diff mbox

[v2,1/7] mmc: sdhci-pxav3: Fix SDR50 and DDR50 capabilities for the Armada 38x flavor

Message ID 1422010594-1735-2-git-send-email-gregory.clement@free-electrons.com (mailing list archive)
State New, archived
Headers show

Commit Message

Gregory CLEMENT Jan. 23, 2015, 10:56 a.m. UTC
According to erratum 'FE-2946959' both SDR50 and DDR50 modes require
specific clock adjustments in SDIO3 Configuration register. However,
this register was not part of the device tree binding. Even if the
binding can (and will) be extended we still need handling the case
where this register was not available. In this case we use the
SDHCI_QUIRK_MISSING_CAPS quirk remove them from the capabilities.

This commit is based on the work done by Marcin Wojtas<mw@semihalf.com>

Fixes: 5491ce3f79ee ("mmc: sdhci-pxav3: add support for the Armada 38x SDHCI controller")
Cc: <stable@vger.kernel.org> # v3.15+
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: Marcin Wojtas <mw@semihalf.com>
---
 drivers/mmc/host/sdhci-pxav3.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

Comments

Ulf Hansson Jan. 28, 2015, 8:36 p.m. UTC | #1
On 23 January 2015 at 11:56, Gregory CLEMENT
<gregory.clement@free-electrons.com> wrote:
> According to erratum 'FE-2946959' both SDR50 and DDR50 modes require
> specific clock adjustments in SDIO3 Configuration register. However,
> this register was not part of the device tree binding. Even if the
> binding can (and will) be extended we still need handling the case
> where this register was not available. In this case we use the
> SDHCI_QUIRK_MISSING_CAPS quirk remove them from the capabilities.
>
> This commit is based on the work done by Marcin Wojtas<mw@semihalf.com>
>
> Fixes: 5491ce3f79ee ("mmc: sdhci-pxav3: add support for the Armada 38x SDHCI controller")
> Cc: <stable@vger.kernel.org> # v3.15+
> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
> Signed-off-by: Marcin Wojtas <mw@semihalf.com>
> ---
>  drivers/mmc/host/sdhci-pxav3.c | 18 ++++++++++++++++++
>  1 file changed, 18 insertions(+)
>
> diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
> index ca3424e7ef71..7b07325b4fba 100644
> --- a/drivers/mmc/host/sdhci-pxav3.c
> +++ b/drivers/mmc/host/sdhci-pxav3.c
> @@ -118,6 +118,20 @@ static int mv_conf_mbus_windows(struct platform_device *pdev,
>         return 0;
>  }
>
> +static int armada_38x_quirks(struct sdhci_host *host)

Seems like this function can be void instead of always returning 0.

> +{
> +       host->quirks |= SDHCI_QUIRK_MISSING_CAPS;
> +       /*
> +        * According to erratum 'FE-2946959' both SDR50 and DDR50
> +        * modes require specific clock adjustments in SDIO3
> +        * Configuration register, if the adjustment is not done,
> +        * remove them from the capabilities.
> +        */
> +       host->caps1 = sdhci_readl(host, SDHCI_CAPABILITIES_1);
> +       host->caps1 &= ~(SDHCI_SUPPORT_SDR50 | SDHCI_SUPPORT_DDR50);
> +       return 0;
> +}
> +
>  static void pxav3_reset(struct sdhci_host *host, u8 mask)
>  {
>         struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
> @@ -319,6 +333,9 @@ static int sdhci_pxav3_probe(struct platform_device *pdev)
>                 clk_prepare_enable(pxa->clk_core);
>
>         if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) {
> +               ret = armada_38x_quirks(host);
> +               if (ret < 0)
> +                       goto err_quirks;
>                 ret = mv_conf_mbus_windows(pdev, mv_mbus_dram_info());
>                 if (ret < 0)
>                         goto err_mbus_win;
> @@ -400,6 +417,7 @@ err_mbus_win:
>         if (!IS_ERR(pxa->clk_core))
>                 clk_disable_unprepare(pxa->clk_core);
>  err_clk_get:
> +err_quirks:
>         sdhci_pltfm_free(pdev);
>         return ret;
>  }
> --
> 2.1.0
>

Kind regards
Uffe
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Gregory CLEMENT Jan. 29, 2015, 8:31 a.m. UTC | #2
Hi Ulf,

