From b7b2d2ccdbc4dce33a882d9c00484a3a2dc10866 Mon Sep 17 00:00:00 2001 From: Baolin Wang Date: Fri, 10 Jul 2026 10:09:22 +0800 Subject: [PATCH] mm: shmem: fix potential livelock issue for shmem direct swapin When skipping swapcache for synchronous IO swap devices, swapcache_prepare() is used to prevent parallel swapin from proceeding with the swap cache flag. However, on PREEMPT kernels this can lead to a livelock, as reported by Chao[1]: Thread A starts direct swapin of a shmem folio and calls swapcache_prepare() to set SWAP_HAS_CACHE. It may then be preempted inside workingset_refault(). Meanwhile, a higher priority thread B also attempts direct swapin of the same shmem swap entry. Since swapcache_prepare() already marks the entry, thread B repeatedly gets -EEXIST and busy-loops waiting for thread A to finish. But as thread B runs at higher priority, thread A cannot preempt it, resulting in starvation and a livelock. Fix it by yielding the CPU with schedule_timeout_uninterruptible(1) when swapcache_prepare() fails, following the same approach used in commit 029c4628b2eb ("mm: swap: get rid of livelock in swapin readahead") and commit 13ddaf26be32 ("mm/swap: fix race when skipping swapcache"). However, commit 01626a1823 ("mm: avoid unconditional one-tick sleep when swapcache_prepare fails") found that the unconditional one-tick sleep can cause UI stuttering on latency-sensitive Android devices. So we can follow the same approach by adding a waitqueue to wake up tasks when needed, instead of always sleeping for a full tick. Note that mainline does not have this potential issue, which has already been resolved by Kairui's swap refactoring work[2]. [1] https://lore.kernel.org/all/700a2cbf90a2484f979aac858f08f5d4@xiaomi.com/ [2] https://lore.kernel.org/all/20260517-swap-table-p4-v5-0-88ae43e064c7@tencent.com/ Fixes: 1dd44c0af4fa ("mm: shmem: skip swapcache for swapin of synchronous swap device") Reported-by: Ma Chao Closes: https://lore.kernel.org/all/700a2cbf90a2484f979aac858f08f5d4@xiaomi.com/ Signed-off-by: Baolin Wang Signed-off-by: Sasha Levin --- mm/shmem.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/mm/shmem.c b/mm/shmem.c index 94c5b0d78ac3..3c329b794ae4 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2005,11 +2005,14 @@ unlock: return ERR_PTR(error); } +static DECLARE_WAIT_QUEUE_HEAD(shmem_swapcache_wq); + static struct folio *shmem_swap_alloc_folio(struct inode *inode, struct vm_area_struct *vma, pgoff_t index, swp_entry_t entry, int order, gfp_t gfp) { struct shmem_inode_info *info = SHMEM_I(inode); + DECLARE_WAITQUEUE(wait, current); int nr_pages = 1 << order; struct folio *new; gfp_t alloc_gfp; @@ -2066,6 +2069,10 @@ retry: if (swapcache_prepare(entry, nr_pages)) { folio_put(new); new = ERR_PTR(-EEXIST); + /* Relax a bit to prevent rapid repeated page faults */ + add_wait_queue(&shmem_swapcache_wq, &wait); + schedule_timeout_uninterruptible(1); + remove_wait_queue(&shmem_swapcache_wq, &wait); /* Try smaller folio to avoid cache conflict */ goto fallback; } @@ -2423,6 +2430,8 @@ static int shmem_swapin_folio(struct inode *inode, pgoff_t index, if (skip_swapcache) { folio->swap.val = 0; swapcache_clear(si, swap, nr_pages); + if (waitqueue_active(&shmem_swapcache_wq)) + wake_up(&shmem_swapcache_wq); } else { swap_cache_del_folio(folio); } @@ -2442,8 +2451,11 @@ unlock: if (folio) folio_unlock(folio); failed_nolock: - if (skip_swapcache) + if (skip_swapcache) { swapcache_clear(si, folio->swap, folio_nr_pages(folio)); + if (waitqueue_active(&shmem_swapcache_wq)) + wake_up(&shmem_swapcache_wq); + } if (folio) folio_put(folio); put_swap_device(si);