[PULL V2] IIO fixes for v3.7-rc1 set 2 (tiny fixes and regressions only)

[PULL V2] IIO fixes for v3.7-rc1 set 2 (tiny fixes and regressions only)

[Jonathan Cameron]

The following changes since commit 98f9c3c137cea016c57c9fd60b3caf88eb34e4cd:

  iio: Remove duplicates for light/ in Kconfig and Makefile (2012-10-19 19:44:06 +0100)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git tags/iio-fixes-for-3.7b-take2

for you to fetch changes up to 6310e59fb73d94ed702912aa3eb14be38619913d:

  iio: hid-sensor: Use __devexit annotation for remove() (2012-11-04 16:00:56 +0000)

----------------------------------------------------------------
Second round (request 2) of fixes for IIO in 3.7 cycle, applies to 3.7-rc1.

A mixed bag of fixes mainly for drivers but with one core
double free fix.

* A couple of little fixlets for the hid-sensors drivers.
* Incorrect return value in adt7410
* Double free in industrialio-event.c
* Ancient bug in tsl2563 in which the interrupt disable code
  was inverted (i.e. it enable the interrupts)

Fixes for the adsi16400 driver have been postponed and will now be queued
up for the next merge window at Greg's suggestion given their large
size and invasive nature. The relevant code never worked so it
wasn't a regression.

Jonathan

----------------------------------------------------------------
Axel Lin (2):
      iio: hid-sensor: Return proper error if kmemdup fails
      iio: hid-sensor: Use __devexit annotation for remove()

Derek Basehore (1):
      tsl2563: fixed bug with disabling interrupts

Julia Lawall (1):
      drivers/iio/industrialio-event.c: eliminate possible double free

Kirill A. Shutemov (1):
      iio: hid-sensors: convert HID_SENSOR_ENUM_BASE_QUIRKS to bool

Lars-Peter Clausen (1):
      staging:iio:adt7410: Fix adt7410_set_mode return value

 drivers/iio/accel/hid-sensor-accel-3d.c             | 10 +++++-----
 drivers/iio/common/hid-sensors/Kconfig              |  2 +-
 drivers/iio/common/hid-sensors/hid-sensor-trigger.c |  6 ++----
 drivers/iio/gyro/hid-sensor-gyro-3d.c               | 10 +++++-----
 drivers/iio/industrialio-event.c                    |  7 +------
 drivers/iio/light/hid-sensor-als.c                  |  9 ++++-----
 drivers/iio/magnetometer/hid-sensor-magn-3d.c       | 10 +++++-----
 drivers/staging/iio/adc/adt7410.c                   |  2 +-
 drivers/staging/iio/light/tsl2563.c                 |  4 ++--
 9 files changed, 26 insertions(+), 34 deletions(-)

Re: [PULL V2] IIO fixes for v3.7-rc1 set 2 (tiny fixes and regressions only)

[Greg KH]

On Sun, Nov 04, 2012 at 04:23:20PM +0000, Jonathan Cameron wrote:
> Axel Lin (2):
>       iio: hid-sensor: Use __devexit annotation for remove()

This one, while maybe nice, sure isn't necessary as CONFIG_HOTPLUG can't
be turned off anymore.  So it's not needed for 3.7, right?

thanks,

greg k-h

Re: [PULL V2] IIO fixes for v3.7-rc1 set 2 (tiny fixes and regressions only)

[Jonathan Cameron]

On 12/11/12 02:30, Greg KH wrote:
> On Sun, Nov 04, 2012 at 04:23:20PM +0000, Jonathan Cameron wrote:
>> Axel Lin (2):
>>        iio: hid-sensor: Use __devexit annotation for remove()
>
> This one, while maybe nice, sure isn't necessary as CONFIG_HOTPLUG can't
> be turned off anymore.  So it's not needed for 3.7, right?
>
Sorry that patch title doesn't mention why this is done. I should have
cleaned it up before merging (or poked Axel to do so)

Right now they are completely incorrectly marked as __devinit rather 
than __devexit.

Hence if any platform is being clever about ditching init.text after
module insertion it's not going to be there on removal.

If this was simply a case of adding __devexit I'd agree.

If this annotation still doesn't actually matter I've definitely missed
something.

Thanks as ever for continued handling of the IIO tree!

Jonathan

Re: [PULL V2] IIO fixes for v3.7-rc1 set 2 (tiny fixes and regressions only)

[Jonathan Cameron]

On 12/11/12 09:04, Jonathan Cameron wrote:
> On 12/11/12 02:30, Greg KH wrote:
>> On Sun, Nov 04, 2012 at 04:23:20PM +0000, Jonathan Cameron wrote:
>>> Axel Lin (2):
>>>        iio: hid-sensor: Use __devexit annotation for remove()
>>
>> This one, while maybe nice, sure isn't necessary as CONFIG_HOTPLUG can't
>> be turned off anymore.  So it's not needed for 3.7, right?
>>
> Sorry that patch title doesn't mention why this is done. I should have
> cleaned it up before merging (or poked Axel to do so)
>
> Right now they are completely incorrectly marked as __devinit rather
> than __devexit.
>
> Hence if any platform is being clever about ditching init.text after
> module insertion it's not going to be there on removal.
>
> If this was simply a case of adding __devexit I'd agree.
>
> If this annotation still doesn't actually matter I've definitely missed
> something.
Of course, if you'd rather pull this in for 3.8 then feel free.
>
> Thanks as ever for continued handling of the IIO tree!
>
> Jonathan
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PULL V2] IIO fixes for v3.7-rc1 set 2 (tiny fixes and regressions only)

[Greg KH]

On Mon, Nov 12, 2012 at 09:04:24AM +0000, Jonathan Cameron wrote:
> On 12/11/12 02:30, Greg KH wrote:
> >On Sun, Nov 04, 2012 at 04:23:20PM +0000, Jonathan Cameron wrote:
> >>Axel Lin (2):
> >>       iio: hid-sensor: Use __devexit annotation for remove()
> >
> >This one, while maybe nice, sure isn't necessary as CONFIG_HOTPLUG can't
> >be turned off anymore.  So it's not needed for 3.7, right?
> >
> Sorry that patch title doesn't mention why this is done. I should have
> cleaned it up before merging (or poked Axel to do so)
> 
> Right now they are completely incorrectly marked as __devinit rather
> than __devexit.
> 
> Hence if any platform is being clever about ditching init.text after
> module insertion it's not going to be there on removal.

No platform that I know of does this.  And also, __devexit is defined
away as nothing as well due to CONFIG_HOTPLUG always being set.

> If this was simply a case of adding __devexit I'd agree.
> 
> If this annotation still doesn't actually matter I've definitely missed
> something.

It doesn't matter anymore :)

So could we hold off on this one until 3.8?

thanks,

greg k-h