Add a specialized agent for NVIDIA Nsys profiling that handles:
- Profile data collection using framework scripts
- Statistical analysis of kernel timing and memory transfers
- Timeline analysis for GPU-CPU overlap efficiency
- Comparative analysis between different configurations
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
- Add "Policy 不能为 None (CRITICAL)" section
- Document that sparse_policy must always be at least FullAttentionPolicy
- Document warmup phase as the only exception where kvcache_manager can be None
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
- Add .claude/rules/gpu-monitor.md requiring gpu-monitor agent for all GPU memory monitoring tasks
- Update CLAUDE.md rules index with reference to new rule
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add a custom agent for continuous GPU monitoring during benchmarks:
- Track GPU utilization, memory usage, and temperature
- Support multi-GPU and configurable sampling intervals
- Generate summary statistics when stopped
Useful for debugging memory leaks and profiling long-running tasks.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add test_xattn_kernels.py demonstrating flat_group_gemm_fuse_reshape
and softmax_fuse_block_sum Triton kernels with structured data
- Update testing.md with new test code style guidelines
- Update xattn.py and xattn_bsa.py with improvements
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add a new Claude command that executes task_plan.md refactoring with:
- GPU isolation via --gpu <id> parameter (required)
- Optional --no-interrupt mode for autonomous execution
- Progress tracking via progress.md and findings.md
- Strict CUDA_VISIBLE_DEVICES enforcement
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Create docs/sparse_policy_implementation_guide.md with comprehensive guide
- Rewrite .claude/rules/sparse-policy.md with mandatory base class requirements
- Add new doc reference to CLAUDE.md documentation index
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Remove cross-layer pipeline from OffloadEngine (saves ~1GB GPU memory for long sequences)
- Delete layer_k/v_buffer_a/b double buffers
- Remove start_decode_pipeline, get_decode_layer_kv, end_decode_pipeline methods
- Remove pipeline state tracking variables
- Simplify decode to use ring buffer pipeline only (more efficient for long sequences)
- Rename compute_chunked_attention → compute_chunked_prefill for clarity
- Add mandatory needle test requirements: --enable-offload --input-len 32768
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add planning files (task_plan.md, findings.md, progress.md) to .gitignore
- Remove existing planning files from git index (keep local)
- Update planning-with-files rule with git management policy
These temporary session files should not be version controlled.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Remove unused claude-flow hooks, permissions, and daemon settings.
Add disabled MCP servers list for claude-flow related servers.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add .claude/settings.json to enable claude-flow MCP in all worktrees.
This configuration includes:
- SessionStart hook to auto-start claude-flow daemon
- Auto-approval for claude-flow MCP tools and CLI commands
- Basic claude-flow settings
Co-Authored-By: Claude <noreply@anthropic.com>
