flac dev - May 2012 - Git branch with compiling fixes for win32

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 07.05.2012 17:11, Miroslav Lichvar wrote:
> On Fri, May 04, 2012 at 11:22:00AM -0400, Cristian Rodr?guez
> wrote:
>> El 03/05/12 12:19, Miroslav Lichvar escribi?:
>> 
>> 
>>> It makes the C function faster than the corresponding asm
>>> routine, so if it's included I'd suggest to just drop the
asm
>>> function to not keep around more asm code than is necessary.
>> 
>> With current compilers it is very likely that those routines are
>> already superflous.
> 
> It seems the current compilers are not that good yet :).
> 
> In a test on a Core 2 machine with gcc-4.6.3, i686 flac build with 
> nasm enabled is about 7% faster in decoding than without nasm.
> x86_64 build is about 2% faster than the i686 build with nasm
> enabled.
> 
Was that with -O2 or -O3?

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJPqAQ3AAoJEOs4Jb6SI2CwsGUIAIrvJUZPPyC6snMLQxvYWCCb
OpxoxApO3+gRyAlk1702im1zBcziECit4AoNmetbLWpaYQNs6WXCEQ8HBJa7Aepq
gN85v/cG9yk0zVL5ntgbttgbDEuLI0OZo7LWtF0dKsBDRjt8U8LftZ9UvVUHmLQk
pw7v2pA6co7UXhNL/PMhHwVRsx6nAkRfHptQ2hbmLlqFlh4WJw1Pzt4rxilOHEtk
3p9PU1QgT5TQmxTuVaNjW/KGy7IgMIMS3s0y1tjLijwqFqIUsdDk5TuTKTD7QEwt
Ymn4ZT20DfyLRq8PMHiGQUc8Jksvb6enNrKe1hMK3/xof4VYDled4PdaVI+EJQg=q0Ag
-----END PGP SIGNATURE-----

On Mon, May 07, 2012 at 09:19:52PM +0400, LRN wrote:
> > In a test on a Core 2 machine with gcc-4.6.3, i686 flac build with 
> > nasm enabled is about 7% faster in decoding than without nasm.
> > x86_64 build is about 2% faster than the i686 build with nasm
> > enabled.
> > 
> Was that with -O2 or -O3?
Not sure, I didn't set CFLAGS which seems to result in using both:
-O3 -funroll-loops -Wall -W -Winline -g -O2

I think funroll-loops can be harmful, the Fedora package is compiled
with -O2 and funroll-loops is used only for the stream_encoder file
and it seems decoding is a tiny bit faster than with the git version.
Unfortunately I don't recall the details.

On Mon, May 07, 2012 at 01:23:03PM -0400, Cristian Rodr?guez
wrote:
> Did you build with -fprofile-generate ... then make check .. then 
> rebuild with -fprofile-use ?
No, I have not tried that.

-- 
Miroslav Lichvar

El 07/05/12 13:37, Miroslav Lichvar escribi?:
> On Mon, May 07, 2012 at 09:19:52PM +0400, LRN wrote:
>>> In a test on a Core 2 machine with gcc-4.6.3, i686 flac build with
>>> nasm enabled is about 7% faster in decoding than without nasm.
>>> x86_64 build is about 2% faster than the i686 build with nasm
>>> enabled.
>>>
>> Was that with -O2 or -O3?
>
> Not sure, I didn't set CFLAGS which seems to result in using both:
> -O3 -funroll-loops -Wall -W -Winline -g -O2
>
> I think funroll-loops can be harmful, the Fedora package is compiled
> with -O2 and funroll-loops is used only for the stream_encoder file
> and it seems decoding is a tiny bit faster than with the git version.
> Unfortunately I don't recall the details.
>
> On Mon, May 07, 2012 at 01:23:03PM -0400, Cristian Rodr?guez wrote:
>> Did you build with -fprofile-generate ... then make check .. then
>> rebuild with -fprofile-use ?
>
> No, I have not tried that.
>
Yes, -funroll-loops might cause code growth and/or make binaries 
actually slower...

BTW, just wondering, how are you measuring the speed in this tests ? I 
would be interested to know so I can reproduce it here with different 
compile/link options