diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/vector-idiv-sdiv-128.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/vector-idiv-sdiv-128.ll | 96 |
1 files changed, 44 insertions, 52 deletions
diff --git a/llvm/test/CodeGen/X86/vector-idiv-sdiv-128.ll b/llvm/test/CodeGen/X86/vector-idiv-sdiv-128.ll index 3f251dd8d62..159bb762a87 100644 --- a/llvm/test/CodeGen/X86/vector-idiv-sdiv-128.ll +++ b/llvm/test/CodeGen/X86/vector-idiv-sdiv-128.ll @@ -78,22 +78,19 @@ define <2 x i64> @test_div7_2i64(<2 x i64> %a) nounwind { define <4 x i32> @test_div7_4i32(<4 x i32> %a) nounwind { ; SSE2-LABEL: test_div7_4i32: ; SSE2: # %bb.0: -; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [2454267027,2454267027,2454267027,2454267027] -; SSE2-NEXT: movdqa %xmm0, %xmm2 -; SSE2-NEXT: psrad $31, %xmm2 -; SSE2-NEXT: pand %xmm1, %xmm2 -; SSE2-NEXT: movdqa %xmm0, %xmm3 -; SSE2-NEXT: pmuludq %xmm1, %xmm3 -; SSE2-NEXT: pshufd {{.*#+}} xmm4 = xmm1[1,1,3,3] -; SSE2-NEXT: psrad $31, %xmm1 -; SSE2-NEXT: pand %xmm0, %xmm1 -; SSE2-NEXT: paddd %xmm1, %xmm2 -; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm3[1,3,2,3] +; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [2454267027,2454267027,2454267027,2454267027] +; SSE2-NEXT: movdqa %xmm0, %xmm1 +; SSE2-NEXT: pmuludq %xmm2, %xmm1 +; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[1,3,2,3] ; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm0[1,1,3,3] -; SSE2-NEXT: pmuludq %xmm4, %xmm3 +; SSE2-NEXT: pmuludq %xmm2, %xmm3 ; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm3[1,3,2,3] ; SSE2-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1] -; SSE2-NEXT: psubd %xmm2, %xmm1 +; SSE2-NEXT: movdqa %xmm0, %xmm3 +; SSE2-NEXT: psrad $31, %xmm3 +; SSE2-NEXT: pand %xmm2, %xmm3 +; SSE2-NEXT: paddd %xmm0, %xmm3 +; SSE2-NEXT: psubd %xmm3, %xmm1 ; SSE2-NEXT: paddd %xmm0, %xmm1 ; SSE2-NEXT: movdqa %xmm1, %xmm0 ; SSE2-NEXT: psrld $31, %xmm0 @@ -134,13 +131,12 @@ define <4 x i32> @test_div7_4i32(<4 x i32> %a) nounwind { ; ; AVX2-LABEL: test_div7_4i32: ; AVX2: # %bb.0: -; AVX2-NEXT: vpbroadcastd {{.*#+}} xmm1 = [2454267027,2454267027,2454267027,2454267027] -; AVX2-NEXT: vpshufd {{.*#+}} xmm2 = xmm1[1,1,3,3] -; AVX2-NEXT: vpshufd {{.*#+}} xmm3 = xmm0[1,1,3,3] -; AVX2-NEXT: vpmuldq %xmm2, %xmm3, %xmm2 -; AVX2-NEXT: vpmuldq %xmm1, %xmm0, %xmm1 -; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[1,1,3,3] -; AVX2-NEXT: vpblendd {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3] +; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,3,3] +; AVX2-NEXT: vpbroadcastd {{.*#+}} xmm2 = [2454267027,2454267027,2454267027,2454267027] +; AVX2-NEXT: vpmuldq %xmm2, %xmm1, %xmm1 +; AVX2-NEXT: vpmuldq %xmm2, %xmm0, %xmm2 +; AVX2-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[1,1,3,3] +; AVX2-NEXT: vpblendd {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2],xmm1[3] ; AVX2-NEXT: vpaddd %xmm0, %xmm1, %xmm0 ; AVX2-NEXT: vpsrld $31, %xmm0, %xmm1 ; AVX2-NEXT: vpsrad $2, %xmm0, %xmm0 @@ -384,33 +380,30 @@ define <4 x i32> @test_rem7_4i32(<4 x i32> %a) nounwind { ; SSE2: # %bb.0: ; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [2454267027,2454267027,2454267027,2454267027] ; SSE2-NEXT: movdqa %xmm0, %xmm2 -; SSE2-NEXT: psrad $31, %xmm2 -; SSE2-NEXT: pand %xmm1, %xmm2 -; SSE2-NEXT: movdqa %xmm0, %xmm3 -; SSE2-NEXT: pmuludq %xmm1, %xmm3 -; SSE2-NEXT: pshufd {{.*#+}} xmm4 = xmm1[1,1,3,3] -; SSE2-NEXT: psrad $31, %xmm1 -; SSE2-NEXT: pand %xmm0, %xmm1 -; SSE2-NEXT: paddd %xmm1, %xmm2 -; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm3[1,3,2,3] +; SSE2-NEXT: pmuludq %xmm1, %xmm2 +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm2[1,3,2,3] ; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm0[1,1,3,3] -; SSE2-NEXT: pmuludq %xmm4, %xmm3 +; SSE2-NEXT: pmuludq %xmm1, %xmm3 ; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm3[1,3,2,3] -; SSE2-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1] -; SSE2-NEXT: psubd %xmm2, %xmm1 -; SSE2-NEXT: paddd %xmm0, %xmm1 -; SSE2-NEXT: movdqa %xmm1, %xmm2 -; SSE2-NEXT: psrld $31, %xmm2 -; SSE2-NEXT: psrad $2, %xmm1 -; SSE2-NEXT: paddd %xmm2, %xmm1 -; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [7,7,7,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm1[1,1,3,3] -; SSE2-NEXT: pmuludq %xmm2, %xmm1 -; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3] -; SSE2-NEXT: pmuludq %xmm2, %xmm3 -; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm3[0,2,2,3] -; SSE2-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1] -; SSE2-NEXT: psubd %xmm1, %xmm0 +; SSE2-NEXT: punpckldq {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[1],xmm3[1] +; SSE2-NEXT: movdqa %xmm0, %xmm3 +; SSE2-NEXT: psrad $31, %xmm3 +; SSE2-NEXT: pand %xmm1, %xmm3 +; SSE2-NEXT: paddd %xmm0, %xmm3 +; SSE2-NEXT: psubd %xmm3, %xmm2 +; SSE2-NEXT: paddd %xmm0, %xmm2 +; SSE2-NEXT: movdqa %xmm2, %xmm1 +; SSE2-NEXT: psrld $31, %xmm1 +; SSE2-NEXT: psrad $2, %xmm2 +; SSE2-NEXT: paddd %xmm1, %xmm2 +; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [7,7,7,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm2[1,1,3,3] +; SSE2-NEXT: pmuludq %xmm1, %xmm2 +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm2[0,2,2,3] +; SSE2-NEXT: pmuludq %xmm1, %xmm3 +; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm3[0,2,2,3] +; SSE2-NEXT: punpckldq {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1] +; SSE2-NEXT: psubd %xmm2, %xmm0 ; SSE2-NEXT: retq ; ; SSE41-LABEL: test_rem7_4i32: @@ -448,13 +441,12 @@ define <4 x i32> @test_rem7_4i32(<4 x i32> %a) nounwind { ; ; AVX2-LABEL: test_rem7_4i32: ; AVX2: # %bb.0: -; AVX2-NEXT: vpbroadcastd {{.*#+}} xmm1 = [2454267027,2454267027,2454267027,2454267027] -; AVX2-NEXT: vpshufd {{.*#+}} xmm2 = xmm1[1,1,3,3] -; AVX2-NEXT: vpshufd {{.*#+}} xmm3 = xmm0[1,1,3,3] -; AVX2-NEXT: vpmuldq %xmm2, %xmm3, %xmm2 -; AVX2-NEXT: vpmuldq %xmm1, %xmm0, %xmm1 -; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[1,1,3,3] -; AVX2-NEXT: vpblendd {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3] +; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,3,3] +; AVX2-NEXT: vpbroadcastd {{.*#+}} xmm2 = [2454267027,2454267027,2454267027,2454267027] +; AVX2-NEXT: vpmuldq %xmm2, %xmm1, %xmm1 +; AVX2-NEXT: vpmuldq %xmm2, %xmm0, %xmm2 +; AVX2-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[1,1,3,3] +; AVX2-NEXT: vpblendd {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2],xmm1[3] ; AVX2-NEXT: vpaddd %xmm0, %xmm1, %xmm1 ; AVX2-NEXT: vpsrld $31, %xmm1, %xmm2 ; AVX2-NEXT: vpsrad $2, %xmm1, %xmm1 |
