Skip to content

Fix silently incorrect CPU matmul for a stride-zero broadcast batch #5889

Fix silently incorrect CPU matmul for a stride-zero broadcast batch

Fix silently incorrect CPU matmul for a stride-zero broadcast batch #5889

Workflow file for this run

name: CI / cuda
on:
workflow_dispatch:
pull_request:
jobs:
test-cuda:
concurrency:
group: ${{ github.workflow }}-test-cuda-${{ github.head_ref || github.run_id }}
cancel-in-progress: true
runs-on:
group: aws-g5-4xlarge-cache
container:
image: nvidia/cuda:13.0.2-cudnn-devel-ubuntu24.04
if: ${{ github.event.pull_request.head.repo.full_name == github.event.pull_request.base.repo.full_name }}
permissions:
contents: read
# This is used to complete the identity challenge
# with sigstore/fulcio when running outside of PRs.
id-token: write
security-events: write
env:
CUDA_COMPUTE_CAP: 86
steps:
- name: Checkout repository
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
- name: Install dependencies
run: apt update && apt install curl build-essential libssl-dev protobuf-compiler pkg-config -y
- name: Install Rust Stable
uses: dtolnay/rust-toolchain@29eef336d9b2848a0b548edc03f92a220660cdb8 # stable
- uses: Swatinem/rust-cache@e18b497796c12c097a38f9edb9d0641fb99eee32 # v2
- name: Test (cuda)
run: |
cargo test --workspace --lib --bins --tests --features cuda
cargo test --workspace --doc --features cuda
test-cutile:
concurrency:
group: ${{ github.workflow }}-test-cutile-${{ github.head_ref || github.run_id }}
cancel-in-progress: true
runs-on:
group: aws-g5-4xlarge-cache
container:
image: nvidia/cuda:13.2.1-devel-ubuntu24.04
if: ${{ github.event.pull_request.head.repo.full_name == github.event.pull_request.base.repo.full_name }}
permissions:
contents: read
env:
CUDA_COMPUTE_CAP: 86
CUDA_TOOLKIT_PATH: /usr/local/cuda
steps:
- name: Checkout repository
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
- name: Install dependencies
run: apt update && apt install curl build-essential clang libclang-dev libssl-dev protobuf-compiler pkg-config -y
- name: Install Rust Stable
uses: dtolnay/rust-toolchain@29eef336d9b2848a0b548edc03f92a220660cdb8 # stable
- uses: Swatinem/rust-cache@e18b497796c12c097a38f9edb9d0641fb99eee32 # v2
- name: Verify tileiras
run: test -x "${CUDA_TOOLKIT_PATH}/bin/tileiras"
- name: Test cuTile interop
run: |
cargo test --package candle-core --features cutile --lib
cargo test --package candle-core --features cutile --test cutile_tests
cargo test --package candle-core --features cutile --doc
- name: Test cuTile routed MoE
run: |
cargo test --package candle-nn --features cutile --lib
cargo test --package candle-nn --features cutile --test cutile_moe
- name: Check cuTile example
run: cargo check --package candle-core --features cutile --example cutile