[PATCH v2 1/4] scsi: ufs: Recheck bkops level if bkops is disabled

[PATCH v2 1/4] scsi: ufs: Recheck bkops level if bkops is disabled

[Can Guo]

From: Asutosh Das <asutoshd@codeaurora.org>

Bkops level should be rechecked upon receiving an exception.
Currently the bkops level is being cached and never updated.

Update the same each time the level is checked.
Also do not use the cached bkops level value if it is disabled
and then enabled.

Fixes: afdfff59a0e0 (scsi: ufs: handle non spec compliant bkops behaviour by device)
Signed-off-by: Asutosh Das <asutoshd@codeaurora.org>
Signed-off-by: Can Guo <cang@codeaurora.org>
Reviewed-by: Bean Huo <beanhuo@micron.com>
---
 drivers/scsi/ufs/ufshcd.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 3910c58..8e7c362 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -5099,6 +5099,7 @@ static int ufshcd_disable_auto_bkops(struct ufs_hba *hba)
 
 	hba->auto_bkops_enabled = false;
 	trace_ufshcd_auto_bkops_state(dev_name(hba->dev), "Disabled");
+	hba->is_urgent_bkops_lvl_checked = false;
 out:
 	return err;
 }
@@ -5123,6 +5124,7 @@ static void ufshcd_force_reset_auto_bkops(struct ufs_hba *hba)
 		hba->ee_ctrl_mask &= ~MASK_EE_URGENT_BKOPS;
 		ufshcd_disable_auto_bkops(hba);
 	}
+	hba->is_urgent_bkops_lvl_checked = false;
 }
 
 static inline int ufshcd_get_bkops_status(struct ufs_hba *hba, u32 *status)
@@ -5169,6 +5171,7 @@ static int ufshcd_bkops_ctrl(struct ufs_hba *hba,
 		err = ufshcd_enable_auto_bkops(hba);
 	else
 		err = ufshcd_disable_auto_bkops(hba);
+	hba->urgent_bkops_lvl = curr_status;
 out:
 	return err;
 }
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

[PATCH v2 2/4] scsi: ufs: Update VCCQ2 and VCCQ min/max voltage hard codes

[Can Guo]

Per UFS 3.0 JEDEC standard, the VCCQ2 min voltage is 1.7v and the VCCQ
voltage range is 1.14v ~ 1.26v. Update their hard codes accordingly to
make sure they work in a safe range compliant for ver 1.0/2.0/2.1/3.0
UFS devices.

Signed-off-by: Can Guo <cang@codeaurora.org>
---
 drivers/scsi/ufs/ufs.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/ufs/ufs.h b/drivers/scsi/ufs/ufs.h
index 385bac8..9df4f4d 100644
--- a/drivers/scsi/ufs/ufs.h
+++ b/drivers/scsi/ufs/ufs.h
@@ -500,9 +500,9 @@ struct ufs_query_res {
 #define UFS_VREG_VCC_MAX_UV	   3600000 /* uV */
 #define UFS_VREG_VCC_1P8_MIN_UV    1700000 /* uV */
 #define UFS_VREG_VCC_1P8_MAX_UV    1950000 /* uV */
-#define UFS_VREG_VCCQ_MIN_UV	   1100000 /* uV */
+#define UFS_VREG_VCCQ_MIN_UV	   1140000 /* uV */
 #define UFS_VREG_VCCQ_MAX_UV	   1300000 /* uV */
-#define UFS_VREG_VCCQ2_MIN_UV	   1650000 /* uV */
+#define UFS_VREG_VCCQ2_MIN_UV	   1700000 /* uV */
 #define UFS_VREG_VCCQ2_MAX_UV	   1950000 /* uV */
 
 /*
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

Re: [PATCH v2 2/4] scsi: ufs: Update VCCQ2 and VCCQ min/max voltage hard codes

[Alim Akhtar]

I agree with Avri, you can just drop this patch, unless you see
anything affected because of not setting VDDQ/VDDQ2 as per spec.
or alternatively you can have  different marco of each spec and handle
the same in the code as per ufs specification version.

On Mon, Nov 18, 2019 at 9:25 AM Can Guo <cang@codeaurora.org> wrote:
>
> Per UFS 3.0 JEDEC standard, the VCCQ2 min voltage is 1.7v and the VCCQ
> voltage range is 1.14v ~ 1.26v. Update their hard codes accordingly to
> make sure they work in a safe range compliant for ver 1.0/2.0/2.1/3.0
> UFS devices.
>
> Signed-off-by: Can Guo <cang@codeaurora.org>
> ---
>  drivers/scsi/ufs/ufs.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/scsi/ufs/ufs.h b/drivers/scsi/ufs/ufs.h
> index 385bac8..9df4f4d 100644
> --- a/drivers/scsi/ufs/ufs.h
> +++ b/drivers/scsi/ufs/ufs.h
> @@ -500,9 +500,9 @@ struct ufs_query_res {
>  #define UFS_VREG_VCC_MAX_UV       3600000 /* uV */
>  #define UFS_VREG_VCC_1P8_MIN_UV    1700000 /* uV */
>  #define UFS_VREG_VCC_1P8_MAX_UV    1950000 /* uV */
> -#define UFS_VREG_VCCQ_MIN_UV      1100000 /* uV */
> +#define UFS_VREG_VCCQ_MIN_UV      1140000 /* uV */
>  #define UFS_VREG_VCCQ_MAX_UV      1300000 /* uV */
> -#define UFS_VREG_VCCQ2_MIN_UV     1650000 /* uV */
> +#define UFS_VREG_VCCQ2_MIN_UV     1700000 /* uV */
>  #define UFS_VREG_VCCQ2_MAX_UV     1950000 /* uV */
>
>  /*
> --
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
> a Linux Foundation Collaborative Project
>


-- 
Regards,
Alim

[PATCH v2 3/4] scsi: ufs: Avoid messing up the compl_time_stamp of lrbs

[Can Guo]

To be on the safe side, do not touch one lrb after clear its slot in the
lrb_in_use bitmap to avoid messing up the next task which would possibly
occupy this lrb.

Signed-off-by: Can Guo <cang@codeaurora.org>
---
 drivers/scsi/ufs/ufshcd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 8e7c362..5950a7c 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -4902,12 +4902,14 @@ static void __ufshcd_transfer_req_compl(struct ufs_hba *hba,
 			cmd->result = result;
 			/* Mark completed command as NULL in LRB */
 			lrbp->cmd = NULL;
+			lrbp->compl_time_stamp = ktime_get();
 			clear_bit_unlock(index, &hba->lrb_in_use);
 			/* Do not touch lrbp after scsi done */
 			cmd->scsi_done(cmd);
 			__ufshcd_release(hba);
 		} else if (lrbp->command_type == UTP_CMD_TYPE_DEV_MANAGE ||
 			lrbp->command_type == UTP_CMD_TYPE_UFS_STORAGE) {
+			lrbp->compl_time_stamp = ktime_get();
 			if (hba->dev_cmd.complete) {
 				ufshcd_add_command_trace(hba, index,
 						"dev_complete");
@@ -4916,8 +4918,6 @@ static void __ufshcd_transfer_req_compl(struct ufs_hba *hba,
 		}
 		if (ufshcd_is_clkscaling_supported(hba))
 			hba->clk_scaling.active_reqs--;
-
-		lrbp->compl_time_stamp = ktime_get();
 	}
 
 	/* clear corresponding bits of completed commands */
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

Re: [PATCH v2 3/4] scsi: ufs: Avoid messing up the compl_time_stamp of lrbs

[Alim Akhtar]

On Mon, Nov 18, 2019 at 9:27 AM Can Guo <cang@codeaurora.org> wrote:
>
> To be on the safe side, do not touch one lrb after clear its slot in the
> lrb_in_use bitmap to avoid messing up the next task which would possibly
> occupy this lrb.
>
> Signed-off-by: Can Guo <cang@codeaurora.org>
> ---
Tested-by: Alim Akhtar <alim.akhtar@samsung.com>

>  drivers/scsi/ufs/ufshcd.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
> index 8e7c362..5950a7c 100644
> --- a/drivers/scsi/ufs/ufshcd.c
> +++ b/drivers/scsi/ufs/ufshcd.c
> @@ -4902,12 +4902,14 @@ static void __ufshcd_transfer_req_compl(struct ufs_hba *hba,
>                         cmd->result = result;
>                         /* Mark completed command as NULL in LRB */
>                         lrbp->cmd = NULL;
> +                       lrbp->compl_time_stamp = ktime_get();
>                         clear_bit_unlock(index, &hba->lrb_in_use);
>                         /* Do not touch lrbp after scsi done */
>                         cmd->scsi_done(cmd);
>                         __ufshcd_release(hba);
>                 } else if (lrbp->command_type == UTP_CMD_TYPE_DEV_MANAGE ||
>                         lrbp->command_type == UTP_CMD_TYPE_UFS_STORAGE) {
> +                       lrbp->compl_time_stamp = ktime_get();
>                         if (hba->dev_cmd.complete) {
>                                 ufshcd_add_command_trace(hba, index,
>                                                 "dev_complete");
> @@ -4916,8 +4918,6 @@ static void __ufshcd_transfer_req_compl(struct ufs_hba *hba,
>                 }
>                 if (ufshcd_is_clkscaling_supported(hba))
>                         hba->clk_scaling.active_reqs--;
> -
> -               lrbp->compl_time_stamp = ktime_get();
>         }
>
>         /* clear corresponding bits of completed commands */
> --
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
> a Linux Foundation Collaborative Project
>


-- 
Regards,
Alim

[PATCH v2 4/4] scsi: ufs: Complete pending requests in host reset and restore path

[Can Guo]

In UFS host reset and restore path, before probe, we stop and start the
host controller once. After host controller is stopped, the pending
requests, if any, are cleared from the doorbell, but no completion IRQ
would be raised due to the hba is stopped.
These pending requests shall be completed along with the first NOP_OUT
command(as it is the first command which can raise a transfer completion
IRQ) sent during probe.
Since the OCSs of these pending requests are not SUCCESS(because they are
not yet literally finished), their UPIUs shall be dumped. When there are
multiple pending requests, the UPIU dump can be overwhelming and may lead
to stability issues because it is in atomic context.
Therefore, before probe, complete these pending requests right after host
controller is stopped and silence the UPIU dump from them.

Signed-off-by: Can Guo <cang@codeaurora.org>
---
 drivers/scsi/ufs/ufshcd.c | 24 ++++++++++--------------
 drivers/scsi/ufs/ufshcd.h |  2 ++
 2 files changed, 12 insertions(+), 14 deletions(-)

diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 5950a7c..b92a3f4 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -4845,7 +4845,7 @@ static void ufshcd_slave_destroy(struct scsi_device *sdev)
 		break;
 	} /* end of switch */
 
-	if (host_byte(result) != DID_OK)
+	if ((host_byte(result) != DID_OK) && !hba->silence_err_logs)
 		ufshcd_print_trs(hba, 1 << lrbp->task_tag, true);
 	return result;
 }
@@ -5404,8 +5404,8 @@ static void ufshcd_err_handler(struct work_struct *work)
 
 	/*
 	 * if host reset is required then skip clearing the pending
-	 * transfers forcefully because they will automatically get
-	 * cleared after link startup.
+	 * transfers forcefully because they will get cleared during
+	 * host reset and restore
 	 */
 	if (needs_reset)
 		goto skip_pending_xfer_clear;
@@ -6333,9 +6333,15 @@ static int ufshcd_host_reset_and_restore(struct ufs_hba *hba)
 	int err;
 	unsigned long flags;
 
-	/* Reset the host controller */
+	/*
+	 * Stop the host controller and complete the requests
+	 * cleared by h/w
+	 */
 	spin_lock_irqsave(hba->host->host_lock, flags);
 	ufshcd_hba_stop(hba, false);
+	hba->silence_err_logs = true;
+	ufshcd_complete_requests(hba);
+	hba->silence_err_logs = false;
 	spin_unlock_irqrestore(hba->host->host_lock, flags);
 
 	/* scale up clocks to max frequency before full reinitialization */
@@ -6369,7 +6375,6 @@ static int ufshcd_host_reset_and_restore(struct ufs_hba *hba)
 static int ufshcd_reset_and_restore(struct ufs_hba *hba)
 {
 	int err = 0;
-	unsigned long flags;
 	int retries = MAX_HOST_RESET_RETRIES;
 
 	do {
@@ -6379,15 +6384,6 @@ static int ufshcd_reset_and_restore(struct ufs_hba *hba)
 		err = ufshcd_host_reset_and_restore(hba);
 	} while (err && --retries);
 
