add lineinfo for all cuda device compilations

[wenchen76]

Purpose

To resolve #22958, add -lineinfo in NVCC flags

Test Plan

WIP

Test Result

(Optional) Documentation Update

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[gemini-code-assist]

Code Review

This pull request adds the -lineinfo NVCC flag to enable line number information for debugging and profiling CUDA kernels. The change is correct, but I've suggested making it conditional based on the CMake build type to avoid impacting release builds. This aligns with best practices for managing build configurations.

[gemini-code-assist]

While adding -lineinfo is useful for debugging and profiling, adding it unconditionally to all build types, including Release, is not standard practice. This can increase binary size and may have a minor performance impact on production builds.

It's better to enable this flag for debug-related build types only (e.g., Debug, RelWithDebInfo, or any build type except Release). This gives developers control over the trade-off between debuggability and performance/size.

if(VLLM_GPU_LANG STREQUAL "CUDA" AND NOT (CMAKE_BUILD_TYPE STREQUAL "Release"))
  list(APPEND VLLM_GPU_FLAGS "-lineinfo")
endif()

[youkaichao]

Suggested change

	# Add line info for debugging.
	# Add line info for cuda core dump.