Skip to contentSkip to conversation

staging: most: dim2: fix error handling in fsl_mx6_enable

PATCH Superseded Artem Lytkin 3 participants 13 messages Feb 16 – Feb 24

Subsystems (2)

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

cover staging: most: dim2: fix error handling in fsl_mx6_enable

Fix error handling issues in fsl_mx6_enable() in the DIM2 driver.

Patch 1 fixes a real bug: the return value of clk_prepare_enable() for the PLL clock is not checked, while the same call for the MLB clock is properly checked earlier in the function.

Patch 2 addresses review feedback from Dan Carpenter on an earlier attempt to fix this function [1]: replace IS_ERR_OR_NULL() with IS_ERR(), use dev_err_probe() to propagate actual error codes and handle probe deferral correctly.

[1] https://lore.kernel.org/linux-staging/ZLUnW2CrhFMBCCa1@moroto/

Artem Lytkin (2):
staging: most: dim2: check return value of clk_prepare_enable for PLL
staging: most: dim2: fix error handling in fsl_mx6_enable

 drivers/staging/most/dim2/dim2.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)
Signature · 1 line
2.43.0
#1 lore raw

PATCH 1/2 staging: most: dim2: check return value of clk_prepare_enable for PLL

The return value of clk_prepare_enable() for the PLL clock is not checked, while the same call for the MLB clock is properly checked earlier in the function. If clk_prepare_enable() fails, the driver continues without the PLL clock enabled, leading to undefined hardware behavior.

Add the missing error check and disable the MLB clock on failure to keep the cleanup consistent with the rest of the function.

---
 drivers/staging/most/dim2/dim2.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/staging/most/dim2/dim2.c +6 -1
index dad2abe6c..1f07c37df 100644
@@ -944,7 +944,12 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 		}
 
 		writel(0x888, dev->io_base + 0x38);
-		clk_prepare_enable(dev->clk_pll);
+		ret = clk_prepare_enable(dev->clk_pll);
+		if (ret) {
+			dev_err(&pdev->dev, "failed to enable pll clock\n");
+			clk_disable_unprepare(dev->clk);
+			return ret;
+		}
 	}
 
 	return 0;
Signature · 1 line
2.43.0
#2 lore raw

PATCH 2/2 staging: most: dim2: fix error handling in fsl_mx6_enable

In reply to Artem Lytkin

Address several error handling issues in fsl_mx6_enable():

- Replace IS_ERR_OR_NULL() with IS_ERR() for devm_clk_get() return value checks, since devm_clk_get() never returns NULL on error.

- Replace hardcoded -EFAULT returns with dev_err_probe() which both propagates the actual error code from devm_clk_get() and suppresses log messages during probe deferral.

- Remove an unnecessary string indirection in the clk_prepare_enable error message.

---
 drivers/staging/most/dim2/dim2.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)
drivers/staging/most/dim2/dim2.c +7 -8
index 1f07c37df..15eaf4ff8 100644
@@ -923,24 +923,23 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 	int ret;
 
 	dev->clk = devm_clk_get(&pdev->dev, "mlb");
-	if (IS_ERR_OR_NULL(dev->clk)) {
-		dev_err(&pdev->dev, "unable to get mlb clock\n");
-		return -EFAULT;
-	}
+	if (IS_ERR(dev->clk))
+		return dev_err_probe(&pdev->dev, PTR_ERR(dev->clk),
+				     "unable to get mlb clock\n");
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
-		dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed");
+		dev_err(&pdev->dev, "clk_prepare_enable failed\n");
 		return ret;
 	}
 
 	if (dev->clk_speed >= CLK_2048FS) {
 		/* enable pll */
 		dev->clk_pll = devm_clk_get(&pdev->dev, "pll8_mlb");
-		if (IS_ERR_OR_NULL(dev->clk_pll)) {
-			dev_err(&pdev->dev, "unable to get mlb pll clock\n");
+		if (IS_ERR(dev->clk_pll)) {
 			clk_disable_unprepare(dev->clk);
-			return -EFAULT;
+			return dev_err_probe(&pdev->dev, PTR_ERR(dev->clk_pll),
+					     "unable to get mlb pll clock\n");
 		}
 
 		writel(0x888, dev->io_base + 0x38);
Signature · 1 line
2.43.0
#3 lore raw
Quoted 5 lines · Artem Lytkin

Address several error handling issues in fsl_mx6_enable():

- Replace IS_ERR_OR_NULL() with IS_ERR() for devm_clk_get() return value checks, since devm_clk_get() never returns NULL on error.

Quoted 7 lines

- Replace hardcoded -EFAULT returns with dev_err_probe() which both propagates the actual error code from devm_clk_get() and suppresses log messages during probe deferral.

- Remove an unnecessary string indirection in the clk_prepare_enable error message.

Some of these are unrelated things.

regards,
dan carpenter

#4 lore raw

v2 cover staging: most: dim2: clean up error handling in fsl_mx6_enable

In reply to Artem Lytkin

This series cleans up error handling in fsl_mx6_enable(), split into separate logical changes per Dan Carpenter's review of v1.

Changes since v1:
- Split combined patch into 3 separate patches (one change per patch)
- Extended string indirection fix to all three enable functions

Patch 1 replaces IS_ERR_OR_NULL() with IS_ERR() since devm_clk_get() never returns NULL.

Patch 2 replaces dev_err() + hardcoded -EFAULT with dev_err_probe() and PTR_ERR() to propagate correct error codes and handle probe deferral properly.

Patch 3 removes unnecessary string indirection in dev_err() calls ("%s\n", "message" -> "message\n").

Artem Lytkin (3):
staging: most: dim2: replace IS_ERR_OR_NULL with IS_ERR for
devm_clk_get
staging: most: dim2: use dev_err_probe and proper error codes for
clock
staging: most: dim2: remove unnecessary string indirection in dev_err

 drivers/staging/most/dim2/dim2.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)
Signature · 1 line
2.43.0
#5 lore raw

PATCH v2 1/3 staging: most: dim2: replace IS_ERR_OR_NULL with IS_ERR for devm_clk_get

devm_clk_get() never returns NULL, so IS_ERR_OR_NULL() checks are unnecessary. Replace them with IS_ERR() for both the "mlb" and "pll8_mlb" clock lookups in fsl_mx6_enable().

---
 drivers/staging/most/dim2/dim2.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/staging/most/dim2/dim2.c +2 -2
index 92c7a7d8fe7e..cea1ba99caf7 100644
@@ -921,7 +921,7 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 	int ret;
 
 	dev->clk = devm_clk_get(&pdev->dev, "mlb");
-	if (IS_ERR_OR_NULL(dev->clk)) {
+	if (IS_ERR(dev->clk)) {
 		dev_err(&pdev->dev, "unable to get mlb clock\n");
 		return -EFAULT;
 	}
@@ -935,7 +935,7 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 	if (dev->clk_speed >= CLK_2048FS) {
 		/* enable pll */
 		dev->clk_pll = devm_clk_get(&pdev->dev, "pll8_mlb");
-		if (IS_ERR_OR_NULL(dev->clk_pll)) {
+		if (IS_ERR(dev->clk_pll)) {
 			dev_err(&pdev->dev, "unable to get mlb pll clock\n");
 			clk_disable_unprepare(dev->clk);
 			return -EFAULT;
Signature · 1 line
2.43.0
#6 lore raw

PATCH v2 2/3 staging: most: dim2: use dev_err_probe and proper error codes for clock

In reply to Artem Lytkin

Replace hardcoded -EFAULT returns with dev_err_probe() and PTR_ERR() when devm_clk_get() fails in fsl_mx6_enable(). This ensures the correct error code is propagated (e.g. -EPROBE_DEFER for deferred probing) and avoids log noise during probe deferral.

---
 drivers/staging/most/dim2/dim2.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/staging/most/dim2/dim2.c +5 -6
index cea1ba99caf7..e4c8b4995c61 100644
@@ -921,10 +921,9 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 	int ret;
 
 	dev->clk = devm_clk_get(&pdev->dev, "mlb");
-	if (IS_ERR(dev->clk)) {
-		dev_err(&pdev->dev, "unable to get mlb clock\n");
-		return -EFAULT;
-	}
+	if (IS_ERR(dev->clk))
+		return dev_err_probe(&pdev->dev, PTR_ERR(dev->clk),
+				     "unable to get mlb clock\n");
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
@@ -936,9 +935,9 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 		/* enable pll */
 		dev->clk_pll = devm_clk_get(&pdev->dev, "pll8_mlb");
 		if (IS_ERR(dev->clk_pll)) {
-			dev_err(&pdev->dev, "unable to get mlb pll clock\n");
 			clk_disable_unprepare(dev->clk);
-			return -EFAULT;
+			return dev_err_probe(&pdev->dev, PTR_ERR(dev->clk_pll),
+					     "unable to get mlb pll clock\n");
 		}
 
 		writel(0x888, dev->io_base + 0x38);
Signature · 1 line
2.43.0
#7 lore raw

PATCH v2 3/3 staging: most: dim2: remove unnecessary string indirection in dev_err

