Joakim Tjernlund wrote:
Noticed while optimizing crc16 that gcc -O performed much better
than gcc -O2 while doing crc16:
I have also seen -O1 beat -O2 when with a large FORTRAN program.
http://www.hep.manchester.ac.uk/u/sam/zgoubi-optimise/
Sam
Jakub Jelinek wrote on 2010/02/14 20:21:50:
>
> On Sun, Feb 14, 2010 at 08:06:20PM +0100, Joakim Tjernlund wrote:
> > the warmup isn't really needed after I added memset and inline or not should
> > not matter that much are only used once so I think my
> > conclusion still stands: gcc 4.3.4 is slo
Hello
On 14.02.10, you wrote:
>
> Glad to hear that newer versions are back on track, does than mean
> that this won't be fixed in gcc 4.3.x series?
maybe you test the GCC 4.5.0 Version.
I notice in zlib (used in PNG image compression) when there is a image that
contain lots same pixel values
On Sun, Feb 14, 2010 at 08:06:20PM +0100, Joakim Tjernlund wrote:
> the warmup isn't really needed after I added memset and inline or not should
> not matter that much are only used once so I think my
> conclusion still stands: gcc 4.3.4 is slower with -O2 than -O1
You mean this particular GCC ver
On Sun, Feb 14, 2010 at 8:06 PM, Joakim Tjernlund
wrote:
> Richard Guenther wrote on 2010/02/14 19:05:24:
>>
>> On Sun, Feb 14, 2010 at 5:51 PM, Joakim Tjernlund
>> wrote:
>> >
>> > Noticed while optimizing crc16 that gcc -O performed much better
>> > than gcc -O2 while doing crc16:
>>
>> Reduci
Richard Guenther wrote on 2010/02/14 19:05:24:
>
> On Sun, Feb 14, 2010 at 5:51 PM, Joakim Tjernlund
> wrote:
> >
> > Noticed while optimizing crc16 that gcc -O performed much better
> > than gcc -O2 while doing crc16:
>
> Reducing the noise by adding a loop with trip count 64, making sure
> my p
On Sun, Feb 14, 2010 at 5:51 PM, Joakim Tjernlund
wrote:
>
> Noticed while optimizing crc16 that gcc -O performed much better
> than gcc -O2 while doing crc16:
Reducing the noise by adding a loop with trip count 64, making sure
my powersaving model is fixed at performance I see
-O1:
crc1:f532 cr
