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

rocr/driver: add PC sampling support to driver interface

Browse Source 
Add PC sampling functionality to the driver interface:

1. Add new PC sampling methods to Driver base class:
   - PcSamplingQueryCapabilities
   - PcSamplingCreate
   - PcSamplingDestroy
   - PcSamplingStart
   - PcSamplingStop

2. Implement PC sampling methods in KfdDriver using HSAKMT APIs:
   - Map HSAKMT status codes to HSA status codes
   - Handle resource busy conditions
   - Proper error handling for all operations

Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com>
This commit is contained in:
Honglei Huang
2025-06-27 11:09:17 +08:00
committed by Huang, Honglei1
parent a47c060d6a
commit 56cb9390ff
4 changed files with 132 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files
runtime/hsa-runtime/core/driver/kfd/amd_kfd_driver.cpp
+59
View File
@@ -730,5 +730,64 @@ hsa_status_t KfdDriver::ReturnAsanHeaderPage(void* mem) const {
return HSA_STATUS_SUCCESS;
}
hsa_status_t KfdDriver::PcSamplingQueryCapabilities(uint32_t node_id, void* sample_info,
uint32_t sample_info_sz,
uint32_t* sz_needed) const {
HSAKMT_STATUS status = HSAKMT_CALL(
hsaKmtPcSamplingQueryCapabilities(node_id, sample_info, sample_info_sz, sz_needed));
if (status == HSAKMT_STATUS_KERNEL_ALREADY_OPENED) {
return static_cast<hsa_status_t>(HSA_STATUS_ERROR_RESOURCE_BUSY);
}
if (status != HSAKMT_STATUS_SUCCESS) {
return HSA_STATUS_ERROR;
}
return HSA_STATUS_SUCCESS;
}
hsa_status_t KfdDriver::PcSamplingCreate(uint32_t node_id, HsaPcSamplingInfo* sample_info,
uint32_t* trace_id) const {
HSAKMT_STATUS status = HSAKMT_CALL(hsaKmtPcSamplingCreate(node_id, sample_info, trace_id));
if (status == HSAKMT_STATUS_KERNEL_ALREADY_OPENED) {
return static_cast<hsa_status_t>(HSA_STATUS_ERROR_RESOURCE_BUSY);
}
if (status != HSAKMT_STATUS_SUCCESS) {
return HSA_STATUS_ERROR;
}
return HSA_STATUS_SUCCESS;
}
hsa_status_t KfdDriver::PcSamplingDestroy(uint32_t node_id, uint32_t trace_id) const {
HSAKMT_STATUS status = HSAKMT_CALL(hsaKmtPcSamplingDestroy(node_id, trace_id));
if (status == HSAKMT_STATUS_KERNEL_ALREADY_OPENED) {
return static_cast<hsa_status_t>(HSA_STATUS_ERROR_RESOURCE_BUSY);
}
if (status != HSAKMT_STATUS_SUCCESS) {
return HSA_STATUS_ERROR;
}
return HSA_STATUS_SUCCESS;
}
hsa_status_t KfdDriver::PcSamplingStart(uint32_t node_id, uint32_t trace_id) const {
HSAKMT_STATUS status = HSAKMT_CALL(hsaKmtPcSamplingStart(node_id, trace_id));
if (status == HSAKMT_STATUS_KERNEL_ALREADY_OPENED) {
return static_cast<hsa_status_t>(HSA_STATUS_ERROR_RESOURCE_BUSY);
}
if (status != HSAKMT_STATUS_SUCCESS) {
return HSA_STATUS_ERROR;
}
return HSA_STATUS_SUCCESS;
}
hsa_status_t KfdDriver::PcSamplingStop(uint32_t node_id, uint32_t trace_id) const {
HSAKMT_STATUS status = HSAKMT_CALL(hsaKmtPcSamplingStop(node_id, trace_id));
if (status == HSAKMT_STATUS_KERNEL_ALREADY_OPENED) {
return static_cast<hsa_status_t>(HSA_STATUS_ERROR_RESOURCE_BUSY);
}
if (status != HSAKMT_STATUS_SUCCESS) {
return HSA_STATUS_ERROR;
}
return HSA_STATUS_SUCCESS;
}
} // namespace AMD
} // namespace rocr