1. Patchew
  2. linux
  3. View series

[PATCH] irqchip/imx-intmux: fix runtime PM teardown on remove

Guangshuo Li posted 1 patch 1 week, 3 days ago 
drivers/irqchip/irq-imx-intmux.c | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
[PATCH] irqchip/imx-intmux: fix runtime PM teardown on remove
Posted by Guangshuo Li 1 week, 3 days ago 
imx_intmux_probe() enables the IPG clock and then drops its runtime PM
reference with pm_runtime_put(). With CONFIG_PM enabled, the runtime
suspend callback disables the IPG clock. With CONFIG_PM disabled, the
clock remains enabled after probe.

imx_intmux_remove() accesses the CHANIER registers without first making
sure that the device is runtime active. The device may therefore be
runtime suspended with the IPG clock disabled when these registers are
accessed. In addition, the remove path only calls pm_runtime_disable()
and does not balance the clock enable when CONFIG_PM is disabled.

Resume the device and acquire a runtime PM reference before accessing
the registers. If resume fails, skip the register accesses but still
tear down the chained handlers and IRQ domains. Disable runtime PM
afterwards, drop the acquired reference without triggering another
runtime suspend, and explicitly disable the IPG clock.

This keeps the clock enabled while the hardware registers are accessed
and balances the clk_prepare_enable() performed during probe for both
CONFIG_PM configurations.

This issue was found by manual code inspection.

Fixes: bb403111e017 ("irqchip/imx-intmux: Implement intmux runtime power management")
Cc: stable@vger.kernel.org
Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
---
 drivers/irqchip/irq-imx-intmux.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/irqchip/irq-imx-intmux.c b/drivers/irqchip/irq-imx-intmux.c
index 47c2681d138a..de2a8f98f625 100644
--- a/drivers/irqchip/irq-imx-intmux.c
+++ b/drivers/irqchip/irq-imx-intmux.c
@@ -288,12 +288,16 @@ static int imx_intmux_probe(struct platform_device *pdev)
 static void imx_intmux_remove(struct platform_device *pdev)
 {
 	struct intmux_data *data = platform_get_drvdata(pdev);
-	int i;
+	int i, ret;
+
+	ret = pm_runtime_resume_and_get(&pdev->dev);
+	if (ret < 0)
+		dev_warn(&pdev->dev, "failed to resume device: %d\n", ret);
 
 	for (i = 0; i < data->channum; i++) {
 		/* disable all interrupt sources of this channel */
-		writel_relaxed(0, data->regs + CHANIER(i));
-
+		if (ret >= 0)
+			writel_relaxed(0, data->regs + CHANIER(i));
 		irq_set_chained_handler_and_data(data->irqchip_data[i].irq,
 						 NULL, NULL);
 
@@ -301,6 +305,11 @@ static void imx_intmux_remove(struct platform_device *pdev)
 	}
 
 	pm_runtime_disable(&pdev->dev);
+	if (ret >= 0) {
+		pm_runtime_put_noidle(&pdev->dev);
+		clk_disable_unprepare(data->ipg_clk);
+		pm_runtime_set_suspended(&pdev->dev);
+	}
 }
 
 #ifdef CONFIG_PM
-- 
2.43.0
Re: [PATCH] irqchip/imx-intmux: fix runtime PM teardown on remove
Posted by Radu Rendec 1 week ago 
On Mon, 2026-09-14 at 21:42 +0800, Guangshuo Li wrote:
> imx_intmux_probe() enables the IPG clock and then drops its runtime PM
> reference with pm_runtime_put(). With CONFIG_PM enabled, the runtime
> suspend callback disables the IPG clock. With CONFIG_PM disabled, the
> clock remains enabled after probe.
> 
> imx_intmux_remove() accesses the CHANIER registers without first making
> sure that the device is runtime active. The device may therefore be
> runtime suspended with the IPG clock disabled when these registers are
> accessed. In addition, the remove path only calls pm_runtime_disable()
> and does not balance the clock enable when CONFIG_PM is disabled.
> 
> Resume the device and acquire a runtime PM reference before accessing
> the registers. If resume fails, skip the register accesses but still
> tear down the chained handlers and IRQ domains. Disable runtime PM
> afterwards, drop the acquired reference without triggering another
> runtime suspend, and explicitly disable the IPG clock.
> 
> This keeps the clock enabled while the hardware registers are accessed
> and balances the clk_prepare_enable() performed during probe for both
> CONFIG_PM configurations.
> 
> This issue was found by manual code inspection.
> 
> Fixes: bb403111e017 ("irqchip/imx-intmux: Implement intmux runtime power management")
> Cc: stable@vger.kernel.org
> Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
> ---
>  drivers/irqchip/irq-imx-intmux.c | 15 ++++++++++++---
>  1 file changed, 12 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/irqchip/irq-imx-intmux.c b/drivers/irqchip/irq-imx-intmux.c
> index 47c2681d138a..de2a8f98f625 100644
> --- a/drivers/irqchip/irq-imx-intmux.c
> +++ b/drivers/irqchip/irq-imx-intmux.c
> @@ -288,12 +288,16 @@ static int imx_intmux_probe(struct platform_device *pdev)
>  static void imx_intmux_remove(struct platform_device *pdev)
>  {
>  	struct intmux_data *data = platform_get_drvdata(pdev);
> -	int i;
> +	int i, ret;
> +
> +	ret = pm_runtime_resume_and_get(&pdev->dev);
> +	if (ret < 0)
> +		dev_warn(&pdev->dev, "failed to resume device: %d\n", ret);
>  
>  	for (i = 0; i < data->channum; i++) {
>  		/* disable all interrupt sources of this channel */
> -		writel_relaxed(0, data->regs + CHANIER(i));
> -
> +		if (ret >= 0)
> +			writel_relaxed(0, data->regs + CHANIER(i));
>  		irq_set_chained_handler_and_data(data->irqchip_data[i].irq,
>  						 NULL, NULL);
>  
> @@ -301,6 +305,11 @@ static void imx_intmux_remove(struct platform_device *pdev)
>  	}
>  
>  	pm_runtime_disable(&pdev->dev);
> +	if (ret >= 0) {
> +		pm_runtime_put_noidle(&pdev->dev);
> +		clk_disable_unprepare(data->ipg_clk);
> +		pm_runtime_set_suspended(&pdev->dev);
> +	}
>  }
>  
>  #ifdef CONFIG_PM

Reviewed-by: Radu Rendec <radu@rendec.net>
Re: [PATCH] irqchip/imx-intmux: fix runtime PM teardown on remove
Posted by Frank Li 1 week, 1 day ago 
On Mon, Sep 14, 2026 at 09:42:11PM +0800, Guangshuo Li wrote:
> imx_intmux_probe() enables the IPG clock and then drops its runtime PM
> reference with pm_runtime_put(). With CONFIG_PM enabled, the runtime
> suspend callback disables the IPG clock. With CONFIG_PM disabled, the
> clock remains enabled after probe.
>
> imx_intmux_remove() accesses the CHANIER registers without first making
> sure that the device is runtime active. The device may therefore be
> runtime suspended with the IPG clock disabled when these registers are
> accessed. In addition, the remove path only calls pm_runtime_disable()
> and does not balance the clock enable when CONFIG_PM is disabled.
>
> Resume the device and acquire a runtime PM reference before accessing
> the registers. If resume fails, skip the register accesses but still
> tear down the chained handlers and IRQ domains. Disable runtime PM
> afterwards, drop the acquired reference without triggering another
> runtime suspend, and explicitly disable the IPG clock.
>
> This keeps the clock enabled while the hardware registers are accessed
> and balances the clk_prepare_enable() performed during probe for both
> CONFIG_PM configurations.
>
> This issue was found by manual code inspection.
>
> Fixes: bb403111e017 ("irqchip/imx-intmux: Implement intmux runtime power management")
> Cc: stable@vger.kernel.org
> Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
> ---

Reviewed-by: Frank Li <Frank.Li@nxp.com>

>  drivers/irqchip/irq-imx-intmux.c | 15 ++++++++++++---
>  1 file changed, 12 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/irqchip/irq-imx-intmux.c b/drivers/irqchip/irq-imx-intmux.c
> index 47c2681d138a..de2a8f98f625 100644
> --- a/drivers/irqchip/irq-imx-intmux.c
> +++ b/drivers/irqchip/irq-imx-intmux.c
> @@ -288,12 +288,16 @@ static int imx_intmux_probe(struct platform_device *pdev)
>  static void imx_intmux_remove(struct platform_device *pdev)
>  {
>  	struct intmux_data *data = platform_get_drvdata(pdev);
> -	int i;
> +	int i, ret;
> +
> +	ret = pm_runtime_resume_and_get(&pdev->dev);
> +	if (ret < 0)
> +		dev_warn(&pdev->dev, "failed to resume device: %d\n", ret);
>
>  	for (i = 0; i < data->channum; i++) {
>  		/* disable all interrupt sources of this channel */
> -		writel_relaxed(0, data->regs + CHANIER(i));
> -
> +		if (ret >= 0)
> +			writel_relaxed(0, data->regs + CHANIER(i));
>  		irq_set_chained_handler_and_data(data->irqchip_data[i].irq,
>  						 NULL, NULL);
>
> @@ -301,6 +305,11 @@ static void imx_intmux_remove(struct platform_device *pdev)
>  	}
>
>  	pm_runtime_disable(&pdev->dev);
> +	if (ret >= 0) {
> +		pm_runtime_put_noidle(&pdev->dev);
> +		clk_disable_unprepare(data->ipg_clk);
> +		pm_runtime_set_suspended(&pdev->dev);
> +	}
>  }
>
>  #ifdef CONFIG_PM
> --
> 2.43.0
>
>
Re: [PATCH] irqchip/imx-intmux: fix runtime PM teardown on remove
Posted by Radu Rendec 1 week, 1 day ago 
On Mon, 2026-09-14 at 21:42 +0800, Guangshuo Li wrote:
> imx_intmux_probe() enables the IPG clock and then drops its runtime PM
> reference with pm_runtime_put(). With CONFIG_PM enabled, the runtime
> suspend callback disables the IPG clock. With CONFIG_PM disabled, the
> clock remains enabled after probe.
> 
> imx_intmux_remove() accesses the CHANIER registers without first making
> sure that the device is runtime active. The device may therefore be
> runtime suspended with the IPG clock disabled when these registers are
> accessed. In addition, the remove path only calls pm_runtime_disable()
> and does not balance the clock enable when CONFIG_PM is disabled.
> 
> Resume the device and acquire a runtime PM reference before accessing
> the registers. If resume fails, skip the register accesses but still
> tear down the chained handlers and IRQ domains. Disable runtime PM
> afterwards, drop the acquired reference without triggering another
> runtime suspend, and explicitly disable the IPG clock.
> 
> This keeps the clock enabled while the hardware registers are accessed
> and balances the clk_prepare_enable() performed during probe for both
> CONFIG_PM configurations.
> 
> This issue was found by manual code inspection.
> 
> Fixes: bb403111e017 ("irqchip/imx-intmux: Implement intmux runtime power management")
> Cc: stable@vger.kernel.org
> Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
> ---
>  drivers/irqchip/irq-imx-intmux.c | 15 ++++++++++++---
>  1 file changed, 12 insertions(+), 3 deletions(-)
> 

+ Zhipeng

The patch looks good to me but I'm wondering if it's not a better idea
to fix it in a similar manner to the imx-irqsteer driver. Please see
https://lore.kernel.org/all/20260821101039.4037925-7-Zhipeng.wang_1@oss.nxp.com/

The two drivers are already similar, so I think it makes sense to keep
them aligned. It would also address some of the issues that sashiko
flagged (probably, I haven't looked very closely).

> diff --git a/drivers/irqchip/irq-imx-intmux.c b/drivers/irqchip/irq-imx-intmux.c
> index 47c2681d138a..de2a8f98f625 100644
> --- a/drivers/irqchip/irq-imx-intmux.c
> +++ b/drivers/irqchip/irq-imx-intmux.c
> @@ -288,12 +288,16 @@ static int imx_intmux_probe(struct platform_device *pdev)
>  static void imx_intmux_remove(struct platform_device *pdev)
>  {
>  	struct intmux_data *data = platform_get_drvdata(pdev);
> -	int i;
> +	int i, ret;
> +
> +	ret = pm_runtime_resume_and_get(&pdev->dev);
> +	if (ret < 0)
> +		dev_warn(&pdev->dev, "failed to resume device: %d\n", ret);
>  
>  	for (i = 0; i < data->channum; i++) {
>  		/* disable all interrupt sources of this channel */
> -		writel_relaxed(0, data->regs + CHANIER(i));
> -
> +		if (ret >= 0)
> +			writel_relaxed(0, data->regs + CHANIER(i));
>  		irq_set_chained_handler_and_data(data->irqchip_data[i].irq,
>  						 NULL, NULL);
>  
> @@ -301,6 +305,11 @@ static void imx_intmux_remove(struct platform_device *pdev)
>  	}
>  
>  	pm_runtime_disable(&pdev->dev);
> +	if (ret >= 0) {
> +		pm_runtime_put_noidle(&pdev->dev);
> +		clk_disable_unprepare(data->ipg_clk);
> +		pm_runtime_set_suspended(&pdev->dev);
> +	}
>  }
>  
>  #ifdef CONFIG_PM

-- 
Regards,
Radu

Patchew 2.1-devel-ea86937

© 2016 - 2026 Red Hat, Inc.