The problem starts with a file backed dirty page which is charged to a
memcg. Then page migration is used to move oldpage to newpage.
Migration:
- copies the oldpage's data to newpage
- clears oldpage.PG_dirty
- sets newpage.PG_dirty
- uncharges oldpage from memcg
- charges newpage to memcg
Clearing oldpage.PG_dirty decrements the charged memcg's dirty page
count.
However, because newpage is not yet charged, setting newpage.PG_dirty
does not increment the memcg's dirty page count. After migration
completes newpage.PG_dirty is eventually cleared, often in
account_page_cleaned(). At this time newpage is charged to a memcg so
the memcg's dirty page count is decremented which causes underflow
because the count was not previously incremented by migration. This
underflow causes balance_dirty_pages() to see a very large unsigned
number of dirty memcg pages which leads to aggressive throttling of
buffered writes by processes in non root memcg.
This issue:
- can harm performance of non root memcg buffered writes.
- can report too small (even negative) values in
memory.stat[(total_)dirty] counters of all memcg, including the root.
To avoid polluting migrate.c with #ifdef CONFIG_MEMCG checks, introduce
page_memcg() and set_page_memcg() helpers.
Test:
0) setup and enter limited memcg
mkdir /sys/fs/cgroup/test
echo 1G > /sys/fs/cgroup/test/memory.limit_in_bytes
echo $$ > /sys/fs/cgroup/test/cgroup.procs
1) buffered writes baseline
dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k
sync
grep ^dirty /sys/fs/cgroup/test/memory.stat
2) buffered writes with compaction antagonist to induce migration
yes 1 > /proc/sys/vm/compact_memory &
rm -rf /data/tmp/foo
dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k
kill %
sync
grep ^dirty /sys/fs/cgroup/test/memory.stat
3) buffered writes without antagonist, should match baseline
rm -rf /data/tmp/foo
dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k
sync
grep ^dirty /sys/fs/cgroup/test/memory.stat
(speed, dirty residue)
unpatched patched
1) 841 MB/s 0 dirty pages 886 MB/s 0 dirty pages
2) 611 MB/s -33427456 dirty pages 793 MB/s 0 dirty pages
3) 114 MB/s -33427456 dirty pages 891 MB/s 0 dirty pages
Notice that unpatched baseline performance (1) fell after
migration (3): 841 -> 114 MB/s. In the patched kernel, post
migration performance matches baseline.
Fixes: c4843a7593 ("memcg: add per cgroup dirty page accounting")
Signed-off-by: Greg Thelen <gthelen@google.com>
Reported-by: Dave Hansen <dave.hansen@intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org> [4.2+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull media updates from Mauro Carvalho Chehab:
"A series of patches that move part of the code used to allocate memory
from the media subsystem to the mm subsystem"
[ The mm parts have been acked by VM people, and the series was
apparently in -mm for a while - Linus ]
* tag 'media/v4.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
[media] drm/exynos: Convert g2d_userptr_get_dma_addr() to use get_vaddr_frames()
[media] media: vb2: Remove unused functions
[media] media: vb2: Convert vb2_dc_get_userptr() to use frame vector
[media] media: vb2: Convert vb2_vmalloc_get_userptr() to use frame vector
[media] media: vb2: Convert vb2_dma_sg_get_userptr() to use frame vector
[media] vb2: Provide helpers for mapping virtual addresses
[media] media: omap_vout: Convert omap_vout_uservirt_to_phys() to use get_vaddr_pfns()
[media] mm: Provide new get_vaddr_frames() helper
[media] vb2: Push mmap_sem down to memops
Add the additional "vm_flags_t vm_flags" argument to do_mmap_pgoff(),
rename it to do_mmap(), and re-introduce do_mmap_pgoff() as a simple
wrapper on top of do_mmap(). Perhaps we should update the callers of
do_mmap_pgoff() and kill it later.
This way mpx_mmap() can simply call do_mmap(vm_flags => VM_MPX) and do not
play with vm internals.
After this change mmap_region() has a single user outside of mmap.c,
arch/tile/mm/elf.c:arch_setup_additional_pages(). It would be nice to
change arch/tile/ and unexport mmap_region().
[kirill@shutemov.name: fix build]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Tested-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge second patch-bomb from Andrew Morton:
"Almost all of the rest of MM. There was an unusually large amount of
MM material this time"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (141 commits)
zpool: remove no-op module init/exit
mm: zbud: constify the zbud_ops
mm: zpool: constify the zpool_ops
mm: swap: zswap: maybe_preload & refactoring
zram: unify error reporting
zsmalloc: remove null check from destroy_handle_cache()
zsmalloc: do not take class lock in zs_shrinker_count()
zsmalloc: use class->pages_per_zspage
zsmalloc: consider ZS_ALMOST_FULL as migrate source
zsmalloc: partial page ordering within a fullness_list
zsmalloc: use shrinker to trigger auto-compaction
zsmalloc: account the number of compacted pages
zsmalloc/zram: introduce zs_pool_stats api
zsmalloc: cosmetic compaction code adjustments
zsmalloc: introduce zs_can_compact() function
zsmalloc: always keep per-class stats
zsmalloc: drop unused variable `nr_to_migrate'
mm/memblock.c: fix comment in __next_mem_range()
mm/page_alloc.c: fix type information of memoryless node
memory-hotplug: fix comments in zone_spanned_pages_in_node() and zone_spanned_pages_in_node()
...
The pair of get/set_freepage_migratetype() functions are used to cache
pageblock migratetype for a page put on a pcplist, so that it does not
have to be retrieved again when the page is put on a free list (e.g.
when pcplists become full). Historically it was also assumed that the
value is accurate for pages on freelists (as the functions' names
unfortunately suggest), but that cannot be guaranteed without affecting
various allocator fast paths. It is in fact not needed and all such
uses have been removed.
The last remaining (but pointless) usage related to pages of freelists
is in move_freepages(), which this patch removes.
To prevent further confusion, rename the functions to
get/set_pcppage_migratetype() and expand their description. Since all
the users are now in mm/page_alloc.c, move the functions there from the
shared header.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Cc: Laura Abbott <lauraa@codeaurora.org>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Seungho Park <seungho1.park@lge.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
special_mapping_fault() is absolutely broken. It seems it was always
wrong, but this didn't matter until vdso/vvar started to use more than
one page.
And after this change vma_is_anonymous() becomes really trivial, it
simply checks vm_ops == NULL. However, I do think the helper makes
sense. There are a lot of ->vm_ops != NULL checks, the helper makes the
caller's code more understandable (self-documented) and this is more
grep-friendly.
This patch (of 3):
Preparation. Add the new simple helper, vma_is_anonymous(vma), and change
handle_pte_fault() to use it. It will have more users.
The name is not accurate, say a hpet_mmap()'ed vma is not anonymous.
Perhaps it should be named vma_has_fault() instead. But it matches the
logic in mmap.c/memory.c (see next changes). "True" just means that a
page fault will use do_anonymous_page().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull libnvdimm updates from Dan Williams:
"This update has successfully completed a 0day-kbuild run and has
appeared in a linux-next release. The changes outside of the typical
drivers/nvdimm/ and drivers/acpi/nfit.[ch] paths are related to the
removal of IORESOURCE_CACHEABLE, the introduction of memremap(), and
the introduction of ZONE_DEVICE + devm_memremap_pages().
Summary:
- Introduce ZONE_DEVICE and devm_memremap_pages() as a generic
mechanism for adding device-driver-discovered memory regions to the
kernel's direct map.
This facility is used by the pmem driver to enable pfn_to_page()
operations on the page frames returned by DAX ('direct_access' in
'struct block_device_operations').
For now, the 'memmap' allocation for these "device" pages comes
from "System RAM". Support for allocating the memmap from device
memory will arrive in a later kernel.
- Introduce memremap() to replace usages of ioremap_cache() and
ioremap_wt(). memremap() drops the __iomem annotation for these
mappings to memory that do not have i/o side effects. The
replacement of ioremap_cache() with memremap() is limited to the
pmem driver to ease merging the api change in v4.3.
Completion of the conversion is targeted for v4.4.
- Similar to the usage of memcpy_to_pmem() + wmb_pmem() in the pmem
driver, update the VFS DAX implementation and PMEM api to provide
persistence guarantees for kernel operations on a DAX mapping.
- Convert the ACPI NFIT 'BLK' driver to map the block apertures as
cacheable to improve performance.
- Miscellaneous updates and fixes to libnvdimm including support for
issuing "address range scrub" commands, clarifying the optimal
'sector size' of pmem devices, a clarification of the usage of the
ACPI '_STA' (status) property for DIMM devices, and other minor
fixes"
* tag 'libnvdimm-for-4.3' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (34 commits)
libnvdimm, pmem: direct map legacy pmem by default
libnvdimm, pmem: 'struct page' for pmem
libnvdimm, pfn: 'struct page' provider infrastructure
x86, pmem: clarify that ARCH_HAS_PMEM_API implies PMEM mapped WB
add devm_memremap_pages
mm: ZONE_DEVICE for "device memory"
mm: move __phys_to_pfn and __pfn_to_phys to asm/generic/memory_model.h
dax: drop size parameter to ->direct_access()
nd_blk: change aperture mapping from WC to WB
nvdimm: change to use generic kvfree()
pmem, dax: have direct_access use __pmem annotation
dax: update I/O path to do proper PMEM flushing
pmem: add copy_from_iter_pmem() and clear_pmem()
pmem, x86: clean up conditional pmem includes
pmem: remove layer when calling arch_has_wmb_pmem()
pmem, x86: move x86 PMEM API to new pmem.h header
libnvdimm, e820: make CONFIG_X86_PMEM_LEGACY a tristate option
pmem: switch to devm_ allocations
devres: add devm_memremap
libnvdimm, btt: write and validate parent_uuid
...
Commit c48a11c7ad ("netvm: propagate page->pfmemalloc to skb") added
checks for page->pfmemalloc to __skb_fill_page_desc():
if (page->pfmemalloc && !page->mapping)
skb->pfmemalloc = true;
It assumes page->mapping == NULL implies that page->pfmemalloc can be
trusted. However, __delete_from_page_cache() can set set page->mapping
to NULL and leave page->index value alone. Due to being in union, a
non-zero page->index will be interpreted as true page->pfmemalloc.
So the assumption is invalid if the networking code can see such a page.
And it seems it can. We have encountered this with a NFS over loopback
setup when such a page is attached to a new skbuf. There is no copying
going on in this case so the page confuses __skb_fill_page_desc which
interprets the index as pfmemalloc flag and the network stack drops
packets that have been allocated using the reserves unless they are to
be queued on sockets handling the swapping which is the case here and
that leads to hangs when the nfs client waits for a response from the
server which has been dropped and thus never arrive.
The struct page is already heavily packed so rather than finding another
hole to put it in, let's do a trick instead. We can reuse the index
again but define it to an impossible value (-1UL). This is the page
index so it should never see the value that large. Replace all direct
users of page->pfmemalloc by page_is_pfmemalloc which will hide this
nastiness from unspoiled eyes.
The information will get lost if somebody wants to use page->index
obviously but that was the case before and the original code expected
that the information should be persisted somewhere else if that is
really needed (e.g. what SLAB and SLUB do).
[akpm@linux-foundation.org: fix blooper in slub]
Fixes: c48a11c7ad ("netvm: propagate page->pfmemalloc to skb")
Signed-off-by: Michal Hocko <mhocko@suse.com>
Debugged-by: Vlastimil Babka <vbabka@suse.com>
Debugged-by: Jiri Bohac <jbohac@suse.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Miller <davem@davemloft.net>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: <stable@vger.kernel.org> [3.6+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Provide new function get_vaddr_frames(). This function maps virtual
addresses from given start and fills given array with page frame numbers of
the corresponding pages. If given start belongs to a normal vma, the function
grabs reference to each of the pages to pin them in memory. If start
belongs to VM_IO | VM_PFNMAP vma, we don't touch page structures. Caller
must make sure pfns aren't reused for anything else while he is using
them.
This function is created for various drivers to simplify handling of
their buffers.
Signed-off-by: Jan Kara <jack@suse.cz>
Acked-by: Mel Gorman <mgorman@suse.de>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
region_is_ram() is used to prevent the establishment of aliased mappings
to physical "System RAM" with incompatible cache settings. However, it
uses "-1" to indicate both "unknown" memory ranges (ranges not described
by platform firmware) and "mixed" ranges (where the parameters describe
a range that partially overlaps "System RAM").
Fix this up by explicitly tracking the "unknown" vs "mixed" resource
cases and returning REGION_INTERSECTS, REGION_MIXED, or REGION_DISJOINT.
This re-write also adds support for detecting when the requested region
completely eclipses all of a resource. Note, the implementation treats
overlaps between "unknown" and the requested memory type as
REGION_INTERSECTS.
Finally, other memory types can be passed in by name, for now the only
usage "System RAM".
Suggested-by: Luis R. Rodriguez <mcgrof@suse.com>
Reviewed-by: Toshi Kani <toshi.kani@hp.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
__early_pfn_to_nid() use static variables to cache recent lookups as
memblock lookups are very expensive but it assumes that memory
initialisation is single-threaded. Parallel initialisation of struct
pages will break that assumption so this patch makes __early_pfn_to_nid()
SMP-safe by requiring the caller to cache recent search information.
early_pfn_to_nid() keeps the same interface but is only safe to use early
in boot due to the use of a global static variable. meminit_pfn_in_nid()
is an SMP-safe version that callers must maintain their own state for.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Tested-by: Nate Zimmer <nzimmer@sgi.com>
Tested-by: Waiman Long <waiman.long@hp.com>
Tested-by: Daniel J Blueman <daniel@numascale.com>
Acked-by: Pekka Enberg <penberg@kernel.org>
Cc: Robin Holt <robinmholt@gmail.com>
Cc: Nate Zimmer <nzimmer@sgi.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Waiman Long <waiman.long@hp.com>
Cc: Scott Norton <scott.norton@hp.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull cgroup writeback support from Jens Axboe:
"This is the big pull request for adding cgroup writeback support.
This code has been in development for a long time, and it has been
simmering in for-next for a good chunk of this cycle too. This is one
of those problems that has been talked about for at least half a
decade, finally there's a solution and code to go with it.
Also see last weeks writeup on LWN:
http://lwn.net/Articles/648292/"
* 'for-4.2/writeback' of git://git.kernel.dk/linux-block: (85 commits)
writeback, blkio: add documentation for cgroup writeback support
vfs, writeback: replace FS_CGROUP_WRITEBACK with SB_I_CGROUPWB
writeback: do foreign inode detection iff cgroup writeback is enabled
v9fs: fix error handling in v9fs_session_init()
bdi: fix wrong error return value in cgwb_create()
buffer: remove unusued 'ret' variable
writeback: disassociate inodes from dying bdi_writebacks
writeback: implement foreign cgroup inode bdi_writeback switching
writeback: add lockdep annotation to inode_to_wb()
writeback: use unlocked_inode_to_wb transaction in inode_congested()
writeback: implement unlocked_inode_to_wb transaction and use it for stat updates
writeback: implement [locked_]inode_to_wb_and_lock_list()
writeback: implement foreign cgroup inode detection
writeback: make writeback_control track the inode being written back
writeback: relocate wb[_try]_get(), wb_put(), inode_{attach|detach}_wb()
mm: vmscan: disable memcg direct reclaim stalling if cgroup writeback support is in use
writeback: implement memcg writeback domain based throttling
writeback: reset wb_domain->dirty_limit[_tstmp] when memcg domain size changes
writeback: implement memcg wb_domain
writeback: update wb_over_bg_thresh() to use wb_domain aware operations
...
memory_failure() can run in 2 different mode (specified by
MF_COUNT_INCREASED) in page refcount perspective. When
MF_COUNT_INCREASED is set, memory_failure() assumes that the caller
takes a refcount of the target page. And if cleared, memory_failure()
takes it in it's own.
In current code, however, refcounting is done differently in each caller.
For example, madvise_hwpoison() uses get_user_pages_fast() and
hwpoison_inject() uses get_page_unless_zero(). So this inconsistent
refcounting causes refcount failure especially for thp tail pages.
Typical user visible effects are like memory leak or
VM_BUG_ON_PAGE(!page_count(page)) in isolate_lru_page().
