Skip to contentSkip to conversation

scsi: ufs: Cleanup and refactor clock scaling

PATCH Applied Stanley Chu 2 participants 6 messages Jan 20, 2021 – Jan 23, 2021

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

v3 cover scsi: ufs: Cleanup and refactor clock scaling

Hi,
This series cleans up and refactors clk-scaling feature, and shall not change any functionality.

This series is based on Can's series "Three changes related with UFS clock scaling" in 5.12/scsi-queue branch in Martin's tree.

Changes since v2:
- Remove patch 4
- Rebase to Can's series: [v11] Three changes related with UFS clock scaling

Changes since v1:
- Refactor ufshcd_clk_scaling_suspend() in patch [3/4]
- Change function name from ufshcd_clk_scaling_pm() to ufshcd_clk_scaling_suspend() in patch [3/4]
- Refine patch titles

Stanley Chu (3):
scsi: ufs: Refactor cancelling clkscaling works
scsi: ufs: Remove redundant null checking of devfreq instance
scsi: ufs: Cleanup and refactor clk-scaling feature

 drivers/scsi/ufs/ufshcd.c | 87 ++++++++++++++++++---------------------
 1 file changed, 41 insertions(+), 46 deletions(-)
Signature · 5 lines
2.18.0
_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek
#1 lore raw

PATCH v3 1/3 scsi: ufs: Refactor cancelling clkscaling works

Cancelling suspend_work and resume_work is only required while suspending clk-scaling. Thus moving these two invokes into ufshcd_suspend_clkscaling() function.

---
 drivers/scsi/ufs/ufshcd.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)
drivers/scsi/ufs/ufshcd.c +6 -11
index 5450e049d1a9..226504d05c14 100644
@@ -1503,6 +1503,9 @@ static void ufshcd_suspend_clkscaling(struct ufs_hba *hba)
 	if (!ufshcd_is_clkscaling_supported(hba))
 		return;
 
+	cancel_work_sync(&hba->clk_scaling.suspend_work);
+	cancel_work_sync(&hba->clk_scaling.resume_work);
+
 	spin_lock_irqsave(hba->host->host_lock, flags);
 	if (!hba->clk_scaling.is_suspended) {
 		suspend = true;
@@ -1564,9 +1567,6 @@ static ssize_t ufshcd_clkscale_enable_store(struct device *dev,
 	pm_runtime_get_sync(hba->dev);
 	ufshcd_hold(hba, false);
 
-	cancel_work_sync(&hba->clk_scaling.suspend_work);
-	cancel_work_sync(&hba->clk_scaling.resume_work);
-
 	hba->clk_scaling.is_enabled = value;
 
 	if (value) {
@@ -5783,11 +5783,8 @@ static void ufshcd_err_handling_prepare(struct ufs_hba *hba)
 		ufshcd_vops_resume(hba, pm_op);
 	} else {
 		ufshcd_hold(hba, false);
-		if (hba->clk_scaling.is_enabled) {
-			cancel_work_sync(&hba->clk_scaling.suspend_work);
-			cancel_work_sync(&hba->clk_scaling.resume_work);
+		if (hba->clk_scaling.is_enabled)
 			ufshcd_suspend_clkscaling(hba);
-		}
 		down_write(&hba->clk_scaling_lock);
 		hba->clk_scaling.is_allowed = false;
 		up_write(&hba->clk_scaling_lock);
@@ -8707,11 +8704,9 @@ static int ufshcd_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op)
 	ufshcd_hold(hba, false);
 	hba->clk_gating.is_suspended = true;
 
-	if (hba->clk_scaling.is_enabled) {
-		cancel_work_sync(&hba->clk_scaling.suspend_work);
-		cancel_work_sync(&hba->clk_scaling.resume_work);
+	if (hba->clk_scaling.is_enabled)
 		ufshcd_suspend_clkscaling(hba);
-	}
+
 	down_write(&hba->clk_scaling_lock);
 	hba->clk_scaling.is_allowed = false;
 	up_write(&hba->clk_scaling_lock);
Signature · 5 lines
2.18.0
_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek
#2 lore raw

PATCH v3 2/3 scsi: ufs: Remove redundant null checking of devfreq instance

In reply to Stanley Chu

hba->devfreq is zero-initialized thus it is not required to check its existence in ufshcd_add_lus() function which is invoked during initialization only.

---
 drivers/scsi/ufs/ufshcd.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)
drivers/scsi/ufs/ufshcd.c +7 -8
index 226504d05c14..f707234580ea 100644
@@ -7777,15 +7777,14 @@ static int ufshcd_add_lus(struct ufs_hba *hba)
 			&hba->pwr_info,
 			sizeof(struct ufs_pa_layer_attr));
 		hba->clk_scaling.saved_pwr_info.is_valid = true;
-		if (!hba->devfreq) {
-			hba->clk_scaling.is_allowed = true;
-			ret = ufshcd_devfreq_init(hba);
-			if (ret)
-				goto out;
+		hba->clk_scaling.is_allowed = true;
 
-			hba->clk_scaling.is_enabled = true;
-			ufshcd_init_clk_scaling_sysfs(hba);
-		}
+		ret = ufshcd_devfreq_init(hba);
+		if (ret)
+			goto out;
+
+		hba->clk_scaling.is_enabled = true;
+		ufshcd_init_clk_scaling_sysfs(hba);
 	}
 
 	ufs_bsg_probe(hba);
Signature · 5 lines
2.18.0
_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek
#3 lore raw

PATCH v3 3/3 scsi: ufs: Cleanup and refactor clk-scaling feature

In reply to Stanley Chu

Manipulate clock scaling related stuff only if the host capability supports clock scaling feature to avoid redundant code execution.

---
 drivers/scsi/ufs/ufshcd.c | 63 ++++++++++++++++++++-------------------
 1 file changed, 32 insertions(+), 31 deletions(-)
drivers/scsi/ufs/ufshcd.c +32 -31
index f707234580ea..edfac8752d33 100644
@@ -1500,9 +1500,6 @@ static void ufshcd_suspend_clkscaling(struct ufs_hba *hba)
 	unsigned long flags;
 	bool suspend = false;
 
-	if (!ufshcd_is_clkscaling_supported(hba))
-		return;
-
 	cancel_work_sync(&hba->clk_scaling.suspend_work);
 	cancel_work_sync(&hba->clk_scaling.resume_work);
 
@@ -1522,9 +1519,6 @@ static void ufshcd_resume_clkscaling(struct ufs_hba *hba)
 	unsigned long flags;
 	bool resume = false;
 
-	if (!ufshcd_is_clkscaling_supported(hba))
-		return;
-
 	spin_lock_irqsave(hba->host->host_lock, flags);
 	if (hba->clk_scaling.is_suspended) {
 		resume = true;
@@ -5759,6 +5753,26 @@ static inline void ufshcd_schedule_eh_work(struct ufs_hba *hba)
 	}
 }
 
+static void ufshcd_clk_scaling_allow(struct ufs_hba *hba, bool allow)
+{
+	down_write(&hba->clk_scaling_lock);
+	hba->clk_scaling.is_allowed = allow;
+	up_write(&hba->clk_scaling_lock);
+}
+
+static void ufshcd_clk_scaling_suspend(struct ufs_hba *hba, bool suspend)
+{
+	if (suspend) {
+		if (hba->clk_scaling.is_enabled)
+			ufshcd_suspend_clkscaling(hba);
+		ufshcd_clk_scaling_allow(hba, false);
+	} else {
+		ufshcd_clk_scaling_allow(hba, true);
+		if (hba->clk_scaling.is_enabled)
+			ufshcd_resume_clkscaling(hba);
+	}
+}
+
 static void ufshcd_err_handling_prepare(struct ufs_hba *hba)
 {
 	pm_runtime_get_sync(hba->dev);
@@ -5783,22 +5797,18 @@ static void ufshcd_err_handling_prepare(struct ufs_hba *hba)
 		ufshcd_vops_resume(hba, pm_op);
 	} else {
 		ufshcd_hold(hba, false);
-		if (hba->clk_scaling.is_enabled)
+		if (ufshcd_is_clkscaling_supported(hba) &&
+		    hba->clk_scaling.is_enabled)
 			ufshcd_suspend_clkscaling(hba);
-		down_write(&hba->clk_scaling_lock);
-		hba->clk_scaling.is_allowed = false;
-		up_write(&hba->clk_scaling_lock);
+		ufshcd_clk_scaling_allow(hba, false);
 	}
 }
 
 static void ufshcd_err_handling_unprepare(struct ufs_hba *hba)
 {
 	ufshcd_release(hba);
-	down_write(&hba->clk_scaling_lock);
-	hba->clk_scaling.is_allowed = true;
-	up_write(&hba->clk_scaling_lock);
-	if (hba->clk_scaling.is_enabled)
-		ufshcd_resume_clkscaling(hba);
+	if (ufshcd_is_clkscaling_supported(hba))
+		ufshcd_clk_scaling_suspend(hba, false);
 	pm_runtime_put(hba->dev);
 }
 
@@ -8703,12 +8713,8 @@ static int ufshcd_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op)
 	ufshcd_hold(hba, false);
 	hba->clk_gating.is_suspended = true;
 
-	if (hba->clk_scaling.is_enabled)
-		ufshcd_suspend_clkscaling(hba);
-
-	down_write(&hba->clk_scaling_lock);
-	hba->clk_scaling.is_allowed = false;
-	up_write(&hba->clk_scaling_lock);
+	if (ufshcd_is_clkscaling_supported(hba))
+		ufshcd_clk_scaling_suspend(hba, true);
 
 	if (req_dev_pwr_mode == UFS_ACTIVE_PWR_MODE &&
 			req_link_state == UIC_LINK_ACTIVE_STATE) {
@@ -8828,11 +8834,9 @@ static int ufshcd_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op)
 	if (!ufshcd_set_dev_pwr_mode(hba, UFS_ACTIVE_PWR_MODE))
 		ufshcd_disable_auto_bkops(hba);
 enable_gating:
-	down_write(&hba->clk_scaling_lock);
-	hba->clk_scaling.is_allowed = true;
-	up_write(&hba->clk_scaling_lock);
-	if (hba->clk_scaling.is_enabled)
-		ufshcd_resume_clkscaling(hba);
+	if (ufshcd_is_clkscaling_supported(hba))
+		ufshcd_clk_scaling_suspend(hba, false);
+
 	hba->clk_gating.is_suspended = false;
 	hba->dev_info.b_rpm_dev_flush_capable = false;
 	ufshcd_release(hba);
@@ -8934,11 +8938,8 @@ static int ufshcd_resume(struct ufs_hba *hba, enum ufs_pm_op pm_op)
 
 	hba->clk_gating.is_suspended = false;
 
-	down_write(&hba->clk_scaling_lock);
-	hba->clk_scaling.is_allowed = true;
-	up_write(&hba->clk_scaling_lock);
-	if (hba->clk_scaling.is_enabled)
-		ufshcd_resume_clkscaling(hba);
+	if (ufshcd_is_clkscaling_supported(hba))
+		ufshcd_clk_scaling_suspend(hba, false);
 
 	/* Enable Auto-Hibernate if configured */
 	ufshcd_auto_hibern8_enable(hba);
Signature · 5 lines
2.18.0
_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek
#4 lore raw
In reply to Stanley Chu

Stanley,

Quoted 2 lines

This series cleans up and refactors clk-scaling feature, and shall not change any functionality.

Applied to 5.12/scsi-staging, thanks!

Signature · 6 lines
Martin K. Petersen	Oracle Linux Engineering

_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek
#5 lore raw
In reply to Stanley Chu
Quoted 9 lines · Stanley Chu

This series cleans up and refactors clk-scaling feature, and shall not change any functionality.

This series is based on Can's series "Three changes related with UFS clock scaling" in 5.12/scsi-queue branch in Martin's tree.

Changes since v2:
- Remove patch 4
- Rebase to Can's series: [v11] Three changes related with UFS clock scaling

[...]

Applied to 5.12/scsi-queue, thanks!

[1/3] scsi: ufs: Refactor cancelling clkscaling works

      https://git.kernel.org/mkp/scsi/c/f9a7fa345aec
[2/3] scsi: ufs: Remove redundant null checking of devfreq instance
      https://git.kernel.org/mkp/scsi/c/b058fa868234
[3/3] scsi: ufs: Cleanup and refactor clk-scaling feature
      https://git.kernel.org/mkp/scsi/c/348e1bc5f4b7
Signature · 6 lines
Martin K. Petersen	Oracle Linux Engineering

_______________________________________________
Linux-mediatek mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-mediatek