- 15 Dec, 2009 40 commits
-
-
Jie Zhang authored
The NOMMU code currently clears all anonymous mmapped memory. While this is what we want in the default case, all memory allocation from userspace under NOMMU has to go through this interface, including malloc() which is allowed to return uninitialized memory. This can easily be a significant performance penalty. So for constrained embedded systems were security is irrelevant, allow people to avoid clearing memory unnecessarily. This also alters the ELF-FDPIC binfmt such that it obtains uninitialised memory for the brk and stack region. Signed-off-by:
Jie Zhang <jie.zhang@analog.com> Signed-off-by:
Robin Getz <rgetz@blackfin.uclinux.org> Signed-off-by:
Mike Frysinger <vapier@gentoo.org> Signed-off-by:
David Howells <dhowells@redhat.com> Acked-by:
Paul Mundt <lethal@linux-sh.org> Acked-by:
Greg Ungerer <gerg@snapgear.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Naoya Horiguchi authored
This patch enables extraction of the pfn of a hugepage from /proc/pid/pagemap in an architecture independent manner. Details ------- My test program (leak_pagemap) works as follows: - creat() and mmap() a file on hugetlbfs (file size is 200MB == 100 hugepages,) - read()/write() something on it, - call page-types with option -p, - munmap() and unlink() the file on hugetlbfs Without my patches ------------------ $ ./leak_pagemap flags page-count MB symbolic-flags long-symbolic-flags 0x0000000000000000 1 0 __________________________________ 0x0000000000000804 1 0 __R________M______________________ referenced,mmap 0x000000000000086c 81 0 __RU_lA____M______________________ referenced,uptodate,lru,active,mmap 0x0000000000005808 5 0 ___U_______Ma_b___________________ uptodate,mmap,anonymous,swapbacked 0x0000000000005868 12 0 ___U_lA____Ma_b___________________ uptodate,lru,active,mmap,anonymous,swapbacked 0x000000000000586c 1 0 __RU_lA____Ma_b___________________ referenced,uptodate,lru,active,mmap,anonymous,swapbacked total 101 0 The output of page-types don't show any hugepage. With my patches --------------- $ ./leak_pagemap flags page-count MB symbolic-flags long-symbolic-flags 0x0000000000000000 1 0 __________________________________ 0x0000000000030000 51100 199 ________________TG________________ compound_tail,huge 0x0000000000028018 100 0 ___UD__________H_G________________ uptodate,dirty,compound_head,huge 0x0000000000000804 1 0 __R________M______________________ referenced,mmap 0x000000000000080c 1 0 __RU_______M______________________ referenced,uptodate,mmap 0x000000000000086c 80 0 __RU_lA____M______________________ referenced,uptodate,lru,active,mmap 0x0000000000005808 4 0 ___U_______Ma_b___________________ uptodate,mmap,anonymous,swapbacked 0x0000000000005868 12 0 ___U_lA____Ma_b___________________ uptodate,lru,active,mmap,anonymous,swapbacked 0x000000000000586c 1 0 __RU_lA____Ma_b___________________ referenced,uptodate,lru,active,mmap,anonymous,swapbacked total 51300 200 The output of page-types shows 51200 pages contributing to hugepages, containing 100 head pages and 51100 tail pages as expected. [akpm@linux-foundation.org: build fix] Signed-off-by:Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: David Rientjes <rientjes@google.com> Signed-off-by:
-
Naoya Horiguchi authored
Most callers of pmd_none_or_clear_bad() check whether the target page is in a hugepage or not, but walk_page_range() do not check it. So if we read /proc/pid/pagemap for the hugepage on x86 machine, the hugepage memory is leaked as shown below. This patch fixes it. Details ======= My test program (leak_pagemap) works as follows: - creat() and mmap() a file on hugetlbfs (file size is 200MB == 100 hugepages,) - read()/write() something on it, - call page-types with option -p (walk around the page tables), - munmap() and unlink() the file on hugetlbfs Without my patches ------------------ $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_pagemap [snip output] $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 900 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs/ $ 100 hugepages are accounted as used while there is no file on hugetlbfs. With my patches --------------- $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_pagemap [snip output] $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs $ No memory leaks. Signed-off-by:
-
Naoya Horiguchi authored
Most callers of pmd_none_or_clear_bad() check whether the target page is in a hugepage or not, but mincore() and walk_page_range() do not check it. So if we use mincore() on a hugepage on x86 machine, the hugepage memory is leaked as shown below. This patch fixes it by extending mincore() system call to support hugepages. Details ======= My test program (leak_mincore) works as follows: - creat() and mmap() a file on hugetlbfs (file size is 200MB == 100 hugepages,) - read()/write() something on it, - call mincore() for first ten pages and printf() the values of *vec - munmap() and unlink() the file on hugetlbfs Without my patch ---------------- $ cat /proc/meminfo| grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_mincore vec[0] 0 vec[1] 0 vec[2] 0 vec[3] 0 vec[4] 0 vec[5] 0 vec[6] 0 vec[7] 0 vec[8] 0 vec[9] 0 $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 999 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs/ $ Return values in *vec from mincore() are set to 0, while the hugepage should be in memory, and 1 hugepage is still accounted as used while there is no file on hugetlbfs. With my patch ------------- $ cat /proc/meminfo| grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_mincore vec[0] 1 vec[1] 1 vec[2] 1 vec[3] 1 vec[4] 1 vec[5] 1 vec[6] 1 vec[7] 1 vec[8] 1 vec[9] 1 $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs/ $ Return value in *vec set to 1 and no memory leaks. [akpm@linux-foundation.org: cleanup] [akpm@linux-foundation.org: build fix] Signed-off-by:
-
Mel Gorman authored
If a user asks for a hugepage pool resize but specified a large number, the machine can begin trashing. In response, they might hit ctrl-c but signals are ignored and the pool resize continues until it fails an allocation. This can take a considerable amount of time so this patch aborts a pool resize if a signal is pending. Suggested by Dave Hansen. Signed-off-by:
Mel Gorman <mel@csn.ul.ie> Cc: Dave Hansen <dave@linux.vnet.ibm.com> Signed-off-by:
-
Lee Schermerhorn authored
Cleanup stale comments on munlock_vma_page(). Signed-off-by:
Lee Schermerhorn <lee.schermerhorn@hp.com> Acked-by:
Hugh Dickins <hugh.dickins@tiscali.co.uk> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by:
-
Lee Schermerhorn authored
unevictable_migrate_page() in mm/internal.h is a relic of the since removed UNEVICTABLE_LRU Kconfig option. This patch removes the function and open codes the test in migrate_page_copy(). Signed-off-by:
Christoph Lameter <cl@linux-foundation.org> Acked-by:
-
Mel Gorman authored
When the owner of a mapping fails COW because a child process is holding a reference, the children VMAs are walked and the page is unmapped. The i_mmap_lock is taken for the unmapping of the page but not the walking of the prio_tree. In theory, that tree could be changing if the lock is not held. This patch takes the i_mmap_lock properly for the duration of the prio_tree walk. [hugh.dickins@tiscali.co.uk: Spotted the problem in the first place] Signed-off-by:
-
Amerigo Wang authored
Jan Engelhardt reported we have this problem: setting max_map_count to a value large enough results in programs dying at first try. This is on 2.6.31.6: 15:59 borg:/proc/sys/vm # echo $[1<<31-1] >max_map_count 15:59 borg:/proc/sys/vm # cat max_map_count 1073741824 15:59 borg:/proc/sys/vm # echo $[1<<31] >max_map_count 15:59 borg:/proc/sys/vm # cat max_map_count Killed This is because we have a chance to make 'max_map_count' negative. but it's meaningless. Make it only accept non-negative values. Reported-by:
Jan Engelhardt <jengelh@medozas.de> Signed-off-by:
WANG Cong <amwang@redhat.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: James Morris <jmorris@namei.org> Cc: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
Huang Shijie authored
The check code for CONFIG_SWAP is redundant, because there is a non-CONFIG_SWAP version for PageSwapCache() which just returns 0. Signed-off-by:
Huang Shijie <shijie8@gmail.com> Signed-off-by:
-
Magnus Damm authored
Modify the generic mmap() code to keep the cache attribute in vma->vm_page_prot regardless if writenotify is enabled or not. Without this patch the cache configuration selected by f_op->mmap() is overwritten if writenotify is enabled, making it impossible to keep the vma uncached. Needed by drivers such as drivers/video/sh_mobile_lcdcfb.c which uses deferred io together with uncached memory. Signed-off-by:
Magnus Damm <damm@opensource.se> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Jaya Kumar <jayakumar.lkml@gmail.com> Cc: Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Huang Shijie authored
Simplify the code for shrink_inactive_list(). Signed-off-by:
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Reviewed-by:
Wu Fengguang <fengguang.wu@intel.com> Signed-off-by:
-
Rik van Riel authored
In AIM7 runs, recent kernels start swapping out anonymous pages well before they should. This is due to shrink_list falling through to shrink_inactive_list if !inactive_anon_is_low(zone, sc), when all we really wanted to do is pre-age some anonymous pages to give them extra time to be referenced while on the inactive list. The obvious fix is to make sure that shrink_list does not fall through to scanning/reclaiming inactive pages when we called it to scan one of the active lists. This change should be safe because the loop in shrink_zone ensures that we will still shrink the anon and file inactive lists whenever we should. [kosaki.motohiro@jp.fujitsu.com: inactive_file_is_low() should be inactive_anon_is_low()] Reported-by:
Larry Woodman <lwoodman@redhat.com> Signed-off-by:
Rik van Riel <riel@redhat.com> Acked-by:
Johannes Weiner <hannes@cmpxchg.org> Cc: Tomasz Chmielewski <mangoo@wpkg.org> Signed-off-by:
-
Jan Beulich authored
Signed-off-by:
Jan Beulich <jbeulich@novell.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by:
-
David Rientjes authored
Nodemasks should not be allocated on the stack for large systems (when it is larger than 256 bytes) since there is a threat of overflow. This patch causes the unregister_mem_sect_under_nodes() nodemask to be allocated on the stack for smaller systems and be allocated by slab for larger systems. GFP_KERNEL is used since remove_memory_block() can block. Cc: Gary Hade <garyhade@us.ibm.com> Cc: Badari Pulavarty <pbadari@us.ibm.com> Cc: Alex Chiang <achiang@hp.com> Signed-off-by:
David Rientjes <rientjes@google.com> Cc: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by:
-
KOSAKI Motohiro authored
SWAP_MLOCK mean "We marked the page as PG_MLOCK, please move it to unevictable-lru". So, following code is easy confusable. if (vma->vm_flags & VM_LOCKED) { ret = SWAP_MLOCK; goto out_unmap; } Plus, if the VMA doesn't have VM_LOCKED, We don't need to check the needed of calling mlock_vma_page(). Also, add some commentary to try_to_unmap_one(). Acked-by: -
Rakib Mullick authored
__free_pages_bootmem() is a __meminit function - which has been called from put_pages_bootmem thus causes a section mismatch warning. We were warned by the following warning: LD mm/built-in.o WARNING: mm/built-in.o(.text+0x26b22): Section mismatch in reference from the function put_page_bootmem() to the function .meminit.text:__free_pages_bootmem() The function put_page_bootmem() references the function __meminit __free_pages_bootmem(). This is often because put_page_bootmem lacks a __meminit annotation or the annotation of __free_pages_bootmem is wrong. Signed-off-by:
Rakib Mullick <rakib.mullick@gmail.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Badari Pulavarty <pbadari@us.ibm.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by:
-
Larry Woodman authored
hugetlb_fault() takes the mm->page_table_lock spinlock then calls hugetlb_cow(). If the alloc_huge_page() in hugetlb_cow() fails due to an insufficient huge page pool it calls unmap_ref_private() with the mm->page_table_lock held. unmap_ref_private() then calls unmap_hugepage_range() which tries to acquire the mm->page_table_lock. [<ffffffff810928c3>] print_circular_bug_tail+0x80/0x9f [<ffffffff8109280b>] ? check_noncircular+0xb0/0xe8 [<ffffffff810935e0>] __lock_acquire+0x956/0xc0e [<ffffffff81093986>] lock_acquire+0xee/0x12e [<ffffffff8111a7a6>] ? unmap_hugepage_range+0x3e/0x84 [<ffffffff8111a7a6>] ? unmap_hugepage_range+0x3e/0x84 [<ffffffff814c348d>] _spin_lock+0x40/0x89 [<ffffffff8111a7a6>] ? unmap_hugepage_range+0x3e/0x84 [<ffffffff8111afee>] ? alloc_huge_page+0x218/0x318 [<ffffffff8111a7a6>] unmap_hugepage_range+0x3e/0x84 [<ffffffff8111b2d0>] hugetlb_cow+0x1e2/0x3f4 [<ffffffff8111b935>] ? hugetlb_fault+0x453/0x4f6 [<ffffffff8111b962>] hugetlb_fault+0x480/0x4f6 [<ffffffff8111baee>] follow_hugetlb_page+0x116/0x2d9 [<ffffffff814c31a7>] ? _spin_unlock_irq+0x3a/0x5c [<ffffffff81107b4d>] __get_user_pages+0x2a3/0x427 [<ffffffff81107d0f>] get_user_pages+0x3e/0x54 [<ffffffff81040b8b>] get_user_pages_fast+0x170/0x1b5 [<ffffffff81160352>] dio_get_page+0x64/0x14a [<ffffffff8116112a>] __blockdev_direct_IO+0x4b7/0xb31 [<ffffffff8115ef91>] blkdev_direct_IO+0x58/0x6e [<ffffffff8115e0a4>] ? blkdev_get_blocks+0x0/0xb8 [<ffffffff810ed2c5>] generic_file_aio_read+0xdd/0x528 [<ffffffff81219da3>] ? avc_has_perm+0x66/0x8c [<ffffffff81132842>] do_sync_read+0xf5/0x146 [<ffffffff8107da00>] ? autoremove_wake_function+0x0/0x5a [<ffffffff81211857>] ? security_file_permission+0x24/0x3a [<ffffffff81132fd8>] vfs_read+0xb5/0x126 [<ffffffff81133f6b>] ? fget_light+0x5e/0xf8 [<ffffffff81133131>] sys_read+0x54/0x8c [<ffffffff81011e42>] system_call_fastpath+0x16/0x1b This can be fixed by dropping the mm->page_table_lock around the call to unmap_ref_private() if alloc_huge_page() fails, its dropped right below in the normal path anyway. However, earlier in the that function, it's also possible to call into the page allocator with the same spinlock held. What this patch does is drop the spinlock before the page allocator is potentially entered. The check for page allocation failure can be made without the page_table_lock as well as the copy of the huge page. Even if the PTE changed while the spinlock was held, the consequence is that a huge page is copied unnecessarily. This resolves both the double taking of the lock and sleeping with the spinlock held. [mel@csn.ul.ie: Cover also the case where process can sleep with spinlock] Signed-off-by:
Larry Woodman <lwooman@redhat.com> Signed-off-by:
Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@shadowen.org> Cc: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk> Cc: David Gibson <david@gibson.dropbear.id.au> Signed-off-by:
-
Andrew Morton authored
It has no references outside memory_hotplug.c. Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Andi Kleen <andi@firstfloor.org> Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Signed-off-by:
-
Hugh Dickins authored
Now that ksm pages are swappable, and the known holes plugged, remove mention of unswappable kernel pages from KSM documentation and comments. Remove the totalram_pages/4 initialization of max_kernel_pages. In fact, remove max_kernel_pages altogether - we can reinstate it if removal turns out to break someone's script; but if we later want to limit KSM's memory usage, limiting the stable nodes would not be an effective approach. Signed-off-by:
-
Hugh Dickins authored
The previous patch enables page migration of ksm pages, but that soon gets into trouble: not surprising, since we're using the ksm page lock to lock operations on its stable_node, but page migration switches the page whose lock is to be used for that. Another layer of locking would fix it, but do we need that yet? Do we actually need page migration of ksm pages? Yes, memory hotremove needs to offline sections of memory: and since we stopped allocating ksm pages with GFP_HIGHUSER, they will tend to be GFP_HIGHUSER_MOVABLE candidates for migration. But KSM is currently unconscious of NUMA issues, happily merging pages from different NUMA nodes: at present the rule must be, not to use MADV_MERGEABLE where you care about NUMA. So no, NUMA page migration of ksm pages does not make sense yet. So, to complete support for ksm swapping we need to make hotremove safe. ksm_memory_callback() take ksm_thread_mutex when MEM_GOING_OFFLINE and release it when MEM_OFFLINE or MEM_CANCEL_OFFLINE. But if mapped pages are freed before migration reaches them, stable_nodes may be left still pointing to struct pages which have been removed from the system: the stable_node needs to identify a page by pfn rather than page pointer, then it can safely prune them when MEM_OFFLINE. And make NUMA migration skip PageKsm pages where it skips PageReserved. But it's only when we reach unmap_and_move() that the page lock is taken and we can be sure that raised pagecount has prevented a PageAnon from being upgraded: so add offlining arg to migrate_pages(), to migrate ksm page when offlining (has sufficient locking) but reject it otherwise. Signed-off-by:
