linux/mm/truncate.c

870 lines
26 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* mm/truncate.c - code for taking down pages from address_spaces
*
* Copyright (C) 2002, Linus Torvalds
*
* 10Sep2002 Andrew Morton
* Initial version.
*/
#include <linux/kernel.h>
#include <linux/backing-dev.h>
dax: support dirty DAX entries in radix tree Add support for tracking dirty DAX entries in the struct address_space radix tree. This tree is already used for dirty page writeback, and it already supports the use of exceptional (non struct page*) entries. In order to properly track dirty DAX pages we will insert new exceptional entries into the radix tree that represent dirty DAX PTE or PMD pages. These exceptional entries will also contain the writeback addresses for the PTE or PMD faults that we can use at fsync/msync time. There are currently two types of exceptional entries (shmem and shadow) that can be placed into the radix tree, and this adds a third. We rely on the fact that only one type of exceptional entry can be found in a given radix tree based on its usage. This happens for free with DAX vs shmem but we explicitly prevent shadow entries from being added to radix trees for DAX mappings. The only shadow entries that would be generated for DAX radix trees would be to track zero page mappings that were created for holes. These pages would receive minimal benefit from having shadow entries, and the choice to have only one type of exceptional entry in a given radix tree makes the logic simpler both in clear_exceptional_entry() and in the rest of DAX. Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Dave Chinner <david@fromorbit.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jeff Layton <jlayton@poochiereds.net> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Matthew Wilcox <matthew.r.wilcox@intel.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-01-22 23:10:40 +00:00
#include <linux/dax.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/gfp.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/export.h>
#include <linux/pagemap.h>
#include <linux/highmem.h>
#include <linux/pagevec.h>
#include <linux/task_io_accounting_ops.h>
#include <linux/buffer_head.h> /* grr. try_to_release_page */
#include <linux/shmem_fs.h>
#include <linux/rmap.h>
#include "internal.h"
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
/*
* Regular page slots are stabilized by the page lock even without the tree
* itself locked. These unlocked entries need verification under the tree
* lock.
*/
static inline void __clear_shadow_entry(struct address_space *mapping,
pgoff_t index, void *entry)
2014-04-03 21:47:46 +00:00
{
XA_STATE(xas, &mapping->i_pages, index);
mm: keep page cache radix tree nodes in check Previously, page cache radix tree nodes were freed after reclaim emptied out their page pointers. But now reclaim stores shadow entries in their place, which are only reclaimed when the inodes themselves are reclaimed. This is problematic for bigger files that are still in use after they have a significant amount of their cache reclaimed, without any of those pages actually refaulting. The shadow entries will just sit there and waste memory. In the worst case, the shadow entries will accumulate until the machine runs out of memory. To get this under control, the VM will track radix tree nodes exclusively containing shadow entries on a per-NUMA node list. Per-NUMA rather than global because we expect the radix tree nodes themselves to be allocated node-locally and we want to reduce cross-node references of otherwise independent cache workloads. A simple shrinker will then reclaim these nodes on memory pressure. A few things need to be stored in the radix tree node to implement the shadow node LRU and allow tree deletions coming from the list: 1. There is no index available that would describe the reverse path from the node up to the tree root, which is needed to perform a deletion. To solve this, encode in each node its offset inside the parent. This can be stored in the unused upper bits of the same member that stores the node's height at no extra space cost. 2. The number of shadow entries needs to be counted in addition to the regular entries, to quickly detect when the node is ready to go to the shadow node LRU list. The current entry count is an unsigned int but the maximum number of entries is 64, so a shadow counter can easily be stored in the unused upper bits. 3. Tree modification needs tree lock and tree root, which are located in the address space, so store an address_space backpointer in the node. The parent pointer of the node is in a union with the 2-word rcu_head, so the backpointer comes at no extra cost as well. 4. The node needs to be linked to an LRU list, which requires a list head inside the node. This does increase the size of the node, but it does not change the number of objects that fit into a slab page. [akpm@linux-foundation.org: export the right function] Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Greg Thelen <gthelen@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Metin Doslu <metin@citusdata.com> Cc: Michel Lespinasse <walken@google.com> Cc: Ozgun Erdogan <ozgun@citusdata.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roman Gushchin <klamm@yandex-team.ru> Cc: Ryan Mallon <rmallon@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:47:56 +00:00
xas_set_update(&xas, workingset_update_node);
if (xas_load(&xas) != entry)
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
return;
xas_store(&xas, NULL);
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
}
static void clear_shadow_entry(struct address_space *mapping, pgoff_t index,
void *entry)
{
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
spin_lock(&mapping->host->i_lock);
xa_lock_irq(&mapping->i_pages);
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
__clear_shadow_entry(mapping, index, entry);
xa_unlock_irq(&mapping->i_pages);
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
if (mapping_shrinkable(mapping))
inode_add_lru(mapping->host);
spin_unlock(&mapping->host->i_lock);
2014-04-03 21:47:46 +00:00
}
/*
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
* Unconditionally remove exceptional entries. Usually called from truncate
* path. Note that the folio_batch may be altered by this function by removing
* exceptional entries similar to what folio_batch_remove_exceptionals() does.
*/
static void truncate_folio_batch_exceptionals(struct address_space *mapping,
struct folio_batch *fbatch, pgoff_t *indices)
{
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
int i, j;
bool dax;
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
/* Handled by shmem itself */
if (shmem_mapping(mapping))
return;
for (j = 0; j < folio_batch_count(fbatch); j++)
if (xa_is_value(fbatch->folios[j]))
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
break;
if (j == folio_batch_count(fbatch))
return;
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
dax = dax_mapping(mapping);
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
if (!dax) {
spin_lock(&mapping->host->i_lock);
xa_lock_irq(&mapping->i_pages);
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
}
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
for (i = j; i < folio_batch_count(fbatch); i++) {
struct folio *folio = fbatch->folios[i];
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
pgoff_t index = indices[i];
if (!xa_is_value(folio)) {
fbatch->folios[j++] = folio;
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
continue;
}
if (unlikely(dax)) {
dax_delete_mapping_entry(mapping, index);
continue;
}
__clear_shadow_entry(mapping, index, folio);
}
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
if (!dax) {
xa_unlock_irq(&mapping->i_pages);
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
if (mapping_shrinkable(mapping))
inode_add_lru(mapping->host);
spin_unlock(&mapping->host->i_lock);
}
fbatch->nr = j;
}
/*
* Invalidate exceptional entry if easily possible. This handles exceptional
dax: prevent invalidation of mapped DAX entries Patch series "mm,dax: Fix data corruption due to mmap inconsistency", v4. This series fixes data corruption that can happen for DAX mounts when page faults race with write(2) and as a result page tables get out of sync with block mappings in the filesystem and thus data seen through mmap is different from data seen through read(2). The series passes testing with t_mmap_stale test program from Ross and also other mmap related tests on DAX filesystem. This patch (of 4): dax_invalidate_mapping_entry() currently removes DAX exceptional entries only if they are clean and unlocked. This is done via: invalidate_mapping_pages() invalidate_exceptional_entry() dax_invalidate_mapping_entry() However, for page cache pages removed in invalidate_mapping_pages() there is an additional criteria which is that the page must not be mapped. This is noted in the comments above invalidate_mapping_pages() and is checked in invalidate_inode_page(). For DAX entries this means that we can can end up in a situation where a DAX exceptional entry, either a huge zero page or a regular DAX entry, could end up mapped but without an associated radix tree entry. This is inconsistent with the rest of the DAX code and with what happens in the page cache case. We aren't able to unmap the DAX exceptional entry because according to its comments invalidate_mapping_pages() isn't allowed to block, and unmap_mapping_range() takes a write lock on the mapping->i_mmap_rwsem. Since we essentially never have unmapped DAX entries to evict from the radix tree, just remove dax_invalidate_mapping_entry(). Fixes: c6dcf52c23d2 ("mm: Invalidate DAX radix tree entries only if appropriate") Link: http://lkml.kernel.org/r/20170510085419.27601-2-jack@suse.cz Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Jan Kara <jack@suse.cz> Reported-by: Jan Kara <jack@suse.cz> Cc: Dan Williams <dan.j.williams@intel.com> Cc: <stable@vger.kernel.org> [4.10+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-05-12 22:46:47 +00:00
* entries for invalidate_inode_pages().
*/
static int invalidate_exceptional_entry(struct address_space *mapping,
pgoff_t index, void *entry)
{
dax: prevent invalidation of mapped DAX entries Patch series "mm,dax: Fix data corruption due to mmap inconsistency", v4. This series fixes data corruption that can happen for DAX mounts when page faults race with write(2) and as a result page tables get out of sync with block mappings in the filesystem and thus data seen through mmap is different from data seen through read(2). The series passes testing with t_mmap_stale test program from Ross and also other mmap related tests on DAX filesystem. This patch (of 4): dax_invalidate_mapping_entry() currently removes DAX exceptional entries only if they are clean and unlocked. This is done via: invalidate_mapping_pages() invalidate_exceptional_entry() dax_invalidate_mapping_entry() However, for page cache pages removed in invalidate_mapping_pages() there is an additional criteria which is that the page must not be mapped. This is noted in the comments above invalidate_mapping_pages() and is checked in invalidate_inode_page(). For DAX entries this means that we can can end up in a situation where a DAX exceptional entry, either a huge zero page or a regular DAX entry, could end up mapped but without an associated radix tree entry. This is inconsistent with the rest of the DAX code and with what happens in the page cache case. We aren't able to unmap the DAX exceptional entry because according to its comments invalidate_mapping_pages() isn't allowed to block, and unmap_mapping_range() takes a write lock on the mapping->i_mmap_rwsem. Since we essentially never have unmapped DAX entries to evict from the radix tree, just remove dax_invalidate_mapping_entry(). Fixes: c6dcf52c23d2 ("mm: Invalidate DAX radix tree entries only if appropriate") Link: http://lkml.kernel.org/r/20170510085419.27601-2-jack@suse.cz Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Jan Kara <jack@suse.cz> Reported-by: Jan Kara <jack@suse.cz> Cc: Dan Williams <dan.j.williams@intel.com> Cc: <stable@vger.kernel.org> [4.10+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-05-12 22:46:47 +00:00
/* Handled by shmem itself, or for DAX we do nothing. */
if (shmem_mapping(mapping) || dax_mapping(mapping))
return 1;
clear_shadow_entry(mapping, index, entry);
return 1;
}
/*
* Invalidate exceptional entry if clean. This handles exceptional entries for
* invalidate_inode_pages2() so for DAX it evicts only clean entries.
*/
static int invalidate_exceptional_entry2(struct address_space *mapping,
pgoff_t index, void *entry)
{
/* Handled by shmem itself */
if (shmem_mapping(mapping))
return 1;
if (dax_mapping(mapping))
return dax_invalidate_mapping_entry_sync(mapping, index);
clear_shadow_entry(mapping, index, entry);
return 1;
}
/**
* folio_invalidate - Invalidate part or all of a folio.
* @folio: The folio which is affected.
* @offset: start of the range to invalidate
* @length: length of the range to invalidate
*
* folio_invalidate() is called when all or part of the folio has become
* invalidated by a truncate operation.
*
* folio_invalidate() does not have to release all buffers, but it must
* ensure that no dirty buffer is left outside @offset and that no I/O
* is underway against any of the blocks which are outside the truncation
* point. Because the caller is about to free (and possibly reuse) those
* blocks on-disk.
*/
void folio_invalidate(struct folio *folio, size_t offset, size_t length)
{
const struct address_space_operations *aops = folio->mapping->a_ops;
if (aops->invalidate_folio)
aops->invalidate_folio(folio, offset, length);
}
EXPORT_SYMBOL_GPL(folio_invalidate);
/*
* If truncate cannot remove the fs-private metadata from the page, the page
* becomes orphaned. It will be left on the LRU and may even be mapped into
mm: merge populate and nopage into fault (fixes nonlinear) Nonlinear mappings are (AFAIKS) simply a virtual memory concept that encodes the virtual address -> file offset differently from linear mappings. ->populate is a layering violation because the filesystem/pagecache code should need to know anything about the virtual memory mapping. The hitch here is that the ->nopage handler didn't pass down enough information (ie. pgoff). But it is more logical to pass pgoff rather than have the ->nopage function calculate it itself anyway (because that's a similar layering violation). Having the populate handler install the pte itself is likewise a nasty thing to be doing. This patch introduces a new fault handler that replaces ->nopage and ->populate and (later) ->nopfn. Most of the old mechanism is still in place so there is a lot of duplication and nice cleanups that can be removed if everyone switches over. The rationale for doing this in the first place is that nonlinear mappings are subject to the pagefault vs invalidate/truncate race too, and it seemed stupid to duplicate the synchronisation logic rather than just consolidate the two. After this patch, MAP_NONBLOCK no longer sets up ptes for pages present in pagecache. Seems like a fringe functionality anyway. NOPAGE_REFAULT is removed. This should be implemented with ->fault, and no users have hit mainline yet. [akpm@linux-foundation.org: cleanup] [randy.dunlap@oracle.com: doc. fixes for readahead] [akpm@linux-foundation.org: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Cc: Mark Fasheh <mark.fasheh@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 08:46:59 +00:00
* user pagetables if we're racing with filemap_fault().
*
* We need to bail out if page->mapping is no longer equal to the original
* mapping. This happens a) when the VM reclaimed the page while we waited on
* its lock, b) when a concurrent invalidate_mapping_pages got there first and
* c) when tmpfs swizzles a page between a tmpfs inode and swapper_space.
*/
static void truncate_cleanup_folio(struct folio *folio)
{
if (folio_mapped(folio))
unmap_mapping_folio(folio);
if (folio_has_private(folio))
folio_invalidate(folio, 0, folio_size(folio));
page_writeback: clean up mess around cancel_dirty_page() This patch replaces cancel_dirty_page() with a helper function account_page_cleaned() which only updates counters. It's called from truncate_complete_page() and from try_to_free_buffers() (hack for ext3). Page is locked in both cases, page-lock protects against concurrent dirtiers: see commit 2d6d7f982846 ("mm: protect set_page_dirty() from ongoing truncation"). Delete_from_page_cache() shouldn't be called for dirty pages, they must be handled by caller (either written or truncated). This patch treats final dirty accounting fixup at the end of __delete_from_page_cache() as a debug check and adds WARN_ON_ONCE() around it. If something removes dirty pages without proper handling that might be a bug and unwritten data might be lost. Hugetlbfs has no dirty pages accounting, ClearPageDirty() is enough here. cancel_dirty_page() in nfs_wb_page_cancel() is redundant. This is helper for nfs_invalidate_page() and it's called only in case complete invalidation. The mess was started in v2.6.20 after commits 46d2277c796f ("Clean up and make try_to_free_buffers() not race with dirty pages") and 3e67c0987d75 ("truncate: clear page dirtiness before running try_to_free_buffers()") first was reverted right in v2.6.20 in commit ecdfc9787fe5 ("Resurrect 'try_to_free_buffers()' VM hackery"), second in v2.6.25 commit a2b345642f53 ("Fix dirty page accounting leak with ext3 data=journal"). Custom fixes were introduced between these points. NFS in v2.6.23, commit 1b3b4a1a2deb ("NFS: Fix a write request leak in nfs_invalidate_page()"). Kludge in __delete_from_page_cache() in v2.6.24, commit 3a6927906f1b ("Do dirty page accounting when removing a page from the page cache"). Since v2.6.25 all of them are redundant. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Cc: Tejun Heo <tj@kernel.org> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-04-14 22:45:27 +00:00
/*
* Some filesystems seem to re-dirty the page even after
* the VM has canceled the dirty bit (eg ext3 journaling).
* Hence dirty accounting check is placed after invalidation.
*/
folio_cancel_dirty(folio);
folio_clear_mappedtodisk(folio);
}
int truncate_inode_folio(struct address_space *mapping, struct folio *folio)
{
if (folio->mapping != mapping)
return -EIO;
truncate_cleanup_folio(folio);
filemap_remove_folio(folio);
return 0;
}
/*
* Handle partial folios. The folio may be entirely within the
* range if a split has raced with us. If not, we zero the part of the
* folio that's within the [start, end] range, and then split the folio if
* it's large. split_page_range() will discard pages which now lie beyond
* i_size, and we rely on the caller to discard pages which lie within a
* newly created hole.
*
* Returns false if splitting failed so the caller can avoid
* discarding the entire folio which is stubbornly unsplit.
*/
bool truncate_inode_partial_folio(struct folio *folio, loff_t start, loff_t end)
{
loff_t pos = folio_pos(folio);
unsigned int offset, length;
if (pos < start)
offset = start - pos;
else
offset = 0;
length = folio_size(folio);
if (pos + length <= (u64)end)
length = length - offset;
else
length = end + 1 - pos - offset;
folio_wait_writeback(folio);
if (length == folio_size(folio)) {
truncate_inode_folio(folio->mapping, folio);
return true;
}
/*
* We may be zeroing pages we're about to discard, but it avoids
* doing a complex calculation here, and then doing the zeroing
* anyway if the page split fails.
*/
folio_zero_range(folio, offset, length);
if (folio_has_private(folio))
folio_invalidate(folio, offset, length);
if (!folio_test_large(folio))
return true;
if (split_folio(folio) == 0)
return true;
if (folio_test_dirty(folio))
return false;
truncate_inode_folio(folio->mapping, folio);
return true;
}
/*
* Used to get rid of pages on hardware memory corruption.
*/
int generic_error_remove_page(struct address_space *mapping, struct page *page)
{
VM_BUG_ON_PAGE(PageTail(page), page);
if (!mapping)
return -EINVAL;
/*
* Only punch for normal data pages for now.
* Handling other types like directories would need more auditing.
*/
if (!S_ISREG(mapping->host->i_mode))
return -EIO;
return truncate_inode_folio(mapping, page_folio(page));
}
EXPORT_SYMBOL(generic_error_remove_page);
static long mapping_evict_folio(struct address_space *mapping,
struct folio *folio)
{
if (folio_test_dirty(folio) || folio_test_writeback(folio))
return 0;
/* The refcount will be elevated if any page in the folio is mapped */
if (folio_ref_count(folio) >
folio_nr_pages(folio) + folio_has_private(folio) + 1)
return 0;
if (folio_has_private(folio) && !filemap_release_folio(folio, 0))
return 0;
return remove_mapping(mapping, folio);
}
/**
* invalidate_inode_page() - Remove an unused page from the pagecache.
* @page: The page to remove.
*
* Safely invalidate one page from its pagecache mapping.
* It only drops clean, unused pages.
*
* Context: Page must be locked.
* Return: The number of pages successfully removed.
*/
long invalidate_inode_page(struct page *page)
{
struct folio *folio = page_folio(page);
struct address_space *mapping = folio_mapping(folio);
/* The page may have been truncated before it was locked */
if (!mapping)
return 0;
return mapping_evict_folio(mapping, folio);
}
/**
* truncate_inode_pages_range - truncate range of pages specified by start & end byte offsets
* @mapping: mapping to truncate
* @lstart: offset from which to truncate
* @lend: offset to which to truncate (inclusive)
*
* Truncate the page cache, removing the pages that are between
* specified offsets (and zeroing out partial pages
* if lstart or lend + 1 is not page aligned).
*
* Truncate takes two passes - the first pass is nonblocking. It will not
* block on page locks and it will not block on writeback. The second pass
* will wait. This is to prevent as much IO as possible in the affected region.
* The first pass will remove most pages, so the search cost of the second pass
* is low.
*
* We pass down the cache-hot hint to the page freeing code. Even if the
* mapping is large, it is probably the case that the final pages are the most
* recently touched, and freeing happens in ascending file offset order.
*
* Note that since ->invalidate_folio() accepts range to invalidate
* truncate_inode_pages_range is able to handle cases where lend + 1 is not
* page aligned properly.
*/
void truncate_inode_pages_range(struct address_space *mapping,
loff_t lstart, loff_t lend)
{
pgoff_t start; /* inclusive */
pgoff_t end; /* exclusive */
struct folio_batch fbatch;
2014-04-03 21:47:46 +00:00
pgoff_t indices[PAGEVEC_SIZE];
pgoff_t index;
int i;
struct folio *folio;
bool same_folio;
if (mapping_empty(mapping))
return;
/*
* 'start' and 'end' always covers the range of pages to be fully
* truncated. Partial pages are covered with 'partial_start' at the
* start of the range and 'partial_end' at the end of the range.
* Note that 'end' is exclusive while 'lend' is inclusive.
*/
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
start = (lstart + PAGE_SIZE - 1) >> PAGE_SHIFT;
if (lend == -1)
/*
* lend == -1 indicates end-of-file so we have to set 'end'
* to the highest possible pgoff_t and since the type is
* unsigned we're using -1.
*/
end = -1;
else
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
end = (lend + 1) >> PAGE_SHIFT;
folio_batch_init(&fbatch);
index = start;
while (index < end && find_lock_entries(mapping, index, end - 1,
&fbatch, indices)) {
index = indices[folio_batch_count(&fbatch) - 1] + 1;
truncate_folio_batch_exceptionals(mapping, &fbatch, indices);
for (i = 0; i < folio_batch_count(&fbatch); i++)
truncate_cleanup_folio(fbatch.folios[i]);
delete_from_page_cache_batch(mapping, &fbatch);
for (i = 0; i < folio_batch_count(&fbatch); i++)
folio_unlock(fbatch.folios[i]);
folio_batch_release(&fbatch);
cond_resched();
}
same_folio = (lstart >> PAGE_SHIFT) == (lend >> PAGE_SHIFT);
folio = __filemap_get_folio(mapping, lstart >> PAGE_SHIFT, FGP_LOCK, 0);
if (folio) {
same_folio = lend < folio_pos(folio) + folio_size(folio);
if (!truncate_inode_partial_folio(folio, lstart, lend)) {
start = folio->index + folio_nr_pages(folio);
if (same_folio)
end = folio->index;
}
folio_unlock(folio);
folio_put(folio);
folio = NULL;
}
if (!same_folio)
folio = __filemap_get_folio(mapping, lend >> PAGE_SHIFT,
FGP_LOCK, 0);
if (folio) {
if (!truncate_inode_partial_folio(folio, lstart, lend))
end = folio->index;
folio_unlock(folio);
folio_put(folio);
}
index = start;
while (index < end) {
cond_resched();
if (!find_get_entries(mapping, index, end - 1, &fbatch,
indices)) {
mm/fs: fix pessimization in hole-punching pagecache I wanted to revert my v3.1 commit d0823576bf4b ("mm: pincer in truncate_inode_pages_range"), to keep truncate_inode_pages_range() in synch with shmem_undo_range(); but have stepped back - a change to hole-punching in truncate_inode_pages_range() is a change to hole-punching in every filesystem (except tmpfs) that supports it. If there's a logical proof why no filesystem can depend for its own correctness on the pincer guarantee in truncate_inode_pages_range() - an instant when the entire hole is removed from pagecache - then let's revisit later. But the evidence is that only tmpfs suffered from the livelock, and we have no intention of extending hole-punch to ramfs. So for now just add a few comments (to match or differ from those in shmem_undo_range()), and fix one silliness noticed in d0823576bf4b... Its "index == start" addition to the hole-punch termination test was incomplete: it opened a way for the end condition to be missed, and the loop go on looking through the radix_tree, all the way to end of file. Fix that pessimization by resetting index when detected in inner loop. Note that it's actually hard to hit this case, without the obsessive concurrent faulting that trinity does: normally all pages are removed in the initial trylock_page() pass, and this loop finds nothing to do. I had to "#if 0" out the initial pass to reproduce bug and test fix. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Sasha Levin <sasha.levin@oracle.com> Cc: Konstantin Khlebnikov <koct9i@gmail.com> Cc: Lukas Czerner <lczerner@redhat.com> Cc: Dave Jones <davej@redhat.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-07-23 21:00:15 +00:00
