MIN_CCK, DEF_TX2_CW_MIN_CCK,
DEF_TX3_CW_MIN_CCK},
{DEF_TX0_CW_MAX_CCK, DEF_TX1_CW_MAX_CCK, DEF_TX2_CW_MAX_CCK,
DEF_TX3_CW_MAX_CCK},
{DEF_TX0_AIFS, DEF_TX1_AIFS, DEF_TX2_AIFS, DEF_TX3_AIFS},
{DEF_TX0_ACM, DEF_TX1_ACM, DEF_TX2_ACM, DEF_TX3_ACM},
{DEF_TX0_TXOP_LIMIT_CCK, DEF_TX1_TXOP_LIMIT_CCK, DEF_TX2_TXOP_LIMIT_CCK,
DEF_TX3_TXOP_LIMIT_CCK}
};
static u8 qos_oui[QOS_OUI_LEN] = { 0x00, 0x50, 0xF2 };
static int from_priority_to_tx_queue[] = {
IPW_TX_QUEUE_1, IPW_TX_QUEUE_2, IPW_TX_QUEUE_2, IPW_TX_QUEUE_1,
IPW_TX_QUEUE_3, IPW_TX_QUEUE_3, IPW_TX_QUEUE_4, IPW_TX_QUEUE_4
};
static u32 ipw_qos_get_burst_duration(struct ipw_priv *priv);
static int ipw_send_qos_params_command(struct ipw_priv *priv, struct ieee80211_qos_parameters
*qos_param);
static int ipw_send_qos_info_command(struct ipw_priv *priv, struct ieee80211_qos_information_element
*qos_param);
#endif /* CONFIG_IPW2200_QOS */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static struct iw_statistics *ipw_get_wireless_stats(struct net_device *dev);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_remove_current_network(struct ipw_priv *priv);
2005-03-25 13:31:53 -05:00
static void ipw_rx(struct ipw_priv *priv);
static int ipw_queue_tx_reclaim(struct ipw_priv *priv,
2005-03-25 13:31:53 -05:00
struct clx2_tx_queue *txq, int qindex);
static int ipw_queue_reset(struct ipw_priv *priv);
static int ipw_queue_tx_hcmd(struct ipw_priv *priv, int hcmd, void *buf,
int len, int sync);
static void ipw_tx_queue_free(struct ipw_priv *);
static struct ipw_rx_queue *ipw_rx_queue_alloc(struct ipw_priv *);
static void ipw_rx_queue_free(struct ipw_priv *, struct ipw_rx_queue *);
static void ipw_rx_queue_replenish(void *);
static int ipw_up(struct ipw_priv *);
static void ipw_bg_up(void *);
2005-03-25 13:31:53 -05:00
static void ipw_down(struct ipw_priv *);
static void ipw_bg_down(void *);
2005-03-25 13:31:53 -05:00
static int ipw_config(struct ipw_priv *);
static int init_supported_rates(struct ipw_priv *priv,
struct ipw_supported_rates *prates);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_set_hwcrypto_keys(struct ipw_priv *);
static void ipw_send_wep_keys(struct ipw_priv *, int);
2005-03-25 13:31:53 -05:00
static int snprint_line(char *buf, size_t count,
const u8 * data, u32 len, u32 ofs)
2005-03-25 13:31:53 -05:00
{
int out, i, j, l;
char c;
2005-03-25 13:31:53 -05:00
out = snprintf(buf, count, "%08X", ofs);
for (l = 0, i = 0; i < 2; i++) {
out += snprintf(buf + out, count - out, " ");
for (j = 0; j < 8 && l < len; j++, l++)
out += snprintf(buf + out, count - out, "%02X ",
2005-03-25 13:31:53 -05:00
data[(i * 8 + j)]);
for (; j < 8; j++)
out += snprintf(buf + out, count - out, " ");
}
2005-03-25 13:31:53 -05:00
out += snprintf(buf + out, count - out, " ");
for (l = 0, i = 0; i < 2; i++) {
out += snprintf(buf + out, count - out, " ");
for (j = 0; j < 8 && l < len; j++, l++) {
c = data[(i * 8 + j)];
if (!isascii(c) || !isprint(c))
c = '.';
2005-03-25 13:31:53 -05:00
out += snprintf(buf + out, count - out, "%c", c);
}
for (; j < 8; j++)
out += snprintf(buf + out, count - out, " ");
}
return out;
2005-03-25 13:31:53 -05:00
}
static void printk_buf(int level, const u8 * data, u32 len)
2005-03-25 13:31:53 -05:00
{
char line[81];
u32 ofs = 0;
if (!(ipw_debug_level & level))
return;
while (len) {
snprint_line(line, sizeof(line), &data[ofs],
min(len, 16U), ofs);
printk(KERN_DEBUG "%s\n", line);
2005-03-25 13:31:53 -05:00
ofs += 16;
len -= min(len, 16U);
}
}
static int snprintk_buf(u8 * output, size_t size, const u8 * data, size_t len)
{
size_t out = size;
u32 ofs = 0;
int total = 0;
while (size && len) {
out = snprint_line(output, size, &data[ofs],
min_t(size_t, len, 16U), ofs);
ofs += 16;
output += out;
size -= out;
len -= min_t(size_t, len, 16U);
total += out;
}
return total;
}
/* alias for 32-bit indirect read (for SRAM/reg above 4K), with debug wrapper */
2005-03-25 13:31:53 -05:00
static u32 _ipw_read_reg32(struct ipw_priv *priv, u32 reg);
#define ipw_read_reg32(a, b) _ipw_read_reg32(a, b)
/* alias for 8-bit indirect read (for SRAM/reg above 4K), with debug wrapper */
2005-03-25 13:31:53 -05:00
static u8 _ipw_read_reg8(struct ipw_priv *ipw, u32 reg);
#define ipw_read_reg8(a, b) _ipw_read_reg8(a, b)
/* 8-bit indirect write (for SRAM/reg above 4K), with debug wrapper */
2005-03-25 13:31:53 -05:00
static void _ipw_write_reg8(struct ipw_priv *priv, u32 reg, u8 value);
static inline void ipw_write_reg8(struct ipw_priv *a, u32 b, u8 c)
{
IPW_DEBUG_IO("%s %d: write_indirect8(0x%08X, 0x%08X)\n", __FILE__,
__LINE__, (u32) (b), (u32) (c));
2005-03-25 13:31:53 -05:00
_ipw_write_reg8(a, b, c);
}
/* 16-bit indirect write (for SRAM/reg above 4K), with debug wrapper */
2005-03-25 13:31:53 -05:00
static void _ipw_write_reg16(struct ipw_priv *priv, u32 reg, u16 value);
static inline void ipw_write_reg16(struct ipw_priv *a, u32 b, u16 c)
{
IPW_DEBUG_IO("%s %d: write_indirect16(0x%08X, 0x%08X)\n", __FILE__,
__LINE__, (u32) (b), (u32) (c));
2005-03-25 13:31:53 -05:00
_ipw_write_reg16(a, b, c);
}
/* 32-bit indirect write (for SRAM/reg above 4K), with debug wrapper */
2005-03-25 13:31:53 -05:00
static void _ipw_write_reg32(struct ipw_priv *priv, u32 reg, u32 value);
static inline void ipw_write_reg32(struct ipw_priv *a, u32 b, u32 c)
{
IPW_DEBUG_IO("%s %d: write_indirect32(0x%08X, 0x%08X)\n", __FILE__,
__LINE__, (u32) (b), (u32) (c));
2005-03-25 13:31:53 -05:00
_ipw_write_reg32(a, b, c);
}
/* 8-bit direct write (low 4K) */
2005-03-25 13:31:53 -05:00
#define _ipw_write8(ipw, ofs, val) writeb((val), (ipw)->hw_base + (ofs))
/* 8-bit direct write (for low 4K of SRAM/regs), with debug wrapper */
2005-03-25 13:31:53 -05:00
#define ipw_write8(ipw, ofs, val) \
IPW_DEBUG_IO("%s %d: write_direct8(0x%08X, 0x%08X)\n", __FILE__, __LINE__, (u32)(ofs), (u32)(val)); \
_ipw_write8(ipw, ofs, val)
/* 16-bit direct write (low 4K) */
2005-03-25 13:31:53 -05:00
#define _ipw_write16(ipw, ofs, val) writew((val), (ipw)->hw_base + (ofs))
/* 16-bit direct write (for low 4K of SRAM/regs), with debug wrapper */
2005-03-25 13:31:53 -05:00
#define ipw_write16(ipw, ofs, val) \
IPW_DEBUG_IO("%s %d: write_direct16(0x%08X, 0x%08X)\n", __FILE__, __LINE__, (u32)(ofs), (u32)(val)); \
_ipw_write16(ipw, ofs, val)
/* 32-bit direct write (low 4K) */
2005-03-25 13:31:53 -05:00
#define _ipw_write32(ipw, ofs, val) writel((val), (ipw)->hw_base + (ofs))
/* 32-bit direct write (for low 4K of SRAM/regs), with debug wrapper */
2005-03-25 13:31:53 -05:00
#define ipw_write32(ipw, ofs, val) \
IPW_DEBUG_IO("%s %d: write_direct32(0x%08X, 0x%08X)\n", __FILE__, __LINE__, (u32)(ofs), (u32)(val)); \
_ipw_write32(ipw, ofs, val)
/* 8-bit direct read (low 4K) */
2005-03-25 13:31:53 -05:00
#define _ipw_read8(ipw, ofs) readb((ipw)->hw_base + (ofs))
/* 8-bit direct read (low 4K), with debug wrapper */
static inline u8 __ipw_read8(char *f, u32 l, struct ipw_priv *ipw, u32 ofs)
{
IPW_DEBUG_IO("%s %d: read_direct8(0x%08X)\n", f, l, (u32) (ofs));
2005-03-25 13:31:53 -05:00
return _ipw_read8(ipw, ofs);
}
/* alias to 8-bit direct read (low 4K of SRAM/regs), with debug wrapper */
2005-03-25 13:31:53 -05:00
#define ipw_read8(ipw, ofs) __ipw_read8(__FILE__, __LINE__, ipw, ofs)
/* 16-bit direct read (low 4K) */
2005-03-25 13:31:53 -05:00
#define _ipw_read16(ipw, ofs) readw((ipw)->hw_base + (ofs))
/* 16-bit direct read (low 4K), with debug wrapper */
static inline u16 __ipw_read16(char *f, u32 l, struct ipw_priv *ipw, u32 ofs)
{
IPW_DEBUG_IO("%s %d: read_direct16(0x%08X)\n", f, l, (u32) (ofs));
2005-03-25 13:31:53 -05:00
return _ipw_read16(ipw, ofs);
}
/* alias to 16-bit direct read (low 4K of SRAM/regs), with debug wrapper */
2005-03-25 13:31:53 -05:00
#define ipw_read16(ipw, ofs) __ipw_read16(__FILE__, __LINE__, ipw, ofs)
/* 32-bit direct read (low 4K) */
2005-03-25 13:31:53 -05:00
#define _ipw_read32(ipw, ofs) readl((ipw)->hw_base + (ofs))
/* 32-bit direct read (low 4K), with debug wrapper */
static inline u32 __ipw_read32(char *f, u32 l, struct ipw_priv *ipw, u32 ofs)
{
IPW_DEBUG_IO("%s %d: read_direct32(0x%08X)\n", f, l, (u32) (ofs));
2005-03-25 13:31:53 -05:00
return _ipw_read32(ipw, ofs);
}
/* alias to 32-bit direct read (low 4K of SRAM/regs), with debug wrapper */
2005-03-25 13:31:53 -05:00
#define ipw_read32(ipw, ofs) __ipw_read32(__FILE__, __LINE__, ipw, ofs)
/* multi-byte read (above 4K), with debug wrapper */
2005-03-25 13:31:53 -05:00
static void _ipw_read_indirect(struct ipw_priv *, u32, u8 *, int);
static inline void __ipw_read_indirect(const char *f, int l,
struct ipw_priv *a, u32 b, u8 * c, int d)
{
IPW_DEBUG_IO("%s %d: read_indirect(0x%08X) %d bytes\n", f, l, (u32) (b),
d);
_ipw_read_indirect(a, b, c, d);
}
/* alias to multi-byte read (SRAM/regs above 4K), with debug wrapper */
#define ipw_read_indirect(a, b, c, d) __ipw_read_indirect(__FILE__, __LINE__, a, b, c, d)
2005-03-25 13:31:53 -05:00
/* alias to multi-byte read (SRAM/regs above 4K), with debug wrapper */
static void _ipw_write_indirect(struct ipw_priv *priv, u32 addr, u8 * data,
int num);
2005-03-25 13:31:53 -05:00
#define ipw_write_indirect(a, b, c, d) \
IPW_DEBUG_IO("%s %d: write_indirect(0x%08X) %d bytes\n", __FILE__, __LINE__, (u32)(b), d); \
_ipw_write_indirect(a, b, c, d)
2005-03-25 13:31:53 -05:00
/* 32-bit indirect write (above 4K) */
static void _ipw_write_reg32(struct ipw_priv *priv, u32 reg, u32 value)
2005-03-25 13:31:53 -05:00
{
IPW_DEBUG_IO(" %p : reg = 0x%8X : value = 0x%8X\n", priv, reg, value);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_INDIRECT_ADDR, reg);
_ipw_write32(priv, IPW_INDIRECT_DATA, value);
2005-03-25 13:31:53 -05:00
}
/* 8-bit indirect write (above 4K) */
2005-03-25 13:31:53 -05:00
static void _ipw_write_reg8(struct ipw_priv *priv, u32 reg, u8 value)
{
u32 aligned_addr = reg & IPW_INDIRECT_ADDR_MASK; /* dword align */
u32 dif_len = reg - aligned_addr;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_IO(" reg = 0x%8X : value = 0x%8X\n", reg, value);
_ipw_write32(priv, IPW_INDIRECT_ADDR, aligned_addr);
_ipw_write8(priv, IPW_INDIRECT_DATA + dif_len, value);
2005-03-25 13:31:53 -05:00
}
/* 16-bit indirect write (above 4K) */
static void _ipw_write_reg16(struct ipw_priv *priv, u32 reg, u16 value)
2005-03-25 13:31:53 -05:00
{
u32 aligned_addr = reg & IPW_INDIRECT_ADDR_MASK; /* dword align */
u32 dif_len = (reg - aligned_addr) & (~0x1ul);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_IO(" reg = 0x%8X : value = 0x%8X\n", reg, value);
_ipw_write32(priv, IPW_INDIRECT_ADDR, aligned_addr);
_ipw_write16(priv, IPW_INDIRECT_DATA + dif_len, value);
2005-03-25 13:31:53 -05:00
}
/* 8-bit indirect read (above 4K) */
2005-03-25 13:31:53 -05:00
static u8 _ipw_read_reg8(struct ipw_priv *priv, u32 reg)
{
u32 word;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_INDIRECT_ADDR, reg & IPW_INDIRECT_ADDR_MASK);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_IO(" reg = 0x%8X : \n", reg);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
word = _ipw_read32(priv, IPW_INDIRECT_DATA);
return (word >> ((reg & 0x3) * 8)) & 0xff;
2005-03-25 13:31:53 -05:00
}
/* 32-bit indirect read (above 4K) */
2005-03-25 13:31:53 -05:00
static u32 _ipw_read_reg32(struct ipw_priv *priv, u32 reg)
{
u32 value;
IPW_DEBUG_IO("%p : reg = 0x%08x\n", priv, reg);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_INDIRECT_ADDR, reg);
value = _ipw_read32(priv, IPW_INDIRECT_DATA);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_IO(" reg = 0x%4X : value = 0x%4x \n", reg, value);
return value;
}
/* General purpose, no alignment requirement, iterative (multi-byte) read, */
/* for area above 1st 4K of SRAM/reg space */
2005-03-25 13:31:53 -05:00
static void _ipw_read_indirect(struct ipw_priv *priv, u32 addr, u8 * buf,
int num)
{
u32 aligned_addr = addr & IPW_INDIRECT_ADDR_MASK; /* dword align */
2005-03-25 13:31:53 -05:00
u32 dif_len = addr - aligned_addr;
u32 i;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_IO("addr = %i, buf = %p, num = %i\n", addr, buf, num);
if (num <= 0) {
return;
}
/* Read the first dword (or portion) byte by byte */
2005-03-25 13:31:53 -05:00
if (unlikely(dif_len)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_INDIRECT_ADDR, aligned_addr);
2005-03-25 13:31:53 -05:00
/* Start reading at aligned_addr + dif_len */
for (i = dif_len; ((i < 4) && (num > 0)); i++, num--)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
*buf++ = _ipw_read8(priv, IPW_INDIRECT_DATA + i);
2005-03-25 13:31:53 -05:00
aligned_addr += 4;
}
/* Read all of the middle dwords as dwords, with auto-increment */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_AUTOINC_ADDR, aligned_addr);
for (; num >= 4; buf += 4, aligned_addr += 4, num -= 4)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
*(u32 *) buf = _ipw_read32(priv, IPW_AUTOINC_DATA);
/* Read the last dword (or portion) byte by byte */
if (unlikely(num)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_INDIRECT_ADDR, aligned_addr);
for (i = 0; num > 0; i++, num--)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
*buf++ = ipw_read8(priv, IPW_INDIRECT_DATA + i);
}
2005-03-25 13:31:53 -05:00
}
/* General purpose, no alignment requirement, iterative (multi-byte) write, */
/* for area above 1st 4K of SRAM/reg space */
static void _ipw_write_indirect(struct ipw_priv *priv, u32 addr, u8 * buf,
2005-03-25 13:31:53 -05:00
int num)
{
u32 aligned_addr = addr & IPW_INDIRECT_ADDR_MASK; /* dword align */
2005-03-25 13:31:53 -05:00
u32 dif_len = addr - aligned_addr;
u32 i;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_IO("addr = %i, buf = %p, num = %i\n", addr, buf, num);
if (num <= 0) {
return;
}
/* Write the first dword (or portion) byte by byte */
2005-03-25 13:31:53 -05:00
if (unlikely(dif_len)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_INDIRECT_ADDR, aligned_addr);
/* Start writing at aligned_addr + dif_len */
for (i = dif_len; ((i < 4) && (num > 0)); i++, num--, buf++)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write8(priv, IPW_INDIRECT_DATA + i, *buf);
2005-03-25 13:31:53 -05:00
aligned_addr += 4;
}
/* Write all of the middle dwords as dwords, with auto-increment */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_AUTOINC_ADDR, aligned_addr);
for (; num >= 4; buf += 4, aligned_addr += 4, num -= 4)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_AUTOINC_DATA, *(u32 *) buf);
/* Write the last dword (or portion) byte by byte */
if (unlikely(num)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_INDIRECT_ADDR, aligned_addr);
for (i = 0; num > 0; i++, num--, buf++)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write8(priv, IPW_INDIRECT_DATA + i, *buf);
}
2005-03-25 13:31:53 -05:00
}
/* General purpose, no alignment requirement, iterative (multi-byte) write, */
/* for 1st 4K of SRAM/regs space */
static void ipw_write_direct(struct ipw_priv *priv, u32 addr, void *buf,
2005-03-25 13:31:53 -05:00
int num)
{
memcpy_toio((priv->hw_base + addr), buf, num);
}
/* Set bit(s) in low 4K of SRAM/regs */
2005-03-25 13:31:53 -05:00
static inline void ipw_set_bit(struct ipw_priv *priv, u32 reg, u32 mask)
{
ipw_write32(priv, reg, ipw_read32(priv, reg) | mask);
}
/* Clear bit(s) in low 4K of SRAM/regs */
2005-03-25 13:31:53 -05:00
static inline void ipw_clear_bit(struct ipw_priv *priv, u32 reg, u32 mask)
{
ipw_write32(priv, reg, ipw_read32(priv, reg) & ~mask);
}
static inline void ipw_enable_interrupts(struct ipw_priv *priv)
{
if (priv->status & STATUS_INT_ENABLED)
return;
priv->status |= STATUS_INT_ENABLED;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_MASK_R, IPW_INTA_MASK_ALL);
2005-03-25 13:31:53 -05:00
}
static inline void ipw_disable_interrupts(struct ipw_priv *priv)
{
if (!(priv->status & STATUS_INT_ENABLED))
return;
priv->status &= ~STATUS_INT_ENABLED;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_MASK_R, ~IPW_INTA_MASK_ALL);
2005-03-25 13:31:53 -05:00
}
#ifdef CONFIG_IPW2200_DEBUG
2005-03-25 13:31:53 -05:00
static char *ipw_error_desc(u32 val)
{
switch (val) {
case IPW_FW_ERROR_OK:
2005-03-25 13:31:53 -05:00
return "ERROR_OK";
case IPW_FW_ERROR_FAIL:
2005-03-25 13:31:53 -05:00
return "ERROR_FAIL";
case IPW_FW_ERROR_MEMORY_UNDERFLOW:
2005-03-25 13:31:53 -05:00
return "MEMORY_UNDERFLOW";
case IPW_FW_ERROR_MEMORY_OVERFLOW:
2005-03-25 13:31:53 -05:00
return "MEMORY_OVERFLOW";
case IPW_FW_ERROR_BAD_PARAM:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "BAD_PARAM";
case IPW_FW_ERROR_BAD_CHECKSUM:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "BAD_CHECKSUM";
case IPW_FW_ERROR_NMI_INTERRUPT:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "NMI_INTERRUPT";
case IPW_FW_ERROR_BAD_DATABASE:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "BAD_DATABASE";
case IPW_FW_ERROR_ALLOC_FAIL:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "ALLOC_FAIL";
case IPW_FW_ERROR_DMA_UNDERRUN:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "DMA_UNDERRUN";
case IPW_FW_ERROR_DMA_STATUS:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "DMA_STATUS";
case IPW_FW_ERROR_DINO_ERROR:
return "DINO_ERROR";
case IPW_FW_ERROR_EEPROM_ERROR:
return "EEPROM_ERROR";
case IPW_FW_ERROR_SYSASSERT:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "SYSASSERT";
case IPW_FW_ERROR_FATAL_ERROR:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "FATAL_ERROR";
default:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return "UNKNOWN_ERROR";
2005-03-25 13:31:53 -05:00
}
}
static void ipw_dump_error_log(struct ipw_priv *priv,
struct ipw_fw_error *error)
2005-03-25 13:31:53 -05:00
{
u32 i;
if (!error) {
IPW_ERROR("Error allocating and capturing error log. "
"Nothing to dump.\n");
return;
2005-03-25 13:31:53 -05:00
}
IPW_ERROR("Start IPW Error Log Dump:\n");
IPW_ERROR("Status: 0x%08X, Config: %08X\n",
error->status, error->config);
2005-03-25 13:31:53 -05:00
for (i = 0; i < error->elem_len; i++)
IPW_ERROR("%s %i 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n",
ipw_error_desc(error->elem[i].desc),
error->elem[i].time,
error->elem[i].blink1,
error->elem[i].blink2,
error->elem[i].link1,
error->elem[i].link2, error->elem[i].data);
for (i = 0; i < error->log_len; i++)
IPW_ERROR("%i\t0x%08x\t%i\n",
error->log[i].time,
error->log[i].data, error->log[i].event);
2005-03-25 13:31:53 -05:00
}
#endif
static inline int ipw_is_init(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
return (priv->status & STATUS_INIT) ? 1 : 0;
2005-03-25 13:31:53 -05:00
}
static int ipw_get_ordinal(struct ipw_priv *priv, u32 ord, void *val, u32 * len)
2005-03-25 13:31:53 -05:00
{
u32 addr, field_info, field_len, field_count, total_len;
IPW_DEBUG_ORD("ordinal = %i\n", ord);
if (!priv || !val || !len) {
IPW_DEBUG_ORD("Invalid argument\n");
return -EINVAL;
}
2005-03-25 13:31:53 -05:00
/* verify device ordinal tables have been initialized */
if (!priv->table0_addr || !priv->table1_addr || !priv->table2_addr) {
IPW_DEBUG_ORD("Access ordinals before initialization\n");
return -EINVAL;
}
switch (IPW_ORD_TABLE_ID_MASK & ord) {
case IPW_ORD_TABLE_0_MASK:
/*
* TABLE 0: Direct access to a table of 32 bit values
*
* This is a very simple table with the data directly
2005-03-25 13:31:53 -05:00
* read from the table
*/
/* remove the table id from the ordinal */
ord &= IPW_ORD_TABLE_VALUE_MASK;
/* boundary check */
if (ord > priv->table0_len) {
IPW_DEBUG_ORD("ordinal value (%i) longer then "
"max (%i)\n", ord, priv->table0_len);
return -EINVAL;
}
/* verify we have enough room to store the value */
if (*len < sizeof(u32)) {
IPW_DEBUG_ORD("ordinal buffer length too small, "
"need %zd\n", sizeof(u32));
2005-03-25 13:31:53 -05:00
return -EINVAL;
}
IPW_DEBUG_ORD("Reading TABLE0[%i] from offset 0x%08x\n",
ord, priv->table0_addr + (ord << 2));
2005-03-25 13:31:53 -05:00
*len = sizeof(u32);
ord <<= 2;
*((u32 *) val) = ipw_read32(priv, priv->table0_addr + ord);
2005-03-25 13:31:53 -05:00
break;
case IPW_ORD_TABLE_1_MASK:
/*
* TABLE 1: Indirect access to a table of 32 bit values
*
* This is a fairly large table of u32 values each
2005-03-25 13:31:53 -05:00
* representing starting addr for the data (which is
* also a u32)
*/
/* remove the table id from the ordinal */
ord &= IPW_ORD_TABLE_VALUE_MASK;
2005-03-25 13:31:53 -05:00
/* boundary check */
if (ord > priv->table1_len) {
IPW_DEBUG_ORD("ordinal value too long\n");
return -EINVAL;
}
/* verify we have enough room to store the value */
if (*len < sizeof(u32)) {
IPW_DEBUG_ORD("ordinal buffer length too small, "
"need %zd\n", sizeof(u32));
2005-03-25 13:31:53 -05:00
return -EINVAL;
}
*((u32 *) val) =
ipw_read_reg32(priv, (priv->table1_addr + (ord << 2)));
2005-03-25 13:31:53 -05:00
*len = sizeof(u32);
break;
case IPW_ORD_TABLE_2_MASK:
/*
* TABLE 2: Indirect access to a table of variable sized values
*
* This table consist of six values, each containing
* - dword containing the starting offset of the data
* - dword containing the lengh in the first 16bits
* and the count in the second 16bits
*/
/* remove the table id from the ordinal */
ord &= IPW_ORD_TABLE_VALUE_MASK;
/* boundary check */
if (ord > priv->table2_len) {
IPW_DEBUG_ORD("ordinal value too long\n");
return -EINVAL;
}
/* get the address of statistic */
addr = ipw_read_reg32(priv, priv->table2_addr + (ord << 3));
/* get the second DW of statistics ;
2005-03-25 13:31:53 -05:00
* two 16-bit words - first is length, second is count */
field_info =
ipw_read_reg32(priv,
priv->table2_addr + (ord << 3) +
sizeof(u32));
2005-03-25 13:31:53 -05:00
/* get each entry length */
field_len = *((u16 *) & field_info);
2005-03-25 13:31:53 -05:00
/* get number of entries */
field_count = *(((u16 *) & field_info) + 1);
2005-03-25 13:31:53 -05:00
/* abort if not enought memory */
total_len = field_len * field_count;
if (total_len > *len) {
*len = total_len;
return -EINVAL;
}
2005-03-25 13:31:53 -05:00
*len = total_len;
if (!total_len)
return 0;
IPW_DEBUG_ORD("addr = 0x%08x, total_len = %i, "
"field_info = 0x%08x\n",
2005-03-25 13:31:53 -05:00
addr, total_len, field_info);
ipw_read_indirect(priv, addr, val, total_len);
break;
default:
IPW_DEBUG_ORD("Invalid ordinal!\n");
return -EINVAL;
}
return 0;
}
static void ipw_init_ordinals(struct ipw_priv *priv)
{
priv->table0_addr = IPW_ORDINALS_TABLE_LOWER;
priv->table0_len = ipw_read32(priv, priv->table0_addr);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ORD("table 0 offset at 0x%08x, len = %i\n",
priv->table0_addr, priv->table0_len);
priv->table1_addr = ipw_read32(priv, IPW_ORDINALS_TABLE_1);
priv->table1_len = ipw_read_reg32(priv, priv->table1_addr);
IPW_DEBUG_ORD("table 1 offset at 0x%08x, len = %i\n",
priv->table1_addr, priv->table1_len);
priv->table2_addr = ipw_read32(priv, IPW_ORDINALS_TABLE_2);
priv->table2_len = ipw_read_reg32(priv, priv->table2_addr);
priv->table2_len &= 0x0000ffff; /* use first two bytes */
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ORD("table 2 offset at 0x%08x, len = %i\n",
priv->table2_addr, priv->table2_len);
}
static u32 ipw_register_toggle(u32 reg)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
reg &= ~IPW_START_STANDBY;
if (reg & IPW_GATE_ODMA)
reg &= ~IPW_GATE_ODMA;
if (reg & IPW_GATE_IDMA)
reg &= ~IPW_GATE_IDMA;
if (reg & IPW_GATE_ADMA)
reg &= ~IPW_GATE_ADMA;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
return reg;
}
/*
* LED behavior:
* - On radio ON, turn on any LEDs that require to be on during start
* - On initialization, start unassociated blink
* - On association, disable unassociated blink
* - On disassociation, start unassociated blink
* - On radio OFF, turn off any LEDs started during radio on
*
*/
#define LD_TIME_LINK_ON msecs_to_jiffies(300)
#define LD_TIME_LINK_OFF msecs_to_jiffies(2700)
#define LD_TIME_ACT_ON msecs_to_jiffies(250)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
static void ipw_led_link_on(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
unsigned long flags;
u32 led;
/* If configured to not use LEDs, or nic_type is 1,
* then we don't toggle a LINK led */
if (priv->config & CFG_NO_LED || priv->nic_type == EEPROM_NIC_TYPE_1)
return;
spin_lock_irqsave(&priv->lock, flags);
if (!(priv->status & STATUS_RF_KILL_MASK) &&
!(priv->status & STATUS_LED_LINK_ON)) {
IPW_DEBUG_LED("Link LED On\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
led = ipw_read_reg32(priv, IPW_EVENT_REG);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
led |= priv->led_association_on;
led = ipw_register_toggle(led);
IPW_DEBUG_LED("Reg: 0x%08X\n", led);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_EVENT_REG, led);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
priv->status |= STATUS_LED_LINK_ON;
/* If we aren't associated, schedule turning the LED off */
if (!(priv->status & STATUS_ASSOCIATED))
queue_delayed_work(priv->workqueue,
&priv->led_link_off,
LD_TIME_LINK_ON);
}
spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipw_bg_led_link_on(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_led_link_on(data);
mutex_unlock(&priv->mutex);
}
static void ipw_led_link_off(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
unsigned long flags;
u32 led;
/* If configured not to use LEDs, or nic type is 1,
* then we don't goggle the LINK led. */
if (priv->config & CFG_NO_LED || priv->nic_type == EEPROM_NIC_TYPE_1)
return;
spin_lock_irqsave(&priv->lock, flags);
if (priv->status & STATUS_LED_LINK_ON) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
led = ipw_read_reg32(priv, IPW_EVENT_REG);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
led &= priv->led_association_off;
led = ipw_register_toggle(led);
IPW_DEBUG_LED("Reg: 0x%08X\n", led);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_EVENT_REG, led);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
IPW_DEBUG_LED("Link LED Off\n");
priv->status &= ~STATUS_LED_LINK_ON;
/* If we aren't associated and the radio is on, schedule
* turning the LED on (blink while unassociated) */
if (!(priv->status & STATUS_RF_KILL_MASK) &&
!(priv->status & STATUS_ASSOCIATED))
queue_delayed_work(priv->workqueue, &priv->led_link_on,
LD_TIME_LINK_OFF);
}
spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipw_bg_led_link_off(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_led_link_off(data);
mutex_unlock(&priv->mutex);
}
static void __ipw_led_activity_on(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
u32 led;
if (priv->config & CFG_NO_LED)
return;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->status & STATUS_RF_KILL_MASK)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
return;
if (!(priv->status & STATUS_LED_ACT_ON)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
led = ipw_read_reg32(priv, IPW_EVENT_REG);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
led |= priv->led_activity_on;
led = ipw_register_toggle(led);
IPW_DEBUG_LED("Reg: 0x%08X\n", led);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_EVENT_REG, led);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
IPW_DEBUG_LED("Activity LED On\n");
priv->status |= STATUS_LED_ACT_ON;
cancel_delayed_work(&priv->led_act_off);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
queue_delayed_work(priv->workqueue, &priv->led_act_off,
LD_TIME_ACT_ON);
} else {
/* Reschedule LED off for full time period */
cancel_delayed_work(&priv->led_act_off);
queue_delayed_work(priv->workqueue, &priv->led_act_off,
LD_TIME_ACT_ON);
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
#if 0
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
void ipw_led_activity_on(struct ipw_priv *priv)
{
unsigned long flags;
spin_lock_irqsave(&priv->lock, flags);
__ipw_led_activity_on(priv);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->lock, flags);
}
#endif /* 0 */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
static void ipw_led_activity_off(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
unsigned long flags;
u32 led;
if (priv->config & CFG_NO_LED)
return;
spin_lock_irqsave(&priv->lock, flags);
if (priv->status & STATUS_LED_ACT_ON) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
led = ipw_read_reg32(priv, IPW_EVENT_REG);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
led &= priv->led_activity_off;
led = ipw_register_toggle(led);
IPW_DEBUG_LED("Reg: 0x%08X\n", led);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_EVENT_REG, led);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
IPW_DEBUG_LED("Activity LED Off\n");
priv->status &= ~STATUS_LED_ACT_ON;
}
spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipw_bg_led_activity_off(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_led_activity_off(data);
mutex_unlock(&priv->mutex);
}
static void ipw_led_band_on(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
unsigned long flags;
u32 led;
/* Only nic type 1 supports mode LEDs */
if (priv->config & CFG_NO_LED ||
priv->nic_type != EEPROM_NIC_TYPE_1 || !priv->assoc_network)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
return;
spin_lock_irqsave(&priv->lock, flags);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
led = ipw_read_reg32(priv, IPW_EVENT_REG);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (priv->assoc_network->mode == IEEE_A) {
led |= priv->led_ofdm_on;
led &= priv->led_association_off;
IPW_DEBUG_LED("Mode LED On: 802.11a\n");
} else if (priv->assoc_network->mode == IEEE_G) {
led |= priv->led_ofdm_on;
led |= priv->led_association_on;
IPW_DEBUG_LED("Mode LED On: 802.11g\n");
} else {
led &= priv->led_ofdm_off;
led |= priv->led_association_on;
IPW_DEBUG_LED("Mode LED On: 802.11b\n");
}
led = ipw_register_toggle(led);
IPW_DEBUG_LED("Reg: 0x%08X\n", led);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_EVENT_REG, led);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipw_led_band_off(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
unsigned long flags;
u32 led;
/* Only nic type 1 supports mode LEDs */
if (priv->config & CFG_NO_LED || priv->nic_type != EEPROM_NIC_TYPE_1)
return;
spin_lock_irqsave(&priv->lock, flags);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
led = ipw_read_reg32(priv, IPW_EVENT_REG);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
led &= priv->led_ofdm_off;
led &= priv->led_association_off;
led = ipw_register_toggle(led);
IPW_DEBUG_LED("Reg: 0x%08X\n", led);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_EVENT_REG, led);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipw_led_radio_on(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
ipw_led_link_on(priv);
}
static void ipw_led_radio_off(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
ipw_led_activity_off(priv);
ipw_led_link_off(priv);
}
static void ipw_led_link_up(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
/* Set the Link Led on for all nic types */
ipw_led_link_on(priv);
}
static void ipw_led_link_down(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
ipw_led_activity_off(priv);
ipw_led_link_off(priv);
if (priv->status & STATUS_RF_KILL_MASK)
ipw_led_radio_off(priv);
}
static void ipw_led_init(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
priv->nic_type = priv->eeprom[EEPROM_NIC_TYPE];
/* Set the default PINs for the link and activity leds */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->led_activity_on = IPW_ACTIVITY_LED;
priv->led_activity_off = ~(IPW_ACTIVITY_LED);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->led_association_on = IPW_ASSOCIATED_LED;
priv->led_association_off = ~(IPW_ASSOCIATED_LED);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
/* Set the default PINs for the OFDM leds */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->led_ofdm_on = IPW_OFDM_LED;
priv->led_ofdm_off = ~(IPW_OFDM_LED);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
switch (priv->nic_type) {
case EEPROM_NIC_TYPE_1:
/* In this NIC type, the LEDs are reversed.... */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->led_activity_on = IPW_ASSOCIATED_LED;
priv->led_activity_off = ~(IPW_ASSOCIATED_LED);
priv->led_association_on = IPW_ACTIVITY_LED;
priv->led_association_off = ~(IPW_ACTIVITY_LED);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (!(priv->config & CFG_NO_LED))
ipw_led_band_on(priv);
/* And we don't blink link LEDs for this nic, so
* just return here */
return;
case EEPROM_NIC_TYPE_3:
case EEPROM_NIC_TYPE_2:
case EEPROM_NIC_TYPE_4:
case EEPROM_NIC_TYPE_0:
break;
default:
IPW_DEBUG_INFO("Unknown NIC type from EEPROM: %d\n",
priv->nic_type);
priv->nic_type = EEPROM_NIC_TYPE_0;
break;
}
if (!(priv->config & CFG_NO_LED)) {
if (priv->status & STATUS_ASSOCIATED)
ipw_led_link_on(priv);
else
ipw_led_link_off(priv);
}
}
static void ipw_led_shutdown(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
ipw_led_activity_off(priv);
ipw_led_link_off(priv);
ipw_led_band_off(priv);
cancel_delayed_work(&priv->led_link_on);
cancel_delayed_work(&priv->led_link_off);
cancel_delayed_work(&priv->led_act_off);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
}
2005-03-25 13:31:53 -05:00
/*
* The following adds a new attribute to the sysfs representation
* of this device driver (i.e. a new file in /sys/bus/pci/drivers/ipw/)
* used for controling the debug level.
*
2005-03-25 13:31:53 -05:00
* See the level definitions in ipw for details.
*/
static ssize_t show_debug_level(struct device_driver *d, char *buf)
{
return sprintf(buf, "0x%08X\n", ipw_debug_level);
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
static ssize_t store_debug_level(struct device_driver *d, const char *buf,
size_t count)
2005-03-25 13:31:53 -05:00
{
char *p = (char *)buf;
u32 val;
if (p[1] == 'x' || p[1] == 'X' || p[0] == 'x' || p[0] == 'X') {
p++;
if (p[0] == 'x' || p[0] == 'X')
p++;
val = simple_strtoul(p, &p, 16);
} else
val = simple_strtoul(p, &p, 10);
if (p == buf)
printk(KERN_INFO DRV_NAME
2005-03-25 13:31:53 -05:00
": %s is not in hex or decimal form.\n", buf);
else
ipw_debug_level = val;
return strnlen(buf, count);
}
static DRIVER_ATTR(debug_level, S_IWUSR | S_IRUGO,
2005-03-25 13:31:53 -05:00
show_debug_level, store_debug_level);
static inline u32 ipw_get_event_log_len(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
/* length = 1st dword in log */
return ipw_read_reg32(priv, ipw_read32(priv, IPW_EVENT_LOG));
2005-03-25 13:31:53 -05:00
}
static void ipw_capture_event_log(struct ipw_priv *priv,
u32 log_len, struct ipw_event *log)
2005-03-25 13:31:53 -05:00
{
u32 base;
if (log_len) {
base = ipw_read32(priv, IPW_EVENT_LOG);
ipw_read_indirect(priv, base + sizeof(base) + sizeof(u32),
(u8 *) log, sizeof(*log) * log_len);
}
}
2005-03-25 13:31:53 -05:00
static struct ipw_fw_error *ipw_alloc_error_log(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
struct ipw_fw_error *error;
u32 log_len = ipw_get_event_log_len(priv);
u32 base = ipw_read32(priv, IPW_ERROR_LOG);
u32 elem_len = ipw_read_reg32(priv, base);
2005-03-25 13:31:53 -05:00
error = kmalloc(sizeof(*error) +
sizeof(*error->elem) * elem_len +
sizeof(*error->log) * log_len, GFP_ATOMIC);
if (!error) {
IPW_ERROR("Memory allocation for firmware error log "
"failed.\n");
return NULL;
2005-03-25 13:31:53 -05:00
}
error->jiffies = jiffies;
error->status = priv->status;
error->config = priv->config;
error->elem_len = elem_len;
error->log_len = log_len;
error->elem = (struct ipw_error_elem *)error->payload;
error->log = (struct ipw_event *)(error->elem + elem_len);
ipw_capture_event_log(priv, log_len, error->log);
if (elem_len)
ipw_read_indirect(priv, base + sizeof(base), (u8 *) error->elem,
sizeof(*error->elem) * elem_len);
return error;
2005-03-25 13:31:53 -05:00
}
static void ipw_free_error_log(struct ipw_fw_error *error)
{
if (error)
kfree(error);
}
2005-03-25 13:31:53 -05:00
static ssize_t show_event_log(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = dev_get_drvdata(d);
u32 log_len = ipw_get_event_log_len(priv);
struct ipw_event log[log_len];
u32 len = 0, i;
2005-03-25 13:31:53 -05:00
ipw_capture_event_log(priv, log_len, log);
2005-03-25 13:31:53 -05:00
len += snprintf(buf + len, PAGE_SIZE - len, "%08X", log_len);
for (i = 0; i < log_len; i++)
len += snprintf(buf + len, PAGE_SIZE - len,
"\n%08X%08X%08X",
log[i].time, log[i].event, log[i].data);
len += snprintf(buf + len, PAGE_SIZE - len, "\n");
return len;
2005-03-25 13:31:53 -05:00
}
static DEVICE_ATTR(event_log, S_IRUGO, show_event_log, NULL);
2005-03-25 13:31:53 -05:00
static ssize_t show_error(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = dev_get_drvdata(d);
u32 len = 0, i;
if (!priv->error)
return 0;
len += snprintf(buf + len, PAGE_SIZE - len,
"%08lX%08X%08X%08X",
priv->error->jiffies,
priv->error->status,
priv->error->config, priv->error->elem_len);
for (i = 0; i < priv->error->elem_len; i++)
len += snprintf(buf + len, PAGE_SIZE - len,
"\n%08X%08X%08X%08X%08X%08X%08X",
priv->error->elem[i].time,
priv->error->elem[i].desc,
priv->error->elem[i].blink1,
priv->error->elem[i].blink2,
priv->error->elem[i].link1,
priv->error->elem[i].link2,
priv->error->elem[i].data);
len += snprintf(buf + len, PAGE_SIZE - len,
"\n%08X", priv->error->log_len);
for (i = 0; i < priv->error->log_len; i++)
len += snprintf(buf + len, PAGE_SIZE - len,
"\n%08X%08X%08X",
priv->error->log[i].time,
priv->error->log[i].event,
priv->error->log[i].data);
len += snprintf(buf + len, PAGE_SIZE - len, "\n");
return len;
}
static ssize_t clear_error(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct ipw_priv *priv = dev_get_drvdata(d);
if (priv->error) {
ipw_free_error_log(priv->error);
priv->error = NULL;
}
return count;
}
2005-03-25 13:31:53 -05:00
static DEVICE_ATTR(error, S_IRUGO | S_IWUSR, show_error, clear_error);
2005-03-25 13:31:53 -05:00
static ssize_t show_cmd_log(struct device *d,
struct device_attribute *attr, char *buf)
{
struct ipw_priv *priv = dev_get_drvdata(d);
u32 len = 0, i;
if (!priv->cmdlog)
return 0;
for (i = (priv->cmdlog_pos + 1) % priv->cmdlog_len;
(i != priv->cmdlog_pos) && (PAGE_SIZE - len);
i = (i + 1) % priv->cmdlog_len) {
len +=
snprintf(buf + len, PAGE_SIZE - len,
"\n%08lX%08X%08X%08X\n", priv->cmdlog[i].jiffies,
priv->cmdlog[i].retcode, priv->cmdlog[i].cmd.cmd,
priv->cmdlog[i].cmd.len);
len +=
snprintk_buf(buf + len, PAGE_SIZE - len,
(u8 *) priv->cmdlog[i].cmd.param,
priv->cmdlog[i].cmd.len);
len += snprintf(buf + len, PAGE_SIZE - len, "\n");
}
len += snprintf(buf + len, PAGE_SIZE - len, "\n");
return len;
2005-03-25 13:31:53 -05:00
}
static DEVICE_ATTR(cmd_log, S_IRUGO, show_cmd_log, NULL);
2005-03-25 13:31:53 -05:00
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
static void ipw_prom_free(struct ipw_priv *priv);
static int ipw_prom_alloc(struct ipw_priv *priv);
static ssize_t store_rtap_iface(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct ipw_priv *priv = dev_get_drvdata(d);
int rc = 0;
if (count < 1)
return -EINVAL;
switch (buf[0]) {
case '0':
if (!rtap_iface)
return count;
if (netif_running(priv->prom_net_dev)) {
IPW_WARNING("Interface is up. Cannot unregister.\n");
return count;
}
ipw_prom_free(priv);
rtap_iface = 0;
break;
case '1':
if (rtap_iface)
return count;
rc = ipw_prom_alloc(priv);
if (!rc)
rtap_iface = 1;
break;
default:
return -EINVAL;
}
if (rc) {
IPW_ERROR("Failed to register promiscuous network "
"device (error %d).\n", rc);
}
return count;
}
static ssize_t show_rtap_iface(struct device *d,
struct device_attribute *attr,
char *buf)
{
struct ipw_priv *priv = dev_get_drvdata(d);
if (rtap_iface)
return sprintf(buf, "%s", priv->prom_net_dev->name);
else {
buf[0] = '-';
buf[1] = '1';
buf[2] = '\0';
return 3;
}
}
static DEVICE_ATTR(rtap_iface, S_IWUSR | S_IRUSR, show_rtap_iface,
store_rtap_iface);
static ssize_t store_rtap_filter(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct ipw_priv *priv = dev_get_drvdata(d);
if (!priv->prom_priv) {
IPW_ERROR("Attempting to set filter without "
"rtap_iface enabled.\n");
return -EPERM;
}
priv->prom_priv->filter = simple_strtol(buf, NULL, 0);
IPW_DEBUG_INFO("Setting rtap filter to " BIT_FMT16 "\n",
BIT_ARG16(priv->prom_priv->filter));
return count;
}
static ssize_t show_rtap_filter(struct device *d,
struct device_attribute *attr,
char *buf)
{
struct ipw_priv *priv = dev_get_drvdata(d);
return sprintf(buf, "0x%04X",
priv->prom_priv ? priv->prom_priv->filter : 0);
}
static DEVICE_ATTR(rtap_filter, S_IWUSR | S_IRUSR, show_rtap_filter,
store_rtap_filter);
#endif
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
static ssize_t show_scan_age(struct device *d, struct device_attribute *attr,
char *buf)
2005-03-25 13:31:53 -05:00
{
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
struct ipw_priv *priv = dev_get_drvdata(d);
return sprintf(buf, "%d\n", priv->ieee->scan_age);
}
static ssize_t store_scan_age(struct device *d, struct device_attribute *attr,
const char *buf, size_t count)
{
struct ipw_priv *priv = dev_get_drvdata(d);
#ifdef CONFIG_IPW2200_DEBUG
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
struct net_device *dev = priv->net_dev;
#endif
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
char buffer[] = "00000000";
unsigned long len =
(sizeof(buffer) - 1) > count ? count : sizeof(buffer) - 1;
unsigned long val;
char *p = buffer;
IPW_DEBUG_INFO("enter\n");
strncpy(buffer, buf, len);
buffer[len] = 0;
if (p[1] == 'x' || p[1] == 'X' || p[0] == 'x' || p[0] == 'X') {
p++;
if (p[0] == 'x' || p[0] == 'X')
p++;
val = simple_strtoul(p, &p, 16);
} else
val = simple_strtoul(p, &p, 10);
if (p == buffer) {
IPW_DEBUG_INFO("%s: user supplied invalid value.\n", dev->name);
} else {
priv->ieee->scan_age = val;
IPW_DEBUG_INFO("set scan_age = %u\n", priv->ieee->scan_age);
}
IPW_DEBUG_INFO("exit\n");
return len;
}
static DEVICE_ATTR(scan_age, S_IWUSR | S_IRUGO, show_scan_age, store_scan_age);
static ssize_t show_led(struct device *d, struct device_attribute *attr,
char *buf)
{
struct ipw_priv *priv = dev_get_drvdata(d);
return sprintf(buf, "%d\n", (priv->config & CFG_NO_LED) ? 0 : 1);
}
static ssize_t store_led(struct device *d, struct device_attribute *attr,
const char *buf, size_t count)
{
struct ipw_priv *priv = dev_get_drvdata(d);
IPW_DEBUG_INFO("enter\n");
if (count == 0)
return 0;
if (*buf == 0) {
IPW_DEBUG_LED("Disabling LED control.\n");
priv->config |= CFG_NO_LED;
ipw_led_shutdown(priv);
} else {
IPW_DEBUG_LED("Enabling LED control.\n");
priv->config &= ~CFG_NO_LED;
ipw_led_init(priv);
}
IPW_DEBUG_INFO("exit\n");
return count;
}
static DEVICE_ATTR(led, S_IWUSR | S_IRUGO, show_led, store_led);
static ssize_t show_status(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
return sprintf(buf, "0x%08x\n", (int)p->status);
}
2005-03-25 13:31:53 -05:00
static DEVICE_ATTR(status, S_IRUGO, show_status, NULL);
static ssize_t show_cfg(struct device *d, struct device_attribute *attr,
char *buf)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
return sprintf(buf, "0x%08x\n", (int)p->config);
}
2005-03-25 13:31:53 -05:00
static DEVICE_ATTR(cfg, S_IRUGO, show_cfg, NULL);
static ssize_t show_nic_type(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
struct ipw_priv *priv = d->driver_data;
return sprintf(buf, "TYPE: %d\n", priv->nic_type);
2005-03-25 13:31:53 -05:00
}
2005-03-25 13:31:53 -05:00
static DEVICE_ATTR(nic_type, S_IRUGO, show_nic_type, NULL);
static ssize_t show_ucode_version(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
u32 len = sizeof(u32), tmp = 0;
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
if (ipw_get_ordinal(p, IPW_ORD_STAT_UCODE_VERSION, &tmp, &len))
2005-03-25 13:31:53 -05:00
return 0;
return sprintf(buf, "0x%08x\n", tmp);
}
static DEVICE_ATTR(ucode_version, S_IWUSR | S_IRUGO, show_ucode_version, NULL);
2005-03-25 13:31:53 -05:00
static ssize_t show_rtc(struct device *d, struct device_attribute *attr,
char *buf)
2005-03-25 13:31:53 -05:00
{
u32 len = sizeof(u32), tmp = 0;
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
if (ipw_get_ordinal(p, IPW_ORD_STAT_RTC, &tmp, &len))
2005-03-25 13:31:53 -05:00
return 0;
return sprintf(buf, "0x%08x\n", tmp);
}
static DEVICE_ATTR(rtc, S_IWUSR | S_IRUGO, show_rtc, NULL);
2005-03-25 13:31:53 -05:00
/*
* Add a device attribute to view/control the delay between eeprom
* operations.
*/
static ssize_t show_eeprom_delay(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
int n = ((struct ipw_priv *)d->driver_data)->eeprom_delay;
2005-03-25 13:31:53 -05:00
return sprintf(buf, "%i\n", n);
}
static ssize_t store_eeprom_delay(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
sscanf(buf, "%i", &p->eeprom_delay);
return strnlen(buf, count);
}
static DEVICE_ATTR(eeprom_delay, S_IWUSR | S_IRUGO,
show_eeprom_delay, store_eeprom_delay);
2005-03-25 13:31:53 -05:00
static ssize_t show_command_event_reg(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
u32 reg = 0;
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
reg = ipw_read_reg32(p, IPW_INTERNAL_CMD_EVENT);
2005-03-25 13:31:53 -05:00
return sprintf(buf, "0x%08x\n", reg);
}
static ssize_t store_command_event_reg(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
2005-03-25 13:31:53 -05:00
{
u32 reg;
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
sscanf(buf, "%x", &reg);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(p, IPW_INTERNAL_CMD_EVENT, reg);
2005-03-25 13:31:53 -05:00
return strnlen(buf, count);
}
static DEVICE_ATTR(command_event_reg, S_IWUSR | S_IRUGO,
show_command_event_reg, store_command_event_reg);
2005-03-25 13:31:53 -05:00
static ssize_t show_mem_gpio_reg(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
u32 reg = 0;
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
reg = ipw_read_reg32(p, 0x301100);
return sprintf(buf, "0x%08x\n", reg);
}
static ssize_t store_mem_gpio_reg(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
2005-03-25 13:31:53 -05:00
{
u32 reg;
struct ipw_priv *p = d->driver_data;
2005-03-25 13:31:53 -05:00
sscanf(buf, "%x", &reg);
ipw_write_reg32(p, 0x301100, reg);
return strnlen(buf, count);
}
static DEVICE_ATTR(mem_gpio_reg, S_IWUSR | S_IRUGO,
show_mem_gpio_reg, store_mem_gpio_reg);
2005-03-25 13:31:53 -05:00
static ssize_t show_indirect_dword(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
u32 reg = 0;
struct ipw_priv *priv = d->driver_data;
if (priv->status & STATUS_INDIRECT_DWORD)
2005-03-25 13:31:53 -05:00
reg = ipw_read_reg32(priv, priv->indirect_dword);
else
2005-03-25 13:31:53 -05:00
reg = 0;
2005-03-25 13:31:53 -05:00
return sprintf(buf, "0x%08x\n", reg);
}
static ssize_t store_indirect_dword(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = d->driver_data;
2005-03-25 13:31:53 -05:00
sscanf(buf, "%x", &priv->indirect_dword);
priv->status |= STATUS_INDIRECT_DWORD;
return strnlen(buf, count);
}
static DEVICE_ATTR(indirect_dword, S_IWUSR | S_IRUGO,
show_indirect_dword, store_indirect_dword);
2005-03-25 13:31:53 -05:00
static ssize_t show_indirect_byte(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
u8 reg = 0;
struct ipw_priv *priv = d->driver_data;
if (priv->status & STATUS_INDIRECT_BYTE)
2005-03-25 13:31:53 -05:00
reg = ipw_read_reg8(priv, priv->indirect_byte);
else
2005-03-25 13:31:53 -05:00
reg = 0;
return sprintf(buf, "0x%02x\n", reg);
}
static ssize_t store_indirect_byte(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = d->driver_data;
2005-03-25 13:31:53 -05:00
sscanf(buf, "%x", &priv->indirect_byte);
priv->status |= STATUS_INDIRECT_BYTE;
return strnlen(buf, count);
}
static DEVICE_ATTR(indirect_byte, S_IWUSR | S_IRUGO,
2005-03-25 13:31:53 -05:00
show_indirect_byte, store_indirect_byte);
static ssize_t show_direct_dword(struct device *d,
struct device_attribute *attr, char *buf)
2005-03-25 13:31:53 -05:00
{
u32 reg = 0;
struct ipw_priv *priv = d->driver_data;
2005-03-25 13:31:53 -05:00
if (priv->status & STATUS_DIRECT_DWORD)
2005-03-25 13:31:53 -05:00
reg = ipw_read32(priv, priv->direct_dword);
else
2005-03-25 13:31:53 -05:00
reg = 0;
return sprintf(buf, "0x%08x\n", reg);
}
static ssize_t store_direct_dword(struct device *d,
struct device_attribute *attr,
const char *buf, size_t count)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = d->driver_data;
2005-03-25 13:31:53 -05:00
sscanf(buf, "%x", &priv->direct_dword);
priv->status |= STATUS_DIRECT_DWORD;
return strnlen(buf, count);
}
static DEVICE_ATTR(direct_dword, S_IWUSR | S_IRUGO,
show_direct_dword, store_direct_dword);
2005-03-25 13:31:53 -05:00
static int rf_kill_active(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
if (0 == (ipw_read32(priv, 0x30) & 0x10000))
priv->status |= STATUS_RF_KILL_HW;
else
priv->status &= ~STATUS_RF_KILL_HW;
return (priv->status & STATUS_RF_KILL_HW) ? 1 : 0;
}
static ssize_t show_rf_kill(struct device *d, struct device_attribute *attr,
char *buf)
2005-03-25 13:31:53 -05:00
{
/* 0 - RF kill not enabled
1 - SW based RF kill active (sysfs)
2005-03-25 13:31:53 -05:00
2 - HW based RF kill active
3 - Both HW and SW baed RF kill active */
struct ipw_priv *priv = d->driver_data;
2005-03-25 13:31:53 -05:00
int val = ((priv->status & STATUS_RF_KILL_SW) ? 0x1 : 0x0) |
(rf_kill_active(priv) ? 0x2 : 0x0);
2005-03-25 13:31:53 -05:00
return sprintf(buf, "%i\n", val);
}
static int ipw_radio_kill_sw(struct ipw_priv *priv, int disable_radio)
{
if ((disable_radio ? 1 : 0) ==
((priv->status & STATUS_RF_KILL_SW) ? 1 : 0))
return 0;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_RF_KILL("Manual SW RF Kill set to: RADIO %s\n",
disable_radio ? "OFF" : "ON");
if (disable_radio) {
priv->status |= STATUS_RF_KILL_SW;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (priv->workqueue)
2005-03-25 13:31:53 -05:00
cancel_delayed_work(&priv->request_scan);
queue_work(priv->workqueue, &priv->down);
} else {
priv->status &= ~STATUS_RF_KILL_SW;
if (rf_kill_active(priv)) {
IPW_DEBUG_RF_KILL("Can not turn radio back on - "
"disabled by HW switch\n");
/* Make sure the RF_KILL check timer is running */
cancel_delayed_work(&priv->rf_kill);
queue_delayed_work(priv->workqueue, &priv->rf_kill,
2005-03-25 13:31:53 -05:00
2 * HZ);
} else
2005-03-25 13:31:53 -05:00
queue_work(priv->workqueue, &priv->up);
}
return 1;
}
static ssize_t store_rf_kill(struct device *d, struct device_attribute *attr,
const char *buf, size_t count)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = d->driver_data;
2005-03-25 13:31:53 -05:00
ipw_radio_kill_sw(priv, buf[0] == '1');
return count;
}
static DEVICE_ATTR(rf_kill, S_IWUSR | S_IRUGO, show_rf_kill, store_rf_kill);
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static ssize_t show_speed_scan(struct device *d, struct device_attribute *attr,
char *buf)
{
struct ipw_priv *priv = (struct ipw_priv *)d->driver_data;
int pos = 0, len = 0;
if (priv->config & CFG_SPEED_SCAN) {
while (priv->speed_scan[pos] != 0)
len += sprintf(&buf[len], "%d ",
priv->speed_scan[pos++]);
return len + sprintf(&buf[len], "\n");
}
return sprintf(buf, "0\n");
}
static ssize_t store_speed_scan(struct device *d, struct device_attribute *attr,
const char *buf, size_t count)
{
struct ipw_priv *priv = (struct ipw_priv *)d->driver_data;
int channel, pos = 0;
const char *p = buf;
/* list of space separated channels to scan, optionally ending with 0 */
while ((channel = simple_strtol(p, NULL, 0))) {
if (pos == MAX_SPEED_SCAN - 1) {
priv->speed_scan[pos] = 0;
break;
}
if (ieee80211_is_valid_channel(priv->ieee, channel))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->speed_scan[pos++] = channel;
else
IPW_WARNING("Skipping invalid channel request: %d\n",
channel);
p = strchr(p, ' ');
if (!p)
break;
while (*p == ' ' || *p == '\t')
p++;
}
if (pos == 0)
priv->config &= ~CFG_SPEED_SCAN;
else {
priv->speed_scan_pos = 0;
priv->config |= CFG_SPEED_SCAN;
}
return count;
}
static DEVICE_ATTR(speed_scan, S_IWUSR | S_IRUGO, show_speed_scan,
store_speed_scan);
static ssize_t show_net_stats(struct device *d, struct device_attribute *attr,
char *buf)
{
struct ipw_priv *priv = (struct ipw_priv *)d->driver_data;
return sprintf(buf, "%c\n", (priv->config & CFG_NET_STATS) ? '1' : '0');
}
static ssize_t store_net_stats(struct device *d, struct device_attribute *attr,
const char *buf, size_t count)
{
struct ipw_priv *priv = (struct ipw_priv *)d->driver_data;
if (buf[0] == '1')
priv->config |= CFG_NET_STATS;
else
priv->config &= ~CFG_NET_STATS;
return count;
}
static DEVICE_ATTR(net_stats, S_IWUSR | S_IRUGO,
show_net_stats, store_net_stats);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void notify_wx_assoc_event(struct ipw_priv *priv)
{
union iwreq_data wrqu;
wrqu.ap_addr.sa_family = ARPHRD_ETHER;
if (priv->status & STATUS_ASSOCIATED)
memcpy(wrqu.ap_addr.sa_data, priv->bssid, ETH_ALEN);
else
memset(wrqu.ap_addr.sa_data, 0, ETH_ALEN);
wireless_send_event(priv->net_dev, SIOCGIWAP, &wrqu, NULL);
}
2005-03-25 13:31:53 -05:00
static void ipw_irq_tasklet(struct ipw_priv *priv)
{
u32 inta, inta_mask, handled = 0;
unsigned long flags;
int rc = 0;
spin_lock_irqsave(&priv->lock, flags);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
inta = ipw_read32(priv, IPW_INTA_RW);
inta_mask = ipw_read32(priv, IPW_INTA_MASK_R);
inta &= (IPW_INTA_MASK_ALL & inta_mask);
2005-03-25 13:31:53 -05:00
/* Add any cached INTA values that need to be handled */
inta |= priv->isr_inta;
/* handle all the justifications for the interrupt */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_RX_TRANSFER) {
2005-03-25 13:31:53 -05:00
ipw_rx(priv);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_RX_TRANSFER;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_TX_CMD_QUEUE) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_HC("Command completed.\n");
rc = ipw_queue_tx_reclaim(priv, &priv->txq_cmd, -1);
2005-03-25 13:31:53 -05:00
priv->status &= ~STATUS_HCMD_ACTIVE;
wake_up_interruptible(&priv->wait_command_queue);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_TX_CMD_QUEUE;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_TX_QUEUE_1) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TX("TX_QUEUE_1\n");
rc = ipw_queue_tx_reclaim(priv, &priv->txq[0], 0);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_TX_QUEUE_1;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_TX_QUEUE_2) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TX("TX_QUEUE_2\n");
rc = ipw_queue_tx_reclaim(priv, &priv->txq[1], 1);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_TX_QUEUE_2;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_TX_QUEUE_3) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TX("TX_QUEUE_3\n");
rc = ipw_queue_tx_reclaim(priv, &priv->txq[2], 2);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_TX_QUEUE_3;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_TX_QUEUE_4) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TX("TX_QUEUE_4\n");
rc = ipw_queue_tx_reclaim(priv, &priv->txq[3], 3);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_TX_QUEUE_4;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_STATUS_CHANGE) {
2005-03-25 13:31:53 -05:00
IPW_WARNING("STATUS_CHANGE\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_STATUS_CHANGE;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_BEACON_PERIOD_EXPIRED) {
2005-03-25 13:31:53 -05:00
IPW_WARNING("TX_PERIOD_EXPIRED\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_BEACON_PERIOD_EXPIRED;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_SLAVE_MODE_HOST_CMD_DONE) {
2005-03-25 13:31:53 -05:00
IPW_WARNING("HOST_CMD_DONE\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_SLAVE_MODE_HOST_CMD_DONE;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_FW_INITIALIZATION_DONE) {
2005-03-25 13:31:53 -05:00
IPW_WARNING("FW_INITIALIZATION_DONE\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_FW_INITIALIZATION_DONE;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_FW_CARD_DISABLE_PHY_OFF_DONE) {
2005-03-25 13:31:53 -05:00
IPW_WARNING("PHY_OFF_DONE\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_FW_CARD_DISABLE_PHY_OFF_DONE;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_RF_KILL_DONE) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_RF_KILL("RF_KILL_DONE\n");
priv->status |= STATUS_RF_KILL_HW;
wake_up_interruptible(&priv->wait_command_queue);
priv->status &= ~(STATUS_ASSOCIATED | STATUS_ASSOCIATING);
2005-03-25 13:31:53 -05:00
cancel_delayed_work(&priv->request_scan);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
schedule_work(&priv->link_down);
2005-03-25 13:31:53 -05:00
queue_delayed_work(priv->workqueue, &priv->rf_kill, 2 * HZ);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_RF_KILL_DONE;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_FATAL_ERROR) {
IPW_WARNING("Firmware error detected. Restarting.\n");
if (priv->error) {
IPW_DEBUG_FW("Sysfs 'error' log already exists.\n");
#ifdef CONFIG_IPW2200_DEBUG
if (ipw_debug_level & IPW_DL_FW_ERRORS) {
struct ipw_fw_error *error =
ipw_alloc_error_log(priv);
ipw_dump_error_log(priv, error);
if (error)
ipw_free_error_log(error);
}
#endif
} else {
priv->error = ipw_alloc_error_log(priv);
if (priv->error)
IPW_DEBUG_FW("Sysfs 'error' log captured.\n");
else
IPW_DEBUG_FW("Error allocating sysfs 'error' "
"log.\n");
#ifdef CONFIG_IPW2200_DEBUG
if (ipw_debug_level & IPW_DL_FW_ERRORS)
ipw_dump_error_log(priv, priv->error);
2005-03-25 13:31:53 -05:00
#endif
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* XXX: If hardware encryption is for WPA/WPA2,
* we have to notify the supplicant. */
if (priv->ieee->sec.encrypt) {
priv->status &= ~STATUS_ASSOCIATED;
notify_wx_assoc_event(priv);
}
/* Keep the restart process from trying to send host
* commands by clearing the INIT status bit */
priv->status &= ~STATUS_INIT;
/* Cancel currently queued command. */
priv->status &= ~STATUS_HCMD_ACTIVE;
wake_up_interruptible(&priv->wait_command_queue);
2005-03-25 13:31:53 -05:00
queue_work(priv->workqueue, &priv->adapter_restart);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_FATAL_ERROR;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (inta & IPW_INTA_BIT_PARITY_ERROR) {
2005-03-25 13:31:53 -05:00
IPW_ERROR("Parity error\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
handled |= IPW_INTA_BIT_PARITY_ERROR;
2005-03-25 13:31:53 -05:00
}
if (handled != inta) {
IPW_ERROR("Unhandled INTA bits 0x%08x\n", inta & ~handled);
2005-03-25 13:31:53 -05:00
}
/* enable all interrupts */
ipw_enable_interrupts(priv);
spin_unlock_irqrestore(&priv->lock, flags);
}
2005-03-25 13:31:53 -05:00
#define IPW_CMD(x) case IPW_CMD_ ## x : return #x
static char *get_cmd_string(u8 cmd)
{
switch (cmd) {
IPW_CMD(HOST_COMPLETE);
IPW_CMD(POWER_DOWN);
IPW_CMD(SYSTEM_CONFIG);
IPW_CMD(MULTICAST_ADDRESS);
IPW_CMD(SSID);
IPW_CMD(ADAPTER_ADDRESS);
IPW_CMD(PORT_TYPE);
IPW_CMD(RTS_THRESHOLD);
IPW_CMD(FRAG_THRESHOLD);
IPW_CMD(POWER_MODE);
IPW_CMD(WEP_KEY);
IPW_CMD(TGI_TX_KEY);
IPW_CMD(SCAN_REQUEST);
IPW_CMD(SCAN_REQUEST_EXT);
IPW_CMD(ASSOCIATE);
IPW_CMD(SUPPORTED_RATES);
IPW_CMD(SCAN_ABORT);
IPW_CMD(TX_FLUSH);
IPW_CMD(QOS_PARAMETERS);
IPW_CMD(DINO_CONFIG);
IPW_CMD(RSN_CAPABILITIES);
IPW_CMD(RX_KEY);
IPW_CMD(CARD_DISABLE);
IPW_CMD(SEED_NUMBER);
IPW_CMD(TX_POWER);
IPW_CMD(COUNTRY_INFO);
IPW_CMD(AIRONET_INFO);
IPW_CMD(AP_TX_POWER);
IPW_CMD(CCKM_INFO);
IPW_CMD(CCX_VER_INFO);
IPW_CMD(SET_CALIBRATION);
IPW_CMD(SENSITIVITY_CALIB);
IPW_CMD(RETRY_LIMIT);
IPW_CMD(IPW_PRE_POWER_DOWN);
IPW_CMD(VAP_BEACON_TEMPLATE);
IPW_CMD(VAP_DTIM_PERIOD);
IPW_CMD(EXT_SUPPORTED_RATES);
IPW_CMD(VAP_LOCAL_TX_PWR_CONSTRAINT);
IPW_CMD(VAP_QUIET_INTERVALS);
IPW_CMD(VAP_CHANNEL_SWITCH);
IPW_CMD(VAP_MANDATORY_CHANNELS);
IPW_CMD(VAP_CELL_PWR_LIMIT);
IPW_CMD(VAP_CF_PARAM_SET);
IPW_CMD(VAP_SET_BEACONING_STATE);
IPW_CMD(MEASUREMENT);
IPW_CMD(POWER_CAPABILITY);
IPW_CMD(SUPPORTED_CHANNELS);
IPW_CMD(TPC_REPORT);
IPW_CMD(WME_INFO);
IPW_CMD(PRODUCTION_COMMAND);
default:
2005-03-25 13:31:53 -05:00
return "UNKNOWN";
}
}
#define HOST_COMPLETE_TIMEOUT HZ
static int __ipw_send_cmd(struct ipw_priv *priv, struct host_cmd *cmd)
2005-03-25 13:31:53 -05:00
{
int rc = 0;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
unsigned long flags;
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_lock_irqsave(&priv->lock, flags);
2005-03-25 13:31:53 -05:00
if (priv->status & STATUS_HCMD_ACTIVE) {
IPW_ERROR("Failed to send %s: Already sending a command.\n",
get_cmd_string(cmd->cmd));
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->lock, flags);
return -EAGAIN;
2005-03-25 13:31:53 -05:00
}
priv->status |= STATUS_HCMD_ACTIVE;
if (priv->cmdlog) {
priv->cmdlog[priv->cmdlog_pos].jiffies = jiffies;
priv->cmdlog[priv->cmdlog_pos].cmd.cmd = cmd->cmd;
priv->cmdlog[priv->cmdlog_pos].cmd.len = cmd->len;
memcpy(priv->cmdlog[priv->cmdlog_pos].cmd.param, cmd->param,
cmd->len);
priv->cmdlog[priv->cmdlog_pos].retcode = -1;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_DEBUG_HC("%s command (#%d) %d bytes: 0x%08X\n",
get_cmd_string(cmd->cmd), cmd->cmd, cmd->len,
priv->status);
#ifndef DEBUG_CMD_WEP_KEY
if (cmd->cmd == IPW_CMD_WEP_KEY)
IPW_DEBUG_HC("WEP_KEY command masked out for secure.\n");
else
#endif
printk_buf(IPW_DL_HOST_COMMAND, (u8 *) cmd->param, cmd->len);
rc = ipw_queue_tx_hcmd(priv, cmd->cmd, cmd->param, cmd->len, 0);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (rc) {
priv->status &= ~STATUS_HCMD_ACTIVE;
IPW_ERROR("Failed to send %s: Reason %d\n",
get_cmd_string(cmd->cmd), rc);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->lock, flags);
goto exit;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
}
spin_unlock_irqrestore(&priv->lock, flags);
2005-03-25 13:31:53 -05:00
rc = wait_event_interruptible_timeout(priv->wait_command_queue,
!(priv->
status & STATUS_HCMD_ACTIVE),
HOST_COMPLETE_TIMEOUT);
2005-03-25 13:31:53 -05:00
if (rc == 0) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_lock_irqsave(&priv->lock, flags);
if (priv->status & STATUS_HCMD_ACTIVE) {
IPW_ERROR("Failed to send %s: Command timed out.\n",
get_cmd_string(cmd->cmd));
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
priv->status &= ~STATUS_HCMD_ACTIVE;
spin_unlock_irqrestore(&priv->lock, flags);
rc = -EIO;
goto exit;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
}
spin_unlock_irqrestore(&priv->lock, flags);
} else
rc = 0;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->status & STATUS_RF_KILL_HW) {
IPW_ERROR("Failed to send %s: Aborted due to RF kill switch.\n",
get_cmd_string(cmd->cmd));
rc = -EIO;
goto exit;
2005-03-25 13:31:53 -05:00
}
exit:
if (priv->cmdlog) {
priv->cmdlog[priv->cmdlog_pos++].retcode = rc;
priv->cmdlog_pos %= priv->cmdlog_len;
}
return rc;
2005-03-25 13:31:53 -05:00
}
static int ipw_send_cmd_simple(struct ipw_priv *priv, u8 command)
{
struct host_cmd cmd = {
.cmd = command,
};
return __ipw_send_cmd(priv, &cmd);
}
static int ipw_send_cmd_pdu(struct ipw_priv *priv, u8 command, u8 len,
void *data)
2005-03-25 13:31:53 -05:00
{
struct host_cmd cmd = {
.cmd = command,
.len = len,
.param = data,
2005-03-25 13:31:53 -05:00
};
return __ipw_send_cmd(priv, &cmd);
}
static int ipw_send_host_complete(struct ipw_priv *priv)
{
2005-03-25 13:31:53 -05:00
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_simple(priv, IPW_CMD_HOST_COMPLETE);
2005-03-25 13:31:53 -05:00
}
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
static int ipw_send_system_config(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
return ipw_send_cmd_pdu(priv, IPW_CMD_SYSTEM_CONFIG,
sizeof(priv->sys_config),
&priv->sys_config);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_ssid(struct ipw_priv *priv, u8 * ssid, int len)
2005-03-25 13:31:53 -05:00
{
if (!priv || !ssid) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_SSID, min(len, IW_ESSID_MAX_SIZE),
ssid);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_adapter_address(struct ipw_priv *priv, u8 * mac)
2005-03-25 13:31:53 -05:00
{
if (!priv || !mac) {
IPW_ERROR("Invalid args\n");
return -1;
}
IPW_DEBUG_INFO("%s: Setting MAC to " MAC_FMT "\n",
priv->net_dev->name, MAC_ARG(mac));
return ipw_send_cmd_pdu(priv, IPW_CMD_ADAPTER_ADDRESS, ETH_ALEN, mac);
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
/*
* NOTE: This must be executed from our workqueue as it results in udelay
* being called which may corrupt the keyboard if executed on default
* workqueue
*/
2005-03-25 13:31:53 -05:00
static void ipw_adapter_restart(void *adapter)
{
struct ipw_priv *priv = adapter;
if (priv->status & STATUS_RF_KILL_MASK)
return;
ipw_down(priv);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->assoc_network &&
(priv->assoc_network->capability & WLAN_CAPABILITY_IBSS))
ipw_remove_current_network(priv);
2005-03-25 13:31:53 -05:00
if (ipw_up(priv)) {
IPW_ERROR("Failed to up device\n");
return;
}
}
static void ipw_bg_adapter_restart(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_adapter_restart(data);
mutex_unlock(&priv->mutex);
}
2005-03-25 13:31:53 -05:00
#define IPW_SCAN_CHECK_WATCHDOG (5 * HZ)
static void ipw_scan_check(void *data)
{
struct ipw_priv *priv = data;
if (priv->status & (STATUS_SCANNING | STATUS_SCAN_ABORTING)) {
IPW_DEBUG_SCAN("Scan completion watchdog resetting "
"adapter after (%dms).\n",
jiffies_to_msecs(IPW_SCAN_CHECK_WATCHDOG));
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
queue_work(priv->workqueue, &priv->adapter_restart);
2005-03-25 13:31:53 -05:00
}
}
static void ipw_bg_scan_check(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_scan_check(data);
mutex_unlock(&priv->mutex);
}
2005-03-25 13:31:53 -05:00
static int ipw_send_scan_request_ext(struct ipw_priv *priv,
struct ipw_scan_request_ext *request)
{
return ipw_send_cmd_pdu(priv, IPW_CMD_SCAN_REQUEST_EXT,
sizeof(*request), request);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_scan_abort(struct ipw_priv *priv)
{
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_simple(priv, IPW_CMD_SCAN_ABORT);
2005-03-25 13:31:53 -05:00
}
static int ipw_set_sensitivity(struct ipw_priv *priv, u16 sens)
{
struct ipw_sensitivity_calib calib = {
.beacon_rssi_raw = sens,
2005-03-25 13:31:53 -05:00
};
return ipw_send_cmd_pdu(priv, IPW_CMD_SENSITIVITY_CALIB, sizeof(calib),
&calib);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_associate(struct ipw_priv *priv,
struct ipw_associate *associate)
{
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
struct ipw_associate tmp_associate;
if (!priv || !associate) {
IPW_ERROR("Invalid args\n");
return -1;
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
memcpy(&tmp_associate, associate, sizeof(*associate));
tmp_associate.policy_support =
cpu_to_le16(tmp_associate.policy_support);
tmp_associate.assoc_tsf_msw = cpu_to_le32(tmp_associate.assoc_tsf_msw);
tmp_associate.assoc_tsf_lsw = cpu_to_le32(tmp_associate.assoc_tsf_lsw);
tmp_associate.capability = cpu_to_le16(tmp_associate.capability);
tmp_associate.listen_interval =
cpu_to_le16(tmp_associate.listen_interval);
tmp_associate.beacon_interval =
cpu_to_le16(tmp_associate.beacon_interval);
tmp_associate.atim_window = cpu_to_le16(tmp_associate.atim_window);
return ipw_send_cmd_pdu(priv, IPW_CMD_ASSOCIATE, sizeof(tmp_associate),
&tmp_associate);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_supported_rates(struct ipw_priv *priv,
struct ipw_supported_rates *rates)
{
if (!priv || !rates) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_SUPPORTED_RATES, sizeof(*rates),
rates);
2005-03-25 13:31:53 -05:00
}
static int ipw_set_random_seed(struct ipw_priv *priv)
{
u32 val;
2005-03-25 13:31:53 -05:00
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
get_random_bytes(&val, sizeof(val));
2005-03-25 13:31:53 -05:00
return ipw_send_cmd_pdu(priv, IPW_CMD_SEED_NUMBER, sizeof(val), &val);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_card_disable(struct ipw_priv *priv, u32 phy_off)
{
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_CARD_DISABLE, sizeof(phy_off),
&phy_off);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_tx_power(struct ipw_priv *priv, struct ipw_tx_power *power)
2005-03-25 13:31:53 -05:00
{
if (!priv || !power) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_TX_POWER, sizeof(*power), power);
2005-03-25 13:31:53 -05:00
}
static int ipw_set_tx_power(struct ipw_priv *priv)
{
const struct ieee80211_geo *geo = ieee80211_get_geo(priv->ieee);
struct ipw_tx_power tx_power;
s8 max_power;
int i;
memset(&tx_power, 0, sizeof(tx_power));
/* configure device for 'G' band */
tx_power.ieee_mode = IPW_G_MODE;
tx_power.num_channels = geo->bg_channels;
for (i = 0; i < geo->bg_channels; i++) {
max_power = geo->bg[i].max_power;
tx_power.channels_tx_power[i].channel_number =
geo->bg[i].channel;
tx_power.channels_tx_power[i].tx_power = max_power ?
min(max_power, priv->tx_power) : priv->tx_power;
2005-03-25 13:31:53 -05:00
}
if (ipw_send_tx_power(priv, &tx_power))
return -EIO;
/* configure device to also handle 'B' band */
tx_power.ieee_mode = IPW_B_MODE;
if (ipw_send_tx_power(priv, &tx_power))
return -EIO;
/* configure device to also handle 'A' band */
if (priv->ieee->abg_true) {
tx_power.ieee_mode = IPW_A_MODE;
tx_power.num_channels = geo->a_channels;
for (i = 0; i < tx_power.num_channels; i++) {
max_power = geo->a[i].max_power;
tx_power.channels_tx_power[i].channel_number =
geo->a[i].channel;
tx_power.channels_tx_power[i].tx_power = max_power ?
min(max_power, priv->tx_power) : priv->tx_power;
}
if (ipw_send_tx_power(priv, &tx_power))
return -EIO;
}
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_send_rts_threshold(struct ipw_priv *priv, u16 rts)
{
struct ipw_rts_threshold rts_threshold = {
.rts_threshold = rts,
};
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_RTS_THRESHOLD,
sizeof(rts_threshold), &rts_threshold);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_frag_threshold(struct ipw_priv *priv, u16 frag)
{
struct ipw_frag_threshold frag_threshold = {
.frag_threshold = frag,
};
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_FRAG_THRESHOLD,
sizeof(frag_threshold), &frag_threshold);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_power_mode(struct ipw_priv *priv, u32 mode)
{
u32 param;
2005-03-25 13:31:53 -05:00
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
2005-03-25 13:31:53 -05:00
/* If on battery, set to 3, if AC set to CAM, else user
* level */
switch (mode) {
case IPW_POWER_BATTERY:
param = IPW_POWER_INDEX_3;
2005-03-25 13:31:53 -05:00
break;
case IPW_POWER_AC:
param = IPW_POWER_MODE_CAM;
2005-03-25 13:31:53 -05:00
break;
default:
param = mode;
2005-03-25 13:31:53 -05:00
break;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_POWER_MODE, sizeof(param),
&param);
2005-03-25 13:31:53 -05:00
}
static int ipw_send_retry_limit(struct ipw_priv *priv, u8 slimit, u8 llimit)
{
struct ipw_retry_limit retry_limit = {
.short_retry_limit = slimit,
.long_retry_limit = llimit
};
if (!priv) {
IPW_ERROR("Invalid args\n");
return -1;
}
return ipw_send_cmd_pdu(priv, IPW_CMD_RETRY_LIMIT, sizeof(retry_limit),
&retry_limit);
}
2005-03-25 13:31:53 -05:00
/*
* The IPW device contains a Microwire compatible EEPROM that stores
* various data like the MAC address. Usually the firmware has exclusive
* access to the eeprom, but during device initialization (before the
* device driver has sent the HostComplete command to the firmware) the
* device driver has read access to the EEPROM by way of indirect addressing
* through a couple of memory mapped registers.
*
* The following is a simplified implementation for pulling data out of the
* the eeprom, along with some helper functions to find information in
* the per device private data's copy of the eeprom.
*
* NOTE: To better understand how these functions work (i.e what is a chip
* select and why do have to keep driving the eeprom clock?), read
* just about any data sheet for a Microwire compatible EEPROM.
*/
/* write a 32 bit value into the indirect accessor register */
static inline void eeprom_write_reg(struct ipw_priv *p, u32 data)
{
ipw_write_reg32(p, FW_MEM_REG_EEPROM_ACCESS, data);
2005-03-25 13:31:53 -05:00
/* the eeprom requires some time to complete the operation */
udelay(p->eeprom_delay);
return;
}
/* perform a chip select operation */
static void eeprom_cs(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
eeprom_write_reg(priv, 0);
eeprom_write_reg(priv, EEPROM_BIT_CS);
eeprom_write_reg(priv, EEPROM_BIT_CS | EEPROM_BIT_SK);
eeprom_write_reg(priv, EEPROM_BIT_CS);
2005-03-25 13:31:53 -05:00
}
/* perform a chip select operation */
static void eeprom_disable_cs(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
eeprom_write_reg(priv, EEPROM_BIT_CS);
eeprom_write_reg(priv, 0);
eeprom_write_reg(priv, EEPROM_BIT_SK);
2005-03-25 13:31:53 -05:00
}
/* push a single bit down to the eeprom */
static inline void eeprom_write_bit(struct ipw_priv *p, u8 bit)
2005-03-25 13:31:53 -05:00
{
int d = (bit ? EEPROM_BIT_DI : 0);
eeprom_write_reg(p, EEPROM_BIT_CS | d);
eeprom_write_reg(p, EEPROM_BIT_CS | d | EEPROM_BIT_SK);
2005-03-25 13:31:53 -05:00
}
/* push an opcode followed by an address down to the eeprom */
static void eeprom_op(struct ipw_priv *priv, u8 op, u8 addr)
2005-03-25 13:31:53 -05:00
{
int i;
eeprom_cs(priv);
eeprom_write_bit(priv, 1);
eeprom_write_bit(priv, op & 2);
eeprom_write_bit(priv, op & 1);
for (i = 7; i >= 0; i--) {
eeprom_write_bit(priv, addr & (1 << i));
2005-03-25 13:31:53 -05:00
}
}
/* pull 16 bits off the eeprom, one bit at a time */
static u16 eeprom_read_u16(struct ipw_priv *priv, u8 addr)
2005-03-25 13:31:53 -05:00
{
int i;
u16 r = 0;
2005-03-25 13:31:53 -05:00
/* Send READ Opcode */
eeprom_op(priv, EEPROM_CMD_READ, addr);
2005-03-25 13:31:53 -05:00
/* Send dummy bit */
eeprom_write_reg(priv, EEPROM_BIT_CS);
2005-03-25 13:31:53 -05:00
/* Read the byte off the eeprom one bit at a time */
for (i = 0; i < 16; i++) {
2005-03-25 13:31:53 -05:00
u32 data = 0;
eeprom_write_reg(priv, EEPROM_BIT_CS | EEPROM_BIT_SK);
eeprom_write_reg(priv, EEPROM_BIT_CS);
data = ipw_read_reg32(priv, FW_MEM_REG_EEPROM_ACCESS);
r = (r << 1) | ((data & EEPROM_BIT_DO) ? 1 : 0);
2005-03-25 13:31:53 -05:00
}
2005-03-25 13:31:53 -05:00
/* Send another dummy bit */
eeprom_write_reg(priv, 0);
2005-03-25 13:31:53 -05:00
eeprom_disable_cs(priv);
2005-03-25 13:31:53 -05:00
return r;
}
/* helper function for pulling the mac address out of the private */
/* data's copy of the eeprom data */
static void eeprom_parse_mac(struct ipw_priv *priv, u8 * mac)
2005-03-25 13:31:53 -05:00
{
memcpy(mac, &priv->eeprom[EEPROM_MAC_ADDRESS], 6);
2005-03-25 13:31:53 -05:00
}
/*
* Either the device driver (i.e. the host) or the firmware can
* load eeprom data into the designated region in SRAM. If neither
* happens then the FW will shutdown with a fatal error.
*
* In order to signal the FW to load the EEPROM, the EEPROM_LOAD_DISABLE
* bit needs region of shared SRAM needs to be non-zero.
*/
static void ipw_eeprom_init_sram(struct ipw_priv *priv)
{
int i;
u16 *eeprom = (u16 *) priv->eeprom;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE(">>\n");
/* read entire contents of eeprom into private buffer */
for (i = 0; i < 128; i++)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
eeprom[i] = le16_to_cpu(eeprom_read_u16(priv, (u8) i));
2005-03-25 13:31:53 -05:00
/*
If the data looks correct, then copy it to our private
2005-03-25 13:31:53 -05:00
copy. Otherwise let the firmware know to perform the operation
on its own.
*/
if (priv->eeprom[EEPROM_VERSION] != 0) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_INFO("Writing EEPROM data into SRAM\n");
/* write the eeprom data to sram */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
for (i = 0; i < IPW_EEPROM_IMAGE_SIZE; i++)
ipw_write8(priv, IPW_EEPROM_DATA + i, priv->eeprom[i]);
2005-03-25 13:31:53 -05:00
/* Do not load eeprom data on fatal error or suspend */
ipw_write32(priv, IPW_EEPROM_LOAD_DISABLE, 0);
} else {
IPW_DEBUG_INFO("Enabling FW initializationg of SRAM\n");
/* Load eeprom data on fatal error or suspend */
ipw_write32(priv, IPW_EEPROM_LOAD_DISABLE, 1);
}
IPW_DEBUG_TRACE("<<\n");
}
static void ipw_zero_memory(struct ipw_priv *priv, u32 start, u32 count)
2005-03-25 13:31:53 -05:00
{
count >>= 2;
if (!count)
return;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_AUTOINC_ADDR, start);
while (count--)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
_ipw_write32(priv, IPW_AUTOINC_DATA, 0);
2005-03-25 13:31:53 -05:00
}
static inline void ipw_fw_dma_reset_command_blocks(struct ipw_priv *priv)
{
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_zero_memory(priv, IPW_SHARED_SRAM_DMA_CONTROL,
CB_NUMBER_OF_ELEMENTS_SMALL *
2005-03-25 13:31:53 -05:00
sizeof(struct command_block));
}
static int ipw_fw_dma_enable(struct ipw_priv *priv)
{ /* start dma engine but no transfers yet */
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW(">> : \n");
2005-03-25 13:31:53 -05:00
/* Start the dma */
ipw_fw_dma_reset_command_blocks(priv);
2005-03-25 13:31:53 -05:00
/* Write CB base address */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_DMA_I_CB_BASE, IPW_SHARED_SRAM_DMA_CONTROL);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW("<< : \n");
return 0;
}
static void ipw_fw_dma_abort(struct ipw_priv *priv)
{
u32 control = 0;
IPW_DEBUG_FW(">> :\n");
//set the Stop and Abort bit
2005-03-25 13:31:53 -05:00
control = DMA_CONTROL_SMALL_CB_CONST_VALUE | DMA_CB_STOP_AND_ABORT;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_DMA_I_DMA_CONTROL, control);
2005-03-25 13:31:53 -05:00
priv->sram_desc.last_cb_index = 0;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW("<< \n");
}
static int ipw_fw_dma_write_command_block(struct ipw_priv *priv, int index,
struct command_block *cb)
2005-03-25 13:31:53 -05:00
{
u32 address =
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_SHARED_SRAM_DMA_CONTROL +
(sizeof(struct command_block) * index);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW(">> :\n");
ipw_write_indirect(priv, address, (u8 *) cb,
(int)sizeof(struct command_block));
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW("<< :\n");
return 0;
}
static int ipw_fw_dma_kick(struct ipw_priv *priv)
{
u32 control = 0;
u32 index = 0;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW(">> :\n");
2005-03-25 13:31:53 -05:00
for (index = 0; index < priv->sram_desc.last_cb_index; index++)
ipw_fw_dma_write_command_block(priv, index,
&priv->sram_desc.cb_list[index]);
2005-03-25 13:31:53 -05:00
/* Enable the DMA in the CSR register */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_clear_bit(priv, IPW_RESET_REG,
IPW_RESET_REG_MASTER_DISABLED |
IPW_RESET_REG_STOP_MASTER);
/* Set the Start bit. */
2005-03-25 13:31:53 -05:00
control = DMA_CONTROL_SMALL_CB_CONST_VALUE | DMA_CB_START;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_DMA_I_DMA_CONTROL, control);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW("<< :\n");
return 0;
}
static void ipw_fw_dma_dump_command_block(struct ipw_priv *priv)
{
u32 address;
u32 register_value = 0;
u32 cb_fields_address = 0;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW(">> :\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
address = ipw_read_reg32(priv, IPW_DMA_I_CURRENT_CB);
IPW_DEBUG_FW_INFO("Current CB is 0x%x \n", address);
2005-03-25 13:31:53 -05:00
/* Read the DMA Controlor register */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
register_value = ipw_read_reg32(priv, IPW_DMA_I_DMA_CONTROL);
IPW_DEBUG_FW_INFO("IPW_DMA_I_DMA_CONTROL is 0x%x \n", register_value);
2005-03-25 13:31:53 -05:00
/* Print the CB values */
2005-03-25 13:31:53 -05:00
cb_fields_address = address;
register_value = ipw_read_reg32(priv, cb_fields_address);
IPW_DEBUG_FW_INFO("Current CB ControlField is 0x%x \n", register_value);
2005-03-25 13:31:53 -05:00
cb_fields_address += sizeof(u32);
register_value = ipw_read_reg32(priv, cb_fields_address);
IPW_DEBUG_FW_INFO("Current CB Source Field is 0x%x \n", register_value);
2005-03-25 13:31:53 -05:00
cb_fields_address += sizeof(u32);
register_value = ipw_read_reg32(priv, cb_fields_address);
IPW_DEBUG_FW_INFO("Current CB Destination Field is 0x%x \n",
register_value);
cb_fields_address += sizeof(u32);
register_value = ipw_read_reg32(priv, cb_fields_address);
IPW_DEBUG_FW_INFO("Current CB Status Field is 0x%x \n", register_value);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW(">> :\n");
}
static int ipw_fw_dma_command_block_index(struct ipw_priv *priv)
{
u32 current_cb_address = 0;
u32 current_cb_index = 0;
IPW_DEBUG_FW("<< :\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
current_cb_address = ipw_read_reg32(priv, IPW_DMA_I_CURRENT_CB);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
current_cb_index = (current_cb_address - IPW_SHARED_SRAM_DMA_CONTROL) /
sizeof(struct command_block);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW_INFO("Current CB index 0x%x address = 0x%X \n",
current_cb_index, current_cb_address);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW(">> :\n");
return current_cb_index;
}
static int ipw_fw_dma_add_command_block(struct ipw_priv *priv,
u32 src_address,
u32 dest_address,
u32 length,
int interrupt_enabled, int is_last)
2005-03-25 13:31:53 -05:00
{
u32 control = CB_VALID | CB_SRC_LE | CB_DEST_LE | CB_SRC_AUTOINC |
CB_SRC_IO_GATED | CB_DEST_AUTOINC | CB_SRC_SIZE_LONG |
CB_DEST_SIZE_LONG;
2005-03-25 13:31:53 -05:00
struct command_block *cb;
u32 last_cb_element = 0;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW_INFO("src_address=0x%x dest_address=0x%x length=0x%x\n",
src_address, dest_address, length);
if (priv->sram_desc.last_cb_index >= CB_NUMBER_OF_ELEMENTS_SMALL)
return -1;
last_cb_element = priv->sram_desc.last_cb_index;
cb = &priv->sram_desc.cb_list[last_cb_element];
priv->sram_desc.last_cb_index++;
/* Calculate the new CB control word */
if (interrupt_enabled)
2005-03-25 13:31:53 -05:00
control |= CB_INT_ENABLED;
if (is_last)
control |= CB_LAST_VALID;
2005-03-25 13:31:53 -05:00
control |= length;
/* Calculate the CB Element's checksum value */
cb->status = control ^ src_address ^ dest_address;
2005-03-25 13:31:53 -05:00
/* Copy the Source and Destination addresses */
cb->dest_addr = dest_address;
cb->source_addr = src_address;
/* Copy the Control Word last */
cb->control = control;
return 0;
}
static int ipw_fw_dma_add_buffer(struct ipw_priv *priv,
u32 src_phys, u32 dest_address, u32 length)
2005-03-25 13:31:53 -05:00
{
u32 bytes_left = length;
u32 src_offset = 0;
u32 dest_offset = 0;
2005-03-25 13:31:53 -05:00
int status = 0;
IPW_DEBUG_FW(">> \n");
IPW_DEBUG_FW_INFO("src_phys=0x%x dest_address=0x%x length=0x%x\n",
src_phys, dest_address, length);
while (bytes_left > CB_MAX_LENGTH) {
status = ipw_fw_dma_add_command_block(priv,
src_phys + src_offset,
dest_address +
dest_offset,
CB_MAX_LENGTH, 0, 0);
2005-03-25 13:31:53 -05:00
if (status) {
IPW_DEBUG_FW_INFO(": Failed\n");
return -1;
} else
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW_INFO(": Added new cb\n");
src_offset += CB_MAX_LENGTH;
dest_offset += CB_MAX_LENGTH;
bytes_left -= CB_MAX_LENGTH;
}
/* add the buffer tail */
if (bytes_left > 0) {
status =
ipw_fw_dma_add_command_block(priv, src_phys + src_offset,
dest_address + dest_offset,
bytes_left, 0, 0);
2005-03-25 13:31:53 -05:00
if (status) {
IPW_DEBUG_FW_INFO(": Failed on the buffer tail\n");
return -1;
} else
IPW_DEBUG_FW_INFO
(": Adding new cb - the buffer tail\n");
2005-03-25 13:31:53 -05:00
}
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW("<< \n");
return 0;
}
static int ipw_fw_dma_wait(struct ipw_priv *priv)
{
u32 current_index = 0, previous_index;
2005-03-25 13:31:53 -05:00
u32 watchdog = 0;
IPW_DEBUG_FW(">> : \n");
current_index = ipw_fw_dma_command_block_index(priv);
IPW_DEBUG_FW_INFO("sram_desc.last_cb_index:0x%08X\n",
(int)priv->sram_desc.last_cb_index);
2005-03-25 13:31:53 -05:00
while (current_index < priv->sram_desc.last_cb_index) {
udelay(50);
previous_index = current_index;
2005-03-25 13:31:53 -05:00
current_index = ipw_fw_dma_command_block_index(priv);
if (previous_index < current_index) {
watchdog = 0;
continue;
}
if (++watchdog > 400) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW_INFO("Timeout\n");
ipw_fw_dma_dump_command_block(priv);
ipw_fw_dma_abort(priv);
return -1;
}
}
ipw_fw_dma_abort(priv);
/*Disable the DMA in the CSR register */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_bit(priv, IPW_RESET_REG,
IPW_RESET_REG_MASTER_DISABLED | IPW_RESET_REG_STOP_MASTER);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_FW("<< dmaWaitSync \n");
return 0;
}
static void ipw_remove_current_network(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
struct list_head *element, *safe;
struct ieee80211_network *network = NULL;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
unsigned long flags;
spin_lock_irqsave(&priv->ieee->lock, flags);
2005-03-25 13:31:53 -05:00
list_for_each_safe(element, safe, &priv->ieee->network_list) {
network = list_entry(element, struct ieee80211_network, list);
if (!memcmp(network->bssid, priv->bssid, ETH_ALEN)) {
list_del(element);
list_add_tail(&network->list,
2005-03-25 13:31:53 -05:00
&priv->ieee->network_free_list);
}
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->ieee->lock, flags);
2005-03-25 13:31:53 -05:00
}
/**
* Check that card is still alive.
2005-03-25 13:31:53 -05:00
* Reads debug register from domain0.
* If card is present, pre-defined value should
* be found there.
*
2005-03-25 13:31:53 -05:00
* @param priv
* @return 1 if card is present, 0 otherwise
*/
static inline int ipw_alive(struct ipw_priv *priv)
{
return ipw_read32(priv, 0x90) == 0xd55555d5;
}
/* timeout in msec, attempted in 10-msec quanta */
static int ipw_poll_bit(struct ipw_priv *priv, u32 addr, u32 mask,
2005-03-25 13:31:53 -05:00
int timeout)
{
int i = 0;
do {
if ((ipw_read32(priv, addr) & mask) == mask)
2005-03-25 13:31:53 -05:00
return i;
mdelay(10);
i += 10;
} while (i < timeout);
2005-03-25 13:31:53 -05:00
return -ETIME;
}
/* These functions load the firmware and micro code for the operation of
2005-03-25 13:31:53 -05:00
* the ipw hardware. It assumes the buffer has all the bits for the
* image and the caller is handling the memory allocation and clean up.
*/
static int ipw_stop_master(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
int rc;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE(">> \n");
/* stop master. typical delay - 0 */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_bit(priv, IPW_RESET_REG, IPW_RESET_REG_STOP_MASTER);
2005-03-25 13:31:53 -05:00
/* timeout is in msec, polled in 10-msec quanta */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
rc = ipw_poll_bit(priv, IPW_RESET_REG,
IPW_RESET_REG_MASTER_DISABLED, 100);
2005-03-25 13:31:53 -05:00
if (rc < 0) {
IPW_ERROR("wait for stop master failed after 100ms\n");
2005-03-25 13:31:53 -05:00
return -1;
}
IPW_DEBUG_INFO("stop master %dms\n", rc);
return rc;
}
static void ipw_arc_release(struct ipw_priv *priv)
{
IPW_DEBUG_TRACE(">> \n");
mdelay(5);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_clear_bit(priv, IPW_RESET_REG, CBD_RESET_REG_PRINCETON_RESET);
2005-03-25 13:31:53 -05:00
/* no one knows timing, for safety add some delay */
mdelay(5);
}
struct fw_chunk {
u32 address;
u32 length;
};
static int ipw_load_ucode(struct ipw_priv *priv, u8 * data, size_t len)
2005-03-25 13:31:53 -05:00
{
int rc = 0, i, addr;
u8 cr = 0;
u16 *image;
image = (u16 *) data;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE(">> \n");
rc = ipw_stop_master(priv);
if (rc < 0)
return rc;
// spin_lock_irqsave(&priv->lock, flags);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
for (addr = IPW_SHARED_LOWER_BOUND;
addr < IPW_REGISTER_DOMAIN1_END; addr += 4) {
2005-03-25 13:31:53 -05:00
ipw_write32(priv, addr, 0);
}
/* no ucode (yet) */
memset(&priv->dino_alive, 0, sizeof(priv->dino_alive));
/* destroy DMA queues */
/* reset sequence */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_MEM_HALT_AND_RESET, IPW_BIT_HALT_RESET_ON);
2005-03-25 13:31:53 -05:00
ipw_arc_release(priv);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_MEM_HALT_AND_RESET, IPW_BIT_HALT_RESET_OFF);
2005-03-25 13:31:53 -05:00
mdelay(1);
/* reset PHY */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_INTERNAL_CMD_EVENT, IPW_BASEBAND_POWER_DOWN);
2005-03-25 13:31:53 -05:00
mdelay(1);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg32(priv, IPW_INTERNAL_CMD_EVENT, 0);
2005-03-25 13:31:53 -05:00
mdelay(1);
2005-03-25 13:31:53 -05:00
/* enable ucode store */
ipw_write_reg8(priv, IPW_BASEBAND_CONTROL_STATUS, 0x0);
ipw_write_reg8(priv, IPW_BASEBAND_CONTROL_STATUS, DINO_ENABLE_CS);
2005-03-25 13:31:53 -05:00
mdelay(1);
/* write ucode */
/**
* @bug
* Do NOT set indirect address register once and then
* store data to indirect data register in the loop.
* It seems very reasonable, but in this case DINO do not
* accept ucode. It is essential to set address each time.
*/
/* load new ipw uCode */
for (i = 0; i < len / 2; i++)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg16(priv, IPW_BASEBAND_CONTROL_STORE,
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
cpu_to_le16(image[i]));
2005-03-25 13:31:53 -05:00
/* enable DINO */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg8(priv, IPW_BASEBAND_CONTROL_STATUS, 0);
ipw_write_reg8(priv, IPW_BASEBAND_CONTROL_STATUS, DINO_ENABLE_SYSTEM);
2005-03-25 13:31:53 -05:00
/* this is where the igx / win driver deveates from the VAP driver. */
2005-03-25 13:31:53 -05:00
/* wait for alive response */
for (i = 0; i < 100; i++) {
/* poll for incoming data */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
cr = ipw_read_reg8(priv, IPW_BASEBAND_CONTROL_STATUS);
2005-03-25 13:31:53 -05:00
if (cr & DINO_RXFIFO_DATA)
break;
mdelay(1);
}
if (cr & DINO_RXFIFO_DATA) {
/* alive_command_responce size is NOT multiple of 4 */
u32 response_buffer[(sizeof(priv->dino_alive) + 3) / 4];
for (i = 0; i < ARRAY_SIZE(response_buffer); i++)
2005-03-25 13:31:53 -05:00
response_buffer[i] =
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
le32_to_cpu(ipw_read_reg32(priv,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_BASEBAND_RX_FIFO_READ));
2005-03-25 13:31:53 -05:00
memcpy(&priv->dino_alive, response_buffer,
sizeof(priv->dino_alive));
if (priv->dino_alive.alive_command == 1
&& priv->dino_alive.ucode_valid == 1) {
rc = 0;
IPW_DEBUG_INFO
("Microcode OK, rev. %d (0x%x) dev. %d (0x%x) "
"of %02d/%02d/%02d %02d:%02d\n",
priv->dino_alive.software_revision,
priv->dino_alive.software_revision,
priv->dino_alive.device_identifier,
priv->dino_alive.device_identifier,
priv->dino_alive.time_stamp[0],
priv->dino_alive.time_stamp[1],
priv->dino_alive.time_stamp[2],
priv->dino_alive.time_stamp[3],
priv->dino_alive.time_stamp[4]);
2005-03-25 13:31:53 -05:00
} else {
IPW_DEBUG_INFO("Microcode is not alive\n");
rc = -EINVAL;
}
} else {
IPW_DEBUG_INFO("No alive response from DINO\n");
rc = -ETIME;
}
/* disable DINO, otherwise for some reason
firmware have problem getting alive resp. */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write_reg8(priv, IPW_BASEBAND_CONTROL_STATUS, 0);
2005-03-25 13:31:53 -05:00
// spin_unlock_irqrestore(&priv->lock, flags);
2005-03-25 13:31:53 -05:00
return rc;
}
static int ipw_load_firmware(struct ipw_priv *priv, u8 * data, size_t len)
2005-03-25 13:31:53 -05:00
{
int rc = -1;
int offset = 0;
struct fw_chunk *chunk;
dma_addr_t shared_phys;
u8 *shared_virt;
IPW_DEBUG_TRACE("<< : \n");
shared_virt = pci_alloc_consistent(priv->pci_dev, len, &shared_phys);
if (!shared_virt)
return -ENOMEM;
memmove(shared_virt, data, len);
/* Start the Dma */
rc = ipw_fw_dma_enable(priv);
if (priv->sram_desc.last_cb_index > 0) {
/* the DMA is already ready this would be a bug. */
BUG();
goto out;
}
do {
chunk = (struct fw_chunk *)(data + offset);
offset += sizeof(struct fw_chunk);
/* build DMA packet and queue up for sending */
/* dma to chunk->address, the chunk->length bytes from data +
2005-03-25 13:31:53 -05:00
* offeset*/
/* Dma loading */
rc = ipw_fw_dma_add_buffer(priv, shared_phys + offset,
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
le32_to_cpu(chunk->address),
le32_to_cpu(chunk->length));
2005-03-25 13:31:53 -05:00
if (rc) {
IPW_DEBUG_INFO("dmaAddBuffer Failed\n");
goto out;
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
offset += le32_to_cpu(chunk->length);
2005-03-25 13:31:53 -05:00
} while (offset < len);
/* Run the DMA and wait for the answer */
2005-03-25 13:31:53 -05:00
rc = ipw_fw_dma_kick(priv);
if (rc) {
IPW_ERROR("dmaKick Failed\n");
goto out;
}
rc = ipw_fw_dma_wait(priv);
if (rc) {
IPW_ERROR("dmaWaitSync Failed\n");
goto out;
}
out:
pci_free_consistent(priv->pci_dev, len, shared_virt, shared_phys);
2005-03-25 13:31:53 -05:00
return rc;
}
/* stop nic */
static int ipw_stop_nic(struct ipw_priv *priv)
{
int rc = 0;
/* stop */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_RESET_REG, IPW_RESET_REG_STOP_MASTER);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
rc = ipw_poll_bit(priv, IPW_RESET_REG,
IPW_RESET_REG_MASTER_DISABLED, 500);
2005-03-25 13:31:53 -05:00
if (rc < 0) {
IPW_ERROR("wait for reg master disabled failed after 500ms\n");
2005-03-25 13:31:53 -05:00
return rc;
}
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_bit(priv, IPW_RESET_REG, CBD_RESET_REG_PRINCETON_RESET);
2005-03-25 13:31:53 -05:00
return rc;
}
static void ipw_start_nic(struct ipw_priv *priv)
{
IPW_DEBUG_TRACE(">>\n");
/* prvHwStartNic release ARC */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_clear_bit(priv, IPW_RESET_REG,
IPW_RESET_REG_MASTER_DISABLED |
IPW_RESET_REG_STOP_MASTER |
2005-03-25 13:31:53 -05:00
CBD_RESET_REG_PRINCETON_RESET);
2005-03-25 13:31:53 -05:00
/* enable power management */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_bit(priv, IPW_GP_CNTRL_RW,
IPW_GP_CNTRL_BIT_HOST_ALLOWS_STANDBY);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE("<<\n");
}
2005-03-25 13:31:53 -05:00
static int ipw_init_nic(struct ipw_priv *priv)
{
int rc;
IPW_DEBUG_TRACE(">>\n");
/* reset */
2005-03-25 13:31:53 -05:00
/*prvHwInitNic */
/* set "initialization complete" bit to move adapter to D0 state */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_bit(priv, IPW_GP_CNTRL_RW, IPW_GP_CNTRL_BIT_INIT_DONE);
2005-03-25 13:31:53 -05:00
/* low-level PLL activation */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_READ_INT_REGISTER,
IPW_BIT_INT_HOST_SRAM_READ_INT_REGISTER);
2005-03-25 13:31:53 -05:00
/* wait for clock stabilization */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
rc = ipw_poll_bit(priv, IPW_GP_CNTRL_RW,
IPW_GP_CNTRL_BIT_CLOCK_READY, 250);
if (rc < 0)
2005-03-25 13:31:53 -05:00
IPW_DEBUG_INFO("FAILED wait for clock stablization\n");
/* assert SW reset */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_bit(priv, IPW_RESET_REG, IPW_RESET_REG_SW_RESET);
2005-03-25 13:31:53 -05:00
udelay(10);
/* set "initialization complete" bit to move adapter to D0 state */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_bit(priv, IPW_GP_CNTRL_RW, IPW_GP_CNTRL_BIT_INIT_DONE);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE(">>\n");
return 0;
}
/* Call this function from process context, it will sleep in request_firmware.
2005-03-25 13:31:53 -05:00
* Probe is an ok place to call this from.
*/
static int ipw_reset_nic(struct ipw_priv *priv)
{
int rc = 0;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
unsigned long flags;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE(">>\n");
2005-03-25 13:31:53 -05:00
rc = ipw_init_nic(priv);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_lock_irqsave(&priv->lock, flags);
2005-03-25 13:31:53 -05:00
/* Clear the 'host command active' bit... */
priv->status &= ~STATUS_HCMD_ACTIVE;
wake_up_interruptible(&priv->wait_command_queue);
priv->status &= ~(STATUS_SCANNING | STATUS_SCAN_ABORTING);
wake_up_interruptible(&priv->wait_state);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->lock, flags);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE("<<\n");
return rc;
}
2005-03-25 13:31:53 -05:00
struct ipw_fw {
__le32 ver;
__le32 boot_size;
__le32 ucode_size;
__le32 fw_size;
u8 data[0];
};
static int ipw_get_fw(struct ipw_priv *priv,
const struct firmware **raw, const char *name)
2005-03-25 13:31:53 -05:00
{
struct ipw_fw *fw;
2005-03-25 13:31:53 -05:00
int rc;
/* ask firmware_class module to get the boot firmware off disk */
rc = request_firmware(raw, name, &priv->pci_dev->dev);
2005-03-25 13:31:53 -05:00
if (rc < 0) {
IPW_ERROR("%s request_firmware failed: Reason %d\n", name, rc);
2005-03-25 13:31:53 -05:00
return rc;
}
2005-03-25 13:31:53 -05:00
if ((*raw)->size < sizeof(*fw)) {
IPW_ERROR("%s is too small (%zd)\n", name, (*raw)->size);
return -EINVAL;
}
fw = (void *)(*raw)->data;
if ((*raw)->size < sizeof(*fw) + le32_to_cpu(fw->boot_size) +
le32_to_cpu(fw->ucode_size) + le32_to_cpu(fw->fw_size)) {
IPW_ERROR("%s is too small or corrupt (%zd)\n",
name, (*raw)->size);
2005-03-25 13:31:53 -05:00
return -EINVAL;
}
IPW_DEBUG_INFO("Read firmware '%s' image v%d.%d (%zd bytes)\n",
2005-03-25 13:31:53 -05:00
name,
le32_to_cpu(fw->ver) >> 16,
le32_to_cpu(fw->ver) & 0xff,
(*raw)->size - sizeof(*fw));
2005-03-25 13:31:53 -05:00
return 0;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#define IPW_RX_BUF_SIZE (3000)
2005-03-25 13:31:53 -05:00
static void ipw_rx_queue_reset(struct ipw_priv *priv,
2005-03-25 13:31:53 -05:00
struct ipw_rx_queue *rxq)
{
unsigned long flags;
int i;
spin_lock_irqsave(&rxq->lock, flags);
INIT_LIST_HEAD(&rxq->rx_free);
INIT_LIST_HEAD(&rxq->rx_used);
/* Fill the rx_used queue with _all_ of the Rx buffers */
for (i = 0; i < RX_FREE_BUFFERS + RX_QUEUE_SIZE; i++) {
/* In the reset function, these buffers may have been allocated
* to an SKB, so we need to unmap and free potential storage */
if (rxq->pool[i].skb != NULL) {
pci_unmap_single(priv->pci_dev, rxq->pool[i].dma_addr,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_RX_BUF_SIZE, PCI_DMA_FROMDEVICE);
2005-03-25 13:31:53 -05:00
dev_kfree_skb(rxq->pool[i].skb);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
rxq->pool[i].skb = NULL;
2005-03-25 13:31:53 -05:00
}
list_add_tail(&rxq->pool[i].list, &rxq->rx_used);
}
2005-03-25 13:31:53 -05:00
/* Set us so that we have processed and used all buffers, but have
* not restocked the Rx queue with fresh buffers */
rxq->read = rxq->write = 0;
rxq->processed = RX_QUEUE_SIZE - 1;
rxq->free_count = 0;
spin_unlock_irqrestore(&rxq->lock, flags);
}
#ifdef CONFIG_PM
static int fw_loaded = 0;
static const struct firmware *raw = NULL;
static void free_firmware(void)
{
if (fw_loaded) {
release_firmware(raw);
raw = NULL;
fw_loaded = 0;
}
}
#else
#define free_firmware() do {} while (0)
2005-03-25 13:31:53 -05:00
#endif
static int ipw_load(struct ipw_priv *priv)
{
#ifndef CONFIG_PM
const struct firmware *raw = NULL;
2005-03-25 13:31:53 -05:00
#endif
struct ipw_fw *fw;
u8 *boot_img, *ucode_img, *fw_img;
u8 *name = NULL;
2005-03-25 13:31:53 -05:00
int rc = 0, retries = 3;
switch (priv->ieee->iw_mode) {
case IW_MODE_ADHOC:
name = "ipw2200-ibss.fw";
break;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
case IW_MODE_MONITOR:
name = "ipw2200-sniffer.fw";
break;
2005-03-25 13:31:53 -05:00
#endif
case IW_MODE_INFRA:
name = "ipw2200-bss.fw";
break;
}
if (!name) {
rc = -EINVAL;
goto error;
}
#ifdef CONFIG_PM
if (!fw_loaded) {
#endif
rc = ipw_get_fw(priv, &raw, name);
if (rc < 0)
goto error;
#ifdef CONFIG_PM
2005-03-25 13:31:53 -05:00
}
#endif
fw = (void *)raw->data;
boot_img = &fw->data[0];
ucode_img = &fw->data[le32_to_cpu(fw->boot_size)];
fw_img = &fw->data[le32_to_cpu(fw->boot_size) +
le32_to_cpu(fw->ucode_size)];
if (rc < 0)
goto error;
2005-03-25 13:31:53 -05:00
if (!priv->rxq)
priv->rxq = ipw_rx_queue_alloc(priv);
else
ipw_rx_queue_reset(priv, priv->rxq);
if (!priv->rxq) {
IPW_ERROR("Unable to initialize Rx queue\n");
goto error;
}
retry:
2005-03-25 13:31:53 -05:00
/* Ensure interrupts are disabled */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_MASK_R, ~IPW_INTA_MASK_ALL);
2005-03-25 13:31:53 -05:00
priv->status &= ~STATUS_INT_ENABLED;
/* ack pending interrupts */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_RW, IPW_INTA_MASK_ALL);
2005-03-25 13:31:53 -05:00
ipw_stop_nic(priv);
rc = ipw_reset_nic(priv);
if (rc < 0) {
2005-03-25 13:31:53 -05:00
IPW_ERROR("Unable to reset NIC\n");
goto error;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_zero_memory(priv, IPW_NIC_SRAM_LOWER_BOUND,
IPW_NIC_SRAM_UPPER_BOUND - IPW_NIC_SRAM_LOWER_BOUND);
2005-03-25 13:31:53 -05:00
/* DMA the initial boot firmware into the device */
rc = ipw_load_firmware(priv, boot_img, le32_to_cpu(fw->boot_size));
2005-03-25 13:31:53 -05:00
if (rc < 0) {
IPW_ERROR("Unable to load boot firmware: %d\n", rc);
2005-03-25 13:31:53 -05:00
goto error;
}
/* kick start the device */
ipw_start_nic(priv);
/* wait for the device to finish its initial startup sequence */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
rc = ipw_poll_bit(priv, IPW_INTA_RW,
IPW_INTA_BIT_FW_INITIALIZATION_DONE, 500);
2005-03-25 13:31:53 -05:00
if (rc < 0) {
IPW_ERROR("device failed to boot initial fw image\n");
goto error;
}
IPW_DEBUG_INFO("initial device response after %dms\n", rc);
/* ack fw init done interrupt */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_RW, IPW_INTA_BIT_FW_INITIALIZATION_DONE);
2005-03-25 13:31:53 -05:00
/* DMA the ucode into the device */
rc = ipw_load_ucode(priv, ucode_img, le32_to_cpu(fw->ucode_size));
2005-03-25 13:31:53 -05:00
if (rc < 0) {
IPW_ERROR("Unable to load ucode: %d\n", rc);
2005-03-25 13:31:53 -05:00
goto error;
}
2005-03-25 13:31:53 -05:00
/* stop nic */
ipw_stop_nic(priv);
/* DMA bss firmware into the device */
rc = ipw_load_firmware(priv, fw_img, le32_to_cpu(fw->fw_size));
if (rc < 0) {
IPW_ERROR("Unable to load firmware: %d\n", rc);
2005-03-25 13:31:53 -05:00
goto error;
}
#ifdef CONFIG_PM
fw_loaded = 1;
#endif
2005-03-25 13:31:53 -05:00
ipw_write32(priv, IPW_EEPROM_LOAD_DISABLE, 0);
rc = ipw_queue_reset(priv);
if (rc < 0) {
2005-03-25 13:31:53 -05:00
IPW_ERROR("Unable to initialize queues\n");
goto error;
}
/* Ensure interrupts are disabled */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_MASK_R, ~IPW_INTA_MASK_ALL);
/* ack pending interrupts */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_RW, IPW_INTA_MASK_ALL);
2005-03-25 13:31:53 -05:00
/* kick start the device */
ipw_start_nic(priv);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (ipw_read32(priv, IPW_INTA_RW) & IPW_INTA_BIT_PARITY_ERROR) {
2005-03-25 13:31:53 -05:00
if (retries > 0) {
IPW_WARNING("Parity error. Retrying init.\n");
retries--;
goto retry;
}
IPW_ERROR("TODO: Handle parity error -- schedule restart?\n");
rc = -EIO;
goto error;
}
/* wait for the device */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
rc = ipw_poll_bit(priv, IPW_INTA_RW,
IPW_INTA_BIT_FW_INITIALIZATION_DONE, 500);
2005-03-25 13:31:53 -05:00
if (rc < 0) {
IPW_ERROR("device failed to start within 500ms\n");
2005-03-25 13:31:53 -05:00
goto error;
}
IPW_DEBUG_INFO("device response after %dms\n", rc);
/* ack fw init done interrupt */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_RW, IPW_INTA_BIT_FW_INITIALIZATION_DONE);
2005-03-25 13:31:53 -05:00
/* read eeprom data and initialize the eeprom region of sram */
priv->eeprom_delay = 1;
ipw_eeprom_init_sram(priv);
2005-03-25 13:31:53 -05:00
/* enable interrupts */
ipw_enable_interrupts(priv);
/* Ensure our queue has valid packets */
ipw_rx_queue_replenish(priv);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_RX_READ_INDEX, priv->rxq->read);
2005-03-25 13:31:53 -05:00
/* ack pending interrupts */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_INTA_RW, IPW_INTA_MASK_ALL);
2005-03-25 13:31:53 -05:00
#ifndef CONFIG_PM
release_firmware(raw);
2005-03-25 13:31:53 -05:00
#endif
return 0;
error:
2005-03-25 13:31:53 -05:00
if (priv->rxq) {
ipw_rx_queue_free(priv, priv->rxq);
priv->rxq = NULL;
}
ipw_tx_queue_free(priv);
if (raw)
release_firmware(raw);
2005-03-25 13:31:53 -05:00
#ifdef CONFIG_PM
fw_loaded = 0;
raw = NULL;
2005-03-25 13:31:53 -05:00
#endif
return rc;
}
/**
2005-03-25 13:31:53 -05:00
* DMA services
*
* Theory of operation
*
* A queue is a circular buffers with 'Read' and 'Write' pointers.
* 2 empty entries always kept in the buffer to protect from overflow.
*
* For Tx queue, there are low mark and high mark limits. If, after queuing
* the packet for Tx, free space become < low mark, Tx queue stopped. When
* reclaiming packets (on 'tx done IRQ), if free space become > high mark,
2005-03-25 13:31:53 -05:00
* Tx queue resumed.
*
* The IPW operates with six queues, one receive queue in the device's
* sram, one transmit queue for sending commands to the device firmware,
* and four transmit queues for data.
2005-03-25 13:31:53 -05:00
*
* The four transmit queues allow for performing quality of service (qos)
2005-03-25 13:31:53 -05:00
* transmissions as per the 802.11 protocol. Currently Linux does not
* provide a mechanism to the user for utilizing prioritized queues, so
2005-03-25 13:31:53 -05:00
* we only utilize the first data transmit queue (queue1).
*/
/**
* Driver allocates buffers of this size for Rx
*/
static inline int ipw_queue_space(const struct clx2_queue *q)
{
int s = q->last_used - q->first_empty;
if (s <= 0)
s += q->n_bd;
s -= 2; /* keep some reserve to not confuse empty and full situations */
if (s < 0)
s = 0;
return s;
}
static inline int ipw_queue_inc_wrap(int index, int n_bd)
{
return (++index == n_bd) ? 0 : index;
}
/**
* Initialize common DMA queue structure
*
2005-03-25 13:31:53 -05:00
* @param q queue to init
* @param count Number of BD's to allocate. Should be power of 2
* @param read_register Address for 'read' register
* (not offset within BAR, full address)
* @param write_register Address for 'write' register
* (not offset within BAR, full address)
* @param base_register Address for 'base' register
* (not offset within BAR, full address)
* @param size Address for 'size' register
* (not offset within BAR, full address)
*/
static void ipw_queue_init(struct ipw_priv *priv, struct clx2_queue *q,
int count, u32 read, u32 write, u32 base, u32 size)
2005-03-25 13:31:53 -05:00
{
q->n_bd = count;
q->low_mark = q->n_bd / 4;
if (q->low_mark < 4)
q->low_mark = 4;
q->high_mark = q->n_bd / 8;
if (q->high_mark < 2)
q->high_mark = 2;
q->first_empty = q->last_used = 0;
q->reg_r = read;
q->reg_w = write;
ipw_write32(priv, base, q->dma_addr);
ipw_write32(priv, size, count);
ipw_write32(priv, read, 0);
ipw_write32(priv, write, 0);
_ipw_read32(priv, 0x90);
}
static int ipw_queue_tx_init(struct ipw_priv *priv,
2005-03-25 13:31:53 -05:00
struct clx2_tx_queue *q,
int count, u32 read, u32 write, u32 base, u32 size)
2005-03-25 13:31:53 -05:00
{
struct pci_dev *dev = priv->pci_dev;
q->txb = kmalloc(sizeof(q->txb[0]) * count, GFP_KERNEL);
if (!q->txb) {
IPW_ERROR("vmalloc for auxilary BD structures failed\n");
return -ENOMEM;
}
q->bd =
pci_alloc_consistent(dev, sizeof(q->bd[0]) * count, &q->q.dma_addr);
2005-03-25 13:31:53 -05:00
if (!q->bd) {
IPW_ERROR("pci_alloc_consistent(%zd) failed\n",
sizeof(q->bd[0]) * count);
2005-03-25 13:31:53 -05:00
kfree(q->txb);
q->txb = NULL;
return -ENOMEM;
}
ipw_queue_init(priv, &q->q, count, read, write, base, size);
return 0;
}
/**
* Free one TFD, those at index [txq->q.last_used].
* Do NOT advance any indexes
*
2005-03-25 13:31:53 -05:00
* @param dev
* @param txq
*/
static void ipw_queue_tx_free_tfd(struct ipw_priv *priv,
struct clx2_tx_queue *txq)
{
struct tfd_frame *bd = &txq->bd[txq->q.last_used];
struct pci_dev *dev = priv->pci_dev;
int i;
2005-03-25 13:31:53 -05:00
/* classify bd */
if (bd->control_flags.message_type == TX_HOST_COMMAND_TYPE)
/* nothing to cleanup after for host commands */
return;
/* sanity check */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (le32_to_cpu(bd->u.data.num_chunks) > NUM_TFD_CHUNKS) {
IPW_ERROR("Too many chunks: %i\n",
le32_to_cpu(bd->u.data.num_chunks));
2005-03-25 13:31:53 -05:00
/** @todo issue fatal error, it is quite serious situation */
return;
}
/* unmap chunks if any */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
for (i = 0; i < le32_to_cpu(bd->u.data.num_chunks); i++) {
pci_unmap_single(dev, le32_to_cpu(bd->u.data.chunk_ptr[i]),
le16_to_cpu(bd->u.data.chunk_len[i]),
PCI_DMA_TODEVICE);
2005-03-25 13:31:53 -05:00
if (txq->txb[txq->q.last_used]) {
ieee80211_txb_free(txq->txb[txq->q.last_used]);
txq->txb[txq->q.last_used] = NULL;
}
}
}
/**
* Deallocate DMA queue.
*
2005-03-25 13:31:53 -05:00
* Empty queue by removing and destroying all BD's.
* Free all buffers.
*
2005-03-25 13:31:53 -05:00
* @param dev
* @param q
*/
static void ipw_queue_tx_free(struct ipw_priv *priv, struct clx2_tx_queue *txq)
2005-03-25 13:31:53 -05:00
{
struct clx2_queue *q = &txq->q;
struct pci_dev *dev = priv->pci_dev;
if (q->n_bd == 0)
return;
2005-03-25 13:31:53 -05:00
/* first, empty all BD's */
for (; q->first_empty != q->last_used;
q->last_used = ipw_queue_inc_wrap(q->last_used, q->n_bd)) {
ipw_queue_tx_free_tfd(priv, txq);
}
2005-03-25 13:31:53 -05:00
/* free buffers belonging to queue itself */
pci_free_consistent(dev, sizeof(txq->bd[0]) * q->n_bd, txq->bd,
2005-03-25 13:31:53 -05:00
q->dma_addr);
kfree(txq->txb);
/* 0 fill whole structure */
memset(txq, 0, sizeof(*txq));
}
/**
* Destroy all DMA queues and structures
*
2005-03-25 13:31:53 -05:00
* @param priv
*/
static void ipw_tx_queue_free(struct ipw_priv *priv)
{
/* Tx CMD queue */
ipw_queue_tx_free(priv, &priv->txq_cmd);
/* Tx queues */
ipw_queue_tx_free(priv, &priv->txq[0]);
ipw_queue_tx_free(priv, &priv->txq[1]);
ipw_queue_tx_free(priv, &priv->txq[2]);
ipw_queue_tx_free(priv, &priv->txq[3]);
}
static void ipw_create_bssid(struct ipw_priv *priv, u8 * bssid)
2005-03-25 13:31:53 -05:00
{
/* First 3 bytes are manufacturer */
bssid[0] = priv->mac_addr[0];
bssid[1] = priv->mac_addr[1];
bssid[2] = priv->mac_addr[2];
/* Last bytes are random */
get_random_bytes(&bssid[3], ETH_ALEN - 3);
2005-03-25 13:31:53 -05:00
bssid[0] &= 0xfe; /* clear multicast bit */
bssid[0] |= 0x02; /* set local assignment bit (IEEE802) */
2005-03-25 13:31:53 -05:00
}
static u8 ipw_add_station(struct ipw_priv *priv, u8 * bssid)
2005-03-25 13:31:53 -05:00
{
struct ipw_station_entry entry;
int i;
for (i = 0; i < priv->num_stations; i++) {
if (!memcmp(priv->stations[i], bssid, ETH_ALEN)) {
/* Another node is active in network */
priv->missed_adhoc_beacons = 0;
if (!(priv->config & CFG_STATIC_CHANNEL))
/* when other nodes drop out, we drop out */
priv->config &= ~CFG_ADHOC_PERSIST;
return i;
}
}
if (i == MAX_STATIONS)
return IPW_INVALID_STATION;
IPW_DEBUG_SCAN("Adding AdHoc station: " MAC_FMT "\n", MAC_ARG(bssid));
entry.reserved = 0;
entry.support_mode = 0;
memcpy(entry.mac_addr, bssid, ETH_ALEN);
memcpy(priv->stations[i], bssid, ETH_ALEN);
ipw_write_direct(priv, IPW_STATION_TABLE_LOWER + i * sizeof(entry),
&entry, sizeof(entry));
2005-03-25 13:31:53 -05:00
priv->num_stations++;
return i;
}
static u8 ipw_find_station(struct ipw_priv *priv, u8 * bssid)
2005-03-25 13:31:53 -05:00
{
int i;
for (i = 0; i < priv->num_stations; i++)
if (!memcmp(priv->stations[i], bssid, ETH_ALEN))
2005-03-25 13:31:53 -05:00
return i;
return IPW_INVALID_STATION;
}
static void ipw_send_disassociate(struct ipw_priv *priv, int quiet)
{
int err;
if (priv->status & STATUS_ASSOCIATING) {
IPW_DEBUG_ASSOC("Disassociating while associating.\n");
queue_work(priv->workqueue, &priv->disassociate);
return;
}
if (!(priv->status & STATUS_ASSOCIATED)) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ASSOC("Disassociating while not associated.\n");
return;
}
IPW_DEBUG_ASSOC("Disassocation attempt from " MAC_FMT " "
"on channel %d.\n",
MAC_ARG(priv->assoc_request.bssid),
2005-03-25 13:31:53 -05:00
priv->assoc_request.channel);
priv->status &= ~(STATUS_ASSOCIATING | STATUS_ASSOCIATED);
priv->status |= STATUS_DISASSOCIATING;
if (quiet)
priv->assoc_request.assoc_type = HC_DISASSOC_QUIET;
else
priv->assoc_request.assoc_type = HC_DISASSOCIATE;
2005-03-25 13:31:53 -05:00
err = ipw_send_associate(priv, &priv->assoc_request);
if (err) {
IPW_DEBUG_HC("Attempt to send [dis]associate command "
"failed.\n");
return;
}
}
static int ipw_disassociate(void *data)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = data;
if (!(priv->status & (STATUS_ASSOCIATED | STATUS_ASSOCIATING)))
return 0;
2005-03-25 13:31:53 -05:00
ipw_send_disassociate(data, 0);
return 1;
2005-03-25 13:31:53 -05:00
}
static void ipw_bg_disassociate(void *data)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_disassociate(data);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
}
static void ipw_system_config(void *data)
{
struct ipw_priv *priv = data;
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
if (priv->prom_net_dev && netif_running(priv->prom_net_dev)) {
priv->sys_config.accept_all_data_frames = 1;
priv->sys_config.accept_non_directed_frames = 1;
priv->sys_config.accept_all_mgmt_bcpr = 1;
priv->sys_config.accept_all_mgmt_frames = 1;
}
#endif
ipw_send_system_config(priv);
2005-03-25 13:31:53 -05:00
}
struct ipw_status_code {
u16 status;
const char *reason;
};
static const struct ipw_status_code ipw_status_codes[] = {
{0x00, "Successful"},
{0x01, "Unspecified failure"},
{0x0A, "Cannot support all requested capabilities in the "
"Capability information field"},
{0x0B, "Reassociation denied due to inability to confirm that "
"association exists"},
{0x0C, "Association denied due to reason outside the scope of this "
"standard"},
{0x0D,
"Responding station does not support the specified authentication "
2005-03-25 13:31:53 -05:00
"algorithm"},
{0x0E,
"Received an Authentication frame with authentication sequence "
2005-03-25 13:31:53 -05:00
"transaction sequence number out of expected sequence"},
{0x0F, "Authentication rejected because of challenge failure"},
{0x10, "Authentication rejected due to timeout waiting for next "
"frame in sequence"},
{0x11, "Association denied because AP is unable to handle additional "
"associated stations"},
{0x12,
"Association denied due to requesting station not supporting all "
2005-03-25 13:31:53 -05:00
"of the datarates in the BSSBasicServiceSet Parameter"},
{0x13,
"Association denied due to requesting station not supporting "
2005-03-25 13:31:53 -05:00
"short preamble operation"},
{0x14,
"Association denied due to requesting station not supporting "
2005-03-25 13:31:53 -05:00
"PBCC encoding"},
{0x15,
"Association denied due to requesting station not supporting "
2005-03-25 13:31:53 -05:00
"channel agility"},
{0x19,
"Association denied due to requesting station not supporting "
2005-03-25 13:31:53 -05:00
"short slot operation"},
{0x1A,
"Association denied due to requesting station not supporting "
2005-03-25 13:31:53 -05:00
"DSSS-OFDM operation"},
{0x28, "Invalid Information Element"},
{0x29, "Group Cipher is not valid"},
{0x2A, "Pairwise Cipher is not valid"},
{0x2B, "AKMP is not valid"},
{0x2C, "Unsupported RSN IE version"},
{0x2D, "Invalid RSN IE Capabilities"},
{0x2E, "Cipher suite is rejected per security policy"},
};
#ifdef CONFIG_IPW2200_DEBUG
static const char *ipw_get_status_code(u16 status)
2005-03-25 13:31:53 -05:00
{
int i;
for (i = 0; i < ARRAY_SIZE(ipw_status_codes); i++)
if (ipw_status_codes[i].status == (status & 0xff))
2005-03-25 13:31:53 -05:00
return ipw_status_codes[i].reason;
return "Unknown status value.";
}
#endif
static void inline average_init(struct average *avg)
{
memset(avg, 0, sizeof(*avg));
}
#define DEPTH_RSSI 8
#define DEPTH_NOISE 16
static s16 exponential_average(s16 prev_avg, s16 val, u8 depth)
{
return ((depth-1)*prev_avg + val)/depth;
}
static void average_add(struct average *avg, s16 val)
2005-03-25 13:31:53 -05:00
{
avg->sum -= avg->entries[avg->pos];
avg->sum += val;
avg->entries[avg->pos++] = val;
if (unlikely(avg->pos == AVG_ENTRIES)) {
avg->init = 1;
avg->pos = 0;
}
}
static s16 average_value(struct average *avg)
2005-03-25 13:31:53 -05:00
{
if (!unlikely(avg->init)) {
if (avg->pos)
return avg->sum / avg->pos;
return 0;
}
return avg->sum / AVG_ENTRIES;
}
static void ipw_reset_stats(struct ipw_priv *priv)
{
u32 len = sizeof(u32);
priv->quality = 0;
average_init(&priv->average_missed_beacons);
priv->exp_avg_rssi = -60;
priv->exp_avg_noise = -85 + 0x100;
2005-03-25 13:31:53 -05:00
priv->last_rate = 0;
priv->last_missed_beacons = 0;
priv->last_rx_packets = 0;
priv->last_tx_packets = 0;
priv->last_tx_failures = 0;
2005-03-25 13:31:53 -05:00
/* Firmware managed, reset only when NIC is restarted, so we have to
* normalize on the current value */
ipw_get_ordinal(priv, IPW_ORD_STAT_RX_ERR_CRC,
2005-03-25 13:31:53 -05:00
&priv->last_rx_err, &len);
ipw_get_ordinal(priv, IPW_ORD_STAT_TX_FAILURE,
2005-03-25 13:31:53 -05:00
&priv->last_tx_failures, &len);
/* Driver managed, reset with each association */
priv->missed_adhoc_beacons = 0;
priv->missed_beacons = 0;
priv->tx_packets = 0;
priv->rx_packets = 0;
}
static u32 ipw_get_max_rate(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
u32 i = 0x80000000;
u32 mask = priv->rates_mask;
/* If currently associated in B mode, restrict the maximum
* rate match to B rates */
if (priv->assoc_request.ieee_mode == IPW_B_MODE)
mask &= IEEE80211_CCK_RATES_MASK;
/* TODO: Verify that the rate is supported by the current rates
* list. */
while (i && !(mask & i))
i >>= 1;
2005-03-25 13:31:53 -05:00
switch (i) {
case IEEE80211_CCK_RATE_1MB_MASK:
return 1000000;
case IEEE80211_CCK_RATE_2MB_MASK:
return 2000000;
case IEEE80211_CCK_RATE_5MB_MASK:
return 5500000;
case IEEE80211_OFDM_RATE_6MB_MASK:
return 6000000;
case IEEE80211_OFDM_RATE_9MB_MASK:
return 9000000;
case IEEE80211_CCK_RATE_11MB_MASK:
return 11000000;
case IEEE80211_OFDM_RATE_12MB_MASK:
return 12000000;
case IEEE80211_OFDM_RATE_18MB_MASK:
return 18000000;
case IEEE80211_OFDM_RATE_24MB_MASK:
return 24000000;
case IEEE80211_OFDM_RATE_36MB_MASK:
return 36000000;
case IEEE80211_OFDM_RATE_48MB_MASK:
return 48000000;
case IEEE80211_OFDM_RATE_54MB_MASK:
return 54000000;
2005-03-25 13:31:53 -05:00
}
if (priv->ieee->mode == IEEE_B)
2005-03-25 13:31:53 -05:00
return 11000000;
else
return 54000000;
}
static u32 ipw_get_current_rate(struct ipw_priv *priv)
{
u32 rate, len = sizeof(rate);
int err;
if (!(priv->status & STATUS_ASSOCIATED))
2005-03-25 13:31:53 -05:00
return 0;
if (priv->tx_packets > IPW_REAL_RATE_RX_PACKET_THRESHOLD) {
err = ipw_get_ordinal(priv, IPW_ORD_STAT_TX_CURR_RATE, &rate,
2005-03-25 13:31:53 -05:00
&len);
if (err) {
IPW_DEBUG_INFO("failed querying ordinals.\n");
return 0;
}
} else
2005-03-25 13:31:53 -05:00
return ipw_get_max_rate(priv);
switch (rate) {
case IPW_TX_RATE_1MB:
return 1000000;
case IPW_TX_RATE_2MB:
return 2000000;
case IPW_TX_RATE_5MB:
return 5500000;
case IPW_TX_RATE_6MB:
return 6000000;
case IPW_TX_RATE_9MB:
return 9000000;
case IPW_TX_RATE_11MB:
return 11000000;
case IPW_TX_RATE_12MB:
return 12000000;
case IPW_TX_RATE_18MB:
return 18000000;
case IPW_TX_RATE_24MB:
return 24000000;
case IPW_TX_RATE_36MB:
return 36000000;
case IPW_TX_RATE_48MB:
return 48000000;
case IPW_TX_RATE_54MB:
return 54000000;
2005-03-25 13:31:53 -05:00
}
return 0;
}
#define IPW_STATS_INTERVAL (2 * HZ)
static void ipw_gather_stats(struct ipw_priv *priv)
{
u32 rx_err, rx_err_delta, rx_packets_delta;
u32 tx_failures, tx_failures_delta, tx_packets_delta;
u32 missed_beacons_percent, missed_beacons_delta;
u32 quality = 0;
u32 len = sizeof(u32);
s16 rssi;
u32 beacon_quality, signal_quality, tx_quality, rx_quality,
rate_quality;
u32 max_rate;
2005-03-25 13:31:53 -05:00
if (!(priv->status & STATUS_ASSOCIATED)) {
priv->quality = 0;
return;
}
/* Update the statistics */
ipw_get_ordinal(priv, IPW_ORD_STAT_MISSED_BEACONS,
2005-03-25 13:31:53 -05:00
&priv->missed_beacons, &len);
missed_beacons_delta = priv->missed_beacons - priv->last_missed_beacons;
2005-03-25 13:31:53 -05:00
priv->last_missed_beacons = priv->missed_beacons;
if (priv->assoc_request.beacon_interval) {
missed_beacons_percent = missed_beacons_delta *
(HZ * priv->assoc_request.beacon_interval) /
(IPW_STATS_INTERVAL * 10);
2005-03-25 13:31:53 -05:00
} else {
missed_beacons_percent = 0;
}
average_add(&priv->average_missed_beacons, missed_beacons_percent);
ipw_get_ordinal(priv, IPW_ORD_STAT_RX_ERR_CRC, &rx_err, &len);
rx_err_delta = rx_err - priv->last_rx_err;
priv->last_rx_err = rx_err;
ipw_get_ordinal(priv, IPW_ORD_STAT_TX_FAILURE, &tx_failures, &len);
tx_failures_delta = tx_failures - priv->last_tx_failures;
priv->last_tx_failures = tx_failures;
rx_packets_delta = priv->rx_packets - priv->last_rx_packets;
priv->last_rx_packets = priv->rx_packets;
tx_packets_delta = priv->tx_packets - priv->last_tx_packets;
priv->last_tx_packets = priv->tx_packets;
/* Calculate quality based on the following:
*
2005-03-25 13:31:53 -05:00
* Missed beacon: 100% = 0, 0% = 70% missed
* Rate: 60% = 1Mbs, 100% = Max
* Rx and Tx errors represent a straight % of total Rx/Tx
* RSSI: 100% = > -50, 0% = < -80
* Rx errors: 100% = 0, 0% = 50% missed
*
2005-03-25 13:31:53 -05:00
* The lowest computed quality is used.
*
*/
#define BEACON_THRESHOLD 5
beacon_quality = 100 - missed_beacons_percent;
if (beacon_quality < BEACON_THRESHOLD)
beacon_quality = 0;
else
beacon_quality = (beacon_quality - BEACON_THRESHOLD) * 100 /
(100 - BEACON_THRESHOLD);
IPW_DEBUG_STATS("Missed beacon: %3d%% (%d%%)\n",
2005-03-25 13:31:53 -05:00
beacon_quality, missed_beacons_percent);
2005-03-25 13:31:53 -05:00
priv->last_rate = ipw_get_current_rate(priv);
max_rate = ipw_get_max_rate(priv);
rate_quality = priv->last_rate * 40 / max_rate + 60;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_STATS("Rate quality : %3d%% (%dMbs)\n",
rate_quality, priv->last_rate / 1000000);
if (rx_packets_delta > 100 && rx_packets_delta + rx_err_delta)
rx_quality = 100 - (rx_err_delta * 100) /
(rx_packets_delta + rx_err_delta);
2005-03-25 13:31:53 -05:00
else
rx_quality = 100;
IPW_DEBUG_STATS("Rx quality : %3d%% (%u errors, %u packets)\n",
rx_quality, rx_err_delta, rx_packets_delta);
if (tx_packets_delta > 100 && tx_packets_delta + tx_failures_delta)
tx_quality = 100 - (tx_failures_delta * 100) /
(tx_packets_delta + tx_failures_delta);
2005-03-25 13:31:53 -05:00
else
tx_quality = 100;
IPW_DEBUG_STATS("Tx quality : %3d%% (%u errors, %u packets)\n",
tx_quality, tx_failures_delta, tx_packets_delta);
rssi = priv->exp_avg_rssi;
signal_quality =
(100 *
(priv->ieee->perfect_rssi - priv->ieee->worst_rssi) *
(priv->ieee->perfect_rssi - priv->ieee->worst_rssi) -
(priv->ieee->perfect_rssi - rssi) *
(15 * (priv->ieee->perfect_rssi - priv->ieee->worst_rssi) +
62 * (priv->ieee->perfect_rssi - rssi))) /
((priv->ieee->perfect_rssi - priv->ieee->worst_rssi) *
(priv->ieee->perfect_rssi - priv->ieee->worst_rssi));
if (signal_quality > 100)
2005-03-25 13:31:53 -05:00
signal_quality = 100;
else if (signal_quality < 1)
2005-03-25 13:31:53 -05:00
signal_quality = 0;
IPW_DEBUG_STATS("Signal level : %3d%% (%d dBm)\n",
2005-03-25 13:31:53 -05:00
signal_quality, rssi);
quality = min(beacon_quality,
2005-03-25 13:31:53 -05:00
min(rate_quality,
min(tx_quality, min(rx_quality, signal_quality))));
if (quality == beacon_quality)
IPW_DEBUG_STATS("Quality (%d%%): Clamped to missed beacons.\n",
quality);
2005-03-25 13:31:53 -05:00
if (quality == rate_quality)
IPW_DEBUG_STATS("Quality (%d%%): Clamped to rate quality.\n",
quality);
2005-03-25 13:31:53 -05:00
if (quality == tx_quality)
IPW_DEBUG_STATS("Quality (%d%%): Clamped to Tx quality.\n",
quality);
2005-03-25 13:31:53 -05:00
if (quality == rx_quality)
IPW_DEBUG_STATS("Quality (%d%%): Clamped to Rx quality.\n",
quality);
2005-03-25 13:31:53 -05:00
if (quality == signal_quality)
IPW_DEBUG_STATS("Quality (%d%%): Clamped to signal quality.\n",
quality);
2005-03-25 13:31:53 -05:00
priv->quality = quality;
queue_delayed_work(priv->workqueue, &priv->gather_stats,
2005-03-25 13:31:53 -05:00
IPW_STATS_INTERVAL);
}
static void ipw_bg_gather_stats(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_gather_stats(data);
mutex_unlock(&priv->mutex);
}
/* Missed beacon behavior:
* 1st missed -> roaming_threshold, just wait, don't do any scan/roam.
* roaming_threshold -> disassociate_threshold, scan and roam for better signal.
* Above disassociate threshold, give up and stop scanning.
* Roaming is disabled if disassociate_threshold <= roaming_threshold */
static void ipw_handle_missed_beacon(struct ipw_priv *priv,
int missed_count)
{
priv->notif_missed_beacons = missed_count;
if (missed_count > priv->disassociate_threshold &&
priv->status & STATUS_ASSOCIATED) {
/* If associated and we've hit the missed
* beacon threshold, disassociate, turn
* off roaming, and abort any active scans */
IPW_DEBUG(IPW_DL_INFO | IPW_DL_NOTIF |
IPW_DL_STATE | IPW_DL_ASSOC,
"Missed beacon: %d - disassociate\n", missed_count);
priv->status &= ~STATUS_ROAMING;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (priv->status & STATUS_SCANNING) {
IPW_DEBUG(IPW_DL_INFO | IPW_DL_NOTIF |
IPW_DL_STATE,
"Aborting scan with missed beacon.\n");
queue_work(priv->workqueue, &priv->abort_scan);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
}
queue_work(priv->workqueue, &priv->disassociate);
return;
}
if (priv->status & STATUS_ROAMING) {
/* If we are currently roaming, then just
* print a debug statement... */
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE,
"Missed beacon: %d - roam in progress\n",
missed_count);
return;
}
if (roaming &&
(missed_count > priv->roaming_threshold &&
missed_count <= priv->disassociate_threshold)) {
/* If we are not already roaming, set the ROAM
* bit in the status and kick off a scan.
* This can happen several times before we reach
* disassociate_threshold. */
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE,
"Missed beacon: %d - initiate "
"roaming\n", missed_count);
if (!(priv->status & STATUS_ROAMING)) {
priv->status |= STATUS_ROAMING;
if (!(priv->status & STATUS_SCANNING))
queue_work(priv->workqueue,
&priv->request_scan);
}
return;
}
if (priv->status & STATUS_SCANNING) {
/* Stop scan to keep fw from getting
* stuck (only if we aren't roaming --
* otherwise we'll never scan more than 2 or 3
* channels..) */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_DEBUG(IPW_DL_INFO | IPW_DL_NOTIF | IPW_DL_STATE,
"Aborting scan with missed beacon.\n");
queue_work(priv->workqueue, &priv->abort_scan);
}
IPW_DEBUG_NOTIF("Missed beacon: %d\n", missed_count);
}
2005-03-25 13:31:53 -05:00
/**
* Handle host notification packet.
* Called from interrupt routine
*/
static void ipw_rx_notification(struct ipw_priv *priv,
2005-03-25 13:31:53 -05:00
struct ipw_rx_notification *notif)
{
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
notif->size = le16_to_cpu(notif->size);
IPW_DEBUG_NOTIF("type = %i (%d bytes)\n", notif->subtype, notif->size);
2005-03-25 13:31:53 -05:00
switch (notif->subtype) {
case HOST_NOTIFICATION_STATUS_ASSOCIATED:{
struct notif_association *assoc = &notif->u.assoc;
switch (assoc->state) {
case CMAS_ASSOCIATED:{
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC,
"associated: '%s' " MAC_FMT
" \n",
escape_essid(priv->essid,
priv->essid_len),
MAC_ARG(priv->bssid));
switch (priv->ieee->iw_mode) {
case IW_MODE_INFRA:
memcpy(priv->ieee->bssid,
priv->bssid, ETH_ALEN);
break;
case IW_MODE_ADHOC:
memcpy(priv->ieee->bssid,
priv->bssid, ETH_ALEN);
/* clear out the station table */
priv->num_stations = 0;
IPW_DEBUG_ASSOC
("queueing adhoc check\n");
queue_delayed_work(priv->
workqueue,
&priv->
adhoc_check,
priv->
assoc_request.
beacon_interval);
break;
}
priv->status &= ~STATUS_ASSOCIATING;
priv->status |= STATUS_ASSOCIATED;
queue_work(priv->workqueue,
&priv->system_config);
#ifdef CONFIG_IPW2200_QOS
#define IPW_GET_PACKET_STYPE(x) WLAN_FC_GET_STYPE( \
le16_to_cpu(((struct ieee80211_hdr *)(x))->frame_ctl))
if ((priv->status & STATUS_AUTH) &&
(IPW_GET_PACKET_STYPE(&notif->u.raw)
== IEEE80211_STYPE_ASSOC_RESP)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if ((sizeof
(struct
ieee80211_assoc_response)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
<= notif->size)
&& (notif->size <= 2314)) {
struct
ieee80211_rx_stats
stats = {
.len =
notif->
size - 1,
};
IPW_DEBUG_QOS
("QoS Associate "
"size %d\n",
notif->size);
ieee80211_rx_mgt(priv->
ieee,
(struct
ieee80211_hdr_4addr
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
*)
&notif->u.raw, &stats);
}
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#endif
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
schedule_work(&priv->link_up);
2005-03-25 13:31:53 -05:00
break;
}
case CMAS_AUTHENTICATED:{
if (priv->
status & (STATUS_ASSOCIATED |
STATUS_AUTH)) {
#ifdef CONFIG_IPW2200_DEBUG
struct notif_authenticate *auth
= &notif->u.auth;
IPW_DEBUG(IPW_DL_NOTIF |
IPW_DL_STATE |
IPW_DL_ASSOC,
"deauthenticated: '%s' "
MAC_FMT
": (0x%04X) - %s \n",
escape_essid(priv->
essid,
priv->
essid_len),
MAC_ARG(priv->bssid),
ntohs(auth->status),
ipw_get_status_code
(ntohs
(auth->status)));
2005-03-25 13:31:53 -05:00
#endif
priv->status &=
~(STATUS_ASSOCIATING |
STATUS_AUTH |
STATUS_ASSOCIATED);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
schedule_work(&priv->link_down);
break;
}
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC,
"authenticated: '%s' " MAC_FMT
"\n",
escape_essid(priv->essid,
priv->essid_len),
MAC_ARG(priv->bssid));
break;
}
case CMAS_INIT:{
if (priv->status & STATUS_AUTH) {
struct
ieee80211_assoc_response
*resp;
resp =
(struct
ieee80211_assoc_response
*)&notif->u.raw;
IPW_DEBUG(IPW_DL_NOTIF |
IPW_DL_STATE |
IPW_DL_ASSOC,
"association failed (0x%04X): %s\n",
ntohs(resp->status),
ipw_get_status_code
(ntohs
(resp->status)));
}
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC,
"disassociated: '%s' " MAC_FMT
" \n",
escape_essid(priv->essid,
priv->essid_len),
MAC_ARG(priv->bssid));
priv->status &=
~(STATUS_DISASSOCIATING |
STATUS_ASSOCIATING |
STATUS_ASSOCIATED | STATUS_AUTH);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->assoc_network
&& (priv->assoc_network->
capability &
WLAN_CAPABILITY_IBSS))
ipw_remove_current_network
(priv);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
schedule_work(&priv->link_down);
break;
}
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
case CMAS_RX_ASSOC_RESP:
break;
default:
IPW_ERROR("assoc: unknown (%d)\n",
assoc->state);
2005-03-25 13:31:53 -05:00
break;
}
2005-03-25 13:31:53 -05:00
break;
}
case HOST_NOTIFICATION_STATUS_AUTHENTICATE:{
struct notif_authenticate *auth = &notif->u.auth;
switch (auth->state) {
case CMAS_AUTHENTICATED:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE,
"authenticated: '%s' " MAC_FMT " \n",
escape_essid(priv->essid,
priv->essid_len),
MAC_ARG(priv->bssid));
priv->status |= STATUS_AUTH;
break;
2005-03-25 13:31:53 -05:00
case CMAS_INIT:
if (priv->status & STATUS_AUTH) {
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC,
"authentication failed (0x%04X): %s\n",
ntohs(auth->status),
ipw_get_status_code(ntohs
(auth->
status)));
}
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC,
"deauthenticated: '%s' " MAC_FMT "\n",
escape_essid(priv->essid,
priv->essid_len),
MAC_ARG(priv->bssid));
priv->status &= ~(STATUS_ASSOCIATING |
STATUS_AUTH |
STATUS_ASSOCIATED);
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
schedule_work(&priv->link_down);
break;
2005-03-25 13:31:53 -05:00
case CMAS_TX_AUTH_SEQ_1:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "AUTH_SEQ_1\n");
break;
case CMAS_RX_AUTH_SEQ_2:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "AUTH_SEQ_2\n");
break;
case CMAS_AUTH_SEQ_1_PASS:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "AUTH_SEQ_1_PASS\n");
break;
case CMAS_AUTH_SEQ_1_FAIL:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "AUTH_SEQ_1_FAIL\n");
break;
case CMAS_TX_AUTH_SEQ_3:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "AUTH_SEQ_3\n");
break;
case CMAS_RX_AUTH_SEQ_4:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "RX_AUTH_SEQ_4\n");
break;
case CMAS_AUTH_SEQ_2_PASS:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "AUTH_SEQ_2_PASS\n");
break;
case CMAS_AUTH_SEQ_2_FAIL:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "AUT_SEQ_2_FAIL\n");
break;
case CMAS_TX_ASSOC:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "TX_ASSOC\n");
break;
case CMAS_RX_ASSOC_RESP:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "RX_ASSOC_RESP\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
case CMAS_ASSOCIATED:
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE |
IPW_DL_ASSOC, "ASSOCIATED\n");
break;
default:
IPW_DEBUG_NOTIF("auth: failure - %d\n",
auth->state);
break;
2005-03-25 13:31:53 -05:00
}
break;
}
case HOST_NOTIFICATION_STATUS_SCAN_CHANNEL_RESULT:{
struct notif_channel_result *x =
&notif->u.channel_result;
2005-03-25 13:31:53 -05:00
if (notif->size == sizeof(*x)) {
IPW_DEBUG_SCAN("Scan result for channel %d\n",
x->channel_num);
} else {
IPW_DEBUG_SCAN("Scan result of wrong size %d "
"(should be %zd)\n",
notif->size, sizeof(*x));
}
2005-03-25 13:31:53 -05:00
break;
}
case HOST_NOTIFICATION_STATUS_SCAN_COMPLETED:{
struct notif_scan_complete *x = &notif->u.scan_complete;
if (notif->size == sizeof(*x)) {
IPW_DEBUG_SCAN
("Scan completed: type %d, %d channels, "
"%d status\n", x->scan_type,
x->num_channels, x->status);
} else {
IPW_ERROR("Scan completed of wrong size %d "
"(should be %zd)\n",
notif->size, sizeof(*x));
}
2005-03-25 13:31:53 -05:00
priv->status &=
~(STATUS_SCANNING | STATUS_SCAN_ABORTING);
wake_up_interruptible(&priv->wait_state);
cancel_delayed_work(&priv->scan_check);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->status & STATUS_EXIT_PENDING)
break;
priv->ieee->scans++;
#ifdef CONFIG_IPW2200_MONITOR
if (priv->ieee->iw_mode == IW_MODE_MONITOR) {
priv->status |= STATUS_SCAN_FORCED;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
queue_work(priv->workqueue,
&priv->request_scan);
break;
}
priv->status &= ~STATUS_SCAN_FORCED;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#endif /* CONFIG_IPW2200_MONITOR */
if (!(priv->status & (STATUS_ASSOCIATED |
STATUS_ASSOCIATING |
STATUS_ROAMING |
STATUS_DISASSOCIATING)))
queue_work(priv->workqueue, &priv->associate);
else if (priv->status & STATUS_ROAMING) {
if (x->status == SCAN_COMPLETED_STATUS_COMPLETE)
/* If a scan completed and we are in roam mode, then
* the scan that completed was the one requested as a
* result of entering roam... so, schedule the
* roam work */
queue_work(priv->workqueue,
&priv->roam);
else
/* Don't schedule if we aborted the scan */
priv->status &= ~STATUS_ROAMING;
} else if (priv->status & STATUS_SCAN_PENDING)
queue_work(priv->workqueue,
&priv->request_scan);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
else if (priv->config & CFG_BACKGROUND_SCAN
&& priv->status & STATUS_ASSOCIATED)
queue_delayed_work(priv->workqueue,
&priv->request_scan, HZ);
/* Send an empty event to user space.
