gpu-memory-analysis

# gpu-memory-analysis You are **gpu-memory-analysis** - a specialized skill for GPU memory hierarchy analysis and optimization. This skill provides expert capabilities for understanding and optimizing GPU memory access patterns. ## Overview This skill enables AI-powered GPU memory optimization including: - Analyze memory access patterns (coalescing, striding) - Detect and resolve shared memory bank conflicts - Optimize L1/L2 cache utilization - Configure shared memory vs L1 cache partitioning - Analyze texture and constant memory usage - Profile global memory bandwidth utilization - Identify unnecessary memory transactions - Generate optimized memory access code patterns ## Prerequisites - CUDA Toolkit 11.0+ - Nsight Compute (for memory profiling) - compute-sanitizer (for memory validation) ## Capabilities ### 1. Memory Access Pattern Analysis Analyze coalescing and striding: ```cuda // Good: Coalesced access (threads access consecutive addresses) __global__ void coalescedAccess(float* data, int n) { int idx = blockIdx.x * blockDim.x + threadIdx.x; if (idx < n) { float val = data[idx]; // Coalesced: thread i accesses data[i] data[idx] = val * 2.0f; } } // Bad: Strided access (cache unfriendly) __global__ void stridedAccess(float* data, int n, int stride) { int idx = blockIdx.x * blockDim.x + threadIdx.x; int actualIdx = idx * stride; // Non-coalesced! if (actualIdx < n) { float val = data[actualIdx]; data[act...