https://bugs.llvm.org/show_bug.cgi?id=47949
Bug ID: 47949
Summary: CTPOP(x) > 0 isn't optimized away before SelectionDAG?
Product: libraries
Version: trunk
Hardware: PC
OS: All
Status: NEW
Severity: normal
Priority: P
Component: Common Code Generator Code
Assignee: unassignedb...@nondot.org
Reporter: d...@znu.io
CC: llvm-bugs@lists.llvm.org
I committed exhaustive x86 CTPOP tests to the repo this morning
(test/CodeGen/X86/vector-popcnt-{128,256,512}-ult-ugt.ll).
Notably, unsigned greater-than-zero tests are generating actual CTPOP emulation
when they don't need to. It should just optimize down to a comparison against
zero.
My limited investigation shows that SelectionDAG can fix this, but shouldn't an
earlier pass have caught this?
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
