[RFC PATCH 0/3] mmc: sdhci: sdhci-msm: Add more debug logs

[RFC PATCH 0/3] mmc: sdhci: sdhci-msm: Add more debug logs

[Ritesh Harjani]

Below is a small set of patches which provide useful info for
debugging issues.

Patch 1 :- This provide callback mechanism for platform drivers
to dump their necessary register info at the time of sdhci_dumpregs.

Patch 2 :- This implements this callback (->platform_dumpregs)
for sdhci-msm.

Patch 3 :- Adds more logs for data errors to aid debugging.

Ritesh Harjani (1):
  mmc: sdhci: Add more debug info in case of data error

Sahitya Tummala (2):
  mmc: sdhci: Add platform_dumpregs callback support to sdhci_ops.
  mmc: sdhci-msm: Implement platform_dumpregs callback in sdhci-msm

 drivers/mmc/host/sdhci-msm.c | 34 ++++++++++++++++++++++++++++++++++
 drivers/mmc/host/sdhci.c     | 24 ++++++++++++++++++++++--
 drivers/mmc/host/sdhci.h     |  1 +
 3 files changed, 57 insertions(+), 2 deletions(-)

-- 
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
a Linux Foundation Collaborative Project.

[RFC PATCH 1/3] mmc: sdhci: Add platform_dumpregs callback support to sdhci_ops.

[Ritesh Harjani]

From: Sahitya Tummala <stummala@codeaurora.org>

Add new host operation ->platform_dumpregs to provide a
mechanism through which host drivers can dump platform
specific registers in addition to SDHC registers
during error conditions.

Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
---
 drivers/mmc/host/sdhci.c | 3 +++
 drivers/mmc/host/sdhci.h | 1 +
 2 files changed, 4 insertions(+)

diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 2390980..73a8918 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -101,6 +101,9 @@ static void sdhci_dumpregs(struct sdhci_host *host)
 			       readl(host->ioaddr + SDHCI_ADMA_ADDRESS));
 	}
 
+	if (host->ops->platform_dumpregs)
+		host->ops->platform_dumpregs(host);
+
 	pr_err(DRIVER_NAME ": ===========================================\n");
 }
 
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 0b66f21..400f3a1 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -564,6 +564,7 @@ struct sdhci_ops {
 					 struct mmc_card *card,
 					 unsigned int max_dtr, int host_drv,
 					 int card_drv, int *drv_type);
+	void	(*platform_dumpregs)(struct sdhci_host *host);
 };
 
 #ifdef CONFIG_MMC_SDHCI_IO_ACCESSORS
-- 
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
a Linux Foundation Collaborative Project.

Re: [RFC PATCH 1/3] mmc: sdhci: Add platform_dumpregs callback support to sdhci_ops.

[Jeremy McNicoll]

On Fri, Dec 30, 2016 at 05:02:09PM +0530, Ritesh Harjani wrote:
> From: Sahitya Tummala <stummala@codeaurora.org>
> 
> Add new host operation ->platform_dumpregs to provide a
> mechanism through which host drivers can dump platform
> specific registers in addition to SDHC registers
> during error conditions.
> 
> Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
> Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
> ---
>  drivers/mmc/host/sdhci.c | 3 +++
>  drivers/mmc/host/sdhci.h | 1 +
>  2 files changed, 4 insertions(+)
>

This change and 2/3 look like they can be squished together into
one patch.  At least that is how I did it, 