In reply to Artem Lytkin

Replace dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed") with the direct format string dev_err(&pdev->dev, "clk_prepare_enable failed\n"). The extra level of indirection through %s is unnecessary.

---
 drivers/staging/most/dim2/dim2.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/staging/most/dim2/dim2.c +3 -3
index e4c8b4995c61..66617e89e028 100644
@@ -927,7 +927,7 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
-		dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed");
+		dev_err(&pdev->dev, "clk_prepare_enable failed\n");
 		return ret;
 	}
 
@@ -975,7 +975,7 @@ static int rcar_gen2_enable(struct platform_device *pdev)
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
-		dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed");
+		dev_err(&pdev->dev, "clk_prepare_enable failed\n");
 		return ret;
 	}
 
@@ -1020,7 +1020,7 @@ static int rcar_gen3_enable(struct platform_device *pdev)
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
-		dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed");
+		dev_err(&pdev->dev, "clk_prepare_enable failed\n");
 		return ret;
 	}
 
Signature · 1 line
2.43.0
#8 lore raw

v2 cover staging: most: dim2: clean up error handling in fsl_mx6_enable

In reply to Artem Lytkin

This series cleans up error handling in fsl_mx6_enable(), split into separate logical changes per Dan Carpenter's review of v1.

Changes since v1:
- Split combined patch into 3 separate patches (one change per patch)
- Extended string indirection fix to all three enable functions

Patch 1 replaces IS_ERR_OR_NULL() with IS_ERR() since devm_clk_get() never returns NULL.

Patch 2 replaces dev_err() + hardcoded -EFAULT with dev_err_probe() and PTR_ERR() to propagate correct error codes and handle probe deferral properly.

Patch 3 removes unnecessary string indirection in dev_err() calls ("%s\n", "message" -> "message\n").

Artem Lytkin (3):
staging: most: dim2: replace IS_ERR_OR_NULL with IS_ERR for
devm_clk_get
staging: most: dim2: use dev_err_probe and proper error codes for
clock
staging: most: dim2: remove unnecessary string indirection in dev_err

 drivers/staging/most/dim2/dim2.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)
Signature · 1 line
2.43.0
#9 lore raw

PATCH v2 1/3 staging: most: dim2: replace IS_ERR_OR_NULL with IS_ERR for devm_clk_get

devm_clk_get() never returns NULL, so IS_ERR_OR_NULL() checks are unnecessary. Replace them with IS_ERR() for both the "mlb" and "pll8_mlb" clock lookups in fsl_mx6_enable().

---
 drivers/staging/most/dim2/dim2.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/staging/most/dim2/dim2.c +2 -2
index 92c7a7d8fe7e..cea1ba99caf7 100644
@@ -921,7 +921,7 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 	int ret;
 
 	dev->clk = devm_clk_get(&pdev->dev, "mlb");
-	if (IS_ERR_OR_NULL(dev->clk)) {
+	if (IS_ERR(dev->clk)) {
 		dev_err(&pdev->dev, "unable to get mlb clock\n");
 		return -EFAULT;
 	}
@@ -935,7 +935,7 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 	if (dev->clk_speed >= CLK_2048FS) {
 		/* enable pll */
 		dev->clk_pll = devm_clk_get(&pdev->dev, "pll8_mlb");
-		if (IS_ERR_OR_NULL(dev->clk_pll)) {
+		if (IS_ERR(dev->clk_pll)) {
 			dev_err(&pdev->dev, "unable to get mlb pll clock\n");
 			clk_disable_unprepare(dev->clk);
 			return -EFAULT;
Signature · 1 line
2.43.0
#10 lore raw

PATCH v2 2/3 staging: most: dim2: use dev_err_probe and proper error codes for clock

In reply to Artem Lytkin

Replace hardcoded -EFAULT returns with dev_err_probe() and PTR_ERR() when devm_clk_get() fails in fsl_mx6_enable(). This ensures the correct error code is propagated (e.g. -EPROBE_DEFER for deferred probing) and avoids log noise during probe deferral.

