From: Peter Zijlstra <peterz@infradead.org>
To: David Laight <david.laight.linux@gmail.com>
Cc: Waiman Long <longman@redhat.com>, Ingo Molnar <mingo@redhat.com>,
Will Deacon <will@kernel.org>, Boqun Feng <boqun@kernel.org>,
linux-kernel@vger.kernel.org,
Linus Torvalds <torvalds@linux-foundation.org>,
Yafang Shao <laoar.shao@gmail.com>,
Steven Rostedt <rostedt@goodmis.org>
Subject: Re: [PATCH v4 next 3/9] locking/osq_lock: Set prev_cpu=0 instead of locked=1
Date: Tue, 15 Sep 2026 16:01:52 +0200 [thread overview]
Message-ID: <20260915140152.GC4121339@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <20260915145513.111c4a23@pumpkin>
On Tue, Sep 15, 2026 at 02:55:13PM +0100, David Laight wrote:
> > That comment *should* obviously have been here. Also, I would *much*
> > rather have this store at the top, next to node->next = NULL; That is
> > much saner.
>
>
> Part of the original reason I started these patches was to remove
> the writes to the current cpu's 'node' in the fast path acquire.
> That completely saves reading the cache line.
> It also means the initial xchg need only have acquire semantics.
That makes no sense... you *have* to write node->prev_cpu, and since you
get to have the line dirty for that, a few more stores to the same line
are 'free', no?
Trying to make the code keep invariants like that also makes it
needlessly convoluted.
If you need a store -- and you do, then you might as well write the
whole thing.
And if you think it makes a difference, your changelogs completely lack
any factual data to that effect. That is, there are no numbers. And
without numbers, we're just not doing this.
next prev parent reply other threads:[~2026-09-15 14:01 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-07 8:41 [PATCH v4 next 0/9] locking/osq_lock: Optimisations to osq_lock code David Laight
2026-09-07 8:41 ` [PATCH v4 next 1/9] locking/osq_lock: Add some comments about how it works David Laight
2026-09-09 14:57 ` Waiman Long
2026-09-07 8:41 ` [PATCH v4 next 2/9] locking/osq_lock: Save the cpu number for 'prev' not the node address David Laight
2026-09-09 17:36 ` Waiman Long
2026-09-15 8:33 ` Peter Zijlstra
2026-09-15 10:26 ` Peter Zijlstra
2026-09-15 11:11 ` David Laight
2026-09-15 17:27 ` Waiman Long
2026-09-07 8:41 ` [PATCH v4 next 3/9] locking/osq_lock: Set prev_cpu=0 instead of locked=1 David Laight
2026-09-09 18:01 ` Waiman Long
2026-09-09 18:52 ` David Laight
2026-09-14 12:01 ` Peter Zijlstra
2026-09-14 13:10 ` David Laight
2026-09-14 12:03 ` Peter Zijlstra
2026-09-14 13:08 ` David Laight
2026-09-15 8:40 ` Peter Zijlstra
2026-09-15 10:14 ` David Laight
2026-09-15 8:50 ` Peter Zijlstra
2026-09-15 10:20 ` David Laight
2026-09-15 10:40 ` Peter Zijlstra
2026-09-15 13:13 ` Peter Zijlstra
2026-09-15 13:15 ` Peter Zijlstra
2026-09-15 13:55 ` David Laight
2026-09-15 14:01 ` Peter Zijlstra [this message]
2026-09-15 17:47 ` David Laight
2026-09-07 8:41 ` [PATCH v4 next 4/9] locking/osq_lock: Delete 'fast path' code from osq_unlock() David Laight
2026-09-15 14:15 ` Peter Zijlstra
2026-09-07 8:41 ` [PATCH v4 next 5/9] locking/osq_lock: Avoid writing to node->next in the osq_lock() fast path David Laight
2026-09-07 8:41 ` [PATCH v4 next 6/9] locking/osq: Use cpu number for 'next' pointer David Laight
2026-09-07 8:41 ` [PATCH v4 next 7/9] locking/osq: Use 'unsigned int' for next/prev/tail David Laight
2026-09-07 8:41 ` [PATCH v4 next 8/9] locking/osq: inline encode_cpu() and rename decode_cpu() David Laight
2026-09-07 8:41 ` [PATCH v4 next 9/9] locking/osq_lock: Swap next<->prev and tail<->head David Laight
2026-09-07 16:08 ` [PATCH v4 next 0/9] locking/osq_lock: Optimisations to osq_lock code Linus Torvalds
2026-09-07 17:27 ` David Laight
2026-09-09 14:15 ` Haakon Bugge
2026-09-09 19:09 ` David Laight
2026-09-09 20:14 ` Waiman Long
2026-09-09 20:33 ` Waiman Long
2026-09-10 9:45 ` Haakon Bugge
2026-09-10 11:00 ` David Laight
2026-09-10 11:31 ` Haakon Bugge
2026-09-10 12:05 ` David Laight
2026-09-10 15:30 ` Haakon Bugge
2026-09-10 16:22 ` Waiman Long
2026-09-11 17:15 ` David Laight
2026-09-14 11:46 ` Haakon Bugge
2026-09-14 13:24 ` David Laight
2026-09-11 10:08 ` David Laight
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=20260915140152.GC4121339@noisy.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=boqun@kernel.org \
--cc=david.laight.linux@gmail.com \
--cc=laoar.shao@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=longman@redhat.com \
--cc=mingo@redhat.com \
--cc=rostedt@goodmis.org \
--cc=torvalds@linux-foundation.org \
--cc=will@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
all inboxes | Powered by JetHome®