[RFCv3 0/7] OMAP SSI driver

[RFCv3 0/7] OMAP SSI driver

[Sebastian Reichel]

Hi,

Here is the third round of the OMAP SSI driver patches.

Changes since RFCv2:
* Move DT data from omap34xx.dtsi to omap3.dtsi
  Nokia N9 uses an omap3630 and ssi for modem connection, thus SSI
  seems to be part of all OMAP3 SoCs.
* Move DT documentation to the driver patch
* Update wording in DT documentation
  I just applied Stephen Warrens suggestions
* Removal of omap_pm_get_dev_context_loss_count as demanded by Tony Lindgren
  For now there is no context loss support in the driver, since I do not
  have enough information about the chip to support this via registers.
* Dropped useless "default n" from KConfig
* Add support for HSI client registration via DT
* Register hsi-char as ssi client for Nokia N900
  Currently only hsi-char client is available in the mainline kernel,
  but for the Nokia N900 two more clients will be needed: ssi-protocol
  and cmt-speech. Those will be added with their own patchset.
* Use ida_simple_get for device id, since DT does not setup pd->id
* Fixed SSI interrupt numbers in omap3.dtsi
  I subtracted OMAP_INTC_START before, but the original patch did
  not yet have OMAP_INTC_START added.
* Fixed compilation as kernel modules
* Fixed reload of kernel modules
* Added module dependency: omap_ssi_port needs omap_port

TODO:
* Central Message Queue
  I did not yet implement a central message queue in the HSI framework.
  I will do this after Nokia N900 modem is working in the mainline kernel,
  since that makes testing easier.
* Get clock data from DT
  This depends on a patchset, which is not yet part of any kernel tree.
  Apart from that I don't know which clocks will be needed. That depends
  on the implementation of the hwmod DTification (see next point)
* Remove the hwmod DT hack
  This depends on some future work merging hwmod data into DT.

Version 2 of the patchset: https://lkml.org/lkml/2013/9/15/98

-- Sebastian

Sebastian Reichel (7):
  ARM: OMAP2+: hwmod-data: Add SSI information
  HSI: add Device Tree support for HSI clients
  HSI: method to unregister clients from an hsi port
  HSI: hsi-char: add Device Tree support
  HSI: hsi-char: fix driver for multiport scenarios
  ARM: OMAP2+: HSI: Introduce OMAP SSI driver
  ARM: dts: N900: Add SSI information

 Documentation/devicetree/bindings/hsi/omap_ssi.txt |   69 +
 .../devicetree/bindings/hsi/trivial-devices.txt    |   36 +
 arch/arm/boot/dts/omap3-n900.dts                   |   12 +
 arch/arm/boot/dts/omap3.dtsi                       |   47 +
 arch/arm/mach-omap2/omap_hwmod_3xxx_data.c         |   48 +
 drivers/hsi/Kconfig                                |    1 +
 drivers/hsi/Makefile                               |    1 +
 drivers/hsi/clients/hsi_char.c                     |   15 +-
 drivers/hsi/controllers/Kconfig                    |   19 +
 drivers/hsi/controllers/Makefile                   |    6 +
 drivers/hsi/controllers/omap_ssi.c                 |  618 +++++++++
 drivers/hsi/controllers/omap_ssi.h                 |  166 +++
 drivers/hsi/controllers/omap_ssi_port.c            | 1400 ++++++++++++++++++++
 drivers/hsi/controllers/omap_ssi_regs.h            |  171 +++
 drivers/hsi/hsi.c                                  |   80 +-
 include/dt-bindings/hsi/hsi.h                      |   17 +
 include/linux/hsi/hsi.h                            |    3 +
 17 files changed, 2707 insertions(+), 2 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/hsi/omap_ssi.txt
 create mode 100644 Documentation/devicetree/bindings/hsi/trivial-devices.txt
 create mode 100644 drivers/hsi/controllers/Kconfig
 create mode 100644 drivers/hsi/controllers/Makefile
 create mode 100644 drivers/hsi/controllers/omap_ssi.c
 create mode 100644 drivers/hsi/controllers/omap_ssi.h
 create mode 100644 drivers/hsi/controllers/omap_ssi_port.c
 create mode 100644 drivers/hsi/controllers/omap_ssi_regs.h
 create mode 100644 include/dt-bindings/hsi/hsi.h

-- 
1.8.4.rc3

[RFCv3 0/7] OMAP SSI driver

[Tony Lindgren]

* Sebastian Reichel <sre@debian.org> [131006 13:36]:
> Hi,
> 
> Here is the third round of the OMAP SSI driver patches.

Thanks for updating these, they look good to me now:

Acked-by: Tony Lindgren <tony@atomide.com>

[RFCv3 1/7] ARM: OMAP2+: hwmod-data: Add SSI information

[Paul Walmsley]

On Sun, 6 Oct 2013, Sebastian Reichel wrote:

> This patch adds Synchronous Serial Interface (SSI) hwmod support for
> OMAP34xx SoCs.
> 
> Signed-off-by: Sebastian Reichel <sre@debian.org>

Thanks, queued this one for v3.13.  You can drop it from any future 
reposts of this series.


- Paul

[RFCv3 1/7] ARM: OMAP2+: hwmod-data: Add SSI information

[Sebastian Reichel]

On Wed, Oct 09, 2013 at 05:43:13AM +0000, Paul Walmsley wrote:
> On Sun, 6 Oct 2013, Sebastian Reichel wrote:
> 
> > This patch adds Synchronous Serial Interface (SSI) hwmod support for
> > OMAP34xx SoCs.
> > 
> > Signed-off-by: Sebastian Reichel <sre@debian.org>
> 
> Thanks, queued this one for v3.13.  You can drop it from any future 
> reposts of this series.

Thanks.

-- Sebastian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131010/647ee67a/attachment-0001.sig>

[RFCv3 0/7] OMAP SSI driver

[Sebastian Reichel]

Hi,

On Tue, Oct 08, 2013 at 10:28:15AM -0700, Tony Lindgren wrote:
> * Sebastian Reichel <sre@debian.org> [131006 13:36]:
> > Here is the third round of the OMAP SSI driver patches.
> 
> Thanks for updating these, they look good to me now:
> 
> Acked-by: Tony Lindgren <tony@atomide.com>

Ok. I guess it's time to check how this gets included in the kernel.

The first patch has already been queued and the last patch could
probably go through Beno?t once the DT people have reviewed the
updated patches. The other ones would normally go through the
hsi tree, but that one was maintained by Carlos Chinea and is 
now more or less orphaned.

My suggestion is:

1. I create another patch, which adds entries for hsi subsystem and
   ssi driver to MAINTAINERS.

   It would be nice to have some co-maintainers for the subsystem.
   Linus Walleij maybe? I think you created the launchpad blueprint [1]?
   Especially because I have no documentation/specification for HSI, which
   brings me to my next slightly off topic request.

   I would like to have access to SSI documentation and HSI specification.
   Can the guys from Texas Instruments please give me some pointers how I
   can get access to the NDA'd documentation? I couldn't find any hints
   on ti.com.

2. I create a patch adding Carlos Chinea to CREDITS

3. I create a git repository and put all of the patches in there, so
   that they can be pulled into linux-next.

4. I add a gpg/pgp signed tag with the key used to signed this mail.

Any issues with this or any other suggestions how to proceed? And if following this
- should I request an account on kernel.org to host the repository there,
or should I just upload it at a random place?

[0] https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/tree/Next/Trees
[1] https://blueprints.launchpad.net/linux-linaro/+spec/hsi-subsystem-maint

-- Sebastian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131010/6f548040/attachment.sig>

[RFCv3 0/7] OMAP SSI driver

[Tony Lindgren]

* Sebastian Reichel <sre@debian.org> [131010 10:30]:
> Hi,
> 
> On Tue, Oct 08, 2013 at 10:28:15AM -0700, Tony Lindgren wrote:
> > * Sebastian Reichel <sre@debian.org> [131006 13:36]:
> > > Here is the third round of the OMAP SSI driver patches.
> > 
> > Thanks for updating these, they look good to me now:
> > 
> > Acked-by: Tony Lindgren <tony@atomide.com>
> 
> Ok. I guess it's time to check how this gets included in the kernel.
> 
> The first patch has already been queued and the last patch could
> probably go through Beno?t once the DT people have reviewed the
> updated patches. The other ones would normally go through the
> hsi tree, but that one was maintained by Carlos Chinea and is 
> now more or less orphaned.

