public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] regulator: da9063: Fix up irq leak
@ 2015-05-31  4:27 Axel Lin
  2015-06-02 14:36 ` Mark Brown
  0 siblings, 1 reply; 2+ messages in thread
From: Axel Lin @ 2015-05-31  4:27 UTC (permalink / raw)
  To: Mark Brown
  Cc: Krystian Garbaciak, Liam Girdwood, Support Opensource,
	linux-kernel

Current code does not set regulators->irq_ldo_lim and regulators->irq_uvov,
so it actually calls free_irq(0, regulators) twice in remove() but does not
free the irq actually used.

Convert to use devm_request_threaded_irq instead and then we don't need to
take care the clean up irq so remove irq_ldo_lim and irq_uvov from
struct da9063_regulators. Note, regulators->irq_uvov is not used at all in
current code.

There is a slightly change in this patch, it will return error in probe()
if devm_request_threaded_irq fails. If the irq is optional, it should be
fine to allow platform_get_irq_byname fails. But current code does not
allow platform_get_irq_byname fails. So I think the reason to allow
request irq failure is just because the irq leak.

Signed-off-by: Axel Lin <axel.lin@ingics.com>
---
 drivers/regulator/da9063-regulator.c | 21 +++------------------
 1 file changed, 3 insertions(+), 18 deletions(-)

diff --git a/drivers/regulator/da9063-regulator.c b/drivers/regulator/da9063-regulator.c
index 31c2c59..aed1ad3 100644
--- a/drivers/regulator/da9063-regulator.c
+++ b/drivers/regulator/da9063-regulator.c
@@ -117,9 +117,6 @@ struct da9063_regulator {
 
 /* Encapsulates all information for the regulators driver */
 struct da9063_regulators {
-	int					irq_ldo_lim;
-	int					irq_uvov;
-
 	unsigned				n_regulators;
 	/* Array size to be defined during init. Keep at end. */
 	struct da9063_regulator			regulator[0];
@@ -867,35 +864,23 @@ static int da9063_regulator_probe(struct platform_device *pdev)
 		return irq;
 	}
 
-	ret = request_threaded_irq(irq,
+	ret = devm_request_threaded_irq(&pdev->dev, irq,
 				NULL, da9063_ldo_lim_event,
 				IRQF_TRIGGER_LOW | IRQF_ONESHOT,
 				"LDO_LIM", regulators);
 	if (ret) {
-		dev_err(&pdev->dev,
-				"Failed to request LDO_LIM IRQ.\n");
-		regulators->irq_ldo_lim = -ENXIO;
+		dev_err(&pdev->dev, "Failed to request LDO_LIM IRQ.\n");
+		return ret;
 	}
 
 	return 0;
 }
 
-static int da9063_regulator_remove(struct platform_device *pdev)
-{
-	struct da9063_regulators *regulators = platform_get_drvdata(pdev);
-
-	free_irq(regulators->irq_ldo_lim, regulators);
-	free_irq(regulators->irq_uvov, regulators);
-
-	return 0;
-}
-
 static struct platform_driver da9063_regulator_driver = {
 	.driver = {
 		.name = DA9063_DRVNAME_REGULATORS,
 	},
 	.probe = da9063_regulator_probe,
-	.remove = da9063_regulator_remove,
 };
 
 static int __init da9063_regulator_init(void)
-- 
2.1.0




^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] regulator: da9063: Fix up irq leak
  2015-05-31  4:27 [PATCH] regulator: da9063: Fix up irq leak Axel Lin
@ 2015-06-02 14:36 ` Mark Brown
  0 siblings, 0 replies; 2+ messages in thread
From: Mark Brown @ 2015-06-02 14:36 UTC (permalink / raw)
  To: Axel Lin
  Cc: Krystian Garbaciak, Liam Girdwood, Support Opensource,
	linux-kernel

[-- Attachment #1: Type: text/plain, Size: 262 bytes --]

On Sun, May 31, 2015 at 12:27:38PM +0800, Axel Lin wrote:
> Current code does not set regulators->irq_ldo_lim and regulators->irq_uvov,
> so it actually calls free_irq(0, regulators) twice in remove() but does not
> free the irq actually used.

Applied, thanks.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2015-06-02 14:36 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-05-31  4:27 [PATCH] regulator: da9063: Fix up irq leak Axel Lin
2015-06-02 14:36 ` Mark Brown

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox