[PATCH 4.6 45/56] Revert "mm, mempool: only set __GFP_NOMEMALLOC if there are free elements"

From	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Newsgroups	linux.kernel
Subject	[PATCH 4.6 45/56] Revert "mm, mempool: only set __GFP_NOMEMALLOC if there are free elements"
Date	2016-08-14 22:50 +0200
Message-ID	<s6aw2-4AX-45@gated-at.bofh.it> (permalink)
References	<s6amm-4wI-9@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

4.6-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Michal Hocko <mhocko@suse.com>

commit 4e390b2b2f34b8daaabf2df1df0cf8f798b87ddb upstream.

This reverts commit f9054c70d28b ("mm, mempool: only set __GFP_NOMEMALLOC
if there are free elements").

There has been a report about OOM killer invoked when swapping out to a
dm-crypt device.  The primary reason seems to be that the swapout out IO
managed to completely deplete memory reserves.  Ondrej was able to
bisect and explained the issue by pointing to f9054c70d28b ("mm,
mempool: only set __GFP_NOMEMALLOC if there are free elements").

The reason is that the swapout path is not throttled properly because
the md-raid layer needs to allocate from the generic_make_request path
which means it allocates from the PF_MEMALLOC context.  dm layer uses
mempool_alloc in order to guarantee a forward progress which used to
inhibit access to memory reserves when using page allocator.  This has
changed by f9054c70d28b ("mm, mempool: only set __GFP_NOMEMALLOC if
there are free elements") which has dropped the __GFP_NOMEMALLOC
protection when the memory pool is depleted.

If we are running out of memory and the only way forward to free memory
is to perform swapout we just keep consuming memory reserves rather than
throttling the mempool allocations and allowing the pending IO to
complete up to a moment when the memory is depleted completely and there
is no way forward but invoking the OOM killer.  This is less than
optimal.

The original intention of f9054c70d28b was to help with the OOM
situations where the oom victim depends on mempool allocation to make a
forward progress.  David has mentioned the following backtrace:

  schedule
  schedule_timeout
  io_schedule_timeout
  mempool_alloc
  __split_and_process_bio
  dm_request
  generic_make_request
  submit_bio
  mpage_readpages
  ext4_readpages
  __do_page_cache_readahead
  ra_submit
  filemap_fault
  handle_mm_fault
  __do_page_fault
  do_page_fault
  page_fault

We do not know more about why the mempool is depleted without being
replenished in time, though.  In any case the dm layer shouldn't depend
on any allocations outside of the dedicated pools so a forward progress
should be guaranteed.  If this is not the case then the dm should be
fixed rather than papering over the problem and postponing it to later
by accessing more memory reserves.

mempools are a mechanism to maintain dedicated memory reserves to
guaratee forward progress.  Allowing them an unbounded access to the
page allocator memory reserves is going against the whole purpose of
this mechanism.

Bisected by Ondrej Kozina.

[akpm@linux-foundation.org: coding-style fixes]
Link: http://lkml.kernel.org/r/20160721145309.GR26379@dhcp22.suse.cz
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reported-by: Ondrej Kozina <okozina@redhat.com>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: NeilBrown <neilb@suse.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Mikulas Patocka <mpatocka@redhat.com>
Cc: Ondrej Kozina <okozina@redhat.com>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 mm/mempool.c |   18 +++---------------
 1 file changed, 3 insertions(+), 15 deletions(-)

--- a/mm/mempool.c
+++ b/mm/mempool.c
@@ -310,7 +310,7 @@ EXPORT_SYMBOL(mempool_resize);
  * returns NULL. Note that due to preallocation, this function
  * *never* fails when called from process contexts. (it might
  * fail if called from an IRQ context.)
- * Note: neither __GFP_NOMEMALLOC nor __GFP_ZERO are supported.
+ * Note: using __GFP_ZERO is not supported.
  */
 void *mempool_alloc(mempool_t *pool, gfp_t gfp_mask)
 {
@@ -319,27 +319,16 @@ void *mempool_alloc(mempool_t *pool, gfp
 	wait_queue_t wait;
 	gfp_t gfp_temp;
 
-	/* If oom killed, memory reserves are essential to prevent livelock */
-	VM_WARN_ON_ONCE(gfp_mask & __GFP_NOMEMALLOC);
-	/* No element size to zero on allocation */
 	VM_WARN_ON_ONCE(gfp_mask & __GFP_ZERO);
-
 	might_sleep_if(gfp_mask & __GFP_DIRECT_RECLAIM);
 
+	gfp_mask |= __GFP_NOMEMALLOC;	/* don't allocate emergency reserves */
 	gfp_mask |= __GFP_NORETRY;	/* don't loop in __alloc_pages */
 	gfp_mask |= __GFP_NOWARN;	/* failures are OK */
 
 	gfp_temp = gfp_mask & ~(__GFP_DIRECT_RECLAIM|__GFP_IO);
 
 repeat_alloc:
-	if (likely(pool->curr_nr)) {
-		/*
-		 * Don't allocate from emergency reserves if there are
-		 * elements available.  This check is racy, but it will
-		 * be rechecked each loop.
-		 */
-		gfp_temp |= __GFP_NOMEMALLOC;
-	}
 
 	element = pool->alloc(gfp_temp, pool->pool_data);
 	if (likely(element != NULL))
@@ -363,12 +352,11 @@ repeat_alloc:
 	 * We use gfp mask w/o direct reclaim or IO for the first round.  If
 	 * alloc failed with that and @pool was empty, retry immediately.
 	 */
-	if ((gfp_temp & ~__GFP_NOMEMALLOC) != gfp_mask) {
+	if (gfp_temp != gfp_mask) {
 		spin_unlock_irqrestore(&pool->lock, flags);
 		gfp_temp = gfp_mask;
 		goto repeat_alloc;
 	}
-	gfp_temp = gfp_mask;
 
 	/* We must not sleep if !__GFP_DIRECT_RECLAIM */
 	if (!(gfp_mask & __GFP_DIRECT_RECLAIM)) {

Thread

[PATCH 4.6 00/56] 4.6.7-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:40 +0200
  [PATCH 4.6 56/56] ext4: fix reference counting bug on block allocation error Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 02/56] libnvdimm, pfn, dax: fix initialization vs autodetect for mode + alignment Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 28/56] x86/xen, pat: Remove PAT table init code from Xen Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 35/56] mm: memcontrol: fix memcg id ref counter on swap charge move Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 06/56] bonding: set carrier off for devices created through netlink Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 34/56] mm: memcontrol: fix swap counter leak on swapout from offline cgroup Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 32/56] powerpc/eeh: Fix invalid cached PE primary bus Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 39/56] radix-tree: account nodes to memcg only if explicitly requested Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 45/56] Revert "mm, mempool: only set __GFP_NOMEMALLOC if there are free elements" Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 41/56] x86/power/64: Fix hibernation return address corruption Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 22/56] random: strengthen input validation for RNDADDTOENTCNT Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 54/56] ext4: validate s_reserved_gdt_blocks on mount Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 37/56] block: fix use-after-free in seq file Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 51/56] ext4: check for extents that wrap around Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 27/56] x86/mtrr: Fix PAT init handling when MTRR is disabled Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 05/56] tcp: enable per-socket rate limiting of all challenge acks Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 25/56] x86/mm/pat: Replace cpu_has_pat with boot_cpu_has() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 21/56] regulator: qcom_smd: Remove list_voltage callback for rpm_smps_ldo_ops_fixed Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 47/56] Revert "cpufreq: pcc-cpufreq: update default value of cpuinfo_transition_latency" Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 50/56] serial: mvebu-uart: free the IRQ in ->shutdown() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 53/56] ext4: dont call ext4_should_journal_data() on the journal inode Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 46/56] fs/dcache.c: avoid soft-lockup in dput() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 48/56] crypto: gcm - Filter out async ghash if necessary Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 09/56] bridge: Fix incorrect re-injection of LLDP packets Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 30/56] x86/mm/pat: Fix BUG_ON() in mmap_mem() on QEMU/i386 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 38/56] sysv, ipc: fix security-layer leaking Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 44/56] fuse: fix wrong assignment of ->flags in fuse_send_init() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 18/56] IB/hfi1: Correct issues with sc5 computation Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 16/56] arm: oabi compat: add missing access checks Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 17/56] KEYS: 64-bit MIPS needs to use compat_sys_keyctl for 32-bit userspace Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 22:50 +0200
  [PATCH 4.6 42/56] fuse: fsync() did not return IO errors Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 36/56] x86/syscalls/64: Add compat_sys_keyctl for 32-bit userspace Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 29/56] x86/pat: Document the PAT initialization sequence Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 40/56] x86/microcode: Fix suspend to RAM with builtin microcode Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 33/56] powerpc/bpf/jit: Disable classic BPF JIT on ppc64le Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 26/56] x86/mtrr: Fix Xorg crashes in Qemu sessions Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 23/56] x86/mm/pat: Add support of non-default PAT MSR setting Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 52/56] ext4: fix deadlock during page writeback Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 55/56] ext4: short-cut orphan cleanup on error Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 31/56] udf: Prevent stack overflow on corrupted filesystem mount Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 43/56] fuse: fuse_flush must check mapping->flags for errors Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 49/56] crypto: scatterwalk - Fix test in scatterwalk_done Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:00 +0200
  [PATCH 4.6 20/56] apparmor: fix ref count leak when profile sha1 hash is read Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:10 +0200
  [PATCH 4.6 04/56] tcp: make challenge acks less predictable Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:10 +0200
  [PATCH 4.6 03/56] ppp: defer netns reference release for ppp channel Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:10 +0200
  [PATCH 4.6 19/56] IB/hfi1: Fix deadlock with txreq allocation slow path Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:10 +0200
  [PATCH 4.6 08/56] vlan: use a valid default mtu value for vlan over macsec Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:10 +0200
  [PATCH 4.6 07/56] net: bgmac: Fix infinite loop in bgmac_dma_tx_add() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:10 +0200
  [PATCH 4.6 24/56] x86/mm/pat: Add pat_disable() interface Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-14 23:10 +0200
  Re: [PATCH 4.6 00/56] 4.6.7-stable review Guenter Roeck <linux@roeck-us.net> - 2016-08-15 15:10 +0200
  Re: [PATCH 4.6 00/56] 4.6.7-stable review Shuah Khan <shuah.kh@samsung.com> - 2016-08-16 06:10 +0200

csiph-web