2005-04-16 18:20:36 -04:00
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/hugetlb.h>
|
|
|
|
#include <linux/mount.h>
|
|
|
|
#include <linux/seq_file.h>
|
2005-09-03 18:55:10 -04:00
|
|
|
#include <linux/highmem.h>
|
2007-05-08 03:26:04 -04:00
|
|
|
#include <linux/ptrace.h>
|
2005-09-03 18:54:45 -04:00
|
|
|
#include <linux/pagemap.h>
|
|
|
|
#include <linux/mempolicy.h>
|
2008-02-05 01:29:04 -05:00
|
|
|
#include <linux/swap.h>
|
|
|
|
#include <linux/swapops.h>
|
2005-09-03 18:55:10 -04:00
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
#include <asm/elf.h>
|
|
|
|
#include <asm/uaccess.h>
|
2005-09-03 18:55:10 -04:00
|
|
|
#include <asm/tlbflush.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
#include "internal.h"
|
|
|
|
|
2008-02-08 07:18:33 -05:00
|
|
|
void task_mem(struct seq_file *m, struct mm_struct *mm)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
unsigned long data, text, lib;
|
[PATCH] mm: update_hiwaters just in time
update_mem_hiwater has attracted various criticisms, in particular from those
concerned with mm scalability. Originally it was called whenever rss or
total_vm got raised. Then many of those callsites were replaced by a timer
tick call from account_system_time. Now Frank van Maarseveen reports that to
be found inadequate. How about this? Works for Frank.
Replace update_mem_hiwater, a poor combination of two unrelated ops, by macros
update_hiwater_rss and update_hiwater_vm. Don't attempt to keep
mm->hiwater_rss up to date at timer tick, nor every time we raise rss (usually
by 1): those are hot paths. Do the opposite, update only when about to lower
rss (usually by many), or just before final accounting in do_exit. Handle
mm->hiwater_vm in the same way, though it's much less of an issue. Demand
that whoever collects these hiwater statistics do the work of taking the
maximum with rss or total_vm.
And there has been no collector of these hiwater statistics in the tree. The
new convention needs an example, so match Frank's usage by adding a VmPeak
line above VmSize to /proc/<pid>/status, and also a VmHWM line above VmRSS
(High-Water-Mark or High-Water-Memory).
There was a particular anomaly during mremap move, that hiwater_vm might be
captured too high. A fleeting such anomaly remains, but it's quickly
corrected now, whereas before it would stick.
What locking? None: if the app is racy then these statistics will be racy,
it's not worth any overhead to make them exact. But whenever it suits,
hiwater_vm is updated under exclusive mmap_sem, and hiwater_rss under
page_table_lock (for now) or with preemption disabled (later on): without
going to any trouble, minimize the time between reading current values and
updating, to minimize those occasions when a racing thread bumps a count up
and back down in between.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29 21:16:18 -04:00
|
|
|
unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Note: to minimize their overhead, mm maintains hiwater_vm and
|
|
|
|
* hiwater_rss only when about to *lower* total_vm or rss. Any
|
|
|
|
* collector of these hiwater stats must therefore get total_vm
|
|
|
|
* and rss too, which will usually be the higher. Barriers? not
|
|
|
|
* worth the effort, such snapshots can always be inconsistent.
|
|
|
|
*/
|
|
|
|
hiwater_vm = total_vm = mm->total_vm;
|
|
|
|
if (hiwater_vm < mm->hiwater_vm)
|
|
|
|
hiwater_vm = mm->hiwater_vm;
|
|
|
|
hiwater_rss = total_rss = get_mm_rss(mm);
|
|
|
|
if (hiwater_rss < mm->hiwater_rss)
|
|
|
|
hiwater_rss = mm->hiwater_rss;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
data = mm->total_vm - mm->shared_vm - mm->stack_vm;
|
|
|
|
text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10;
|
|
|
|
lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text;
|
2008-02-08 07:18:33 -05:00
|
|
|
seq_printf(m,
|
[PATCH] mm: update_hiwaters just in time
update_mem_hiwater has attracted various criticisms, in particular from those
concerned with mm scalability. Originally it was called whenever rss or
total_vm got raised. Then many of those callsites were replaced by a timer
tick call from account_system_time. Now Frank van Maarseveen reports that to
be found inadequate. How about this? Works for Frank.
Replace update_mem_hiwater, a poor combination of two unrelated ops, by macros
update_hiwater_rss and update_hiwater_vm. Don't attempt to keep
mm->hiwater_rss up to date at timer tick, nor every time we raise rss (usually
by 1): those are hot paths. Do the opposite, update only when about to lower
rss (usually by many), or just before final accounting in do_exit. Handle
mm->hiwater_vm in the same way, though it's much less of an issue. Demand
that whoever collects these hiwater statistics do the work of taking the
maximum with rss or total_vm.
And there has been no collector of these hiwater statistics in the tree. The
new convention needs an example, so match Frank's usage by adding a VmPeak
line above VmSize to /proc/<pid>/status, and also a VmHWM line above VmRSS
(High-Water-Mark or High-Water-Memory).
There was a particular anomaly during mremap move, that hiwater_vm might be
captured too high. A fleeting such anomaly remains, but it's quickly
corrected now, whereas before it would stick.
What locking? None: if the app is racy then these statistics will be racy,
it's not worth any overhead to make them exact. But whenever it suits,
hiwater_vm is updated under exclusive mmap_sem, and hiwater_rss under
page_table_lock (for now) or with preemption disabled (later on): without
going to any trouble, minimize the time between reading current values and
updating, to minimize those occasions when a racing thread bumps a count up
and back down in between.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29 21:16:18 -04:00
|
|
|
"VmPeak:\t%8lu kB\n"
|
2005-04-16 18:20:36 -04:00
|
|
|
"VmSize:\t%8lu kB\n"
|
|
|
|
"VmLck:\t%8lu kB\n"
|
[PATCH] mm: update_hiwaters just in time
update_mem_hiwater has attracted various criticisms, in particular from those
concerned with mm scalability. Originally it was called whenever rss or
total_vm got raised. Then many of those callsites were replaced by a timer
tick call from account_system_time. Now Frank van Maarseveen reports that to
be found inadequate. How about this? Works for Frank.
Replace update_mem_hiwater, a poor combination of two unrelated ops, by macros
update_hiwater_rss and update_hiwater_vm. Don't attempt to keep
mm->hiwater_rss up to date at timer tick, nor every time we raise rss (usually
by 1): those are hot paths. Do the opposite, update only when about to lower
rss (usually by many), or just before final accounting in do_exit. Handle
mm->hiwater_vm in the same way, though it's much less of an issue. Demand
that whoever collects these hiwater statistics do the work of taking the
maximum with rss or total_vm.
And there has been no collector of these hiwater statistics in the tree. The
new convention needs an example, so match Frank's usage by adding a VmPeak
line above VmSize to /proc/<pid>/status, and also a VmHWM line above VmRSS
(High-Water-Mark or High-Water-Memory).
There was a particular anomaly during mremap move, that hiwater_vm might be
captured too high. A fleeting such anomaly remains, but it's quickly
corrected now, whereas before it would stick.
