public inbox for linux-mmc@vger.kernel.org
 help / color / mirror / Atom feed
From: Zhoujie Wu <zjwu@marvell.com>
To: Shawn Lin <shawn.lin@rock-chips.com>
Cc: linux-mmc@vger.kernel.org, ulf.hansson@linaro.org,
	adrian.hunter@intel.com, zmxu@marvell.com, jszhang@marvell.com,
	nadavh@marvell.com, xigu@marvell.com, dingwei@marvell.com,
	kostap@marvell.com, hannah@marvell.com, hongd@marvell.com,
	dougj@marvell.com, ygao@marvell.com, liuw@marvell.com,
	gregory.clement@free-electrons.com,
	thomas.petazzoni@free-electrons.com
Subject: Re: [EXT] Re: [PATCH] mmc: sdhci-xenon: add gpio hard reset support
Date: Thu, 17 Aug 2017 11:34:41 -0700	[thread overview]
Message-ID: <5995E1C1.9000008@marvell.com> (raw)
In-Reply-To: <bd8b35c6-7af4-56dc-0949-ca6deb2016a9@rock-chips.com>

Hi Shawn,

On 08/16/2017 05:47 PM, Shawn Lin wrote:
>
> On 2017/8/17 6:46, Zhoujie Wu wrote:
>> Hi ,
>>
>> On 08/15/2017 05:32 PM, Shawn Lin wrote:
>>> Hi
>>>
>>> On 2017/8/16 6:40, Zhoujie Wu wrote:
>>>> Hi Shawn,
>>>
>>>>> ---------------------------------------------------------------------- 
>>>>>
>>>>> Hi
>>>>>
>>>>> On 2017/8/15 6:19, Zhoujie Wu wrote:
>>>>>> On some platforms, like armada3700, SD card need to
>>>>>> do hard reset by gpio toggling to make it work properly
>>>>>> after warm reset the board.
>>>>>
>>>>> I don't get this that SD card need to do hard reset...
>>>>>
>>>>> I assume what you talk about is either for eMMC or a power-cycle
>>>>> for SD card.
>>>>
>>>> The subject of the patch confused you. What I want is a power-cycle 
>>>> for the SD card. The gpio is  used to enable/disable the vdd power 
>>>> supply for sd card.
>>>> Actually on a3700, when warm reset the board, their is no 
>>>> power-cycle for SD card, which will lead sd card can't response 
>>>> correct ocr and never set S18A unless a power-cycle.
>>>> This is the purpose I submit this patch.
>>>>
>>>
>>> Well, if that is the case, I suggest you to use regulator-gpio.
>>>
>>> i.e:
>>>
>>>        vcc_sd: regulator {
>>>                 compatible = "regulator-gpio";
>>>                 regulator-name = "vcc_sd";
>>>                 regulator-min-microvolt = <3300000>;
>>>                 regulator-max-microvolt = <3300000>;
>>>
>>>                 gpios = <&gpio1 23 GPIO_ACTIVE_HIGH>;
>>>                 enable-active-high;
>>>         };
>>>
>>>     &sdhci {
>>>         vmmc-supply = <&vcc_sd>;
>>>     }
>>>
>>>
>>
>> I tried to use the vmmc-supply regulator before I submit this patch, 
>> I met a issue in this case.
>> sdhci_set_power_reg will be called instead, and I saw 
>> mmc_regulator_set_ocr do enabled the regulator, after that SW will 
>> set 0xf to power controller register, but the register is 
>> self-cleared to 0 soon which lead to later cmd timeout all the time.
>
> I don't parse it from the SDHCI spec that SD bus power bit is
> self-cleared one. Is it sdhci-xenon specific?
>
> If yes, you need to hook you set_power callback and avoid touching this
> bit with some proper description for the reason.
>
>
Thanks for your great help. Today I debugged this issue and found that 
if I used sdhci_set_poewr_noreg right after I enabled the vmmc 
regulator, the whole thing worked well. I can just write power on bit in 
pwr_ctrl_reg.
I think this might be the xenon limitation, even with external vmmc 
power supply, the xenon sdh controller still need to program voltage 
select bits. I will check with our design guys to get more information.
As you said, I have to add set_power callback in our driver to make it 
work perfectly.  And the regulator I defined as a fixed regulator, which 
is always on to make sure the card detection can work without issue.
I abandon this patch and nice to have this talk. Really appreciate.

>> Have you ever met similar issue before?
>
> Haven't.
>
>>
>>>>>
>>>>>> Add gpio hard reset feature for this purpose.
>>>>>>
>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>
>>
>>
>>
>


  reply	other threads:[~2017-08-17 18:34 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-08-14 22:19 [PATCH] mmc: sdhci-xenon: add gpio hard reset support Zhoujie Wu
2017-08-15  2:11 ` Shawn Lin
2017-08-15 22:40   ` [EXT] " Zhoujie Wu
2017-08-15 23:27     ` Zhoujie Wu
2017-08-16  0:32     ` Shawn Lin
2017-08-16  2:44       ` Jisheng Zhang
2017-08-16 22:46       ` Zhoujie Wu
2017-08-17  0:47         ` Shawn Lin
2017-08-17 18:34           ` Zhoujie Wu [this message]
2017-08-15  2:56 ` Jisheng Zhang
2017-08-15 22:43   ` Zhoujie Wu
2017-08-16  2:22     ` Jisheng Zhang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5995E1C1.9000008@marvell.com \
    --to=zjwu@marvell.com \
    --cc=adrian.hunter@intel.com \
    --cc=dingwei@marvell.com \
    --cc=dougj@marvell.com \
    --cc=gregory.clement@free-electrons.com \
    --cc=hannah@marvell.com \
    --cc=hongd@marvell.com \
    --cc=jszhang@marvell.com \
    --cc=kostap@marvell.com \
    --cc=linux-mmc@vger.kernel.org \
    --cc=liuw@marvell.com \
    --cc=nadavh@marvell.com \
    --cc=shawn.lin@rock-chips.com \
    --cc=thomas.petazzoni@free-electrons.com \
    --cc=ulf.hansson@linaro.org \
    --cc=xigu@marvell.com \
    --cc=ygao@marvell.com \
    --cc=zmxu@marvell.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox