Re: [PATCH 5/5] io_uring: batch completion in prior_task_list

[Jens Axboe <axboe@kernel.dk>]

On 12/7/21 2:39 AM, Hao Xu wrote:
> In previous patches, we have already gathered some tw with
> io_req_task_complete() as callback in prior_task_list, let's complete
> them in batch while we cannot grab uring lock. In this way, we batch
> the req_complete_post path.
> 
> Tested-by: Pavel Begunkov <asml.silence@gmail.com>
> Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
> ---
> 
> Hi Pavel,
> May I add the above Test-by tag here?

When you fold in Pavel's fixed, please also address the below.

>  fs/io_uring.c | 70 +++++++++++++++++++++++++++++++++++++++++++--------
>  1 file changed, 60 insertions(+), 10 deletions(-)
> 
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index 21738ed7521e..f224f8df77a1 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -2225,6 +2225,49 @@ static void ctx_flush_and_put(struct io_ring_ctx *ctx, bool *locked)
>  	percpu_ref_put(&ctx->refs);
>  }
>  
> +static inline void ctx_commit_and_unlock(struct io_ring_ctx *ctx)
> +{
> +	io_commit_cqring(ctx);
> +	spin_unlock(&ctx->completion_lock);
> +	io_cqring_ev_posted(ctx);
> +}
> +
> +static void handle_prior_tw_list(struct io_wq_work_node *node, struct io_ring_ctx **ctx,
> +				 bool *uring_locked, bool *compl_locked)
> +{

Please wrap at 80 lines. And let's name this one 'handle_prev_tw_list'
instead.

-- 
Jens Axboe