What locking? None: if the app is racy then these statistics will be racy,
it's not worth any overhead to make them exact. But whenever it suits,
hiwater_vm is updated under exclusive mmap_sem, and hiwater_rss under
page_table_lock (for now) or with preemption disabled (later on): without
going to any trouble, minimize the time between reading current values and
updating, to minimize those occasions when a racing thread bumps a count up
and back down in between.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29 21:16:18 -04:00
|
|
|
"VmHWM:\t%8lu kB\n"
|
2005-04-16 18:20:36 -04:00
|
|
|
"VmRSS:\t%8lu kB\n"
|
|
|
|
"VmData:\t%8lu kB\n"
|
|
|
|
"VmStk:\t%8lu kB\n"
|
|
|
|
"VmExe:\t%8lu kB\n"
|
|
|
|
"VmLib:\t%8lu kB\n"
|
|
|
|
"VmPTE:\t%8lu kB\n",
|
[PATCH] mm: update_hiwaters just in time
update_mem_hiwater has attracted various criticisms, in particular from those
concerned with mm scalability. Originally it was called whenever rss or
total_vm got raised. Then many of those callsites were replaced by a timer
tick call from account_system_time. Now Frank van Maarseveen reports that to
be found inadequate. How about this? Works for Frank.
Replace update_mem_hiwater, a poor combination of two unrelated ops, by macros
update_hiwater_rss and update_hiwater_vm. Don't attempt to keep
mm->hiwater_rss up to date at timer tick, nor every time we raise rss (usually
by 1): those are hot paths. Do the opposite, update only when about to lower
rss (usually by many), or just before final accounting in do_exit. Handle
mm->hiwater_vm in the same way, though it's much less of an issue. Demand
that whoever collects these hiwater statistics do the work of taking the
maximum with rss or total_vm.
And there has been no collector of these hiwater statistics in the tree. The
new convention needs an example, so match Frank's usage by adding a VmPeak
line above VmSize to /proc/<pid>/status, and also a VmHWM line above VmRSS
(High-Water-Mark or High-Water-Memory).
There was a particular anomaly during mremap move, that hiwater_vm might be
captured too high. A fleeting such anomaly remains, but it's quickly
corrected now, whereas before it would stick.
What locking? None: if the app is racy then these statistics will be racy,
it's not worth any overhead to make them exact. But whenever it suits,
hiwater_vm is updated under exclusive mmap_sem, and hiwater_rss under
page_table_lock (for now) or with preemption disabled (later on): without
going to any trouble, minimize the time between reading current values and
updating, to minimize those occasions when a racing thread bumps a count up
and back down in between.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29 21:16:18 -04:00
|
|
|
hiwater_vm << (PAGE_SHIFT-10),
|
|
|
|
(total_vm - mm->reserved_vm) << (PAGE_SHIFT-10),
|
2005-04-16 18:20:36 -04:00
|
|
|
mm->locked_vm << (PAGE_SHIFT-10),
|
[PATCH] mm: update_hiwaters just in time
update_mem_hiwater has attracted various criticisms, in particular from those
concerned with mm scalability. Originally it was called whenever rss or
total_vm got raised. Then many of those callsites were replaced by a timer
tick call from account_system_time. Now Frank van Maarseveen reports that to
be found inadequate. How about this? Works for Frank.
Replace update_mem_hiwater, a poor combination of two unrelated ops, by macros
update_hiwater_rss and update_hiwater_vm. Don't attempt to keep
mm->hiwater_rss up to date at timer tick, nor every time we raise rss (usually
by 1): those are hot paths. Do the opposite, update only when about to lower
rss (usually by many), or just before final accounting in do_exit. Handle
mm->hiwater_vm in the same way, though it's much less of an issue. Demand
that whoever collects these hiwater statistics do the work of taking the
maximum with rss or total_vm.
And there has been no collector of these hiwater statistics in the tree. The
new convention needs an example, so match Frank's usage by adding a VmPeak
line above VmSize to /proc/<pid>/status, and also a VmHWM line above VmRSS
(High-Water-Mark or High-Water-Memory).
There was a particular anomaly during mremap move, that hiwater_vm might be
captured too high. A fleeting such anomaly remains, but it's quickly
corrected now, whereas before it would stick.
What locking? None: if the app is racy then these statistics will be racy,
it's not worth any overhead to make them exact. But whenever it suits,
hiwater_vm is updated under exclusive mmap_sem, and hiwater_rss under
page_table_lock (for now) or with preemption disabled (later on): without
going to any trouble, minimize the time between reading current values and
updating, to minimize those occasions when a racing thread bumps a count up
and back down in between.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29 21:16:18 -04:00
|
|
|
hiwater_rss << (PAGE_SHIFT-10),
|
|
|
|
total_rss << (PAGE_SHIFT-10),
|
2005-04-16 18:20:36 -04:00
|
|
|
data << (PAGE_SHIFT-10),
|
|
|
|
mm->stack_vm << (PAGE_SHIFT-10), text, lib,
|
|
|
|
(PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10);
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned long task_vsize(struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
return PAGE_SIZE * mm->total_vm;
|
|
|
|
}
|
|
|
|
|
|
|
|
int task_statm(struct mm_struct *mm, int *shared, int *text,
|
|
|
|
int *data, int *resident)
|
|
|
|
{
|
2005-10-29 21:16:05 -04:00
|
|
|
*shared = get_mm_counter(mm, file_rss);
|
2005-04-16 18:20:36 -04:00
|
|
|
*text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK))
|
|
|
|
>> PAGE_SHIFT;
|
|
|
|
*data = mm->total_vm - mm->shared_vm;
|
2005-10-29 21:16:05 -04:00
|
|
|
*resident = *shared + get_mm_counter(mm, anon_rss);
|
2005-04-16 18:20:36 -04:00
|
|
|
return mm->total_vm;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void pad_len_spaces(struct seq_file *m, int len)
|
|
|
|
{
|
|
|
|
len = 25 + sizeof(void*) * 6 - len;
|
|
|
|
if (len < 1)
|
|
|
|
len = 1;
|
|
|
|
seq_printf(m, "%*c", len, ' ');
|
|
|
|
}
|
|
|
|
|
2008-02-05 01:29:03 -05:00
|
|
|
static void vma_stop(struct proc_maps_private *priv, struct vm_area_struct *vma)
|
|
|
|
{
|
|
|
|
if (vma && vma != priv->tail_vma) {
|
|
|
|
struct mm_struct *mm = vma->vm_mm;
|
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
mmput(mm);
|
|
|
|
}
|
|
|
|
}
|
2008-02-05 01:28:56 -05:00
|
|
|
|
2008-02-05 01:29:03 -05:00
|
|
|
static void *m_start(struct seq_file *m, loff_t *pos)
|
2005-09-03 18:55:10 -04:00
|
|
|
{
|
2008-02-05 01:29:03 -05:00
|
|
|
struct proc_maps_private *priv = m->private;
|
|
|
|
unsigned long last_addr = m->version;
|
|
|
|
struct mm_struct *mm;
|
|
|
|
struct vm_area_struct *vma, *tail_vma = NULL;
|
|
|
|
loff_t l = *pos;
|
|
|
|
|
|
|
|
/* Clear the per syscall fields in priv */
|
|
|
|
priv->task = NULL;
|
|
|
|
priv->tail_vma = NULL;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We remember last_addr rather than next_addr to hit with
|
|
|
|
* mmap_cache most of the time. We have zero last_addr at
|
|
|
|
* the beginning and also after lseek. We will have -1 last_addr
|
|
|
|
* after the end of the vmas.
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (last_addr == -1UL)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
priv->task = get_pid_task(priv->pid, PIDTYPE_PID);
|
|
|
|
if (!priv->task)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
mm = mm_for_maps(priv->task);
|
|
|
|
if (!mm)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
tail_vma = get_gate_vma(priv->task);
|
|
|
|
priv->tail_vma = tail_vma;
|
|
|
|
|
|
|
|
/* Start with last addr hint */
|
|
|
|
vma = find_vma(mm, last_addr);
|
|
|
|
if (last_addr && vma) {
|
|
|
|
vma = vma->vm_next;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check the vma index is within the range and do
|
|
|
|
* sequential scan until m_index.
|
|
|
|
*/
|
|
|
|
vma = NULL;
|
|
|
|
if ((unsigned long)l < mm->map_count) {
|
|
|
|
vma = mm->mmap;
|
|
|
|
while (l-- && vma)
|
|
|
|
vma = vma->vm_next;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (l != mm->map_count)
|
|
|
|
tail_vma = NULL; /* After gate vma */
|
|
|
|
|
|
|
|
out:
|
|
|
|
if (vma)
|
|
|
|
return vma;
|
|
|
|
|
|
|
|
/* End of vmas has been reached */
|
|
|
|
m->version = (tail_vma != NULL)? 0: -1UL;
|
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
mmput(mm);
|
|
|
|
return tail_vma;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void *m_next(struct seq_file *m, void *v, loff_t *pos)
|
|
|
|
{
|
|
|
|
struct proc_maps_private *priv = m->private;
|
|
|
|
struct vm_area_struct *vma = v;
|
|
|
|
struct vm_area_struct *tail_vma = priv->tail_vma;
|
|
|
|
|
|
|
|
(*pos)++;
|
|
|
|
if (vma && (vma != tail_vma) && vma->vm_next)
|
|
|
|
return vma->vm_next;
|
|
|
|
vma_stop(priv, vma);
|
|
|
|
return (vma != tail_vma)? tail_vma: NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void m_stop(struct seq_file *m, void *v)
|
|
|
|
{
|
|
|
|
struct proc_maps_private *priv = m->private;
|
|
|
|
struct vm_area_struct *vma = v;
|
|
|
|
|
|
|
|
vma_stop(priv, vma);
|
|
|
|
if (priv->task)
|
|
|
|
put_task_struct(priv->task);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int do_maps_open(struct inode *inode, struct file *file,
|
2008-02-08 07:21:19 -05:00
|
|
|
const struct seq_operations *ops)
|
2008-02-05 01:29:03 -05:00
|
|
|
{
|
|
|
|
struct proc_maps_private *priv;
|
|
|
|
int ret = -ENOMEM;
|
|
|
|
priv = kzalloc(sizeof(*priv), GFP_KERNEL);
|
|
|
|
if (priv) {
|
|
|
|
priv->pid = proc_pid(inode);
|
|
|
|
ret = seq_open(file, ops);
|
|
|
|
if (!ret) {
|
|
|
|
struct seq_file *m = file->private_data;
|
|
|
|
m->private = priv;
|
|
|
|
} else {
|
|
|
|
kfree(priv);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
2005-09-03 18:55:10 -04:00
|
|
|
|
2008-02-05 01:29:02 -05:00
|
|
|
static int show_map(struct seq_file *m, void *v)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2006-06-26 03:25:55 -04:00
|
|
|
struct proc_maps_private *priv = m->private;
|
|
|
|
struct task_struct *task = priv->task;
|
2005-09-03 18:55:10 -04:00
|
|
|
struct vm_area_struct *vma = v;
|
|
|
|
struct mm_struct *mm = vma->vm_mm;
|
|
|
|
struct file *file = vma->vm_file;
|
|
|
|
int flags = vma->vm_flags;
|
2005-04-16 18:20:36 -04:00
|
|
|
unsigned long ino = 0;
|
|
|
|
dev_t dev = 0;
|
|
|
|
int len;
|
|
|
|
|
2007-05-08 03:26:04 -04:00
|
|
|
if (maps_protect && !ptrace_may_attach(task))
|
|
|
|
return -EACCES;
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
if (file) {
|
2006-12-08 05:36:36 -05:00
|
|
|
struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
|
2005-04-16 18:20:36 -04:00
|
|
|
dev = inode->i_sb->s_dev;
|
|
|
|
ino = inode->i_ino;
|
|
|
|
}
|
|
|
|
|
|
|
|
seq_printf(m, "%08lx-%08lx %c%c%c%c %08lx %02x:%02x %lu %n",
|
2005-09-03 18:55:10 -04:00
|
|
|
vma->vm_start,
|
|
|
|
vma->vm_end,
|
2005-04-16 18:20:36 -04:00
|
|
|
flags & VM_READ ? 'r' : '-',
|
|
|
|
flags & VM_WRITE ? 'w' : '-',
|
|
|
|
flags & VM_EXEC ? 'x' : '-',
|
|
|
|
flags & VM_MAYSHARE ? 's' : 'p',
|
2005-09-03 18:55:10 -04:00
|
|
|
vma->vm_pgoff << PAGE_SHIFT,
|
2005-04-16 18:20:36 -04:00
|
|
|
MAJOR(dev), MINOR(dev), ino, &len);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Print the dentry name for named mappings, and a
|
|
|
|
* special [heap] marker for the heap:
|
|
|
|
*/
|
2005-09-03 18:55:10 -04:00
|
|
|
if (file) {
|
2005-04-16 18:20:36 -04:00
|
|
|
pad_len_spaces(m, len);
|
2008-02-14 22:38:43 -05:00
|
|
|
seq_path(m, &file->f_path, "\n");
|
2005-04-16 18:20:36 -04:00
|
|
|
} else {
|
[PATCH] vdso: randomize the i386 vDSO by moving it into a vma
Move the i386 VDSO down into a vma and thus randomize it.
Besides the security implications, this feature also helps debuggers, which
can COW a vma-backed VDSO just like a normal DSO and can thus do
single-stepping and other debugging features.
It's good for hypervisors (Xen, VMWare) too, which typically live in the same
high-mapped address space as the VDSO, hence whenever the VDSO is used, they
get lots of guest pagefaults and have to fix such guest accesses up - which
slows things down instead of speeding things up (the primary purpose of the
VDSO).
There's a new CONFIG_COMPAT_VDSO (default=y) option, which provides support
for older glibcs that still rely on a prelinked high-mapped VDSO. Newer
distributions (using glibc 2.3.3 or later) can turn this option off. Turning
it off is also recommended for security reasons: attackers cannot use the
predictable high-mapped VDSO page as syscall trampoline anymore.
There is a new vdso=[0|1] boot option as well, and a runtime
/proc/sys/vm/vdso_enabled sysctl switch, that allows the VDSO to be turned
on/off.
(This version of the VDSO-randomization patch also has working ELF
coredumping, the previous patch crashed in the coredumping code.)
This code is a combined work of the exec-shield VDSO randomization
code and Gerd Hoffmann's hypervisor-centric VDSO patch. Rusty Russell
started this patch and i completed it.
[akpm@osdl.org: cleanups]
[akpm@osdl.org: compile fix]
[akpm@osdl.org: compile fix 2]
[akpm@osdl.org: compile fix 3]
[akpm@osdl.org: revernt MAXMEM change]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 05:53:50 -04:00
|
|
|
const char *name = arch_vma_name(vma);
|
|
|
|
if (!name) {
|
|
|
|
if (mm) {
|
|
|
|
if (vma->vm_start <= mm->start_brk &&
|
2005-09-03 18:55:10 -04:00
|
|
|
vma->vm_end >= mm->brk) {
|
[PATCH] vdso: randomize the i386 vDSO by moving it into a vma
Move the i386 VDSO down into a vma and thus randomize it.
Besides the security implications, this feature also helps debuggers, which
can COW a vma-backed VDSO just like a normal DSO and can thus do
single-stepping and other debugging features.
It's good for hypervisors (Xen, VMWare) too, which typically live in the same
high-mapped address space as the VDSO, hence whenever the VDSO is used, they
get lots of guest pagefaults and have to fix such guest accesses up - which
slows things down instead of speeding things up (the primary purpose of the
VDSO).
There's a new CONFIG_COMPAT_VDSO (default=y) option, which provides support
for older glibcs that still rely on a prelinked high-mapped VDSO. Newer
distributions (using glibc 2.3.3 or later) can turn this option off. Turning
it off is also recommended for security reasons: attackers cannot use the
predictable high-mapped VDSO page as syscall trampoline anymore.
There is a new vdso=[0|1] boot option as well, and a runtime
/proc/sys/vm/vdso_enabled sysctl switch, that allows the VDSO to be turned
on/off.
(This version of the VDSO-randomization patch also has working ELF
coredumping, the previous patch crashed in the coredumping code.)
This code is a combined work of the exec-shield VDSO randomization
code and Gerd Hoffmann's hypervisor-centric VDSO patch. Rusty Russell
started this patch and i completed it.
[akpm@osdl.org: cleanups]
[akpm@osdl.org: compile fix]
[akpm@osdl.org: compile fix 2]
[akpm@osdl.org: compile fix 3]
[akpm@osdl.org: revernt MAXMEM change]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 05:53:50 -04:00
|
|
|
name = "[heap]";
|
|
|
|
} else if (vma->vm_start <= mm->start_stack &&
|
|
|
|
vma->vm_end >= mm->start_stack) {
|
|
|
|
name = "[stack]";
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
[PATCH] vdso: randomize the i386 vDSO by moving it into a vma
Move the i386 VDSO down into a vma and thus randomize it.
Besides the security implications, this feature also helps debuggers, which
can COW a vma-backed VDSO just like a normal DSO and can thus do
single-stepping and other debugging features.
It's good for hypervisors (Xen, VMWare) too, which typically live in the same
high-mapped address space as the VDSO, hence whenever the VDSO is used, they
get lots of guest pagefaults and have to fix such guest accesses up - which
slows things down instead of speeding things up (the primary purpose of the
VDSO).
There's a new CONFIG_COMPAT_VDSO (default=y) option, which provides support
for older glibcs that still rely on a prelinked high-mapped VDSO. Newer
distributions (using glibc 2.3.3 or later) can turn this option off. Turning
it off is also recommended for security reasons: attackers cannot use the
predictable high-mapped VDSO page as syscall trampoline anymore.
There is a new vdso=[0|1] boot option as well, and a runtime
/proc/sys/vm/vdso_enabled sysctl switch, that allows the VDSO to be turned
on/off.
(This version of the VDSO-randomization patch also has working ELF
coredumping, the previous patch crashed in the coredumping code.)
This code is a combined work of the exec-shield VDSO randomization
code and Gerd Hoffmann's hypervisor-centric VDSO patch. Rusty Russell
started this patch and i completed it.
[akpm@osdl.org: cleanups]
[akpm@osdl.org: compile fix]
[akpm@osdl.org: compile fix 2]
[akpm@osdl.org: compile fix 3]
[akpm@osdl.org: revernt MAXMEM change]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 05:53:50 -04:00
|
|
|
} else {
|
|
|
|
name = "[vdso]";
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
[PATCH] vdso: randomize the i386 vDSO by moving it into a vma
Move the i386 VDSO down into a vma and thus randomize it.
Besides the security implications, this feature also helps debuggers, which
can COW a vma-backed VDSO just like a normal DSO and can thus do
single-stepping and other debugging features.
It's good for hypervisors (Xen, VMWare) too, which typically live in the same
high-mapped address space as the VDSO, hence whenever the VDSO is used, they
get lots of guest pagefaults and have to fix such guest accesses up - which
slows things down instead of speeding things up (the primary purpose of the
VDSO).
There's a new CONFIG_COMPAT_VDSO (default=y) option, which provides support
for older glibcs that still rely on a prelinked high-mapped VDSO. Newer
distributions (using glibc 2.3.3 or later) can turn this option off. Turning
it off is also recommended for security reasons: attackers cannot use the
predictable high-mapped VDSO page as syscall trampoline anymore.
There is a new vdso=[0|1] boot option as well, and a runtime
/proc/sys/vm/vdso_enabled sysctl switch, that allows the VDSO to be turned
on/off.
(This version of the VDSO-randomization patch also has working ELF
coredumping, the previous patch crashed in the coredumping code.)
This code is a combined work of the exec-shield VDSO randomization
code and Gerd Hoffmann's hypervisor-centric VDSO patch. Rusty Russell
started this patch and i completed it.
[akpm@osdl.org: cleanups]
[akpm@osdl.org: compile fix]
[akpm@osdl.org: compile fix 2]
[akpm@osdl.org: compile fix 3]
[akpm@osdl.org: revernt MAXMEM change]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 05:53:50 -04:00
|
|
|
}
|
|
|
|
if (name) {
|
2005-04-16 18:20:36 -04:00
|
|
|
pad_len_spaces(m, len);
|
[PATCH] vdso: randomize the i386 vDSO by moving it into a vma
Move the i386 VDSO down into a vma and thus randomize it.
Besides the security implications, this feature also helps debuggers, which
can COW a vma-backed VDSO just like a normal DSO and can thus do
single-stepping and other debugging features.
It's good for hypervisors (Xen, VMWare) too, which typically live in the same
high-mapped address space as the VDSO, hence whenever the VDSO is used, they
get lots of guest pagefaults and have to fix such guest accesses up - which
slows things down instead of speeding things up (the primary purpose of the
VDSO).
There's a new CONFIG_COMPAT_VDSO (default=y) option, which provides support
for older glibcs that still rely on a prelinked high-mapped VDSO. Newer
distributions (using glibc 2.3.3 or later) can turn this option off. Turning
it off is also recommended for security reasons: attackers cannot use the
predictable high-mapped VDSO page as syscall trampoline anymore.
There is a new vdso=[0|1] boot option as well, and a runtime
/proc/sys/vm/vdso_enabled sysctl switch, that allows the VDSO to be turned
on/off.
(This version of the VDSO-randomization patch also has working ELF
coredumping, the previous patch crashed in the coredumping code.)
This code is a combined work of the exec-shield VDSO randomization
code and Gerd Hoffmann's hypervisor-centric VDSO patch. Rusty Russell
started this patch and i completed it.
[akpm@osdl.org: cleanups]
[akpm@osdl.org: compile fix]
[akpm@osdl.org: compile fix 2]
[akpm@osdl.org: compile fix 3]
[akpm@osdl.org: revernt MAXMEM change]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 05:53:50 -04:00
|
|
|
seq_puts(m, name);
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
seq_putc(m, '\n');
|
2005-09-03 18:55:10 -04:00
|
|
|
|
|
|
|
if (m->count < m->size) /* vma is copied successfully */
|
|
|
|
m->version = (vma != get_gate_vma(task))? vma->vm_start: 0;
|
2005-04-16 18:20:36 -04:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-02-08 07:21:19 -05:00
|
|
|
static const struct seq_operations proc_pid_maps_op = {
|
2008-02-05 01:29:03 -05:00
|
|
|
.start = m_start,
|
|
|
|
.next = m_next,
|
|
|
|
.stop = m_stop,
|
|
|
|
.show = show_map
|
|
|
|
};
|
|
|
|
|
|
|
|
static int maps_open(struct inode *inode, struct file *file)
|
|
|
|
{
|
|
|
|
return do_maps_open(inode, file, &proc_pid_maps_op);
|
|
|
|
}
|
|
|
|
|
|
|
|
const struct file_operations proc_maps_operations = {
|
|
|
|
.open = maps_open,
|
|
|
|
.read = seq_read,
|
|
|
|
.llseek = seq_lseek,
|
|
|
|
.release = seq_release_private,
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Proportional Set Size(PSS): my share of RSS.
|
|
|
|
*
|
|
|
|
* PSS of a process is the count of pages it has in memory, where each
|
|
|
|
* page is divided by the number of processes sharing it. So if a
|
|
|
|
* process has 1000 pages all to itself, and 1000 shared with one other
|
|
|
|
* process, its PSS will be 1500.
|
|
|
|
*
|
|
|
|
* To keep (accumulated) division errors low, we adopt a 64bit
|
|
|
|
* fixed-point pss counter to minimize division errors. So (pss >>
|
|
|
|
* PSS_SHIFT) would be the real byte count.
|
|
|
|
*
|
|
|
|
* A shift of 12 before division means (assuming 4K page size):
|
|
|
|
* - 1M 3-user-pages add up to 8KB errors;
|
|
|
|
* - supports mapcount up to 2^24, or 16M;
|
|
|
|
* - supports PSS up to 2^52 bytes, or 4PB.
|
|
|
|
*/
|
|
|
|
#define PSS_SHIFT 12
|
|
|
|
|
2008-02-05 01:29:07 -05:00
|
|
|
#ifdef CONFIG_PROC_PAGE_MONITOR
|
2008-04-28 05:12:55 -04:00
|
|
|
struct mem_size_stats {
|
2008-02-05 01:29:03 -05:00
|
|
|
struct vm_area_struct *vma;
|
|
|
|
unsigned long resident;
|
|
|
|
unsigned long shared_clean;
|
|
|
|
unsigned long shared_dirty;
|
|
|
|
unsigned long private_clean;
|
|
|
|
unsigned long private_dirty;
|
|
|
|
unsigned long referenced;
|
2008-04-28 05:12:55 -04:00
|
|
|
unsigned long swap;
|
2008-02-05 01:29:03 -05:00
|
|
|
u64 pss;
|
|
|
|
};
|
|
|
|
|
2008-02-05 01:29:01 -05:00
|
|
|
static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
|
2008-06-12 18:21:47 -04:00
|
|
|
struct mm_walk *walk)
|
2005-09-03 18:55:10 -04:00
|
|
|
{
|
2008-06-12 18:21:47 -04:00
|
|
|
struct mem_size_stats *mss = walk->private;
|
2008-02-05 01:29:01 -05:00
|
|
|
struct vm_area_struct *vma = mss->vma;
|
2005-09-03 18:55:10 -04:00
|
|
|
pte_t *pte, ptent;
|
2005-10-29 21:16:27 -04:00
|
|
|
spinlock_t *ptl;
|
2005-09-03 18:55:10 -04:00
|
|
|
struct page *page;
|
2008-02-05 01:28:56 -05:00
|
|
|
int mapcount;
|
2005-09-03 18:55:10 -04:00
|
|
|
|
2005-10-29 21:16:27 -04:00
|
|
|
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
|
smaps: extract pmd walker from smaps code
Extracts the pmd walker from smaps-specific code in fs/proc/task_mmu.c.
The new struct pmd_walker includes the struct vm_area_struct of the memory to
walk over. Iteration begins at the vma->vm_start and completes at
vma->vm_end. A pointer to another data structure may be stored in the private
field such as struct mem_size_stats, which acts as the smaps accumulator. For
each pmd in the VMA, the action function is called with a pointer to its
struct vm_area_struct, a pointer to the pmd_t, its start and end addresses,
and the private field.
The interface for walking pmd's in a VMA for fs/proc/task_mmu.c is now:
void for_each_pmd(struct vm_area_struct *vma,
void (*action)(struct vm_area_struct *vma,
pmd_t *pmd, unsigned long addr,
unsigned long end,
void *private),
void *private);
Since the pmd walker is now extracted from the smaps code, smaps_one_pmd() is
invoked for each pmd in the VMA. Its behavior and efficiency is identical to
the existing implementation.
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-06 17:49:21 -04:00
|
|
|
for (; addr != end; pte++, addr += PAGE_SIZE) {
|
2005-09-03 18:55:10 -04:00
|
|
|
ptent = *pte;
|
2008-04-28 05:12:55 -04:00
|
|
|
|
|
|
|
if (is_swap_pte(ptent)) {
|
|
|
|
mss->swap += PAGE_SIZE;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2005-10-29 21:16:27 -04:00
|
|
|
if (!pte_present(ptent))
|
2005-09-03 18:55:10 -04:00
|
|
|
continue;
|
|
|
|
|
|
|
|
mss->resident += PAGE_SIZE;
|
2006-03-06 18:42:58 -05:00
|
|
|
|
|
|
|
page = vm_normal_page(vma, addr, ptent);
|
|
|
|
if (!page)
|
2005-09-03 18:55:10 -04:00
|
|
|
continue;
|
|
|
|
|
2007-05-06 17:49:22 -04:00
|
|
|
/* Accumulate the size in pages that have been accessed. */
|
|
|
|
if (pte_young(ptent) || PageReferenced(page))
|
|
|
|
mss->referenced += PAGE_SIZE;
|
2008-02-05 01:28:56 -05:00
|
|
|
mapcount = page_mapcount(page);
|
|
|
|
if (mapcount >= 2) {
|
2005-09-03 18:55:10 -04:00
|
|
|
if (pte_dirty(ptent))
|
|
|
|
mss->shared_dirty += PAGE_SIZE;
|
|
|
|
else
|
|
|
|
mss->shared_clean += PAGE_SIZE;
|
2008-02-05 01:28:56 -05:00
|
|
|
mss->pss += (PAGE_SIZE << PSS_SHIFT) / mapcount;
|
2005-09-03 18:55:10 -04:00
|
|
|
} else {
|
|
|
|
if (pte_dirty(ptent))
|
|
|
|
mss->private_dirty += PAGE_SIZE;
|
|
|
|
else
|
|
|
|
mss->private_clean += PAGE_SIZE;
|
2008-02-05 01:28:56 -05:00
|
|
|
mss->pss += (PAGE_SIZE << PSS_SHIFT);
|
2005-09-03 18:55:10 -04:00
|
|
|
}
|
smaps: extract pmd walker from smaps code
Extracts the pmd walker from smaps-specific code in fs/proc/task_mmu.c.
The new struct pmd_walker includes the struct vm_area_struct of the memory to
walk over. Iteration begins at the vma->vm_start and completes at
vma->vm_end. A pointer to another data structure may be stored in the private
field such as struct mem_size_stats, which acts as the smaps accumulator. For
each pmd in the VMA, the action function is called with a pointer to its
struct vm_area_struct, a pointer to the pmd_t, its start and end addresses,
and the private field.
The interface for walking pmd's in a VMA for fs/proc/task_mmu.c is now:
void for_each_pmd(struct vm_area_struct *vma,
void (*action)(struct vm_area_struct *vma,
pmd_t *pmd, unsigned long addr,
unsigned long end,
void *private),
void *private);
Since the pmd walker is now extracted from the smaps code, smaps_one_pmd() is
invoked for each pmd in the VMA. Its behavior and efficiency is identical to
the existing implementation.
