This is the mail archive of the 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 v3] powerpc: strstr optimization

On 07/27/2015 05:19 AM, OndÅej BÃlka wrote:
> 1. Reviews, it would be better to always write reviews publictly. But as
> you still have them could you send reviews here to clarify.

As machine maintainers they can commit their code as-if they had consensus.

Your after-the-fact reviews are good, and should be considered follow-on
work. You should work closely with IBM in a professional and technical
manner, present your own patches to the existing IBM code and discus
the testing and changes that you made. All of this *on top* of whatever
is in the existing master branch. You must present this work in a clear
and concise manner, providing IBM the tools with which to evaluate and test
your implementation.

The IBM maintainers do not need to convince you. They have consensus as
maintainers. You need to convince *them* that your solution is better rather
than attempting to block their patches, which is not your responsibility.

Lastly, beware that your single dissenting opinion may not constitute
an important part of the concerned interests of the glibc community[1].
Therefore, even if you are correct, the community may tell you that your
comments will not be considered until you find a way to work with the
machine maintainer.

I for one would like to see you working *with* IBM instead of what appears
to be an antagonistic realtionship surrounding these performance-related
changes. I think both sides should look to the positive aspects of having
more people reviewing existing implementations for performance benefits.



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