Skip to contentSkip to conversation

two cleanups in atcspi200

PATCH In review Pei Xiao 2 participants 8 messages Mar 11 – Mar 16

Subsystems (2)
lore
Share & tools
lore

Pei Xiao

author
#0 lore raw

v2 cover two cleanups in atcspi200

1.Use helper function devm_clk_get_enabled() to simple code. 2.fix mutex initialization order
---
changlog in v2: fix return code error and leak the controller
---
Pei Xiao (2):
spi: atcspi200: Use helper function devm_clk_get_enabled()
spi: atcspi200: fix mutex initialization order

 drivers/spi/spi-atcspi200.c | 29 ++++++++++-------------------
 1 file changed, 10 insertions(+), 19 deletions(-)
Signature · 1 line
2.25.1
author
#1 lore raw

PATCH v2 1/2 spi: atcspi200: Use helper function devm_clk_get_enabled()

Since commit 7ef9651e9792 ("clk: Provide new devm_clk helpers for prepared and enabled clocks"), devm_clk_get() and clk_prepare_enable() can now be replaced by devm_clk_get_enabled(). Moreover, it is no longer necessary to unprepare and disable the clocks explicitly.

---
 drivers/spi/spi-atcspi200.c | 24 ++++++------------------
 1 file changed, 6 insertions(+), 18 deletions(-)
drivers/spi/spi-atcspi200.c +6 -18
index dd19fe9f9ee7..c0607193be8d 100644
@@ -486,11 +486,6 @@ static int atcspi_init_resources(struct platform_device *pdev,
 		return dev_err_probe(spi->dev, PTR_ERR(spi->regmap),
 				     "Failed to init regmap\n");
 
-	spi->clk = devm_clk_get(spi->dev, NULL);
-	if (IS_ERR(spi->clk))
-		return dev_err_probe(spi->dev, PTR_ERR(spi->clk),
-				     "Failed to get SPI clock\n");
-
 	spi->sclk_rate = ATCSPI_MAX_SPEED_HZ;
 	return 0;
 }
@@ -512,13 +507,10 @@ static int atcspi_configure_dma(struct atcspi_dev *spi)
 
 static int atcspi_enable_clk(struct atcspi_dev *spi)
 {
-	int ret;
-
-	ret = clk_prepare_enable(spi->clk);
-	if (ret)
-		return dev_err_probe(spi->dev, ret,
-				     "Failed to enable clock\n");
-
+	spi->clk = devm_clk_get_enabled(spi->dev, NULL);
+	if (IS_ERR(spi->clk))
+		return dev_err_probe(spi->dev, PTR_ERR(spi->clk),
+				     "Failed to get SPI clock\n");
 	spi->clk_rate = clk_get_rate(spi->clk);
 	if (!spi->clk_rate)
 		return dev_err_probe(spi->dev, -EINVAL,
@@ -571,15 +563,14 @@ static int atcspi_probe(struct platform_device *pdev)
 
 	ret = atcspi_setup(spi);
 	if (ret)
-		goto disable_clk;
+		goto free_controller;
 
 	ret = devm_spi_register_controller(&pdev->dev, host);
 	if (ret) {
 		dev_err_probe(spi->dev, ret,
 			      "Failed to register SPI controller\n");
-		goto disable_clk;
+		goto free_controller;
 	}
-
 	spi->use_dma = false;
 	if (ATCSPI_DMA_SUPPORT) {
 		ret = atcspi_configure_dma(spi);
@@ -593,9 +584,6 @@ static int atcspi_probe(struct platform_device *pdev)
 
 	return 0;
 
-disable_clk:
-	clk_disable_unprepare(spi->clk);
-
 free_controller:
 	spi_controller_put(host);
 	return ret;
Signature · 1 line
2.25.1
author
#2 lore raw

PATCH v2 2/2 spi: atcspi200: fix mutex initialization order

In reply to Pei Xiao

The atcspi_exec_mem_op() function may call mutex_lock() on the driver's mutex before it is properly initialized if a SPI memory operation is initiated immediately after devm_spi_register_controller() is called. The mutex initialization currently occurs after the controller registration, which leaves a window where the mutex could be used uninitialized.

Move the mutex initialization to the beginning of the probe function, before any registration or resource allocation. Also replace the plain mutex_init() with devm_mutex_init() to benefit from automatic cleanup on driver unbind, preventing potential resource leaks.

