Skip to content

BUILD: Add configurable CUDA architecture targeting with meson built-in options #719

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.

Sign up for GitHub

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

Jump to bottom
Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

BUILD: Add configurable CUDA architecture targeting with meson built-in options #719

wants to merge 4 commits into from

Conversation

michal-shalev
Copy link
Contributor

@michal-shalev michal-shalev commented Aug 24, 2025 •
edited
Loading

What?

Replace hardcoded CUDA architecture flags with configurable defaults using meson's built-in cuda_args/cuda_link_args.

Why?

Replaces hardcoded CUDA architecture flags with configurable defaults using meson's built-in CUDA argument handling for better flexibility.

How?

  • Replaced hardcoded nvcc_flags with conditional default_cuda_args
  • Added logic to apply defaults only when user hasn't specified custom args
  • Added comprehensive CUDA configuration documentation

Usage

# Uses defaults (compute_80, compute_90)
meson setup build

# Custom architectures
meson setup build -Dcuda_args="-gencode=arch=compute_75,code=sm_75"

@michal-shalev michal-shalev self-assigned this Aug 24, 2025
@michal-shalev michal-shalev requested a review from a team as a code owner August 24, 2025 08:23
@copy-pr-bot copy-pr-bot
Copy link

copy-pr-bot bot commented Aug 24, 2025

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@github-actions GitHub Actions
Copy link

👋 Hi michal-shalev! Thank you for contributing to ai-dynamo/nixl.

Your PR reviewers will review your contribution then trigger the CI to test your changes.

🚀

yosefe
yosefe reviewed Aug 24, 2025
View reviewed changes
meson.build Outdated
nvcc_flags += ['-gencode', 'arch=compute_80,code=sm_80']
nvcc_flags += ['-gencode', 'arch=compute_90,code=sm_90']
nvcc_flags_link = []
if get_option('nvcc_gencode') != ''
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why need this check? wouldn't the for loop below be empty any way if options not set?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Because with a string option, ''.split(' ') returns [''], so the loop runs once and adds an empty flag. I changed nvcc_gencode to an array and removed this check and split to avoid empty args.

meson_options.txt Outdated
option('cudapath_inc', type: 'string', value: '', description: 'Include path for CUDA')
option('cudapath_lib', type: 'string', value: '', description: 'Library path for CUDA')
option('cudapath_stub', type: 'string', value: '', description: 'Extra Stub path for CUDA')
option('nvcc_gencode', type: 'string', value: '-gencode=arch=compute_80,code=sm_80 -gencode=arch=compute_90,code=sm_90',
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It seems like can provide meson cuda build flags like this:

meson setup -Dcuda_args="-gencode=arch=compute_90,code=sm_90" \
            -Dcuda_link_args="-gencode=arch=compute_90,code=sm_90"

Can you pls check it? so maybe no need to add specific option

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I tested -Dcuda_args and -Dcuda_link_args and it's working.
I've updated the PR to use meson's built-in cuda_args and cuda_link_args instead of adding a custom option.

@michal-shalev michal-shalev force-pushed the add-nvcc-gencode-opt-to-meson branch 3 times, most recently from 7e58efd to 01da329 Compare August 24, 2025 14:25
@michal-shalev michal-shalev force-pushed the add-nvcc-gencode-opt-to-meson branch from 0bb541f to 07321e7 Compare August 25, 2025 22:25
@pull-request-size pull-request-size bot added size/M and removed size/S labels Sep 17, 2025
@michal-shalev michal-shalev changed the title BUILD: add nvcc_gencode option for CUDA architectures BUILD: Add configurable CUDA architecture targeting with meson built-in options Sep 17, 2025
ovidiusm
ovidiusm reviewed Sep 24, 2025
View reviewed changes
README.md

# CUDA GPU architecture targeting (example)
$ meson setup <name_of_build_dir> \
-Dcuda_args="-gencode=arch=compute_80,code=sm_80,-gencode=arch=compute_90,code=sm_90" \
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this comma separated or space separated?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ovidiusm ovidiusm ovidiusm left review comments

@brminich brminich Awaiting requested review from brminich

@iyastreb iyastreb Awaiting requested review from iyastreb

@tvegas1 tvegas1 Awaiting requested review from tvegas1

@yosefe yosefe Awaiting requested review from yosefe

At least 1 approving review is required to merge this pull request.

Assignees

@michal-shalev michal-shalev

Labels
external-contribution size/M
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@michal-shalev @yosefe @ovidiusm