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

CI: Relax timestamp checking (#1189)

Browse Source 
* Relax timestamp checking

- Prevent recurring CI failures that have no remedy until HSA/driver issues are resolved

* Replace "cc" abbreviation in tests with "counter-collection"

* Update CODEOWNERS to explicitly include jrmadsen for source/include

* Extra logging in rocprofiler tool library

* Tweak aborted-app test

- remove counter collection as part of the test
This commit is contained in:
Jonathan R. Madsen
2024-11-06 23:32:47 -06:00
committed by GitHub
parent 6564419357
commit 98858b60ec
12 changed files with 127 additions and 87 deletions
Download Patch File Download Diff File Expand all files Collapse all files
source/lib/rocprofiler-sdk/kernel_dispatch/profiling_time.cpp
+8 -13
View File
@@ -55,27 +55,22 @@ get_dispatch_time(hsa_agent_t _hsa_agent,
if(_profile_time.status == HSA_STATUS_SUCCESS)
{
// if we encounter this in CI, it will cause test to fail
ROCP_CI_LOG_IF(ERROR, _profile_time.end < _profile_time.start)
<< "hsa_amd_profiling_get_dispatch_time for kernel_id=" << _kernel_id
<< " on rocprofiler_agent="
<< CHECK_NOTNULL(agent::get_rocprofiler_agent(_hsa_agent))->node_id
<< " returned dispatch times where the end time (" << _profile_time.end
<< ") was less than the start time (" << _profile_time.start << ")";
_profile_time = tracing::adjust_profiling_time(
"dispatch",
"hsa_amd_profiling_get_dispatch_time",
_profile_time,
tracing::profiling_time{
HSA_STATUS_SUCCESS, _baseline.value_or(dispatch_time.start), ts});
}
else
{
ROCP_CI_LOG(ERROR) << "hsa_amd_profiling_get_dispatch_time for kernel id=" << _kernel_id
<< " on rocprofiler_agent="
<< CHECK_NOTNULL(agent::get_rocprofiler_agent(_hsa_agent))->id.handle
<< " returned status=" << dispatch_time_status
<< " :: " << hsa::get_hsa_status_string(dispatch_time_status);
ROCP_CI_LOG(ERROR) << fmt::format(
"hsa_amd_profiling_get_dispatch_time for kernel id={} on agent-{} returned status={} "
":: {}",
_kernel_id,
CHECK_NOTNULL(agent::get_rocprofiler_agent(_hsa_agent))->node_id,
static_cast<int>(dispatch_time_status),
hsa::get_hsa_status_string(dispatch_time_status));
}
return _profile_time;