-	/*
-	 * After reset the door-bell might be cleared, complete
-	 * outstanding requests in s/w here.
-	 */
-	spin_lock_irqsave(hba->host->host_lock, flags);
-	ufshcd_transfer_req_compl(hba);
-	ufshcd_tmc_handler(hba);
-	spin_unlock_irqrestore(hba->host->host_lock, flags);
-
 	return err;
 }
 
diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h
index e0fe247..1e51034 100644
--- a/drivers/scsi/ufs/ufshcd.h
+++ b/drivers/scsi/ufs/ufshcd.h
@@ -513,6 +513,7 @@ struct ufs_stats {
  * @uic_error: UFS interconnect layer error status
  * @saved_err: sticky error mask
  * @saved_uic_err: sticky UIC error mask
+ * @silence_err_logs: flag to silence error logs
  * @dev_cmd: ufs device management command information
  * @last_dme_cmd_tstamp: time stamp of the last completed DME command
  * @auto_bkops_enabled: to track whether bkops is enabled in device
@@ -670,6 +671,7 @@ struct ufs_hba {
 	u32 saved_err;
 	u32 saved_uic_err;
 	struct ufs_stats ufs_stats;
+	bool silence_err_logs;
 
 	/* Device management request data */
 	struct ufs_dev_cmd dev_cmd;
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

RE: [EXT] [PATCH v2 4/4] scsi: ufs: Complete pending requests in host reset and restore path

[Bean Huo (beanhuo)]

> Therefore, before probe, complete these pending requests right after host
> controller is stopped and silence the UPIU dump from them.
> 
> Signed-off-by: Can Guo <cang@codeaurora.org>
Reviewed-by: Bean Huo <beanhuo@micron.com>
Tested-by: Bean Huo <beanhuo@micon.com>

[PATCH v2 3/4] scsi: ufs: Avoid messing up the compl_time_stamp of lrbs

[Can Guo]

From: Can Guo <cang@codeaurora.org>

To be on the safe side, do not touch one lrb after clear its slot in the
lrb_in_use bitmap to avoid messing up the next task which would possibly
occupy this lrb.

Signed-off-by: Can Guo <cang@codeaurora.org>
---
 drivers/scsi/ufs/ufshcd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 8e7c362..5950a7c 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -4902,12 +4902,14 @@ static void __ufshcd_transfer_req_compl(struct ufs_hba *hba,
 			cmd->result = result;
 			/* Mark completed command as NULL in LRB */
 			lrbp->cmd = NULL;
+			lrbp->compl_time_stamp = ktime_get();
 			clear_bit_unlock(index, &hba->lrb_in_use);
 			/* Do not touch lrbp after scsi done */
 			cmd->scsi_done(cmd);
 			__ufshcd_release(hba);
 		} else if (lrbp->command_type == UTP_CMD_TYPE_DEV_MANAGE ||
 			lrbp->command_type == UTP_CMD_TYPE_UFS_STORAGE) {
+			lrbp->compl_time_stamp = ktime_get();
 			if (hba->dev_cmd.complete) {
 				ufshcd_add_command_trace(hba, index,
 						"dev_complete");
@@ -4916,8 +4918,6 @@ static void __ufshcd_transfer_req_compl(struct ufs_hba *hba,
 		}
 		if (ufshcd_is_clkscaling_supported(hba))
 			hba->clk_scaling.active_reqs--;
-
-		lrbp->compl_time_stamp = ktime_get();
 	}
 
 	/* clear corresponding bits of completed commands */
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

RE: [PATCH v2 3/4] scsi: ufs: Avoid messing up the compl_time_stamp of lrbs

[Avri Altman]

 
> 
> From: Can Guo <cang@codeaurora.org>
> 
> To be on the safe side, do not touch one lrb after clear its slot in the lrb_in_use
> bitmap to avoid messing up the next task which would possibly occupy this lrb.
> 
> Signed-off-by: Can Guo <cang@codeaurora.org>
Reviewed by: Avri Altman <avri.altman@wdc.com>