Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH V4 5/7] mmc: sdhci: add Auto CMD Auto Select support
To:     Chunyan Zhang <zhang.lyra@gmail.com>
Cc:     Chunyan Zhang <zhang.chunyan@linaro.org>,
        Ulf Hansson <ulf.hansson@linaro.org>,
        linux-mmc@vger.kernel.org,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Orson Zhai <orsonzhai@gmail.com>,
        Baolin Wang <baolin.wang@linaro.org>,
        Billows Wu <billows.wu@spreadtrum.com>,
        Jason Wu <jason.wu@unisoc.com>
References: <1532340508-8749-1-git-send-email-zhang.chunyan@linaro.org>
 <1532340508-8749-6-git-send-email-zhang.chunyan@linaro.org>
 <33ceb61c-3e09-c336-e9b9-87c01336f56c@intel.com>
 <CAAfSe-vVEve63y-Pc0QF6w7vc37cbbbOknGkp9SmocXyNX3J4g@mail.gmail.com>
 <f810df5a-2cc4-a534-9fdc-2575ae81cf4b@intel.com>
 <CAAfSe-uyjnBGk7w2KMgrgEYVgYY6p6EqCWTUW+r2BWDoYmPpLg@mail.gmail.com>
From:   Adrian Hunter <adrian.hunter@intel.com>
Organization: Intel Finland Oy, Registered Address: PL 281, 00181 Helsinki,
 Business Identity Code: 0357606 - 4, Domiciled in Helsinki
Message-ID: <23282747-4ae1-d627-a537-590b9315b941@intel.com>
Date:   Tue, 31 Jul 2018 11:56:27 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.0
MIME-Version: 1.0
In-Reply-To: <CAAfSe-uyjnBGk7w2KMgrgEYVgYY6p6EqCWTUW+r2BWDoYmPpLg@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 31/07/18 11:36, Chunyan Zhang wrote:
> On 31 July 2018 at 16:05, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> On 31/07/18 10:04, Chunyan Zhang wrote:
>>> Hi Adrian,
>>>
>>> On 30 July 2018 at 21:06, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>> On 23/07/18 13:08, Chunyan Zhang wrote:
>>>>> As SD Host Controller Specification v4.10 documents:
>>>>> Host Controller Version 4.10 defines this "Auto CMD Auto Select" mode.
>>>>> Selection of Auto CMD depends on setting of CMD23 Enable in the Host
>>>>> Control 2 register which indicates whether card supports CMD23. If CMD23
>>>>> Enable =1, Auto CMD23 is used and if CMD23 Enable =0, Auto CMD12 is
>>>>> used. In case of Version 4.10 or later, use of Auto CMD Auto Select is
>>>>> recommended rather than use of Auto CMD12 Enable or Auto CMD23
>>>>> Enable.
>>>>>
>>>>> This patch add this new mode support.
>>>>>
>>>>> Signed-off-by: Chunyan Zhang <zhang.chunyan@linaro.org>
>>>>> ---
>>>>>  drivers/mmc/host/sdhci.c | 61 +++++++++++++++++++++++++++++++++++++++---------
>>>>>  drivers/mmc/host/sdhci.h |  2 ++
>>>>>  2 files changed, 52 insertions(+), 11 deletions(-)
>>>>>
>>>>> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
>>>>> index 5acea3d..5c60590 100644
>>>>> --- a/drivers/mmc/host/sdhci.c
>>>>> +++ b/drivers/mmc/host/sdhci.c
>>>>> @@ -311,6 +311,23 @@ static void sdhci_config_dma(struct sdhci_host *host)
>>>>>       sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
>>>>>  }
>>>>>
>>>>> +static void sdhci_enable_cmd23(struct sdhci_host *host)
>>>>> +{
>>>>> +     u16 ctrl2;
>>>>> +
>>>>> +     /*
>>>>> +      * This is used along with "Auto CMD Auto Select" feature,
>>>>> +      * which is introduced from v4.10, if card supports CMD23,
>>>>> +      * Auto CMD23 should be used instead of Auto CMD12.
>>>>> +      */
>>>>> +     if (host->version >= SDHCI_SPEC_410 &&
>>>>> +         (host->mmc->caps & MMC_CAP_CMD23)) {
>>>>
>>>> That is the host capability.  It needs to be the card capability.
>>>
>>> Could you please elaborate the issue?
>>>
>>> I thought we're setting for host here.  Do you mean we need to see the
>>> card capability?
>>
>> Yes. SDHCI_TRNS_AUTO_SEL selects auto-CMD23 or auto-CMD12 based on this
>> setting, so this must reflect the card's capability.
> 
> Got it, but how to know if the card supports CMD23?

At the moment the only way of knowing is if a request is received with mrq->sbc

> 
>>
>>>
>>>>
>>>>> +             ctrl2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
>>>>> +             ctrl2 |= SDHCI_CMD23_ENABLE;
>>>>> +             sdhci_writew(host, ctrl2, SDHCI_HOST_CONTROL2);
>>>>> +     }
>>>>> +}
>>>>> +
>>>>>  static void sdhci_init(struct sdhci_host *host, int soft)
>>>>>  {
>>>>>       struct mmc_host *mmc = host->mmc;
>>>>> @@ -329,6 +346,8 @@ static void sdhci_init(struct sdhci_host *host, int soft)
>>>>>               host->clock = 0;
>>>>>               mmc->ops->set_ios(mmc, &mmc->ios);
>>>>>       }
>>>>> +
>>>>> +     sdhci_enable_cmd23(host);
>>>>>  }
>>>>>
>>>>>  static void sdhci_reinit(struct sdhci_host *host)
>>>>> @@ -1093,6 +1112,36 @@ static inline bool sdhci_auto_cmd12(struct sdhci_host *host,
>>>>>              !mrq->cap_cmd_during_tfr;
>>>>>  }
>>>>>
>>>>> +static inline void sdhci_auto_cmd_select(struct sdhci_host *host,
>>>>> +                                      struct mmc_command *cmd,
>>>>> +                                      u16 *mode)
>>>>> +{
>>>>> +     bool use_cmd12 = sdhci_auto_cmd12(host, cmd->mrq) &&
>>>>> +                      (cmd->opcode != SD_IO_RW_EXTENDED);
>>>>> +     bool use_cmd23 = cmd->mrq->sbc && (host->flags & SDHCI_AUTO_CMD23);
>>>>> +
>>>>> +     /*
>>>>> +      * In case of Version 4.10 or later, use of 'Auto CMD Auto
>>>>> +      * Select' is recommended rather than use of 'Auto CMD12
>>>>> +      * Enable' or 'Auto CMD23 Enable'.
>>>>> +      */
>>>>> +     if (host->version >= SDHCI_SPEC_410 && (use_cmd12 || use_cmd23)) {
>>>>> +             *mode |= SDHCI_TRNS_AUTO_SEL;
>>>>
>>>> As noted in patch 4, the CMD23 argument is not just the block count for eMMC.
>>>
>>> Probably I haven't got your point...
>>>
>>> >From what I understand, auto_sel mode doesn't need argument2. Doesn't
>>> this work for eMMC?
>>
>> CMD23 always needs an argument
> 
> But setting argument for CMD23 will cover the block count value we set
> before, that will lead mounting emmc to fail.

Does it fail because it contains cmd->mrq->sbc->arg or does it fail because
it gets written twice?

> 
>>
>>>
>>> The test platform I used was just eMMC only, haven't found problem.
>>
>> Because the bits that are missing from the CMD23 argument (reliable write,
>> context id, etc) will not make the command fail.
>>
>>>
>>> Thanks,
>>> Chunyan
>>>
>>>>
>>>>> +             return;
>>>>> +     }
>>>>> +
>>>>> +     /*
>>>>> +      * If we are sending CMD23, CMD12 never gets sent
>>>>> +      * on successful completion (so no Auto-CMD12).
>>>>> +      */
>>>>> +     if (use_cmd12) {
>>>>> +             *mode |= SDHCI_TRNS_AUTO_CMD12;
>>>>> +     } else if (use_cmd23) {
>>>>> +             *mode |= SDHCI_TRNS_AUTO_CMD23;
>>>>> +             sdhci_writel(host, cmd->mrq->sbc->arg, SDHCI_ARGUMENT2);
>>>>> +     }
>>>>> +}
>>>>> +
>>>>>  static void sdhci_set_transfer_mode(struct sdhci_host *host,
>>>>>       struct mmc_command *cmd)
>>>>>  {
>>>>> @@ -1119,17 +1168,7 @@ static void sdhci_set_transfer_mode(struct sdhci_host *host,
>>>>>
>>>>>       if (mmc_op_multi(cmd->opcode) || data->blocks > 1) {
>>>>>               mode = SDHCI_TRNS_BLK_CNT_EN | SDHCI_TRNS_MULTI;
>>>>> -             /*
>>>>> -              * If we are sending CMD23, CMD12 never gets sent
>>>>> -              * on successful completion (so no Auto-CMD12).
>>>>> -              */
>>>>> -             if (sdhci_auto_cmd12(host, cmd->mrq) &&
>>>>> -                 (cmd->opcode != SD_IO_RW_EXTENDED))
>>>>> -                     mode |= SDHCI_TRNS_AUTO_CMD12;
>>>>> -             else if (cmd->mrq->sbc && (host->flags & SDHCI_AUTO_CMD23)) {
>>>>> -                     mode |= SDHCI_TRNS_AUTO_CMD23;
>>>>> -                     sdhci_writel(host, cmd->mrq->sbc->arg, SDHCI_ARGUMENT2);
>>>>> -             }
>>>>> +             sdhci_auto_cmd_select(host, cmd, &mode);
>>>>>       }
>>>>>
>>>>>       if (data->flags & MMC_DATA_READ)
>>>>> diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
>>>>> index 81aae07..a8f4ec2 100644
>>>>> --- a/drivers/mmc/host/sdhci.h
>>>>> +++ b/drivers/mmc/host/sdhci.h
>>>>> @@ -42,6 +42,7 @@
>>>>>  #define  SDHCI_TRNS_BLK_CNT_EN       0x02
>>>>>  #define  SDHCI_TRNS_AUTO_CMD12       0x04
>>>>>  #define  SDHCI_TRNS_AUTO_CMD23       0x08
>>>>> +#define  SDHCI_TRNS_AUTO_SEL 0x0C
>>>>>  #define  SDHCI_TRNS_READ     0x10
>>>>>  #define  SDHCI_TRNS_MULTI    0x20
>>>>>
>>>>> @@ -185,6 +186,7 @@
>>>>>  #define   SDHCI_CTRL_DRV_TYPE_D              0x0030
>>>>>  #define  SDHCI_CTRL_EXEC_TUNING              0x0040
>>>>>  #define  SDHCI_CTRL_TUNED_CLK                0x0080
>>>>> +#define  SDHCI_CMD23_ENABLE          0x0800
>>>>>  #define  SDHCI_CTRL_V4_MODE          0x1000
>>>>>  #define  SDHCI_CTRL_64BIT_ADDR               0x2000
>>>>>  #define  SDHCI_CTRL_PRESET_VAL_ENABLE        0x8000
>>>>>
>>>>
>>>
>>
>