Re: [PATCH for-next] io_uring: fix CQE reordering

From: Jens Axboe <axboe@kernel.dk>
To: Pavel Begunkov <asml.silence@gmail.com>, io-uring@vger.kernel.org
Cc: Dylan Yudaken <dylany@fb.com>
Subject: Re: [PATCH for-next] io_uring: fix CQE reordering
Date: Fri, 23 Sep 2022 08:19:49 -0600	[thread overview]
Message-ID: <30718187-8558-4b32-b8d1-5c1f4f322d4f@kernel.dk> (raw)
In-Reply-To: <ec3bc55687b0768bbe20fb62d7d06cfced7d7e70.1663892031.git.asml.silence@gmail.com>

On 9/23/22 7:53 AM, Pavel Begunkov wrote:
> Overflowing CQEs may result in reordeing, which is buggy in case of
> links, F_MORE and so.
> 
> Reported-by: Dylan Yudaken <dylany@fb.com>
> Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
> ---
>  io_uring/io_uring.c | 12 ++++++++++--
>  io_uring/io_uring.h | 12 +++++++++---
>  2 files changed, 19 insertions(+), 5 deletions(-)
> 
> diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
> index f359e24b46c3..62d1f55fde55 100644
> --- a/io_uring/io_uring.c
> +++ b/io_uring/io_uring.c
> @@ -609,7 +609,7 @@ static bool __io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force)
>  
>  	io_cq_lock(ctx);
>  	while (!list_empty(&ctx->cq_overflow_list)) {
> -		struct io_uring_cqe *cqe = io_get_cqe(ctx);
> +		struct io_uring_cqe *cqe = io_get_cqe_overflow(ctx, true);
>  		struct io_overflow_cqe *ocqe;
>  
>  		if (!cqe && !force)
> @@ -736,12 +736,19 @@ bool io_req_cqe_overflow(struct io_kiocb *req)
>   * control dependency is enough as we're using WRITE_ONCE to
>   * fill the cq entry
>   */
> -struct io_uring_cqe *__io_get_cqe(struct io_ring_ctx *ctx)
> +struct io_uring_cqe *__io_get_cqe(struct io_ring_ctx *ctx, bool overflow)
>  {
>  	struct io_rings *rings = ctx->rings;
>  	unsigned int off = ctx->cached_cq_tail & (ctx->cq_entries - 1);
>  	unsigned int free, queued, len;
>  
> +	/*
> +	 * Posting into the CQ when there are pending overflowed CQEs may break
> +	 * ordering guarantees, which will affect links, F_MORE users and more.
> +	 * Force overflow the completion.
> +	 */
> +	if (!overflow && (ctx->check_cq & BIT(IO_CHECK_CQ_OVERFLOW_BIT)))
> +		return NULL;

Rather than pass this bool around for the hot path, why not add a helper
for the case where 'overflow' isn't known? That can leave the regular
io_get_cqe() avoiding this altogether.

-- 
Jens Axboe