[U-Boot] [PATCH] MMC: DWMMC: Fix FIFO_DEPTH calculation

[Jagan Teki]

On Fri, May 24, 2013 at 12:16 PM, Jaehoon Chung <jh80.chung at samsung.com> wrote:
> Hi Jagan,
>
> On 05/24/2013 03:15 PM, Rajeshwari Birje wrote:
>> Hi Jagan,
>>
>> Please find my comments below.
>> On Fri, May 24, 2013 at 10:31 AM, Jagan Teki <jagannadh.teki at gmail.com> wrote:
>>> On Fri, May 24, 2013 at 7:12 AM, Jaehoon Chung <jh80.chung at samsung.com> wrote:
>>>> On 05/24/2013 03:27 AM, Jagan Teki wrote:
>>>>> On Thu, May 23, 2013 at 6:45 PM, Rajeshwari Shinde
>>>>> <rajeshwari.s at samsung.com> wrote:
>>>>>> Current DWMMC driver used to give FIFO underrun/overrun error every 3rd time
>>>>>> for mmc rescan command.
>>>>>> In current code FIFO_DEPTH is getting calculated after reading the FIFOTH
>>>>>> register and extracting the RX_WMARK bits from it i.e (RX_WMARK = FIFO_DEPTH/2 -1).
>>>>>> Instead of storing the correct value, we were recalculating the FIFO_DEPT each
>>>>>> time which is not correct.
>>>>>>
>>>>>> Signed-off-by: Hatim Ali <hatim.rv at samsung.com>
>>>>>> Signed-off-by: Rajeshwari Shinde <rajeshwari.s at samsung.com>
>>>>>> ---
>>>>>>  drivers/mmc/dw_mmc.c |    6 ++++--
>>>>>>  1 files changed, 4 insertions(+), 2 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/mmc/dw_mmc.c b/drivers/mmc/dw_mmc.c
>>>>>> index 4070d4e..be590a4 100644
>>>>>> --- a/drivers/mmc/dw_mmc.c
>>>>>> +++ b/drivers/mmc/dw_mmc.c
>>>>>> @@ -332,11 +332,13 @@ static int dwmci_init(struct mmc *mmc)
>>>>>>         dwmci_writel(host, DWMCI_BMOD, 1);
>>>>>>
>>>>>>         fifo_size = dwmci_readl(host, DWMCI_FIFOTH);
>>>>>> -       if (host->fifoth_val)
>>>>>> +       if (host->fifoth_val) {
>>>>>
>>>>> What is the inital value for host->fifoth_val, for the first time call.?
>>>> It should be set into board-specific file(dw-mmc_exynos.c) or others.
>>>
>>> I am unable to find dw-mmc_exynos.c on the master, and also I haven't
>>> see fifoth_val used other than dw_mmc.c
>>> Could you please help me out.
>> dw-mmc_exynos.c is merged in mmc tree. Yes  FIFO_DEPTH is set only in dw_mmc.c
> host->fifoth_val should be set to hard-coding by developer.
> In general case, it's right that fifoth_Val is calculated with register value.
> But sometime, it needs to set to other value, not register value.
> Then we can use the host->fifoth_val in board-specific file.
> Maybe it didn't use the host->fifoth_val anywhere. In future, we can use this.

Ok, thanks for your response.
If that is the case may be we can initialize host->fifoth_val on
add_dwmci() for as of now.
Once the proper init of this is done in board specific file in future
may be we can remove this.

Thanks,
Jagan.

>
> Best Regards,
> Jaehoon Chung
>>>
>>> Thanks,
>>> Jagan.
>>>
>>>>
>>>> Best Regards,
>>>> Jaehoon Chung
>>>>>
>>>>>>                 fifoth_val = host->fifoth_val;
>>>>>> -       else
>>>>>> +       } else {
>>>>>>                 fifoth_val = MSIZE(0x2) | RX_WMARK(fifo_size/2 -1) |
>>>>>>                         TX_WMARK(fifo_size/2);
>>>>>> +               host->fifoth_val = fifoth_val;
>>>>>> +       }
>>>>>>         dwmci_writel(host, DWMCI_FIFOTH, fifoth_val);
>>>>>>
>>>>>>         dwmci_writel(host, DWMCI_CLKENA, 0);
>>>>>> --
>>>>>> 1.7.4.4
>>>>>>
>>>>>
>>>>> Thanks,
>>>>> Jagan.
>>>>>
>>>>
>>> _______________________________________________
>>> U-Boot mailing list
>>> U-Boot at lists.denx.de
>>> http://lists.denx.de/mailman/listinfo/u-boot
>>
>