5 Commits

Include renames

Author	SHA1	Message	Date
wyj	e1ec480669	HIP: accelerate PSF accumulation and restore parallel producers ... Cooperate across 32 lanes per PSF and use two completion-protected staging slots with complete batch timing. Restore coarse OpenMP event production while serializing shared GPU submissions and direct fallback boundaries. Add bounded benchmarks, streaming and renderer regressions, and preserve validation evidence and ownership documentation.	2026-09-06 21:38:53 -04:00
wyj	a18ebfbf1a	HIP: record direct atomic PSF timings	2026-09-05 22:13:14 -04:00
wyj	7a80086f63	Feat: add initial HIP PSF backend	2026-09-05 04:23:57 -04:00
wyj	e48c490337	Doc: update GPU PSF acceleration status	2026-09-05 03:47:42 -04:00
wyj	dd8cf29dac	Doc: add GPU PSF acceleration plan	2026-08-31 15:26:47 -04:00