mbox series

[V3,0/4] mmc: add error statistics for eMMC and SD card

Message ID 1642699582-14785-1-git-send-email-quic_c_sbhanu@quicinc.com
Headers show
Series mmc: add error statistics for eMMC and SD card | expand

Message

Shaik Sajida Bhanu Jan. 20, 2022, 5:26 p.m. UTC 
Changes since V2:
	-Removed userspace error stats clear debug fs entry as suggested by Adrain Hunter.
	-Split patch into 4 patches
		[PATCH V3 1/4] : sdhci driver
		[PATCH V3 2/4] : debug fs entries
		[PATCH V3 3/4] : core driver
		[PATCH V3 4/4] : cqhci driver
	-Used for loop to print error messages instead of using printf
	 statements for all error messages as suggested by Adrain Hunter.
	-Introduced one flag to enable error stats feature, if any other
	 client wants to use this feature, they need to enable that flag.
	-Moved reset command timeout error statement to card init flow
	 as suggested by Adrain Hunter.

Changes since V1:
	-Removed sysfs entry for eMMC and SD card error statistics and added
	 debugfs entry as suggested by Adrian Hunter and Ulf Hansson.

Shaik Sajida Bhanu (4):
  mmc: sdhci: Capture eMMC and SD card errors
  mmc: debugfs: Add debug fs entry for mmc driver
  mmc: core: Capture eMMC and SD card errors
  mmc: cqhci: Capture eMMC and SD card errors

 drivers/mmc/core/core.c       |  8 +++++
 drivers/mmc/core/debugfs.c    | 81 +++++++++++++++++++++++++++++++++++++++++++
 drivers/mmc/core/queue.c      |  3 ++
 drivers/mmc/host/cqhci-core.c |  9 ++++-
 drivers/mmc/host/sdhci-msm.c  |  3 ++
 drivers/mmc/host/sdhci.c      | 72 +++++++++++++++++++++++++++++++-------
 include/linux/mmc/host.h      | 31 +++++++++++++++++
 7 files changed, 194 insertions(+), 13 deletions(-)

Comments

Adrian Hunter Jan. 21, 2022, 8:20 a.m. UTC | #1 
On 20/01/2022 19:26, Shaik Sajida Bhanu wrote:
> Add changes to capture eMMC and SD card errors.
> This is useful for debug and testing.
> 
> Signed-off-by: Shaik Sajida Bhanu <quic_c_sbhanu@quicinc.com>
> Signed-off-by: Liangliang Lu <luliang@codeaurora.org>
> Signed-off-by: Sayali Lokhande <sayalil@codeaurora.org>
> Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
> Signed-off-by: Ram Prakash Gupta <quic_rampraka@quicinc.com>
> ---
>  drivers/mmc/core/core.c  | 8 ++++++++
>  drivers/mmc/core/queue.c | 3 +++
>  2 files changed, 11 insertions(+)
> 
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
> index 368f104..c586d69 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -2242,6 +2242,14 @@ void mmc_rescan(struct work_struct *work)
>  		if (freqs[i] <= host->f_min)
>  			break;
>  	}
> +
> +	/*
> +	 * Ignore the command timeout errors observed during
> +	 * the card init as those are excepted.
> +	 */
> +

Please remove blank line here.

> +	if (host && host->err_stats_enabled)

The condition is not needed.

> +		host->err_stats[MMC_ERR_CMD_TIMEOUT] = 0;

Please put this after successful call to mmc_rescan_try_freq

>  	mmc_release_host(host);
>  
>   out:
> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> index c69b2d9..7dc9dfb 100644
> --- a/drivers/mmc/core/queue.c
> +++ b/drivers/mmc/core/queue.c
> @@ -100,6 +100,9 @@ static enum blk_eh_timer_return mmc_cqe_timed_out(struct request *req)
>  	enum mmc_issue_type issue_type = mmc_issue_type(mq, req);
>  	bool recovery_needed = false;
>  
> +	if (host->err_stats_enabled)
> +		mmc_debugfs_err_stats_inc(host, MMC_ERR_CMDQ_REQ_TIMEOUT);

Doesn't this get covered by the drivers.  It seems like this should not be needed.