To fix this refcounting issue, this patch introduces get_hwpoison_page()
to handle thp tail pages in the same manner for each caller of hwpoison
code.
memory_failure() might fail to split thp and in such case it returns
without completing page isolation. This is not good because PageHWPoison
on the thp is still set and there's no easy way to unpoison such thps. So
this patch try to roll back any action to the thp in "non anonymous thp"
case and "thp split failed" case, expecting an MCE(SRAR) generated by
later access afterward will properly free such thps.
[akpm@linux-foundation.org: fix CONFIG_HWPOISON_INJECT=m]
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Reintroduce 8d63d99a5d ("mm: avoid tail page refcounting on non-THP
compound pages") after removing bogus VM_BUG_ON_PAGE() in
put_unrefcounted_compound_page().
THP uses tail page refcounting to be able to split huge pages at any time.
Tail page refcounting is not needed for other users of compound pages and
it's harmful because of overhead.
We try to exclude non-THP pages from tail page refcounting using
__compound_tail_refcounted() check. It excludes most common non-THP
compound pages: SL*B and hugetlb, but it doesn't catch rest of __GFP_COMP
users -- drivers.
And it's not only about overhead.
Drivers might want to use compound pages to get refcounting semantics
suitable for mapping high-order pages to userspace. But tail page
refcounting breaks it.
Tail page refcounting uses ->_mapcount in tail pages to store GUP pins on
them. It means GUP pins would affect page_mapcount() for tail pages.
It's not a problem for THP, because it never maps tail pages. But unlike
THP, drivers map parts of compound pages with PTEs and it makes
page_mapcount() be called for tail pages.
In particular, GUP pins would shift PSS up and affect /proc/kpagecount for
such pages. But, I'm not aware about anything which can lead to crash or
other serious misbehaviour.
Since currently all THP pages are anonymous and all drivers pages are not,
we can fix the __compound_tail_refcounted() check by requiring PageAnon()
to enable tail page refcounting.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Andrea Arcangeli <aarcange@redhat.com>
Reported-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The mechanism for detecting whether an inode should switch its wb
(bdi_writeback) association is now in place. This patch build the
framework for the actual switching.
This patch adds a new inode flag I_WB_SWITCHING, which has two
functions. First, the easy one, it ensures that there's only one
switching in progress for a give inode. Second, it's used as a
mechanism to synchronize wb stat updates.
The two stats, WB_RECLAIMABLE and WB_WRITEBACK, aren't event counters
but track the current number of dirty pages and pages under writeback
respectively. As such, when an inode is moved from one wb to another,
the inode's portion of those stats have to be transferred together;
unfortunately, this is a bit tricky as those stat updates are percpu
operations which are performed without holding any lock in some
places.
This patch solves the problem in a similar way as memcg. Each such
lockless stat updates are wrapped in transaction surrounded by
unlocked_inode_to_wb_begin/end(). During normal operation, they map
to rcu_read_lock/unlock(); however, if I_WB_SWITCHING is asserted,
mapping->tree_lock is grabbed across the transaction.
In turn, the switching path sets I_WB_SWITCHING and waits for a RCU
grace period to pass before actually starting to switch, which
guarantees that all stat update paths are synchronizing against
mapping->tree_lock.
This patch still doesn't implement the actual switching.
v3: Updated on top of the recent cancel_dirty_page() updates.
unlocked_inode_to_wb_begin() now nests inside
mem_cgroup_begin_page_stat() to match the locking order.
v2: The i_wb access transaction will be used for !stat accesses too.
Function names and comments updated accordingly.
s/inode_wb_stat_unlocked_{begin|end}/unlocked_inode_to_wb_{begin|end}/
s/switch_wb/switch_wbs/
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Jan Kara <jack@suse.cz>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Greg Thelen <gthelen@google.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
When modifying PG_Dirty on cached file pages, update the new
MEM_CGROUP_STAT_DIRTY counter. This is done in the same places where
global NR_FILE_DIRTY is managed. The new memcg stat is visible in the
per memcg memory.stat cgroupfs file. The most recent past attempt at
this was http://thread.gmane.org/gmane.linux.kernel.cgroups/8632
The new accounting supports future efforts to add per cgroup dirty
page throttling and writeback. It also helps an administrator break
down a container's memory usage and provides evidence to understand
memcg oom kills (the new dirty count is included in memcg oom kill
messages).
