Skip to content

Refresh bench tooling and Claude profiles#1

Open
brettheap wants to merge 3 commits into
mainfrom
agent/bench-refresh-2026-07
Open

Refresh bench tooling and Claude profiles#1
brettheap wants to merge 3 commits into
mainfrom
agent/bench-refresh-2026-07

Conversation

@brettheap

@brettheap brettheap commented Jul 13, 2026

Copy link
Copy Markdown
Contributor

Summary

  • mount portable Claude account profiles where applicable
  • refresh the bench-specific runtime and administration toolchain
  • keep the bench aligned with the synchronized workBenches parent

Validation

  • git diff --check
  • changed shell scripts: bash -n
  • changed Dockerfile: docker buildx build --check

Summary by Sourcery

Align bench tooling and devcontainer configuration with updated ROCm WSL setup and portable Claude profiles.

New Features:

  • Mount local Claude profiles into the development container for portable account configuration.

Bug Fixes:

  • Update AMD ROCm WSL configuration to use a dedicated ROCM_HSA_PRELOAD variable instead of LD_PRELOAD to improve environment handling.

Enhancements:

  • Refresh auxiliary bench scripts and Dockerfile layer to stay in sync with the parent workBenches configuration.

Copilot AI review requested due to automatic review settings July 13, 2026 00:14
@brettheap

Copy link
Copy Markdown
Contributor Author

@codex review

@sourcery-ai

sourcery-ai Bot commented Jul 13, 2026

Copy link
Copy Markdown
Reviewer's guide (collapsed on small PRs)

Reviewer's Guide

Updates ROCm WSL configuration, mounts portable Claude profiles into the devcontainer, and refreshes bench tooling and container/runtime definitions to stay aligned with the parent workBenches setup.

Flow diagram for updated ROCm WSL configuration preload variable

flowchart TD
  start[run configure-amd-rocm-wsl.sh] --> detectROCM[detect ROCM_ROOT]
  detectROCM --> writeYAML[append ROCm config YAML]
  writeYAML --> setPreload[set ROCM_HSA_PRELOAD in YAML]
Loading

File-Level Changes

Change Details Files
Adjust ROCm-on-WSL environment configuration to use a dedicated preload variable rather than LD_PRELOAD.
  • Replace use of LD_PRELOAD with ROCM_HSA_PRELOAD for the HSA runtime library in the generated YAML configuration.
  • Keep existing HIP/HSA-related environment variables and library path configuration intact while narrowing the preload behavior.
scripts/configure-amd-rocm-wsl.sh
Mount Claude profile configuration into the devcontainer to support portable account profiles.
  • Add a bind mount for the ~/.claude-profiles directory into the devcontainer user home.
  • Retain existing mounts for Claude, Codex, PI, SonarQube, and other tooling configs to keep bench behavior consistent.
.devcontainer/docker-compose.yml
Refresh devcontainer and Docker layer/tooling definitions to align with the synchronized workBenches parent.
  • Update devcontainer metadata and configuration (e.g., features, settings, postCreate/postStart hooks) as needed for the refreshed bench runtime.
  • Adjust Dockerfile.layer2 to match the updated base/tooling stack from the parent bench setup.
  • Revise the AMD ROCm WSL check script to be compatible with the updated configuration and tooling expectations.
.devcontainer/devcontainer.json
Dockerfile.layer2
scripts/check-amd-rocm-wsl

Tips and commands

Interacting with Sourcery

  • Trigger a new review: Comment @sourcery-ai review on the pull request.
  • Continue discussions: Reply directly to Sourcery's review comments.
  • Generate a GitHub issue from a review comment: Ask Sourcery to create an
    issue from a review comment by replying to it. You can also reply to a
    review comment with @sourcery-ai issue to create an issue from it.
  • Generate a pull request title: Write @sourcery-ai anywhere in the pull
    request title to generate a title at any time. You can also comment
    @sourcery-ai title on the pull request to (re-)generate the title at any time.
  • Generate a pull request summary: Write @sourcery-ai summary anywhere in
    the pull request body to generate a PR summary at any time exactly where you
    want it. You can also comment @sourcery-ai summary on the pull request to
    (re-)generate the summary at any time.
  • Generate reviewer's guide: Comment @sourcery-ai guide on the pull
    request to (re-)generate the reviewer's guide at any time.
  • Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
    pull request to resolve all Sourcery comments. Useful if you've already
    addressed all the comments and don't want to see them anymore.
  • Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
    request to dismiss all existing Sourcery reviews. Especially useful if you
    want to start fresh with a new review - don't forget to comment
    @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

  • Enable or disable review features such as the Sourcery-generated pull request
    summary, the reviewer's guide, and others.
  • Change the review language.
  • Add, remove or edit custom review instructions.
  • Adjust other review settings.

Getting Help

@sourcery-ai sourcery-ai Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Hey - I've left some high level feedback:

  • For the new ROCM_HSA_PRELOAD environment variable, consider adding a brief comment or helper script note explaining why LD_PRELOAD was avoided so future updates don’t inadvertently revert this change.
  • The new .claude-profiles volume is mounted with :cached; if these profiles are not expected to be modified from inside the container, you may want to mount them read-only to avoid accidental edits.
Prompt for AI Agents
Please address the comments from this code review:

## Overall Comments
- For the new ROCM_HSA_PRELOAD environment variable, consider adding a brief comment or helper script note explaining why LD_PRELOAD was avoided so future updates don’t inadvertently revert this change.
- The new .claude-profiles volume is mounted with :cached; if these profiles are not expected to be modified from inside the container, you may want to mount them read-only to avoid accidental edits.

Sourcery is free for open source - if you like our reviews please consider sharing them ✨
Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Pull request overview

Updates pyBench’s devcontainer/tooling setup to align with upstream workBenches changes, including portable Claude profile mounting and refreshed ROCm WSL handling.

Changes:

  • Switch WSL ROCm preload handling from LD_PRELOAD to a dedicated ROCM_HSA_PRELOAD variable in the generated compose override and related diagnostics.
  • Refresh Python web stack tooling in Dockerfile.layer2 (use fastapi[standard]).
  • Mount ~/.claude-profiles into the devcontainer (compose + devcontainer mounts) for portable Claude account/profile configuration.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 2 comments.

Show a summary per file
File Description
scripts/configure-amd-rocm-wsl.sh Generates ROCm WSL docker-compose override; now exports ROCM_HSA_PRELOAD instead of setting LD_PRELOAD globally.
scripts/check-amd-rocm-wsl Prints ROCM_HSA_PRELOAD and warns if LD_PRELOAD is globally set.
Dockerfile.layer2 Updates pip-installed Python tooling to use fastapi[standard].
.devcontainer/docker-compose.yml Adds bind mount for ~/.claude-profiles into the container.
.devcontainer/devcontainer.json Adds corresponding additionalMounts entry for ~/.claude-profiles.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment on lines 57 to 60
HSA_ENABLE_DXG_DETECTION: "1"
HSA_ENABLE_SDMA: "0"
LD_PRELOAD: ${ROCM_ROOT}/lib/libhsa-runtime64.so.1
ROCM_HSA_PRELOAD: ${ROCM_ROOT}/lib/libhsa-runtime64.so.1
LD_LIBRARY_PATH: ${ROCM_ROOT}/lib:/usr/lib/x86_64-linux-gnu:/usr/lib
Comment thread Dockerfile.layer2
Comment on lines 26 to +28
RUN python3 -m pip install --break-system-packages \
pipx poetry black flake8 mypy pytest pytest-cov coverage pysonar ipython jupyter \
django flask fastapi uvicorn pandas numpy
django flask 'fastapi[standard]' uvicorn pandas numpy

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: acf0f50a15

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

HSA_ENABLE_DXG_DETECTION: "1"
HSA_ENABLE_SDMA: "0"
LD_PRELOAD: ${ROCM_ROOT}/lib/libhsa-runtime64.so.1
ROCM_HSA_PRELOAD: ${ROCM_ROOT}/lib/libhsa-runtime64.so.1

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P2 Badge Keep preloading HSA for ROCm containers

On WSL AMD ROCm hosts, this generated override now only exports ROCM_HSA_PRELOAD, but neither Docker Compose nor the dynamic linker consumes that variable; only LD_PRELOAD actually causes libhsa-runtime64.so.1 to be preloaded. Since there is no wrapper in the repo that maps ROCM_HSA_PRELOAD back to LD_PRELOAD, ROCm/Paddle commands launched normally in the devcontainer lose the preload that the previous generated config provided and can fail unless every invocation is manually prefixed.

Useful? React with 👍 / 👎.

"source=${localEnv:HOME}/.agents,target=/home/${localEnv:USER}/.agents,type=bind,consistency=cached",
"source=${localEnv:HOME}/.claude,target=/home/${localEnv:USER}/.claude,type=bind,consistency=cached",
"source=${localEnv:HOME}/.claude.json,target=/home/${localEnv:USER}/.claude.json,type=bind,consistency=cached",
"source=${localEnv:HOME}/.claude-profiles,target=/home/${localEnv:USER}/.claude-profiles,type=bind,consistency=cached",

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P2 Badge Avoid requiring optional Claude profile paths

For hosts that have the existing credential paths but have not created ~/.claude-profiles, this new Dev Containers bind mount makes startup fail because entries in mounts use Docker --mount semantics and the bind source must already exist. Since the profiles are only useful for users who opted into portable Claude profiles, create the host directory before startup or make this mount conditional rather than requiring every devcontainer user to have it.

Useful? React with 👍 / 👎.

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants