RPCS3/llvm
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm.git synced 2025-02-12 13:48:45 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[X86] Add tests for missed opportunities to fold a 128-bit vector load into vfpclassss and vpfpclasssd.

Browse Source 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@318018 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Craig Topper 2017-11-13 06:46:46 +00:00
parent f5c667f88c
commit 1d8fbacd7d
1 changed files with 26 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
test/CodeGen/X86/avx512dq-intrinsics.ll
View File

@ -396,6 +396,19 @@ define i8 @test_int_x86_avx512_mask_fpclass_sd(<2 x double> %x0, i8 %x1) {
ret i8 %res2
}
define i8 @test_int_x86_avx512_mask_fpclass_sd_load(<2 x double>* %x0ptr) {
; CHECK-LABEL: test_int_x86_avx512_mask_fpclass_sd_load:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovapd (%rdi), %xmm0
; CHECK-NEXT: vfpclasssd $4, %xmm0, %k0
; CHECK-NEXT: kmovw %k0, %eax
; CHECK-NEXT: ## kill: %AL<def> %AL<kill> %EAX<kill>
; CHECK-NEXT: retq
%x0 = load <2 x double>, <2 x double>* %x0ptr
%res = call i8 @llvm.x86.avx512.mask.fpclass.sd(<2 x double> %x0, i32 4, i8 -1)
ret i8 %res
}
declare i8 @llvm.x86.avx512.mask.fpclass.ss(<4 x float>, i32, i8)
define i8 @test_int_x86_avx512_mask_fpclass_ss(<4 x float> %x0, i8 %x1) {
@ -415,6 +428,19 @@ define i8 @test_int_x86_avx512_mask_fpclass_ss(<4 x float> %x0, i8 %x1) {
ret i8 %res2
}
define i8 @test_int_x86_avx512_mask_fpclass_ss_load(<4 x float>* %x0ptr, i8 %x1) {
; CHECK-LABEL: test_int_x86_avx512_mask_fpclass_ss_load:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovaps (%rdi), %xmm0
; CHECK-NEXT: vfpclassss $4, %xmm0, %k0
; CHECK-NEXT: kmovw %k0, %eax
; CHECK-NEXT: ## kill: %AL<def> %AL<kill> %EAX<kill>
; CHECK-NEXT: retq
%x0 = load <4 x float>, <4 x float>* %x0ptr
%res = call i8 @llvm.x86.avx512.mask.fpclass.ss(<4 x float> %x0, i32 4, i8 -1)
ret i8 %res
}
declare i16 @llvm.x86.avx512.cvtd2mask.512(<16 x i32>)
define i16@test_int_x86_avx512_cvtd2mask_512(<16 x i32> %x0) {