[PATCH] D34336: [x86] transform vector inc/dec to use -1 constant (PR33483)

[Sanjay Patel via Phabricator via llvm-commits]

spatel added a comment.

Here's a test loop:
F3453503: inc_test.c <https://reviews.llvm.org/F3453503>

And asm (Linux x86 flavor):
F3453504: inc.s <https://reviews.llvm.org/F3453504>

compile with:
$ clang -O1 inc_test.c inc.s

I used this to benchmark this transform on Haswell and Jaguar, but I couldn't measure any difference. I also had expanded versions of the asm that repeated the padd/psub up to 128 times...but still no perf diff on either chip.


https://reviews.llvm.org/D34336