[U-Boot] [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board

[U-Boot] [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board

[Matwey V. Kornilov]

Hello,

I am running Zynq Z-turn board and I face the following issue with MMC
initialization in SPL.
With u-boot master, I see the message similar to the following:

U-Boot SPL 2019.07-00352-gb5f3eb3393 (Jul 31 2019 - 20:03:42 +0300)
mmc boot
Trying to boot from MMC1

Then, the u-boot waits forever. I've tried to add debug prints and
found that execution is stalled somewhere inside mmc_init().
Using bisect I've found that the following broken commit is the following:

commit 3d296365e4e8823c7c0d4b568fa7accfae4bf895 (refs/bisect/bad)
Author: Faiz Abbas <faiz_abbas@ti.com>
Date:   Tue Jun 11 00:43:34 2019 +0530

    mmc: sdhci: Add support for sdhci-caps-mask

    Add Support for masking some bits in the capabilities
    register of a host controller.

    Also remove the redundant readl() into caps1.

    Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
    Reviewed-by: Tom Rini <trini@konsulko.com>

Until that commit the behavior was the following:

U-Boot SPL 2019.07-00351-g889a4dfc55 (Jul 31 2019 - 20:01:41 +0300)
mmc boot
Trying to boot from MMC1
spl_load_image_fat_os: error reading image system.dtb, err - -2
spl_load_image_fat: error reading image u-boot.img, err - -2
SPL: failed to boot from all boot devices
### ERROR ### Please RESET the board ###

There were no u-boot.img at the SD card while testing, so this error
message is expected here.
5456935a1da3 ("ARM: zynq: Add configuration for Z-turn board") was
applied at the top of every testing commit to allow the board
initialization in SPL.

What could be wrong with that commit and how could I fix the board?

-- 
With best regards,
Matwey V. Kornilov

[U-Boot] [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board

[Peng Fan]

> Subject: [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board
> 
> Hello,
> 
> I am running Zynq Z-turn board and I face the following issue with MMC
> initialization in SPL.
> With u-boot master, I see the message similar to the following:
> 
> U-Boot SPL 2019.07-00352-gb5f3eb3393 (Jul 31 2019 - 20:03:42 +0300) mmc
> boot Trying to boot from MMC1
> 
> Then, the u-boot waits forever. I've tried to add debug prints and found that
> execution is stalled somewhere inside mmc_init().
> Using bisect I've found that the following broken commit is the following:

https://gitlab.denx.de/u-boot/custodians/u-boot-mmc/commit/41a9fab8dac841afb70a059668abaf14d47cdf59

Would this patch help?

Regards,
Peng.

> 
> commit 3d296365e4e8823c7c0d4b568fa7accfae4bf895 (refs/bisect/bad)
> Author: Faiz Abbas <faiz_abbas@ti.com>
> Date:   Tue Jun 11 00:43:34 2019 +0530
> 
>     mmc: sdhci: Add support for sdhci-caps-mask
> 
>     Add Support for masking some bits in the capabilities
>     register of a host controller.
> 
>     Also remove the redundant readl() into caps1.
> 
>     Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
>     Reviewed-by: Tom Rini <trini@konsulko.com>
> 
> Until that commit the behavior was the following:
> 
> U-Boot SPL 2019.07-00351-g889a4dfc55 (Jul 31 2019 - 20:01:41 +0300) mmc
> boot Trying to boot from MMC1
> spl_load_image_fat_os: error reading image system.dtb, err - -2
> spl_load_image_fat: error reading image u-boot.img, err - -2
> SPL: failed to boot from all boot devices ### ERROR ### Please RESET the
> board ###
> 
> There were no u-boot.img at the SD card while testing, so this error message
> is expected here.
> 5456935a1da3 ("ARM: zynq: Add configuration for Z-turn board") was applied
> at the top of every testing commit to allow the board initialization in SPL.
> 
> What could be wrong with that commit and how could I fix the board?
> 
> --
> With best regards,
> Matwey V. Kornilov

[U-Boot] [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board

[Faiz Abbas]

Hi,

On 01/08/19 7:51 AM, Peng Fan wrote:
>> Subject: [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board
>>
>> Hello,
>>
>> I am running Zynq Z-turn board and I face the following issue with MMC
>> initialization in SPL.
>> With u-boot master, I see the message similar to the following:
>>
>> U-Boot SPL 2019.07-00352-gb5f3eb3393 (Jul 31 2019 - 20:03:42 +0300) mmc
>> boot Trying to boot from MMC1
>>
>> Then, the u-boot waits forever. I've tried to add debug prints and found that
>> execution is stalled somewhere inside mmc_init().
>> Using bisect I've found that the following broken commit is the following:
> 
> https://gitlab.denx.de/u-boot/custodians/u-boot-mmc/commit/41a9fab8dac841afb70a059668abaf14d47cdf59
> 
> Would this patch help?
> 

Yeah, this mostly looks like a NULL pointer access issue.

Thanks,
Faiz

[U-Boot] [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board

[Matwey V. Kornilov]

Hi all,

чт, 1 авг. 2019 г. в 09:48, Faiz Abbas <faiz_abbas@ti.com>:
>
> Hi,
>
> On 01/08/19 7:51 AM, Peng Fan wrote:
> >> Subject: [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board
> >>
> >> Hello,
> >>
> >> I am running Zynq Z-turn board and I face the following issue with MMC
> >> initialization in SPL.
> >> With u-boot master, I see the message similar to the following:
> >>
> >> U-Boot SPL 2019.07-00352-gb5f3eb3393 (Jul 31 2019 - 20:03:42 +0300) mmc
> >> boot Trying to boot from MMC1
> >>
> >> Then, the u-boot waits forever. I've tried to add debug prints and found that
> >> execution is stalled somewhere inside mmc_init().
> >> Using bisect I've found that the following broken commit is the following:
> >
> > https://gitlab.denx.de/u-boot/custodians/u-boot-mmc/commit/41a9fab8dac841afb70a059668abaf14d47cdf59
> >
> > Would this patch help?
> >
>
> Yeah, this mostly looks like a NULL pointer access issue.

Unfortunately, 41a9fab8 doesn't help me.... :-(
Behavior remains the same.

>
> Thanks,
> Faiz



-- 
With best regards,
Matwey V. Kornilov

[U-Boot] [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board

[Faiz Abbas]

Hi Matwey,

On 01/08/19 1:03 PM, Matwey V. Kornilov wrote:
> Hi all,
> 
> чт, 1 авг. 2019 г. в 09:48, Faiz Abbas <faiz_abbas@ti.com>:
>>
>> Hi,
>>
>> On 01/08/19 7:51 AM, Peng Fan wrote:
>>>> Subject: [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board
>>>>
>>>> Hello,
>>>>
>>>> I am running Zynq Z-turn board and I face the following issue with MMC
>>>> initialization in SPL.
>>>> With u-boot master, I see the message similar to the following:
>>>>
>>>> U-Boot SPL 2019.07-00352-gb5f3eb3393 (Jul 31 2019 - 20:03:42 +0300) mmc
>>>> boot Trying to boot from MMC1
>>>>
>>>> Then, the u-boot waits forever. I've tried to add debug prints and found that
>>>> execution is stalled somewhere inside mmc_init().
>>>> Using bisect I've found that the following broken commit is the following:
>>>
>>> https://gitlab.denx.de/u-boot/custodians/u-boot-mmc/commit/41a9fab8dac841afb70a059668abaf14d47cdf59
>>>
>>> Would this patch help?
>>>
>>
>> Yeah, this mostly looks like a NULL pointer access issue.
> 
> Unfortunately, 41a9fab8 doesn't help me.... :-(
> Behavior remains the same.
> 

You need to implement the same thing with your driver (I assume you are
using zynq_sdhci.c?). sdhci_setup_cfg() requires a valid host->mmc->dev.
Move that assignment to above the sdhci_set_cfg() call in your driver.

Thanks,
Faiz

[U-Boot] [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board

[Matwey V. Kornilov]

чт, 1 авг. 2019 г. в 10:40, Faiz Abbas <faiz_abbas@ti.com>:
>
> Hi Matwey,
>
> On 01/08/19 1:03 PM, Matwey V. Kornilov wrote:
> > Hi all,
> >
> > чт, 1 авг. 2019 г. в 09:48, Faiz Abbas <faiz_abbas@ti.com>:
> >>
> >> Hi,
> >>
> >> On 01/08/19 7:51 AM, Peng Fan wrote:
> >>>> Subject: [BISECTED] [BUG]: MMC initialization hang at Zynq Z-turn board
> >>>>
> >>>> Hello,
> >>>>
> >>>> I am running Zynq Z-turn board and I face the following issue with MMC
> >>>> initialization in SPL.
> >>>> With u-boot master, I see the message similar to the following:
> >>>>
> >>>> U-Boot SPL 2019.07-00352-gb5f3eb3393 (Jul 31 2019 - 20:03:42 +0300) mmc
> >>>> boot Trying to boot from MMC1
> >>>>
> >>>> Then, the u-boot waits forever. I've tried to add debug prints and found that
> >>>> execution is stalled somewhere inside mmc_init().
> >>>> Using bisect I've found that the following broken commit is the following:
> >>>
> >>> https://gitlab.denx.de/u-boot/custodians/u-boot-mmc/commit/41a9fab8dac841afb70a059668abaf14d47cdf59
> >>>
> >>> Would this patch help?
> >>>
> >>
> >> Yeah, this mostly looks like a NULL pointer access issue.
> >
> > Unfortunately, 41a9fab8 doesn't help me.... :-(
> > Behavior remains the same.
> >
>
> You need to implement the same thing with your driver (I assume you are
> using zynq_sdhci.c?). sdhci_setup_cfg() requires a valid host->mmc->dev.
> Move that assignment to above the sdhci_set_cfg() call in your driver.
>

Now it works. Thank you a lot! I'll prepare fixing patch for
zynq_sdhci.c shortly.

> Thanks,
> Faiz
>


-- 
With best regards,
Matwey V. Kornilov

[U-Boot] [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe

[Matwey V. Kornilov]

Since commit 3d296365e4e8 ("mmc: sdhci: Add support for
sdhci-caps-mask") sdhci_setup_cfg() expects a valid sdhci_host mmc
field. Move the mmc field initialization before sdhci_setup_cfg()
call to avoid crash on mmc pointer dereference.

[this patch is based on commit 41a9fab8dac8 ("mmc: mv_sdhci: fix
uninitialized pointer deref on probe") by Baruch Siach]

Fixes: 3d296365e4e8 ("mmc: sdhci: Add support for sdhci-caps-mask")
Cc: Faiz Abbas <faiz_abbas@ti.com>
Cc: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
---
 drivers/mmc/zynq_sdhci.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/mmc/zynq_sdhci.c b/drivers/mmc/zynq_sdhci.c
index c525084250..3225a7ac93 100644
--- a/drivers/mmc/zynq_sdhci.c
+++ b/drivers/mmc/zynq_sdhci.c
@@ -242,13 +242,14 @@ static int arasan_sdhci_probe(struct udevice *dev)
 
 	host->max_clk = clock;
 
+	host->mmc = &plat->mmc;
+	host->mmc->dev = dev;
+	host->mmc->priv = host;
+
 	ret = sdhci_setup_cfg(&plat->cfg, host, plat->f_max,
 			      CONFIG_ZYNQ_SDHCI_MIN_FREQ);
-	host->mmc = &plat->mmc;
 	if (ret)
 		return ret;
-	host->mmc->priv = host;
-	host->mmc->dev = dev;
 	upriv->mmc = host->mmc;
 
 	return sdhci_probe(dev);
-- 
2.16.4

[U-Boot] [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe

[Peng Fan]

> Subject: [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe
> 
> Since commit 3d296365e4e8 ("mmc: sdhci: Add support for
> sdhci-caps-mask") sdhci_setup_cfg() expects a valid sdhci_host mmc field.
> Move the mmc field initialization before sdhci_setup_cfg() call to avoid crash
> on mmc pointer dereference.
> 
> [this patch is based on commit 41a9fab8dac8 ("mmc: mv_sdhci: fix
> uninitialized pointer deref on probe") by Baruch Siach]
> 
> Fixes: 3d296365e4e8 ("mmc: sdhci: Add support for sdhci-caps-mask")
> Cc: Faiz Abbas <faiz_abbas@ti.com>
> Cc: Baruch Siach <baruch@tkos.co.il>
> Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
> ---
>  drivers/mmc/zynq_sdhci.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/mmc/zynq_sdhci.c b/drivers/mmc/zynq_sdhci.c index
> c525084250..3225a7ac93 100644
> --- a/drivers/mmc/zynq_sdhci.c
> +++ b/drivers/mmc/zynq_sdhci.c
> @@ -242,13 +242,14 @@ static int arasan_sdhci_probe(struct udevice *dev)
> 
>  	host->max_clk = clock;
> 
> +	host->mmc = &plat->mmc;
> +	host->mmc->dev = dev;
> +	host->mmc->priv = host;
> +
>  	ret = sdhci_setup_cfg(&plat->cfg, host, plat->f_max,
>  			      CONFIG_ZYNQ_SDHCI_MIN_FREQ);
> -	host->mmc = &plat->mmc;
>  	if (ret)
>  		return ret;
> -	host->mmc->priv = host;
> -	host->mmc->dev = dev;
>  	upriv->mmc = host->mmc;
> 
>  	return sdhci_probe(dev);

Reviewed-by: Peng Fan <peng.fan@nxp.com>

> --
> 2.16.4

[U-Boot] [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe

[Michal Simek]

On 01. 08. 19 17:00, Matwey V. Kornilov wrote:
> Since commit 3d296365e4e8 ("mmc: sdhci: Add support for
> sdhci-caps-mask") sdhci_setup_cfg() expects a valid sdhci_host mmc
> field. Move the mmc field initialization before sdhci_setup_cfg()
> call to avoid crash on mmc pointer dereference.
> 
> [this patch is based on commit 41a9fab8dac8 ("mmc: mv_sdhci: fix
> uninitialized pointer deref on probe") by Baruch Siach]
> 
> Fixes: 3d296365e4e8 ("mmc: sdhci: Add support for sdhci-caps-mask")
> Cc: Faiz Abbas <faiz_abbas@ti.com>
> Cc: Baruch Siach <baruch@tkos.co.il>
> Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
> ---
>  drivers/mmc/zynq_sdhci.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/mmc/zynq_sdhci.c b/drivers/mmc/zynq_sdhci.c
> index c525084250..3225a7ac93 100644
> --- a/drivers/mmc/zynq_sdhci.c
> +++ b/drivers/mmc/zynq_sdhci.c
> @@ -242,13 +242,14 @@ static int arasan_sdhci_probe(struct udevice *dev)
>  
>  	host->max_clk = clock;
>  
> +	host->mmc = &plat->mmc;
> +	host->mmc->dev = dev;
> +	host->mmc->priv = host;
> +
>  	ret = sdhci_setup_cfg(&plat->cfg, host, plat->f_max,
>  			      CONFIG_ZYNQ_SDHCI_MIN_FREQ);
> -	host->mmc = &plat->mmc;
>  	if (ret)
>  		return ret;
> -	host->mmc->priv = host;
> -	host->mmc->dev = dev;
>  	upriv->mmc = host->mmc;
>  
>  	return sdhci_probe(dev);
> 

Would be good to mentioned that this was tested on Z-turn board.

Anyway I have tested it on zcu102 and issue is not visible there but it
is visible on zc706. That's why I expect this is issue only for Zynq device.

Anyway
Tested-by: Michal Simek <michal.simek@xilinx.com> (on zcu102/zc706)

Peng: Feel free to take it via your tree or I will take it.

Thanks,
Michal

[U-Boot] [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe

[Peng Fan]

> Subject: Re: [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe
> 
> On 01. 08. 19 17:00, Matwey V. Kornilov wrote:
> > Since commit 3d296365e4e8 ("mmc: sdhci: Add support for
> > sdhci-caps-mask") sdhci_setup_cfg() expects a valid sdhci_host mmc
> > field. Move the mmc field initialization before sdhci_setup_cfg() call
> > to avoid crash on mmc pointer dereference.
> >
> > [this patch is based on commit 41a9fab8dac8 ("mmc: mv_sdhci: fix
> > uninitialized pointer deref on probe") by Baruch Siach]
> >
> > Fixes: 3d296365e4e8 ("mmc: sdhci: Add support for sdhci-caps-mask")
> > Cc: Faiz Abbas <faiz_abbas@ti.com>
> > Cc: Baruch Siach <baruch@tkos.co.il>
> > Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
> > ---
> >  drivers/mmc/zynq_sdhci.c | 7 ++++---
> >  1 file changed, 4 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/mmc/zynq_sdhci.c b/drivers/mmc/zynq_sdhci.c index
> > c525084250..3225a7ac93 100644
> > --- a/drivers/mmc/zynq_sdhci.c
> > +++ b/drivers/mmc/zynq_sdhci.c
> > @@ -242,13 +242,14 @@ static int arasan_sdhci_probe(struct udevice
> > *dev)
> >
> >  	host->max_clk = clock;
> >
> > +	host->mmc = &plat->mmc;
> > +	host->mmc->dev = dev;
> > +	host->mmc->priv = host;
> > +
> >  	ret = sdhci_setup_cfg(&plat->cfg, host, plat->f_max,
> >  			      CONFIG_ZYNQ_SDHCI_MIN_FREQ);
> > -	host->mmc = &plat->mmc;
> >  	if (ret)
> >  		return ret;
> > -	host->mmc->priv = host;
> > -	host->mmc->dev = dev;
> >  	upriv->mmc = host->mmc;
> >
> >  	return sdhci_probe(dev);
> >
> 
> Would be good to mentioned that this was tested on Z-turn board.
> 
> Anyway I have tested it on zcu102 and issue is not visible there but it is visible
> on zc706. That's why I expect this is issue only for Zynq device.
> 
> Anyway
> Tested-by: Michal Simek <michal.simek@xilinx.com> (on zcu102/zc706)
> 
> Peng: Feel free to take it via your tree or I will take it.

I'll take it.

Thanks,
Peng.

> 
> Thanks,
> Michal

[U-Boot] [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe

[Michal Simek]

On 02. 08. 19 11:23, Peng Fan wrote:
> 
>> Subject: Re: [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe
>>
>> On 01. 08. 19 17:00, Matwey V. Kornilov wrote:
>>> Since commit 3d296365e4e8 ("mmc: sdhci: Add support for
>>> sdhci-caps-mask") sdhci_setup_cfg() expects a valid sdhci_host mmc
>>> field. Move the mmc field initialization before sdhci_setup_cfg() call
>>> to avoid crash on mmc pointer dereference.
>>>
>>> [this patch is based on commit 41a9fab8dac8 ("mmc: mv_sdhci: fix
>>> uninitialized pointer deref on probe") by Baruch Siach]
>>>
>>> Fixes: 3d296365e4e8 ("mmc: sdhci: Add support for sdhci-caps-mask")
>>> Cc: Faiz Abbas <faiz_abbas@ti.com>
>>> Cc: Baruch Siach <baruch@tkos.co.il>
>>> Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
>>> ---
>>>  drivers/mmc/zynq_sdhci.c | 7 ++++---
>>>  1 file changed, 4 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/mmc/zynq_sdhci.c b/drivers/mmc/zynq_sdhci.c index
>>> c525084250..3225a7ac93 100644
>>> --- a/drivers/mmc/zynq_sdhci.c
>>> +++ b/drivers/mmc/zynq_sdhci.c
>>> @@ -242,13 +242,14 @@ static int arasan_sdhci_probe(struct udevice
>>> *dev)
>>>
>>>  	host->max_clk = clock;
>>>
>>> +	host->mmc = &plat->mmc;
>>> +	host->mmc->dev = dev;
>>> +	host->mmc->priv = host;
>>> +
>>>  	ret = sdhci_setup_cfg(&plat->cfg, host, plat->f_max,
>>>  			      CONFIG_ZYNQ_SDHCI_MIN_FREQ);
>>> -	host->mmc = &plat->mmc;
>>>  	if (ret)
>>>  		return ret;
>>> -	host->mmc->priv = host;
>>> -	host->mmc->dev = dev;
>>>  	upriv->mmc = host->mmc;
>>>
>>>  	return sdhci_probe(dev);
>>>
>>
>> Would be good to mentioned that this was tested on Z-turn board.
>>
>> Anyway I have tested it on zcu102 and issue is not visible there but it is visible
>> on zc706. That's why I expect this is issue only for Zynq device.
>>
>> Anyway
>> Tested-by: Michal Simek <michal.simek@xilinx.com> (on zcu102/zc706)
>>
>> Peng: Feel free to take it via your tree or I will take it.
> 
> I'll take it.

thx.
M

[U-Boot] [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe

[Peng Fan]

> Subject: [PATCH] mmc: zynq_sdhci: fix uninitialized pointer deref on probe
> 
> Since commit 3d296365e4e8 ("mmc: sdhci: Add support for
> sdhci-caps-mask") sdhci_setup_cfg() expects a valid sdhci_host mmc field.
> Move the mmc field initialization before sdhci_setup_cfg() call to avoid crash
> on mmc pointer dereference.
> 
> [this patch is based on commit 41a9fab8dac8 ("mmc: mv_sdhci: fix
> uninitialized pointer deref on probe") by Baruch Siach]
> 
> Fixes: 3d296365e4e8 ("mmc: sdhci: Add support for sdhci-caps-mask")
> Cc: Faiz Abbas <faiz_abbas@ti.com>
> Cc: Baruch Siach <baruch@tkos.co.il>
> Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
> ---
>  drivers/mmc/zynq_sdhci.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/mmc/zynq_sdhci.c b/drivers/mmc/zynq_sdhci.c index
> c525084250..3225a7ac93 100644
> --- a/drivers/mmc/zynq_sdhci.c
> +++ b/drivers/mmc/zynq_sdhci.c
> @@ -242,13 +242,14 @@ static int arasan_sdhci_probe(struct udevice *dev)
> 
>  	host->max_clk = clock;
> 
> +	host->mmc = &plat->mmc;
> +	host->mmc->dev = dev;
> +	host->mmc->priv = host;
> +
>  	ret = sdhci_setup_cfg(&plat->cfg, host, plat->f_max,
>  			      CONFIG_ZYNQ_SDHCI_MIN_FREQ);
> -	host->mmc = &plat->mmc;
>  	if (ret)
>  		return ret;
> -	host->mmc->priv = host;
> -	host->mmc->dev = dev;
>  	upriv->mmc = host->mmc;
> 
>  	return sdhci_probe(dev);

Applied to mmc/master.

Thanks,
Peng.

> --
> 2.16.4