* We don't send the received data on the event because
* it would require us to do complex transcoding, and
* we want to minimise the work done in the irq handler
* Use a request to extract the data.
* Also, we generate this even for any scan, regardless
* on how the scan was initiated. User space can just
* sync on periodic scan to get fresh data...
* Jean II */
if (x->status == SCAN_COMPLETED_STATUS_COMPLETE) {
union iwreq_data wrqu;
wrqu.data.length = 0;
wrqu.data.flags = 0;
wireless_send_event(priv->net_dev, SIOCGIWSCAN,
&wrqu, NULL);
}
break;
2005-03-25 13:31:53 -05:00
}
case HOST_NOTIFICATION_STATUS_FRAG_LENGTH:{
struct notif_frag_length *x = &notif->u.frag_len;
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (notif->size == sizeof(*x))
IPW_ERROR("Frag length: %d\n",
le16_to_cpu(x->frag_length));
else
IPW_ERROR("Frag length of wrong size %d "
"(should be %zd)\n",
notif->size, sizeof(*x));
break;
2005-03-25 13:31:53 -05:00
}
case HOST_NOTIFICATION_STATUS_LINK_DETERIORATION:{
struct notif_link_deterioration *x =
&notif->u.link_deterioration;
if (notif->size == sizeof(*x)) {
IPW_DEBUG(IPW_DL_NOTIF | IPW_DL_STATE,
"link deterioration: type %d, cnt %d\n",
x->silence_notification_type,
x->silence_count);
memcpy(&priv->last_link_deterioration, x,
sizeof(*x));
} else {
IPW_ERROR("Link Deterioration of wrong size %d "
"(should be %zd)\n",
notif->size, sizeof(*x));
}
2005-03-25 13:31:53 -05:00
break;
}
case HOST_NOTIFICATION_DINO_CONFIG_RESPONSE:{
IPW_ERROR("Dino config\n");
if (priv->hcmd
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
&& priv->hcmd->cmd != HOST_CMD_DINO_CONFIG)
IPW_ERROR("Unexpected DINO_CONFIG_RESPONSE\n");
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
break;
}
2005-03-25 13:31:53 -05:00
case HOST_NOTIFICATION_STATUS_BEACON_STATE:{
struct notif_beacon_state *x = &notif->u.beacon_state;
if (notif->size != sizeof(*x)) {
IPW_ERROR
("Beacon state of wrong size %d (should "
"be %zd)\n", notif->size, sizeof(*x));
break;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (le32_to_cpu(x->state) ==
HOST_NOTIFICATION_STATUS_BEACON_MISSING)
ipw_handle_missed_beacon(priv,
le32_to_cpu(x->
number));
2005-03-25 13:31:53 -05:00
break;
}
2005-03-25 13:31:53 -05:00
case HOST_NOTIFICATION_STATUS_TGI_TX_KEY:{
struct notif_tgi_tx_key *x = &notif->u.tgi_tx_key;
if (notif->size == sizeof(*x)) {
IPW_ERROR("TGi Tx Key: state 0x%02x sec type "
"0x%02x station %d\n",
x->key_state, x->security_type,
x->station_index);
break;
}
2005-03-25 13:31:53 -05:00
IPW_ERROR
("TGi Tx Key of wrong size %d (should be %zd)\n",
notif->size, sizeof(*x));
2005-03-25 13:31:53 -05:00
break;
}
2005-03-25 13:31:53 -05:00
case HOST_NOTIFICATION_CALIB_KEEP_RESULTS:{
struct notif_calibration *x = &notif->u.calibration;
2005-03-25 13:31:53 -05:00
if (notif->size == sizeof(*x)) {
memcpy(&priv->calib, x, sizeof(*x));
IPW_DEBUG_INFO("TODO: Calibration\n");
break;
}
2005-03-25 13:31:53 -05:00
IPW_ERROR
("Calibration of wrong size %d (should be %zd)\n",
notif->size, sizeof(*x));
2005-03-25 13:31:53 -05:00
break;
}
case HOST_NOTIFICATION_NOISE_STATS:{
if (notif->size == sizeof(u32)) {
priv->exp_avg_noise =
exponential_average(priv->exp_avg_noise,
(u8) (le32_to_cpu(notif->u.noise.value) & 0xff),
DEPTH_NOISE);
break;
}
2005-03-25 13:31:53 -05:00
IPW_ERROR
("Noise stat is wrong size %d (should be %zd)\n",
notif->size, sizeof(u32));
2005-03-25 13:31:53 -05:00
break;
}
default:
IPW_DEBUG_NOTIF("Unknown notification: "
"subtype=%d,flags=0x%2x,size=%d\n",
notif->subtype, notif->flags, notif->size);
2005-03-25 13:31:53 -05:00
}
}
/**
* Destroys all DMA structures and initialise them again
*
2005-03-25 13:31:53 -05:00
* @param priv
* @return error code
*/
static int ipw_queue_reset(struct ipw_priv *priv)
{
int rc = 0;
/** @todo customize queue sizes */
int nTx = 64, nTxCmd = 8;
ipw_tx_queue_free(priv);
/* Tx CMD queue */
rc = ipw_queue_tx_init(priv, &priv->txq_cmd, nTxCmd,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_TX_CMD_QUEUE_READ_INDEX,
IPW_TX_CMD_QUEUE_WRITE_INDEX,
IPW_TX_CMD_QUEUE_BD_BASE,
IPW_TX_CMD_QUEUE_BD_SIZE);
2005-03-25 13:31:53 -05:00
if (rc) {
IPW_ERROR("Tx Cmd queue init failed\n");
goto error;
}
/* Tx queue(s) */
rc = ipw_queue_tx_init(priv, &priv->txq[0], nTx,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_TX_QUEUE_0_READ_INDEX,
IPW_TX_QUEUE_0_WRITE_INDEX,
IPW_TX_QUEUE_0_BD_BASE, IPW_TX_QUEUE_0_BD_SIZE);
2005-03-25 13:31:53 -05:00
if (rc) {
IPW_ERROR("Tx 0 queue init failed\n");
goto error;
}
rc = ipw_queue_tx_init(priv, &priv->txq[1], nTx,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_TX_QUEUE_1_READ_INDEX,
IPW_TX_QUEUE_1_WRITE_INDEX,
IPW_TX_QUEUE_1_BD_BASE, IPW_TX_QUEUE_1_BD_SIZE);
2005-03-25 13:31:53 -05:00
if (rc) {
IPW_ERROR("Tx 1 queue init failed\n");
goto error;
}
rc = ipw_queue_tx_init(priv, &priv->txq[2], nTx,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_TX_QUEUE_2_READ_INDEX,
IPW_TX_QUEUE_2_WRITE_INDEX,
IPW_TX_QUEUE_2_BD_BASE, IPW_TX_QUEUE_2_BD_SIZE);
2005-03-25 13:31:53 -05:00
if (rc) {
IPW_ERROR("Tx 2 queue init failed\n");
goto error;
}
rc = ipw_queue_tx_init(priv, &priv->txq[3], nTx,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_TX_QUEUE_3_READ_INDEX,
IPW_TX_QUEUE_3_WRITE_INDEX,
IPW_TX_QUEUE_3_BD_BASE, IPW_TX_QUEUE_3_BD_SIZE);
2005-03-25 13:31:53 -05:00
if (rc) {
IPW_ERROR("Tx 3 queue init failed\n");
goto error;
}
/* statistics */
priv->rx_bufs_min = 0;
priv->rx_pend_max = 0;
return rc;
error:
2005-03-25 13:31:53 -05:00
ipw_tx_queue_free(priv);
return rc;
}
/**
* Reclaim Tx queue entries no more used by NIC.
*
2005-03-25 13:31:53 -05:00
* When FW adwances 'R' index, all entries between old and
* new 'R' index need to be reclaimed. As result, some free space
* forms. If there is enough free space (> low mark), wake Tx queue.
*
2005-03-25 13:31:53 -05:00
* @note Need to protect against garbage in 'R' index
* @param priv
* @param txq
* @param qindex
* @return Number of used entries remains in the queue
*/
static int ipw_queue_tx_reclaim(struct ipw_priv *priv,
2005-03-25 13:31:53 -05:00
struct clx2_tx_queue *txq, int qindex)
{
u32 hw_tail;
int used;
struct clx2_queue *q = &txq->q;
hw_tail = ipw_read32(priv, q->reg_r);
if (hw_tail >= q->n_bd) {
IPW_ERROR
("Read index for DMA queue (%d) is out of range [0-%d)\n",
hw_tail, q->n_bd);
2005-03-25 13:31:53 -05:00
goto done;
}
for (; q->last_used != hw_tail;
q->last_used = ipw_queue_inc_wrap(q->last_used, q->n_bd)) {
ipw_queue_tx_free_tfd(priv, txq);
priv->tx_packets++;
}
done:
if ((ipw_queue_space(q) > q->low_mark) &&
(qindex >= 0) &&
(priv->status & STATUS_ASSOCIATED) && netif_running(priv->net_dev))
netif_wake_queue(priv->net_dev);
2005-03-25 13:31:53 -05:00
used = q->first_empty - q->last_used;
if (used < 0)
used += q->n_bd;
return used;
}
static int ipw_queue_tx_hcmd(struct ipw_priv *priv, int hcmd, void *buf,
int len, int sync)
{
struct clx2_tx_queue *txq = &priv->txq_cmd;
struct clx2_queue *q = &txq->q;
struct tfd_frame *tfd;
if (ipw_queue_space(q) < (sync ? 1 : 2)) {
IPW_ERROR("No space for Tx\n");
return -EBUSY;
}
tfd = &txq->bd[q->first_empty];
txq->txb[q->first_empty] = NULL;
memset(tfd, 0, sizeof(*tfd));
tfd->control_flags.message_type = TX_HOST_COMMAND_TYPE;
tfd->control_flags.control_bits = TFD_NEED_IRQ_MASK;
priv->hcmd_seq++;
tfd->u.cmd.index = hcmd;
tfd->u.cmd.length = len;
memcpy(tfd->u.cmd.payload, buf, len);
q->first_empty = ipw_queue_inc_wrap(q->first_empty, q->n_bd);
ipw_write32(priv, q->reg_w, q->first_empty);
_ipw_read32(priv, 0x90);
return 0;
}
/*
2005-03-25 13:31:53 -05:00
* Rx theory of operation
*
* The host allocates 32 DMA target addresses and passes the host address
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
* to the firmware at register IPW_RFDS_TABLE_LOWER + N * RFD_SIZE where N is
2005-03-25 13:31:53 -05:00
* 0 to 31
*
* Rx Queue Indexes
* The host/firmware share two index registers for managing the Rx buffers.
*
* The READ index maps to the first position that the firmware may be writing
* to -- the driver can read up to (but not including) this position and get
* good data.
2005-03-25 13:31:53 -05:00
* The READ index is managed by the firmware once the card is enabled.
*
* The WRITE index maps to the last position the driver has read from -- the
* position preceding WRITE is the last slot the firmware can place a packet.
*
* The queue is empty (no good data) if WRITE = READ - 1, and is full if
* WRITE = READ.
2005-03-25 13:31:53 -05:00
*
* During initialization the host sets up the READ queue position to the first
2005-03-25 13:31:53 -05:00
* INDEX position, and WRITE to the last (READ - 1 wrapped)
*
* When the firmware places a packet in a buffer it will advance the READ index
* and fire the RX interrupt. The driver can then query the READ index and
* process as many packets as possible, moving the WRITE index forward as it
* resets the Rx queue buffers with new memory.
*
2005-03-25 13:31:53 -05:00
* The management in the driver is as follows:
* + A list of pre-allocated SKBs is stored in ipw->rxq->rx_free. When
2005-03-25 13:31:53 -05:00
* ipw->rxq->free_count drops to or below RX_LOW_WATERMARK, work is scheduled
* to replensish the ipw->rxq->rx_free.
2005-03-25 13:31:53 -05:00
* + In ipw_rx_queue_replenish (scheduled) if 'processed' != 'read' then the
* ipw->rxq is replenished and the READ INDEX is updated (updating the
* 'processed' and 'read' driver indexes as well)
* + A received packet is processed and handed to the kernel network stack,
* detached from the ipw->rxq. The driver 'processed' index is updated.
* + The Host/Firmware ipw->rxq is replenished at tasklet time from the rx_free
* list. If there are no allocated buffers in ipw->rxq->rx_free, the READ
* INDEX is not incremented and ipw->status(RX_STALLED) is set. If there
2005-03-25 13:31:53 -05:00
* were enough free buffers and RX_STALLED is set it is cleared.
*
*
* Driver sequence:
*
* ipw_rx_queue_alloc() Allocates rx_free
2005-03-25 13:31:53 -05:00
* ipw_rx_queue_replenish() Replenishes rx_free list from rx_used, and calls
* ipw_rx_queue_restock
* ipw_rx_queue_restock() Moves available buffers from rx_free into Rx
* queue, updates firmware pointers, and updates
* the WRITE index. If insufficient rx_free buffers
* are available, schedules ipw_rx_queue_replenish
*
* -- enable interrupts --
* ISR - ipw_rx() Detach ipw_rx_mem_buffers from pool up to the
* READ INDEX, detaching the SKB from the pool.
2005-03-25 13:31:53 -05:00
* Moves the packet buffer from queue to rx_used.
* Calls ipw_rx_queue_restock to refill any empty
* slots.
* ...
*
*/
/*
2005-03-25 13:31:53 -05:00
* If there are slots in the RX queue that need to be restocked,
* and we have free pre-allocated buffers, fill the ranks as much
* as we can pulling from rx_free.
*
* This moves the 'write' index forward to catch up with 'processed', and
* also updates the memory address in the firmware to reference the new
* target buffer.
*/
static void ipw_rx_queue_restock(struct ipw_priv *priv)
{
struct ipw_rx_queue *rxq = priv->rxq;
struct list_head *element;
struct ipw_rx_mem_buffer *rxb;
unsigned long flags;
int write;
spin_lock_irqsave(&rxq->lock, flags);
write = rxq->write;
while ((rxq->write != rxq->processed) && (rxq->free_count)) {
element = rxq->rx_free.next;
rxb = list_entry(element, struct ipw_rx_mem_buffer, list);
list_del(element);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_RFDS_TABLE_LOWER + rxq->write * RFD_SIZE,
2005-03-25 13:31:53 -05:00
rxb->dma_addr);
rxq->queue[rxq->write] = rxb;
rxq->write = (rxq->write + 1) % RX_QUEUE_SIZE;
rxq->free_count--;
}
spin_unlock_irqrestore(&rxq->lock, flags);
/* If the pre-allocated buffer pool is dropping low, schedule to
2005-03-25 13:31:53 -05:00
* refill it */
if (rxq->free_count <= RX_LOW_WATERMARK)
queue_work(priv->workqueue, &priv->rx_replenish);
/* If we've added more space for the firmware to place data, tell it */
if (write != rxq->write)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_write32(priv, IPW_RX_WRITE_INDEX, rxq->write);
2005-03-25 13:31:53 -05:00
}
/*
* Move all used packet from rx_used to rx_free, allocating a new SKB for each.
* Also restock the Rx queue via ipw_rx_queue_restock.
*
2005-03-25 13:31:53 -05:00
* This is called as a scheduled work item (except for during intialization)
*/
static void ipw_rx_queue_replenish(void *data)
{
struct ipw_priv *priv = data;
struct ipw_rx_queue *rxq = priv->rxq;
struct list_head *element;
struct ipw_rx_mem_buffer *rxb;
unsigned long flags;
spin_lock_irqsave(&rxq->lock, flags);
while (!list_empty(&rxq->rx_used)) {
element = rxq->rx_used.next;
rxb = list_entry(element, struct ipw_rx_mem_buffer, list);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
rxb->skb = alloc_skb(IPW_RX_BUF_SIZE, GFP_ATOMIC);
2005-03-25 13:31:53 -05:00
if (!rxb->skb) {
printk(KERN_CRIT "%s: Can not allocate SKB buffers.\n",
priv->net_dev->name);
/* We don't reschedule replenish work here -- we will
* call the restock method and if it still needs
* more buffers it will schedule replenish */
break;
}
list_del(element);
2005-03-25 13:31:53 -05:00
rxb->rxb = (struct ipw_rx_buffer *)rxb->skb->data;
rxb->dma_addr =
pci_map_single(priv->pci_dev, rxb->skb->data,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_RX_BUF_SIZE, PCI_DMA_FROMDEVICE);
2005-03-25 13:31:53 -05:00
list_add_tail(&rxb->list, &rxq->rx_free);
rxq->free_count++;
}
spin_unlock_irqrestore(&rxq->lock, flags);
ipw_rx_queue_restock(priv);
}
static void ipw_bg_rx_queue_replenish(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_rx_queue_replenish(data);
mutex_unlock(&priv->mutex);
}
2005-03-25 13:31:53 -05:00
/* Assumes that the skb field of the buffers in 'pool' is kept accurate.
* If an SKB has been detached, the POOL needs to have its SKB set to NULL
* This free routine walks the list of POOL entries and if SKB is set to
2005-03-25 13:31:53 -05:00
* non NULL it is unmapped and freed
*/
static void ipw_rx_queue_free(struct ipw_priv *priv, struct ipw_rx_queue *rxq)
2005-03-25 13:31:53 -05:00
{
int i;
if (!rxq)
return;
2005-03-25 13:31:53 -05:00
for (i = 0; i < RX_QUEUE_SIZE + RX_FREE_BUFFERS; i++) {
if (rxq->pool[i].skb != NULL) {
pci_unmap_single(priv->pci_dev, rxq->pool[i].dma_addr,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_RX_BUF_SIZE, PCI_DMA_FROMDEVICE);
2005-03-25 13:31:53 -05:00
dev_kfree_skb(rxq->pool[i].skb);
}
}
kfree(rxq);
}
static struct ipw_rx_queue *ipw_rx_queue_alloc(struct ipw_priv *priv)
{
struct ipw_rx_queue *rxq;
int i;
rxq = kzalloc(sizeof(*rxq), GFP_KERNEL);
if (unlikely(!rxq)) {
IPW_ERROR("memory allocation failed\n");
return NULL;
}
2005-03-25 13:31:53 -05:00
spin_lock_init(&rxq->lock);
INIT_LIST_HEAD(&rxq->rx_free);
INIT_LIST_HEAD(&rxq->rx_used);
/* Fill the rx_used queue with _all_ of the Rx buffers */
for (i = 0; i < RX_FREE_BUFFERS + RX_QUEUE_SIZE; i++)
2005-03-25 13:31:53 -05:00
list_add_tail(&rxq->pool[i].list, &rxq->rx_used);
/* Set us so that we have processed and used all buffers, but have
* not restocked the Rx queue with fresh buffers */
rxq->read = rxq->write = 0;
rxq->processed = RX_QUEUE_SIZE - 1;
rxq->free_count = 0;
return rxq;
}
static int ipw_is_rate_in_mask(struct ipw_priv *priv, int ieee_mode, u8 rate)
{
rate &= ~IEEE80211_BASIC_RATE_MASK;
if (ieee_mode == IEEE_A) {
switch (rate) {
case IEEE80211_OFDM_RATE_6MB:
return priv->rates_mask & IEEE80211_OFDM_RATE_6MB_MASK ?
1 : 0;
case IEEE80211_OFDM_RATE_9MB:
return priv->rates_mask & IEEE80211_OFDM_RATE_9MB_MASK ?
1 : 0;
case IEEE80211_OFDM_RATE_12MB:
return priv->
rates_mask & IEEE80211_OFDM_RATE_12MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_18MB:
return priv->
rates_mask & IEEE80211_OFDM_RATE_18MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_24MB:
return priv->
rates_mask & IEEE80211_OFDM_RATE_24MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_36MB:
return priv->
rates_mask & IEEE80211_OFDM_RATE_36MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_48MB:
return priv->
rates_mask & IEEE80211_OFDM_RATE_48MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_54MB:
return priv->
rates_mask & IEEE80211_OFDM_RATE_54MB_MASK ? 1 : 0;
2005-03-25 13:31:53 -05:00
default:
return 0;
}
}
2005-03-25 13:31:53 -05:00
/* B and G mixed */
switch (rate) {
case IEEE80211_CCK_RATE_1MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_CCK_RATE_1MB_MASK ? 1 : 0;
case IEEE80211_CCK_RATE_2MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_CCK_RATE_2MB_MASK ? 1 : 0;
case IEEE80211_CCK_RATE_5MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_CCK_RATE_5MB_MASK ? 1 : 0;
case IEEE80211_CCK_RATE_11MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_CCK_RATE_11MB_MASK ? 1 : 0;
}
/* If we are limited to B modulations, bail at this point */
if (ieee_mode == IEEE_B)
return 0;
/* G */
switch (rate) {
case IEEE80211_OFDM_RATE_6MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_6MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_9MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_9MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_12MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_12MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_18MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_18MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_24MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_24MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_36MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_36MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_48MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_48MB_MASK ? 1 : 0;
case IEEE80211_OFDM_RATE_54MB:
2005-03-25 13:31:53 -05:00
return priv->rates_mask & IEEE80211_OFDM_RATE_54MB_MASK ? 1 : 0;
}
return 0;
}
static int ipw_compatible_rates(struct ipw_priv *priv,
2005-03-25 13:31:53 -05:00
const struct ieee80211_network *network,
struct ipw_supported_rates *rates)
{
int num_rates, i;
memset(rates, 0, sizeof(*rates));
num_rates = min(network->rates_len, (u8) IPW_MAX_RATES);
2005-03-25 13:31:53 -05:00
rates->num_rates = 0;
for (i = 0; i < num_rates; i++) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (!ipw_is_rate_in_mask(priv, network->mode,
network->rates[i])) {
if (network->rates[i] & IEEE80211_BASIC_RATE_MASK) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
IPW_DEBUG_SCAN("Adding masked mandatory "
"rate %02X\n",
network->rates[i]);
rates->supported_rates[rates->num_rates++] =
network->rates[i];
continue;
}
2005-03-25 13:31:53 -05:00
IPW_DEBUG_SCAN("Rate %02X masked : 0x%08X\n",
network->rates[i], priv->rates_mask);
continue;
}
2005-03-25 13:31:53 -05:00
rates->supported_rates[rates->num_rates++] = network->rates[i];
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
num_rates = min(network->rates_ex_len,
(u8) (IPW_MAX_RATES - num_rates));
2005-03-25 13:31:53 -05:00
for (i = 0; i < num_rates; i++) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (!ipw_is_rate_in_mask(priv, network->mode,
network->rates_ex[i])) {
if (network->rates_ex[i] & IEEE80211_BASIC_RATE_MASK) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
IPW_DEBUG_SCAN("Adding masked mandatory "
"rate %02X\n",
network->rates_ex[i]);
rates->supported_rates[rates->num_rates++] =
network->rates[i];
continue;
}
2005-03-25 13:31:53 -05:00
IPW_DEBUG_SCAN("Rate %02X masked : 0x%08X\n",
network->rates_ex[i], priv->rates_mask);
continue;
}
rates->supported_rates[rates->num_rates++] =
network->rates_ex[i];
2005-03-25 13:31:53 -05:00
}
return 1;
2005-03-25 13:31:53 -05:00
}
static void ipw_copy_rates(struct ipw_supported_rates *dest,
2005-03-25 13:31:53 -05:00
const struct ipw_supported_rates *src)
{
u8 i;
for (i = 0; i < src->num_rates; i++)
dest->supported_rates[i] = src->supported_rates[i];
dest->num_rates = src->num_rates;
}
/* TODO: Look at sniffed packets in the air to determine if the basic rate
* mask should ever be used -- right now all callers to add the scan rates are
* set with the modulation = CCK, so BASIC_RATE_MASK is never set... */
static void ipw_add_cck_scan_rates(struct ipw_supported_rates *rates,
u8 modulation, u32 rate_mask)
2005-03-25 13:31:53 -05:00
{
u8 basic_mask = (IEEE80211_OFDM_MODULATION == modulation) ?
IEEE80211_BASIC_RATE_MASK : 0;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_CCK_RATE_1MB_MASK)
rates->supported_rates[rates->num_rates++] =
IEEE80211_BASIC_RATE_MASK | IEEE80211_CCK_RATE_1MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_CCK_RATE_2MB_MASK)
rates->supported_rates[rates->num_rates++] =
IEEE80211_BASIC_RATE_MASK | IEEE80211_CCK_RATE_2MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_CCK_RATE_5MB_MASK)
rates->supported_rates[rates->num_rates++] = basic_mask |
IEEE80211_CCK_RATE_5MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_CCK_RATE_11MB_MASK)
rates->supported_rates[rates->num_rates++] = basic_mask |
IEEE80211_CCK_RATE_11MB;
2005-03-25 13:31:53 -05:00
}
static void ipw_add_ofdm_scan_rates(struct ipw_supported_rates *rates,
u8 modulation, u32 rate_mask)
2005-03-25 13:31:53 -05:00
{
u8 basic_mask = (IEEE80211_OFDM_MODULATION == modulation) ?
IEEE80211_BASIC_RATE_MASK : 0;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_6MB_MASK)
rates->supported_rates[rates->num_rates++] = basic_mask |
IEEE80211_OFDM_RATE_6MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_9MB_MASK)
rates->supported_rates[rates->num_rates++] =
IEEE80211_OFDM_RATE_9MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_12MB_MASK)
rates->supported_rates[rates->num_rates++] = basic_mask |
IEEE80211_OFDM_RATE_12MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_18MB_MASK)
rates->supported_rates[rates->num_rates++] =
IEEE80211_OFDM_RATE_18MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_24MB_MASK)
rates->supported_rates[rates->num_rates++] = basic_mask |
IEEE80211_OFDM_RATE_24MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_36MB_MASK)
rates->supported_rates[rates->num_rates++] =
IEEE80211_OFDM_RATE_36MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_48MB_MASK)
rates->supported_rates[rates->num_rates++] =
IEEE80211_OFDM_RATE_48MB;
2005-03-25 13:31:53 -05:00
if (rate_mask & IEEE80211_OFDM_RATE_54MB_MASK)
rates->supported_rates[rates->num_rates++] =
IEEE80211_OFDM_RATE_54MB;
2005-03-25 13:31:53 -05:00
}
struct ipw_network_match {
struct ieee80211_network *network;
struct ipw_supported_rates rates;
};
static int ipw_find_adhoc_network(struct ipw_priv *priv,
struct ipw_network_match *match,
struct ieee80211_network *network,
int roaming)
2005-03-25 13:31:53 -05:00
{
struct ipw_supported_rates rates;
/* Verify that this network's capability is compatible with the
* current mode (AdHoc or Infrastructure) */
if ((priv->ieee->iw_mode == IW_MODE_ADHOC &&
2005-03-25 13:31:53 -05:00
!(network->capability & WLAN_CAPABILITY_IBSS))) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded due to "
"capability mismatch.\n",
2005-03-25 13:31:53 -05:00
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
/* If we do not have an ESSID for this AP, we can not associate with
* it */
if (network->flags & NETWORK_EMPTY_ESSID) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
2005-03-25 13:31:53 -05:00
"because of hidden ESSID.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
2005-03-25 13:31:53 -05:00
if (unlikely(roaming)) {
/* If we are roaming, then ensure check if this is a valid
* network to try and roam to */
if ((network->ssid_len != match->network->ssid_len) ||
memcmp(network->ssid, match->network->ssid,
2005-03-25 13:31:53 -05:00
network->ssid_len)) {
IPW_DEBUG_MERGE("Netowrk '%s (" MAC_FMT ")' excluded "
2005-03-25 13:31:53 -05:00
"because of non-network ESSID.\n",
escape_essid(network->ssid,
2005-03-25 13:31:53 -05:00
network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
} else {
/* If an ESSID has been configured then compare the broadcast
* ESSID to ours */
if ((priv->config & CFG_STATIC_ESSID) &&
2005-03-25 13:31:53 -05:00
((network->ssid_len != priv->essid_len) ||
memcmp(network->ssid, priv->essid,
2005-03-25 13:31:53 -05:00
min(network->ssid_len, priv->essid_len)))) {
char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
strncpy(escaped,
escape_essid(network->ssid, network->ssid_len),
2005-03-25 13:31:53 -05:00
sizeof(escaped));
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
"because of ESSID mismatch: '%s'.\n",
2005-03-25 13:31:53 -05:00
escaped, MAC_ARG(network->bssid),
escape_essid(priv->essid,
priv->essid_len));
2005-03-25 13:31:53 -05:00
return 0;
}
}
/* If the old network rate is better than this one, don't bother
* testing everything else. */
if (network->time_stamp[0] < match->network->time_stamp[0]) {
IPW_DEBUG_MERGE("Network '%s excluded because newer than "
"current network.\n",
2005-03-25 13:31:53 -05:00
escape_essid(match->network->ssid,
match->network->ssid_len));
2005-03-25 13:31:53 -05:00
return 0;
} else if (network->time_stamp[1] < match->network->time_stamp[1]) {
IPW_DEBUG_MERGE("Network '%s excluded because newer than "
"current network.\n",
escape_essid(match->network->ssid,
match->network->ssid_len));
2005-03-25 13:31:53 -05:00
return 0;
}
/* Now go through and see if the requested network is valid... */
if (priv->ieee->scan_age != 0 &&
time_after(jiffies, network->last_scanned + priv->ieee->scan_age)) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
"because of age: %ums.\n",
2005-03-25 13:31:53 -05:00
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid),
jiffies_to_msecs(jiffies -
network->last_scanned));
2005-03-25 13:31:53 -05:00
return 0;
}
2005-03-25 13:31:53 -05:00
if ((priv->config & CFG_STATIC_CHANNEL) &&
2005-03-25 13:31:53 -05:00
(network->channel != priv->channel)) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
2005-03-25 13:31:53 -05:00
"because of channel mismatch: %d != %d.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid),
network->channel, priv->channel);
return 0;
}
2005-03-25 13:31:53 -05:00
/* Verify privacy compatability */
if (((priv->capability & CAP_PRIVACY_ON) ? 1 : 0) !=
2005-03-25 13:31:53 -05:00
((network->capability & WLAN_CAPABILITY_PRIVACY) ? 1 : 0)) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
2005-03-25 13:31:53 -05:00
"because of privacy mismatch: %s != %s.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid),
priv->
capability & CAP_PRIVACY_ON ? "on" : "off",
network->
capability & WLAN_CAPABILITY_PRIVACY ? "on" :
"off");
2005-03-25 13:31:53 -05:00
return 0;
}
if (!memcmp(network->bssid, priv->bssid, ETH_ALEN)) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
"because of the same BSSID match: " MAC_FMT
".\n", escape_essid(network->ssid,
network->ssid_len),
MAC_ARG(network->bssid), MAC_ARG(priv->bssid));
2005-03-25 13:31:53 -05:00
return 0;
}
2005-03-25 13:31:53 -05:00
/* Filter out any incompatible freq / mode combinations */
if (!ieee80211_is_valid_mode(priv->ieee, network->mode)) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
2005-03-25 13:31:53 -05:00
"because of invalid frequency/mode "
"combination.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
/* Ensure that the rates supported by the driver are compatible with
* this AP, including verification of basic rates (mandatory) */
if (!ipw_compatible_rates(priv, network, &rates)) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
"because configured rate mask excludes "
"AP mandatory rate.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
2005-03-25 13:31:53 -05:00
if (rates.num_rates == 0) {
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' excluded "
2005-03-25 13:31:53 -05:00
"because of no compatible rates.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
2005-03-25 13:31:53 -05:00
/* TODO: Perform any further minimal comparititive tests. We do not
* want to put too much policy logic here; intelligent scan selection
* should occur within a generic IEEE 802.11 user space tool. */
/* Set up 'new' AP to this network */
ipw_copy_rates(&match->rates, &rates);
match->network = network;
IPW_DEBUG_MERGE("Network '%s (" MAC_FMT ")' is a viable match.\n",
2005-03-25 13:31:53 -05:00
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 1;
}
static void ipw_merge_adhoc_network(void *data)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = data;
struct ieee80211_network *network = NULL;
struct ipw_network_match match = {
.network = priv->assoc_network
};
if ((priv->status & STATUS_ASSOCIATED) &&
(priv->ieee->iw_mode == IW_MODE_ADHOC)) {
/* First pass through ROAM process -- look for a better
* network */
unsigned long flags;
spin_lock_irqsave(&priv->ieee->lock, flags);
list_for_each_entry(network, &priv->ieee->network_list, list) {
if (network != priv->assoc_network)
ipw_find_adhoc_network(priv, &match, network,
1);
}
spin_unlock_irqrestore(&priv->ieee->lock, flags);
if (match.network == priv->assoc_network) {
IPW_DEBUG_MERGE("No better ADHOC in this network to "
"merge to.\n");
return;
}
mutex_lock(&priv->mutex);
if ((priv->ieee->iw_mode == IW_MODE_ADHOC)) {
IPW_DEBUG_MERGE("remove network %s\n",
escape_essid(priv->essid,
priv->essid_len));
ipw_remove_current_network(priv);
2005-03-25 13:31:53 -05:00
}
ipw_disassociate(priv);
priv->assoc_network = match.network;
mutex_unlock(&priv->mutex);
return;
2005-03-25 13:31:53 -05:00
}
}
2005-03-25 13:31:53 -05:00
static int ipw_best_network(struct ipw_priv *priv,
struct ipw_network_match *match,
struct ieee80211_network *network, int roaming)
2005-03-25 13:31:53 -05:00
{
struct ipw_supported_rates rates;
/* Verify that this network's capability is compatible with the
* current mode (AdHoc or Infrastructure) */
if ((priv->ieee->iw_mode == IW_MODE_INFRA &&
!(network->capability & WLAN_CAPABILITY_ESS)) ||
2005-03-25 13:31:53 -05:00
(priv->ieee->iw_mode == IW_MODE_ADHOC &&
!(network->capability & WLAN_CAPABILITY_IBSS))) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded due to "
"capability mismatch.\n",
2005-03-25 13:31:53 -05:00
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
/* If we do not have an ESSID for this AP, we can not associate with
* it */
if (network->flags & NETWORK_EMPTY_ESSID) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of hidden ESSID.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
2005-03-25 13:31:53 -05:00
if (unlikely(roaming)) {
/* If we are roaming, then ensure check if this is a valid
* network to try and roam to */
if ((network->ssid_len != match->network->ssid_len) ||
memcmp(network->ssid, match->network->ssid,
2005-03-25 13:31:53 -05:00
network->ssid_len)) {
IPW_DEBUG_ASSOC("Netowrk '%s (" MAC_FMT ")' excluded "
"because of non-network ESSID.\n",
escape_essid(network->ssid,
2005-03-25 13:31:53 -05:00
network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
} else {
/* If an ESSID has been configured then compare the broadcast
* ESSID to ours */
if ((priv->config & CFG_STATIC_ESSID) &&
2005-03-25 13:31:53 -05:00
((network->ssid_len != priv->essid_len) ||
memcmp(network->ssid, priv->essid,
2005-03-25 13:31:53 -05:00
min(network->ssid_len, priv->essid_len)))) {
char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
strncpy(escaped,
escape_essid(network->ssid, network->ssid_len),
2005-03-25 13:31:53 -05:00
sizeof(escaped));
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of ESSID mismatch: '%s'.\n",
2005-03-25 13:31:53 -05:00
escaped, MAC_ARG(network->bssid),
escape_essid(priv->essid,
priv->essid_len));
2005-03-25 13:31:53 -05:00
return 0;
}
}
/* If the old network rate is better than this one, don't bother
* testing everything else. */
if (match->network && match->network->stats.rssi > network->stats.rssi) {
2005-03-25 13:31:53 -05:00
char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
strncpy(escaped,
escape_essid(network->ssid, network->ssid_len),
2005-03-25 13:31:53 -05:00
sizeof(escaped));
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded because "
"'%s (" MAC_FMT ")' has a stronger signal.\n",
escaped, MAC_ARG(network->bssid),
escape_essid(match->network->ssid,
match->network->ssid_len),
MAC_ARG(match->network->bssid));
return 0;
}
2005-03-25 13:31:53 -05:00
/* If this network has already had an association attempt within the
* last 3 seconds, do not try and associate again... */
if (network->last_associate &&
time_after(network->last_associate + (HZ * 3UL), jiffies)) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of storming (%ums since last "
2005-03-25 13:31:53 -05:00
"assoc attempt).\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid),
jiffies_to_msecs(jiffies -
network->last_associate));
2005-03-25 13:31:53 -05:00
return 0;
}
/* Now go through and see if the requested network is valid... */
if (priv->ieee->scan_age != 0 &&
time_after(jiffies, network->last_scanned + priv->ieee->scan_age)) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of age: %ums.\n",
2005-03-25 13:31:53 -05:00
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid),
jiffies_to_msecs(jiffies -
network->last_scanned));
2005-03-25 13:31:53 -05:00
return 0;
}
2005-03-25 13:31:53 -05:00
if ((priv->config & CFG_STATIC_CHANNEL) &&
2005-03-25 13:31:53 -05:00
(network->channel != priv->channel)) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of channel mismatch: %d != %d.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid),
network->channel, priv->channel);
return 0;
}
2005-03-25 13:31:53 -05:00
/* Verify privacy compatability */
if (((priv->capability & CAP_PRIVACY_ON) ? 1 : 0) !=
2005-03-25 13:31:53 -05:00
((network->capability & WLAN_CAPABILITY_PRIVACY) ? 1 : 0)) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of privacy mismatch: %s != %s.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid),
priv->capability & CAP_PRIVACY_ON ? "on" :
2005-03-25 13:31:53 -05:00
"off",
network->capability &
WLAN_CAPABILITY_PRIVACY ? "on" : "off");
2005-03-25 13:31:53 -05:00
return 0;
}
if ((priv->config & CFG_STATIC_BSSID) &&
2005-03-25 13:31:53 -05:00
memcmp(network->bssid, priv->bssid, ETH_ALEN)) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of BSSID mismatch: " MAC_FMT ".\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid), MAC_ARG(priv->bssid));
2005-03-25 13:31:53 -05:00
return 0;
}
2005-03-25 13:31:53 -05:00
/* Filter out any incompatible freq / mode combinations */
if (!ieee80211_is_valid_mode(priv->ieee, network->mode)) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of invalid frequency/mode "
"combination.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
/* Filter out invalid channel in current GEO */
if (!ieee80211_is_valid_channel(priv->ieee, network->channel)) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of invalid channel in current GEO\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
/* Ensure that the rates supported by the driver are compatible with
* this AP, including verification of basic rates (mandatory) */
if (!ipw_compatible_rates(priv, network, &rates)) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because configured rate mask excludes "
"AP mandatory rate.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
2005-03-25 13:31:53 -05:00
if (rates.num_rates == 0) {
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' excluded "
"because of no compatible rates.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 0;
}
2005-03-25 13:31:53 -05:00
/* TODO: Perform any further minimal comparititive tests. We do not
* want to put too much policy logic here; intelligent scan selection
* should occur within a generic IEEE 802.11 user space tool. */
/* Set up 'new' AP to this network */
ipw_copy_rates(&match->rates, &rates);
match->network = network;
IPW_DEBUG_ASSOC("Network '%s (" MAC_FMT ")' is a viable match.\n",
escape_essid(network->ssid, network->ssid_len),
MAC_ARG(network->bssid));
return 1;
}
static void ipw_adhoc_create(struct ipw_priv *priv,
struct ieee80211_network *network)
2005-03-25 13:31:53 -05:00
{
const struct ieee80211_geo *geo = ieee80211_get_geo(priv->ieee);
int i;
2005-03-25 13:31:53 -05:00
/*
* For the purposes of scanning, we can set our wireless mode
* to trigger scans across combinations of bands, but when it
* comes to creating a new ad-hoc network, we have tell the FW
* exactly which band to use.
*
* We also have the possibility of an invalid channel for the
2005-03-25 13:31:53 -05:00
* chossen band. Attempting to create a new ad-hoc network
* with an invalid channel for wireless mode will trigger a
* FW fatal error.
*
2005-03-25 13:31:53 -05:00
*/
switch (ieee80211_is_valid_channel(priv->ieee, priv->channel)) {
case IEEE80211_52GHZ_BAND:
network->mode = IEEE_A;
i = ieee80211_channel_to_index(priv->ieee, priv->channel);
BUG_ON(i == -1);
if (geo->a[i].flags & IEEE80211_CH_PASSIVE_ONLY) {
IPW_WARNING("Overriding invalid channel\n");
priv->channel = geo->a[0].channel;
}
break;
case IEEE80211_24GHZ_BAND:
if (priv->ieee->mode & IEEE_G)
network->mode = IEEE_G;
else
network->mode = IEEE_B;
i = ieee80211_channel_to_index(priv->ieee, priv->channel);
BUG_ON(i == -1);
if (geo->bg[i].flags & IEEE80211_CH_PASSIVE_ONLY) {
IPW_WARNING("Overriding invalid channel\n");
priv->channel = geo->bg[0].channel;
}
break;
default:
2005-03-25 13:31:53 -05:00
IPW_WARNING("Overriding invalid channel\n");
if (priv->ieee->mode & IEEE_A) {
network->mode = IEEE_A;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->channel = geo->a[0].channel;
2005-03-25 13:31:53 -05:00
} else if (priv->ieee->mode & IEEE_G) {
network->mode = IEEE_G;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->channel = geo->bg[0].channel;
2005-03-25 13:31:53 -05:00
} else {
network->mode = IEEE_B;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->channel = geo->bg[0].channel;
2005-03-25 13:31:53 -05:00
}
break;
}
2005-03-25 13:31:53 -05:00
network->channel = priv->channel;
priv->config |= CFG_ADHOC_PERSIST;
ipw_create_bssid(priv, network->bssid);
network->ssid_len = priv->essid_len;
memcpy(network->ssid, priv->essid, priv->essid_len);
memset(&network->stats, 0, sizeof(network->stats));
network->capability = WLAN_CAPABILITY_IBSS;
if (!(priv->config & CFG_PREAMBLE_LONG))
network->capability |= WLAN_CAPABILITY_SHORT_PREAMBLE;
2005-03-25 13:31:53 -05:00
if (priv->capability & CAP_PRIVACY_ON)
network->capability |= WLAN_CAPABILITY_PRIVACY;
network->rates_len = min(priv->rates.num_rates, MAX_RATES_LENGTH);
memcpy(network->rates, priv->rates.supported_rates, network->rates_len);
2005-03-25 13:31:53 -05:00
network->rates_ex_len = priv->rates.num_rates - network->rates_len;
memcpy(network->rates_ex,
2005-03-25 13:31:53 -05:00
&priv->rates.supported_rates[network->rates_len],
network->rates_ex_len);
network->last_scanned = 0;
network->flags = 0;
network->last_associate = 0;
network->time_stamp[0] = 0;
network->time_stamp[1] = 0;
network->beacon_interval = 100; /* Default */
network->listen_interval = 10; /* Default */
network->atim_window = 0; /* Default */
2005-03-25 13:31:53 -05:00
network->wpa_ie_len = 0;
network->rsn_ie_len = 0;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_send_tgi_tx_key(struct ipw_priv *priv, int type, int index)
{
struct ipw_tgi_tx_key key;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (!(priv->ieee->sec.flags & (1 << index)))
return;
key.key_id = index;
memcpy(key.key, priv->ieee->sec.keys[index], SCM_TEMPORAL_KEY_LENGTH);
key.security_type = type;
key.station_index = 0; /* always 0 for BSS */
key.flags = 0;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* 0 for new key; previous value of counter (after fatal error) */
key.tx_counter[0] = 0;
key.tx_counter[1] = 0;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_send_cmd_pdu(priv, IPW_CMD_TGI_TX_KEY, sizeof(key), &key);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
static void ipw_send_wep_keys(struct ipw_priv *priv, int type)
2005-03-25 13:31:53 -05:00
{
struct ipw_wep_key key;
2005-03-25 13:31:53 -05:00
int i;
key.cmd_id = DINO_CMD_WEP_KEY;
key.seq_num = 0;
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* Note: AES keys cannot be set for multiple times.
* Only set it at the first time. */
for (i = 0; i < 4; i++) {
key.key_index = i | type;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (!(priv->ieee->sec.flags & (1 << i))) {
key.key_size = 0;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
continue;
2005-03-25 13:31:53 -05:00
}
key.key_size = priv->ieee->sec.key_sizes[i];
memcpy(key.key, priv->ieee->sec.keys[i], key.key_size);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_send_cmd_pdu(priv, IPW_CMD_WEP_KEY, sizeof(key), &key);
}
2005-03-25 13:31:53 -05:00
}
static void ipw_set_hw_decrypt_unicast(struct ipw_priv *priv, int level)
2005-03-25 13:31:53 -05:00
{
if (priv->ieee->host_encrypt)
2005-03-25 13:31:53 -05:00
return;
switch (level) {
case SEC_LEVEL_3:
priv->sys_config.disable_unicast_decryption = 0;
priv->ieee->host_decrypt = 0;
break;
case SEC_LEVEL_2:
priv->sys_config.disable_unicast_decryption = 1;
priv->ieee->host_decrypt = 1;
break;
case SEC_LEVEL_1:
priv->sys_config.disable_unicast_decryption = 0;
priv->ieee->host_decrypt = 0;
break;
case SEC_LEVEL_0:
priv->sys_config.disable_unicast_decryption = 1;
break;
default:
break;
}
}
static void ipw_set_hw_decrypt_multicast(struct ipw_priv *priv, int level)
{
if (priv->ieee->host_encrypt)
return;
switch (level) {
case SEC_LEVEL_3:
priv->sys_config.disable_multicast_decryption = 0;
break;
case SEC_LEVEL_2:
priv->sys_config.disable_multicast_decryption = 1;
break;
case SEC_LEVEL_1:
priv->sys_config.disable_multicast_decryption = 0;
break;
case SEC_LEVEL_0:
priv->sys_config.disable_multicast_decryption = 1;
break;
default:
break;
}
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_set_hwcrypto_keys(struct ipw_priv *priv)
{
switch (priv->ieee->sec.level) {
case SEC_LEVEL_3:
if (priv->ieee->sec.flags & SEC_ACTIVE_KEY)
ipw_send_tgi_tx_key(priv,
DCT_FLAG_EXT_SECURITY_CCM,
priv->ieee->sec.active_key);
if (!priv->ieee->host_mc_decrypt)
ipw_send_wep_keys(priv, DCW_WEP_KEY_SEC_TYPE_CCM);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
case SEC_LEVEL_2:
if (priv->ieee->sec.flags & SEC_ACTIVE_KEY)
ipw_send_tgi_tx_key(priv,
DCT_FLAG_EXT_SECURITY_TKIP,
priv->ieee->sec.active_key);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
case SEC_LEVEL_1:
ipw_send_wep_keys(priv, DCW_WEP_KEY_SEC_TYPE_WEP);
ipw_set_hw_decrypt_unicast(priv, priv->ieee->sec.level);
ipw_set_hw_decrypt_multicast(priv, priv->ieee->sec.level);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
case SEC_LEVEL_0:
default:
break;
}
}
2005-03-25 13:31:53 -05:00
static void ipw_adhoc_check(void *data)
{
struct ipw_priv *priv = data;
if (priv->missed_adhoc_beacons++ > priv->disassociate_threshold &&
2005-03-25 13:31:53 -05:00
!(priv->config & CFG_ADHOC_PERSIST)) {
IPW_DEBUG(IPW_DL_INFO | IPW_DL_NOTIF |
IPW_DL_STATE | IPW_DL_ASSOC,
"Missed beacon: %d - disassociate\n",
priv->missed_adhoc_beacons);
2005-03-25 13:31:53 -05:00
ipw_remove_current_network(priv);
ipw_disassociate(priv);
return;
}
queue_delayed_work(priv->workqueue, &priv->adhoc_check,
2005-03-25 13:31:53 -05:00
priv->assoc_request.beacon_interval);
}
static void ipw_bg_adhoc_check(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_adhoc_check(data);
mutex_unlock(&priv->mutex);
}
#ifdef CONFIG_IPW2200_DEBUG
2005-03-25 13:31:53 -05:00
static void ipw_debug_config(struct ipw_priv *priv)
{
IPW_DEBUG_INFO("Scan completed, no valid APs matched "
"[CFG 0x%08X]\n", priv->config);
if (priv->config & CFG_STATIC_CHANNEL)
IPW_DEBUG_INFO("Channel locked to %d\n", priv->channel);
2005-03-25 13:31:53 -05:00
else
IPW_DEBUG_INFO("Channel unlocked.\n");
if (priv->config & CFG_STATIC_ESSID)
IPW_DEBUG_INFO("ESSID locked to '%s'\n",
escape_essid(priv->essid, priv->essid_len));
2005-03-25 13:31:53 -05:00
else
IPW_DEBUG_INFO("ESSID unlocked.\n");
if (priv->config & CFG_STATIC_BSSID)
IPW_DEBUG_INFO("BSSID locked to " MAC_FMT "\n",
MAC_ARG(priv->bssid));
2005-03-25 13:31:53 -05:00
else
IPW_DEBUG_INFO("BSSID unlocked.\n");
if (priv->capability & CAP_PRIVACY_ON)
IPW_DEBUG_INFO("PRIVACY on\n");
else
IPW_DEBUG_INFO("PRIVACY off\n");
IPW_DEBUG_INFO("RATE MASK: 0x%08X\n", priv->rates_mask);
}
#else
#define ipw_debug_config(x) do {} while (0)
2005-03-25 13:31:53 -05:00
#endif
static void ipw_set_fixed_rate(struct ipw_priv *priv, int mode)
2005-03-25 13:31:53 -05:00
{
/* TODO: Verify that this works... */
struct ipw_fixed_rate fr = {
.tx_rates = priv->rates_mask
};
u32 reg;
u16 mask = 0;
/* Identify 'current FW band' and match it with the fixed
2005-03-25 13:31:53 -05:00
* Tx rates */
2005-03-25 13:31:53 -05:00
switch (priv->ieee->freq_band) {
case IEEE80211_52GHZ_BAND: /* A only */
2005-03-25 13:31:53 -05:00
/* IEEE_A */
if (priv->rates_mask & ~IEEE80211_OFDM_RATES_MASK) {
/* Invalid fixed rate mask */
IPW_DEBUG_WX
("invalid fixed rate mask in ipw_set_fixed_rate\n");
2005-03-25 13:31:53 -05:00
fr.tx_rates = 0;
break;
}
2005-03-25 13:31:53 -05:00
fr.tx_rates >>= IEEE80211_OFDM_SHIFT_MASK_A;
break;
default: /* 2.4Ghz or Mixed */
2005-03-25 13:31:53 -05:00
/* IEEE_B */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (mode == IEEE_B) {
2005-03-25 13:31:53 -05:00
if (fr.tx_rates & ~IEEE80211_CCK_RATES_MASK) {
/* Invalid fixed rate mask */
IPW_DEBUG_WX
("invalid fixed rate mask in ipw_set_fixed_rate\n");
2005-03-25 13:31:53 -05:00
fr.tx_rates = 0;
}
break;
}
2005-03-25 13:31:53 -05:00
/* IEEE_G */
if (fr.tx_rates & ~(IEEE80211_CCK_RATES_MASK |
IEEE80211_OFDM_RATES_MASK)) {
/* Invalid fixed rate mask */
IPW_DEBUG_WX
("invalid fixed rate mask in ipw_set_fixed_rate\n");
2005-03-25 13:31:53 -05:00
fr.tx_rates = 0;
break;
}
2005-03-25 13:31:53 -05:00
if (IEEE80211_OFDM_RATE_6MB_MASK & fr.tx_rates) {
mask |= (IEEE80211_OFDM_RATE_6MB_MASK >> 1);
fr.tx_rates &= ~IEEE80211_OFDM_RATE_6MB_MASK;
}
2005-03-25 13:31:53 -05:00
if (IEEE80211_OFDM_RATE_9MB_MASK & fr.tx_rates) {
mask |= (IEEE80211_OFDM_RATE_9MB_MASK >> 1);
fr.tx_rates &= ~IEEE80211_OFDM_RATE_9MB_MASK;
}
2005-03-25 13:31:53 -05:00
if (IEEE80211_OFDM_RATE_12MB_MASK & fr.tx_rates) {
mask |= (IEEE80211_OFDM_RATE_12MB_MASK >> 1);
fr.tx_rates &= ~IEEE80211_OFDM_RATE_12MB_MASK;
}
2005-03-25 13:31:53 -05:00
fr.tx_rates |= mask;
break;
}
reg = ipw_read32(priv, IPW_MEM_FIXED_OVERRIDE);
ipw_write_reg32(priv, reg, *(u32 *) & fr);
2005-03-25 13:31:53 -05:00
}
static void ipw_abort_scan(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
int err;
if (priv->status & STATUS_SCAN_ABORTING) {
IPW_DEBUG_HC("Ignoring concurrent scan abort request.\n");
return;
}
priv->status |= STATUS_SCAN_ABORTING;
2005-03-25 13:31:53 -05:00
err = ipw_send_scan_abort(priv);
if (err)
IPW_DEBUG_HC("Request to abort scan failed.\n");
}
static void ipw_add_scan_channels(struct ipw_priv *priv,
struct ipw_scan_request_ext *scan,
int scan_type)
{
int channel_index = 0;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
const struct ieee80211_geo *geo;
int i;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
geo = ieee80211_get_geo(priv->ieee);
2005-03-25 13:31:53 -05:00
if (priv->ieee->freq_band & IEEE80211_52GHZ_BAND) {
int start = channel_index;
for (i = 0; i < geo->a_channels; i++) {
if ((priv->status & STATUS_ASSOCIATED) &&
geo->a[i].channel == priv->channel)
continue;
channel_index++;
scan->channels_list[channel_index] = geo->a[i].channel;
ipw_set_scan_type(scan, channel_index,
geo->a[i].
flags & IEEE80211_CH_PASSIVE_ONLY ?
IPW_SCAN_PASSIVE_FULL_DWELL_SCAN :
scan_type);
}
if (start != channel_index) {
scan->channels_list[start] = (u8) (IPW_A_MODE << 6) |
(channel_index - start);
channel_index++;
}
}
if (priv->ieee->freq_band & IEEE80211_24GHZ_BAND) {
int start = channel_index;
if (priv->config & CFG_SPEED_SCAN) {
int index;
u8 channels[IEEE80211_24GHZ_CHANNELS] = {
/* nop out the list */
[0] = 0
};
u8 channel;
while (channel_index < IPW_SCAN_CHANNELS) {
channel =
priv->speed_scan[priv->speed_scan_pos];
if (channel == 0) {
priv->speed_scan_pos = 0;
channel = priv->speed_scan[0];
}
if ((priv->status & STATUS_ASSOCIATED) &&
channel == priv->channel) {
priv->speed_scan_pos++;
continue;
}
/* If this channel has already been
* added in scan, break from loop
* and this will be the first channel
* in the next scan.
*/
if (channels[channel - 1] != 0)
break;
channels[channel - 1] = 1;
priv->speed_scan_pos++;
channel_index++;
scan->channels_list[channel_index] = channel;
index =
ieee80211_channel_to_index(priv->ieee, channel);
ipw_set_scan_type(scan, channel_index,
geo->bg[index].
flags &
IEEE80211_CH_PASSIVE_ONLY ?
IPW_SCAN_PASSIVE_FULL_DWELL_SCAN
: scan_type);
}
} else {
for (i = 0; i < geo->bg_channels; i++) {
if ((priv->status & STATUS_ASSOCIATED) &&
geo->bg[i].channel == priv->channel)
continue;
channel_index++;
scan->channels_list[channel_index] =
geo->bg[i].channel;
ipw_set_scan_type(scan, channel_index,
geo->bg[i].
flags &
IEEE80211_CH_PASSIVE_ONLY ?
IPW_SCAN_PASSIVE_FULL_DWELL_SCAN
: scan_type);
}
}
if (start != channel_index) {
scan->channels_list[start] = (u8) (IPW_B_MODE << 6) |
(channel_index - start);
}
}
}
static int ipw_request_scan(struct ipw_priv *priv)
{
struct ipw_scan_request_ext scan;
int err = 0, scan_type;
if (!(priv->status & STATUS_INIT) ||
(priv->status & STATUS_EXIT_PENDING))
return 0;
mutex_lock(&priv->mutex);
if (priv->status & STATUS_SCANNING) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
IPW_DEBUG_HC("Concurrent scan requested. Ignoring.\n");
priv->status |= STATUS_SCAN_PENDING;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
goto done;
}
2005-03-25 13:31:53 -05:00
if (!(priv->status & STATUS_SCAN_FORCED) &&
priv->status & STATUS_SCAN_ABORTING) {
IPW_DEBUG_HC("Scan request while abort pending. Queuing.\n");
priv->status |= STATUS_SCAN_PENDING;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
goto done;
2005-03-25 13:31:53 -05:00
}
if (priv->status & STATUS_RF_KILL_MASK) {
IPW_DEBUG_HC("Aborting scan due to RF Kill activation\n");
priv->status |= STATUS_SCAN_PENDING;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
goto done;
}
2005-03-25 13:31:53 -05:00
memset(&scan, 0, sizeof(scan));
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->config & CFG_SPEED_SCAN)
scan.dwell_time[IPW_SCAN_ACTIVE_BROADCAST_SCAN] =
cpu_to_le16(30);
else
scan.dwell_time[IPW_SCAN_ACTIVE_BROADCAST_SCAN] =
cpu_to_le16(20);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
scan.dwell_time[IPW_SCAN_ACTIVE_BROADCAST_AND_DIRECT_SCAN] =
cpu_to_le16(20);
scan.dwell_time[IPW_SCAN_PASSIVE_FULL_DWELL_SCAN] = cpu_to_le16(120);
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
scan.full_scan_index = cpu_to_le32(ieee80211_get_scans(priv->ieee));
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
if (priv->ieee->iw_mode == IW_MODE_MONITOR) {
u8 channel;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
u8 band = 0;
2005-03-25 13:31:53 -05:00
switch (ieee80211_is_valid_channel(priv->ieee, priv->channel)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
case IEEE80211_52GHZ_BAND:
band = (u8) (IPW_A_MODE << 6) | 1;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
channel = priv->channel;
break;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
case IEEE80211_24GHZ_BAND:
band = (u8) (IPW_B_MODE << 6) | 1;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
channel = priv->channel;
break;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
default:
band = (u8) (IPW_B_MODE << 6) | 1;
channel = 9;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
scan.channels_list[0] = band;
scan.channels_list[1] = channel;
ipw_set_scan_type(&scan, 1, IPW_SCAN_PASSIVE_FULL_DWELL_SCAN);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* NOTE: The card will sit on this channel for this time
* period. Scan aborts are timing sensitive and frequently
* result in firmware restarts. As such, it is best to
* set a small dwell_time here and just keep re-issuing
* scans. Otherwise fast channel hopping will not actually
* hop channels.
*
* TODO: Move SPEED SCAN support to all modes and bands */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
scan.dwell_time[IPW_SCAN_PASSIVE_FULL_DWELL_SCAN] =
cpu_to_le16(2000);
2005-03-25 13:31:53 -05:00
} else {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#endif /* CONFIG_IPW2200_MONITOR */
/* If we are roaming, then make this a directed scan for the
* current network. Otherwise, ensure that every other scan
* is a fast channel hop scan */
if ((priv->status & STATUS_ROAMING)
|| (!(priv->status & STATUS_ASSOCIATED)
&& (priv->config & CFG_STATIC_ESSID)
&& (le32_to_cpu(scan.full_scan_index) % 2))) {
err = ipw_send_ssid(priv, priv->essid, priv->essid_len);
if (err) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_DEBUG_HC("Attempt to send SSID command "
"failed.\n");
goto done;
}
2005-03-25 13:31:53 -05:00
scan_type = IPW_SCAN_ACTIVE_BROADCAST_AND_DIRECT_SCAN;
} else
scan_type = IPW_SCAN_ACTIVE_BROADCAST_SCAN;
ipw_add_scan_channels(priv, &scan, scan_type);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
2005-03-25 13:31:53 -05:00
}
#endif
err = ipw_send_scan_request_ext(priv, &scan);
2005-03-25 13:31:53 -05:00
if (err) {
IPW_DEBUG_HC("Sending scan command failed: %08X\n", err);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
goto done;
2005-03-25 13:31:53 -05:00
}
priv->status |= STATUS_SCANNING;
priv->status &= ~STATUS_SCAN_PENDING;
queue_delayed_work(priv->workqueue, &priv->scan_check,
IPW_SCAN_CHECK_WATCHDOG);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
done:
mutex_unlock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return err;
}
static void ipw_bg_abort_scan(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_abort_scan(data);
mutex_unlock(&priv->mutex);
}
static int ipw_wpa_enable(struct ipw_priv *priv, int value)
{
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* This is called when wpa_supplicant loads and closes the driver
* interface. */
priv->ieee->wpa_enabled = value;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return 0;
}
static int ipw_wpa_set_auth_algs(struct ipw_priv *priv, int value)
{
struct ieee80211_device *ieee = priv->ieee;
struct ieee80211_security sec = {
.flags = SEC_AUTH_MODE,
};
int ret = 0;
if (value & IW_AUTH_ALG_SHARED_KEY) {
sec.auth_mode = WLAN_AUTH_SHARED_KEY;
ieee->open_wep = 0;
} else if (value & IW_AUTH_ALG_OPEN_SYSTEM) {
sec.auth_mode = WLAN_AUTH_OPEN;
ieee->open_wep = 1;
} else if (value & IW_AUTH_ALG_LEAP) {
sec.auth_mode = WLAN_AUTH_LEAP;
ieee->open_wep = 1;
} else
return -EINVAL;
if (ieee->set_security)
ieee->set_security(ieee->dev, &sec);
else
ret = -EOPNOTSUPP;
return ret;
}
static void ipw_wpa_assoc_frame(struct ipw_priv *priv, char *wpa_ie,
int wpa_ie_len)
{
/* make sure WPA is enabled */
ipw_wpa_enable(priv, 1);
ipw_disassociate(priv);
}
static int ipw_set_rsn_capa(struct ipw_priv *priv,
char *capabilities, int length)
{
IPW_DEBUG_HC("HOST_CMD_RSN_CAPABILITIES\n");
return ipw_send_cmd_pdu(priv, IPW_CMD_RSN_CAPABILITIES, length,
capabilities);
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/*
* WE-18 support
*/
/* SIOCSIWGENIE */
static int ipw_wx_set_genie(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct ieee80211_device *ieee = priv->ieee;
u8 *buf;
int err = 0;
if (wrqu->data.length > MAX_WPA_IE_LEN ||
(wrqu->data.length && extra == NULL))
return -EINVAL;
//mutex_lock(&priv->mutex);
//if (!ieee->wpa_enabled) {
// err = -EOPNOTSUPP;
// goto out;
//}
if (wrqu->data.length) {
buf = kmalloc(wrqu->data.length, GFP_KERNEL);
if (buf == NULL) {
err = -ENOMEM;
goto out;
}
memcpy(buf, extra, wrqu->data.length);
kfree(ieee->wpa_ie);
ieee->wpa_ie = buf;
ieee->wpa_ie_len = wrqu->data.length;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
} else {
kfree(ieee->wpa_ie);
ieee->wpa_ie = NULL;
ieee->wpa_ie_len = 0;
}
ipw_wpa_assoc_frame(priv, ieee->wpa_ie, ieee->wpa_ie_len);
out:
//mutex_unlock(&priv->mutex);
return err;
}
/* SIOCGIWGENIE */
static int ipw_wx_get_genie(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct ieee80211_device *ieee = priv->ieee;
int err = 0;
//mutex_lock(&priv->mutex);
//if (!ieee->wpa_enabled) {
// err = -EOPNOTSUPP;
// goto out;
//}
if (ieee->wpa_ie_len == 0 || ieee->wpa_ie == NULL) {
wrqu->data.length = 0;
goto out;
}
if (wrqu->data.length < ieee->wpa_ie_len) {
err = -E2BIG;
goto out;
}
wrqu->data.length = ieee->wpa_ie_len;
memcpy(extra, ieee->wpa_ie, ieee->wpa_ie_len);
out:
//mutex_unlock(&priv->mutex);
return err;
}
static int wext_cipher2level(int cipher)
{
switch (cipher) {
case IW_AUTH_CIPHER_NONE:
return SEC_LEVEL_0;
case IW_AUTH_CIPHER_WEP40:
case IW_AUTH_CIPHER_WEP104:
return SEC_LEVEL_1;
case IW_AUTH_CIPHER_TKIP:
return SEC_LEVEL_2;
case IW_AUTH_CIPHER_CCMP:
return SEC_LEVEL_3;
default:
return -1;
}
}
/* SIOCSIWAUTH */
static int ipw_wx_set_auth(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct ieee80211_device *ieee = priv->ieee;
struct iw_param *param = &wrqu->param;
struct ieee80211_crypt_data *crypt;
unsigned long flags;
int ret = 0;
switch (param->flags & IW_AUTH_INDEX) {
case IW_AUTH_WPA_VERSION:
break;
case IW_AUTH_CIPHER_PAIRWISE:
ipw_set_hw_decrypt_unicast(priv,
wext_cipher2level(param->value));
break;
case IW_AUTH_CIPHER_GROUP:
ipw_set_hw_decrypt_multicast(priv,
wext_cipher2level(param->value));
break;
case IW_AUTH_KEY_MGMT:
/*
* ipw2200 does not use these parameters
*/
break;
case IW_AUTH_TKIP_COUNTERMEASURES:
crypt = priv->ieee->crypt[priv->ieee->tx_keyidx];
if (!crypt || !crypt->ops->set_flags || !crypt->ops->get_flags)
break;
flags = crypt->ops->get_flags(crypt->priv);
if (param->value)
flags |= IEEE80211_CRYPTO_TKIP_COUNTERMEASURES;
else
flags &= ~IEEE80211_CRYPTO_TKIP_COUNTERMEASURES;
crypt->ops->set_flags(flags, crypt->priv);
break;
case IW_AUTH_DROP_UNENCRYPTED:{
/* HACK:
*
* wpa_supplicant calls set_wpa_enabled when the driver
* is loaded and unloaded, regardless of if WPA is being
* used. No other calls are made which can be used to
* determine if encryption will be used or not prior to
* association being expected. If encryption is not being
* used, drop_unencrypted is set to false, else true -- we
* can use this to determine if the CAP_PRIVACY_ON bit should
* be set.
*/
struct ieee80211_security sec = {
.flags = SEC_ENABLED,
.enabled = param->value,
};
priv->ieee->drop_unencrypted = param->value;
/* We only change SEC_LEVEL for open mode. Others
* are set by ipw_wpa_set_encryption.
*/
if (!param->value) {
sec.flags |= SEC_LEVEL;
sec.level = SEC_LEVEL_0;
} else {
sec.flags |= SEC_LEVEL;
sec.level = SEC_LEVEL_1;
}
if (priv->ieee->set_security)
priv->ieee->set_security(priv->ieee->dev, &sec);
break;
}
case IW_AUTH_80211_AUTH_ALG:
ret = ipw_wpa_set_auth_algs(priv, param->value);
break;
case IW_AUTH_WPA_ENABLED:
ret = ipw_wpa_enable(priv, param->value);
break;
case IW_AUTH_RX_UNENCRYPTED_EAPOL:
ieee->ieee802_1x = param->value;
break;
//case IW_AUTH_ROAMING_CONTROL:
case IW_AUTH_PRIVACY_INVOKED:
ieee->privacy_invoked = param->value;
break;
default:
return -EOPNOTSUPP;
}
return ret;
}
/* SIOCGIWAUTH */
static int ipw_wx_get_auth(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct ieee80211_device *ieee = priv->ieee;
struct ieee80211_crypt_data *crypt;
struct iw_param *param = &wrqu->param;
int ret = 0;
switch (param->flags & IW_AUTH_INDEX) {
case IW_AUTH_WPA_VERSION:
case IW_AUTH_CIPHER_PAIRWISE:
case IW_AUTH_CIPHER_GROUP:
case IW_AUTH_KEY_MGMT:
/*
* wpa_supplicant will control these internally
*/
ret = -EOPNOTSUPP;
break;
case IW_AUTH_TKIP_COUNTERMEASURES:
crypt = priv->ieee->crypt[priv->ieee->tx_keyidx];
if (!crypt || !crypt->ops->get_flags)
break;
param->value = (crypt->ops->get_flags(crypt->priv) &
IEEE80211_CRYPTO_TKIP_COUNTERMEASURES) ? 1 : 0;
break;
case IW_AUTH_DROP_UNENCRYPTED:
param->value = ieee->drop_unencrypted;
break;
case IW_AUTH_80211_AUTH_ALG:
param->value = ieee->sec.auth_mode;
break;
case IW_AUTH_WPA_ENABLED:
param->value = ieee->wpa_enabled;
break;
case IW_AUTH_RX_UNENCRYPTED_EAPOL:
param->value = ieee->ieee802_1x;
break;
case IW_AUTH_ROAMING_CONTROL:
case IW_AUTH_PRIVACY_INVOKED:
param->value = ieee->privacy_invoked;
break;
default:
return -EOPNOTSUPP;
}
return 0;
}
/* SIOCSIWENCODEEXT */
static int ipw_wx_set_encodeext(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct iw_encode_ext *ext = (struct iw_encode_ext *)extra;
if (hwcrypto) {
if (ext->alg == IW_ENCODE_ALG_TKIP) {
/* IPW HW can't build TKIP MIC,
host decryption still needed */
if (ext->ext_flags & IW_ENCODE_EXT_GROUP_KEY)
priv->ieee->host_mc_decrypt = 1;
else {
priv->ieee->host_encrypt = 0;
priv->ieee->host_encrypt_msdu = 1;
priv->ieee->host_decrypt = 1;
}
} else {
priv->ieee->host_encrypt = 0;
priv->ieee->host_encrypt_msdu = 0;
priv->ieee->host_decrypt = 0;
priv->ieee->host_mc_decrypt = 0;
}
}
return ieee80211_wx_set_encodeext(priv->ieee, info, wrqu, extra);
}
/* SIOCGIWENCODEEXT */
static int ipw_wx_get_encodeext(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
return ieee80211_wx_get_encodeext(priv->ieee, info, wrqu, extra);
}
/* SIOCSIWMLME */
static int ipw_wx_set_mlme(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct iw_mlme *mlme = (struct iw_mlme *)extra;
u16 reason;
reason = cpu_to_le16(mlme->reason_code);
switch (mlme->cmd) {
case IW_MLME_DEAUTH:
// silently ignore
break;
case IW_MLME_DISASSOC:
ipw_disassociate(priv);
break;
default:
return -EOPNOTSUPP;
}
return 0;
}
#ifdef CONFIG_IPW2200_QOS
/* QoS */
/*
* get the modulation type of the current network or
* the card current mode
*/
static u8 ipw_qos_current_mode(struct ipw_priv * priv)
{
u8 mode = 0;
if (priv->status & STATUS_ASSOCIATED) {
unsigned long flags;
spin_lock_irqsave(&priv->ieee->lock, flags);
mode = priv->assoc_network->mode;
spin_unlock_irqrestore(&priv->ieee->lock, flags);
} else {
mode = priv->ieee->mode;
}
IPW_DEBUG_QOS("QoS network/card mode %d \n", mode);
return mode;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/*
* Handle management frame beacon and probe response
*/
static int ipw_qos_handle_probe_response(struct ipw_priv *priv,
int active_network,
struct ieee80211_network *network)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
{
u32 size = sizeof(struct ieee80211_qos_parameters);
if (network->capability & WLAN_CAPABILITY_IBSS)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
network->qos_data.active = network->qos_data.supported;
if (network->flags & NETWORK_HAS_QOS_MASK) {
if (active_network &&
(network->flags & NETWORK_HAS_QOS_PARAMETERS))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
network->qos_data.active = network->qos_data.supported;
if ((network->qos_data.active == 1) && (active_network == 1) &&
(network->flags & NETWORK_HAS_QOS_PARAMETERS) &&
(network->qos_data.old_param_count !=
network->qos_data.param_count)) {
network->qos_data.old_param_count =
network->qos_data.param_count;
schedule_work(&priv->qos_activate);
IPW_DEBUG_QOS("QoS parameters change call "
"qos_activate\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
} else {
if ((priv->ieee->mode == IEEE_B) || (network->mode == IEEE_B))
memcpy(&network->qos_data.parameters,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
&def_parameters_CCK, size);
else
memcpy(&network->qos_data.parameters,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
&def_parameters_OFDM, size);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if ((network->qos_data.active == 1) && (active_network == 1)) {
IPW_DEBUG_QOS("QoS was disabled call qos_activate \n");
schedule_work(&priv->qos_activate);
}
network->qos_data.active = 0;
network->qos_data.supported = 0;
}
if ((priv->status & STATUS_ASSOCIATED) &&
(priv->ieee->iw_mode == IW_MODE_ADHOC) && (active_network == 0)) {
if (memcmp(network->bssid, priv->bssid, ETH_ALEN))
if ((network->capability & WLAN_CAPABILITY_IBSS) &&
!(network->flags & NETWORK_EMPTY_ESSID))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if ((network->ssid_len ==
priv->assoc_network->ssid_len) &&
!memcmp(network->ssid,
priv->assoc_network->ssid,
network->ssid_len)) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
queue_work(priv->workqueue,
&priv->merge_networks);
}
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return 0;
}
/*
* This function set up the firmware to support QoS. It sends
* IPW_CMD_QOS_PARAMETERS and IPW_CMD_WME_INFO
*/
static int ipw_qos_activate(struct ipw_priv *priv,
struct ieee80211_qos_data *qos_network_data)
{
int err;
struct ieee80211_qos_parameters qos_parameters[QOS_QOS_SETS];
struct ieee80211_qos_parameters *active_one = NULL;
u32 size = sizeof(struct ieee80211_qos_parameters);
u32 burst_duration;
int i;
u8 type;
type = ipw_qos_current_mode(priv);
active_one = &(qos_parameters[QOS_PARAM_SET_DEF_CCK]);
memcpy(active_one, priv->qos_data.def_qos_parm_CCK, size);
active_one = &(qos_parameters[QOS_PARAM_SET_DEF_OFDM]);
memcpy(active_one, priv->qos_data.def_qos_parm_OFDM, size);
if (qos_network_data == NULL) {
if (type == IEEE_B) {
IPW_DEBUG_QOS("QoS activate network mode %d\n", type);
active_one = &def_parameters_CCK;
} else
active_one = &def_parameters_OFDM;
memcpy(&qos_parameters[QOS_PARAM_SET_ACTIVE], active_one, size);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
burst_duration = ipw_qos_get_burst_duration(priv);
for (i = 0; i < QOS_QUEUE_NUM; i++)
qos_parameters[QOS_PARAM_SET_ACTIVE].tx_op_limit[i] =
(u16) burst_duration;
} else if (priv->ieee->iw_mode == IW_MODE_ADHOC) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (type == IEEE_B) {
IPW_DEBUG_QOS("QoS activate IBSS nework mode %d\n",
type);
if (priv->qos_data.qos_enable == 0)
active_one = &def_parameters_CCK;
else
active_one = priv->qos_data.def_qos_parm_CCK;
} else {
if (priv->qos_data.qos_enable == 0)
active_one = &def_parameters_OFDM;
else
active_one = priv->qos_data.def_qos_parm_OFDM;
}
memcpy(&qos_parameters[QOS_PARAM_SET_ACTIVE], active_one, size);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
} else {
unsigned long flags;
int active;
spin_lock_irqsave(&priv->ieee->lock, flags);
active_one = &(qos_network_data->parameters);
qos_network_data->old_param_count =
qos_network_data->param_count;
memcpy(&qos_parameters[QOS_PARAM_SET_ACTIVE], active_one, size);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
active = qos_network_data->supported;
spin_unlock_irqrestore(&priv->ieee->lock, flags);
if (active == 0) {
burst_duration = ipw_qos_get_burst_duration(priv);
for (i = 0; i < QOS_QUEUE_NUM; i++)
qos_parameters[QOS_PARAM_SET_ACTIVE].
tx_op_limit[i] = (u16) burst_duration;
}
}
IPW_DEBUG_QOS("QoS sending IPW_CMD_QOS_PARAMETERS\n");
err = ipw_send_qos_params_command(priv,
(struct ieee80211_qos_parameters *)
&(qos_parameters[0]));
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (err)
IPW_DEBUG_QOS("QoS IPW_CMD_QOS_PARAMETERS failed\n");
return err;
}
/*
* send IPW_CMD_WME_INFO to the firmware
*/
static int ipw_qos_set_info_element(struct ipw_priv *priv)
{
int ret = 0;
struct ieee80211_qos_information_element qos_info;
if (priv == NULL)
return -1;
qos_info.elementID = QOS_ELEMENT_ID;
qos_info.length = sizeof(struct ieee80211_qos_information_element) - 2;
qos_info.version = QOS_VERSION_1;
qos_info.ac_info = 0;
memcpy(qos_info.qui, qos_oui, QOS_OUI_LEN);
qos_info.qui_type = QOS_OUI_TYPE;
qos_info.qui_subtype = QOS_OUI_INFO_SUB_TYPE;
ret = ipw_send_qos_info_command(priv, &qos_info);
if (ret != 0) {
IPW_DEBUG_QOS("QoS error calling ipw_send_qos_info_command\n");
}
return ret;
}
/*
* Set the QoS parameter with the association request structure
*/
static int ipw_qos_association(struct ipw_priv *priv,
struct ieee80211_network *network)
{
int err = 0;
struct ieee80211_qos_data *qos_data = NULL;
struct ieee80211_qos_data ibss_data = {
.supported = 1,
.active = 1,
};
switch (priv->ieee->iw_mode) {
case IW_MODE_ADHOC:
BUG_ON(!(network->capability & WLAN_CAPABILITY_IBSS));
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
qos_data = &ibss_data;
break;
case IW_MODE_INFRA:
qos_data = &network->qos_data;
break;
default:
BUG();
break;
}
err = ipw_qos_activate(priv, qos_data);
if (err) {
priv->assoc_request.policy_support &= ~HC_QOS_SUPPORT_ASSOC;
return err;
}
if (priv->qos_data.qos_enable && qos_data->supported) {
IPW_DEBUG_QOS("QoS will be enabled for this association\n");
priv->assoc_request.policy_support |= HC_QOS_SUPPORT_ASSOC;
return ipw_qos_set_info_element(priv);
}
return 0;
}
/*
* handling the beaconing responces. if we get different QoS setting
* of the network from the the associated setting adjust the QoS
* setting
*/
static int ipw_qos_association_resp(struct ipw_priv *priv,
struct ieee80211_network *network)
{
int ret = 0;
unsigned long flags;
u32 size = sizeof(struct ieee80211_qos_parameters);
int set_qos_param = 0;
if ((priv == NULL) || (network == NULL) ||
(priv->assoc_network == NULL))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return ret;
if (!(priv->status & STATUS_ASSOCIATED))
return ret;
if ((priv->ieee->iw_mode != IW_MODE_INFRA))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return ret;
spin_lock_irqsave(&priv->ieee->lock, flags);
if (network->flags & NETWORK_HAS_QOS_PARAMETERS) {
memcpy(&priv->assoc_network->qos_data, &network->qos_data,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
sizeof(struct ieee80211_qos_data));
priv->assoc_network->qos_data.active = 1;
if ((network->qos_data.old_param_count !=
network->qos_data.param_count)) {
set_qos_param = 1;
network->qos_data.old_param_count =
network->qos_data.param_count;
}
} else {
if ((network->mode == IEEE_B) || (priv->ieee->mode == IEEE_B))
memcpy(&priv->assoc_network->qos_data.parameters,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
&def_parameters_CCK, size);
else
memcpy(&priv->assoc_network->qos_data.parameters,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
&def_parameters_OFDM, size);
priv->assoc_network->qos_data.active = 0;
priv->assoc_network->qos_data.supported = 0;
set_qos_param = 1;
}
spin_unlock_irqrestore(&priv->ieee->lock, flags);
if (set_qos_param == 1)
schedule_work(&priv->qos_activate);
return ret;
}
static u32 ipw_qos_get_burst_duration(struct ipw_priv *priv)
{
u32 ret = 0;
if ((priv == NULL))
return 0;
if (!(priv->ieee->modulation & IEEE80211_OFDM_MODULATION))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ret = priv->qos_data.burst_duration_CCK;
else
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ret = priv->qos_data.burst_duration_OFDM;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return ret;
}
/*
* Initialize the setting of QoS global
*/
static void ipw_qos_init(struct ipw_priv *priv, int enable,
int burst_enable, u32 burst_duration_CCK,
u32 burst_duration_OFDM)
{
priv->qos_data.qos_enable = enable;
if (priv->qos_data.qos_enable) {
priv->qos_data.def_qos_parm_CCK = &def_qos_parameters_CCK;
priv->qos_data.def_qos_parm_OFDM = &def_qos_parameters_OFDM;
IPW_DEBUG_QOS("QoS is enabled\n");
} else {
priv->qos_data.def_qos_parm_CCK = &def_parameters_CCK;
priv->qos_data.def_qos_parm_OFDM = &def_parameters_OFDM;
IPW_DEBUG_QOS("QoS is not enabled\n");
}
priv->qos_data.burst_enable = burst_enable;
if (burst_enable) {
priv->qos_data.burst_duration_CCK = burst_duration_CCK;
priv->qos_data.burst_duration_OFDM = burst_duration_OFDM;
} else {
priv->qos_data.burst_duration_CCK = 0;
priv->qos_data.burst_duration_OFDM = 0;
}
}
/*
* map the packet priority to the right TX Queue
*/
static int ipw_get_tx_queue_number(struct ipw_priv *priv, u16 priority)
{
if (priority > 7 || !priv->qos_data.qos_enable)
priority = 0;
return from_priority_to_tx_queue[priority] - 1;
}
static int ipw_is_qos_active(struct net_device *dev,
struct sk_buff *skb)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
struct ieee80211_qos_data *qos_data = NULL;
int active, supported;
u8 *daddr = skb->data + ETH_ALEN;
int unicast = !is_multicast_ether_addr(daddr);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (!(priv->status & STATUS_ASSOCIATED))
return 0;
qos_data = &priv->assoc_network->qos_data;
if (priv->ieee->iw_mode == IW_MODE_ADHOC) {
if (unicast == 0)
qos_data->active = 0;
else
qos_data->active = qos_data->supported;
}
active = qos_data->active;
supported = qos_data->supported;
IPW_DEBUG_QOS("QoS %d network is QoS active %d supported %d "
"unicast %d\n",
priv->qos_data.qos_enable, active, supported, unicast);
if (active && priv->qos_data.qos_enable)
return 1;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return 0;
}
/*
* add QoS parameter to the TX command
*/
static int ipw_qos_set_tx_queue_command(struct ipw_priv *priv,
u16 priority,
struct tfd_data *tfd)
{
int tx_queue_id = 0;
tx_queue_id = from_priority_to_tx_queue[priority] - 1;
tfd->tx_flags_ext |= DCT_FLAG_EXT_QOS_ENABLED;
if (priv->qos_data.qos_no_ack_mask & (1UL << tx_queue_id)) {
tfd->tx_flags &= ~DCT_FLAG_ACK_REQD;
tfd->tfd.tfd_26.mchdr.qos_ctrl |= CTRL_QOS_NO_ACK;
}
return 0;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
/*
* background support to run QoS activate functionality
*/
static void ipw_bg_qos_activate(void *data)
{
struct ipw_priv *priv = data;
if (priv == NULL)
return;
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->status & STATUS_ASSOCIATED)
ipw_qos_activate(priv, &(priv->assoc_network->qos_data));
mutex_unlock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
static int ipw_handle_probe_response(struct net_device *dev,
struct ieee80211_probe_response *resp,
struct ieee80211_network *network)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
int active_network = ((priv->status & STATUS_ASSOCIATED) &&
(network == priv->assoc_network));
2005-03-25 13:31:53 -05:00
ipw_qos_handle_probe_response(priv, active_network, network);
2005-03-25 13:31:53 -05:00
return 0;
}
2005-03-25 13:31:53 -05:00
static int ipw_handle_beacon(struct net_device *dev,
struct ieee80211_beacon *resp,
struct ieee80211_network *network)
{
struct ipw_priv *priv = ieee80211_priv(dev);
int active_network = ((priv->status & STATUS_ASSOCIATED) &&
(network == priv->assoc_network));
ipw_qos_handle_probe_response(priv, active_network, network);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return 0;
}
static int ipw_handle_assoc_response(struct net_device *dev,
struct ieee80211_assoc_response *resp,
struct ieee80211_network *network)
{
struct ipw_priv *priv = ieee80211_priv(dev);
ipw_qos_association_resp(priv, network);
return 0;
}
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static int ipw_send_qos_params_command(struct ipw_priv *priv, struct ieee80211_qos_parameters
*qos_param)
{
return ipw_send_cmd_pdu(priv, IPW_CMD_QOS_PARAMETERS,
sizeof(*qos_param) * 3, qos_param);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
static int ipw_send_qos_info_command(struct ipw_priv *priv, struct ieee80211_qos_information_element
*qos_param)
{
return ipw_send_cmd_pdu(priv, IPW_CMD_WME_INFO, sizeof(*qos_param),
qos_param);
2005-03-25 13:31:53 -05:00
}
#endif /* CONFIG_IPW2200_QOS */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
2005-03-25 13:31:53 -05:00
static int ipw_associate_network(struct ipw_priv *priv,
struct ieee80211_network *network,
struct ipw_supported_rates *rates, int roaming)
2005-03-25 13:31:53 -05:00
{
int err;
if (priv->config & CFG_FIXED_RATE)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_fixed_rate(priv, network->mode);
2005-03-25 13:31:53 -05:00
if (!(priv->config & CFG_STATIC_ESSID)) {
priv->essid_len = min(network->ssid_len,
(u8) IW_ESSID_MAX_SIZE);
2005-03-25 13:31:53 -05:00
memcpy(priv->essid, network->ssid, priv->essid_len);
}
network->last_associate = jiffies;
memset(&priv->assoc_request, 0, sizeof(priv->assoc_request));
priv->assoc_request.channel = network->channel;
priv->assoc_request.auth_key = 0;
2005-03-25 13:31:53 -05:00
if ((priv->capability & CAP_PRIVACY_ON) &&
(priv->ieee->sec.auth_mode == WLAN_AUTH_SHARED_KEY)) {
2005-03-25 13:31:53 -05:00
priv->assoc_request.auth_type = AUTH_SHARED_KEY;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->assoc_request.auth_key = priv->ieee->sec.active_key;
if (priv->ieee->sec.level == SEC_LEVEL_1)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_send_wep_keys(priv, DCW_WEP_KEY_SEC_TYPE_WEP);
} else if ((priv->capability & CAP_PRIVACY_ON) &&
(priv->ieee->sec.auth_mode == WLAN_AUTH_LEAP))
priv->assoc_request.auth_type = AUTH_LEAP;
else
2005-03-25 13:31:53 -05:00
priv->assoc_request.auth_type = AUTH_OPEN;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->ieee->wpa_ie_len) {
priv->assoc_request.policy_support = 0x02; /* RSN active */
ipw_set_rsn_capa(priv, priv->ieee->wpa_ie,
priv->ieee->wpa_ie_len);
}
2005-03-25 13:31:53 -05:00
/*
* It is valid for our ieee device to support multiple modes, but
* when it comes to associating to a given network we have to choose
2005-03-25 13:31:53 -05:00
* just one mode.
*/
if (network->mode & priv->ieee->mode & IEEE_A)
priv->assoc_request.ieee_mode = IPW_A_MODE;
else if (network->mode & priv->ieee->mode & IEEE_G)
priv->assoc_request.ieee_mode = IPW_G_MODE;
else if (network->mode & priv->ieee->mode & IEEE_B)
priv->assoc_request.ieee_mode = IPW_B_MODE;
priv->assoc_request.capability = network->capability;
if ((network->capability & WLAN_CAPABILITY_SHORT_PREAMBLE)
&& !(priv->config & CFG_PREAMBLE_LONG)) {
priv->assoc_request.preamble_length = DCT_FLAG_SHORT_PREAMBLE;
} else {
priv->assoc_request.preamble_length = DCT_FLAG_LONG_PREAMBLE;
/* Clear the short preamble if we won't be supporting it */
priv->assoc_request.capability &=
~WLAN_CAPABILITY_SHORT_PREAMBLE;
}
/* Clear capability bits that aren't used in Ad Hoc */
if (priv->ieee->iw_mode == IW_MODE_ADHOC)
priv->assoc_request.capability &=
~WLAN_CAPABILITY_SHORT_SLOT_TIME;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ASSOC("%sssocation attempt: '%s', channel %d, "
"802.11%c [%d], %s[:%s], enc=%s%s%s%c%c\n",
2005-03-25 13:31:53 -05:00
roaming ? "Rea" : "A",
escape_essid(priv->essid, priv->essid_len),
network->channel,
ipw_modes[priv->assoc_request.ieee_mode],
rates->num_rates,
(priv->assoc_request.preamble_length ==
DCT_FLAG_LONG_PREAMBLE) ? "long" : "short",
network->capability &
WLAN_CAPABILITY_SHORT_PREAMBLE ? "short" : "long",
2005-03-25 13:31:53 -05:00
priv->capability & CAP_PRIVACY_ON ? "on " : "off",
priv->capability & CAP_PRIVACY_ON ?
(priv->capability & CAP_SHARED_KEY ? "(shared)" :
2005-03-25 13:31:53 -05:00
"(open)") : "",
priv->capability & CAP_PRIVACY_ON ? " key=" : "",
priv->capability & CAP_PRIVACY_ON ?
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
'1' + priv->ieee->sec.active_key : '.',
priv->capability & CAP_PRIVACY_ON ? '.' : ' ');
2005-03-25 13:31:53 -05:00
priv->assoc_request.beacon_interval = network->beacon_interval;
if ((priv->ieee->iw_mode == IW_MODE_ADHOC) &&
(network->time_stamp[0] == 0) && (network->time_stamp[1] == 0)) {
2005-03-25 13:31:53 -05:00
priv->assoc_request.assoc_type = HC_IBSS_START;
priv->assoc_request.assoc_tsf_msw = 0;
priv->assoc_request.assoc_tsf_lsw = 0;
} else {
if (unlikely(roaming))
priv->assoc_request.assoc_type = HC_REASSOCIATE;
else
priv->assoc_request.assoc_type = HC_ASSOCIATE;
priv->assoc_request.assoc_tsf_msw = network->time_stamp[1];
priv->assoc_request.assoc_tsf_lsw = network->time_stamp[0];
}
memcpy(priv->assoc_request.bssid, network->bssid, ETH_ALEN);
2005-03-25 13:31:53 -05:00
if (priv->ieee->iw_mode == IW_MODE_ADHOC) {
memset(&priv->assoc_request.dest, 0xFF, ETH_ALEN);
priv->assoc_request.atim_window = network->atim_window;
} else {
memcpy(priv->assoc_request.dest, network->bssid, ETH_ALEN);
2005-03-25 13:31:53 -05:00
priv->assoc_request.atim_window = 0;
}
priv->assoc_request.listen_interval = network->listen_interval;
2005-03-25 13:31:53 -05:00
err = ipw_send_ssid(priv, priv->essid, priv->essid_len);
if (err) {
IPW_DEBUG_HC("Attempt to send SSID command failed.\n");
return err;
}
rates->ieee_mode = priv->assoc_request.ieee_mode;
rates->purpose = IPW_RATE_CONNECT;
ipw_send_supported_rates(priv, rates);
2005-03-25 13:31:53 -05:00
if (priv->assoc_request.ieee_mode == IPW_G_MODE)
priv->sys_config.dot11g_auto_detection = 1;
else
priv->sys_config.dot11g_auto_detection = 0;
if (priv->ieee->iw_mode == IW_MODE_ADHOC)
priv->sys_config.answer_broadcast_ssid_probe = 1;
else
priv->sys_config.answer_broadcast_ssid_probe = 0;
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
err = ipw_send_system_config(priv);
2005-03-25 13:31:53 -05:00
if (err) {
IPW_DEBUG_HC("Attempt to send sys config command failed.\n");
return err;
}
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ASSOC("Association sensitivity: %d\n", network->stats.rssi);
err = ipw_set_sensitivity(priv, network->stats.rssi + IPW_RSSI_TO_DBM);
2005-03-25 13:31:53 -05:00
if (err) {
IPW_DEBUG_HC("Attempt to send associate command failed.\n");
return err;
}
/*
* If preemption is enabled, it is possible for the association
* to complete before we return from ipw_send_associate. Therefore
* we have to be sure and update our priviate data first.
*/
priv->channel = network->channel;
memcpy(priv->bssid, network->bssid, ETH_ALEN);
priv->status |= STATUS_ASSOCIATING;
2005-03-25 13:31:53 -05:00
priv->status &= ~STATUS_SECURITY_UPDATED;
priv->assoc_network = network;
#ifdef CONFIG_IPW2200_QOS
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_qos_association(priv, network);
#endif
2005-03-25 13:31:53 -05:00
err = ipw_send_associate(priv, &priv->assoc_request);
if (err) {
IPW_DEBUG_HC("Attempt to send associate command failed.\n");
return err;
}
IPW_DEBUG(IPW_DL_STATE, "associating: '%s' " MAC_FMT " \n",
2005-03-25 13:31:53 -05:00
escape_essid(priv->essid, priv->essid_len),
MAC_ARG(priv->bssid));
return 0;
}
static void ipw_roam(void *data)
{
struct ipw_priv *priv = data;
struct ieee80211_network *network = NULL;
struct ipw_network_match match = {
.network = priv->assoc_network
};
/* The roaming process is as follows:
*
* 1. Missed beacon threshold triggers the roaming process by
2005-03-25 13:31:53 -05:00
* setting the status ROAM bit and requesting a scan.
* 2. When the scan completes, it schedules the ROAM work
* 3. The ROAM work looks at all of the known networks for one that
* is a better network than the currently associated. If none
* found, the ROAM process is over (ROAM bit cleared)
* 4. If a better network is found, a disassociation request is
* sent.
* 5. When the disassociation completes, the roam work is again
* scheduled. The second time through, the driver is no longer
* associated, and the newly selected network is sent an
* association request.
2005-03-25 13:31:53 -05:00
* 6. At this point ,the roaming process is complete and the ROAM
* status bit is cleared.
*/
/* If we are no longer associated, and the roaming bit is no longer
* set, then we are not actively roaming, so just return */
if (!(priv->status & (STATUS_ASSOCIATED | STATUS_ROAMING)))
return;
2005-03-25 13:31:53 -05:00
if (priv->status & STATUS_ASSOCIATED) {
/* First pass through ROAM process -- look for a better
2005-03-25 13:31:53 -05:00
* network */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
unsigned long flags;
2005-03-25 13:31:53 -05:00
u8 rssi = priv->assoc_network->stats.rssi;
priv->assoc_network->stats.rssi = -128;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_lock_irqsave(&priv->ieee->lock, flags);
2005-03-25 13:31:53 -05:00
list_for_each_entry(network, &priv->ieee->network_list, list) {
if (network != priv->assoc_network)
ipw_best_network(priv, &match, network, 1);
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->ieee->lock, flags);
2005-03-25 13:31:53 -05:00
priv->assoc_network->stats.rssi = rssi;
2005-03-25 13:31:53 -05:00
if (match.network == priv->assoc_network) {
IPW_DEBUG_ASSOC("No better APs in this network to "
"roam to.\n");
priv->status &= ~STATUS_ROAMING;
ipw_debug_config(priv);
return;
}
2005-03-25 13:31:53 -05:00
ipw_send_disassociate(priv, 1);
priv->assoc_network = match.network;
return;
}
2005-03-25 13:31:53 -05:00
/* Second pass through ROAM process -- request association */
ipw_compatible_rates(priv, priv->assoc_network, &match.rates);
ipw_associate_network(priv, priv->assoc_network, &match.rates, 1);
priv->status &= ~STATUS_ROAMING;
}
static void ipw_bg_roam(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_roam(data);
mutex_unlock(&priv->mutex);
}
static int ipw_associate(void *data)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = data;
struct ieee80211_network *network = NULL;
struct ipw_network_match match = {
.network = NULL
};
struct ipw_supported_rates *rates;
struct list_head *element;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
unsigned long flags;
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->ieee->iw_mode == IW_MODE_MONITOR) {
IPW_DEBUG_ASSOC("Not attempting association (monitor mode)\n");
return 0;
}
if (priv->status & (STATUS_ASSOCIATED | STATUS_ASSOCIATING)) {
IPW_DEBUG_ASSOC("Not attempting association (already in "
"progress)\n");
return 0;
}
if (priv->status & STATUS_DISASSOCIATING) {
IPW_DEBUG_ASSOC("Not attempting association (in "
"disassociating)\n ");
queue_work(priv->workqueue, &priv->associate);
return 0;
}
if (!ipw_is_init(priv) || (priv->status & STATUS_SCANNING)) {
IPW_DEBUG_ASSOC("Not attempting association (scanning or not "
"initialized)\n");
return 0;
}
2005-03-25 13:31:53 -05:00
if (!(priv->config & CFG_ASSOCIATE) &&
!(priv->config & (CFG_STATIC_ESSID |
CFG_STATIC_CHANNEL | CFG_STATIC_BSSID))) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ASSOC("Not attempting association (associate=0)\n");
return 0;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
/* Protect our use of the network_list */
spin_lock_irqsave(&priv->ieee->lock, flags);
list_for_each_entry(network, &priv->ieee->network_list, list)
ipw_best_network(priv, &match, network, 0);
2005-03-25 13:31:53 -05:00
network = match.network;
rates = &match.rates;
if (network == NULL &&
priv->ieee->iw_mode == IW_MODE_ADHOC &&
priv->config & CFG_ADHOC_CREATE &&
priv->config & CFG_STATIC_ESSID &&
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
priv->config & CFG_STATIC_CHANNEL &&
2005-03-25 13:31:53 -05:00
!list_empty(&priv->ieee->network_free_list)) {
element = priv->ieee->network_free_list.next;
network = list_entry(element, struct ieee80211_network, list);
2005-03-25 13:31:53 -05:00
ipw_adhoc_create(priv, network);
rates = &priv->rates;
list_del(element);
list_add_tail(&network->list, &priv->ieee->network_list);
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
spin_unlock_irqrestore(&priv->ieee->lock, flags);
2005-03-25 13:31:53 -05:00
/* If we reached the end of the list, then we don't have any valid
* matching APs */
if (!network) {
ipw_debug_config(priv);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (!(priv->status & STATUS_SCANNING)) {
if (!(priv->config & CFG_SPEED_SCAN))
queue_delayed_work(priv->workqueue,
&priv->request_scan,
SCAN_INTERVAL);
else
queue_work(priv->workqueue,
&priv->request_scan);
}
return 0;
2005-03-25 13:31:53 -05:00
}
ipw_associate_network(priv, network, rates, 0);
return 1;
}
static void ipw_bg_associate(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_associate(data);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_rebuild_decrypted_skb(struct ipw_priv *priv,
struct sk_buff *skb)
{
struct ieee80211_hdr *hdr;
u16 fc;
hdr = (struct ieee80211_hdr *)skb->data;
fc = le16_to_cpu(hdr->frame_ctl);
if (!(fc & IEEE80211_FCTL_PROTECTED))
return;
fc &= ~IEEE80211_FCTL_PROTECTED;
hdr->frame_ctl = cpu_to_le16(fc);
switch (priv->ieee->sec.level) {
case SEC_LEVEL_3:
/* Remove CCMP HDR */
memmove(skb->data + IEEE80211_3ADDR_LEN,
skb->data + IEEE80211_3ADDR_LEN + 8,
skb->len - IEEE80211_3ADDR_LEN - 8);
skb_trim(skb, skb->len - 16); /* CCMP_HDR_LEN + CCMP_MIC_LEN */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
case SEC_LEVEL_2:
break;
case SEC_LEVEL_1:
/* Remove IV */
memmove(skb->data + IEEE80211_3ADDR_LEN,
skb->data + IEEE80211_3ADDR_LEN + 4,
skb->len - IEEE80211_3ADDR_LEN - 4);
skb_trim(skb, skb->len - 8); /* IV + ICV */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
case SEC_LEVEL_0:
break;
default:
printk(KERN_ERR "Unknow security level %d\n",
priv->ieee->sec.level);
break;
}
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_handle_data_packet(struct ipw_priv *priv,
struct ipw_rx_mem_buffer *rxb,
struct ieee80211_rx_stats *stats)
2005-03-25 13:31:53 -05:00
{
struct ieee80211_hdr_4addr *hdr;
2005-03-25 13:31:53 -05:00
struct ipw_rx_packet *pkt = (struct ipw_rx_packet *)rxb->skb->data;
/* We received data from the HW, so stop the watchdog */
priv->net_dev->trans_start = jiffies;
/* We only process data packets if the
2005-03-25 13:31:53 -05:00
* interface is open */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (unlikely((le16_to_cpu(pkt->u.frame.length) + IPW_RX_FRAME_SIZE) >
2005-03-25 13:31:53 -05:00
skb_tailroom(rxb->skb))) {
priv->ieee->stats.rx_errors++;
priv->wstats.discard.misc++;
IPW_DEBUG_DROP("Corruption detected! Oh no!\n");
return;
} else if (unlikely(!netif_running(priv->net_dev))) {
priv->ieee->stats.rx_dropped++;
priv->wstats.discard.misc++;
IPW_DEBUG_DROP("Dropping packet while interface is not up.\n");
return;
}
/* Advance skb->data to the start of the actual payload */
skb_reserve(rxb->skb, offsetof(struct ipw_rx_packet, u.frame.data));
2005-03-25 13:31:53 -05:00
/* Set the size of the skb to the size of the frame */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
skb_put(rxb->skb, le16_to_cpu(pkt->u.frame.length));
2005-03-25 13:31:53 -05:00
IPW_DEBUG_RX("Rx packet of %d bytes.\n", rxb->skb->len);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* HW decrypt will not clear the WEP bit, MIC, PN, etc. */
hdr = (struct ieee80211_hdr_4addr *)rxb->skb->data;
if (priv->ieee->iw_mode != IW_MODE_MONITOR &&
(is_multicast_ether_addr(hdr->addr1) ?
