What drives the choice to include a specific version of ath10k-firmware OpenWrt-master when other versions are out there?

With all the different versions of the ath10k-firmware available upstream, I am wondering what drives the choice to include a particular version in the OW package.

For example:

So version 3.5.3/firmware-5.bin_10.4-3.5.3-00053 is included, but there is a high version number within the 3.5.3 directory firmware-5.bin_10.4-3.5.3-00057... further, there are higher versions outside of that directory.

For example, there are larger versions like 3.5.3.1, 3.9.0.2, 3.10, etc.Thanks!

Afair - nothing particular. Usually the version was bumped based on user feedback or when PRs were submitted.

Thanks for the info. Do I understand correctly that the directories correspond to the version number?
So even though 3.9.0.2 was updated 2 months ago, 3.10 (updated 12 months ago) is the current release?

I don't know. I never understood ath10k firmware versioning :slight_smile:

1 Like

:upside_down_face:

If you look closer (and lower, line 181), ath10k-firmware-qca9984 isn't even built from the ath10k-firmware source anymore, but has been switched to linux-firmware instead.

Thanks for pointing that out. Someone else did the same for me in another thread. I just wanted to understand the selection strategy for one of them given all the versions out there whether pulled form the linux-firmware tarball or from the upstream github repo.

With the switch to the linux-firmware source, the selection has been offloaded to the linux-firmware maintainer's on kernel.org (which effectively means QCA via their driver maintainers pushing their preferred version).

1 Like