[mlir][AMDGPU] Support vector<2xbf16> packed atomic fadd #113929
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Member
|
@llvm/pr-subscribers-mlir @llvm/pr-subscribers-mlir-amdgpu Author: Krzysztof Drewniak (krzysz00) ChangesNow that we use LLVM's native bfloat types in the AMDGPU lowering, enable vector<2xbf16> for AMDGPU. Full diff: https://github.com/llvm/llvm-project/pull/113929.diff 2 Files Affected:
diff --git a/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td b/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
index cba35bbca1f83b..69745addfd748e 100644
--- a/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
+++ b/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
@@ -254,7 +254,7 @@ def AMDGPU_RawBufferAtomicCmpswapOp :
def AMDGPU_RawBufferAtomicFaddOp :
AMDGPU_Op<"raw_buffer_atomic_fadd", [AllElementTypesMatch<["value", "memref"]>,
AttrSizedOperandSegments]>,
- Arguments<(ins AnyTypeOf<[F32, VectorOfLengthAndType<[2], [F16]>]>:$value,
+ Arguments<(ins AnyTypeOf<[F32, VectorOfLengthAndType<[2], [F16, BF16]>]>:$value,
Arg<AnyMemRef, "buffer to operate on", [MemRead, MemWrite]>:$memref,
Variadic<I32>:$indices,
DefaultValuedAttr<BoolAttr, "true">:$boundsCheck,
diff --git a/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir b/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir
index 7fd5610a88913e..a9ea44925e9140 100644
--- a/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir
+++ b/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir
@@ -163,6 +163,17 @@ func.func @gpu_gcn_raw_buffer_atomic_fadd_v2f16(%value: vector<2xf16>, %buf: mem
func.return
}
+// CHECK-LABEL: func @gpu_gcn_raw_buffer_atomic_fadd_v2bf16
+func.func @gpu_gcn_raw_buffer_atomic_fadd_v2bf16(%value: vector<2xbf16>, %buf: memref<64xbf16>, %idx: i32) {
+ // CHECK: %[[numRecords:.*]] = llvm.mlir.constant(128 : i32)
+ // GFX9: %[[flags:.*]] = llvm.mlir.constant(159744 : i32)
+ // RDNA: %[[flags:.*]] = llvm.mlir.constant(822243328 : i32)
+ // CHECK: %[[resource:.*]] = rocdl.make.buffer.rsrc %{{.*}}, %{{.*}}, %[[numRecords]], %[[flags]]
+ // CHECK: rocdl.raw.ptr.buffer.atomic.fadd %{{.*}}, %[[resource]], %{{.*}}, %{{.*}}, %{{.*}} : vector<2xbf16>
+ amdgpu.raw_buffer_atomic_fadd {boundsCheck = true} %value -> %buf[%idx] : vector<2xbf16> -> memref<64xbf16>, i32
+ func.return
+}
+
// CHECK-LABEL: func @gpu_gcn_raw_buffer_atomic_fmax_f32
func.func @gpu_gcn_raw_buffer_atomic_fmax_f32(%value: f32, %buf: memref<64xf32>, %idx: i32) {
// CHECK: %[[numRecords:.*]] = llvm.mlir.constant(256 : i32)
|
Member
|
@llvm/pr-subscribers-backend-amdgpu Author: Krzysztof Drewniak (krzysz00) ChangesNow that we use LLVM's native bfloat types in the AMDGPU lowering, enable vector<2xbf16> for AMDGPU. Full diff: https://github.com/llvm/llvm-project/pull/113929.diff 2 Files Affected:
diff --git a/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td b/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
index cba35bbca1f83b..69745addfd748e 100644
--- a/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
+++ b/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
@@ -254,7 +254,7 @@ def AMDGPU_RawBufferAtomicCmpswapOp :
def AMDGPU_RawBufferAtomicFaddOp :
AMDGPU_Op<"raw_buffer_atomic_fadd", [AllElementTypesMatch<["value", "memref"]>,
AttrSizedOperandSegments]>,
- Arguments<(ins AnyTypeOf<[F32, VectorOfLengthAndType<[2], [F16]>]>:$value,
+ Arguments<(ins AnyTypeOf<[F32, VectorOfLengthAndType<[2], [F16, BF16]>]>:$value,
Arg<AnyMemRef, "buffer to operate on", [MemRead, MemWrite]>:$memref,
Variadic<I32>:$indices,
DefaultValuedAttr<BoolAttr, "true">:$boundsCheck,
diff --git a/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir b/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir
index 7fd5610a88913e..a9ea44925e9140 100644
--- a/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir
+++ b/mlir/test/Conversion/AMDGPUToROCDL/amdgpu-to-rocdl.mlir
@@ -163,6 +163,17 @@ func.func @gpu_gcn_raw_buffer_atomic_fadd_v2f16(%value: vector<2xf16>, %buf: mem
func.return
}
+// CHECK-LABEL: func @gpu_gcn_raw_buffer_atomic_fadd_v2bf16
+func.func @gpu_gcn_raw_buffer_atomic_fadd_v2bf16(%value: vector<2xbf16>, %buf: memref<64xbf16>, %idx: i32) {
+ // CHECK: %[[numRecords:.*]] = llvm.mlir.constant(128 : i32)
+ // GFX9: %[[flags:.*]] = llvm.mlir.constant(159744 : i32)
+ // RDNA: %[[flags:.*]] = llvm.mlir.constant(822243328 : i32)
+ // CHECK: %[[resource:.*]] = rocdl.make.buffer.rsrc %{{.*}}, %{{.*}}, %[[numRecords]], %[[flags]]
+ // CHECK: rocdl.raw.ptr.buffer.atomic.fadd %{{.*}}, %[[resource]], %{{.*}}, %{{.*}}, %{{.*}} : vector<2xbf16>
+ amdgpu.raw_buffer_atomic_fadd {boundsCheck = true} %value -> %buf[%idx] : vector<2xbf16> -> memref<64xbf16>, i32
+ func.return
+}
+
// CHECK-LABEL: func @gpu_gcn_raw_buffer_atomic_fmax_f32
func.func @gpu_gcn_raw_buffer_atomic_fmax_f32(%value: f32, %buf: memref<64xf32>, %idx: i32) {
// CHECK: %[[numRecords:.*]] = llvm.mlir.constant(256 : i32)
|
Now that we use LLVM's native bfloat types in the AMDGPU lowering, enable vector<2xbf16> for AMDGPU.
krzysz00
force-pushed
the
packed-atomic-bf16
branch
from
dc58053 to
4fd02f3
Compare
kuhar
approved these changes
Oct 30, 2024
smallp-o-p
pushed a commit
to smallp-o-p/llvm-project
that referenced
this pull request
Nov 3, 2024
Now that we use LLVM's native bfloat types in the AMDGPU lowering, enable vector<2xbf16> for AMDGPU.
NoumanAmir657
pushed a commit
to NoumanAmir657/llvm-project
that referenced
this pull request
Nov 4, 2024
Now that we use LLVM's native bfloat types in the AMDGPU lowering, enable vector<2xbf16> for AMDGPU.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Now that we use LLVM's native bfloat types in the AMDGPU lowering, enable vector<2xbf16> for AMDGPU.