-
Hugh Dickins authored
A side-effect of making ksm pages swappable is that they have to be placed on the LRUs: which then exposes them to isolate_lru_page() and hence to page migration. Add rmap_walk() for remove_migration_ptes() to use: rmap_walk_anon() and rmap_walk_file() in rmap.c, but rmap_walk_ksm() in ksm.c. Perhaps some consolidation with existing code is possible, but don't attempt that yet (try_to_unmap needs to handle nonlinears, but migration pte removal does not). rmap_walk() is sadly less general than it appears: rmap_walk_anon(), like remove_anon_migration_ptes() which it replaces, avoids calling page_lock_anon_vma(), because that includes a page_mapped() test which fails when all migration ptes are in place. That was valid when NUMA page migration was introduced (holding mmap_sem provided the missing guarantee that anon_vma's slab had not already been destroyed), but I believe not valid in the memory hotremove case added since. For now do the same as before, and consider the best way to fix that unlikely race later on. When fixed, we can probably use rmap_walk() on hwpoisoned ksm pages too: for now, they remain among hwpoison's various exceptions (its PageKsm test comes before the page is locked, but its page_lock_anon_vma fails safely if an anon gets upgraded). Signed-off-by:
-
Hugh Dickins authored
But ksm swapping does require one small change in mem cgroup handling. When do_swap_page()'s call to ksm_might_need_to_copy() does indeed substitute a duplicate page to accommodate a different anon_vma (or a the !PageSwapCache check in mem_cgroup_try_charge_swapin(). That was returning success without charging, on the assumption that pte_same() would fail after, which is not the case here. Originally I proposed that success, so that an unshrinkable mem cgroup at its limit would not fail unnecessarily; but that's a minor point, and there are plenty of other places where we may fail an overallocation which might later prove unnecessary. So just go ahead and do what all the other exceptions do: proceed to charge current mm. Signed-off-by:
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by:
Balbir Singh <balbir@linux.vnet.ibm.com> Signed-off-by:
-
Hugh Dickins authored
When ksm pages were unswappable, it made no sense to include them in mem cgroup accounting; but now that they are swappable (although I see no strict logical connection) the principle of least surprise implies that they should be accounted (with the usual dissatisfaction, that a shared page is accounted to only one of the cgroups using it). This patch was intended to add mem cgroup accounting where necessary; but turned inside out, it now avoids allocating a ksm page, instead upgrading an anon page to ksm - which brings its existing mem cgroup accounting with it. Thus mem cgroups don't appear in the patch at all. This upgrade from PageAnon to PageKsm takes place under page lock (via a somewhat hacky NULL kpage interface), and audit showed only one place which needed to cope with the race - page_referenced() is sometimes used without page lock, so page_lock_anon_vma() needs an ACCESS_ONCE() to be sure of getting anon_vma and flags together (no problem if the page goes ksm an instant after, the integrity of that anon_vma list is unaffected). Signed-off-by:
-
Hugh Dickins authored
There's a lamentable flaw in KSM swapping: the stable_node holds a reference to the ksm page, so the page to be freed cannot actually be freed until ksmd works its way around to removing the last rmap_item from its stable_node. Which in some configurations may take minutes: not quite responsive enough for memory reclaim. And we don't want to twist KSM and its locking more tightly into the rest of mm. What a pity. But although the stable_node needs to hold a pointer to the ksm page, does it actually need to raise the reference count of that page? No. It would need to do so if struct pages were ordinary kmalloc'ed objects; but they are more stable than that, and reused in particular ways according to particular rules. Access to stable_node from its pointer in struct page is no problem, so long as we never free a stable_node before the ksm page itself has been freed. Access to struct page from its pointer in stable_node: reintroduce get_ksm_page(), and let that peep out through its keyhole (the stable_node pointer to ksm page), to see if that struct page still holds the right key to open it (the ksm page mapping pointer back to this stable_node). This relies upon the established way in which free_hot_cold_page() sets an anon (including ksm) page->mapping to NULL; and relies upon no other user of a struct page to put something which looks like the original stable_node pointer (with two low bits also set) into page->mapping. It also needs get_page_unless_zero() technique pioneered by speculative pagecache; and uses rcu_read_lock() to keep the guarantees that gives. There are several drivers which put pointers of their own into page-> mapping; but none of those could coincide with our stable_node pointers, since KSM won't free a stable_node until it sees that the page has gone. The only problem case found is the pagetable spinlock USE_SPLIT_PTLOCKS places in struct page (my own abuse): to accommodate GENERIC_LOCKBREAK's break_lock on 32-bit, that spans both page->private and page->mapping. Since break_lock is only 0 or 1, again no confusion for get_ksm_page(). But what of DEBUG_SPINLOCK on 64-bit bigendian? When owner_cpu is 3 (matching PageKsm low bits), it might see 0xdead4ead00000003 in page-> mapping, which might coincide? We could get around that by... but a better answer is to suppress USE_SPLIT_PTLOCKS when DEBUG_SPINLOCK or DEBUG_LOCK_ALLOC, to stop bloating sizeof(struct page) in their case - already proposed in an earlier mm/Kconfig patch. Signed-off-by:
-
Hugh Dickins authored
For full functionality, page_referenced_one() and try_to_unmap_one() need to know the vma: to pass vma down to arch-dependent flushes, or to observe VM_LOCKED or VM_EXEC. But KSM keeps no record of vma: nor can it, since vmas get split and merged without its knowledge. Instead, note page's anon_vma in its rmap_item when adding to stable tree: all the vmas which might map that page are listed by its anon_vma. page_referenced_ksm() and try_to_unmap_ksm() then traverse the anon_vma, first to find the probable vma, that which matches rmap_item's mm; but if that is not enough to locate all instances, traverse again to try the others. This catches those occasions when fork has duplicated a pte of a ksm page, but ksmd has not yet come around to assign it an rmap_item. But each rmap_item in the stable tree which refers to an anon_vma needs to take a reference to it. Andrea's anon_vma design cleverly avoided a reference count (an anon_vma was free when its list of vmas was empty), but KSM now needs to add that. Is a 32-bit count sufficient? I believe so - the anon_vma is only free when both count is 0 and list is empty. Signed-off-by:
-
Hugh Dickins authored
Initial implementation for swapping out KSM's shared pages: add page_referenced_ksm() and try_to_unmap_ksm(), which rmap.c calls when faced with a PageKsm page. Most of what's needed can be got from the rmap_items listed from the stable_node of the ksm page, without discovering the actual vma: so in this patch just fake up a struct vma for page_referenced_one() or try_to_unmap_one(), then refine that in the next patch. Add VM_NONLINEAR to ksm_madvise()'s list of exclusions: it has always been implicit there (being only set with VM_SHARED, already excluded), but let's make it explicit, to help justify the lack of nonlinear unmap. Rely on the page lock to protect against concurrent modifications to that page's node of the stable tree. The awkward part is not swapout but swapin: do_swap_page() and page_add_anon_rmap() now have to allow for new possibilities - perhaps a ksm page still in swapcache, perhaps a swapcache page associated with one location in one anon_vma now needed for another location or anon_vma. (And the vma might even be no longer VM_MERGEABLE when that happens.) ksm_might_need_to_copy() checks for that case, and supplies a duplicate page when necessary, simply leaving it to a subsequent pass of ksmd to rediscover the identity and merge them back into one ksm page. Disappointingly primitive: but the alternative would have to accumulate unswappable info about the swapped out ksm pages, limiting swappability. Remove page_add_ksm_rmap(): page_add_anon_rmap() now has to allow for the particular case it was handling, so just use it instead. Signed-off-by:
