llehtahw

看起来rpyc的进程在prefill过程中崩掉了，可能是异常被吞了，也可能是个segfault

Hi @leiwen83, Were you able to use the fast tokenizer in LightLLM? If yes, would you like to open a pull request ? 😄

1. ```py import re text = re.sub(r'\n+', '\n', text) ``` 2.推完一个问题，然后推下一个问题 2.趁router没反应过来，把多个问题提上去

@xxm1668 你可以尝试编译一下这个代码 `gcc a.c -o a -lcuda` ```c #include extern void cuInit(); int main() { printf("%x\n", &cuInit); return 0; } ``` 看看编译是否报错，运行是否报错。 如果编译运行正常，可以看一下`ldd ./a`的输出结果，也许可以定位到正确的`libcuda.so`的位置。这里是我的输出 ``` linux-vdso.so.1 => (0x00007ffe235ef000) libcuda.so.1 =>...

> `gcc a.c -o a -lcuda` > > ```c > #include > extern void cuInit(); > int main() > { > printf("%x\n", &cuInit); > return 0; > } > ```...

> https://github.com/ModelTC/lightllm/issues/16#issuecomment-1670362288 May I have your error log please? - If `/usr/bin/ld: cannot find -lcuda: No such file or directory` was reported - check if there's any `libcuda.so`(without any suffix...

> `RuntimeError: Triton Error [CUDA]: invalid argument` This seems to be another problem. Could you please open a new issue on it, and post your system, cuda driver version, `pip...