[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20241025080200.GD14555@noisy.programming.kicks-ass.net>
Date: Fri, 25 Oct 2024 10:02:00 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Uros Bizjak <ubizjak@...il.com>
Cc: linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...nel.org>,
Will Deacon <will@...nel.org>, Waiman Long <longman@...hat.com>,
Boqun Feng <boqun.feng@...il.com>
Subject: Re: [PATCH] locking/osq_lock: Use atomic_try_cmpxchg_release() in
osq_unlock()
On Tue, Oct 01, 2024 at 01:45:57PM +0200, Uros Bizjak wrote:
> Replace this pattern in osq_unlock():
>
> atomic_cmpxchg(*ptr, old, new) == old
>
> ... with the simpler and faster:
>
> atomic_try_cmpxchg(*ptr, &old, new)
>
> The x86 CMPXCHG instruction returns success in the ZF flag,
> so this change saves a compare after the CMPXCHG. The code
> in the fast path of osq_unlock() improves from:
>
> 11b: 31 c9 xor %ecx,%ecx
> 11d: 8d 50 01 lea 0x1(%rax),%edx
> 120: 89 d0 mov %edx,%eax
> 122: f0 0f b1 0f lock cmpxchg %ecx,(%rdi)
> 126: 39 c2 cmp %eax,%edx
> 128: 75 05 jne 12f <...>
>
> to:
>
> 12b: 31 d2 xor %edx,%edx
> 12d: 83 c0 01 add $0x1,%eax
> 130: f0 0f b1 17 lock cmpxchg %edx,(%rdi)
> 134: 75 05 jne 13b <...>
>
> Signed-off-by: Uros Bizjak <ubizjak@...il.com>
Thanks!
Powered by blists - more mailing lists