[PATCH 0/1][linux-firmware][SRU][B/C/D] linux-firmware: Update firmware file for Intel Bluetooth AX201

Seth Forshee seth.forshee at canonical.com
Mon Jul 15 12:37:29 UTC 2019


On Mon, Jul 15, 2019 at 04:07:57PM +0800, You-Sheng Yang wrote:
> BugLink: https://bugs.launchpad.net/bugs/1835879
> 
> [Impact]
> Bluetooth on Intel Wireless-AC 9462 on IceLake requires new firmware
> blobs intel/ibt-19-32-0.{sfi,ddc} to be enabled.
> 
> [Fix]
> Backport newly released firmware blobs from linux-firmware commit
> d52556e4592e64023157a83fb0f483661f23ac0e. `git cherry-pick` works for
> each target branch, but the generated patch doesn't apply to each other.
> B/C/D would requires this explicitly and Eoan should be rebased onto
> linux-firmware master branch HEAD as usual.
> 
> [Test Case]
> Verified with hardware platforms with corresponding devices. Completely
> power off and power on to trigger firmware reload, check from dmesg to
> see if Bluetooth firmware has been correctly loaded.
> 
> [Regression Risk]
> Low. This adds new binary firmware blobs that are only loaded by target
> configuration. For others, they have been verified individually on the
> platforms that have corresponding hardwares.
> 
> Amit K Bag (1):
>   linux-firmware: Update firmware file for Intel Bluetooth AX201
> 
>  WHENCE                |  25 +++++++++++++++----------
>  intel/ibt-19-0-0.sfi  | Bin 715376 -> 722576 bytes
>  intel/ibt-19-0-1.sfi  | Bin 715376 -> 722576 bytes
>  intel/ibt-19-0-4.sfi  | Bin 715376 -> 722576 bytes
>  intel/ibt-19-16-4.sfi | Bin 715896 -> 722576 bytes
>  intel/ibt-19-32-0.ddc | Bin 0 -> 70 bytes
>  intel/ibt-19-32-0.sfi | Bin 0 -> 722576 bytes
>  intel/ibt-19-32-1.sfi | Bin 715896 -> 722576 bytes
>  8 files changed, 15 insertions(+), 10 deletions(-)
>  create mode 100644 intel/ibt-19-32-0.ddc
>  create mode 100644 intel/ibt-19-32-0.sfi

First of all, cosmic is nearly EOL and I see no reason to enable new
hardware there, so I do not plan to apply this there.

Second, for SRU every change needs to have a justification. I cannot
tell whether the updates to the existing files are actually necessary
for enabling the hardware as described in the SRU justification. If they
are not necessary to enable that hardware, they either should not be
part of the SRU, or there should be another bug explaining what issues
are being fixed by these updates. We cannot just bring them along becuse
they were part of the upstream commit which introduced the new files you
need.

Thanks,
Seth



More information about the kernel-team mailing list