Adding GPU index as a parameter for ATT (#547)

* Adding GPU index as a parameter for ATT

* Tidy fix

* Using tokenize

* Update tests/rocprofv3/advanced-thread-trace/CMakeLists.txt

Co-authored-by: Indic, Vladimir <Vladimir.Indic@amd.com>

* Update tests/rocprofv3/advanced-thread-trace/CMakeLists.txt

* Adding error logging. Using idx instead of id.

---------

Co-authored-by: Giovanni <gbaraldi@amd.com>
Co-authored-by: Indic, Vladimir <Vladimir.Indic@amd.com>
Šī revīzija ir iekļauta:
Baraldi, Giovanni
2025-08-04 23:15:50 +02:00
revīziju iesūtīja GitHub
vecāks 533a8329d8
revīzija fd6f96ffb5
5 mainīti faili ar 67 papildinājumiem un 0 dzēšanām
+13
Parādīt failu
@@ -781,6 +781,13 @@ For MPI applications (or other job launchers such as SLURM), place rocprofv3 ins
type=str,
)
att_options.add_argument(
"--att-gpu-index",
help="Comma-separated list of GPU index(es) to enable thread trace. Default: All",
default=None,
type=str,
)
att_options.add_argument(
"--att-perfcounters",
help="(gfx9) List of performance counters, and optionally their SIMD mask.",
@@ -1553,6 +1560,12 @@ def run(app_args, args, **kwargs):
args.att_serialize_all,
overwrite=True,
)
if args.att_gpu_index:
update_env(
"ROCPROF_ATT_PARAM_GPU_INDEX",
args.att_gpu_index,
overwrite=True,
)
if check_att_capability(args):
update_env(
"ROCPROF_ATT_LIBRARY_PATH",