RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-24 19:52:54 +01:00
Code Issues Projects Releases Wiki Activity

[X86][SSE] Test showing missing EXTRACT_SUBVECTOR/CONCAT_VECTORS demandedelts support in ComputeNumSignBits

Browse Source 
llvm-svn: 302994
This commit is contained in:
Simon Pilgrim 2017-05-13 21:50:18 +00:00
parent dd6308c45b
commit e3e00995f1
1 changed files with 55 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
test/CodeGen/X86/known-signbits-vector.ll
View File

@ -169,3 +169,58 @@ define <4 x double> @signbits_sext_shuffle_sitofp(<4 x i32> %a0, <4 x i64> %a1)
%3 = sitofp <4 x i64> %2 to <4 x double>
ret <4 x double> %3
}
define <2 x double> @signbits_ashr_concat_ashr_extract_sitofp(<2 x i64> %a0, <4 x i64> %a1) nounwind {
; X32-LABEL: signbits_ashr_concat_ashr_extract_sitofp:
; X32: # BB#0:
; X32-NEXT: pushl %ebp
; X32-NEXT: movl %esp, %ebp
; X32-NEXT: andl $-8, %esp
; X32-NEXT: subl $32, %esp
; X32-NEXT: vpsrad $16, %xmm0, %xmm1
; X32-NEXT: vpsrlq $16, %xmm0, %xmm0
; X32-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5],xmm1[6,7]
; X32-NEXT: vmovaps {{.*#+}} ymm1 = [16,0,16,0,16,0,16,0]
; X32-NEXT: vextractf128 $1, %ymm1, %xmm1
; X32-NEXT: vmovdqa {{.*#+}} xmm2 = [0,2147483648,0,2147483648]
; X32-NEXT: vpsrlq %xmm1, %xmm2, %xmm2
; X32-NEXT: vpsrlq %xmm1, %xmm0, %xmm0
; X32-NEXT: vpxor %xmm2, %xmm0, %xmm0
; X32-NEXT: vpsubq %xmm2, %xmm0, %xmm0
; X32-NEXT: vmovq {{.*#+}} xmm1 = xmm0[0],zero
; X32-NEXT: vmovq %xmm1, {{[0-9]+}}(%esp)
; X32-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
; X32-NEXT: vmovq %xmm0, {{[0-9]+}}(%esp)
; X32-NEXT: fildll {{[0-9]+}}(%esp)
; X32-NEXT: fstpl {{[0-9]+}}(%esp)
; X32-NEXT: fildll {{[0-9]+}}(%esp)
; X32-NEXT: fstpl (%esp)
; X32-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X32-NEXT: vmovhpd {{.*#+}} xmm0 = xmm0[0],mem[0]
; X32-NEXT: movl %ebp, %esp
; X32-NEXT: popl %ebp
; X32-NEXT: vzeroupper
; X32-NEXT: retl
;
; X64-LABEL: signbits_ashr_concat_ashr_extract_sitofp:
; X64: # BB#0:
; X64-NEXT: vpsrad $16, %xmm0, %xmm1
; X64-NEXT: vpsrlq $16, %xmm0, %xmm0
; X64-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5],xmm1[6,7]
; X64-NEXT: vpsrad $16, %xmm0, %xmm1
; X64-NEXT: vpsrlq $16, %xmm0, %xmm0
; X64-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5],xmm1[6,7]
; X64-NEXT: vpextrq $1, %xmm0, %rax
; X64-NEXT: vcvtsi2sdq %rax, %xmm2, %xmm1
; X64-NEXT: vmovq %xmm0, %rax
; X64-NEXT: vcvtsi2sdq %rax, %xmm2, %xmm0
; X64-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; X64-NEXT: retq
%1 = ashr <2 x i64> %a0, <i64 16, i64 16>
%2 = shufflevector <2 x i64> %1, <2 x i64> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
%3 = shufflevector <4 x i64> %a1, <4 x i64> %2, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
%4 = ashr <4 x i64> %3, <i64 16, i64 16, i64 16, i64 16>
%5 = shufflevector <4 x i64> %4, <4 x i64> undef, <2 x i32> <i32 2, i32 3>
%6 = sitofp <2 x i64> %5 to <2 x double>
ret <2 x double> %6
}