FileNotFoundError when using eGPU

[shade-belisar]

I'm running Linux Mint 21.3 on a GPD Win MAX 2 (2024) with a RX 7900 XT attached via Oculink. In this constellation the hdd daemon does not start, instead it throws this error:

Jun 22 21:29:01 My-Machine hhd[4857]: ╭───────────────────── Traceback (most recent call last) ──────────────────>
Jun 22 21:29:01 My-Machine hhd[4857]: │ /home/me/.local/share/hhd/venv/bin/hhd:8 in <module>                  >
Jun 22 21:29:01 My-Machine hhd[4857]: │                                                                          >
Jun 22 21:29:01 My-Machine hhd[4857]: │   5 from hhd.__main__ import main                                        >
Jun 22 21:29:01 My-Machine hhd[4857]: │   6 if __name__ == '__main__':                                           >
Jun 22 21:29:01 My-Machine hhd[4857]: │   7 │   sys.argv[0] = re.sub(r'(-script\.pyw|\.exe)?$', '', sys.argv[0]) >
Jun 22 21:29:01 My-Machine hhd[4857]: │ ❱ 8 │   sys.exit(main())                                                 >
Jun 22 21:29:01 My-Machine hhd[4857]: │   9                                                                      >
Jun 22 21:29:01 My-Machine hhd[4857]: │                                                                          >
Jun 22 21:29:01 My-Machine hhd[4857]: │ /home/me/.local/share/hhd/venv/lib/python3.12/site-packages/hhd/__main>
Jun 22 21:29:01 My-Machine hhd[4857]: │ py:564 in main                                                           >
Jun 22 21:29:01 My-Machine hhd[4857]: │                                                                          >
Jun 22 21:29:01 My-Machine hhd[4857]: │   561 │   │   │   │   except Exception as e:                             >
Jun 22 21:29:01 My-Machine hhd[4857]: │   562 │   │   │   │   │   logger.warning(f"Could not hide update settings>
Jun 22 21:29:01 My-Machine hhd[4857]: │   563 │   │   │   │   settings = merge_settings(                         >
Jun 22 21:29:01 My-Machine hhd[4857]: │ ❱ 564 │   │   │   │   │   [*[p.settings() for p in sorted_plugins], hhd_s>
Jun 22 21:29:01 My-Machine hhd[4857]: │   565 │   │   │   │   )                                                  >
Jun 22 21:29:01 My-Machine hhd[4857]: │   566 │   │   │   │   # Force general settings to be last                >
Jun 22 21:29:01 My-Machine hhd[4857]: │   567 │   │   │   │   if "hhd" in settings:                              >
Jun 22 21:29:01 My-Machine hhd[4857]: │                                                                          >
Jun 22 21:29:01 My-Machine hhd[4857]: │ /home/me/.local/share/hhd/venv/lib/python3.12/site-packages/adjustor/d>
Jun 22 21:29:01 My-Machine hhd[4857]: │ ers/amd/__init__.py:99 in settings                                       >
Jun 22 21:29:01 My-Machine hhd[4857]: │                                                                          >
Jun 22 21:29:01 My-Machine hhd[4857]: │    96 │   │   │   self.core_available = False                            >
Jun 22 21:29:01 My-Machine hhd[4857]: │    97 │   │   │   return {}                                              >
Jun 22 21:29:01 My-Machine hhd[4857]: │    98 │   │                                                              >
Jun 22 21:29:01 My-Machine hhd[4857]: │ ❱  99 │   │   status = get_igpu_status()                                 >
Jun 22 21:29:01 My-Machine hhd[4857]: │   100 │   │   if not status:                                             >
Jun 22 21:29:01 My-Machine hhd[4857]: │   101 │   │   │   self.core_available = False                            >
Jun 22 21:29:01 My-Machine hhd[4857]: │   102 │   │   │   if not self.logged_error:                              >
Jun 22 21:29:01 My-Machine hhd[4857]: │                                                                          >
Jun 22 21:29:01 My-Machine hhd[4857]: │ /home/me/.local/share/hhd/venv/lib/python3.12/site-packages/adjustor/f>
Jun 22 21:29:01 My-Machine hhd[4857]: │ /gpu.py:52 in get_igpu_status                                            >
Jun 22 21:29:01 My-Machine hhd[4857]: │                                                                          >
Jun 22 21:29:01 My-Machine hhd[4857]: │    49 │   epp_avail = None                                               >
Jun 22 21:29:01 My-Machine hhd[4857]: │    50 │   epp = None                                                     >
Jun 22 21:29:01 My-Machine hhd[4857]: │    51 │                                                                  >
Jun 22 21:29:01 My-Machine hhd[4857]: │ ❱  52 │   with open(os.path.join(hwmon, GPU_FREQUENCY_PATH), "r") as f:  >
Jun 22 21:29:01 My-Machine hhd[4857]: │    53 │   │   for line in f.readlines():                                 >
Jun 22 21:29:01 My-Machine hhd[4857]: │    54 │   │   │   if line.startswith("0:"):                              >
Jun 22 21:29:01 My-Machine hhd[4857]: │    55 │   │   │   │   freq = int(line.split()[1].replace("Mhz", ""))     >
Jun 22 21:29:01 My-Machine hhd[4857]: ╰──────────────────────────────────────────────────────────────────────────>
Jun 22 21:29:01 My-Machine hhd[4857]: FileNotFoundError: [Errno 2] No such file or directory:
Jun 22 21:29:01 My-Machine hhd[4857]: '/sys/devices/pci0000:00/0000:00:01.1/0000:01:00.0/0000:02:00.0/0000:03:00.>
Jun 22 21:29:01 My-Machine hhd[4857]: on/hwmon4/device/pp_od_clk_voltage'
Jun 22 21:29:02 My-Machine systemd[1]: hhd_local@me.service: Main process exited, code=exited, status=1/FAILURE
Jun 22 21:29:02 My-Machine systemd[1]: hhd_local@me.service: Failed with result 'exit-code'.

This only happens when the eGPU is attached, without it there is no issue. Is this a bug in hhd, or would I need to configure something for this case?

[antheas]

Linked issue: hhd-dev/adjustor#15

I fixed the crash, but APU detection is still spotty. I will push a release later today.