All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pavel Begunkov <asml.silence@gmail.com>
To: io-uring@vger.kernel.org
Cc: Jens Axboe <axboe@kernel.dk>, asml.silence@gmail.com
Subject: [PATCH 5/6] io_uring: conditional ->async_data allocation
Date: Wed, 24 Aug 2022 13:07:42 +0100	[thread overview]
Message-ID: <9dc62be9e88dd0ed63c48365340e8922d2498293.1661342812.git.asml.silence@gmail.com> (raw)
In-Reply-To: <cover.1661342812.git.asml.silence@gmail.com>

There are opcodes that need ->async_data only in some cases and
allocation it unconditionally may hurt performance. Add an option to
opdef to make move the allocation part from the core io_uring to opcode
specific code.
Note, we can't just set opdef->async_size to zero because there are
other helpers that rely on it, e.g. io_alloc_async_data().

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 io_uring/io_uring.c | 7 ++++---
 io_uring/opdef.h    | 2 ++
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index ebfdb2212ec2..77616279000b 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -1450,9 +1450,10 @@ int io_req_prep_async(struct io_kiocb *req)
 		return 0;
 	if (WARN_ON_ONCE(req_has_async_data(req)))
 		return -EFAULT;
-	if (io_alloc_async_data(req))
-		return -EAGAIN;
-
+	if (!io_op_defs[req->opcode].manual_alloc) {
+		if (io_alloc_async_data(req))
+			return -EAGAIN;
+	}
 	return def->prep_async(req);
 }
 
diff --git a/io_uring/opdef.h b/io_uring/opdef.h
index ece8ed4f96c4..763c6e54e2ee 100644
--- a/io_uring/opdef.h
+++ b/io_uring/opdef.h
@@ -25,6 +25,8 @@ struct io_op_def {
 	unsigned		ioprio : 1;
 	/* supports iopoll */
 	unsigned		iopoll : 1;
+	/* opcode specific path will handle ->async_data allocation if needed */
+	unsigned		manual_alloc : 1;
 	/* size of async data needed, if any */
 	unsigned short		async_size;
 
-- 
2.37.2


  parent reply	other threads:[~2022-08-24 12:10 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-24 12:07 [PATCH 0/6] bunch of zerocopy send changes Pavel Begunkov
2022-08-24 12:07 ` [PATCH 1/6] io_uring/net: fix must_hold annotation Pavel Begunkov
2022-08-24 12:07 ` [PATCH 2/6] io_uring/net: fix zc send link failing Pavel Begunkov
2022-08-24 12:07 ` [PATCH 3/6] io_uring/net: fix indention Pavel Begunkov
2022-08-24 12:07 ` [PATCH 4/6] io_uring/notif: order notif vs send CQEs Pavel Begunkov
2022-08-24 12:07 ` Pavel Begunkov [this message]
2022-08-24 12:07 ` [PATCH 6/6] io_uring/net: save address for sendzc async execution Pavel Begunkov
2022-08-24 14:57 ` [PATCH 0/6] bunch of zerocopy send changes 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=9dc62be9e88dd0ed63c48365340e8922d2498293.1661342812.git.asml.silence@gmail.com \
    --to=asml.silence@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=io-uring@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.