> +
>  	switch (issue_type) {
>  	case MMC_ISSUE_ASYNC:
>  	case MMC_ISSUE_DCMD:
>
Adrian Hunter Jan. 21, 2022, 8:22 a.m. UTC | #2 
On 20/01/2022 19:26, Shaik Sajida Bhanu wrote:
> Add changes to capture eMMC and SD card errors.
> This is useful for debug and testing.
> 
> Signed-off-by: Shaik Sajida Bhanu <quic_c_sbhanu@quicinc.com>
> Signed-off-by: Liangliang Lu <luliang@codeaurora.org>
> Signed-off-by: Sayali Lokhande <sayalil@codeaurora.org>
> Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
> Signed-off-by: Ram Prakash Gupta <quic_rampraka@quicinc.com>
> ---
>  drivers/mmc/host/cqhci-core.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/mmc/host/cqhci-core.c b/drivers/mmc/host/cqhci-core.c
> index b0d30c3..2908d30 100644
> --- a/drivers/mmc/host/cqhci-core.c
> +++ b/drivers/mmc/host/cqhci-core.c
> @@ -822,8 +822,15 @@ irqreturn_t cqhci_irq(struct mmc_host *mmc, u32 intmask, int cmd_error,
>  	pr_debug("%s: cqhci: IRQ status: 0x%08x\n", mmc_hostname(mmc), status);
>  
>  	if ((status & (CQHCI_IS_RED | CQHCI_IS_GCE | CQHCI_IS_ICCE)) ||
> -	    cmd_error || data_error)
> +	    cmd_error || data_error) {
> +		if ((status & CQHCI_IS_RED) && mmc->err_stats_enabled)
> +			mmc_debugfs_err_stats_inc(mmc, MMC_ERR_CMDQ_RED);
> +		if ((status & CQHCI_IS_GCE) && (mmc->err_stats_enabled))
> +			mmc_debugfs_err_stats_inc(mmc, MMC_ERR_CMDQ_GCE);
> +		if ((status & CQHCI_IS_ICCE) && mmc->err_stats_enabled)
> +			mmc_debugfs_err_stats_inc(mmc, MMC_ERR_CMDQ_ICCE);

Please don't check mmc->err_stats_enabled

>  		cqhci_error_irq(mmc, status, cmd_error, data_error);
> +	}
>  
>  	if (status & CQHCI_IS_TCC) {
>  		/* read TCN and complete the request */
>
Shaik Sajida Bhanu Jan. 25, 2022, 6:40 p.m. UTC | #3 
Hi,

Thanks for the review.

Please find the inline comments.

Thanks,
Sajida

-----Original Message-----
From: Adrian Hunter <adrian.hunter@intel.com> 
Sent: Friday, January 21, 2022 1:50 PM
To: Sajida Bhanu (Temp) (QUIC) <quic_c_sbhanu@quicinc.com>; Asutosh Das (QUIC) <quic_asutoshd@quicinc.com>; ulf.hansson@linaro.org; agross@kernel.org; bjorn.andersson@linaro.org; linux-mmc@vger.kernel.org; linux-arm-msm@vger.kernel.org; linux-kernel@vger.kernel.org
Cc: stummala@codeaurora.org; vbadigan@codeaurora.org; Ram Prakash Gupta (QUIC) <quic_rampraka@quicinc.com>; Pradeep Pragallapati (QUIC) <quic_pragalla@quicinc.com>; sartgarg@codeaurora.org; nitirawa@codeaurora.org; sayalil@codeaurora.org; Liangliang Lu <luliang@codeaurora.org>; Bao D . Nguyen <nguyenb@codeaurora.org>
Subject: Re: [PATCH V3 3/4] mmc: core: Capture eMMC and SD card errors

On 20/01/2022 19:26, Shaik Sajida Bhanu wrote:
> Add changes to capture eMMC and SD card errors.
> This is useful for debug and testing.
> 
> Signed-off-by: Shaik Sajida Bhanu <quic_c_sbhanu@quicinc.com>
> Signed-off-by: Liangliang Lu <luliang@codeaurora.org>
> Signed-off-by: Sayali Lokhande <sayalil@codeaurora.org>
> Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
> Signed-off-by: Ram Prakash Gupta <quic_rampraka@quicinc.com>
> ---
>  drivers/mmc/core/core.c  | 8 ++++++++  drivers/mmc/core/queue.c | 3 
> +++
>  2 files changed, 11 insertions(+)
> 
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 
> 368f104..c586d69 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -2242,6 +2242,14 @@ void mmc_rescan(struct work_struct *work)
>  		if (freqs[i] <= host->f_min)
>  			break;
>  	}
> +
> +	/*
> +	 * Ignore the command timeout errors observed during
> +	 * the card init as those are excepted.
> +	 */
> +

Please remove blank line here.

>>>>Sure.

> +	if (host && host->err_stats_enabled)

The condition is not needed.

>>>>Sure .

> +		host->err_stats[MMC_ERR_CMD_TIMEOUT] = 0;

Please put this after successful call to mmc_rescan_try_freq

>>>>>Sure.

>  	mmc_release_host(host);
>  
>   out:
> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c index 
> c69b2d9..7dc9dfb 100644
> --- a/drivers/mmc/core/queue.c
> +++ b/drivers/mmc/core/queue.c
> @@ -100,6 +100,9 @@ static enum blk_eh_timer_return mmc_cqe_timed_out(struct request *req)
>  	enum mmc_issue_type issue_type = mmc_issue_type(mq, req);
>  	bool recovery_needed = false;
>  
> +	if (host->err_stats_enabled)
> +		mmc_debugfs_err_stats_inc(host, MMC_ERR_CMDQ_REQ_TIMEOUT);

Doesn't this get covered by the drivers.  It seems like this should not be needed.

>>>>>>Okay

> +
>  	switch (issue_type) {
>  	case MMC_ISSUE_ASYNC:
>  	case MMC_ISSUE_DCMD:
>
Shaik Sajida Bhanu Jan. 25, 2022, 6:41 p.m. UTC | #4 
Hi,

Thanks for the review.

Please find the inline comments.

Thanks,
Sajida
-----Original Message-----
From: Adrian Hunter <adrian.hunter@intel.com> 
Sent: Friday, January 21, 2022 1:52 PM
To: Sajida Bhanu (Temp) (QUIC) <quic_c_sbhanu@quicinc.com>; Asutosh Das (QUIC) <quic_asutoshd@quicinc.com>; ulf.hansson@linaro.org; agross@kernel.org; bjorn.andersson@linaro.org; linux-mmc@vger.kernel.org; linux-arm-msm@vger.kernel.org; linux-kernel@vger.kernel.org
Cc: stummala@codeaurora.org; vbadigan@codeaurora.org; Ram Prakash Gupta (QUIC) <quic_rampraka@quicinc.com>; Pradeep Pragallapati (QUIC) <quic_pragalla@quicinc.com>; sartgarg@codeaurora.org; nitirawa@codeaurora.org; sayalil@codeaurora.org; Liangliang Lu <luliang@codeaurora.org>; Bao D . Nguyen <nguyenb@codeaurora.org>
Subject: Re: [PATCH V3 4/4] mmc: cqhci: Capture eMMC and SD card errors

On 20/01/2022 19:26, Shaik Sajida Bhanu wrote:
> Add changes to capture eMMC and SD card errors.
> This is useful for debug and testing.
> 
> Signed-off-by: Shaik Sajida Bhanu <quic_c_sbhanu@quicinc.com>
> Signed-off-by: Liangliang Lu <luliang@codeaurora.org>
> Signed-off-by: Sayali Lokhande <sayalil@codeaurora.org>
> Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
> Signed-off-by: Ram Prakash Gupta <quic_rampraka@quicinc.com>
> ---
>  drivers/mmc/host/cqhci-core.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/mmc/host/cqhci-core.c 
> b/drivers/mmc/host/cqhci-core.c index b0d30c3..2908d30 100644
> --- a/drivers/mmc/host/cqhci-core.c
> +++ b/drivers/mmc/host/cqhci-core.c
> @@ -822,8 +822,15 @@ irqreturn_t cqhci_irq(struct mmc_host *mmc, u32 intmask, int cmd_error,
>  	pr_debug("%s: cqhci: IRQ status: 0x%08x\n", mmc_hostname(mmc), 
> status);
>  
>  	if ((status & (CQHCI_IS_RED | CQHCI_IS_GCE | CQHCI_IS_ICCE)) ||
> -	    cmd_error || data_error)
> +	    cmd_error || data_error) {
> +		if ((status & CQHCI_IS_RED) && mmc->err_stats_enabled)
> +			mmc_debugfs_err_stats_inc(mmc, MMC_ERR_CMDQ_RED);
> +		if ((status & CQHCI_IS_GCE) && (mmc->err_stats_enabled))
> +			mmc_debugfs_err_stats_inc(mmc, MMC_ERR_CMDQ_GCE);
> +		if ((status & CQHCI_IS_ICCE) && mmc->err_stats_enabled)
> +			mmc_debugfs_err_stats_inc(mmc, MMC_ERR_CMDQ_ICCE);

Please don't check mmc->err_stats_enabled

>>>>>>>Sure.

>  		cqhci_error_irq(mmc, status, cmd_error, data_error);
> +	}
>  
>  	if (status & CQHCI_IS_TCC) {
>  		/* read TCN and complete the request */
>