replace device param with bounds_check_warning of inputs_to_device function #3831
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
|
hi, @842974287 @q10 @jiayisuse @aporialiao could you take a look? |
|
@q10 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
Hi @tiankongdeguiji , thanks for opening this PR. We will look into this and get back to you if we have questions, and if no questions, we will merge the PR. |
thx! |
liligwu
pushed a commit
to ROCm/FBGEMM
that referenced
this pull request
Mar 19, 2025
…nction (pytorch#3831) Summary: X-link: https://github.com/facebookresearch/FBGEMM/pull/930 When using fbgemm-gpu version 1.1.0 or later, the fx trace model will be bound to the device cuda:0. As a result, deploying the model to a CPU or a different device, such as cuda:1, is not possible. We can reproduce this with the case in pytorch#3830 we fix it, the `fbgemm_gpu_split_table_batched_embeddings_ops_inference_inputs_to_device` do not need `device(type='cuda', index=0))` as input. ``` ··· getitem_1 = _fx_trec_unwrap_kjt[0] getitem_2 = _fx_trec_unwrap_kjt[1]; _fx_trec_unwrap_kjt = None _tensor_constant0 = self._tensor_constant0 inputs_to_device = fbgemm_gpu_split_table_batched_embeddings_ops_inference_inputs_to_device(getitem_1, getitem_2, None, _tensor_constant0); getitem_1 = getitem_2 = _tensor_constant0 = None getitem_3 = inputs_to_device[0] getitem_4 = inputs_to_device[1] getitem_5 = inputs_to_device[2]; inputs_to_device = None _tensor_constant1 = self._tensor_constant1 _tensor_constant0_1 = self._tensor_constant0 bounds_check_indices = torch.ops.fbgemm.bounds_check_indices(_tensor_constant1, getitem_3, getitem_4, 1, _tensor_constant0_1, getitem_5); _tensor_constant1 = _tensor_constant0_1 = bounds_check_indices = None _tensor_constant2 = self._tensor_constant2 _tensor_constant3 = self._tensor_constant3 ··· ``` Pull Request resolved: pytorch#3831 Reviewed By: sryap Differential Revision: D71370666 Pulled By: q10 fbshipit-source-id: e8f65a534bf8235534ff861d1f135497f4660820
q10
pushed a commit
to q10/FBGEMM
that referenced
this pull request
Apr 10, 2025
…nction (pytorch#930) Summary: Pull Request resolved: facebookresearch/FBGEMM#930 When using fbgemm-gpu version 1.1.0 or later, the fx trace model will be bound to the device cuda:0. As a result, deploying the model to a CPU or a different device, such as cuda:1, is not possible. We can reproduce this with the case in pytorch#3830 we fix it, the `fbgemm_gpu_split_table_batched_embeddings_ops_inference_inputs_to_device` do not need `device(type='cuda', index=0))` as input. ``` ··· getitem_1 = _fx_trec_unwrap_kjt[0] getitem_2 = _fx_trec_unwrap_kjt[1]; _fx_trec_unwrap_kjt = None _tensor_constant0 = self._tensor_constant0 inputs_to_device = fbgemm_gpu_split_table_batched_embeddings_ops_inference_inputs_to_device(getitem_1, getitem_2, None, _tensor_constant0); getitem_1 = getitem_2 = _tensor_constant0 = None getitem_3 = inputs_to_device[0] getitem_4 = inputs_to_device[1] getitem_5 = inputs_to_device[2]; inputs_to_device = None _tensor_constant1 = self._tensor_constant1 _tensor_constant0_1 = self._tensor_constant0 bounds_check_indices = torch.ops.fbgemm.bounds_check_indices(_tensor_constant1, getitem_3, getitem_4, 1, _tensor_constant0_1, getitem_5); _tensor_constant1 = _tensor_constant0_1 = bounds_check_indices = None _tensor_constant2 = self._tensor_constant2 _tensor_constant3 = self._tensor_constant3 ··· ``` X-link: pytorch#3831 Reviewed By: sryap Differential Revision: D71370666 Pulled By: q10 fbshipit-source-id: e8f65a534bf8235534ff861d1f135497f4660820
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When using fbgemm-gpu version 1.1.0 or later, the fx trace model will be bound to the device cuda:0. As a result, deploying the model to a CPU or a different device, such as cuda:1, is not possible. We can reproduce this with the case in #3830
we fix it, the
fbgemm_gpu_split_table_batched_embeddings_ops_inference_inputs_to_devicedo not needdevice(type='cuda', index=0))as input.