When I ran a model with SGLang on 4 B60 GPUs with tensor parallel (TP), the unitrace raised segfault errors. The command I ran is unitrace -d -h <my command to run model with SGLang>. So, does unitrace support such scenarios? If yes, how should I use it?
When I ran a model with SGLang on 4 B60 GPUs with tensor parallel (TP), the unitrace raised
segfaulterrors. The command I ran isunitrace -d -h <my command to run model with SGLang>. So, does unitrace support such scenarios? If yes, how should I use it?