The patch titled
Subject: mm: throttle on IO only when there are too many dirty and writeback pages
has been removed from the -mm tree. Its filename was
mm-throttle-on-io-only-when-there-are-too-many-dirty-and-writeback-pages.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: Michal Hocko <mhocko@xxxxxxxx>
Subject: mm: throttle on IO only when there are too many dirty and writeback pages
wait_iff_congested has been used to throttle allocator before it retried
another round of direct reclaim to allow the writeback to make some
progress and prevent reclaim from looping over dirty/writeback pages
without making any progress. We used to do congestion_wait before
0e093d99763e ("writeback: do not sleep on the congestion queue if there
are no congested BDIs or if significant congestion is not being
encountered in the current zone") but that led to undesirable stalls and
sleeping for the full timeout even when the BDI wasn't congested. Hence
wait_iff_congested was used instead. But it seems that even
wait_iff_congested doesn't work as expected. We might have a small file
LRU list with all pages dirty/writeback and yet the bdi is not congested
so this is just a cond_resched in the end and can end up triggering pre
mature OOM.
This patch replaces the unconditional wait_iff_congested by
congestion_wait which is executed only if we _know_ that the last round of
direct reclaim didn't make any progress and dirty+writeback pages are more
than a half of the reclaimable pages on the zone which might be usable for
our target allocation. This shouldn't reintroduce stalls fixed by
0e093d99763e because congestion_wait is called only when we are getting
hopeless when sleeping is a better choice than OOM with many pages under
IO.
We have to preserve logic introduced by 373ccbe59270 ("mm, vmstat: allow
WQ concurrency to discover memory reclaim doesn't make any progress") into
the __alloc_pages_slowpath now that wait_iff_congested is not used
anymore. As the only remaining user of wait_iff_congested is
shrink_inactive_list we can remove the WQ specific short sleep from
wait_iff_congested because the sleep is needed to be done only once in the
allocation retry cycle.
[mhocko@xxxxxxxx: high_zoneidx->ac_classzone_idx to evaluate memory reserves properly]
Link: http://lkml.kernel.org/r/1463051677-29418-2-git-send-email-mhocko@xxxxxxxxxx
Signed-off-by: Michal Hocko <mhocko@xxxxxxxx>
Acked-by: Hillf Danton <hillf.zj@xxxxxxxxxxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Joonsoo Kim <js1304@xxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx>
Cc: Vladimir Davydov <vdavydov@xxxxxxxxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/backing-dev.c | 20 +++-----------------
mm/page_alloc.c | 41 +++++++++++++++++++++++++++++++++++++----
2 files changed, 40 insertions(+), 21 deletions(-)
diff -puN mm/backing-dev.c~mm-throttle-on-io-only-when-there-are-too-many-dirty-and-writeback-pages mm/backing-dev.c
--- a/mm/backing-dev.c~mm-throttle-on-io-only-when-there-are-too-many-dirty-and-writeback-pages
+++ a/mm/backing-dev.c
@@ -957,9 +957,8 @@ EXPORT_SYMBOL(congestion_wait);
* jiffies for either a BDI to exit congestion of the given @sync queue
* or a write to complete.
*
- * In the absence of zone congestion, a short sleep or a cond_resched is
- * performed to yield the processor and to allow other subsystems to make
- * a forward progress.
+ * In the absence of zone congestion, cond_resched() is called to yield
+ * the processor if necessary but otherwise does not sleep.
*
* The return value is 0 if the sleep is for the full timeout. Otherwise,
* it is the number of jiffies that were still remaining when the function
@@ -979,20 +978,7 @@ long wait_iff_congested(struct zone *zon
*/
if (atomic_read(&nr_wb_congested[sync]) == 0 ||
!test_bit(ZONE_CONGESTED, &zone->flags)) {
-
- /*
- * Memory allocation/reclaim might be called from a WQ
- * context and the current implementation of the WQ
- * concurrency control doesn't recognize that a particular
- * WQ is congested if the worker thread is looping without
- * ever sleeping. Therefore we have to do a short sleep
- * here rather than calling cond_resched().
- */
- if (current->flags & PF_WQ_WORKER)
- schedule_timeout_uninterruptible(1);
- else
- cond_resched();
-
+ cond_resched();
/* In case we scheduled, work out time remaining */
ret = timeout - (jiffies - start);
if (ret < 0)
diff -puN mm/page_alloc.c~mm-throttle-on-io-only-when-there-are-too-many-dirty-and-writeback-pages mm/page_alloc.c
--- a/mm/page_alloc.c~mm-throttle-on-io-only-when-there-are-too-many-dirty-and-writeback-pages
+++ a/mm/page_alloc.c
@@ -3436,8 +3436,9 @@ should_reclaim_retry(gfp_t gfp_mask, uns
for_each_zone_zonelist_nodemask(zone, z, ac->zonelist, ac->high_zoneidx,
ac->nodemask) {
unsigned long available;
+ unsigned long reclaimable;
- available = zone_reclaimable_pages(zone);
+ available = reclaimable = zone_reclaimable_pages(zone);
available -= DIV_ROUND_UP(no_progress_loops * available,
MAX_RECLAIM_RETRIES);
available += zone_page_state_snapshot(zone, NR_FREE_PAGES);
@@ -3447,9 +3448,41 @@ should_reclaim_retry(gfp_t gfp_mask, uns
* available?
*/
if (__zone_watermark_ok(zone, order, min_wmark_pages(zone),
- ac->high_zoneidx, alloc_flags, available)) {
- /* Wait for some write requests to complete then retry */
- wait_iff_congested(zone, BLK_RW_ASYNC, HZ/50);
+ ac_classzone_idx(ac), alloc_flags, available)) {
+ /*
+ * If we didn't make any progress and have a lot of
+ * dirty + writeback pages then we should wait for
+ * an IO to complete to slow down the reclaim and
+ * prevent from pre mature OOM
+ */
+ if (!did_some_progress) {
+ unsigned long writeback;
+ unsigned long dirty;
+
+ writeback = zone_page_state_snapshot(zone,
+ NR_WRITEBACK);
+ dirty = zone_page_state_snapshot(zone, NR_FILE_DIRTY);
+
+ if (2*(writeback + dirty) > reclaimable) {
+ congestion_wait(BLK_RW_ASYNC, HZ/10);
+ return true;
+ }
+ }
+
+ /*
+ * Memory allocation/reclaim might be called from a WQ
+ * context and the current implementation of the WQ
+ * concurrency control doesn't recognize that
+ * a particular WQ is congested if the worker thread is
+ * looping without ever sleeping. Therefore we have to
+ * do a short sleep here rather than calling
+ * cond_resched().
+ */
+ if (current->flags & PF_WQ_WORKER)
+ schedule_timeout_uninterruptible(1);
+ else
+ cond_resched();
+
return true;
}
}
_
Patches currently in -mm which might be from mhocko@xxxxxxxx are
mm-oom-protect-costly-allocations-some-more.patch
mm-consider-compaction-feedback-also-for-costly-allocation.patch
mm-oom-compaction-prevent-from-should_compact_retry-looping-for-ever-for-costly-orders.patch
mm-oom-protect-costly-allocations-some-more-for-config_compaction.patch
mm-oom_reaper-hide-oom-reaped-tasks-from-oom-killer-more-carefully.patch
mm-oom_reaper-do-not-mmput-synchronously-from-the-oom-reaper-context.patch
oom-consider-multi-threaded-tasks-in-task_will_free_mem.patch
mm-make-mmap_sem-for-write-waits-killable-for-mm-syscalls.patch
mm-make-vm_mmap-killable.patch
mm-make-vm_munmap-killable.patch
mm-aout-handle-vm_brk-failures.patch
mm-elf-handle-vm_brk-error.patch
mm-make-vm_brk-killable.patch
mm-proc-make-clear_refs-killable.patch
mm-fork-make-dup_mmap-wait-for-mmap_sem-for-write-killable.patch
ipc-shm-make-shmem-attach-detach-wait-for-mmap_sem-killable.patch
vdso-make-arch_setup_additional_pages-wait-for-mmap_sem-for-write-killable.patch
coredump-make-coredump_wait-wait-for-mmap_sem-for-write-killable.patch
aio-make-aio_setup_ring-killable.patch
exec-make-exec-path-waiting-for-mmap_sem-killable.patch
prctl-make-pr_set_thp_disable-wait-for-mmap_sem-killable.patch
uprobes-wait-for-mmap_sem-for-write-killable.patch
drm-i915-make-i915_gem_mmap_ioctl-wait-for-mmap_sem-killable.patch
drm-radeon-make-radeon_mn_get-wait-for-mmap_sem-killable.patch
drm-amdgpu-make-amdgpu_mn_get-wait-for-mmap_sem-killable.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
