On 2018-05-12 10:01, Michael Büsch wrote:
On Sat, 12 May 2018 10:50:42 +0300
Kalle Valo <kvalo@xxxxxxxxxxxxxx> wrote:
Larry Finger <Larry.Finger@xxxxxxxxxxxx> writes:
> On 05/11/2018 05:13 AM, Kalle Valo wrote:
>> Rafał Miłecki <zajec5@xxxxxxxxx> writes:
>>
>>> On 11 May 2018 at 11:17, Rafał Miłecki <zajec5@xxxxxxxxx> wrote:
[...]
>>>
>>> As these patches fix regression/build error, I believe both should get
>>> into 4.17.
>>
>> How much confidence do we have that we don't need to end up reverting
>> patch 2 as well? I rather be pushing patch 2 to 4.18 so that there's
>> more time for testing and waiting for feedback.
>
> Although I do not have the hardware to test the builds, I worked
> closely with the OP in the bug at b.r.c noted above. From that effort,
> it became clear what configuration variables were missing to cause the
> x86 failures. Patch 2 satisfies the requirement, and prevents the
> build problems found by the MIPS users. Both patches are needed in
> 4.17.
And I assume Michael is ok with this approach as well as I haven't
heard
from him. I'll then push both of these to 4.17.
Yes, I'm OK with the patch, if we have a third patch that cleans up the
PCI_DRIVERS_LEGACY dependency by moving it to SSB_PCICORE_HOSTMODE
where it belongs. (This doesn't need to go into the stable tree.)
We currently implicitly get that via dependency chain, so this is OK
for now as-is.
I'm planning to handle PCI_DRIVERS_LEGACY cleanup once my patches hit
net-next.git and then wireless-drivers-next.git. It's to avoid
conflicts.
