-
```
I have Intel 82599 10GE Nic with 4 hardware queues enabled and Debian 7 Wheezy.
cat /proc/interrupts |grep eth4
74: 808231153 0 0 0 PCI-MSI-edge eth4-TxRx-0
…
-
### 🐛 Describe the bug
I train and inference a classifier using autocast. Result is different accross diffenent GPUs (same .venv, code and data).
The result on A100 is much superior than on RTX A…
-
In
https://github.com/ExaESM-WP4/dask-jobqueue-configs/blob/8318f739d97abc5064151121a2fd95c3120f4e6d/nesh/nesh-dask-jobqueue-config.yaml#L13
we might decide to go for `cores` higher than the req…
-
### 🐛 Describe the bug
Background of the issue:
DeepSpeed depends a lot on param.data = other.data for ZeRO3 parameter offload. And ZeRO3 also depends on register a hook on param AccumulateGrad ob…
-
**Is your enhancement proposal related to a problem? Please describe.**
I'm using an RPi4B since almost 2 years now with OMR and 3 4G router (no wired WAN) and willing to upgrade my setup
RPi4…
-
### 🐛 Describe the bug
I am having trouble working with torch.topk. Its behaviour is inconsistent. Here is the relevant code.
```
import torch
errors = torch.tensor([-0.1944, -0.1944, -0.1945,…
-
| | |
| --- | --- |
| Bugzilla Link | [39027](https://llvm.org/bz39027) |
| Version | 6.0 |
| OS | Linux |
| Attachments | [C code and generated assembly](https://user-images.githubusercontent.…
-
### 🐛 Describe the bug
When training a large model on H100s, we are seeing an illegal memory access error when using AdamW `fused=True`. I suspect the root cause may be related to https://github.co…
-
```
What steps will reproduce the problem?
1) cd netmap/examples/
2) ./pkt-gen -i vale-2:b -frx
3) ./pkt-gen -i vale-2:a -ftx
What is the expected output? What do you see instead?
I expect traffic to…
-
```
What steps will reproduce the problem?
1) cd netmap/examples/
2) ./pkt-gen -i vale-2:b -frx
3) ./pkt-gen -i vale-2:a -ftx
What is the expected output? What do you see instead?
I expect traffic to…