Skip to contentSkip to conversation

net/rds: use wq_has_sleeper() in rds_cong_map_updated()

PATCH Applied Allison Henderson 3 participants 3 messages Aug 22 – Aug 24

Subsystems (4)
lore
Share & tools
lore

Allison Henderson

author maintainer
#0 lore raw

PATCH 1/1 net/rds: use wq_has_sleeper() in rds_cong_map_updated()

rds_cong_map_updated() runs after a peer's congestion map has been rewritten (by rds_tcp_cong_recv() and rds_ib_cong_recv(), or the clear-all in the loopback and IB send-completion paths). It bumps rds_cong_generation and then checks waitqueue_active() on map->m_waitq and on rds_poll_waitq to decide whether anyone needs waking. atomic_inc() carries no ordering and waitqueue_active() is a plain load, so nothing orders the map and generation stores before the wait queue reads. The waiters do the mirror image: rds_cong_wait() adds itself to m_waitq and then tests the port bit, and rds_poll() registers on rds_poll_waitq and then reads the generation. That is the store-buffering pattern described above waitqueue_active() in include/linux/wait.h - the updater can observe an empty wait queue while the waiter still observes the port as congested, and no wake-up is issued.

rds_cong_wait() is an interruptible sleep with no timeout, so a sender blocked on a congested port stays blocked until the next congestion update from that peer arrives or a signal is delivered. A poll() waiter misses the map-updated notification the same way.

Use wq_has_sleeper(), which is waitqueue_active() preceded by the required full barrier, as rds_tcp_state_change() already does for the same pattern.

Fixes 922cb17a5c81 ("RDS: Congestion-handling code")
Assisted-by Claude-Code:claude-fable-5

---
Raised during review of "net/rds: own the fastpath locks across connection teardown", whose first patch fixes the same pattern in release_in_xmit():
https://lore.kernel.org/netdev/[email protected]/
This patch is independent of that set and applies on its own.

 net/rds/cong.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/rds/cong.c +2 -2
index 3133b91f9e69..f7634ce3ffc1 100644
@@ -256,9 +256,9 @@ void rds_cong_map_updated(struct rds_cong_map *map, uint64_t portmask)
 	  map, &map->m_addr);
 	rds_stats_inc(s_cong_update_received);
 	atomic_inc(&rds_cong_generation);
-	if (waitqueue_active(&map->m_waitq))
+	if (wq_has_sleeper(&map->m_waitq))
 		wake_up(&map->m_waitq);
-	if (waitqueue_active(&rds_poll_waitq))
+	if (wq_has_sleeper(&rds_poll_waitq))
 		wake_up_all(&rds_poll_waitq);
 
 	if (portmask && !list_empty(&rds_cong_monitor)) {

base-commit: 4e15e89faac9f308baeb01f46c13a051814d2449

Signature · 1 line
2.25.1
reviewer Reviewed-by on v1 1/1
#1 lore raw
Quoted 27 lines · Allison Henderson

rds_cong_map_updated() runs after a peer's congestion map has been rewritten (by rds_tcp_cong_recv() and rds_ib_cong_recv(), or the clear-all in the loopback and IB send-completion paths). It bumps rds_cong_generation and then checks waitqueue_active() on map->m_waitq and on rds_poll_waitq to decide whether anyone needs waking. atomic_inc() carries no ordering and waitqueue_active() is a plain load, so nothing orders the map and generation stores before the wait queue reads. The waiters do the mirror image: rds_cong_wait() adds itself to m_waitq and then tests the port bit, and rds_poll() registers on rds_poll_waitq and then reads the generation. That is the store-buffering pattern described above waitqueue_active() in include/linux/wait.h - the updater can observe an empty wait queue while the waiter still observes the port as congested, and no wake-up is issued.

rds_cong_wait() is an interruptible sleep with no timeout, so a sender blocked on a congested port stays blocked until the next congestion update from that peer arrives or a signal is delivered. A poll() waiter misses the map-updated notification the same way.

Use wq_has_sleeper(), which is waitqueue_active() preceded by the required full barrier, as rds_tcp_state_change() already does for the same pattern.

Fixes 922cb17a5c81 ("RDS: Congestion-handling code")
Assisted-by Claude-Code:claude-fable-5
Quoted 6 lines

---
Raised during review of "net/rds: own the fastpath locks across connection teardown", whose first patch fixes the same pattern in release_in_xmit():
https://lore.kernel.org/netdev/[email protected]/
This patch is independent of that set and applies on its own.

#2 lore raw
In reply to Allison Henderson
This patch was applied to netdev/net.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <[email protected]>:

Quoted 15 lines · On Fri, 21 Aug 2026 22:26:47 -0700 you wrote:

rds_cong_map_updated() runs after a peer's congestion map has been rewritten (by rds_tcp_cong_recv() and rds_ib_cong_recv(), or the clear-all in the loopback and IB send-completion paths). It bumps rds_cong_generation and then checks waitqueue_active() on map->m_waitq and on rds_poll_waitq to decide whether anyone needs waking. atomic_inc() carries no ordering and waitqueue_active() is a plain load, so nothing orders the map and generation stores before the wait queue reads. The waiters do the mirror image: rds_cong_wait() adds itself to m_waitq and then tests the port bit, and rds_poll() registers on rds_poll_waitq and then reads the generation. That is the store-buffering pattern described above waitqueue_active() in include/linux/wait.h - the updater can observe an empty wait queue while the waiter still observes the port as congested, and no wake-up is issued.

Quoted 1 line

[...]

Here is the summary with links:
- [net] net/rds: use wq_has_sleeper() in rds_cong_map_updated()

    https://git.kernel.org/netdev/net/c/d4f484661961

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html