From: Pavel Begunkov <[email protected]>
To: [email protected]
Cc: Jens Axboe <[email protected]>, [email protected]
Subject: [PATCH for-next v3 02/16] io_uring: kill REQ_F_COMPLETE_INLINE
Date: Thu, 16 Jun 2022 10:21:58 +0100 [thread overview]
Message-ID: <600ba20a9338b8a39b249b23d3d177803613dde4.1655371007.git.asml.silence@gmail.com> (raw)
In-Reply-To: <[email protected]>
REQ_F_COMPLETE_INLINE is only needed to delay queueing into the
completion list to io_queue_sqe() as __io_req_complete() is inlined and
we don't want to bloat the kernel.
As now we complete in a more centralised fashion in io_issue_sqe() we
can get rid of the flag and queue to the list directly.
Signed-off-by: Pavel Begunkov <[email protected]>
---
io_uring/io_uring.c | 18 +++++++-----------
io_uring/io_uring.h | 5 -----
io_uring/io_uring_types.h | 3 ---
3 files changed, 7 insertions(+), 19 deletions(-)
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index f1ecfdd0166e..02a70e7eb774 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -742,10 +742,7 @@ void io_req_complete_post(struct io_kiocb *req)
inline void __io_req_complete(struct io_kiocb *req, unsigned issue_flags)
{
- if (issue_flags & IO_URING_F_COMPLETE_DEFER)
- req->flags |= REQ_F_COMPLETE_INLINE;
- else
- io_req_complete_post(req);
+ io_req_complete_post(req);
}
void io_req_complete_failed(struct io_kiocb *req, s32 res)
@@ -1581,9 +1578,12 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
if (creds)
revert_creds(creds);
- if (ret == IOU_OK)
- __io_req_complete(req, issue_flags);
- else if (ret != IOU_ISSUE_SKIP_COMPLETE)
+ if (ret == IOU_OK) {
+ if (issue_flags & IO_URING_F_COMPLETE_DEFER)
+ io_req_add_compl_list(req);
+ else
+ io_req_complete_post(req);
+ } else if (ret != IOU_ISSUE_SKIP_COMPLETE)
return ret;
/* If the op doesn't have a file, we're not polling for it */
@@ -1748,10 +1748,6 @@ static inline void io_queue_sqe(struct io_kiocb *req)
ret = io_issue_sqe(req, IO_URING_F_NONBLOCK|IO_URING_F_COMPLETE_DEFER);
- if (req->flags & REQ_F_COMPLETE_INLINE) {
- io_req_add_compl_list(req);
- return;
- }
/*
* We async punt it if the file wasn't marked NOWAIT, or if the file
* doesn't support non-blocking read/write attempts
diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h
index e30e639c2822..3f6cad3d356c 100644
--- a/io_uring/io_uring.h
+++ b/io_uring/io_uring.h
@@ -217,11 +217,6 @@ static inline bool io_run_task_work(void)
return false;
}
-static inline void io_req_complete_state(struct io_kiocb *req)
-{
- req->flags |= REQ_F_COMPLETE_INLINE;
-}
-
static inline void io_tw_lock(struct io_ring_ctx *ctx, bool *locked)
{
if (!*locked) {
diff --git a/io_uring/io_uring_types.h b/io_uring/io_uring_types.h
index ef1cf86e8932..4576ea8cad2e 100644
--- a/io_uring/io_uring_types.h
+++ b/io_uring/io_uring_types.h
@@ -301,7 +301,6 @@ enum {
REQ_F_POLLED_BIT,
REQ_F_BUFFER_SELECTED_BIT,
REQ_F_BUFFER_RING_BIT,
- REQ_F_COMPLETE_INLINE_BIT,
REQ_F_REISSUE_BIT,
REQ_F_CREDS_BIT,
REQ_F_REFCOUNT_BIT,
@@ -356,8 +355,6 @@ enum {
REQ_F_BUFFER_SELECTED = BIT(REQ_F_BUFFER_SELECTED_BIT),
/* buffer selected from ring, needs commit */
REQ_F_BUFFER_RING = BIT(REQ_F_BUFFER_RING_BIT),
- /* completion is deferred through io_comp_state */
- REQ_F_COMPLETE_INLINE = BIT(REQ_F_COMPLETE_INLINE_BIT),
/* caller should reissue async */
REQ_F_REISSUE = BIT(REQ_F_REISSUE_BIT),
/* supports async reads/writes */
--
2.36.1
next prev parent reply other threads:[~2022-06-16 9:22 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-16 9:21 [PATCH for-next v3 00/16] 5.20 cleanups and poll optimisations Pavel Begunkov
2022-06-16 9:21 ` [PATCH for-next v3 01/16] io_uring: rw: delegate sync completions to core io_uring Pavel Begunkov
2022-06-16 9:21 ` Pavel Begunkov [this message]
2022-06-16 9:21 ` [PATCH for-next v3 03/16] io_uring: refactor io_req_task_complete() Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 04/16] io_uring: don't inline io_put_kbuf Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 05/16] io_uring: poll: remove unnecessary req->ref set Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 06/16] io_uring: switch cancel_hash to use per entry spinlock Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 07/16] io_uring: pass poll_find lock back Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 08/16] io_uring: clean up io_try_cancel Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 09/16] io_uring: limit the number of cancellation buckets Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 10/16] io_uring: clean up io_ring_ctx_alloc Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 11/16] io_uring: use state completion infra for poll reqs Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 12/16] io_uring: add IORING_SETUP_SINGLE_ISSUER Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 13/16] io_uring: pass hash table into poll_find Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 14/16] io_uring: introduce a struct for hash table Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 15/16] io_uring: propagate locking state to poll cancel Pavel Begunkov
2022-06-16 9:22 ` [PATCH for-next v3 16/16] io_uring: mutex locked poll hashing Pavel Begunkov
2022-06-17 15:35 ` Nathan Chancellor
2022-06-18 12:07 ` Pavel Begunkov
2022-06-16 13:18 ` [PATCH for-next v3 00/16] 5.20 cleanups and poll optimisations Jens Axboe
2022-06-16 15:58 ` Hao Xu
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 \
--in-reply-to=600ba20a9338b8a39b249b23d3d177803613dde4.1655371007.git.asml.silence@gmail.com \
[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