Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Pali Rohár]

On Monday 28 December 2015 15:55:28 Arnd Bergmann wrote:
> On Monday 28 December 2015 15:54:35 Pali Rohár wrote:
> > On Monday 28 December 2015 15:41:01 Arnd Bergmann wrote:
> > > On Monday 28 December 2015 15:28:48 Pali Rohár wrote:
> > > > On Monday 28 December 2015 15:14:50 Arnd Bergmann wrote:
> > > > > On Friday 25 December 2015 13:53:11 Pali Rohár wrote:
> > > > > > On Monday 18 May 2015 17:07:57 Arnd Bergmann wrote:
> > > > > > > On Monday 18 May 2015 08:06:07 Tony Lindgren wrote:
> > > > > > > > * Arnd Bergmann  [150515 14:26]:
> > > > > > > > > On Friday 15 May 2015 23:22:37 Pali Rohár wrote:
> > > > > > > > If setting up the generic binding is expected to take a
> > > > > > > > while, you can naturally pass it in pdata while waiting
> > > > > > > > for the generic binding to get merged.
> > > > > > > 
> > > > > > > Yes, good idea. So the n900 machine can use auxdata to
> > > > > > > pass this for the time being, while the binding and
> > > > > > > generic implementation is being worked out.
> > > > > > 
> > > > > > Ok, so what is needed to finish this patch series?
> > > > > 
> > > > > I don't know where we are at this point. Has either the
> > > > > auxdata approach or the generic binding been worked on at
> > > > > all?
> > > > 
> > > > What are auxdata data?
> > > 
> > > I mean you can add the platform data to the omap_auxdata_lookup[]
> > > table for this board.
> > 
> > But can I mix data from omap3-n900.dts and omap_auxdata_lookup[]?
> 
> Yes.
> 
>   Arnd

Hm... looks like it is not possible. omap_hsmmc driver ignores any 
supplied platform data if there are device tree data. So array 
omap_auxdata_lookup[] does not help.

-- 
Pali Rohár
pali.ro...@gmail.com


signature.asc
Description: This is a digitally signed message part.

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Arnd Bergmann]

On Saturday 02 January 2016 16:22:03 Pali Rohár wrote:
> On Monday 28 December 2015 15:55:28 Arnd Bergmann wrote:
> > On Monday 28 December 2015 15:54:35 Pali Rohár wrote:
> > > > 
> > > > I mean you can add the platform data to the omap_auxdata_lookup[]
> > > > table for this board.
> > > 
> > > But can I mix data from omap3-n900.dts and omap_auxdata_lookup[]?
> > 
> Hm... looks like it is not possible. omap_hsmmc driver ignores any 
> supplied platform data if there are device tree data. So array 
> omap_auxdata_lookup[] does not help.

Obviously you need to the driver to work with that setting. Maybe
something as simple as

diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index e06b1881b6a1..4fa35fc84b8b 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -2006,7 +2006,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
void __iomem *base;
 
match = of_match_device(of_match_ptr(omap_mmc_of_match), >dev);
-   if (match) {
+   if (!pdata && match) {
pdata = of_get_hsmmc_pdata(>dev);
 
if (IS_ERR(pdata))


Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Pali Rohár]

On Saturday 02 January 2016 23:57:47 Arnd Bergmann wrote:
> On Saturday 02 January 2016 16:22:03 Pali Rohár wrote:
> > On Monday 28 December 2015 15:55:28 Arnd Bergmann wrote:
> > > On Monday 28 December 2015 15:54:35 Pali Rohár wrote:
> > > > > I mean you can add the platform data to the
> > > > > omap_auxdata_lookup[] table for this board.
> > > > 
> > > > But can I mix data from omap3-n900.dts and
> > > > omap_auxdata_lookup[]?
> > 
> > Hm... looks like it is not possible. omap_hsmmc driver ignores any
> > supplied platform data if there are device tree data. So array
> > omap_auxdata_lookup[] does not help.
> 
> Obviously you need to the driver to work with that setting. Maybe
> something as simple as
> 
> diff --git a/drivers/mmc/host/omap_hsmmc.c
> b/drivers/mmc/host/omap_hsmmc.c index e06b1881b6a1..4fa35fc84b8b
> 100644
> --- a/drivers/mmc/host/omap_hsmmc.c
> +++ b/drivers/mmc/host/omap_hsmmc.c
> @@ -2006,7 +2006,7 @@ static int omap_hsmmc_probe(struct
> platform_device *pdev) void __iomem *base;
> 
>   match = of_match_device(of_match_ptr(omap_mmc_of_match),
> >dev); -if (match) {
> + if (!pdata && match) {
>   pdata = of_get_hsmmc_pdata(>dev);
> 
>   if (IS_ERR(pdata))
> 
> 
>   Arnd

But in this case I must copy mmc definition from omap3-n900.dts file 
back to board code to omap_auxdata_lookup[]. And mmc definitions in 
omap3-n900.dts will be ignored. This is step back.

Mixing mmc definitions from DTS file together with omap_auxdata_lookup[] 
just will not work.

-- 
Pali Rohár
pali.ro...@gmail.com


signature.asc
Description: This is a digitally signed message part.

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Arnd Bergmann]

On Sunday 03 January 2016 00:03:54 Pali Rohár wrote:
> On Saturday 02 January 2016 23:57:47 Arnd Bergmann wrote:
> > On Saturday 02 January 2016 16:22:03 Pali Rohár wrote:
> > > On Monday 28 December 2015 15:55:28 Arnd Bergmann wrote:
> > > > On Monday 28 December 2015 15:54:35 Pali Rohár wrote:
> > > > > > I mean you can add the platform data to the
> > > > > > omap_auxdata_lookup[] table for this board.
> > > > > 
> > > > > But can I mix data from omap3-n900.dts and
> > > > > omap_auxdata_lookup[]?
> > > 
> > > Hm... looks like it is not possible. omap_hsmmc driver ignores any
> > > supplied platform data if there are device tree data. So array
> > > omap_auxdata_lookup[] does not help.
> > 
> > Obviously you need to the driver to work with that setting. Maybe
> > something as simple as
> > 
> > diff --git a/drivers/mmc/host/omap_hsmmc.c
> > b/drivers/mmc/host/omap_hsmmc.c index e06b1881b6a1..4fa35fc84b8b
> > 100644
> > --- a/drivers/mmc/host/omap_hsmmc.c
> > +++ b/drivers/mmc/host/omap_hsmmc.c
> > @@ -2006,7 +2006,7 @@ static int omap_hsmmc_probe(struct
> > platform_device *pdev) void __iomem *base;
> > 
> >   match = of_match_device(of_match_ptr(omap_mmc_of_match),
> > >dev); -if (match) {
> > + if (!pdata && match) {
> >   pdata = of_get_hsmmc_pdata(>dev);
> > 
> >   if (IS_ERR(pdata))
> > 
> 
> But in this case I must copy mmc definition from omap3-n900.dts file 
> back to board code to omap_auxdata_lookup[]. And mmc definitions in 
> omap3-n900.dts will be ignored. This is step back.
> 
> Mixing mmc definitions from DTS file together with omap_auxdata_lookup[] 
> just will not work.

As I said earlier, if you prefer to avoid the auxdata hack, you are
also welcome to work on support for named slots in the MMC core code,
it will just be more work and will take time to get consensus on.

Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Arnd Bergmann]

On Friday 25 December 2015 13:53:11 Pali Rohár wrote:
> On Monday 18 May 2015 17:07:57 Arnd Bergmann wrote:
> > On Monday 18 May 2015 08:06:07 Tony Lindgren wrote:
> > > * Arnd Bergmann  [150515 14:26]:
> > > > On Friday 15 May 2015 23:22:37 Pali Rohár wrote:
> > > If setting up the generic binding is expected to take a while,
> > > you can naturally pass it in pdata while waiting for the generic
> > > binding to get merged.
> > 
> > Yes, good idea. So the n900 machine can use auxdata to pass this for
> > the time being, while the binding and generic implementation is
> > being worked out.
> > 
> 
> Ok, so what is needed to finish this patch series?

I don't know where we are at this point. Has either the auxdata approach
or the generic binding been worked on at all?

If not, please try to get the auxdata variant to work, it should not be hard
at all because there is no dependency on a generic binding.

Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Pali Rohár]

On Monday 28 December 2015 15:14:50 Arnd Bergmann wrote:
> On Friday 25 December 2015 13:53:11 Pali Rohár wrote:
> > On Monday 18 May 2015 17:07:57 Arnd Bergmann wrote:
> > > On Monday 18 May 2015 08:06:07 Tony Lindgren wrote:
> > > > * Arnd Bergmann  [150515 14:26]:
> > > > > On Friday 15 May 2015 23:22:37 Pali Rohár wrote:
> > > > If setting up the generic binding is expected to take a while,
> > > > you can naturally pass it in pdata while waiting for the
> > > > generic binding to get merged.
> > > 
> > > Yes, good idea. So the n900 machine can use auxdata to pass this
> > > for the time being, while the binding and generic implementation
> > > is being worked out.
> > 
> > Ok, so what is needed to finish this patch series?
> 
> I don't know where we are at this point. Has either the auxdata
> approach or the generic binding been worked on at all?

What are auxdata data? And what do you mean with "generic binding" 
approach?

> If not, please try to get the auxdata variant to work, it should not
> be hard at all because there is no dependency on a generic binding.
> 
>   Arnd


-- 
Pali Rohár
pali.ro...@gmail.com


signature.asc
Description: This is a digitally signed message part.

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Arnd Bergmann]

On Monday 28 December 2015 15:28:48 Pali Rohár wrote:
> On Monday 28 December 2015 15:14:50 Arnd Bergmann wrote:
> > On Friday 25 December 2015 13:53:11 Pali Rohár wrote:
> > > On Monday 18 May 2015 17:07:57 Arnd Bergmann wrote:
> > > > On Monday 18 May 2015 08:06:07 Tony Lindgren wrote:
> > > > > * Arnd Bergmann  [150515 14:26]:
> > > > > > On Friday 15 May 2015 23:22:37 Pali Rohár wrote:
> > > > > If setting up the generic binding is expected to take a while,
> > > > > you can naturally pass it in pdata while waiting for the
> > > > > generic binding to get merged.
> > > > 
> > > > Yes, good idea. So the n900 machine can use auxdata to pass this
> > > > for the time being, while the binding and generic implementation
> > > > is being worked out.
> > > 
> > > Ok, so what is needed to finish this patch series?
> > 
> > I don't know where we are at this point. Has either the auxdata
> > approach or the generic binding been worked on at all?
> 
> What are auxdata data? 

I mean you can add the platform data to the omap_auxdata_lookup[] table
for this board.

> And what do you mean with "generic binding" approach?

Start a discussion to specify slot names and implement that in the
mmc driver core, so we can remove the hack from the OMAP driver and
make it work the same way for any machine.

Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Arnd Bergmann]

On Monday 28 December 2015 15:54:35 Pali Rohár wrote:
> On Monday 28 December 2015 15:41:01 Arnd Bergmann wrote:
> > On Monday 28 December 2015 15:28:48 Pali Rohár wrote:
> > > On Monday 28 December 2015 15:14:50 Arnd Bergmann wrote:
> > > > On Friday 25 December 2015 13:53:11 Pali Rohár wrote:
> > > > > On Monday 18 May 2015 17:07:57 Arnd Bergmann wrote:
> > > > > > On Monday 18 May 2015 08:06:07 Tony Lindgren wrote:
> > > > > > > * Arnd Bergmann  [150515 14:26]:
> > > > > > > > On Friday 15 May 2015 23:22:37 Pali Rohár wrote:
> > > > > > > If setting up the generic binding is expected to take a
> > > > > > > while, you can naturally pass it in pdata while waiting
> > > > > > > for the generic binding to get merged.
> > > > > > 
> > > > > > Yes, good idea. So the n900 machine can use auxdata to pass
> > > > > > this for the time being, while the binding and generic
> > > > > > implementation is being worked out.
> > > > > 
> > > > > Ok, so what is needed to finish this patch series?
> > > > 
> > > > I don't know where we are at this point. Has either the auxdata
> > > > approach or the generic binding been worked on at all?
> > > 
> > > What are auxdata data?
> > 
> > I mean you can add the platform data to the omap_auxdata_lookup[]
> > table for this board.
> 
> But can I mix data from omap3-n900.dts and omap_auxdata_lookup[]?

Yes.

Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Pali Rohár]

On Monday 28 December 2015 15:41:01 Arnd Bergmann wrote:
> On Monday 28 December 2015 15:28:48 Pali Rohár wrote:
> > On Monday 28 December 2015 15:14:50 Arnd Bergmann wrote:
> > > On Friday 25 December 2015 13:53:11 Pali Rohár wrote:
> > > > On Monday 18 May 2015 17:07:57 Arnd Bergmann wrote:
> > > > > On Monday 18 May 2015 08:06:07 Tony Lindgren wrote:
> > > > > > * Arnd Bergmann  [150515 14:26]:
> > > > > > > On Friday 15 May 2015 23:22:37 Pali Rohár wrote:
> > > > > > If setting up the generic binding is expected to take a
> > > > > > while, you can naturally pass it in pdata while waiting
> > > > > > for the generic binding to get merged.
> > > > > 
> > > > > Yes, good idea. So the n900 machine can use auxdata to pass
> > > > > this for the time being, while the binding and generic
> > > > > implementation is being worked out.
> > > > 
> > > > Ok, so what is needed to finish this patch series?
> > > 
> > > I don't know where we are at this point. Has either the auxdata
> > > approach or the generic binding been worked on at all?
> > 
> > What are auxdata data?
> 
> I mean you can add the platform data to the omap_auxdata_lookup[]
> table for this board.

But can I mix data from omap3-n900.dts and omap_auxdata_lookup[]?

-- 
Pali Rohár
pali.ro...@gmail.com


signature.asc
Description: This is a digitally signed message part.

Re: [PATCH 1/2] mmc: omap_hsmmc: Add support for slot-name property in DT

[Pali Rohár]

On Monday 18 May 2015 17:07:57 Arnd Bergmann wrote:
> On Monday 18 May 2015 08:06:07 Tony Lindgren wrote:
> > * Arnd Bergmann  [150515 14:26]:
> > > On Friday 15 May 2015 23:22:37 Pali Rohár wrote:
> > If setting up the generic binding is expected to take a while,
> > you can naturally pass it in pdata while waiting for the generic
> > binding to get merged.
> 
> Yes, good idea. So the n900 machine can use auxdata to pass this for
> the time being, while the binding and generic implementation is
> being worked out.
> 
>   Arnd

Ok, so what is needed to finish this patch series?

-- 
Pali Rohár
pali.ro...@gmail.com


signature.asc
Description: This is a digitally signed message part.