mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-23 19:23:23 +01:00
a0a83cc9da
Instead of creating a pattern like "(p && a) || ((!p) && b)", just expand the i8 operands to i32 and perform the selp on them. Fixes PR22246 llvm-svn: 227123
15 lines
500 B
LLVM
15 lines
500 B
LLVM
; RUN: llc < %s -march=nvptx -mcpu=sm_20 | FileCheck %s
|
|
|
|
target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
|
|
target triple = "nvptx64-nvidia-cuda"
|
|
|
|
; CHECK-LABEL: _Z3foobbbPb
|
|
define void @_Z3foobbbPb(i1 zeroext %p1, i1 zeroext %p2, i1 zeroext %p3, i8* nocapture %output) {
|
|
entry:
|
|
; CHECK: selp.b32 %r{{[0-9]+}}, %r{{[0-9]+}}, %r{{[0-9]+}}, %p{{[0-9]+}}
|
|
%.sink.v = select i1 %p1, i1 %p2, i1 %p3
|
|
%frombool5 = zext i1 %.sink.v to i8
|
|
store i8 %frombool5, i8* %output, align 1
|
|
ret void
|
|
}
|