[llvm] r268539 - [AArch64] Use the reciprocal estimation machinery

[Eric Christopher via llvm-commits]

I've replied with the problems in another email, also you have two long
term committers and two people who have extensively touched both the
aarch64 backend and the target machinery in general asking you to revert.

Thanks!

-eric

On Tue, Sep 20, 2016 at 7:09 AM Evandro Menezes <e.menezes at samsung.com>
wrote:

> Renato,
>
> I understand that, but I don't think that there's been enough discussion
> after the commit to support its reversal.
>
> Thank you,
>
> --
> Evandro Menezes
>
> On 09/20/16 09:04, Renato Golin wrote:
> > On 20 September 2016 at 14:13, Evandro Menezes <e.menezes at samsung.com>
> wrote:
> >> But improvements to existing implementations should not imply their
> >> reversal.
> > You're taking the position that the commit is in, therefore it's
> > accepted, and any reversal would be a "regression". This is not how
> > the community behaves. Everyone is expected to follow through
> > post-commit review as serious as pre-commit review.
> >
> > cheers,
> > --renato
> >
> >
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160920/6a8556cc/attachment.html>