[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Compilers flags
|
From: |
Fredrik Lingvall |
|
Subject: |
Re: Compilers flags |
|
Date: |
Mon, 17 Jan 2011 08:02:39 +0100 |
|
User-agent: |
Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.13) Gecko/20110105 Lightning/1.0b3pre Thunderbird/3.1.7 |
On 01/14/11 16:45, Jaroslav Hajek wrote:
> On Fri, Jan 14, 2011 at 4:18 PM, CdeMills <address@hidden> wrote:
>> Hello,
>>
>> Web pages about High Performance Computing mention gcc flag -ftree-vectorize
>> as a performance gain by vectorizing loops. I compiled a version from hg
>> tip, configured as
>>
>> env CFLAGS="-O3 -ftree-vectorize" CXXFLAGS="-O3 -ftree-vectorize"
>> FFLAGS="-O3 -ftree-vectorize" ../configure --prefix=$HOME/usr
>> --enable-strict-warning-flags
>>
>> [In most architectures except Power, -ftree-vectorize is implicitelly
>> available in -O3 ]
>> The tests done on this version were all successfull. Are there some kind of
>> benchmarks to quantify the speed gain ?
>>
>> Regards
>>
>> Pascal
>>
> Try the "benchmark" OctaveForge package.
This may also be of interest:
http://www.gentoo.org/doc/en/gcc-optimization.xml
See in particular the info about using -O3 and gcc 4.
/Fredrik