android_kernel_xiaomi_sm8350/drivers/net/wireless/iwlwifi/iwl-tx.c

639 lines
18 KiB
C
Raw Normal View History

/******************************************************************************
*
* Copyright(c) 2003 - 2010 Intel Corporation. All rights reserved.
*
* Portions of this file are derived from the ipw3945 project, as well
* as portions of the ieee80211 subsystem header files.
*
* 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
* 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
* more details.
*
* 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.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110, USA
*
* The full GNU General Public License is included in this distribution in the
* file called LICENSE.
*
* Contact Information:
* Intel Linux Wireless <ilw@linux.intel.com>
* Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
*
*****************************************************************************/
#include <linux/etherdevice.h>
#include <linux/sched.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 04:04:11 -04:00
#include <linux/slab.h>
#include <net/mac80211.h>
#include "iwl-eeprom.h"
#include "iwl-dev.h"
#include "iwl-core.h"
#include "iwl-sta.h"
#include "iwl-io.h"
#include "iwl-helpers.h"
/**
* iwl_txq_update_write_ptr - Send new write index to hardware
*/
void iwl_txq_update_write_ptr(struct iwl_priv *priv, struct iwl_tx_queue *txq)
{
u32 reg = 0;
int txq_id = txq->q.id;
if (txq->need_update == 0)
return;
/* if we're trying to save power */
if (test_bit(STATUS_POWER_PMI, &priv->status)) {
/* wake up nic if it's powered down ...
* uCode will wake up, and interrupt us again, so next
* time we'll skip this part. */
reg = iwl_read32(priv, CSR_UCODE_DRV_GP1);
if (reg & CSR_UCODE_DRV_GP1_BIT_MAC_SLEEP) {
IWL_DEBUG_INFO(priv, "Tx queue %d requesting wakeup, GP1 = 0x%x\n",
txq_id, reg);
iwl_set_bit(priv, CSR_GP_CNTRL,
CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
return;
}
iwl_write_direct32(priv, HBUS_TARG_WRPTR,
txq->q.write_ptr | (txq_id << 8));
/* else not in power-save mode, uCode will never sleep when we're
* trying to tx (during RFKILL, we're not trying to tx). */
} else
iwl_write32(priv, HBUS_TARG_WRPTR,
txq->q.write_ptr | (txq_id << 8));
txq->need_update = 0;
}
EXPORT_SYMBOL(iwl_txq_update_write_ptr);
/**
* iwl_tx_queue_free - Deallocate DMA queue.
* @txq: Transmit queue to deallocate.
*
* Empty queue by removing and destroying all BD's.
* Free all buffers.
* 0-fill, but do not free "txq" descriptor structure.
*/
void iwl_tx_queue_free(struct iwl_priv *priv, int txq_id)
{
struct iwl_tx_queue *txq = &priv->txq[txq_id];
struct iwl_queue *q = &txq->q;
struct device *dev = &priv->pci_dev->dev;
int i;
if (q->n_bd == 0)
return;
/* first, empty all BD's */
for (; q->write_ptr != q->read_ptr;
q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd))
priv->cfg->ops->lib->txq_free_tfd(priv, txq);
/* De-alloc array of command/tx buffers */
for (i = 0; i < TFD_TX_CMD_SLOTS; i++)
kfree(txq->cmd[i]);
/* De-alloc circular buffer of TFDs */
if (txq->q.n_bd)
dma_free_coherent(dev, priv->hw_params.tfd_size *
txq->q.n_bd, txq->tfds, txq->q.dma_addr);
/* De-alloc array of per-TFD driver data */
kfree(txq->txb);
txq->txb = NULL;
/* deallocate arrays */
kfree(txq->cmd);
kfree(txq->meta);
txq->cmd = NULL;
txq->meta = NULL;
/* 0-fill queue descriptor structure */
memset(txq, 0, sizeof(*txq));
}
EXPORT_SYMBOL(iwl_tx_queue_free);
/**
* iwl_cmd_queue_free - Deallocate DMA queue.
* @txq: Transmit queue to deallocate.
*
* Empty queue by removing and destroying all BD's.
* Free all buffers.
* 0-fill, but do not free "txq" descriptor structure.
