[llvm] r309330 - [ConstantFolder] Don't try to fold gep when the idx is a vector.

[Hans Wennborg via llvm-commits]

r309595. Thanks!

On Mon, Jul 31, 2017 at 10:39 AM, David Majnemer
<david.majnemer at gmail.com> wrote:
> Yes, let's merge it.
>
> On Mon, Jul 31, 2017 at 10:34 AM, Hans Wennborg <hans at chromium.org> wrote:
>>
>> On Thu, Jul 27, 2017 at 3:23 PM, Davide Italiano <davide at freebsd.org>
>> wrote:
>> > On Thu, Jul 27, 2017 at 3:20 PM, Davide Italiano via llvm-commits
>> > <llvm-commits at lists.llvm.org> wrote:
>> >> Author: davide
>> >> Date: Thu Jul 27 15:20:44 2017
>> >> New Revision: 309330
>> >>
>> >> URL: http://llvm.org/viewvc/llvm-project?rev=309330&view=rev
>> >> Log:
>> >> [ConstantFolder] Don't try to fold gep when the idx is a vector.
>> >>
>> >> The code in ConstantFoldGetElementPtr() assumes integers, and
>> >> therefore it crashes trying to get the integer bidwith of a vector
>> >> type (in this case <4 x i32>. I just changed the code to prevent
>> >> the folding in case of vectors and I didn't bother to generalize
>> >> as this doesn't seem to me something that really happens in
>> >> practice, but I'm willing to change the patch if you think
>> >> it's worth it.
>> >> This is hard to trigger from -instsimplify or -instcombine
>> >> only as the second instruction is dead, so the test uses loop-unroll.
>> >>
>> >
>> > Hans, David, can we consider merging this to 5.0?
>>
>> David: what do you think?
>
>