[llvm-commits] [llvm] r71013 - in /llvm/trunk: lib/Transforms/Scalar/LoopStrengthReduce.cpp test/Transforms/LoopStrengthReduce/2009-05-04-Overflow.ll

[Bill Wendling]

On Wed, May 6, 2009 at 12:45 PM, David Greene <dag at cray.com> wrote:
> On Wednesday 06 May 2009 13:30, Chris Lattner wrote:
>
>> Having a patch be reverted should not be seen as a negative thing.
>
> It's not a negative thing as in, "I'm taking this personally."  It IS
> a negative thing in terms of time-to-fix.  Dealing with conflicts,
> etc. just means we take more time to fix problems.
>
>> There are many people pulling from mainline svn: if there is a problem
>> with a patch, I think it makes sense to revert it immediately.
>
> I really disagree with this.  If someone isn't responding for days, sure,
> but reverting a patch means that when I update I lose some functionality
> I thought would be there.  Things start failing that I don't expect to
> fail, etc.  I then end up tracking down a bug I've already tracked down
> before.
>
> A simple personal e-mail asking for a fix should be sufficient.
>
I normally send out an email saying that something failed and that I
reverted the patch. I apologize if you didn't get it or it got
rerouted.

-bw