Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Sat, 12 Jun 2021 22:31:38 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Fangrui Song <maskray@google.com>
Cc:     Bill Wendling <morbo@google.com>, Kees Cook <keescook@google.com>,
        Jonathan Corbet <corbet@lwn.net>,
        Masahiro Yamada <masahiroy@kernel.org>,
        Linux Doc Mailing List <linux-doc@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
        clang-built-linux <clang-built-linux@googlegroups.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Nathan Chancellor <natechancellor@gmail.com>,
        Nick Desaulniers <ndesaulniers@google.com>,
        Sami Tolvanen <samitolvanen@google.com>,
        "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@kernel.org>
Subject: Re: [PATCH v9] pgo: add clang's Profile Guided Optimization
 infrastructure
Message-ID: <20210612203138.GH68208@worktop.programming.kicks-ass.net>
References: <20210111081821.3041587-1-morbo@google.com>
 <20210407211704.367039-1-morbo@google.com>
 <YMTn9yjuemKFLbws@hirez.programming.kicks-ass.net>
 <CAGG=3QXjD1DQjACu=CQQSP=whue-14Pw8FcNcXrJZfLC_E+y9w@mail.gmail.com>
 <YMT5xZsZMX0PpDKQ@hirez.programming.kicks-ass.net>
 <20210612202015.s4743sr6d3lv3lgf@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210612202015.s4743sr6d3lv3lgf@google.com>
Precedence: bulk

On Sat, Jun 12, 2021 at 01:20:15PM -0700, Fangrui Song wrote:

> For applications, I think instrumentation based PGO can be 1%~4% faster
> than sample-based PGO (e.g. AutoFDO) on x86.

Why? What specifically is missed by sample-based? I thought that LBR
augmented samples were very useful for exactly this.

> Sample-based PGO has CPU requirement (e.g. Performance Monitoring Unit).
> (my gut feeling is that there may be larger gap between instrumentation
> based PGO and sample-based PGO for aarch64/ppc64, even though they can
> use sample-based PGO.)
> Instrumentation based PGO can be ported to more architectures.

Every architecture that cares about performance had better have a
hardware PMU. Both argh64 and ppc64 have one.

> In addition, having an infrastructure for instrumentation based PGO
> makes it easy to deploy newer techniques like context-sensitive PGO
> (just changed compile options; it doesn't need new source level
> annotation).

What's this context sensitive stuff you speak of? The link provided
earlier is devoid of useful information.