forkless/XPUSYS-Monitor-NG
1
0
Fork 0
mirror of https://github.com/forkless/XPUSYS-Monitor-NG.git synced 2026-08-16 08:46:38 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix: use typeperf for AMD GPU util, skip PDH

Browse Source
This commit is contained in:
forkless
2026-06-08 23:50:41 +02:00
parent 6d8a385ca3
commit 03e2ae3f40
1 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
providers/amd.py
+5 -4
View File
@@ -18,7 +18,7 @@ import sys
from typing import Tuple
from .base import BaseGPUProvider, GPUSnapshot
from ._utils import _get_cpu_info, _read_cpu_ram_stats, _PdhQuery, _TypeperfGpuQuery, _is_admin
from ._utils import _get_cpu_info, _read_cpu_ram_stats, _TypeperfGpuQuery, _is_admin
logger = logging.getLogger("XPUSYSMonitor")
@@ -52,11 +52,12 @@ class AMDProvider(BaseGPUProvider):
self._check_torch()
self._check_psutil()
# Windows GPU utilisation — chain: PDH (fast) -> typeperf (fallback)
# Windows GPU utilisation — typeperf (primary, reliable CSV output)
# PDH has wildcard-counter issues with AMD drivers, so skip it.
self._pdh = _PdhQuery()
self._pdh_ok = self._pdh.init()
self._pdh_ok = False # PDH disabled for AMD (wildcard handling unreliable)
self._tp_gpu = _TypeperfGpuQuery()
self._tp_gpu_ok = self._tp_gpu.init() if not self._pdh_ok else False
self._tp_gpu_ok = self._tp_gpu.init()
# BaseGPUProvider.__init__ starts the polling thread — call last
super().__init__(interval_ms=interval_ms)