[llvm-dev] Need help with code generation

[Hal Finkel via llvm-dev]

----- Original Message -----
> From: "Rafael Espíndola" <rafael.espindola at gmail.com>
> To: "Hal Finkel" <hfinkel at anl.gov>
> Cc: "llvm-dev" <llvm-dev at lists.llvm.org>, "Bruce Hoult" <bruce at hoult.org>, "Mehdi Amini" <mehdi.amini at apple.com>
> Sent: Tuesday, March 22, 2016 2:09:21 PM
> Subject: Re: [llvm-dev] Need help with code generation
> 
> > That's because we seem to be debating whether we'd actively reject
> > a patch to fix such issues, not how important they are to us to
> > fix.
> 
> I would not work on it. Including not review it while there are
> actual
> missing features to be implemented.

I disagree with your characterization. "Actual missing features" are anything for which there is community support and there are contributors.

> 
> If you want to call that a low priority bug, go for it. I don't find
> it honest to do that myself.

That's fine. No one can force you to review anything. Generically, Rui, as the code owner, is responsible for serving as reviewer of last resort. However, lld is an LLVM project, and any contributor familiar with our general conventions and the current design of the code in question is qualified to approve a patch to remove undefined behavior from our code.

Rafael, please don't misunderstand me. I think you (and Rui and George and several others) have done an excellent job moving lld forward. I'm very excited about it. I'm concerned, however, that you would effectively turn away potential contributors (especially contributors interested in overall code quality, not just some particular bell or whistle) by implying that reviews of their patches would be indefinitely delayed. I don't think that's healthy, either for lld in particular, or for the LLVM community more generally.

Thanks again,
Hal

> 
> Cheers,
> Rafael
> 

-- 
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory