[PATCH 0/2] Regarding NAND core and drivers maintenance

[PATCH 0/2] Regarding NAND core and drivers maintenance

[Ezequiel Garcia]

Hi all,

As it's obvious from the mailing list activity, in the past months Brian
Norris took the NAND maintainership role. In my opinion, he's been doing
a terrific work and as a result the reviewing cycle is now considerably
shorter.

Great work, Brian!

This position is not at all easy, due to each NAND driver being filled
with hardware controller specificities. For that reason, I think it
would be healthy to take another step in setting maintainership
responsabilites.

My proposal (copy-pasted from the media subsystem) is to define
sub-maintainers, one for each NAND driver, in addition to the top-level
NAND subsystem maintainer.

When a patch comes for a given driver, the _ideal_ situation would be
for the sub-maintainer to Ack-it first.

And to formalize this proposal, I'm pushing a couple patches: the first
adds Brian to MAINTAINERS, as the NAND maintainer.

Then, I'm adding myself as sub-maintainer to the pxa3xx-nand driver.
This way, scripts/get_maintainers.pl will show my mail and I'll get
CCed by occasional or new contributors.

How does this sound?

Ezequiel Garcia (2):
  mtd: Add NAND device drivers to MAINTAINERS
  mtd: Add PXA3xx NAND driver to MAINTAINERS

 MAINTAINERS | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

-- 
1.8.1.5

[PATCH 1/2] mtd: Add NAND device drivers to MAINTAINERS

[Ezequiel Garcia]

Add drivers/mtd/nand/ to MAINTAINERS file with Brian Norris
as the subsystem maintainer.

Cc: Brian Norris <computersforpeace@gmail.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
---
 MAINTAINERS | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 167e188..d4145f7 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5554,6 +5554,18 @@ F:	drivers/mtd/
 F:	include/linux/mtd/
 F:	include/uapi/mtd/
 
+NAND DEVICE DRIVERS
+M:	Brian Norris <computersforpeace@gmail.com>
+L:	linux-mtd@lists.infradead.org
+W:	http://www.linux-mtd.infradead.org/
+Q:	http://patchwork.ozlabs.org/project/linux-mtd/list/
+T:	git http://git.infradead.org/l2-mtd.git
+S:	Maintained
+F:	drivers/mtd/nand/
+F:	include/linux/mtd/nand.h
+F:	include/linux/mtd/nand_ecc.h
+F:	include/linux/mtd/nand_bch.h
+
 MEN A21 WATCHDOG DRIVER
 M:  	Johannes Thumshirn <johannes.thumshirn@men.de>
 L:	linux-watchdog@vger.kernel.org
-- 
1.8.1.5

Re: [PATCH 1/2] mtd: Add NAND device drivers to MAINTAINERS

[Brian Norris]

On Wed, Nov 20, 2013 at 09:20:08AM -0300, Ezequiel Garcia wrote:
> Add drivers/mtd/nand/ to MAINTAINERS file with Brian Norris
> as the subsystem maintainer.
> 
> Cc: Brian Norris <computersforpeace@gmail.com>
> Cc: David Woodhouse <dwmw2@infradead.org>
> Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
> ---
>  MAINTAINERS | 12 ++++++++++++
>  1 file changed, 12 insertions(+)
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 167e188..d4145f7 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -5554,6 +5554,18 @@ F:	drivers/mtd/
>  F:	include/linux/mtd/
>  F:	include/uapi/mtd/
>  
> +NAND DEVICE DRIVERS

While it would be nice for MTD and NAND to be listed together, I think
this file is in alphabetical order ("NAND" doesn't come between "MTD"
and "MEN").

Also, I might recommend including "FLASH" in the title: NAND FLASH
DEVICE DRIVERS.

> +M:	Brian Norris <computersforpeace@gmail.com>
> +L:	linux-mtd@lists.infradead.org
> +W:	http://www.linux-mtd.infradead.org/
> +Q:	http://patchwork.ozlabs.org/project/linux-mtd/list/
> +T:	git http://git.infradead.org/l2-mtd.git

The git tree is an interesting case. l2-mtd.git was really only set up
as a secondary tree, not as the official tree. It has moved to being the
primary place for development, though, so your entry does reflect
reality...

> +S:	Maintained
> +F:	drivers/mtd/nand/
> +F:	include/linux/mtd/nand.h
> +F:	include/linux/mtd/nand_ecc.h
> +F:	include/linux/mtd/nand_bch.h

include/linux/mtd/nand*.h?
include/linux/mtd/bbm.h?

> +
>  MEN A21 WATCHDOG DRIVER
>  M:  	Johannes Thumshirn <johannes.thumshirn@men.de>
>  L:	linux-watchdog@vger.kernel.org

Brian

Re: [PATCH 1/2] mtd: Add NAND device drivers to MAINTAINERS

[Ezequiel Garcia]

On Thu, Nov 21, 2013 at 10:53:14AM -0800, Brian Norris wrote:
> On Wed, Nov 20, 2013 at 09:20:08AM -0300, Ezequiel Garcia wrote:
> > Add drivers/mtd/nand/ to MAINTAINERS file with Brian Norris
> > as the subsystem maintainer.
> > 
> > Cc: Brian Norris <computersforpeace@gmail.com>
> > Cc: David Woodhouse <dwmw2@infradead.org>
> > Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
> > ---
> >  MAINTAINERS | 12 ++++++++++++
> >  1 file changed, 12 insertions(+)
> > 
> > diff --git a/MAINTAINERS b/MAINTAINERS
> > index 167e188..d4145f7 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -5554,6 +5554,18 @@ F:	drivers/mtd/
> >  F:	include/linux/mtd/
> >  F:	include/uapi/mtd/
> >  
> > +NAND DEVICE DRIVERS
> 
> While it would be nice for MTD and NAND to be listed together, I think
> this file is in alphabetical order ("NAND" doesn't come between "MTD"
> and "MEN").
> 

Right, I haven't noticed order.

> Also, I might recommend including "FLASH" in the title: NAND FLASH
> DEVICE DRIVERS.
> 

Sure...

> > +M:	Brian Norris <computersforpeace@gmail.com>
> > +L:	linux-mtd@lists.infradead.org
> > +W:	http://www.linux-mtd.infradead.org/
> > +Q:	http://patchwork.ozlabs.org/project/linux-mtd/list/
> > +T:	git http://git.infradead.org/l2-mtd.git
> 
> The git tree is an interesting case. l2-mtd.git was really only set up
> as a secondary tree, not as the official tree. It has moved to being the
> primary place for development, though, so your entry does reflect
> reality...
> 

Maybe create a new git repository, specially for this?

> > +S:	Maintained
> > +F:	drivers/mtd/nand/
> > +F:	include/linux/mtd/nand.h
> > +F:	include/linux/mtd/nand_ecc.h
> > +F:	include/linux/mtd/nand_bch.h
> 
> include/linux/mtd/nand*.h?
> include/linux/mtd/bbm.h?
> 

Yes, that's probably better.
-- 
Ezequiel García, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com

Re: [PATCH 1/2] mtd: Add NAND device drivers to MAINTAINERS

[Brian Norris]

On Thu, Nov 21, 2013 at 04:11:18PM -0300, Ezequiel Garcia wrote:
> On Thu, Nov 21, 2013 at 10:53:14AM -0800, Brian Norris wrote:
> > On Wed, Nov 20, 2013 at 09:20:08AM -0300, Ezequiel Garcia wrote:
> > > Add drivers/mtd/nand/ to MAINTAINERS file with Brian Norris
> > > as the subsystem maintainer.
> > > 
> > > Cc: Brian Norris <computersforpeace@gmail.com>
> > > Cc: David Woodhouse <dwmw2@infradead.org>
> > > Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
> > > ---
> > >  MAINTAINERS | 12 ++++++++++++
> > >  1 file changed, 12 insertions(+)
> > > 
> > > diff --git a/MAINTAINERS b/MAINTAINERS
> > > index 167e188..d4145f7 100644
> > > --- a/MAINTAINERS
> > > +++ b/MAINTAINERS
> > > @@ -5554,6 +5554,18 @@ F:	drivers/mtd/
...
> > > +M:	Brian Norris <computersforpeace@gmail.com>
> > > +L:	linux-mtd@lists.infradead.org
> > > +W:	http://www.linux-mtd.infradead.org/
> > > +Q:	http://patchwork.ozlabs.org/project/linux-mtd/list/
> > > +T:	git http://git.infradead.org/l2-mtd.git
> > 
> > The git tree is an interesting case. l2-mtd.git was really only set up
> > as a secondary tree, not as the official tree. It has moved to being the
> > primary place for development, though, so your entry does reflect
> > reality...
> > 
> 
> Maybe create a new git repository, specially for this?

I think we're due for more consolidation rather than adding new git
trees, so I'd still list linux-mtd.git as the official NAND repo. Plus,
David has previously mentioned that I could eventually just be
committing straight to linux-mtd.git in the future. As it stands now, I
have commit access to it, and for 3.13-rc1, I just pushed l2-mtd.git
straight into linux-mtd.git and then to Linus. So the multi-repository
make-up may eventually make less sense.

> > > +S:	Maintained
> > > +F:	drivers/mtd/nand/
> > > +F:	include/linux/mtd/nand.h
> > > +F:	include/linux/mtd/nand_ecc.h
> > > +F:	include/linux/mtd/nand_bch.h

Brian

[PATCH 2/2] mtd: Add PXA3xx NAND driver to MAINTAINERS

[Ezequiel Garcia]

Add the pxa3xx-nand driver which supports PXA3xx and Armada 370/XP SoC
family to MAINTAINERS to ensure I get proper Cc on patches.

Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
---
 MAINTAINERS | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index d4145f7..e596b93 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5566,6 +5566,12 @@ F:	include/linux/mtd/nand.h
 F:	include/linux/mtd/nand_ecc.h
 F:	include/linux/mtd/nand_bch.h
 
+PXA3xx NAND DRIVER
+M:	Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
+L:	linux-mtd@lists.infradead.org
+S:	Maintained
+F:	drivers/mtd/nand/pxa3xx-nand.c
+
 MEN A21 WATCHDOG DRIVER
 M:  	Johannes Thumshirn <johannes.thumshirn@men.de>
 L:	linux-watchdog@vger.kernel.org
-- 
1.8.1.5

Re: [PATCH 0/2] Regarding NAND core and drivers maintenance

[Brian Norris]

On Wed, Nov 20, 2013 at 09:20:07AM -0300, Ezequiel Garcia wrote:
> As it's obvious from the mailing list activity, in the past months Brian
> Norris took the NAND maintainership role. In my opinion, he's been doing
> a terrific work and as a result the reviewing cycle is now considerably
> shorter.
> 
> Great work, Brian!

Thanks!

> This position is not at all easy, due to each NAND driver being filled
> with hardware controller specificities. For that reason, I think it
> would be healthy to take another step in setting maintainership
> responsabilites.
> 
> My proposal (copy-pasted from the media subsystem) is to define
> sub-maintainers, one for each NAND driver, in addition to the top-level
> NAND subsystem maintainer.

I believe the offering is open, for any who are in a situation to
maintain particular drivers. You may note that there are a few NAND
drivers which are listed as "maintained" by others, either under their
corresponding ARM architecture, or standalone. And as you mention, this
arrangement is common in some other subsystems.

I don't suppose that every NAND driver will need a dedicated maintainer,
but active ones which have interested, knowledgeable developers might be
good candidates.

> When a patch comes for a given driver, the _ideal_ situation would be
> for the sub-maintainer to Ack-it first.
> 
> And to formalize this proposal, I'm pushing a couple patches: the first
> adds Brian to MAINTAINERS, as the NAND maintainer.
> 
> Then, I'm adding myself as sub-maintainer to the pxa3xx-nand driver.
> This way, scripts/get_maintainers.pl will show my mail and I'll get
> CCed by occasional or new contributors.
> 
> How does this sound?

For the idea:

Acked-by: Brian Norris <computersforpeace@gmail.com>

I think patch 1 would need a little tweaking, though.

I'd also like David's or Artem's comments. I know that Artem, for one,
claimed to not want to put himself in an official role as MTD
maintainer, even though he was taking much of this responsibility
previously.

Brian

RE: [PATCH 0/2] Regarding NAND core and drivers maintenance

[Gupta, Pekon]

> From: Ezequiel Garcia
> As it's obvious from the mailing list activity, in the past months Brian
> Norris took the NAND maintainership role. In my opinion, he's been doing
> a terrific work and as a result the reviewing cycle is now considerably
> shorter.
> 
> Great work, Brian!
> 
> This position is not at all easy, due to each NAND driver being filled
> with hardware controller specificities. For that reason, I think it
> would be healthy to take another step in setting maintainership
> responsabilites.
> 
Absolutely Agree.. Thanks Brian..

> My proposal (copy-pasted from the media subsystem) is to define
> sub-maintainers, one for each NAND driver, in addition to the top-level
> NAND subsystem maintainer.
> 
> When a patch comes for a given driver, the _ideal_ situation would be
> for the sub-maintainer to Ack-it first.
> 
Specifically, we need someone to keep a watch on m25p80 (apart from
David and Brian). This is because I see lot of patches coming in but not
on generic framework approach. Serial Flash driver might be next big
component of MTD sub-system in coming times.


with regards, pekon

Re: [PATCH 0/2] Regarding NAND core and drivers maintenance

[Huang Shijie]

于 2013年11月22日 15:07, Gupta, Pekon 写道:
> on generic framework approach. Serial Flash driver might be next big
> component of MTD sub-system in coming times.
I am coding it now. maybe sent out next week. it's a big change.

thanks
Huang Shijie

Re: [PATCH 0/2] Regarding NAND core and drivers maintenance

[Brian Norris]

On Fri, Nov 22, 2013 at 07:07:09AM +0000, Pekon Gupta wrote:
> > My proposal (copy-pasted from the media subsystem) is to define
> > sub-maintainers, one for each NAND driver, in addition to the top-level
> > NAND subsystem maintainer.
> > 
> > When a patch comes for a given driver, the _ideal_ situation would be
> > for the sub-maintainer to Ack-it first.
> > 
> Specifically, we need someone to keep a watch on m25p80 (apart from
> David and Brian). This is because I see lot of patches coming in but not
> on generic framework approach. Serial Flash driver might be next big
> component of MTD sub-system in coming times.

Hmm, I don't know if we really need a separate SPI NOR maintainer, and I
don't know if we have anyone with enough experience with SPI NOR to
qualify anyway. I know Marek Vasut has been a good reviewer, but I'd be
surprised if he really wanted to take official responsibility. And
anyway, implementors and reviewers don't have to be official
maintainers, so this shouldn't stop any pending work on the subject.

Also, I have some HW that can do quad-SPI, but only in limited
circumstances. Right now, my driver is just faking being a generic SPI
driver when it really scrapes the commands and performs accelerated
transactions on ones it recognizes... so it shares some similarities
with what Huang is trying to do. I hope to contribute more to the
generic framework of SPI NOR in the future (as time allows, of course).

So while I might be officially responsible for driver/mtd/nand/, I will
be involved in the rest of MTD and will likely continue to queue up
patches anyway.

Brian

Re: [PATCH 0/2] Regarding NAND core and drivers maintenance

[Marek Vasut]

Hi Brian,

> On Fri, Nov 22, 2013 at 07:07:09AM +0000, Pekon Gupta wrote:
> > > My proposal (copy-pasted from the media subsystem) is to define
> > > sub-maintainers, one for each NAND driver, in addition to the top-level
> > > NAND subsystem maintainer.
> > > 
> > > When a patch comes for a given driver, the _ideal_ situation would be
> > > for the sub-maintainer to Ack-it first.
> > 
> > Specifically, we need someone to keep a watch on m25p80 (apart from
> > David and Brian). This is because I see lot of patches coming in but not
> > on generic framework approach. Serial Flash driver might be next big
> > component of MTD sub-system in coming times.
> 
> Hmm, I don't know if we really need a separate SPI NOR maintainer, and I
> don't know if we have anyone with enough experience with SPI NOR to
> qualify anyway. I know Marek Vasut has been a good reviewer, but I'd be
> surprised if he really wanted to take official responsibility. And
> anyway, implementors and reviewers don't have to be official
> maintainers, so this shouldn't stop any pending work on the subject.

I'm honestly pleased by you saying this ;-) I don't mind continue reviewing, but 
if things were to blow out of proportion, I don't mind helping either.

> Also, I have some HW that can do quad-SPI, but only in limited
> circumstances. Right now, my driver is just faking being a generic SPI
> driver when it really scrapes the commands and performs accelerated
> transactions on ones it recognizes... so it shares some similarities
> with what Huang is trying to do. I hope to contribute more to the
> generic framework of SPI NOR in the future (as time allows, of course).
> 
> So while I might be officially responsible for driver/mtd/nand/, I will
> be involved in the rest of MTD and will likely continue to queue up
> patches anyway.

Thanks for that !

Best regards,
Marek Vasut