Skip to content

Conversation

@dzhulgakov
Copy link
Collaborator

Spotted while debugging some problem

Before

>>> torch.empty(10**15)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
RuntimeError: [enforce fail at CPUAllocator.cpp:56] posix_memalign(&data, gAlignment, nbytes) == 0. 12 vs 0

After

>>> torch.empty(10**15)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
RuntimeError: [enforce fail at CPUAllocator.cpp:65] . DefaultCPUAllocator: can't allocate memory: you tried to allocate 4000000000000000 bytes. Error code 12 (Cannot allocate memory)

@pytorchbot pytorchbot added the module: internals Related to internal abstractions in c10 and ATen label May 17, 2019
Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@dzhulgakov has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

"DefaultCPUAllocator: not enough memory: you tried to allocate %dGB. Buy new RAM!",
nbytes / 1073741824);
"DefaultCPUAllocator: not enough memory: you tried to allocate ",
nbytes,
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wouldn't it be simpler for the user to print the amount in MB or GB?

@colesbury
Copy link
Member

But how will users know now that they need to Buy new RAM? R.I.P. Samsung, SK Hynix, Micron

@facebook-github-bot
Copy link
Contributor

@dzhulgakov merged this pull request in d9dcfac.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Merged module: internals Related to internal abstractions in c10 and ATen

Projects

None yet

Development

Successfully merging this pull request may close these issues.

8 participants