io_uring: use cond_resched() in io_ring_ctx_wait_and_kill()
PATCH 1/1 io_uring: use cond_resched() in io_ring_ctx_wait_and_kill()
While working on to make io_uring sqpoll mode support syscalls that need struct files_struct, I got cpu soft lockup in io_ring_ctx_wait_and_kill(),
while (ctx->sqo_thread && !wq_has_sleeper(&ctx->sqo_wait))
cpu_relax();above loop never has an chance to exit, it's because preempt isn't enabled in the kernel, and the context calling io_ring_ctx_wait_and_kill() and io_sq_thread() run in the same cpu, if io_sq_thread calls a cond_resched() yield cpu and another context enters above loop, then io_sq_thread() will always in runqueue and never exit.
Use cond_resched() can fix this issue.
--- fs/io_uring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/io_uring.c +1 -1
@@ -7340,7 +7340,7 @@ static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)* it could cause shutdown to hang. */ while (ctx->sqo_thread && !wq_has_sleeper (&ctx->sqo_wait)) -cpu_relax (); +cond_resched ();io_kill_timeouts (ctx);io_poll_remove_all (ctx);
Signature · 1 line
2.17.2
Jens Axboe
maintainerQuoted 11 lines · Xiaoguang Wang
While working on to make
io_uringsqpoll mode support syscalls that need structfiles_struct, I got cpu soft lockup inio_ring_ctx_wait_and_kill(),while (ctx->sqo_thread && !wq_has_sleeper(&ctx->sqo_wait)) cpu_relax();above loop never has an chance to exit, it's because preempt isn't enabled in the kernel, and the context calling
io_ring_ctx_wait_and_kill()andio_sq_thread()run in the same cpu, ifio_sq_threadcalls acond_resched()yield cpu and another context enters above loop, thenio_sq_thread()will always in runqueue and never exit.Use
cond_resched()can fix this issue.
Thanks, this looks good, I believe it should also fix the syzbot reported issue so I added the tag for that.
Signature · 1 line
Jens Axboe