f2fs: Set GF_NOFS in read_cache_page_gfp while doing f2fs_quota_read

Hello,

Observed below deadlock with quota enabled on f2fs during low memory situation. I think during f2fs_quota_enable -> which is doing read_cache_page with __GFP_FS flag set, due to low memory this is causing shrinker to get invoked and thus going into a deadlock due to mutex lock in dquota. Commit details in next patch contains the call stack.

Since I did not had the quota enabled on f2fs configuration, I could not test this on my setup yet.

Could the community kindly review and let me know if the patch is correct or not ?

Ritesh Harjani (1):
f2fs: Set GF_NOFS in read_cache_page_gfp while doing f2fs_quota_read

 fs/f2fs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
a Linux Foundation Collaborative Project.

Quota code itself is serializing the operations by taking mutex_lock. It seems a below deadlock can happen if GF_NOFS is not used in f2fs_quota_read

__switch_to+0x88
__schedule+0x5b0
schedule+0x78
schedule_preempt_disabled+0x20
__mutex_lock_slowpath+0xdc //mutex owner is itself
mutex_lock+0x2c
dquot_commit+0x30 //mutex_lock(&dqopt->dqio_mutex);
dqput+0xe0
__dquot_drop+0x80
dquot_drop+0x48
f2fs_evict_inode+0x218
evict+0xa8
dispose_list+0x3c
prune_icache_sb+0x58
super_cache_scan+0xf4
do_shrink_slab+0x208
shrink_slab.part.40+0xac
shrink_zone+0x1b0
do_try_to_free_pages+0x25c
try_to_free_pages+0x164
__alloc_pages_nodemask+0x534
do_read_cache_page+0x6c
read_cache_page+0x14
f2fs_quota_read+0xa4
read_blk+0x54
find_tree_dqentry+0xe4
find_tree_dqentry+0xb8
find_tree_dqentry+0xb8
find_tree_dqentry+0xb8
qtree_read_dquot+0x68
v2_read_dquot+0x24
dquot_acquire+0x5c // mutex_lock(&dqopt->dqio_mutex);
dqget+0x238
__dquot_initialize+0xd4
dquot_initialize+0x10
dquot_file_open+0x34
f2fs_file_open+0x6c
do_dentry_open+0x1e4
vfs_open+0x6c
path_openat+0xa20
do_filp_open+0x4c
do_sys_open+0x178

---
 fs/f2fs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 708155d..3a9a924 100644
@@ -1382,7 +1382,7 @@ static ssize_t f2fs_quota_read(struct super_block *sb, int type, char *data,
 	while (toread > 0) {
 		tocopy = min_t(unsigned long, sb->s_blocksize - offset, toread);
 repeat:
-		page = read_mapping_page(mapping, blkidx, NULL);
+		page = read_cache_page_gfp(mapping, blkidx, GFP_NOFS);
 		if (IS_ERR(page)) {
 			if (PTR_ERR(page) == -ENOMEM) {
 				congestion_wait(BLK_RW_ASYNC, HZ/50);

The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
a Linux Foundation Collaborative Project.

Thanks,