Dear Thorsten, Sorry for the misunderstanding. MT7920 and MT7921 are different chip (hardware) but both use same chip_id 0x7961 and we distinguish between these two chip through the 8th bit of the "fw_flavor" register. Therefore, I think we have not violated this rule: > Users switching to a newer kernel should *not* have to install newer > firmware files to keep their hardware working. because MT7921 still be MT7921 and use same FW bin BT_RAM_MT7961_1_2_hdr.bin. About MT7920, we will push the stable FW bin after internal test done. BRs Peter On Wed, 2024-04-24 at 08:06 +0200, Linux regression tracking (Thorsten Leemhuis) wrote: > On 23.04.24 12:23, Artem S. Tashkinov wrote: > > > > Could you please push this firmware ASAP? > > > > It's been reported to be missing: > > https://urldefense.com/v3/__https://bugzilla.kernel.org/show_bug.cgi?id=218757__;!!CTRNKA9wMg0ARbw!jL88juHMcB9CkhfQy00tGVjWsRXXdMN-_T28ANIP1xdw0C5Yr2dENm0mM3n9Egv5xqdT7oDyEXtOswJ6o2EYZSw9h_v5$ > > > > FWIW, that can't be the only solution for that problem, as > Documentation/driver-api/firmware/firmware-usage-guidelines.rst > clearly > states: > > """ > Users switching to a newer kernel should *not* have to install newer > firmware files to keep their hardware working. > """ > > Could anyone from mediatek please confirm that this rule is adhered? > > Side note: I wonder if Peter's patch > https://lore.kernel.org/all/20240415141922.25055-1-peter.tsao@xxxxxxxxxxxx/ > ("Bluetooth: btusb: Fix the patch for MT7920 the affected to MT7921") > is > relevant for this. > > Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' > hat) > -- > Everything you wanna know about Linux kernel regression tracking: > https://urldefense.com/v3/__https://linux-regtracking.leemhuis.info/about/*tldr__;Iw!!CTRNKA9wMg0ARbw!jL88juHMcB9CkhfQy00tGVjWsRXXdMN-_T28ANIP1xdw0C5Yr2dENm0mM3n9Egv5xqdT7oDyEXtOswJ6o2EYZXhunwKm$ > > If I did something stupid, please tell me, as explained on that page.
