Betr: faster bzip2

cedric.dewijs at telfort.nl cedric.dewijs at telfort.nl
Tue Mar 2 06:56:28 PST 2010


>-- Oorspronkelijk bericht --
>From: Robert Connolly <robert at linuxfromscratch.org>
>To: General chatter list <lfs-chat at linuxfromscratch.org>
>Subject: faster bzip2
>Date: Mon, 01 Mar 2010 04:32:06 -0500
>Reply-To: General chatter list <lfs-chat at linuxfromscratch.org>
>
>
>Hi.
>
>This patch builds bzip2 with profile guided optimization, the --combine

>option, and -fomit-frame-pointer:
>http://www.linuxfromscratch.org/~robert/new/patches/bzip2-1.0.5-faster.diff
>
>The non-shared bzip2 program decompresses 7% faster for me, without accounting
>
>for -fomit-frame-pointer. The libbz2.so library runs 4.5% faster for me.
>The 
>profiling optimization works a lot better when the all the code is built
>
>together, so the non-shared bzip2 program is suggested. So:
>cp -v bzip2 /bin/bzip2 # Not bzip2-shared
>
>I also tried other options like -fwhole-program, -O3, and -finline-functions,
>
>and they made bzip2 and the library slower. My tests were only with 
>decompression, of a gcc tarball, but compression should also be faster.
>
>There are also other bzip2 packages that do parallel/multithread compression,
>
>but not parallel decompression.
>
>robert

Hi Robert,

I have nothing usefull (on-topic) to say, my programming skills are not high
enough to even understand what your compiling flags do. I do however understood
that you have managed to speed up a widely used program.

I would like to salute you. This is exactly what drives me to use linux instead
of windows. Instead of thinking: "It runs good enough to sell" I see "let's
spend a large amout of time to see if it can run even better!"

This will eventually create the perfect operating system.

Sorry for wasting bandwith, but I had to say it.

Best regards,
Cedric

       






More information about the lfs-chat mailing list