summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen/AMDGPU/fmul.ll
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/fmul.ll')
-rw-r--r--llvm/test/CodeGen/AMDGPU/fmul.ll49
1 files changed, 24 insertions, 25 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/fmul.ll b/llvm/test/CodeGen/AMDGPU/fmul.ll
index addc409c9eb..e851d900de4 100644
--- a/llvm/test/CodeGen/AMDGPU/fmul.ll
+++ b/llvm/test/CodeGen/AMDGPU/fmul.ll
@@ -1,12 +1,11 @@
-; RUN: llc -march=amdgcn -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
-; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=FUNC %s
; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
-
; FUNC-LABEL: {{^}}fmul_f32:
-; R600: MUL_IEEE {{\** *}}{{T[0-9]+\.[XYZW]}}, KC0[2].Z, KC0[2].W
+; GCN: v_mul_f32
-; SI: v_mul_f32
+; R600: MUL_IEEE {{\** *}}{{T[0-9]+\.[XYZW]}}, KC0[2].Z, KC0[2].W
define void @fmul_f32(float addrspace(1)* %out, float %a, float %b) {
entry:
%0 = fmul float %a, %b
@@ -19,11 +18,11 @@ declare float @llvm.R600.load.input(i32) readnone
declare void @llvm.AMDGPU.store.output(float, i32)
; FUNC-LABEL: {{^}}fmul_v2f32:
+; GCN: v_mul_f32
+; GCN: v_mul_f32
+
; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}
; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}
-
-; SI: v_mul_f32
-; SI: v_mul_f32
define void @fmul_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %a, <2 x float> %b) {
entry:
%0 = fmul <2 x float> %a, %b
@@ -32,15 +31,15 @@ entry:
}
; FUNC-LABEL: {{^}}fmul_v4f32:
+; GCN: v_mul_f32
+; GCN: v_mul_f32
+; GCN: v_mul_f32
+; GCN: v_mul_f32
+
; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
-
-; SI: v_mul_f32
-; SI: v_mul_f32
-; SI: v_mul_f32
-; SI: v_mul_f32
define void @fmul_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)* %in) {
%b_ptr = getelementptr <4 x float>, <4 x float> addrspace(1)* %in, i32 1
%a = load <4 x float>, <4 x float> addrspace(1) * %in
@@ -51,9 +50,9 @@ define void @fmul_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)
}
; FUNC-LABEL: {{^}}test_mul_2_k:
-; SI: v_mul_f32
-; SI-NOT: v_mul_f32
-; SI: s_endpgm
+; GCN: v_mul_f32
+; GCN-NOT: v_mul_f32
+; GCN: s_endpgm
define void @test_mul_2_k(float addrspace(1)* %out, float %x) #0 {
%y = fmul float %x, 2.0
%z = fmul float %y, 3.0
@@ -62,10 +61,10 @@ define void @test_mul_2_k(float addrspace(1)* %out, float %x) #0 {
}
; FUNC-LABEL: {{^}}test_mul_2_k_inv:
-; SI: v_mul_f32
-; SI-NOT: v_mul_f32
-; SI-NOT: v_mad_f32
-; SI: s_endpgm
+; GCN: v_mul_f32
+; GCN-NOT: v_mul_f32
+; GCN-NOT: v_mad_f32
+; GCN: s_endpgm
define void @test_mul_2_k_inv(float addrspace(1)* %out, float %x) #0 {
%y = fmul float %x, 3.0
%z = fmul float %y, 2.0
@@ -76,10 +75,10 @@ define void @test_mul_2_k_inv(float addrspace(1)* %out, float %x) #0 {
; There should be three multiplies here; %a should be used twice (once
; negated), not duplicated into mul x, 5.0 and mul x, -5.0.
; FUNC-LABEL: {{^}}test_mul_twouse:
-; SI: v_mul_f32
-; SI: v_mul_f32
-; SI: v_mul_f32
-; SI-NOT: v_mul_f32
+; GCN: v_mul_f32
+; GCN: v_mul_f32
+; GCN: v_mul_f32
+; GCN-NOT: v_mul_f32
define void @test_mul_twouse(float addrspace(1)* %out, float %x, float %y) #0 {
%a = fmul float %x, 5.0
%b = fsub float -0.0, %a
@@ -89,4 +88,4 @@ define void @test_mul_twouse(float addrspace(1)* %out, float %x, float %y) #0 {
ret void
}
-attributes #0 = { "less-precise-fpmad"="true" "no-infs-fp-math"="true" "no-nans-fp-math"="true" "unsafe-fp-math"="true" }
+attributes #0 = { nounwind }
OpenPOWER on IntegriCloud