From: Pavel Begunkov <[email protected]>
To: Jens Axboe <[email protected]>,
[email protected], [email protected]
Cc: Tejun Heo <[email protected]>, Dennis Zhou <[email protected]>,
Christoph Lameter <[email protected]>
Subject: Re: [PATCH RFC v2 3/3] io_uring: batch get(ctx->ref) across submits
Date: Sat, 21 Dec 2019 19:20:26 +0300 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <925d8fe5406779bbfa108caa3d1f9fd16e3434b5.1576944502.git.asml.silence@gmail.com>
[-- Attachment #1.1: Type: text/plain, Size: 4211 bytes --]
On 21/12/2019 19:15, Pavel Begunkov wrote:
> Double account ctx->refs keeping number of taken refs in ctx. As
> io_uring gets per-request ctx->refs during submission, while holding
> ctx->uring_lock, this allows in most of the time to bypass
> percpu_ref_get*() and its overhead.
Jens, could you please benchmark with this one? Especially for offloaded QD1
case. I haven't got any difference for nops test and don't have a decent SSD
at hands to test it myself. We could drop it, if there is no benefit.
This rewrites that @extra_refs from the second one, so I left it for now.
>
> Signed-off-by: Pavel Begunkov <[email protected]>
> ---
> fs/io_uring.c | 32 +++++++++++++++++++++++++-------
> 1 file changed, 25 insertions(+), 7 deletions(-)
>
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index 5392134f042f..eef09de94609 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -84,6 +84,9 @@
> #define IORING_MAX_ENTRIES 32768
> #define IORING_MAX_CQ_ENTRIES (2 * IORING_MAX_ENTRIES)
>
> +/* Not less than IORING_MAX_ENTRIES, so can grab once per submission loop */
> +#define IORING_REFS_THRESHOLD IORING_MAX_ENTRIES
> +
> /*
> * Shift of 9 is 512 entries, or exactly one page on 64-bit archs
> */
> @@ -197,6 +200,7 @@ struct fixed_file_data {
> struct io_ring_ctx {
> struct {
> struct percpu_ref refs;
> + unsigned long taken_refs; /* used under @uring_lock */
> } ____cacheline_aligned_in_smp;
>
> struct {
> @@ -690,6 +694,13 @@ static void io_ring_ctx_ref_free(struct percpu_ref *ref)
> complete(&ctx->completions[0]);
> }
>
> +static void io_free_taken_refs(struct io_ring_ctx *ctx)
> +{
> + if (ctx->taken_refs)
> + percpu_ref_put_many(&ctx->refs, ctx->taken_refs);
> + ctx->taken_refs = 0;
> +}
> +
> static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
> {
> struct io_ring_ctx *ctx;
> @@ -4388,7 +4399,6 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
> struct io_submit_state state, *statep = NULL;
> struct io_kiocb *link = NULL;
> int i, submitted = 0;
> - unsigned int extra_refs;
> bool mm_fault = false;
>
> /* if we have a backlog and couldn't flush it all, return BUSY */
> @@ -4398,9 +4408,15 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
> return -EBUSY;
> }
>
> - if (!percpu_ref_tryget_many(&ctx->refs, nr))
> - return -EAGAIN;
> - extra_refs = nr;
> + if (ctx->taken_refs < IORING_REFS_THRESHOLD) {
> + if (unlikely(percpu_ref_is_dying(&ctx->refs))) {
> + io_free_taken_refs(ctx);
> + return -ENXIO;
> + }
> + if (!percpu_ref_tryget_many(&ctx->refs, IORING_REFS_THRESHOLD))
> + return -EAGAIN;
> + ctx->taken_refs += IORING_REFS_THRESHOLD;
> + }
>
> if (nr > IO_PLUG_THRESHOLD) {
> io_submit_state_start(&state, nr);
> @@ -4417,8 +4433,9 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
> submitted = -EAGAIN;
> break;
> }
> - --extra_refs;
> if (!io_get_sqring(ctx, req, &sqe)) {
> + /* not submitted, but a ref is freed */
> + ctx->taken_refs--;
> __io_free_req(req);
> break;
> }
> @@ -4454,8 +4471,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
> io_queue_link_head(link);
> if (statep)
> io_submit_state_end(&state);
> - if (extra_refs)
> - percpu_ref_put_many(&ctx->refs, extra_refs);
> + ctx->taken_refs -= submitted;
>
> /* Commit SQ ring head once we've consumed and submitted all SQEs */
> io_commit_sqring(ctx);
> @@ -5731,6 +5747,7 @@ static int io_uring_fasync(int fd, struct file *file, int on)
> static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
> {
> mutex_lock(&ctx->uring_lock);
> + io_free_taken_refs(ctx);
> percpu_ref_kill(&ctx->refs);
> mutex_unlock(&ctx->uring_lock);
>
> @@ -6196,6 +6213,7 @@ static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
>
> if (opcode != IORING_UNREGISTER_FILES &&
> opcode != IORING_REGISTER_FILES_UPDATE) {
> + io_free_taken_refs(ctx);
> percpu_ref_kill(&ctx->refs);
>
> /*
>
--
Pavel Begunkov
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2019-12-21 16:21 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-17 22:28 [PATCH 0/2] optimise ctx's refs grabbing in io_uring Pavel Begunkov
2019-12-17 22:28 ` [PATCH 1/2] pcpu_ref: add percpu_ref_tryget_many() Pavel Begunkov
2019-12-17 23:42 ` Jens Axboe
2019-12-18 16:26 ` Tejun Heo
2019-12-18 17:49 ` Dennis Zhou
2019-12-21 15:36 ` Pavel Begunkov
2019-12-17 22:28 ` [PATCH 2/2] io_uring: batch getting pcpu references Pavel Begunkov
2019-12-17 23:21 ` Jens Axboe
2019-12-17 23:31 ` Jens Axboe
2019-12-18 9:25 ` Pavel Begunkov
2019-12-18 9:23 ` Pavel Begunkov
2019-12-18 0:02 ` Jens Axboe
2019-12-18 10:41 ` Pavel Begunkov
2019-12-21 16:15 ` [PATCH v2 0/3] optimise ctx's refs grabbing in io_uring Pavel Begunkov
2019-12-21 16:15 ` [PATCH v2 1/3] pcpu_ref: add percpu_ref_tryget_many() Pavel Begunkov
2019-12-21 16:15 ` [PATCH v2 2/3] io_uring: batch getting pcpu references Pavel Begunkov
2019-12-21 16:15 ` [PATCH RFC v2 3/3] io_uring: batch get(ctx->ref) across submits Pavel Begunkov
2019-12-21 16:20 ` Pavel Begunkov [this message]
2019-12-21 16:38 ` Jens Axboe
2019-12-21 16:48 ` Pavel Begunkov
2019-12-21 17:01 ` Jens Axboe
2019-12-21 17:26 ` Pavel Begunkov
2019-12-21 20:12 ` [PATCH v3 0/2] optimise ctx's refs grabbing in io_uring Pavel Begunkov
2019-12-21 20:12 ` [PATCH v3 1/2] pcpu_ref: add percpu_ref_tryget_many() Pavel Begunkov
2019-12-21 20:12 ` [PATCH v3 2/2] io_uring: batch getting pcpu references Pavel Begunkov
2019-12-21 21:56 ` Pavel Begunkov
2019-12-28 11:13 ` [PATCH v4 0/2] optimise ctx's refs grabbing in io_uring Pavel Begunkov
2019-12-28 11:13 ` [PATCH v4 1/2] pcpu_ref: add percpu_ref_tryget_many() Pavel Begunkov
2019-12-28 11:13 ` [PATCH v4 2/2] io_uring: batch getting pcpu references Pavel Begunkov
2019-12-28 11:15 ` Pavel Begunkov
2019-12-28 17:03 ` Jens Axboe
2019-12-28 18:37 ` Pavel Begunkov
2019-12-30 3:33 ` Brian Gianforcaro
2019-12-30 18:45 ` 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 \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[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