[X86] Use vector widening to support sign extend from i1 when the dest type is not 512-bits and vlx is not enabled.
Previously we used a wider element type and truncated. But its more efficient to keep the element type and drop unused elements.
If BWI isn't supported and we have a i16 or i8 type, we'll extend it to be i32 and still use a truncate.
llvm-svn: 319740
diff --git a/llvm/test/CodeGen/X86/avg-mask.ll b/llvm/test/CodeGen/X86/avg-mask.ll
index 4eacbdd..578d7aa 100644
--- a/llvm/test/CodeGen/X86/avg-mask.ll
+++ b/llvm/test/CodeGen/X86/avg-mask.ll
@@ -252,8 +252,8 @@
; AVX512F: # %bb.0:
; AVX512F-NEXT: vpavgw %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: kmovw %edi, %k1
-; AVX512F-NEXT: vpternlogq $255, %zmm1, %zmm1, %zmm1 {%k1} {z}
-; AVX512F-NEXT: vpmovqw %zmm1, %xmm1
+; AVX512F-NEXT: vpternlogd $255, %zmm1, %zmm1, %zmm1 {%k1} {z}
+; AVX512F-NEXT: vpmovdw %zmm1, %ymm1
; AVX512F-NEXT: vpblendvb %xmm1, %xmm0, %xmm2, %xmm0
; AVX512F-NEXT: vzeroupper
; AVX512F-NEXT: retq
@@ -280,8 +280,8 @@
; AVX512F: # %bb.0:
; AVX512F-NEXT: vpavgw %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: kmovw %edi, %k1
-; AVX512F-NEXT: vpternlogq $255, %zmm1, %zmm1, %zmm1 {%k1} {z}
-; AVX512F-NEXT: vpmovqw %zmm1, %xmm1
+; AVX512F-NEXT: vpternlogd $255, %zmm1, %zmm1, %zmm1 {%k1} {z}
+; AVX512F-NEXT: vpmovdw %zmm1, %ymm1
; AVX512F-NEXT: vpand %xmm0, %xmm1, %xmm0
; AVX512F-NEXT: vzeroupper
; AVX512F-NEXT: retq