1. 25 Feb, 2025 1 commit
    • Lei Wang's avatar
      [Example] Implement TileLang Native Sparse Attention Kernel (#121) · 3cbf8cbc
      Lei Wang authored
      * Add DeepSeek MLA decode example with Flash Attention implementation
      
      * Add GEMM SplitK and StreamK example implementations
      
      This commit introduces two new example scripts demonstrating advanced GEMM (matrix multiplication) techniques:
      - `example_tilelang_gemm_splitk.py`: Implements a Split-K GEMM kernel using TileLang
      - `example_tilelang_gemm_streamk.py`: Implements a Stream-K GEMM kernel using TileLang
      
      Both examples showcase different parallel computation strategies for matrix multiplication, with comprehensive testing using PyTorch reference implementations.
      
      * Refactor GEMM SplitK and StreamK example implementations
      
      Clean up and improve code formatting for the SplitK and StreamK GEMM example scripts:
      - Remove unused import (Profiler) in splitk example
      - Simplify line breaks and improve code readability
      - Standardize indentation and remove unnecessary whitespace
      - Optimize atomic add and copy operations for better clarity
      
      * Add block sparse attention benchmarks for multiple libraries
      
      This commit introduces comprehensive block sparse attention benchmarks for different libraries:
      - TileLang block sparse FMHA implementation
      - Triton block sparse FMHA implementation
      - PyTorch reference block sparse FMHA implementation
      - FlashAttention dense FMHA reference implementation
      
      The benchmarks include:
      - Configurable benchmark parameters (batch size, heads, sequence length, etc.)
      - Sparse mask generation using top-k and threshold methods
      - Performance measurement for different sparse attention configurations
      - Utility functions for mask generation and benchmarking
      
      * Refactor block sparse attention benchmarks with code style improvements
      
      - Add Ruff linter ignore comments to benchmark files
      - Improve code formatting and line breaks
      - Remove unused imports
      - Standardize print statement formatting
      - Enhance code readability across multiple library benchmarks
      
      * lint fix
      
      * Add CUDA atomic operations for BFLOAT16 and update function naming
      
      - Implement AtomicAdd functions for BFLOAT16 and BFLOAT16x2 in CUDA common header
      - Rename existing atomic add functions to use PascalCase (atomicAdd -> AtomicAdd)
      - Add a new __pack_nv_bfloat162 function for packing BFLOAT16 values
      - Update kernel and language customization to use new function names
      - Add return type annotations in profiler module
      
      * lint fix
      
      * Add example for Group Query Attention (GQA) forward pass using Flash Attention in TileLang
      
      This commit introduces a new example script `example_gqa_fwd_bshd.py` that demonstrates:
      - Group Query Attention (GQA) implementation
      - Flash Attention forward pass
      - Performance benchmarking
      - Configurable parameters for batch, heads, sequence length, and dimension
      - Autotuning support
      - Reference implementation comparison
      
      * Refactor IR lowering pipeline into modular phases
      
      This commit introduces a new module `phase.py` to modularize the IR lowering process by splitting the complex lowering pipeline into two distinct phases:
      - `LowerAndLegalize`: Handles initial IR legalization and transformation
      - `OptimizeForTarget`: Applies target-specific optimizations
      
      The changes simplify the lowering logic in multiple files by extracting the transformation steps into reusable functions, improving code readability and maintainability.
      
      * lintfix
      
      * nas kernel
      
      * Enhance Native Sparse Attention Examples with Code Improvements and Parameter Updates
      
      - Updated example_tilelang_nsa.py and example_triton_nsa.py with code formatting and style improvements
      - Increased default number of heads and selected blocks in TileLang NSA example
      - Added Ruff linter ignore comments to reference.py
      - Standardized function signatures and improved code readability across NSA implementations
      
      * Add utility math functions for integer operations
      
      - Implement `next_power_of_2()` to calculate the next power of 2 for an integer
      - Add `cdiv()` function for ceiling division of integers
      3cbf8cbc