Re: [PATCH v5 04/28] mtd: spi-nor: swp: Improve locking user experience

[Miquel Raynal]

Hi Pratyush,

>> I know what the maintainer load can be, sometimes it does not play well
>> with the rest of the your personal and professional duties. But the
>> series has already been on the list for about 8 months, it's been looked
>> at by other people, the ones who had enough time to dedicate to it. From
>> my perspective, asking such contributions to wait indefinitely and then
>> suggesting partial application without a technical reason is not a
>> sustainable way to handle contributions. This series has not moved much,
>> it could have been applied *much* earlier. I've now addressed most of
>> the comments from Sashiko, v6 is coming, further improving the quality
>> for sure, as there were bugs - there are always. I am of course happy to
>> address further technical concerns, if there are any, but I would
>> strongly prefer merging the series as the coherent set it was intended
>> to be, rather than only taking the preparatory parts.
>
> Unfortunately I am only a patch monkey for SPI NOR these days and do not
> have any time to do reviews, especially for big series. I mainly check
> for reviewed patches and try to apply them. Even that is becoming harder
> these days since I am more busy at new $DAYJOB.

I fully understand that.

> I understand your frustration, and do take the blame for this, but
> unfortunately can't promise anything better in the future. We pretty
> much don't have any active reviewers in SPI NOR. Michael and Tudor are
> also short on time these days. If your employer cares about SPI NOR,
> perhaps you can encourage them to support developers in helping out with
> reviews and maintenance.

Well, I believe I already take a faire share in reviewing and
maintaining the subsystem :-) I even started getting a look into spi-nor
patches recently since I ramped up on the subsystem.

Michel and Tudor regularly make useful feedback, Takahiro is ramping up,
I try to keep enough availability for maintaining NAND, SPI NAND and
anything that is MTD but not in one of the three main subsystems. I
don't think we are in a bad maintainership situation.

> Applying the reviewed patches reduces the total patchset size and does
> make things easier to review and land. So I don't get why you are so
> opposed to the idea.

Applying series partially is okay; specifically for this series,
patches 1-8 could have been taken alone, but a bit earlier IMHO.

> But anyway, I'll bite the bullet here and apply the series. You
> generally have high quality patches so I am not super worried about this
> to begin with. We can deal with the bugs or shortcomings later I
> suppose...

Just to be clear, I am not asking for a fast path here. We need to
follow the contribution and review model, because that is the saniest
model that worked well for us until now. I am however opposed to the
statut quo situation that was progressively settling because of the lack
of availability.

Also, I am always following closely when patches get applied, so I can
respond quickly in case of breakage.

Thanks for your work Pratyush, don't get me wrong, your work is highly
appreciated, and your availability is a parameter we will comply with.

Cheers ;-)
Miquèl