https://bugs.llvm.org/show_bug.cgi?id=46897
Bug ID: 46897
Summary: Failure to recognise commutable binop intrinsics
Product: libraries
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: Scalar Optimizations
Assignee: unassignedb...@nondot.org
Reporter: llvm-...@redking.me.uk
CC: lebedev...@gmail.com, llvm-bugs@lists.llvm.org,
spatel+l...@rotateright.com
define <2 x i64> @smax(<2 x i64> %a, <2 x i64> %b) {
%x = call <2 x i64> @llvm.smax.v2i64(<2 x i64> %a, <2 x i64> %b)
%y = call <2 x i64> @llvm.smax.v2i64(<2 x i64> %b, <2 x i64> %a)
%o = or <2 x i64> %x, %y
ret <2 x i64> %o
}
declare <2 x i64> @llvm.smax.v2i64(<2 x i64>, <2 x i64>)
opt -O3
define <2 x i64> @smax(<2 x i64> %a, <2 x i64> %b) {
%x = tail call <2 x i64> @llvm.smax.v2i64(<2 x i64> %a, <2 x i64> %b)
%y = tail call <2 x i64> @llvm.smax.v2i64(<2 x i64> %b, <2 x i64> %a)
%o = or <2 x i64> %x, %y
ret <2 x i64> %o
}
but we should be able to recognise that %x and %y are equivalent:
define <2 x i64> @smax(<2 x i64> %a, <2 x i64> %b) {
%x = tail call <2 x i64> @llvm.smax.v2i64(<2 x i64> %a, <2 x i64> %b)
ret <2 x i64> %x
}
--
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
