mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-17 03:30:28 +00:00
cdbdfa28d1
to the zero-extend-vector-inreg node introduced previously for the same purpose: manage the type legalization of widened extend operations, especially to support the experimental widening mode for x86. I'm adding both because sign-extend is expanded in terms of any-extend with shifts to propagate the sign bit. This removes the last fundamental scalarization from vec_cast2.ll (a test case that hit many really bad edge cases for widening legalization), although the trunc tests in that file still appear scalarized because the the shuffle legalization is scalarizing. Funny thing, I've been working on that. Some initial experiments with this and SSE2 scenarios is showing moderately good behavior already for sign extension. Still some work to do on the shuffle combining on X86 before we're generating optimal sequences, but avoiding scalarization is a huge step forward. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@212714 91177308-0d34-0410-b5e6-96231b3b80d8
77 lines
2.0 KiB
LLVM
77 lines
2.0 KiB
LLVM
; RUN: llc < %s -mtriple=i386-apple-darwin10 -mcpu=corei7-avx -mattr=+avx | FileCheck %s
|
|
; RUN: llc < %s -mtriple=i386-apple-darwin10 -mcpu=corei7-avx -mattr=+avx -x86-experimental-vector-widening-legalization | FileCheck %s --check-prefix=CHECK-WIDE
|
|
|
|
;CHECK-LABEL: foo1_8:
|
|
;CHECK: vcvtdq2ps
|
|
;CHECK: ret
|
|
;
|
|
;CHECK-WIDE-LABEL: foo1_8:
|
|
;CHECK-WIDE: vpmovzxbd %xmm0, %xmm1
|
|
;CHECK-WIDE-NEXT: vpslld $24, %xmm1, %xmm1
|
|
;CHECK-WIDE-NEXT: vpsrad $24, %xmm1, %xmm1
|
|
;CHECK-WIDE-NEXT: vpshufb {{.*}}, %xmm0, %xmm0
|
|
;CHECK-WIDE-NEXT: vpslld $24, %xmm0, %xmm0
|
|
;CHECK-WIDE-NEXT: vpsrad $24, %xmm0, %xmm0
|
|
;CHECK-WIDE-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
|
|
;CHECK-WIDE-NEXT: vcvtdq2ps %ymm0, %ymm0
|
|
;CHECK-WIDE-NEXT: ret
|
|
define <8 x float> @foo1_8(<8 x i8> %src) {
|
|
%res = sitofp <8 x i8> %src to <8 x float>
|
|
ret <8 x float> %res
|
|
}
|
|
|
|
;CHECK-LABEL: foo1_4:
|
|
;CHECK: vcvtdq2ps
|
|
;CHECK: ret
|
|
;
|
|
;CHECK-WIDE-LABEL: foo1_4:
|
|
;CHECK-WIDE: vpmovzxbd %xmm0, %xmm0
|
|
;CHECK-WIDE-NEXT: vpslld $24, %xmm0, %xmm0
|
|
;CHECK-WIDE-NEXT: vpsrad $24, %xmm0, %xmm0
|
|
;CHECK-WIDE-NEXT: vcvtdq2ps %xmm0, %xmm0
|
|
;CHECK-WIDE-NEXT: ret
|
|
define <4 x float> @foo1_4(<4 x i8> %src) {
|
|
%res = sitofp <4 x i8> %src to <4 x float>
|
|
ret <4 x float> %res
|
|
}
|
|
|
|
;CHECK-LABEL: foo2_8:
|
|
;CHECK: vcvtdq2ps
|
|
;CHECK: ret
|
|
;
|
|
;CHECK-WIDE-LABEL: foo2_8:
|
|
;CHECK-WIDE: vcvtdq2ps %ymm{{.*}}, %ymm{{.*}}
|
|
;CHECK-WIDE: ret
|
|
define <8 x float> @foo2_8(<8 x i8> %src) {
|
|
%res = uitofp <8 x i8> %src to <8 x float>
|
|
ret <8 x float> %res
|
|
}
|
|
|
|
;CHECK-LABEL: foo2_4:
|
|
;CHECK: vcvtdq2ps
|
|
;CHECK: ret
|
|
;
|
|
;CHECK-WIDE-LABEL: foo2_4:
|
|
;CHECK-WIDE: vcvtdq2ps %xmm{{.*}}, %xmm{{.*}}
|
|
;CHECK-WIDE: ret
|
|
define <4 x float> @foo2_4(<4 x i8> %src) {
|
|
%res = uitofp <4 x i8> %src to <4 x float>
|
|
ret <4 x float> %res
|
|
}
|
|
|
|
;CHECK-LABEL: foo3_8:
|
|
;CHECK: vcvttps2dq
|
|
;CHECK: ret
|
|
define <8 x i8> @foo3_8(<8 x float> %src) {
|
|
%res = fptosi <8 x float> %src to <8 x i8>
|
|
ret <8 x i8> %res
|
|
}
|
|
;CHECK-LABEL: foo3_4:
|
|
;CHECK: vcvttps2dq
|
|
;CHECK: ret
|
|
define <4 x i8> @foo3_4(<4 x float> %src) {
|
|
%res = fptosi <4 x float> %src to <4 x i8>
|
|
ret <4 x i8> %res
|
|
}
|
|
|