On 28/01/2015 21:36, Ulf Hansson wrote:
> On 23 January 2015 at 11:56, Gregory CLEMENT
> <gregory.clement@free-electrons.com> wrote:
>> According to erratum 'FE-2946959' both SDR50 and DDR50 modes require
>> specific clock adjustments in SDIO3 Configuration register. However,
>> this register was not part of the device tree binding. Even if the
>> binding can (and will) be extended we still need handling the case
>> where this register was not available. In this case we use the
>> SDHCI_QUIRK_MISSING_CAPS quirk remove them from the capabilities.
>>
>> This commit is based on the work done by Marcin Wojtas<mw@semihalf.com>
>>
>> Fixes: 5491ce3f79ee ("mmc: sdhci-pxav3: add support for the Armada 38x SDHCI controller")
>> Cc: <stable@vger.kernel.org> # v3.15+
>> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
>> Signed-off-by: Marcin Wojtas <mw@semihalf.com>
>> ---
>>  drivers/mmc/host/sdhci-pxav3.c | 18 ++++++++++++++++++
>>  1 file changed, 18 insertions(+)
>>
>> diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
>> index ca3424e7ef71..7b07325b4fba 100644
>> --- a/drivers/mmc/host/sdhci-pxav3.c
>> +++ b/drivers/mmc/host/sdhci-pxav3.c
>> @@ -118,6 +118,20 @@ static int mv_conf_mbus_windows(struct platform_device *pdev,
>>         return 0;
>>  }
>>
>> +static int armada_38x_quirks(struct sdhci_host *host)
> 
> Seems like this function can be void instead of always returning 0.

In patch 4 "mmc: sdhci-pxav3: Modify clock settings for the SDR50 and
DDR50 modes", this function can return other values than 0.

I could change the prototype in patch 4, but it would also imply
removing the test of the return value in this patch and adding in back
patch 4. By returning a value in this patch, it reduced the amount of
change over the patches.

But if you still prefer than I this function return void in this
patch, I can do it.


Thanks,

Gregory


> 
>> +{
>> +       host->quirks |= SDHCI_QUIRK_MISSING_CAPS;
>> +       /*
>> +        * According to erratum 'FE-2946959' both SDR50 and DDR50
>> +        * modes require specific clock adjustments in SDIO3
>> +        * Configuration register, if the adjustment is not done,
>> +        * remove them from the capabilities.
>> +        */
>> +       host->caps1 = sdhci_readl(host, SDHCI_CAPABILITIES_1);
>> +       host->caps1 &= ~(SDHCI_SUPPORT_SDR50 | SDHCI_SUPPORT_DDR50);
>> +       return 0;
>> +}
>> +
>>  static void pxav3_reset(struct sdhci_host *host, u8 mask)
>>  {
>>         struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
>> @@ -319,6 +333,9 @@ static int sdhci_pxav3_probe(struct platform_device *pdev)
>>                 clk_prepare_enable(pxa->clk_core);
>>
>>         if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) {
>> +               ret = armada_38x_quirks(host);
>> +               if (ret < 0)
>> +                       goto err_quirks;
>>                 ret = mv_conf_mbus_windows(pdev, mv_mbus_dram_info());
>>                 if (ret < 0)
>>                         goto err_mbus_win;
>> @@ -400,6 +417,7 @@ err_mbus_win:
>>         if (!IS_ERR(pxa->clk_core))
>>                 clk_disable_unprepare(pxa->clk_core);
>>  err_clk_get:
>> +err_quirks:
>>         sdhci_pltfm_free(pdev);
>>         return ret;
>>  }
>> --
>> 2.1.0
>>
> 
> Kind regards
> Uffe
>
Ulf Hansson Jan. 29, 2015, 9:31 a.m. UTC | #3
[...]

>> Seems like this function can be void instead of always returning 0.
>
> In patch 4 "mmc: sdhci-pxav3: Modify clock settings for the SDR50 and
> DDR50 modes", this function can return other values than 0.
>
> I could change the prototype in patch 4, but it would also imply
> removing the test of the return value in this patch and adding in back
> patch 4. By returning a value in this patch, it reduced the amount of
> change over the patches.
>
> But if you still prefer than I this function return void in this
> patch, I can do it.

No worries, let's keep it as an int. But then I have a few other
comments, see below.

