review.tizen.org Git - platform/upstream/pytorch.git/commit

projects / platform / upstream / pytorch.git / commit

author	Jongsoo Park <jongsoo@fb.com>
	Thu, 31 Jan 2019 06:46:07 +0000 (22:46 -0800)
committer	Facebook Github Bot <facebook-github-bot@users.noreply.github.com>
	Thu, 31 Jan 2019 06:49:37 +0000 (22:49 -0800)
commit	db121375e7c07d43f48cfbc2a57e39e5f16e7d3b
tree	86139ff3947757f7a550038db3bbe7cfa20a8572	tree \| snapshot
parent	26200ebf56e7727a60f461bce9f964674dd5c9fb	commit \| diff

more careful use of inline/template function in perfkernels (#15388)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15388

This is another pass to make perfkernels code safer from illegal instruction error.
Removed dependency to c10/util/Logging.h
We're err on the safer side at the expense of some verbosity.

Reviewed By: dskhudia

Differential Revision: D13502902

fbshipit-source-id: 4f833115df885c5b4f8c1ca83b9badea1553f944

14 files changed:

caffe2/perfkernels/adagrad.cc		diff \| blob \| history
caffe2/perfkernels/adagrad.h		diff \| blob \| history
caffe2/perfkernels/adagrad_avx.cc		diff \| blob \| history
caffe2/perfkernels/embedding_lookup.cc		diff \| blob \| history
caffe2/perfkernels/embedding_lookup.h		diff \| blob \| history
caffe2/perfkernels/embedding_lookup_avx2.cc		diff \| blob \| history
caffe2/perfkernels/embedding_lookup_fused_8bit_rowwise_avx2.cc		diff \| blob \| history
caffe2/perfkernels/fused_8bit_rowwise_embedding_lookup.cc		diff \| blob \| history
caffe2/perfkernels/fused_8bit_rowwise_embedding_lookup.h		diff \| blob \| history
caffe2/perfkernels/hp_emblookup_codegen.py		diff \| blob \| history
caffe2/perfkernels/math_cpu_avx2.cc		diff \| blob \| history
caffe2/perfkernels/typed_axpy_avx.cc		diff \| blob \| history
caffe2/perfkernels/typed_axpy_avx2.cc		diff \| blob \| history
caffe2/python/lengths_reducer_fused_8bit_rowwise_ops_test.py		diff \| blob \| history

Domain: Machine Learning / ML Framework;

RSS Atom