6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-07-26 20:26:07 +00:00
Code Issues Projects Releases Wiki Activity

Fix PR12529. The Vxx family of instructions are only supported by AVX.

Browse Source 
Use non-vex instructions for SSE4.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154770 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Nadav Rotem
2012-04-15 19:36:44 +00:00
parent e747fadedf
commit f16af0a053
2 changed files with 14 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
lib/Target/X86/X86InstrSSE.td
View File

@@ -6742,6 +6742,16 @@ let Predicates = [HasAVX] in {
def : Pat<(v4f64 (X86Blendpd (v4f64 VR256:$src1), (v4f64 VR256:$src2), def : Pat<(v4f64 (X86Blendpd (v4f64 VR256:$src1), (v4f64 VR256:$src2),
(imm:$mask))), (imm:$mask))),
(VBLENDPDYrri VR256:$src2, VR256:$src1, imm:$mask)>; (VBLENDPDYrri VR256:$src2, VR256:$src1, imm:$mask)>;
def : Pat<(v8i16 (X86Blendpw (v8i16 VR128:$src1), (v8i16 VR128:$src2),
(imm:$mask))),
(VPBLENDWrri VR128:$src2, VR128:$src1, imm:$mask)>;
def : Pat<(v4f32 (X86Blendps (v4f32 VR128:$src1), (v4f32 VR128:$src2),
(imm:$mask))),
(VBLENDPSrri VR128:$src2, VR128:$src1, imm:$mask)>;
def : Pat<(v2f64 (X86Blendpd (v2f64 VR128:$src1), (v2f64 VR128:$src2),
(imm:$mask))),
(VBLENDPDrri VR128:$src2, VR128:$src1, imm:$mask)>;
} }
let Predicates = [HasAVX2] in { let Predicates = [HasAVX2] in {
@@ -6802,13 +6812,13 @@ let Predicates = [HasSSE41] in {
def : Pat<(v8i16 (X86Blendpw (v8i16 VR128:$src1), (v8i16 VR128:$src2), def : Pat<(v8i16 (X86Blendpw (v8i16 VR128:$src1), (v8i16 VR128:$src2),
(imm:$mask))), (imm:$mask))),
(VPBLENDWrri VR128:$src2, VR128:$src1, imm:$mask)>; (PBLENDWrri VR128:$src2, VR128:$src1, imm:$mask)>;
def : Pat<(v4f32 (X86Blendps (v4f32 VR128:$src1), (v4f32 VR128:$src2), def : Pat<(v4f32 (X86Blendps (v4f32 VR128:$src1), (v4f32 VR128:$src2),
(imm:$mask))), (imm:$mask))),
(VBLENDPSrri VR128:$src2, VR128:$src1, imm:$mask)>; (BLENDPSrri VR128:$src2, VR128:$src1, imm:$mask)>;
def : Pat<(v2f64 (X86Blendpd (v2f64 VR128:$src1), (v2f64 VR128:$src2), def : Pat<(v2f64 (X86Blendpd (v2f64 VR128:$src1), (v2f64 VR128:$src2),
(imm:$mask))), (imm:$mask))),
(VBLENDPDrri VR128:$src2, VR128:$src1, imm:$mask)>; (BLENDPDrri VR128:$src2, VR128:$src1, imm:$mask)>;
} }

2
test/CodeGen/X86/vec_shuffle-20.ll
View File

@@ -1,4 +1,4 @@
; RUN: llc < %s -o /dev/null -march=x86 -mcpu=corei7 -mtriple=i686-apple-darwin9 -stats -info-output-file - | grep asm-printer | grep 2 ; RUN: llc < %s -o /dev/null -march=x86 -mcpu=corei7 -mtriple=i686-apple-darwin9 -stats -info-output-file - | grep asm-printer | grep 3
define <4 x float> @func(<4 x float> %fp0, <4 x float> %fp1) nounwind { define <4 x float> @func(<4 x float> %fp0, <4 x float> %fp1) nounwind {
entry: entry: