-
Happens for example with rhel7 kernel which doesn't have offload support.
Libreswan 4.15 - completely stable.
Libreswan 5.0 - causes time to time adaptor to hang for 5 minutes.
We should real…
bleve updated
4 weeks ago
-
per:
- #1697
in 4.x and 5.0 the code probes the kernel to see if it supports nic_offload while adding an interface; which effectively means during startup (when `ipsec listen` is run).
The co…
-
I have the same problem like this [cannot use unsloth](https://github.com/unslothai/unsloth/issues/820), but when I run the code below it is still got the same error :
`os.environ['CUDA_VISIBLE_DEVIC…
-
For some reason I cannot just `Unblock::new()` a TcpStream to offload it to a thread pool. I need it because I want to target Windows platform.
-
Hi, Thanks for your great work! When I fine-tune InternLM-XComposer2(unfreeze the proj and the whole LLM, freeze vit). In order to avoid OOM, I use zero3 and offload the optimizer to CPU(by modifying …
-
I have a 3090 with 24GB VRAM. i tried to geneate with 2 images and i get this error
Traceback (most recent call last):
File "F:\OmniGen\venv\lib\site-packages\gradio\queueing.py", line 624, in p…
-
您好！我在64卡上外推72b模型时一直遇到OOM的问题，是不是multi_node.yaml中配置错了？
multi_node.yaml
`debug: false
deepspeed_config:
deepspeed_config_file: utils/accelerate_configs/zero3_offload.json
deepspeed_multinode_lau…
-
When using fw3 and software offload on, I got around 680Mbps in speedtest test.
When using fw3 and software offload off, I got around 260Mbps in speedtest test.
When using fw4 and software offlo…
-
We have to figure out the best approach.
One approach is using our (to be implemented) MLIR backend (#4319) and possibly using an OpenMP MLIR dialect that AMD compilers might be able to offload.
…
-
Union is currently implemented in Java and the operator could break up one fused Velox's task into two. We'd find a way to offload the operator to Velox so Velox could get a bigger plan for applying m…