Speedup Function::getIntrinsicID() with caching of result

[Michael Ilseman]

The "opt" run is a better isolation of this change. If we were curious, we could do an opt run with non-overloadable intrinsics to see if that regressed at all. It's clearly a massive win for overloadable intrinsics, which were O(n).

On Mar 2, 2013, at 2:48 AM, Renato Golin <renato.golin at linaro.org> wrote:

> On 1 March 2013 19:41, Jean-Luc Duprat <jduprat at apple.com> wrote:
> After this change:
> $ time ./bin/llc  -march=arm -mattr=+neon vsub_JLD2.ll -O0
> real    0m10.440s (-5%)
> $ time ./bin/llc  -march=arm -mattr=+neon vsub_JLD2.ll -O2
> real    0m11.383s (-5%)
> 
> Wow, that's a big difference from the original 10%, but the times are also different, so I wonder how much it is really the changes and how much was noise...
> 
> --renato

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130302/39ba4d14/attachment.html>