Re: linux-5.13.2: warning from kernel/rcu/tree_plugin.h:359

[Matthew Wilcox]

On Mon, Jul 19, 2021 at 02:12:15PM +0200, Oleksandr Natalenko wrote:
> On pondělí 19. července 2021 14:08:37 CEST Miaohe Lin wrote:
> > On 2021/7/19 19:59, Oleksandr Natalenko wrote:
> > > On pondělí 19. července 2021 13:50:07 CEST Miaohe Lin wrote:
> > >> On 2021/7/19 19:22, Matthew Wilcox wrote:
> > >>> On Mon, Jul 19, 2021 at 07:12:58PM +0800, Miaohe Lin wrote:
> > >>>> When in the commit 2799e77529c2a, we're using the percpu_ref to
> > >>>> serialize
> > >>>> against concurrent swapoff, i.e. there's percpu_ref inside
> > >>>> get_swap_device() instead of rcu_read_lock(). Please see commit
> > >>>> 63d8620ecf93 ("mm/swapfile: use percpu_ref to serialize against
> > >>>> concurrent swapoff") for detail.
> > >>> 
> > >>> Oh, so this is a backport problem.  2799e77529c2 was backported without
> > >>> its prerequisite 63d8620ecf93.  Greg, probably best to just drop
> > >> 
> > >> Yes, they're posted as a patch set:
> > >> 
> > >> https://lkml.kernel.org/r/20210426123316.806267-1-linmiaohe@xxxxxxxxxx
> > >> 
> > >>> 2799e77529c2 from all stable trees; the race described is not very
> > >>> important (swapoff vs reading a page back from that swap device).
> > >>> .
> > >> 
> > >> The swapoff races with reading a page back from that swap device should
> > >> be
> > >> really uncommon as most users only do swapoff when the system is going to
> > >> shutdown.
> > >> 
> > >> Sorry for the trouble!
> > > 
> > > git log --oneline v5.13..v5.13.3 --author="Miaohe Lin"
> > > 11ebc09e50dc mm/zswap.c: fix two bugs in zswap_writeback_entry()
> > > 95d192da198d mm/z3fold: use release_z3fold_page_locked() to release locked
> > > z3fold page
> > > ccb7848e2344 mm/z3fold: fix potential memory leak in z3fold_destroy_pool()
> > > 9f7229c901c1 mm/huge_memory.c: don't discard hugepage if other processes
> > > are mapping it
> > > f13259175e4f mm/huge_memory.c: add missing read-only THP checking in
> > > transparent_hugepage_enabled()
> > > afafd371e7de mm/huge_memory.c: remove dedicated macro
> > > HPAGE_CACHE_INDEX_MASK a533a21b692f mm/shmem: fix shmem_swapin() race
> > > with swapoff
> > > c3b39134bbd0 swap: fix do_swap_page() race with swapoff
> > > 
> > > Do you suggest reverting "mm/shmem: fix shmem_swapin() race with swapoff"
> > > as well?
> > 
> > This patch also rely on its prerequisite 63d8620ecf93. I think we should
> > either revert any commit in this series or just backport the entire series.
> 
> Then why not just pick up 2 more patches instead of dropping 2 patches. Greg, 
> could you please make sure the whole series from [1] gets pulled?

Because none of these patches should have been backported in the first
place.  It's just not worth the destabilisation.