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:   Wed, 24 Jun 2020 16:37:24 -0700
From:   Sami Tolvanen <samitolvanen@google.com>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Masahiro Yamada <masahiroy@kernel.org>,
        Will Deacon <will@kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        "Paul E. McKenney" <paulmck@kernel.org>,
        Kees Cook <keescook@chromium.org>,
        Nick Desaulniers <ndesaulniers@google.com>,
        clang-built-linux@googlegroups.com,
        kernel-hardening@lists.openwall.com, linux-arch@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org, linux-kbuild@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
        x86@kernel.org, George Burgess IV <gbiv@google.com>
Subject: Re: [PATCH 06/22] kbuild: lto: limit inlining
Message-ID: <20200624233724.GA94769@google.com>
References: <20200624203200.78870-1-samitolvanen@google.com>
 <20200624203200.78870-7-samitolvanen@google.com>
 <20200624212055.GU4817@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200624212055.GU4817@hirez.programming.kicks-ass.net>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Jun 24, 2020 at 11:20:55PM +0200, Peter Zijlstra wrote:
> On Wed, Jun 24, 2020 at 01:31:44PM -0700, Sami Tolvanen wrote:
> > This change limits function inlining across translation unit
> > boundaries in order to reduce the binary size with LTO.
> > 
> > The -import-instr-limit flag defines a size limit, as the number
> > of LLVM IR instructions, for importing functions from other TUs.
> > The default value is 100, and decreasing it to 5 reduces the size
> > of a stripped arm64 defconfig vmlinux by 11%.
> 
> Is that also the right number for x86? What about the effect on
> performance? What did 6 do? or 4?

This is the size limit we decided on for Android after testing on
arm64, but the number is obviously a compromise between code size
and performance. I'd be happy to benchmark this further once other
concerns have been resolved.

Sami