https://patchwork.kernel.org/patch/9442449/

-jeremy


> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
> index 2390980..73a8918 100644
> --- a/drivers/mmc/host/sdhci.c
> +++ b/drivers/mmc/host/sdhci.c
> @@ -101,6 +101,9 @@ static void sdhci_dumpregs(struct sdhci_host *host)
>  			       readl(host->ioaddr + SDHCI_ADMA_ADDRESS));
>  	}
>  
> +	if (host->ops->platform_dumpregs)
> +		host->ops->platform_dumpregs(host);
> +
>  	pr_err(DRIVER_NAME ": ===========================================\n");
>  }
>  
> diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
> index 0b66f21..400f3a1 100644
> --- a/drivers/mmc/host/sdhci.h
> +++ b/drivers/mmc/host/sdhci.h
> @@ -564,6 +564,7 @@ struct sdhci_ops {
>  					 struct mmc_card *card,
>  					 unsigned int max_dtr, int host_drv,
>  					 int card_drv, int *drv_type);
> +	void	(*platform_dumpregs)(struct sdhci_host *host);
>  };
>  
>  #ifdef CONFIG_MMC_SDHCI_IO_ACCESSORS
> -- 
> The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
> a Linux Foundation Collaborative Project.
> 

Re: [RFC PATCH 1/3] mmc: sdhci: Add platform_dumpregs callback support to sdhci_ops.

[Ritesh Harjani]

Hi Jeremy,

Thanks for the review.

On 1/14/2017 3:51 AM, Jeremy McNicoll wrote:
> On Fri, Dec 30, 2016 at 05:02:09PM +0530, Ritesh Harjani wrote:
>> From: Sahitya Tummala <stummala@codeaurora.org>
>>
>> Add new host operation ->platform_dumpregs to provide a
>> mechanism through which host drivers can dump platform
>> specific registers in addition to SDHC registers
>> during error conditions.
>>
>> Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
>> Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
>> ---
>>  drivers/mmc/host/sdhci.c | 3 +++
>>  drivers/mmc/host/sdhci.h | 1 +
>>  2 files changed, 4 insertions(+)
>>
>
> This change and 2/3 look like they can be squished together into
> one patch.  At least that is how I did it,

I felt it is better this way to keep sdhci and sdhci-msm patches 
separate. This patch provides mechanism and patch 3/3 provides
implementation of ->platform_dumpregs in sdhci-msm.

Regards
Ritesh


>
> https://patchwork.kernel.org/patch/9442449/
>
> -jeremy
>
>
>> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
>> index 2390980..73a8918 100644
>> --- a/drivers/mmc/host/sdhci.c
>> +++ b/drivers/mmc/host/sdhci.c
>> @@ -101,6 +101,9 @@ static void sdhci_dumpregs(struct sdhci_host *host)
>>  			       readl(host->ioaddr + SDHCI_ADMA_ADDRESS));
>>  	}
>>
>> +	if (host->ops->platform_dumpregs)
>> +		host->ops->platform_dumpregs(host);
>> +
>>  	pr_err(DRIVER_NAME ": ===========================================\n");
>>  }
>>
>> diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
>> index 0b66f21..400f3a1 100644
>> --- a/drivers/mmc/host/sdhci.h
>> +++ b/drivers/mmc/host/sdhci.h
>> @@ -564,6 +564,7 @@ struct sdhci_ops {
>>  					 struct mmc_card *card,
>>  					 unsigned int max_dtr, int host_drv,
>>  					 int card_drv, int *drv_type);
>> +	void	(*platform_dumpregs)(struct sdhci_host *host);
>>  };
>>
>>  #ifdef CONFIG_MMC_SDHCI_IO_ACCESSORS
>> --
>> The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
>> a Linux Foundation Collaborative Project.
>>

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

Re: [RFC PATCH 1/3] mmc: sdhci: Add platform_dumpregs callback support to sdhci_ops.

[Jeremy McNicoll]

On Mon, Jan 16, 2017 at 07:45:48AM +0530, Ritesh Harjani wrote:
> Hi Jeremy,
> 
> Thanks for the review.
> 
> On 1/14/2017 3:51 AM, Jeremy McNicoll wrote:
> >On Fri, Dec 30, 2016 at 05:02:09PM +0530, Ritesh Harjani wrote:
> >>From: Sahitya Tummala <stummala@codeaurora.org>
> >>
> >>Add new host operation ->platform_dumpregs to provide a
> >>mechanism through which host drivers can dump platform
> >>specific registers in addition to SDHC registers
> >>during error conditions.
> >>
> >>Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
> >>Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
> >>---
> >> drivers/mmc/host/sdhci.c | 3 +++
> >> drivers/mmc/host/sdhci.h | 1 +
> >> 2 files changed, 4 insertions(+)
> >>
> >
> >This change and 2/3 look like they can be squished together into
> >one patch.  At least that is how I did it,
> 
> I felt it is better this way to keep sdhci and sdhci-msm patches separate.
> This patch provides mechanism and patch 3/3 provides
> implementation of ->platform_dumpregs in sdhci-msm.
>

Alright, lets go with that approach.  

-jeremy

> Regards
> Ritesh
> 
> 
> >
> >https://patchwork.kernel.org/patch/9442449/
> >
> >-jeremy
> >
> >
> >>diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
> >>index 2390980..73a8918 100644
> >>--- a/drivers/mmc/host/sdhci.c
> >>+++ b/drivers/mmc/host/sdhci.c
> >>@@ -101,6 +101,9 @@ static void sdhci_dumpregs(struct sdhci_host *host)
> >> 			       readl(host->ioaddr + SDHCI_ADMA_ADDRESS));
> >> 	}
> >>
> >>+	if (host->ops->platform_dumpregs)
> >>+		host->ops->platform_dumpregs(host);
> >>+
> >> 	pr_err(DRIVER_NAME ": ===========================================\n");
> >> }
> >>
> >>diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
> >>index 0b66f21..400f3a1 100644
> >>--- a/drivers/mmc/host/sdhci.h
> >>+++ b/drivers/mmc/host/sdhci.h
> >>@@ -564,6 +564,7 @@ struct sdhci_ops {
> >> 					 struct mmc_card *card,
> >> 					 unsigned int max_dtr, int host_drv,
> >> 					 int card_drv, int *drv_type);
> >>+	void	(*platform_dumpregs)(struct sdhci_host *host);
> >> };
> >>
> >> #ifdef CONFIG_MMC_SDHCI_IO_ACCESSORS
> >>--
> >>The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
> >>a Linux Foundation Collaborative Project.
> >>
> 
> -- 
> Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
> a Linux Foundation Collaborative Project

[RFC PATCH 2/3] mmc: sdhci-msm: Implement platform_dumpregs callback in sdhci-msm

[Ritesh Harjani]

From: Sahitya Tummala <stummala@codeaurora.org>

Implement ->platform_dumpregs host operation to print the
platform specific registers in addition to standard SDHC
register during error conditions.

Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
---
 drivers/mmc/host/sdhci-msm.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c
index 32879b8..1241dbd 100644
--- a/drivers/mmc/host/sdhci-msm.c
+++ b/drivers/mmc/host/sdhci-msm.c
@@ -29,6 +29,11 @@
 #define CORE_VERSION_MAJOR_MASK		(0xf << CORE_VERSION_MAJOR_SHIFT)
 #define CORE_VERSION_MINOR_MASK		0xff
 
+#define CORE_MCI_DATA_CNT	0x30
+#define CORE_MCI_STATUS		0x34
+#define CORE_MCI_FIFO_CNT	0x44
+#define CORE_MCI_STATUS2	0x6c
+
 #define CORE_HC_MODE		0x78
 #define HC_MODE_EN		0x1
 #define CORE_POWER		0x0
@@ -77,6 +82,10 @@
 #define CORE_HC_SELECT_IN_HS400	(6 << 19)
 #define CORE_HC_SELECT_IN_MASK	(7 << 19)
 
+#define CORE_VENDOR_SPEC_FUNC2		0x110
+#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR0	0x114
+#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR1	0x118
+
 #define CORE_CSR_CDC_CTLR_CFG0		0x130
 #define CORE_SW_TRIG_FULL_CALIB		BIT(16)
 #define CORE_HW_AUTOCAL_ENA		BIT(17)
@@ -658,6 +667,30 @@ static int sdhci_msm_hs400_dll_calibration(struct sdhci_host *host)
 	return ret;
 }
 
+static void sdhci_msm_dumpregs(struct sdhci_host *host)
+{
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct sdhci_msm_host *msm_host = sdhci_pltfm_priv(pltfm_host);
+
+	pr_err("----------- PLATFORM REGISTER DUMP -----------\n");
+
+	pr_err("Data cnt: 0x%08x | Fifo cnt: 0x%08x | Int sts: 0x%08x | Int sts2: 0x%08x\n",
+	       readl_relaxed(msm_host->core_mem + CORE_MCI_DATA_CNT),
+	       readl_relaxed(msm_host->core_mem + CORE_MCI_FIFO_CNT),
+	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS),
+	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS2));
+	pr_err("DLL cfg:  0x%08x | DLL sts:  0x%08x | SDCC ver: 0x%08x\n",
+	       readl_relaxed(host->ioaddr + CORE_DLL_CONFIG),
+	       readl_relaxed(host->ioaddr + CORE_DLL_STATUS),
+	       readl_relaxed(msm_host->core_mem + CORE_MCI_VERSION));
+	pr_err("Vndr func: 0x%08x | Vndr adma err : addr0: 0x%08x addr1: 0x%08x\n",
+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC),
+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR0),
+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR1));
+	pr_err("Vndr func2: 0x%08x\n",
+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_FUNC2));
+}
+
 static int sdhci_msm_execute_tuning(struct sdhci_host *host, u32 opcode)
 {
 	int tuning_seq_cnt = 3;
@@ -1035,6 +1068,7 @@ static void sdhci_msm_set_clock(struct sdhci_host *host, unsigned int clock)
 	.set_bus_width = sdhci_set_bus_width,
 	.set_uhs_signaling = sdhci_msm_set_uhs_signaling,
 	.voltage_switch = sdhci_msm_voltage_switch,
+	.platform_dumpregs = sdhci_msm_dumpregs,
 };
 
 static const struct sdhci_pltfm_data sdhci_msm_pdata = {
-- 
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
a Linux Foundation Collaborative Project.

Re: [RFC PATCH 2/3] mmc: sdhci-msm: Implement platform_dumpregs callback in sdhci-msm

[Jeremy McNicoll]

On Fri, Dec 30, 2016 at 05:02:10PM +0530, Ritesh Harjani wrote:
> From: Sahitya Tummala <stummala@codeaurora.org>
> 
> Implement ->platform_dumpregs host operation to print the
> platform specific registers in addition to standard SDHC
> register during error conditions.
> 
> Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
> Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
> ---
>  drivers/mmc/host/sdhci-msm.c | 34 ++++++++++++++++++++++++++++++++++
>  1 file changed, 34 insertions(+)
> 
> diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c
> index 32879b8..1241dbd 100644
> --- a/drivers/mmc/host/sdhci-msm.c
> +++ b/drivers/mmc/host/sdhci-msm.c
> @@ -29,6 +29,11 @@
>  #define CORE_VERSION_MAJOR_MASK		(0xf << CORE_VERSION_MAJOR_SHIFT)
>  #define CORE_VERSION_MINOR_MASK		0xff
>  
> +#define CORE_MCI_DATA_CNT	0x30
> +#define CORE_MCI_STATUS		0x34
> +#define CORE_MCI_FIFO_CNT	0x44
> +#define CORE_MCI_STATUS2	0x6c
> +
>  #define CORE_HC_MODE		0x78
>  #define HC_MODE_EN		0x1
>  #define CORE_POWER		0x0
> @@ -77,6 +82,10 @@
>  #define CORE_HC_SELECT_IN_HS400	(6 << 19)
>  #define CORE_HC_SELECT_IN_MASK	(7 << 19)
>  
> +#define CORE_VENDOR_SPEC_FUNC2		0x110
> +#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR0	0x114
> +#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR1	0x118
> +
>  #define CORE_CSR_CDC_CTLR_CFG0		0x130
>  #define CORE_SW_TRIG_FULL_CALIB		BIT(16)
>  #define CORE_HW_AUTOCAL_ENA		BIT(17)
> @@ -658,6 +667,30 @@ static int sdhci_msm_hs400_dll_calibration(struct sdhci_host *host)
>  	return ret;
>  }
>  
> +static void sdhci_msm_dumpregs(struct sdhci_host *host)
> +{
> +	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
> +	struct sdhci_msm_host *msm_host = sdhci_pltfm_priv(pltfm_host);
> +
> +	pr_err("----------- PLATFORM REGISTER DUMP -----------\n");
> +
> +	pr_err("Data cnt: 0x%08x | Fifo cnt: 0x%08x | Int sts: 0x%08x | Int sts2: 0x%08x\n",
> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_DATA_CNT),
> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_FIFO_CNT),
> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS),
> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS2));
> +	pr_err("DLL cfg:  0x%08x | DLL sts:  0x%08x | SDCC ver: 0x%08x\n",
> +	       readl_relaxed(host->ioaddr + CORE_DLL_CONFIG),
> +	       readl_relaxed(host->ioaddr + CORE_DLL_STATUS),
> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_VERSION));
> +	pr_err("Vndr func: 0x%08x | Vndr adma err : addr0: 0x%08x addr1: 0x%08x\n",
> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC),
> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR0),
> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR1));
> +	pr_err("Vndr func2: 0x%08x\n",
> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_FUNC2));
> +}
> +

Why did you omit the printing of CORE_TESTBUS_CONFIG like that of 
https://patchwork.kernel.org/patch/9442449/

and sdhci_dump_state() was dropped too.  Being able to dump the rpm_info
was very handy for issues I was running into.

-jeremy

>  static int sdhci_msm_execute_tuning(struct sdhci_host *host, u32 opcode)
>  {
>  	int tuning_seq_cnt = 3;
> @@ -1035,6 +1068,7 @@ static void sdhci_msm_set_clock(struct sdhci_host *host, unsigned int clock)
>  	.set_bus_width = sdhci_set_bus_width,
>  	.set_uhs_signaling = sdhci_msm_set_uhs_signaling,
>  	.voltage_switch = sdhci_msm_voltage_switch,
> +	.platform_dumpregs = sdhci_msm_dumpregs,
>  };
>  
>  static const struct sdhci_pltfm_data sdhci_msm_pdata = {
> -- 
> The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
> a Linux Foundation Collaborative Project.
> 

Re: [RFC PATCH 2/3] mmc: sdhci-msm: Implement platform_dumpregs callback in sdhci-msm

[Ritesh Harjani]

Hi Jeremy,

On 1/14/2017 4:01 AM, Jeremy McNicoll wrote:
> On Fri, Dec 30, 2016 at 05:02:10PM +0530, Ritesh Harjani wrote:
>> From: Sahitya Tummala <stummala@codeaurora.org>
>>
>> Implement ->platform_dumpregs host operation to print the
>> platform specific registers in addition to standard SDHC
>> register during error conditions.
>>
>> Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
>> Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
>> ---
>>  drivers/mmc/host/sdhci-msm.c | 34 ++++++++++++++++++++++++++++++++++
>>  1 file changed, 34 insertions(+)
>>
>> diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c
>> index 32879b8..1241dbd 100644
>> --- a/drivers/mmc/host/sdhci-msm.c
>> +++ b/drivers/mmc/host/sdhci-msm.c
>> @@ -29,6 +29,11 @@
>>  #define CORE_VERSION_MAJOR_MASK		(0xf << CORE_VERSION_MAJOR_SHIFT)
>>  #define CORE_VERSION_MINOR_MASK		0xff
>>
>> +#define CORE_MCI_DATA_CNT	0x30
>> +#define CORE_MCI_STATUS		0x34
>> +#define CORE_MCI_FIFO_CNT	0x44
>> +#define CORE_MCI_STATUS2	0x6c
>> +
>>  #define CORE_HC_MODE		0x78
>>  #define HC_MODE_EN		0x1
>>  #define CORE_POWER		0x0
>> @@ -77,6 +82,10 @@
>>  #define CORE_HC_SELECT_IN_HS400	(6 << 19)
>>  #define CORE_HC_SELECT_IN_MASK	(7 << 19)
>>
>> +#define CORE_VENDOR_SPEC_FUNC2		0x110
>> +#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR0	0x114
>> +#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR1	0x118
>> +
>>  #define CORE_CSR_CDC_CTLR_CFG0		0x130
>>  #define CORE_SW_TRIG_FULL_CALIB		BIT(16)
>>  #define CORE_HW_AUTOCAL_ENA		BIT(17)
