[MLIR][NVVM] Fix lowering logic after fddf7b05 #170545
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 Author: Sohaib Iftikhar (sohaibiftikhar) ChangesWithout this mapping fails when there is no result specified. Full diff: https://github.com/llvm/llvm-project/pull/170545.diff 1 Files Affected:
diff --git a/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td b/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
index a96d65d3fcacd..cb83ec23bc76e 100644
--- a/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
+++ b/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
@@ -948,8 +948,7 @@ def NVVM_MBarrierArriveExpectTxOp : NVVM_PTXBuilder_Op<"mbarrier.arrive.expect_t
auto [id, args] = NVVM::MBarrierArriveExpectTxOp::getIntrinsicIDAndArgs(
*op, moduleTranslation, builder);
- int addrSpace = llvm::cast<LLVMPointerType>(op.getAddr().getType()).getAddressSpace();
- if (addrSpace != NVVM::NVVMMemorySpace::SharedCluster)
+ if (op.getNumResults() > 0)
$res = createIntrinsicCall(builder, id, args);
else
createIntrinsicCall(builder, id, args);
@@ -985,9 +984,7 @@ def NVVM_MBarrierArriveDropExpectTxOp : NVVM_Op<"mbarrier.arrive_drop.expect_tx"
string llvmBuilder = [{
auto [id, args] = NVVM::MBarrierArriveDropExpectTxOp::getIntrinsicIDAndArgs(
*op, moduleTranslation, builder);
-
- int addrSpace = llvm::cast<LLVMPointerType>(op.getAddr().getType()).getAddressSpace();
- if (addrSpace != NVVM::NVVMMemorySpace::SharedCluster)
+ if (op.getNumResults() > 0)
$res = createIntrinsicCall(builder, id, args);
else
createIntrinsicCall(builder, id, args);
|
Member
|
@llvm/pr-subscribers-mlir-llvm Author: Sohaib Iftikhar (sohaibiftikhar) ChangesWithout this mapping fails when there is no result specified. Full diff: https://github.com/llvm/llvm-project/pull/170545.diff 1 Files Affected:
diff --git a/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td b/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
index a96d65d3fcacd..cb83ec23bc76e 100644
--- a/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
+++ b/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
@@ -948,8 +948,7 @@ def NVVM_MBarrierArriveExpectTxOp : NVVM_PTXBuilder_Op<"mbarrier.arrive.expect_t
auto [id, args] = NVVM::MBarrierArriveExpectTxOp::getIntrinsicIDAndArgs(
*op, moduleTranslation, builder);
- int addrSpace = llvm::cast<LLVMPointerType>(op.getAddr().getType()).getAddressSpace();
- if (addrSpace != NVVM::NVVMMemorySpace::SharedCluster)
+ if (op.getNumResults() > 0)
$res = createIntrinsicCall(builder, id, args);
else
createIntrinsicCall(builder, id, args);
@@ -985,9 +984,7 @@ def NVVM_MBarrierArriveDropExpectTxOp : NVVM_Op<"mbarrier.arrive_drop.expect_tx"
string llvmBuilder = [{
auto [id, args] = NVVM::MBarrierArriveDropExpectTxOp::getIntrinsicIDAndArgs(
*op, moduleTranslation, builder);
-
- int addrSpace = llvm::cast<LLVMPointerType>(op.getAddr().getType()).getAddressSpace();
- if (addrSpace != NVVM::NVVMMemorySpace::SharedCluster)
+ if (op.getNumResults() > 0)
$res = createIntrinsicCall(builder, id, args);
else
createIntrinsicCall(builder, id, args);
|
durga4github
approved these changes
Dec 4, 2025
Contributor
durga4github
left a comment
durga4github
left a comment
There was a problem hiding this comment.
Thank you for the fix, LGTM.
I will try this out on my end too.
kcloudy0717
pushed a commit
to kcloudy0717/llvm-project
that referenced
this pull request
Dec 4, 2025
Without this mapping fails when there is no result specified. See: llvm#169922 (comment) To reproduce error on `main`: ```bash mkdir -p build && cd build cmake -G Ninja ../llvm \ -DLLVM_ENABLE_PROJECTS=mlir \ -DLLVM_TARGETS_TO_BUILD="host;NVPTX" \ -DMLIR_ENABLE_CUDA_RUNNER=ON \ -DMLIR_RUN_CUDA_TENSOR_CORE_TESTS=ON \ -DMLIR_RUN_CUDA_SM90_TESTS=ON \ -DMLIR_GPU_COMPILATION_TEST_FORMAT=fatbin \ -DMLIR_INCLUDE_INTEGRATION_TESTS=ON \ -DLLVM_ENABLE_ASSERTIONS=ON \ -DLLVM_INSTALL_UTILS=ON \ -DCMAKE_BUILD_TYPE=Release ninja bin/mlir-opt ../mlir/test/Integration/GPU/CUDA/sm90/tma_load_128x64_swizzle128b.mlir -gpu-lower-to-nvvm-pipeline="cubin-chip=sm_90 cubin-features=+ptx80 opt-level=3" ```
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.
Without this mapping fails when there is no result specified.
See: #169922 (comment)
To reproduce error on
main: