mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-16 11:30:51 +00:00
7c9c6ed761
Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230794 91177308-0d34-0410-b5e6-96231b3b80d8
303 lines
12 KiB
LLVM
303 lines
12 KiB
LLVM
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=knl | FileCheck %s
|
|
|
|
define <16 x i32> @_inreg16xi32(i32 %a) {
|
|
; CHECK-LABEL: _inreg16xi32:
|
|
; CHECK: ## BB#0:
|
|
; CHECK-NEXT: vpbroadcastd %edi, %zmm0
|
|
; CHECK-NEXT: retq
|
|
%b = insertelement <16 x i32> undef, i32 %a, i32 0
|
|
%c = shufflevector <16 x i32> %b, <16 x i32> undef, <16 x i32> zeroinitializer
|
|
ret <16 x i32> %c
|
|
}
|
|
|
|
define <8 x i64> @_inreg8xi64(i64 %a) {
|
|
; CHECK-LABEL: _inreg8xi64:
|
|
; CHECK: ## BB#0:
|
|
; CHECK-NEXT: vpbroadcastq %rdi, %zmm0
|
|
; CHECK-NEXT: retq
|
|
%b = insertelement <8 x i64> undef, i64 %a, i32 0
|
|
%c = shufflevector <8 x i64> %b, <8 x i64> undef, <8 x i32> zeroinitializer
|
|
ret <8 x i64> %c
|
|
}
|
|
|
|
;CHECK-LABEL: _ss16xfloat_v4
|
|
;CHECK: vbroadcastss %xmm0, %zmm0
|
|
;CHECK: ret
|
|
define <16 x float> @_ss16xfloat_v4(<4 x float> %a) {
|
|
%b = shufflevector <4 x float> %a, <4 x float> undef, <16 x i32> zeroinitializer
|
|
ret <16 x float> %b
|
|
}
|
|
|
|
define <16 x float> @_inreg16xfloat(float %a) {
|
|
; CHECK-LABEL: _inreg16xfloat:
|
|
; CHECK: ## BB#0:
|
|
; CHECK-NEXT: vbroadcastss %xmm0, %zmm0
|
|
; CHECK-NEXT: retq
|
|
%b = insertelement <16 x float> undef, float %a, i32 0
|
|
%c = shufflevector <16 x float> %b, <16 x float> undef, <16 x i32> zeroinitializer
|
|
ret <16 x float> %c
|
|
}
|
|
|
|
;CHECK-LABEL: _ss16xfloat_mask:
|
|
;CHECK: vbroadcastss %xmm0, %zmm1 {%k1}
|
|
;CHECK: ret
|
|
define <16 x float> @_ss16xfloat_mask(float %a, <16 x float> %i, <16 x i32> %mask1) {
|
|
%mask = icmp ne <16 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <16 x float> undef, float %a, i32 0
|
|
%c = shufflevector <16 x float> %b, <16 x float> undef, <16 x i32> zeroinitializer
|
|
%r = select <16 x i1> %mask, <16 x float> %c, <16 x float> %i
|
|
ret <16 x float> %r
|
|
}
|
|
|
|
;CHECK-LABEL: _ss16xfloat_maskz:
|
|
;CHECK: vbroadcastss %xmm0, %zmm0 {%k1} {z}
|
|
;CHECK: ret
|
|
define <16 x float> @_ss16xfloat_maskz(float %a, <16 x i32> %mask1) {
|
|
%mask = icmp ne <16 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <16 x float> undef, float %a, i32 0
|
|
%c = shufflevector <16 x float> %b, <16 x float> undef, <16 x i32> zeroinitializer
|
|
%r = select <16 x i1> %mask, <16 x float> %c, <16 x float> zeroinitializer
|
|
ret <16 x float> %r
|
|
}
|
|
|
|
;CHECK-LABEL: _ss16xfloat_load:
|
|
;CHECK: vbroadcastss (%{{.*}}, %zmm
|
|
;CHECK: ret
|
|
define <16 x float> @_ss16xfloat_load(float* %a.ptr) {