!priv->ieee->host_mc_decrypt : !priv->ieee->host_decrypt))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_rebuild_decrypted_skb(priv, rxb->skb);
if (!ieee80211_rx(priv->ieee, rxb->skb, stats))
2005-03-25 13:31:53 -05:00
priv->ieee->stats.rx_errors++;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
else { /* ieee80211_rx succeeded, so it now owns the SKB */
2005-03-25 13:31:53 -05:00
rxb->skb = NULL;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
__ipw_led_activity_on(priv);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
}
2005-03-25 13:31:53 -05:00
}
#ifdef CONFIG_IEEE80211_RADIOTAP
static void ipw_handle_data_packet_monitor(struct ipw_priv *priv,
struct ipw_rx_mem_buffer *rxb,
struct ieee80211_rx_stats *stats)
{
struct ipw_rx_packet *pkt = (struct ipw_rx_packet *)rxb->skb->data;
struct ipw_rx_frame *frame = &pkt->u.frame;
/* initial pull of some data */
u16 received_channel = frame->received_channel;
u8 antennaAndPhy = frame->antennaAndPhy;
s8 antsignal = frame->rssi_dbm - IPW_RSSI_TO_DBM; /* call it signed anyhow */
u16 pktrate = frame->rate;
/* Magic struct that slots into the radiotap header -- no reason
* to build this manually element by element, we can write it much
* more efficiently than we can parse it. ORDER MATTERS HERE */
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
struct ipw_rt_hdr *ipw_rt;
short len = le16_to_cpu(pkt->u.frame.length);
/* We received data from the HW, so stop the watchdog */
priv->net_dev->trans_start = jiffies;
/* We only process data packets if the
* interface is open */
if (unlikely((le16_to_cpu(pkt->u.frame.length) + IPW_RX_FRAME_SIZE) >
skb_tailroom(rxb->skb))) {
priv->ieee->stats.rx_errors++;
priv->wstats.discard.misc++;
IPW_DEBUG_DROP("Corruption detected! Oh no!\n");
return;
} else if (unlikely(!netif_running(priv->net_dev))) {
priv->ieee->stats.rx_dropped++;
priv->wstats.discard.misc++;
IPW_DEBUG_DROP("Dropping packet while interface is not up.\n");
return;
}
/* Libpcap 0.9.3+ can handle variable length radiotap, so we'll use
* that now */
if (len > IPW_RX_BUF_SIZE - sizeof(struct ipw_rt_hdr)) {
/* FIXME: Should alloc bigger skb instead */
priv->ieee->stats.rx_dropped++;
priv->wstats.discard.misc++;
IPW_DEBUG_DROP("Dropping too large packet in monitor\n");
return;
}
/* copy the frame itself */
memmove(rxb->skb->data + sizeof(struct ipw_rt_hdr),
rxb->skb->data + IPW_RX_FRAME_SIZE, len);
/* Zero the radiotap static buffer ... We only need to zero the bytes NOT
* part of our real header, saves a little time.
*
* No longer necessary since we fill in all our data. Purge before merging
* patch officially.
* memset(rxb->skb->data + sizeof(struct ipw_rt_hdr), 0,
* IEEE80211_RADIOTAP_HDRLEN - sizeof(struct ipw_rt_hdr));
*/
ipw_rt = (struct ipw_rt_hdr *)rxb->skb->data;
ipw_rt->rt_hdr.it_version = PKTHDR_RADIOTAP_VERSION;
ipw_rt->rt_hdr.it_pad = 0; /* always good to zero */
ipw_rt->rt_hdr.it_len = sizeof(struct ipw_rt_hdr); /* total header+data */
/* Big bitfield of all the fields we provide in radiotap */
ipw_rt->rt_hdr.it_present =
((1 << IEEE80211_RADIOTAP_FLAGS) |
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
(1 << IEEE80211_RADIOTAP_TSFT) |
(1 << IEEE80211_RADIOTAP_RATE) |
(1 << IEEE80211_RADIOTAP_CHANNEL) |
(1 << IEEE80211_RADIOTAP_DBM_ANTSIGNAL) |
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
(1 << IEEE80211_RADIOTAP_DBM_ANTNOISE) |
(1 << IEEE80211_RADIOTAP_ANTENNA));
/* Zero the flags, we'll add to them as we go */
ipw_rt->rt_flags = 0;
/* Convert signal to DBM */
ipw_rt->rt_dbmsignal = antsignal;
/* Convert the channel data and set the flags */
ipw_rt->rt_channel = cpu_to_le16(ieee80211chan2mhz(received_channel));
if (received_channel > 14) { /* 802.11a */
ipw_rt->rt_chbitmask =
cpu_to_le16((IEEE80211_CHAN_OFDM | IEEE80211_CHAN_5GHZ));
} else if (antennaAndPhy & 32) { /* 802.11b */
ipw_rt->rt_chbitmask =
cpu_to_le16((IEEE80211_CHAN_CCK | IEEE80211_CHAN_2GHZ));
} else { /* 802.11g */
ipw_rt->rt_chbitmask =
(IEEE80211_CHAN_OFDM | IEEE80211_CHAN_2GHZ);
}
/* set the rate in multiples of 500k/s */
switch (pktrate) {
case IPW_TX_RATE_1MB:
ipw_rt->rt_rate = 2;
break;
case IPW_TX_RATE_2MB:
ipw_rt->rt_rate = 4;
break;
case IPW_TX_RATE_5MB:
ipw_rt->rt_rate = 10;
break;
case IPW_TX_RATE_6MB:
ipw_rt->rt_rate = 12;
break;
case IPW_TX_RATE_9MB:
ipw_rt->rt_rate = 18;
break;
case IPW_TX_RATE_11MB:
ipw_rt->rt_rate = 22;
break;
case IPW_TX_RATE_12MB:
ipw_rt->rt_rate = 24;
break;
case IPW_TX_RATE_18MB:
ipw_rt->rt_rate = 36;
break;
case IPW_TX_RATE_24MB:
ipw_rt->rt_rate = 48;
break;
case IPW_TX_RATE_36MB:
ipw_rt->rt_rate = 72;
break;
case IPW_TX_RATE_48MB:
ipw_rt->rt_rate = 96;
break;
case IPW_TX_RATE_54MB:
ipw_rt->rt_rate = 108;
break;
default:
ipw_rt->rt_rate = 0;
break;
}
/* antenna number */
ipw_rt->rt_antenna = (antennaAndPhy & 3); /* Is this right? */
/* set the preamble flag if we have it */
if ((antennaAndPhy & 64))
ipw_rt->rt_flags |= IEEE80211_RADIOTAP_F_SHORTPRE;
/* Set the size of the skb to the size of the frame */
skb_put(rxb->skb, len + sizeof(struct ipw_rt_hdr));
2005-03-25 13:31:53 -05:00
IPW_DEBUG_RX("Rx packet of %d bytes.\n", rxb->skb->len);
if (!ieee80211_rx(priv->ieee, rxb->skb, stats))
2005-03-25 13:31:53 -05:00
priv->ieee->stats.rx_errors++;
else { /* ieee80211_rx succeeded, so it now owns the SKB */
rxb->skb = NULL;
/* no LED during capture */
}
}
#endif
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
#define ieee80211_is_probe_response(fc) \
((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT && \
(fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PROBE_RESP )
#define ieee80211_is_management(fc) \
((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)
#define ieee80211_is_control(fc) \
((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL)
#define ieee80211_is_data(fc) \
((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA)
#define ieee80211_is_assoc_request(fc) \
((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ)
#define ieee80211_is_reassoc_request(fc) \
((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_REQ)
static void ipw_handle_promiscuous_rx(struct ipw_priv *priv,
struct ipw_rx_mem_buffer *rxb,
struct ieee80211_rx_stats *stats)
{
struct ipw_rx_packet *pkt = (struct ipw_rx_packet *)rxb->skb->data;
struct ipw_rx_frame *frame = &pkt->u.frame;
struct ipw_rt_hdr *ipw_rt;
/* First cache any information we need before we overwrite
* the information provided in the skb from the hardware */
struct ieee80211_hdr *hdr;
u16 channel = frame->received_channel;
u8 phy_flags = frame->antennaAndPhy;
s8 signal = frame->rssi_dbm - IPW_RSSI_TO_DBM;
s8 noise = frame->noise;
u8 rate = frame->rate;
short len = le16_to_cpu(pkt->u.frame.length);
u64 tsf = 0;
struct sk_buff *skb;
int hdr_only = 0;
u16 filter = priv->prom_priv->filter;
/* If the filter is set to not include Rx frames then return */
if (filter & IPW_PROM_NO_RX)
return;
if (!noise)
noise = priv->last_noise;
/* We received data from the HW, so stop the watchdog */
priv->prom_net_dev->trans_start = jiffies;
if (unlikely((len + IPW_RX_FRAME_SIZE) > skb_tailroom(rxb->skb))) {
priv->prom_priv->ieee->stats.rx_errors++;
IPW_DEBUG_DROP("Corruption detected! Oh no!\n");
return;
}
/* We only process data packets if the interface is open */
if (unlikely(!netif_running(priv->prom_net_dev))) {
priv->prom_priv->ieee->stats.rx_dropped++;
IPW_DEBUG_DROP("Dropping packet while interface is not up.\n");
return;
}
/* Libpcap 0.9.3+ can handle variable length radiotap, so we'll use
* that now */
if (len > IPW_RX_BUF_SIZE - sizeof(struct ipw_rt_hdr)) {
/* FIXME: Should alloc bigger skb instead */
priv->prom_priv->ieee->stats.rx_dropped++;
IPW_DEBUG_DROP("Dropping too large packet in monitor\n");
return;
}
hdr = (void *)rxb->skb->data + IPW_RX_FRAME_SIZE;
if (ieee80211_is_management(hdr->frame_ctl)) {
if (filter & IPW_PROM_NO_MGMT)
return;
if (filter & IPW_PROM_MGMT_HEADER_ONLY)
hdr_only = 1;
} else if (ieee80211_is_control(hdr->frame_ctl)) {
if (filter & IPW_PROM_NO_CTL)
return;
if (filter & IPW_PROM_CTL_HEADER_ONLY)
hdr_only = 1;
} else if (ieee80211_is_data(hdr->frame_ctl)) {
if (filter & IPW_PROM_NO_DATA)
return;
if (filter & IPW_PROM_DATA_HEADER_ONLY)
hdr_only = 1;
}
/* Copy the SKB since this is for the promiscuous side */
skb = skb_copy(rxb->skb, GFP_ATOMIC);
if (skb == NULL) {
IPW_ERROR("skb_clone failed for promiscuous copy.\n");
return;
}
/* copy the frame data to write after where the radiotap header goes */
ipw_rt = (void *)skb->data;
if (hdr_only)
len = ieee80211_get_hdrlen(hdr->frame_ctl);
memcpy(ipw_rt->payload, hdr, len);
/* Zero the radiotap static buffer ... We only need to zero the bytes
* NOT part of our real header, saves a little time.
*
* No longer necessary since we fill in all our data. Purge before
* merging patch officially.
* memset(rxb->skb->data + sizeof(struct ipw_rt_hdr), 0,
* IEEE80211_RADIOTAP_HDRLEN - sizeof(struct ipw_rt_hdr));
*/
ipw_rt->rt_hdr.it_version = PKTHDR_RADIOTAP_VERSION;
ipw_rt->rt_hdr.it_pad = 0; /* always good to zero */
ipw_rt->rt_hdr.it_len = sizeof(*ipw_rt); /* total header+data */
/* Set the size of the skb to the size of the frame */
skb_put(skb, ipw_rt->rt_hdr.it_len + len);
/* Big bitfield of all the fields we provide in radiotap */
ipw_rt->rt_hdr.it_present =
((1 << IEEE80211_RADIOTAP_FLAGS) |
(1 << IEEE80211_RADIOTAP_TSFT) |
(1 << IEEE80211_RADIOTAP_RATE) |
(1 << IEEE80211_RADIOTAP_CHANNEL) |
(1 << IEEE80211_RADIOTAP_DBM_ANTSIGNAL) |
(1 << IEEE80211_RADIOTAP_DBM_ANTNOISE) |
(1 << IEEE80211_RADIOTAP_ANTENNA));
/* Zero the flags, we'll add to them as we go */
ipw_rt->rt_flags = 0;
ipw_rt->rt_tsf = tsf;
/* Convert to DBM */
ipw_rt->rt_dbmsignal = signal;
ipw_rt->rt_dbmnoise = noise;
/* Convert the channel data and set the flags */
ipw_rt->rt_channel = cpu_to_le16(ieee80211chan2mhz(channel));
if (channel > 14) { /* 802.11a */
ipw_rt->rt_chbitmask =
cpu_to_le16((IEEE80211_CHAN_OFDM | IEEE80211_CHAN_5GHZ));
} else if (phy_flags & (1 << 5)) { /* 802.11b */
ipw_rt->rt_chbitmask =
cpu_to_le16((IEEE80211_CHAN_CCK | IEEE80211_CHAN_2GHZ));
} else { /* 802.11g */
ipw_rt->rt_chbitmask =
(IEEE80211_CHAN_OFDM | IEEE80211_CHAN_2GHZ);
}
/* set the rate in multiples of 500k/s */
switch (rate) {
case IPW_TX_RATE_1MB:
ipw_rt->rt_rate = 2;
break;
case IPW_TX_RATE_2MB:
ipw_rt->rt_rate = 4;
break;
case IPW_TX_RATE_5MB:
ipw_rt->rt_rate = 10;
break;
case IPW_TX_RATE_6MB:
ipw_rt->rt_rate = 12;
break;
case IPW_TX_RATE_9MB:
ipw_rt->rt_rate = 18;
break;
case IPW_TX_RATE_11MB:
ipw_rt->rt_rate = 22;
break;
case IPW_TX_RATE_12MB:
ipw_rt->rt_rate = 24;
break;
case IPW_TX_RATE_18MB:
ipw_rt->rt_rate = 36;
break;
case IPW_TX_RATE_24MB:
ipw_rt->rt_rate = 48;
break;
case IPW_TX_RATE_36MB:
ipw_rt->rt_rate = 72;
break;
case IPW_TX_RATE_48MB:
ipw_rt->rt_rate = 96;
break;
case IPW_TX_RATE_54MB:
ipw_rt->rt_rate = 108;
break;
default:
ipw_rt->rt_rate = 0;
break;
}
/* antenna number */
ipw_rt->rt_antenna = (phy_flags & 3);
/* set the preamble flag if we have it */
if (phy_flags & (1 << 6))
ipw_rt->rt_flags |= IEEE80211_RADIOTAP_F_SHORTPRE;
IPW_DEBUG_RX("Rx packet of %d bytes.\n", skb->len);
if (!ieee80211_rx(priv->prom_priv->ieee, skb, stats)) {
priv->prom_priv->ieee->stats.rx_errors++;
dev_kfree_skb_any(skb);
}
}
#endif
static int is_network_packet(struct ipw_priv *priv,
struct ieee80211_hdr_4addr *header)
{
/* Filter incoming packets to determine if they are targetted toward
* this network, discarding packets coming from ourselves */
switch (priv->ieee->iw_mode) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
case IW_MODE_ADHOC: /* Header: Dest. | Source | BSSID */
/* packets from our adapter are dropped (echo) */
if (!memcmp(header->addr2, priv->net_dev->dev_addr, ETH_ALEN))
return 0;
/* {broad,multi}cast packets to our BSSID go through */
if (is_multicast_ether_addr(header->addr1))
return !memcmp(header->addr3, priv->bssid, ETH_ALEN);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
/* packets to our adapter go through */
return !memcmp(header->addr1, priv->net_dev->dev_addr,
ETH_ALEN);
case IW_MODE_INFRA: /* Header: Dest. | BSSID | Source */
/* packets from our adapter are dropped (echo) */
if (!memcmp(header->addr3, priv->net_dev->dev_addr, ETH_ALEN))
return 0;
/* {broad,multi}cast packets to our BSS go through */
if (is_multicast_ether_addr(header->addr1))
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
return !memcmp(header->addr2, priv->bssid, ETH_ALEN);
/* packets to our adapter go through */
return !memcmp(header->addr1, priv->net_dev->dev_addr,
ETH_ALEN);
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
return 1;
}
#define IPW_PACKET_RETRY_TIME HZ
static int is_duplicate_packet(struct ipw_priv *priv,
struct ieee80211_hdr_4addr *header)
{
u16 sc = le16_to_cpu(header->seq_ctl);
u16 seq = WLAN_GET_SEQ_SEQ(sc);
u16 frag = WLAN_GET_SEQ_FRAG(sc);
u16 *last_seq, *last_frag;
unsigned long *last_time;
switch (priv->ieee->iw_mode) {
case IW_MODE_ADHOC:
{
struct list_head *p;
struct ipw_ibss_seq *entry = NULL;
u8 *mac = header->addr2;
int index = mac[5] % IPW_IBSS_MAC_HASH_SIZE;
__list_for_each(p, &priv->ibss_mac_hash[index]) {
entry =
list_entry(p, struct ipw_ibss_seq, list);
if (!memcmp(entry->mac, mac, ETH_ALEN))
break;
}
if (p == &priv->ibss_mac_hash[index]) {
entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
if (!entry) {
IPW_ERROR
("Cannot malloc new mac entry\n");
return 0;
}
memcpy(entry->mac, mac, ETH_ALEN);
entry->seq_num = seq;
entry->frag_num = frag;
entry->packet_time = jiffies;
list_add(&entry->list,
&priv->ibss_mac_hash[index]);
return 0;
}
last_seq = &entry->seq_num;
last_frag = &entry->frag_num;
last_time = &entry->packet_time;
break;
}
case IW_MODE_INFRA:
last_seq = &priv->last_seq_num;
last_frag = &priv->last_frag_num;
last_time = &priv->last_packet_time;
break;
default:
return 0;
}
if ((*last_seq == seq) &&
time_after(*last_time + IPW_PACKET_RETRY_TIME, jiffies)) {
if (*last_frag == frag)
goto drop;
if (*last_frag + 1 != frag)
/* out-of-order fragment */
goto drop;
} else
*last_seq = seq;
*last_frag = frag;
*last_time = jiffies;
return 0;
drop:
/* Comment this line now since we observed the card receives
* duplicate packets but the FCTL_RETRY bit is not set in the
* IBSS mode with fragmentation enabled.
BUG_ON(!(le16_to_cpu(header->frame_ctl) & IEEE80211_FCTL_RETRY)); */
return 1;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_handle_mgmt_packet(struct ipw_priv *priv,
struct ipw_rx_mem_buffer *rxb,
struct ieee80211_rx_stats *stats)
{
struct sk_buff *skb = rxb->skb;
struct ipw_rx_packet *pkt = (struct ipw_rx_packet *)skb->data;
struct ieee80211_hdr_4addr *header = (struct ieee80211_hdr_4addr *)
(skb->data + IPW_RX_FRAME_SIZE);
ieee80211_rx_mgt(priv->ieee, header, stats);
if (priv->ieee->iw_mode == IW_MODE_ADHOC &&
((WLAN_FC_GET_STYPE(le16_to_cpu(header->frame_ctl)) ==
IEEE80211_STYPE_PROBE_RESP) ||
(WLAN_FC_GET_STYPE(le16_to_cpu(header->frame_ctl)) ==
IEEE80211_STYPE_BEACON))) {
if (!memcmp(header->addr3, priv->bssid, ETH_ALEN))
ipw_add_station(priv, header->addr2);
}
if (priv->config & CFG_NET_STATS) {
IPW_DEBUG_HC("sending stat packet\n");
/* Set the size of the skb to the size of the full
* ipw header and 802.11 frame */
skb_put(skb, le16_to_cpu(pkt->u.frame.length) +
IPW_RX_FRAME_SIZE);
/* Advance past the ipw packet header to the 802.11 frame */
skb_pull(skb, IPW_RX_FRAME_SIZE);
/* Push the ieee80211_rx_stats before the 802.11 frame */
memcpy(skb_push(skb, sizeof(*stats)), stats, sizeof(*stats));
skb->dev = priv->ieee->dev;
/* Point raw at the ieee80211_stats */
skb->mac.raw = skb->data;
skb->pkt_type = PACKET_OTHERHOST;
skb->protocol = __constant_htons(ETH_P_80211_STATS);
memset(skb->cb, 0, sizeof(rxb->skb->cb));
netif_rx(skb);
2005-03-25 13:31:53 -05:00
rxb->skb = NULL;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
2005-03-25 13:31:53 -05:00
}
/*
* Main entry function for recieving a packet with 80211 headers. This
* should be called when ever the FW has notified us that there is a new
* skb in the recieve queue.
*/
static void ipw_rx(struct ipw_priv *priv)
{
struct ipw_rx_mem_buffer *rxb;
struct ipw_rx_packet *pkt;
struct ieee80211_hdr_4addr *header;
2005-03-25 13:31:53 -05:00
u32 r, w, i;
u8 network_packet;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
r = ipw_read32(priv, IPW_RX_READ_INDEX);
w = ipw_read32(priv, IPW_RX_WRITE_INDEX);
2005-03-25 13:31:53 -05:00
i = (priv->rxq->processed + 1) % RX_QUEUE_SIZE;
while (i != r) {
rxb = priv->rxq->queue[i];
if (unlikely(rxb == NULL)) {
printk(KERN_CRIT "Queue not allocated!\n");
break;
}
priv->rxq->queue[i] = NULL;
pci_dma_sync_single_for_cpu(priv->pci_dev, rxb->dma_addr,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_RX_BUF_SIZE,
2005-03-25 13:31:53 -05:00
PCI_DMA_FROMDEVICE);
pkt = (struct ipw_rx_packet *)rxb->skb->data;
IPW_DEBUG_RX("Packet: type=%02X seq=%02X bits=%02X\n",
pkt->header.message_type,
pkt->header.rx_seq_num, pkt->header.control_bits);
2005-03-25 13:31:53 -05:00
switch (pkt->header.message_type) {
case RX_FRAME_TYPE: /* 802.11 frame */ {
struct ieee80211_rx_stats stats = {
.rssi =
le16_to_cpu(pkt->u.frame.rssi_dbm) -
IPW_RSSI_TO_DBM,
.signal =
le16_to_cpu(pkt->u.frame.rssi_dbm) -
IPW_RSSI_TO_DBM + 0x100,
.noise =
le16_to_cpu(pkt->u.frame.noise),
.rate = pkt->u.frame.rate,
.mac_time = jiffies,
.received_channel =
pkt->u.frame.received_channel,
.freq =
(pkt->u.frame.
control & (1 << 0)) ?
IEEE80211_24GHZ_BAND :
IEEE80211_52GHZ_BAND,
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
.len = le16_to_cpu(pkt->u.frame.length),
};
if (stats.rssi != 0)
stats.mask |= IEEE80211_STATMASK_RSSI;
if (stats.signal != 0)
stats.mask |= IEEE80211_STATMASK_SIGNAL;
if (stats.noise != 0)
stats.mask |= IEEE80211_STATMASK_NOISE;
if (stats.rate != 0)
stats.mask |= IEEE80211_STATMASK_RATE;
priv->rx_packets++;
2005-03-25 13:31:53 -05:00
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
if (priv->prom_net_dev && netif_running(priv->prom_net_dev))
ipw_handle_promiscuous_rx(priv, rxb, &stats);
#endif
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
if (priv->ieee->iw_mode == IW_MODE_MONITOR) {
#ifdef CONFIG_IEEE80211_RADIOTAP
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
ipw_handle_data_packet_monitor(priv,
rxb,
&stats);
#else
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
ipw_handle_data_packet(priv, rxb,
&stats);
#endif
break;
}
2005-03-25 13:31:53 -05:00
#endif
header =
(struct ieee80211_hdr_4addr *)(rxb->skb->
data +
IPW_RX_FRAME_SIZE);
2005-03-25 13:31:53 -05:00
/* TODO: Check Ad-Hoc dest/source and make sure
* that we are actually parsing these packets
* correctly -- we should probably use the
2005-03-25 13:31:53 -05:00
* frame control of the packet and disregard
* the current iw_mode */
network_packet =
is_network_packet(priv, header);
if (network_packet && priv->assoc_network) {
priv->assoc_network->stats.rssi =
stats.rssi;
priv->exp_avg_rssi =
exponential_average(priv->exp_avg_rssi,
stats.rssi, DEPTH_RSSI);
}
IPW_DEBUG_RX("Frame: len=%u\n",
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
le16_to_cpu(pkt->u.frame.length));
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (le16_to_cpu(pkt->u.frame.length) <
ieee80211_get_hdrlen(le16_to_cpu(
header->frame_ctl))) {
IPW_DEBUG_DROP
("Received packet is too small. "
"Dropping.\n");
priv->ieee->stats.rx_errors++;
priv->wstats.discard.misc++;
break;
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
switch (WLAN_FC_GET_TYPE
(le16_to_cpu(header->frame_ctl))) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
case IEEE80211_FTYPE_MGMT:
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_handle_mgmt_packet(priv, rxb,
&stats);
break;
case IEEE80211_FTYPE_CTL:
break;
case IEEE80211_FTYPE_DATA:
if (unlikely(!network_packet ||
is_duplicate_packet(priv,
header)))
{
IPW_DEBUG_DROP("Dropping: "
MAC_FMT ", "
MAC_FMT ", "
MAC_FMT "\n",
MAC_ARG(header->
addr1),
MAC_ARG(header->
addr2),
MAC_ARG(header->
addr3));
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
break;
}
ipw_handle_data_packet(priv, rxb,
&stats);
break;
}
2005-03-25 13:31:53 -05:00
break;
}
case RX_HOST_NOTIFICATION_TYPE:{
IPW_DEBUG_RX
("Notification: subtype=%02X flags=%02X size=%d\n",
2005-03-25 13:31:53 -05:00
pkt->u.notification.subtype,
pkt->u.notification.flags,
pkt->u.notification.size);
ipw_rx_notification(priv, &pkt->u.notification);
break;
}
2005-03-25 13:31:53 -05:00
default:
IPW_DEBUG_RX("Bad Rx packet of type %d\n",
pkt->header.message_type);
break;
}
/* For now we just don't re-use anything. We can tweak this
* later to try and re-use notification packets and SKBs that
2005-03-25 13:31:53 -05:00
* fail to Rx correctly */
if (rxb->skb != NULL) {
dev_kfree_skb_any(rxb->skb);
rxb->skb = NULL;
}
2005-03-25 13:31:53 -05:00
pci_unmap_single(priv->pci_dev, rxb->dma_addr,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_RX_BUF_SIZE, PCI_DMA_FROMDEVICE);
2005-03-25 13:31:53 -05:00
list_add_tail(&rxb->list, &priv->rxq->rx_used);
2005-03-25 13:31:53 -05:00
i = (i + 1) % RX_QUEUE_SIZE;
}
/* Backtrack one entry */
priv->rxq->processed = (i ? i : RX_QUEUE_SIZE) - 1;
ipw_rx_queue_restock(priv);
}
#define DEFAULT_RTS_THRESHOLD 2304U
#define MIN_RTS_THRESHOLD 1U
#define MAX_RTS_THRESHOLD 2304U
#define DEFAULT_BEACON_INTERVAL 100U
#define DEFAULT_SHORT_RETRY_LIMIT 7U
#define DEFAULT_LONG_RETRY_LIMIT 4U
/**
* ipw_sw_reset
* @option: options to control different reset behaviour
* 0 = reset everything except the 'disable' module_param
* 1 = reset everything and print out driver info (for probe only)
* 2 = reset everything
*/
static int ipw_sw_reset(struct ipw_priv *priv, int option)
2005-03-25 13:31:53 -05:00
{
int band, modulation;
int old_mode = priv->ieee->iw_mode;
2005-03-25 13:31:53 -05:00
/* Initialize module parameter values here */
priv->config = 0;
2005-03-25 13:31:53 -05:00
/* We default to disabling the LED code as right now it causes
* too many systems to lock up... */
if (!led)
priv->config |= CFG_NO_LED;
2005-03-25 13:31:53 -05:00
if (associate)
priv->config |= CFG_ASSOCIATE;
else
IPW_DEBUG_INFO("Auto associate disabled.\n");
if (auto_create)
priv->config |= CFG_ADHOC_CREATE;
else
IPW_DEBUG_INFO("Auto adhoc creation disabled.\n");
2005-03-25 13:31:53 -05:00
priv->config &= ~CFG_STATIC_ESSID;
priv->essid_len = 0;
memset(priv->essid, 0, IW_ESSID_MAX_SIZE);
if (disable && option) {
priv->status |= STATUS_RF_KILL_SW;
IPW_DEBUG_INFO("Radio disabled.\n");
2005-03-25 13:31:53 -05:00
}
if (channel != 0) {
priv->config |= CFG_STATIC_CHANNEL;
priv->channel = channel;
IPW_DEBUG_INFO("Bind to static channel %d\n", channel);
/* TODO: Validate that provided channel is in range */
2005-03-25 13:31:53 -05:00
}
#ifdef CONFIG_IPW2200_QOS
ipw_qos_init(priv, qos_enable, qos_burst_enable,
burst_duration_CCK, burst_duration_OFDM);
#endif /* CONFIG_IPW2200_QOS */
2005-03-25 13:31:53 -05:00
switch (mode) {
case 1:
priv->ieee->iw_mode = IW_MODE_ADHOC;
priv->net_dev->type = ARPHRD_ETHER;
break;
#ifdef CONFIG_IPW2200_MONITOR
case 2:
priv->ieee->iw_mode = IW_MODE_MONITOR;
#ifdef CONFIG_IEEE80211_RADIOTAP
priv->net_dev->type = ARPHRD_IEEE80211_RADIOTAP;
#else
priv->net_dev->type = ARPHRD_IEEE80211;
#endif
break;
#endif
default:
case 0:
priv->net_dev->type = ARPHRD_ETHER;
priv->ieee->iw_mode = IW_MODE_INFRA;
break;
2005-03-25 13:31:53 -05:00
}
if (hwcrypto) {
priv->ieee->host_encrypt = 0;
priv->ieee->host_encrypt_msdu = 0;
priv->ieee->host_decrypt = 0;
priv->ieee->host_mc_decrypt = 0;
}
IPW_DEBUG_INFO("Hardware crypto [%s]\n", hwcrypto ? "on" : "off");
2005-03-25 13:31:53 -05:00
/* IPW2200/2915 is abled to do hardware fragmentation. */
priv->ieee->host_open_frag = 0;
if ((priv->pci_dev->device == 0x4223) ||
(priv->pci_dev->device == 0x4224)) {
if (option == 1)
printk(KERN_INFO DRV_NAME
": Detected Intel PRO/Wireless 2915ABG Network "
"Connection\n");
priv->ieee->abg_true = 1;
band = IEEE80211_52GHZ_BAND | IEEE80211_24GHZ_BAND;
modulation = IEEE80211_OFDM_MODULATION |
IEEE80211_CCK_MODULATION;
priv->adapter = IPW_2915ABG;
priv->ieee->mode = IEEE_A | IEEE_G | IEEE_B;
2005-03-25 13:31:53 -05:00
} else {
if (option == 1)
printk(KERN_INFO DRV_NAME
": Detected Intel PRO/Wireless 2200BG Network "
"Connection\n");
priv->ieee->abg_true = 0;
band = IEEE80211_24GHZ_BAND;
modulation = IEEE80211_OFDM_MODULATION |
IEEE80211_CCK_MODULATION;
priv->adapter = IPW_2200BG;
priv->ieee->mode = IEEE_G | IEEE_B;
2005-03-25 13:31:53 -05:00
}
priv->ieee->freq_band = band;
priv->ieee->modulation = modulation;
2005-03-25 13:31:53 -05:00
priv->rates_mask = IEEE80211_DEFAULT_RATES_MASK;
priv->disassociate_threshold = IPW_MB_DISASSOCIATE_THRESHOLD_DEFAULT;
priv->roaming_threshold = IPW_MB_ROAMING_THRESHOLD_DEFAULT;
2005-03-25 13:31:53 -05:00
priv->rts_threshold = DEFAULT_RTS_THRESHOLD;
priv->short_retry_limit = DEFAULT_SHORT_RETRY_LIMIT;
priv->long_retry_limit = DEFAULT_LONG_RETRY_LIMIT;
2005-03-25 13:31:53 -05:00
/* If power management is turned on, default to AC mode */
priv->power_mode = IPW_POWER_AC;
priv->tx_power = IPW_TX_POWER_DEFAULT;
return old_mode == priv->ieee->iw_mode;
2005-03-25 13:31:53 -05:00
}
/*
* This file defines the Wireless Extension handlers. It does not
* define any methods of hardware manipulation and relies on the
* functions defined in ipw_main to provide the HW interaction.
*
* The exception to this is the use of the ipw_get_ordinal()
2005-03-25 13:31:53 -05:00
* function used to poll the hardware vs. making unecessary calls.
*
*/
static int ipw_wx_get_name(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
if (priv->status & STATUS_RF_KILL_MASK)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
strcpy(wrqu->name, "radio off");
else if (!(priv->status & STATUS_ASSOCIATED))
2005-03-25 13:31:53 -05:00
strcpy(wrqu->name, "unassociated");
else
2005-03-25 13:31:53 -05:00
snprintf(wrqu->name, IFNAMSIZ, "IEEE 802.11%c",
ipw_modes[priv->assoc_request.ieee_mode]);
IPW_DEBUG_WX("Name: %s\n", wrqu->name);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_set_channel(struct ipw_priv *priv, u8 channel)
{
if (channel == 0) {
IPW_DEBUG_INFO("Setting channel to ANY (0)\n");
priv->config &= ~CFG_STATIC_CHANNEL;
IPW_DEBUG_ASSOC("Attempting to associate with new "
"parameters.\n");
ipw_associate(priv);
2005-03-25 13:31:53 -05:00
return 0;
}
priv->config |= CFG_STATIC_CHANNEL;
if (priv->channel == channel) {
IPW_DEBUG_INFO("Request to set channel to current value (%d)\n",
channel);
2005-03-25 13:31:53 -05:00
return 0;
}
IPW_DEBUG_INFO("Setting channel to %i\n", (int)channel);
priv->channel = channel;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
if (priv->ieee->iw_mode == IW_MODE_MONITOR) {
int i;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->status & STATUS_SCANNING) {
IPW_DEBUG_SCAN("Scan abort triggered due to "
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
"channel change.\n");
ipw_abort_scan(priv);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
for (i = 1000; i && (priv->status & STATUS_SCANNING); i--)
udelay(10);
if (priv->status & STATUS_SCANNING)
IPW_DEBUG_SCAN("Still scanning...\n");
else
IPW_DEBUG_SCAN("Took %dms to abort current scan\n",
1000 - i);
return 0;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#endif /* CONFIG_IPW2200_MONITOR */
/* Network configuration changed -- force [re]association */
IPW_DEBUG_ASSOC("[re]association triggered due to channel change.\n");
if (!ipw_disassociate(priv))
2005-03-25 13:31:53 -05:00
ipw_associate(priv);
return 0;
}
static int ipw_wx_set_freq(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
const struct ieee80211_geo *geo = ieee80211_get_geo(priv->ieee);
2005-03-25 13:31:53 -05:00
struct iw_freq *fwrq = &wrqu->freq;
int ret = 0, i;
u8 channel, flags;
int band;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (fwrq->m == 0) {
IPW_DEBUG_WX("SET Freq/Channel -> any\n");
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ret = ipw_set_channel(priv, 0);
mutex_unlock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return ret;
}
2005-03-25 13:31:53 -05:00
/* if setting by freq convert to channel */
if (fwrq->e == 1) {
channel = ieee80211_freq_to_channel(priv->ieee, fwrq->m);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (channel == 0)
return -EINVAL;
} else
channel = fwrq->m;
if (!(band = ieee80211_is_valid_channel(priv->ieee, channel)))
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return -EINVAL;
2005-03-25 13:31:53 -05:00
if (priv->ieee->iw_mode == IW_MODE_ADHOC) {
i = ieee80211_channel_to_index(priv->ieee, channel);
if (i == -1)
return -EINVAL;
flags = (band == IEEE80211_24GHZ_BAND) ?
geo->bg[i].flags : geo->a[i].flags;
if (flags & IEEE80211_CH_PASSIVE_ONLY) {
IPW_DEBUG_WX("Invalid Ad-Hoc channel for 802.11a\n");
return -EINVAL;
2005-03-25 13:31:53 -05:00
}
}
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("SET Freq/Channel -> %d \n", fwrq->m);
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ret = ipw_set_channel(priv, channel);
mutex_unlock(&priv->mutex);
return ret;
2005-03-25 13:31:53 -05:00
}
static int ipw_wx_get_freq(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
wrqu->freq.e = 0;
/* If we are associated, trying to associate, or have a statically
* configured CHANNEL then return that; otherwise return ANY */
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (priv->config & CFG_STATIC_CHANNEL ||
priv->status & (STATUS_ASSOCIATING | STATUS_ASSOCIATED))
wrqu->freq.m = priv->channel;
else
2005-03-25 13:31:53 -05:00
wrqu->freq.m = 0;
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("GET Freq/Channel -> %d \n", priv->channel);
return 0;
}
static int ipw_wx_set_mode(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
int err = 0;
IPW_DEBUG_WX("Set MODE: %d\n", wrqu->mode);
switch (wrqu->mode) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
2005-03-25 13:31:53 -05:00
case IW_MODE_MONITOR:
#endif
case IW_MODE_ADHOC:
case IW_MODE_INFRA:
break;
case IW_MODE_AUTO:
wrqu->mode = IW_MODE_INFRA;
break;
default:
return -EINVAL;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (wrqu->mode == priv->ieee->iw_mode)
return 0;
2005-03-25 13:31:53 -05:00
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
ipw_sw_reset(priv, 0);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
if (priv->ieee->iw_mode == IW_MODE_MONITOR)
2005-03-25 13:31:53 -05:00
priv->net_dev->type = ARPHRD_ETHER;
if (wrqu->mode == IW_MODE_MONITOR)
#ifdef CONFIG_IEEE80211_RADIOTAP
priv->net_dev->type = ARPHRD_IEEE80211_RADIOTAP;
#else
2005-03-25 13:31:53 -05:00
priv->net_dev->type = ARPHRD_IEEE80211;
#endif
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#endif /* CONFIG_IPW2200_MONITOR */
/* Free the existing firmware and reset the fw_loaded
2005-03-25 13:31:53 -05:00
* flag so ipw_load() will bring in the new firmawre */
free_firmware();
2005-03-25 13:31:53 -05:00
priv->ieee->iw_mode = wrqu->mode;
queue_work(priv->workqueue, &priv->adapter_restart);
mutex_unlock(&priv->mutex);
return err;
2005-03-25 13:31:53 -05:00
}
static int ipw_wx_get_mode(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
wrqu->mode = priv->ieee->iw_mode;
IPW_DEBUG_WX("Get MODE -> %d\n", wrqu->mode);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
/* Values are in microsecond */
static const s32 timeout_duration[] = {
350000,
250000,
75000,
37000,
25000,
};
static const s32 period_duration[] = {
400000,
700000,
1000000,
1000000,
1000000
};
static int ipw_wx_get_range(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct iw_range *range = (struct iw_range *)extra;
const struct ieee80211_geo *geo = ieee80211_get_geo(priv->ieee);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
int i = 0, j;
2005-03-25 13:31:53 -05:00
wrqu->data.length = sizeof(*range);
memset(range, 0, sizeof(*range));
/* 54Mbs == ~27 Mb/s real (802.11g) */
range->throughput = 27 * 1000 * 1000;
2005-03-25 13:31:53 -05:00
range->max_qual.qual = 100;
/* TODO: Find real max RSSI and stick here */
range->max_qual.level = 0;
range->max_qual.noise = 0;
range->max_qual.updated = 7; /* Updated all three */
2005-03-25 13:31:53 -05:00
range->avg_qual.qual = 70;
/* TODO: Find real 'good' to 'bad' threshol value for RSSI */
range->avg_qual.level = 0; /* FIXME to real average level */
2005-03-25 13:31:53 -05:00
range->avg_qual.noise = 0;
range->avg_qual.updated = 7; /* Updated all three */
mutex_lock(&priv->mutex);
range->num_bitrates = min(priv->rates.num_rates, (u8) IW_MAX_BITRATES);
2005-03-25 13:31:53 -05:00
for (i = 0; i < range->num_bitrates; i++)
range->bitrate[i] = (priv->rates.supported_rates[i] & 0x7F) *
500000;
2005-03-25 13:31:53 -05:00
range->max_rts = DEFAULT_RTS_THRESHOLD;
range->min_frag = MIN_FRAG_THRESHOLD;
range->max_frag = MAX_FRAG_THRESHOLD;
range->encoding_size[0] = 5;
range->encoding_size[1] = 13;
2005-03-25 13:31:53 -05:00
range->num_encoding_sizes = 2;
range->max_encoding_tokens = WEP_KEYS;
/* Set the Wireless Extension versions */
range->we_version_compiled = WIRELESS_EXT;
range->we_version_source = 18;
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
i = 0;
if (priv->ieee->mode & (IEEE_B | IEEE_G)) {
for (j = 0; j < geo->bg_channels && i < IW_MAX_FREQUENCIES; j++) {
if ((priv->ieee->iw_mode == IW_MODE_ADHOC) &&
(geo->bg[j].flags & IEEE80211_CH_PASSIVE_ONLY))
continue;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
range->freq[i].i = geo->bg[j].channel;
range->freq[i].m = geo->bg[j].freq * 100000;
range->freq[i].e = 1;
i++;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
}
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->ieee->mode & IEEE_A) {
for (j = 0; j < geo->a_channels && i < IW_MAX_FREQUENCIES; j++) {
if ((priv->ieee->iw_mode == IW_MODE_ADHOC) &&
(geo->a[j].flags & IEEE80211_CH_PASSIVE_ONLY))
continue;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
range->freq[i].i = geo->a[j].channel;
range->freq[i].m = geo->a[j].freq * 100000;
range->freq[i].e = 1;
i++;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
range->num_channels = i;
range->num_frequency = i;
mutex_unlock(&priv->mutex);
/* Event capability (kernel + driver) */
range->event_capa[0] = (IW_EVENT_CAPA_K_0 |
IW_EVENT_CAPA_MASK(SIOCGIWTHRSPY) |
IW_EVENT_CAPA_MASK(SIOCGIWAP) |
IW_EVENT_CAPA_MASK(SIOCGIWSCAN));
range->event_capa[1] = IW_EVENT_CAPA_K_1;
2005-03-25 13:31:53 -05:00
range->enc_capa = IW_ENC_CAPA_WPA | IW_ENC_CAPA_WPA2 |
IW_ENC_CAPA_CIPHER_TKIP | IW_ENC_CAPA_CIPHER_CCMP;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("GET Range\n");
return 0;
}
static int ipw_wx_set_wap(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
static const unsigned char any[] = {
0xff, 0xff, 0xff, 0xff, 0xff, 0xff
};
static const unsigned char off[] = {
0x00, 0x00, 0x00, 0x00, 0x00, 0x00
};
if (wrqu->ap_addr.sa_family != ARPHRD_ETHER)
2005-03-25 13:31:53 -05:00
return -EINVAL;
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (!memcmp(any, wrqu->ap_addr.sa_data, ETH_ALEN) ||
!memcmp(off, wrqu->ap_addr.sa_data, ETH_ALEN)) {
/* we disable mandatory BSSID association */
IPW_DEBUG_WX("Setting AP BSSID to ANY\n");
priv->config &= ~CFG_STATIC_BSSID;
IPW_DEBUG_ASSOC("Attempting to associate with new "
"parameters.\n");
ipw_associate(priv);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
priv->config |= CFG_STATIC_BSSID;
if (!memcmp(priv->bssid, wrqu->ap_addr.sa_data, ETH_ALEN)) {
IPW_DEBUG_WX("BSSID set to current BSSID.\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
IPW_DEBUG_WX("Setting mandatory BSSID to " MAC_FMT "\n",
MAC_ARG(wrqu->ap_addr.sa_data));
memcpy(priv->bssid, wrqu->ap_addr.sa_data, ETH_ALEN);
/* Network configuration changed -- force [re]association */
IPW_DEBUG_ASSOC("[re]association triggered due to BSSID change.\n");
if (!ipw_disassociate(priv))
2005-03-25 13:31:53 -05:00
ipw_associate(priv);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_get_wap(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
/* If we are associated, trying to associate, or have a statically
* configured BSSID then return that; otherwise return ANY */
mutex_lock(&priv->mutex);
if (priv->config & CFG_STATIC_BSSID ||
2005-03-25 13:31:53 -05:00
priv->status & (STATUS_ASSOCIATED | STATUS_ASSOCIATING)) {
wrqu->ap_addr.sa_family = ARPHRD_ETHER;
memcpy(wrqu->ap_addr.sa_data, priv->bssid, ETH_ALEN);
2005-03-25 13:31:53 -05:00
} else
memset(wrqu->ap_addr.sa_data, 0, ETH_ALEN);
IPW_DEBUG_WX("Getting WAP BSSID: " MAC_FMT "\n",
MAC_ARG(wrqu->ap_addr.sa_data));
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_set_essid(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
char *essid = ""; /* ANY */
2005-03-25 13:31:53 -05:00
int length = 0;
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (wrqu->essid.flags && wrqu->essid.length) {
length = wrqu->essid.length - 1;
essid = extra;
}
if (length == 0) {
IPW_DEBUG_WX("Setting ESSID to ANY\n");
if ((priv->config & CFG_STATIC_ESSID) &&
!(priv->status & (STATUS_ASSOCIATED |
2005-03-25 13:31:53 -05:00
STATUS_ASSOCIATING))) {
IPW_DEBUG_ASSOC("Attempting to associate with new "
"parameters.\n");
priv->config &= ~CFG_STATIC_ESSID;
2005-03-25 13:31:53 -05:00
ipw_associate(priv);
}
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
length = min(length, IW_ESSID_MAX_SIZE);
priv->config |= CFG_STATIC_ESSID;
if (priv->essid_len == length && !memcmp(priv->essid, extra, length)) {
IPW_DEBUG_WX("ESSID set to current ESSID.\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
IPW_DEBUG_WX("Setting ESSID: '%s' (%d)\n", escape_essid(essid, length),
length);
priv->essid_len = length;
memcpy(priv->essid, essid, priv->essid_len);
/* Network configuration changed -- force [re]association */
IPW_DEBUG_ASSOC("[re]association triggered due to ESSID change.\n");
if (!ipw_disassociate(priv))
2005-03-25 13:31:53 -05:00
ipw_associate(priv);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_get_essid(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
/* If we are associated, trying to associate, or have a statically
* configured ESSID then return that; otherwise return ANY */
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (priv->config & CFG_STATIC_ESSID ||
priv->status & (STATUS_ASSOCIATED | STATUS_ASSOCIATING)) {
IPW_DEBUG_WX("Getting essid: '%s'\n",
2005-03-25 13:31:53 -05:00
escape_essid(priv->essid, priv->essid_len));
memcpy(extra, priv->essid, priv->essid_len);
2005-03-25 13:31:53 -05:00
wrqu->essid.length = priv->essid_len;
wrqu->essid.flags = 1; /* active */
2005-03-25 13:31:53 -05:00
} else {
IPW_DEBUG_WX("Getting essid: ANY\n");
wrqu->essid.length = 0;
wrqu->essid.flags = 0; /* active */
2005-03-25 13:31:53 -05:00
}
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_set_nick(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
IPW_DEBUG_WX("Setting nick to '%s'\n", extra);
if (wrqu->data.length > IW_ESSID_MAX_SIZE)
return -E2BIG;
mutex_lock(&priv->mutex);
wrqu->data.length = min((size_t) wrqu->data.length, sizeof(priv->nick));
2005-03-25 13:31:53 -05:00
memset(priv->nick, 0, sizeof(priv->nick));
memcpy(priv->nick, extra, wrqu->data.length);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TRACE("<<\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_get_nick(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
IPW_DEBUG_WX("Getting nick\n");
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
wrqu->data.length = strlen(priv->nick) + 1;
memcpy(extra, priv->nick, wrqu->data.length);
wrqu->data.flags = 1; /* active */
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_set_sens(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
int err = 0;
IPW_DEBUG_WX("Setting roaming threshold to %d\n", wrqu->sens.value);
IPW_DEBUG_WX("Setting disassociate threshold to %d\n", 3*wrqu->sens.value);
mutex_lock(&priv->mutex);
if (wrqu->sens.fixed == 0)
{
priv->roaming_threshold = IPW_MB_ROAMING_THRESHOLD_DEFAULT;
priv->disassociate_threshold = IPW_MB_DISASSOCIATE_THRESHOLD_DEFAULT;
goto out;
}
if ((wrqu->sens.value > IPW_MB_ROAMING_THRESHOLD_MAX) ||
(wrqu->sens.value < IPW_MB_ROAMING_THRESHOLD_MIN)) {
err = -EINVAL;
goto out;
}
priv->roaming_threshold = wrqu->sens.value;
priv->disassociate_threshold = 3*wrqu->sens.value;
out:
mutex_unlock(&priv->mutex);
return err;
}
static int ipw_wx_get_sens(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
wrqu->sens.fixed = 1;
wrqu->sens.value = priv->roaming_threshold;
mutex_unlock(&priv->mutex);
IPW_DEBUG_WX("GET roaming threshold -> %s %d \n",
wrqu->power.disabled ? "OFF" : "ON", wrqu->power.value);
return 0;
}
2005-03-25 13:31:53 -05:00
static int ipw_wx_set_rate(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
/* TODO: We should use semaphores or locks for access to priv */
struct ipw_priv *priv = ieee80211_priv(dev);
u32 target_rate = wrqu->bitrate.value;
u32 fixed, mask;
/* value = -1, fixed = 0 means auto only, so we should use all rates offered by AP */
/* value = X, fixed = 1 means only rate X */
/* value = X, fixed = 0 means all rates lower equal X */
if (target_rate == -1) {
fixed = 0;
mask = IEEE80211_DEFAULT_RATES_MASK;
/* Now we should reassociate */
goto apply;
}
mask = 0;
fixed = wrqu->bitrate.fixed;
if (target_rate == 1000000 || !fixed)
mask |= IEEE80211_CCK_RATE_1MB_MASK;
if (target_rate == 1000000)
goto apply;
if (target_rate == 2000000 || !fixed)
mask |= IEEE80211_CCK_RATE_2MB_MASK;
if (target_rate == 2000000)
goto apply;
if (target_rate == 5500000 || !fixed)
mask |= IEEE80211_CCK_RATE_5MB_MASK;
if (target_rate == 5500000)
goto apply;
if (target_rate == 6000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_6MB_MASK;
if (target_rate == 6000000)
goto apply;
if (target_rate == 9000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_9MB_MASK;
if (target_rate == 9000000)
goto apply;
if (target_rate == 11000000 || !fixed)
mask |= IEEE80211_CCK_RATE_11MB_MASK;
if (target_rate == 11000000)
goto apply;
if (target_rate == 12000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_12MB_MASK;
if (target_rate == 12000000)
goto apply;
if (target_rate == 18000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_18MB_MASK;
if (target_rate == 18000000)
goto apply;
if (target_rate == 24000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_24MB_MASK;
if (target_rate == 24000000)
goto apply;
if (target_rate == 36000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_36MB_MASK;
if (target_rate == 36000000)
goto apply;
if (target_rate == 48000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_48MB_MASK;
if (target_rate == 48000000)
goto apply;
if (target_rate == 54000000 || !fixed)
mask |= IEEE80211_OFDM_RATE_54MB_MASK;
if (target_rate == 54000000)
goto apply;
IPW_DEBUG_WX("invalid rate specified, returning error\n");
return -EINVAL;
apply:
IPW_DEBUG_WX("Setting rate mask to 0x%08X [%s]\n",
mask, fixed ? "fixed" : "sub-rates");
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (mask == IEEE80211_DEFAULT_RATES_MASK) {
priv->config &= ~CFG_FIXED_RATE;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_set_fixed_rate(priv, priv->ieee->mode);
} else
priv->config |= CFG_FIXED_RATE;
if (priv->rates_mask == mask) {
IPW_DEBUG_WX("Mask set to current mask.\n");
mutex_unlock(&priv->mutex);
return 0;
}
priv->rates_mask = mask;
/* Network configuration changed -- force [re]association */
IPW_DEBUG_ASSOC("[re]association triggered due to rates change.\n");
if (!ipw_disassociate(priv))
ipw_associate(priv);
mutex_unlock(&priv->mutex);
return 0;
2005-03-25 13:31:53 -05:00
}
static int ipw_wx_get_rate(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
wrqu->bitrate.value = priv->last_rate;
wrqu->bitrate.fixed = (priv->config & CFG_FIXED_RATE) ? 1 : 0;
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("GET Rate -> %d \n", wrqu->bitrate.value);
return 0;
}
static int ipw_wx_set_rts(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (wrqu->rts.disabled)
priv->rts_threshold = DEFAULT_RTS_THRESHOLD;
else {
if (wrqu->rts.value < MIN_RTS_THRESHOLD ||
wrqu->rts.value > MAX_RTS_THRESHOLD) {
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return -EINVAL;
}
2005-03-25 13:31:53 -05:00
priv->rts_threshold = wrqu->rts.value;
}
ipw_send_rts_threshold(priv, priv->rts_threshold);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("SET RTS Threshold -> %d \n", priv->rts_threshold);
return 0;
}
static int ipw_wx_get_rts(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
wrqu->rts.value = priv->rts_threshold;
wrqu->rts.fixed = 0; /* no auto select */
wrqu->rts.disabled = (wrqu->rts.value == DEFAULT_RTS_THRESHOLD);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("GET RTS Threshold -> %d \n", wrqu->rts.value);
return 0;
}
static int ipw_wx_set_txpow(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
int err = 0;
2005-03-25 13:31:53 -05:00
mutex_lock(&priv->mutex);
if (ipw_radio_kill_sw(priv, wrqu->power.disabled)) {
err = -EINPROGRESS;
goto out;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (!wrqu->power.fixed)
wrqu->power.value = IPW_TX_POWER_DEFAULT;
if (wrqu->power.flags != IW_TXPOW_DBM) {
err = -EINVAL;
goto out;
}
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if ((wrqu->power.value > IPW_TX_POWER_MAX) ||
(wrqu->power.value < IPW_TX_POWER_MIN)) {
err = -EINVAL;
goto out;
}
2005-03-25 13:31:53 -05:00
priv->tx_power = wrqu->power.value;
err = ipw_set_tx_power(priv);
out:
mutex_unlock(&priv->mutex);
return err;
2005-03-25 13:31:53 -05:00
}
static int ipw_wx_get_txpow(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
wrqu->power.value = priv->tx_power;
wrqu->power.fixed = 1;
wrqu->power.flags = IW_TXPOW_DBM;
wrqu->power.disabled = (priv->status & STATUS_RF_KILL_MASK) ? 1 : 0;
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("GET TX Power -> %s %d \n",
wrqu->power.disabled ? "OFF" : "ON", wrqu->power.value);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_set_frag(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (wrqu->frag.disabled)
priv->ieee->fts = DEFAULT_FTS;
else {
if (wrqu->frag.value < MIN_FRAG_THRESHOLD ||
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
wrqu->frag.value > MAX_FRAG_THRESHOLD) {
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return -EINVAL;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
2005-03-25 13:31:53 -05:00
priv->ieee->fts = wrqu->frag.value & ~0x1;
}
ipw_send_frag_threshold(priv, wrqu->frag.value);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("SET Frag Threshold -> %d \n", wrqu->frag.value);
return 0;
}
static int ipw_wx_get_frag(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
wrqu->frag.value = priv->ieee->fts;
wrqu->frag.fixed = 0; /* no auto select */
wrqu->frag.disabled = (wrqu->frag.value == DEFAULT_FTS);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("GET Frag Threshold -> %d \n", wrqu->frag.value);
return 0;
}
static int ipw_wx_set_retry(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
if (wrqu->retry.flags & IW_RETRY_LIFETIME || wrqu->retry.disabled)
return -EINVAL;
if (!(wrqu->retry.flags & IW_RETRY_LIMIT))
return 0;
if (wrqu->retry.value < 0 || wrqu->retry.value > 255)
return -EINVAL;
mutex_lock(&priv->mutex);
if (wrqu->retry.flags & IW_RETRY_MIN)
priv->short_retry_limit = (u8) wrqu->retry.value;
else if (wrqu->retry.flags & IW_RETRY_MAX)
priv->long_retry_limit = (u8) wrqu->retry.value;
else {
priv->short_retry_limit = (u8) wrqu->retry.value;
priv->long_retry_limit = (u8) wrqu->retry.value;
}
ipw_send_retry_limit(priv, priv->short_retry_limit,
priv->long_retry_limit);
mutex_unlock(&priv->mutex);
IPW_DEBUG_WX("SET retry limit -> short:%d long:%d\n",
priv->short_retry_limit, priv->long_retry_limit);
return 0;
2005-03-25 13:31:53 -05:00
}
static int ipw_wx_get_retry(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
wrqu->retry.disabled = 0;
if ((wrqu->retry.flags & IW_RETRY_TYPE) == IW_RETRY_LIFETIME) {
mutex_unlock(&priv->mutex);
return -EINVAL;
}
if (wrqu->retry.flags & IW_RETRY_MAX) {
wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
wrqu->retry.value = priv->long_retry_limit;
} else if (wrqu->retry.flags & IW_RETRY_MIN) {
wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_MIN;
wrqu->retry.value = priv->short_retry_limit;
} else {
wrqu->retry.flags = IW_RETRY_LIMIT;
wrqu->retry.value = priv->short_retry_limit;
}
mutex_unlock(&priv->mutex);
IPW_DEBUG_WX("GET retry -> %d \n", wrqu->retry.value);
return 0;
}
static int ipw_request_direct_scan(struct ipw_priv *priv, char *essid,
int essid_len)
{
struct ipw_scan_request_ext scan;
int err = 0, scan_type;
if (!(priv->status & STATUS_INIT) ||
(priv->status & STATUS_EXIT_PENDING))
return 0;
mutex_lock(&priv->mutex);
if (priv->status & STATUS_RF_KILL_MASK) {
IPW_DEBUG_HC("Aborting scan due to RF kill activation\n");
priv->status |= STATUS_SCAN_PENDING;
goto done;
}
IPW_DEBUG_HC("starting request direct scan!\n");
if (priv->status & (STATUS_SCANNING | STATUS_SCAN_ABORTING)) {
/* We should not sleep here; otherwise we will block most
* of the system (for instance, we hold rtnl_lock when we
* get here).
