projects / platform / upstream / tvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1dcf8a1) | patch
[Codegen][cuda-fp16] fallback to fp32 simulation when cuda arch < sm53 (#4268)
authorYizhi Liu <liuyizhi@apache.org>
Sun, 10 Nov 2019 06:16:34 +0000 (22:16 -0800)
committerTianqi Chen <tqchen@users.noreply.github.com>
Sun, 10 Nov 2019 06:16:34 +0000 (22:16 -0800)
commit801cf0e8d3427b12d017ba7bcf0a17b9da2d2408
tree23e9ba483e71f47174db089cb458c4f058c7b095tree | snapshot
parent1dcf8a16ee3a93dff5ffc1ad1a66892eda03ef13commit | diff
[Codegen][cuda-fp16] fallback to fp32 simulation when cuda arch < sm53 (#4268)
src/codegen/codegen_cuda.cc diff | blob | history
src/codegen/literal/cuda_half_t.h [new file with mode: 0644] blob
Domain: Machine Learning / ML Framework;