|
|
%a = load float, float* %a.ptr
|
|
%b = insertelement <16 x float> undef, float %a, i32 0
|
|
%c = shufflevector <16 x float> %b, <16 x float> undef, <16 x i32> zeroinitializer
|
|
ret <16 x float> %c
|
|
}
|
|
|
|
;CHECK-LABEL: _ss16xfloat_mask_load:
|
|
;CHECK: vbroadcastss (%rdi), %zmm0 {%k1}
|
|
;CHECK: ret
|
|
define <16 x float> @_ss16xfloat_mask_load(float* %a.ptr, <16 x float> %i, <16 x i32> %mask1) {
|
|
%a = load float, float* %a.ptr
|
|
%mask = icmp ne <16 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <16 x float> undef, float %a, i32 0
|
|
%c = shufflevector <16 x float> %b, <16 x float> undef, <16 x i32> zeroinitializer
|
|
%r = select <16 x i1> %mask, <16 x float> %c, <16 x float> %i
|
|
ret <16 x float> %r
|
|
}
|
|
|
|
;CHECK-LABEL: _ss16xfloat_maskz_load:
|
|
;CHECK: vbroadcastss (%rdi), %zmm0 {%k1} {z}
|
|
;CHECK: ret
|
|
define <16 x float> @_ss16xfloat_maskz_load(float* %a.ptr, <16 x i32> %mask1) {
|
|
%a = load float, float* %a.ptr
|
|
%mask = icmp ne <16 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <16 x float> undef, float %a, i32 0
|
|
%c = shufflevector <16 x float> %b, <16 x float> undef, <16 x i32> zeroinitializer
|
|
%r = select <16 x i1> %mask, <16 x float> %c, <16 x float> zeroinitializer
|
|
ret <16 x float> %r
|
|
}
|
|
|
|
define <8 x double> @_inreg8xdouble(double %a) {
|
|
; CHECK-LABEL: _inreg8xdouble:
|
|
; CHECK: ## BB#0:
|
|
; CHECK-NEXT: vbroadcastsd %xmm0, %zmm0
|
|
; CHECK-NEXT: retq
|
|
%b = insertelement <8 x double> undef, double %a, i32 0
|
|
%c = shufflevector <8 x double> %b, <8 x double> undef, <8 x i32> zeroinitializer
|
|
ret <8 x double> %c
|
|
}
|
|
|
|
;CHECK-LABEL: _sd8xdouble_mask:
|
|
;CHECK: vbroadcastsd %xmm0, %zmm1 {%k1}
|
|
;CHECK: ret
|
|
define <8 x double> @_sd8xdouble_mask(double %a, <8 x double> %i, <8 x i32> %mask1) {
|
|
%mask = icmp ne <8 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <8 x double> undef, double %a, i32 0
|
|
%c = shufflevector <8 x double> %b, <8 x double> undef, <8 x i32> zeroinitializer
|
|
%r = select <8 x i1> %mask, <8 x double> %c, <8 x double> %i
|
|
ret <8 x double> %r
|
|
}
|
|
|
|
;CHECK-LABEL: _sd8xdouble_maskz:
|
|
;CHECK: vbroadcastsd %xmm0, %zmm0 {%k1} {z}
|
|
;CHECK: ret
|
|
define <8 x double> @_sd8xdouble_maskz(double %a, <8 x i32> %mask1) {
|
|
%mask = icmp ne <8 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <8 x double> undef, double %a, i32 0
|
|
%c = shufflevector <8 x double> %b, <8 x double> undef, <8 x i32> zeroinitializer
|
|
%r = select <8 x i1> %mask, <8 x double> %c, <8 x double> zeroinitializer
|
|
ret <8 x double> %r
|
|
}
|
|
|
|
;CHECK-LABEL: _sd8xdouble_load:
|
|
;CHECK: vbroadcastsd (%rdi), %zmm
|
|
;CHECK: ret
|
|
define <8 x double> @_sd8xdouble_load(double* %a.ptr) {
|
|
%a = load double, double* %a.ptr
|
|
%b = insertelement <8 x double> undef, double %a, i32 0
|
|
%c = shufflevector <8 x double> %b, <8 x double> undef, <8 x i32> zeroinitializer
|
|
ret <8 x double> %c
|
|
}
|
|
|
|
;CHECK-LABEL: _sd8xdouble_mask_load:
|
|
;CHECK: vbroadcastsd (%rdi), %zmm0 {%k1}
|
|
;CHECK: ret
|
|
define <8 x double> @_sd8xdouble_mask_load(double* %a.ptr, <8 x double> %i, <8 x i32> %mask1) {
|
|
%a = load double, double* %a.ptr
|
|
%mask = icmp ne <8 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <8 x double> undef, double %a, i32 0
|
|
%c = shufflevector <8 x double> %b, <8 x double> undef, <8 x i32> zeroinitializer
|
|
%r = select <8 x i1> %mask, <8 x double> %c, <8 x double> %i
|
|
ret <8 x double> %r
|
|
}
|
|
|
|
define <8 x double> @_sd8xdouble_maskz_load(double* %a.ptr, <8 x i32> %mask1) {
|
|
; CHECK-LABEL: _sd8xdouble_maskz_load:
|
|
; CHECK: vbroadcastsd (%rdi), %zmm0 {%k1} {z}
|
|
; CHECK: ret
|
|
%a = load double, double* %a.ptr
|
|
%mask = icmp ne <8 x i32> %mask1, zeroinitializer
|
|
%b = insertelement <8 x double> undef, double %a, i32 0
|
|
%c = shufflevector <8 x double> %b, <8 x double> undef, <8 x i32> zeroinitializer
|
|
%r = select <8 x i1> %mask, <8 x double> %c, <8 x double> zeroinitializer
|
|
ret <8 x double> %r
|
|
}
|
|
|
|
define <16 x i32> @_xmm16xi32(<16 x i32> %a) {
|
|
; CHECK-LABEL: _xmm16xi32:
|
|
; CHECK: ## BB#0:
|
|
; CHECK-NEXT: vpbroadcastd %xmm0, %zmm0
|
|
; CHECK-NEXT: retq
|
|
%b = shufflevector <16 x i32> %a, <16 x i32> undef, <16 x i32> zeroinitializer
|
|
ret <16 x i32> %b
|
|
}
|
|
|
|
define <16 x float> @_xmm16xfloat(<16 x float> %a) {
|
|
; CHECK-LABEL: _xmm16xfloat:
|
|
; CHECK: ## BB#0:
|
|
; CHECK-NEXT: vbroadcastss %xmm0, %zmm0
|
|
; CHECK-NEXT: retq
|
|
%b = shufflevector <16 x float> %a, <16 x float> undef, <16 x i32> zeroinitializer
|
|
ret <16 x float> %b
|
|
}
|
|
|
|
define <16 x i32> @test_vbroadcast() {
|
|
; CHECK-LABEL: test_vbroadcast:
|
|
; CHECK: ## BB#0: ## %entry
|
|
; CHECK-NEXT: vpxord %zmm0, %zmm0, %zmm0
|
|
; CHECK-NEXT: vcmpunordps %zmm0, %zmm0, %k1
|
|
; CHECK-NEXT: vpbroadcastd {{.*}}(%rip), %zmm0 {%k1} {z}
|
|
; CHECK-NEXT: knotw %k1, %k1
|
|
; CHECK-NEXT: vmovdqu32 %zmm0, %zmm0 {%k1} {z}
|
|
; CHECK-NEXT: retq
|
|
entry:
|
|
%0 = sext <16 x i1> zeroinitializer to <16 x i32>
|
|
%1 = fcmp uno <16 x float> undef, zeroinitializer
|
|
%2 = sext <16 x i1> %1 to <16 x i32>
|
|
%3 = select <16 x i1> %1, <16 x i32> %0, <16 x i32> %2
|
|
ret <16 x i32> %3
|
|
}
|
|
|
|
; We implement the set1 intrinsics with vector initializers. Verify that the
|
|
; IR generated will produce broadcasts at the end.