Yes, I suggest you post the .dts changes completely seprately
so driver maintainers don't accidentally pick those up.
 
> My suggestion is:
> 
> 1. I create another patch, which adds entries for hsi subsystem and
>    ssi driver to MAINTAINERS.
> 
>    It would be nice to have some co-maintainers for the subsystem.
>    Linus Walleij maybe? I think you created the launchpad blueprint [1]?
>    Especially because I have no documentation/specification for HSI, which
>    brings me to my next slightly off topic request.
> 
>    I would like to have access to SSI documentation and HSI specification.
>    Can the guys from Texas Instruments please give me some pointers how I
>    can get access to the NDA'd documentation? I couldn't find any hints
>    on ti.com.
> 
> 2. I create a patch adding Carlos Chinea to CREDITS
> 
> 3. I create a git repository and put all of the patches in there, so
>    that they can be pulled into linux-next.
> 
> 4. I add a gpg/pgp signed tag with the key used to signed this mail.
> 
> Any issues with this or any other suggestions how to proceed? And if following this
> - should I request an account on kernel.org to host the repository there,
> or should I just upload it at a random place?

That all sounds good to me. It is a *good thing* to have a maintainer
for a driver subsystem, and if you're willing to pick it up, that's
the way to go.

Regards,

Tony
 
> [0] https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/tree/Next/Trees
> [1] https://blueprints.launchpad.net/linux-linaro/+spec/hsi-subsystem-maint
> 
> -- Sebastian

[RFCv3 0/7] OMAP SSI driver

[Aaro Koskinen]

Hi,

On Thu, Oct 10, 2013 at 07:21:30PM +0200, Sebastian Reichel wrote:
> Any issues with this or any other suggestions how to proceed?

Maybe you could provide some brief instructions/description of how this
was tested?

Thanks,

A.

[RFCv3 0/7] OMAP SSI driver

[Sebastian Reichel]

On Thu, Oct 10, 2013 at 09:28:01PM +0300, Aaro Koskinen wrote:
> On Thu, Oct 10, 2013 at 07:21:30PM +0200, Sebastian Reichel wrote:
> > Any issues with this or any other suggestions how to proceed?
> 
> Maybe you could provide some brief instructions/description of how this
> was tested?

Sure. I used Carlos' original patch, which has been tested to be
working. Then I checked with some debug prints and the debugfs
support what the registers and resources should look like.
Next I updated the patch to use the new kernel frameworks and
made sure everything is behaves as before.

For actually using the modem of the Nokia N900 three more drivers
are needed in the mainline kernel in addtion to this one:

1. cmt

 This is a small driver handling the modem's reset line, which
 is connected via a gpio pin to the system.

2. ssi_protocol
 
 This is an hsi client driver, which actually talks with the modem
 and exports an phonet interface for userspace access. It makes
 use of the cmt driver.

 I plan on porting these to the new kernel frameworks next. If
 I'm not mistaken this should be enough to communicate with the
 modem.

3. cmt_speech

 This is an hsi client driver, which takes care of interchanging
 speech data with the modem. This one is needed for calling. I
 will have a look at it once the basic stuff is working.

P.S.: You can get a mainline kernel status matrix for the Nokia
      N900 on this page: http://elinux.org/N900

-- Sebastian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131010/c6047523/attachment-0001.sig>

[RFCv3 0/7] OMAP SSI driver

[Aaro Koskinen]

On Thu, Oct 10, 2013 at 10:02:36PM +0200, Sebastian Reichel wrote:
> P.S.: You can get a mainline kernel status matrix for the Nokia
>       N900 on this page: http://elinux.org/N900

Thanks for summary, and the above page looks very useful. I wonder
would it make sense to add also N9/N950 there (some functionality is
shared)? Also TWL4030 DT status could be updated, e.g. watchdog support
was done already last year with 8899b8d93ec64b7a8e54807a68a958e1206535e2.

A.

[RFCv3 0/7] OMAP SSI driver

[Sebastian Reichel]

On Thu, Oct 10, 2013 at 11:19:14PM +0300, Aaro Koskinen wrote:
> On Thu, Oct 10, 2013 at 10:02:36PM +0200, Sebastian Reichel wrote:
> > P.S.: You can get a mainline kernel status matrix for the Nokia
> >       N900 on this page: http://elinux.org/N900
> 
> Thanks for summary,

You're welcome.

> and the above page looks very useful. I wonder would it make sense
> to add also N9/N950 there (some functionality is shared)?

mh maybe. I haven't looked @ N9/N950 so far, since I don't own them.
I would be interested in a N950, but I don't own one and it's almost
impossible to get one :(

Apart from that I'm not sure if it's a good idea to make the table
even more complex. Maybe just add a new page?

> Also TWL4030 DT status could be updated, e.g. watchdog support
> was done already last year with 8899b8d93ec64b7a8e54807a68a958e1206535e2.

Yes. I only recently added the DT column and have not yet checked
all the drivers. Probably it's also a good idea to split this
column into "driver supports DT" and "DT file supports it".

-- Sebastian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131011/6e3fd846/attachment.sig>

[RFCv3 0/7] OMAP SSI driver

[Linus Walleij]

On Thu, Oct 10, 2013 at 7:21 PM, Sebastian Reichel <sre@debian.org> wrote:

> 1. I create another patch, which adds entries for hsi subsystem and
>    ssi driver to MAINTAINERS.
>
>    It would be nice to have some co-maintainers for the subsystem.
>    Linus Walleij maybe? I think you created the launchpad blueprint [1]?

Well yeah I did... but I do not think this is big enough a subsystem to
require two maintainers right now, also I have no easily available HSI
hardware that I can use to test things. (I think.)

Yours,
Linus Walleij

[RFCv3 0/7] OMAP SSI driver

[Sebastian Reichel]

Hi,

On Sun, Oct 06, 2013 at 10:27:08PM +0200, Sebastian Reichel wrote:
> Here is the third round of the OMAP SSI driver patches.

ping!

I'm still waiting for feedback from DT maintainers.
Can you review the changes and Ack them or tell me how to proceed?

-- Sebastian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131021/59571ce5/attachment.sig>

[RFCv3 7/7] ARM: dts: N900: Add SSI information

[Benoit Cousson]

Hi Sebastian,

Thanks to Tony, I've just realized that I missed all your patches, that 
for some reason ended-up in my junk folder :-(

That being said, I cannot apply any of your DTS patches! What base 
branch are you using?

Could you repost all your pending DTS patches in one series rebased on 
top of my for_3.13/dts?

Thanks,
Benoit

On 06/10/2013 22:27, Sebastian Reichel wrote:
> Add SSI device tree data for OMAP3 and Nokia N900.
>
> Signed-off-by: Sebastian Reichel <sre@debian.org>
> ---
>   arch/arm/boot/dts/omap3-n900.dts | 12 ++++++++++
>   arch/arm/boot/dts/omap3.dtsi     | 47 ++++++++++++++++++++++++++++++++++++++++
>   2 files changed, 59 insertions(+)
>
> diff --git a/arch/arm/boot/dts/omap3-n900.dts b/arch/arm/boot/dts/omap3-n900.dts
> index 0582356..0fbb77e 100644
> --- a/arch/arm/boot/dts/omap3-n900.dts
> +++ b/arch/arm/boot/dts/omap3-n900.dts
> @@ -186,6 +186,18 @@
>   	power = <50>;
>   };
>
> +&ssi_port1 {
> +	ti,ssi-cawake-gpio = <&gpio5 23 GPIO_ACTIVE_HIGH>; /* 151 */
> +
> +	ssi-char {
> +		compatible = "ssi-char";
> +	};
> +};
> +
> +&ssi_port2 {
> +	status = "disabled";
> +};
> +
>   &uart1 {
>   	pinctrl-names = "default";
>   	pinctrl-0 = <&uart1_pins>;
> diff --git a/arch/arm/boot/dts/omap3.dtsi b/arch/arm/boot/dts/omap3.dtsi
> index 7d95cda..9f60b82 100644
> --- a/arch/arm/boot/dts/omap3.dtsi
> +++ b/arch/arm/boot/dts/omap3.dtsi
> @@ -532,5 +532,52 @@
>   			num-eps = <16>;
>   			ram-bits = <12>;
>   		};
> +
> +		ssi: ssi-controller at 48058000 {
> +			compatible = "ti,omap3-ssi";
> +			ti,hwmods = "ssi";
> +
> +			reg = <0x48058000 0x1000>,
> +			      <0x48059000 0x1000>;
> +			reg-names = "sys",
> +				    "gdd";
> +
> +			interrupts = <71>;
> +			interrupt-names = "gdd_mpu";
> +
> +			#address-cells = <1>;
> +			#size-cells = <1>;
> +			ranges;
> +
> +			ssi_port1: ssi-port at 0 {
> +				compatible = "ti,omap3-ssi-port";
> +
> +				reg = <0x4805a000 0x800>,
> +				      <0x4805a800 0x800>;
> +				reg-names = "tx",
> +					    "rx";
> +
> +				interrupt-parent = <&intc>;
> +				interrupts = <67>,
> +					     <68>;
> +				interrupt-names = "mpu_irq0",
> +						  "mpu_irq1";
> +			};
> +
> +			ssi_port2: ssi-port at 1 {
> +				compatible = "ti,omap3-ssi-port";
> +
> +				reg = <0x4805b000 0x800>,
> +				      <0x4805b800 0x800>;
> +				reg-names = "tx",
> +					    "rx";
> +
> +				interrupt-parent = <&intc>;
> +				interrupts = <69>,
> +					     <70>;
> +				interrupt-names = "mpu_irq0",
> +						  "mpu_irq1";
> +			};
> +		};
>   	};
>   };
>


