[PATCH] spi: erase pointer to drvdata on removal

[PATCH] spi: erase pointer to drvdata on removal

[Vivien Didelot]

As for i2c-core, let the SPI core handle the removal of the device's
drvdata, after a remove() or a probe() failure.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
---
 drivers/spi/spi.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 84c2861..fe636fe 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -233,15 +233,25 @@ EXPORT_SYMBOL_GPL(spi_bus_type);
 static int spi_drv_probe(struct device *dev)
 {
 	const struct spi_driver		*sdrv = to_spi_driver(dev->driver);
+	struct spi_device		*sdev = to_spi_device(dev);
+	int				status;
 
-	return sdrv->probe(to_spi_device(dev));
+	status = sdrv->probe(sdev);
+	if (status)
+		spi_set_drvdata(sdev, NULL);
+	return status;
 }
 
 static int spi_drv_remove(struct device *dev)
 {
 	const struct spi_driver		*sdrv = to_spi_driver(dev->driver);
+	struct spi_device		*sdev = to_spi_device(dev);
+	int				status;
 
-	return sdrv->remove(to_spi_device(dev));
+	status = sdrv->remove(sdev);
+	if (status == 0)
+		spi_set_drvdata(sdev, NULL);
+	return status;
 }
 
 static void spi_drv_shutdown(struct device *dev)
-- 
1.7.11.7

Re: [PATCH] spi: erase pointer to drvdata on removal

[Mark Brown]

On Thu, Nov 01, 2012 at 02:05:36PM -0400, Vivien Didelot wrote:
> As for i2c-core, let the SPI core handle the removal of the device's
> drvdata, after a remove() or a probe() failure.

Any driver that notices this change is buggy, the driver shouldn't
use a drvdata value that it didn't set.  I had thought this stuff had
all been removed from I2C and either dropped or factored out into the
driver core...

------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. SALE $99.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122412

Re: [PATCH] spi: erase pointer to drvdata on removal

[Grant Likely]

On Mon, 14 Jan 2013 02:51:45 +0000, Mark Brown <broonie@opensource.wolfsonmicro.com> wrote:
> On Thu, Nov 01, 2012 at 02:05:36PM -0400, Vivien Didelot wrote:
> > As for i2c-core, let the SPI core handle the removal of the device's
> > drvdata, after a remove() or a probe() failure.
> 
> Any driver that notices this change is buggy, the driver shouldn't
> use a drvdata value that it didn't set.  I had thought this stuff had
> all been removed from I2C and either dropped or factored out into the
> driver core...

Looks to me like __device_release_driver() in drivers/base/dd.c already
does this.

g.


-- 
Grant Likely, B.Sc, P.Eng.
Secret Lab Technologies, Ltd.

Re: [PATCH] spi: erase pointer to drvdata on removal

[Vivien Didelot]

Hi Mark, Grant,

> On Mon, 14 Jan 2013 02:51:45 +0000, Mark Brown
> <broonie@opensource.wolfsonmicro.com> wrote:
> > On Thu, Nov 01, 2012 at 02:05:36PM -0400, Vivien Didelot wrote:
> > > As for i2c-core, let the SPI core handle the removal of the
> > > device's
> > > drvdata, after a remove() or a probe() failure.
> > 
> > Any driver that notices this change is buggy, the driver shouldn't
> > use a drvdata value that it didn't set.  I had thought this stuff
> > had
> > all been removed from I2C and either dropped or factored out into
> > the
> > driver core...
> 
> Looks to me like __device_release_driver() in drivers/base/dd.c
> already
> does this.
> 
> g.

You are right, I figured this out after the resend of this patch, see my reply here:
https://lkml.org/lkml/2012/12/4/224

    -Vivien