[PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Tomasz Figa]

Some boards use fixed voltage regulator for vmmc supply (e.g. for eMMC
memories). MMC_CAP2_BROKEN_VOLTAGE must be enabled for them to operate
correctly.

Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Chris Ball <cjb@laptop.org>
CC: linux-mmc@vger.kernel.org
Signed-off-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
---
 Documentation/devicetree/bindings/mmc/mmc.txt | 1 +
 drivers/mmc/host/sdhci-s3c.c                  | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/mmc/mmc.txt b/Documentation/devicetree/bindings/mmc/mmc.txt
index 8a6811f..ecbde68 100644
--- a/Documentation/devicetree/bindings/mmc/mmc.txt
+++ b/Documentation/devicetree/bindings/mmc/mmc.txt
@@ -16,6 +16,7 @@ Optional properties:
 - wp-inverted: when present, polarity on the wp gpio line is inverted
 - non-removable: non-removable slot (like eMMC)
 - max-frequency: maximum operating clock frequency
+- broken-voltage: vmmc regulator does not allow voltage control
 
 Example:
 
diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index 445910e..39715b8 100644
--- a/drivers/mmc/host/sdhci-s3c.c
+++ b/drivers/mmc/host/sdhci-s3c.c
@@ -443,6 +443,9 @@ static int __devinit sdhci_s3c_parse_dt(struct device *dev,
 	if (!ourhost->gpios)
 		return -ENOMEM;
 
+	if (of_get_property(node, "broken-voltage", 0))
+		pdata->host_caps2 |= MMC_CAP2_BROKEN_VOLTAGE;
+
 	/* get the card detection method */
 	if (of_get_property(node, "broken-cd", 0)) {
 		pdata->cd_type = S3C_SDHCI_CD_NONE;
-- 
1.7.12

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Jaehoon Chung]

Acked-by: Jaehoon Chung <jh80.chung@samsung.com>

On 09/04/2012 05:50 PM, Tomasz Figa wrote:
> Some boards use fixed voltage regulator for vmmc supply (e.g. for eMMC
> memories). MMC_CAP2_BROKEN_VOLTAGE must be enabled for them to operate
> correctly.
> 
> Cc: Ben Dooks <ben-linux@fluff.org>
> Cc: Chris Ball <cjb@laptop.org>
> CC: linux-mmc@vger.kernel.org
> Signed-off-by: Tomasz Figa <t.figa@samsung.com>
> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
> ---
>  Documentation/devicetree/bindings/mmc/mmc.txt | 1 +
>  drivers/mmc/host/sdhci-s3c.c                  | 3 +++
>  2 files changed, 4 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/mmc/mmc.txt b/Documentation/devicetree/bindings/mmc/mmc.txt
> index 8a6811f..ecbde68 100644
> --- a/Documentation/devicetree/bindings/mmc/mmc.txt
> +++ b/Documentation/devicetree/bindings/mmc/mmc.txt
> @@ -16,6 +16,7 @@ Optional properties:
>  - wp-inverted: when present, polarity on the wp gpio line is inverted
>  - non-removable: non-removable slot (like eMMC)
>  - max-frequency: maximum operating clock frequency
> +- broken-voltage: vmmc regulator does not allow voltage control
>  
>  Example:
>  
> diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
> index 445910e..39715b8 100644
> --- a/drivers/mmc/host/sdhci-s3c.c
> +++ b/drivers/mmc/host/sdhci-s3c.c
> @@ -443,6 +443,9 @@ static int __devinit sdhci_s3c_parse_dt(struct device *dev,
>  	if (!ourhost->gpios)
>  		return -ENOMEM;
>  
> +	if (of_get_property(node, "broken-voltage", 0))
> +		pdata->host_caps2 |= MMC_CAP2_BROKEN_VOLTAGE;
> +
>  	/* get the card detection method */
>  	if (of_get_property(node, "broken-cd", 0)) {
>  		pdata->cd_type = S3C_SDHCI_CD_NONE;
> 

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Chris Ball]

