6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chao Yu chao@kernel.org
[ Upstream commit f7f8932ca6bb22494ef6db671633ad3b4d982271 ]
If checkpoint is disabled, GC can not reclaim any segments, we need to detect such condition and bail out from fallocate() of a pinfile, rather than letting allocator running out of free segment, which may cause f2fs to be shutdown.
reproducer: mkfs.f2fs -f /dev/vda 16777216 mount -o checkpoint=disable:10% /dev/vda /mnt/f2fs for ((i=0;i<4096;i++)) do { dd if=/dev/zero of=/mnt/f2fs/$i bs=1M count=1; } done sync for ((i=0;i<4096;i+=2)) do { rm /mnt/f2fs/$i; } done sync touch /mnt/f2fs/pinfile f2fs_io pinfile set /mnt/f2fs/pinfile f2fs_io fallocate 0 0 4201644032 /mnt/f2fs/pinfile
cat /sys/kernel/debug/f2fs/status output: - Free: 0 (0)
Fixes: f5a53edcf01e ("f2fs: support aligned pinned file") Signed-off-by: Chao Yu chao@kernel.org Signed-off-by: Jaegeuk Kim jaegeuk@kernel.org Stable-dep-of: e462fc48ceb8 ("f2fs: maintain one time GC mode is enabled during whole zoned GC cycle") Signed-off-by: Sasha Levin sashal@kernel.org --- fs/f2fs/file.c | 12 ++++++++++++ 1 file changed, 12 insertions(+)
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index b8d24c1c48bd4..be32f672497d6 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1838,6 +1838,18 @@ static int f2fs_expand_inode_data(struct inode *inode, loff_t offset, next_alloc: f2fs_down_write(&sbi->pin_sem);
+ if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) { + if (has_not_enough_free_secs(sbi, 0, 0)) { + f2fs_up_write(&sbi->pin_sem); + err = -ENOSPC; + f2fs_warn_ratelimited(sbi, + "ino:%lu, start:%lu, end:%lu, need to trigger GC to " + "reclaim enough free segment when checkpoint is enabled", + inode->i_ino, pg_start, pg_end); + goto out_err; + } + } + if (has_not_enough_free_secs(sbi, 0, f2fs_sb_has_blkzoned(sbi) ? ZONED_PIN_SEC_REQUIRED_COUNT : GET_SEC_FROM_SEG(sbi, overprovision_segments(sbi)))) {