[llvm] r201275 - Remove a very old instcombine where we would turn sequences of selects into

Nadav Rotem nrotem at apple.com
Wed Feb 12 16:10:34 PST 2014


On Feb 12, 2014, at 4:08 PM, Chandler Carruth <chandlerc at google.com> wrote:

> 
> On Wed, Feb 12, 2014 at 4:05 PM, Nadav Rotem <nrotem at apple.com> wrote:
> On Feb 12, 2014, at 3:54 PM, Owen Anderson <resistor at mac.com> wrote:
> 
> > Author: resistor
> > Date: Wed Feb 12 17:54:07 2014
> > New Revision: 201275
> >
> > URL: http://llvm.org/viewvc/llvm-project?rev=201275&view=rev
> > Log:
> > Remove a very old instcombine where
> 
> For vectors this optimization is always a win.  Can you please bring it back and add a check that the comparison type is a vector?
> 
> Do you have test cases which regressed? I find it odd that I saw no regressions from any of the heavily auto-vectorized benchmarks on x86-64.

I don’t have a specific testcase that regressed, this is purely theoretical.  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140212/84d28daf/attachment.html>


More information about the llvm-commits mailing list