Re: mm: mkfs.ext4 invoked oom-killer on i386 - pagecache_get_page

On Thu, 21 May 2020, Johannes Weiner wrote:

Very much appreciate you guys tracking it down so quickly. Sorry about the breakage.

I think mem_cgroup_disabled() checks are pretty good markers of public entry points to the memcg API, so I'd prefer that even if a bit more verbose. What do you think?

An explicit mem_cgroup_disabled() check would be fine, but I must admit, the patch below is rather too verbose for my own taste. Your call.

---

From cd373ec232942a9bc43ee5e7d2171352019a58fb Mon Sep 17 00:00:00 2001 From: Hugh Dickins hughd@google.com Date: Thu, 21 May 2020 14:58:36 -0400 Subject: [PATCH] mm: memcontrol: prepare swap controller setup for integration fix

Fix crash with cgroup_disable=memory:

...

...

...

...

• mkfs -t ext4 /dev/disk/by-id/ata-TOSHIBA_MG04ACA100N_Y8NRK0BPF6XF

mke2fs 1.43.8 (1-Jan-2018) Creating filesystem with 244190646 4k blocks and 61054976 inodes Filesystem UUID: 3bb1a285-2cb4-44b4-b6e8-62548f3ac620 Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968, 102400000, 214990848 Allocating group tables: 0/7453 done Writing inode tables: 0/7453 done Creating journal (262144 blocks): [ 35.502102] BUG: kernel NULL pointer dereference, address: 000000c8 [ 35.508372] #PF: supervisor read access in kernel mode [ 35.513506] #PF: error_code(0x0000) - not-present page [ 35.518638] *pde = 00000000 [ 35.521514] Oops: 0000 [#1] SMP [ 35.524652] CPU: 0 PID: 145 Comm: kswapd0 Not tainted 5.7.0-rc6-next-20200519+ #1 [ 35.532121] Hardware name: Supermicro SYS-5019S-ML/X11SSH-F, BIOS 2.2 05/23/2018 [ 35.539507] EIP: mem_cgroup_get_nr_swap_pages+0x28/0x60

do_memsw_account() used to be automatically false when the cgroup controller was disabled. Now that it's replaced by cgroup_memory_noswap, for which this isn't true, make the mem_cgroup_disabled() checks explicit in the swap control API.

[hannes@cmpxchg.org: use mem_cgroup_disabled() in all API functions] Reported-by: Naresh Kamboju naresh.kamboju@linaro.org Debugged-by: Hugh Dickins hughd@google.com Debugged-by: Michal Hocko mhocko@kernel.org Signed-off-by: Johannes Weiner hannes@cmpxchg.org

---

mm/memcontrol.c | 47 +++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 41 insertions(+), 6 deletions(-)

I'm certainly not against a mem_cgroup_disabled() check in the only place that's been observed to need it, as a fixup to merge into your original patch; but this seems rather an over-reaction - and I'm a little surprised that setting mem_cgroup_disabled() doesn't just force cgroup_memory_noswap, saving repetitious checks elsewhere (perhaps there's a difficulty in that, I haven't looked).

Historically, I think we've added mem_cgroup_disabled() checks (accessing a cacheline we'd rather avoid) where they're necessary, rather than at every "interface".

And you seem to be in a very "goto out" mood today - we all have our "goto out" days, alternating with our "return 0" days :)

Hugh

diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 3e000a316b59..850bca380562 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -6811,6 +6811,9 @@ void mem_cgroup_swapout(struct page *page, swp_entry_t entry) VM_BUG_ON_PAGE(PageLRU(page), page); VM_BUG_ON_PAGE(page_count(page), page);

• if (mem_cgroup_disabled())

• return;
  

• if (cgroup_subsys_on_dfl(memory_cgrp_subsys)) return;

@@ -6876,6 +6879,10 @@ int mem_cgroup_try_charge_swap(struct page *page, swp_entry_t entry) struct mem_cgroup *memcg; unsigned short oldid;

• if (mem_cgroup_disabled())

• return 0;
  

• /* Only cgroup2 has swap.max */ if (!cgroup_subsys_on_dfl(memory_cgrp_subsys)) return 0;

@@ -6920,6 +6927,9 @@ void mem_cgroup_uncharge_swap(swp_entry_t entry, unsigned int nr_pages) struct mem_cgroup *memcg; unsigned short id;

• if (mem_cgroup_disabled())

• return;
  

• id = swap_cgroup_record(entry, 0, nr_pages); rcu_read_lock(); memcg = mem_cgroup_from_id(id);

@@ -6940,12 +6950,25 @@ long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg) { long nr_swap_pages = get_nr_swap_pages();

• if (cgroup_memory_noswap || !cgroup_subsys_on_dfl(memory_cgrp_subsys))

• return nr_swap_pages;
  

• if (mem_cgroup_disabled())

• goto out;
  

• /* Swap control disabled */
• if (cgroup_memory_noswap)

• goto out;
  

• /*

• * Only cgroup2 has swap.max, cgroup1 does mem+sw accounting,
  

• * which does not place restrictions specifically on swap.
  

• */
  

• if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))

• goto out;
  

• for (; memcg != root_mem_cgroup; memcg = parent_mem_cgroup(memcg)) nr_swap_pages = min_t(long, nr_swap_pages, READ_ONCE(memcg->swap.max) - page_counter_read(&memcg->swap));

+out: return nr_swap_pages; } @@ -6957,18 +6980,30 @@ bool mem_cgroup_swap_full(struct page *page) if (vm_swap_full()) return true;

• if (cgroup_memory_noswap || !cgroup_subsys_on_dfl(memory_cgrp_subsys))

• return false;
  

• if (mem_cgroup_disabled())

• goto out;
  

• /* Swap control disabled */
• if (cgroup_memory_noswap)

• goto out;
  

• /*

• * Only cgroup2 has swap.max, cgroup1 does mem+sw accounting,
  

• * which does not place restrictions specifically on swap.
  

• */
  

• if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))

• goto out;
  

memcg = page->mem_cgroup; if (!memcg)

• return false;
  

• goto out;
  

for (; memcg != root_mem_cgroup; memcg = parent_mem_cgroup(memcg)) if (page_counter_read(&memcg->swap) * 2 >= READ_ONCE(memcg->swap.max)) return true;

+out: return false; } -- 2.26.2