Function: create_config

Atlas 350 Accelerator Card

√

Atlas A3 training product/Atlas A3 inference product

√

Atlas A2 training product/Atlas A2 inference product

√

Atlas training product

√

Atlas inference product

√

Atlas 200I/500 A2 inference product

√

device_list

List, device ID list. Set it based on the device ID in the actual environment.

aicore_metrics

Int, aclprofAicoreMetrics.

aicore_events

Int, AI Core event, which is set to 0.

data_type_config

Int. Select from the following aclproftype values in logical OR format (for example, ACL_PROF_ACL_API|ACL_PROF_AICORE_METRICS) as the parameter value of data_type_config. Each value indicates a type of profile data.

• ACL_PROF_ACL_API: collects profile data of APIs, including the synchronous/asynchronous memory copy latencies between the host and devices.
• ACL_PROF_TASK_TIME: collects operator delivery and execution duration data, as well as basic operator information, to provide more comprehensive performance analysis data.
• ACL_PROF_TASK_TIME_L0: collects operator delivery and execution duration data. Compared with ACL_PROF_TASK_TIME, ACL_PROF_TASK_TIME_L0 does not collect basic operator information, so the performance overhead during collection is smaller, and this enables more accurate collection of statistics on time duration data.
• ACL_PROF_AICORE_METRICS: collects AI Core metrics. This macro must be included in the OR logic for aicore_metrics to take effect.
• ACL_PROF_TASK_MEMORY: specifies whether to collect the memory usage of CANN operators. Only data of GE operators is collected.
• ACL_PROF_AICPU: collects traces of AI CPU tasks, including the start and end of each task.
• ACL_PROF_L2CACHE: L2 Cache data collection.
• ACL_PROF_HCCL_TRACE: collects communication data.
• ACL_PROF_MSPROFTX: profile data output by the user and upper-layer framework program. In the application script, you need to add one of the following APIs:
  • For details about the mstx API (MindStudio Tools Extension API) operations, see mstx API Usage Examples in Performance Tuning Tool.
  • msproftx Extended APIs.
• ACL_PROF_TRAINING_TRACE: collects iteration traces.
• ACL_PROF_RUNTIME_API: collects runtime API profile data.

prof_config

int.

• If a non-zero value is returned, the operation is successful. The return value is the pointer address of the aclprofConfig type.
• If 0 is returned, the operation fails.