Re: [PATCH] xen/arm: fix SBDF calculation for vPCI MMIO handlers [and 2 more messages]

[Oleksandr Andrushchenko <Oleksandr_Andrushchenko@xxxxxxxx>]

On 02.11.21 11:32, Julien Grall wrote:
> Hi Oleksandr,
>
> On 02/11/2021 07:16, Oleksandr Andrushchenko wrote:
>>
>>
>> On 01.11.21 23:06, Stefano Stabellini wrote:
>>> On Mon, 1 Nov 2021, Ian Jackson wrote:
>>>> Julien Grall writes ("Re: [PATCH] xen/arm: fix SBDF calculation for vPCI 
>>>> MMIO handlers"):
>>>>> On 28/10/2021 16:54, Ian Jackson wrote:
>>>>>> There are a number of patches that I'm getting CC'd on related to ARM
>>>>>> and vpci (according to the Subject).  Are these targeted for 4.16 ?
>>>>>> Most of them don't have 4.16 Subject tags.
>>>>> Oleksandr wants this patch to be included for 4.16 but forgot to tag it
>>>>> properly.
>>>> Oh yes.  However,
>>>>
>>>> 1. I also wrote this:
>>>>
>>>>>> I am finding it difficult to see the wood for the trees.
>>>>>> It would be really helpful if these vpci fixes were collected
>>>>>> together into a series.
>>>> Can someone please confirm whether this is the only vpci-related patch
>>>> that ought to be on my radar for 4.16 ?
>>>>
>>>> 2. I have not had a reply to my question on Wednesday in
>>>> <24953.34635.645112.279110@xxxxxxxxxxxxxxxxxxxxxxxx>:
>>>>
>>>>     Um, can you explain what the practical impact is of not taking this
>>>>     patch for 4.16 ?  As I understand it vpci for ARM is non-functional in
>>>>     4.16 and this is not expected to change ?  So there would be no
>>>>     benefit to users, and taking the patch would add small but nonzero
>>>>     risk ?
>>>>
>>>> I need this information to decide whether a release-ack is
>>>> appropriate.
>>>>
>>>> Note that we are in code freeze so all patches, including bugfixes,
>>>> need my ack.
>>> Hi Ian,
>>>
>>> This patch [1] is a straightforward 2 lines fix for vpci on ARM.  There
>>> is no risk for the release as the source file affected only builds when
>>> CONFIG_HAS_VPCI is enabled, and it is currently disabled on ARM.
>>>
>>> At the same time, as we know vpci is not complete in 4.16 anyway, so the
>>> counter argument is that we don't need to fix it.
>>>
>>> Given how trivial the fix is, and that it cannot break the build or
>>> runtime, I would take it.
>> Thank you,
>> I can re-send the patch with the updated commit message (Julien),
>> but I still have no R-b's for the patch, so not sure if it is worth it
>
> I can't speak for the others. In my case, I didn't give my reviewed-by 
> because the commit message needs to be updated. If you resend, I will have 
> another look.
Sure
>
> Cheers,
>
Thanks,
Oleksandr