- User Since
- May 5 2014, 7:26 AM (327 w, 4 d)
@vitalybuka Please can you investigate this buildbot failure: http://lab.llvm.org:8011/builders/llvm-clang-x86_64-expensive-checks-ubuntu/builds/8198 - it appears to be due to an APInt being self-assigned
a few minors - you might also want to keep an eye on D85794
Wed, Aug 12
The -misched-print-dags approach seems to be doable, and not too unpretty.
Tue, Aug 11
@andreadb How well could we test this using the llvm-mca resource schedule tests?
@greened Any luck with a test case?
Pre-commit the test case and rebase to show the diff?
@asb Any thoughts on the RISCV changes? This is the last blocker on the patch.
Mon, Aug 10
Abandon this now D85553 has landed?
Sun, Aug 9
Sat, Aug 8
LGTM - the getScalarizationOverhead() change is OK
Fri, Aug 7
LGTM with one minor @nikic Does this look OK now?
Are you saying you think we can get away with removing X86ISD::ANDNP entirely? We don't have it for the BMI scalar variant, and it /usually/ works.
Thu, Aug 6
Thanks @lebedev.ri !
Thanks, no more comments from me - the reverse_hadd_v4f32 fix will be very useful to x86 backend hadd formation!
Wed, Aug 5
Thanks for checking! LGTM
A couple of minor style comments - trivial scope reductions
Tue, Aug 4
This landed at rGf8cc94a61afe48a4b77f111a8ad313fc3b9417de
Maybe make it clear at the start of InstCombinerImpl::foldSelectOpOp where we use the matchers that we can safely use the matchers for the select(icmp) pattern OR the intrinsics?