[PATCH 01/10] i2c: add and export of_get_i2c_adapter_by_node() interface

[Thierry Reding]

On Wed, Jul 08, 2015 at 04:31:37PM +0300, Vladimir Zapolskiy wrote:
> Hi Thierry,
> 
> On 08.07.2015 16:11, Thierry Reding wrote:
> > On Wed, Jul 08, 2015 at 03:59:12PM +0300, Vladimir Zapolskiy wrote:
> >> of_find_i2c_adapter_by_node() call requires quite often missing
> >> put_device(), and i2c_put_adapter() releases a device locked by
> >> i2c_get_adapter() only. In general module_put(adapter->owner) and
> >> put_device(dev) are not interchangeable.
> >>
> >> This is a common error reproduction scenario as a result of the
> >> misusage described above (for clearness this is run on iMX6 platform
> >> with HDMI and I2C bus drivers compiled as kernel modules):
> >>
> >>     root at mx6q:~# lsmod | grep i2c
> >>     i2c_imx                10213  0
> >>     root at mx6q:~# lsmod | grep dw_hdmi_imx
> >>     dw_hdmi_imx             3631  0
> >>     dw_hdmi                11846  1 dw_hdmi_imx
> >>     imxdrm                  8674  3 dw_hdmi_imx,imx_ipuv3_crtc,imx_ldb
> >>     drm_kms_helper        113765  5 dw_hdmi,imxdrm,imx_ipuv3_crtc,imx_ldb
> >>     root at mx6q:~# rmmod dw_hdmi_imx
> >>     root at mx6q:~# lsmod | grep i2c
> >>     i2c_imx                10213  -1
> >>
> >>                                  ^^^^^
> >>
> >>     root at mx6q:~# rmmod i2c_imx
> >>     rmmod: ERROR: Module i2c_imx is in use
> >>
> >> To fix existing users of these interfaces and to avoid any further
> >> confusion and misusage in future, add one more interface
> >> of_get_i2c_adapter_by_node(), it is similar to i2c_get_adapter() in
> >> sense that an I2C bus device driver found and locked by user can be
> >> correctly unlocked by i2c_put_adapter().
> >>
> >> Signed-off-by: Vladimir Zapolskiy <vladimir_zapolskiy at mentor.com>
> >> ---
> >> The change is based on RFC http://www.spinics.net/lists/linux-i2c/msg20257.html
> >>
> >> * added new exported function declaration in include/linux/i2c.h
> >> * added put_device(dev) call right inside of_get_i2c_adapter_by_node()
> >> * corrected authorship of the change
> >>
> >>  drivers/i2c/i2c-core.c | 20 ++++++++++++++++++++
> >>  include/linux/i2c.h    |  6 ++++++
> >>  2 files changed, 26 insertions(+)
> >>
> >> diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
> >> index 069a41f..0d902ab 100644
> >> --- a/drivers/i2c/i2c-core.c
> >> +++ b/drivers/i2c/i2c-core.c
> >> @@ -1356,6 +1356,26 @@ struct i2c_adapter *of_find_i2c_adapter_by_node(struct device_node *node)
> >>  	return i2c_verify_adapter(dev);
> >>  }
> >>  EXPORT_SYMBOL(of_find_i2c_adapter_by_node);
> >> +
> >> +struct i2c_adapter *of_get_i2c_adapter_by_node(struct device_node *node)
> >> +{
> >> +	struct device *dev;
> >> +	struct i2c_adapter *adapter;
> >> +
> >> +	dev = bus_find_device(&i2c_bus_type, NULL, node,
> >> +			      of_dev_node_match);
> >> +	if (!dev)
> >> +		return NULL;
> >> +
> >> +	adapter = i2c_verify_adapter(dev);
> >> +	if (adapter && !try_module_get(adapter->owner))
> >> +		adapter = NULL;
> >> +
> >> +	put_device(dev);
> > 
> > I don't think this is correct. Users still need to keep a reference to
> > the device, otherwise it can simply disappear even if the module stays
> > around (think sysfs bind/unbind attributes).
> > 
> > Looking at i2c_put_adapter() it seems like it would need to do more than
> > just drop the module reference. Then again, that probably means that we
> > need to add a get_device() somewhere in i2c_get_adapter() to balance the
> > put_device() in i2c_put_adapter().
> 
> it makes sense for me, thanks for momentary review.
> 
> I'm hesitating to add put_device(dev) to i2c_put_adapter() etc. in this
> series though. After development and testing I would like to send
> another preceding independent change updating i2c_get_adapter(),
> i2c_put_adapter() and clients (or if you wish you can do it), then I'll
> rebase 01/10 on top of it, the rest most probably is unchanged.

I think that would make sense, yes.

Thierry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20150708/06f627f2/attachment.sig>