Re: [PATCH 2/2] MAINTAINERS: Mark "vhost-user-scmi" orphaned

"Michael S. Tsirkin" <[email protected]> writes:

> On Mon, Jul 13, 2026 at 10:17:07AM +0200, Markus Armbruster wrote:
>> Philippe Mathieu-Daudé <[email protected]> writes:
>> 
>> > On 13/7/26 08:52, Michael S. Tsirkin wrote:
>> >> On Mon, Jul 13, 2026 at 08:13:38AM +0200, Philippe Mathieu-Daudé wrote:
>> >>> On 11/7/26 10:00, Michael S. Tsirkin wrote:
>> >>>> On Fri, Jul 10, 2026 at 01:43:05PM +0200, Markus Armbruster wrote:
>> >>>>> Commit a5dab090e1 (hw/virtio: Add boilerplate for vhost-user-scmi
>> >>>>> device, 2023-06-28) added the section with S: Supported, but no M:.
>> >>>>> This makes no sense.  Downgrade to S: Orphan.
>> >>>>>
>> >>>>> Signed-off-by: Markus Armbruster <[email protected]>
>> >>>>
>> >>>>
>> >>>> No, this makes sense. The reason this section was added is so that Milan can
>> >>>> get copied on vhost-user-scmi but not other virtio stuff:
>> >>>>
>> >>>> ./scripts/get_maintainer.pl -f hw/virtio/vhost-user-scmi.c
>> >>>> Milan Zamazal <[email protected]> (reviewer:vhost-user-scmi)
>> >>>> "Michael S. Tsirkin" <[email protected]> (supporter:vhost)
>> >>>> Stefano Garzarella <[email protected]> (reviewer:vhost)
>> >>>> [email protected] (open list:All patches CC here)
>> >>>>
>> >>>> Pls drop this patch.
>> >>>>
>> >>>>> ---
>> >>>>>    MAINTAINERS | 2 +-
>> >>>>>    1 file changed, 1 insertion(+), 1 deletion(-)
>> >>>>>
>> >>>>> diff --git a/MAINTAINERS b/MAINTAINERS
>> >>>>> index aa2df9cbe3..4d558ddee9 100644
>> >>>>> --- a/MAINTAINERS
>> >>>>> +++ b/MAINTAINERS
>> >>>>> @@ -2622,7 +2622,7 @@ F: include/hw/virtio/vhost-user-snd.h
>> >>>>>    vhost-user-scmi
>> >>>
>> >>> So what we want is to add here:
>> >>>
>> >>>    M: Michael S. Tsirkin <[email protected]>
>> >>>    R: Stefano Garzarella <[email protected]>
>> >>
>> >> But these files already get M from the match against hw/virtio/ -
>> >> what am I missing?
>> >
>> > Yeah, just trying to make Markus script happy ;)
>> 
>> Appreciate the kindness!
>> 
>> In decreasing importance:
>> 
>> 1. Correct get_maintainers.pl output
>
> Is anything wrong with it?

No, it looks good to me.

>> 2. Legible MAINTAINERS
>
> We can always come up with a new rule for such cases,
> where we have a reviewer just for some specific files
> in a project, but the supporters are from the bigger
> project.
>
> How about we drop S: completely in such cases?

Could that be differently confusing?  Honest question!

1. Correct get_maintainers.pl output

   Appears to be fine.  As far as I can tell, it accumulates S: for all
   matching sections.  A matching section without S: should be harmless
   as long as another matching section provides one.

   By the way, this is another reason why my patch was bad.  The S:
   Orphan gets accumulated, and can lead to problematic output, e.g.

    $ scripts/get_maintainer.pl --status -f hw/virtio/vhost-user-scmi.c 
    Milan Zamazal <[email protected]> (reviewer:vhost-user-scmi)
    "Michael S. Tsirkin" <[email protected]> (supporter:vhost)
    Stefano Garzarella <[email protected]> (reviewer:vhost)
    [email protected] (open list:All patches CC here)
    Orphan
    Supported

2. Legible MAINTAINERS

   S: Supported is without supporters is mildly confusing unless you
   understand that the supporters are listed elsewhere.  Evidence: it
   confused me.

   No S: looks like we forgot to add it.  Feels less bad to me, but I
   might be biased.  Could it use a comment?  Not sure.

   Hmm, S: Supported could even be misleading.  Say Milan had agreed to
   look after this in his spare time.  That's status Maintained.  We'd
   use M: to record his commitment.  But what S: to use?

   S: Maintained gives M: the correct meaning, but mislead readers about
   the status of the matched files: these are still Supported.

   S: Supported is locally correct about the status of the matched
   files, but misleads about the M:.

   I think in this case we'd use M: Maintained as the lesser evil.

   The reliable way to find out status is running get_maintainers.pl.

> We can also tweak the name:
> vhost-user-scmi (part of vhost)

I guess the "vhost-" in "vhost-user-scmi" is enough of a hint.

>> 99. My script is happy.
>> 
>> [...]