From: Arnd Bergmann <arnd@arndb.de>
To: Huacai Chen <chenhuacai@loongson.cn>
Cc: Arnd Bergmann <arnd@arndb.de>,
Huacai Chen <chenhuacai@kernel.org>,
loongarch@lists.linux.dev,
linux-arch <linux-arch@vger.kernel.org>,
Xuefeng Li <lixuefeng@loongson.cn>, Guo Ren <guoren@kernel.org>,
Xuerui Wang <kernel@xen0n.name>,
Jiaxun Yang <jiaxun.yang@flygoat.com>,
Peter Zijlstra <peterz@infradead.org>,
Will Deacon <will@kernel.org>, Ingo Molnar <mingo@redhat.com>
Subject: Re: [PATCH] LoongArch: Add qspinlock support
Date: Fri, 17 Jun 2022 18:10:51 +0200 [thread overview]
Message-ID: <CAK8P3a2nD_Zxv5X_LB7AbO=kxHQyk3vz09fQZ-TTX4PL0b3g1g@mail.gmail.com> (raw)
In-Reply-To: <20220617145705.581985-1-chenhuacai@loongson.cn>
On Fri, Jun 17, 2022 at 4:57 PM Huacai Chen <chenhuacai@loongson.cn> wrote:
>
> On NUMA system, the performance of qspinlock is better than generic
> spinlock. Below is the UnixBench test results on a 8 nodes (4 cores
> per node, 32 cores in total) machine.
>
The performance increase is nice, but this is only half the story we need here.
I think the more important bit is how you can guarantee that the xchg16()
implementation is correct and always allows forward progress.
>@@ -123,6 +123,10 @@ static inline unsigned long __percpu_xchg(void *ptr, unsigned long val,
> int size)
> {
> switch (size) {
>+ case 1:
>+ case 2:
>+ return __xchg_small((volatile void *)ptr, val, size);
>+
Do you actually need the size 1 as well?
Generally speaking, I would like to rework the xchg()/cmpxchg() logic
to only cover the 32-bit and word-sized (possibly 64-bit) case, while
having separate optional 8-bit and 16-bit functions. I had a patch for
this in the past, and can try to dig that out, this may be the time to
finally do that.
I see that the qspinlock() code actually calls a 'relaxed' version of xchg16(),
but you only implement the one with the full barrier. Is it possible to
directly provide a relaxed version that has something less than the
__WEAK_LLSC_MB?
Arnd
next prev parent reply other threads:[~2022-06-17 16:11 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-17 14:57 [PATCH] LoongArch: Add qspinlock support Huacai Chen
2022-06-17 16:10 ` Arnd Bergmann [this message]
2022-06-17 17:45 ` Guo Ren
2022-06-17 18:59 ` Arnd Bergmann
2022-06-17 23:19 ` Guo Ren
2022-06-18 5:40 ` Arnd Bergmann
2022-06-19 15:48 ` Guo Ren
2022-06-19 16:10 ` Arnd Bergmann
2022-06-20 9:49 ` Huacai Chen
2022-06-20 16:00 ` Guo Ren
2022-06-21 0:59 ` Huacai Chen
2022-06-21 2:11 ` Guo Ren
2022-06-18 12:50 ` WANG Xuerui
2022-06-19 4:28 ` hev
2022-06-19 15:06 ` Guo Ren
2022-06-19 15:38 ` hev
2022-06-19 15:23 ` Guo Ren
2022-06-17 16:35 ` Guo Ren
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='CAK8P3a2nD_Zxv5X_LB7AbO=kxHQyk3vz09fQZ-TTX4PL0b3g1g@mail.gmail.com' \
--to=arnd@arndb.de \
--cc=chenhuacai@kernel.org \
--cc=chenhuacai@loongson.cn \
--cc=guoren@kernel.org \
--cc=jiaxun.yang@flygoat.com \
--cc=kernel@xen0n.name \
--cc=linux-arch@vger.kernel.org \
--cc=lixuefeng@loongson.cn \
--cc=loongarch@lists.linux.dev \
--cc=mingo@redhat.com \
--cc=peterz@infradead.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;
as well as URLs for NNTP newsgroup(s).