|
|
define <8 x double> @test_set1_pd(double %d) #2 {
|
|
; CHECK-LABEL: test_set1_pd:
|
|
; CHECK: ## BB#0: ## %entry
|
|
; CHECK-NEXT: vbroadcastsd %xmm0, %zmm0
|
|
; CHECK-NEXT: retq
|
|
entry:
|
|
%vecinit.i = insertelement <8 x double> undef, double %d, i32 0
|
|
%vecinit1.i = insertelement <8 x double> %vecinit.i, double %d, i32 1
|
|
%vecinit2.i = insertelement <8 x double> %vecinit1.i, double %d, i32 2
|
|
%vecinit3.i = insertelement <8 x double> %vecinit2.i, double %d, i32 3
|
|
%vecinit4.i = insertelement <8 x double> %vecinit3.i, double %d, i32 4
|
|
%vecinit5.i = insertelement <8 x double> %vecinit4.i, double %d, i32 5
|
|
%vecinit6.i = insertelement <8 x double> %vecinit5.i, double %d, i32 6
|
|
%vecinit7.i = insertelement <8 x double> %vecinit6.i, double %d, i32 7
|
|
ret <8 x double> %vecinit7.i
|
|
}
|
|
|
|
define <8 x i64> @test_set1_epi64(i64 %d) #2 {
|
|
; CHECK-LABEL: test_set1_epi64:
|
|
; CHECK: ## BB#0: ## %entry
|
|
; CHECK-NEXT: vpbroadcastq %rdi, %zmm0
|
|
; CHECK-NEXT: retq
|
|
entry:
|
|
%vecinit.i = insertelement <8 x i64> undef, i64 %d, i32 0
|
|
%vecinit1.i = insertelement <8 x i64> %vecinit.i, i64 %d, i32 1
|
|
%vecinit2.i = insertelement <8 x i64> %vecinit1.i, i64 %d, i32 2
|
|
%vecinit3.i = insertelement <8 x i64> %vecinit2.i, i64 %d, i32 3
|
|
%vecinit4.i = insertelement <8 x i64> %vecinit3.i, i64 %d, i32 4
|
|
%vecinit5.i = insertelement <8 x i64> %vecinit4.i, i64 %d, i32 5
|
|
%vecinit6.i = insertelement <8 x i64> %vecinit5.i, i64 %d, i32 6
|
|
%vecinit7.i = insertelement <8 x i64> %vecinit6.i, i64 %d, i32 7
|
|
ret <8 x i64> %vecinit7.i
|
|
}
|
|
|
|
define <16 x float> @test_set1_ps(float %f) #2 {
|
|
; CHECK-LABEL: test_set1_ps:
|
|
; CHECK: ## BB#0: ## %entry
|
|
; CHECK-NEXT: vbroadcastss %xmm0, %zmm0
|
|
; CHECK-NEXT: retq
|
|
entry:
|
|
%vecinit.i = insertelement <16 x float> undef, float %f, i32 0
|
|
%vecinit1.i = insertelement <16 x float> %vecinit.i, float %f, i32 1
|
|
%vecinit2.i = insertelement <16 x float> %vecinit1.i, float %f, i32 2
|
|
%vecinit3.i = insertelement <16 x float> %vecinit2.i, float %f, i32 3
|
|
%vecinit4.i = insertelement <16 x float> %vecinit3.i, float %f, i32 4
|
|
%vecinit5.i = insertelement <16 x float> %vecinit4.i, float %f, i32 5
|
|
%vecinit6.i = insertelement <16 x float> %vecinit5.i, float %f, i32 6
|
|
%vecinit7.i = insertelement <16 x float> %vecinit6.i, float %f, i32 7
|
|
%vecinit8.i = insertelement <16 x float> %vecinit7.i, float %f, i32 8
|
|
%vecinit9.i = insertelement <16 x float> %vecinit8.i, float %f, i32 9
|
|
%vecinit10.i = insertelement <16 x float> %vecinit9.i, float %f, i32 10
|
|
%vecinit11.i = insertelement <16 x float> %vecinit10.i, float %f, i32 11
|
|
%vecinit12.i = insertelement <16 x float> %vecinit11.i, float %f, i32 12
|
|
%vecinit13.i = insertelement <16 x float> %vecinit12.i, float %f, i32 13
|
|
%vecinit14.i = insertelement <16 x float> %vecinit13.i, float %f, i32 14
|
|
%vecinit15.i = insertelement <16 x float> %vecinit14.i, float %f, i32 15
|
|
ret <16 x float> %vecinit15.i
|
|
}
|
|
|
|
define <16 x i32> @test_set1_epi32(i32 %f) #2 {
|
|
; CHECK-LABEL: test_set1_epi32:
|
|
; CHECK: ## BB#0: ## %entry
|
|
; CHECK-NEXT: vpbroadcastd %edi, %zmm0
|
|
; CHECK-NEXT: retq
|
|
entry:
|
|
%vecinit.i = insertelement <16 x i32> undef, i32 %f, i32 0
|
|
%vecinit1.i = insertelement <16 x i32> %vecinit.i, i32 %f, i32 1
|
|
%vecinit2.i = insertelement <16 x i32> %vecinit1.i, i32 %f, i32 2
|
|
%vecinit3.i = insertelement <16 x i32> %vecinit2.i, i32 %f, i32 3
|
|
%vecinit4.i = insertelement <16 x i32> %vecinit3.i, i32 %f, i32 4
|
|
%vecinit5.i = insertelement <16 x i32> %vecinit4.i, i32 %f, i32 5
|
|
%vecinit6.i = insertelement <16 x i32> %vecinit5.i, i32 %f, i32 6
|
|
%vecinit7.i = insertelement <16 x i32> %vecinit6.i, i32 %f, i32 7
|
|
%vecinit8.i = insertelement <16 x i32> %vecinit7.i, i32 %f, i32 8
|
|
%vecinit9.i = insertelement <16 x i32> %vecinit8.i, i32 %f, i32 9
|
|
%vecinit10.i = insertelement <16 x i32> %vecinit9.i, i32 %f, i32 10
|
|
%vecinit11.i = insertelement <16 x i32> %vecinit10.i, i32 %f, i32 11
|
|
%vecinit12.i = insertelement <16 x i32> %vecinit11.i, i32 %f, i32 12
|
|
%vecinit13.i = insertelement <16 x i32> %vecinit12.i, i32 %f, i32 13
|
|
%vecinit14.i = insertelement <16 x i32> %vecinit13.i, i32 %f, i32 14
|
|
%vecinit15.i = insertelement <16 x i32> %vecinit14.i, i32 %f, i32 15
|
|
ret <16 x i32> %vecinit15.i
|
|
}
|
|
|
|
; We implement the scalar broadcast intrinsics with vector initializers.
|
|
; Verify that the IR generated will produce the broadcast at the end.
|
|
define <8 x double> @test_mm512_broadcastsd_pd(<2 x double> %a) {
|
|
; CHECK-LABEL: test_mm512_broadcastsd_pd:
|
|
; CHECK: ## BB#0: ## %entry
|
|
; CHECK-NEXT: vbroadcastsd %xmm0, %zmm0
|
|
; CHECK-NEXT: retq
|
|
entry:
|
|
%0 = extractelement <2 x double> %a, i32 0
|
|
%vecinit.i = insertelement <8 x double> undef, double %0, i32 0
|
|
%vecinit1.i = insertelement <8 x double> %vecinit.i, double %0, i32 1
|
|
%vecinit2.i = insertelement <8 x double> %vecinit1.i, double %0, i32 2
|
|
%vecinit3.i = insertelement <8 x double> %vecinit2.i, double %0, i32 3
|
|
%vecinit4.i = insertelement <8 x double> %vecinit3.i, double %0, i32 4
|
|
%vecinit5.i = insertelement <8 x double> %vecinit4.i, double %0, i32 5
|
|
%vecinit6.i = insertelement <8 x double> %vecinit5.i, double %0, i32 6
|
|
%vecinit7.i = insertelement <8 x double> %vecinit6.i, double %0, i32 7
|
|
ret <8 x double> %vecinit7.i
|
|
}
|