>> @@ -658,6 +667,30 @@ static int sdhci_msm_hs400_dll_calibration(struct sdhci_host *host)
>>  	return ret;
>>  }
>>
>> +static void sdhci_msm_dumpregs(struct sdhci_host *host)
>> +{
>> +	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
>> +	struct sdhci_msm_host *msm_host = sdhci_pltfm_priv(pltfm_host);
>> +
>> +	pr_err("----------- PLATFORM REGISTER DUMP -----------\n");
>> +
>> +	pr_err("Data cnt: 0x%08x | Fifo cnt: 0x%08x | Int sts: 0x%08x | Int sts2: 0x%08x\n",
>> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_DATA_CNT),
>> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_FIFO_CNT),
>> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS),
>> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS2));
>> +	pr_err("DLL cfg:  0x%08x | DLL sts:  0x%08x | SDCC ver: 0x%08x\n",
>> +	       readl_relaxed(host->ioaddr + CORE_DLL_CONFIG),
>> +	       readl_relaxed(host->ioaddr + CORE_DLL_STATUS),
>> +	       readl_relaxed(msm_host->core_mem + CORE_MCI_VERSION));
>> +	pr_err("Vndr func: 0x%08x | Vndr adma err : addr0: 0x%08x addr1: 0x%08x\n",
>> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC),
>> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR0),
>> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR1));
>> +	pr_err("Vndr func2: 0x%08x\n",
>> +	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_FUNC2));
>> +}
>> +
>
> Why did you omit the printing of CORE_TESTBUS_CONFIG like that of
> https://patchwork.kernel.org/patch/9442449/
Actually testbus registers aren't something we analyze every-time.
So I have skipped it for now. We can add it later if and when required.


>
> and sdhci_dump_state() was dropped too.  Being able to dump the rpm_info
> was very handy for issues I was running into.
Sure, let me add that info as well.

>
> -jeremy
>
>>  static int sdhci_msm_execute_tuning(struct sdhci_host *host, u32 opcode)
>>  {
>>  	int tuning_seq_cnt = 3;
>> @@ -1035,6 +1068,7 @@ static void sdhci_msm_set_clock(struct sdhci_host *host, unsigned int clock)
>>  	.set_bus_width = sdhci_set_bus_width,
>>  	.set_uhs_signaling = sdhci_msm_set_uhs_signaling,
>>  	.voltage_switch = sdhci_msm_voltage_switch,
>> +	.platform_dumpregs = sdhci_msm_dumpregs,
>>  };
>>
>>  static const struct sdhci_pltfm_data sdhci_msm_pdata = {
>> --
>> The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
>> a Linux Foundation Collaborative Project.
>>

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

Re: [RFC PATCH 2/3] mmc: sdhci-msm: Implement platform_dumpregs callback in sdhci-msm

[Jeremy McNicoll]

On Mon, Jan 16, 2017 at 07:51:40AM +0530, Ritesh Harjani wrote:
> Hi Jeremy,
> 
> On 1/14/2017 4:01 AM, Jeremy McNicoll wrote:
> >On Fri, Dec 30, 2016 at 05:02:10PM +0530, Ritesh Harjani wrote:
> >>From: Sahitya Tummala <stummala@codeaurora.org>
> >>
> >>Implement ->platform_dumpregs host operation to print the
> >>platform specific registers in addition to standard SDHC
> >>register during error conditions.
> >>
> >>Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
> >>Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
> >>---
> >> drivers/mmc/host/sdhci-msm.c | 34 ++++++++++++++++++++++++++++++++++
> >> 1 file changed, 34 insertions(+)
> >>
> >>diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c
> >>index 32879b8..1241dbd 100644
> >>--- a/drivers/mmc/host/sdhci-msm.c
> >>+++ b/drivers/mmc/host/sdhci-msm.c
> >>@@ -29,6 +29,11 @@
> >> #define CORE_VERSION_MAJOR_MASK		(0xf << CORE_VERSION_MAJOR_SHIFT)
> >> #define CORE_VERSION_MINOR_MASK		0xff
> >>
> >>+#define CORE_MCI_DATA_CNT	0x30
> >>+#define CORE_MCI_STATUS		0x34
> >>+#define CORE_MCI_FIFO_CNT	0x44
> >>+#define CORE_MCI_STATUS2	0x6c
> >>+
> >> #define CORE_HC_MODE		0x78
> >> #define HC_MODE_EN		0x1
> >> #define CORE_POWER		0x0
> >>@@ -77,6 +82,10 @@
> >> #define CORE_HC_SELECT_IN_HS400	(6 << 19)
> >> #define CORE_HC_SELECT_IN_MASK	(7 << 19)
> >>
> >>+#define CORE_VENDOR_SPEC_FUNC2		0x110
> >>+#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR0	0x114
> >>+#define CORE_VENDOR_SPEC_ADMA_ERR_ADDR1	0x118
> >>+
> >> #define CORE_CSR_CDC_CTLR_CFG0		0x130
> >> #define CORE_SW_TRIG_FULL_CALIB		BIT(16)
> >> #define CORE_HW_AUTOCAL_ENA		BIT(17)
> >>@@ -658,6 +667,30 @@ static int sdhci_msm_hs400_dll_calibration(struct sdhci_host *host)
> >> 	return ret;
> >> }
> >>
> >>+static void sdhci_msm_dumpregs(struct sdhci_host *host)
> >>+{
> >>+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
> >>+	struct sdhci_msm_host *msm_host = sdhci_pltfm_priv(pltfm_host);
> >>+
> >>+	pr_err("----------- PLATFORM REGISTER DUMP -----------\n");
> >>+
> >>+	pr_err("Data cnt: 0x%08x | Fifo cnt: 0x%08x | Int sts: 0x%08x | Int sts2: 0x%08x\n",
> >>+	       readl_relaxed(msm_host->core_mem + CORE_MCI_DATA_CNT),
> >>+	       readl_relaxed(msm_host->core_mem + CORE_MCI_FIFO_CNT),
> >>+	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS),
> >>+	       readl_relaxed(msm_host->core_mem + CORE_MCI_STATUS2));
> >>+	pr_err("DLL cfg:  0x%08x | DLL sts:  0x%08x | SDCC ver: 0x%08x\n",
> >>+	       readl_relaxed(host->ioaddr + CORE_DLL_CONFIG),
> >>+	       readl_relaxed(host->ioaddr + CORE_DLL_STATUS),
> >>+	       readl_relaxed(msm_host->core_mem + CORE_MCI_VERSION));
> >>+	pr_err("Vndr func: 0x%08x | Vndr adma err : addr0: 0x%08x addr1: 0x%08x\n",
> >>+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC),
> >>+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR0),
> >>+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_ADMA_ERR_ADDR1));
> >>+	pr_err("Vndr func2: 0x%08x\n",
> >>+	       readl_relaxed(host->ioaddr + CORE_VENDOR_SPEC_FUNC2));
> >>+}
> >>+
> >
> >Why did you omit the printing of CORE_TESTBUS_CONFIG like that of
> >https://patchwork.kernel.org/patch/9442449/
> Actually testbus registers aren't something we analyze every-time.
> So I have skipped it for now. We can add it later if and when required.
> 

Sounds reasonable.  

> 
> >
> >and sdhci_dump_state() was dropped too.  Being able to dump the rpm_info
> >was very handy for issues I was running into.
> Sure, let me add that info as well.
> 

Great, as it was useful when I was debugging SDHCI enablement. 

