Skip to contentSkip to conversation

io_uring: honor I/O nowait flag for read/write

RFC PATCH In review Chaitanya Kulkarni 2 participants 3 messages Apr 21, 2023 io_uring

lore
Share & tools
lore
#0 lore raw

cover io_uring: honor I/O nowait flag for read/write

Hi,

When IO_URING_F_NONBLOCK is set on io_kiocb req->flag in io_write() or io_read() IOCB_NOWAIT is set for kiocb when passed it to the respective rw_iter callback. This sets REQ_NOWAIT for underlaying I/O. The result is low level driver always sees block layer request as REQ_NOWAIT even if user has submitted request with nowait = 0 e.g. fio nowait=0.

That is not consistent behaviour with other fio ioengine such as libaio as it will issue non REQ_NOWAIT request with REQ_NOWAIT:-

libaio nowait = 0:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=FALSE* REQ_OP_WRITE

libaio nowait = 1:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

* Without this patch with fio ioengine io_uring :-
---------------------------------------------------

iouring nowait = 0:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

iouring nowait = 1:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

* With this patch with fio ioengine io_uring :-
---------------------------------------------------

iouring nowait = 0:-
null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=FALSE* WRITE

iouring nowait = 1:
null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=TRUE* WRITE

Instead of only relying on IO_URING_F_NONBLOCK blindly in io_read() and io_write(), also make sure io_kiocb->io_rw->flags is set to RWF_NOWAIT before we mark kiocb->ki_flags = IOCB_NOWAIT.

Below is the deatailed testing log.

-ck

Chaitanya Kulkarni (1):
io_uring: honor I/O nowait flag for read/write

 io_uring/rw.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
$ sh test-iouring-no-wait
* Without this patch :-
#######################
-------------------libaio nowait = 0--------------------------

++ grep -e ioengine -e rw -e nowait fio/randwrite-libaio-nowait-0.fio ioengine=libaio
rw=randwrite
nowait=0 <-----------------------------
overwrite=0
++ fio fio/randwrite-libaio-nowait-0.fio --filename=/dev/nullb0 ++ grep err
RANDWRITE: (groupid=0, jobs=1): err= 0: pid=40835: Thu Apr 20 21:31:09 2023 ++ dmesg -c

[17978.613789] null_blk: disk nullb0 created
[17978.613814] null_blk: module loaded
__________________________________________________________________________
[17978.796560] null_blk: fio:null_handle_rq 1288 *NOWAIT=FALSE* REQ_OP_WRITE

-------------------libaio nowait = 1--------------------------

++ blkdiscard /dev/nullb0
++ grep -e ioengine -e rw -e nowait fio/randwrite-libaio-nowait-1.fio ioengine=libaio
rw=randwrite
nowait=1 <-----------------------------
overwrite=0
++ fio fio/randwrite-libaio-nowait-1.fio --filename=/dev/nullb0 ++ grep err
RANDWRITE: (groupid=0, jobs=1): err= 0: pid=40842: Thu Apr 20 21:31:09 2023 ++ dmesg -c
__________________________________________________________________________

[17979.019595] null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

-------------------iouring nowait = 0--------------------------

++ blkdiscard /dev/nullb0
++ grep -e ioengine -e rw -e nowait fio/randwrite-iouring-nowait-0.fio ioengine=io_uring
rw=randwrite
nowait=0 <-----------------------------
overwrite=0
++ fio fio/randwrite-iouring-nowait-0.fio --filename=/dev/nullb0 ++ grep err
RANDWRITE: (groupid=0, jobs=1): err= 0: pid=40849: Thu Apr 20 21:31:10 2023 ++ dmesg -c
__________________________________________________________________________

[17979.242849] null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

-------------------iouring nowait = 1--------------------------

++ blkdiscard /dev/nullb0
++ grep -e ioengine -e rw -e nowait fio/randwrite-iouring-nowait-1.fio ioengine=io_uring
rw=randwrite
nowait=1 <-----------------------------
overwrite=0
++ fio fio/randwrite-iouring-nowait-1.fio --filename=/dev/nullb0 ++ grep err
RANDWRITE: (groupid=0, jobs=1): err= 0: pid=40856: Thu Apr 20 21:31:10 2023 ++ dmesg -c
__________________________________________________________________________

[17979.454102] null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

* With this patch :-
####################
-------------------iouring nowait = 0--------------------------
+ blkdiscard /dev/nullb0
+ grep -e ioengine -e rw -e nowait fio/randwrite-iouring-nowait-0.fio ioengine=io_uring
rw=randwrite
nowait=0 <-----------------------------
overwrite=0
+ fio fio/randwrite-iouring-nowait-0.fio --filename=/dev/nullb0
+ grep err
RANDWRITE: (groupid=0, jobs=1): err= 0: pid=2788: Thu Apr 20 23:35:40 2023
+ dmesg -c
__________________________________________________________________________

[  164.255136] null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=FALSE* WRITE

-------------------iouring nowait = 1--------------------------
+ blkdiscard /dev/nullb0
+ grep -e ioengine -e rw -e nowait fio/randwrite-iouring-nowait-1.fio ioengine=io_uring
rw=randwrite
nowait=1 <-----------------------------
overwrite=0
+ fio fio/randwrite-iouring-nowait-1.fio --filename=/dev/nullb0
+ grep err
RANDWRITE: (groupid=0, jobs=1): err= 0: pid=2795: Thu Apr 20 23:35:41 2023
+ dmesg -c
__________________________________________________________________________

[  164.467420] null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=TRUE* WRITE
Signature · 1 line
2.40.0
#1 lore raw

RFC PATCH 1/1 io_uring: honor I/O nowait flag for read/write

When IO_URING_F_NONBLOCK is set on io_kiocb req->flag in io_write() or io_read() IOCB_NOWAIT is set for kiocb when passed it to the respective rw_iter callback. This sets REQ_NOWAIT for underlaying I/O. The result is low level driver always sees block layer request as REQ_NOWAIT even if user has submitted request with nowait = 0 e.g. fio nowait=0.

