From: David Howells <dhowells@redhat.com>
To: torvalds@linux-foundation.org
Cc: dhowells@redhat.com, Rasmus Villemoes <linux@rasmusvillemoes.dk>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Peter Zijlstra <peterz@infradead.org>,
nicolas.dichtel@6wind.com, raven@themaw.net,
Christian Brauner <christian@brauner.io>,
dhowells@redhat.com, keyrings@vger.kernel.org,
linux-usb@vger.kernel.org, linux-block@vger.kernel.org,
linux-security-module@vger.kernel.org,
linux-fsdevel@vger.kernel.org, linux-api@vger.kernel.org,
linux-security-module@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: [RFC PATCH 11/11] pipe: Increase the writer-wakeup threshold to reduce context-switch count [ver #3]
Date: Fri, 01 Nov 2019 17:35:41 +0000 [thread overview]
Message-ID: <157262974145.13142.5910403713310791673.stgit@warthog.procyon.org.uk> (raw)
In-Reply-To: <157262963995.13142.5568934007158044624.stgit@warthog.procyon.org.uk>
Increase the threshold at which the reader sends a wake event to the
writers in the queue such that the queue must be half empty before the wake
is issued rather than the wake being issued when just a single slot
available.
This reduces the number of context switches in the tests significantly,
without altering the amount of work achieved. With my pipe-bench program,
there's a 20% reduction versus an unpatched kernel.
Suggested-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: David Howells <dhowells@redhat.com>
---
fs/pipe.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/fs/pipe.c b/fs/pipe.c
index aba2455caabe..9cd5cbef9552 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -324,16 +324,18 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to)
}
if (!buf->len) {
+ bool wake;
pipe_buf_release(pipe, buf);
spin_lock_irq(&pipe->wait.lock);
tail++;
pipe->tail = tail;
do_wakeup = 1;
- if (head - (tail - 1) == pipe->max_usage)
+ wake = head - (tail - 1) == pipe->max_usage / 2;
+ if (wake)
wake_up_interruptible_sync_poll_locked(
&pipe->wait, EPOLLOUT | EPOLLWRNORM);
spin_unlock_irq(&pipe->wait.lock);
- if (head - (tail - 1) == pipe->max_usage)
+ if (wake)
kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
}
total_len -= chars;
next prev parent reply other threads:[~2019-11-01 17:35 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-01 17:34 [RFC PATCH 00/11] pipe: Notification queue preparation [ver #3] David Howells
2019-11-01 17:34 ` [RFC PATCH 01/11] pipe: Reduce #inclusion of pipe_fs_i.h " David Howells
2019-11-01 17:34 ` [RFC PATCH 02/11] Remove the nr_exclusive argument from __wake_up_sync_key() " David Howells
2019-11-01 17:34 ` [RFC PATCH 03/11] Add wake_up_interruptible_sync_poll_locked() " David Howells
2019-11-01 17:34 ` [RFC PATCH 04/11] pipe: Use head and tail pointers for the ring, not cursor and length " David Howells
2019-11-07 9:03 ` [pipe] d60337eff1: phoronix-test-suite.noise-level.0.activity_level 144.0% improvement lkp report check
2019-11-07 16:51 ` Linus Torvalds
2019-11-07 17:42 ` David Howells
2019-11-10 3:13 ` [pipe] d60337eff1: BUG:kernel_NULL_pointer_dereference,address kernel test robot
2019-11-15 13:28 ` David Howells
2019-11-15 16:22 ` David Howells
2019-11-18 7:53 ` [LKP] " kernel test robot
2019-11-01 17:34 ` [RFC PATCH 05/11] pipe: Allow pipes to have kernel-reserved slots [ver #3] David Howells
2019-11-01 17:34 ` [RFC PATCH 06/11] pipe: Advance tail pointer inside of wait spinlock in pipe_read() " David Howells
2019-11-01 17:35 ` [RFC PATCH 07/11] pipe: Conditionalise wakeup " David Howells
2019-11-01 17:35 ` [RFC PATCH 08/11] pipe: Rearrange sequence in pipe_write() to preallocate slot " David Howells
2019-11-01 17:35 ` [RFC PATCH 09/11] pipe: Remove redundant wakeup from pipe_write() " David Howells
2019-11-01 17:35 ` [RFC PATCH 10/11] pipe: Check for ring full inside of the spinlock in " David Howells
2019-11-01 17:35 ` David Howells [this message]
2019-11-01 19:24 ` [RFC PATCH 00/11] pipe: Notification queue preparation " Linus Torvalds
2019-11-01 22:05 ` David Howells
2019-11-01 22:12 ` Linus Torvalds
2019-11-05 16:02 ` Details on the UAPI of implementing notifications on pipes David Howells
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=157262974145.13142.5910403713310791673.stgit@warthog.procyon.org.uk \
--to=dhowells@redhat.com \
--cc=christian@brauner.io \
--cc=gregkh@linuxfoundation.org \
--cc=keyrings@vger.kernel.org \
--cc=linux-api@vger.kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=linux@rasmusvillemoes.dk \
--cc=nicolas.dichtel@6wind.com \
--cc=peterz@infradead.org \
--cc=raven@themaw.net \
--cc=torvalds@linux-foundation.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).