[Intel-wired-lan] [PATCH 2/2] ethernet: igb: e1000_phy: Check for ops.force_speed_duplex existence

[Jakub Kicinski]

On Tue,  3 Nov 2020 00:13:07 +0100 Paul Menzel wrote:
> From: Jeffrey Townsend <jeffrey.townsend at bigswitch.com>
> 
> The ops field might no be defined, so add a check.

This change should be first, otherwise AFAIU if someone builds the
kernel in between the commits (e.g. for bisection) it will crash.

> The patch is taken from Open Network Linux (ONL), and it was added there
> as part of the patch
> 
>     packages/base/any/kernels/3.16+deb8/patches/driver-support-intel-igb-bcm5461X-phy.patch
> 
> in ONL commit f32316c63c (Support the BCM54616 and BCM5461S.) [1]. Part
> of this commit was already upstreamed in Linux commit eeb0149660 (igb:
> support BCM54616 PHY) in 2017.
> 
> I applied the forward-ported
> 
>     packages/base/any/kernels/5.4-lts/patches/0002-driver-support-intel-igb-bcm5461S-phy.patch
> 
> added in ONL commit 5ace6bcdb3 (Add 5.4 LTS kernel build.) [2].
> 
> [1]: https://github.com/opencomputeproject/OpenNetworkLinux/commit/f32316c63ce3a64de125b7429115c6d45e942bd1
> [2]: https://github.com/opencomputeproject/OpenNetworkLinux/commit/5ace6bcdb37cb8065dcd1d4404b3dcb6424f6331

No need to put this in every commit message.

We preserve the cover letter in tree as a merge commit message, so
explaining things once in the cover letter is sufficient.

> Cc: Jeffrey Townsend <jeffrey.townsend at bigswitch.com>

Jefferey will need to provide a sign-off as the author.

> Cc: John W Linville <linville at tuxdriver.com>
> Signed-off-by: Paul Menzel <pmenzel at molgen.mpg.de>