Update: pulled the full boot-time log via journalctl -b <id> -k, since
dmesg had rotated past it after several suspend/resume cycles. The
actual failure sequence is more specific than "GPIO chip not found yet":

int3472-discrete INT3472:00: cannot find GPIO chip INTC10B5:00, deferring   
(repeats ~20x over ~1s)
usbcore: registered new interface driver usbio-bridge
gpio_usbio.usbio-gpio usbio.usbio-gpio.0: [Firmware Bug]: GPIO 4 is not in FW 
pins bitmap
int3472-discrete INT3472:00: GPIO type 0x02 unknown; the sensor may not work

The repeated "deferring" resolves normally within ~1 second once usbio-
bridge/gpio_usbio register — that part of probing is not the failure.
The actual failure is gpio_usbio itself flagging a firmware bug:
something (int3472, requesting the sensor's power/reset line) asks for
pin 4 on the USBIO GPIO controller, but the platform's own advertised
pin bitmap for that controller doesn't include pin 4. int3472 then logs
"GPIO type 0x02 unknown" and gives up on properly sequencing power/reset
for the OV02C10 sensor — which matches the black-frame symptom (sensor
enumerates and streams, but was never actually powered/reset correctly).

Both intel_ipu7 and intel_ipu7_isys log "module is from the staging
directory, the quality is unknown" — this whole stack (ipu7, gpio-usbio,
the int3472 GPIO-type-0x02 path) is still very new/actively developed,
so a pin-bitmap mismatch for this SoC's GPIO expander is plausible as a
driver-side parsing bug, a genuine BIOS/ACPI firmware descriptor bug on
this laptop model, or simply unhandled in this driver revision.

modinfo confirms intel_ipu7/gpio_usbio are built correctly for the
running 7.0.0-1013-oem kernel (vermagic matches) — not a stale-module
issue. Not yet tested against the 7.0.0-31-generic kernel also installed
on this system, which carries a separately-built ipu7/usbio module set;
will follow up if that flavor behaves differently, since that would
point at a backport gap between the oem and hwe/generic kernel lines
rather than a firmware bug.

** Summary changed:

- Intel IPU7 OV02C10 camera produces black frames: INT3472 cannot resolve 
INTC10B5 GPIO
+ [Firmware Bug]: GPIO 4 is not in FW pins bitmap + GPIO type 0x02 unknown

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2167477

Title:
  [Firmware Bug]: GPIO 4 is not in FW pins bitmap + GPIO type 0x02
  unknown

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-signed-oem-7.0/+bug/2167477/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to