Re: [PATCH] mmc: dw_mmc-exynos: fixed wrong sample-clock selection

[Jaehoon Chung <jh80.chung@samsung.com>]

Hi, Alim.

On 09/17/2014 09:49 PM, Alim Akhtar wrote:
> Hi Jaehoon
> 
> On Wed, Sep 17, 2014 at 8:06 AM, Jaehoon Chung <jh80.chung@samsung.com> wrote:
>> Hi, Alim.
>>
>> On 09/17/2014 07:27 AM, Alim Akhtar wrote:
>>> Hi Jaehoon,
>>>
>>> On Mon, Sep 15, 2014 at 3:56 PM, Jaehoon Chung <jh80.chung@samsung.com> wrote:
>>>> Almost all case is selected to 0.
>>>> (It's not correct sample-clock value.)
>>>> Since it set to wrong value, HS200 mode don't work fine.
>>> Can you please explain what problem you are facing here?
>>
>> When i try to use HS200 with some eMMC card, Some card is working fine and other isn't working fine.
>> (DWMMC is occured the DCRC error.)
>> DCRC error might be occurred when the wrong sample-clock is set.
>> So i tried to see what value is selected to sample-clock.
>> (Selected value - Working fine card : 0, Not working fine card : 0)
>>
>> When i skip to 0, it's selected to other value for best sample-clock.
>> (Selected value - Working fine card : 4, Not working fine card : 6)
>>
>> After working fine card is selected to 6, it's working fine.
>> I didn't know which value is set for other card. But in my case, best sample-clock is selected to 0.
> I am not if this is only reason for the failure on you board.
> When I tested this patch on my exynos5800 based Pi board, HS200 does
> not work at all. Even the eMMC card detection does not happening,
> forget about getting a DCRC error.
> Can you confirm what base kernel you are testing on? FYI, I am testing
> on Ulf's -next branch, which has two recently merged patches specially
> for HS200 on dw_mmc controller.
> In case you are using the same branch as mine, then some more debug
> log will be good for more analysis of the problem.

Thanks for pointing out. Well, i didn't test with other boards.
So your testing is helpful to me. I think it's not my board's problem.
Do you ensure that sample-clock is correct value?

I tested with 3.16-kerenl. Which value is selected at exynos5800 board?
I will test with kernel on Ulf's repository.

Best Regards,
Jaehoon Chung

>>
>> I think it has the some problem for selecting the sample-clock.
>>
>> Best Regards,
>> Jaehoon Chung
>>
>>> It is not clear from your patch description.
>>>> If we want to select the correct value, it has to check from 1 to 7.(skip 0)
>>>>
>>>> Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>
>>>> ---
>>>>  drivers/mmc/host/dw_mmc-exynos.c |    9 ++++++++-
>>>>  1 file changed, 8 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/mmc/host/dw_mmc-exynos.c b/drivers/mmc/host/dw_mmc-exynos.c
>>>> index 0fbc53a..5294035 100644
>>>> --- a/drivers/mmc/host/dw_mmc-exynos.c
>>>> +++ b/drivers/mmc/host/dw_mmc-exynos.c
>>>> @@ -25,6 +25,7 @@
>>>>  #define NUM_PINS(x)                    (x + 2)
>>>>
>>>>  #define SDMMC_CLKSEL                   0x09C
>>>> +#define SDMMC_CLKSEL_CCLK_SAMPLE_MASK  0x7
>>>>  #define SDMMC_CLKSEL_CCLK_SAMPLE(x)    (((x) & 7) << 0)
>>>>  #define SDMMC_CLKSEL_CCLK_DRIVE(x)     (((x) & 7) << 16)
>>>>  #define SDMMC_CLKSEL_CCLK_DIVIDER(x)   (((x) & 7) << 24)
>>>> @@ -330,6 +331,12 @@ static int dw_mci_exynos_execute_tuning(struct dw_mci_slot *slot, u32 opcode,
>>>>         if (!blk_test)
>>>>                 return -ENOMEM;
>>>>
>>>> +       /*
>>>> +        * In order to check all selclk_sample clock,
>>>> +        * it needs to reset to 0.
>>>> +        */
>>>> +       dw_mci_exynos_set_clksmpl(host, 0);
>>>> +
>>>>         start_smpl = dw_mci_exynos_get_clksmpl(host);
>>>>
>>>>         do {
>>>> @@ -372,7 +379,7 @@ static int dw_mci_exynos_execute_tuning(struct dw_mci_slot *slot, u32 opcode,
>>>>                                 "Tuning error: cmd.error:%d, data.error:%d\n",
>>>>                                 cmd.error, data.error);
>>>>                 }
>>>> -       } while (start_smpl != smpl);
>>>> +       } while (start_smpl < SDMMC_CLKSEL_CCLK_SAMPLE_MASK);
>>>>
>>>>         found = dw_mci_exynos_get_best_clksmpl(candiates);
>>>>         if (found >= 0)
>>>> --
>>>> 1.7.9.5
>>>>
>>>> --
>>>> 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
>>>
>>>
>>>
>>
> 
> 
>