*/
err = -EAGAIN;
goto done;
}
memset(&scan, 0, sizeof(scan));
if (priv->config & CFG_SPEED_SCAN)
scan.dwell_time[IPW_SCAN_ACTIVE_BROADCAST_SCAN] =
cpu_to_le16(30);
else
scan.dwell_time[IPW_SCAN_ACTIVE_BROADCAST_SCAN] =
cpu_to_le16(20);
scan.dwell_time[IPW_SCAN_ACTIVE_BROADCAST_AND_DIRECT_SCAN] =
cpu_to_le16(20);
scan.dwell_time[IPW_SCAN_PASSIVE_FULL_DWELL_SCAN] = cpu_to_le16(120);
scan.dwell_time[IPW_SCAN_ACTIVE_DIRECT_SCAN] = cpu_to_le16(20);
scan.full_scan_index = cpu_to_le32(ieee80211_get_scans(priv->ieee));
err = ipw_send_ssid(priv, essid, essid_len);
if (err) {
IPW_DEBUG_HC("Attempt to send SSID command failed\n");
goto done;
}
scan_type = IPW_SCAN_ACTIVE_BROADCAST_AND_DIRECT_SCAN;
ipw_add_scan_channels(priv, &scan, scan_type);
err = ipw_send_scan_request_ext(priv, &scan);
if (err) {
IPW_DEBUG_HC("Sending scan command failed: %08X\n", err);
goto done;
}
priv->status |= STATUS_SCANNING;
done:
mutex_unlock(&priv->mutex);
return err;
2005-03-25 13:31:53 -05:00
}
static int ipw_wx_set_scan(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct iw_scan_req *req = NULL;
if (wrqu->data.length
&& wrqu->data.length == sizeof(struct iw_scan_req)) {
req = (struct iw_scan_req *)extra;
if (wrqu->data.flags & IW_SCAN_THIS_ESSID) {
ipw_request_direct_scan(priv, req->essid,
req->essid_len);
return 0;
}
}
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("Start scan\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
queue_work(priv->workqueue, &priv->request_scan);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_get_scan(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
return ieee80211_wx_get_scan(priv->ieee, info, wrqu, extra);
}
static int ipw_wx_set_encode(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *key)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
int ret;
u32 cap = priv->capability;
mutex_lock(&priv->mutex);
ret = ieee80211_wx_set_encode(priv->ieee, info, wrqu, key);
/* In IBSS mode, we need to notify the firmware to update
* the beacon info after we changed the capability. */
if (cap != priv->capability &&
priv->ieee->iw_mode == IW_MODE_ADHOC &&
priv->status & STATUS_ASSOCIATED)
ipw_disassociate(priv);
mutex_unlock(&priv->mutex);
return ret;
2005-03-25 13:31:53 -05:00
}
static int ipw_wx_get_encode(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *key)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
return ieee80211_wx_get_encode(priv->ieee, info, wrqu, key);
}
static int ipw_wx_set_power(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
int err;
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (wrqu->power.disabled) {
priv->power_mode = IPW_POWER_LEVEL(priv->power_mode);
err = ipw_send_power_mode(priv, IPW_POWER_MODE_CAM);
if (err) {
IPW_DEBUG_WX("failed setting power mode.\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return err;
}
IPW_DEBUG_WX("SET Power Management Mode -> off\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
2005-03-25 13:31:53 -05:00
switch (wrqu->power.flags & IW_POWER_MODE) {
case IW_POWER_ON: /* If not specified */
case IW_POWER_MODE: /* If set all mask */
case IW_POWER_ALL_R: /* If explicitely state all */
2005-03-25 13:31:53 -05:00
break;
default: /* Otherwise we don't support it */
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("SET PM Mode: %X not supported.\n",
wrqu->power.flags);
mutex_unlock(&priv->mutex);
return -EOPNOTSUPP;
2005-03-25 13:31:53 -05:00
}
2005-03-25 13:31:53 -05:00
/* If the user hasn't specified a power management mode yet, default
* to BATTERY */
if (IPW_POWER_LEVEL(priv->power_mode) == IPW_POWER_AC)
2005-03-25 13:31:53 -05:00
priv->power_mode = IPW_POWER_ENABLED | IPW_POWER_BATTERY;
else
2005-03-25 13:31:53 -05:00
priv->power_mode = IPW_POWER_ENABLED | priv->power_mode;
err = ipw_send_power_mode(priv, IPW_POWER_LEVEL(priv->power_mode));
if (err) {
IPW_DEBUG_WX("failed setting power mode.\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return err;
}
IPW_DEBUG_WX("SET Power Management Mode -> 0x%02X\n", priv->power_mode);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_get_power(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
if (!(priv->power_mode & IPW_POWER_ENABLED))
2005-03-25 13:31:53 -05:00
wrqu->power.disabled = 1;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
else
2005-03-25 13:31:53 -05:00
wrqu->power.disabled = 0;
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("GET Power Management Mode -> %02X\n", priv->power_mode);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_set_powermode(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
int mode = *(int *)extra;
int err;
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if ((mode < 1) || (mode > IPW_POWER_LIMIT)) {
mode = IPW_POWER_AC;
priv->power_mode = mode;
} else {
priv->power_mode = IPW_POWER_ENABLED | mode;
}
2005-03-25 13:31:53 -05:00
if (priv->power_mode != mode) {
err = ipw_send_power_mode(priv, mode);
2005-03-25 13:31:53 -05:00
if (err) {
IPW_DEBUG_WX("failed setting power mode.\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return err;
}
}
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
#define MAX_WX_STRING 80
static int ipw_wx_get_powermode(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
int level = IPW_POWER_LEVEL(priv->power_mode);
char *p = extra;
p += snprintf(p, MAX_WX_STRING, "Power save level: %d ", level);
switch (level) {
case IPW_POWER_AC:
p += snprintf(p, MAX_WX_STRING - (p - extra), "(AC)");
break;
case IPW_POWER_BATTERY:
p += snprintf(p, MAX_WX_STRING - (p - extra), "(BATTERY)");
break;
default:
p += snprintf(p, MAX_WX_STRING - (p - extra),
"(Timeout %dms, Period %dms)",
2005-03-25 13:31:53 -05:00
timeout_duration[level - 1] / 1000,
period_duration[level - 1] / 1000);
}
if (!(priv->power_mode & IPW_POWER_ENABLED))
p += snprintf(p, MAX_WX_STRING - (p - extra), " OFF");
2005-03-25 13:31:53 -05:00
wrqu->data.length = p - extra + 1;
return 0;
}
static int ipw_wx_set_wireless_mode(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
2005-03-25 13:31:53 -05:00
int mode = *(int *)extra;
u8 band = 0, modulation = 0;
if (mode == 0 || mode & ~IEEE_MODE_MASK) {
IPW_WARNING("Attempt to set invalid wireless mode: %d\n", mode);
2005-03-25 13:31:53 -05:00
return -EINVAL;
}
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (priv->adapter == IPW_2915ABG) {
priv->ieee->abg_true = 1;
2005-03-25 13:31:53 -05:00
if (mode & IEEE_A) {
band |= IEEE80211_52GHZ_BAND;
modulation |= IEEE80211_OFDM_MODULATION;
} else
priv->ieee->abg_true = 0;
2005-03-25 13:31:53 -05:00
} else {
if (mode & IEEE_A) {
IPW_WARNING("Attempt to set 2200BG into "
"802.11a mode\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return -EINVAL;
}
priv->ieee->abg_true = 0;
2005-03-25 13:31:53 -05:00
}
if (mode & IEEE_B) {
band |= IEEE80211_24GHZ_BAND;
modulation |= IEEE80211_CCK_MODULATION;
} else
priv->ieee->abg_true = 0;
2005-03-25 13:31:53 -05:00
if (mode & IEEE_G) {
band |= IEEE80211_24GHZ_BAND;
modulation |= IEEE80211_OFDM_MODULATION;
} else
priv->ieee->abg_true = 0;
2005-03-25 13:31:53 -05:00
priv->ieee->mode = mode;
priv->ieee->freq_band = band;
priv->ieee->modulation = modulation;
init_supported_rates(priv, &priv->rates);
2005-03-25 13:31:53 -05:00
/* Network configuration changed -- force [re]association */
IPW_DEBUG_ASSOC("[re]association triggered due to mode change.\n");
if (!ipw_disassociate(priv)) {
2005-03-25 13:31:53 -05:00
ipw_send_supported_rates(priv, &priv->rates);
ipw_associate(priv);
}
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
/* Update the band LEDs */
ipw_led_band_on(priv);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("PRIV SET MODE: %c%c%c\n",
2005-03-25 13:31:53 -05:00
mode & IEEE_A ? 'a' : '.',
mode & IEEE_B ? 'b' : '.', mode & IEEE_G ? 'g' : '.');
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_wx_get_wireless_mode(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
switch (priv->ieee->mode) {
case IEEE_A:
2005-03-25 13:31:53 -05:00
strncpy(extra, "802.11a (1)", MAX_WX_STRING);
break;
case IEEE_B:
strncpy(extra, "802.11b (2)", MAX_WX_STRING);
break;
case IEEE_A | IEEE_B:
strncpy(extra, "802.11ab (3)", MAX_WX_STRING);
break;
case IEEE_G:
strncpy(extra, "802.11g (4)", MAX_WX_STRING);
break;
case IEEE_A | IEEE_G:
strncpy(extra, "802.11ag (5)", MAX_WX_STRING);
break;
case IEEE_B | IEEE_G:
strncpy(extra, "802.11bg (6)", MAX_WX_STRING);
break;
case IEEE_A | IEEE_B | IEEE_G:
strncpy(extra, "802.11abg (7)", MAX_WX_STRING);
break;
default:
strncpy(extra, "unknown", MAX_WX_STRING);
2005-03-25 13:31:53 -05:00
break;
}
2005-03-25 13:31:53 -05:00
IPW_DEBUG_WX("PRIV GET MODE: %s\n", extra);
wrqu->data.length = strlen(extra) + 1;
mutex_unlock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return 0;
}
static int ipw_wx_set_preamble(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
int mode = *(int *)extra;
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* Switching from SHORT -> LONG requires a disassociation */
if (mode == 1) {
if (!(priv->config & CFG_PREAMBLE_LONG)) {
priv->config |= CFG_PREAMBLE_LONG;
/* Network configuration changed -- force [re]association */
IPW_DEBUG_ASSOC
("[re]association triggered due to preamble change.\n");
if (!ipw_disassociate(priv))
ipw_associate(priv);
}
goto done;
}
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (mode == 0) {
priv->config &= ~CFG_PREAMBLE_LONG;
goto done;
}
mutex_unlock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return -EINVAL;
done:
mutex_unlock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return 0;
}
static int ipw_wx_get_preamble(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->config & CFG_PREAMBLE_LONG)
snprintf(wrqu->name, IFNAMSIZ, "long (1)");
else
snprintf(wrqu->name, IFNAMSIZ, "auto (0)");
mutex_unlock(&priv->mutex);
return 0;
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
static int ipw_wx_set_monitor(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
int *parms = (int *)extra;
int enable = (parms[0] > 0);
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_DEBUG_WX("SET MONITOR: %d %d\n", enable, parms[1]);
2005-03-25 13:31:53 -05:00
if (enable) {
if (priv->ieee->iw_mode != IW_MODE_MONITOR) {
#ifdef CONFIG_IEEE80211_RADIOTAP
priv->net_dev->type = ARPHRD_IEEE80211_RADIOTAP;
#else
2005-03-25 13:31:53 -05:00
priv->net_dev->type = ARPHRD_IEEE80211;
#endif
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
queue_work(priv->workqueue, &priv->adapter_restart);
2005-03-25 13:31:53 -05:00
}
2005-03-25 13:31:53 -05:00
ipw_set_channel(priv, parms[1]);
} else {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->ieee->iw_mode != IW_MODE_MONITOR) {
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
}
2005-03-25 13:31:53 -05:00
priv->net_dev->type = ARPHRD_ETHER;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
queue_work(priv->workqueue, &priv->adapter_restart);
2005-03-25 13:31:53 -05:00
}
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#endif // CONFIG_IPW2200_MONITOR
static int ipw_wx_reset(struct net_device *dev,
struct iw_request_info *info,
2005-03-25 13:31:53 -05:00
union iwreq_data *wrqu, char *extra)
{
2005-03-25 13:31:53 -05:00
struct ipw_priv *priv = ieee80211_priv(dev);
IPW_DEBUG_WX("RESET\n");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
queue_work(priv->workqueue, &priv->adapter_restart);
return 0;
}
static int ipw_wx_sw_reset(struct net_device *dev,
struct iw_request_info *info,
union iwreq_data *wrqu, char *extra)
{
struct ipw_priv *priv = ieee80211_priv(dev);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
union iwreq_data wrqu_sec = {
.encoding = {
.flags = IW_ENCODE_DISABLED,
},
};
int ret;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_DEBUG_WX("SW_RESET\n");
mutex_lock(&priv->mutex);
ret = ipw_sw_reset(priv, 2);
if (!ret) {
free_firmware();
ipw_adapter_restart(priv);
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
/* The SW reset bit might have been toggled on by the 'disable'
* module parameter, so take appropriate action */
ipw_radio_kill_sw(priv, priv->status & STATUS_RF_KILL_SW);
mutex_unlock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ieee80211_wx_set_encode(priv->ieee, info, &wrqu_sec, NULL);
mutex_lock(&priv->mutex);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (!(priv->status & STATUS_RF_KILL_MASK)) {
/* Configuration likely changed -- force [re]association */
IPW_DEBUG_ASSOC("[re]association triggered due to sw "
"reset.\n");
if (!ipw_disassociate(priv))
ipw_associate(priv);
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
/* Rebase the WE IOCTLs to zero for the handler array */
#define IW_IOCTL(x) [(x)-SIOCSIWCOMMIT]
static iw_handler ipw_wx_handlers[] = {
IW_IOCTL(SIOCGIWNAME) = ipw_wx_get_name,
IW_IOCTL(SIOCSIWFREQ) = ipw_wx_set_freq,
IW_IOCTL(SIOCGIWFREQ) = ipw_wx_get_freq,
IW_IOCTL(SIOCSIWMODE) = ipw_wx_set_mode,
IW_IOCTL(SIOCGIWMODE) = ipw_wx_get_mode,
IW_IOCTL(SIOCSIWSENS) = ipw_wx_set_sens,
IW_IOCTL(SIOCGIWSENS) = ipw_wx_get_sens,
IW_IOCTL(SIOCGIWRANGE) = ipw_wx_get_range,
IW_IOCTL(SIOCSIWAP) = ipw_wx_set_wap,
IW_IOCTL(SIOCGIWAP) = ipw_wx_get_wap,
IW_IOCTL(SIOCSIWSCAN) = ipw_wx_set_scan,
IW_IOCTL(SIOCGIWSCAN) = ipw_wx_get_scan,
IW_IOCTL(SIOCSIWESSID) = ipw_wx_set_essid,
IW_IOCTL(SIOCGIWESSID) = ipw_wx_get_essid,
IW_IOCTL(SIOCSIWNICKN) = ipw_wx_set_nick,
IW_IOCTL(SIOCGIWNICKN) = ipw_wx_get_nick,
IW_IOCTL(SIOCSIWRATE) = ipw_wx_set_rate,
IW_IOCTL(SIOCGIWRATE) = ipw_wx_get_rate,
IW_IOCTL(SIOCSIWRTS) = ipw_wx_set_rts,
IW_IOCTL(SIOCGIWRTS) = ipw_wx_get_rts,
IW_IOCTL(SIOCSIWFRAG) = ipw_wx_set_frag,
IW_IOCTL(SIOCGIWFRAG) = ipw_wx_get_frag,
IW_IOCTL(SIOCSIWTXPOW) = ipw_wx_set_txpow,
IW_IOCTL(SIOCGIWTXPOW) = ipw_wx_get_txpow,
IW_IOCTL(SIOCSIWRETRY) = ipw_wx_set_retry,
IW_IOCTL(SIOCGIWRETRY) = ipw_wx_get_retry,
IW_IOCTL(SIOCSIWENCODE) = ipw_wx_set_encode,
IW_IOCTL(SIOCGIWENCODE) = ipw_wx_get_encode,
IW_IOCTL(SIOCSIWPOWER) = ipw_wx_set_power,
IW_IOCTL(SIOCGIWPOWER) = ipw_wx_get_power,
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
IW_IOCTL(SIOCSIWSPY) = iw_handler_set_spy,
IW_IOCTL(SIOCGIWSPY) = iw_handler_get_spy,
IW_IOCTL(SIOCSIWTHRSPY) = iw_handler_set_thrspy,
IW_IOCTL(SIOCGIWTHRSPY) = iw_handler_get_thrspy,
IW_IOCTL(SIOCSIWGENIE) = ipw_wx_set_genie,
IW_IOCTL(SIOCGIWGENIE) = ipw_wx_get_genie,
IW_IOCTL(SIOCSIWMLME) = ipw_wx_set_mlme,
IW_IOCTL(SIOCSIWAUTH) = ipw_wx_set_auth,
IW_IOCTL(SIOCGIWAUTH) = ipw_wx_get_auth,
IW_IOCTL(SIOCSIWENCODEEXT) = ipw_wx_set_encodeext,
IW_IOCTL(SIOCGIWENCODEEXT) = ipw_wx_get_encodeext,
2005-03-25 13:31:53 -05:00
};
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
enum {
IPW_PRIV_SET_POWER = SIOCIWFIRSTPRIV,
IPW_PRIV_GET_POWER,
IPW_PRIV_SET_MODE,
IPW_PRIV_GET_MODE,
IPW_PRIV_SET_PREAMBLE,
IPW_PRIV_GET_PREAMBLE,
IPW_PRIV_RESET,
IPW_PRIV_SW_RESET,
#ifdef CONFIG_IPW2200_MONITOR
IPW_PRIV_SET_MONITOR,
#endif
};
2005-03-25 13:31:53 -05:00
static struct iw_priv_args ipw_priv_args[] = {
2005-03-25 13:31:53 -05:00
{
.cmd = IPW_PRIV_SET_POWER,
.set_args = IW_PRIV_TYPE_INT | IW_PRIV_SIZE_FIXED | 1,
.name = "set_power"},
2005-03-25 13:31:53 -05:00
{
.cmd = IPW_PRIV_GET_POWER,
.get_args = IW_PRIV_TYPE_CHAR | IW_PRIV_SIZE_FIXED | MAX_WX_STRING,
.name = "get_power"},
2005-03-25 13:31:53 -05:00
{
.cmd = IPW_PRIV_SET_MODE,
.set_args = IW_PRIV_TYPE_INT | IW_PRIV_SIZE_FIXED | 1,
.name = "set_mode"},
2005-03-25 13:31:53 -05:00
{
.cmd = IPW_PRIV_GET_MODE,
.get_args = IW_PRIV_TYPE_CHAR | IW_PRIV_SIZE_FIXED | MAX_WX_STRING,
.name = "get_mode"},
2005-03-25 13:31:53 -05:00
{
.cmd = IPW_PRIV_SET_PREAMBLE,
.set_args = IW_PRIV_TYPE_INT | IW_PRIV_SIZE_FIXED | 1,
.name = "set_preamble"},
{
.cmd = IPW_PRIV_GET_PREAMBLE,
.get_args = IW_PRIV_TYPE_CHAR | IW_PRIV_SIZE_FIXED | IFNAMSIZ,
.name = "get_preamble"},
2005-03-25 13:31:53 -05:00
{
IPW_PRIV_RESET,
IW_PRIV_TYPE_INT | IW_PRIV_SIZE_FIXED | 0, 0, "reset"},
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
{
IPW_PRIV_SW_RESET,
IW_PRIV_TYPE_INT | IW_PRIV_SIZE_FIXED | 0, 0, "sw_reset"},
#ifdef CONFIG_IPW2200_MONITOR
{
IPW_PRIV_SET_MONITOR,
IW_PRIV_TYPE_INT | IW_PRIV_SIZE_FIXED | 2, 0, "monitor"},
#endif /* CONFIG_IPW2200_MONITOR */
2005-03-25 13:31:53 -05:00
};
static iw_handler ipw_priv_handler[] = {
ipw_wx_set_powermode,
ipw_wx_get_powermode,
ipw_wx_set_wireless_mode,
ipw_wx_get_wireless_mode,
ipw_wx_set_preamble,
ipw_wx_get_preamble,
ipw_wx_reset,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_wx_sw_reset,
#ifdef CONFIG_IPW2200_MONITOR
ipw_wx_set_monitor,
2005-03-25 13:31:53 -05:00
#endif
};
static struct iw_handler_def ipw_wx_handler_def = {
.standard = ipw_wx_handlers,
.num_standard = ARRAY_SIZE(ipw_wx_handlers),
.num_private = ARRAY_SIZE(ipw_priv_handler),
.num_private_args = ARRAY_SIZE(ipw_priv_args),
.private = ipw_priv_handler,
.private_args = ipw_priv_args,
.get_wireless_stats = ipw_get_wireless_stats,
2005-03-25 13:31:53 -05:00
};
/*
* Get wireless statistics.
* Called by /proc/net/wireless
* Also called by SIOCGIWSTATS
*/
static struct iw_statistics *ipw_get_wireless_stats(struct net_device *dev)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct iw_statistics *wstats;
2005-03-25 13:31:53 -05:00
wstats = &priv->wstats;
/* if hw is disabled, then ipw_get_ordinal() can't be called.
* netdev->get_wireless_stats seems to be called before fw is
2005-03-25 13:31:53 -05:00
* initialized. STATUS_ASSOCIATED will only be set if the hw is up
* and associated; if not associcated, the values are all meaningless
* anyway, so set them all to NULL and INVALID */
if (!(priv->status & STATUS_ASSOCIATED)) {
wstats->miss.beacon = 0;
wstats->discard.retries = 0;
wstats->qual.qual = 0;
wstats->qual.level = 0;
wstats->qual.noise = 0;
wstats->qual.updated = 7;
wstats->qual.updated |= IW_QUAL_NOISE_INVALID |
IW_QUAL_QUAL_INVALID | IW_QUAL_LEVEL_INVALID;
2005-03-25 13:31:53 -05:00
return wstats;
}
2005-03-25 13:31:53 -05:00
wstats->qual.qual = priv->quality;
wstats->qual.level = priv->exp_avg_rssi;
wstats->qual.noise = priv->exp_avg_noise;
2005-03-25 13:31:53 -05:00
wstats->qual.updated = IW_QUAL_QUAL_UPDATED | IW_QUAL_LEVEL_UPDATED |
IW_QUAL_NOISE_UPDATED | IW_QUAL_DBM;
2005-03-25 13:31:53 -05:00
wstats->miss.beacon = average_value(&priv->average_missed_beacons);
wstats->discard.retries = priv->last_tx_failures;
wstats->discard.code = priv->ieee->ieee_stats.rx_discards_undecryptable;
2005-03-25 13:31:53 -05:00
/* if (ipw_get_ordinal(priv, IPW_ORD_STAT_TX_RETRY, &tx_retry, &len))
goto fail_get_ordinal;
wstats->discard.retries += tx_retry; */
2005-03-25 13:31:53 -05:00
return wstats;
}
/* net device stuff */
static void init_sys_config(struct ipw_sys_config *sys_config)
2005-03-25 13:31:53 -05:00
{
memset(sys_config, 0, sizeof(struct ipw_sys_config));
sys_config->bt_coexistence = 0;
2005-03-25 13:31:53 -05:00
sys_config->answer_broadcast_ssid_probe = 0;
sys_config->accept_all_data_frames = 0;
sys_config->accept_non_directed_frames = 1;
sys_config->exclude_unicast_unencrypted = 0;
sys_config->disable_unicast_decryption = 1;
sys_config->exclude_multicast_unencrypted = 0;
sys_config->disable_multicast_decryption = 1;
if (antenna < CFG_SYS_ANTENNA_BOTH || antenna > CFG_SYS_ANTENNA_B)
antenna = CFG_SYS_ANTENNA_BOTH;
sys_config->antenna_diversity = antenna;
sys_config->pass_crc_to_host = 0; /* TODO: See if 1 gives us FCS */
2005-03-25 13:31:53 -05:00
sys_config->dot11g_auto_detection = 0;
sys_config->enable_cts_to_self = 0;
2005-03-25 13:31:53 -05:00
sys_config->bt_coexist_collision_thr = 0;
sys_config->pass_noise_stats_to_host = 1; //1 -- fix for 256
sys_config->silence_threshold = 0x1e;
2005-03-25 13:31:53 -05:00
}
static int ipw_net_open(struct net_device *dev)
{
struct ipw_priv *priv = ieee80211_priv(dev);
IPW_DEBUG_INFO("dev->open\n");
/* we should be verifying the device is ready to be opened */
mutex_lock(&priv->mutex);
if (!(priv->status & STATUS_RF_KILL_MASK) &&
(priv->status & STATUS_ASSOCIATED))
2005-03-25 13:31:53 -05:00
netif_start_queue(dev);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_net_stop(struct net_device *dev)
{
IPW_DEBUG_INFO("dev->close\n");
netif_stop_queue(dev);
return 0;
}
/*
todo:
modify to send one tfd per fragment instead of using chunking. otherwise
we need to heavily modify the ieee80211_skb_to_txb.
*/
static int ipw_tx_skb(struct ipw_priv *priv, struct ieee80211_txb *txb,
int pri)
2005-03-25 13:31:53 -05:00
{
struct ieee80211_hdr_3addrqos *hdr = (struct ieee80211_hdr_3addrqos *)
txb->fragments[0]->data;
2005-03-25 13:31:53 -05:00
int i = 0;
struct tfd_frame *tfd;
#ifdef CONFIG_IPW2200_QOS
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
int tx_id = ipw_get_tx_queue_number(priv, pri);
struct clx2_tx_queue *txq = &priv->txq[tx_id];
#else
2005-03-25 13:31:53 -05:00
struct clx2_tx_queue *txq = &priv->txq[0];
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#endif
2005-03-25 13:31:53 -05:00
struct clx2_queue *q = &txq->q;
u8 id, hdr_len, unicast;
u16 remaining_bytes;
int fc;
2005-03-25 13:31:53 -05:00
hdr_len = ieee80211_get_hdrlen(le16_to_cpu(hdr->frame_ctl));
2005-03-25 13:31:53 -05:00
switch (priv->ieee->iw_mode) {
case IW_MODE_ADHOC:
unicast = !is_multicast_ether_addr(hdr->addr1);
2005-03-25 13:31:53 -05:00
id = ipw_find_station(priv, hdr->addr1);
if (id == IPW_INVALID_STATION) {
id = ipw_add_station(priv, hdr->addr1);
if (id == IPW_INVALID_STATION) {
IPW_WARNING("Attempt to send data to "
"invalid cell: " MAC_FMT "\n",
2005-03-25 13:31:53 -05:00
MAC_ARG(hdr->addr1));
goto drop;
}
}
break;
case IW_MODE_INFRA:
default:
unicast = !is_multicast_ether_addr(hdr->addr3);
2005-03-25 13:31:53 -05:00
id = 0;
break;
}
tfd = &txq->bd[q->first_empty];
txq->txb[q->first_empty] = txb;
memset(tfd, 0, sizeof(*tfd));
tfd->u.data.station_number = id;
tfd->control_flags.message_type = TX_FRAME_TYPE;
tfd->control_flags.control_bits = TFD_NEED_IRQ_MASK;
tfd->u.data.cmd_id = DINO_CMD_TX;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
tfd->u.data.len = cpu_to_le16(txb->payload_size);
2005-03-25 13:31:53 -05:00
remaining_bytes = txb->payload_size;
2005-03-25 13:31:53 -05:00
if (priv->assoc_request.ieee_mode == IPW_B_MODE)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
tfd->u.data.tx_flags_ext |= DCT_FLAG_EXT_MODE_CCK;
2005-03-25 13:31:53 -05:00
else
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
tfd->u.data.tx_flags_ext |= DCT_FLAG_EXT_MODE_OFDM;
2005-03-25 13:31:53 -05:00
if (priv->assoc_request.preamble_length == DCT_FLAG_SHORT_PREAMBLE)
tfd->u.data.tx_flags |= DCT_FLAG_SHORT_PREAMBLE;
2005-03-25 13:31:53 -05:00
fc = le16_to_cpu(hdr->frame_ctl);
hdr->frame_ctl = cpu_to_le16(fc & ~IEEE80211_FCTL_MOREFRAGS);
2005-03-25 13:31:53 -05:00
memcpy(&tfd->u.data.tfd.tfd_24.mchdr, hdr, hdr_len);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (likely(unicast))
tfd->u.data.tx_flags |= DCT_FLAG_ACK_REQD;
if (txb->encrypted && !priv->ieee->host_encrypt) {
switch (priv->ieee->sec.level) {
case SEC_LEVEL_3:
tfd->u.data.tfd.tfd_24.mchdr.frame_ctl |=
IEEE80211_FCTL_PROTECTED;
/* XXX: ACK flag must be set for CCMP even if it
* is a multicast/broadcast packet, because CCMP
* group communication encrypted by GTK is
* actually done by the AP. */
if (!unicast)
tfd->u.data.tx_flags |= DCT_FLAG_ACK_REQD;
tfd->u.data.tx_flags &= ~DCT_FLAG_NO_WEP;
tfd->u.data.tx_flags_ext |= DCT_FLAG_EXT_SECURITY_CCM;
tfd->u.data.key_index = 0;
tfd->u.data.key_index |= DCT_WEP_INDEX_USE_IMMEDIATE;
break;
case SEC_LEVEL_2:
tfd->u.data.tfd.tfd_24.mchdr.frame_ctl |=
IEEE80211_FCTL_PROTECTED;
tfd->u.data.tx_flags &= ~DCT_FLAG_NO_WEP;
tfd->u.data.tx_flags_ext |= DCT_FLAG_EXT_SECURITY_TKIP;
tfd->u.data.key_index = DCT_WEP_INDEX_USE_IMMEDIATE;
break;
case SEC_LEVEL_1:
tfd->u.data.tfd.tfd_24.mchdr.frame_ctl |=
IEEE80211_FCTL_PROTECTED;
tfd->u.data.key_index = priv->ieee->tx_keyidx;
if (priv->ieee->sec.key_sizes[priv->ieee->tx_keyidx] <=
40)
tfd->u.data.key_index |= DCT_WEP_KEY_64Bit;
else
tfd->u.data.key_index |= DCT_WEP_KEY_128Bit;
break;
case SEC_LEVEL_0:
break;
default:
printk(KERN_ERR "Unknow security level %d\n",
priv->ieee->sec.level);
break;
}
} else
/* No hardware encryption */
tfd->u.data.tx_flags |= DCT_FLAG_NO_WEP;
#ifdef CONFIG_IPW2200_QOS
if (fc & IEEE80211_STYPE_QOS_DATA)
ipw_qos_set_tx_queue_command(priv, pri, &(tfd->u.data));
#endif /* CONFIG_IPW2200_QOS */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
2005-03-25 13:31:53 -05:00
/* payload */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
tfd->u.data.num_chunks = cpu_to_le32(min((u8) (NUM_TFD_CHUNKS - 2),
txb->nr_frags));
IPW_DEBUG_FRAG("%i fragments being sent as %i chunks.\n",
txb->nr_frags, le32_to_cpu(tfd->u.data.num_chunks));
for (i = 0; i < le32_to_cpu(tfd->u.data.num_chunks); i++) {
IPW_DEBUG_FRAG("Adding fragment %i of %i (%d bytes).\n",
i, le32_to_cpu(tfd->u.data.num_chunks),
txb->fragments[i]->len - hdr_len);
IPW_DEBUG_TX("Dumping TX packet frag %i of %i (%d bytes):\n",
2005-03-25 13:31:53 -05:00
i, tfd->u.data.num_chunks,
txb->fragments[i]->len - hdr_len);
printk_buf(IPW_DL_TX, txb->fragments[i]->data + hdr_len,
2005-03-25 13:31:53 -05:00
txb->fragments[i]->len - hdr_len);
tfd->u.data.chunk_ptr[i] =
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
cpu_to_le32(pci_map_single
(priv->pci_dev,
txb->fragments[i]->data + hdr_len,
txb->fragments[i]->len - hdr_len,
PCI_DMA_TODEVICE));
tfd->u.data.chunk_len[i] =
cpu_to_le16(txb->fragments[i]->len - hdr_len);
2005-03-25 13:31:53 -05:00
}
if (i != txb->nr_frags) {
struct sk_buff *skb;
u16 remaining_bytes = 0;
int j;
for (j = i; j < txb->nr_frags; j++)
remaining_bytes += txb->fragments[j]->len - hdr_len;
printk(KERN_INFO "Trying to reallocate for %d bytes\n",
remaining_bytes);
skb = alloc_skb(remaining_bytes, GFP_ATOMIC);
if (skb != NULL) {
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
tfd->u.data.chunk_len[i] = cpu_to_le16(remaining_bytes);
2005-03-25 13:31:53 -05:00
for (j = i; j < txb->nr_frags; j++) {
int size = txb->fragments[j]->len - hdr_len;
2005-03-25 13:31:53 -05:00
printk(KERN_INFO "Adding frag %d %d...\n",
j, size);
2005-03-25 13:31:53 -05:00
memcpy(skb_put(skb, size),
txb->fragments[j]->data + hdr_len, size);
2005-03-25 13:31:53 -05:00
}
dev_kfree_skb_any(txb->fragments[i]);
txb->fragments[i] = skb;
tfd->u.data.chunk_ptr[i] =
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
cpu_to_le32(pci_map_single
(priv->pci_dev, skb->data,
tfd->u.data.chunk_len[i],
PCI_DMA_TODEVICE));
tfd->u.data.num_chunks =
cpu_to_le32(le32_to_cpu(tfd->u.data.num_chunks) +
1);
}
2005-03-25 13:31:53 -05:00
}
/* kick DMA */
q->first_empty = ipw_queue_inc_wrap(q->first_empty, q->n_bd);
ipw_write32(priv, q->reg_w, q->first_empty);
if (ipw_queue_space(q) < q->high_mark)
netif_stop_queue(priv->net_dev);
return NETDEV_TX_OK;
2005-03-25 13:31:53 -05:00
drop:
2005-03-25 13:31:53 -05:00
IPW_DEBUG_DROP("Silently dropping Tx packet.\n");
ieee80211_txb_free(txb);
return NETDEV_TX_OK;
}
static int ipw_net_is_queue_full(struct net_device *dev, int pri)
{
struct ipw_priv *priv = ieee80211_priv(dev);
#ifdef CONFIG_IPW2200_QOS
int tx_id = ipw_get_tx_queue_number(priv, pri);
struct clx2_tx_queue *txq = &priv->txq[tx_id];
#else
struct clx2_tx_queue *txq = &priv->txq[0];
#endif /* CONFIG_IPW2200_QOS */
if (ipw_queue_space(&txq->q) < txq->q.high_mark)
return 1;
return 0;
2005-03-25 13:31:53 -05:00
}
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
static void ipw_handle_promiscuous_tx(struct ipw_priv *priv,
struct ieee80211_txb *txb)
{
struct ieee80211_rx_stats dummystats;
struct ieee80211_hdr *hdr;
u8 n;
u16 filter = priv->prom_priv->filter;
int hdr_only = 0;
if (filter & IPW_PROM_NO_TX)
return;
memset(&dummystats, 0, sizeof(dummystats));
/* Filtering of fragment chains is done agains the first fragment */
hdr = (void *)txb->fragments[0]->data;
if (ieee80211_is_management(hdr->frame_ctl)) {
if (filter & IPW_PROM_NO_MGMT)
return;
if (filter & IPW_PROM_MGMT_HEADER_ONLY)
hdr_only = 1;
} else if (ieee80211_is_control(hdr->frame_ctl)) {
if (filter & IPW_PROM_NO_CTL)
return;
if (filter & IPW_PROM_CTL_HEADER_ONLY)
hdr_only = 1;
} else if (ieee80211_is_data(hdr->frame_ctl)) {
if (filter & IPW_PROM_NO_DATA)
return;
if (filter & IPW_PROM_DATA_HEADER_ONLY)
hdr_only = 1;
}
for(n=0; n<txb->nr_frags; ++n) {
struct sk_buff *src = txb->fragments[n];
struct sk_buff *dst;
struct ieee80211_radiotap_header *rt_hdr;
int len;
if (hdr_only) {
hdr = (void *)src->data;
len = ieee80211_get_hdrlen(hdr->frame_ctl);
} else
len = src->len;
dst = alloc_skb(
len + IEEE80211_RADIOTAP_HDRLEN, GFP_ATOMIC);
if (!dst) continue;
rt_hdr = (void *)skb_put(dst, sizeof(*rt_hdr));
rt_hdr->it_version = PKTHDR_RADIOTAP_VERSION;
rt_hdr->it_pad = 0;
rt_hdr->it_present = 0; /* after all, it's just an idea */
rt_hdr->it_present |= (1 << IEEE80211_RADIOTAP_CHANNEL);
*(u16*)skb_put(dst, sizeof(u16)) = cpu_to_le16(
ieee80211chan2mhz(priv->channel));
if (priv->channel > 14) /* 802.11a */
*(u16*)skb_put(dst, sizeof(u16)) =
cpu_to_le16(IEEE80211_CHAN_OFDM |
IEEE80211_CHAN_5GHZ);
else if (priv->ieee->mode == IEEE_B) /* 802.11b */
*(u16*)skb_put(dst, sizeof(u16)) =
cpu_to_le16(IEEE80211_CHAN_CCK |
IEEE80211_CHAN_2GHZ);
else /* 802.11g */
*(u16*)skb_put(dst, sizeof(u16)) =
cpu_to_le16(IEEE80211_CHAN_OFDM |
IEEE80211_CHAN_2GHZ);
rt_hdr->it_len = dst->len;
memcpy(skb_put(dst, len), src->data, len);
if (!ieee80211_rx(priv->prom_priv->ieee, dst, &dummystats))
dev_kfree_skb_any(dst);
}
}
#endif
2005-03-25 13:31:53 -05:00
static int ipw_net_hard_start_xmit(struct ieee80211_txb *txb,
struct net_device *dev, int pri)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = ieee80211_priv(dev);
unsigned long flags;
int ret;
2005-03-25 13:31:53 -05:00
IPW_DEBUG_TX("dev->xmit(%d bytes)\n", txb->payload_size);
spin_lock_irqsave(&priv->lock, flags);
if (!(priv->status & STATUS_ASSOCIATED)) {
IPW_DEBUG_INFO("Tx attempt while not associated.\n");
priv->ieee->stats.tx_carrier_errors++;
netif_stop_queue(dev);
goto fail_unlock;
}
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
if (rtap_iface && netif_running(priv->prom_net_dev))
ipw_handle_promiscuous_tx(priv, txb);
#endif
ret = ipw_tx_skb(priv, txb, pri);
if (ret == NETDEV_TX_OK)
__ipw_led_activity_on(priv);
2005-03-25 13:31:53 -05:00
spin_unlock_irqrestore(&priv->lock, flags);
return ret;
2005-03-25 13:31:53 -05:00
fail_unlock:
2005-03-25 13:31:53 -05:00
spin_unlock_irqrestore(&priv->lock, flags);
return 1;
}
static struct net_device_stats *ipw_net_get_stats(struct net_device *dev)
{
struct ipw_priv *priv = ieee80211_priv(dev);
2005-03-25 13:31:53 -05:00
priv->ieee->stats.tx_packets = priv->tx_packets;
priv->ieee->stats.rx_packets = priv->rx_packets;
return &priv->ieee->stats;
}
static void ipw_net_set_multicast_list(struct net_device *dev)
{
}
static int ipw_net_set_mac_address(struct net_device *dev, void *p)
{
struct ipw_priv *priv = ieee80211_priv(dev);
struct sockaddr *addr = p;
if (!is_valid_ether_addr(addr->sa_data))
return -EADDRNOTAVAIL;
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
priv->config |= CFG_CUSTOM_MAC;
memcpy(priv->mac_addr, addr->sa_data, ETH_ALEN);
printk(KERN_INFO "%s: Setting MAC to " MAC_FMT "\n",
priv->net_dev->name, MAC_ARG(priv->mac_addr));
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
queue_work(priv->workqueue, &priv->adapter_restart);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static void ipw_ethtool_get_drvinfo(struct net_device *dev,
2005-03-25 13:31:53 -05:00
struct ethtool_drvinfo *info)
{
struct ipw_priv *p = ieee80211_priv(dev);
char vers[64];
char date[32];
u32 len;
strcpy(info->driver, DRV_NAME);
strcpy(info->version, DRV_VERSION);
len = sizeof(vers);
ipw_get_ordinal(p, IPW_ORD_STAT_FW_VERSION, vers, &len);
len = sizeof(date);
ipw_get_ordinal(p, IPW_ORD_STAT_FW_DATE, date, &len);
snprintf(info->fw_version, sizeof(info->fw_version), "%s (%s)",
2005-03-25 13:31:53 -05:00
vers, date);
strcpy(info->bus_info, pci_name(p->pci_dev));
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
info->eedump_len = IPW_EEPROM_IMAGE_SIZE;
2005-03-25 13:31:53 -05:00
}
static u32 ipw_ethtool_get_link(struct net_device *dev)
{
struct ipw_priv *priv = ieee80211_priv(dev);
return (priv->status & STATUS_ASSOCIATED) != 0;
}
static int ipw_ethtool_get_eeprom_len(struct net_device *dev)
{
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
return IPW_EEPROM_IMAGE_SIZE;
2005-03-25 13:31:53 -05:00
}
static int ipw_ethtool_get_eeprom(struct net_device *dev,
struct ethtool_eeprom *eeprom, u8 * bytes)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *p = ieee80211_priv(dev);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (eeprom->offset + eeprom->len > IPW_EEPROM_IMAGE_SIZE)
2005-03-25 13:31:53 -05:00
return -EINVAL;
mutex_lock(&p->mutex);
memcpy(bytes, &p->eeprom[eeprom->offset], eeprom->len);
mutex_unlock(&p->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_ethtool_set_eeprom(struct net_device *dev,
struct ethtool_eeprom *eeprom, u8 * bytes)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *p = ieee80211_priv(dev);
int i;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (eeprom->offset + eeprom->len > IPW_EEPROM_IMAGE_SIZE)
2005-03-25 13:31:53 -05:00
return -EINVAL;
mutex_lock(&p->mutex);
memcpy(&p->eeprom[eeprom->offset], bytes, eeprom->len);
for (i = 0; i < IPW_EEPROM_IMAGE_SIZE; i++)
ipw_write8(p, i + IPW_EEPROM_DATA, p->eeprom[i]);
mutex_unlock(&p->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
static struct ethtool_ops ipw_ethtool_ops = {
.get_link = ipw_ethtool_get_link,
.get_drvinfo = ipw_ethtool_get_drvinfo,
.get_eeprom_len = ipw_ethtool_get_eeprom_len,
.get_eeprom = ipw_ethtool_get_eeprom,
.set_eeprom = ipw_ethtool_set_eeprom,
2005-03-25 13:31:53 -05:00
};
static irqreturn_t ipw_isr(int irq, void *data, struct pt_regs *regs)
{
struct ipw_priv *priv = data;
u32 inta, inta_mask;
2005-03-25 13:31:53 -05:00
if (!priv)
return IRQ_NONE;
spin_lock(&priv->lock);
if (!(priv->status & STATUS_INT_ENABLED)) {
/* Shared IRQ */
goto none;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
inta = ipw_read32(priv, IPW_INTA_RW);
inta_mask = ipw_read32(priv, IPW_INTA_MASK_R);
2005-03-25 13:31:53 -05:00
if (inta == 0xFFFFFFFF) {
/* Hardware disappeared */
IPW_WARNING("IRQ INTA == 0xFFFFFFFF\n");
goto none;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (!(inta & (IPW_INTA_MASK_ALL & inta_mask))) {
2005-03-25 13:31:53 -05:00
/* Shared interrupt */
goto none;
}
/* tell the device to stop sending interrupts */
ipw_disable_interrupts(priv);
2005-03-25 13:31:53 -05:00
/* ack current interrupts */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
inta &= (IPW_INTA_MASK_ALL & inta_mask);
ipw_write32(priv, IPW_INTA_RW, inta);
2005-03-25 13:31:53 -05:00
/* Cache INTA value for our tasklet */
priv->isr_inta = inta;
tasklet_schedule(&priv->irq_tasklet);
spin_unlock(&priv->lock);
2005-03-25 13:31:53 -05:00
return IRQ_HANDLED;
none:
2005-03-25 13:31:53 -05:00
spin_unlock(&priv->lock);
return IRQ_NONE;
}
static void ipw_rf_kill(void *adapter)
{
struct ipw_priv *priv = adapter;
unsigned long flags;
2005-03-25 13:31:53 -05:00
spin_lock_irqsave(&priv->lock, flags);
if (rf_kill_active(priv)) {
IPW_DEBUG_RF_KILL("RF Kill active, rescheduling GPIO check\n");
if (priv->workqueue)
queue_delayed_work(priv->workqueue,
&priv->rf_kill, 2 * HZ);
goto exit_unlock;
}
/* RF Kill is now disabled, so bring the device back up */
if (!(priv->status & STATUS_RF_KILL_MASK)) {
IPW_DEBUG_RF_KILL("HW RF Kill no longer active, restarting "
"device\n");
/* we can not do an adapter restart while inside an irq lock */
queue_work(priv->workqueue, &priv->adapter_restart);
} else
2005-03-25 13:31:53 -05:00
IPW_DEBUG_RF_KILL("HW RF Kill deactivated. SW RF Kill still "
"enabled\n");
exit_unlock:
2005-03-25 13:31:53 -05:00
spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipw_bg_rf_kill(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_rf_kill(data);
mutex_unlock(&priv->mutex);
}
static void ipw_link_up(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
priv->last_seq_num = -1;
priv->last_frag_num = -1;
priv->last_packet_time = 0;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
netif_carrier_on(priv->net_dev);
if (netif_queue_stopped(priv->net_dev)) {
IPW_DEBUG_NOTIF("waking queue\n");
netif_wake_queue(priv->net_dev);
} else {
IPW_DEBUG_NOTIF("starting queue\n");
netif_start_queue(priv->net_dev);
}
cancel_delayed_work(&priv->request_scan);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
ipw_reset_stats(priv);
/* Ensure the rate is updated immediately */
priv->last_rate = ipw_get_current_rate(priv);
ipw_gather_stats(priv);
ipw_led_link_up(priv);
notify_wx_assoc_event(priv);
if (priv->config & CFG_BACKGROUND_SCAN)
queue_delayed_work(priv->workqueue, &priv->request_scan, HZ);
}
static void ipw_bg_link_up(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_link_up(data);
mutex_unlock(&priv->mutex);
}
static void ipw_link_down(struct ipw_priv *priv)
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{
ipw_led_link_down(priv);
netif_carrier_off(priv->net_dev);
netif_stop_queue(priv->net_dev);
notify_wx_assoc_event(priv);
/* Cancel any queued work ... */
cancel_delayed_work(&priv->request_scan);
cancel_delayed_work(&priv->adhoc_check);
cancel_delayed_work(&priv->gather_stats);
ipw_reset_stats(priv);
if (!(priv->status & STATUS_EXIT_PENDING)) {
/* Queue up another scan... */
queue_work(priv->workqueue, &priv->request_scan);
}
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
}
static void ipw_bg_link_down(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_link_down(data);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
}
static int ipw_setup_deferred_work(struct ipw_priv *priv)
{
int ret = 0;
priv->workqueue = create_workqueue(DRV_NAME);
init_waitqueue_head(&priv->wait_command_queue);
init_waitqueue_head(&priv->wait_state);
2005-03-25 13:31:53 -05:00
INIT_WORK(&priv->adhoc_check, ipw_bg_adhoc_check, priv);
INIT_WORK(&priv->associate, ipw_bg_associate, priv);
INIT_WORK(&priv->disassociate, ipw_bg_disassociate, priv);
INIT_WORK(&priv->system_config, ipw_system_config, priv);
INIT_WORK(&priv->rx_replenish, ipw_bg_rx_queue_replenish, priv);
INIT_WORK(&priv->adapter_restart, ipw_bg_adapter_restart, priv);
INIT_WORK(&priv->rf_kill, ipw_bg_rf_kill, priv);
INIT_WORK(&priv->up, (void (*)(void *))ipw_bg_up, priv);
INIT_WORK(&priv->down, (void (*)(void *))ipw_bg_down, priv);
INIT_WORK(&priv->request_scan,
2005-03-25 13:31:53 -05:00
(void (*)(void *))ipw_request_scan, priv);
INIT_WORK(&priv->gather_stats,
(void (*)(void *))ipw_bg_gather_stats, priv);
INIT_WORK(&priv->abort_scan, (void (*)(void *))ipw_bg_abort_scan, priv);
INIT_WORK(&priv->roam, ipw_bg_roam, priv);
INIT_WORK(&priv->scan_check, ipw_bg_scan_check, priv);
INIT_WORK(&priv->link_up, (void (*)(void *))ipw_bg_link_up, priv);
INIT_WORK(&priv->link_down, (void (*)(void *))ipw_bg_link_down, priv);
INIT_WORK(&priv->led_link_on, (void (*)(void *))ipw_bg_led_link_on,
priv);
INIT_WORK(&priv->led_link_off, (void (*)(void *))ipw_bg_led_link_off,
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
priv);
INIT_WORK(&priv->led_act_off, (void (*)(void *))ipw_bg_led_activity_off,
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
priv);
INIT_WORK(&priv->merge_networks,
(void (*)(void *))ipw_merge_adhoc_network, priv);
2005-03-25 13:31:53 -05:00
#ifdef CONFIG_IPW2200_QOS
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
INIT_WORK(&priv->qos_activate, (void (*)(void *))ipw_bg_qos_activate,
priv);
#endif /* CONFIG_IPW2200_QOS */
2005-03-25 13:31:53 -05:00
tasklet_init(&priv->irq_tasklet, (void (*)(unsigned long))
ipw_irq_tasklet, (unsigned long)priv);
return ret;
}
static void shim__set_security(struct net_device *dev,
struct ieee80211_security *sec)
{
struct ipw_priv *priv = ieee80211_priv(dev);
int i;
for (i = 0; i < 4; i++) {
2005-03-25 13:31:53 -05:00
if (sec->flags & (1 << i)) {
priv->ieee->sec.encode_alg[i] = sec->encode_alg[i];
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->ieee->sec.key_sizes[i] = sec->key_sizes[i];
2005-03-25 13:31:53 -05:00
if (sec->key_sizes[i] == 0)
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->ieee->sec.flags &= ~(1 << i);
else {
memcpy(priv->ieee->sec.keys[i], sec->keys[i],
2005-03-25 13:31:53 -05:00
sec->key_sizes[i]);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->ieee->sec.flags |= (1 << i);
}
2005-03-25 13:31:53 -05:00
priv->status |= STATUS_SECURITY_UPDATED;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
} else if (sec->level != SEC_LEVEL_1)
priv->ieee->sec.flags &= ~(1 << i);
2005-03-25 13:31:53 -05:00
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (sec->flags & SEC_ACTIVE_KEY) {
2005-03-25 13:31:53 -05:00
if (sec->active_key <= 3) {
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->ieee->sec.active_key = sec->active_key;
priv->ieee->sec.flags |= SEC_ACTIVE_KEY;
} else
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->ieee->sec.flags &= ~SEC_ACTIVE_KEY;
2005-03-25 13:31:53 -05:00
priv->status |= STATUS_SECURITY_UPDATED;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
} else
priv->ieee->sec.flags &= ~SEC_ACTIVE_KEY;
2005-03-25 13:31:53 -05:00
if ((sec->flags & SEC_AUTH_MODE) &&
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
(priv->ieee->sec.auth_mode != sec->auth_mode)) {
priv->ieee->sec.auth_mode = sec->auth_mode;
priv->ieee->sec.flags |= SEC_AUTH_MODE;
2005-03-25 13:31:53 -05:00
if (sec->auth_mode == WLAN_AUTH_SHARED_KEY)
priv->capability |= CAP_SHARED_KEY;
else
priv->capability &= ~CAP_SHARED_KEY;
priv->status |= STATUS_SECURITY_UPDATED;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (sec->flags & SEC_ENABLED && priv->ieee->sec.enabled != sec->enabled) {
priv->ieee->sec.flags |= SEC_ENABLED;
priv->ieee->sec.enabled = sec->enabled;
2005-03-25 13:31:53 -05:00
priv->status |= STATUS_SECURITY_UPDATED;
if (sec->enabled)
2005-03-25 13:31:53 -05:00
priv->capability |= CAP_PRIVACY_ON;
else
priv->capability &= ~CAP_PRIVACY_ON;
}
if (sec->flags & SEC_ENCRYPT)
priv->ieee->sec.encrypt = sec->encrypt;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (sec->flags & SEC_LEVEL && priv->ieee->sec.level != sec->level) {
priv->ieee->sec.level = sec->level;
priv->ieee->sec.flags |= SEC_LEVEL;
2005-03-25 13:31:53 -05:00
priv->status |= STATUS_SECURITY_UPDATED;
}
if (!priv->ieee->host_encrypt && (sec->flags & SEC_ENCRYPT))
ipw_set_hwcrypto_keys(priv);
/* To match current functionality of ipw2100 (which works well w/
* various supplicants, we don't force a disassociate if the
2005-03-25 13:31:53 -05:00
* privacy capability changes ... */
#if 0
if ((priv->status & (STATUS_ASSOCIATED | STATUS_ASSOCIATING)) &&
(((priv->assoc_request.capability &
2005-03-25 13:31:53 -05:00
WLAN_CAPABILITY_PRIVACY) && !sec->enabled) ||
(!(priv->assoc_request.capability &
WLAN_CAPABILITY_PRIVACY) && sec->enabled))) {
2005-03-25 13:31:53 -05:00
IPW_DEBUG_ASSOC("Disassociating due to capability "
"change.\n");
ipw_disassociate(priv);
}
#endif
}
static int init_supported_rates(struct ipw_priv *priv,
2005-03-25 13:31:53 -05:00
struct ipw_supported_rates *rates)
{
/* TODO: Mask out rates based on priv->rates_mask */
memset(rates, 0, sizeof(*rates));
/* configure supported rates */
2005-03-25 13:31:53 -05:00
switch (priv->ieee->freq_band) {
case IEEE80211_52GHZ_BAND:
rates->ieee_mode = IPW_A_MODE;
rates->purpose = IPW_RATE_CAPABILITIES;
ipw_add_ofdm_scan_rates(rates, IEEE80211_CCK_MODULATION,
IEEE80211_OFDM_DEFAULT_RATES_MASK);
break;
default: /* Mixed or 2.4Ghz */
2005-03-25 13:31:53 -05:00
rates->ieee_mode = IPW_G_MODE;
rates->purpose = IPW_RATE_CAPABILITIES;
ipw_add_cck_scan_rates(rates, IEEE80211_CCK_MODULATION,
IEEE80211_CCK_DEFAULT_RATES_MASK);
if (priv->ieee->modulation & IEEE80211_OFDM_MODULATION) {
ipw_add_ofdm_scan_rates(rates, IEEE80211_CCK_MODULATION,
IEEE80211_OFDM_DEFAULT_RATES_MASK);
}
break;
}
return 0;
}
static int ipw_config(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
/* This is only called from ipw_up, which resets/reloads the firmware
so, we don't need to first disable the card before we configure
it */
if (ipw_set_tx_power(priv))
2005-03-25 13:31:53 -05:00
goto error;
/* initialize adapter address */
if (ipw_send_adapter_address(priv, priv->net_dev->dev_addr))
goto error;
/* set basic system config settings */
init_sys_config(&priv->sys_config);
/* Support Bluetooth if we have BT h/w on board, and user wants to.
