From: Ming Lei <[email protected]>
To: Jens Axboe <[email protected]>,
[email protected], [email protected]
Cc: [email protected],
Miklos Szeredi <[email protected]>,
ZiyangZhang <[email protected]>,
Xiaoguang Wang <[email protected]>,
Bernd Schubert <[email protected]>,
Pavel Begunkov <[email protected]>,
Stefan Hajnoczi <[email protected]>,
Dan Williams <[email protected]>,
Ming Lei <[email protected]>
Subject: [PATCH V6 02/17] io_uring: use ctx->cached_sq_head to calculate left sqes
Date: Thu, 30 Mar 2023 19:36:15 +0800 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
Use ctx->cached_sq_head to calculate 'left' sqes, and prepare for supporting
fused requests, which need to get req/sqe in its own ->prep() callback.
ctx->cached_sq_head should always be cached or to be fetched, so this change
is just fine.
Signed-off-by: Ming Lei <[email protected]>
---
io_uring/io_uring.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 693558c4b10b..25a940f0ab68 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -2429,15 +2429,16 @@ int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr)
__must_hold(&ctx->uring_lock)
{
unsigned int entries = io_sqring_entries(ctx);
- unsigned int left;
+ unsigned old_head = ctx->cached_sq_head;
+ unsigned int left = 0;
int ret;
if (unlikely(!entries))
return 0;
/* make sure SQ entry isn't read before tail */
- ret = left = min3(nr, ctx->sq_entries, entries);
- io_get_task_refs(left);
- io_submit_state_start(&ctx->submit_state, left);
+ ret = min3(nr, ctx->sq_entries, entries);
+ io_get_task_refs(ret);
+ io_submit_state_start(&ctx->submit_state, ret);
do {
const struct io_uring_sqe *sqe;
@@ -2456,11 +2457,12 @@ int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr)
*/
if (unlikely(io_submit_sqe(ctx, req, sqe)) &&
!(ctx->flags & IORING_SETUP_SUBMIT_ALL)) {
- left--;
+ left = 1;
break;
}
- } while (--left);
+ } while ((ctx->cached_sq_head - old_head) < ret);
+ left = ret - (ctx->cached_sq_head - old_head) - left;
if (unlikely(left)) {
ret -= left;
/* try again if it submitted nothing and can't allocate a req */
--
2.39.2
next prev parent reply other threads:[~2023-03-30 11:38 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-30 11:36 [PATCH V6 00/17] io_uring/ublk: add generic IORING_OP_FUSED_CMD Ming Lei
2023-03-30 11:36 ` [PATCH V6 01/17] io_uring: increase io_kiocb->flags into 64bit Ming Lei
2023-03-30 11:36 ` Ming Lei [this message]
2023-03-30 11:36 ` [PATCH V6 03/17] io_uring: add generic IORING_OP_FUSED_CMD Ming Lei
2023-04-01 14:35 ` Ming Lei
2023-03-30 11:36 ` [PATCH V6 04/17] io_uring: support providing buffer by IORING_OP_FUSED_CMD Ming Lei
2023-03-30 11:36 ` [PATCH V6 05/17] io_uring: support OP_READ/OP_WRITE for fused secondary request Ming Lei
2023-03-30 11:36 ` [PATCH V6 06/17] io_uring: support OP_SEND_ZC/OP_RECV " Ming Lei
2023-03-30 11:36 ` [PATCH V6 07/17] block: ublk_drv: add common exit handling Ming Lei
2023-03-30 11:36 ` [PATCH V6 08/17] block: ublk_drv: don't consider flush request in map/unmap io Ming Lei
2023-03-30 11:36 ` [PATCH V6 09/17] block: ublk_drv: add two helpers to clean up map/unmap request Ming Lei
2023-03-30 11:36 ` [PATCH V6 10/17] block: ublk_drv: clean up several helpers Ming Lei
2023-03-30 11:36 ` [PATCH V6 11/17] block: ublk_drv: cleanup 'struct ublk_map_data' Ming Lei
2023-03-30 11:36 ` [PATCH V6 12/17] block: ublk_drv: cleanup ublk_copy_user_pages Ming Lei
2023-03-31 16:22 ` Bernd Schubert
2023-03-30 11:36 ` [PATCH V6 13/17] block: ublk_drv: grab request reference when the request is handled by userspace Ming Lei
2023-03-30 11:36 ` [PATCH V6 14/17] block: ublk_drv: support to copy any part of request pages Ming Lei
2023-03-30 11:36 ` [PATCH V6 15/17] block: ublk_drv: add read()/write() support for ublk char device Ming Lei
2023-03-30 11:36 ` [PATCH V6 16/17] block: ublk_drv: don't check buffer in case of zero copy Ming Lei
2023-03-30 11:36 ` [PATCH V6 17/17] block: ublk_drv: apply io_uring FUSED_CMD for supporting " Ming Lei
2023-03-31 19:13 ` Bernd Schubert
2023-04-01 13:19 ` Ming Lei
2023-03-31 19:55 ` Bernd Schubert
2023-04-01 13:22 ` Ming Lei
2023-04-03 9:25 ` Bernd Schubert
2023-04-03 1:11 ` [PATCH V6 00/17] io_uring/ublk: add generic IORING_OP_FUSED_CMD Ming Lei
2023-04-03 1:24 ` Jens Axboe
2023-04-04 7:48 ` Ming Lei
2023-04-03 1:23 ` (subset) " Jens Axboe
2023-04-18 19:38 ` Bernd Schubert
2023-04-19 1:51 ` Ming Lei
2023-04-19 9:56 ` Bernd Schubert
2023-04-19 11:19 ` Ming Lei
2023-04-19 15:42 ` Bernd Schubert
2023-04-20 1:18 ` Pavel Begunkov
2023-04-20 1:38 ` Ming Lei
2023-04-21 22:38 ` Bernd Schubert
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox