[PATCH v2] mmc: dw_mmc: add quirk for broken data transfer over scheme

[addy ke]

Hi,
On 2014/11/27 06:46, Doug Anderson wrote:
> Hi,
> 
> On Tue, Nov 25, 2014 at 12:10 AM, Addy Ke <addy.ke at rock-chips.com> wrote:
>> This patch add a new quirk to add a s/w timer to notify the driver
>> to terminate current transfer and report a data timeout to the core,
>> if DTO interrupt does NOT come within the given time.
>>
>> dw_mmc call mmc_request_done func to finish transfer depends on
>> DTO interrupt. If DTO interrupt does not come in sending data state,
>> the current transfer will be blocked.
>>
>> But this case really exists, when driver reads tuning data from
>> card on RK3288-pink2 board. I measured waveforms by oscilloscope
>> and found that card clock was always on and data lines were always
>> holded high level in sending data state.
>>
>> We got the reply from synopsys:
>> There are two counters but both use the same value of [31:8] bits.
>> Data timeout counter doesn't wait for stop clock and you should get
>> DRTO even when the clock is not stopped.
>> Host Starvation timeout counter is triggered with stop clock condition.
>>
>> This means that host should get DRTO and DTO interrupt.
>>
>> But we really don't get any data-related interrupt in RK3X SoCs.
>> And driver can't get data transfer state, it can do nothing but wait for.
> 
> Have you asked someone on your IC team to confirm this is an SoC
> errata on your SoC?  ...or is there something else we could be doing
> wrong (overclocking?  jitter in the clock?  bad dividers?) that could
> be causing this problem?
> 
> 
>>  #ifdef CONFIG_OF
>>  static struct dw_mci_of_quirks {
>>         char *quirk;
>> @@ -2513,6 +2549,9 @@ static struct dw_mci_of_quirks {
>>         }, {
>>                 .quirk  = "disable-wp",
>>                 .id     = DW_MCI_QUIRK_NO_WRITE_PROTECT,
>> +       }, {
>> +               .quirk  = "broken-dto",
>> +               .id     = DW_MCI_QUIRK_BROKEN_DTO,
> 
> You're adding a device tree property without any binding.  If you need
> to add this please send a patch before this one modifying the device
> tree bindings.
> 
> ...but that brings up the question: do you _really_ need to add a
> property?  You already know that all rk3288 SoCs need this and you
> already know that you're an rk3288 SoC.  Just add this quirk in the
> rk3288 code always and be done with it.  ...and if this is also needed
> on other Rockchip parts, add it there too.
> 
> -Doug

We don't know why we have this problem,
but this problem is really exist, and we need patch to fix this problem now.
I will post a follow up change when we find the root cause.

And there is a little probability of this problem on RK SoC, such as RK3188, RK3066,
when worse card inserted in.
Maybe the other SoCs have the similar problem.

So I will add this quirk in rockchip code(dw_mmc-rockchip.c) as follows:
static int dw_mci_rockchip_parse_dt(struct dw_mci *host)
{
	host->quirk |= DW_MCI_QUIRK_BROKEN_DTO;

	return 0;
}

......
	.parse_dt = dw_mci_rockchip_parse_dt,
......

is right?

> 
> 
>