trunk/1021ddaa95dbaec8775442af63d998591b50a353: Key sparse Triton tuning tables by the current accelerator (#194176)
- PyTorch 近 90 天出现 2635 次
- PyTorch 近 90 天第 2595 次发版
- 上一次:同一天稍早 · trunk/41003927e76c4e4254fd45fa83f2ade160d2617a
发生了什么
Related to this RFC: #194355 Summary _get_device_name() supplies the device component of the key under which _operation_device_version_data stores tuned Triton parameters. It was a two-branch chain over torch.cuda and torch.xpu returning "" for anything else, and it took no arguments -- so the key named whatever the process was configured for, never the device the inputs were actually on. An accelerator outside that…
摘要按规则整理自下方来源原文