Re: renesas 1912:0014 failures

[Michal Pecio <michal.pecio@gmail.com>]

On Tue, 12 May 2026 16:40:30 +0200, Anders Thomson wrote:
> Adding that quirk to the boot cmdline, caused this on resume:
> 
> [ 1357.247864] usb usb10: root hub lost power or was reset
> [ 1385.611310] xhci-pci-renesas 0000:03:00.0: PCI post-resume error -110!
> [ 1385.611438] xhci-pci-renesas 0000:03:00.0: HC died; cleaning up
> [ 1385.611491] xhci-pci-renesas 0000:03:00.0: PM: dpm_run_callback():
> pci_pm_resume returns -110
> [ 1385.611506] xhci-pci-renesas 0000:03:00.0: PM: failed to resume
> async: error -110

Looks like the quirk worked, but the chip is still stuck for good.

Sounds like it happens every time?
Does it happen without any connected devices?

Any chance that something goes weird already during suspend?
Maybe try to collect suspend/resume log with extra debug info:

echo 'module xhci_hcd +p' >/proc/dynamic_debug/control
dmesg -W |tee log.txt
# suspend and resume
echo 'module xhci_hcd -p' >/proc/dynamic_debug/control # disable
# stop dmesg, send log.txt

It's odd, I have a uPD720202 (the 2 port variant) and it never had
similar problems. But my card doesn't use the standby supply and
hence it fully power cycles the chip during S3 sleep, so it may be
apples to oranges if your card keeps aux power applied.

Regards,
Michal