| Message ID | 1a79362b9c10b8523ef70b061d96523650a23344.1672795998.git.asml.silence@gmail.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | [RFC,1/1] io_uring: pin context while queueing deferred tw | expand |
On 1/4/23 01:34, Pavel Begunkov wrote: Damn, the subj is wrong, it's not an RFC. Same with the other patch > Unlike normal tw, nothing prevents deferred tw to be executed right > after an tw item added to ->work_llist in io_req_local_work_add(). For > instance, the waiting task may get waken up by CQ posting or a normal > tw. Thus we need to pin the ring for the rest of io_req_local_work_add() > > Cc: stable@vger.kernel.org > Fixes: c0e0d6ba25f18 ("io_uring: add IORING_SETUP_DEFER_TASKRUN")
On Wed, 04 Jan 2023 01:34:02 +0000, Pavel Begunkov wrote: > Unlike normal tw, nothing prevents deferred tw to be executed right > after an tw item added to ->work_llist in io_req_local_work_add(). For > instance, the waiting task may get waken up by CQ posting or a normal > tw. Thus we need to pin the ring for the rest of io_req_local_work_add() > > Applied, thanks! [1/1] io_uring: pin context while queueing deferred tw (no commit info) Best regards,
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 58ac13b69dc8..6bed44855679 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -1236,13 +1236,18 @@ static void io_req_local_work_add(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; - if (!llist_add(&req->io_task_work.node, &ctx->work_llist)) + percpu_ref_get(&ctx->refs); + + if (!llist_add(&req->io_task_work.node, &ctx->work_llist)) { + percpu_ref_put(&ctx->refs); return; + } /* need it for the following io_cqring_wake() */ smp_mb__after_atomic(); if (unlikely(atomic_read(&req->task->io_uring->in_idle))) { io_move_task_work_from_local(ctx); + percpu_ref_put(&ctx->refs); return; } @@ -1252,6 +1257,7 @@ static void io_req_local_work_add(struct io_kiocb *req) if (ctx->has_evfd) io_eventfd_signal(ctx); __io_cqring_wake(ctx); + percpu_ref_put(&ctx->refs); } void __io_req_task_work_add(struct io_kiocb *req, bool allow_local)
Unlike normal tw, nothing prevents deferred tw to be executed right after an tw item added to ->work_llist in io_req_local_work_add(). For instance, the waiting task may get waken up by CQ posting or a normal tw. Thus we need to pin the ring for the rest of io_req_local_work_add() Cc: stable@vger.kernel.org Fixes: c0e0d6ba25f18 ("io_uring: add IORING_SETUP_DEFER_TASKRUN") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> --- io_uring/io_uring.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)