This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Don't use SSE4_2 instructions on Intel Silvermont Micro Architecture.


Ondrej,

> Here,
> http://sourceware.org/bugzilla/show_bug.cgi?id=15424 
> I am not sure which original mistake was but there was problem that gcc
> optimized call away.

That's great, we found the problem, filed an issue and it's fixed.

The framework is in its infancy, and we need the help of experienced
people, including you, if we're going to have a chance at making the
contributed code better.

Are you suggesting we could we have done something better?

Do you have any recommendations?

>> In my LFCS2013 talk I came out and directly asked for people
>> with performance measurement experience to help glibc.
>>
> Could you also invite them to say their opinion?

Certainly.

>> Surely the implementation can't offend you so much that you
>> aren't willing to help us make it better? :-)
>>
> A question is if I staring from scratch would be faster than writing
> patch, waiting few weeks for review and meanwhile I need to rewrite
> patch because half internals changed and more bugs were introduced.

Write *new* tests from scratch, contribute them, and then we'll
phase out the old tests as we have new tests that cover the old
test behaviours?

We can and should have multiple tests.

Cheers,
Carlos.
 


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]