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-06 17:49:21 -04:00
|
|
|
}
|
2005-10-29 21:16:27 -04:00
|
|
|
pte_unmap_unlock(pte - 1, ptl);
|
|
|
|
cond_resched();
|
2008-02-05 01:29:01 -05:00
|
|
|
return 0;
|
2005-09-03 18:55:10 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
static int show_smap(struct seq_file *m, void *v)
|
|
|
|
{
|
|
|
|
struct vm_area_struct *vma = v;
|
|
|
|
struct mem_size_stats mss;
|
2008-02-05 01:29:02 -05:00
|
|
|
int ret;
|
2008-06-12 18:21:47 -04:00
|
|
|
struct mm_walk smaps_walk = {
|
|
|
|
.pmd_entry = smaps_pte_range,
|
|
|
|
.mm = vma->vm_mm,
|
|
|
|
.private = &mss,
|
|
|
|
};
|
2005-09-03 18:55:10 -04:00
|
|
|
|
|
|
|
memset(&mss, 0, sizeof mss);
|
2008-02-05 01:29:01 -05:00
|
|
|
mss.vma = vma;
|
2006-03-06 18:42:57 -05:00
|
|
|
if (vma->vm_mm && !is_vm_hugetlb_page(vma))
|
2008-06-12 18:21:47 -04:00
|
|
|
walk_page_range(vma->vm_start, vma->vm_end, &smaps_walk);
|
2008-02-05 01:29:02 -05:00
|
|
|
|
|
|
|
ret = show_map(m, v);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
seq_printf(m,
|
|
|
|
"Size: %8lu kB\n"
|
|
|
|
"Rss: %8lu kB\n"
|
|
|
|
"Pss: %8lu kB\n"
|
|
|
|
"Shared_Clean: %8lu kB\n"
|
|
|
|
"Shared_Dirty: %8lu kB\n"
|
|
|
|
"Private_Clean: %8lu kB\n"
|
|
|
|
"Private_Dirty: %8lu kB\n"
|
2008-04-28 05:12:55 -04:00
|
|
|
"Referenced: %8lu kB\n"
|
|
|
|
"Swap: %8lu kB\n",
|
2008-02-05 01:29:02 -05:00
|
|
|
(vma->vm_end - vma->vm_start) >> 10,
|
|
|
|
mss.resident >> 10,
|
|
|
|
(unsigned long)(mss.pss >> (10 + PSS_SHIFT)),
|
|
|
|
mss.shared_clean >> 10,
|
|
|
|
mss.shared_dirty >> 10,
|
|
|
|
mss.private_clean >> 10,
|
|
|
|
mss.private_dirty >> 10,
|
2008-04-28 05:12:55 -04:00
|
|
|
mss.referenced >> 10,
|
|
|
|
mss.swap >> 10);
|
2008-02-05 01:29:02 -05:00
|
|
|
|
|
|
|
return ret;
|
2005-09-03 18:55:10 -04:00
|
|
|
}
|
|
|
|
|
2008-02-08 07:21:19 -05:00
|
|
|
static const struct seq_operations proc_pid_smaps_op = {
|
2008-02-05 01:29:03 -05:00
|
|
|
.start = m_start,
|
|
|
|
.next = m_next,
|
|
|
|
.stop = m_stop,
|
|
|
|
.show = show_smap
|
|
|
|
};
|
|
|
|
|
|
|
|
static int smaps_open(struct inode *inode, struct file *file)
|
|
|
|
{
|
|
|
|
return do_maps_open(inode, file, &proc_pid_smaps_op);
|
|
|
|
}
|
|
|
|
|
|
|
|
const struct file_operations proc_smaps_operations = {
|
|
|
|
.open = smaps_open,
|
|
|
|
.read = seq_read,
|
|
|
|
.llseek = seq_lseek,
|
|
|
|
.release = seq_release_private,
|
|
|
|
};
|
|
|
|
|
|
|
|
static int clear_refs_pte_range(pmd_t *pmd, unsigned long addr,
|
2008-06-12 18:21:47 -04:00
|
|
|
unsigned long end, struct mm_walk *walk)
|
2008-02-05 01:29:03 -05:00
|
|
|
{
|
2008-06-12 18:21:47 -04:00
|
|
|
struct vm_area_struct *vma = walk->private;
|
2008-02-05 01:29:03 -05:00
|
|
|
pte_t *pte, ptent;
|
|
|
|
spinlock_t *ptl;
|
|
|
|
struct page *page;
|
|
|
|
|
|
|
|
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
|
|
|
|
for (; addr != end; pte++, addr += PAGE_SIZE) {
|
|
|
|
ptent = *pte;
|
|
|
|
if (!pte_present(ptent))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
page = vm_normal_page(vma, addr, ptent);
|
|
|
|
if (!page)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
/* Clear accessed and referenced bits. */
|
|
|
|
ptep_test_and_clear_young(vma, addr, pte);
|
|
|
|
ClearPageReferenced(page);
|
|
|
|
}
|
|
|
|
pte_unmap_unlock(pte - 1, ptl);
|
|
|
|
cond_resched();
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-02-05 01:29:03 -05:00
|
|
|
static ssize_t clear_refs_write(struct file *file, const char __user *buf,
|
|
|
|
size_t count, loff_t *ppos)
|
2007-05-06 17:49:24 -04:00
|
|
|
{
|
2008-02-05 01:29:03 -05:00
|
|
|
struct task_struct *task;
|
|
|
|
char buffer[PROC_NUMBUF], *end;
|
|
|
|
struct mm_struct *mm;
|
2007-05-06 17:49:24 -04:00
|
|
|
struct vm_area_struct *vma;
|
|
|
|
|
2008-02-05 01:29:03 -05:00
|
|
|
memset(buffer, 0, sizeof(buffer));
|
|
|
|
if (count > sizeof(buffer) - 1)
|
|
|
|
count = sizeof(buffer) - 1;
|
|
|
|
if (copy_from_user(buffer, buf, count))
|
|
|
|
return -EFAULT;
|
|
|
|
if (!simple_strtol(buffer, &end, 0))
|
|
|
|
return -EINVAL;
|
|
|
|
if (*end == '\n')
|
|
|
|
end++;
|
|
|
|
task = get_proc_task(file->f_path.dentry->d_inode);
|
|
|
|
if (!task)
|
|
|
|
return -ESRCH;
|
|
|
|
mm = get_task_mm(task);
|
|
|
|
if (mm) {
|
2008-06-12 18:21:47 -04:00
|
|
|
static struct mm_walk clear_refs_walk;
|
|
|
|
memset(&clear_refs_walk, 0, sizeof(clear_refs_walk));
|
|
|
|
clear_refs_walk.pmd_entry = clear_refs_pte_range;
|
|
|
|
clear_refs_walk.mm = mm;
|
2008-02-05 01:29:03 -05:00
|
|
|
down_read(&mm->mmap_sem);
|
2008-06-12 18:21:47 -04:00
|
|
|
for (vma = mm->mmap; vma; vma = vma->vm_next) {
|
|
|
|
clear_refs_walk.private = vma;
|
2008-02-05 01:29:03 -05:00
|
|
|
if (!is_vm_hugetlb_page(vma))
|
2008-06-12 18:21:47 -04:00
|
|
|
walk_page_range(vma->vm_start, vma->vm_end,
|
|
|
|
&clear_refs_walk);
|
|
|
|
}
|
2008-02-05 01:29:03 -05:00
|
|
|
flush_tlb_mm(mm);
|
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
mmput(mm);
|
|
|
|
}
|
|
|
|
put_task_struct(task);
|
|
|
|
if (end - buffer == 0)
|
|
|
|
return -EIO;
|
|
|
|
return end - buffer;
|
2007-05-06 17:49:24 -04:00
|
|
|
}
|
|
|
|
|
2008-02-05 01:29:03 -05:00
|
|
|
const struct file_operations proc_clear_refs_operations = {
|
|
|
|
.write = clear_refs_write,
|
|
|
|
};
|
|
|
|
|
2008-02-05 01:29:04 -05:00
|
|
|
struct pagemapread {
|
2008-06-06 01:46:31 -04:00
|
|
|
u64 __user *out, *end;
|
2008-02-05 01:29:04 -05:00
|
|
|
};
|
|
|
|
|
2008-03-21 19:46:59 -04:00
|
|
|
#define PM_ENTRY_BYTES sizeof(u64)
|
|
|
|
#define PM_STATUS_BITS 3
|
|
|
|
#define PM_STATUS_OFFSET (64 - PM_STATUS_BITS)
|
|
|
|
#define PM_STATUS_MASK (((1LL << PM_STATUS_BITS) - 1) << PM_STATUS_OFFSET)
|
|
|
|
#define PM_STATUS(nr) (((nr) << PM_STATUS_OFFSET) & PM_STATUS_MASK)
|
|
|
|
#define PM_PSHIFT_BITS 6
|
|
|
|
#define PM_PSHIFT_OFFSET (PM_STATUS_OFFSET - PM_PSHIFT_BITS)
|
|
|
|
#define PM_PSHIFT_MASK (((1LL << PM_PSHIFT_BITS) - 1) << PM_PSHIFT_OFFSET)
|
|
|
|
#define PM_PSHIFT(x) (((u64) (x) << PM_PSHIFT_OFFSET) & PM_PSHIFT_MASK)
|
|
|
|
#define PM_PFRAME_MASK ((1LL << PM_PSHIFT_OFFSET) - 1)
|
|
|
|
#define PM_PFRAME(x) ((x) & PM_PFRAME_MASK)
|
|
|
|
|
|
|
|
#define PM_PRESENT PM_STATUS(4LL)
|
|
|
|
#define PM_SWAP PM_STATUS(2LL)
|
|
|
|
#define PM_NOT_PRESENT PM_PSHIFT(PAGE_SHIFT)
|
2008-02-05 01:29:04 -05:00
|
|
|
#define PM_END_OF_BUFFER 1
|
|
|
|
|
|
|
|
static int add_to_pagemap(unsigned long addr, u64 pfn,
|
|
|
|
struct pagemapread *pm)
|
|
|
|
{
|
|
|
|
if (put_user(pfn, pm->out))
|
|
|
|
return -EFAULT;
|
2008-06-06 01:46:31 -04:00
|
|
|
pm->out++;
|
|
|
|
if (pm->out >= pm->end)
|
|
|
|
return PM_END_OF_BUFFER;
|
2008-02-05 01:29:04 -05:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int pagemap_pte_hole(unsigned long start, unsigned long end,
|
2008-06-12 18:21:47 -04:00
|
|
|
struct mm_walk *walk)
|
2008-02-05 01:29:04 -05:00
|
|
|
{
|
2008-06-12 18:21:47 -04:00
|
|
|
struct pagemapread *pm = walk->private;
|
2008-02-05 01:29:04 -05:00
|
|
|
unsigned long addr;
|
|
|
|
int err = 0;
|
|
|
|
for (addr = start; addr < end; addr += PAGE_SIZE) {
|
|
|
|
err = add_to_pagemap(addr, PM_NOT_PRESENT, pm);
|
|
|
|
if (err)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2008-04-28 05:12:11 -04:00
|
|
|
static u64 swap_pte_to_pagemap_entry(pte_t pte)
|
2008-02-05 01:29:04 -05:00
|
|
|
{
|
|
|
|
swp_entry_t e = pte_to_swp_entry(pte);
|
2008-03-21 19:46:59 -04:00
|
|
|
return swp_type(e) | (swp_offset(e) << MAX_SWAPFILES_SHIFT);
|
2008-02-05 01:29:04 -05:00
|
|
|
}
|
|
|
|
|
2008-06-12 18:21:48 -04:00
|
|
|
static unsigned long pte_to_pagemap_entry(pte_t pte)
|
|
|
|
{
|
|
|
|
unsigned long pme = 0;
|
|
|
|
if (is_swap_pte(pte))
|
|
|
|
pme = PM_PFRAME(swap_pte_to_pagemap_entry(pte))
|
|
|
|
| PM_PSHIFT(PAGE_SHIFT) | PM_SWAP;
|
|
|
|
else if (pte_present(pte))
|
|
|
|
pme = PM_PFRAME(pte_pfn(pte))
|
|
|
|
| PM_PSHIFT(PAGE_SHIFT) | PM_PRESENT;
|
|
|
|
return pme;
|
|
|
|
}
|
|
|
|
|
2008-02-05 01:29:04 -05:00
|
|
|
static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
|
2008-06-12 18:21:47 -04:00
|
|
|
struct mm_walk *walk)
|
2008-02-05 01:29:04 -05:00
|
|
|
{
|
2008-06-12 18:21:48 -04:00
|
|
|
struct vm_area_struct *vma;
|
2008-06-12 18:21:47 -04:00
|
|
|
struct pagemapread *pm = walk->private;
|
2008-02-05 01:29:04 -05:00
|
|
|
pte_t *pte;
|
|
|
|
int err = 0;
|
|
|
|
|
2008-06-12 18:21:48 -04:00
|
|
|
/* find the first VMA at or above 'addr' */
|
|
|
|
vma = find_vma(walk->mm, addr);
|
2008-02-05 01:29:04 -05:00
|
|
|
for (; addr != end; addr += PAGE_SIZE) {
|
|
|
|
u64 pfn = PM_NOT_PRESENT;
|
2008-06-12 18:21:48 -04:00
|
|
|
|
|
|
|
/* check to see if we've left 'vma' behind
|
|
|
|
* and need a new, higher one */
|
|
|
|
if (vma && (addr >= vma->vm_end))
|
|
|
|
vma = find_vma(walk->mm, addr);
|
|
|
|
|
|
|
|
/* check that 'vma' actually covers this address,
|
|
|
|
* and that it isn't a huge page vma */
|
|
|
|
if (vma && (vma->vm_start <= addr) &&
|
|
|
|
!is_vm_hugetlb_page(vma)) {
|
|
|
|
pte = pte_offset_map(pmd, addr);
|
|
|
|
pfn = pte_to_pagemap_entry(*pte);
|
|
|
|
/* unmap before userspace copy */
|
|
|
|
pte_unmap(pte);
|
|
|
|
}
|
2008-02-05 01:29:04 -05:00
|
|
|
err = add_to_pagemap(addr, pfn, pm);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
cond_resched();
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct mm_walk pagemap_walk = {
|
|
|
|
.pmd_entry = pagemap_pte_range,
|
|
|
|
.pte_hole = pagemap_pte_hole
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* /proc/pid/pagemap - an array mapping virtual pages to pfns
|
|
|
|
*
|
2008-03-21 19:46:59 -04:00
|
|
|
* For each page in the address space, this file contains one 64-bit entry
|
|
|
|
* consisting of the following:
|
|
|
|
*
|
|
|
|
* Bits 0-55 page frame number (PFN) if present
|
|
|
|
* Bits 0-4 swap type if swapped
|
|
|
|
* Bits 5-55 swap offset if swapped
|
|
|
|
* Bits 55-60 page shift (page size = 1<<page shift)
|
|
|
|
* Bit 61 reserved for future use
|
|
|
|
* Bit 62 page swapped
|
|
|
|
* Bit 63 page present
|
|
|
|
*
|
|
|
|
* If the page is not present but in swap, then the PFN contains an
|
|
|
|
* encoding of the swap file number and the page's offset into the
|
|
|
|
* swap. Unmapped pages return a null PFN. This allows determining
|
2008-02-05 01:29:04 -05:00
|
|
|
* precisely which pages are mapped (or in swap) and comparing mapped
|
|
|
|
* pages between processes.
|
|
|
|
*
|
|
|
|
* Efficient users of this interface will use /proc/pid/maps to
|
|
|
|
* determine which areas of memory are actually mapped and llseek to
|
|
|
|
* skip over unmapped regions.
|
|
|
|
*/
|
|
|
|
static ssize_t pagemap_read(struct file *file, char __user *buf,
|
|
|
|
size_t count, loff_t *ppos)
|
|
|
|
{
|
|
|
|
struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
|
|
|
|
struct page **pages, *page;
|
|
|
|
unsigned long uaddr, uend;
|
|
|
|
struct mm_struct *mm;
|
|
|
|
struct pagemapread pm;
|
|
|
|
int pagecount;
|
|
|
|
int ret = -ESRCH;
|
|
|
|
|
|
|
|
if (!task)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
ret = -EACCES;
|
|
|
|
if (!ptrace_may_attach(task))
|
2008-03-13 15:32:35 -04:00
|
|
|
goto out_task;
|
2008-02-05 01:29:04 -05:00
|
|
|
|
|
|
|
ret = -EINVAL;
|
|
|
|
/* file position must be aligned */
|
2008-06-06 01:46:31 -04:00
|
|
|
if ((*ppos % PM_ENTRY_BYTES) || (count % PM_ENTRY_BYTES))
|
2008-03-13 15:32:35 -04:00
|
|
|
goto out_task;
|
2008-02-05 01:29:04 -05:00
|
|
|
|
|
|
|
ret = 0;
|
|
|
|
mm = get_task_mm(task);
|
|
|
|
if (!mm)
|
2008-03-13 15:32:35 -04:00
|
|
|
goto out_task;
|
2008-02-05 01:29:04 -05:00
|
|
|
|
|
|
|
ret = -ENOMEM;
|
|
|
|
uaddr = (unsigned long)buf & PAGE_MASK;
|
|
|
|
uend = (unsigned long)(buf + count);
|
|
|
|
pagecount = (PAGE_ALIGN(uend) - uaddr) / PAGE_SIZE;
|
|
|
|
pages = kmalloc(pagecount * sizeof(struct page *), GFP_KERNEL);
|
|
|
|
if (!pages)
|
2008-03-13 15:32:35 -04:00
|
|
|
goto out_mm;
|
2008-02-05 01:29:04 -05:00
|
|
|
|
|
|
|
down_read(¤t->mm->mmap_sem);
|
|
|
|
ret = get_user_pages(current, current->mm, uaddr, pagecount,
|
|
|
|
1, 0, pages, NULL);
|
|
|
|
up_read(¤t->mm->mmap_sem);
|
|
|
|
|
|
|
|
if (ret < 0)
|
|
|
|
goto out_free;
|
|
|
|
|
2008-03-13 15:32:35 -04:00
|
|
|
if (ret != pagecount) {
|
|
|
|
pagecount = ret;
|
|
|
|
ret = -EFAULT;
|
|
|
|
goto out_pages;
|
|
|
|
}
|
|
|
|
|
2008-06-06 01:46:31 -04:00
|
|
|
pm.out = (u64 *)buf;
|
|
|
|
pm.end = (u64 *)(buf + count);
|
2008-02-05 01:29:04 -05:00
|
|
|
|
|
|
|
if (!ptrace_may_attach(task)) {
|
|
|
|
ret = -EIO;
|
|
|
|
} else {
|
|
|
|
unsigned long src = *ppos;
|
|
|
|
unsigned long svpfn = src / PM_ENTRY_BYTES;
|
|
|
|
unsigned long start_vaddr = svpfn << PAGE_SHIFT;
|
|
|
|
unsigned long end_vaddr = TASK_SIZE_OF(task);
|
|
|
|
|
|
|
|
/* watch out for wraparound */
|
|
|
|
if (svpfn > TASK_SIZE_OF(task) >> PAGE_SHIFT)
|
|
|
|
start_vaddr = end_vaddr;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The odds are that this will stop walking way
|
|
|
|
* before end_vaddr, because the length of the
|
|
|
|
* user buffer is tracked in "pm", and the walk
|
|
|
|
* will stop when we hit the end of the buffer.
|
|
|
|
*/
|
2008-06-12 18:21:47 -04:00
|
|
|
ret = walk_page_range(start_vaddr, end_vaddr,
|
|
|
|
&pagemap_walk);
|
2008-02-05 01:29:04 -05:00
|
|
|
if (ret == PM_END_OF_BUFFER)
|
|
|
|
ret = 0;
|
|
|
|
/* don't need mmap_sem for these, but this looks cleaner */
|
2008-06-06 01:46:31 -04:00
|
|
|
*ppos += (char *)pm.out - buf;
|
2008-02-05 01:29:04 -05:00
|
|
|
if (!ret)
|
2008-06-06 01:46:31 -04:00
|
|
|
ret = (char *)pm.out - buf;
|
2008-02-05 01:29:04 -05:00
|
|
|
}
|
|
|
|
|
2008-03-13 15:32:35 -04:00
|
|
|
out_pages:
|
2008-02-05 01:29:04 -05:00
|
|
|
for (; pagecount; pagecount--) {
|
|
|
|
page = pages[pagecount-1];
|
|
|
|
if (!PageReserved(page))
|
|
|
|
SetPageDirty(page);
|
|
|
|
page_cache_release(page);
|
|
|
|
}
|
|
|
|
out_free:
|
|
|
|
kfree(pages);
|
2008-03-13 15:32:35 -04:00
|
|
|
out_mm:
|
|
|
|
mmput(mm);
|
2008-02-05 01:29:04 -05:00
|
|
|
out_task:
|
|
|
|
put_task_struct(task);
|
|
|
|
out:
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
const struct file_operations proc_pagemap_operations = {
|
|
|
|
.llseek = mem_lseek, /* borrow this */
|
|
|
|
.read = pagemap_read,
|
|
|
|
};
|
2008-02-05 01:29:07 -05:00
|
|
|
#endif /* CONFIG_PROC_PAGE_MONITOR */
|
2008-02-05 01:29:04 -05:00
|
|
|
|
2005-09-03 18:54:45 -04:00
|
|
|
#ifdef CONFIG_NUMA
|
2006-01-08 04:01:02 -05:00
|
|
|
extern int show_numa_map(struct seq_file *m, void *v);
|
2005-09-03 18:54:45 -04:00
|
|
|
|
2007-05-08 03:26:04 -04:00
|
|
|
static int show_numa_map_checked(struct seq_file *m, void *v)
|
|
|
|
{
|
|
|
|
struct proc_maps_private *priv = m->private;
|
|
|
|
struct task_struct *task = priv->task;
|
|
|
|
|
|
|
|
if (maps_protect && !ptrace_may_attach(task))
|
|
|
|
return -EACCES;
|
|
|
|
|
|
|
|
return show_numa_map(m, v);
|
|
|
|
}
|
|
|
|
|
2008-02-08 07:21:19 -05:00
|
|
|
static const struct seq_operations proc_pid_numa_maps_op = {
|
2006-01-08 04:01:02 -05:00
|
|
|
.start = m_start,
|
|
|
|
.next = m_next,
|
|
|
|
.stop = m_stop,
|
2007-05-08 03:26:04 -04:00
|
|
|
.show = show_numa_map_checked
|
2005-09-03 18:54:45 -04:00
|
|
|
};
|
2006-06-26 03:25:48 -04:00
|
|
|
|
|
|
|
static int numa_maps_open(struct inode *inode, struct file *file)
|
|
|
|
{
|
|
|
|
return do_maps_open(inode, file, &proc_pid_numa_maps_op);
|
|
|
|
}
|
|
|
|
|
2007-02-12 03:55:34 -05:00
|
|
|
const struct file_operations proc_numa_maps_operations = {
|
2006-06-26 03:25:48 -04:00
|
|
|
.open = numa_maps_open,
|
|
|
|
.read = seq_read,
|
|
|
|
.llseek = seq_lseek,
|
2006-06-26 03:25:55 -04:00
|
|
|
.release = seq_release_private,
|
2006-06-26 03:25:48 -04:00
|
|
|
};
|
2005-09-03 18:54:45 -04:00
|
|
|
#endif
|