The ability to move page accounting between memcg
(memory.move_charge_at_immigrate) makes this accounting more
complicated than the global counter. The existing
mem_cgroup_{begin,end}_page_stat() lock is used to serialize move
accounting with stat updates.
Typical update operation:
memcg = mem_cgroup_begin_page_stat(page)
if (TestSetPageDirty()) {
[...]
mem_cgroup_update_page_stat(memcg)
}
mem_cgroup_end_page_stat(memcg)
Summary of mem_cgroup_end_page_stat() overhead:
- Without CONFIG_MEMCG it's a no-op
- With CONFIG_MEMCG and no inter memcg task movement, it's just
rcu_read_lock()
- With CONFIG_MEMCG and inter memcg task movement, it's
rcu_read_lock() + spin_lock_irqsave()
A memcg parameter is added to several routines because their callers
now grab mem_cgroup_begin_page_stat() which returns the memcg later
needed by for mem_cgroup_update_page_stat().
Because mem_cgroup_begin_page_stat() may disable interrupts, some
adjustments are needed:
- move __mark_inode_dirty() from __set_page_dirty() to its caller.
__mark_inode_dirty() locking does not want interrupts disabled.
- use spin_lock_irqsave(tree_lock) rather than spin_lock_irq() in
__delete_from_page_cache(), replace_page_cache_page(),
invalidate_complete_page2(), and __remove_mapping().
text data bss dec hex filename
8925147 1774832 1785856 12485835 be84cb vmlinux-!CONFIG_MEMCG-before
8925339 1774832 1785856 12486027 be858b vmlinux-!CONFIG_MEMCG-after
+192 text bytes
8965977 1784992 1785856 12536825 bf4bf9 vmlinux-CONFIG_MEMCG-before
8966750 1784992 1785856 12537598 bf4efe vmlinux-CONFIG_MEMCG-after
+773 text bytes
Performance tests run on v4.0-rc1-36-g4f671fe2f952. Lower is better for
all metrics, they're all wall clock or cycle counts. The read and write
fault benchmarks just measure fault time, they do not include I/O time.
* CONFIG_MEMCG not set:
baseline patched
kbuild 1m25.030000(+-0.088% 3 samples) 1m25.426667(+-0.120% 3 samples)
dd write 100 MiB 0.859211561 +-15.10% 0.874162885 +-15.03%
dd write 200 MiB 1.670653105 +-17.87% 1.669384764 +-11.99%
dd write 1000 MiB 8.434691190 +-14.15% 8.474733215 +-14.77%
read fault cycles 254.0(+-0.000% 10 samples) 253.0(+-0.000% 10 samples)
write fault cycles 2021.2(+-3.070% 10 samples) 1984.5(+-1.036% 10 samples)
* CONFIG_MEMCG=y root_memcg:
baseline patched
kbuild 1m25.716667(+-0.105% 3 samples) 1m25.686667(+-0.153% 3 samples)
dd write 100 MiB 0.855650830 +-14.90% 0.887557919 +-14.90%
dd write 200 MiB 1.688322953 +-12.72% 1.667682724 +-13.33%
dd write 1000 MiB 8.418601605 +-14.30% 8.673532299 +-15.00%
read fault cycles 266.0(+-0.000% 10 samples) 266.0(+-0.000% 10 samples)
write fault cycles 2051.7(+-1.349% 10 samples) 2049.6(+-1.686% 10 samples)
* CONFIG_MEMCG=y non-root_memcg:
baseline patched
kbuild 1m26.120000(+-0.273% 3 samples) 1m25.763333(+-0.127% 3 samples)
dd write 100 MiB 0.861723964 +-15.25% 0.818129350 +-14.82%
dd write 200 MiB 1.669887569 +-13.30% 1.698645885 +-13.27%
dd write 1000 MiB 8.383191730 +-14.65% 8.351742280 +-14.52%
read fault cycles 265.7(+-0.172% 10 samples) 267.0(+-0.000% 10 samples)
write fault cycles 2070.6(+-1.512% 10 samples) 2084.4(+-2.148% 10 samples)
As expected anon page faults are not affected by this patch.
tj: Updated to apply on top of the recent cancel_dirty_page() changes.
Signed-off-by: Sha Zhengju <handai.szj@gmail.com>
Signed-off-by: Greg Thelen <gthelen@google.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
