From: Peter Zijlstra <peterz@infradead.org>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Nathan Chancellor <natechancellor@gmail.com>,
Kees Cook <keescook@chromium.org>,
Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>,
"Gustavo A. R. Silva" <gustavo@embeddedor.com>,
Joe Perches <joe@perches.com>, Ingo Molnar <mingo@redhat.com>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Jiri Olsa <jolsa@redhat.com>, Namhyung Kim <namhyung@kernel.org>,
Borislav Petkov <bp@alien8.de>, "H. Peter Anvin" <hpa@zytor.com>,
"maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)"
<x86@kernel.org>, Kan Liang <kan.liang@linux.intel.com>,
linux-kernel <linux-kernel@vger.kernel.org>,
Shawn Landden <shawn@git.icu>,
clang-built-linux@googlegroups.com,
Josh Poimboeuf <jpoimboe@redhat.com>
Subject: Re: [PATCH] perf/x86/intel: Mark expected switch fall-throughs
Date: Wed, 26 Jun 2019 12:43:21 +0200 [thread overview]
Message-ID: <20190626104321.GC3463@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20190626092432.GJ3419@hirez.programming.kicks-ass.net>
On Wed, Jun 26, 2019 at 11:24:32AM +0200, Peter Zijlstra wrote:
> That is, would something like this:
>
> diff --git a/arch/x86/include/asm/jump_label.h b/arch/x86/include/asm/jump_label.h
> index 06c3cc22a058..1761b1e76ddc 100644
> --- a/arch/x86/include/asm/jump_label.h
> +++ b/arch/x86/include/asm/jump_label.h
> @@ -32,7 +32,7 @@ static __always_inline bool arch_static_branch(struct static_key *key, bool bran
> : : "i" (key), "i" (branch) : : l_yes);
>
> return false;
> -l_yes:
> +l_yes: __attribute__((cold));
> return true;
> }
>
> @@ -49,7 +49,7 @@ static __always_inline bool arch_static_branch_jump(struct static_key *key, bool
> : : "i" (key), "i" (branch) : : l_yes);
>
> return false;
> -l_yes:
> +l_yes: __attribute__((hot));
> return true;
> }
>
> Help LLVM?
No, that's broken. What we do is the likely() and unlikely() hints in
static_branch_{,un}likely():
#define static_branch_likely(x) \
({ \
bool branch; \
if (__builtin_types_compatible_p(typeof(*x), struct static_key_true)) \
branch = !arch_static_branch(&(x)->key, true); \
else if (__builtin_types_compatible_p(typeof(*x), struct static_key_false)) \
branch = !arch_static_branch_jump(&(x)->key, true); \
else \
branch = ____wrong_branch_error(); \
likely(branch); \
})
#define static_branch_unlikely(x) \
({ \
bool branch; \
if (__builtin_types_compatible_p(typeof(*x), struct static_key_true)) \
branch = arch_static_branch_jump(&(x)->key, false); \
else if (__builtin_types_compatible_p(typeof(*x), struct static_key_false)) \
branch = arch_static_branch(&(x)->key, false); \
else \
branch = ____wrong_branch_error(); \
unlikely(branch); \
})
That should convey out-of-line vs in-line 'hint'. And clearly that's
broken for LLVM.
next prev parent reply other threads:[~2019-06-26 10:43 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-24 16:19 [PATCH] perf/x86/intel: Mark expected switch fall-throughs Gustavo A. R. Silva
2019-06-24 19:31 ` Peter Zijlstra
2019-06-24 19:45 ` Joe Perches
2019-06-24 20:37 ` Peter Zijlstra
2019-06-24 20:53 ` Gustavo A. R. Silva
2019-06-24 20:57 ` Joe Perches
2019-06-25 7:20 ` Peter Zijlstra
2019-06-24 22:28 ` Miguel Ojeda
2019-06-25 7:18 ` Peter Zijlstra
2019-06-25 12:47 ` Miguel Ojeda
2019-06-25 18:15 ` Nick Desaulniers
2019-06-25 22:29 ` Joe Perches
2019-06-25 22:57 ` Nick Desaulniers
2019-06-25 23:25 ` Joe Perches
2019-06-26 8:49 ` Peter Zijlstra
2019-06-26 22:14 ` Nick Desaulniers
2019-06-27 7:12 ` Peter Zijlstra
2019-06-28 13:31 ` Peter Zijlstra
2019-06-28 18:44 ` Nick Desaulniers
2019-06-29 7:10 ` Peter Zijlstra
2019-06-25 17:12 ` Kees Cook
2019-06-25 18:05 ` Nathan Chancellor
2019-06-25 19:53 ` Thomas Gleixner
2019-06-25 20:27 ` Nathan Chancellor
2019-06-25 20:37 ` Nick Desaulniers
2019-06-25 21:47 ` Thomas Gleixner
2019-06-26 5:10 ` Nathan Chancellor
2019-06-26 15:18 ` Thomas Gleixner
2019-06-26 19:00 ` Nathan Chancellor
2019-06-26 19:46 ` Thomas Gleixner
2019-06-26 20:03 ` Nathan Chancellor
2019-06-26 9:24 ` Peter Zijlstra
2019-06-26 9:55 ` Peter Zijlstra
2019-06-26 22:23 ` Nick Desaulniers
2019-06-27 7:35 ` Peter Zijlstra
2019-06-26 10:43 ` Peter Zijlstra [this message]
2019-06-26 22:15 ` Nick Desaulniers
2019-06-27 7:16 ` Peter Zijlstra
2019-06-26 16:30 ` Peter Zijlstra
2019-06-26 22:33 ` Nick Desaulniers
2019-06-26 23:11 ` Thomas Gleixner
2019-06-27 7:11 ` Peter Zijlstra
2019-06-25 23:46 ` Arnaldo Carvalho de Melo
2019-06-26 5:14 ` Nathan Chancellor
2019-06-25 20:09 ` Kees Cook
2019-06-26 8:06 ` Peter Zijlstra
2019-06-25 7:15 ` Peter Zijlstra
2019-07-25 16:27 ` [tip:perf/urgent] " tip-bot for Gustavo A. R. Silva
2019-07-25 17:06 ` Borislav Petkov
2019-07-25 17:35 ` Peter Zijlstra
2019-07-25 23:18 ` Joe Perches
2019-07-25 23:28 ` Kees Cook
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=20190626104321.GC3463@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=acme@kernel.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=bp@alien8.de \
--cc=clang-built-linux@googlegroups.com \
--cc=gustavo@embeddedor.com \
--cc=hpa@zytor.com \
--cc=joe@perches.com \
--cc=jolsa@redhat.com \
--cc=jpoimboe@redhat.com \
--cc=kan.liang@linux.intel.com \
--cc=keescook@chromium.org \
--cc=linux-kernel@vger.kernel.org \
--cc=miguel.ojeda.sandonis@gmail.com \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=natechancellor@gmail.com \
--cc=shawn@git.icu \
--cc=tglx@linutronix.de \
--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