My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Guennadi Liakhovetski]

Hi Chris

There are a number of MMC patches from me outstanding for 3.6. 
Appreciating, that it might not be very easy to keep on overview - what 
latest versions are and in which order they should be applied - would you 
like me to either assemble a list of patches with links to ML archives or 
maybe set up a git branch somewhere on github? In the latter case - would 
you require some signing process - if it is at all supported on github. 
No, I never used one, but I'd have a look if you prefer me to use one.

We still have some time until 3.6 merge window, but it would be nice to 
let patches simmer for a while in -next.

Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/

Re: My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Chris Ball]

Hi Guennadi,

On Thu, Jun 28 2012, Guennadi Liakhovetski wrote:
> There are a number of MMC patches from me outstanding for 3.6. 
> Appreciating, that it might not be very easy to keep on overview - what 
> latest versions are and in which order they should be applied - would you 
> like me to either assemble a list of patches with links to ML archives or 
> maybe set up a git branch somewhere on github? In the latter case - would 
> you require some signing process - if it is at all supported on github. 
> No, I never used one, but I'd have a look if you prefer me to use one.
>
> We still have some time until 3.6 merge window, but it would be nice to 
> let patches simmer for a while in -next.

Thanks, a github branch based on mmc-next would be great, and you don't
need to sign it.

If you have gpg setup but don't want to learn how to use git signed tags
yet, something you could do is sign the e-mail that "git request-pull"
generates that says "pull from url <foo> to go from commit <hash>
to a new HEAD hash of <hash>".

(This effectively signs all the commits, because applying all of the
correct commits is the only way to get to that new hash, and I can
verify that I have that hash as my new HEAD after pulling.)

- Chris.
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child

Re: My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Guennadi Liakhovetski]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Chris

On Thu, 28 Jun 2012, Chris Ball wrote:

> Hi Guennadi,
> 
> On Thu, Jun 28 2012, Guennadi Liakhovetski wrote:
> > There are a number of MMC patches from me outstanding for 3.6. 
> > Appreciating, that it might not be very easy to keep on overview - what 
> > latest versions are and in which order they should be applied - would you 
> > like me to either assemble a list of patches with links to ML archives or 
> > maybe set up a git branch somewhere on github? In the latter case - would 
> > you require some signing process - if it is at all supported on github. 
> > No, I never used one, but I'd have a look if you prefer me to use one.
> >
> > We still have some time until 3.6 merge window, but it would be nice to 
> > let patches simmer for a while in -next.
> 
> Thanks, a github branch based on mmc-next would be great, and you don't
> need to sign it.
> 
> If you have gpg setup but don't want to learn how to use git signed tags
> yet, something you could do is sign the e-mail that "git request-pull"
> generates that says "pull from url <foo> to go from commit <hash>
> to a new HEAD hash of <hash>".
> 
> (This effectively signs all the commits, because applying all of the
> correct commits is the only way to get to that new hash, and I can
> verify that I have that hash as my new HEAD after pulling.)

Ok, let's try, no guarantee though :) Let me know if there are any 
problems with my signing exercise.

The following changes since commit ed99260f748d4c1894a23a4519fd59156a27a308:

  mmc: tmio: remove a duplicated comment line (2012-06-26 16:10:40 -0400)

are available in the git repository at:
  https://github.com/lyakh/linux.git mmc-for-3.6

Guennadi Liakhovetski (19):
      mmc: always save error codes in mmc_regulator_get_supply()
      mmc: sh_mmcif: simplify and use meaningful label names in error-handling
      mmc: sh_mmcif: fix clock management
      mmc: sh_mmcif: re-read the clock frequency every time the clock is turned on
      mmc: sh_mmcif: remove redundant .down_pwr() callback
      mmc: sh_mmcif: add regulator support
      mmc: extend and rename cd-gpio helpers to handle more slot GPIO functions
      mmc: use a more generic name for slot function types and fields
      mmc: add two capability flags for CD and WP signal polarity
      mmc: add CD GPIO polling support to slot functions
      mmc: convert slot functions to managed allocation
      mmc: add WP pin handler to slot functions
      mmc: tmio: support caps2 flags
      mmc: sh_mobile_sdhi: support caps2 flags
      mmc: tmio: use generic GPIO CD and WP handlers
      mmc: sdhi: add OF support, make platform data optional
      mmc: sh-mmcif: add OF support, make platform data optional
      mmc: prohibit card detection when host is not ready
      mmc: sh_mmcif: support generic card-detection

 drivers/mmc/core/Makefile          |    2 +-
 drivers/mmc/core/cd-gpio.c         |   83 ----------------
 drivers/mmc/core/core.c            |    7 +-
 drivers/mmc/core/host.c            |    6 +
 drivers/mmc/core/slot-gpio.c       |  188 ++++++++++++++++++++++++++++++++++++
 drivers/mmc/host/sh_mmcif.c        |  171 ++++++++++++++++++++++----------
 drivers/mmc/host/sh_mobile_sdhi.c  |   36 +++++---
 drivers/mmc/host/tmio_mmc_pio.c    |   13 ++-
 include/linux/mfd/tmio.h           |    1 +
 include/linux/mmc/cd-gpio.h        |   18 ----
 include/linux/mmc/host.h           |   22 ++++-
 include/linux/mmc/sh_mmcif.h       |    2 +
 include/linux/mmc/sh_mobile_sdhi.h |    1 +
 include/linux/mmc/slot-gpio.h      |   24 +++++
 14 files changed, 398 insertions(+), 176 deletions(-)
 delete mode 100644 drivers/mmc/core/cd-gpio.c
 create mode 100644 drivers/mmc/core/slot-gpio.c
 delete mode 100644 include/linux/mmc/cd-gpio.h
 create mode 100644 include/linux/mmc/slot-gpio.h

Thanks
Guennadi

> 
> - Chris.
> -- 
> Chris Ball   <cjb@laptop.org>   <http://printf.net/>
> One Laptop Per Child
> 

- ---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEARECAAYFAk/tw60ACgkQU6Nrc+zHHwhG1QCghrAxwVEmedkIoQPpwfH9de5k
/90An2Ww9OeOxayGnUr9pbtACcLZwS5G
=UtWG
-----END PGP SIGNATURE-----

Re: My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Chris Ball]

Hi,

On Fri, Jun 29 2012, Guennadi Liakhovetski wrote:
> Hi Chris
>
> On Thu, 28 Jun 2012, Chris Ball wrote:
>> If you have gpg setup but don't want to learn how to use git signed tags
>> yet, something you could do is sign the e-mail that "git request-pull"
>> generates that says "pull from url <foo> to go from commit <hash>
>> to a new HEAD hash of <hash>".
>> 
>> (This effectively signs all the commits, because applying all of the
>> correct commits is the only way to get to that new hash, and I can
>> verify that I have that hash as my new HEAD after pulling.)
>
> Ok, let's try, no guarantee though :) Let me know if there are any 
> problems with my signing exercise.
>
> The following changes since commit ed99260f748d4c1894a23a4519fd59156a27a308:
>
>   mmc: tmio: remove a duplicated comment line (2012-06-26 16:10:40 -0400)
>
> are available in the git repository at:
>   https://github.com/lyakh/linux.git mmc-for-3.6

Close, but you didn't tell me what the new HEAD hash should be after
pulling.  This means that the patches could be modified on github
and as long as the diffstat doesn't change, there's no way for me
to know that I'm not pulling what you think I'm pulling.

So, just tell me (in a signed mail) what the top commit's hash is for
you, and that'll prove I'm getting what you're sending.

Thanks,

- Chris.
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child

Re: My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Guennadi Liakhovetski]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Chris

On Fri, 29 Jun 2012, Chris Ball wrote:

> Hi,
> 
> On Fri, Jun 29 2012, Guennadi Liakhovetski wrote:
> > Hi Chris
> >
> > On Thu, 28 Jun 2012, Chris Ball wrote:
> >> If you have gpg setup but don't want to learn how to use git signed tags
> >> yet, something you could do is sign the e-mail that "git request-pull"
> >> generates that says "pull from url <foo> to go from commit <hash>
> >> to a new HEAD hash of <hash>".
> >> 
> >> (This effectively signs all the commits, because applying all of the
> >> correct commits is the only way to get to that new hash, and I can
> >> verify that I have that hash as my new HEAD after pulling.)
> >
> > Ok, let's try, no guarantee though :) Let me know if there are any 
> > problems with my signing exercise.
> >
> > The following changes since commit ed99260f748d4c1894a23a4519fd59156a27a308:
> >
> >   mmc: tmio: remove a duplicated comment line (2012-06-26 16:10:40 -0400)
> >
> > are available in the git repository at:
> >   https://github.com/lyakh/linux.git mmc-for-3.6
> 
> Close, but you didn't tell me what the new HEAD hash should be after
> pulling.  This means that the patches could be modified on github
> and as long as the diffstat doesn't change, there's no way for me
> to know that I'm not pulling what you think I'm pulling.
> 
> So, just tell me (in a signed mail) what the top commit's hash is for
> you, and that'll prove I'm getting what you're sending.