Hi,

On Tue, Sep 04 2012, Tomasz Figa wrote:
> Some boards use fixed voltage regulator for vmmc supply (e.g. for eMMC
> memories). MMC_CAP2_BROKEN_VOLTAGE must be enabled for them to operate
> correctly.
>
> Cc: Ben Dooks <ben-linux@fluff.org>
> Cc: Chris Ball <cjb@laptop.org>
> CC: linux-mmc@vger.kernel.org
> Signed-off-by: Tomasz Figa <t.figa@samsung.com>
> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
> ---
>  Documentation/devicetree/bindings/mmc/mmc.txt | 1 +
>  drivers/mmc/host/sdhci-s3c.c                  | 3 +++
>  2 files changed, 4 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/mmc/mmc.txt b/Documentation/devicetree/bindings/mmc/mmc.txt
> index 8a6811f..ecbde68 100644
> --- a/Documentation/devicetree/bindings/mmc/mmc.txt
> +++ b/Documentation/devicetree/bindings/mmc/mmc.txt
> @@ -16,6 +16,7 @@ Optional properties:
>  - wp-inverted: when present, polarity on the wp gpio line is inverted
>  - non-removable: non-removable slot (like eMMC)
>  - max-frequency: maximum operating clock frequency
> +- broken-voltage: vmmc regulator does not allow voltage control
>  
>  Example:
>  
> diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
> index 445910e..39715b8 100644
> --- a/drivers/mmc/host/sdhci-s3c.c
> +++ b/drivers/mmc/host/sdhci-s3c.c
> @@ -443,6 +443,9 @@ static int __devinit sdhci_s3c_parse_dt(struct device *dev,
>  	if (!ourhost->gpios)
>  		return -ENOMEM;
>  
> +	if (of_get_property(node, "broken-voltage", 0))
> +		pdata->host_caps2 |= MMC_CAP2_BROKEN_VOLTAGE;
> +
>  	/* get the card detection method */
>  	if (of_get_property(node, "broken-cd", 0)) {
>  		pdata->cd_type = S3C_SDHCI_CD_NONE;

Is there a reason we can't make this a property on the regulator instead?

Thanks,

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

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Tomasz Figa]

Hi Chris,

On Wednesday 19 of September 2012 01:42:01 Chris Ball wrote:
> On Tue, Sep 04 2012, Tomasz Figa wrote:
> > Some boards use fixed voltage regulator for vmmc supply (e.g. for eMMC
> > memories). MMC_CAP2_BROKEN_VOLTAGE must be enabled for them to operate
> > correctly.
>>
> Is there a reason we can't make this a property on the regulator instead?

Is there a reason we can't make this a property of the mmc subsystem? ;)

Now, seriously, could you elaborate on this a bit more? Do you mean that a 
regulator should provide a dummy set voltage operation that would accept 
any voltage?

Best regards,
-- 
Tomasz Figa
Samsung Poland R&D Center

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Chris Ball]

Hi Tomasz,

On Wed, Sep 19 2012, Tomasz Figa wrote:
> Hi Chris,
>
> On Wednesday 19 of September 2012 01:42:01 Chris Ball wrote:
>> On Tue, Sep 04 2012, Tomasz Figa wrote:
>> > Some boards use fixed voltage regulator for vmmc supply (e.g. for eMMC
>> > memories). MMC_CAP2_BROKEN_VOLTAGE must be enabled for them to operate
>> > correctly.
>>>
>> Is there a reason we can't make this a property on the regulator instead?
>
> Is there a reason we can't make this a property of the mmc subsystem? ;)
>
> Now, seriously, could you elaborate on this a bit more? Do you mean that a 
> regulator should provide a dummy set voltage operation that would accept 
> any voltage?

Sorry for the terseness.

It seems like we're encoding exactly the same information twice in two
different subsystems -- I don't see the point, so I'd like to think
about how we could do better.

For example, if we're only concerned about fixed regulators, could we
just detect a fixed regulator in the driver and avoid the failing call
to regulator_set_voltage() directly, without needing to go via this
capability?  Seems like the capability doesn't tell us anything we
couldn't already have known.

Thanks,

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

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Tomasz Figa]

Hi Chris,

On Wednesday 19 of September 2012 06:24:46 Chris Ball wrote:
> Hi Tomasz,
> 
> On Wed, Sep 19 2012, Tomasz Figa wrote:
> > Hi Chris,
> > 
> > On Wednesday 19 of September 2012 01:42:01 Chris Ball wrote:
> >> On Tue, Sep 04 2012, Tomasz Figa wrote:
> >> > Some boards use fixed voltage regulator for vmmc supply (e.g. for
> >> > eMMC
> >> > memories). MMC_CAP2_BROKEN_VOLTAGE must be enabled for them to
> >> > operate
> >> > correctly.
> >> 
> >> Is there a reason we can't make this a property on the regulator
> >> instead?> 
> > Is there a reason we can't make this a property of the mmc subsystem?
> > ;)
> > 
> > Now, seriously, could you elaborate on this a bit more? Do you mean
> > that a regulator should provide a dummy set voltage operation that
> > would accept any voltage?
> 
> Sorry for the terseness.
> 
> It seems like we're encoding exactly the same information twice in two
> different subsystems -- I don't see the point, so I'd like to think
> about how we could do better.
> 
> For example, if we're only concerned about fixed regulators, could we
> just detect a fixed regulator in the driver and avoid the failing call
> to regulator_set_voltage() directly, without needing to go via this
> capability?  Seems like the capability doesn't tell us anything we
> couldn't already have known.

We could just check if the regulator provides the capability to change the 
voltage.

I don't see any direct way of querying the regulator for provided 
capabilities (correct me if I'm just blind), but calling 
regulator_count_voltages() on the regulator and checking if the returned 
value is 1 should be enough to assume that the regulator is fixed.

What do you think?

Best regards,
-- 
Tomasz Figa
Samsung Poland R&D Center

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Chris Ball]

Hi,

On Wed, Sep 19 2012, Tomasz Figa wrote:
> We could just check if the regulator provides the capability to change the 
> voltage.
>
> I don't see any direct way of querying the regulator for provided 
> capabilities (correct me if I'm just blind), but calling 
> regulator_count_voltages() on the regulator and checking if the returned 
> value is 1 should be enough to assume that the regulator is fixed.

Sounds good, I agree.  Are you able to test that the obvious patch below
works on your fixed-regulator board?

Jaehoon and Adrian, can you think of any reason why we shouldn't replace
MMC_CAP2_BROKEN_VOLTAGE with the regulator_count_voltages() call below?
Thanks.


diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 044cd01..a3cc740 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -1132,7 +1132,7 @@ int mmc_regulator_set_ocr(struct mmc_host *mmc,
 		 */
 		voltage = regulator_get_voltage(supply);
 
-		if (mmc->caps2 & MMC_CAP2_BROKEN_VOLTAGE)
+		if (regulator_count_voltages(supply) == 1)
 			min_uV = max_uV = voltage;
 
 		if (voltage < 0)

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

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Tomasz Figa]

hi,

On Wednesday 19 of September 2012 06:47:02 Chris Ball wrote:
> Hi,
> 
> On Wed, Sep 19 2012, Tomasz Figa wrote:
> > We could just check if the regulator provides the capability to change
> > the voltage.
> > 
> > I don't see any direct way of querying the regulator for provided
> > capabilities (correct me if I'm just blind), but calling
> > regulator_count_voltages() on the regulator and checking if the
> > returned
> > value is 1 should be enough to assume that the regulator is fixed.
> 
> Sounds good, I agree.  Are you able to test that the obvious patch below
> works on your fixed-regulator board?

Yes. I will report the result as soon as I test it.

Best regards,
-- 
Tomasz Figa
Samsung Poland R&D Center

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Jaehoon Chung]

Hi,

On 09/19/2012 07:47 PM, Chris Ball wrote:
> Hi,
> 
> On Wed, Sep 19 2012, Tomasz Figa wrote:
>> We could just check if the regulator provides the capability to change the 
>> voltage.
>>
>> I don't see any direct way of querying the regulator for provided 
>> capabilities (correct me if I'm just blind), but calling 
>> regulator_count_voltages() on the regulator and checking if the returned 
>> value is 1 should be enough to assume that the regulator is fixed.
> 
> Sounds good, I agree.  Are you able to test that the obvious patch below
> works on your fixed-regulator board?
> 
> Jaehoon and Adrian, can you think of any reason why we shouldn't replace
> MMC_CAP2_BROKEN_VOLTAGE with the regulator_count_voltages() call below?
> Thanks.
I think this is better than using MMC_CAP2_BROKEN_VOLTAGE.
I tested with this..and working fine.

Best Regards,
Jaehoon Chung
> 
> 
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
> index 044cd01..a3cc740 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -1132,7 +1132,7 @@ int mmc_regulator_set_ocr(struct mmc_host *mmc,
>  		 */
>  		voltage = regulator_get_voltage(supply);
>  
> -		if (mmc->caps2 & MMC_CAP2_BROKEN_VOLTAGE)
> +		if (regulator_count_voltages(supply) == 1)
>  			min_uV = max_uV = voltage;
>  
>  		if (voltage < 0)
> 

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Chris Ball]

Hi,

On Wed, Sep 19 2012, Jaehoon Chung wrote:
>> On Wed, Sep 19 2012, Tomasz Figa wrote:
>>> We could just check if the regulator provides the capability to change the 
>>> voltage.
>>>
>>> I don't see any direct way of querying the regulator for provided 
>>> capabilities (correct me if I'm just blind), but calling 
>>> regulator_count_voltages() on the regulator and checking if the returned 
>>> value is 1 should be enough to assume that the regulator is fixed.
>> 
>> Sounds good, I agree.  Are you able to test that the obvious patch below
>> works on your fixed-regulator board?
>> 
>> Jaehoon and Adrian, can you think of any reason why we shouldn't replace
>> MMC_CAP2_BROKEN_VOLTAGE with the regulator_count_voltages() call below?
>> Thanks.
>
> I think this is better than using MMC_CAP2_BROKEN_VOLTAGE.
> I tested with this..and working fine.

Great, here's the patch.  Jaehoon, once this is merged, maybe you could
help remove the uses of MMC_CAP2_BROKEN_VOLTAGE from arch/arm/mach-exynos
and arch/arm/mach-s5pv210 now that they're no longer needed?

Thanks,

- Chris.


Subject: [PATCH] mmc: core: Replace MMC_CAP2_BROKEN_VOLTAGE with test for fixed regulator

Before this patch, we were using MMC_CAP2_BROKEN_VOLTAGE as a way to
avoid calling regulator_set_voltage() on a fixed regulator, but that's
just duplicating information that already exists -- we should test
whether the regulator is fixed directly, instead of via a capability.

This patch implements that test.  We can't reclaim the capability bit
just yet, since there are still boards in arch/arm/ that reference it;
those references can be removed now.

Reported-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
---
 drivers/mmc/core/core.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 044cd01..6612163 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -1113,7 +1113,8 @@ int mmc_regulator_set_ocr(struct mmc_host *mmc,
 		int		tmp;
 		int		voltage;
 
-		/* REVISIT mmc_vddrange_to_ocrmask() may have set some
+		/*
+		 * REVISIT mmc_vddrange_to_ocrmask() may have set some
 		 * bits this regulator doesn't quite support ... don't
 		 * be too picky, most cards and regulators are OK with
 		 * a 0.1V range goof (it's a small error percentage).
@@ -1127,12 +1128,13 @@ int mmc_regulator_set_ocr(struct mmc_host *mmc,
 			max_uV = min_uV + 100 * 1000;
 		}
 
-		/* avoid needless changes to this voltage; the regulator
-		 * might not allow this operation
+		/*
+		 * If we're using a fixed/static regulator, don't call
+		 * regulator_set_voltage; it would fail.
 		 */
 		voltage = regulator_get_voltage(supply);
 
-		if (mmc->caps2 & MMC_CAP2_BROKEN_VOLTAGE)
+		if (regulator_count_voltages(supply) == 1)
 			min_uV = max_uV = voltage;
 
 		if (voltage < 0)
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child

Re: [PATCH v2 1/5] mmc: host: sdhci-s3c: Add broken-voltage DT property for broken voltage quirk

[Jaehoon Chung]

On 09/19/2012 11:39 PM, Chris Ball wrote:
> Hi,
> 
> On Wed, Sep 19 2012, Jaehoon Chung wrote:
>>> On Wed, Sep 19 2012, Tomasz Figa wrote:
>>>> We could just check if the regulator provides the capability to change the 
>>>> voltage.
>>>>
>>>> I don't see any direct way of querying the regulator for provided 
>>>> capabilities (correct me if I'm just blind), but calling 
>>>> regulator_count_voltages() on the regulator and checking if the returned 
>>>> value is 1 should be enough to assume that the regulator is fixed.
>>>
>>> Sounds good, I agree.  Are you able to test that the obvious patch below
>>> works on your fixed-regulator board?
>>>
>>> Jaehoon and Adrian, can you think of any reason why we shouldn't replace
>>> MMC_CAP2_BROKEN_VOLTAGE with the regulator_count_voltages() call below?
>>> Thanks.
>>
>> I think this is better than using MMC_CAP2_BROKEN_VOLTAGE.
>> I tested with this..and working fine.
> 
> Great, here's the patch.  Jaehoon, once this is merged, maybe you could
> help remove the uses of MMC_CAP2_BROKEN_VOLTAGE from arch/arm/mach-exynos
> and arch/arm/mach-s5pv210 now that they're no longer needed?
Right, I will remove them and send the patch.

Best Regards,
Jaehoon Chung
> 
> Thanks,
> 
> - Chris.
> 
> 
> Subject: [PATCH] mmc: core: Replace MMC_CAP2_BROKEN_VOLTAGE with test for fixed regulator
> 
> Before this patch, we were using MMC_CAP2_BROKEN_VOLTAGE as a way to
> avoid calling regulator_set_voltage() on a fixed regulator, but that's
> just duplicating information that already exists -- we should test
> whether the regulator is fixed directly, instead of via a capability.
> 
> This patch implements that test.  We can't reclaim the capability bit
> just yet, since there are still boards in arch/arm/ that reference it;
> those references can be removed now.
> 
> Reported-by: Tomasz Figa <t.figa@samsung.com>
> Signed-off-by: Chris Ball <cjb@laptop.org>
> ---
>  drivers/mmc/core/core.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
> index 044cd01..6612163 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -1113,7 +1113,8 @@ int mmc_regulator_set_ocr(struct mmc_host *mmc,
>  		int		tmp;
>  		int		voltage;
>  
> -		/* REVISIT mmc_vddrange_to_ocrmask() may have set some
> +		/*
> +		 * REVISIT mmc_vddrange_to_ocrmask() may have set some
>  		 * bits this regulator doesn't quite support ... don't
>  		 * be too picky, most cards and regulators are OK with
>  		 * a 0.1V range goof (it's a small error percentage).
> @@ -1127,12 +1128,13 @@ int mmc_regulator_set_ocr(struct mmc_host *mmc,
>  			max_uV = min_uV + 100 * 1000;
>  		}
>  
> -		/* avoid needless changes to this voltage; the regulator
> -		 * might not allow this operation
> +		/*
> +		 * If we're using a fixed/static regulator, don't call
> +		 * regulator_set_voltage; it would fail.
>  		 */
>  		voltage = regulator_get_voltage(supply);
>  
> -		if (mmc->caps2 & MMC_CAP2_BROKEN_VOLTAGE)
> +		if (regulator_count_voltages(supply) == 1)
>  			min_uV = max_uV = voltage;
>  
>  		if (voltage < 0)
>