On Mon, Jan 01, 2007 at 02:30:33PM -0800, dean gaudet wrote:
> these files have assembly:
> ./build-tree/zlib-1.2.3/contrib/asm586/match.S
> ./build-tree/zlib-1.2.3/contrib/asm686/match.S
> ./build-tree/zlib-1.2.3/contrib/inflate86/inffast.S
> but it doesn't appear that they're actually being us
On Mon, 1 Jan 2007, dean gaudet wrote:
> and i can't even reproduce my results... here's the averages of the user
> cpu seconds for 10 runs of "minizip -9o a.zip linux-2.6.19.tar":
>
> baseline -DUNALIGNED_OK
> k8 revF26.62 26.59
> core2 28.43 28.44
you know, gzip
On Mon, 1 Jan 2007, Mark Brown wrote:
> On Wed, Sep 06, 2006 at 05:51:39PM -0700, dean gaudet wrote:
>
> > note that this define wasn't necessary on 32-bit x86 because there's
> > custom 32-bit assembly which uses unaligneds even more aggressively than
> > the C code does even when given UNALIG
On Wed, Sep 06, 2006 at 05:51:39PM -0700, dean gaudet wrote:
> note that this define wasn't necessary on 32-bit x86 because there's
> custom 32-bit assembly which uses unaligneds even more aggressively than
> the C code does even when given UNALIGNED_OK.
Which custom 32 bit assembly are you ref
Package: zlib
Version: 1.2.3-13
please define UNALIGNED_OK when building the amd64 target... unaligneds
are very inexpensive on all intel and amd cpus.
i benchmarked gzip -9 on linux-2.6.17.tar with this define and i see a
2.5% speedup on p4, a64, and a 9% speedup on core2. the zlib source cod
5 matches
Mail list logo