On Thu, Dec 11, 2025 at 09:45:54AM +0100, Cezary Rojewski wrote:
On 2025-12-10 12:55 PM, Andy Shevchenko wrote:
While the used GML is consistent with the pattern for other Intel * Lake SoCs, the de facto use is GLK. Update the acronym and users accordingly.
Note, a handful of the drivers for Gemini Lake in the Linux kernel use GLK already (LPC, MEI, pin control, SDHCI, ...) and even some in ASoC. The only ones in this patch used the inconsistent one.
A number of times I've fought for the 'GLK' to disappear from the Intel's audio subsystem as clearly the "right" shortcut is 'GML'.
Probably you were late as that boat sailed and we have the only your driver (and SOF which was put into position to reuse the existing ID) use the correct acronym.
However, I do understand where are you coming from - the corrections came late and the "mistake" has been widely spread.
Yeah... With all understanding and support for the correctness, the easiest way to get rid of inconsistency is to change only these couple of drivers and not the entire world.
...
-#define PCI_DEVICE_ID_INTEL_HDA_GML 0x3198 +#define PCI_DEVICE_ID_INTEL_HDA_GLK 0x3198
If two #defines are no-go (PCI_DEVICE_ID_INTEL_HDA_GLK and _GML), then perhaps at least a comment to the right of the ID mentioning the "GML" would help.
Bjorn, are you agree on adding a comment to PCI IDs to point out that in a couple of Intel documents the machine's acronym is GML?
...
--- a/sound/soc/intel/avs/board_selection.c +++ b/sound/soc/intel/avs/board_selection.c
- AVS_MACH_ENTRY(HDA_GML, avs_gml_i2s_machines),
- AVS_MACH_ENTRY(HDA_GLK, avs_glk_i2s_machines),
To be honest, I'd leave 'avs_gml_i2s_machines' as is.
Sure, your driver your rules :), I will not touch that in v2.