-
Hugh Dickins authored
When KSM merges an mlocked page, it has been forgetting to munlock it: that's been left to free_page_mlock(), which reports it in /proc/vmstat as unevictable_pgs_mlockfreed instead of unevictable_pgs_munlocked (and whinges "Page flag mlocked set for process" in mmotm, whereas mainline is silently forgiving). Call munlock_vma_page() to fix that. Signed-off-by:
-
Hugh Dickins authored
Add a pointer to the ksm page into struct stable_node, holding a reference to the page while the node exists. Put a pointer to the stable_node into the ksm page's ->mapping. Then we don't need get_ksm_page() while traversing the stable tree: the page to compare against is sure to be present and correct, even if it's no longer visible through any of its existing rmap_items. And we can handle the forked ksm page case more efficiently: no need to memcmp our way through the tree to find its match. Signed-off-by:
-
Hugh Dickins authored
Though we still do well to keep rmap_items in the unstable tree without a separate tree_item at the node, for several reasons it becomes awkward to keep rmap_items in the stable tree without a separate stable_node: lack of space in the nicely-sized rmap_item, the need for an anchor as rmap_items are removed, the need for a node even when temporarily no rmap_items are attached to it. So declare struct stable_node (rb_node to place it in the tree and hlist_head for the rmap_items hanging off it), and convert stable tree handling to use it: without yet taking advantage of it. Note how one stable_tree_insert() of a node now has _two_ stable_tree_append()s of the two rmap_items being merged. Signed-off-by:
-
Hugh Dickins authored
Free up a pointer in struct rmap_item, by making the mm_slot's rmap_list a singly-linked list: we always traverse that list sequentially, and we don't even lose any prefetches (but should consider adding a few later). Name it rmap_list throughout. Do we need to free up that pointer? Not immediately, and in the end, we could continue to avoid it with a union; but having done the conversion, let's keep it this way, since there's no downside, and maybe we'll want more in future (struct rmap_item is a cache-friendly 32 bytes on 32-bit and 64 bytes on 64-bit, so we shall want to avoid expanding it). Signed-off-by:
-
Hugh Dickins authored
Cleanup: make argument names more consistent from cmp_and_merge_page() down to replace_page(), so that it's easier to follow the rmap_item's page and the matching tree_page and the merged kpage through that code. In some places, e.g. break_cow(), pass rmap_item instead of separate mm and address. cmp_and_merge_page() initialize tree_page to NULL, to avoid a "may be used uninitialized" warning seen in one config by Anil SB. Signed-off-by:
-
Hugh Dickins authored
There is no need for replace_page() to calculate a write-protected prot vm_page_prot must already be write-protected for an anonymous page (see mm/memory.c do_anonymous_page() for similar reliance on vm_page_prot). There is no need for try_to_merge_one_page() to get_page and put_page on newpage and oldpage: in every case we already hold a reference to each of them. But some instinct makes me move try_to_merge_one_page()'s unlock_page of oldpage down after replace_page(): that doesn't increase contention on the ksm page, and makes thinking about the transition easier. Signed-off-by:
-
Hugh Dickins authored
1. remove_rmap_item_from_tree() is called as a precaution from various places: don't dirty the rmap_item cacheline unnecessarily, just mask the flags out of the address when they have been set. 2. First get_next_rmap_item() removes an unstable rmap_item from its tree, then shortly afterwards cmp_and_merge_page() removes a stable rmap_item from its tree: it's easier just to do both at once (but definitely keep the BUG_ON(age > 1) which guards against a future omission). 3. When cmp_and_merge_page() moves an rmap_item from unstable to stable tree, it does its own rb_erase() and accounting: that's better expressed by remove_rmap_item_from_tree(). Signed-off-by:
-
KOSAKI Motohiro authored
Fix small inconsistent of ">" and ">=". Signed-off-by:
Minchan Kim <minchan.kim@gmail.com> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by:
-
KOSAKI Motohiro authored
Now, All caller of reclaim use swap_cluster_max as SWAP_CLUSTER_MAX. Then, we can remove it perfectly. Signed-off-by:
-
KOSAKI Motohiro authored
In old days, we didn't have sc.nr_to_reclaim and it brought sc.swap_cluster_max misuse. huge sc.swap_cluster_max might makes unnecessary OOM risk and no performance benefit. Now, we can stop its insane thing. Signed-off-by:
-
KOSAKI Motohiro authored
shrink_all_zone() was introduced by commit d6277db4 (swsusp: rework memory shrinker) for hibernate performance improvement. and sc.swap_cluster_max was introduced by commit a06fe4d307 (Speed freeing memory for suspend). commit a06fe4d307 said Without the patch: Freed 14600 pages in 1749 jiffies = 32.61 MB/s (Anomolous!) Freed 88563 pages in 14719 jiffies = 23.50 MB/s Freed 205734 pages in 32389 jiffies = 24.81 MB/s With the patch: Freed 68252 pages in 496 jiffies = 537.52 MB/s Freed 116464 pages in 569 jiffies = 798.54 MB/s Freed 209699 pages in 705 jiffies = 1161.89 MB/s At that time, their patch was pretty worth. However, Modern Hardware trend and recent VM improvement broke its worth. From several reason, I think we should remove shrink_all_zones() at all. detail: 1) Old days, shrink_zone()'s slowness was mainly caused by stupid io-throttle at no i/o congestion. but current shrink_zone() is sane, not slow. 2) shrink_all_zone() try to shrink all pages at a time. but it doesn't works fine on numa system. example) System has 4GB memory and each node have 2GB. and hibernate need 1GB. optimal) steal 500MB from each node. shrink_all_zones) steal 1GB from node-0. Oh, Cache balancing logic was broken. ;) Unfortunately, Desktop system moved ahead NUMA at nowadays. (Side note, if hibernate require 2GB, shrink_all_zones() never success on above machine) 3) if the node has several I/O flighting pages, shrink_all_zones() makes pretty bad result. schenario) hibernate need 1GB 1) shrink_all_zones() try to reclaim 1GB from Node-0 2) but it only reclaimed 990MB 3) stupidly, shrink_all_zones() try to reclaim 1GB from Node-1 4) it reclaimed 990MB Oh, well. it reclaimed twice much than required. In the other hand, current shrink_zone() has sane baling out logic. then, it doesn't make overkill reclaim. then, we lost shrink_zones()'s risk. 4) SplitLRU VM always keep active/inactive ratio very carefully. inactive list only shrinking break its assumption. it makes unnecessary OOM risk. it obviously suboptimal. Now, shrink_all_memory() is only the wrapper function of do_try_to_free_pages(). it bring good reviewability and debuggability, and solve above problems. side note: Reclaim logic unificication makes two good side effect. - Fix recursive reclaim bug on shrink_all_memory(). it did forgot to use PF_MEMALLOC. it mean the system be able to stuck into deadlock. - Now, shrink_all_memory() got lockdep awareness. it bring good debuggability. Signed-off-by:
Rafael J. Wysocki <rjw@sisk.pl> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by:
-
KOSAKI Motohiro authored
Currently, sc.scap_cluster_max has double meanings. 1) reclaim batch size as isolate_lru_pages()'s argument 2) reclaim baling out thresolds The two meanings pretty unrelated. Thus, Let's separate it. this patch doesn't change any behavior. Signed-off-by:
-
Alex Chiang authored
Describe NUMA node symlink created for CPUs when CONFIG_NUMA is set. Signed-off-by:
Alex Chiang <achiang@hp.com> Cc: Greg KH <greg@kroah.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Gary Hade <garyhade@us.ibm.com> Cc: Badari Pulavarty <pbadari@us.ibm.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: David Rientjes <rientjes@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by:
-