-- 
Beno?t Cousson
BayLibre
Embedded Linux Technology Lab
www.baylibre.com

[RFCv3 7/7] ARM: dts: N900: Add SSI information

[Sebastian Reichel]

Hi Benoit,

On Tue, Oct 22, 2013 at 02:48:45PM +0200, Benoit Cousson wrote:
> Thanks to Tony, I've just realized that I missed all your patches,
> that for some reason ended-up in my junk folder :-(

oh :( Can you check why they ended up in junk? I would like to
prevent more mails going there.

> That being said, I cannot apply any of your DTS patches! What base
> branch are you using?

Currently torvalds/master with pavel's n900 patch and some more
patches from me.

> Could you repost all your pending DTS patches in one series rebased
> on top of my for_3.13/dts?

Sure. DT maintainers haven't given feedback since the last patchset
[0]. Do you want to wait for feedback from them?

[0] https://lkml.org/lkml/2013/9/23/529

-- Sebastian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131022/d075c54f/attachment.sig>

[RFCv3 7/7] ARM: dts: N900: Add SSI information

[Benoit Cousson]

Hi Sebastian,

On 22/10/2013 16:44, Sebastian Reichel wrote:
> Hi Benoit,
>
> On Tue, Oct 22, 2013 at 02:48:45PM +0200, Benoit Cousson wrote:
>> Thanks to Tony, I've just realized that I missed all your patches,
>> that for some reason ended-up in my junk folder :-(
>
> oh :( Can you check why they ended up in junk? I would like to
> prevent more mails going there.
>
>> That being said, I cannot apply any of your DTS patches! What base
>> branch are you using?
>
> Currently torvalds/master with pavel's n900 patch and some more
> patches from me.

Mmm, looks like I missed these patches :-)

>
>> Could you repost all your pending DTS patches in one series rebased
>> on top of my for_3.13/dts?
>
> Sure. DT maintainers haven't given feedback since the last patchset
> [0]. Do you want to wait for feedback from them?

OK, maybe not the SSI series that is still RFC, but I guess all the 
others ones are fine to be merged.

Could you repost the whole N900 DTS patches (beside SSI) in one series 
rebased on top of mine?

Thanks,
Benoit

>
> [0] https://lkml.org/lkml/2013/9/23/529
>
> -- Sebastian
>


-- 
Beno?t Cousson
BayLibre
Embedded Linux Technology Lab
www.baylibre.com

[RFCv3 0/7] OMAP SSI driver

[Pavel Machek]

Hi!

> 3. cmt_speech
> 
>  This is an hsi client driver, which takes care of interchanging
>  speech data with the modem. This one is needed for calling. I
>  will have a look at it once the basic stuff is working.
> 
> P.S.: You can get a mainline kernel status matrix for the Nokia
>       N900 on this page: http://elinux.org/N900

Thanks a lot for the page. I did not realize how well is N900 supported in
latest mainline...

What do you use as userland? Is there specific userland 3.13 is expected
to work with?
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html