On 8/2/23 10:57, Takashi Iwai wrote: > On Wed, 02 Aug 2023 17:52:26 +0200, > Bjorn Helgaas wrote: >> >> On Wed, Aug 02, 2023 at 10:01:01AM -0500, Pierre-Louis Bossart wrote: >>> Add part ID to common include file >> >> Please drop period at end of subject and add one at the end of the >> commit log. >> >> Also mention the drivers that will use this new #define; looks like >> hda_intel.c and ... >> >> Well, actually, I only see that one use, which means we probably >> shouldn't add this #define to pci_ids.h, per the comment at the top of >> the file. If there's only one use, use the hex ID in the driver (or >> add a #define in the driver itself). > > Judging from the previous patterns, the same ID could be required for > ASoC SOF driver, too, which isn't included in this patch set. In > that case, it's worth to put to pci_ids.h. > (OTOH, it can be done at a later stage, too.) I am not following. we just agreed a couple of weeks ago to record ALL Intel/HDaudio PCI IDs in the same pci_ids.h include file. ArrowLake-S is the first addition to first file after the work done by Cezary/Amadeusz. Yes it's required to be added since it'll be used in other parts later on. But even if there was ONE use of this PCI ID, why would we not add it for consistency to the global pci_ids.h file? Takashi's hda_intel.c file would look really bad if we have a mix of single-use PCIs and shared ones... Oh and heads-up that I have a change for LunarLake that will require Mark to pull the branch from Takashi :-)