/* If all gone from start onwards, we're done */
if (index == start)
break;
mm/fs: fix pessimization in hole-punching pagecache I wanted to revert my v3.1 commit d0823576bf4b ("mm: pincer in truncate_inode_pages_range"), to keep truncate_inode_pages_range() in synch with shmem_undo_range(); but have stepped back - a change to hole-punching in truncate_inode_pages_range() is a change to hole-punching in every filesystem (except tmpfs) that supports it. If there's a logical proof why no filesystem can depend for its own correctness on the pincer guarantee in truncate_inode_pages_range() - an instant when the entire hole is removed from pagecache - then let's revisit later. But the evidence is that only tmpfs suffered from the livelock, and we have no intention of extending hole-punch to ramfs. So for now just add a few comments (to match or differ from those in shmem_undo_range()), and fix one silliness noticed in d0823576bf4b... Its "index == start" addition to the hole-punch termination test was incomplete: it opened a way for the end condition to be missed, and the loop go on looking through the radix_tree, all the way to end of file. Fix that pessimization by resetting index when detected in inner loop. Note that it's actually hard to hit this case, without the obsessive concurrent faulting that trinity does: normally all pages are removed in the initial trylock_page() pass, and this loop finds nothing to do. I had to "#if 0" out the initial pass to reproduce bug and test fix. Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Sasha Levin <sasha.levin@oracle.com> Cc: Konstantin Khlebnikov <koct9i@gmail.com> Cc: Lukas Czerner <lczerner@redhat.com> Cc: Dave Jones <davej@redhat.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-07-23 21:00:15 +00:00
/* Otherwise restart to make sure all gone */
index = start;
continue;
}
mm, truncate: remove all exceptional entries from pagevec under one lock During truncate each entry in a pagevec is checked to see if it is an exceptional entry and if so, the shadow entry is cleaned up. This is potentially expensive as multiple entries for a mapping locks/unlocks the tree lock. This batches the operation such that any exceptional entries removed from a pagevec only acquire the mapping tree lock once. The corner case where this is more expensive is where there is only one exceptional entry but this is unlikely due to temporal locality and how it affects LRU ordering. Note that for truncations of small files created recently, this patch should show no gain because it only batches the handling of exceptional entries. sparsetruncate (large) 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Min Time 38.00 ( 0.00%) 27.00 ( 28.95%) 1st-qrtle Time 40.00 ( 0.00%) 28.00 ( 30.00%) 2nd-qrtle Time 44.00 ( 0.00%) 41.00 ( 6.82%) 3rd-qrtle Time 146.00 ( 0.00%) 147.00 ( -0.68%) Max-90% Time 153.00 ( 0.00%) 153.00 ( 0.00%) Max-95% Time 155.00 ( 0.00%) 156.00 ( -0.65%) Max-99% Time 181.00 ( 0.00%) 171.00 ( 5.52%) Amean Time 93.04 ( 0.00%) 88.43 ( 4.96%) Best99%Amean Time 92.08 ( 0.00%) 86.13 ( 6.46%) Best95%Amean Time 89.19 ( 0.00%) 83.13 ( 6.80%) Best90%Amean Time 85.60 ( 0.00%) 79.15 ( 7.53%) Best75%Amean Time 72.95 ( 0.00%) 65.09 ( 10.78%) Best50%Amean Time 39.86 ( 0.00%) 28.20 ( 29.25%) Best25%Amean Time 39.44 ( 0.00%) 27.70 ( 29.77%) bonnie 4.14.0-rc4 4.14.0-rc4 pickhelper-v1r1 batchshadow-v1r1 Hmean SeqCreate ops 71.92 ( 0.00%) 76.78 ( 6.76%) Hmean SeqCreate read 42.42 ( 0.00%) 45.01 ( 6.10%) Hmean SeqCreate del 26519.88 ( 0.00%) 27191.87 ( 2.53%) Hmean RandCreate ops 71.92 ( 0.00%) 76.95 ( 7.00%) Hmean RandCreate read 44.44 ( 0.00%) 49.23 ( 10.78%) Hmean RandCreate del 24948.62 ( 0.00%) 24764.97 ( -0.74%) Truncation of a large number of files shows a substantial gain with 99% of files being truncated 6.46% faster. bonnie shows a modest gain of 2.53% [jack@suse.cz: fix truncate_exceptional_pvec_entries()] Link: http://lkml.kernel.org/r/20171108164226.26788-1-jack@suse.cz Link: http://lkml.kernel.org/r/20171018075952.10627-4-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 01:37:44 +00:00
for (i = 0; i < folio_batch_count(&fbatch); i++) {
struct folio *folio = fbatch.folios[i];
/* We rely upon deletion not changing page->index */
2014-04-03 21:47:46 +00:00
index = indices[i];
if (xa_is_value(folio))
2014-04-03 21:47:46 +00:00
continue;
folio_lock(folio);
VM_BUG_ON_FOLIO(!folio_contains(folio, index), folio);
folio_wait_writeback(folio);
truncate_inode_folio(mapping, folio);
folio_unlock(folio);
index = folio_index(folio) + folio_nr_pages(folio) - 1;
}
truncate_folio_batch_exceptionals(mapping, &fbatch, indices);
folio_batch_release(&fbatch);
index++;
}
}
EXPORT_SYMBOL(truncate_inode_pages_range);
/**
* truncate_inode_pages - truncate *all* the pages from an offset
* @mapping: mapping to truncate
* @lstart: offset from which to truncate
*
* Called under (and serialised by) inode->i_rwsem and
* mapping->invalidate_lock.
*
* Note: When this function returns, there can be a page in the process of
* deletion (inside __filemap_remove_folio()) in the specified range. Thus
* mapping->nrpages can be non-zero when this function returns even after
* truncation of the whole mapping.
*/
void truncate_inode_pages(struct address_space *mapping, loff_t lstart)
{
truncate_inode_pages_range(mapping, lstart, (loff_t)-1);
}
EXPORT_SYMBOL(truncate_inode_pages);
mm + fs: store shadow entries in page cache Reclaim will be leaving shadow entries in the page cache radix tree upon evicting the real page. As those pages are found from the LRU, an iput() can lead to the inode being freed concurrently. At this point, reclaim must no longer install shadow pages because the inode freeing code needs to ensure the page tree is really empty. Add an address_space flag, AS_EXITING, that the inode freeing code sets under the tree lock before doing the final truncate. Reclaim will check for this flag before installing shadow pages. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Greg Thelen <gthelen@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Metin Doslu <metin@citusdata.com> Cc: Michel Lespinasse <walken@google.com> Cc: Ozgun Erdogan <ozgun@citusdata.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roman Gushchin <klamm@yandex-team.ru> Cc: Ryan Mallon <rmallon@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:47:49 +00:00
/**
* truncate_inode_pages_final - truncate *all* pages before inode dies
* @mapping: mapping to truncate
*
* Called under (and serialized by) inode->i_rwsem.
mm + fs: store shadow entries in page cache Reclaim will be leaving shadow entries in the page cache radix tree upon evicting the real page. As those pages are found from the LRU, an iput() can lead to the inode being freed concurrently. At this point, reclaim must no longer install shadow pages because the inode freeing code needs to ensure the page tree is really empty. Add an address_space flag, AS_EXITING, that the inode freeing code sets under the tree lock before doing the final truncate. Reclaim will check for this flag before installing shadow pages. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Greg Thelen <gthelen@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Metin Doslu <metin@citusdata.com> Cc: Michel Lespinasse <walken@google.com> Cc: Ozgun Erdogan <ozgun@citusdata.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roman Gushchin <klamm@yandex-team.ru> Cc: Ryan Mallon <rmallon@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:47:49 +00:00
*
* Filesystems have to use this in the .evict_inode path to inform the
* VM that this is the final truncate and the inode is going away.
*/
void truncate_inode_pages_final(struct address_space *mapping)
{
/*
* Page reclaim can not participate in regular inode lifetime
* management (can't call iput()) and thus can race with the
* inode teardown. Tell it when the address space is exiting,
* so that it does not install eviction information after the
* final truncate has begun.
*/
mapping_set_exiting(mapping);
if (!mapping_empty(mapping)) {
mm + fs: store shadow entries in page cache Reclaim will be leaving shadow entries in the page cache radix tree upon evicting the real page. As those pages are found from the LRU, an iput() can lead to the inode being freed concurrently. At this point, reclaim must no longer install shadow pages because the inode freeing code needs to ensure the page tree is really empty. Add an address_space flag, AS_EXITING, that the inode freeing code sets under the tree lock before doing the final truncate. Reclaim will check for this flag before installing shadow pages. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Greg Thelen <gthelen@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Metin Doslu <metin@citusdata.com> Cc: Michel Lespinasse <walken@google.com> Cc: Ozgun Erdogan <ozgun@citusdata.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roman Gushchin <klamm@yandex-team.ru> Cc: Ryan Mallon <rmallon@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:47:49 +00:00
/*
* As truncation uses a lockless tree lookup, cycle
* the tree lock to make sure any ongoing tree
* modification that does not see AS_EXITING is
* completed before starting the final truncate.
*/
xa_lock_irq(&mapping->i_pages);
xa_unlock_irq(&mapping->i_pages);
mm + fs: store shadow entries in page cache Reclaim will be leaving shadow entries in the page cache radix tree upon evicting the real page. As those pages are found from the LRU, an iput() can lead to the inode being freed concurrently. At this point, reclaim must no longer install shadow pages because the inode freeing code needs to ensure the page tree is really empty. Add an address_space flag, AS_EXITING, that the inode freeing code sets under the tree lock before doing the final truncate. Reclaim will check for this flag before installing shadow pages. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Greg Thelen <gthelen@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Metin Doslu <metin@citusdata.com> Cc: Michel Lespinasse <walken@google.com> Cc: Ozgun Erdogan <ozgun@citusdata.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roman Gushchin <klamm@yandex-team.ru> Cc: Ryan Mallon <rmallon@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:47:49 +00:00
}
mm: cleancache: fix corruption on missed inode invalidation If all pages are deleted from the mapping by memory reclaim and also moved to the cleancache: __delete_from_page_cache (no shadow case) unaccount_page_cache_page cleancache_put_page page_cache_delete mapping->nrpages -= nr (nrpages becomes 0) We don't clean the cleancache for an inode after final file truncation (removal). truncate_inode_pages_final check (nrpages || nrexceptional) is false no truncate_inode_pages no cleancache_invalidate_inode(mapping) These way when reading the new file created with same inode we may get these trash leftover pages from cleancache and see wrong data instead of the contents of the new file. Fix it by always doing truncate_inode_pages which is already ready for nrpages == 0 && nrexceptional == 0 case and just invalidates inode. [akpm@linux-foundation.org: add comment, per Jan] Link: http://lkml.kernel.org/r/20181112095734.17979-1-ptikhomirov@virtuozzo.com Fixes: commit 91b0abe36a7b ("mm + fs: store shadow entries in page cache") Signed-off-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com> Reviewed-by: Vasily Averin <vvs@virtuozzo.com> Reviewed-by: Andrey Ryabinin <aryabinin@virtuozzo.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-11-30 22:09:00 +00:00
truncate_inode_pages(mapping, 0);
mm + fs: store shadow entries in page cache Reclaim will be leaving shadow entries in the page cache radix tree upon evicting the real page. As those pages are found from the LRU, an iput() can lead to the inode being freed concurrently. At this point, reclaim must no longer install shadow pages because the inode freeing code needs to ensure the page tree is really empty. Add an address_space flag, AS_EXITING, that the inode freeing code sets under the tree lock before doing the final truncate. Reclaim will check for this flag before installing shadow pages. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Greg Thelen <gthelen@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Metin Doslu <metin@citusdata.com> Cc: Michel Lespinasse <walken@google.com> Cc: Ozgun Erdogan <ozgun@citusdata.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roman Gushchin <klamm@yandex-team.ru> Cc: Ryan Mallon <rmallon@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:47:49 +00:00
}
EXPORT_SYMBOL(truncate_inode_pages_final);
/**
* invalidate_mapping_pagevec - Invalidate all the unlocked pages of one inode
* @mapping: the address_space which holds the pages to invalidate
* @start: the offset 'from' which to invalidate
* @end: the offset 'to' which to invalidate (inclusive)
* @nr_pagevec: invalidate failed page number for caller
*
* This helper is similar to invalidate_mapping_pages(), except that it accounts
* for pages that are likely on a pagevec and counts them in @nr_pagevec, which
* will be used by the caller.
*/
unsigned long invalidate_mapping_pagevec(struct address_space *mapping,
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
pgoff_t start, pgoff_t end, unsigned long *nr_pagevec)
{
2014-04-03 21:47:46 +00:00
pgoff_t indices[PAGEVEC_SIZE];
struct folio_batch fbatch;
pgoff_t index = start;
mm: deactivate invalidated pages 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 <minchan.kim@gmail.com> Reported-by: Ben Gamari <bgamari.foss@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:32:52 +00:00
unsigned long ret;
unsigned long count = 0;
int i;
folio_batch_init(&fbatch);
while (find_lock_entries(mapping, index, end, &fbatch, indices)) {
for (i = 0; i < folio_batch_count(&fbatch); i++) {
struct folio *folio = fbatch.folios[i];
/* We rely upon deletion not changing folio->index */
2014-04-03 21:47:46 +00:00
index = indices[i];
if (xa_is_value(folio)) {
count += invalidate_exceptional_entry(mapping,
index,
folio);
2014-04-03 21:47:46 +00:00
continue;
}
index += folio_nr_pages(folio) - 1;
ret = mapping_evict_folio(mapping, folio);
folio_unlock(folio);
mm: deactivate invalidated pages 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 <minchan.kim@gmail.com> Reported-by: Ben Gamari <bgamari.foss@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:32:52 +00:00
/*
* Invalidation is a hint that the folio is no longer
mm: deactivate invalidated pages 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 <minchan.kim@gmail.com> Reported-by: Ben Gamari <bgamari.foss@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:32:52 +00:00
* of interest and try to speed up its reclaim.
*/
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
if (!ret) {
deactivate_file_folio(folio);
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
/* It is likely on the pagevec of a remote CPU */
if (nr_pagevec)
(*nr_pagevec)++;
}
mm: deactivate invalidated pages 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 <minchan.kim@gmail.com> Reported-by: Ben Gamari <bgamari.foss@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:32:52 +00:00
count += ret;
}
folio_batch_remove_exceptionals(&fbatch);
folio_batch_release(&fbatch);
cond_resched();
index++;
}
mm: deactivate invalidated pages 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 <minchan.kim@gmail.com> Reported-by: Ben Gamari <bgamari.foss@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 23:32:52 +00:00
return count;
}
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
/**
* invalidate_mapping_pages - Invalidate all clean, unlocked cache of one inode
* @mapping: the address_space which holds the cache to invalidate
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
* @start: the offset 'from' which to invalidate
* @end: the offset 'to' which to invalidate (inclusive)
*
* This function removes pages that are clean, unmapped and unlocked,
* as well as shadow entries. It will not block on IO activity.
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
*
* If you want to remove all the pages of one inode, regardless of
* their use and writeback state, use truncate_inode_pages().
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
*
* Return: the number of the cache entries that were invalidated
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
*/
unsigned long invalidate_mapping_pages(struct address_space *mapping,
pgoff_t start, pgoff_t end)
{
return invalidate_mapping_pagevec(mapping, start, end, NULL);
mm, fadvise: improve the expensive remote LRU cache draining after FADV_DONTNEED Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Link: https://lkml.kernel.org/r/20200923133318.14373-1-laoar.shao@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-13 23:51:47 +00:00
}
EXPORT_SYMBOL(invalidate_mapping_pages);
/*
* This is like invalidate_inode_page(), except it ignores the page's
* refcount. We do this because invalidate_inode_pages2() needs stronger
* invalidation guarantees, and cannot afford to leave pages behind because
* shrink_page_list() has a temp ref on them, or because they're transiently
* sitting in the lru_cache_add() pagevecs.
*/
static int invalidate_complete_folio2(struct address_space *mapping,
struct folio *folio)
{
if (folio->mapping != mapping)
return 0;
if (folio_has_private(folio) &&
!filemap_release_folio(folio, GFP_KERNEL))
return 0;
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
spin_lock(&mapping->host->i_lock);
xa_lock_irq(&mapping->i_pages);
if (folio_test_dirty(folio))
goto failed;
BUG_ON(folio_has_private(folio));
__filemap_remove_folio(folio, NULL);
xa_unlock_irq(&mapping->i_pages);
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
if (mapping_shrinkable(mapping))
inode_add_lru(mapping->host);
spin_unlock(&mapping->host->i_lock);
filemap_free_folio(mapping, folio);
return 1;
failed:
xa_unlock_irq(&mapping->i_pages);
vfs: keep inodes with page cache off the inode shrinker LRU Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Link: https://lkml.kernel.org/r/20210614211904.14420-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-09 02:31:24 +00:00
spin_unlock(&mapping->host->i_lock);
return 0;
}
static int folio_launder(struct address_space *mapping, struct folio *folio)
{
if (!folio_test_dirty(folio))
return 0;
if (folio->mapping != mapping || mapping->a_ops->launder_folio == NULL)
return 0;
return mapping->a_ops->launder_folio(folio);
}
/**
* invalidate_inode_pages2_range - remove range of pages from an address_space
* @mapping: the address_space
* @start: the page offset 'from' which to invalidate
* @end: the page offset 'to' which to invalidate (inclusive)
*
* Any pages which are found to be mapped into pagetables are unmapped prior to
* invalidation.
*
* Return: -EBUSY if any pages could not be invalidated.
*/
int invalidate_inode_pages2_range(struct address_space *mapping,
pgoff_t start, pgoff_t end)
{
2014-04-03 21:47:46 +00:00
pgoff_t indices[PAGEVEC_SIZE];
struct folio_batch fbatch;
pgoff_t index;
int i;
int ret = 0;
int ret2 = 0;
int did_range_unmap = 0;
if (mapping_empty(mapping))
return 0;
folio_batch_init(&fbatch);
index = start;
while (find_get_entries(mapping, index, end, &fbatch, indices)) {
for (i = 0; i < folio_batch_count(&fbatch); i++) {
struct folio *folio = fbatch.folios[i];
/* We rely upon deletion not changing folio->index */
2014-04-03 21:47:46 +00:00
index = indices[i];
if (xa_is_value(folio)) {
if (!invalidate_exceptional_entry2(mapping,
index, folio))
ret = -EBUSY;
2014-04-03 21:47:46 +00:00
continue;
}
if (!did_range_unmap && folio_mapped(folio)) {
mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page() There is a race between THP unmapping and truncation, when truncate sees pmd_none() and skips the entry, after munmap's zap_huge_pmd() cleared it, but before its page_remove_rmap() gets to decrement compound_mapcount: generating false "BUG: Bad page cache" reports that the page is still mapped when deleted. This commit fixes that, but not in the way I hoped. The first attempt used try_to_unmap(page, TTU_SYNC|TTU_IGNORE_MLOCK) instead of unmap_mapping_range() in truncate_cleanup_page(): it has often been an annoyance that we usually call unmap_mapping_range() with no pages locked, but there apply it to a single locked page. try_to_unmap() looks more suitable for a single locked page. However, try_to_unmap_one() contains a VM_BUG_ON_PAGE(!pvmw.pte,page): it is used to insert THP migration entries, but not used to unmap THPs. Copy zap_huge_pmd() and add THP handling now? Perhaps, but their TLB needs are different, I'm too ignorant of the DAX cases, and couldn't decide how far to go for anon+swap. Set that aside. The second attempt took a different tack: make no change in truncate.c, but modify zap_huge_pmd() to insert an invalidated huge pmd instead of clearing it initially, then pmd_clear() between page_remove_rmap() and unlocking at the end. Nice. But powerpc blows that approach out of the water, with its serialize_against_pte_lookup(), and interesting pgtable usage. It would need serious help to get working on powerpc (with a minor optimization issue on s390 too). Set that aside. Just add an "if (page_mapped(page)) synchronize_rcu();" or other such delay, after unmapping in truncate_cleanup_page()? Perhaps, but though that's likely to reduce or eliminate the number of incidents, it would give less assurance of whether we had identified the problem correctly. This successful iteration introduces "unmap_mapping_page(page)" instead of try_to_unmap(), and goes the usual unmap_mapping_range_tree() route, with an addition to details. Then zap_pmd_range() watches for this case, and does spin_unlock(pmd_lock) if so - just like page_vma_mapped_walk() now does in the PVMW_SYNC case. Not pretty, but safe. Note that unmap_mapping_page() is doing a VM_BUG_ON(!PageLocked) to assert its interface; but currently that's only used to make sure that page->mapping is stable, and zap_pmd_range() doesn't care if the page is locked or not. Along these lines, in invalidate_inode_pages2_range() move the initial unmap_mapping_range() out from under page lock, before then calling unmap_mapping_page() under page lock if still mapped. Link: https://lkml.kernel.org/r/a2a4a148-cdd8-942c-4ef8-51b77f643dbe@google.com Fixes: fc127da085c2 ("truncate: handle file thp") Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Alistair Popple <apopple@nvidia.com> Cc: Jan Kara <jack@suse.cz> Cc: Jue Wang <juew@google.com> Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org> Cc: Miaohe Lin <linmiaohe@huawei.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Peter Xu <peterx@redhat.com> Cc: Ralph Campbell <rcampbell@nvidia.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Wang Yugui <wangyugui@e16-tech.com> Cc: Zi Yan <ziy@nvidia.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-16 01:24:03 +00:00
/*
* If folio is mapped, before taking its lock,
mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page() There is a race between THP unmapping and truncation, when truncate sees pmd_none() and skips the entry, after munmap's zap_huge_pmd() cleared it, but before its page_remove_rmap() gets to decrement compound_mapcount: generating false "BUG: Bad page cache" reports that the page is still mapped when deleted. This commit fixes that, but not in the way I hoped. The first attempt used try_to_unmap(page, TTU_SYNC|TTU_IGNORE_MLOCK) instead of unmap_mapping_range() in truncate_cleanup_page(): it has often been an annoyance that we usually call unmap_mapping_range() with no pages locked, but there apply it to a single locked page. try_to_unmap() looks more suitable for a single locked page. However, try_to_unmap_one() contains a VM_BUG_ON_PAGE(!pvmw.pte,page): it is used to insert THP migration entries, but not used to unmap THPs. Copy zap_huge_pmd() and add THP handling now? Perhaps, but their TLB needs are different, I'm too ignorant of the DAX cases, and couldn't decide how far to go for anon+swap. Set that aside. The second attempt took a different tack: make no change in truncate.c, but modify zap_huge_pmd() to insert an invalidated huge pmd instead of clearing it initially, then pmd_clear() between page_remove_rmap() and unlocking at the end. Nice. But powerpc blows that approach out of the water, with its serialize_against_pte_lookup(), and interesting pgtable usage. It would need serious help to get working on powerpc (with a minor optimization issue on s390 too). Set that aside. Just add an "if (page_mapped(page)) synchronize_rcu();" or other such delay, after unmapping in truncate_cleanup_page()? Perhaps, but though that's likely to reduce or eliminate the number of incidents, it would give less assurance of whether we had identified the problem correctly. This successful iteration introduces "unmap_mapping_page(page)" instead of try_to_unmap(), and goes the usual unmap_mapping_range_tree() route, with an addition to details. Then zap_pmd_range() watches for this case, and does spin_unlock(pmd_lock) if so - just like page_vma_mapped_walk() now does in the PVMW_SYNC case. Not pretty, but safe. Note that unmap_mapping_page() is doing a VM_BUG_ON(!PageLocked) to assert its interface; but currently that's only used to make sure that page->mapping is stable, and zap_pmd_range() doesn't care if the page is locked or not. Along these lines, in invalidate_inode_pages2_range() move the initial unmap_mapping_range() out from under page lock, before then calling unmap_mapping_page() under page lock if still mapped. Link: https://lkml.kernel.org/r/a2a4a148-cdd8-942c-4ef8-51b77f643dbe@google.com Fixes: fc127da085c2 ("truncate: handle file thp") Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Alistair Popple <apopple@nvidia.com> Cc: Jan Kara <jack@suse.cz> Cc: Jue Wang <juew@google.com> Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org> Cc: Miaohe Lin <linmiaohe@huawei.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Peter Xu <peterx@redhat.com> Cc: Ralph Campbell <rcampbell@nvidia.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Wang Yugui <wangyugui@e16-tech.com> Cc: Zi Yan <ziy@nvidia.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-16 01:24:03 +00:00
* zap the rest of the file in one hit.
