ublk: build batch from IOs in same io_ring_ctx and io task

ublk_queue_cmd_list() dispatches the whole batch list by scheduling task
work via the tail request's io_uring_cmd, this way is fine even though
more than one io_ring_ctx are involved for this batch since it is just
one running context.

However, the task work handler ublk_cmd_list_tw_cb() takes `issue_flags`
of tail uring_cmd's io_ring_ctx for completing all commands. This way is
wrong if any uring_cmd is issued from different io_ring_ctx.

Fixes it by always building batch IOs from same io_ring_ctx and io task
because ublk_dispatch_req() does validate task context, and IO needs to
be aborted in case of running from fallback task work context.

For typical per-queue or per-io daemon implementation, this way shouldn't
make difference from performance viewpoint, because single io_ring_ctx is
taken in each daemon for normal use case.

Fixes: d796cea ("ublk: implement ->queue_rqs()")
Signed-off-by: Ming Lei <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

Author: Ming Lei

drivers/block/ublk_drv.c

@@ -1416,6 +1416,14 @@ static blk_status_t ublk_queue_rq(struct blk_mq_hw_ctx *hctx,
 	return BLK_STS_OK;
 }
 
+static inline bool ublk_belong_to_same_batch(const struct ublk_io *io,
+					     const struct ublk_io *io2)
+{
+	return (io_uring_cmd_ctx_handle(io->cmd) ==
+		io_uring_cmd_ctx_handle(io2->cmd)) &&
+		(io->task == io2->task);
+}
+
 static void ublk_queue_rqs(struct rq_list *rqlist)
 {
 	struct rq_list requeue_list = { };
@@ -1427,7 +1435,8 @@ static void ublk_queue_rqs(struct rq_list *rqlist)
 		struct ublk_queue *this_q = req->mq_hctx->driver_data;
 		struct ublk_io *this_io = &this_q->ios[req->tag];
 
-		if (io && io->task != this_io->task && !rq_list_empty(&submit_list))
+		if (io && !ublk_belong_to_same_batch(io, this_io) &&
+				!rq_list_empty(&submit_list))
 			ublk_queue_cmd_list(io, &submit_list);
 		io = this_io;