Fix incorrect usage of CUDACachingAllocator #46605
Conversation
|
cc @ptrblck |
zou3519
added
the
triaged
| _apply_single_inverse_helper<scalar_t>( | ||
| &self_data[i * self_mat_stride], &self_inv_data[i * self_inv_mat_stride], pivot, p_infos + i, n); | ||
| &self_data[i * self_mat_stride], &self_inv_data[i * self_inv_mat_stride], | ||
| reinterpret_cast<int*>(dataPtr.get()), p_infos + i, n); |
There was a problem hiding this comment.
this predates this pr, but, shouldn't need a reinterpret cast here, static cast from void* should be fine
There was a problem hiding this comment.
@ezyang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
|
need to merge |
|
It conflicted with my own changes. 😂 Resolved. Thanks! |
There was a problem hiding this comment.
@ezyang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
💊 CI failures summary and remediationsAs of commit 072391b (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker or post in the (internal) Dr. CI Users group. This comment has been revised 4 times. |
There was a problem hiding this comment.
@ezyang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
1 similar comment
Summary: We need an object to hold the ownership of allocated memory in the scope, instead of directly using the raw pointer. Pull Request resolved: pytorch#46605 Reviewed By: zou3519 Differential Revision: D24453548 Pulled By: ezyang fbshipit-source-id: d29e5a69afa6c0d9e519849910e04524667d0a26
We need an object to hold the ownership of allocated memory in the scope, instead of directly using the raw pointer.