04 Apr, 2014
1 commit
-
commit 668f9abbd4334e6c29fa8acd71635c4f9101caa7 upstream.
Commit bf6bddf1924e ("mm: introduce compaction and migration for
ballooned pages") introduces page_count(page) into memory compaction
which dereferences page->first_page if PageTail(page).This results in a very rare NULL pointer dereference on the
aforementioned page_count(page). Indeed, anything that does
compound_head(), including page_count() is susceptible to racing with
prep_compound_page() and seeing a NULL or dangling page->first_page
pointer.This patch uses Andrea's implementation of compound_trans_head() that
deals with such a race and makes it the default compound_head()
implementation. This includes a read memory barrier that ensures that
if PageTail(head) is true that we return a head page that is neither
NULL nor dangling. The patch then adds a store memory barrier to
prep_compound_page() to ensure page->first_page is set.This is the safest way to ensure we see the head page that we are
expecting, PageTail(page) is already in the unlikely() path and the
memory barriers are unfortunately required.Hugetlbfs is the exception, we don't enforce a store memory barrier
during init since no race is possible.Signed-off-by: David Rientjes
Cc: Holger Kiehl
Cc: Christoph Lameter
Cc: Rafael Aquini
Cc: Vlastimil Babka
Cc: Michal Hocko
Cc: Mel Gorman
Cc: Andrea Arcangeli
Cc: Rik van Riel
Cc: "Kirill A. Shutemov"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
Signed-off-by: Greg Kroah-Hartman
07 Feb, 2014
1 commit
-
commit 27c73ae759774e63313c1fbfeb17ba076cea64c5 upstream.
Commit 7cb2ef56e6a8 ("mm: fix aio performance regression for database
caused by THP") can cause dereference of a dangling pointer if
split_huge_page runs during PageHuge() if there are updates to the
tail_page->private field.Also it is repeating compound_head twice for hugetlbfs and it is running
compound_head+compound_trans_head for THP when a single one is needed in
both cases.The new code within the PageSlab() check doesn't need to verify that the
THP page size is never bigger than the smallest hugetlbfs page size, to
avoid memory corruption.A longstanding theoretical race condition was found while fixing the
above (see the change right after the skip_unlock label, that is
relevant for the compound_lock path too).By re-establishing the _mapcount tail refcounting for all compound
pages, this also fixes the below problem:echo 0 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
BUG: Bad page state in process bash pfn:59a01
page:ffffea000139b038 count:0 mapcount:10 mapping: (null) index:0x0
page flags: 0x1c00000000008000(tail)
Modules linked in:
CPU: 6 PID: 2018 Comm: bash Not tainted 3.12.0+ #25
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
Call Trace:
dump_stack+0x55/0x76
bad_page+0xd5/0x130
free_pages_prepare+0x213/0x280
__free_pages+0x36/0x80
update_and_free_page+0xc1/0xd0
free_pool_huge_page+0xc2/0xe0
set_max_huge_pages.part.58+0x14c/0x220
nr_hugepages_store_common.isra.60+0xd0/0xf0
nr_hugepages_store+0x13/0x20
kobj_attr_store+0xf/0x20
sysfs_write_file+0x189/0x1e0
vfs_write+0xc5/0x1f0
SyS_write+0x55/0xb0
system_call_fastpath+0x16/0x1bSigned-off-by: Khalid Aziz
Signed-off-by: Andrea Arcangeli
Tested-by: Khalid Aziz
Cc: Pravin Shelar
Cc: Greg Kroah-Hartman
Cc: Ben Hutchings
Cc: Christoph Lameter
Cc: Johannes Weiner
Cc: Mel Gorman
Cc: Rik van Riel
Cc: Andi Kleen
Cc: Minchan Kim
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
Cc: Guillaume Morin
Signed-off-by: Greg Kroah-Hartman
02 Oct, 2013
1 commit
-
commit 7cb2ef56e6a8b7b368b2e883a0a47d02fed66911 upstream.
I am working with a tool that simulates oracle database I/O workload.
This tool (orion to be specific -
)
allocates hugetlbfs pages using shmget() with SHM_HUGETLB flag. It then
does aio into these pages from flash disks using various common block
sizes used by database. I am looking at performance with two of the most
common block sizes - 1M and 64K. aio performance with these two block
sizes plunged after Transparent HugePages was introduced in the kernel.
Here are performance numbers:pre-THP 2.6.39 3.11-rc5
1M read 8384 MB/s 5629 MB/s 6501 MB/s
64K read 7867 MB/s 4576 MB/s 4251 MB/sI have narrowed the performance impact down to the overheads introduced by
THP in __get_page_tail() and put_compound_page() routines. perf top shows
>40% of cycles being spent in these two routines. Every time direct I/O
to hugetlbfs pages starts, kernel calls get_page() to grab a reference to
the pages and calls put_page() when I/O completes to put the reference
away. THP introduced significant amount of locking overhead to get_page()
and put_page() when dealing with compound pages because hugepages can be
split underneath get_page() and put_page(). It added this overhead
irrespective of whether it is dealing with hugetlbfs pages or transparent
hugepages. This resulted in 20%-45% drop in aio performance when using
hugetlbfs pages.Since hugetlbfs pages can not be split, there is no reason to go through
all the locking overhead for these pages from what I can see. I added
code to __get_page_tail() and put_compound_page() to bypass all the
locking code when working with hugetlbfs pages. This improved performance
significantly. Performance numbers with this patch:pre-THP 3.11-rc5 3.11-rc5 + Patch
1M read 8384 MB/s 6501 MB/s 8371 MB/s
64K read 7867 MB/s 4251 MB/s 6510 MB/sPerformance with 64K read is still lower than what it was before THP, but
still a 53% improvement. It does mean there is more work to be done but I
will take a 53% improvement for now.Please take a look at the following patch and let me know if it looks
reasonable.[akpm@linux-foundation.org: tweak comments]
Signed-off-by: Khalid Aziz
Cc: Pravin B Shelar
Cc: Christoph Lameter
Cc: Andrea Arcangeli
Cc: Johannes Weiner
Cc: Mel Gorman
Cc: Rik van Riel
Cc: Minchan Kim
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
Signed-off-by: Greg Kroah-Hartman
08 May, 2013
1 commit
-
Faster kernel compiles by way of fewer unnecessary includes.
[akpm@linux-foundation.org: fix fallout]
[akpm@linux-foundation.org: fix build]
Signed-off-by: Kent Overstreet
Cc: Zach Brown
Cc: Felipe Balbi
Cc: Greg Kroah-Hartman
Cc: Mark Fasheh
Cc: Joel Becker
Cc: Rusty Russell
Cc: Jens Axboe
Cc: Asai Thambi S P
Cc: Selvan Mani
Cc: Sam Bradshaw
Cc: Jeff Moyer
Cc: Al Viro
Cc: Benjamin LaHaise
Reviewed-by: "Theodore Ts'o"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
30 Apr, 2013
1 commit
-
In page reclaim, huge page is split. split_huge_page() adds tail pages
to LRU list. Since we are reclaiming a huge page, it's better we
reclaim all subpages of the huge page instead of just the head page.
This patch adds split tail pages to shrink page list so the tail pages
can be reclaimed soon.Before this patch, run a swap workload:
thp_fault_alloc 3492
thp_fault_fallback 608
thp_collapse_alloc 6
thp_collapse_alloc_failed 0
thp_split 916With this patch:
thp_fault_alloc 4085
thp_fault_fallback 16
thp_collapse_alloc 90
thp_collapse_alloc_failed 0
thp_split 1272fallback allocation is reduced a lot.
[akpm@linux-foundation.org: fix CONFIG_SWAP=n build]
Signed-off-by: Shaohua Li
Acked-by: Rik van Riel
Acked-by: Minchan Kim
Acked-by: Johannes Weiner
Reviewed-by: Wanpeng Li
Cc: Andrea Arcangeli
Cc: Hugh Dickins
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
24 Feb, 2013
1 commit
-
When I use several fast SSD to do swap, swapper_space.tree_lock is
heavily contended. This makes each swap partition have one
address_space to reduce the lock contention. There is an array of
address_space for swap. The swap entry type is the index to the array.In my test with 3 SSD, this increases the swapout throughput 20%.
[akpm@linux-foundation.org: revert unneeded change to __add_to_swap_cache]
Signed-off-by: Shaohua Li
Cc: Hugh Dickins
Acked-by: Rik van Riel
Acked-by: Minchan Kim
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
09 Oct, 2012
2 commits
-
page_evictable(page, vma) is an irritant: almost all its callers pass
NULL for vma. Remove the vma arg and use mlocked_vma_newpage(vma, page)
explicitly in the couple of places it's needed. But in those places we
don't even need page_evictable() itself! They're dealing with a freshly
allocated anonymous page, which has no "mapping" and cannot be mlocked yet.Signed-off-by: Hugh Dickins
Acked-by: Mel Gorman
Cc: Rik van Riel
Acked-by: Johannes Weiner
Cc: Michel Lespinasse
Cc: Ying Han
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
When writing a new file with 2048 bytes buffer, such as write(fd, buffer,
2048), it will call generic_perform_write() twice for every page:write_begin
mark_page_accessed(page)
write_endwrite_begin
mark_page_accessed(page)
write_endPages 1-13 will be added to lru-pvecs in write_begin() and will *NOT* be
added to active_list even they have be accessed twice because they are not
PageLRU(page). But when page 14th comes, all pages in lru-pvecs will be
moved to inactive_list (by __lru_cache_add() ) in first write_begin(), now
page 14th *is* PageLRU(page). And after second write_end() only page 14th
will be in active_list.In Hadoop environment, we do comes to this situation: after writing a
file, we find out that only 14th, 28th, 42th... page are in active_list
and others in inactive_list. Now kswapd works, shrinks the inactive_list,
the file only have 14th, 28th...pages in memory, the readahead request
size will be broken to only 52k (13*4k), system's performance falls
dramatically.This problem can also replay by below steps (the machine has 8G memory):
1. dd if=/dev/zero of=/test/file.out bs=1024 count=1048576
2. cat another 7.5G file to /dev/null
3. vmtouch -m 1G -v /test/file.out, it will show:/test/file.out
[oooooooooooooooooooOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO] 187847/262144the 'o' means same pages are in memory but same are not.
The solution for this problem is simple: the 14th page should be added to
lru_add_pvecs before mark_page_accessed() just as other pages.[akpm@linux-foundation.org: tweak comment]
[akpm@linux-foundation.org: grab better comment from the v3 patch]
Signed-off-by: Robin Dong
Reviewed-by: Minchan Kim
Cc: KOSAKI Motohiro
Reviewed-by: Johannes Weiner
Cc: Wu Fengguang
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
01 Aug, 2012
2 commits
-
The patch "mm: add support for a filesystem to activate swap files and use
direct_IO for writing swap pages" added support for using direct_IO to
write swap pages but it is insufficient for highmem pages.To support highmem pages, this patch kmaps() the page before calling the
direct_IO() handler. As direct_IO deals with virtual addresses an
additional helper is necessary for get_kernel_pages() to lookup the struct
page for a kmap virtual address.Signed-off-by: Mel Gorman
Acked-by: Rik van Riel
Cc: Christoph Hellwig
Cc: David S. Miller
Cc: Eric B Munson
Cc: Eric Paris
Cc: James Morris
Cc: Mel Gorman
Cc: Mike Christie
Cc: Neil Brown
Cc: Peter Zijlstra
Cc: Sebastian Andrzej Siewior
Cc: Trond Myklebust
Cc: Xiaotian Feng
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch adds two new APIs get_kernel_pages() and get_kernel_page() that
may be used to pin a vector of kernel addresses for IO. The initial user
is expected to be NFS for allowing pages to be written to swap using
aops->direct_IO(). Strictly speaking, swap-over-NFS only needs to pin one
page for IO but it makes sense to express the API in terms of a vector and
add a helper for pinning single pages.Signed-off-by: Mel Gorman
Reviewed-by: Rik van Riel
Cc: Christoph Hellwig
Cc: David S. Miller
Cc: Eric B Munson
Cc: Eric Paris
Cc: James Morris
Cc: Mel Gorman
Cc: Mike Christie
Cc: Neil Brown
Cc: Peter Zijlstra
Cc: Sebastian Andrzej Siewior
Cc: Trond Myklebust
Cc: Xiaotian Feng
Cc: Mark Salter
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
30 May, 2012
3 commits
-
Take lruvec further: pass it instead of zone to add_page_to_lru_list() and
del_page_from_lru_list(); and pagevec_lru_move_fn() pass lruvec down to
its target functions.This cleanup eliminates a swathe of cruft in memcontrol.c, including
mem_cgroup_lru_add_list(), mem_cgroup_lru_del_list() and
mem_cgroup_lru_move_lists() - which never actually touched the lists.In their place, mem_cgroup_page_lruvec() to decide the lruvec, previously
a side-effect of add, and mem_cgroup_update_lru_size() to maintain the
lru_size stats.Whilst these are simplifications in their own right, the goal is to bring
the evaluation of lruvec next to the spin_locking of the lrus, in
preparation for a future patch.Signed-off-by: Hugh Dickins
Cc: KOSAKI Motohiro
Acked-by: KAMEZAWA Hiroyuki
Acked-by: Michal Hocko
Acked-by: Konstantin Khlebnikov
Cc: Johannes Weiner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
With mem_cgroup_disabled() now explicit, it becomes clear that the
zone_reclaim_stat structure actually belongs in lruvec, per-zone when
memcg is disabled but per-memcg per-zone when it's enabled.We can delete mem_cgroup_get_reclaim_stat(), and change
update_page_reclaim_stat() to update just the one set of stats, the one
which get_scan_count() will actually use.Signed-off-by: Hugh Dickins
Signed-off-by: Konstantin Khlebnikov
Acked-by: KAMEZAWA Hiroyuki
Acked-by: Michal Hocko
Reviewed-by: Minchan Kim
Reviewed-by: Michal Hocko
Cc: Glauber Costa
Cc: Johannes Weiner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Transparent huge pages can change page->flags (PG_compound_lock) without
taking Slab lock. Since THP can not break slab pages we can safely access
compound page without taking compound lock.Specifically this patch fixes a race between compound_unlock() and slab
functions which perform page-flags updates. This can occur when
get_page()/put_page() is called on a page from slab.[akpm@linux-foundation.org: tweak comment text, fix comment layout, fix label indenting]
Reported-by: Amey Bhide
Signed-off-by: Pravin B Shelar
Reviewed-by: Christoph Lameter
Acked-by: Andrea Arcangeli
Cc: Pekka Enberg
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
22 Mar, 2012
1 commit
-
This cpu hotplug hook was accidentally removed in commit 00a62ce91e55
("mm: fix Committed_AS underflow on large NR_CPUS environment")The visible effect of this accident: some pages are borrowed in per-cpu
page-vectors. Truncate can deal with it, but these pages cannot be
reused while this cpu is offline. So this is like a temporary memory
leak.Signed-off-by: Konstantin Khlebnikov
Cc: Dave Hansen
Cc: KOSAKI Motohiro
Cc: Eric B Munson
Cc: Mel Gorman
Cc: Christoph Lameter
Cc: KAMEZAWA Hiroyuki
Cc: Johannes Weiner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
06 Mar, 2012
1 commit
-
When moving tasks from old memcg (with move_charge_at_immigrate on new
memcg), followed by removal of old memcg, hit General Protection Fault in
mem_cgroup_lru_del_list() (called from release_pages called from
free_pages_and_swap_cache from tlb_flush_mmu from tlb_finish_mmu from
exit_mmap from mmput from exit_mm from do_exit).Somewhat reproducible, takes a few hours: the old struct mem_cgroup has
been freed and poisoned by SLAB_DEBUG, but mem_cgroup_lru_del_list() is
still trying to update its stats, and take page off lru before freeing.A task, or a charge, or a page on lru: each secures a memcg against
removal. In this case, the last task has been moved out of the old memcg,
and it is exiting: anonymous pages are uncharged one by one from the
memcg, as they are zapped from its pagetables, so the charge gets down to
0; but the pages themselves are queued in an mmu_gather for freeing.Most of those pages will be on lru (and force_empty is careful to
lru_add_drain_all, to add pages from pagevec to lru first), but not
necessarily all: perhaps some have been isolated for page reclaim, perhaps
some isolated for other reasons. So, force_empty may find no task, no
charge and no page on lru, and let the removal proceed.There would still be no problem if these pages were immediately freed; but
typically (and the put_page_testzero protocol demands it) they have to be
added back to lru before they are found freeable, then removed from lru
and freed. We don't see the issue when adding, because the
mem_cgroup_iter() loops keep their own reference to the memcg being
scanned; but when it comes to mem_cgroup_lru_del_list().I believe this was not an issue in v3.2: there, PageCgroupAcctLRU and
PageCgroupUsed flags were used (like a trick with mirrors) to deflect view
of pc->mem_cgroup to the stable root_mem_cgroup when neither set.
38c5d72f3ebe ("memcg: simplify LRU handling by new rule") mercifully
removed those convolutions, but left this General Protection Fault.But it's surprisingly easy to restore the old behaviour: just check
PageCgroupUsed in mem_cgroup_lru_add_list() (which decides on which lruvec
to add), and reset pc to root_mem_cgroup if page is uncharged. A risky
change? just going back to how it worked before; testing, and an audit of
uses of pc->mem_cgroup, show no problem.And there's a nice bonus: with mem_cgroup_lru_add_list() itself making
sure that an uncharged page goes to root lru, mem_cgroup_reset_owner() no
longer has any purpose, and we can safely revert 4e5f01c2b9b9 ("memcg:
clear pc->mem_cgroup if necessary").Calling update_page_reclaim_stat() after add_page_to_lru_list() in swap.c
is not strictly necessary: the lru_lock there, with RCU before memcg
structures are freed, makes mem_cgroup_get_reclaim_stat_from_page safe
without that; but it seems cleaner to rely on one dependency less.Signed-off-by: Hugh Dickins
Cc: KAMEZAWA Hiroyuki
Cc: Johannes Weiner
Cc: Konstantin Khlebnikov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
09 Feb, 2012
1 commit
-
Fix CONFIG_TRANSPARENT_HUGEPAGE=y CONFIG_SMP=n CONFIG_DEBUG_VM=y
CONFIG_DEBUG_SPINLOCK=n kernel: spin_is_locked() is then always false,
and so triggers some BUGs in Transparent HugePage codepaths.asm-generic/bug.h mentions this problem, and provides a WARN_ON_SMP(x);
but being too lazy to add VM_BUG_ON_SMP, BUG_ON_SMP, WARN_ON_SMP_ONCE,
VM_WARN_ON_SMP_ONCE, just test NR_CPUS != 1 in the existing VM_BUG_ONs.Signed-off-by: Hugh Dickins
Cc: Andrea Arcangeli
Cc:
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
13 Jan, 2012
8 commits
-
del_page_from_lru() repeats del_page_from_lru_list(), also working out
which LRU the page was on, clearing the relevant bits. Decouple those
functions: remove del_page_from_lru() and add page_off_lru().Signed-off-by: Hugh Dickins
Reviewed-by: KOSAKI Motohiro
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
checkpatch rightly protests
WARNING: EXPORT_SYMBOL(foo); should immediately follow its function/variable
so fix the five offenders in mm/swap.c.
Signed-off-by: Hugh Dickins
Reviewed-by: KOSAKI Motohiro
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
What's so special about ____pagevec_lru_add() that it needs four leading
underscores? Nothing, it just helped to distinguish from
__pagevec_lru_add() in 2.6.28 development. Cut two leading underscores.Signed-off-by: Hugh Dickins
Reviewed-by: KOSAKI Motohiro
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Replace pagevecs in putback_lru_pages() and move_active_pages_to_lru()
by lists of pages_to_free: then apply Konstantin Khlebnikov's
free_hot_cold_page_list() to them instead of pagevec_release().Which simplifies the flow (no need to drop and retake lock whenever
pagevec fills up) and reduces stale addresses in stack backtraces
(which often showed through the pagevecs); but more importantly,
removes another 120 bytes from the deepest stacks in page reclaim.
Although I've not recently seen an actual stack overflow here with
a vanilla kernel, move_active_pages_to_lru() has often featured in
deep backtraces.However, free_hot_cold_page_list() does not handle compound pages
(nor need it: a Transparent HugePage would have been split by the
time it reaches the call in shrink_page_list()), but it is possible
for putback_lru_pages() or move_active_pages_to_lru() to be left
holding the last reference on a THP, so must exclude the unlikely
compound case before putting on pages_to_free.Remove pagevec_strip(), its work now done in move_active_pages_to_lru().
The pagevec in scan_mapping_unevictable_pages() remains in mm/vmscan.c,
but that is never on the reclaim path, and cannot be replaced by a list.Signed-off-by: Hugh Dickins
Reviewed-by: KOSAKI Motohiro
Reviewed-by: Konstantin Khlebnikov
Cc: KAMEZAWA Hiroyuki
Cc: Mel Gorman
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch started off as a cleanup: __split_huge_page_refcounts() has to
cope with two scenarios, when the hugepage being split is already on LRU,
and when it is not; but why does it have to split that accounting across
three different sites? Consolidate it in lru_add_page_tail(), handling
evictable and unevictable alike, and use standard add_page_to_lru_list()
when accounting is needed (when the head is not yet on LRU).But a recent regression in -next, I guess the removal of PageCgroupAcctLRU
test from mem_cgroup_split_huge_fixup(), makes this now a necessary fix:
under load, the MEM_CGROUP_ZSTAT count was wrapping to a huge number,
messing up reclaim calculations and causing a freeze at rmdir of cgroup.Add a VM_BUG_ON to mem_cgroup_lru_del_list() when we're about to wrap that
count - this has not been the only such incident. Document that
lru_add_page_tail() is for Transparent HugePages by #ifdef around it.Signed-off-by: Hugh Dickins
Cc: Daisuke Nishimura
Cc: KAMEZAWA Hiroyuki
Cc: Johannes Weiner
Cc: Michal Hocko
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Put the tail subpages of an isolated hugepage under splitting in the lru
reclaim head as they supposedly should be isolated too next.Queues the subpages in physical order in the lru for non isolated
hugepages under splitting. That might provide some theoretical cache
benefit to the buddy allocator later.Signed-off-by: Shaohua Li
Signed-off-by: Andrea Arcangeli
Cc: David Rientjes
Cc: Johannes Weiner
Cc: Minchan Kim
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Now that all code that operated on global per-zone LRU lists is
converted to operate on per-memory cgroup LRU lists instead, there is no
reason to keep the double-LRU scheme around any longer.The pc->lru member is removed and page->lru is linked directly to the
per-memory cgroup LRU lists, which removes two pointers from a
descriptor that exists for every page frame in the system.Signed-off-by: Johannes Weiner
Signed-off-by: Hugh Dickins
Signed-off-by: Ying Han
Reviewed-by: KAMEZAWA Hiroyuki
Reviewed-by: Michal Hocko
Reviewed-by: Kirill A. Shutemov
Cc: Daisuke Nishimura
Cc: Balbir Singh
Cc: Greg Thelen
Cc: Michel Lespinasse
Cc: Rik van Riel
Cc: Minchan Kim
Cc: Christoph Hellwig
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Having a unified structure with a LRU list set for both global zones and
per-memcg zones allows to keep that code simple which deals with LRU
lists and does not care about the container itself.Once the per-memcg LRU lists directly link struct pages, the isolation
function and all other list manipulations are shared between the memcg
case and the global LRU case.Signed-off-by: Johannes Weiner
Reviewed-by: KAMEZAWA Hiroyuki
Reviewed-by: Michal Hocko
Reviewed-by: Kirill A. Shutemov
Cc: Daisuke Nishimura
Cc: Balbir Singh
Cc: Ying Han
Cc: Greg Thelen
Cc: Michel Lespinasse
Cc: Rik van Riel
Cc: Minchan Kim
Cc: Christoph Hellwig
Cc: Hugh Dickins
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
11 Jan, 2012
1 commit
-
This patch adds helper free_hot_cold_page_list() to free list of 0-order
pages. It frees pages directly from list without temporary page-vector.
It also calls trace_mm_pagevec_free() to simulate pagevec_free()
behaviour.bloat-o-meter:
add/remove: 1/1 grow/shrink: 1/3 up/down: 267/-295 (-28)
function old new delta
free_hot_cold_page_list - 264 +264
get_page_from_freelist 2129 2132 +3
__pagevec_free 243 239 -4
split_free_page 380 373 -7
release_pages 606 510 -96
free_page_list 188 - -188Signed-off-by: Konstantin Khlebnikov
Cc: Mel Gorman
Cc: KOSAKI Motohiro
Acked-by: Minchan Kim
Acked-by: Hugh Dickins
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
07 Nov, 2011
1 commit
-
* 'modsplit-Oct31_2011' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux: (230 commits)
Revert "tracing: Include module.h in define_trace.h"
irq: don't put module.h into irq.h for tracking irqgen modules.
bluetooth: macroize two small inlines to avoid module.h
ip_vs.h: fix implicit use of module_get/module_put from module.h
nf_conntrack.h: fix up fallout from implicit moduleparam.h presence
include: replace linux/module.h with "struct module" wherever possible
include: convert various register fcns to macros to avoid include chaining
crypto.h: remove unused crypto_tfm_alg_modname() inline
uwb.h: fix implicit use of asm/page.h for PAGE_SIZE
pm_runtime.h: explicitly requires notifier.h
linux/dmaengine.h: fix implicit use of bitmap.h and asm/page.h
miscdevice.h: fix up implicit use of lists and types
stop_machine.h: fix implicit use of smp.h for smp_processor_id
of: fix implicit use of errno.h in include/linux/of.h
of_platform.h: delete needless include
acpi: remove module.h include from platform/aclinux.h
miscdevice.h: delete unnecessary inclusion of module.h
device_cgroup.h: delete needless include
net: sch_generic remove redundant use of
net: inet_timewait_sock doesnt need
...Fix up trivial conflicts (other header files, and removal of the ab3550 mfd driver) in
- drivers/media/dvb/frontends/dibx000_common.c
- drivers/media/video/{mt9m111.c,ov6650.c}
- drivers/mfd/ab3550-core.c
- include/linux/dmaengine.h
03 Nov, 2011
1 commit
-
Michel while working on the working set estimation code, noticed that
calling get_page_unless_zero() on a random pfn_to_page(random_pfn)
wasn't safe, if the pfn ended up being a tail page of a transparent
hugepage under splitting by __split_huge_page_refcount().He then found the problem could also theoretically materialize with
page_cache_get_speculative() during the speculative radix tree lookups
that uses get_page_unless_zero() in SMP if the radix tree page is freed
and reallocated and get_user_pages is called on it before
page_cache_get_speculative has a chance to call get_page_unless_zero().So the best way to fix the problem is to keep page_tail->_count zero at
all times. This will guarantee that get_page_unless_zero() can never
succeed on any tail page. page_tail->_mapcount is guaranteed zero and
is unused for all tail pages of a compound page, so we can simply
account the tail page references there and transfer them to
tail_page->_count in __split_huge_page_refcount() (in addition to the
head_page->_mapcount).While debugging this s/_count/_mapcount/ change I also noticed get_page is
called by direct-io.c on pages returned by get_user_pages. That wasn't
entirely safe because the two atomic_inc in get_page weren't atomic. As
opposed to other get_user_page users like secondary-MMU page fault to
establish the shadow pagetables would never call any superflous get_page
after get_user_page returns. It's safer to make get_page universally safe
for tail pages and to use get_page_foll() within follow_page (inside
get_user_pages()). get_page_foll() is safe to do the refcounting for tail
pages without taking any locks because it is run within PT lock protected
critical sections (PT lock for pte and page_table_lock for
pmd_trans_huge).The standard get_page() as invoked by direct-io instead will now take
the compound_lock but still only for tail pages. The direct-io paths
are usually I/O bound and the compound_lock is per THP so very
finegrined, so there's no risk of scalability issues with it. A simple
direct-io benchmarks with all lockdep prove locking and spinlock
debugging infrastructure enabled shows identical performance and no
overhead. So it's worth it. Ideally direct-io should stop calling
get_page() on pages returned by get_user_pages(). The spinlock in
get_page() is already optimized away for no-THP builds but doing
get_page() on tail pages returned by GUP is generally a rare operation
and usually only run in I/O paths.This new refcounting on page_tail->_mapcount in addition to avoiding new
RCU critical sections will also allow the working set estimation code to
work without any further complexity associated to the tail page
refcounting with THP.Signed-off-by: Andrea Arcangeli
Reported-by: Michel Lespinasse
Reviewed-by: Michel Lespinasse
Reviewed-by: Minchan Kim
Cc: Peter Zijlstra
Cc: Hugh Dickins
Cc: Johannes Weiner
Cc: Rik van Riel
Cc: Mel Gorman
Cc: KOSAKI Motohiro
Cc: Benjamin Herrenschmidt
Cc: David Gibson
Cc:
Cc:
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
31 Oct, 2011
1 commit
-
The files changed within are only using the EXPORT_SYMBOL
macro variants. They are not using core modular infrastructure
and hence don't need module.h but only the export.h header.Signed-off-by: Paul Gortmaker
25 May, 2011
2 commits
-
The zone->lru_lock is heavily contented in workload where activate_page()
is frequently used. We could do batch activate_page() to reduce the lock
contention. The batched pages will be added into zone list when the pool
is full or page reclaim is trying to drain them.For example, in a 4 socket 64 CPU system, create a sparse file and 64
processes, processes shared map to the file. Each process read access the
whole file and then exit. The process exit will do unmap_vmas() and cause
a lot of activate_page() call. In such workload, we saw about 58% total
time reduction with below patch. Other workloads with a lot of
activate_page also benefits a lot too.Andrew Morton suggested activate_page() and putback_lru_pages() should
follow the same path to active pages, but this is hard to implement (see
commit 7a608572a282a ("Revert "mm: batch activate_page() to reduce lock
contention")). On the other hand, do we really need putback_lru_pages()
to follow the same path? I tested several FIO/FFSB benchmark (about 20
scripts for each benchmark) in 3 machines here from 2 sockets to 4
sockets. My test doesn't show anything significant with/without below
patch (there is slight difference but mostly some noise which we found
even without below patch before). Below patch basically returns to the
same as my first post.I tested some microbenchmarks:
case-anon-cow-rand-mt 0.58%
case-anon-cow-rand -3.30%
case-anon-cow-seq-mt -0.51%
case-anon-cow-seq -5.68%
case-anon-r-rand-mt 0.23%
case-anon-r-rand 0.81%
case-anon-r-seq-mt -0.71%
case-anon-r-seq -1.99%
case-anon-rx-rand-mt 2.11%
case-anon-rx-seq-mt 3.46%
case-anon-w-rand-mt -0.03%
case-anon-w-rand -0.50%
case-anon-w-seq-mt -1.08%
case-anon-w-seq -0.12%
case-anon-wx-rand-mt -5.02%
case-anon-wx-seq-mt -1.43%
case-fork 1.65%
case-fork-sleep -0.07%
case-fork-withmem 1.39%
case-hugetlb -0.59%
case-lru-file-mmap-read-mt -0.54%
case-lru-file-mmap-read 0.61%
case-lru-file-mmap-read-rand -2.24%
case-lru-file-readonce -0.64%
case-lru-file-readtwice -11.69%
case-lru-memcg -1.35%
case-mmap-pread-rand-mt 1.88%
case-mmap-pread-rand -15.26%
case-mmap-pread-seq-mt 0.89%
case-mmap-pread-seq -69.72%
case-mmap-xread-rand-mt 0.71%
case-mmap-xread-seq-mt 0.38%The most significent are:
case-lru-file-readtwice -11.69%
case-mmap-pread-rand -15.26%
case-mmap-pread-seq -69.72%which use activate_page a lot. others are basically variations because
each run has slightly difference.In UP case, 'size mm/swap.o'
before the two patches:
text data bss dec hex filename
6466 896 4 7366 1cc6 mm/swap.o
after the two patches:
text data bss dec hex filename
6343 896 4 7243 1c4b mm/swap.oSigned-off-by: Shaohua Li
Cc: KOSAKI Motohiro
Cc: Hiroyuki Kamezawa
Cc: Andi Kleen
Cc: Minchan Kim
Cc: Rik van Riel
Cc: Mel Gorman
Cc: Johannes Weiner
Cc: Hugh Dickins
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
It's pointless that deactive_page's operates on unevictable pages. This
patch removes unnecessary overhead which might be a bit problem in case
that there are many unevictable page in system(ex, mprotect workload)[akpm@linux-foundation.org: tidy up comment]
Reviewed-by: KOSAKI Motohiro
Signed-off-by: Minchan Kim
Reviewed-by: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
12 May, 2011
1 commit
-
The lru_deactivate_fn should not move page which in on unevictable lru
into inactive list. Otherwise, we can meet BUG when we use
isolate_lru_pages as __isolate_lru_page could return -EINVAL.Reported-by: Ying Han
Tested-by: Ying Han
Signed-off-by: Minchan Kim
Reviewed-by: KOSAKI Motohiro
Reviewed-by: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
23 Mar, 2011
4 commits
-
Clean up code and remove duplicate code. Next patch will use
pagevec_lru_move_fn introduced here too.Signed-off-by: Shaohua Li
Cc: KOSAKI Motohiro
Cc: Hiroyuki Kamezawa
Cc: Andi Kleen
Reviewed-by: Minchan Kim
Cc: Rik van Riel
Cc: Mel Gorman
Cc: Johannes Weiner
Cc: Hugh Dickins
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
invalidate_mapping_pages is very big hint to reclaimer. It means user
doesn't want to use the page any more. So in order to prevent working set
page eviction, this patch move the page into tail of inactive list by
PG_reclaim.Please, remember that pages in inactive list are working set as well as
active list. If we don't move pages into inactive list's tail, pages near
by tail of inactive list can be evicted although we have a big clue about
useless pages. It's totally bad.Now PG_readahead/PG_reclaim is shared. fe3cba17 added ClearPageReclaim
into clear_page_dirty_for_io for preventing fast reclaiming readahead
marker page.In this series, PG_reclaim is used by invalidated page, too. If VM find
the page is invalidated and it's dirty, it sets PG_reclaim to reclaim
asap. Then, when the dirty page will be writeback,
clear_page_dirty_for_io will clear PG_reclaim unconditionally. It
disturbs this serie's goal.I think it's okay to clear PG_readahead when the page is dirty, not
writeback time. So this patch moves ClearPageReadahead. In v4,
ClearPageReadahead in set_page_dirty has a problem which is reported by
Steven Barrett. It's due to compound page. Some driver(ex, audio) calls
set_page_dirty with compound page which isn't on LRU. but my patch does
ClearPageRelcaim on compound page. In non-CONFIG_PAGEFLAGS_EXTENDED, it
breaks PageTail flag.I think it doesn't affect THP and pass my test with THP enabling but Cced
Andrea for double check.Signed-off-by: Minchan Kim
Reported-by: Steven Barrett
Reviewed-by: Johannes Weiner
Acked-by: Rik van Riel
Acked-by: Mel Gorman
Cc: Wu Fengguang
Cc: KOSAKI Motohiro
Cc: Nick Piggin
Cc: Andrea Arcangeli
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The rotate_reclaimable_page function moves just written out pages, which
the VM wanted to reclaim, to the end of the inactive list. That way the
VM will find those pages first next time it needs to free memory.This patch applies the rule in memcg. It can help to prevent unnecessary
working page eviction of memcg.Signed-off-by: Minchan Kim
Acked-by: Balbir Singh
Acked-by: KAMEZAWA Hiroyuki
Reviewed-by: Rik van Riel
Cc: KOSAKI Motohiro
Acked-by: Johannes Weiner
Cc: Andrea Arcangeli
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Recently, there are reported problem about thrashing.
(http://marc.info/?l=rsync&m=128885034930933&w=2) It happens by backup
workloads(ex, nightly rsync). That's because the workload makes just
use-once pages and touches pages twice. It promotes the page into active
list so that it results in working set page eviction.Some app developer want to support POSIX_FADV_NOREUSE. But other OSes
don't support it, either.
(http://marc.info/?l=linux-mm&m=128928979512086&w=2)By other approach, app developers use POSIX_FADV_DONTNEED. But it has a
problem. If kernel meets page is writing during invalidate_mapping_pages,
it can't work. It makes for application programmer to use it since they
always have to sync data before calling fadivse(..POSIX_FADV_DONTNEED) to
make sure the pages could be discardable. At last, they can't use
deferred write of kernel so that they could see performance loss.
(http://insights.oetiker.ch/linux/fadvise.html)In fact, invalidation is very big hint to reclaimer. It means we don't
use the page any more. So let's move the writing page into inactive
list's head if we can't truncate it right now.Why I move page to head of lru on this patch, Dirty/Writeback page would
be flushed sooner or later. It can prevent writeout of pageout which is
less effective than flusher's writeout.Originally, I reused lru_demote of Peter with some change so added his
Signed-off-by.Signed-off-by: Minchan Kim
Reported-by: Ben Gamari
Signed-off-by: Peter Zijlstra
Acked-by: Rik van Riel
Acked-by: Mel Gorman
Reviewed-by: KOSAKI Motohiro
Cc: Wu Fengguang
Acked-by: Johannes Weiner
Cc: Nick Piggin
Cc: Andrea Arcangeli
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
18 Jan, 2011
2 commits
-
This reverts commit d8505dee1a87b8d41b9c4ee1325cd72258226fbc.
Chris Mason ended up chasing down some page allocation errors and pages
stuck waiting on the IO scheduler, and was able to narrow it down to two
commits: commit 744ed1442757 ("mm: batch activate_page() to reduce lock
contention") and d8505dee1a87 ("mm: simplify code of swap.c").This reverts the second one.
Reported-and-debugged-by: Chris Mason
Cc: Mel Gorman
Cc: Andrew Morton
Cc: Jens Axboe
Cc: linux-mm
Cc: KAMEZAWA Hiroyuki
Cc: Andrea Arcangeli
Cc: Shaohua Li
Signed-off-by: Linus Torvalds -
This reverts commit 744ed1442757767ffede5008bb13e0805085902e.
Chris Mason ended up chasing down some page allocation errors and pages
stuck waiting on the IO scheduler, and was able to narrow it down to two
commits: commit 744ed1442757 ("mm: batch activate_page() to reduce lock
contention") and d8505dee1a87 ("mm: simplify code of swap.c").This reverts the first of them.
Reported-and-debugged-by: Chris Mason
Cc: Mel Gorman
Cc: Andrew Morton
Cc: Jens Axboe
Cc: linux-mm
Cc: KAMEZAWA Hiroyuki
Cc: Andrea Arcangeli
Cc: Shaohua Li
Signed-off-by: Linus Torvalds
14 Jan, 2011
3 commits
-
The zone->lru_lock is heavily contented in workload where activate_page()
is frequently used. We could do batch activate_page() to reduce the lock
contention. The batched pages will be added into zone list when the pool
is full or page reclaim is trying to drain them.For example, in a 4 socket 64 CPU system, create a sparse file and 64
processes, processes shared map to the file. Each process read access the
whole file and then exit. The process exit will do unmap_vmas() and cause
a lot of activate_page() call. In such workload, we saw about 58% total
time reduction with below patch. Other workloads with a lot of
activate_page also benefits a lot too.I tested some microbenchmarks:
case-anon-cow-rand-mt 0.58%
case-anon-cow-rand -3.30%
case-anon-cow-seq-mt -0.51%
case-anon-cow-seq -5.68%
case-anon-r-rand-mt 0.23%
case-anon-r-rand 0.81%
case-anon-r-seq-mt -0.71%
case-anon-r-seq -1.99%
case-anon-rx-rand-mt 2.11%
case-anon-rx-seq-mt 3.46%
case-anon-w-rand-mt -0.03%
case-anon-w-rand -0.50%
case-anon-w-seq-mt -1.08%
case-anon-w-seq -0.12%
case-anon-wx-rand-mt -5.02%
case-anon-wx-seq-mt -1.43%
case-fork 1.65%
case-fork-sleep -0.07%
case-fork-withmem 1.39%
case-hugetlb -0.59%
case-lru-file-mmap-read-mt -0.54%
case-lru-file-mmap-read 0.61%
case-lru-file-mmap-read-rand -2.24%
case-lru-file-readonce -0.64%
case-lru-file-readtwice -11.69%
case-lru-memcg -1.35%
case-mmap-pread-rand-mt 1.88%
case-mmap-pread-rand -15.26%
case-mmap-pread-seq-mt 0.89%
case-mmap-pread-seq -69.72%
case-mmap-xread-rand-mt 0.71%
case-mmap-xread-seq-mt 0.38%The most significent are:
case-lru-file-readtwice -11.69%
case-mmap-pread-rand -15.26%
case-mmap-pread-seq -69.72%which use activate_page a lot. others are basically variations because
each run has slightly difference.[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Shaohua Li
Cc: Andi Kleen
Cc: Minchan Kim
Cc: KOSAKI Motohiro
Cc: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Clean up code and remove duplicate code. Next patch will use
pagevec_lru_move_fn introduced here too.Signed-off-by: Shaohua Li
Cc: Andi Kleen
Cc: Minchan Kim
Cc: KOSAKI Motohiro
Cc: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Lately I've been working to make KVM use hugepages transparently without
the usual restrictions of hugetlbfs. Some of the restrictions I'd like to
see removed:1) hugepages have to be swappable or the guest physical memory remains
locked in RAM and can't be paged out to swap2) if a hugepage allocation fails, regular pages should be allocated
instead and mixed in the same vma without any failure and without
userland noticing3) if some task quits and more hugepages become available in the
buddy, guest physical memory backed by regular pages should be
relocated on hugepages automatically in regions under
madvise(MADV_HUGEPAGE) (ideally event driven by waking up the
kernel deamon if the order=HPAGE_PMD_SHIFT-PAGE_SHIFT list becomes
not null)4) avoidance of reservation and maximization of use of hugepages whenever
possible. Reservation (needed to avoid runtime fatal faliures) may be ok for
1 machine with 1 database with 1 database cache with 1 database cache size
known at boot time. It's definitely not feasible with a virtualization
hypervisor usage like RHEV-H that runs an unknown number of virtual machines
with an unknown size of each virtual machine with an unknown amount of
pagecache that could be potentially useful in the host for guest not using
O_DIRECT (aka cache=off).hugepages in the virtualization hypervisor (and also in the guest!) are
much more important than in a regular host not using virtualization,
becasue with NPT/EPT they decrease the tlb-miss cacheline accesses from 24
to 19 in case only the hypervisor uses transparent hugepages, and they
decrease the tlb-miss cacheline accesses from 19 to 15 in case both the
linux hypervisor and the linux guest both uses this patch (though the
guest will limit the addition speedup to anonymous regions only for
now...). Even more important is that the tlb miss handler is much slower
on a NPT/EPT guest than for a regular shadow paging or no-virtualization
scenario. So maximizing the amount of virtual memory cached by the TLB
pays off significantly more with NPT/EPT than without (even if there would
be no significant speedup in the tlb-miss runtime).The first (and more tedious) part of this work requires allowing the VM to
handle anonymous hugepages mixed with regular pages transparently on
regular anonymous vmas. This is what this patch tries to achieve in the
least intrusive possible way. We want hugepages and hugetlb to be used in
a way so that all applications can benefit without changes (as usual we
leverage the KVM virtualization design: by improving the Linux VM at
large, KVM gets the performance boost too).The most important design choice is: always fallback to 4k allocation if
the hugepage allocation fails! This is the _very_ opposite of some large
pagecache patches that failed with -EIO back then if a 64k (or similar)
allocation failed...Second important decision (to reduce the impact of the feature on the
existing pagetable handling code) is that at any time we can split an
hugepage into 512 regular pages and it has to be done with an operation
that can't fail. This way the reliability of the swapping isn't decreased
(no need to allocate memory when we are short on memory to swap) and it's
trivial to plug a split_huge_page* one-liner where needed without
polluting the VM. Over time we can teach mprotect, mremap and friends to
handle pmd_trans_huge natively without calling split_huge_page*. The fact
it can't fail isn't just for swap: if split_huge_page would return -ENOMEM
(instead of the current void) we'd need to rollback the mprotect from the
middle of it (ideally including undoing the split_vma) which would be a
big change and in the very wrong direction (it'd likely be simpler not to
call split_huge_page at all and to teach mprotect and friends to handle
hugepages instead of rolling them back from the middle). In short the
very value of split_huge_page is that it can't fail.The collapsing and madvise(MADV_HUGEPAGE) part will remain separated and
incremental and it'll just be an "harmless" addition later if this initial
part is agreed upon. It also should be noted that locking-wise replacing
regular pages with hugepages is going to be very easy if compared to what
I'm doing below in split_huge_page, as it will only happen when
page_count(page) matches page_mapcount(page) if we can take the PG_lock
and mmap_sem in write mode. collapse_huge_page will be a "best effort"
that (unlike split_huge_page) can fail at the minimal sign of trouble and
we can try again later. collapse_huge_page will be similar to how KSM
works and the madvise(MADV_HUGEPAGE) will work similar to
madvise(MADV_MERGEABLE).The default I like is that transparent hugepages are used at page fault
time. This can be changed with
/sys/kernel/mm/transparent_hugepage/enabled. The control knob can be set
to three values "always", "madvise", "never" which mean respectively that
hugepages are always used, or only inside madvise(MADV_HUGEPAGE) regions,
or never used. /sys/kernel/mm/transparent_hugepage/defrag instead
controls if the hugepage allocation should defrag memory aggressively
"always", only inside "madvise" regions, or "never".The pmd_trans_splitting/pmd_trans_huge locking is very solid. The
put_page (from get_user_page users that can't use mmu notifier like
O_DIRECT) that runs against a __split_huge_page_refcount instead was a
pain to serialize in a way that would result always in a coherent page
count for both tail and head. I think my locking solution with a
compound_lock taken only after the page_first is valid and is still a
PageHead should be safe but it surely needs review from SMP race point of
view. In short there is no current existing way to serialize the O_DIRECT
final put_page against split_huge_page_refcount so I had to invent a new
one (O_DIRECT loses knowledge on the mapping status by the time gup_fast
returns so...). And I didn't want to impact all gup/gup_fast users for
now, maybe if we change the gup interface substantially we can avoid this
locking, I admit I didn't think too much about it because changing the gup
unpinning interface would be invasive.If we ignored O_DIRECT we could stick to the existing compound refcounting
code, by simply adding a get_user_pages_fast_flags(foll_flags) where KVM
(and any other mmu notifier user) would call it without FOLL_GET (and if
FOLL_GET isn't set we'd just BUG_ON if nobody registered itself in the
current task mmu notifier list yet). But O_DIRECT is fundamental for
decent performance of virtualized I/O on fast storage so we can't avoid it
to solve the race of put_page against split_huge_page_refcount to achieve
a complete hugepage feature for KVM.Swap and oom works fine (well just like with regular pages ;). MMU
notifier is handled transparently too, with the exception of the young bit
on the pmd, that didn't have a range check but I think KVM will be fine
because the whole point of hugepages is that EPT/NPT will also use a huge
pmd when they notice gup returns pages with PageCompound set, so they
won't care of a range and there's just the pmd young bit to check in that
case.NOTE: in some cases if the L2 cache is small, this may slowdown and waste
memory during COWs because 4M of memory are accessed in a single fault
instead of 8k (the payoff is that after COW the program can run faster).
So we might want to switch the copy_huge_page (and clear_huge_page too) to
not temporal stores. I also extensively researched ways to avoid this
cache trashing with a full prefault logic that would cow in 8k/16k/32k/64k
up to 1M (I can send those patches that fully implemented prefault) but I
concluded they're not worth it and they add an huge additional complexity
and they remove all tlb benefits until the full hugepage has been faulted
in, to save a little bit of memory and some cache during app startup, but
they still don't improve substantially the cache-trashing during startup
if the prefault happens in >4k chunks. One reason is that those 4k pte
entries copied are still mapped on a perfectly cache-colored hugepage, so
the trashing is the worst one can generate in those copies (cow of 4k page
copies aren't so well colored so they trashes less, but again this results
in software running faster after the page fault). Those prefault patches
allowed things like a pte where post-cow pages were local 4k regular anon
pages and the not-yet-cowed pte entries were pointing in the middle of
some hugepage mapped read-only. If it doesn't payoff substantially with
todays hardware it will payoff even less in the future with larger l2
caches, and the prefault logic would blot the VM a lot. If one is
emebdded transparent_hugepage can be disabled during boot with sysfs or
with the boot commandline parameter transparent_hugepage=0 (or
transparent_hugepage=2 to restrict hugepages inside madvise regions) that
will ensure not a single hugepage is allocated at boot time. It is simple
enough to just disable transparent hugepage globally and let transparent
hugepages be allocated selectively by applications in the MADV_HUGEPAGE
region (both at page fault time, and if enabled with the
collapse_huge_page too through the kernel daemon).This patch supports only hugepages mapped in the pmd, archs that have
smaller hugepages will not fit in this patch alone. Also some archs like
power have certain tlb limits that prevents mixing different page size in
the same regions so they will not fit in this framework that requires
"graceful fallback" to basic PAGE_SIZE in case of physical memory
fragmentation. hugetlbfs remains a perfect fit for those because its
software limits happen to match the hardware limits. hugetlbfs also
remains a perfect fit for hugepage sizes like 1GByte that cannot be hoped
to be found not fragmented after a certain system uptime and that would be
very expensive to defragment with relocation, so requiring reservation.
hugetlbfs is the "reservation way", the point of transparent hugepages is
not to have any reservation at all and maximizing the use of cache and
hugepages at all times automatically.Some performance result:
vmx andrea # LD_PRELOAD=/usr/lib64/libhugetlbfs.so HUGETLB_MORECORE=yes HUGETLB_PATH=/mnt/huge/ ./largep
ages3
memset page fault 1566023
memset tlb miss 453854
memset second tlb miss 453321
random access tlb miss 41635
random access second tlb miss 41658
vmx andrea # LD_PRELOAD=/usr/lib64/libhugetlbfs.so HUGETLB_MORECORE=yes HUGETLB_PATH=/mnt/huge/ ./largepages3
memset page fault 1566471
memset tlb miss 453375
memset second tlb miss 453320
random access tlb miss 41636
random access second tlb miss 41637
vmx andrea # ./largepages3
memset page fault 1566642
memset tlb miss 453417
memset second tlb miss 453313
random access tlb miss 41630
random access second tlb miss 41647
vmx andrea # ./largepages3
memset page fault 1566872
memset tlb miss 453418
memset second tlb miss 453315
random access tlb miss 41618
random access second tlb miss 41659
vmx andrea # echo 0 > /proc/sys/vm/transparent_hugepage
vmx andrea # ./largepages3
memset page fault 2182476
memset tlb miss 460305
memset second tlb miss 460179
random access tlb miss 44483
random access second tlb miss 44186
vmx andrea # ./largepages3
memset page fault 2182791
memset tlb miss 460742
memset second tlb miss 459962
random access tlb miss 43981
random access second tlb miss 43988============
#include
#include
#include
#include#define SIZE (3UL*1024*1024*1024)
int main()
{
char *p = malloc(SIZE), *p2;
struct timeval before, after;gettimeofday(&before, NULL);
memset(p, 0, SIZE);
gettimeofday(&after, NULL);
printf("memset page fault %Lu\n",
(after.tv_sec-before.tv_sec)*1000000UL +
after.tv_usec-before.tv_usec);gettimeofday(&before, NULL);
memset(p, 0, SIZE);
gettimeofday(&after, NULL);
printf("memset tlb miss %Lu\n",
(after.tv_sec-before.tv_sec)*1000000UL +
after.tv_usec-before.tv_usec);gettimeofday(&before, NULL);
memset(p, 0, SIZE);
gettimeofday(&after, NULL);
printf("memset second tlb miss %Lu\n",
(after.tv_sec-before.tv_sec)*1000000UL +
after.tv_usec-before.tv_usec);gettimeofday(&before, NULL);
for (p2 = p; p2 < p+SIZE; p2 += 4096)
*p2 = 0;
gettimeofday(&after, NULL);
printf("random access tlb miss %Lu\n",
(after.tv_sec-before.tv_sec)*1000000UL +
after.tv_usec-before.tv_usec);gettimeofday(&before, NULL);
for (p2 = p; p2 < p+SIZE; p2 += 4096)
*p2 = 0;
gettimeofday(&after, NULL);
printf("random access second tlb miss %Lu\n",
(after.tv_sec-before.tv_sec)*1000000UL +
after.tv_usec-before.tv_usec);return 0;
}
============Signed-off-by: Andrea Arcangeli
Acked-by: Rik van Riel
Signed-off-by: Johannes Weiner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds