[Mlir-commits] [mlir] Restore: [mlir][ROCDL] Stop setting amdgpu-implicitarg-num-bytes (PR #79129)
llvmlistbot at llvm.org
llvmlistbot at llvm.org
Tue Jan 23 04:52:34 PST 2024
llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-mlir-llvm
Author: Saiyedul Islam (saiislam)
<details>
<summary>Changes</summary>
This patch restores PR#<!-- -->78498
---
Full diff: https://github.com/llvm/llvm-project/pull/79129.diff
2 Files Affected:
- (modified) mlir/lib/Target/LLVMIR/Dialect/ROCDL/ROCDLToLLVMIRTranslation.cpp (+1-1)
- (modified) mlir/test/Target/LLVMIR/rocdl.mlir (+1-1)
``````````diff
diff --git a/mlir/lib/Target/LLVMIR/Dialect/ROCDL/ROCDLToLLVMIRTranslation.cpp b/mlir/lib/Target/LLVMIR/Dialect/ROCDL/ROCDLToLLVMIRTranslation.cpp
index a230ead7c188314..0cbb3da79d151e0 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/ROCDL/ROCDLToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/ROCDL/ROCDLToLLVMIRTranslation.cpp
@@ -99,7 +99,7 @@ class ROCDLDialectLLVMIRTranslationInterface
if (!llvmFunc->hasFnAttribute("amdgpu-flat-work-group-size")) {
llvmFunc->addFnAttr("amdgpu-flat-work-group-size", "1,256");
}
- llvmFunc->addFnAttr("amdgpu-implicitarg-num-bytes", "256");
+
}
// Override flat-work-group-size
// TODO: update clients to rocdl.flat_work_group_size instead,
diff --git a/mlir/test/Target/LLVMIR/rocdl.mlir b/mlir/test/Target/LLVMIR/rocdl.mlir
index f831d7bba864c8f..3c9c70711ae2304 100644
--- a/mlir/test/Target/LLVMIR/rocdl.mlir
+++ b/mlir/test/Target/LLVMIR/rocdl.mlir
@@ -489,7 +489,7 @@ llvm.func @rocdl_8bit_floats(%source: i32, %stoch: i32) -> i32 {
llvm.return %source5 : i32
}
-// CHECK-DAG: attributes #[[$KERNEL_ATTRS]] = { "amdgpu-flat-work-group-size"="1,256" "amdgpu-implicitarg-num-bytes"="256" }
+// CHECK-DAG: attributes #[[$KERNEL_ATTRS]] = { "amdgpu-flat-work-group-size"="1,256" }
// CHECK-DAG: attributes #[[$KERNEL_WORKGROUP_ATTRS]] = { "amdgpu-flat-work-group-size"="1,1024"
// CHECK-DAG: attributes #[[$KNOWN_BLOCK_SIZE_ATTRS]] = { "amdgpu-flat-work-group-size"="128,128"
// CHECK-DAG: ![[$RANGE]] = !{i32 0, i32 64}
``````````
</details>
https://github.com/llvm/llvm-project/pull/79129
More information about the Mlir-commits
mailing list