[PATCH] Don't generate 64-bit movd after cmpneqsd in 32-bit mode (PR19059)
hans at chromium.org
Mon Mar 10 18:14:42 PDT 2014
On Mon, Mar 10, 2014 at 5:17 PM, Jim Grosbach <grosbach at apple.com> wrote:
> While it does seem a bit roundabout, it does actually capture the semantics
> of what’s happening more accurately, which is good, and without a
> target-specific ISD node or a pseudo-instruction. I like the idea.
> On Mar 10, 2014, at 5:11 PM, Craig Topper <craig.topper at gmail.com> wrote:
> I was sort of wondering if we could scalar_to_vector this to a vector with
> the right sized elements, bitcast that to an f32 vector, then extract
> element on that. I'll admit its rather convoluted.
That seems to work :) New patch uploaded.
More information about the llvm-commits