io-uring.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Pavel Begunkov <asml.silence@gmail.com>
To: Jens Axboe <axboe@kernel.dk>,
	io-uring@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v3 6/6] io_uring: optimise req bulk allocation cache
Date: Sat,  1 Feb 2020 01:15:55 +0300	[thread overview]
Message-ID: <3811fb8e56319d8f1b7952bcaf2c73abd2acd770.1580508735.git.asml.silence@gmail.com> (raw)
In-Reply-To: <cover.1580508735.git.asml.silence@gmail.com>

Traverse backward through @reqs in struct io_submit_state, so it's
possible to remove cur_req from it and easier to handle in general.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 fs/io_uring.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 66742d5772fa..799e80e85027 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -206,7 +206,6 @@ struct io_submit_state {
 	 */
 	void			*reqs[IO_IOPOLL_BATCH];
 	unsigned int		free_reqs;
-	unsigned int		cur_req;
 
 	/*
 	 * File reference cache
@@ -850,7 +849,6 @@ static void io_init_submit_state(struct io_ring_ctx *ctx)
 	struct io_submit_state *state = &ctx->submit_state;
 
 	state->free_reqs = 0;
-	state->cur_req = 0;
 }
 
 static void io_clear_submit_state(struct io_ring_ctx *ctx)
@@ -858,8 +856,7 @@ static void io_clear_submit_state(struct io_ring_ctx *ctx)
 	struct io_submit_state *state = &ctx->submit_state;
 
 	if (state->free_reqs)
-		kmem_cache_free_bulk(req_cachep, state->free_reqs,
-					&state->reqs[state->cur_req]);
+		kmem_cache_free_bulk(req_cachep, state->free_reqs, state->reqs);
 }
 
 static inline bool __req_need_defer(struct io_kiocb *req)
@@ -1204,12 +1201,10 @@ static struct io_kiocb *io_get_req(struct io_ring_ctx *ctx)
 			ret = 1;
 		}
 		state->free_reqs = ret - 1;
-		state->cur_req = 1;
-		req = state->reqs[0];
+		req = state->reqs[ret - 1];
 	} else {
-		req = state->reqs[state->cur_req];
 		state->free_reqs--;
-		state->cur_req++;
+		req = state->reqs[state->free_reqs];
 	}
 
 got_it:
-- 
2.24.0


  parent reply	other threads:[~2020-01-31 22:17 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-31 22:15 [PATCH v3 0/6] add persistent submission state Pavel Begunkov
2020-01-31 22:15 ` [PATCH v3 1/6] io_uring: always pass non-null io_submit_state Pavel Begunkov
2020-01-31 22:15 ` [PATCH v3 2/6] io_uring: place io_submit_state into ctx Pavel Begunkov
2020-01-31 22:15 ` [PATCH v3 3/6] io_uring: move ring_fd into io_submit_state Pavel Begunkov
2020-01-31 22:15 ` [PATCH v3 4/6] io_uring: move *link " Pavel Begunkov
2020-01-31 22:15 ` [PATCH v3 5/6] io_uring: persistent req bulk allocation cache Pavel Begunkov
2020-01-31 22:15 ` Pavel Begunkov [this message]
2020-01-31 22:22 ` [PATCH v3 0/6] add persistent submission state Jens Axboe
2020-01-31 22:32   ` Pavel Begunkov
2020-02-01  0:31     ` Pavel Begunkov
2020-02-01  2:10       ` Jens Axboe
2020-02-01 11:42         ` Pavel Begunkov

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=3811fb8e56319d8f1b7952bcaf2c73abd2acd770.1580508735.git.asml.silence@gmail.com \
    --to=asml.silence@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=io-uring@vger.kernel.org \
    --cc=linux-kernel@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).