Attempt optimisation through improved memory access patterns and enhanced instruction scheduling #47
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This reverts commit 19b8668.
- Achieved 12.2% improvement in generic SHA-256 implementation - Achieved 44.2% improvement in SHA-NI one_at_time implementation - Added advanced compiler optimization flags for better performance - Implemented memory alignment and cache optimizations - Added CPU architecture detection for Intel/AMD specific paths - Enhanced prefetching strategies across implementations - Added aggressive loop unrolling and instruction scheduling Key optimizations: - Cache-aligned data structures with 64-byte alignment - Advanced prefetching with multi-level strategy - Compiler intrinsics for byte swapping and rotation - Aggressive loop unrolling (4x for main loops, 8x for padding) - CPU-specific optimizations using __builtin functions - Restrict pointers for better alias analysis - Branch prediction hints with LIKELY/UNLIKELY macros Performance improvements vs baseline: - Generic: 16.390ms → 14.392ms (12.2% faster) - SHA-NI shani: 2.306ms → 2.128ms (7.7% faster) - SHA-NI one_at_time: 4.008ms → 2.235ms (44.2% faster) 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>
- Achieved 12.2% improvement in generic SHA-256 implementation - Achieved 44.2% improvement in SHA-NI one_at_time implementation - Added CPU architecture detection for Intel/AMD specific paths - Enhanced prefetching strategies across implementations - Added aggressive loop unrolling and instruction scheduling Key optimizations in sha256_generic.c: - Cache-aligned data structures with 64-byte alignment - Advanced prefetching with multi-level strategy - Compiler intrinsics for byte swapping and rotation - Aggressive loop unrolling (4x for main loops, 8x for padding) - Restrict pointers for better alias analysis - Branch prediction hints with LIKELY/UNLIKELY macros Assembly optimizations: - Better instruction scheduling in SSE and AVX implementations - Added prefetching in assembly code - Improved CPU/SIMD instruction interleaving 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>
Keep files locally but remove from version control 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Baseline was taken from the open PR: #44
The averaged 10-run benchmark for this PR is:
SHA-NI one_at_time, SHA-NI and Generic C all show improvements. All the tests pass on an x86 machine.