AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Async memory copy callback tracing + memory copy size (#791)

Browse Source 
* Async memory copy tracing update

- rocprofiler_buffer_tracing_memory_copy_record_t: thread_id and bytes
- support ROCPROFILER_CALLBACK_TRACING_MEMORY_COPY
- init_public_api_struct can fully construct

* Testing for callback async copy tracing
This commit is contained in:
Jonathan R. Madsen
2024-04-18 04:31:59 -05:00
committed by GitHub
parent edb1883a05
commit 12c836f95f
13 changed files with 378 additions and 170 deletions
Download Patch File Download Diff File Expand all files Collapse all files
source/include/rocprofiler-sdk/fwd.h
+6 -5
View File
@@ -155,6 +155,7 @@ typedef enum // NOLINT(performance-enum-size)
ROCPROFILER_CALLBACK_TRACING_CODE_OBJECT, ///< @see ::rocprofiler_code_object_operation_t
ROCPROFILER_CALLBACK_TRACING_SCRATCH_MEMORY, ///< @see ::rocprofiler_scratch_memory_operation_t
ROCPROFILER_CALLBACK_TRACING_KERNEL_DISPATCH, ///< Callbacks for kernel dispatches
ROCPROFILER_CALLBACK_TRACING_MEMORY_COPY, ///< @see ::rocprofiler_memory_copy_operation_t
ROCPROFILER_CALLBACK_TRACING_LAST,
} rocprofiler_callback_tracing_kind_t;
@@ -613,11 +614,11 @@ rocprofiler_record_header_compute_hash(uint32_t category, uint32_t kind)
*/
typedef struct rocprofiler_kernel_dispatch_info_t
{
uint64_t size; ///< Size of this struct
rocprofiler_agent_id_t agent_id; ///< Agent ID where kernel is launched
rocprofiler_queue_id_t queue_id; ///< Queue ID where kernel packet is enqueued
rocprofiler_kernel_id_t kernel_id; ///< Kernel identifier
rocprofiler_dispatch_id_t dispatch_id; ///< unique id for each dispatch
uint64_t size; ///< Size of this struct (minus reserved padding)
rocprofiler_agent_id_t agent_id; ///< Agent ID where kernel is launched
rocprofiler_queue_id_t queue_id; ///< Queue ID where kernel packet is enqueued
rocprofiler_kernel_id_t kernel_id; ///< Kernel identifier
rocprofiler_dispatch_id_t dispatch_id; ///< unique id for each dispatch
uint32_t private_segment_size; ///< runtime private memory segment size
uint32_t group_segment_size; ///< runtime group memory segment size
rocprofiler_dim3_t workgroup_size; ///< runtime workgroup size (grid * threads)