From: Uros Bizjak <ubizjak@gmail.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: x86@kernel.org, linux-kernel@vger.kernel.org,
Andy Lutomirski <luto@kernel.org>, Ingo Molnar <mingo@kernel.org>,
Nadav Amit <namit@vmware.com>, Brian Gerst <brgerst@gmail.com>,
Denys Vlasenko <dvlasenk@redhat.com>,
"H . Peter Anvin" <hpa@zytor.com>,
Peter Zijlstra <peterz@infradead.org>,
Thomas Gleixner <tglx@linutronix.de>,
Borislav Petkov <bp@alien8.de>,
Josh Poimboeuf <jpoimboe@redhat.com>
Subject: Re: [PATCH v2 4/4] x86/percpu: Use C for percpu read/write accessors
Date: Wed, 4 Oct 2023 22:07:54 +0200 [thread overview]
Message-ID: <CAFULd4YRHmQVnwaORm7=7kUs7DYG7SfwdTXAitDt=bxiMU5AoQ@mail.gmail.com> (raw)
In-Reply-To: <CAHk-=wjuRGzhuETLYDoi4hM6RAxHVL0ptuRb3TH-od+348Y8zA@mail.gmail.com>
On Wed, Oct 4, 2023 at 9:42 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> Unrelated reaction..
>
> On Wed, 4 Oct 2023 at 12:24, Uros Bizjak <ubizjak@gmail.com> wrote:
> >
> > the code improves from:
> >
> > 65 8b 05 00 00 00 00 mov %gs:0x0(%rip),%eax
> > a9 00 00 0f 00 test $0xf0000,%eax
> >
> > to:
> >
> > 65 f7 05 00 00 00 00 testl $0xf0000,%gs:0x0(%rip)
> > 00 00 0f 00
>
> Funky.
>
> Why does gcc generate that full-width load from memory, and not demote
> it to a byte test?
It does when LSB is accessed at the same address. For example:
int m;
_Bool foo (void) { return m & 0x0f; }
compiles to:
0: f6 05 00 00 00 00 0f testb $0xf,0x0(%rip) # 7 <foo+0x7>
>
> IOW, it should not be
>
> 65 f7 05 00 00 00 00 testl $0xf0000,%gs:0x0(%rip)
> 00 00 0f 00
>
> after optimizing it, it should be three bytes shorter at
>
> 65 f6 05 00 00 00 00 testb $0xf,%gs:0x0(%rip)
> 0f
>
> instead (this is "objdump", so it doesn't show that the relocation
> entry has changed by +2 to compensate).
>
> Now, doing the access narrowing is a bad idea for stores (because it
> can cause subsequent loads to have conflicts in the store buffer), but
> for loads it should always be a win to narrow the access.
>
> I wonder why gcc doesn't do it. This is not related to __seg_gs - I
> tried it with regular memory accesses too, and gcc kept those as
> 32-bit accesses too.
>
> And no, the assembler can't optimize that operation either, since I
> think changing the testl to a testb would change the 'P' bit in the
> resulting eflags, so this is a "the compiler could pick a better
> instruction choice" thing.
>
> I'm probably missing some reason why gcc wouldn't do this. But clang
> does seem to do this obvious optimization.
You get a store forwarding stall when you write a bigger operand to
memory and then read part of it, if the smaller part doesn't start at
the same
address.
Uros.
next prev parent reply other threads:[~2023-10-04 20:08 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-04 14:49 [PATCH 0/4] x86/percpu: Use segment qualifiers Uros Bizjak
2023-10-04 14:49 ` [PATCH 1/4] x86/percpu: Update arch/x86/include/asm/percpu.h to the current tip Uros Bizjak
2023-10-04 14:49 ` [PATCH 2/4] x86/percpu: Enable named address spaces with known compiler version Uros Bizjak
2023-10-05 7:20 ` [tip: x86/percpu] " tip-bot2 for Uros Bizjak
2023-10-04 14:49 ` [PATCH 3/4] x86/percpu: Use compiler segment prefix qualifier Uros Bizjak
2023-10-05 7:20 ` [tip: x86/percpu] " tip-bot2 for Nadav Amit
2023-10-04 14:49 ` [PATCH 4/4] x86/percpu: Use C for percpu read/write accessors Uros Bizjak
2023-10-04 16:37 ` Ingo Molnar
2023-10-04 16:40 ` Ingo Molnar
2023-10-04 19:23 ` [PATCH v2 " Uros Bizjak
2023-10-04 19:42 ` Linus Torvalds
2023-10-04 20:07 ` Uros Bizjak [this message]
2023-10-04 20:12 ` Linus Torvalds
2023-10-04 20:19 ` Linus Torvalds
2023-10-04 20:22 ` Uros Bizjak
2023-10-05 7:06 ` Ingo Molnar
2023-10-05 7:40 ` Uros Bizjak
2023-10-05 7:20 ` [tip: x86/percpu] " tip-bot2 for Uros Bizjak
2023-10-08 17:59 ` [PATCH 4/4] " Linus Torvalds
2023-10-08 19:17 ` Uros Bizjak
2023-10-08 20:13 ` Linus Torvalds
2023-10-08 20:48 ` Linus Torvalds
2023-10-08 21:41 ` Uros Bizjak
2023-10-09 11:41 ` Ingo Molnar
2023-10-09 11:51 ` Ingo Molnar
2023-10-09 12:00 ` Uros Bizjak
2023-10-09 12:20 ` Ingo Molnar
2023-10-09 12:21 ` Nadav Amit
2023-10-09 12:42 ` Uros Bizjak
2023-10-09 12:53 ` Nadav Amit
2023-10-09 12:27 ` Uros Bizjak
2023-10-09 14:35 ` Uros Bizjak
2024-04-10 11:11 ` Andrey Konovalov
2024-04-10 11:21 ` Uros Bizjak
2024-04-10 11:24 ` Andrey Konovalov
2023-10-09 11:42 ` Ingo Molnar
2023-10-10 6:37 ` Uros Bizjak
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='CAFULd4YRHmQVnwaORm7=7kUs7DYG7SfwdTXAitDt=bxiMU5AoQ@mail.gmail.com' \
--to=ubizjak@gmail.com \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=dvlasenk@redhat.com \
--cc=hpa@zytor.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=namit@vmware.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=x86@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).