* Does not support BT priority yet (don't abort or defer our Tx) */
if (bt_coexist) {
unsigned char bt_caps = priv->eeprom[EEPROM_SKU_CAPABILITY];
if (bt_caps & EEPROM_SKU_CAP_BT_CHANNEL_SIG)
priv->sys_config.bt_coexistence
|= CFG_BT_COEXISTENCE_SIGNAL_CHNL;
if (bt_caps & EEPROM_SKU_CAP_BT_OOB)
priv->sys_config.bt_coexistence
|= CFG_BT_COEXISTENCE_OOB;
}
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
if (priv->prom_net_dev && netif_running(priv->prom_net_dev)) {
priv->sys_config.accept_all_data_frames = 1;
priv->sys_config.accept_non_directed_frames = 1;
priv->sys_config.accept_all_mgmt_bcpr = 1;
priv->sys_config.accept_all_mgmt_frames = 1;
}
#endif
if (priv->ieee->iw_mode == IW_MODE_ADHOC)
priv->sys_config.answer_broadcast_ssid_probe = 1;
else
priv->sys_config.answer_broadcast_ssid_probe = 0;
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
if (ipw_send_system_config(priv))
2005-03-25 13:31:53 -05:00
goto error;
init_supported_rates(priv, &priv->rates);
if (ipw_send_supported_rates(priv, &priv->rates))
2005-03-25 13:31:53 -05:00
goto error;
/* Set request-to-send threshold */
if (priv->rts_threshold) {
if (ipw_send_rts_threshold(priv, priv->rts_threshold))
goto error;
}
#ifdef CONFIG_IPW2200_QOS
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
IPW_DEBUG_QOS("QoS: call ipw_qos_activate\n");
ipw_qos_activate(priv, NULL);
#endif /* CONFIG_IPW2200_QOS */
2005-03-25 13:31:53 -05:00
if (ipw_set_random_seed(priv))
goto error;
2005-03-25 13:31:53 -05:00
/* final state transition to the RUN state */
if (ipw_send_host_complete(priv))
goto error;
priv->status |= STATUS_INIT;
ipw_led_init(priv);
ipw_led_radio_on(priv);
priv->notif_missed_beacons = 0;
/* Set hardware WEP key if it is configured. */
if ((priv->capability & CAP_PRIVACY_ON) &&
(priv->ieee->sec.level == SEC_LEVEL_1) &&
!(priv->ieee->host_encrypt || priv->ieee->host_decrypt))
ipw_set_hwcrypto_keys(priv);
2005-03-25 13:31:53 -05:00
return 0;
error:
2005-03-25 13:31:53 -05:00
return -EIO;
}
/*
* NOTE:
*
* These tables have been tested in conjunction with the
* Intel PRO/Wireless 2200BG and 2915ABG Network Connection Adapters.
*
* Altering this values, using it on other hardware, or in geographies
* not intended for resale of the above mentioned Intel adapters has
* not been tested.
*
* Remember to update the table in README.ipw2200 when changing this
* table.
*
*/
static const struct ieee80211_geo ipw_geos[] = {
{ /* Restricted */
"---",
.bg_channels = 11,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}},
},
{ /* Custom US/Canada */
"ZZF",
.bg_channels = 11,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}},
.a_channels = 8,
.a = {{5180, 36},
{5200, 40},
{5220, 44},
{5240, 48},
{5260, 52, IEEE80211_CH_PASSIVE_ONLY},
{5280, 56, IEEE80211_CH_PASSIVE_ONLY},
{5300, 60, IEEE80211_CH_PASSIVE_ONLY},
{5320, 64, IEEE80211_CH_PASSIVE_ONLY}},
},
{ /* Rest of World */
"ZZD",
.bg_channels = 13,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}, {2467, 12},
{2472, 13}},
},
{ /* Custom USA & Europe & High */
"ZZA",
.bg_channels = 11,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}},
.a_channels = 13,
.a = {{5180, 36},
{5200, 40},
{5220, 44},
{5240, 48},
{5260, 52, IEEE80211_CH_PASSIVE_ONLY},
{5280, 56, IEEE80211_CH_PASSIVE_ONLY},
{5300, 60, IEEE80211_CH_PASSIVE_ONLY},
{5320, 64, IEEE80211_CH_PASSIVE_ONLY},
{5745, 149},
{5765, 153},
{5785, 157},
{5805, 161},
{5825, 165}},
},
{ /* Custom NA & Europe */
"ZZB",
.bg_channels = 11,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}},
.a_channels = 13,
.a = {{5180, 36},
{5200, 40},
{5220, 44},
{5240, 48},
{5260, 52, IEEE80211_CH_PASSIVE_ONLY},
{5280, 56, IEEE80211_CH_PASSIVE_ONLY},
{5300, 60, IEEE80211_CH_PASSIVE_ONLY},
{5320, 64, IEEE80211_CH_PASSIVE_ONLY},
{5745, 149, IEEE80211_CH_PASSIVE_ONLY},
{5765, 153, IEEE80211_CH_PASSIVE_ONLY},
{5785, 157, IEEE80211_CH_PASSIVE_ONLY},
{5805, 161, IEEE80211_CH_PASSIVE_ONLY},
{5825, 165, IEEE80211_CH_PASSIVE_ONLY}},
},
{ /* Custom Japan */
"ZZC",
.bg_channels = 11,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}},
.a_channels = 4,
.a = {{5170, 34}, {5190, 38},
{5210, 42}, {5230, 46}},
},
{ /* Custom */
"ZZM",
.bg_channels = 11,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}},
},
{ /* Europe */
"ZZE",
.bg_channels = 13,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}, {2467, 12},
{2472, 13}},
.a_channels = 19,
.a = {{5180, 36},
{5200, 40},
{5220, 44},
{5240, 48},
{5260, 52, IEEE80211_CH_PASSIVE_ONLY},
{5280, 56, IEEE80211_CH_PASSIVE_ONLY},
{5300, 60, IEEE80211_CH_PASSIVE_ONLY},
{5320, 64, IEEE80211_CH_PASSIVE_ONLY},
{5500, 100, IEEE80211_CH_PASSIVE_ONLY},
{5520, 104, IEEE80211_CH_PASSIVE_ONLY},
{5540, 108, IEEE80211_CH_PASSIVE_ONLY},
{5560, 112, IEEE80211_CH_PASSIVE_ONLY},
{5580, 116, IEEE80211_CH_PASSIVE_ONLY},
{5600, 120, IEEE80211_CH_PASSIVE_ONLY},
{5620, 124, IEEE80211_CH_PASSIVE_ONLY},
{5640, 128, IEEE80211_CH_PASSIVE_ONLY},
{5660, 132, IEEE80211_CH_PASSIVE_ONLY},
{5680, 136, IEEE80211_CH_PASSIVE_ONLY},
{5700, 140, IEEE80211_CH_PASSIVE_ONLY}},
},
{ /* Custom Japan */
"ZZJ",
.bg_channels = 14,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}, {2467, 12},
{2472, 13}, {2484, 14, IEEE80211_CH_B_ONLY}},
.a_channels = 4,
.a = {{5170, 34}, {5190, 38},
{5210, 42}, {5230, 46}},
},
{ /* Rest of World */
"ZZR",
.bg_channels = 14,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}, {2467, 12},
{2472, 13}, {2484, 14, IEEE80211_CH_B_ONLY |
IEEE80211_CH_PASSIVE_ONLY}},
},
{ /* High Band */
"ZZH",
.bg_channels = 13,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11},
{2467, 12, IEEE80211_CH_PASSIVE_ONLY},
{2472, 13, IEEE80211_CH_PASSIVE_ONLY}},
.a_channels = 4,
.a = {{5745, 149}, {5765, 153},
{5785, 157}, {5805, 161}},
},
{ /* Custom Europe */
"ZZG",
.bg_channels = 13,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11},
{2467, 12}, {2472, 13}},
.a_channels = 4,
.a = {{5180, 36}, {5200, 40},
{5220, 44}, {5240, 48}},
},
{ /* Europe */
"ZZK",
.bg_channels = 13,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11},
{2467, 12, IEEE80211_CH_PASSIVE_ONLY},
{2472, 13, IEEE80211_CH_PASSIVE_ONLY}},
.a_channels = 24,
.a = {{5180, 36, IEEE80211_CH_PASSIVE_ONLY},
{5200, 40, IEEE80211_CH_PASSIVE_ONLY},
{5220, 44, IEEE80211_CH_PASSIVE_ONLY},
{5240, 48, IEEE80211_CH_PASSIVE_ONLY},
{5260, 52, IEEE80211_CH_PASSIVE_ONLY},
{5280, 56, IEEE80211_CH_PASSIVE_ONLY},
{5300, 60, IEEE80211_CH_PASSIVE_ONLY},
{5320, 64, IEEE80211_CH_PASSIVE_ONLY},
{5500, 100, IEEE80211_CH_PASSIVE_ONLY},
{5520, 104, IEEE80211_CH_PASSIVE_ONLY},
{5540, 108, IEEE80211_CH_PASSIVE_ONLY},
{5560, 112, IEEE80211_CH_PASSIVE_ONLY},
{5580, 116, IEEE80211_CH_PASSIVE_ONLY},
{5600, 120, IEEE80211_CH_PASSIVE_ONLY},
{5620, 124, IEEE80211_CH_PASSIVE_ONLY},
{5640, 128, IEEE80211_CH_PASSIVE_ONLY},
{5660, 132, IEEE80211_CH_PASSIVE_ONLY},
{5680, 136, IEEE80211_CH_PASSIVE_ONLY},
{5700, 140, IEEE80211_CH_PASSIVE_ONLY},
{5745, 149, IEEE80211_CH_PASSIVE_ONLY},
{5765, 153, IEEE80211_CH_PASSIVE_ONLY},
{5785, 157, IEEE80211_CH_PASSIVE_ONLY},
{5805, 161, IEEE80211_CH_PASSIVE_ONLY},
{5825, 165, IEEE80211_CH_PASSIVE_ONLY}},
},
{ /* Europe */
"ZZL",
.bg_channels = 11,
.bg = {{2412, 1}, {2417, 2}, {2422, 3},
{2427, 4}, {2432, 5}, {2437, 6},
{2442, 7}, {2447, 8}, {2452, 9},
{2457, 10}, {2462, 11}},
.a_channels = 13,
.a = {{5180, 36, IEEE80211_CH_PASSIVE_ONLY},
{5200, 40, IEEE80211_CH_PASSIVE_ONLY},
{5220, 44, IEEE80211_CH_PASSIVE_ONLY},
{5240, 48, IEEE80211_CH_PASSIVE_ONLY},
{5260, 52, IEEE80211_CH_PASSIVE_ONLY},
{5280, 56, IEEE80211_CH_PASSIVE_ONLY},
{5300, 60, IEEE80211_CH_PASSIVE_ONLY},
{5320, 64, IEEE80211_CH_PASSIVE_ONLY},
{5745, 149, IEEE80211_CH_PASSIVE_ONLY},
{5765, 153, IEEE80211_CH_PASSIVE_ONLY},
{5785, 157, IEEE80211_CH_PASSIVE_ONLY},
{5805, 161, IEEE80211_CH_PASSIVE_ONLY},
{5825, 165, IEEE80211_CH_PASSIVE_ONLY}},
}
};
2005-03-25 13:31:53 -05:00
#define MAX_HW_RESTARTS 5
static int ipw_up(struct ipw_priv *priv)
{
int rc, i, j;
2005-03-25 13:31:53 -05:00
if (priv->status & STATUS_EXIT_PENDING)
return -EIO;
if (cmdlog && !priv->cmdlog) {
priv->cmdlog = kmalloc(sizeof(*priv->cmdlog) * cmdlog,
GFP_KERNEL);
if (priv->cmdlog == NULL) {
IPW_ERROR("Error allocating %d command log entries.\n",
cmdlog);
return -ENOMEM;
} else {
memset(priv->cmdlog, 0, sizeof(*priv->cmdlog) * cmdlog);
priv->cmdlog_len = cmdlog;
}
}
for (i = 0; i < MAX_HW_RESTARTS; i++) {
/* Load the microcode, firmware, and eeprom.
2005-03-25 13:31:53 -05:00
* Also start the clocks. */
rc = ipw_load(priv);
if (rc) {
IPW_ERROR("Unable to load firmware: %d\n", rc);
2005-03-25 13:31:53 -05:00
return rc;
}
ipw_init_ordinals(priv);
if (!(priv->config & CFG_CUSTOM_MAC))
eeprom_parse_mac(priv, priv->mac_addr);
memcpy(priv->net_dev->dev_addr, priv->mac_addr, ETH_ALEN);
for (j = 0; j < ARRAY_SIZE(ipw_geos); j++) {
if (!memcmp(&priv->eeprom[EEPROM_COUNTRY_CODE],
ipw_geos[j].name, 3))
break;
}
if (j == ARRAY_SIZE(ipw_geos)) {
IPW_WARNING("SKU [%c%c%c] not recognized.\n",
priv->eeprom[EEPROM_COUNTRY_CODE + 0],
priv->eeprom[EEPROM_COUNTRY_CODE + 1],
priv->eeprom[EEPROM_COUNTRY_CODE + 2]);
j = 0;
}
if (ieee80211_set_geo(priv->ieee, &ipw_geos[j])) {
IPW_WARNING("Could not set geography.");
return 0;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
if (priv->status & STATUS_RF_KILL_SW) {
IPW_WARNING("Radio disabled by module parameter.\n");
return 0;
} else if (rf_kill_active(priv)) {
IPW_WARNING("Radio Frequency Kill Switch is On:\n"
"Kill switch must be turned off for "
"wireless networking to work.\n");
queue_delayed_work(priv->workqueue, &priv->rf_kill,
2 * HZ);
2005-03-25 13:31:53 -05:00
return 0;
}
2005-03-25 13:31:53 -05:00
rc = ipw_config(priv);
if (!rc) {
IPW_DEBUG_INFO("Configured device on count %i\n", i);
/* If configure to try and auto-associate, kick
* off a scan. */
queue_work(priv->workqueue, &priv->request_scan);
2005-03-25 13:31:53 -05:00
return 0;
}
IPW_DEBUG_INFO("Device configuration failed: 0x%08X\n", rc);
2005-03-25 13:31:53 -05:00
IPW_DEBUG_INFO("Failed to config device on retry %d of %d\n",
i, MAX_HW_RESTARTS);
/* We had an error bringing up the hardware, so take it
* all the way back down so we can try again */
ipw_down(priv);
}
/* tried to restart and config the device for as long as our
2005-03-25 13:31:53 -05:00
* patience could withstand */
IPW_ERROR("Unable to initialize device after %d attempts.\n", i);
2005-03-25 13:31:53 -05:00
return -EIO;
}
static void ipw_bg_up(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_up(data);
mutex_unlock(&priv->mutex);
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
static void ipw_deinit(struct ipw_priv *priv)
2005-03-25 13:31:53 -05:00
{
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
int i;
if (priv->status & STATUS_SCANNING) {
IPW_DEBUG_INFO("Aborting scan during shutdown.\n");
ipw_abort_scan(priv);
}
if (priv->status & STATUS_ASSOCIATED) {
IPW_DEBUG_INFO("Disassociating during shutdown.\n");
ipw_disassociate(priv);
}
ipw_led_shutdown(priv);
/* Wait up to 1s for status to change to not scanning and not
* associated (disassociation can take a while for a ful 802.11
* exchange */
for (i = 1000; i && (priv->status &
(STATUS_DISASSOCIATING |
STATUS_ASSOCIATED | STATUS_SCANNING)); i--)
udelay(10);
if (priv->status & (STATUS_DISASSOCIATING |
STATUS_ASSOCIATED | STATUS_SCANNING))
IPW_DEBUG_INFO("Still associated or scanning...\n");
else
IPW_DEBUG_INFO("Took %dms to de-init\n", 1000 - i);
2005-03-25 13:31:53 -05:00
/* Attempt to disable the card */
ipw_send_card_disable(priv, 0);
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->status &= ~STATUS_INIT;
}
static void ipw_down(struct ipw_priv *priv)
{
int exit_pending = priv->status & STATUS_EXIT_PENDING;
priv->status |= STATUS_EXIT_PENDING;
if (ipw_is_init(priv))
ipw_deinit(priv);
/* Wipe out the EXIT_PENDING status bit if we are not actually
* exiting the module */
if (!exit_pending)
priv->status &= ~STATUS_EXIT_PENDING;
2005-03-25 13:31:53 -05:00
/* tell the device to stop sending interrupts */
ipw_disable_interrupts(priv);
/* Clear all bits but the RF Kill */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->status &= STATUS_RF_KILL_MASK | STATUS_EXIT_PENDING;
2005-03-25 13:31:53 -05:00
netif_carrier_off(priv->net_dev);
netif_stop_queue(priv->net_dev);
ipw_stop_nic(priv);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
ipw_led_radio_off(priv);
2005-03-25 13:31:53 -05:00
}
static void ipw_bg_down(void *data)
{
struct ipw_priv *priv = data;
mutex_lock(&priv->mutex);
ipw_down(data);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
}
/* Called by register_netdev() */
static int ipw_net_init(struct net_device *dev)
{
struct ipw_priv *priv = ieee80211_priv(dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (ipw_up(priv)) {
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return -EIO;
}
2005-03-25 13:31:53 -05:00
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
return 0;
}
/* PCI driver stuff */
static struct pci_device_id card_ids[] = {
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2701, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2702, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2711, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2712, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2721, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2722, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2731, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2732, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2741, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x103c, 0x2741, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2742, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2751, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2752, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2753, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2754, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2761, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2762, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x104f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x4220, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, /* BG */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
{PCI_VENDOR_ID_INTEL, 0x4221, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, /* BG */
{PCI_VENDOR_ID_INTEL, 0x4223, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, /* ABG */
{PCI_VENDOR_ID_INTEL, 0x4224, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, /* ABG */
2005-03-25 13:31:53 -05:00
/* required last entry */
{0,}
};
MODULE_DEVICE_TABLE(pci, card_ids);
static struct attribute *ipw_sysfs_entries[] = {
&dev_attr_rf_kill.attr,
&dev_attr_direct_dword.attr,
&dev_attr_indirect_byte.attr,
&dev_attr_indirect_dword.attr,
&dev_attr_mem_gpio_reg.attr,
&dev_attr_command_event_reg.attr,
&dev_attr_nic_type.attr,
&dev_attr_status.attr,
&dev_attr_cfg.attr,
&dev_attr_error.attr,
&dev_attr_event_log.attr,
&dev_attr_cmd_log.attr,
2005-03-25 13:31:53 -05:00
&dev_attr_eeprom_delay.attr,
&dev_attr_ucode_version.attr,
&dev_attr_rtc.attr,
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
&dev_attr_scan_age.attr,
&dev_attr_led.attr,
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
&dev_attr_speed_scan.attr,
&dev_attr_net_stats.attr,
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
&dev_attr_rtap_iface.attr,
&dev_attr_rtap_filter.attr,
#endif
2005-03-25 13:31:53 -05:00
NULL
};
static struct attribute_group ipw_attribute_group = {
.name = NULL, /* put in device directory */
.attrs = ipw_sysfs_entries,
2005-03-25 13:31:53 -05:00
};
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
static int ipw_prom_open(struct net_device *dev)
{
struct ipw_prom_priv *prom_priv = ieee80211_priv(dev);
struct ipw_priv *priv = prom_priv->priv;
IPW_DEBUG_INFO("prom dev->open\n");
netif_carrier_off(dev);
netif_stop_queue(dev);
if (priv->ieee->iw_mode != IW_MODE_MONITOR) {
priv->sys_config.accept_all_data_frames = 1;
priv->sys_config.accept_non_directed_frames = 1;
priv->sys_config.accept_all_mgmt_bcpr = 1;
priv->sys_config.accept_all_mgmt_frames = 1;
ipw_send_system_config(priv);
}
return 0;
}
static int ipw_prom_stop(struct net_device *dev)
{
struct ipw_prom_priv *prom_priv = ieee80211_priv(dev);
struct ipw_priv *priv = prom_priv->priv;
IPW_DEBUG_INFO("prom dev->stop\n");
if (priv->ieee->iw_mode != IW_MODE_MONITOR) {
priv->sys_config.accept_all_data_frames = 0;
priv->sys_config.accept_non_directed_frames = 0;
priv->sys_config.accept_all_mgmt_bcpr = 0;
priv->sys_config.accept_all_mgmt_frames = 0;
ipw_send_system_config(priv);
}
return 0;
}
static int ipw_prom_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
IPW_DEBUG_INFO("prom dev->xmit\n");
netif_stop_queue(dev);
return -EOPNOTSUPP;
}
static struct net_device_stats *ipw_prom_get_stats(struct net_device *dev)
{
struct ipw_prom_priv *prom_priv = ieee80211_priv(dev);
return &prom_priv->ieee->stats;
}
static int ipw_prom_alloc(struct ipw_priv *priv)
{
int rc = 0;
if (priv->prom_net_dev)
return -EPERM;
priv->prom_net_dev = alloc_ieee80211(sizeof(struct ipw_prom_priv));
if (priv->prom_net_dev == NULL)
return -ENOMEM;
priv->prom_priv = ieee80211_priv(priv->prom_net_dev);
priv->prom_priv->ieee = netdev_priv(priv->prom_net_dev);
priv->prom_priv->priv = priv;
strcpy(priv->prom_net_dev->name, "rtap%d");
priv->prom_net_dev->type = ARPHRD_IEEE80211_RADIOTAP;
priv->prom_net_dev->open = ipw_prom_open;
priv->prom_net_dev->stop = ipw_prom_stop;
priv->prom_net_dev->get_stats = ipw_prom_get_stats;
priv->prom_net_dev->hard_start_xmit = ipw_prom_hard_start_xmit;
priv->prom_priv->ieee->iw_mode = IW_MODE_MONITOR;
rc = register_netdev(priv->prom_net_dev);
if (rc) {
free_ieee80211(priv->prom_net_dev);
priv->prom_net_dev = NULL;
return rc;
}
return 0;
}
static void ipw_prom_free(struct ipw_priv *priv)
{
if (!priv->prom_net_dev)
return;
unregister_netdev(priv->prom_net_dev);
free_ieee80211(priv->prom_net_dev);
priv->prom_net_dev = NULL;
}
#endif
static int ipw_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
2005-03-25 13:31:53 -05:00
{
int err = 0;
struct net_device *net_dev;
void __iomem *base;
u32 length, val;
struct ipw_priv *priv;
int i;
2005-03-25 13:31:53 -05:00
net_dev = alloc_ieee80211(sizeof(struct ipw_priv));
if (net_dev == NULL) {
err = -ENOMEM;
goto out;
}
priv = ieee80211_priv(net_dev);
priv->ieee = netdev_priv(net_dev);
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
2005-03-25 13:31:53 -05:00
priv->net_dev = net_dev;
priv->pci_dev = pdev;
#ifdef CONFIG_IPW2200_DEBUG
2005-03-25 13:31:53 -05:00
ipw_debug_level = debug;
#endif
spin_lock_init(&priv->lock);
for (i = 0; i < IPW_IBSS_MAC_HASH_SIZE; i++)
INIT_LIST_HEAD(&priv->ibss_mac_hash[i]);
2005-03-25 13:31:53 -05:00
mutex_init(&priv->mutex);
2005-03-25 13:31:53 -05:00
if (pci_enable_device(pdev)) {
err = -ENODEV;
goto out_free_ieee80211;
}
pci_set_master(pdev);
err = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
if (!err)
err = pci_set_consistent_dma_mask(pdev, DMA_32BIT_MASK);
2005-03-25 13:31:53 -05:00
if (err) {
printk(KERN_WARNING DRV_NAME ": No suitable DMA available.\n");
goto out_pci_disable_device;
}
pci_set_drvdata(pdev, priv);
err = pci_request_regions(pdev, DRV_NAME);
if (err)
2005-03-25 13:31:53 -05:00
goto out_pci_disable_device;
/* We disable the RETRY_TIMEOUT register (0x41) to keep
2005-03-25 13:31:53 -05:00
* PCI Tx retries from interfering with C3 CPU state */
pci_read_config_dword(pdev, 0x40, &val);
if ((val & 0x0000ff00) != 0)
2005-03-25 13:31:53 -05:00
pci_write_config_dword(pdev, 0x40, val & 0xffff00ff);
2005-03-25 13:31:53 -05:00
length = pci_resource_len(pdev, 0);
priv->hw_len = length;
2005-03-25 13:31:53 -05:00
base = ioremap_nocache(pci_resource_start(pdev, 0), length);
if (!base) {
err = -ENODEV;
goto out_pci_release_regions;
}
priv->hw_base = base;
IPW_DEBUG_INFO("pci_resource_len = 0x%08x\n", length);
IPW_DEBUG_INFO("pci_resource_base = %p\n", base);
err = ipw_setup_deferred_work(priv);
if (err) {
IPW_ERROR("Unable to setup deferred work\n");
goto out_iounmap;
}
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
ipw_sw_reset(priv, 1);
2005-03-25 13:31:53 -05:00
err = request_irq(pdev->irq, ipw_isr, SA_SHIRQ, DRV_NAME, priv);
2005-03-25 13:31:53 -05:00
if (err) {
IPW_ERROR("Error allocating IRQ %d\n", pdev->irq);
goto out_destroy_workqueue;
}
SET_MODULE_OWNER(net_dev);
SET_NETDEV_DEV(net_dev, &pdev->dev);
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
priv->ieee->hard_start_xmit = ipw_net_hard_start_xmit;
priv->ieee->set_security = shim__set_security;
priv->ieee->is_queue_full = ipw_net_is_queue_full;
2005-03-25 13:31:53 -05:00
#ifdef CONFIG_IPW2200_QOS
priv->ieee->is_qos_active = ipw_is_qos_active;
priv->ieee->handle_probe_response = ipw_handle_beacon;
priv->ieee->handle_beacon = ipw_handle_probe_response;
priv->ieee->handle_assoc_response = ipw_handle_assoc_response;
#endif /* CONFIG_IPW2200_QOS */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
priv->ieee->perfect_rssi = -20;
priv->ieee->worst_rssi = -85;
2005-03-25 13:31:53 -05:00
net_dev->open = ipw_net_open;
net_dev->stop = ipw_net_stop;
net_dev->init = ipw_net_init;
net_dev->get_stats = ipw_net_get_stats;
net_dev->set_multicast_list = ipw_net_set_multicast_list;
net_dev->set_mac_address = ipw_net_set_mac_address;
priv->wireless_data.spy_data = &priv->ieee->spy_data;
net_dev->wireless_data = &priv->wireless_data;
2005-03-25 13:31:53 -05:00
net_dev->wireless_handlers = &ipw_wx_handler_def;
net_dev->ethtool_ops = &ipw_ethtool_ops;
net_dev->irq = pdev->irq;
net_dev->base_addr = (unsigned long)priv->hw_base;
2005-03-25 13:31:53 -05:00
net_dev->mem_start = pci_resource_start(pdev, 0);
net_dev->mem_end = net_dev->mem_start + pci_resource_len(pdev, 0) - 1;
err = sysfs_create_group(&pdev->dev.kobj, &ipw_attribute_group);
if (err) {
IPW_ERROR("failed to create sysfs device attributes\n");
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
goto out_release_irq;
}
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
err = register_netdev(net_dev);
if (err) {
IPW_ERROR("failed to register network device\n");
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
goto out_remove_sysfs;
2005-03-25 13:31:53 -05:00
}
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
if (rtap_iface) {
err = ipw_prom_alloc(priv);
if (err) {
IPW_ERROR("Failed to register promiscuous network "
"device (error %d).\n", err);
unregister_netdev(priv->net_dev);
goto out_remove_sysfs;
}
}
#endif
printk(KERN_INFO DRV_NAME ": Detected geography %s (%d 802.11bg "
"channels, %d 802.11a channels)\n",
priv->ieee->geo.name, priv->ieee->geo.bg_channels,
priv->ieee->geo.a_channels);
2005-03-25 13:31:53 -05:00
return 0;
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
out_remove_sysfs:
2005-03-25 13:31:53 -05:00
sysfs_remove_group(&pdev->dev.kobj, &ipw_attribute_group);
out_release_irq:
2005-03-25 13:31:53 -05:00
free_irq(pdev->irq, priv);
out_destroy_workqueue:
2005-03-25 13:31:53 -05:00
destroy_workqueue(priv->workqueue);
priv->workqueue = NULL;
out_iounmap:
2005-03-25 13:31:53 -05:00
iounmap(priv->hw_base);
out_pci_release_regions:
2005-03-25 13:31:53 -05:00
pci_release_regions(pdev);
out_pci_disable_device:
2005-03-25 13:31:53 -05:00
pci_disable_device(pdev);
pci_set_drvdata(pdev, NULL);
out_free_ieee80211:
2005-03-25 13:31:53 -05:00
free_ieee80211(priv->net_dev);
out:
2005-03-25 13:31:53 -05:00
return err;
}
static void ipw_pci_remove(struct pci_dev *pdev)
{
struct ipw_priv *priv = pci_get_drvdata(pdev);
struct list_head *p, *q;
int i;
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
2005-03-25 13:31:53 -05:00
if (!priv)
return;
mutex_lock(&priv->mutex);
2005-03-25 13:31:53 -05:00
priv->status |= STATUS_EXIT_PENDING;
2005-03-25 13:31:53 -05:00
ipw_down(priv);
sysfs_remove_group(&pdev->dev.kobj, &ipw_attribute_group);
mutex_unlock(&priv->mutex);
2005-03-25 13:31:53 -05:00
unregister_netdev(priv->net_dev);
if (priv->rxq) {
ipw_rx_queue_free(priv, priv->rxq);
priv->rxq = NULL;
}
ipw_tx_queue_free(priv);
if (priv->cmdlog) {
kfree(priv->cmdlog);
priv->cmdlog = NULL;
}
2005-03-25 13:31:53 -05:00
/* ipw_down will ensure that there is no more pending work
* in the workqueue's, so we can safely remove them now. */
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
cancel_delayed_work(&priv->adhoc_check);
cancel_delayed_work(&priv->gather_stats);
cancel_delayed_work(&priv->request_scan);
cancel_delayed_work(&priv->rf_kill);
cancel_delayed_work(&priv->scan_check);
destroy_workqueue(priv->workqueue);
priv->workqueue = NULL;
2005-03-25 13:31:53 -05:00
/* Free MAC hash list for ADHOC */
for (i = 0; i < IPW_IBSS_MAC_HASH_SIZE; i++) {
list_for_each_safe(p, q, &priv->ibss_mac_hash[i]) {
list_del(p);
kfree(list_entry(p, struct ipw_ibss_seq, list));
}
}
if (priv->error) {
ipw_free_error_log(priv->error);
priv->error = NULL;
2005-03-25 13:31:53 -05:00
}
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
ipw_prom_free(priv);
#endif
2005-03-25 13:31:53 -05:00
free_irq(pdev->irq, priv);
iounmap(priv->hw_base);
pci_release_regions(pdev);
pci_disable_device(pdev);
pci_set_drvdata(pdev, NULL);
free_ieee80211(priv->net_dev);
free_firmware();
2005-03-25 13:31:53 -05:00
}
#ifdef CONFIG_PM
static int ipw_pci_suspend(struct pci_dev *pdev, pm_message_t state)
2005-03-25 13:31:53 -05:00
{
struct ipw_priv *priv = pci_get_drvdata(pdev);
struct net_device *dev = priv->net_dev;
printk(KERN_INFO "%s: Going into suspend...\n", dev->name);
/* Take down the device; powers it off, etc. */
2005-03-25 13:31:53 -05:00
ipw_down(priv);
/* Remove the PRESENT state of the device */
netif_device_detach(dev);
pci_save_state(pdev);
pci_disable_device(pdev);
pci_set_power_state(pdev, pci_choose_state(pdev, state));
2005-03-25 13:31:53 -05:00
return 0;
}
static int ipw_pci_resume(struct pci_dev *pdev)
{
struct ipw_priv *priv = pci_get_drvdata(pdev);
struct net_device *dev = priv->net_dev;
u32 val;
2005-03-25 13:31:53 -05:00
printk(KERN_INFO "%s: Coming out of suspend...\n", dev->name);
pci_set_power_state(pdev, PCI_D0);
2005-03-25 13:31:53 -05:00
pci_enable_device(pdev);
pci_restore_state(pdev);
2005-03-25 13:31:53 -05:00
/*
* Suspend/Resume resets the PCI configuration space, so we have to
* re-disable the RETRY_TIMEOUT register (0x41) to keep PCI Tx retries
* from interfering with C3 CPU state. pci_restore_state won't help
* here since it only restores the first 64 bytes pci config header.
*/
pci_read_config_dword(pdev, 0x40, &val);
if ((val & 0x0000ff00) != 0)
2005-03-25 13:31:53 -05:00
pci_write_config_dword(pdev, 0x40, val & 0xffff00ff);
/* Set the device back into the PRESENT state; this will also wake
* the queue of needed */
netif_device_attach(dev);
/* Bring the device back up */
queue_work(priv->workqueue, &priv->up);
2005-03-25 13:31:53 -05:00
return 0;
}
#endif
/* driver initialization stuff */
static struct pci_driver ipw_driver = {
.name = DRV_NAME,
.id_table = card_ids,
.probe = ipw_pci_probe,
.remove = __devexit_p(ipw_pci_remove),
#ifdef CONFIG_PM
.suspend = ipw_pci_suspend,
.resume = ipw_pci_resume,
#endif
};
static int __init ipw_init(void)
{
int ret;
printk(KERN_INFO DRV_NAME ": " DRV_DESCRIPTION ", " DRV_VERSION "\n");
printk(KERN_INFO DRV_NAME ": " DRV_COPYRIGHT "\n");
ret = pci_module_init(&ipw_driver);
if (ret) {
IPW_ERROR("Unable to initialize PCI module\n");
return ret;
}
ret = driver_create_file(&ipw_driver.driver, &driver_attr_debug_level);
2005-03-25 13:31:53 -05:00
if (ret) {
IPW_ERROR("Unable to create driver sysfs file\n");
pci_unregister_driver(&ipw_driver);
return ret;
}
return ret;
}
static void __exit ipw_exit(void)
{
driver_remove_file(&ipw_driver.driver, &driver_attr_debug_level);
pci_unregister_driver(&ipw_driver);
}
module_param(disable, int, 0444);
MODULE_PARM_DESC(disable, "manually disable the radio (default 0 [radio on])");
module_param(associate, int, 0444);
MODULE_PARM_DESC(associate, "auto associate when scanning (default on)");
module_param(auto_create, int, 0444);
MODULE_PARM_DESC(auto_create, "auto create adhoc network (default on)");
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
module_param(led, int, 0444);
MODULE_PARM_DESC(led, "enable led control on some systems (default 0 off)\n");
Catch ipw2200 up to equivelancy with v1.0.2 Removed unneeded parenthesis around numeric constant defines Added support for iwspy Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH) Put in fix for fragmentation not working for fragment sizes between 441-464 bytes (thanks to Mohamed Abbas) Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA support into the driver -- fixed as a result of no longer limiting WPAs inclusion Fixed #594 problem with user rates mask causing lack of association if AP mandatory rate is masked out. We now add back in as a supported rate any mandatory rate. Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times. Added code to control LEDs that can be controlled through the wireless NIC (vs. non-wireless HW interfaces) -- this is currently disabled by default due to reports by some users of it hanging their laptop. Added some more debug messages around fragmentation logic Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race conditions Moved ipw_adapter_restart to only execute on the priv->workqueue to keep keyboard errors from occuring during adapter restart Added CFG_BACKGROUND_SCAN to easily allow people to play with background scanning implementations Modified WPA logic to send WPA IE if one is set (vs. being based on wpa_enabled) Modified scan result logic to report WPA and RSN IEs if set (vs. being based on wpa_enabled) Fixed issues with endianess compatability between the host and wireless adapter (thanks to York Liu and Yi Zhu) Fixed problem with Ad-Hoc network creation causing a firmware error if a scan was actively running (thanks to Mohamed Abbas) Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 22:43:11 -04:00
#ifdef CONFIG_IPW2200_DEBUG
2005-03-25 13:31:53 -05:00
module_param(debug, int, 0444);
MODULE_PARM_DESC(debug, "debug output mask");
#endif
2005-03-25 13:31:53 -05:00
module_param(channel, int, 0444);
MODULE_PARM_DESC(channel, "channel to limit associate to (default 0 [ANY])");
2005-03-25 13:31:53 -05:00
[PATCH] ipw2200: Enable rtap interface for RF promiscuous mode while associated With this patch, a new promiscuous mode is enabled. If the module is loaded with the rtap_iface=1 module parameter, two interfaces will be created (instead of just one). The second interface is prefixed 'rtap' and provides received 802.11 frames on the current channel to user space in a radiotap header format. Example usage: % modprobe ipw2200 rtap_iface=1 % iwconfig eth1 essid MyNetwork % dhcpcd eth1 % tcpdump -i rtap0 If you do not specify 'rtap_iface=1' then the rtap interface will not be created and you will need to turn it on via: % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface You can filter out what type of information is passed to user space via the rtap_filter sysfs entry. Currently you can tell the driver to transmit just the headers (which will provide the RADIOTAP and IEEE 802.11 header but not the payload), to filter based on frame control type (Management, Control, or Data), and whether to report transmitted frames, received frames, or both. The transmit frame reporting is based on a patch by Stefan Rompf. Filters can be get and set via a sysfs interface. For example, set the filter to only send headers (0x7), don't report Tx'd frames (0x10), and don't report data frames (0x100): % echo 0x117 > /sys/bus/pci/drivers/ipw2200/*/rtap_filter All your packets are belong to us: % tethereal -n -i rtap0 Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-04-13 05:20:27 -04:00
#ifdef CONFIG_IPW2200_PROMISCUOUS
module_param(rtap_iface, int, 0444);
MODULE_PARM_DESC(rtap_iface, "create the rtap interface (1 - create, default 0)");
#endif
#ifdef CONFIG_IPW2200_QOS
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
module_param(qos_enable, int, 0444);
MODULE_PARM_DESC(qos_enable, "enable all QoS functionalitis");
module_param(qos_burst_enable, int, 0444);
MODULE_PARM_DESC(qos_burst_enable, "enable QoS burst mode");
module_param(qos_no_ack_mask, int, 0444);
MODULE_PARM_DESC(qos_no_ack_mask, "mask Tx_Queue to no ack");
2005-03-25 13:31:53 -05:00
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
module_param(burst_duration_CCK, int, 0444);
MODULE_PARM_DESC(burst_duration_CCK, "set CCK burst value");
module_param(burst_duration_OFDM, int, 0444);
MODULE_PARM_DESC(burst_duration_OFDM, "set OFDM burst value");
#endif /* CONFIG_IPW2200_QOS */
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
#ifdef CONFIG_IPW2200_MONITOR
2005-03-25 13:31:53 -05:00
module_param(mode, int, 0444);
MODULE_PARM_DESC(mode, "network mode (0=BSS,1=IBSS,2=Monitor)");
#else
module_param(mode, int, 0444);
MODULE_PARM_DESC(mode, "network mode (0=BSS,1=IBSS)");
#endif
module_param(bt_coexist, int, 0444);
MODULE_PARM_DESC(bt_coexist, "enable bluetooth coexistence (default off)");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
module_param(hwcrypto, int, 0444);
MODULE_PARM_DESC(hwcrypto, "enable hardware crypto (default off)");
Catch ipw2200 up to equivelancy with v1.0.4 * Fixed #627 problem with open APs not working with wpa_supplicant * Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks to Kai Groner) * Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell * Fixed problem with adapter not fully powering off during suspend to RAM or when module unloaded. * Fixed #645 problem with turning fixed rates off not taking effect until you reload the driver * Fixed problem with firmware restart if wpa_supplicant was used to set a key that wasn't exactly 5 or 13 bytes in length. * Fixed #623 Added iwpriv sw_reset extension to reset sw parameters * Added managment frame export to user space with frame statistics * Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is active during driver load * Global s:CX2_:IPW_:g to make code more consistent * Fixed #572 problem with setting txpower to auto * Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200 * Added QoS (CONFIG_IPW_QOS) support. This is being actively developed but is the first step in getting WMM support into the driver and the kernel. * Fixed some race conditions with channel changes, association, and scan abort that could periodically cause a firmware restart. * Added some extensions to export scan and network statistics to user space (exposed through speed_scan and net_stats sysfs entries) * Fixed a few bugs in how monitor mode was supported (scan lists weren't quite right) * Updated the firmware requirement from 2.2 to 2.3 which supports monitor mode. Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
2005-08-24 23:04:42 -04:00
module_param(cmdlog, int, 0444);
MODULE_PARM_DESC(cmdlog,
"allocate a ring buffer for logging firmware commands");
module_param(roaming, int, 0444);
MODULE_PARM_DESC(roaming, "enable roaming support (default on)");
module_param(antenna, int, 0444);
MODULE_PARM_DESC(antenna, "select antenna 1=Main, 3=Aux, default 0 [both], 2=slow_diversity (choose the one with lower background noise)");
2005-03-25 13:31:53 -05:00
module_exit(ipw_exit);
module_init(ipw_init);