*/
unmap_mapping_pages(mapping, index,
(1 + end - index), false);
did_range_unmap = 1;
}
folio_lock(folio);
VM_BUG_ON_FOLIO(!folio_contains(folio, index), folio);
if (folio->mapping != mapping) {
folio_unlock(folio);
continue;
}
folio_wait_writeback(folio);
mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page() There is a race between THP unmapping and truncation, when truncate sees pmd_none() and skips the entry, after munmap's zap_huge_pmd() cleared it, but before its page_remove_rmap() gets to decrement compound_mapcount: generating false "BUG: Bad page cache" reports that the page is still mapped when deleted. This commit fixes that, but not in the way I hoped. The first attempt used try_to_unmap(page, TTU_SYNC|TTU_IGNORE_MLOCK) instead of unmap_mapping_range() in truncate_cleanup_page(): it has often been an annoyance that we usually call unmap_mapping_range() with no pages locked, but there apply it to a single locked page. try_to_unmap() looks more suitable for a single locked page. However, try_to_unmap_one() contains a VM_BUG_ON_PAGE(!pvmw.pte,page): it is used to insert THP migration entries, but not used to unmap THPs. Copy zap_huge_pmd() and add THP handling now? Perhaps, but their TLB needs are different, I'm too ignorant of the DAX cases, and couldn't decide how far to go for anon+swap. Set that aside. The second attempt took a different tack: make no change in truncate.c, but modify zap_huge_pmd() to insert an invalidated huge pmd instead of clearing it initially, then pmd_clear() between page_remove_rmap() and unlocking at the end. Nice. But powerpc blows that approach out of the water, with its serialize_against_pte_lookup(), and interesting pgtable usage. It would need serious help to get working on powerpc (with a minor optimization issue on s390 too). Set that aside. Just add an "if (page_mapped(page)) synchronize_rcu();" or other such delay, after unmapping in truncate_cleanup_page()? Perhaps, but though that's likely to reduce or eliminate the number of incidents, it would give less assurance of whether we had identified the problem correctly. This successful iteration introduces "unmap_mapping_page(page)" instead of try_to_unmap(), and goes the usual unmap_mapping_range_tree() route, with an addition to details. Then zap_pmd_range() watches for this case, and does spin_unlock(pmd_lock) if so - just like page_vma_mapped_walk() now does in the PVMW_SYNC case. Not pretty, but safe. Note that unmap_mapping_page() is doing a VM_BUG_ON(!PageLocked) to assert its interface; but currently that's only used to make sure that page->mapping is stable, and zap_pmd_range() doesn't care if the page is locked or not. Along these lines, in invalidate_inode_pages2_range() move the initial unmap_mapping_range() out from under page lock, before then calling unmap_mapping_page() under page lock if still mapped. Link: https://lkml.kernel.org/r/a2a4a148-cdd8-942c-4ef8-51b77f643dbe@google.com Fixes: fc127da085c2 ("truncate: handle file thp") Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Alistair Popple <apopple@nvidia.com> Cc: Jan Kara <jack@suse.cz> Cc: Jue Wang <juew@google.com> Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org> Cc: Miaohe Lin <linmiaohe@huawei.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Peter Xu <peterx@redhat.com> Cc: Ralph Campbell <rcampbell@nvidia.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Wang Yugui <wangyugui@e16-tech.com> Cc: Zi Yan <ziy@nvidia.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-16 01:24:03 +00:00
if (folio_mapped(folio))
unmap_mapping_folio(folio);
BUG_ON(folio_mapped(folio));
mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page() There is a race between THP unmapping and truncation, when truncate sees pmd_none() and skips the entry, after munmap's zap_huge_pmd() cleared it, but before its page_remove_rmap() gets to decrement compound_mapcount: generating false "BUG: Bad page cache" reports that the page is still mapped when deleted. This commit fixes that, but not in the way I hoped. The first attempt used try_to_unmap(page, TTU_SYNC|TTU_IGNORE_MLOCK) instead of unmap_mapping_range() in truncate_cleanup_page(): it has often been an annoyance that we usually call unmap_mapping_range() with no pages locked, but there apply it to a single locked page. try_to_unmap() looks more suitable for a single locked page. However, try_to_unmap_one() contains a VM_BUG_ON_PAGE(!pvmw.pte,page): it is used to insert THP migration entries, but not used to unmap THPs. Copy zap_huge_pmd() and add THP handling now? Perhaps, but their TLB needs are different, I'm too ignorant of the DAX cases, and couldn't decide how far to go for anon+swap. Set that aside. The second attempt took a different tack: make no change in truncate.c, but modify zap_huge_pmd() to insert an invalidated huge pmd instead of clearing it initially, then pmd_clear() between page_remove_rmap() and unlocking at the end. Nice. But powerpc blows that approach out of the water, with its serialize_against_pte_lookup(), and interesting pgtable usage. It would need serious help to get working on powerpc (with a minor optimization issue on s390 too). Set that aside. Just add an "if (page_mapped(page)) synchronize_rcu();" or other such delay, after unmapping in truncate_cleanup_page()? Perhaps, but though that's likely to reduce or eliminate the number of incidents, it would give less assurance of whether we had identified the problem correctly. This successful iteration introduces "unmap_mapping_page(page)" instead of try_to_unmap(), and goes the usual unmap_mapping_range_tree() route, with an addition to details. Then zap_pmd_range() watches for this case, and does spin_unlock(pmd_lock) if so - just like page_vma_mapped_walk() now does in the PVMW_SYNC case. Not pretty, but safe. Note that unmap_mapping_page() is doing a VM_BUG_ON(!PageLocked) to assert its interface; but currently that's only used to make sure that page->mapping is stable, and zap_pmd_range() doesn't care if the page is locked or not. Along these lines, in invalidate_inode_pages2_range() move the initial unmap_mapping_range() out from under page lock, before then calling unmap_mapping_page() under page lock if still mapped. Link: https://lkml.kernel.org/r/a2a4a148-cdd8-942c-4ef8-51b77f643dbe@google.com Fixes: fc127da085c2 ("truncate: handle file thp") Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Alistair Popple <apopple@nvidia.com> Cc: Jan Kara <jack@suse.cz> Cc: Jue Wang <juew@google.com> Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org> Cc: Miaohe Lin <linmiaohe@huawei.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Peter Xu <peterx@redhat.com> Cc: Ralph Campbell <rcampbell@nvidia.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Wang Yugui <wangyugui@e16-tech.com> Cc: Zi Yan <ziy@nvidia.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-16 01:24:03 +00:00
ret2 = folio_launder(mapping, folio);
if (ret2 == 0) {
if (!invalidate_complete_folio2(mapping, folio))
VFS: fix dio write returning EIO when try_to_release_page fails Dio write returns EIO when try_to_release_page fails because bh is still referenced. The patch commit 3f31fddfa26b7594b44ff2b34f9a04ba409e0f91 Author: Mingming Cao <cmm@us.ibm.com> Date: Fri Jul 25 01:46:22 2008 -0700 jbd: fix race between free buffer and commit transaction was merged into 2.6.27-rc1, but I noticed that this patch is not enough to fix the race. I did fsstress test heavily to 2.6.27-rc1, and found that dio write still sometimes got EIO through this test. The patch above fixed race between freeing buffer(dio) and committing transaction(jbd) but I discovered that there is another race, freeing buffer(dio) and ext3/4_ordered_writepage. : background_writeout() ->write_cache_pages() ->ext3_ordered_writepage() walk_page_buffers() -> take a bh ref block_write_full_page() -> unlock_page : <- end_page_writeback : <- race! (dio write->try_to_release_page fails) walk_page_buffers() ->release a bh ref ext3_ordered_writepage holds bh ref and does unlock_page remaining taking a bh ref, so this causes the race and failure of try_to_release_page. To fix this race, I used the approach of falling back to buffered writes if try_to_release_page() fails on a page. [akpm@linux-foundation.org: cleanups] Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Chris Mason <chris.mason@oracle.com> Cc: Jan Kara <jack@suse.cz> Cc: Mingming Cao <cmm@us.ibm.com> Cc: Zach Brown <zach.brown@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-09-02 21:35:40 +00:00
ret2 = -EBUSY;
}
if (ret2 < 0)
ret = ret2;
folio_unlock(folio);
}
folio_batch_remove_exceptionals(&fbatch);
folio_batch_release(&fbatch);
cond_resched();
index++;
}
/*
* For DAX we invalidate page tables after invalidating page cache. We
* could invalidate page tables while invalidating each entry however
* that would be expensive. And doing range unmapping before doesn't
* work as we have no cheap way to find whether page cache entry didn't
* get remapped later.
*/
if (dax_mapping(mapping)) {
unmap_mapping_pages(mapping, start, end - start + 1, false);
}
return ret;
}
EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range);
/**
* invalidate_inode_pages2 - remove all pages from an address_space
* @mapping: the address_space
*
* Any pages which are found to be mapped into pagetables are unmapped prior to
* invalidation.
*
* Return: -EBUSY if any pages could not be invalidated.
*/
int invalidate_inode_pages2(struct address_space *mapping)
{
return invalidate_inode_pages2_range(mapping, 0, -1);
}
EXPORT_SYMBOL_GPL(invalidate_inode_pages2);
/**
* truncate_pagecache - unmap and remove pagecache that has been truncated
* @inode: inode
* @newsize: new file size
*
* inode's new i_size must already be written before truncate_pagecache
* is called.
*
* This function should typically be called before the filesystem
* releases resources associated with the freed range (eg. deallocates
* blocks). This way, pagecache will always stay logically coherent
* with on-disk format, and the filesystem would not have to deal with
* situations such as writepage being called for a page that has already
* had its underlying blocks deallocated.
*/
void truncate_pagecache(struct inode *inode, loff_t newsize)
{
struct address_space *mapping = inode->i_mapping;
loff_t holebegin = round_up(newsize, PAGE_SIZE);
/*
* unmap_mapping_range is called twice, first simply for
* efficiency so that truncate_inode_pages does fewer
* single-page unmaps. However after this first call, and
* before truncate_inode_pages finishes, it is possible for
* private pages to be COWed, which remain after
* truncate_inode_pages finishes, hence the second
* unmap_mapping_range call must be made for correctness.
*/
unmap_mapping_range(mapping, holebegin, 0, 1);
truncate_inode_pages(mapping, newsize);
unmap_mapping_range(mapping, holebegin, 0, 1);
}
EXPORT_SYMBOL(truncate_pagecache);
/**
* truncate_setsize - update inode and pagecache for a new file size
* @inode: inode
* @newsize: new file size
*
* truncate_setsize updates i_size and performs pagecache truncation (if
* necessary) to @newsize. It will be typically be called from the filesystem's
* setattr function when ATTR_SIZE is passed in.
*
* Must be called with a lock serializing truncates and writes (generally
* i_rwsem but e.g. xfs uses a different lock) and before all filesystem
* specific block truncation has been performed.
*/
void truncate_setsize(struct inode *inode, loff_t newsize)
{
loff_t oldsize = inode->i_size;
i_size_write(inode, newsize);
if (newsize > oldsize)
pagecache_isize_extended(inode, oldsize, newsize);
truncate_pagecache(inode, newsize);
}
EXPORT_SYMBOL(truncate_setsize);
/**
* pagecache_isize_extended - update pagecache after extension of i_size
* @inode: inode for which i_size was extended
* @from: original inode size
* @to: new inode size
*
* Handle extension of inode size either caused by extending truncate or by
* write starting after current i_size. We mark the page straddling current
* i_size RO so that page_mkwrite() is called on the nearest write access to
* the page. This way filesystem can be sure that page_mkwrite() is called on
* the page before user writes to the page via mmap after the i_size has been
* changed.
*
* The function must be called after i_size is updated so that page fault
* coming after we unlock the page will already see the new i_size.
* The function must be called while we still hold i_rwsem - this not only
* makes sure i_size is stable but also that userspace cannot observe new
* i_size value before we are prepared to store mmap writes at new inode size.
*/
void pagecache_isize_extended(struct inode *inode, loff_t from, loff_t to)
{
int bsize = i_blocksize(inode);
loff_t rounded_from;
struct page *page;
pgoff_t index;
WARN_ON(to > inode->i_size);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
if (from >= to || bsize == PAGE_SIZE)
return;
/* Page straddling @from will not have any hole block created? */
rounded_from = round_up(from, bsize);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
if (to <= rounded_from || !(rounded_from & (PAGE_SIZE - 1)))
return;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
index = from >> PAGE_SHIFT;
page = find_lock_page(inode->i_mapping, index);
/* Page not cached? Nothing to do */
if (!page)
return;
/*
* See clear_page_dirty_for_io() for details why set_page_dirty()
* is needed.
*/
if (page_mkclean(page))
set_page_dirty(page);
unlock_page(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(page);
}
EXPORT_SYMBOL(pagecache_isize_extended);
/**
* truncate_pagecache_range - unmap and remove pagecache that is hole-punched
* @inode: inode
* @lstart: offset of beginning of hole
* @lend: offset of last byte of hole
*
* This function should typically be called before the filesystem
* releases resources associated with the freed range (eg. deallocates
* blocks). This way, pagecache will always stay logically coherent
* with on-disk format, and the filesystem would not have to deal with
* situations such as writepage being called for a page that has already
* had its underlying blocks deallocated.
*/
void truncate_pagecache_range(struct inode *inode, loff_t lstart, loff_t lend)
{
struct address_space *mapping = inode->i_mapping;
loff_t unmap_start = round_up(lstart, PAGE_SIZE);
loff_t unmap_end = round_down(1 + lend, PAGE_SIZE) - 1;
/*
* This rounding is currently just for example: unmap_mapping_range
* expands its hole outwards, whereas we want it to contract the hole
* inwards. However, existing callers of truncate_pagecache_range are
* doing their own page rounding first. Note that unmap_mapping_range
* allows holelen 0 for all, and we allow lend -1 for end of file.
*/
/*
* Unlike in truncate_pagecache, unmap_mapping_range is called only
* once (before truncating pagecache), and without "even_cows" flag:
* hole-punching should not remove private COWed pages from the hole.
*/
if ((u64)unmap_end > (u64)unmap_start)
unmap_mapping_range(mapping, unmap_start,
1 + unmap_end - unmap_start, 0);
truncate_inode_pages_range(mapping, lstart, lend);
}
EXPORT_SYMBOL(truncate_pagecache_range);