Re: [PATCH 4/5] scsi: ufs: Log error handler activity

[Adrian Hunter <adrian.hunter@intel.com>]

On 13/10/2021 00:54, Bart Van Assche wrote:
> Kernel logs are hard to comprehend without information about what the
> UFS error handler is doing. Hence this patch that logs information
> about error handler activity.
> 
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
>  drivers/scsi/ufs/ufshcd.c | 24 ++++++++++++++++++++++++
>  1 file changed, 24 insertions(+)
> 
> diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
> index cb55ba3cb3e6..ecfe1f124f8a 100644
> --- a/drivers/scsi/ufs/ufshcd.c
> +++ b/drivers/scsi/ufs/ufshcd.c
> @@ -134,6 +134,14 @@ enum {
>  	UFSHCD_CAN_QUEUE	= 32,
>  };
>  
> +static const char *const ufshcd_state_name[] = {
> +	[UFSHCD_STATE_RESET]			= "reset",
> +	[UFSHCD_STATE_OPERATIONAL]		= "operational",
> +	[UFSHCD_STATE_ERROR]			= "error",
> +	[UFSHCD_STATE_EH_SCHEDULED_FATAL]	= "eh_will_reset",
> +	[UFSHCD_STATE_EH_SCHEDULED_NON_FATAL]	= "eh_wont_reset",

Currently, the error handler can do a reset for 
UFSHCD_STATE_EH_SCHEDULED_NON_FATAL, so that description is
misleading.

There is code like:

	if (hba->force_reset || ufshcd_is_link_broken(hba) ||
	    ufshcd_is_saved_err_fatal(hba) ||
	    ((hba->saved_err & UIC_ERROR) &&
	     (hba->saved_uic_err & (UFSHCD_UIC_DL_NAC_RECEIVED_ERROR |
				    UFSHCD_UIC_DL_TCx_REPLAY_ERROR)))) {
		needs_reset = true;
		goto do_reset;
	}

where UFSHCD_UIC_DL_NAC_RECEIVED_ERROR and UFSHCD_UIC_DL_TCx_REPLAY_ERROR
are non-fatal errors.  I think the spec. says they should not need a reset
but the driver does anyway.

> +};
> +
>  /* UFSHCD error handling flags */
>  enum {
>  	UFSHCD_EH_IN_PROGRESS = (1 << 0),
> @@ -6065,6 +6073,13 @@ static void ufshcd_err_handler(struct Scsi_Host *host)
>  	int pmc_err;
>  	int tag;
>  
> +	dev_info(hba->dev,
> +		 "%s started; HBA state %s; powered %d; shutting down %d; saved_err = %d; saved_uic_err = %d; force_reset = %d%s\n",
> +		 __func__, ufshcd_state_name[hba->ufshcd_state],
> +		 hba->is_powered, hba->shutting_down, hba->saved_err,
> +		 hba->saved_uic_err, hba->force_reset,
> +		 ufshcd_is_link_broken(hba) ? "; link is broken" : "");
> +
>  	down(&hba->host_sem);
>  	spin_lock_irqsave(hba->host->host_lock, flags);
>  	hba->host->host_eh_scheduled = 0;
> @@ -6160,6 +6175,8 @@ static void ufshcd_err_handler(struct Scsi_Host *host)
>  			err_xfer = true;
>  			goto lock_skip_pending_xfer_clear;
>  		}
> +		dev_err(hba->dev, "Aborted tag %d / CDB %#02x\n", tag,
> +			hba->lrb[tag].cmd ? hba->lrb[tag].cmd->cmnd[0] : -1);
>  	}
>  
>  	/* Clear pending task management requests */
> @@ -6240,6 +6257,9 @@ static void ufshcd_err_handler(struct Scsi_Host *host)
>  	spin_unlock_irqrestore(hba->host->host_lock, flags);
>  	ufshcd_err_handling_unprepare(hba);
>  	up(&hba->host_sem);
> +
> +	dev_info(hba->dev, "%s finished; HBA state %s\n", __func__,
> +		 ufshcd_state_name[hba->ufshcd_state]);
>  }
>  
>  /**
> @@ -6554,6 +6574,10 @@ static int ufshcd_clear_tm_cmd(struct ufs_hba *hba, int tag)
>  	err = ufshcd_wait_for_register(hba,
>  			REG_UTP_TASK_REQ_DOOR_BELL,
>  			mask, 0, 1000, 1000);
> +
> +	dev_err(hba->dev, "Clearing task management function with tag %d %s\n",
> +		tag, err ? "succeeded" : "failed");
> +
>  out:
>  	return err;
>  }
>