That is not consistent behaviour with other fio ioengine such as libaio as it will issue non REQ_NOWAIT request with REQ_NOWAIT:-

libaio nowait = 0:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=FALSE* REQ_OP_WRITE

libaio nowait = 1:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

* Without this patch with fio ioengine io_uring :-
---------------------------------------------------

iouring nowait = 0:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

iouring nowait = 1:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

* With this patch with fio ioengine io_uring :-
---------------------------------------------------

iouring nowait = 0:-
null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=FALSE* WRITE

iouring nowait = 1:
null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=TRUE* WRITE

Instead of only relying on IO_URING_F_NONBLOCK blindly in io_read() and io_write(), also make sure io_kiocb->io_rw->flags is set to RWF_NOWAIT before we mark kiocb->ki_flags = IOCB_NOWAIT.

---
 io_uring/rw.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
io_uring/rw.c +2 -2
index 3f118ed46e4f..4b3a2c1df5f2 100644
@@ -745,7 +745,7 @@ int io_read(struct io_kiocb *req, unsigned int issue_flags)
 	}
 	req->cqe.res = iov_iter_count(&s->iter);
 
-	if (force_nonblock) {
+	if (force_nonblock && (rw->flags & RWF_NOWAIT)) {
 		/* If the file doesn't support async, just async punt */
 		if (unlikely(!io_file_supports_nowait(req))) {
 			ret = io_setup_async_rw(req, iovec, s, true);
@@ -877,7 +877,7 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
 	}
 	req->cqe.res = iov_iter_count(&s->iter);
 
-	if (force_nonblock) {
+	if (force_nonblock && (rw->flags & RWF_NOWAIT)) {
 		/* If the file doesn't support async, just async punt */
 		if (unlikely(!io_file_supports_nowait(req)))
 			goto copy_iov;
Signature · 1 line
2.40.0
maintainer
#2 lore raw
Quoted 56 lines · Chaitanya Kulkarni

When IO_URING_F_NONBLOCK is set on io_kiocb req->flag in io_write() or io_read() IOCB_NOWAIT is set for kiocb when passed it to the respective rw_iter callback. This sets REQ_NOWAIT for underlaying I/O. The result is low level driver always sees block layer request as REQ_NOWAIT even if user has submitted request with nowait = 0 e.g. fio nowait=0.

That is not consistent behaviour with other fio ioengine such as libaio as it will issue non REQ_NOWAIT request with REQ_NOWAIT:-

libaio nowait = 0:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=FALSE* REQ_OP_WRITE

libaio nowait = 1:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

* Without this patch with fio ioengine io_uring :-
---------------------------------------------------

iouring nowait = 0:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

iouring nowait = 1:-
null_blk: fio:null_handle_rq 1288 *NOWAIT=TRUE* REQ_OP_WRITE

* With this patch with fio ioengine io_uring :-
---------------------------------------------------

iouring nowait = 0:-
null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=FALSE* WRITE

iouring nowait = 1:
null_blk: fio:null_handle_rq 1307 *REQ_NOWAIT=TRUE* WRITE

Instead of only relying on IO_URING_F_NONBLOCK blindly in io_read() and io_write(), also make sure io_kiocb->io_rw->flags is set to RWF_NOWAIT before we mark kiocb->ki_flags = IOCB_NOWAIT.

---
 io_uring/rw.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
io_uring/rw.c +1 -1
index 3f118ed46e4f..4b3a2c1df5f2 100644
@@ -745,7 +745,7 @@ int io_read(struct io_kiocb *req, unsigned int issue_flags)
 	}
 	req->cqe.res = iov_iter_count(&s->iter);
 
-	if (force_nonblock) {
+	if (force_nonblock && (rw->flags & RWF_NOWAIT)) {
 		/* If the file doesn't support async, just async punt */
 		if (unlikely(!io_file_supports_nowait(req))) {
 			ret = io_setup_async_rw(req, iovec, s, true);

io_uring/rw.c int io_write(struct io_kiocb *req, unsigned int issue_flags) · line 883

Quoted 8 lines
io_uring/rw.c +1 -1
@@ -877,7 +877,7 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
 	}
 	req->cqe.res = iov_iter_count(&s->iter);
 
-	if (force_nonblock) {
+	if (force_nonblock && (rw->flags & RWF_NOWAIT)) {
 		/* If the file doesn't support async, just async punt */
 		if (unlikely(!io_file_supports_nowait(req)))
 			goto copy_iov;

This is wrong. libaio doesn't care if it blocks for submission, and this is actually one of the complains of aio/libaio. Is it async? Maybe? io_uring, on the other hand, tries much harder to not block for submission. Because if you do block, you've now starved your issue pipeline. And how do you do that? By always having NOWAIT set on the request, unless you are in a context (io-wq) where you want to block in case you have to.

This has _nothing_ to do with the user setting RWF_NOWAIT or not, only change for that is that if we did have that set, then we should obviously not retry from blocking context. Rather, the io should be complete with whatever we originally got (typically -EAGAIN).

Signature · 1 line
Jens Axboe