(In reply to Daniel Drake from comment #22)
> Created attachment 136369 [deta=
ils] [review] [review]
> GP104: enable HDMI audio device function
>=20
> On Asus GL502VS I investigated why the gfx device must be removed befo=
re
> rescan in the above workarounds.
>=20
> The reason is that when Linux first probes the device (before you atte=
mpt
> the workaround), pci_setup_device() notes that the device is not
> multifunction capable. This causes pci_scan_slot() to not bother scann=
ing
> the non-zero functions (per the behaviour inside next_fn()).
>=20
> I checked and I found that when the 0x488 magic bit is not set, the gfx
> device advertises as non-multifunction. After the bit is set, the devi=
ce
> advertises as multi-function. So, after setting the magic bit, removin=
g the
> device will cause Linux to re-probe it during the next rescan, taking =
note
> at that point that it is a multi-function device, and proceeding to sc=
an the
> functions, finding the audio device at function 1.
>=20
> Based on that I have a first attempt at a fix. It's not working though,
> audio output is silent (but I did have it working with the previous
> workarounds). I'll look closer next week.

I think this check against 0x134 can be something like < 0x82 instead. T=
his
thing is there since like G82 GPUs and I am sure the HDMI audio device is
controlled like this on all those GPUs.