>
>
> Thanks,
>
> Gregory
>
>
>>
>>> +{
>>> +       host->quirks |= SDHCI_QUIRK_MISSING_CAPS;
>>> +       /*
>>> +        * According to erratum 'FE-2946959' both SDR50 and DDR50
>>> +        * modes require specific clock adjustments in SDIO3
>>> +        * Configuration register, if the adjustment is not done,
>>> +        * remove them from the capabilities.
>>> +        */
>>> +       host->caps1 = sdhci_readl(host, SDHCI_CAPABILITIES_1);
>>> +       host->caps1 &= ~(SDHCI_SUPPORT_SDR50 | SDHCI_SUPPORT_DDR50);
>>> +       return 0;
>>> +}
>>> +
>>>  static void pxav3_reset(struct sdhci_host *host, u8 mask)
>>>  {
>>>         struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
>>> @@ -319,6 +333,9 @@ static int sdhci_pxav3_probe(struct platform_device *pdev)
>>>                 clk_prepare_enable(pxa->clk_core);
>>>
>>>         if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) {
>>> +               ret = armada_38x_quirks(host);
>>> +               if (ret < 0)

Since in patch 4 you return a proper error code, let's also adopt to
that here by changing to:

"if (IS_ERR(ret))

>>> +                       goto err_quirks;
>>>                 ret = mv_conf_mbus_windows(pdev, mv_mbus_dram_info());
>>>                 if (ret < 0)
>>>                         goto err_mbus_win;
>>> @@ -400,6 +417,7 @@ err_mbus_win:
>>>         if (!IS_ERR(pxa->clk_core))
>>>                 clk_disable_unprepare(pxa->clk_core);
>>>  err_clk_get:
>>> +err_quirks:

You don't need the new label, you could use "err_clk_get".

>>>         sdhci_pltfm_free(pdev);
>>>         return ret;
>>>  }
>>> --
>>> 2.1.0
>>>

Kind regards
Uffe
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Gregory CLEMENT Jan. 29, 2015, 9:42 a.m. UTC | #4
Hi Ulf,

On 29/01/2015 10:31, Ulf Hansson wrote:
> [...]
> 
>>> Seems like this function can be void instead of always returning 0.
>>
>> In patch 4 "mmc: sdhci-pxav3: Modify clock settings for the SDR50 and
>> DDR50 modes", this function can return other values than 0.
>>
>> I could change the prototype in patch 4, but it would also imply
>> removing the test of the return value in this patch and adding in back
>> patch 4. By returning a value in this patch, it reduced the amount of
>> change over the patches.
>>
>> But if you still prefer than I this function return void in this
>> patch, I can do it.
> 
> No worries, let's keep it as an int. But then I have a few other
> comments, see below.

OK

> 
>>
>>
>> Thanks,
>>
>> Gregory
>>
>>
>>>
>>>> +{
>>>> +       host->quirks |= SDHCI_QUIRK_MISSING_CAPS;
>>>> +       /*
>>>> +        * According to erratum 'FE-2946959' both SDR50 and DDR50
>>>> +        * modes require specific clock adjustments in SDIO3
>>>> +        * Configuration register, if the adjustment is not done,
>>>> +        * remove them from the capabilities.
>>>> +        */
>>>> +       host->caps1 = sdhci_readl(host, SDHCI_CAPABILITIES_1);
>>>> +       host->caps1 &= ~(SDHCI_SUPPORT_SDR50 | SDHCI_SUPPORT_DDR50);
>>>> +       return 0;
>>>> +}
>>>> +
>>>>  static void pxav3_reset(struct sdhci_host *host, u8 mask)
>>>>  {
>>>>         struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
>>>> @@ -319,6 +333,9 @@ static int sdhci_pxav3_probe(struct platform_device *pdev)
>>>>                 clk_prepare_enable(pxa->clk_core);
>>>>
>>>>         if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) {
>>>> +               ret = armada_38x_quirks(host);
>>>> +               if (ret < 0)
> 
> Since in patch 4 you return a proper error code, let's also adopt to
> that here by changing to:
> 
> "if (IS_ERR(ret))

The function returns an int and IS_ERR expects a pointer. I am not sure
this macro would be appropriate here.


> 
>>>> +                       goto err_quirks;
>>>>                 ret = mv_conf_mbus_windows(pdev, mv_mbus_dram_info());
>>>>                 if (ret < 0)
>>>>                         goto err_mbus_win;
>>>> @@ -400,6 +417,7 @@ err_mbus_win:
>>>>         if (!IS_ERR(pxa->clk_core))
>>>>                 clk_disable_unprepare(pxa->clk_core);
>>>>  err_clk_get:
>>>> +err_quirks:
> 
> You don't need the new label, you could use "err_clk_get".

Right.

Thanks,

Gregory
Ulf Hansson Jan. 29, 2015, 10:01 a.m. UTC | #5
On 29 January 2015 at 10:42, Gregory CLEMENT
<gregory.clement@free-electrons.com> wrote:
> Hi Ulf,
>
> On 29/01/2015 10:31, Ulf Hansson wrote:
>> [...]
>>
>>>> Seems like this function can be void instead of always returning 0.
>>>
>>> In patch 4 "mmc: sdhci-pxav3: Modify clock settings for the SDR50 and
>>> DDR50 modes", this function can return other values than 0.
>>>
>>> I could change the prototype in patch 4, but it would also imply
>>> removing the test of the return value in this patch and adding in back
>>> patch 4. By returning a value in this patch, it reduced the amount of
>>> change over the patches.
>>>
>>> But if you still prefer than I this function return void in this
>>> patch, I can do it.
>>
>> No worries, let's keep it as an int. But then I have a few other
>> comments, see below.
>
> OK
>
>>
>>>
>>>
>>> Thanks,
>>>
>>> Gregory
>>>
>>>
>>>>
>>>>> +{
>>>>> +       host->quirks |= SDHCI_QUIRK_MISSING_CAPS;
>>>>> +       /*
>>>>> +        * According to erratum 'FE-2946959' both SDR50 and DDR50
>>>>> +        * modes require specific clock adjustments in SDIO3
>>>>> +        * Configuration register, if the adjustment is not done,
>>>>> +        * remove them from the capabilities.
>>>>> +        */
>>>>> +       host->caps1 = sdhci_readl(host, SDHCI_CAPABILITIES_1);
>>>>> +       host->caps1 &= ~(SDHCI_SUPPORT_SDR50 | SDHCI_SUPPORT_DDR50);
>>>>> +       return 0;
>>>>> +}
>>>>> +
>>>>>  static void pxav3_reset(struct sdhci_host *host, u8 mask)
>>>>>  {
>>>>>         struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
>>>>> @@ -319,6 +333,9 @@ static int sdhci_pxav3_probe(struct platform_device *pdev)
>>>>>                 clk_prepare_enable(pxa->clk_core);
>>>>>
>>>>>         if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) {
>>>>> +               ret = armada_38x_quirks(host);
>>>>> +               if (ret < 0)
>>
>> Since in patch 4 you return a proper error code, let's also adopt to
>> that here by changing to:
>>
>> "if (IS_ERR(ret))
>
> The function returns an int and IS_ERR expects a pointer. I am not sure
> this macro would be appropriate here.

You are right. Don't know what I was thinking. :-)

Kind regards
Uffe
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
index ca3424e7ef71..7b07325b4fba 100644
--- a/drivers/mmc/host/sdhci-pxav3.c
+++ b/drivers/mmc/host/sdhci-pxav3.c
@@ -118,6 +118,20 @@  static int mv_conf_mbus_windows(struct platform_device *pdev,
 	return 0;
 }
 
+static int armada_38x_quirks(struct sdhci_host *host)
+{
+	host->quirks |= SDHCI_QUIRK_MISSING_CAPS;
+	/*
+	 * According to erratum 'FE-2946959' both SDR50 and DDR50
+	 * modes require specific clock adjustments in SDIO3
+	 * Configuration register, if the adjustment is not done,
+	 * remove them from the capabilities.
+	 */
+	host->caps1 = sdhci_readl(host, SDHCI_CAPABILITIES_1);
+	host->caps1 &= ~(SDHCI_SUPPORT_SDR50 | SDHCI_SUPPORT_DDR50);
+	return 0;
+}
+
 static void pxav3_reset(struct sdhci_host *host, u8 mask)
 {
 	struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
@@ -319,6 +333,9 @@  static int sdhci_pxav3_probe(struct platform_device *pdev)
 		clk_prepare_enable(pxa->clk_core);
 
 	if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) {
+		ret = armada_38x_quirks(host);
+		if (ret < 0)
+			goto err_quirks;
 		ret = mv_conf_mbus_windows(pdev, mv_mbus_dram_info());
 		if (ret < 0)
 			goto err_mbus_win;
@@ -400,6 +417,7 @@  err_mbus_win:
 	if (!IS_ERR(pxa->clk_core))
 		clk_disable_unprepare(pxa->clk_core);
 err_clk_get:
+err_quirks:
 	sdhci_pltfm_free(pdev);
 	return ret;
 }