---
 drivers/staging/most/dim2/dim2.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/staging/most/dim2/dim2.c +5 -6
index cea1ba99caf7..e4c8b4995c61 100644
@@ -921,10 +921,9 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 	int ret;
 
 	dev->clk = devm_clk_get(&pdev->dev, "mlb");
-	if (IS_ERR(dev->clk)) {
-		dev_err(&pdev->dev, "unable to get mlb clock\n");
-		return -EFAULT;
-	}
+	if (IS_ERR(dev->clk))
+		return dev_err_probe(&pdev->dev, PTR_ERR(dev->clk),
+				     "unable to get mlb clock\n");
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
@@ -936,9 +935,9 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 		/* enable pll */
 		dev->clk_pll = devm_clk_get(&pdev->dev, "pll8_mlb");
 		if (IS_ERR(dev->clk_pll)) {
-			dev_err(&pdev->dev, "unable to get mlb pll clock\n");
 			clk_disable_unprepare(dev->clk);
-			return -EFAULT;
+			return dev_err_probe(&pdev->dev, PTR_ERR(dev->clk_pll),
+					     "unable to get mlb pll clock\n");
 		}
 
 		writel(0x888, dev->io_base + 0x38);
Signature · 1 line
2.43.0
#11 lore raw

PATCH v2 3/3 staging: most: dim2: remove unnecessary string indirection in dev_err

In reply to Artem Lytkin

Replace dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed") with the direct format string dev_err(&pdev->dev, "clk_prepare_enable failed\n"). The extra level of indirection through %s is unnecessary.

---
 drivers/staging/most/dim2/dim2.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/staging/most/dim2/dim2.c +3 -3
index e4c8b4995c61..66617e89e028 100644
@@ -927,7 +927,7 @@ static int fsl_mx6_enable(struct platform_device *pdev)
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
-		dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed");
+		dev_err(&pdev->dev, "clk_prepare_enable failed\n");
 		return ret;
 	}
 
@@ -975,7 +975,7 @@ static int rcar_gen2_enable(struct platform_device *pdev)
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
-		dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed");
+		dev_err(&pdev->dev, "clk_prepare_enable failed\n");
 		return ret;
 	}
 
@@ -1020,7 +1020,7 @@ static int rcar_gen3_enable(struct platform_device *pdev)
 
 	ret = clk_prepare_enable(dev->clk);
 	if (ret) {
-		dev_err(&pdev->dev, "%s\n", "clk_prepare_enable failed");
+		dev_err(&pdev->dev, "clk_prepare_enable failed\n");
 		return ret;
 	}
 
Signature · 1 line
2.43.0
#12 lore raw

Re: [PATCH v2 0/3] staging: most: dim2: clean up error handling in fsl_mx6_enable

In reply to Artem Lytkin
Quoted 24 lines · Artem Lytkin

This series cleans up error handling in fsl_mx6_enable(), split into separate logical changes per Dan Carpenter's review of v1.

Changes since v1:
- Split combined patch into 3 separate patches (one change per patch)
- Extended string indirection fix to all three enable functions

Patch 1 replaces IS_ERR_OR_NULL() with IS_ERR() since devm_clk_get() never returns NULL.

Patch 2 replaces dev_err() + hardcoded -EFAULT with dev_err_probe() and PTR_ERR() to propagate correct error codes and handle probe deferral properly.

Patch 3 removes unnecessary string indirection in dev_err() calls ("%s\n", "message" -> "message\n").

Artem Lytkin (3):
staging: most: dim2: replace IS_ERR_OR_NULL with IS_ERR for
devm_clk_get
staging: most: dim2: use dev_err_probe and proper error codes for
clock
staging: most: dim2: remove unnecessary string indirection in dev_err

Quoted 6 lines
 drivers/staging/most/dim2/dim2.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)
Signature · 1 line
2.43.0

I see 2 different copies of this series, as can be seen on lore.kernel.org:

	https://lore.kernel.org/all/[email protected]/

which is "correct"?

Can you send a v3 so I know what to do?

thanks,

greg k-h