This is the mail archive of the
mailing list for the GCC project.
Re: gcc -S vs clang -S
- From: Martin Sebor <msebor at redhat dot com>
- To: Fei Ding <fdingiit at gmail dot com>, Thiago Farina <tfransosi at gmail dot com>
- Cc: Andrew Pinski <pinskia at gmail dot com>, Eric Botcazou <ebotcazou at libertysurf dot fr>, Olivier Galibert <galibert at pobox dot com>, GCC Mailing List <gcc at gcc dot gnu dot org>
- Date: Wed, 13 May 2015 08:41:39 -0600
- Subject: Re: gcc -S vs clang -S
- Authentication-results: sourceware.org; auth=none
- References: <CACnwZYe=psc2jqUeva3ZxFHvADz3D7SHqRAVyGcHyVOuRca+NA at mail dot gmail dot com> <8AF3EC7F-E86D-462D-8E07-B51162E9271E at gmail dot com> <CAMFK0gscitxWu70QHneBgF7dtYXcysWF7nngooOcztyvn8BBjA at mail dot gmail dot com> <1462026 dot YT44ftsO3U at polaris> <CAGmPkfJMjdg_U40oWKnbZRNqQS+j1ksqZ6U_+wRrHuPO=uneWQ at mail dot gmail dot com> <CA+=Sn1k9+82hUqe9OC32ZbEfDYmK6_fqVERQjL7U1hwucSMudA at mail dot gmail dot com>
On 05/12/2015 07:40 PM, Andrew Pinski wrote:
On Tue, May 12, 2015 at 6:36 PM, Fei Ding <firstname.lastname@example.org> wrote:
I think Thiago and Eric just want to know which code-gen is better and why...
You need to understand for a complex process (CISC ISAs) like x86,
there is no one right answer sometimes. You need to look at each
micro-arch and understand the pipeline. Sometimes different code
stream will performance the same but it also depends on the code size
A good place to start is the Intel 64 and IA-32 Architectures
Optimization Reference Manual. It lists the throughput and
latencies of x86 instructions and gives guidance for which
ones might be more efficient on which processors. For example,
in the section titled Using LEA it discusses why the three
operand form of the instruction is slower on the Sandy Bridge
microarchitecture than on others:
2015-05-12 23:29 GMT+08:00 Eric Botcazou <email@example.com>:
Note that at -O3 there is a difference still:
addl %esi, %edi
movl %edi, %eax
leal (%rdi,%rsi), %eax
Can't tell which is best, if any.
But what's your point exactly here? You cannot expect different compilers to
generate exactly the same code on a given testcase for non-toy architectures.
Note that this kind of discussion is more appropriate for firstname.lastname@example.org