*/
void iwl_cmd_queue_free(struct iwl_priv *priv)
{
struct iwl_tx_queue *txq = &priv->txq[priv->cmd_queue];
struct iwl_queue *q = &txq->q;
struct device *dev = &priv->pci_dev->dev;
int i;
iwlwifi: fix DMA allocation warnings Below warning is triggered sometimes at module removal time when CONFIG_DMA_API_DEBUG is enabled. This should be caused by we didn't unmap pending commands (enqueued, but no complete notification received) for the Tx command queue. [ 1583.107469] ------------[ cut here ]------------ [ 1583.107539] WARNING: at lib/dma-debug.c:688 dma_debug_device_change+0x13c/0x180() [ 1583.107617] Hardware name: ... [ 1583.107664] pci 0000:04:00.0: DMA-API: device driver has pending DMA allocations while released from device [count=1] [ 1583.107713] Modules linked in: ... [ 1583.111661] Pid: 16970, comm: modprobe Tainted: G W 2.6.34-rc1-wl #33 [ 1583.111727] Call Trace: [ 1583.111779] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111833] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111908] [<c0138e11>] warn_slowpath_common+0x71/0xd0 [ 1583.111963] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.112016] [<c0138ebb>] warn_slowpath_fmt+0x2b/0x30 [ 1583.112086] [<c02a281c>] dma_debug_device_change+0x13c/0x180 [ 1583.112142] [<c03e6c33>] notifier_call_chain+0x53/0x90 [ 1583.112198] [<c03e1ebe>] ? down_read+0x6e/0x90 [ 1583.112271] [<c015b229>] __blocking_notifier_call_chain+0x49/0x70 [ 1583.112326] [<c015b26f>] blocking_notifier_call_chain+0x1f/0x30 [ 1583.112380] [<c031931c>] __device_release_driver+0x8c/0xa0 [ 1583.112451] [<c03193bf>] driver_detach+0x8f/0xa0 [ 1583.112538] [<c0318382>] bus_remove_driver+0x82/0x100 [ 1583.112595] [<c0319ad9>] driver_unregister+0x49/0x80 [ 1583.112671] [<c024feb2>] ? sysfs_remove_file+0x12/0x20 [ 1583.112727] [<c02aa292>] pci_unregister_driver+0x32/0x80 [ 1583.112791] [<fc13a3c1>] iwl_exit+0x12/0x19 [iwlagn] [ 1583.112848] [<c017940a>] sys_delete_module+0x15a/0x210 [ 1583.112870] [<c015a5db>] ? up_read+0x1b/0x30 [ 1583.112893] [<c029600c>] ? trace_hardirqs_off_thunk+0xc/0x10 [ 1583.112924] [<c0295ffc>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 1583.112947] [<c03e6a1f>] ? do_page_fault+0x1ff/0x3c0 [ 1583.112978] [<c03e36f6>] ? restore_all_notrace+0x0/0x18 [ 1583.113002] [<c016aa70>] ? trace_hardirqs_on_caller+0x20/0x190 [ 1583.113025] [<c0102d58>] sysenter_do_call+0x12/0x38 [ 1583.113054] ---[ end trace fc23e059cc4c2ced ]--- Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
2010-03-22 05:28:41 -04:00
bool huge = false;
if (q->n_bd == 0)
return;
iwlwifi: fix DMA allocation warnings Below warning is triggered sometimes at module removal time when CONFIG_DMA_API_DEBUG is enabled. This should be caused by we didn't unmap pending commands (enqueued, but no complete notification received) for the Tx command queue. [ 1583.107469] ------------[ cut here ]------------ [ 1583.107539] WARNING: at lib/dma-debug.c:688 dma_debug_device_change+0x13c/0x180() [ 1583.107617] Hardware name: ... [ 1583.107664] pci 0000:04:00.0: DMA-API: device driver has pending DMA allocations while released from device [count=1] [ 1583.107713] Modules linked in: ... [ 1583.111661] Pid: 16970, comm: modprobe Tainted: G W 2.6.34-rc1-wl #33 [ 1583.111727] Call Trace: [ 1583.111779] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111833] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111908] [<c0138e11>] warn_slowpath_common+0x71/0xd0 [ 1583.111963] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.112016] [<c0138ebb>] warn_slowpath_fmt+0x2b/0x30 [ 1583.112086] [<c02a281c>] dma_debug_device_change+0x13c/0x180 [ 1583.112142] [<c03e6c33>] notifier_call_chain+0x53/0x90 [ 1583.112198] [<c03e1ebe>] ? down_read+0x6e/0x90 [ 1583.112271] [<c015b229>] __blocking_notifier_call_chain+0x49/0x70 [ 1583.112326] [<c015b26f>] blocking_notifier_call_chain+0x1f/0x30 [ 1583.112380] [<c031931c>] __device_release_driver+0x8c/0xa0 [ 1583.112451] [<c03193bf>] driver_detach+0x8f/0xa0 [ 1583.112538] [<c0318382>] bus_remove_driver+0x82/0x100 [ 1583.112595] [<c0319ad9>] driver_unregister+0x49/0x80 [ 1583.112671] [<c024feb2>] ? sysfs_remove_file+0x12/0x20 [ 1583.112727] [<c02aa292>] pci_unregister_driver+0x32/0x80 [ 1583.112791] [<fc13a3c1>] iwl_exit+0x12/0x19 [iwlagn] [ 1583.112848] [<c017940a>] sys_delete_module+0x15a/0x210 [ 1583.112870] [<c015a5db>] ? up_read+0x1b/0x30 [ 1583.112893] [<c029600c>] ? trace_hardirqs_off_thunk+0xc/0x10 [ 1583.112924] [<c0295ffc>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 1583.112947] [<c03e6a1f>] ? do_page_fault+0x1ff/0x3c0 [ 1583.112978] [<c03e36f6>] ? restore_all_notrace+0x0/0x18 [ 1583.113002] [<c016aa70>] ? trace_hardirqs_on_caller+0x20/0x190 [ 1583.113025] [<c0102d58>] sysenter_do_call+0x12/0x38 [ 1583.113054] ---[ end trace fc23e059cc4c2ced ]--- Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
2010-03-22 05:28:41 -04:00
for (; q->read_ptr != q->write_ptr;
q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd)) {
/* we have no way to tell if it is a huge cmd ATM */
i = get_cmd_index(q, q->read_ptr, 0);
if (txq->meta[i].flags & CMD_SIZE_HUGE) {
huge = true;
continue;
}
pci_unmap_single(priv->pci_dev,
dma_unmap_addr(&txq->meta[i], mapping),
dma_unmap_len(&txq->meta[i], len),
iwlwifi: fix DMA allocation warnings Below warning is triggered sometimes at module removal time when CONFIG_DMA_API_DEBUG is enabled. This should be caused by we didn't unmap pending commands (enqueued, but no complete notification received) for the Tx command queue. [ 1583.107469] ------------[ cut here ]------------ [ 1583.107539] WARNING: at lib/dma-debug.c:688 dma_debug_device_change+0x13c/0x180() [ 1583.107617] Hardware name: ... [ 1583.107664] pci 0000:04:00.0: DMA-API: device driver has pending DMA allocations while released from device [count=1] [ 1583.107713] Modules linked in: ... [ 1583.111661] Pid: 16970, comm: modprobe Tainted: G W 2.6.34-rc1-wl #33 [ 1583.111727] Call Trace: [ 1583.111779] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111833] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111908] [<c0138e11>] warn_slowpath_common+0x71/0xd0 [ 1583.111963] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.112016] [<c0138ebb>] warn_slowpath_fmt+0x2b/0x30 [ 1583.112086] [<c02a281c>] dma_debug_device_change+0x13c/0x180 [ 1583.112142] [<c03e6c33>] notifier_call_chain+0x53/0x90 [ 1583.112198] [<c03e1ebe>] ? down_read+0x6e/0x90 [ 1583.112271] [<c015b229>] __blocking_notifier_call_chain+0x49/0x70 [ 1583.112326] [<c015b26f>] blocking_notifier_call_chain+0x1f/0x30 [ 1583.112380] [<c031931c>] __device_release_driver+0x8c/0xa0 [ 1583.112451] [<c03193bf>] driver_detach+0x8f/0xa0 [ 1583.112538] [<c0318382>] bus_remove_driver+0x82/0x100 [ 1583.112595] [<c0319ad9>] driver_unregister+0x49/0x80 [ 1583.112671] [<c024feb2>] ? sysfs_remove_file+0x12/0x20 [ 1583.112727] [<c02aa292>] pci_unregister_driver+0x32/0x80 [ 1583.112791] [<fc13a3c1>] iwl_exit+0x12/0x19 [iwlagn] [ 1583.112848] [<c017940a>] sys_delete_module+0x15a/0x210 [ 1583.112870] [<c015a5db>] ? up_read+0x1b/0x30 [ 1583.112893] [<c029600c>] ? trace_hardirqs_off_thunk+0xc/0x10 [ 1583.112924] [<c0295ffc>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 1583.112947] [<c03e6a1f>] ? do_page_fault+0x1ff/0x3c0 [ 1583.112978] [<c03e36f6>] ? restore_all_notrace+0x0/0x18 [ 1583.113002] [<c016aa70>] ? trace_hardirqs_on_caller+0x20/0x190 [ 1583.113025] [<c0102d58>] sysenter_do_call+0x12/0x38 [ 1583.113054] ---[ end trace fc23e059cc4c2ced ]--- Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
2010-03-22 05:28:41 -04:00
PCI_DMA_BIDIRECTIONAL);
}
if (huge) {
i = q->n_window;
pci_unmap_single(priv->pci_dev,
dma_unmap_addr(&txq->meta[i], mapping),
dma_unmap_len(&txq->meta[i], len),
iwlwifi: fix DMA allocation warnings Below warning is triggered sometimes at module removal time when CONFIG_DMA_API_DEBUG is enabled. This should be caused by we didn't unmap pending commands (enqueued, but no complete notification received) for the Tx command queue. [ 1583.107469] ------------[ cut here ]------------ [ 1583.107539] WARNING: at lib/dma-debug.c:688 dma_debug_device_change+0x13c/0x180() [ 1583.107617] Hardware name: ... [ 1583.107664] pci 0000:04:00.0: DMA-API: device driver has pending DMA allocations while released from device [count=1] [ 1583.107713] Modules linked in: ... [ 1583.111661] Pid: 16970, comm: modprobe Tainted: G W 2.6.34-rc1-wl #33 [ 1583.111727] Call Trace: [ 1583.111779] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111833] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111908] [<c0138e11>] warn_slowpath_common+0x71/0xd0 [ 1583.111963] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.112016] [<c0138ebb>] warn_slowpath_fmt+0x2b/0x30 [ 1583.112086] [<c02a281c>] dma_debug_device_change+0x13c/0x180 [ 1583.112142] [<c03e6c33>] notifier_call_chain+0x53/0x90 [ 1583.112198] [<c03e1ebe>] ? down_read+0x6e/0x90 [ 1583.112271] [<c015b229>] __blocking_notifier_call_chain+0x49/0x70 [ 1583.112326] [<c015b26f>] blocking_notifier_call_chain+0x1f/0x30 [ 1583.112380] [<c031931c>] __device_release_driver+0x8c/0xa0 [ 1583.112451] [<c03193bf>] driver_detach+0x8f/0xa0 [ 1583.112538] [<c0318382>] bus_remove_driver+0x82/0x100 [ 1583.112595] [<c0319ad9>] driver_unregister+0x49/0x80 [ 1583.112671] [<c024feb2>] ? sysfs_remove_file+0x12/0x20 [ 1583.112727] [<c02aa292>] pci_unregister_driver+0x32/0x80 [ 1583.112791] [<fc13a3c1>] iwl_exit+0x12/0x19 [iwlagn] [ 1583.112848] [<c017940a>] sys_delete_module+0x15a/0x210 [ 1583.112870] [<c015a5db>] ? up_read+0x1b/0x30 [ 1583.112893] [<c029600c>] ? trace_hardirqs_off_thunk+0xc/0x10 [ 1583.112924] [<c0295ffc>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 1583.112947] [<c03e6a1f>] ? do_page_fault+0x1ff/0x3c0 [ 1583.112978] [<c03e36f6>] ? restore_all_notrace+0x0/0x18 [ 1583.113002] [<c016aa70>] ? trace_hardirqs_on_caller+0x20/0x190 [ 1583.113025] [<c0102d58>] sysenter_do_call+0x12/0x38 [ 1583.113054] ---[ end trace fc23e059cc4c2ced ]--- Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
2010-03-22 05:28:41 -04:00
PCI_DMA_BIDIRECTIONAL);
}
/* De-alloc array of command/tx buffers */
for (i = 0; i <= TFD_CMD_SLOTS; i++)
kfree(txq->cmd[i]);
/* De-alloc circular buffer of TFDs */
if (txq->q.n_bd)
dma_free_coherent(dev, priv->hw_params.tfd_size * txq->q.n_bd,
txq->tfds, txq->q.dma_addr);
/* deallocate arrays */
kfree(txq->cmd);
kfree(txq->meta);
txq->cmd = NULL;
txq->meta = NULL;
/* 0-fill queue descriptor structure */
memset(txq, 0, sizeof(*txq));
}
EXPORT_SYMBOL(iwl_cmd_queue_free);
/*************** DMA-QUEUE-GENERAL-FUNCTIONS *****
* DMA services
*
* Theory of operation
*
* A Tx or Rx queue resides in host DRAM, and is comprised of a circular buffer
* of buffer descriptors, each of which points to one or more data buffers for
* the device to read from or fill. Driver and device exchange status of each
* queue via "read" and "write" pointers. Driver keeps minimum of 2 empty
* entries in each circular buffer, to protect against confusing empty and full
* queue states.
*
* The device reads or writes the data in the queues via the device's several
* DMA/FIFO channels. Each queue is mapped to a single DMA channel.
*
* 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,
* Tx queue resumed.
*
* See more detailed info in iwl-4965-hw.h.
***************************************************/
int iwl_queue_space(const struct iwl_queue *q)
{
int s = q->read_ptr - q->write_ptr;
if (q->read_ptr > q->write_ptr)
s -= q->n_bd;
if (s <= 0)
s += q->n_window;
/* keep some reserve to not confuse empty and full situations */
s -= 2;
if (s < 0)
s = 0;
return s;
}
EXPORT_SYMBOL(iwl_queue_space);
/**
* iwl_queue_init - Initialize queue's high/low-water and read/write indexes
*/
static int iwl_queue_init(struct iwl_priv *priv, struct iwl_queue *q,
int count, int slots_num, u32 id)
{
q->n_bd = count;
q->n_window = slots_num;
q->id = id;
/* count must be power-of-two size, otherwise iwl_queue_inc_wrap
* and iwl_queue_dec_wrap are broken. */
BUG_ON(!is_power_of_2(count));
/* slots_num must be power-of-two size, otherwise
* get_cmd_index is broken. */
BUG_ON(!is_power_of_2(slots_num));
q->low_mark = q->n_window / 4;
if (q->low_mark < 4)
q->low_mark = 4;
q->high_mark = q->n_window / 8;
if (q->high_mark < 2)
q->high_mark = 2;
q->write_ptr = q->read_ptr = 0;
q->last_read_ptr = 0;
q->repeat_same_read_ptr = 0;
return 0;
}
/**
* iwl_tx_queue_alloc - Alloc driver data and TFD CB for one Tx/cmd queue
*/
static int iwl_tx_queue_alloc(struct iwl_priv *priv,
struct iwl_tx_queue *txq, u32 id)
{
struct device *dev = &priv->pci_dev->dev;
size_t tfd_sz = priv->hw_params.tfd_size * TFD_QUEUE_SIZE_MAX;
/* Driver private data, only for Tx (not command) queues,
* not shared with device. */
if (id != priv->cmd_queue) {
txq->txb = kzalloc(sizeof(txq->txb[0]) *
TFD_QUEUE_SIZE_MAX, GFP_KERNEL);
if (!txq->txb) {
IWL_ERR(priv, "kmalloc for auxiliary BD "
"structures failed\n");
goto error;
}
} else {
txq->txb = NULL;
}
/* Circular buffer of transmit frame descriptors (TFDs),
* shared with device */
txq->tfds = dma_alloc_coherent(dev, tfd_sz, &txq->q.dma_addr,
GFP_KERNEL);
if (!txq->tfds) {
IWL_ERR(priv, "pci_alloc_consistent(%zd) failed\n", tfd_sz);
goto error;
}
txq->q.id = id;
return 0;
error:
kfree(txq->txb);
txq->txb = NULL;
return -ENOMEM;
}
/**
* iwl_tx_queue_init - Allocate and initialize one tx/cmd queue
*/
int iwl_tx_queue_init(struct iwl_priv *priv, struct iwl_tx_queue *txq,
int slots_num, u32 txq_id)
{
int i, len;
int ret;
int actual_slots = slots_num;
/*
* Alloc buffer array for commands (Tx or other types of commands).
* For the command queue (#4/#9), allocate command space + one big
* command for scan, since scan command is very huge; the system will
* not have two scans at the same time, so only one is needed.
* For normal Tx queues (all other queues), no super-size command
* space is needed.
*/
if (txq_id == priv->cmd_queue)
actual_slots++;
txq->meta = kzalloc(sizeof(struct iwl_cmd_meta) * actual_slots,
GFP_KERNEL);
txq->cmd = kzalloc(sizeof(struct iwl_device_cmd *) * actual_slots,
GFP_KERNEL);
if (!txq->meta || !txq->cmd)
goto out_free_arrays;
len = sizeof(struct iwl_device_cmd);
for (i = 0; i < actual_slots; i++) {
/* only happens for cmd queue */
if (i == slots_num)
len = IWL_MAX_CMD_SIZE;
txq->cmd[i] = kmalloc(len, GFP_KERNEL);
if (!txq->cmd[i])
goto err;
}
/* Alloc driver data array and TFD circular buffer */
ret = iwl_tx_queue_alloc(priv, txq, txq_id);
if (ret)
goto err;
txq->need_update = 0;
/*
* Aggregation TX queues will get their ID when aggregation begins;
* they overwrite the setting done here. The command FIFO doesn't
* need an swq_id so don't set one to catch errors, all others can
* be set up to the identity mapping.
*/
if (txq_id != priv->cmd_queue)
txq->swq_id = txq_id;
/* TFD_QUEUE_SIZE_MAX must be power-of-two size, otherwise
* iwl_queue_inc_wrap and iwl_queue_dec_wrap are broken. */
BUILD_BUG_ON(TFD_QUEUE_SIZE_MAX & (TFD_QUEUE_SIZE_MAX - 1));
/* Initialize queue's high/low-water marks, and head/tail indexes */
iwl_queue_init(priv, &txq->q, TFD_QUEUE_SIZE_MAX, slots_num, txq_id);
/* Tell device where to find queue */
priv->cfg->ops->lib->txq_init(priv, txq);
return 0;
err:
for (i = 0; i < actual_slots; i++)
kfree(txq->cmd[i]);
out_free_arrays:
kfree(txq->meta);
kfree(txq->cmd);
return -ENOMEM;
}
EXPORT_SYMBOL(iwl_tx_queue_init);
void iwl_tx_queue_reset(struct iwl_priv *priv, struct iwl_tx_queue *txq,
int slots_num, u32 txq_id)
{
int actual_slots = slots_num;
if (txq_id == priv->cmd_queue)
actual_slots++;
memset(txq->meta, 0, sizeof(struct iwl_cmd_meta) * actual_slots);
txq->need_update = 0;
/* Initialize queue's high/low-water marks, and head/tail indexes */
iwl_queue_init(priv, &txq->q, TFD_QUEUE_SIZE_MAX, slots_num, txq_id);
/* Tell device where to find queue */
priv->cfg->ops->lib->txq_init(priv, txq);
}
EXPORT_SYMBOL(iwl_tx_queue_reset);
/*************** HOST COMMAND QUEUE FUNCTIONS *****/
/**
* iwl_enqueue_hcmd - enqueue a uCode command
* @priv: device private data point
* @cmd: a point to the ucode command structure
*
* The function returns < 0 values to indicate the operation is
* failed. On success, it turns the index (> 0) of command in the
* command queue.
*/
int iwl_enqueue_hcmd(struct iwl_priv *priv, struct iwl_host_cmd *cmd)
{
struct iwl_tx_queue *txq = &priv->txq[priv->cmd_queue];
struct iwl_queue *q = &txq->q;
struct iwl_device_cmd *out_cmd;
struct iwl_cmd_meta *out_meta;
dma_addr_t phys_addr;
unsigned long flags;
int len;
u32 idx;
u16 fix_size;
bool is_ct_kill = false;
cmd->len = priv->cfg->ops->utils->get_hcmd_size(cmd->id, cmd->len);
fix_size = (u16)(cmd->len + sizeof(out_cmd->hdr));
/* If any of the command structures end up being larger than
* the TFD_MAX_PAYLOAD_SIZE, and it sent as a 'small' command then
* we will need to increase the size of the TFD entries
* Also, check to see if command buffer should not exceed the size
* of device_cmd and max_cmd_size. */
BUG_ON((fix_size > TFD_MAX_PAYLOAD_SIZE) &&
!(cmd->flags & CMD_SIZE_HUGE));
BUG_ON(fix_size > IWL_MAX_CMD_SIZE);
iwlwifi: reliable entering of critical temperature state When uCode detects critical temperature it should send "card state notification" interrupt to driver and then shut itself down to prevent overheating. There is a race condition where uCode shuts down before it can deliver the interrupt to driver. Additional method provided here for driver to enter CT_KILL state based on temperature reading. How it works: Method 1: If driver receive "card state notification" interrupt from uCode; it enters "CT_KILL" state immediately Method 2: If the last temperature report by Card reach Critical temperature, driver will send "statistic notification" request to uCode to verify the temperature reading, if driver can not get reply from uCode within 300ms, driver will enter CT_KILL state automatically. Method 3: If the last temperature report by Card did not reach Critical temperature, but uCode already shut down due to critical temperature. All the host commands send to uCode will not get process by uCode; when command queue reach the limit, driver will check the last reported temperature reading, if it is within pre-defined margin, enter "CT_KILL" state immediately. In this case, when uCode ready to exit from "CT_KILL" state, driver need to restart the adapter in order to reset all the queues and resume normal operation. One additional issue being address here, when system is in CT_KILL state, both tx and rx already stopped, but driver still can send host command to uCode, it will flood the command queue since card was not responding; adding STATUS_CT_KILL flag to reject enqueue host commands to uCode if it is in CT_KILL state, when uCode is ready to come out of CT_KILL, driver will clear the STATUS_CT_KILL bit and allow enqueue the host commands to uCode to recover from CT_KILL state. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-10-02 16:43:58 -04:00
if (iwl_is_rfkill(priv) || iwl_is_ctkill(priv)) {
IWL_WARN(priv, "Not sending command - %s KILL\n",
iwl_is_rfkill(priv) ? "RF" : "CT");
return -EIO;
}
if (iwl_queue_space(q) < ((cmd->flags & CMD_ASYNC) ? 2 : 1)) {
IWL_ERR(priv, "No space in command queue\n");
if (priv->cfg->ops->lib->tt_ops.ct_kill_check) {
is_ct_kill =
priv->cfg->ops->lib->tt_ops.ct_kill_check(priv);
}
if (!is_ct_kill) {
iwlwifi: reliable entering of critical temperature state When uCode detects critical temperature it should send "card state notification" interrupt to driver and then shut itself down to prevent overheating. There is a race condition where uCode shuts down before it can deliver the interrupt to driver. Additional method provided here for driver to enter CT_KILL state based on temperature reading. How it works: Method 1: If driver receive "card state notification" interrupt from uCode; it enters "CT_KILL" state immediately Method 2: If the last temperature report by Card reach Critical temperature, driver will send "statistic notification" request to uCode to verify the temperature reading, if driver can not get reply from uCode within 300ms, driver will enter CT_KILL state automatically. Method 3: If the last temperature report by Card did not reach Critical temperature, but uCode already shut down due to critical temperature. All the host commands send to uCode will not get process by uCode; when command queue reach the limit, driver will check the last reported temperature reading, if it is within pre-defined margin, enter "CT_KILL" state immediately. In this case, when uCode ready to exit from "CT_KILL" state, driver need to restart the adapter in order to reset all the queues and resume normal operation. One additional issue being address here, when system is in CT_KILL state, both tx and rx already stopped, but driver still can send host command to uCode, it will flood the command queue since card was not responding; adding STATUS_CT_KILL flag to reject enqueue host commands to uCode if it is in CT_KILL state, when uCode is ready to come out of CT_KILL, driver will clear the STATUS_CT_KILL bit and allow enqueue the host commands to uCode to recover from CT_KILL state. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-10-02 16:43:58 -04:00
IWL_ERR(priv, "Restarting adapter due to queue full\n");
queue_work(priv->workqueue, &priv->restart);
}
return -ENOSPC;
}
spin_lock_irqsave(&priv->hcmd_lock, flags);
iwlwifi: fix DMA allocation warnings Below warning is triggered sometimes at module removal time when CONFIG_DMA_API_DEBUG is enabled. This should be caused by we didn't unmap pending commands (enqueued, but no complete notification received) for the Tx command queue. [ 1583.107469] ------------[ cut here ]------------ [ 1583.107539] WARNING: at lib/dma-debug.c:688 dma_debug_device_change+0x13c/0x180() [ 1583.107617] Hardware name: ... [ 1583.107664] pci 0000:04:00.0: DMA-API: device driver has pending DMA allocations while released from device [count=1] [ 1583.107713] Modules linked in: ... [ 1583.111661] Pid: 16970, comm: modprobe Tainted: G W 2.6.34-rc1-wl #33 [ 1583.111727] Call Trace: [ 1583.111779] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111833] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111908] [<c0138e11>] warn_slowpath_common+0x71/0xd0 [ 1583.111963] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.112016] [<c0138ebb>] warn_slowpath_fmt+0x2b/0x30 [ 1583.112086] [<c02a281c>] dma_debug_device_change+0x13c/0x180 [ 1583.112142] [<c03e6c33>] notifier_call_chain+0x53/0x90 [ 1583.112198] [<c03e1ebe>] ? down_read+0x6e/0x90 [ 1583.112271] [<c015b229>] __blocking_notifier_call_chain+0x49/0x70 [ 1583.112326] [<c015b26f>] blocking_notifier_call_chain+0x1f/0x30 [ 1583.112380] [<c031931c>] __device_release_driver+0x8c/0xa0 [ 1583.112451] [<c03193bf>] driver_detach+0x8f/0xa0 [ 1583.112538] [<c0318382>] bus_remove_driver+0x82/0x100 [ 1583.112595] [<c0319ad9>] driver_unregister+0x49/0x80 [ 1583.112671] [<c024feb2>] ? sysfs_remove_file+0x12/0x20 [ 1583.112727] [<c02aa292>] pci_unregister_driver+0x32/0x80 [ 1583.112791] [<fc13a3c1>] iwl_exit+0x12/0x19 [iwlagn] [ 1583.112848] [<c017940a>] sys_delete_module+0x15a/0x210 [ 1583.112870] [<c015a5db>] ? up_read+0x1b/0x30 [ 1583.112893] [<c029600c>] ? trace_hardirqs_off_thunk+0xc/0x10 [ 1583.112924] [<c0295ffc>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 1583.112947] [<c03e6a1f>] ? do_page_fault+0x1ff/0x3c0 [ 1583.112978] [<c03e36f6>] ? restore_all_notrace+0x0/0x18 [ 1583.113002] [<c016aa70>] ? trace_hardirqs_on_caller+0x20/0x190 [ 1583.113025] [<c0102d58>] sysenter_do_call+0x12/0x38 [ 1583.113054] ---[ end trace fc23e059cc4c2ced ]--- Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
2010-03-22 05:28:41 -04:00
/* If this is a huge cmd, mark the huge flag also on the meta.flags
* of the _original_ cmd. This is used for DMA mapping clean up.
*/
if (cmd->flags & CMD_SIZE_HUGE) {
idx = get_cmd_index(q, q->write_ptr, 0);
txq->meta[idx].flags = CMD_SIZE_HUGE;
}
idx = get_cmd_index(q, q->write_ptr, cmd->flags & CMD_SIZE_HUGE);
out_cmd = txq->cmd[idx];
out_meta = &txq->meta[idx];
memset(out_meta, 0, sizeof(*out_meta)); /* re-initialize to NULL */
out_meta->flags = cmd->flags;
if (cmd->flags & CMD_WANT_SKB)
out_meta->source = cmd;
if (cmd->flags & CMD_ASYNC)
out_meta->callback = cmd->callback;
out_cmd->hdr.cmd = cmd->id;
memcpy(&out_cmd->cmd.payload, cmd->data, cmd->len);
/* At this point, the out_cmd now has all of the incoming cmd
* information */
out_cmd->hdr.flags = 0;
out_cmd->hdr.sequence = cpu_to_le16(QUEUE_TO_SEQ(priv->cmd_queue) |
INDEX_TO_SEQ(q->write_ptr));
if (cmd->flags & CMD_SIZE_HUGE)
out_cmd->hdr.sequence |= SEQ_HUGE_FRAME;
len = sizeof(struct iwl_device_cmd);
if (idx == TFD_CMD_SLOTS)
len = IWL_MAX_CMD_SIZE;
#ifdef CONFIG_IWLWIFI_DEBUG
switch (out_cmd->hdr.cmd) {
case REPLY_TX_LINK_QUALITY_CMD:
case SENSITIVITY_CMD:
IWL_DEBUG_HC_DUMP(priv, "Sending command %s (#%x), seq: 0x%04X, "
"%d bytes at %d[%d]:%d\n",
get_cmd_string(out_cmd->hdr.cmd),
out_cmd->hdr.cmd,
le16_to_cpu(out_cmd->hdr.sequence), fix_size,
q->write_ptr, idx, priv->cmd_queue);
break;
default:
IWL_DEBUG_HC(priv, "Sending command %s (#%x), seq: 0x%04X, "
"%d bytes at %d[%d]:%d\n",
get_cmd_string(out_cmd->hdr.cmd),
out_cmd->hdr.cmd,
le16_to_cpu(out_cmd->hdr.sequence), fix_size,
q->write_ptr, idx, priv->cmd_queue);
}
#endif
txq->need_update = 1;
if (priv->cfg->ops->lib->txq_update_byte_cnt_tbl)
/* Set up entry in queue's byte count circular buffer */
priv->cfg->ops->lib->txq_update_byte_cnt_tbl(priv, txq, 0);
phys_addr = pci_map_single(priv->pci_dev, &out_cmd->hdr,
fix_size, PCI_DMA_BIDIRECTIONAL);
dma_unmap_addr_set(out_meta, mapping, phys_addr);
dma_unmap_len_set(out_meta, len, fix_size);
trace_iwlwifi_dev_hcmd(priv, &out_cmd->hdr, fix_size, cmd->flags);
priv->cfg->ops->lib->txq_attach_buf_to_tfd(priv, txq,
phys_addr, fix_size, 1,
U32_PAD(cmd->len));
/* Increment and update queue's write index */
q->write_ptr = iwl_queue_inc_wrap(q->write_ptr, q->n_bd);
iwl_txq_update_write_ptr(priv, txq);
spin_unlock_irqrestore(&priv->hcmd_lock, flags);
return idx;
}
/**
* iwl_hcmd_queue_reclaim - Reclaim TX command queue entries already Tx'd
*
* When FW advances '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 the stack that feeds us.
*/
static void iwl_hcmd_queue_reclaim(struct iwl_priv *priv, int txq_id,
int idx, int cmd_idx)
{
struct iwl_tx_queue *txq = &priv->txq[txq_id];
struct iwl_queue *q = &txq->q;
int nfreed = 0;
if ((idx >= q->n_bd) || (iwl_queue_used(q, idx) == 0)) {
IWL_ERR(priv, "Read index for DMA queue txq id (%d), index %d, "
"is out of range [0-%d] %d %d.\n", txq_id,
idx, q->n_bd, q->write_ptr, q->read_ptr);
return;
}
for (idx = iwl_queue_inc_wrap(idx, q->n_bd); q->read_ptr != idx;
q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd)) {
if (nfreed++ > 0) {
IWL_ERR(priv, "HCMD skipped: index (%d) %d %d\n", idx,
q->write_ptr, q->read_ptr);
queue_work(priv->workqueue, &priv->restart);
}
}
}
/**
* iwl_tx_cmd_complete - Pull unused buffers off the queue and reclaim them
* @rxb: Rx buffer to reclaim
*
* If an Rx buffer has an async callback associated with it the callback
* will be executed. The attached skb (if present) will only be freed
* if the callback returns 1
*/
void iwl_tx_cmd_complete(struct iwl_priv *priv, struct iwl_rx_mem_buffer *rxb)
{
struct iwl_rx_packet *pkt = rxb_addr(rxb);
u16 sequence = le16_to_cpu(pkt->hdr.sequence);
int txq_id = SEQ_TO_QUEUE(sequence);
int index = SEQ_TO_INDEX(sequence);
int cmd_index;
bool huge = !!(pkt->hdr.sequence & SEQ_HUGE_FRAME);
struct iwl_device_cmd *cmd;
struct iwl_cmd_meta *meta;
struct iwl_tx_queue *txq = &priv->txq[priv->cmd_queue];
/* If a Tx command is being handled and it isn't in the actual
* command queue then there a command routing bug has been introduced
* in the queue management code. */
if (WARN(txq_id != priv->cmd_queue,
"wrong command queue %d (should be %d), sequence 0x%X readp=%d writep=%d\n",
txq_id, priv->cmd_queue, sequence,
priv->txq[priv->cmd_queue].q.read_ptr,
priv->txq[priv->cmd_queue].q.write_ptr)) {
iwl_print_hex_error(priv, pkt, 32);
return;
}
iwlwifi: fix DMA allocation warnings Below warning is triggered sometimes at module removal time when CONFIG_DMA_API_DEBUG is enabled. This should be caused by we didn't unmap pending commands (enqueued, but no complete notification received) for the Tx command queue. [ 1583.107469] ------------[ cut here ]------------ [ 1583.107539] WARNING: at lib/dma-debug.c:688 dma_debug_device_change+0x13c/0x180() [ 1583.107617] Hardware name: ... [ 1583.107664] pci 0000:04:00.0: DMA-API: device driver has pending DMA allocations while released from device [count=1] [ 1583.107713] Modules linked in: ... [ 1583.111661] Pid: 16970, comm: modprobe Tainted: G W 2.6.34-rc1-wl #33 [ 1583.111727] Call Trace: [ 1583.111779] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111833] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111908] [<c0138e11>] warn_slowpath_common+0x71/0xd0 [ 1583.111963] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.112016] [<c0138ebb>] warn_slowpath_fmt+0x2b/0x30 [ 1583.112086] [<c02a281c>] dma_debug_device_change+0x13c/0x180 [ 1583.112142] [<c03e6c33>] notifier_call_chain+0x53/0x90 [ 1583.112198] [<c03e1ebe>] ? down_read+0x6e/0x90 [ 1583.112271] [<c015b229>] __blocking_notifier_call_chain+0x49/0x70 [ 1583.112326] [<c015b26f>] blocking_notifier_call_chain+0x1f/0x30 [ 1583.112380] [<c031931c>] __device_release_driver+0x8c/0xa0 [ 1583.112451] [<c03193bf>] driver_detach+0x8f/0xa0 [ 1583.112538] [<c0318382>] bus_remove_driver+0x82/0x100 [ 1583.112595] [<c0319ad9>] driver_unregister+0x49/0x80 [ 1583.112671] [<c024feb2>] ? sysfs_remove_file+0x12/0x20 [ 1583.112727] [<c02aa292>] pci_unregister_driver+0x32/0x80 [ 1583.112791] [<fc13a3c1>] iwl_exit+0x12/0x19 [iwlagn] [ 1583.112848] [<c017940a>] sys_delete_module+0x15a/0x210 [ 1583.112870] [<c015a5db>] ? up_read+0x1b/0x30 [ 1583.112893] [<c029600c>] ? trace_hardirqs_off_thunk+0xc/0x10 [ 1583.112924] [<c0295ffc>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 1583.112947] [<c03e6a1f>] ? do_page_fault+0x1ff/0x3c0 [ 1583.112978] [<c03e36f6>] ? restore_all_notrace+0x0/0x18 [ 1583.113002] [<c016aa70>] ? trace_hardirqs_on_caller+0x20/0x190 [ 1583.113025] [<c0102d58>] sysenter_do_call+0x12/0x38 [ 1583.113054] ---[ end trace fc23e059cc4c2ced ]--- Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
2010-03-22 05:28:41 -04:00
/* If this is a huge cmd, clear the huge flag on the meta.flags
* of the _original_ cmd. So that iwl_cmd_queue_free won't unmap
* the DMA buffer for the scan (huge) command.
*/
if (huge) {
cmd_index = get_cmd_index(&txq->q, index, 0);
txq->meta[cmd_index].flags = 0;
}
cmd_index = get_cmd_index(&txq->q, index, huge);
cmd = txq->cmd[cmd_index];
meta = &txq->meta[cmd_index];
pci_unmap_single(priv->pci_dev,
dma_unmap_addr(meta, mapping),
dma_unmap_len(meta, len),
PCI_DMA_BIDIRECTIONAL);
/* Input error checking is done when commands are added to queue. */
if (meta->flags & CMD_WANT_SKB) {
meta->source->reply_page = (unsigned long)rxb_addr(rxb);
rxb->page = NULL;
} else if (meta->callback)
meta->callback(priv, cmd, pkt);
iwl_hcmd_queue_reclaim(priv, txq_id, index, cmd_index);
if (!(meta->flags & CMD_ASYNC)) {
clear_bit(STATUS_HCMD_ACTIVE, &priv->status);
IWL_DEBUG_INFO(priv, "Clearing HCMD_ACTIVE for command %s\n",
get_cmd_string(cmd->hdr.cmd));
wake_up_interruptible(&priv->wait_command_queue);
}
iwlwifi: fix DMA allocation warnings Below warning is triggered sometimes at module removal time when CONFIG_DMA_API_DEBUG is enabled. This should be caused by we didn't unmap pending commands (enqueued, but no complete notification received) for the Tx command queue. [ 1583.107469] ------------[ cut here ]------------ [ 1583.107539] WARNING: at lib/dma-debug.c:688 dma_debug_device_change+0x13c/0x180() [ 1583.107617] Hardware name: ... [ 1583.107664] pci 0000:04:00.0: DMA-API: device driver has pending DMA allocations while released from device [count=1] [ 1583.107713] Modules linked in: ... [ 1583.111661] Pid: 16970, comm: modprobe Tainted: G W 2.6.34-rc1-wl #33 [ 1583.111727] Call Trace: [ 1583.111779] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111833] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.111908] [<c0138e11>] warn_slowpath_common+0x71/0xd0 [ 1583.111963] [<c02a281c>] ? dma_debug_device_change+0x13c/0x180 [ 1583.112016] [<c0138ebb>] warn_slowpath_fmt+0x2b/0x30 [ 1583.112086] [<c02a281c>] dma_debug_device_change+0x13c/0x180 [ 1583.112142] [<c03e6c33>] notifier_call_chain+0x53/0x90 [ 1583.112198] [<c03e1ebe>] ? down_read+0x6e/0x90 [ 1583.112271] [<c015b229>] __blocking_notifier_call_chain+0x49/0x70 [ 1583.112326] [<c015b26f>] blocking_notifier_call_chain+0x1f/0x30 [ 1583.112380] [<c031931c>] __device_release_driver+0x8c/0xa0 [ 1583.112451] [<c03193bf>] driver_detach+0x8f/0xa0 [ 1583.112538] [<c0318382>] bus_remove_driver+0x82/0x100 [ 1583.112595] [<c0319ad9>] driver_unregister+0x49/0x80 [ 1583.112671] [<c024feb2>] ? sysfs_remove_file+0x12/0x20 [ 1583.112727] [<c02aa292>] pci_unregister_driver+0x32/0x80 [ 1583.112791] [<fc13a3c1>] iwl_exit+0x12/0x19 [iwlagn] [ 1583.112848] [<c017940a>] sys_delete_module+0x15a/0x210 [ 1583.112870] [<c015a5db>] ? up_read+0x1b/0x30 [ 1583.112893] [<c029600c>] ? trace_hardirqs_off_thunk+0xc/0x10 [ 1583.112924] [<c0295ffc>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 1583.112947] [<c03e6a1f>] ? do_page_fault+0x1ff/0x3c0 [ 1583.112978] [<c03e36f6>] ? restore_all_notrace+0x0/0x18 [ 1583.113002] [<c016aa70>] ? trace_hardirqs_on_caller+0x20/0x190 [ 1583.113025] [<c0102d58>] sysenter_do_call+0x12/0x38 [ 1583.113054] ---[ end trace fc23e059cc4c2ced ]--- Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
2010-03-22 05:28:41 -04:00
meta->flags = 0;
}
EXPORT_SYMBOL(iwl_tx_cmd_complete);