Skip to contentSkip to conversation

mm: memcg: fix memcg reclaim soft lockup

PATCH In review Julius Hemanth Pitti 2 participants 3 messages Sep 18, 2020 – Sep 21, 2020

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 mm: memcg: fix memcg reclaim soft lockup

From: Xunlei Pang <[email protected]>

commit e3336cab2579012b1e72b5265adf98e2d6e244ad upstream.

We've met softlockup with "CONFIG_PREEMPT_NONE=y", when the target memcg doesn't have any reclaimable memory.

It can be easily reproduced as below:

  watchdog: BUG: soft lockup - CPU#0 stuck for 111s![memcg_test:2204]
  CPU: 0 PID: 2204 Comm: memcg_test Not tainted 5.9.0-rc2+ #12
  Call Trace:
    shrink_lruvec+0x49f/0x640
    shrink_node+0x2a6/0x6f0
    do_try_to_free_pages+0xe9/0x3e0
    try_to_free_mem_cgroup_pages+0xef/0x1f0
    try_charge+0x2c1/0x750
    mem_cgroup_charge+0xd7/0x240
    __add_to_page_cache_locked+0x2fd/0x370
    add_to_page_cache_lru+0x4a/0xc0
    pagecache_get_page+0x10b/0x2f0
    filemap_fault+0x661/0xad0
    ext4_filemap_fault+0x2c/0x40
    __do_fault+0x4d/0xf9
    handle_mm_fault+0x1080/0x1790

It only happens on our 1-vcpu instances, because there's no chance for oom reaper to run to reclaim the to-be-killed process.

Add a cond_resched() at the upper shrink_node_memcgs() to solve this issue, this will mean that we will get a scheduling point for each memcg in the reclaimed hierarchy without any dependency on the reclaimable memory in that memcg thus making it more predictable.

---
 mm/vmscan.c | 8 ++++++++
 1 file changed, 8 insertions(+)
mm/vmscan.c +8 -0
index 749d239c62b2..8b97bc615d8c 100644
@@ -2619,6 +2619,14 @@ static void shrink_node_memcgs(pg_data_t *pgdat, struct scan_control *sc)
 		unsigned long reclaimed;
 		unsigned long scanned;
 
+		/*
+		 * This loop can become CPU-bound when target memcgs
+		 * aren't eligible for reclaim - either because they
+		 * don't have any reclaimable pages, or because their
+		 * memory is explicitly protected. Avoid soft lockups.
+		 */
+		cond_resched();
+
 		switch (mem_cgroup_protected(target_memcg, memcg)) {
 		case MEMCG_PROT_MIN:
 			/*
Signature · 1 line
2.17.1
#1 lore raw
Quoted 45 lines · Julius Hemanth Pitti

From: Xunlei Pang <[email protected]>

commit e3336cab2579012b1e72b5265adf98e2d6e244ad upstream.

We've met softlockup with "CONFIG_PREEMPT_NONE=y", when the target memcg doesn't have any reclaimable memory.

It can be easily reproduced as below:

  watchdog: BUG: soft lockup - CPU#0 stuck for 111s![memcg_test:2204]
  CPU: 0 PID: 2204 Comm: memcg_test Not tainted 5.9.0-rc2+ #12
  Call Trace:
    shrink_lruvec+0x49f/0x640
    shrink_node+0x2a6/0x6f0
    do_try_to_free_pages+0xe9/0x3e0
    try_to_free_mem_cgroup_pages+0xef/0x1f0
    try_charge+0x2c1/0x750
    mem_cgroup_charge+0xd7/0x240
    __add_to_page_cache_locked+0x2fd/0x370
    add_to_page_cache_lru+0x4a/0xc0
    pagecache_get_page+0x10b/0x2f0
    filemap_fault+0x661/0xad0
    ext4_filemap_fault+0x2c/0x40
    __do_fault+0x4d/0xf9
    handle_mm_fault+0x1080/0x1790

It only happens on our 1-vcpu instances, because there's no chance for oom reaper to run to reclaim the to-be-killed process.

Add a cond_resched() at the upper shrink_node_memcgs() to solve this issue, this will mean that we will get a scheduling point for each memcg in the reclaimed hierarchy without any dependency on the reclaimable memory in that memcg thus making it more predictable.

Quoted 3 lines
---
 mm/vmscan.c | 8 ++++++++
 1 file changed, 8 insertions(+)

The Fixes: tag you show here goes back to 4.19, can you provide a 4.19.y and 5.4.y version of this as well?

thanks,

greg k-h

#2 lore raw
Quoted 58 lines · Greg KH
Quoted 48 lines · Julius Hemanth Pitti

From: Xunlei Pang <[email protected]>

commit e3336cab2579012b1e72b5265adf98e2d6e244ad upstream.

We've met softlockup with "CONFIG_PREEMPT_NONE=y", when the target memcg
doesn't have any reclaimable memory.

It can be easily reproduced as below:

  watchdog: BUG: soft lockup - CPU#0 stuck for
111s![memcg_test:2204]
  CPU: 0 PID: 2204 Comm: memcg_test Not tainted 5.9.0-rc2+ #12
  Call Trace:
    shrink_lruvec+0x49f/0x640
    shrink_node+0x2a6/0x6f0
    do_try_to_free_pages+0xe9/0x3e0
    try_to_free_mem_cgroup_pages+0xef/0x1f0
    try_charge+0x2c1/0x750
    mem_cgroup_charge+0xd7/0x240
    __add_to_page_cache_locked+0x2fd/0x370
    add_to_page_cache_lru+0x4a/0xc0
    pagecache_get_page+0x10b/0x2f0
    filemap_fault+0x661/0xad0
    ext4_filemap_fault+0x2c/0x40
    __do_fault+0x4d/0xf9
    handle_mm_fault+0x1080/0x1790

It only happens on our 1-vcpu instances, because there's no chance for
oom reaper to run to reclaim the to-be-killed process.

Add a cond_resched() at the upper shrink_node_memcgs() to solve this
issue, this will mean that we will get a scheduling point for each memcg
in the reclaimed hierarchy without any dependency on the reclaimable
memory in that memcg thus making it more predictable.

Link:
http://lkml.kernel.org/r/[email protected]

Quoted 8 lines
Fixes b0dedc49a2da ("mm/vmscan.c: iterate only over charged

shrinkers during memcg shrink_slab()")

---
 mm/vmscan.c | 8 ++++++++
 1 file changed, 8 insertions(+)
Quoted 3 lines

The Fixes: tag you show here goes back to 4.19, can you provide a 4.19.y
and 5.4.y version of this as well?

Sure. Will send for both 5.4.y and 4.19.y.

Quoted 4 lines

thanks,

greg k-h