From: Pavel Begunkov <[email protected]>
To: [email protected]
Cc: [email protected]
Subject: [PATCH v2 2/4] io_uring/rw: don't directly use ki_complete
Date: Wed, 19 Feb 2025 01:33:38 +0000 [thread overview]
Message-ID: <4eb4bdab8cbcf5bc87083f7047edc81e920ab83c.1739919038.git.asml.silence@gmail.com> (raw)
In-Reply-To: <[email protected]>
We want to avoid checking ->ki_complete directly in the io_uring
completion path. Fortunately we have only two callback the selection
of which depend on the ring constant flags, i.e. IOPOLL, so use that
to infer the function.
Signed-off-by: Pavel Begunkov <[email protected]>
---
io_uring/rw.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/io_uring/rw.c b/io_uring/rw.c
index e8efd97fdee5..27ccc82d7843 100644
--- a/io_uring/rw.c
+++ b/io_uring/rw.c
@@ -563,8 +563,10 @@ static void io_complete_rw_iopoll(struct kiocb *kiocb, long res)
smp_store_release(&req->iopoll_completed, 1);
}
-static inline void io_rw_done(struct kiocb *kiocb, ssize_t ret)
+static inline void io_rw_done(struct io_kiocb *req, ssize_t ret)
{
+ struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
+
/* IO was queued async, completion will happen later */
if (ret == -EIOCBQUEUED)
return;
@@ -586,8 +588,10 @@ static inline void io_rw_done(struct kiocb *kiocb, ssize_t ret)
}
}
- INDIRECT_CALL_2(kiocb->ki_complete, io_complete_rw_iopoll,
- io_complete_rw, kiocb, ret);
+ if (req->ctx->flags & IORING_SETUP_IOPOLL)
+ io_complete_rw_iopoll(&rw->kiocb, ret);
+ else
+ io_complete_rw(&rw->kiocb, ret);
}
static int kiocb_done(struct io_kiocb *req, ssize_t ret,
@@ -598,7 +602,7 @@ static int kiocb_done(struct io_kiocb *req, ssize_t ret,
if (ret >= 0 && req->flags & REQ_F_CUR_POS)
req->file->f_pos = rw->kiocb.ki_pos;
- if (ret >= 0 && (rw->kiocb.ki_complete == io_complete_rw)) {
+ if (ret >= 0 && !(req->ctx->flags & IORING_SETUP_IOPOLL)) {
__io_complete_rw_common(req, ret);
/*
* Safe to call io_end from here as we're inline
@@ -609,7 +613,7 @@ static int kiocb_done(struct io_kiocb *req, ssize_t ret,
io_req_rw_cleanup(req, issue_flags);
return IOU_OK;
} else {
- io_rw_done(&rw->kiocb, ret);
+ io_rw_done(req, ret);
}
return IOU_ISSUE_SKIP_COMPLETE;
--
2.48.1
next prev parent reply other threads:[~2025-02-19 1:32 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-19 1:33 [PATCH v2 0/4] forced sync mshot read + cleanups Pavel Begunkov
2025-02-19 1:33 ` [PATCH v2 1/4] io_uring/rw: forbid multishot async reads Pavel Begunkov
2025-02-19 1:33 ` Pavel Begunkov [this message]
2025-02-19 1:33 ` [PATCH v2 3/4] io_uring/rw: move ki_complete init into prep Pavel Begunkov
2025-02-19 1:33 ` [PATCH v2 4/4] io_uring/rw: clean up mshot forced sync mode Pavel Begunkov
2025-02-19 20:56 ` [PATCH v2 0/4] forced sync mshot read + cleanups Jens Axboe
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=4eb4bdab8cbcf5bc87083f7047edc81e920ab83c.1739919038.git.asml.silence@gmail.com \
[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