2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* linux/mm/mlock.c
|
|
|
|
*
|
|
|
|
* (C) Copyright 1995 Linus Torvalds
|
|
|
|
* (C) Copyright 2002 Christoph Hellwig
|
|
|
|
*/
|
|
|
|
|
2006-01-11 21:17:46 +01:00
|
|
|
#include <linux/capability.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <linux/mman.h>
|
|
|
|
#include <linux/mm.h>
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
#include <linux/swap.h>
|
|
|
|
#include <linux/swapops.h>
|
|
|
|
#include <linux/pagemap.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <linux/mempolicy.h>
|
|
|
|
#include <linux/syscalls.h>
|
Detach sched.h from mm.h
First thing mm.h does is including sched.h solely for can_do_mlock() inline
function which has "current" dereference inside. By dealing with can_do_mlock()
mm.h can be detached from sched.h which is good. See below, why.
This patch
a) removes unconditional inclusion of sched.h from mm.h
b) makes can_do_mlock() normal function in mm/mlock.c
c) exports can_do_mlock() to not break compilation
d) adds sched.h inclusions back to files that were getting it indirectly.
e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were
getting them indirectly
Net result is:
a) mm.h users would get less code to open, read, preprocess, parse, ... if
they don't need sched.h
b) sched.h stops being dependency for significant number of files:
on x86_64 allmodconfig touching sched.h results in recompile of 4083 files,
after patch it's only 3744 (-8.3%).
Cross-compile tested on
all arm defconfigs, all mips defconfigs, all powerpc defconfigs,
alpha alpha-up
arm
i386 i386-up i386-defconfig i386-allnoconfig
ia64 ia64-up
m68k
mips
parisc parisc-up
powerpc powerpc-up
s390 s390-up
sparc sparc-up
sparc64 sparc64-up
um-x86_64
x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfig
as well as my two usual configs.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-20 23:22:52 +02:00
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/module.h>
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
#include <linux/rmap.h>
|
|
|
|
#include <linux/mmzone.h>
|
|
|
|
#include <linux/hugetlb.h>
|
|
|
|
|
|
|
|
#include "internal.h"
|
2005-04-17 00:20:36 +02:00
|
|
|
|
Detach sched.h from mm.h
First thing mm.h does is including sched.h solely for can_do_mlock() inline
function which has "current" dereference inside. By dealing with can_do_mlock()
mm.h can be detached from sched.h which is good. See below, why.
This patch
a) removes unconditional inclusion of sched.h from mm.h
b) makes can_do_mlock() normal function in mm/mlock.c
c) exports can_do_mlock() to not break compilation
d) adds sched.h inclusions back to files that were getting it indirectly.
e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were
getting them indirectly
Net result is:
a) mm.h users would get less code to open, read, preprocess, parse, ... if
they don't need sched.h
b) sched.h stops being dependency for significant number of files:
on x86_64 allmodconfig touching sched.h results in recompile of 4083 files,
after patch it's only 3744 (-8.3%).
Cross-compile tested on
all arm defconfigs, all mips defconfigs, all powerpc defconfigs,
alpha alpha-up
arm
i386 i386-up i386-defconfig i386-allnoconfig
ia64 ia64-up
m68k
mips
parisc parisc-up
powerpc powerpc-up
s390 s390-up
sparc sparc-up
sparc64 sparc64-up
um-x86_64
x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfig
as well as my two usual configs.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-20 23:22:52 +02:00
|
|
|
int can_do_mlock(void)
|
|
|
|
{
|
|
|
|
if (capable(CAP_IPC_LOCK))
|
|
|
|
return 1;
|
2010-03-05 22:41:44 +01:00
|
|
|
if (rlimit(RLIMIT_MEMLOCK) != 0)
|
Detach sched.h from mm.h
First thing mm.h does is including sched.h solely for can_do_mlock() inline
function which has "current" dereference inside. By dealing with can_do_mlock()
mm.h can be detached from sched.h which is good. See below, why.
This patch
a) removes unconditional inclusion of sched.h from mm.h
b) makes can_do_mlock() normal function in mm/mlock.c
c) exports can_do_mlock() to not break compilation
d) adds sched.h inclusions back to files that were getting it indirectly.
e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were
getting them indirectly
Net result is:
a) mm.h users would get less code to open, read, preprocess, parse, ... if
they don't need sched.h
b) sched.h stops being dependency for significant number of files:
on x86_64 allmodconfig touching sched.h results in recompile of 4083 files,
after patch it's only 3744 (-8.3%).
Cross-compile tested on
all arm defconfigs, all mips defconfigs, all powerpc defconfigs,
alpha alpha-up
arm
i386 i386-up i386-defconfig i386-allnoconfig
ia64 ia64-up
m68k
mips
parisc parisc-up
powerpc powerpc-up
s390 s390-up
sparc sparc-up
sparc64 sparc64-up
um-x86_64
x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfig
as well as my two usual configs.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-20 23:22:52 +02:00
|
|
|
return 1;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(can_do_mlock);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
/*
|
|
|
|
* Mlocked pages are marked with PageMlocked() flag for efficient testing
|
|
|
|
* in vmscan and, possibly, the fault path; and to support semi-accurate
|
|
|
|
* statistics.
|
|
|
|
*
|
|
|
|
* An mlocked page [PageMlocked(page)] is unevictable. As such, it will
|
|
|
|
* be placed on the LRU "unevictable" list, rather than the [in]active lists.
|
|
|
|
* The unevictable list is an LRU sibling list to the [in]active lists.
|
|
|
|
* PageUnevictable is set to indicate the unevictable state.
|
|
|
|
*
|
|
|
|
* When lazy mlocking via vmscan, it is important to ensure that the
|
|
|
|
* vma's VM_LOCKED status is not concurrently being modified, otherwise we
|
|
|
|
* may have mlocked a page that is being munlocked. So lazy mlock must take
|
|
|
|
* the mmap_sem for read, and verify that the vma really is locked
|
|
|
|
* (see mm/rmap.c).
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* LRU accounting for clear_page_mlock()
|
|
|
|
*/
|
|
|
|
void __clear_page_mlock(struct page *page)
|
|
|
|
{
|
|
|
|
VM_BUG_ON(!PageLocked(page));
|
|
|
|
|
|
|
|
if (!page->mapping) { /* truncated ? */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2008-10-19 05:26:51 +02:00
|
|
|
dec_zone_page_state(page, NR_MLOCK);
|
|
|
|
count_vm_event(UNEVICTABLE_PGCLEARED);
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
if (!isolate_lru_page(page)) {
|
|
|
|
putback_lru_page(page);
|
|
|
|
} else {
|
|
|
|
/*
|
2008-11-12 22:26:53 +01:00
|
|
|
* We lost the race. the page already moved to evictable list.
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*/
|
2008-11-12 22:26:53 +01:00
|
|
|
if (PageUnevictable(page))
|
2008-10-19 05:26:51 +02:00
|
|
|
count_vm_event(UNEVICTABLE_PGSTRANDED);
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Mark page as mlocked if not already.
|
|
|
|
* If page on LRU, isolate and putback to move to unevictable list.
|
|
|
|
*/
|
|
|
|
void mlock_vma_page(struct page *page)
|
|
|
|
{
|
|
|
|
BUG_ON(!PageLocked(page));
|
|
|
|
|
2008-10-19 05:26:51 +02:00
|
|
|
if (!TestSetPageMlocked(page)) {
|
|
|
|
inc_zone_page_state(page, NR_MLOCK);
|
|
|
|
count_vm_event(UNEVICTABLE_PGMLOCKED);
|
|
|
|
if (!isolate_lru_page(page))
|
|
|
|
putback_lru_page(page);
|
|
|
|
}
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
}
|
|
|
|
|
2009-12-15 02:59:55 +01:00
|
|
|
/**
|
|
|
|
* munlock_vma_page - munlock a vma page
|
|
|
|
* @page - page to be unlocked
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*
|
2009-12-15 02:59:55 +01:00
|
|
|
* called from munlock()/munmap() path with page supposedly on the LRU.
|
|
|
|
* When we munlock a page, because the vma where we found the page is being
|
|
|
|
* munlock()ed or munmap()ed, we want to check whether other vmas hold the
|
|
|
|
* page locked so that we can leave it on the unevictable lru list and not
|
|
|
|
* bother vmscan with it. However, to walk the page's rmap list in
|
|
|
|
* try_to_munlock() we must isolate the page from the LRU. If some other
|
|
|
|
* task has removed the page from the LRU, we won't be able to do that.
|
|
|
|
* So we clear the PageMlocked as we might not get another chance. If we
|
|
|
|
* can't isolate the page, we leave it for putback_lru_page() and vmscan
|
|
|
|
* [page_referenced()/try_to_unmap()] to deal with.
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*/
|
2009-12-15 02:59:22 +01:00
|
|
|
void munlock_vma_page(struct page *page)
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
{
|
|
|
|
BUG_ON(!PageLocked(page));
|
|
|
|
|
2008-10-19 05:26:51 +02:00
|
|
|
if (TestClearPageMlocked(page)) {
|
|
|
|
dec_zone_page_state(page, NR_MLOCK);
|
|
|
|
if (!isolate_lru_page(page)) {
|
|
|
|
int ret = try_to_munlock(page);
|
|
|
|
/*
|
|
|
|
* did try_to_unlock() succeed or punt?
|
|
|
|
*/
|
2009-12-15 02:58:58 +01:00
|
|
|
if (ret != SWAP_MLOCK)
|
2008-10-19 05:26:51 +02:00
|
|
|
count_vm_event(UNEVICTABLE_PGMUNLOCKED);
|
|
|
|
|
|
|
|
putback_lru_page(page);
|
|
|
|
} else {
|
|
|
|
/*
|
2009-12-15 02:59:55 +01:00
|
|
|
* Some other task has removed the page from the LRU.
|
|
|
|
* putback_lru_page() will take care of removing the
|
|
|
|
* page from the unevictable list, if necessary.
|
|
|
|
* vmscan [page_referenced()] will move the page back
|
|
|
|
* to the unevictable list if some other vma has it
|
|
|
|
* mlocked.
|
2008-10-19 05:26:51 +02:00
|
|
|
*/
|
|
|
|
if (PageUnevictable(page))
|
|
|
|
count_vm_event(UNEVICTABLE_PGSTRANDED);
|
|
|
|
else
|
|
|
|
count_vm_event(UNEVICTABLE_PGMUNLOCKED);
|
|
|
|
}
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-08-21 01:39:25 +02:00
|
|
|
static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr)
|
|
|
|
{
|
|
|
|
return (vma->vm_flags & VM_GROWSDOWN) &&
|
|
|
|
(vma->vm_start == addr) &&
|
|
|
|
!vma_stack_continue(vma->vm_prev, addr);
|
|
|
|
}
|
|
|
|
|
2008-10-19 05:26:50 +02:00
|
|
|
/**
|
2009-09-22 02:03:23 +02:00
|
|
|
* __mlock_vma_pages_range() - mlock a range of pages in the vma.
|
2008-10-19 05:26:50 +02:00
|
|
|
* @vma: target vma
|
|
|
|
* @start: start address
|
|
|
|
* @end: end address
|
|
|
|
*
|
2009-09-22 02:03:23 +02:00
|
|
|
* This takes care of making the pages present too.
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*
|
2008-10-19 05:26:50 +02:00
|
|
|
* return 0 on success, negative error code on error.
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*
|
2008-10-19 05:26:50 +02:00
|
|
|
* vma->vm_mm->mmap_sem must be held for at least read.
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*/
|
2008-10-19 05:26:50 +02:00
|
|
|
static long __mlock_vma_pages_range(struct vm_area_struct *vma,
|
2011-01-14 00:46:14 +01:00
|
|
|
unsigned long start, unsigned long end,
|
|
|
|
int *nonblocking)
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
{
|
|
|
|
struct mm_struct *mm = vma->vm_mm;
|
|
|
|
unsigned long addr = start;
|
|
|
|
int nr_pages = (end - start) / PAGE_SIZE;
|
2009-09-22 02:03:23 +02:00
|
|
|
int gup_flags;
|
2008-10-19 05:26:50 +02:00
|
|
|
|
|
|
|
VM_BUG_ON(start & ~PAGE_MASK);
|
|
|
|
VM_BUG_ON(end & ~PAGE_MASK);
|
|
|
|
VM_BUG_ON(start < vma->vm_start);
|
|
|
|
VM_BUG_ON(end > vma->vm_end);
|
2009-09-22 02:03:23 +02:00
|
|
|
VM_BUG_ON(!rwsem_is_locked(&mm->mmap_sem));
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
|
2011-01-14 00:46:12 +01:00
|
|
|
gup_flags = FOLL_TOUCH;
|
2011-01-14 00:46:09 +01:00
|
|
|
/*
|
|
|
|
* We want to touch writable mappings with a write fault in order
|
|
|
|
* to break COW, except for shared mappings because these don't COW
|
|
|
|
* and we would not want to dirty them for nothing.
|
|
|
|
*/
|
|
|
|
if ((vma->vm_flags & (VM_WRITE | VM_SHARED)) == VM_WRITE)
|
2009-09-22 02:03:31 +02:00
|
|
|
gup_flags |= FOLL_WRITE;
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
|
2011-02-01 02:03:41 +01:00
|
|
|
/*
|
|
|
|
* We want mlock to succeed for regions that have any permissions
|
|
|
|
* other than PROT_NONE.
|
|
|
|
*/
|
|
|
|
if (vma->vm_flags & (VM_READ | VM_WRITE | VM_EXEC))
|
|
|
|
gup_flags |= FOLL_FORCE;
|
|
|
|
|
2011-01-14 00:46:12 +01:00
|
|
|
if (vma->vm_flags & VM_LOCKED)
|
|
|
|
gup_flags |= FOLL_MLOCK;
|
|
|
|
|
2010-08-15 20:35:52 +02:00
|
|
|
/* We don't try to access the guard page of a stack vma */
|
2010-08-21 01:39:25 +02:00
|
|
|
if (stack_guard_page(vma, start)) {
|
|
|
|
addr += PAGE_SIZE;
|
|
|
|
nr_pages--;
|
2010-08-15 20:35:52 +02:00
|
|
|
}
|
|
|
|
|
2011-01-14 00:46:14 +01:00
|
|
|
return __get_user_pages(current, mm, addr, nr_pages, gup_flags,
|
|
|
|
NULL, NULL, nonblocking);
|
2008-10-19 05:26:56 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* convert get_user_pages() return value to posix mlock() error
|
|
|
|
*/
|
|
|
|
static int __mlock_posix_error_return(long retval)
|
|
|
|
{
|
|
|
|
if (retval == -EFAULT)
|
|
|
|
retval = -ENOMEM;
|
|
|
|
else if (retval == -ENOMEM)
|
|
|
|
retval = -EAGAIN;
|
|
|
|
return retval;
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
}
|
|
|
|
|
2008-10-19 05:26:50 +02:00
|
|
|
/**
|
|
|
|
* mlock_vma_pages_range() - mlock pages in specified vma range.
|
|
|
|
* @vma - the vma containing the specfied address range
|
|
|
|
* @start - starting address in @vma to mlock
|
|
|
|
* @end - end address [+1] in @vma to mlock
|
|
|
|
*
|
|
|
|
* For mmap()/mremap()/expansion of mlocked vma.
|
|
|
|
*
|
|
|
|
* return 0 on success for "normal" vmas.
|
|
|
|
*
|
|
|
|
* return number of pages [> 0] to be removed from locked_vm on success
|
|
|
|
* of "special" vmas.
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*/
|
2008-10-19 05:26:50 +02:00
|
|
|
long mlock_vma_pages_range(struct vm_area_struct *vma,
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
unsigned long start, unsigned long end)
|
|
|
|
{
|
|
|
|
int nr_pages = (end - start) / PAGE_SIZE;
|
|
|
|
BUG_ON(!(vma->vm_flags & VM_LOCKED));
|
|
|
|
|
|
|
|
/*
|
|
|
|
* filter unlockable vmas
|
|
|
|
*/
|
|
|
|
if (vma->vm_flags & (VM_IO | VM_PFNMAP))
|
|
|
|
goto no_mlock;
|
|
|
|
|
|
|
|
if (!((vma->vm_flags & (VM_DONTEXPAND | VM_RESERVED)) ||
|
|
|
|
is_vm_hugetlb_page(vma) ||
|
mlock: downgrade mmap sem while populating mlocked regions
We need to hold the mmap_sem for write to initiatate mlock()/munlock()
because we may need to merge/split vmas. However, this can lead to very
long lock hold times attempting to fault in a large memory region to mlock
it into memory. This can hold off other faults against the mm
[multithreaded tasks] and other scans of the mm, such as via /proc. To
alleviate this, downgrade the mmap_sem to read mode during the population
of the region for locking. This is especially the case if we need to
reclaim memory to lock down the region. We [probably?] don't need to do
this for unlocking as all of the pages should be resident--they're already
mlocked.
Now, the caller's of the mlock functions [mlock_fixup() and
mlock_vma_pages_range()] expect the mmap_sem to be returned in write mode.
Changing all callers appears to be way too much effort at this point.
So, restore write mode before returning. Note that this opens a window
where the mmap list could change in a multithreaded process. So, at least
for mlock_fixup(), where we could be called in a loop over multiple vmas,
we check that a vma still exists at the start address and that vma still
covers the page range [start,end). If not, we return an error, -EAGAIN,
and let the caller deal with it.
Return -EAGAIN from mlock_vma_pages_range() function and mlock_fixup() if
the vma at 'start' disappears or changes so that the page range
[start,end) is no longer contained in the vma. Again, let the caller deal
with it. Looks like only sys_remap_file_pages() [via mmap_region()]
should actually care.
With this patch, I no longer see processes like ps(1) blocked for seconds
or minutes at a time waiting for a large [multiple gigabyte] region to be
locked down. However, I occassionally see delays while unlocking or
unmapping a large mlocked region. Should we also downgrade the mmap_sem
for the unlock path?
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:49 +02:00
|
|
|
vma == get_gate_vma(current))) {
|
|
|
|
|
2011-01-14 00:46:14 +01:00
|
|
|
__mlock_vma_pages_range(vma, start, end, NULL);
|
2009-02-08 21:56:58 +01:00
|
|
|
|
|
|
|
/* Hide errors from mmap() and other callers */
|
|
|
|
return 0;
|
mlock: downgrade mmap sem while populating mlocked regions
We need to hold the mmap_sem for write to initiatate mlock()/munlock()
because we may need to merge/split vmas. However, this can lead to very
long lock hold times attempting to fault in a large memory region to mlock
it into memory. This can hold off other faults against the mm
[multithreaded tasks] and other scans of the mm, such as via /proc. To
alleviate this, downgrade the mmap_sem to read mode during the population
of the region for locking. This is especially the case if we need to
reclaim memory to lock down the region. We [probably?] don't need to do
this for unlocking as all of the pages should be resident--they're already
mlocked.
Now, the caller's of the mlock functions [mlock_fixup() and
mlock_vma_pages_range()] expect the mmap_sem to be returned in write mode.
Changing all callers appears to be way too much effort at this point.
So, restore write mode before returning. Note that this opens a window
where the mmap list could change in a multithreaded process. So, at least
for mlock_fixup(), where we could be called in a loop over multiple vmas,
we check that a vma still exists at the start address and that vma still
covers the page range [start,end). If not, we return an error, -EAGAIN,
and let the caller deal with it.
Return -EAGAIN from mlock_vma_pages_range() function and mlock_fixup() if
the vma at 'start' disappears or changes so that the page range
[start,end) is no longer contained in the vma. Again, let the caller deal
with it. Looks like only sys_remap_file_pages() [via mmap_region()]
should actually care.
With this patch, I no longer see processes like ps(1) blocked for seconds
or minutes at a time waiting for a large [multiple gigabyte] region to be
locked down. However, I occassionally see delays while unlocking or
unmapping a large mlocked region. Should we also downgrade the mmap_sem
for the unlock path?
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:49 +02:00
|
|
|
}
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* User mapped kernel pages or huge pages:
|
|
|
|
* make these pages present to populate the ptes, but
|
|
|
|
* fall thru' to reset VM_LOCKED--no need to unlock, and
|
|
|
|
* return nr_pages so these don't get counted against task's
|
|
|
|
* locked limit. huge pages are already counted against
|
|
|
|
* locked vm limit.
|
|
|
|
*/
|
|
|
|
make_pages_present(start, end);
|
|
|
|
|
|
|
|
no_mlock:
|
|
|
|
vma->vm_flags &= ~VM_LOCKED; /* and don't come back! */
|
2008-10-19 05:26:50 +02:00
|
|
|
return nr_pages; /* error or pages NOT mlocked */
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2008-10-19 05:26:50 +02:00
|
|
|
* munlock_vma_pages_range() - munlock all pages in the vma range.'
|
|
|
|
* @vma - vma containing range to be munlock()ed.
|
|
|
|
* @start - start address in @vma of the range
|
|
|
|
* @end - end of range in @vma.
|
|
|
|
*
|
|
|
|
* For mremap(), munmap() and exit().
|
|
|
|
*
|
|
|
|
* Called with @vma VM_LOCKED.
|
|
|
|
*
|
|
|
|
* Returns with VM_LOCKED cleared. Callers must be prepared to
|
|
|
|
* deal with this.
|
|
|
|
*
|
|
|
|
* We don't save and restore VM_LOCKED here because pages are
|
|
|
|
* still on lru. In unmap path, pages might be scanned by reclaim
|
|
|
|
* and re-mlocked by try_to_{munlock|unmap} before we unmap and
|
|
|
|
* free them. This will result in freeing mlocked pages.
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*/
|
2008-10-19 05:26:50 +02:00
|
|
|
void munlock_vma_pages_range(struct vm_area_struct *vma,
|
2009-09-22 02:03:23 +02:00
|
|
|
unsigned long start, unsigned long end)
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
{
|
2009-09-22 02:03:23 +02:00
|
|
|
unsigned long addr;
|
|
|
|
|
|
|
|
lru_add_drain();
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
vma->vm_flags &= ~VM_LOCKED;
|
2009-09-22 02:03:23 +02:00
|
|
|
|
|
|
|
for (addr = start; addr < end; addr += PAGE_SIZE) {
|
2009-09-22 02:03:32 +02:00
|
|
|
struct page *page;
|
|
|
|
/*
|
|
|
|
* Although FOLL_DUMP is intended for get_dump_page(),
|
|
|
|
* it just so happens that its special treatment of the
|
|
|
|
* ZERO_PAGE (returning an error instead of doing get_page)
|
|
|
|
* suits munlock very well (and if somehow an abnormal page
|
|
|
|
* has sneaked into the range, we won't oops here: great).
|
|
|
|
*/
|
|
|
|
page = follow_page(vma, addr, FOLL_GET | FOLL_DUMP);
|
|
|
|
if (page && !IS_ERR(page)) {
|
2009-09-22 02:03:23 +02:00
|
|
|
lock_page(page);
|
2009-09-22 02:03:32 +02:00
|
|
|
/*
|
|
|
|
* Like in __mlock_vma_pages_range(),
|
|
|
|
* because we lock page here and migration is
|
|
|
|
* blocked by the elevated reference, we need
|
|
|
|
* only check for file-cache page truncation.
|
|
|
|
*/
|
2009-09-22 02:03:23 +02:00
|
|
|
if (page->mapping)
|
|
|
|
munlock_vma_page(page);
|
|
|
|
unlock_page(page);
|
|
|
|
put_page(page);
|
|
|
|
}
|
|
|
|
cond_resched();
|
|
|
|
}
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* mlock_fixup - handle mlock[all]/munlock[all] requests.
|
|
|
|
*
|
|
|
|
* Filters out "special" vmas -- VM_LOCKED never gets set for these, and
|
|
|
|
* munlock is a no-op. However, for some special vmas, we go ahead and
|
|
|
|
* populate the ptes via make_pages_present().
|
|
|
|
*
|
|
|
|
* For vmas that pass the filters, merge/split as appropriate.
|
|
|
|
*/
|
2005-04-17 00:20:36 +02:00
|
|
|
static int mlock_fixup(struct vm_area_struct *vma, struct vm_area_struct **prev,
|
|
|
|
unsigned long start, unsigned long end, unsigned int newflags)
|
|
|
|
{
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
struct mm_struct *mm = vma->vm_mm;
|
2005-04-17 00:20:36 +02:00
|
|
|
pgoff_t pgoff;
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
int nr_pages;
|
2005-04-17 00:20:36 +02:00
|
|
|
int ret = 0;
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
int lock = newflags & VM_LOCKED;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
if (newflags == vma->vm_flags || (vma->vm_flags & VM_SPECIAL) ||
|
|
|
|
is_vm_hugetlb_page(vma) || vma == get_gate_vma(current))
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
goto out; /* don't set VM_LOCKED, don't count */
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
|
|
|
|
*prev = vma_merge(mm, *prev, start, end, newflags, vma->anon_vma,
|
|
|
|
vma->vm_file, pgoff, vma_policy(vma));
|
|
|
|
if (*prev) {
|
|
|
|
vma = *prev;
|
|
|
|
goto success;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (start != vma->vm_start) {
|
|
|
|
ret = split_vma(mm, vma, start, 1);
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (end != vma->vm_end) {
|
|
|
|
ret = split_vma(mm, vma, end, 0);
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
success:
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
/*
|
|
|
|
* Keep track of amount of locked VM.
|
|
|
|
*/
|
|
|
|
nr_pages = (end - start) >> PAGE_SHIFT;
|
|
|
|
if (!lock)
|
|
|
|
nr_pages = -nr_pages;
|
|
|
|
mm->locked_vm += nr_pages;
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* vm_flags is protected by the mmap_sem held in write mode.
|
|
|
|
* It's okay if try_to_unmap_one unmaps a page just after we
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
* set VM_LOCKED, __mlock_vma_pages_range will bring it back.
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
|
|
|
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
if (lock)
|
2009-09-22 02:03:23 +02:00
|
|
|
vma->vm_flags = newflags;
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
else
|
2009-09-22 02:03:23 +02:00
|
|
|
munlock_vma_pages_range(vma, start, end);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
out:
|
mlock: mlocked pages are unevictable
Make sure that mlocked pages also live on the unevictable LRU, so kswapd
will not scan them over and over again.
This is achieved through various strategies:
1) add yet another page flag--PG_mlocked--to indicate that
the page is locked for efficient testing in vmscan and,
optionally, fault path. This allows early culling of
unevictable pages, preventing them from getting to
page_referenced()/try_to_unmap(). Also allows separate
accounting of mlock'd pages, as Nick's original patch
did.
Note: Nick's original mlock patch used a PG_mlocked
flag. I had removed this in favor of the PG_unevictable
flag + an mlock_count [new page struct member]. I
restored the PG_mlocked flag to eliminate the new
count field.
2) add the mlock/unevictable infrastructure to mm/mlock.c,
with internal APIs in mm/internal.h. This is a rework
of Nick's original patch to these files, taking into
account that mlocked pages are now kept on unevictable
LRU list.
3) update vmscan.c:page_evictable() to check PageMlocked()
and, if vma passed in, the vm_flags. Note that the vma
will only be passed in for new pages in the fault path;
and then only if the "cull unevictable pages in fault
path" patch is included.
4) add try_to_unlock() to rmap.c to walk a page's rmap and
ClearPageMlocked() if no other vmas have it mlocked.
Reuses as much of try_to_unmap() as possible. This
effectively replaces the use of one of the lru list links
as an mlock count. If this mechanism let's pages in mlocked
vmas leak through w/o PG_mlocked set [I don't know that it
does], we should catch them later in try_to_unmap(). One
hopes this will be rare, as it will be relatively expensive.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
Signed-off-by: Nick Piggin <npiggin@suse.de>
splitlru: introduce __get_user_pages():
New munlock processing need to GUP_FLAGS_IGNORE_VMA_PERMISSIONS.
because current get_user_pages() can't grab PROT_NONE pages theresore it
cause PROT_NONE pages can't munlock.
[akpm@linux-foundation.org: fix this for pagemap-pass-mm-into-pagewalkers.patch]
[akpm@linux-foundation.org: untangle patch interdependencies]
[akpm@linux-foundation.org: fix things after out-of-order merging]
[hugh@veritas.com: fix page-flags mess]
[lee.schermerhorn@hp.com: fix munlock page table walk - now requires 'mm']
[kosaki.motohiro@jp.fujitsu.com: build fix]
[kosaki.motohiro@jp.fujitsu.com: fix truncate race and sevaral comments]
[kosaki.motohiro@jp.fujitsu.com: splitlru: introduce __get_user_pages()]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 05:26:44 +02:00
|
|
|
*prev = vma;
|
2005-04-17 00:20:36 +02:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int do_mlock(unsigned long start, size_t len, int on)
|
|
|
|
{
|
|
|
|
unsigned long nstart, end, tmp;
|
|
|
|
struct vm_area_struct * vma, * prev;
|
|
|
|
int error;
|
|
|
|
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
VM_BUG_ON(start & ~PAGE_MASK);
|
|
|
|
VM_BUG_ON(len != PAGE_ALIGN(len));
|
2005-04-17 00:20:36 +02:00
|
|
|
end = start + len;
|
|
|
|
if (end < start)
|
|
|
|
return -EINVAL;
|
|
|
|
if (end == start)
|
|
|
|
return 0;
|
|
|
|
vma = find_vma_prev(current->mm, start, &prev);
|
|
|
|
if (!vma || vma->vm_start > start)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
if (start > vma->vm_start)
|
|
|
|
prev = vma;
|
|
|
|
|
|
|
|
for (nstart = start ; ; ) {
|
|
|
|
unsigned int newflags;
|
|
|
|
|
|
|
|
/* Here we know that vma->vm_start <= nstart < vma->vm_end. */
|
|
|
|
|
|
|
|
newflags = vma->vm_flags | VM_LOCKED;
|
|
|
|
if (!on)
|
|
|
|
newflags &= ~VM_LOCKED;
|
|
|
|
|
|
|
|
tmp = vma->vm_end;
|
|
|
|
if (tmp > end)
|
|
|
|
tmp = end;
|
|
|
|
error = mlock_fixup(vma, &prev, nstart, tmp, newflags);
|
|
|
|
if (error)
|
|
|
|
break;
|
|
|
|
nstart = tmp;
|
|
|
|
if (nstart < prev->vm_end)
|
|
|
|
nstart = prev->vm_end;
|
|
|
|
if (nstart >= end)
|
|
|
|
break;
|
|
|
|
|
|
|
|
vma = prev->vm_next;
|
|
|
|
if (!vma || vma->vm_start != nstart) {
|
|
|
|
error = -ENOMEM;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return error;
|
|
|
|
}
|
|
|
|
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
static int do_mlock_pages(unsigned long start, size_t len, int ignore_errors)
|
|
|
|
{
|
|
|
|
struct mm_struct *mm = current->mm;
|
|
|
|
unsigned long end, nstart, nend;
|
|
|
|
struct vm_area_struct *vma = NULL;
|
2011-01-14 00:46:14 +01:00
|
|
|
int locked = 0;
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
VM_BUG_ON(start & ~PAGE_MASK);
|
|
|
|
VM_BUG_ON(len != PAGE_ALIGN(len));
|
|
|
|
end = start + len;
|
|
|
|
|
|
|
|
for (nstart = start; nstart < end; nstart = nend) {
|
|
|
|
/*
|
|
|
|
* We want to fault in pages for [nstart; end) address range.
|
|
|
|
* Find first corresponding VMA.
|
|
|
|
*/
|
2011-01-14 00:46:14 +01:00
|
|
|
if (!locked) {
|
|
|
|
locked = 1;
|
|
|
|
down_read(&mm->mmap_sem);
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
vma = find_vma(mm, nstart);
|
2011-01-14 00:46:14 +01:00
|
|
|
} else if (nstart >= vma->vm_end)
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
vma = vma->vm_next;
|
|
|
|
if (!vma || vma->vm_start >= end)
|
|
|
|
break;
|
|
|
|
/*
|
|
|
|
* Set [nstart; nend) to intersection of desired address
|
|
|
|
* range with the first VMA. Also, skip undesirable VMA types.
|
|
|
|
*/
|
|
|
|
nend = min(end, vma->vm_end);
|
|
|
|
if (vma->vm_flags & (VM_IO | VM_PFNMAP))
|
|
|
|
continue;
|
|
|
|
if (nstart < vma->vm_start)
|
|
|
|
nstart = vma->vm_start;
|
|
|
|
/*
|
2011-01-14 00:46:14 +01:00
|
|
|
* Now fault in a range of pages. __mlock_vma_pages_range()
|
|
|
|
* double checks the vma flags, so that it won't mlock pages
|
|
|
|
* if the vma was already munlocked.
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
*/
|
2011-01-14 00:46:14 +01:00
|
|
|
ret = __mlock_vma_pages_range(vma, nstart, nend, &locked);
|
|
|
|
if (ret < 0) {
|
|
|
|
if (ignore_errors) {
|
|
|
|
ret = 0;
|
|
|
|
continue; /* continue at next VMA */
|
|
|
|
}
|
2011-01-14 00:46:12 +01:00
|
|
|
ret = __mlock_posix_error_return(ret);
|
|
|
|
break;
|
|
|
|
}
|
2011-01-14 00:46:14 +01:00
|
|
|
nend = nstart + ret * PAGE_SIZE;
|
|
|
|
ret = 0;
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
}
|
2011-01-14 00:46:14 +01:00
|
|
|
if (locked)
|
|
|
|
up_read(&mm->mmap_sem);
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
return ret; /* 0 or negative error code */
|
|
|
|
}
|
|
|
|
|
2009-01-14 14:14:15 +01:00
|
|
|
SYSCALL_DEFINE2(mlock, unsigned long, start, size_t, len)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
unsigned long locked;
|
|
|
|
unsigned long lock_limit;
|
|
|
|
int error = -ENOMEM;
|
|
|
|
|
|
|
|
if (!can_do_mlock())
|
|
|
|
return -EPERM;
|
|
|
|
|
2008-11-12 22:26:53 +01:00
|
|
|
lru_add_drain_all(); /* flush pagevec */
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
down_write(¤t->mm->mmap_sem);
|
|
|
|
len = PAGE_ALIGN(len + (start & ~PAGE_MASK));
|
|
|
|
start &= PAGE_MASK;
|
|
|
|
|
|
|
|
locked = len >> PAGE_SHIFT;
|
|
|
|
locked += current->mm->locked_vm;
|
|
|
|
|
2010-03-05 22:41:44 +01:00
|
|
|
lock_limit = rlimit(RLIMIT_MEMLOCK);
|
2005-04-17 00:20:36 +02:00
|
|
|
lock_limit >>= PAGE_SHIFT;
|
|
|
|
|
|
|
|
/* check against resource limits */
|
|
|
|
if ((locked <= lock_limit) || capable(CAP_IPC_LOCK))
|
|
|
|
error = do_mlock(start, len, 1);
|
|
|
|
up_write(¤t->mm->mmap_sem);
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
if (!error)
|
|
|
|
error = do_mlock_pages(start, len, 0);
|
2005-04-17 00:20:36 +02:00
|
|
|
return error;
|
|
|
|
}
|
|
|
|
|
2009-01-14 14:14:15 +01:00
|
|
|
SYSCALL_DEFINE2(munlock, unsigned long, start, size_t, len)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
down_write(¤t->mm->mmap_sem);
|
|
|
|
len = PAGE_ALIGN(len + (start & ~PAGE_MASK));
|
|
|
|
start &= PAGE_MASK;
|
|
|
|
ret = do_mlock(start, len, 0);
|
|
|
|
up_write(¤t->mm->mmap_sem);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int do_mlockall(int flags)
|
|
|
|
{
|
|
|
|
struct vm_area_struct * vma, * prev = NULL;
|
|
|
|
unsigned int def_flags = 0;
|
|
|
|
|
|
|
|
if (flags & MCL_FUTURE)
|
|
|
|
def_flags = VM_LOCKED;
|
|
|
|
current->mm->def_flags = def_flags;
|
|
|
|
if (flags == MCL_FUTURE)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
for (vma = current->mm->mmap; vma ; vma = prev->vm_next) {
|
|
|
|
unsigned int newflags;
|
|
|
|
|
|
|
|
newflags = vma->vm_flags | VM_LOCKED;
|
|
|
|
if (!(flags & MCL_CURRENT))
|
|
|
|
newflags &= ~VM_LOCKED;
|
|
|
|
|
|
|
|
/* Ignore errors */
|
|
|
|
mlock_fixup(vma, &prev, vma->vm_start, vma->vm_end, newflags);
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-01-14 14:14:16 +01:00
|
|
|
SYSCALL_DEFINE1(mlockall, int, flags)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
unsigned long lock_limit;
|
|
|
|
int ret = -EINVAL;
|
|
|
|
|
|
|
|
if (!flags || (flags & ~(MCL_CURRENT | MCL_FUTURE)))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
ret = -EPERM;
|
|
|
|
if (!can_do_mlock())
|
|
|
|
goto out;
|
|
|
|
|
2008-11-12 22:26:53 +01:00
|
|
|
lru_add_drain_all(); /* flush pagevec */
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
down_write(¤t->mm->mmap_sem);
|
|
|
|
|
2010-03-05 22:41:44 +01:00
|
|
|
lock_limit = rlimit(RLIMIT_MEMLOCK);
|
2005-04-17 00:20:36 +02:00
|
|
|
lock_limit >>= PAGE_SHIFT;
|
|
|
|
|
|
|
|
ret = -ENOMEM;
|
|
|
|
if (!(flags & MCL_CURRENT) || (current->mm->total_vm <= lock_limit) ||
|
|
|
|
capable(CAP_IPC_LOCK))
|
|
|
|
ret = do_mlockall(flags);
|
|
|
|
up_write(¤t->mm->mmap_sem);
|
mlock: only hold mmap_sem in shared mode when faulting in pages
Currently mlock() holds mmap_sem in exclusive mode while the pages get
faulted in. In the case of a large mlock, this can potentially take a
very long time, during which various commands such as 'ps auxw' will
block. This makes sysadmins unhappy:
real 14m36.232s
user 0m0.003s
sys 0m0.015s
(output from 'time ps auxw' while a 20GB file was being mlocked without
being previously preloaded into page cache)
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages, in small batches, releasing mmap_sem when we
block on disk access or when we detect some contention.
This patch:
Before this change, mlock() holds mmap_sem in exclusive mode while the
pages get faulted in. In the case of a large mlock, this can potentially
take a very long time. Various things will block while mmap_sem is held,
including 'ps auxw'. This can make sysadmins angry.
I propose that mlock() could release mmap_sem after the VM_LOCKED bits
have been set in all appropriate VMAs. Then a second pass could be done
to actually mlock the pages with mmap_sem held for reads only. We need to
recheck the vma flags after we re-acquire mmap_sem, but this is easy.
In the case where a vma has been munlocked before mlock completes, pages
that were already marked as PageMlocked() are handled by the munlock()
call, and mlock() is careful to not mark new page batches as PageMlocked()
after the munlock() call has cleared the VM_LOCKED vma flags. So, the end
result will be identical to what'd happen if munlock() had executed after
the mlock() call.
In a later change, I will allow the second pass to release mmap_sem when
blocking on disk accesses or when it is otherwise contended, so that it
won't be held for long periods of time even in shared mode.
Signed-off-by: Michel Lespinasse <walken@google.com>
Tested-by: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-01-14 00:46:10 +01:00
|
|
|
if (!ret && (flags & MCL_CURRENT)) {
|
|
|
|
/* Ignore errors */
|
|
|
|
do_mlock_pages(0, TASK_SIZE, 1);
|
|
|
|
}
|
2005-04-17 00:20:36 +02:00
|
|
|
out:
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2009-01-14 14:14:16 +01:00
|
|
|
SYSCALL_DEFINE0(munlockall)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
down_write(¤t->mm->mmap_sem);
|
|
|
|
ret = do_mlockall(0);
|
|
|
|
up_write(¤t->mm->mmap_sem);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Objects with different lifetime than processes (SHM_LOCK and SHM_HUGETLB
|
|
|
|
* shm segments) get accounted against the user_struct instead.
|
|
|
|
*/
|
|
|
|
static DEFINE_SPINLOCK(shmlock_user_lock);
|
|
|
|
|
|
|
|
int user_shm_lock(size_t size, struct user_struct *user)
|
|
|
|
{
|
|
|
|
unsigned long lock_limit, locked;
|
|
|
|
int allowed = 0;
|
|
|
|
|
|
|
|
locked = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
|
2010-03-05 22:41:44 +01:00
|
|
|
lock_limit = rlimit(RLIMIT_MEMLOCK);
|
2007-07-16 08:38:25 +02:00
|
|
|
if (lock_limit == RLIM_INFINITY)
|
|
|
|
allowed = 1;
|
2005-04-17 00:20:36 +02:00
|
|
|
lock_limit >>= PAGE_SHIFT;
|
|
|
|
spin_lock(&shmlock_user_lock);
|
2007-07-16 08:38:25 +02:00
|
|
|
if (!allowed &&
|
|
|
|
locked + user->locked_shm > lock_limit && !capable(CAP_IPC_LOCK))
|
2005-04-17 00:20:36 +02:00
|
|
|
goto out;
|
|
|
|
get_uid(user);
|
|
|
|
user->locked_shm += locked;
|
|
|
|
allowed = 1;
|
|
|
|
out:
|
|
|
|
spin_unlock(&shmlock_user_lock);
|
|
|
|
return allowed;
|
|
|
|
}
|
|
|
|
|
|
|
|
void user_shm_unlock(size_t size, struct user_struct *user)
|
|
|
|
{
|
|
|
|
spin_lock(&shmlock_user_lock);
|
|
|
|
user->locked_shm -= (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
|
|
|
|
spin_unlock(&shmlock_user_lock);
|
|
|
|
free_uid(user);
|
|
|
|
}
|