Attention is currently required from: Arthur Heymans, Michael Niewöhner.
Nicholas Sudsgaard has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/80727?usp=email )
Change subject: include/device/azalia_device.h: Fix incorrect bit shift ......................................................................
Patch Set 2:
(1 comment)
Commit Message:
https://review.coreboot.org/c/coreboot/+/80727/comment/aa61ff88_f00bda48 : PS2, Line 9: TEST=Timeless build using AZALIA_PIN_DESC() and without now produce the : same binary.
Only one board uses this in the tree (mainboard/clevo/tgl-u/variants/l140mu/hda_verb. […]
The connected pins have these values: ``` 0x12 0x9fa60130 0x14 0x9e170110 0x15 0x04211020 0x1a 0x04a1113c 0x1d 0x88f70145 ```
These values decoded would turn into: ``` AZALIA_PIN_CFG(0, 0x12, AZALIA_PIN_DESC( INTEGRATED, INTERNAL, 0xf, MIC_IN, OTHER_DIGITAL, COLOR_UNKNOWN, NO_JACK_PRESENCE_DETECT, 0x3, 0x0 )), AZALIA_PIN_CFG(0, 0x14, AZALIA_PIN_DESC( INTEGRATED, INTERNAL, 0xe, SPEAKER, OTHER_ANALOG, COLOR_UNKNOWN, NO_JACK_PRESENCE_DETECT, 0x1, 0x0 )), AZALIA_PIN_CFG(0, 0x15, AZALIA_PIN_DESC( JACK, EXTERNAL_PRIMARY_CHASSIS, RIGHT, HP_OUT, STEREO_MONO_1_8, BLACK, JACK_PRESENCE_DETECT, 0x2, 0x0 )), AZALIA_PIN_CFG(0, 0x1a, AZALIA_PIN_DESC( JACK, EXTERNAL_PRIMARY_CHASSIS, RIGHT, MIC_IN, STEREO_MONO_1_8, BLACK, NO_JACK_PRESENCE_DETECT, 0x3, 0xc )), AZALIA_PIN_CFG(0, 0x1d, AZALIA_PIN_DESC( INTEGRATED, EXTERNAL_PRIMARY_CHASSIS, SPECIAL8, DEVICE_OTHER, OTHER_ANALOG, COLOR_UNKNOWN, NO_JACK_PRESENCE_DETECT, 0x4, 0x5 )), ``` Using this would produce the same binary with timeless builds. However, I find it very odd to be using values like '0xf' and '0xe' in `location_1` as they are reserved values.
I don't think we have to change any of the enums in this case. My assumption is the author assumed `AZALIA_PIN_DESC()` was correct and used the enums that made sense. However, I also understand that it may be better to just keep the tested values even if they are incorrect. I'll leave that decision to you.