io_uring/timeout: don't put a sequence target behind the last flush

io_timeout_fn() and io_kill_timeout() bump ->cq_timeouts when a timeout
fires, but its CQE is only posted later from task_work. Until then the
sequence io_timeout() computes is one below the last flush per fired
timeout. Now that ->cq_last_tm_flush is kept while sequence timeouts are
pending, a timeout armed in that window with a small offset gets a target
behind the base. Its distance from the base wraps, and it's never
completed by events, only by its timer. With DEFER_TASKRUN that window
is as wide as the application makes it, as submission runs before the
local task_work.

Nothing but those in flight CQEs separates the two, so use the last
flush as the base for the target when the sequence is behind it.

Fixes: c6c0b3eba8 ("io_uring/timeout: don't skip over reached sequences")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Jens Axboe 2026-10-15 14:18:22 -06:00
parent 937d38dcac
commit fdfce905a5

View File

@ -690,7 +690,6 @@ int io_timeout(struct io_kiocb *req, unsigned int issue_flags)
}
tail = data_race(ctx->cached_cq_tail) - atomic_read(&ctx->cq_timeouts);
timeout->target_seq = tail + off;
/*
* Update the last seq here in case io_flush_timeouts() hasn't, it only
@ -704,6 +703,15 @@ int io_timeout(struct io_kiocb *req, unsigned int issue_flags)
ctx->cq_last_tm_flush = tail;
base = ctx->cq_last_tm_flush;
/*
* A fired timeout bumps ->cq_timeouts before its CQE is posted, which
* leaves the sequence behind the last flush until it is. Nothing else
* separates the two then, so count from the flush.
*/
if ((s32)(tail - base) < 0)
tail = base;
timeout->target_seq = tail + off;
/*
* Insertion sort, ensuring the first entry in the list is always
* the one we need first.