trunk/312ef3ee1cd45fbee99f2e4e8f6035f05ceeebed: Add BF16x9 precision mode for CUDA FP32 matmul (#195301)
- PyTorch 近 90 天出现 414 次
- PyTorch 近 90 天第 411 次发版
- 上一次：同一天稍早 · trunk/129e6f1d40be5e428a664436d97803f70aa747c0: [inductor] Enable MXFP8 nested-reduction test on XPU (#194864)
发生了什么
Human Note Cublas has a cool bf16x9 mode for fp32 gemms; in theory it should be more accurate than tf32; ill keep prodding to get some intersting data showing this SOL went crazy (although in a good way) and made sure this doesnt break things hence the largish amount of changes .. indeed these flags are insidiuous Most of the changes were in the pt2 stack where we need to properly guard/ensure caches are working cor…
摘要按规则整理自下方来源原文