Ah, ok, right, sorry, I thought, it would be a part of the "git 
request-pull" output. So, the top commit has hash

7c4cf5b31b98f51ebfc8db89b022e4f52f93f5db

Hope, this is what's required :-)

Thanks
Guennadi
- ---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEARECAAYFAk/t96IACgkQU6Nrc+zHHwgIUgCeNZFGp71QoqiZ1VB2oE+Pe5KA
VlcAoJwdgZ59L6l52QBxuC8CYFSSvddX
=WTuB
-----END PGP SIGNATURE-----

Re: My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Chris Ball]

Hi Guennadi,

On Fri, Jun 29 2012, Guennadi Liakhovetski wrote:
> Ah, ok, right, sorry, I thought, it would be a part of the "git 
> request-pull" output.

It is for newer versions of git, FWIW.

> So, the top commit has hash
>
> 7c4cf5b31b98f51ebfc8db89b022e4f52f93f5db
>
> Hope, this is what's required :-)

Yes, perfect -- pulled and pushed out to mmc-next.  Thanks!

Could we get a file in Documentation/devicetree/bindings/mmc/ for the
new sh_mmcif DT bindings you've added, please?  It can come as a new
patch on top of these.  I know it'll be very short right now; you can
at least show an example of a DT node that works.

- Chris.
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child

Re: My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Guennadi Liakhovetski]

On Fri, 29 Jun 2012, Chris Ball wrote:

> Hi Guennadi,
> 
> On Fri, Jun 29 2012, Guennadi Liakhovetski wrote:
> > Ah, ok, right, sorry, I thought, it would be a part of the "git 
> > request-pull" output.
> 
> It is for newer versions of git, FWIW.
> 
> > So, the top commit has hash
> >
> > 7c4cf5b31b98f51ebfc8db89b022e4f52f93f5db
> >
> > Hope, this is what's required :-)
> 
> Yes, perfect -- pulled and pushed out to mmc-next.  Thanks!

Great, thanks!

> Could we get a file in Documentation/devicetree/bindings/mmc/ for the
> new sh_mmcif DT bindings you've added, please?  It can come as a new
> patch on top of these.  I know it'll be very short right now; you can
> at least show an example of a DT node that works.

I added OF support to both sh_mmcif and sh_mobile_sdhi (tmio_mmc), but in 
both cases it's just at the most primitive level. This means, those 
drivers will only use reg and interrupts DT properties, nothing else so 
far. More shall be added in the future. Does this level of support also 
have to be documented?

Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/

Re: My outstanding MMC / SDHI/TMIO / MMCIF patches for 3.6

[Chris Ball]

Hi,

On Sat, Jun 30 2012, Guennadi Liakhovetski wrote:
>> Could we get a file in Documentation/devicetree/bindings/mmc/ for the
>> new sh_mmcif DT bindings you've added, please?  It can come as a new
>> patch on top of these.  I know it'll be very short right now; you can
>> at least show an example of a DT node that works.
>
> I added OF support to both sh_mmcif and sh_mobile_sdhi (tmio_mmc), but in 
> both cases it's just at the most primitive level. This means, those 
> drivers will only use reg and interrupts DT properties, nothing else so 
> far. More shall be added in the future.

Right.

> Does this level of support also have to be documented?

Well, I think the idea is that someone should know how to hook up their
DT to the driver without hunting through the source code; right now they
at least have to read the source to find out that the required
compatible nodes are "renesas,sh-mmcif" and "renesas,shmobile-sdhi".

So the point of adding minimal documentation now would be to show how to
hook it up, and have some placeholder documentation that can be added to
once the driver supports more of the core MMC DT bindings (bus-width,
non-removable, ..).  Seems like it couldn't hurt to add it?

Thanks,

- Chris.
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child