Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1517921

[PATCH 4.4 07/69] mm: memcontrol: do not recurse in direct reclaim

From Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Newsgroups linux.kernel
Subject [PATCH 4.4 07/69] mm: memcontrol: do not recurse in direct reclaim
Date 2016-11-09 12:00 +0100
Message-ID <sByLM-6cP-25@gated-at.bofh.it> (permalink)
References <sByC5-69v-3@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


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

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

From: Johannes Weiner <hannes@cmpxchg.org>

commit 89a2848381b5fcd9c4d9c0cd97680e3b28730e31 upstream.

On 4.0, we saw a stack corruption from a page fault entering direct
memory cgroup reclaim, calling into btrfs_releasepage(), which then
tried to allocate an extent and recursed back into a kmem charge ad
nauseam:

  [...]
  btrfs_releasepage+0x2c/0x30
  try_to_release_page+0x32/0x50
  shrink_page_list+0x6da/0x7a0
  shrink_inactive_list+0x1e5/0x510
  shrink_lruvec+0x605/0x7f0
  shrink_zone+0xee/0x320
  do_try_to_free_pages+0x174/0x440
  try_to_free_mem_cgroup_pages+0xa7/0x130
  try_charge+0x17b/0x830
  memcg_charge_kmem+0x40/0x80
  new_slab+0x2d9/0x5a0
  __slab_alloc+0x2fd/0x44f
  kmem_cache_alloc+0x193/0x1e0
  alloc_extent_state+0x21/0xc0
  __clear_extent_bit+0x2b5/0x400
  try_release_extent_mapping+0x1a3/0x220
  __btrfs_releasepage+0x31/0x70
  btrfs_releasepage+0x2c/0x30
  try_to_release_page+0x32/0x50
  shrink_page_list+0x6da/0x7a0
  shrink_inactive_list+0x1e5/0x510
  shrink_lruvec+0x605/0x7f0
  shrink_zone+0xee/0x320
  do_try_to_free_pages+0x174/0x440
  try_to_free_mem_cgroup_pages+0xa7/0x130
  try_charge+0x17b/0x830
  mem_cgroup_try_charge+0x65/0x1c0
  handle_mm_fault+0x117f/0x1510
  __do_page_fault+0x177/0x420
  do_page_fault+0xc/0x10
  page_fault+0x22/0x30

On later kernels, kmem charging is opt-in rather than opt-out, and that
particular kmem allocation in btrfs_releasepage() is no longer being
charged and won't recurse and overrun the stack anymore.

But it's not impossible for an accounted allocation to happen from the
memcg direct reclaim context, and we needed to reproduce this crash many
times before we even got a useful stack trace out of it.

Like other direct reclaimers, mark tasks in memcg reclaim PF_MEMALLOC to
avoid recursing into any other form of direct reclaim.  Then let
recursive charges from PF_MEMALLOC contexts bypass the cgroup limit.

Link: http://lkml.kernel.org/r/20161025141050.GA13019@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
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/memcontrol.c |    9 +++++++++
 mm/vmscan.c     |    2 ++
 2 files changed, 11 insertions(+)

--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2055,6 +2055,15 @@ retry:
 		     current->flags & PF_EXITING))
 		goto force;
 
+	/*
+	 * Prevent unbounded recursion when reclaim operations need to
+	 * allocate memory. This might exceed the limits temporarily,
+	 * but we prefer facilitating memory reclaim and getting back
+	 * under the limit over triggering OOM kills in these cases.
+	 */
+	if (unlikely(current->flags & PF_MEMALLOC))
+		goto force;
+
 	if (unlikely(task_in_memcg_oom(current)))
 		goto nomem;
 
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2910,7 +2910,9 @@ unsigned long try_to_free_mem_cgroup_pag
 					    sc.may_writepage,
 					    sc.gfp_mask);
 
+	current->flags |= PF_MEMALLOC;
 	nr_reclaimed = do_try_to_free_pages(zonelist, &sc);
+	current->flags &= ~PF_MEMALLOC;
 
 	trace_mm_vmscan_memcg_reclaim_end(nr_reclaimed);
 

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH 4.4 00/69] 4.4.31-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 23/69] USB: serial: ftdi_sio: add support for Infineon TriBoard TC2X7 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 27/69] Fix potential infoleak in older kernels Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 32/69] USB: serial: cp210x: fix tiocmget error handling Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 04/69] h8300: fix syscall restarting Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 01/69] i2c: xgene: Avoid dma_buffer overrun Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 19/69] ubifs: Fix regression in ubifs_readdir() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 29/69] hv: do not lose pending heartbeat vmbus packets Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 31/69] tty: limit terminal size to 4M chars Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 11/69] ALSA: hda - Raise AZX_DCAPS_RIRB_DELAY handling into top drivers Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 06/69] mm/list_lru.c: avoid error-path NULL pointer deref Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 20/69] mei: txe: dont clean an unprocessed interrupt cause. Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 22/69] USB: serial: fix potential NULL-dereference at probe Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 18/69] ubifs: Abort readdir upon error Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 26/69] GenWQE: Fix bad page access during abort of resource allocation Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 11:50 +0100
  [PATCH 4.4 08/69] KEYS: Fix short sprintf buffer in /proc/keys show function Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:00 +0100
  [PATCH 4.4 02/69] i2c: core: fix NULL pointer dereference under race condition Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:00 +0100
  [PATCH 4.4 07/69] mm: memcontrol: do not recurse in direct reclaim Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:00 +0100
  [PATCH 4.4 67/69] kvm: x86: Check memopp before dereference (CVE-2016-8630) Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 60/69] cgroup: avoid false positive gcc-6 warning Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 37/69] parisc: Ensure consistent state when switching to kernel stack at syscall entry Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 46/69] scsi: scsi_debug: Fix memory leak if LBP enabled and module is unloaded Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 39/69] virtio: console: Unlock vqs while freeing buffers Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 47/69] scsi: arcmsr: Send SYNCHRONIZE_CACHE command to firmware Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 65/69] usb: dwc3: Fix size used in dma_free_coherent() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 54/69] perf build: Fix traceevent plugins build race Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 38/69] virtio_ring: Make interrupt suppression spec compliant Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 57/69] ARM: 8584/1: floppy: avoid gcc-6 warning Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 51/69] drm/radeon/si_dpm: workaround for SI kickers Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 64/69] pwm: Unexport children before chip removal Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 36/69] ovl: fsync after copy-up Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 42/69] firewire: net: guard against rx buffer overflows Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:10 +0100
  [PATCH 4.4 68/69] ubi: fastmap: Fix add_vol() return value test in ubi_attach_fastmap() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:40 +0100
  [PATCH 4.4 44/69] mac80211: discard multicast and 4-addr A-MSDUs Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:40 +0100
  [PATCH 4.4 43/69] firewire: net: fix fragmented datagram_size off-by-one Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:40 +0100
  [PATCH 4.4 45/69] scsi: megaraid_sas: Fix data integrity failure for JBOD (passthrough) devices Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:40 +0100
  [PATCH 4.4 69/69] HID: usbhid: add ATEN CS962 to list of quirky devices Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:40 +0100
  [PATCH 4.4 50/69] drm/radeon/si_dpm: Limit clocks on HD86xx part Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 58/69] mm/cma: silence warnings due to max() usage Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 59/69] drm/exynos: fix error handling in exynos_drm_subdrv_open Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 41/69] Input: i8042 - add XMG C504 to keyboard reset table Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 48/69] mmc: dw_mmc-pltfm: fix the potential NULL pointer dereference Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 63/69] UBI: fastmap: scrub PEB when bitflips are detected in a free PEB EC header Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 55/69] x86/xen: fix upper bound of pmd loop in xen_cleanhighmap() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 56/69] powerpc/ptrace: Fix out of bounds array access warning Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 40/69] dm mirror: fix read error on recovery after default leg failure Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  [PATCH 4.4 61/69] smc91x: avoid self-comparison warning Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-09 12:50 +0100
  Re: [PATCH 4.4 00/69] 4.4.31-stable review Shuah Khan <shuah.kh@samsung.com> - 2016-11-09 19:30 +0100
  Re: [PATCH 4.4 00/69] 4.4.31-stable review Guenter Roeck <linux@roeck-us.net> - 2016-11-09 20:40 +0100

csiph-web