Skip to content

[mlir][gpu] Add host-supports-nvptx to test gpu-lower-to-nvvm-pipeline #80155

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from

Conversation

grypp
Copy link
Member

@grypp grypp commented Jan 31, 2024

PR #78556 added a new mlir test with gpu-lower-to-nvvm-pipeline that checks the generated PTX. However, it causes a problem on host without cuda support.

This PR adds REQUIRES: host-supports-nvptx.

PR llvm#78556 added a new mlir test with gpu-lower-to-nvvm-pipeline that checks the generated PTX. However, it causes a problem on host without cuda support.

This PR adds `REQUIRES: host-supports-nvptx`.
@llvmbot
Copy link
Member

llvmbot commented Jan 31, 2024

@llvm/pr-subscribers-mlir

Author: Guray Ozen (grypp)

Changes

PR #78556 added a new mlir test with gpu-lower-to-nvvm-pipeline that checks the generated PTX. However, it causes a problem on host without cuda support.

This PR adds REQUIRES: host-supports-nvptx.


Full diff: https://github.com/llvm/llvm-project/pull/80155.diff

1 Files Affected:

  • (modified) mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir (+1)
diff --git a/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir b/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir
index fde99ab105d68..70587592a3efa 100644
--- a/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir
+++ b/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir
@@ -1,3 +1,4 @@
+// REQUIRES: host-supports-nvptx
 // RUN: mlir-opt %s \
 // RUN:  | mlir-opt -gpu-lower-to-nvvm-pipeline="cubin-format=isa" \ 
 // RUN:    -split-input-file | FileCheck %s

@llvmbot
Copy link
Member

llvmbot commented Jan 31, 2024

@llvm/pr-subscribers-mlir-gpu

Author: Guray Ozen (grypp)

Changes

PR #78556 added a new mlir test with gpu-lower-to-nvvm-pipeline that checks the generated PTX. However, it causes a problem on host without cuda support.

This PR adds REQUIRES: host-supports-nvptx.


Full diff: https://github.com/llvm/llvm-project/pull/80155.diff

1 Files Affected:

  • (modified) mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir (+1)
diff --git a/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir b/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir
index fde99ab105d68..70587592a3efa 100644
--- a/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir
+++ b/mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir
@@ -1,3 +1,4 @@
+// REQUIRES: host-supports-nvptx
 // RUN: mlir-opt %s \
 // RUN:  | mlir-opt -gpu-lower-to-nvvm-pipeline="cubin-format=isa" \ 
 // RUN:    -split-input-file | FileCheck %s

@grypp
Copy link
Member Author

grypp commented Feb 12, 2024

This was a fix for the original PR, but it's reverted.
I'll put a complete PR, so I abandon this.

@grypp grypp closed this Feb 12, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants