Skip to contentSkip to conversation

sunrpc: start cache request seqno at 1 to fix netlink GET_REQS

PATCH In review Jeff Layton 2 participants 2 messages Apr 11 – Apr 13

Subsystems (3)
lore
Share & tools
lore

Jeff Layton

author maintainer
#0 lore raw

PATCH 1/1 sunrpc: start cache request seqno at 1 to fix netlink GET_REQS

sunrpc_cache_requests_snapshot() filters requests with crq->seqno <= min_seqno. The min_seqno for the first netlink dump call is cb->args[0] which is 0. Since next_seqno was initialized to 0, the very first cache request got seqno=0 and was silently skipped by the snapshot (0 <= 0 is true).

This caused netlink-based GET_REQS to return 0 pending requests even when a request was queued, preventing mountd from resolving cache entries (particularly expkey/nfsd.fh). The unresolved CACHE_PENDING state blocked all further notifications for the entry, leading to permanent NFS4ERR_DELAY hangs.

Start next_seqno at 1 so all requests have seqno >= 1 and pass the snapshot filter when min_seqno is 0.

Fixes facc4e3c8042 ("sunrpc: split cache_detail queue into request and reader lists")

---
I started hitting a persistent hang in mountd upcalls just after reboot, and this turns out to be the cause. It's probably best to fold this into the patch in the Fixes: line.

---
 net/sunrpc/cache.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/sunrpc/cache.c +1 -1
index d477b19dbfa1..305c6e67f052 100644
@@ -405,7 +405,7 @@ void sunrpc_init_cache_detail(struct cache_detail *cd)
 	INIT_LIST_HEAD(&cd->readers);
 	spin_lock_init(&cd->queue_lock);
 	init_waitqueue_head(&cd->queue_wait);
-	cd->next_seqno = 0;
+	cd->next_seqno = 1;
 	spin_lock(&cache_list_lock);
 	cd->nextcheck = 0;
 	cd->entries = 0;

---
base-commit: 68f3218e45ab644ed37d5020a4a25e523fc0e30e
change-id: 20260411-exportd-nl-2cd2e9d451bc

Best regards,

Signature · 1 line
Jeff Layton <[email protected]>
maintainer
#1 lore raw

From: Chuck Lever <[email protected]>

Quoted 6 lines · Jeff Layton

sunrpc_cache_requests_snapshot() filters requests with crq->seqno <= min_seqno. The min_seqno for the first netlink dump call is cb->args[0] which is 0. Since next_seqno was initialized to 0, the very first cache request got seqno=0 and was silently skipped by the snapshot (0 <= 0 is true).

Quoted 7 lines

This caused netlink-based GET_REQS to return 0 pending requests even when a request was queued, preventing mountd from resolving cache entries (particularly expkey/nfsd.fh). The unresolved CACHE_PENDING state blocked all further notifications for the entry, leading to permanent NFS4ERR_DELAY hangs.

[...]

Applied to nfsd-testing, thanks!

[1/1] sunrpc: start cache request seqno at 1 to fix netlink GET_REQS

      commit: 2dd2484661eb089e5eb2ba9da4b87decb0f3be36

--
Chuck Lever <[email protected]>