0
I Use This!
Moderate Activity

Commits : Listings

Analyzed about 14 hours ago. based on code collected about 14 hours ago.
Aug 01, 2025 — Aug 01, 2026
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
Merge Aevum v0.3.78 workload-aware plan policy More... 8 days ago
Release Aevum v0.3.78 workload-aware plan policy More... 8 days ago
Merge Aevum v0.3.77 stable safety and plan audit More... 9 days ago
Release Aevum v0.3.77 stable safety and plan audit More... 9 days ago
Select FFT plans by throughput cost More... 12 days ago
Enable power-of-two FFT323161 lead caching More... 12 days ago
Optimize forced PFA9 type-4 requests More... 12 days ago
Make Aevum version audit release-independent More... 13 days ago
Fix native PFA OpenCL syntax matrix More... 13 days ago
Add automatic radix-3 and radix-9 native PFA More... 13 days ago
Disable incomplete Apple tailSquare bridge More... 15 days ago
Aevum v0.3.56: fuse Apple width pipelines More... 15 days ago
Aevum v0.3.54: fuse Apple GF61 tailSquare stages More... 15 days ago
Aevum v0.3.53: fix Apple arithmetic and queue pacing More... 15 days ago
Aevum v0.3.53: fix Apple arithmetic and queue pacing More... 15 days ago
Aevum v0.3.53: fix Apple arithmetic and queue pacing More... 15 days ago
Fix Aevum v0.3.4 macOS AppleClang build More... 19 days ago
Improve formatting in README.md More... 19 days ago
Delete README_AEVUM.md More... 19 days ago
Delete GITHUB_RELEASE_GUIDE.md More... 19 days ago
Delete MODIFICATIONS.md More... 19 days ago
Update README with project description More... 19 days ago
Remove project layout and publication recommendations More... 19 days ago
Create Aevum register engine from GPUOwl/PRPLL More... 19 days ago
Workaround AMD's Windows OpenCL compiler whining about always true if statements when WMUL=1. More... about 1 month ago
Restrict usage of WMUL=4 for WIDTH=1K and WMUL=2 for WIDTH=4K More... 2 months ago
Another attempt at fixing the GPU read failed error. More... 2 months ago
Detect nVidia compute capability to set HAS_PTX properly. Lets nVidia 1080 Gpus use asm by default. More... 2 months ago
Fixed yet another Windws treats long as 32-bits bug. More... 2 months ago
Reworked kernel calls. The ping-pong nature of !in_place and the favorable L2 cache locality of working on NTT data one data type at a time has been moved to a central location. Code is much more readable. Paves the way for further improvements in L2 cache locality work. More... 2 months ago