[llvm-branch-commits] [clang] [llvm] AMDGPU: Define v_mfma_f32_32x32x16_bf16 for gfx950 (PR #116679)
Matt Arsenault via llvm-branch-commits
llvm-branch-commits at lists.llvm.org
Mon Nov 18 12:07:36 PST 2024
https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/116679
>From 330427ffa20582990fe6cf7b65fba33b210e8ae5 Mon Sep 17 00:00:00 2001
From: Matt Arsenault <Matthew.Arsenault at amd.com>
Date: Tue, 19 Dec 2023 12:46:00 +0700
Subject: [PATCH] AMDGPU: Define v_mfma_f32_32x32x16_bf16 for gfx950
Unlike the existing gfx940 intrinsics using short/i16 in place of
bfloat, this uses the natural bfloat type.
---
clang/include/clang/Basic/BuiltinsAMDGPU.def | 2 +
.../CodeGenOpenCL/builtins-amdgcn-mfma.cl | 6 +
.../builtins-amdgcn-error-gfx950-param.cl | 7 +
.../builtins-amdgcn-error-gfx950.cl | 5 +-
llvm/include/llvm/IR/IntrinsicsAMDGPU.td | 2 +
llvm/lib/Target/AMDGPU/SIInstrInfo.td | 1 +
llvm/lib/Target/AMDGPU/VOP3PInstructions.td | 6 +
.../UniformityAnalysis/AMDGPU/intrinsics.ll | 8 +
.../AMDGPU/llvm.amdgcn.mfma.gfx950.bf16.ll | 474 ++++++++++++++++++
llvm/test/MC/AMDGPU/mai-gfx950.s | 56 ++-
.../MC/Disassembler/AMDGPU/gfx950_mai.txt | 27 +
llvm/test/tools/llvm-mca/AMDGPU/gfx950.s | 10 +-
12 files changed, 596 insertions(+), 8 deletions(-)
create mode 100644 llvm/test/CodeGen/AMDGPU/llvm.amdgcn.mfma.gfx950.bf16.ll
diff --git a/clang/include/clang/Basic/BuiltinsAMDGPU.def b/clang/include/clang/Basic/BuiltinsAMDGPU.def
index 6917d8d1aca69d..7ce8f2c1669d67 100644
--- a/clang/include/clang/Basic/BuiltinsAMDGPU.def
+++ b/clang/include/clang/Basic/BuiltinsAMDGPU.def
@@ -437,6 +437,8 @@ TARGET_BUILTIN(__builtin_amdgcn_cvt_sr_fp8_f32, "ifiiIi", "nc", "fp8-conversion-
TARGET_BUILTIN(__builtin_amdgcn_mfma_f32_16x16x32_f16, "V4fV8hV8hV4fIiIiIi", "nc", "gfx950-insts")
TARGET_BUILTIN(__builtin_amdgcn_mfma_f32_32x32x16_f16, "V16fV8hV8hV16fIiIiIi", "nc", "gfx950-insts")
+TARGET_BUILTIN(__builtin_amdgcn_mfma_f32_32x32x16_bf16, "V16fV8yV8yV16fIiIiIi", "nc", "gfx950-insts")
+
//===----------------------------------------------------------------------===//
// GFX12+ only builtins.
//===----------------------------------------------------------------------===//
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-mfma.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-mfma.cl
index a644a60f9ec381..841d8fcad0fee0 100644
--- a/clang/test/CodeGenOpenCL/builtins-amdgcn-mfma.cl
+++ b/clang/test/CodeGenOpenCL/builtins-amdgcn-mfma.cl
@@ -24,6 +24,7 @@ typedef short v8s __attribute__((ext_vector_type(8)));
typedef short v16s __attribute__((ext_vector_type(16)));
typedef short v32s __attribute__((ext_vector_type(32)));
typedef double v4d __attribute__((ext_vector_type(4)));
+typedef __bf16 v8bf16 __attribute__((ext_vector_type(8)));
#ifdef MFMA_GFX908_TESTS
@@ -424,5 +425,10 @@ v16f test_mfma_f32_32x32x16_f16(v8h a, v8h b, v16f c)
return __builtin_amdgcn_mfma_f32_32x32x16_f16(a, b, c, 1, 2, 3);
}
+// CHECK-GFX950-LABEL: @test_mfma_f32_32x32x16_bf16(
+// CHECK-GFX950: tail call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %a, <8 x bfloat> %b, <16 x float> %c, i32 1, i32 2, i32 3)
+v16f test_mfma_f32_32x32x16_bf16(v8bf16 a, v8bf16 b, v16f c) {
+ return __builtin_amdgcn_mfma_f32_32x32x16_bf16(a, b, c, 1, 2, 3);
+}
#endif
diff --git a/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950-param.cl b/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950-param.cl
index 4c267e2cac5cad..4af67763c40dd2 100644
--- a/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950-param.cl
+++ b/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950-param.cl
@@ -4,6 +4,7 @@
typedef float float4 __attribute__((ext_vector_type(4)));
typedef float float16 __attribute__((ext_vector_type(16)));
typedef half half8 __attribute__((ext_vector_type(8)));
+typedef __bf16 bfloat8 __attribute__((ext_vector_type(8)));
void test_mfma_f32_16x16x32_f16(__global float4* out, half8 a, half8 b, float4 c, int X) {
@@ -19,3 +20,9 @@ void test_mfma_f32_32x32x16_f16(__global float16* out, half8 a, half8 b, float16
*out = __builtin_amdgcn_mfma_f32_32x32x16_f16(a, b, c, 0, X, 0); // expected-error{{argument to '__builtin_amdgcn_mfma_f32_32x32x16_f16' must be a constant integer}}
*out = __builtin_amdgcn_mfma_f32_32x32x16_f16(a, b, c, 0, 0, X); // expected-error{{argument to '__builtin_amdgcn_mfma_f32_32x32x16_f16' must be a constant integer}}
}
+
+void test_mfma_f32_32x32x16_bf16(__global float16* out, bfloat8 a, bfloat8 b, float16 c, int X) {
+ *out = __builtin_amdgcn_mfma_f32_32x32x16_bf16(a, b, c, X, 0, 0); // expected-error{{argument to '__builtin_amdgcn_mfma_f32_32x32x16_bf16' must be a constant integer}}
+ *out = __builtin_amdgcn_mfma_f32_32x32x16_bf16(a, b, c, 0, X, 0); // expected-error{{argument to '__builtin_amdgcn_mfma_f32_32x32x16_bf16' must be a constant integer}}
+ *out = __builtin_amdgcn_mfma_f32_32x32x16_bf16(a, b, c, 0, 0, X); // expected-error{{argument to '__builtin_amdgcn_mfma_f32_32x32x16_bf16' must be a constant integer}}
+}
diff --git a/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950.cl b/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950.cl
index 0b3a8e78e1c795..e0fd2aa5c58a02 100644
--- a/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950.cl
+++ b/clang/test/SemaOpenCL/builtins-amdgcn-error-gfx950.cl
@@ -4,9 +4,12 @@
typedef float float4 __attribute__((ext_vector_type(4)));
typedef float float16 __attribute__((ext_vector_type(16)));
typedef half half8 __attribute__((ext_vector_type(8)));
+typedef __bf16 bfloat8 __attribute__((ext_vector_type(8)));
void test(__global float4* out0, half8 a0, half8 b0, float4 c0,
- __global float16* out1, half8 a1, half8 b1, float16 c1) {
+ __global float16* out1, half8 a1, half8 b1, float16 c1,
+ __global float16* out2, bfloat8 a2, bfloat8 b2, float16 c2) {
*out0 = __builtin_amdgcn_mfma_f32_16x16x32_f16(a0, b0, c0, 0, 0, 0); // expected-error{{'__builtin_amdgcn_mfma_f32_16x16x32_f16' needs target feature gfx950-insts}}
*out1 = __builtin_amdgcn_mfma_f32_32x32x16_f16(a1, b1, c1, 0, 0, 0); // expected-error{{'__builtin_amdgcn_mfma_f32_32x32x16_f16' needs target feature gfx950-insts}}
+ *out2 = __builtin_amdgcn_mfma_f32_32x32x16_bf16(a2, b2, c2, 0, 0, 0); // expected-error{{'__builtin_amdgcn_mfma_f32_32x32x16_bf16' needs target feature gfx950-insts}}
}
diff --git a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
index ec1234e7bc7d94..15f33cdbf92e6e 100644
--- a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
+++ b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
@@ -3117,6 +3117,8 @@ def int_amdgcn_cvt_sr_fp8_f32 : ClangBuiltin<"__builtin_amdgcn_cvt_sr_fp8_f32">,
defset list<Intrinsic> AMDGPUMFMAIntrinsics950 = {
def int_amdgcn_mfma_f32_16x16x32_f16 : AMDGPUMfmaIntrinsic<llvm_v4f32_ty, llvm_v8f16_ty>;
def int_amdgcn_mfma_f32_32x32x16_f16 : AMDGPUMfmaIntrinsic<llvm_v16f32_ty, llvm_v8f16_ty>;
+
+def int_amdgcn_mfma_f32_32x32x16_bf16 : AMDGPUMfmaIntrinsic<llvm_v16f32_ty, llvm_v8bf16_ty>;
}
//===----------------------------------------------------------------------===//
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.td b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
index 7df9be5c6f7a0b..2079b34d0448f4 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
@@ -2848,6 +2848,7 @@ def VOP_V16F32_V2I32_V4I32_I32 : VOPProfile <[v16f32, v2i32, v4i32, i32]>;
def VOP_V4F32_V8F16_V8F16_V4F32 : VOPProfile <[v4f32, v8f16, v8f16, v4f32]>;
def VOP_V16F32_V8F16_V8F16_V16F32 : VOPProfile <[v16f32, v8f16, v8f16, v16f32]>;
+def VOP_V16F32_V8BF16_V8BF16_V16F32 : VOPProfile <[v16f32, v8bf16, v8bf16, v16f32]>;
class Commutable_REV <string revOp, bit isOrig> {
diff --git a/llvm/lib/Target/AMDGPU/VOP3PInstructions.td b/llvm/lib/Target/AMDGPU/VOP3PInstructions.td
index 58e26a96ece202..08882e41d863a1 100644
--- a/llvm/lib/Target/AMDGPU/VOP3PInstructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP3PInstructions.td
@@ -631,6 +631,9 @@ def VOPProfileMAI_F32_V8F16_X32_VCD : VOPProfileMAI<VOP_V4F32_V8F16_V8F16_V4F32,
def VOPProfileMAI_F32_V8F16_X16 : VOPProfileMAI<VOP_V16F32_V8F16_V8F16_V16F32, AISrc_512_f32, ADst_512, AVSrc_128>;
def VOPProfileMAI_F32_V8F16_X16_VCD : VOPProfileMAI<VOP_V16F32_V8F16_V8F16_V16F32, VISrc_512_f32, VDst_512, AVSrc_128>;
+def VOPProfileMAI_F32_V8BF16_X16 : VOPProfileMAI<VOP_V16F32_V8BF16_V8BF16_V16F32, AISrc_512_f32, ADst_512, AVSrc_128>;
+def VOPProfileMAI_F32_V8BF16_X16_VCD : VOPProfileMAI<VOP_V16F32_V8BF16_V8BF16_V16F32, VISrc_512_f32, VDst_512, AVSrc_128>;
+
class MFMATable <bit is_mac, string Name> {
bit IsMac = is_mac;
string FMAOp = Name;
@@ -747,6 +750,7 @@ defm V_MFMA_F32_32X32X4BF16 : MAIInst<"v_mfma_f32_32x32x4bf16", "F32_V2I16_X16",
let SubtargetPredicate = HasGFX950Insts, is_gfx940_xdl = 1 in {
defm V_MFMA_F32_16X16X32_F16 : MAIInst<"v_mfma_f32_16x16x32f16", "F32_V8F16_X32", int_amdgcn_mfma_f32_16x16x32_f16>;
defm V_MFMA_F32_32X32X16_F16 : MAIInst<"v_mfma_f32_32x32x16f16", "F32_V8F16_X16", int_amdgcn_mfma_f32_32x32x16_f16>;
+defm V_MFMA_F32_32X32X16_BF16 : MAIInst<"v_mfma_f32_32x32x16bf16", "F32_V8BF16_X16", int_amdgcn_mfma_f32_32x32x16_bf16>;
}
let Predicates = [isGFX90APlus] in {
@@ -1786,6 +1790,8 @@ defm V_MFMA_F64_4X4X4F64 : VOP3P_Real_MFMA_gfx90a <0x6f>;
defm V_MFMA_F32_16X16X32_F16 : VOP3P_Real_MFMA_gfx950 <0x54, "v_mfma_f32_16x16x32_f16">;
defm V_MFMA_F32_32X32X16_F16 : VOP3P_Real_MFMA_gfx950 <0x55, "v_mfma_f32_32x32x16_f16">;
+defm V_MFMA_F32_32X32X16_BF16 : VOP3P_Real_MFMA_gfx950 <0x37, "v_mfma_f32_32x32x16_bf16">;
+
defm V_MFMA_I32_32X32X16I8 : VOP3P_Real_MFMA_gfx940 <0x56, "v_mfma_i32_32x32x16_i8">;
defm V_MFMA_I32_16X16X32I8 : VOP3P_Real_MFMA_gfx940 <0x57, "v_mfma_i32_16x16x32_i8">;
let SubtargetPredicate = HasXF32Insts in {
diff --git a/llvm/test/Analysis/UniformityAnalysis/AMDGPU/intrinsics.ll b/llvm/test/Analysis/UniformityAnalysis/AMDGPU/intrinsics.ll
index c457d867af361e..00a3aaf77f9003 100644
--- a/llvm/test/Analysis/UniformityAnalysis/AMDGPU/intrinsics.ll
+++ b/llvm/test/Analysis/UniformityAnalysis/AMDGPU/intrinsics.ll
@@ -278,6 +278,14 @@ define amdgpu_kernel void @mfma_f32_32x32x16_f16(<8 x half> %arg0, <8 x half> %a
ret void
}
+; CHECK: DIVERGENT: %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 immarg 0, i32 immarg 0, i32 immarg 0)
+define amdgpu_kernel void @mfma_f32_32x32x16_bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, ptr addrspace(1) %out) {
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 immarg 0, i32 immarg 0, i32 immarg 0)
+ store <16 x float> %result, ptr addrspace(1) %out
+ ret void
+}
+
+
declare i32 @llvm.amdgcn.ds.swizzle(i32, i32) #1
declare i32 @llvm.amdgcn.permlane16.i32(i32, i32, i32, i32, i1, i1) #1
declare i32 @llvm.amdgcn.permlanex16.i32(i32, i32, i32, i32, i1, i1) #1
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.mfma.gfx950.bf16.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.mfma.gfx950.bf16.ll
new file mode 100644
index 00000000000000..2da602713d72c4
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.mfma.gfx950.bf16.ll
@@ -0,0 +1,474 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -march=amdgcn -mcpu=gfx950 < %s | FileCheck -enable-var-scope --check-prefix=GCN %s
+
+; FIXME: bfloat vector arguments are broken in globalisel.
+; https://github.com/llvm/llvm-project/issues/77055
+
+declare <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat>, <8 x bfloat>, <16 x float>, i32 immarg, i32 immarg, i32 immarg)
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.mfma.f32.32x32x16.bf16
+; --------------------------------------------------------------------
+
+define amdgpu_kernel void @test_mfma_f32_32x32x16_bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2) #1 {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_dwordx8 s[24:31], s[4:5], 0x24
+; GCN-NEXT: s_load_dwordx16 s[8:23], s[4:5], 0x64
+; GCN-NEXT: v_mov_b64_e32 v[12:13], 48
+; GCN-NEXT: v_mov_b64_e32 v[14:15], 32
+; GCN-NEXT: v_mov_b64_e32 v[16:17], 16
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: v_mov_b64_e32 v[0:1], s[24:25]
+; GCN-NEXT: v_mov_b64_e32 v[2:3], s[26:27]
+; GCN-NEXT: v_mov_b64_e32 v[4:5], s[28:29]
+; GCN-NEXT: v_accvgpr_write_b32 a0, s8
+; GCN-NEXT: v_mov_b64_e32 v[6:7], s[30:31]
+; GCN-NEXT: v_accvgpr_write_b32 a1, s9
+; GCN-NEXT: v_accvgpr_write_b32 a2, s10
+; GCN-NEXT: v_accvgpr_write_b32 a3, s11
+; GCN-NEXT: v_accvgpr_write_b32 a4, s12
+; GCN-NEXT: v_accvgpr_write_b32 a5, s13
+; GCN-NEXT: v_accvgpr_write_b32 a6, s14
+; GCN-NEXT: v_accvgpr_write_b32 a7, s15
+; GCN-NEXT: v_accvgpr_write_b32 a8, s16
+; GCN-NEXT: v_accvgpr_write_b32 a9, s17
+; GCN-NEXT: v_accvgpr_write_b32 a10, s18
+; GCN-NEXT: v_accvgpr_write_b32 a11, s19
+; GCN-NEXT: v_accvgpr_write_b32 a12, s20
+; GCN-NEXT: v_accvgpr_write_b32 a13, s21
+; GCN-NEXT: v_accvgpr_write_b32 a14, s22
+; GCN-NEXT: v_accvgpr_write_b32 a15, s23
+; GCN-NEXT: v_mov_b64_e32 v[18:19], 0
+; GCN-NEXT: v_mov_b32_e32 v8, s16
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[16:31], v[0:3], v[4:7], a[0:15]
+; GCN-NEXT: v_mov_b32_e32 v0, s20
+; GCN-NEXT: v_mov_b32_e32 v1, s21
+; GCN-NEXT: v_mov_b32_e32 v2, s22
+; GCN-NEXT: v_mov_b32_e32 v3, s23
+; GCN-NEXT: v_mov_b32_e32 v9, s17
+; GCN-NEXT: v_mov_b32_e32 v10, s18
+; GCN-NEXT: v_mov_b32_e32 v11, s19
+; GCN-NEXT: s_nop 3
+; GCN-NEXT: global_store_dwordx4 v[12:13], a[28:31], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[14:15], a[24:27], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[16:17], a[20:23], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[18:19], a[16:19], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[14:15], v[8:11], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[12:13], v[0:3], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s8
+; GCN-NEXT: v_mov_b32_e32 v1, s9
+; GCN-NEXT: v_mov_b32_e32 v2, s10
+; GCN-NEXT: v_mov_b32_e32 v3, s11
+; GCN-NEXT: global_store_dwordx4 v[18:19], v[0:3], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s12
+; GCN-NEXT: v_mov_b32_e32 v1, s13
+; GCN-NEXT: v_mov_b32_e32 v2, s14
+; GCN-NEXT: v_mov_b32_e32 v3, s15
+; GCN-NEXT: global_store_dwordx4 v[16:17], v[0:3], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_endpgm
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 0, i32 0, i32 0)
+ store volatile <16 x float> %result, ptr addrspace(1) null
+ store volatile <16 x float> %arg2, ptr addrspace(1) null
+ ret void
+}
+
+define amdgpu_kernel void @test_mfma_f32_32x32x16_bf16__flags(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2) #1 {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16__flags:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_dwordx8 s[24:31], s[4:5], 0x24
+; GCN-NEXT: s_load_dwordx16 s[8:23], s[4:5], 0x64
+; GCN-NEXT: v_mov_b64_e32 v[12:13], 48
+; GCN-NEXT: v_mov_b64_e32 v[14:15], 32
+; GCN-NEXT: v_mov_b64_e32 v[16:17], 16
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: v_mov_b64_e32 v[0:1], s[24:25]
+; GCN-NEXT: v_mov_b64_e32 v[2:3], s[26:27]
+; GCN-NEXT: v_mov_b64_e32 v[4:5], s[28:29]
+; GCN-NEXT: v_accvgpr_write_b32 a0, s8
+; GCN-NEXT: v_mov_b64_e32 v[6:7], s[30:31]
+; GCN-NEXT: v_accvgpr_write_b32 a1, s9
+; GCN-NEXT: v_accvgpr_write_b32 a2, s10
+; GCN-NEXT: v_accvgpr_write_b32 a3, s11
+; GCN-NEXT: v_accvgpr_write_b32 a4, s12
+; GCN-NEXT: v_accvgpr_write_b32 a5, s13
+; GCN-NEXT: v_accvgpr_write_b32 a6, s14
+; GCN-NEXT: v_accvgpr_write_b32 a7, s15
+; GCN-NEXT: v_accvgpr_write_b32 a8, s16
+; GCN-NEXT: v_accvgpr_write_b32 a9, s17
+; GCN-NEXT: v_accvgpr_write_b32 a10, s18
+; GCN-NEXT: v_accvgpr_write_b32 a11, s19
+; GCN-NEXT: v_accvgpr_write_b32 a12, s20
+; GCN-NEXT: v_accvgpr_write_b32 a13, s21
+; GCN-NEXT: v_accvgpr_write_b32 a14, s22
+; GCN-NEXT: v_accvgpr_write_b32 a15, s23
+; GCN-NEXT: v_mov_b64_e32 v[18:19], 0
+; GCN-NEXT: v_mov_b32_e32 v8, s16
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[16:31], v[0:3], v[4:7], a[0:15] cbsz:2 abid:3 blgp:1
+; GCN-NEXT: v_mov_b32_e32 v0, s20
+; GCN-NEXT: v_mov_b32_e32 v1, s21
+; GCN-NEXT: v_mov_b32_e32 v2, s22
+; GCN-NEXT: v_mov_b32_e32 v3, s23
+; GCN-NEXT: v_mov_b32_e32 v9, s17
+; GCN-NEXT: v_mov_b32_e32 v10, s18
+; GCN-NEXT: v_mov_b32_e32 v11, s19
+; GCN-NEXT: s_nop 3
+; GCN-NEXT: global_store_dwordx4 v[12:13], a[28:31], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[14:15], a[24:27], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[16:17], a[20:23], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[18:19], a[16:19], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[14:15], v[8:11], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v[12:13], v[0:3], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s8
+; GCN-NEXT: v_mov_b32_e32 v1, s9
+; GCN-NEXT: v_mov_b32_e32 v2, s10
+; GCN-NEXT: v_mov_b32_e32 v3, s11
+; GCN-NEXT: global_store_dwordx4 v[18:19], v[0:3], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s12
+; GCN-NEXT: v_mov_b32_e32 v1, s13
+; GCN-NEXT: v_mov_b32_e32 v2, s14
+; GCN-NEXT: v_mov_b32_e32 v3, s15
+; GCN-NEXT: global_store_dwordx4 v[16:17], v[0:3], off sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_endpgm
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 2, i32 3, i32 1)
+ store volatile <16 x float> %result, ptr addrspace(1) null
+ store volatile <16 x float> %arg2, ptr addrspace(1) null
+ ret void
+}
+
+define <16 x float> @test_mfma_f32_32x32x16_bf16__mac(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2) {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16__mac:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GCN-NEXT: v_accvgpr_write_b32 a0, v8
+; GCN-NEXT: v_accvgpr_write_b32 a1, v9
+; GCN-NEXT: v_accvgpr_write_b32 a2, v10
+; GCN-NEXT: v_accvgpr_write_b32 a3, v11
+; GCN-NEXT: v_accvgpr_write_b32 a4, v12
+; GCN-NEXT: v_accvgpr_write_b32 a5, v13
+; GCN-NEXT: v_accvgpr_write_b32 a6, v14
+; GCN-NEXT: v_accvgpr_write_b32 a7, v15
+; GCN-NEXT: v_accvgpr_write_b32 a8, v16
+; GCN-NEXT: v_accvgpr_write_b32 a9, v17
+; GCN-NEXT: v_accvgpr_write_b32 a10, v18
+; GCN-NEXT: v_accvgpr_write_b32 a11, v19
+; GCN-NEXT: v_accvgpr_write_b32 a12, v20
+; GCN-NEXT: v_accvgpr_write_b32 a13, v21
+; GCN-NEXT: v_accvgpr_write_b32 a14, v22
+; GCN-NEXT: v_accvgpr_write_b32 a15, v23
+; GCN-NEXT: s_nop 1
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[0:15], v[0:3], v[4:7], a[0:15]
+; GCN-NEXT: s_nop 7
+; GCN-NEXT: s_nop 2
+; GCN-NEXT: v_accvgpr_read_b32 v0, a0
+; GCN-NEXT: v_accvgpr_read_b32 v1, a1
+; GCN-NEXT: v_accvgpr_read_b32 v2, a2
+; GCN-NEXT: v_accvgpr_read_b32 v3, a3
+; GCN-NEXT: v_accvgpr_read_b32 v4, a4
+; GCN-NEXT: v_accvgpr_read_b32 v5, a5
+; GCN-NEXT: v_accvgpr_read_b32 v6, a6
+; GCN-NEXT: v_accvgpr_read_b32 v7, a7
+; GCN-NEXT: v_accvgpr_read_b32 v8, a8
+; GCN-NEXT: v_accvgpr_read_b32 v9, a9
+; GCN-NEXT: v_accvgpr_read_b32 v10, a10
+; GCN-NEXT: v_accvgpr_read_b32 v11, a11
+; GCN-NEXT: v_accvgpr_read_b32 v12, a12
+; GCN-NEXT: v_accvgpr_read_b32 v13, a13
+; GCN-NEXT: v_accvgpr_read_b32 v14, a14
+; GCN-NEXT: v_accvgpr_read_b32 v15, a15
+; GCN-NEXT: s_setpc_b64 s[30:31]
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 0, i32 0, i32 0)
+ ret <16 x float> %result
+}
+
+define <16 x float> @test_mfma_f32_32x32x16_bf16__mac__flags(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2) {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16__mac__flags:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GCN-NEXT: v_accvgpr_write_b32 a0, v8
+; GCN-NEXT: v_accvgpr_write_b32 a1, v9
+; GCN-NEXT: v_accvgpr_write_b32 a2, v10
+; GCN-NEXT: v_accvgpr_write_b32 a3, v11
+; GCN-NEXT: v_accvgpr_write_b32 a4, v12
+; GCN-NEXT: v_accvgpr_write_b32 a5, v13
+; GCN-NEXT: v_accvgpr_write_b32 a6, v14
+; GCN-NEXT: v_accvgpr_write_b32 a7, v15
+; GCN-NEXT: v_accvgpr_write_b32 a8, v16
+; GCN-NEXT: v_accvgpr_write_b32 a9, v17
+; GCN-NEXT: v_accvgpr_write_b32 a10, v18
+; GCN-NEXT: v_accvgpr_write_b32 a11, v19
+; GCN-NEXT: v_accvgpr_write_b32 a12, v20
+; GCN-NEXT: v_accvgpr_write_b32 a13, v21
+; GCN-NEXT: v_accvgpr_write_b32 a14, v22
+; GCN-NEXT: v_accvgpr_write_b32 a15, v23
+; GCN-NEXT: s_nop 1
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[0:15], v[0:3], v[4:7], a[0:15] cbsz:1 abid:1 blgp:1
+; GCN-NEXT: s_nop 7
+; GCN-NEXT: s_nop 2
+; GCN-NEXT: v_accvgpr_read_b32 v0, a0
+; GCN-NEXT: v_accvgpr_read_b32 v1, a1
+; GCN-NEXT: v_accvgpr_read_b32 v2, a2
+; GCN-NEXT: v_accvgpr_read_b32 v3, a3
+; GCN-NEXT: v_accvgpr_read_b32 v4, a4
+; GCN-NEXT: v_accvgpr_read_b32 v5, a5
+; GCN-NEXT: v_accvgpr_read_b32 v6, a6
+; GCN-NEXT: v_accvgpr_read_b32 v7, a7
+; GCN-NEXT: v_accvgpr_read_b32 v8, a8
+; GCN-NEXT: v_accvgpr_read_b32 v9, a9
+; GCN-NEXT: v_accvgpr_read_b32 v10, a10
+; GCN-NEXT: v_accvgpr_read_b32 v11, a11
+; GCN-NEXT: v_accvgpr_read_b32 v12, a12
+; GCN-NEXT: v_accvgpr_read_b32 v13, a13
+; GCN-NEXT: v_accvgpr_read_b32 v14, a14
+; GCN-NEXT: v_accvgpr_read_b32 v15, a15
+; GCN-NEXT: s_setpc_b64 s[30:31]
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 1, i32 1, i32 1)
+ ret <16 x float> %result
+}
+
+define amdgpu_kernel void @test_mfma_f32_32x32x16_bf16__vgprcd(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, ptr addrspace(1) %out) #0 {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16__vgprcd:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_dwordx8 s[24:31], s[4:5], 0x24
+; GCN-NEXT: s_load_dwordx16 s[8:23], s[4:5], 0x64
+; GCN-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0xa4
+; GCN-NEXT: v_mov_b32_e32 v12, 0
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: v_mov_b64_e32 v[0:1], s[24:25]
+; GCN-NEXT: v_mov_b64_e32 v[2:3], s[26:27]
+; GCN-NEXT: v_mov_b64_e32 v[4:5], s[28:29]
+; GCN-NEXT: v_accvgpr_write_b32 a31, s23
+; GCN-NEXT: v_mov_b64_e32 v[6:7], s[30:31]
+; GCN-NEXT: v_accvgpr_write_b32 a30, s22
+; GCN-NEXT: v_accvgpr_write_b32 a29, s21
+; GCN-NEXT: v_accvgpr_write_b32 a28, s20
+; GCN-NEXT: v_accvgpr_write_b32 a27, s19
+; GCN-NEXT: v_accvgpr_write_b32 a26, s18
+; GCN-NEXT: v_accvgpr_write_b32 a25, s17
+; GCN-NEXT: v_accvgpr_write_b32 a24, s16
+; GCN-NEXT: v_accvgpr_write_b32 a23, s15
+; GCN-NEXT: v_accvgpr_write_b32 a22, s14
+; GCN-NEXT: v_accvgpr_write_b32 a21, s13
+; GCN-NEXT: v_accvgpr_write_b32 a20, s12
+; GCN-NEXT: v_accvgpr_write_b32 a19, s11
+; GCN-NEXT: v_accvgpr_write_b32 a18, s10
+; GCN-NEXT: v_accvgpr_write_b32 a17, s9
+; GCN-NEXT: v_accvgpr_write_b32 a16, s8
+; GCN-NEXT: v_mov_b32_e32 v8, s20
+; GCN-NEXT: v_mov_b32_e32 v9, s21
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[0:15], v[0:3], v[4:7], a[16:31]
+; GCN-NEXT: v_mov_b32_e32 v10, s22
+; GCN-NEXT: v_mov_b32_e32 v11, s23
+; GCN-NEXT: v_mov_b32_e32 v0, s16
+; GCN-NEXT: v_mov_b32_e32 v1, s17
+; GCN-NEXT: v_mov_b32_e32 v2, s18
+; GCN-NEXT: v_mov_b32_e32 v3, s19
+; GCN-NEXT: global_store_dwordx4 v12, v[8:11], s[0:1] offset:48 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, v[0:3], s[0:1] offset:32 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s12
+; GCN-NEXT: v_mov_b32_e32 v1, s13
+; GCN-NEXT: v_mov_b32_e32 v2, s14
+; GCN-NEXT: v_mov_b32_e32 v3, s15
+; GCN-NEXT: global_store_dwordx4 v12, v[0:3], s[0:1] offset:16 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s8
+; GCN-NEXT: v_mov_b32_e32 v1, s9
+; GCN-NEXT: v_mov_b32_e32 v2, s10
+; GCN-NEXT: v_mov_b32_e32 v3, s11
+; GCN-NEXT: global_store_dwordx4 v12, v[0:3], s[0:1] sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[8:11], s[0:1] offset:32 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[12:15], s[0:1] offset:48 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[0:3], s[0:1] sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[4:7], s[0:1] offset:16 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_endpgm
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 0, i32 0, i32 0)
+ store volatile <16 x float> %arg2, ptr addrspace(1) %out
+ store volatile <16 x float> %result, ptr addrspace(1) %out
+ ret void
+}
+
+define amdgpu_kernel void @test_mfma_f32_32x32x16_bf16__vgprcd__flags(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, ptr addrspace(1) %out) #0 {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16__vgprcd__flags:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_dwordx8 s[24:31], s[4:5], 0x24
+; GCN-NEXT: s_load_dwordx16 s[8:23], s[4:5], 0x64
+; GCN-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0xa4
+; GCN-NEXT: v_mov_b32_e32 v12, 0
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: v_mov_b64_e32 v[0:1], s[24:25]
+; GCN-NEXT: v_mov_b64_e32 v[2:3], s[26:27]
+; GCN-NEXT: v_mov_b64_e32 v[4:5], s[28:29]
+; GCN-NEXT: v_accvgpr_write_b32 a31, s23
+; GCN-NEXT: v_mov_b64_e32 v[6:7], s[30:31]
+; GCN-NEXT: v_accvgpr_write_b32 a30, s22
+; GCN-NEXT: v_accvgpr_write_b32 a29, s21
+; GCN-NEXT: v_accvgpr_write_b32 a28, s20
+; GCN-NEXT: v_accvgpr_write_b32 a27, s19
+; GCN-NEXT: v_accvgpr_write_b32 a26, s18
+; GCN-NEXT: v_accvgpr_write_b32 a25, s17
+; GCN-NEXT: v_accvgpr_write_b32 a24, s16
+; GCN-NEXT: v_accvgpr_write_b32 a23, s15
+; GCN-NEXT: v_accvgpr_write_b32 a22, s14
+; GCN-NEXT: v_accvgpr_write_b32 a21, s13
+; GCN-NEXT: v_accvgpr_write_b32 a20, s12
+; GCN-NEXT: v_accvgpr_write_b32 a19, s11
+; GCN-NEXT: v_accvgpr_write_b32 a18, s10
+; GCN-NEXT: v_accvgpr_write_b32 a17, s9
+; GCN-NEXT: v_accvgpr_write_b32 a16, s8
+; GCN-NEXT: v_mov_b32_e32 v8, s20
+; GCN-NEXT: v_mov_b32_e32 v9, s21
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[0:15], v[0:3], v[4:7], a[16:31] cbsz:1 abid:2 blgp:3
+; GCN-NEXT: v_mov_b32_e32 v10, s22
+; GCN-NEXT: v_mov_b32_e32 v11, s23
+; GCN-NEXT: v_mov_b32_e32 v0, s16
+; GCN-NEXT: v_mov_b32_e32 v1, s17
+; GCN-NEXT: v_mov_b32_e32 v2, s18
+; GCN-NEXT: v_mov_b32_e32 v3, s19
+; GCN-NEXT: global_store_dwordx4 v12, v[8:11], s[0:1] offset:48 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, v[0:3], s[0:1] offset:32 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s12
+; GCN-NEXT: v_mov_b32_e32 v1, s13
+; GCN-NEXT: v_mov_b32_e32 v2, s14
+; GCN-NEXT: v_mov_b32_e32 v3, s15
+; GCN-NEXT: global_store_dwordx4 v12, v[0:3], s[0:1] offset:16 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_nop 0
+; GCN-NEXT: v_mov_b32_e32 v0, s8
+; GCN-NEXT: v_mov_b32_e32 v1, s9
+; GCN-NEXT: v_mov_b32_e32 v2, s10
+; GCN-NEXT: v_mov_b32_e32 v3, s11
+; GCN-NEXT: global_store_dwordx4 v12, v[0:3], s[0:1] sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[8:11], s[0:1] offset:32 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[12:15], s[0:1] offset:48 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[0:3], s[0:1] sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: global_store_dwordx4 v12, a[4:7], s[0:1] offset:16 sc0 sc1
+; GCN-NEXT: s_waitcnt vmcnt(0)
+; GCN-NEXT: s_endpgm
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 1, i32 2, i32 3)
+ store volatile <16 x float> %arg2, ptr addrspace(1) %out
+ store volatile <16 x float> %result, ptr addrspace(1) %out
+ ret void
+}
+
+define amdgpu_kernel void @test_mfma_f32_32x32x16_bf16__vgprcd_mac(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, ptr addrspace(1) %out) #0 {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16__vgprcd_mac:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_dwordx8 s[24:31], s[4:5], 0x24
+; GCN-NEXT: s_load_dwordx16 s[8:23], s[4:5], 0x64
+; GCN-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0xa4
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: v_mov_b64_e32 v[0:1], s[24:25]
+; GCN-NEXT: v_mov_b64_e32 v[2:3], s[26:27]
+; GCN-NEXT: v_mov_b64_e32 v[4:5], s[28:29]
+; GCN-NEXT: v_accvgpr_write_b32 a0, s8
+; GCN-NEXT: v_mov_b64_e32 v[6:7], s[30:31]
+; GCN-NEXT: v_accvgpr_write_b32 a1, s9
+; GCN-NEXT: v_accvgpr_write_b32 a2, s10
+; GCN-NEXT: v_accvgpr_write_b32 a3, s11
+; GCN-NEXT: v_accvgpr_write_b32 a4, s12
+; GCN-NEXT: v_accvgpr_write_b32 a5, s13
+; GCN-NEXT: v_accvgpr_write_b32 a6, s14
+; GCN-NEXT: v_accvgpr_write_b32 a7, s15
+; GCN-NEXT: v_accvgpr_write_b32 a8, s16
+; GCN-NEXT: v_accvgpr_write_b32 a9, s17
+; GCN-NEXT: v_accvgpr_write_b32 a10, s18
+; GCN-NEXT: v_accvgpr_write_b32 a11, s19
+; GCN-NEXT: v_accvgpr_write_b32 a12, s20
+; GCN-NEXT: v_accvgpr_write_b32 a13, s21
+; GCN-NEXT: v_accvgpr_write_b32 a14, s22
+; GCN-NEXT: v_accvgpr_write_b32 a15, s23
+; GCN-NEXT: s_nop 1
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[0:15], v[0:3], v[4:7], a[0:15]
+; GCN-NEXT: v_mov_b32_e32 v0, 0
+; GCN-NEXT: s_nop 7
+; GCN-NEXT: s_nop 1
+; GCN-NEXT: global_store_dwordx4 v0, a[12:15], s[0:1] offset:48
+; GCN-NEXT: global_store_dwordx4 v0, a[8:11], s[0:1] offset:32
+; GCN-NEXT: global_store_dwordx4 v0, a[4:7], s[0:1] offset:16
+; GCN-NEXT: global_store_dwordx4 v0, a[0:3], s[0:1]
+; GCN-NEXT: s_endpgm
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 0, i32 0, i32 0)
+ store <16 x float> %result, ptr addrspace(1) %out
+ ret void
+}
+
+define amdgpu_kernel void @test_mfma_f32_32x32x16_bf16__vgprcd_mac_flags(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, ptr addrspace(1) %out) #0 {
+; GCN-LABEL: test_mfma_f32_32x32x16_bf16__vgprcd_mac_flags:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_dwordx8 s[24:31], s[4:5], 0x24
+; GCN-NEXT: s_load_dwordx16 s[8:23], s[4:5], 0x64
+; GCN-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0xa4
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: v_mov_b64_e32 v[0:1], s[24:25]
+; GCN-NEXT: v_mov_b64_e32 v[2:3], s[26:27]
+; GCN-NEXT: v_mov_b64_e32 v[4:5], s[28:29]
+; GCN-NEXT: v_accvgpr_write_b32 a0, s8
+; GCN-NEXT: v_mov_b64_e32 v[6:7], s[30:31]
+; GCN-NEXT: v_accvgpr_write_b32 a1, s9
+; GCN-NEXT: v_accvgpr_write_b32 a2, s10
+; GCN-NEXT: v_accvgpr_write_b32 a3, s11
+; GCN-NEXT: v_accvgpr_write_b32 a4, s12
+; GCN-NEXT: v_accvgpr_write_b32 a5, s13
+; GCN-NEXT: v_accvgpr_write_b32 a6, s14
+; GCN-NEXT: v_accvgpr_write_b32 a7, s15
+; GCN-NEXT: v_accvgpr_write_b32 a8, s16
+; GCN-NEXT: v_accvgpr_write_b32 a9, s17
+; GCN-NEXT: v_accvgpr_write_b32 a10, s18
+; GCN-NEXT: v_accvgpr_write_b32 a11, s19
+; GCN-NEXT: v_accvgpr_write_b32 a12, s20
+; GCN-NEXT: v_accvgpr_write_b32 a13, s21
+; GCN-NEXT: v_accvgpr_write_b32 a14, s22
+; GCN-NEXT: v_accvgpr_write_b32 a15, s23
+; GCN-NEXT: s_nop 1
+; GCN-NEXT: v_mfma_f32_32x32x16_bf16 a[0:15], v[0:3], v[4:7], a[0:15] cbsz:3 abid:2 blgp:1
+; GCN-NEXT: v_mov_b32_e32 v0, 0
+; GCN-NEXT: s_nop 7
+; GCN-NEXT: s_nop 1
+; GCN-NEXT: global_store_dwordx4 v0, a[12:15], s[0:1] offset:48
+; GCN-NEXT: global_store_dwordx4 v0, a[8:11], s[0:1] offset:32
+; GCN-NEXT: global_store_dwordx4 v0, a[4:7], s[0:1] offset:16
+; GCN-NEXT: global_store_dwordx4 v0, a[0:3], s[0:1]
+; GCN-NEXT: s_endpgm
+ %result = call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %arg0, <8 x bfloat> %arg1, <16 x float> %arg2, i32 3, i32 2, i32 1)
+ store <16 x float> %result, ptr addrspace(1) %out
+ ret void
+}
+
+attributes #0 = { "amdgpu-flat-work-group-size"="512,512" }
+attributes #1 = { "amdgpu-flat-work-group-size"="1,64" }
diff --git a/llvm/test/MC/AMDGPU/mai-gfx950.s b/llvm/test/MC/AMDGPU/mai-gfx950.s
index deba548b6ae8e1..1d4902e293bb10 100644
--- a/llvm/test/MC/AMDGPU/mai-gfx950.s
+++ b/llvm/test/MC/AMDGPU/mai-gfx950.s
@@ -47,11 +47,11 @@ v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] blgp:1
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] cbsz:3
-// GFX950: v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] abid:1 ; encoding: [0x00,0x88,0xd4,0xd3,0x00,0x01,0x02,0x1c]
+// GFX950: v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] abid:1 ; encoding: [0x00,0x88,0xd4,0xd3,0x00,0x01,0x02,0x1c]
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] abid:1
-// GFX950: v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] cbsz:3 abid:1 ; encoding: [0x00,0x8b,0xd4,0xd3,0x00,0x01,0x02,0x1c]
+// GFX950: v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] cbsz:3 abid:1 ; encoding: [0x00,0x8b,0xd4,0xd3,0x00,0x01,0x02,0x1c]
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] cbsz:3 abid:1
@@ -75,7 +75,7 @@ v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15]
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_32x32x16_f16 a[0:15], a[0:3], a[0:3], a[0:15]
-// GFX950: v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15] ; encoding: [0x00,0x00,0xd5,0xd3,0x00,0x01,0x02,0x04]
+// GFX950: v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15] ; encoding: [0x00,0x00,0xd5,0xd3,0x00,0x01,0x02,0x04]
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_32x32x16f16 v[0:15], v[0:3], v[0:3], v[0:15]
@@ -91,7 +91,7 @@ v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], 1.0
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_32x32x16_f16 a[0:15], a[0:3], a[0:3], 1.0
-// GFX950: v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15] blgp:5 ; encoding: [0x00,0x00,0xd5,0xd3,0x00,0x01,0x02,0xa4]
+// GFX950: v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15] blgp:5 ; encoding: [0x00,0x00,0xd5,0xd3,0x00,0x01,0x02,0xa4]
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15] blgp:5
@@ -110,3 +110,51 @@ v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15] abid:1
// GFX950: v_mfma_f32_32x32x16_f16 a[0:15], a[0:3], a[0:3], a[0:15] cbsz:3 abid:1 ; encoding: [0x00,0x8b,0xd5,0xd3,0x00,0x01,0x02,0x1c]
// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
v_mfma_f32_32x32x16_f16 a[0:15], a[0:3], a[0:3], a[0:15] cbsz:3 abid:1
+
+//===----------------------------------------------------------------------===//
+// v_mfma_f32_32x32x16_bf16
+//===----------------------------------------------------------------------===//
+
+// GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] ; encoding: [0x00,0x00,0xb7,0xd3,0x00,0x01,0x02,0x04]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15]
+
+// GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] ; encoding: [0x00,0x80,0xb7,0xd3,0x00,0x01,0x02,0x1c]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15]
+
+// GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] ; encoding: [0x00,0x00,0xb7,0xd3,0x00,0x01,0x02,0x04]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16bf16 v[0:15], v[0:3], v[0:3], v[0:15]
+
+// GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] ; encoding: [0x00,0x80,0xb7,0xd3,0x00,0x01,0x02,0x1c]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16bf16 a[0:15], a[0:3], a[0:3], a[0:15]
+
+// GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], 1.0 ; encoding: [0x00,0x00,0xb7,0xd3,0x00,0x01,0xca,0x03]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], 1.0
+
+// GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], 1.0 ; encoding: [0x00,0x80,0xb7,0xd3,0x00,0x01,0xca,0x1b]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], 1.0
+
+// GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] blgp:5 ; encoding: [0x00,0x00,0xb7,0xd3,0x00,0x01,0x02,0xa4]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] blgp:5
+
+// GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] blgp:2 ; encoding: [0x00,0x80,0xb7,0xd3,0x00,0x01,0x02,0x5c]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] blgp:2
+
+// GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] cbsz:3 ; encoding: [0x00,0x03,0xb7,0xd3,0x00,0x01,0x02,0x04]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] cbsz:3
+
+// GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] abid:1 ; encoding: [0x00,0x08,0xb7,0xd3,0x00,0x01,0x02,0x04]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] abid:1
+
+// GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] cbsz:3 abid:1 ; encoding: [0x00,0x8b,0xb7,0xd3,0x00,0x01,0x02,0x1c]
+// ERR: :[[@LINE+1]]:{{[0-9]+}}: error: instruction not supported on this GPU
+v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] cbsz:3 abid:1
diff --git a/llvm/test/MC/Disassembler/AMDGPU/gfx950_mai.txt b/llvm/test/MC/Disassembler/AMDGPU/gfx950_mai.txt
index 68b52672b445de..292f2a348df2ef 100644
--- a/llvm/test/MC/Disassembler/AMDGPU/gfx950_mai.txt
+++ b/llvm/test/MC/Disassembler/AMDGPU/gfx950_mai.txt
@@ -59,3 +59,30 @@
# GFX950: v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15] cbsz:3 ; encoding: [0x00,0x03,0xd5,0xd3,0x00,0x01,0x02,0x04]
0x00,0x03,0xd5,0xd3,0x00,0x01,0x02,0x04
+
+# GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], 1.0 ; encoding: [0x00,0x80,0xb7,0xd3,0x00,0x01,0xca,0x1b]
+0x00,0x80,0xb7,0xd3,0x00,0x01,0xca,0x1b
+
+# GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] ; encoding: [0x00,0x80,0xb7,0xd3,0x00,0x01,0x02,0x1c]
+0x00,0x80,0xb7,0xd3,0x00,0x01,0x02,0x1c
+
+# GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] blgp:2 ; encoding: [0x00,0x80,0xb7,0xd3,0x00,0x01,0x02,0x5c]
+0x00,0x80,0xb7,0xd3,0x00,0x01,0x02,0x5c
+
+# GFX950: v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] cbsz:3 abid:1 ; encoding: [0x00,0x8b,0xb7,0xd3,0x00,0x01,0x02,0x1c]
+0x00,0x8b,0xb7,0xd3,0x00,0x01,0x02,0x1c
+
+# GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], 1.0 ; encoding: [0x00,0x00,0xb7,0xd3,0x00,0x01,0xca,0x03]
+0x00,0x00,0xb7,0xd3,0x00,0x01,0xca,0x03
+
+# GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] ; encoding: [0x00,0x00,0xb7,0xd3,0x00,0x01,0x02,0x04]
+0x00,0x00,0xb7,0xd3,0x00,0x01,0x02,0x04
+
+# GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] abid:1 ; encoding: [0x00,0x08,0xb7,0xd3,0x00,0x01,0x02,0x04]
+0x00,0x08,0xb7,0xd3,0x00,0x01,0x02,0x04
+
+# GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] blgp:5 ; encoding: [0x00,0x00,0xb7,0xd3,0x00,0x01,0x02,0xa4]
+0x00,0x00,0xb7,0xd3,0x00,0x01,0x02,0xa4
+
+# GFX950: v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15] cbsz:3 ; encoding: [0x00,0x03,0xb7,0xd3,0x00,0x01,0x02,0x04]
+0x00,0x03,0xb7,0xd3,0x00,0x01,0x02,0x04
diff --git a/llvm/test/tools/llvm-mca/AMDGPU/gfx950.s b/llvm/test/tools/llvm-mca/AMDGPU/gfx950.s
index 66affe8f930afb..667fb7d78a87bd 100644
--- a/llvm/test/tools/llvm-mca/AMDGPU/gfx950.s
+++ b/llvm/test/tools/llvm-mca/AMDGPU/gfx950.s
@@ -1,18 +1,22 @@
# RUN: llvm-mca -mtriple=amdgcn -mcpu=gfx950 --timeline --iterations=1 --timeline-max-cycles=0 < %s | FileCheck %s
# CHECK: Iterations: 1
-# CHECK: Instructions: 4
-# CHECK: Total Cycles: 25
-# CHECK: Total uOps: 4
+# CHECK: Instructions: 6
+# CHECK: Total Cycles: 41
+# CHECK: Total uOps: 6
v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] blgp:1
v_mfma_f32_16x16x32_f16 a[0:3], v[0:3], v[0:3], a[4:7]
v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15]
v_mfma_f32_32x32x16_f16 a[0:15], a[0:3], a[0:3], a[0:15] blgp:2
+v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15]
+v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] blgp:2
# CHECK: [0] [1] [2] [3] [4] [5] [6] Instructions:
# CHECK-NEXT: - - - - - - 4.00 v_mfma_f32_16x16x32_f16 a[0:3], a[0:3], a[0:3], a[0:3] blgp:1
# CHECK-NEXT: - - - - - - 4.00 v_mfma_f32_16x16x32_f16 a[0:3], v[0:3], v[0:3], a[4:7]
# CHECK-NEXT: - - - - - - 8.00 v_mfma_f32_32x32x16_f16 v[0:15], v[0:3], v[0:3], v[0:15]
# CHECK-NEXT: - - - - - - 8.00 v_mfma_f32_32x32x16_f16 a[0:15], a[0:3], a[0:3], a[0:15] blgp:2
+# CHECK-NEXT: - - - - - - 8.00 v_mfma_f32_32x32x16_bf16 v[0:15], v[0:3], v[0:3], v[0:15]
+# CHECK-NEXT: - - - - - - 8.00 v_mfma_f32_32x32x16_bf16 a[0:15], a[0:3], a[0:3], a[0:15] blgp:2
More information about the llvm-branch-commits
mailing list