[PATCH] ARM FastISel fix load register classes

[Renato Golin]

On 9 Jun 2013 05:20, "Chandler Carruth" <chandlerc at google.com> wrote:
> Renato, while in extreme cases (IE, when we simply have zero way of
constructing test cases at this point) this is true, I think it should be a
last resort. If at all possible, a test case should still go in with the
patch

Hi Chandler,

This is certainly true for most cases, but when writing big changes (like
Tim on AArch64 backend), sometimes it's far easier to fix a future problem
now and test later than try to come up with a test case that will have to
be changed again.

But if you think this case strongly needs a test, than, by all means, I'm
not against it.

cheers,

--renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130609/3413f0fe/attachment.html>