-jeremy

> >
> >-jeremy
> >
> >> static int sdhci_msm_execute_tuning(struct sdhci_host *host, u32 opcode)
> >> {
> >> 	int tuning_seq_cnt = 3;
> >>@@ -1035,6 +1068,7 @@ static void sdhci_msm_set_clock(struct sdhci_host *host, unsigned int clock)
> >> 	.set_bus_width = sdhci_set_bus_width,
> >> 	.set_uhs_signaling = sdhci_msm_set_uhs_signaling,
> >> 	.voltage_switch = sdhci_msm_voltage_switch,
> >>+	.platform_dumpregs = sdhci_msm_dumpregs,
> >> };
> >>
> >> static const struct sdhci_pltfm_data sdhci_msm_pdata = {
> >>--
> >>The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
> >>a Linux Foundation Collaborative Project.
> >>
> 
> -- 
> Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
> a Linux Foundation Collaborative Project

[RFC PATCH 3/3] mmc: sdhci: Add more debug info in case of data error

[Ritesh Harjani]

print error log message and dump sdhc registers for debugging
purpose in case of data errors (except when tuning commands
generate CRC/timeout/end bit errors).

Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
---
 drivers/mmc/host/sdhci.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 73a8918..2e51e49 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -2616,9 +2616,26 @@ static void sdhci_data_irq(struct sdhci_host *host, u32 intmask)
 			host->ops->adma_workaround(host, intmask);
 	}
 
-	if (host->data->error)
+	if (host->data->error) {
+		bool pr_msg = true;
+
+		if (intmask & (SDHCI_INT_DATA_CRC | SDHCI_INT_DATA_TIMEOUT |
+		    SDHCI_INT_DATA_END_BIT)) {
+			command = SDHCI_GET_CMD(sdhci_readw(host,
+							    SDHCI_COMMAND));
+			if (command == MMC_SEND_TUNING_BLOCK_HS200 ||
+			    command == MMC_SEND_TUNING_BLOCK)
+				pr_msg = false;
+		}
+		if (pr_msg) {
+			pr_err("%s: data txfr (0x%08x) error: %d\n",
+			       mmc_hostname(host->mmc), intmask,
+			       host->data->error);
+			sdhci_dumpregs(host);
+		}
+
 		sdhci_finish_data(host);
-	else {
+	} else {
 		if (intmask & (SDHCI_INT_DATA_AVAIL | SDHCI_INT_SPACE_AVAIL))
 			sdhci_transfer_pio(host);
 
-- 
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
a Linux Foundation Collaborative Project.

Re: [RFC PATCH 0/3] mmc: sdhci: sdhci-msm: Add more debug logs

[Jeremy McNicoll]

On Fri, Dec 30, 2016 at 05:02:08PM +0530, Ritesh Harjani wrote:
> Below is a small set of patches which provide useful info for
> debugging issues.
> 
> Patch 1 :- This provide callback mechanism for platform drivers
> to dump their necessary register info at the time of sdhci_dumpregs.
> 
> Patch 2 :- This implements this callback (->platform_dumpregs)
> for sdhci-msm.
> 
> Patch 3 :- Adds more logs for data errors to aid debugging.
> 
> Ritesh Harjani (1):
>   mmc: sdhci: Add more debug info in case of data error
> 
> Sahitya Tummala (2):
>   mmc: sdhci: Add platform_dumpregs callback support to sdhci_ops.
>   mmc: sdhci-msm: Implement platform_dumpregs callback in sdhci-msm
> 
>  drivers/mmc/host/sdhci-msm.c | 34 ++++++++++++++++++++++++++++++++++
>  drivers/mmc/host/sdhci.c     | 24 ++++++++++++++++++++++--
>  drivers/mmc/host/sdhci.h     |  1 +
>  3 files changed, 57 insertions(+), 2 deletions(-)
>

This series was part of my SDHCI enablement on my Nexus 5X and
everything still seems to work ok.

Tested-by: Jeremy McNicoll <jeremymc@redhat.com>


> -- 
> The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
> a Linux Foundation Collaborative Project.
>