Skip to content
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

[Issue]: Compiling with amdclang fails due to -Werror #708

Open
tbennun opened this issue Jan 21, 2025 · 2 comments
Open

[Issue]: Compiling with amdclang fails due to -Werror #708

tbennun opened this issue Jan 21, 2025 · 2 comments

Comments

@tbennun
Copy link

tbennun commented Jan 21, 2025

Problem Description

When building PyTorch from source on an AMD system with AMD's clang (and I imagine clang in general) as the compiler, ROCm triton (as a third party of a third party library) fails to build due to -Werror always being enabled (see https://github.com/ROCm/triton/blob/main_perf/CMakeLists.txt#L110). This prohibits building the project with clang.

Thanks in advance.

Operating System

Any

CPU

Any

GPU

Any AMD Instinct GPU

ROCm Version

ROCm 6.2.1

ROCm Component

llvm-project

Steps to Reproduce

No response

(Optional for Linux users) Output of /opt/rocm/bin/rocminfo --support

No response

Additional Information

No response

@harkgill-amd
Copy link

Hi @tbennun, an internal ticket has been created to investigate the failure when compiling with -Werror.

@lucbruni-amd
Copy link

lucbruni-amd commented Jan 29, 2025

Hi @tbennun, could you provide your build logs and clang version to ensure I'm on the same page and hit the same stage of the build that you do? Also, any environment variables and build steps for Pytorch would be greatly appreciated.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants