From: Nick Desaulniers <ndesaulniers@google.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Kees Cook <keescook@chromium.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Bill Wendling <morbo@google.com>, Bill Wendling <wcw@google.com>,
Catalin Marinas <catalin.marinas@arm.com>,
clang-built-linux <clang-built-linux@googlegroups.com>,
Fangrui Song <maskray@google.com>,
Heiko Carstens <hca@linux.ibm.com>,
Jarmo Tiitto <jarmo.tiitto@gmail.com>,
Lukas Bulwahn <lukas.bulwahn@gmail.com>,
Mark Rutland <mark.rutland@arm.com>,
Masahiro Yamada <masahiroy@kernel.org>,
Miguel Ojeda <ojeda@kernel.org>,
Nathan Chancellor <nathan@kernel.org>,
Peter Oberparleiter <oberpar@linux.ibm.com>,
Peter Zijlstra <peterz@infradead.org>,
Sami Tolvanen <samitolvanen@google.com>,
Will Deacon <will@kernel.org>
Subject: Re: [GIT PULL] Clang feature updates for v5.14-rc1
Date: Tue, 29 Jun 2021 14:27:34 -0700 [thread overview]
Message-ID: <CAKwvOdkcKU4K9LWTymmzi_c0wKPTQjWEbNu04WOd6D-EcnWDSg@mail.gmail.com> (raw)
In-Reply-To: <CAHk-=wg8M2DyA=bWtnGsAOVqYU-AusxYcpXubUO2htb6qPU9dg@mail.gmail.com>
On Tue, Jun 29, 2021 at 2:04 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> On Tue, Jun 29, 2021 at 1:44 PM Kees Cook <keescook@chromium.org> wrote:
> > >
> > > And it causes the kernel to be bigger and run slower.
> >
> > Right -- that's expected. It's not designed to be the final kernel
> > someone uses. :)
>
> Well, from what I've seen, you actually want to run real loads in
> production environments for PGO to actually be anything but a bogus
> "performance benchmarks only" kind of thing.
>
> Of course, "performance benchmarks only" is very traditional, and
> we've seen that used over and over in the past in this industry. That
> doesn't make it _right_, though.
The current major use case is ensuring that production kernels have
been "trained" with specific workloads in mind.
> And if you actually want to have it usable in production environments,
> you really should strive to run code as closely as possible to a
> production kernel too.
You could do both. There is a line of research internally using
multiple training rounds ("CSPGO").
> You'd want to run something that you can sample over time, and in
> production, not something that you have to build a special kernels for
> that then gets used for a benchmark run, but can't be kept in
> production because it performs so much worse.
>
> Real proper profiles will tell you what *really* matters - and if you
> don't have enough samples to give you good information, then that
> particular code clearly is not important enough to waste PGO on.
>
> This is not all that dissimilar to using gprof information for
> traditional - manual - optimizations.
>
> Sure, instrumented gprof output is better than nothing, but it is
> *hugely* worse than actual proper sampled profiles that actually show
> what matters for performance (as opposed to what runs a lot - the two
> are not necessarily all that closely correlated, with cache misses
> being a thing).
>
> And I really hate how pretty much all of the PGO support seems to be
> just about this inferior method of getting the data.
Right now we're having trouble with hardware performance counters on
non-intel chips; I don't think we have working LBR equivalents on AMD
until zen3, and our ETM based samples on ARM are hung up on a few last
minute issues requiring new hardware (from multiple different chipset
vendors).
It would be good to have some form profile based optimizations that
aren't architecture or microarchitecture dependent.
--
Thanks,
~Nick Desaulniers
next prev parent reply other threads:[~2021-06-29 21:27 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-28 19:32 [GIT PULL] Clang feature updates for v5.14-rc1 Kees Cook
2021-06-29 2:49 ` Linus Torvalds
2021-06-29 20:44 ` Kees Cook
2021-06-29 21:03 ` Linus Torvalds
2021-06-29 21:27 ` Nick Desaulniers [this message]
2021-06-29 21:57 ` Linus Torvalds
2021-07-07 8:10 ` Ingo Molnar
2021-07-02 12:46 ` Bill Wendling
2021-07-02 12:56 ` Peter Zijlstra
2021-07-02 17:26 ` Nick Desaulniers
2021-07-02 18:57 ` Peter Zijlstra
2021-07-02 19:53 ` Nick Desaulniers
2022-10-31 23:55 ` Bill Wendling
2021-06-29 13:14 ` Mark Rutland
2021-06-29 20:11 ` Kees Cook
2021-06-29 21:05 ` Nick Desaulniers
2021-10-05 13:10 ` ARCH_WANTS_NO_INSTR (Re: [GIT PULL] Clang feature updates for v5.14-rc1) Daniel Axtens
2021-10-05 13:45 ` Daniel Axtens
2021-10-05 14:30 ` Mark Rutland
2021-10-07 6:19 ` Jarmo Tiitto
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=CAKwvOdkcKU4K9LWTymmzi_c0wKPTQjWEbNu04WOd6D-EcnWDSg@mail.gmail.com \
--to=ndesaulniers@google.com \
--cc=catalin.marinas@arm.com \
--cc=clang-built-linux@googlegroups.com \
--cc=hca@linux.ibm.com \
--cc=jarmo.tiitto@gmail.com \
--cc=keescook@chromium.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lukas.bulwahn@gmail.com \
--cc=mark.rutland@arm.com \
--cc=masahiroy@kernel.org \
--cc=maskray@google.com \
--cc=morbo@google.com \
--cc=nathan@kernel.org \
--cc=oberpar@linux.ibm.com \
--cc=ojeda@kernel.org \
--cc=peterz@infradead.org \
--cc=samitolvanen@google.com \
--cc=torvalds@linux-foundation.org \
--cc=wcw@google.com \
--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).