+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+sse3 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE3
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+ssse3 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSSE3
; SSE-NEXT: shufpd {{.*#+}} xmm0 = xmm0[1,0]
; SSE-NEXT: retq
;
-; AVX1-LABEL: shuffle_v2f64_10:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v2f64_10:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v2f64_10:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $1, %xmm0, %xmm0
-; AVX512VL-NEXT: retq
+; AVX-LABEL: shuffle_v2f64_10:
+; AVX: # BB#0:
+; AVX-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX-NEXT: retq
%shuffle = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 1, i32 0>
ret <2 x double> %shuffle
; SSE-NEXT: movhlps {{.*#+}} xmm0 = xmm0[1,1]
; SSE-NEXT: retq
;
-; AVX1-LABEL: shuffle_v2f64_11:
-; AVX1: # BB#0:
-; AVX1-NEXT: vmovhlps {{.*#+}} xmm0 = xmm0[1,1]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v2f64_11:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovhlps {{.*#+}} xmm0 = xmm0[1,1]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v2f64_11:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovhlps %xmm0, %xmm0, %xmm0
-; AVX512VL-NEXT: retq
+; AVX-LABEL: shuffle_v2f64_11:
+; AVX: # BB#0:
+; AVX-NEXT: vmovhlps {{.*#+}} xmm0 = xmm0[1,1]
+; AVX-NEXT: retq
%shuffle = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 1, i32 1>
ret <2 x double> %shuffle
}
; SSE-NEXT: movapd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX1-LABEL: shuffle_v2f64_32:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v2f64_32:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v2f64_32:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $1, %xmm1, %xmm0
-; AVX512VL-NEXT: retq
+; AVX-LABEL: shuffle_v2f64_32:
+; AVX: # BB#0:
+; AVX-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
+; AVX-NEXT: retq
%shuffle = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 3, i32 2>
ret <2 x double> %shuffle
; SSE-NEXT: movaps %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX1-LABEL: shuffle_v2f64_33:
-; AVX1: # BB#0:
-; AVX1-NEXT: vmovhlps {{.*#+}} xmm0 = xmm1[1,1]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v2f64_33:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovhlps {{.*#+}} xmm0 = xmm1[1,1]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v2f64_33:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovhlps %xmm1, %xmm1, %xmm0
-; AVX512VL-NEXT: retq
+; AVX-LABEL: shuffle_v2f64_33:
+; AVX: # BB#0:
+; AVX-NEXT: vmovhlps {{.*#+}} xmm0 = xmm1[1,1]
+; AVX-NEXT: retq
%shuffle = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 3, i32 3>
ret <2 x double> %shuffle
}
; AVX512VL-LABEL: shuffle_v2f64_bitcast_1z:
; AVX512VL: # BB#0:
; AVX512VL-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512VL-NEXT: vshufpd $1, %xmm1, %xmm0, %xmm0
+; AVX512VL-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1],xmm1[0]
; AVX512VL-NEXT: retq
%shuffle64 = shufflevector <2 x double> %a, <2 x double> zeroinitializer, <2 x i32> <i32 2, i32 1>
%bitcast32 = bitcast <2 x double> %shuffle64 to <4 x float>
; SSE41-NEXT: movddup {{.*#+}} xmm0 = mem[0,0]
; SSE41-NEXT: retq
;
-; AVX1-LABEL: insert_dup_mem_v2f64:
-; AVX1: # BB#0:
-; AVX1-NEXT: vmovddup {{.*#+}} xmm0 = mem[0,0]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: insert_dup_mem_v2f64:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovddup {{.*#+}} xmm0 = mem[0,0]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: insert_dup_mem_v2f64:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovsd (%rdi), %xmm0
-; AVX512VL-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
-; AVX512VL-NEXT: retq
+; AVX-LABEL: insert_dup_mem_v2f64:
+; AVX: # BB#0:
+; AVX-NEXT: vmovddup {{.*#+}} xmm0 = mem[0,0]
+; AVX-NEXT: retq
%a = load double, double* %ptr
%v = insertelement <2 x double> undef, double %a, i32 0
%shuffle = shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> <i32 0, i32 0>
ret <2 x double> %shuffle
}
+define <2 x double> @insert_dup_mem128_v2f64(<2 x double>* %ptr) nounwind {
+; SSE2-LABEL: insert_dup_mem128_v2f64:
+; SSE2: # BB#0:
+; SSE2-NEXT: movaps (%rdi), %xmm0
+; SSE2-NEXT: movlhps {{.*#+}} xmm0 = xmm0[0,0]
+; SSE2-NEXT: retq
+;
+; SSE3-LABEL: insert_dup_mem128_v2f64:
+; SSE3: # BB#0:
+; SSE3-NEXT: movddup {{.*#+}} xmm0 = mem[0,0]
+; SSE3-NEXT: retq
+;
+; SSSE3-LABEL: insert_dup_mem128_v2f64:
+; SSSE3: # BB#0:
+; SSSE3-NEXT: movddup {{.*#+}} xmm0 = mem[0,0]
+; SSSE3-NEXT: retq
+;
+; SSE41-LABEL: insert_dup_mem128_v2f64:
+; SSE41: # BB#0:
+; SSE41-NEXT: movddup {{.*#+}} xmm0 = mem[0,0]
+; SSE41-NEXT: retq
+;
+; AVX-LABEL: insert_dup_mem128_v2f64:
+; AVX: # BB#0:
+; AVX-NEXT: vmovddup {{.*#+}} xmm0 = mem[0,0]
+; AVX-NEXT: retq
+ %v = load <2 x double>, <2 x double>* %ptr
+ %shuffle = shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> <i32 0, i32 0>
+ ret <2 x double> %shuffle
+}
+
+
define <2 x i64> @insert_dup_mem_v2i64(i64* %ptr) {
; SSE-LABEL: insert_dup_mem_v2i64:
; SSE: # BB#0:
; SSE-NEXT: shufpd {{.*#+}} xmm0 = xmm0[1,0]
; SSE-NEXT: retq
;
-; AVX1-LABEL: shuffle_mem_v2f64_10:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = mem[1,0]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_mem_v2f64_10:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} xmm0 = mem[1,0]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_mem_v2f64_10:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $1, (%rdi), %xmm0
-; AVX512VL-NEXT: retq
+; AVX-LABEL: shuffle_mem_v2f64_10:
+; AVX: # BB#0:
+; AVX-NEXT: vpermilpd {{.*#+}} xmm0 = mem[1,0]
+; AVX-NEXT: retq
%a = load <2 x double>, <2 x double>* %ptr
%shuffle = shufflevector <2 x double> %a, <2 x double> undef, <2 x i32> <i32 1, i32 0>
projects / oota-llvm.git / blobdiff