OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

Guard test_lapack_empty with has_magma. (#9936)

Browse Source 
Summary:
CUDA lapack functions generally don't work unless has_magma is true.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/9936

Differential Revision: D9028579

Pulled By: gchanan

fbshipit-source-id: 9b77e3b05253fd49bcabf604d0924ffa0e116055
This commit is contained in:
Gregory Chanan 2018-07-27 10:08:08 -07:00 committed by Facebook Github Bot
parent bf32ea8094
commit c0bacc6284
1 changed files with 5 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
test/test_torch.py
View File

@ -6394,6 +6394,11 @@ class TestTorch(TestCase):
devices = ['cpu'] if not torch.cuda.is_available() else ['cpu', 'cuda']
for device in devices:
# need to init cuda to check has_magma
empty = torch.randn((0, 0), device=device)
if device == 'cuda' and not torch.cuda.has_magma:
continue
def fn(torchfn, *args):
return torchfn(*tuple(torch.randn(shape, device=device) if isinstance(shape, tuple) else shape
for shape in args))