alexmsettle 00b8c7e63b New feature for issue #85575. (#86514) ... Introduced RECORD_OUTPUTS() macro that goes with RECORD_FUNCTION(). It is used to capture the output tensors from a kernel launch. The tensors automatically get passed to the profiler using record_function methods. This allows the profiler to track the tensors that flow into and out of each op. Fixes #85575 cc @robieta @chaekit @aaronenyeshi @ngimel @nbcsm @guotuofeng @guyang3532 @gaoteng-git @tiffzhaofb Pull Request resolved: https://github.com/pytorch/pytorch/pull/86514 Approved by: https://github.com/robieta		2022-10-24 20:02:56 +00:00
..
cuda	New feature for issue #85575. (#86514)	2022-10-24 20:02:56 +00:00
fuser	attempted fix for nvrtc with lovelace (#87611)	2022-10-24 18:41:38 +00:00
onednn	[BE] Extend linter to detect DOS newlines (#86973)	2022-10-15 00:20:42 +00:00