Add the RdcSmiDiagnostic module

Provides a RdcSmiDiagnostic module, which will call rocm_smi_lib.

It will support following diagnostics: Get GPU Topology, Check GPU
parameters and check processes running on the GPUs.

The grpc client and server side diagnostics function is added.

The diag module is added to the rdci.

Change-Id: I10a0cf3c20556a61373ab686f82cae75acaa40dd
这个提交包含在:
Bill(Shuzhou) Liu
2021-06-21 15:35:06 -04:00
父节点 5a4bf97327
当前提交 76ccf58008
修改 22 个文件,包含 1330 行新增13 行删除
+8
查看文件
@@ -120,6 +120,14 @@ class RdcAPIServiceImpl final : public ::rdc::RdcAPI::Service {
const ::rdc::Empty* request,
::rdc::RemoveAllJobResponse* reply) override;
::grpc::Status DiagnosticRun(::grpc::ServerContext* context,
const ::rdc::DiagnosticRunRequest* request,
::rdc::DiagnosticRunResponse* reply) override;
::grpc::Status DiagnosticTestCaseRun(::grpc::ServerContext* context,
const ::rdc::DiagnosticTestCaseRunRequest* request,
::rdc::DiagnosticTestCaseRunResponse* reply) override;
private:
bool copy_gpu_usage_info(const rdc_gpu_usage_info_t& src,
::rdc::GpuUsageInfo* target);