AMDGPU: Fix missing immarg for mfma intrinsics

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@366230 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/include/llvm/IR/IntrinsicsAMDGPU.td b/include/llvm/IR/IntrinsicsAMDGPU.td
index 1cde3af..bad4216 100644 (file)
--- a/include/llvm/IR/IntrinsicsAMDGPU.td
+++ b/include/llvm/IR/IntrinsicsAMDGPU.td
@@ -1675,83 +1675,103 @@ def int_amdgcn_global_atomic_fadd    : AMDGPUGlobalAtomicNoRtn;
 // llvm.amdgcn.mfma.f32.* vdst, srcA, srcB, srcC, cbsz, abid, blgp
 def int_amdgcn_mfma_f32_32x32x1f32 : Intrinsic<[llvm_v32f32_ty],
   [llvm_float_ty, llvm_float_ty, llvm_v32f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_16x16x1f32 : Intrinsic<[llvm_v16f32_ty],
   [llvm_float_ty, llvm_float_ty, llvm_v16f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_4x4x1f32 : Intrinsic<[llvm_v4f32_ty],
   [llvm_float_ty, llvm_float_ty, llvm_v4f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_32x32x2f32 : Intrinsic<[llvm_v16f32_ty],
   [llvm_float_ty, llvm_float_ty, llvm_v16f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_16x16x4f32 : Intrinsic<[llvm_v4f32_ty],
   [llvm_float_ty, llvm_float_ty, llvm_v4f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_32x32x4f16 : Intrinsic<[llvm_v32f32_ty],
   [llvm_v4f16_ty, llvm_v4f16_ty, llvm_v32f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_16x16x4f16 : Intrinsic<[llvm_v16f32_ty],
   [llvm_v4f16_ty, llvm_v4f16_ty, llvm_v16f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_4x4x4f16 : Intrinsic<[llvm_v4f32_ty],
   [llvm_v4f16_ty, llvm_v4f16_ty, llvm_v4f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_32x32x8f16 : Intrinsic<[llvm_v16f32_ty],
   [llvm_v4f16_ty, llvm_v4f16_ty, llvm_v16f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_16x16x16f16 : Intrinsic<[llvm_v4f32_ty],
   [llvm_v4f16_ty, llvm_v4f16_ty, llvm_v4f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_i32_32x32x4i8 : Intrinsic<[llvm_v32i32_ty],
   [llvm_i32_ty, llvm_i32_ty, llvm_v32i32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_i32_16x16x4i8 : Intrinsic<[llvm_v16i32_ty],
   [llvm_i32_ty, llvm_i32_ty, llvm_v16i32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_i32_4x4x4i8 : Intrinsic<[llvm_v4i32_ty],
   [llvm_i32_ty, llvm_i32_ty, llvm_v4i32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_i32_32x32x8i8 : Intrinsic<[llvm_v16i32_ty],
   [llvm_i32_ty, llvm_i32_ty, llvm_v16i32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_i32_16x16x16i8 : Intrinsic<[llvm_v4i32_ty],
   [llvm_i32_ty, llvm_i32_ty, llvm_v4i32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_32x32x2bf16 : Intrinsic<[llvm_v32f32_ty],
   [llvm_v2i16_ty, llvm_v2i16_ty, llvm_v32f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_16x16x2bf16 : Intrinsic<[llvm_v16f32_ty],
   [llvm_v2i16_ty, llvm_v2i16_ty, llvm_v16f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_4x4x2bf16 : Intrinsic<[llvm_v4f32_ty],
   [llvm_v2i16_ty, llvm_v2i16_ty, llvm_v4f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_32x32x4bf16 : Intrinsic<[llvm_v16f32_ty],
   [llvm_v2i16_ty, llvm_v2i16_ty, llvm_v16f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 def int_amdgcn_mfma_f32_16x16x8bf16 : Intrinsic<[llvm_v4f32_ty],
   [llvm_v2i16_ty, llvm_v2i16_ty, llvm_v4f32_ty,
-   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
+   llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+   [IntrConvergent, IntrNoMem, ImmArg<3>, ImmArg<4>, ImmArg<5>]>;
 
 //===----------------------------------------------------------------------===//
 // Special Intrinsics for backend internal use only. No frontend

diff --git a/test/Verifier/AMDGPU/intrinsic-immarg.ll b/test/Verifier/AMDGPU/intrinsic-immarg.ll
index a72d1da..7609838 100644 (file)
--- a/test/Verifier/AMDGPU/intrinsic-immarg.ll
+++ b/test/Verifier/AMDGPU/intrinsic-immarg.ll
@@ -674,3 +674,26 @@ define void @test_interp_p2_f16(float %arg0, float %arg1, i32 %arg2, i32 %arg3,
 
   ret void
 }
+
+declare <32 x i32> @llvm.amdgcn.mfma.f32.32x32x1f32(float, float, <32 x i32>, i32, i32, i32)
+define void @test_mfma_f32_32x32x1f32(float %arg0, float %arg1, <32 x i32> %arg2, i32 %arg3, i32 %arg4, i32 %arg5) {
+  ; CHECK: immarg operand has non-immediate parameter
+  ; CHECK-NEXT: i32 %arg3
+  ; CHECK-NEXT: %val0 = call <32 x i32> @llvm.amdgcn.mfma.f32.32x32x1f32(float %arg0, float %arg1, <32 x i32> %arg2, i32 %arg3, i32 2, i32 3)
+  %val0 = call <32 x i32> @llvm.amdgcn.mfma.f32.32x32x1f32(float %arg0, float %arg1, <32 x i32> %arg2, i32 %arg3, i32 2, i32 3)
+  store volatile <32 x i32> %val0, <32 x i32> addrspace(1)* undef
+
+  ; CHECK: immarg operand has non-immediate parameter
+  ; CHECK-NEXT: i32 %arg4
+  ; CHECK-NEXT: %val1 = call <32 x i32> @llvm.amdgcn.mfma.f32.32x32x1f32(float %arg0, float %arg1, <32 x i32> %arg2, i32 1, i32 %arg4, i32 3)
+  %val1 = call <32 x i32> @llvm.amdgcn.mfma.f32.32x32x1f32(float %arg0, float %arg1, <32 x i32> %arg2, i32 1, i32 %arg4, i32 3)
+  store volatile <32 x i32> %val1, <32 x i32> addrspace(1)* undef
+
+  ; CHECK: immarg operand has non-immediate parameter
+  ; CHECK-NEXT: i32 %arg5
+  ; CHECK-NEXT: %val2 = call <32 x i32> @llvm.amdgcn.mfma.f32.32x32x1f32(float %arg0, float %arg1, <32 x i32> %arg2, i32 1, i32 2, i32 %arg5)
+  %val2 = call <32 x i32> @llvm.amdgcn.mfma.f32.32x32x1f32(float %arg0, float %arg1, <32 x i32> %arg2, i32 1, i32 2, i32 %arg5)
+  store volatile <32 x i32> %val2, <32 x i32> addrspace(1)* undef
+
+  ret void
+}