On Thu, 3 Jan 2013 22:18:51 +0100 Jérémy Zurcher said:
good man! i like your stats. graphs were not really needed, but nice additions.
the tables below told me what i need... that its a benefit - small one, no
regressions worth mentioning etc. etc.
in svn it is now! :) trunk :) tnks muchly man!
the patches, the bench code and the git based script I used to collect them are
attached
charts too
here are the numbers,
they are not very impressive like that,
the optimized path is not easily triggered,
but when it does, it worth it.
not easy to speed up such a good work !
add
I'm ready,
numbers and tests will be send in a few hours,
when the kids are in bed and the kitchen is cleaned ;)
On Thursday 03 January 2013 17:17, Carsten Haitzler wrote :
> On Thu, 29 Nov 2012 22:38:35 +0100 Jérémy Zurcher said:
>
> > > the optimized path implise that we are in hash collision
On Thu, 29 Nov 2012 22:38:35 +0100 Jérémy Zurcher said:
> > the optimized path implise that we are in hash collision.
> WRONG !
> > I'll try to come back with something this week-end.
> yes I will
i'm living in suspense for your numbers on perf etc :) (this is still
pending in my queue fyi -
> the optimized path implise that we are in hash collision.
WRONG !
> I'll try to come back with something this week-end.
yes I will
--
Keep yourself connected to Go Parallel:
VERIFY Test and improve your parallel project
On Monday 26 November 2012 17:37, Cedric BAIL wrote :
> Hello,
>
> On Wed, Nov 21, 2012 at 7:39 AM, Jérémy Zurcher wrote:
> > here are 2 patches to speed up eina_share_common_del with minimal costs on
> > eina_share_common_add side.
> >
> > the second might be controversial
> >
> > 0001-eina_sha
Hello,
On Wed, Nov 21, 2012 at 7:39 AM, Jérémy Zurcher wrote:
> here are 2 patches to speed up eina_share_common_del with minimal costs on
> eina_share_common_add side.
>
> the second might be controversial
>
> 0001-eina_share_common_del-speed-up.patch
> builtin node is never unlinked even if e
Hi,
here are 2 patches to speed up eina_share_common_del with minimal costs on
eina_share_common_add side.
the second might be controversial
0001-eina_share_common_del-speed-up.patch
builtin node is never unlinked even if empty,
always is the last of the queue,
so that it can be used to ge