-
### 🐛 Describe the bug
I may have discovered a bug when I registered the corresponding ProcessGroup development.
I found that the destructor of the ProcessGroup subclass object was not called when…
-
### Your current environment
```text
The output of `python collect_env.py`
Collecting environment information...
PyTorch version: 2.3.1+cu121
Is debug build: False
CUDA used to build PyTorch: 12…
-
I follow the guide to installed all the related and build SNP kernel successfully. And my host also installed the SNP kernel.
But once I launch guest OS with SEV_SNP, it will failed as below:
Any he…
-
### 🐛 Describe the bug
Hello, I met some problems recently. I have three third party shared libraries which is compiled with C++17. And I want to extend operators in these libraries to Pytorch. The…
-
### 🐛 Describe the bug
I'm using @torch.compile() for the initialization of a class.
Hopefully it will speed it up.
sys.audit doesn't seem to be supported, but I don't explicitly call it in my cod…
-
After applying the sriov manifest files on the cluster deployed using ZTP playbooks, we found that vfio-pci driver module was not loaded on the worker nodes and hence CU workload VM deployment failed.
-
## Issue description
Setting `timeout` in `torch.distributed.init_process_group` doesn't take actual effect
## Code example
I set timeout with
```python
torch.distributed.init_process_grou…
-
Installed snapshot this week on a nanoHD and support for beamforming and mu-mimo seems to be missing just by checking iw list or grep hostapd confs. The ubiquiti device specsheet for the nanoHD does s…
-
### 🐛 Describe the bug
In PyTorch CI of https://github.com/pytorch/pytorch/pull/130915, in which we want to add OpInfo for `_convert_weight_to_int4pack`, we encounter `RuntimeError: CUDA error: inval…
-
Starting this issue thread to document some of the issues, or maybe more correctly limitations, we've found with GSF (Generic Sensor Format) files. As background, we've used GSF as our preferred proce…