Don't have Telegram yet? Try it now!
https://hgpu.org/?p=9892
SIMD Divergence Optimization through Intra-Warp Compaction