Waiting on a page migration entry has used wait_on_page_locked() all along
since 2006: but you cannot safely wait_on_page_locked() without holding a
reference to the page, and that extra reference is enough to make
migrate_page_move_mapping() fail with -EAGAIN, when a racing task faults
on the entry before migrate_page_move_mapping() gets there.
And that failure is retried nine times, amplifying the pain when trying to
migrate a popular page. With a single persistent faulter, migration
sometimes succeeds; with two or three concurrent faulters, success becomes
much less likely (and the more the page was mapped, the worse the overhead
of unmapping and remapping it on each try).
This is especially a problem for memory offlining, where the outer level
retries forever (or until terminated from userspace), because a heavy
refault workload can trigger an endless loop of migration failures.
wait_on_page_locked() is the wrong tool for the job.
David Herrmann (but was he the first?) noticed this issue in 2014:
https://marc.info/?l=linux-mm&m=140110465608116&w=2
Tim Chen started a thread in August 2017 which appears relevant:
https://marc.info/?l=linux-mm&m=150275941014915&w=2 where Kan Liang went
on to implicate __migration_entry_wait():
https://marc.info/?l=linux-mm&m=150300268411980&w=2 and the thread ended
up with the v4.14 commits: 2554db9165 ("sched/wait: Break up long wake
list walk") 11a19c7b09 ("sched/wait: Introduce wakeup boomark in
wake_up_page_bit")
Baoquan He reported "Memory hotplug softlock issue" 14 November 2018:
https://marc.info/?l=linux-mm&m=154217936431300&w=2
We have all assumed that it is essential to hold a page reference while
waiting on a page lock: partly to guarantee that there is still a struct
page when MEMORY_HOTREMOVE is configured, but also to protect against
reuse of the struct page going to someone who then holds the page locked
indefinitely, when the waiter can reasonably expect timely unlocking.
But in fact, so long as wait_on_page_bit_common() does the put_page(), and
is careful not to rely on struct page contents thereafter, there is no
need to hold a reference to the page while waiting on it. That does mean
that this case cannot go back through the loop: but that's fine for the
page migration case, and even if used more widely, is limited by the "Stop
walking if it's locked" optimization in wake_page_function().
Add interface put_and_wait_on_page_locked() to do this, using "behavior"
enum in place of "lock" arg to wait_on_page_bit_common() to implement it.
No interruptible or killable variant needed yet, but they might follow: I
have a vague notion that reporting -EINTR should take precedence over
return from wait_on_page_bit_common() without knowing the page state, so
arrange it accordingly - but that may be nothing but pedantic.
__migration_entry_wait() still has to take a brief reference to the page,
prior to calling put_and_wait_on_page_locked(): but now that it is dropped
before waiting, the chance of impeding page migration is very much
reduced. Should we perhaps disable preemption across this?
shrink_page_list()'s __ClearPageLocked(): that was a surprise! This
survived a lot of testing before that showed up. PageWaiters may have
been set by wait_on_page_bit_common(), and the reference dropped, just
before shrink_page_list() succeeds in freezing its last page reference: in
such a case, unlock_page() must be used. Follow the suggestion from
Michal Hocko, just revert a978d6f521 ("mm: unlockless reclaim") now:
that optimization predates PageWaiters, and won't buy much these days; but
we can reinstate it for the !PageWaiters case if anyone notices.
It does raise the question: should vmscan.c's is_page_cache_freeable() and
__remove_mapping() now treat a PageWaiters page as if an extra reference
were held? Perhaps, but I don't think it matters much, since
shrink_page_list() already had to win its trylock_page(), so waiters are
not very common there: I noticed no difference when trying the bigger
change, and it's surely not needed while put_and_wait_on_page_locked() is
only used for page migration.
[willy@infradead.org: add put_and_wait_on_page_locked() kerneldoc]
Link: http://lkml.kernel.org/r/alpine.LSU.2.11.1811261121330.1116@eggly.anvils
Signed-off-by: Hugh Dickins <hughd@google.com>
Reported-by: Baoquan He <bhe@redhat.com>
Tested-by: Baoquan He <bhe@redhat.com>
Reviewed-by: Andrea Arcangeli <aarcange@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Baoquan He <bhe@redhat.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: David Herrmann <dh.herrmann@gmail.com>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Nick Piggin <npiggin@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs dedup fixes from Dave Chinner:
"This reworks the vfs data cloning infrastructure.
We discovered many issues with these interfaces late in the 4.19 cycle
- the worst of them (data corruption, setuid stripping) were fixed for
XFS in 4.19-rc8, but a larger rework of the infrastructure fixing all
the problems was needed. That rework is the contents of this pull
request.
Rework the vfs_clone_file_range and vfs_dedupe_file_range
infrastructure to use a common .remap_file_range method and supply
generic bounds and sanity checking functions that are shared with the
data write path. The current VFS infrastructure has problems with
rlimit, LFS file sizes, file time stamps, maximum filesystem file
sizes, stripping setuid bits, etc and so they are addressed in these
commits.
We also introduce the ability for the ->remap_file_range methods to
return short clones so that clones for vfs_copy_file_range() don't get
rejected if the entire range can't be cloned. It also allows
filesystems to sliently skip deduplication of partial EOF blocks if
they are not capable of doing so without requiring errors to be thrown
to userspace.
Existing filesystems are converted to user the new remap_file_range
method, and both XFS and ocfs2 are modified to make use of the new
generic checking infrastructure"
* tag 'xfs-4.20-merge-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (28 commits)
xfs: remove [cm]time update from reflink calls
xfs: remove xfs_reflink_remap_range
xfs: remove redundant remap partial EOF block checks
xfs: support returning partial reflink results
xfs: clean up xfs_reflink_remap_blocks call site
xfs: fix pagecache truncation prior to reflink
ocfs2: remove ocfs2_reflink_remap_range
ocfs2: support partial clone range and dedupe range
ocfs2: fix pagecache truncation prior to reflink
ocfs2: truncate page cache for clone destination file before remapping
vfs: clean up generic_remap_file_range_prep return value
vfs: hide file range comparison function
vfs: enable remap callers that can handle short operations
vfs: plumb remap flags through the vfs dedupe functions
vfs: plumb remap flags through the vfs clone functions
vfs: make remap_file_range functions take and return bytes completed
vfs: remap helper should update destination inode metadata
vfs: pass remap flags to generic_remap_checks
vfs: pass remap flags to generic_remap_file_range_prep
vfs: combine the clone and dedupe into a single remap_file_range
...
Pull AFS updates from Al Viro:
"AFS series, with some iov_iter bits included"
* 'work.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (26 commits)
missing bits of "iov_iter: Separate type from direction and use accessor functions"
afs: Probe multiple fileservers simultaneously
afs: Fix callback handling
afs: Eliminate the address pointer from the address list cursor
afs: Allow dumping of server cursor on operation failure
afs: Implement YFS support in the fs client
afs: Expand data structure fields to support YFS
afs: Get the target vnode in afs_rmdir() and get a callback on it
afs: Calc callback expiry in op reply delivery
afs: Fix FS.FetchStatus delivery from updating wrong vnode
afs: Implement the YFS cache manager service
afs: Remove callback details from afs_callback_break struct
afs: Commit the status on a new file/dir/symlink
afs: Increase to 64-bit volume ID and 96-bit vnode ID for YFS
afs: Don't invoke the server to read data beyond EOF
afs: Add a couple of tracepoints to log I/O errors
afs: Handle EIO from delivery function
afs: Fix TTL on VL server and address lists
afs: Implement VL server rotation
afs: Improve FS server rotation error handling
...
Plumb in a remap flag that enables the filesystem remap handler to
shorten remapping requests for callers that can handle it. Now
copy_file_range can report partial success (in case we run up against
alignment problems, resource limits, etc.).
We also enable CAN_SHORTEN for fideduperange to maintain existing
userspace-visible behavior where xfs/btrfs shorten the dedupe range to
avoid stale post-eof data exposure.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Change the remap_file_range functions to take a number of bytes to
operate upon and return the number of bytes they operated on. This is a
requirement for allowing fs implementations to return short clone/dedupe
results to the user, which will enable us to obey resource limits in a
graceful manner.
A subsequent patch will enable copy_file_range to signal to the
->clone_file_range implementation that it can handle a short length,
which will be returned in the function's return value. For now the
short return is not implemented anywhere so the behavior won't change --
either copy_file_range manages to clone the entire range or it tries an
alternative.
Neither clone ioctl can take advantage of this, alas.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
File range remapping, if allowed to run past the destination file's EOF,
is an optimization on a regular file write. Regular file writes that
extend the file length are subject to various constraints which are not
checked by range cloning.
This is a correctness problem because we're never allowed to touch
ranges that the page cache can't support (s_maxbytes); we're not
supposed to deal with large offsets (MAX_NON_LFS) if O_LARGEFILE isn't
set; and we must obey resource limits (RLIMIT_FSIZE).
Therefore, add these checks to the new generic_remap_checks function so
that we curtail unexpected behavior.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Move the file range checks from vfs_clone_file_prep into a separate
generic_remap_checks function so that all the checks are collected in a
central location. This forms the basis for adding more checks from
generic_write_checks that will make cloning's input checking more
consistent with write input checking.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Pull XArray conversion from Matthew Wilcox:
"The XArray provides an improved interface to the radix tree data
structure, providing locking as part of the API, specifying GFP flags
at allocation time, eliminating preloading, less re-walking the tree,
more efficient iterations and not exposing RCU-protected pointers to
its users.
This patch set
1. Introduces the XArray implementation
2. Converts the pagecache to use it
3. Converts memremap to use it
The page cache is the most complex and important user of the radix
tree, so converting it was most important. Converting the memremap
code removes the only other user of the multiorder code, which allows
us to remove the radix tree code that supported it.
I have 40+ followup patches to convert many other users of the radix
tree over to the XArray, but I'd like to get this part in first. The
other conversions haven't been in linux-next and aren't suitable for
applying yet, but you can see them in the xarray-conv branch if you're
interested"
* 'xarray' of git://git.infradead.org/users/willy/linux-dax: (90 commits)
radix tree: Remove multiorder support
radix tree test: Convert multiorder tests to XArray
radix tree tests: Convert item_delete_rcu to XArray
radix tree tests: Convert item_kill_tree to XArray
radix tree tests: Move item_insert_order
radix tree test suite: Remove multiorder benchmarking
radix tree test suite: Remove __item_insert
memremap: Convert to XArray
xarray: Add range store functionality
xarray: Move multiorder_check to in-kernel tests
xarray: Move multiorder_shrink to kernel tests
xarray: Move multiorder account test in-kernel
radix tree test suite: Convert iteration test to XArray
radix tree test suite: Convert tag_tagged_items to XArray
radix tree: Remove radix_tree_clear_tags
radix tree: Remove radix_tree_maybe_preload_order
radix tree: Remove split/join code
radix tree: Remove radix_tree_update_node_t
page cache: Finish XArray conversion
dax: Convert page fault handlers to XArray
...
When systems are overcommitted and resources become contended, it's hard
to tell exactly the impact this has on workload productivity, or how close
the system is to lockups and OOM kills. In particular, when machines work
multiple jobs concurrently, the impact of overcommit in terms of latency
and throughput on the individual job can be enormous.
In order to maximize hardware utilization without sacrificing individual
job health or risk complete machine lockups, this patch implements a way
to quantify resource pressure in the system.
A kernel built with CONFIG_PSI=y creates files in /proc/pressure/ that
expose the percentage of time the system is stalled on CPU, memory, or IO,
respectively. Stall states are aggregate versions of the per-task delay
accounting delays:
cpu: some tasks are runnable but not executing on a CPU
memory: tasks are reclaiming, or waiting for swapin or thrashing cache
io: tasks are waiting for io completions
These percentages of walltime can be thought of as pressure percentages,
and they give a general sense of system health and productivity loss
incurred by resource overcommit. They can also indicate when the system
is approaching lockup scenarios and OOMs.
To do this, psi keeps track of the task states associated with each CPU
and samples the time they spend in stall states. Every 2 seconds, the
samples are averaged across CPUs - weighted by the CPUs' non-idle time to
eliminate artifacts from unused CPUs - and translated into percentages of
walltime. A running average of those percentages is maintained over 10s,
1m, and 5m periods (similar to the loadaverage).
[hannes@cmpxchg.org: doc fixlet, per Randy]
Link: http://lkml.kernel.org/r/20180828205625.GA14030@cmpxchg.org
[hannes@cmpxchg.org: code optimization]
Link: http://lkml.kernel.org/r/20180907175015.GA8479@cmpxchg.org
[hannes@cmpxchg.org: rename psi_clock() to psi_update_work(), per Peter]
Link: http://lkml.kernel.org/r/20180907145404.GB11088@cmpxchg.org
[hannes@cmpxchg.org: fix build]
Link: http://lkml.kernel.org/r/20180913014222.GA2370@cmpxchg.org
Link: http://lkml.kernel.org/r/20180828172258.3185-9-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Daniel Drake <drake@endlessm.com>
Tested-by: Suren Baghdasaryan <surenb@google.com>
Cc: Christopher Lameter <cl@linux.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Johannes Weiner <jweiner@fb.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Enderborg <peter.enderborg@sony.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vinayak Menon <vinmenon@codeaurora.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Delay accounting already measures the time a task spends in direct reclaim
and waiting for swapin, but in low memory situations tasks spend can spend
a significant amount of their time waiting on thrashing page cache. This
isn't tracked right now.
To know the full impact of memory contention on an individual task,
measure the delay when waiting for a recently evicted active cache page to
read back into memory.
Also update tools/accounting/getdelays.c:
[hannes@computer accounting]$ sudo ./getdelays -d -p 1
print delayacct stats ON
PID 1
CPU count real total virtual total delay total delay average
50318 745000000 847346785 400533713 0.008ms
IO count delay total delay average
435 122601218 0ms
SWAP count delay total delay average
0 0 0ms
RECLAIM count delay total delay average
0 0 0ms
THRASHING count delay total delay average
19 12621439 0ms
Link: http://lkml.kernel.org/r/20180828172258.3185-4-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Daniel Drake <drake@endlessm.com>
Tested-by: Suren Baghdasaryan <surenb@google.com>
Cc: Christopher Lameter <cl@linux.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Johannes Weiner <jweiner@fb.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Enderborg <peter.enderborg@sony.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vinayak Menon <vinmenon@codeaurora.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Refaults happen during transitions between workingsets as well as in-place
thrashing. Knowing the difference between the two has a range of
applications, including measuring the impact of memory shortage on the
system performance, as well as the ability to smarter balance pressure
between the filesystem cache and the swap-backed workingset.
During workingset transitions, inactive cache refaults and pushes out
established active cache. When that active cache isn't stale, however,
and also ends up refaulting, that's bonafide thrashing.
Introduce a new page flag that tells on eviction whether the page has been
active or not in its lifetime. This bit is then stored in the shadow
entry, to classify refaults as transitioning or thrashing.
How many page->flags does this leave us with on 32-bit?
20 bits are always page flags
21 if you have an MMU
23 with the zone bits for DMA, Normal, HighMem, Movable
29 with the sparsemem section bits
30 if PAE is enabled
31 with this patch.
So on 32-bit PAE, that leaves 1 bit for distinguishing two NUMA nodes. If
that's not enough, the system can switch to discontigmem and re-gain the 6
or 7 sparsemem section bits.
Link: http://lkml.kernel.org/r/20180828172258.3185-3-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Daniel Drake <drake@endlessm.com>
Tested-by: Suren Baghdasaryan <surenb@google.com>
Cc: Christopher Lameter <cl@linux.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Johannes Weiner <jweiner@fb.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Enderborg <peter.enderborg@sony.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vinayak Menon <vinmenon@codeaurora.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use accessor functions to access an iterator's type and direction. This
allows for the possibility of using some other method of determining the
type of iterator than if-chains with bitwise-AND conditions.
Signed-off-by: David Howells <dhowells@redhat.com>
Instead of calling find_get_pages_range() and putting any reference,
use xas_find() to iterate over any entries in the range, skipping the
shadow/swap entries.
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Slight change of strategy here; if we have trouble getting hold of a
page for whatever reason (eg a compound page is split underneath us),
don't spin to stabilise the page, just continue the iteration, like we
would if we failed to trylock the page. Since this is a speculative
optimisation, it feels like we should allow the process to take an extra
fault if it turns out to need this page instead of spending time to pin
down a page it may not need.
Signed-off-by: Matthew Wilcox <willy@infradead.org>
The 'end' parameter of the xas_for_each iterator avoids a useless
iteration at the end of the range.
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Davidlohr Bueso
Michal Hocko
Kirill Tkhai
Hugh Dickins
Linus Torvalds
Darrick J. Wong
Souptick Joarder
haiqing.shq
Johannes Weiner
David Howells
Matthew Wilcox