From: Nick Desaulniers <ndesaulniers@google.com>
To: Segher Boessenkool <segher@kernel.crashing.org>
Cc: Nathan Chancellor <natechancellor@gmail.com>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Paul Mackerras <paulus@samba.org>,
Michael Ellerman <mpe@ellerman.id.au>,
clang-built-linux <clang-built-linux@googlegroups.com>,
linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v4 3/3] powerpc/prom_init: Use -ffreestanding to avoid a reference to bcmp
Date: Mon, 14 Oct 2019 08:56:12 -0700 [thread overview]
Message-ID: <CAKwvOdmcUT2A9FG0JD9jd0s=gAavRc_h+RLG6O3mBz4P1FfF8w@mail.gmail.com> (raw)
In-Reply-To: <20191014093501.GE28442@gate.crashing.org>
On Mon, Oct 14, 2019 at 2:35 AM Segher Boessenkool
<segher@kernel.crashing.org> wrote:
>
> On Sun, Oct 13, 2019 at 07:51:01PM -0700, Nathan Chancellor wrote:
> > r374662 gives LLVM the ability to convert certain loops into a reference
> > to bcmp as an optimization; this breaks prom_init_check.sh:
>
> When/why does LLVM think this is okay? This function has been removed
> from POSIX over a decade ago (and before that it always was marked as
> legacy).
Segher, do you have links for any of the above? If so, that would be
helpful to me. I'm arguing against certain transforms that assume that
one library function is faster than another, when such claims are
based on measurements from one stdlib implementation. (There's others
in the pipeline I'm not too thrilled about, too).
The rationale for why it was added was that memcmp takes a measurable
amount of time in Google's fleet, and most calls to memcmp don't care
about the position of the mismatch; bcmp is lower overhead (or at
least for our libc implementation, not sure about others).
--
Thanks,
~Nick Desaulniers
WARNING: multiple messages have this Message-ID (diff)
From: Nick Desaulniers <ndesaulniers@google.com>
To: Segher Boessenkool <segher@kernel.crashing.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
clang-built-linux <clang-built-linux@googlegroups.com>,
Paul Mackerras <paulus@samba.org>,
Nathan Chancellor <natechancellor@gmail.com>,
linuxppc-dev <linuxppc-dev@lists.ozlabs.org>
Subject: Re: [PATCH v4 3/3] powerpc/prom_init: Use -ffreestanding to avoid a reference to bcmp
Date: Mon, 14 Oct 2019 08:56:12 -0700 [thread overview]
Message-ID: <CAKwvOdmcUT2A9FG0JD9jd0s=gAavRc_h+RLG6O3mBz4P1FfF8w@mail.gmail.com> (raw)
In-Reply-To: <20191014093501.GE28442@gate.crashing.org>
On Mon, Oct 14, 2019 at 2:35 AM Segher Boessenkool
<segher@kernel.crashing.org> wrote:
>
> On Sun, Oct 13, 2019 at 07:51:01PM -0700, Nathan Chancellor wrote:
> > r374662 gives LLVM the ability to convert certain loops into a reference
> > to bcmp as an optimization; this breaks prom_init_check.sh:
>
> When/why does LLVM think this is okay? This function has been removed
> from POSIX over a decade ago (and before that it always was marked as
> legacy).
Segher, do you have links for any of the above? If so, that would be
helpful to me. I'm arguing against certain transforms that assume that
one library function is faster than another, when such claims are
based on measurements from one stdlib implementation. (There's others
in the pipeline I'm not too thrilled about, too).
The rationale for why it was added was that memcmp takes a measurable
amount of time in Google's fleet, and most calls to memcmp don't care
about the position of the mismatch; bcmp is lower overhead (or at
least for our libc implementation, not sure about others).
--
Thanks,
~Nick Desaulniers
next prev parent reply other threads:[~2019-10-14 15:56 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-11 18:20 [PATCH v3 0/3] LLVM/Clang fixes for pseries_defconfig Nathan Chancellor
2019-09-11 18:20 ` Nathan Chancellor
2019-09-11 18:20 ` [PATCH v3 1/3] powerpc: Don't add -mabi= flags when building with Clang Nathan Chancellor
2019-09-11 18:20 ` Nathan Chancellor
2019-09-11 18:20 ` [PATCH v3 2/3] powerpc: Avoid clang warnings around setjmp and longjmp Nathan Chancellor
2019-09-11 18:20 ` Nathan Chancellor
2019-09-11 20:55 ` Nick Desaulniers
2019-09-11 20:55 ` Nick Desaulniers
2019-09-11 18:20 ` [PATCH v3 3/3] powerpc/prom_init: Use -ffreestanding to avoid a reference to bcmp Nathan Chancellor
2019-09-11 18:20 ` Nathan Chancellor
2019-09-11 21:01 ` Nick Desaulniers
2019-09-11 21:01 ` Nick Desaulniers
2019-09-12 5:43 ` Nathan Chancellor
2019-09-12 5:43 ` Nathan Chancellor
2019-09-12 17:30 ` Nick Desaulniers
2019-09-12 17:30 ` Nick Desaulniers
2019-10-14 2:50 ` [PATCH v4 0/3] LLVM/Clang fixes for pseries_defconfig Nathan Chancellor
2019-10-14 2:50 ` Nathan Chancellor
2019-10-14 2:50 ` [PATCH v4 1/3] powerpc: Don't add -mabi= flags when building with Clang Nathan Chancellor
2019-10-14 2:50 ` Nathan Chancellor
2019-10-14 2:51 ` [PATCH v4 2/3] powerpc: Avoid clang warnings around setjmp and longjmp Nathan Chancellor
2019-10-14 2:51 ` Nathan Chancellor
2019-10-14 2:51 ` [PATCH v4 3/3] powerpc/prom_init: Use -ffreestanding to avoid a reference to bcmp Nathan Chancellor
2019-10-14 2:51 ` Nathan Chancellor
2019-10-14 9:35 ` Segher Boessenkool
2019-10-14 9:35 ` Segher Boessenkool
2019-10-14 15:56 ` Nick Desaulniers [this message]
2019-10-14 15:56 ` Nick Desaulniers
2019-10-14 19:11 ` Segher Boessenkool
2019-10-14 19:11 ` Segher Boessenkool
2019-10-18 19:00 ` Nathan Chancellor
2019-10-18 19:00 ` Nathan Chancellor
2019-10-18 20:02 ` Segher Boessenkool
2019-10-18 20:02 ` Segher Boessenkool
2019-10-22 5:15 ` Nathan Chancellor
2019-10-22 5:15 ` Nathan Chancellor
2019-10-22 8:57 ` Segher Boessenkool
2019-10-22 8:57 ` Segher Boessenkool
2019-10-30 4:12 ` Nathan Chancellor
2019-10-30 4:12 ` Nathan Chancellor
2019-10-14 16:03 ` [PATCH v4 0/3] LLVM/Clang fixes for pseries_defconfig Nick Desaulniers
2019-10-14 16:03 ` Nick Desaulniers
2019-11-19 4:57 ` [PATCH v5 0/3] LLVM/Clang fixes for a few defconfigs Nathan Chancellor
2019-11-19 4:57 ` Nathan Chancellor
2019-11-19 4:57 ` [PATCH v5 1/3] powerpc: Don't add -mabi= flags when building with Clang Nathan Chancellor
2019-11-19 4:57 ` Nathan Chancellor
2019-11-26 1:13 ` Michael Ellerman
2019-11-19 4:57 ` [PATCH v5 2/3] powerpc: Avoid clang warnings around setjmp and longjmp Nathan Chancellor
2019-11-19 4:57 ` Nathan Chancellor
2019-11-19 4:57 ` [PATCH v5 3/3] powerpc/prom_init: Use -ffreestanding to avoid a reference to bcmp Nathan Chancellor
2019-11-19 4:57 ` Nathan Chancellor
2019-11-25 20:15 ` [PATCH v5 0/3] LLVM/Clang fixes for a few defconfigs Nick Desaulniers
2019-11-25 20:15 ` Nick Desaulniers
2019-11-28 4:59 ` Michael Ellerman
2019-11-28 4:59 ` Michael Ellerman
2019-11-28 7:45 ` Nathan Chancellor
2019-11-28 7:45 ` Nathan Chancellor
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='CAKwvOdmcUT2A9FG0JD9jd0s=gAavRc_h+RLG6O3mBz4P1FfF8w@mail.gmail.com' \
--to=ndesaulniers@google.com \
--cc=benh@kernel.crashing.org \
--cc=clang-built-linux@googlegroups.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mpe@ellerman.id.au \
--cc=natechancellor@gmail.com \
--cc=paulus@samba.org \
--cc=segher@kernel.crashing.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.