[PATCH 1/2] mmc: dw_mmc: poll if CD is broken

[PATCH 1/2] mmc: dw_mmc: poll if CD is broken

[Sebastian Andrzej Siewior]

My CD pin is not available at all (DW_MCI_QUIRK_BROKEN_CARD_DETECTION is
set). The result was in v3.12 that if MMC card was inserted at load
time, then the MMC card was detected. If it wasn't, it wasn't detected
since nobody polled for it and I no irq was wired up.
I haven't noticed any change in that area, therefore I'm adding the
MMC_CAP_NEEDS_POLL flag in case the detection is broken so the core will
check for the MMC card.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 drivers/mmc/host/dw_mmc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index cced599..82f5ea2 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -2227,6 +2227,9 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	}
 
 	slot->wp_gpio = dw_mci_of_get_wp_gpio(host->dev, slot->id);
+
+	if (host->quirks & DW_MCI_QUIRK_BROKEN_CARD_DETECTION)
+		mmc->caps |= MMC_CAP_NEEDS_POLL;
 	dw_mci_of_get_cd_gpio(host->dev, slot->id, mmc);
 
 	ret = mmc_add_host(mmc);
-- 
1.9.2

[PATCH 2/2] mmc: dw_mmc: print card frequency in debug level

[Sebastian Andrzej Siewior]

Printing this at info level can be very noisy the CD is not working and
we poll for the card.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 drivers/mmc/host/dw_mmc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index 82f5ea2..b8afc70 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -800,7 +800,7 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, bool force_clkinit)
 		div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
 
 		if ((clock << div) != slot->__clk_old || force_clkinit)
-			dev_info(&slot->mmc->class_dev,
+			dev_dbg(&slot->mmc->class_dev,
 				 "Bus speed (slot %d) = %dHz (slot req %dHz, actual %dHZ div = %d)\n",
 				 slot->id, host->bus_hz, clock,
 				 div ? ((host->bus_hz / div) >> 1) :
-- 
1.9.2

Re: [PATCH 1/2] mmc: dw_mmc: poll if CD is broken

[Sebastian Andrzej Siewior]

On 04/30/2014 04:57 PM, Jae hoon Chung wrote:
> Hi, Sebastian.

Hi Jaehoon,

> I didn't know which Soc you used.

Altera socfpa.

> Do you use the Device-tree? if you use the device-tree, just add the
> "broken-cd" property.

and with this property the code will rescan the card every few secs? I
didn't see this in v3.12. I do have "broken-cd" set. And the card will
not be detected if it is inserted after the driver probed.

But now that you pointed me at it. You say that "broken-cd" will be
scanned by the core, right? Shouldn't the driver then call
mmc_of_parse() or am I missing something?

> Best Regards,
> Jaehoon Chung

Sebastian

Re: [PATCH 1/2] mmc: dw_mmc: poll if CD is broken

[Jaehoon Chung]

Hi, Sebastian.

Your patch didn't base on latest mmc-next repository.
You can refer to below git.

http://git.kernel.org/cgit/linux/kernel/git/cjb/mmc.git/

I think you made the patch on v3.12.

Best Regards,
Jaehoon Chung

On 05/01/2014 12:17 AM, Sebastian Andrzej Siewior wrote:
> On 04/30/2014 04:57 PM, Jae hoon Chung wrote:
>> Hi, Sebastian.
> 
> Hi Jaehoon,
> 
>> I didn't know which Soc you used.
> 
> Altera socfpa.
> 
>> Do you use the Device-tree? if you use the device-tree, just add the
>> "broken-cd" property.
> 
> and with this property the code will rescan the card every few secs? I
> didn't see this in v3.12. I do have "broken-cd" set. And the card will
> not be detected if it is inserted after the driver probed.
> 
> But now that you pointed me at it. You say that "broken-cd" will be
> scanned by the core, right? Shouldn't the driver then call
> mmc_of_parse() or am I missing something?
> 
>> Best Regards,
>> Jaehoon Chung
> 
> Sebastian
> 

Re: [PATCH 2/2] mmc: dw_mmc: print card frequency in debug level

[Jaehoon Chung]

Hi, Sebastian.

You have to work on latest mmc-next repository.
And that message is useful information for debugging.
I didn't want to change message level.

Best Regards,
Jaehoon Chung

On 04/30/2014 10:44 PM, Sebastian Andrzej Siewior wrote:
> Printing this at info level can be very noisy the CD is not working and
> we poll for the card.
> 
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> ---
>  drivers/mmc/host/dw_mmc.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
> index 82f5ea2..b8afc70 100644
> --- a/drivers/mmc/host/dw_mmc.c
> +++ b/drivers/mmc/host/dw_mmc.c
> @@ -800,7 +800,7 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, bool force_clkinit)
>  		div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
>  
>  		if ((clock << div) != slot->__clk_old || force_clkinit)
> -			dev_info(&slot->mmc->class_dev,
> +			dev_dbg(&slot->mmc->class_dev,
>  				 "Bus speed (slot %d) = %dHz (slot req %dHz, actual %dHZ div = %d)\n",
>  				 slot->id, host->bus_hz, clock,
>  				 div ? ((host->bus_hz / div) >> 1) :
>