Re: [PATCH v2 0/8] HID: iio: Avoid race between callback setup and device exposure

[Jonathan Cameron <jic23@kernel.org>]

On Wed, 24 Jun 2026 14:18:17 +0300
Andy Shevchenko <andriy.shevchenko@intel.com> wrote:

> On Wed, Jun 24, 2026 at 07:13:12AM +0530, Sanjay Chitroda wrote:
> > On 23 June 2026 4:00:27 pm IST, Andy Shevchenko <andriy.shevchenko@intel.com> wrote:  
> > >On Mon, Jun 22, 2026 at 10:59:56AM +0530, Sanjay Chitroda wrote:  
> > >> 
> > >> This series avoid a race condition in HID IIO drivers related to the
> > >> ordering between callback registration and device exposure.
> > >> 
> > >> Currently, several HID IIO drivers register the IIO device (making it
> > >> visible to userspace and other kernel consumers) before all required
> > >> callbacks and resources are fully initialized, or rely on devm-based
> > >> cleanup in a way that does not guarantee correct teardown ordering.
> > >> This creates a window where the device can be accessed while it is  
> > >
> > >There is a difference between "this creates" and "this might create".
> > >I believe Srinivas and others were asking for the proof. So, what path
> > >in the code makes this happen or possible to happen?
> > >  
> > iio_device_register() exposes the IIO device to user space, while
> > sensor_hub_register_callback() registers callbacks for buffered IIO(streaming
> > mode).
> > 
> > This might create window where from userspace buffer mode is enabled and
> > callbacks are not registered which would result into loss of samples until
> > callback registration completes, although no explicit failure. In teardown
> > path which can resulting in stale/no data.
> > 
> > This was discussed in the v1 thread and v2 was posted based on discussion and
> > agreement:
> >  https://lore.kernel.org/all/3FED088A-651B-4E8B-840B-1B92CB4DF6F4@gmail.com/
> >   
> > >> not fully initialized or is being torn down, potentially leading to
> > >> sample drop or stale/no data.
> > >> 
> > >> To handle this, the series ensures that:
> > >>   - All required callbacks and resources are set up before the device
> > >>     is registered with the IIO core
> > >>   - Resource cleanup is performed explicitly where ordering matters
> > >> 
> > >> PS: This is prepratory series to convert all HID IIO driver to devm.
> > >> 
> > >> Testing:
> > >>   - Compiled with W=1 for each patch in series
> > >> 
> > >> ---
> > >> Changes in v2:
> > >> - Drop fixes tag and rectify commit message with reference to that  
> > >
> > >You also dropped my tag. Why?
> > >  
> > Thank you for the review and tag on v1.
> > 
> > While code changes are intact in v2, the rational and commit message were
> > updated substantially. Since commit message is as important as change which
> > will be permanent in history for future reference, I chose to drop the tag to
> > request a fresh review.  
> 
> Now it's clear, thanks.
> This version with changed commit messages seems good to me.
> Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Applied. Thanks,

> 
> > I shall highlight the same in change log. I'll make sure to note in future revision.  
> 
> Yes, please.
> 
> >   
> > >> - Link to v1: https://patch.msgid.link/20260606-5-june-hid-iio-race-fixes-v1-0-27a848c5758f@gmail.com  
>