Index: lib/Target/AMDGPU/SIInstrFormats.td =================================================================== --- lib/Target/AMDGPU/SIInstrFormats.td +++ lib/Target/AMDGPU/SIInstrFormats.td @@ -229,7 +229,19 @@ class VINTRPCommon pattern> : InstSI { - let mayLoad = 1; + + // VINTRP instructions read parameter values from LDS, but these parameter + // values are stored outside of the LDS memory that is allocated to the + // shader for general purpose use. + // + // While it may be possible for ds_read/ds_write instructions to access + // the parameter values in LDS, this would essentially be an out-of-bounds + // memory access which we consider to be undefined behavior. + // + // So even though these instructions read memory, this memory is outside the + // addressable memory space for the shader, and we consider these instructions + // to be readnone. + let mayLoad = 0; let mayStore = 0; let hasSideEffects = 0; } Index: test/CodeGen/AMDGPU/llvm.amdgcn.interp.ll =================================================================== --- test/CodeGen/AMDGPU/llvm.amdgcn.interp.ll +++ test/CodeGen/AMDGPU/llvm.amdgcn.interp.ll @@ -1,5 +1,5 @@ ;RUN: llc < %s -march=amdgcn -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=GCN %s -;RUN: llc < %s -march=amdgcn -mcpu=tonga -verify-machineinstrs | FileCheck --check-prefix=GCN %s +;RUN: llc < %s -march=amdgcn -mcpu=tonga -verify-machineinstrs | FileCheck --check-prefixes=GCN,VI %s ;GCN-LABEL: {{^}}v_interp: ;GCN-NOT: s_wqm @@ -20,6 +20,20 @@ ret void } +; SI won't merge ds memory operations, because of the signed offset bug, so +; we only have check lines for VI. +; VI-LABEL: v_interp_readnone: +; VI: v_mov_b32_e32 [[ZERO:v[0-9]+]], 0 +; VI: ds_write2_b32 v{{[0-9]+}}, [[ZERO]], [[ZERO]] offset1:4 +define amdgpu_ps void @v_interp_readnone(float addrspace(3)* %lds) { + store float 0.0, float addrspace(3)* %lds + %tmp1 = call float @llvm.amdgcn.interp.mov(i32 2, i32 0, i32 0, i32 0) + %tmp2 = getelementptr float, float addrspace(3)* %lds, i32 4 + store float 0.0, float addrspace(3)* %tmp2 + call void @llvm.SI.export(i32 15, i32 1, i32 1, i32 0, i32 1, float %tmp1, float %tmp1, float %tmp1, float %tmp1) + ret void +} + ; Function Attrs: nounwind readnone declare float @llvm.amdgcn.interp.p1(float, i32, i32, i32) #0