Fixes 34e3815ea459 ("spi: atcspi200: Add ATCSPI200 SPI controller driver")
---
 drivers/spi/spi-atcspi200.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/spi/spi-atcspi200.c +4 -1
index c0607193be8d..dd1292c5bb98 100644
@@ -551,6 +551,10 @@ static int atcspi_probe(struct platform_device *pdev)
 	spi->dev = &pdev->dev;
 	dev_set_drvdata(&pdev->dev, host);
 
+	ret = devm_mutex_init(spi->dev, &spi->mutex_lock);
+	if (ret)
+		goto free_controller;
+
 	ret = atcspi_init_resources(pdev, spi, &mem_res);
 	if (ret)
 		goto free_controller;
@@ -580,7 +584,6 @@ static int atcspi_probe(struct platform_device *pdev)
 		else
 			spi->use_dma = true;
 	}
-	mutex_init(&spi->mutex_lock);
 
 	return 0;
 
Signature · 1 line
2.25.1
maintainer
#3 lore raw

Re: [PATCH v2 2/2] spi: atcspi200: fix mutex initialization order

drivers/spi/spi-atcspi200.c static int atcspi_probe(struct platform_device *pdev) · line 557

Quoted 8 lines · Pei Xiao
drivers/spi/spi-atcspi200.c +4 -0
@@ -551,6 +551,10 @@ static int atcspi_probe(struct platform_device *pdev)
 	spi->dev = &pdev->dev;
 	dev_set_drvdata(&pdev->dev, host);
 
+	ret = devm_mutex_init(spi->dev, &spi->mutex_lock);
+	if (ret)
+		goto free_controller;
+

This means we're now using devm to free the mutex that is in spi, but when we pass spi into devm_spi_register_controller() we'll cause spi to be freed before we free the mutex. This pattern gets a bit clunky with devm...

author
#4 lore raw

Re: [PATCH v2 2/2] spi: atcspi200: fix mutex initialization order

Quoted 6 lines · Mark Brown
Quoted 4 lines · Pei Xiao
drivers/spi/spi-atcspi200.c +0 -0
@@ -551,6 +551,10 @@ static int atcspi_probe(struct platform_device *pdev)
 	spi->dev = &pdev->dev;
 	dev_set_drvdata(&pdev->dev, host);
 
Quoted 8 lines

drivers/spi/spi-atcspi200.c static int atcspi_probe(struct platform_device *pdev) · line 557

Quoted 4 lines
drivers/spi/spi-atcspi200.c +4 -0
+	ret = devm_mutex_init(spi->dev, &spi->mutex_lock);
+	if (ret)
+		goto free_controller;
+

This means we're now using devm to free the mutex that is in spi, but when we pass spi into devm_spi_register_controller() we'll cause spi to be freed before we free the mutex. This pattern gets a bit clunky with devm...

Thanks for review! I will use mutex_init() for V3 version.

maintainer
#5 lore raw

Re: [PATCH v2 2/2] spi: atcspi200: fix mutex initialization order

Quoted 13 lines · Pei Xiao
Quoted 11 lines · Mark Brown

drivers/spi/spi-atcspi200.c static int atcspi_probe(struct platform_device *pdev) · line 557

Quoted 4 lines · Pei Xiao
drivers/spi/spi-atcspi200.c +4 -0
+	ret = devm_mutex_init(spi->dev, &spi->mutex_lock);
+	if (ret)
+		goto free_controller;
+

This means we're now using devm to free the mutex that is in spi, but when we pass spi into devm_spi_register_controller() we'll cause spi to be freed before we free the mutex. This pattern gets a bit clunky with devm...

Quoted 1 line

Thanks for review! I will use mutex_init() for V3 version.

That's still going to have trouble because if you do the free in the removal function the lock will go away before the controller which is it's own issue (this is what I mean about clunky). It's especially annoying as it's only an issue for debugging builds. I can't actually see a good way of dealing with this, perhaps just a plain mutex_init() without a free is the least bad option.

maintainer
#6 lore raw

Re: (subset) [PATCH v2 0/2] two cleanups in atcspi200

In reply to Pei Xiao
Quoted 2 lines · Pei Xiao

1.Use helper function devm_clk_get_enabled() to simple code. 2.fix mutex initialization order

Applied to

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-next

Thanks!

[1/2] spi: atcspi200: Use helper function devm_clk_get_enabled()

      commit: 97545e37234fdbe457f5104a09f55033550b3d84

All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying to this mail.

Thanks,
Mark

maintainer
#7 lore raw

Re: (subset) [PATCH v2 0/2] two cleanups in atcspi200

In reply to Pei Xiao
Quoted 4 lines · Pei Xiao

two cleanups in atcspi200

1.Use helper function devm_clk_get_enabled() to simple code. 2.fix mutex initialization order

Applied to

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-7.1

Thanks!

[1/2] spi: atcspi200: Use helper function devm_clk_get_enabled()

      https://git.kernel.org/broonie/misc/c/97545e37234f

All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying to this mail.

Thanks,
Mark