public inbox for [email protected]
 help / color / mirror / Atom feed
From: Pavel Begunkov <[email protected]>
To: Jens Axboe <[email protected]>, [email protected]
Subject: [PATCH 6/6] io_uring: pre-initialise some of req fields
Date: Sat, 26 Jun 2021 21:40:49 +0100	[thread overview]
Message-ID: <892ba0e71309bba9fe9e0142472330bbf9d8f05d.1624739600.git.asml.silence@gmail.com> (raw)
In-Reply-To: <[email protected]>

Most of requests are allocated from an internal cache, so it's waste of
time fully initialising them every time. Instead, let's pre-init some of
the fields we can during initial allocation (e.g. kmalloc(), see
io_alloc_req()) and keep them valid on request recycling. There are four
of them in this patch:

->ctx is always stays the same
->link is NULL on free, it's an invariant
->result is not even needed to init, just a precaution
->async_data we now clean in io_dismantle_req() as it's likely to
   never be allocated.

Signed-off-by: Pavel Begunkov <[email protected]>
---
 fs/io_uring.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 873cfd4a8761..6cfbf72340ab 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -1740,7 +1740,7 @@ static struct io_kiocb *io_alloc_req(struct io_ring_ctx *ctx)
 
 	if (!state->free_reqs) {
 		gfp_t gfp = GFP_KERNEL | __GFP_NOWARN;
-		int ret;
+		int ret, i;
 
 		if (io_flush_cached_reqs(ctx))
 			goto got_req;
@@ -1758,6 +1758,20 @@ static struct io_kiocb *io_alloc_req(struct io_ring_ctx *ctx)
 				return NULL;
 			ret = 1;
 		}
+
+		/*
+		 * Don't initialise the fields below on every allocation, but
+		 * do that in advance and keep valid on free.
+		 */
+		for (i = 0; i < ret; i++) {
+			struct io_kiocb *req = state->reqs[i];
+
+			req->ctx = ctx;
+			req->link = NULL;
+			req->async_data = NULL;
+			/* not necessary, but safer to zero */
+			req->result = 0;
+		}
 		state->free_reqs = ret;
 	}
 got_req:
@@ -1781,8 +1795,10 @@ static void io_dismantle_req(struct io_kiocb *req)
 		io_put_file(req->file);
 	if (req->fixed_rsrc_refs)
 		percpu_ref_put(req->fixed_rsrc_refs);
-	if (req->async_data)
+	if (req->async_data) {
 		kfree(req->async_data);
+		req->async_data = NULL;
+	}
 }
 
 /* must to be called somewhat shortly after putting a request */
@@ -6730,15 +6746,11 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	/* same numerical values with corresponding REQ_F_*, safe to copy */
 	req->flags = sqe_flags = READ_ONCE(sqe->flags);
 	req->user_data = READ_ONCE(sqe->user_data);
-	req->async_data = NULL;
 	req->file = NULL;
-	req->ctx = ctx;
-	req->link = NULL;
 	req->fixed_rsrc_refs = NULL;
 	/* one is dropped after submission, the other at completion */
 	atomic_set(&req->refs, 2);
 	req->task = current;
-	req->result = 0;
 
 	/* enforce forwards compatibility on users */
 	if (unlikely(sqe_flags & ~SQE_VALID_FLAGS))
-- 
2.32.0


  parent reply	other threads:[~2021-06-26 20:41 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-26 20:40 [PATCH for-next 0/6] small for-next optimisations Pavel Begunkov
2021-06-26 20:40 ` [PATCH 1/6] io_uring: refactor io_arm_poll_handler() Pavel Begunkov
2021-06-26 20:40 ` [PATCH 2/6] io_uring: mainstream sqpoll task_work running Pavel Begunkov
2021-06-26 20:40 ` [PATCH 3/6] io_uring: remove not needed PF_EXITING check Pavel Begunkov
2021-06-26 20:40 ` [PATCH 4/6] io_uring: optimise hot path restricted checks Pavel Begunkov
2021-06-26 20:40 ` [PATCH 5/6] io_uring: refactor io_submit_flush_completions Pavel Begunkov
2021-06-26 20:40 ` Pavel Begunkov [this message]
2021-06-27 22:23 ` [PATCH for-next 0/6] small for-next optimisations 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=892ba0e71309bba9fe9e0142472330bbf9d8f05d.1624739600.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