From: Jens Axboe <axboe@kernel.dk>
Date: Tue, 24 Oct 2023 20:44:21 +0000 (-0600)
Subject: Revert "io_uring/rw: disable IOCB_DIO_CALLER_COMP"
X-Git-Url: https://git.kernel.dk/?a=commitdiff_plain;h=refs%2Fheads%2Finode-dio-wait;p=linux-2.6-block.git

Revert "io_uring/rw: disable IOCB_DIO_CALLER_COMP"

This reverts commit 9f4e9d79e89bd7ffc1dfbe7901f2b1f19d73c1d4.

Now that inode_dio_wait() with deferred task_work is properly handled,
we can re-introduce this optimization.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---

diff --git a/io_uring/rw.c b/io_uring/rw.c
index 64390d4e20c1..ed321ac2e157 100644
--- a/io_uring/rw.c
+++ b/io_uring/rw.c
@@ -1025,6 +1025,15 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
 		kiocb_start_write(kiocb);
 	kiocb->ki_flags |= IOCB_WRITE;
 
+	/*
+	 * For non-polled IO, set IOCB_DIO_CALLER_COMP, stating that our handler
+	 * groks deferring the completion to task context. This isn't
+	 * necessary and useful for polled IO as that can always complete
+	 * directly.
+	 */
+	if (!(kiocb->ki_flags & IOCB_HIPRI))
+		kiocb->ki_flags |= IOCB_DIO_CALLER_COMP;
+
 	if (likely(req->file->f_op->write_iter))
 		ret2 = call_write_iter(req->file, kiocb, &s->iter);
 	else if (req->file->f_op->write)