Ma Mingfei
Ma Mingfei
Created new label to further track this task more easily: we can either track from [cpu](https://github.com/sgl-project/sglang/issues?q=state%3Aopen%20label%3A%22cpu%22) or [intel](https://github.com/sgl-project/sglang/issues?q=state%3Aopen%20label%3A%22intel%22) at the moment.
> Note that the following code, to check for linux kernel support, does not work in chromium sandbox? > > ``` > #if (defined(__i386__) || defined(__x86_64__)) && defined(__linux__) > #define...
@fbarchard I think `enabling AMX` is not in the scope of pytorch/cpuinfo. The thing is we are using some detection method from onednn in pytorch which is not aligned with...
> @mingfeima I will suggest we add enabling AMX to pytorch/cpuinfo. If an app is not using OneDNN, it will be helpful to all those apps and user base. Tying...
@YangQun1 3ks~
re-trigger CI.
@YangQun1 @yanbing-j can we take a look the CI failures? seems to be able to reproduce.
Any idea why the AMD CIs failed? is this related to this PR?
* to enable amx build, you will need **gcc11 or above**, and amx optimizations will be compiled if the machine has amx-int8 support * for decoding scenarios (when M=1 in...