Re: [PATCH 07/18] io_uring: support for IO polling

On 1/28/19 8:02 AM, Christoph Hellwig wrote:

quoted

 	struct {
 		spinlock_t		completion_lock;
+		unsigned		poll_multi_file;

This seems to be used as a bool.

Fixed

quoted

-static void __io_cqring_add_event(struct io_ring_ctx *ctx, u64 ki_user_data,
-				  long res, unsigned ev_flags)
+static void io_cqring_fill_event(struct io_ring_ctx *ctx, u64 ki_user_data,
+				 long res, unsigned ev_flags)
 {
 	struct io_uring_cqe *cqe;
 

@@ -206,9 +213,15 @@ static void __io_cqring_add_event(struct io_ring_ctx *ctx, u64 ki_user_data,
 		cqe->user_data = ki_user_data;
 		cqe->res = res;
 		cqe->flags = ev_flags;
-		io_commit_cqring(ctx);
 	} else
 		ctx->cq_ring->overflow++;
+}

Maybe we should have created this low-level helper already in the
patch introducing the code?

I think that can go both ways. If we create it early, it'll just be a
pointless helper. Don't really care too much about this particular case.

-- 
Jens Axboe