Skip to contentSkip to conversation

BUG: VIA HD Audio sound card support regressed

DISCUSSION Alexander Andrejevic 2 participants 2 messages Mar 9, 2016

lore
Share & tools
lore
#0 lore raw

Hi,

A regression in the Intel HD Audio driver was introduced by commit 12daef65fd868cf30be5afe3e6be6689c44c7940 (2011-06-20 14:24:07 GMT). Namely, a VIA VT1708-based card with the PCI ID 1106:3288 stopped working entirely, even though no apparent error message is produced in the log by the driver.
I've debugged this issue and determined that the card needs two additional initialization verbs to be sent to it:

{0x26, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}
{0x26, AC_VERB_SET_EAPD_BTLENABLE, 0x02}

However, I don't know whether this applies to all VT1708-based cards (it most likely doesn't), nor how would they be affected by these additional commands, so I'm not sure how to properly patch this issue myself. Perhaps the correct way to do it is to add an SND_PCI_QUIRK for this card?

Regards,
Alex

NOTE: I'm not subscribed to the Linux Kernel Mailing List, so please CC your replies to my e-mail address.

Signature · 1 line
SDF Public Access UNIX System - http://sdf.lonestar.org
#1 lore raw
Quoted 11 lines · Alexander Andrejevic

Hi,

A regression in the Intel HD Audio driver was introduced by commit 12daef65fd868cf30be5afe3e6be6689c44c7940 (2011-06-20 14:24:07 GMT). Namely, a VIA VT1708-based card with the PCI ID 1106:3288 stopped working entirely, even though no apparent error message is produced in the log by the driver.
I've debugged this issue and determined that the card needs two additional initialization verbs to be sent to it:

Quoted 7 lines

{0x26, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}
{0x26, AC_VERB_SET_EAPD_BTLENABLE, 0x02}

However, I don't know whether this applies to all VT1708-based cards (it most likely doesn't), nor how would they be affected by these additional commands, so I'm not sure how to properly patch this issue myself. Perhaps the correct way to do it is to add an SND_PCI_QUIRK for this card?

Could you open a bug report in bugzilla.kernel.org and attach the alsa-info.sh outputs taken from both the old (good) kernel and the recent broken one? Run the script with --no-upload option and use attachments.

Takashi