Home | History | Annotate | Download | only in AMDGPU
      1 ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG --check-prefix=FUNC %s
      2 ;RUN: llc < %s -march=amdgcn -mcpu=SI -verify-machineinstrs | FileCheck --check-prefix=SI --check-prefix=FUNC %s
      3 ;RUN: llc < %s -march=amdgcn -mcpu=tonga -verify-machineinstrs | FileCheck --check-prefix=SI --check-prefix=VI --check-prefix=FUNC %s
      4 
      5 ;FUNC-LABEL: {{^}}test_select_v2i32:
      6 
      7 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
      8 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
      9 
     10 ;SI: v_cndmask_b32_e64
     11 ;SI: v_cndmask_b32_e32
     12 
     13 define void @test_select_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in0, <2 x i32> addrspace(1)* %in1) {
     14 entry:
     15   %0 = load <2 x i32>, <2 x i32> addrspace(1)* %in0
     16   %1 = load <2 x i32>, <2 x i32> addrspace(1)* %in1
     17   %cmp = icmp ne <2 x i32> %0, %1
     18   %result = select <2 x i1> %cmp, <2 x i32> %0, <2 x i32> %1
     19   store <2 x i32> %result, <2 x i32> addrspace(1)* %out
     20   ret void
     21 }
     22 
     23 ;FUNC-LABEL: {{^}}test_select_v2f32:
     24 
     25 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     26 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     27 
     28 ;SI: v_cndmask_b32_e64
     29 ;SI: v_cndmask_b32_e32
     30 
     31 define void @test_select_v2f32(<2 x float> addrspace(1)* %out, <2 x float> addrspace(1)* %in0, <2 x float> addrspace(1)* %in1) {
     32 entry:
     33   %0 = load <2 x float>, <2 x float> addrspace(1)* %in0
     34   %1 = load <2 x float>, <2 x float> addrspace(1)* %in1
     35   %cmp = fcmp une <2 x float> %0, %1
     36   %result = select <2 x i1> %cmp, <2 x float> %0, <2 x float> %1
     37   store <2 x float> %result, <2 x float> addrspace(1)* %out
     38   ret void
     39 }
     40 
     41 ;FUNC-LABEL: {{^}}test_select_v4i32:
     42 
     43 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     44 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     45 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     46 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     47 
     48 ; FIXME: The shrinking does not happen on tonga
     49 
     50 ;SI: v_cndmask_b32
     51 ;SI: v_cndmask_b32
     52 ;SI: v_cndmask_b32
     53 ;SI: v_cndmask_b32
     54 
     55 define void @test_select_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in0, <4 x i32> addrspace(1)* %in1) {
     56 entry:
     57   %0 = load <4 x i32>, <4 x i32> addrspace(1)* %in0
     58   %1 = load <4 x i32>, <4 x i32> addrspace(1)* %in1
     59   %cmp = icmp ne <4 x i32> %0, %1
     60   %result = select <4 x i1> %cmp, <4 x i32> %0, <4 x i32> %1
     61   store <4 x i32> %result, <4 x i32> addrspace(1)* %out
     62   ret void
     63 }
     64 
     65 ;FUNC-LABEL: {{^}}test_select_v4f32:
     66 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     67 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     68 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     69 ;EG: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
     70 
     71 define void @test_select_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)* %in0, <4 x float> addrspace(1)* %in1) {
     72 entry:
     73   %0 = load <4 x float>, <4 x float> addrspace(1)* %in0
     74   %1 = load <4 x float>, <4 x float> addrspace(1)* %in1
     75   %cmp = fcmp une <4 x float> %0, %1
     76   %result = select <4 x i1> %cmp, <4 x float> %0, <4 x float> %1
     77   store <4 x float> %result, <4 x float> addrspace(1)* %out
     78   ret void
     79 }
     80