Re: [PATCH v5 7/7] gpiolib: let the character device free its IRQs first on chip removal

[Herve Codina]

Hi Andy, Adrian,

On Tue, 3 Nov 2026 12:52:10 +0200
Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:

> On Tue, Nov 03, 2026 at 12:32:40PM +0200, Adrian Ambrozewicz wrote:
> > gpiochip_remove() force-frees the remaining IRQs, removes the irqchip,
> > and only then calls the device notifier on which the character device
> > frees its own IRQs. The character device can then free an IRQ it no
> > longer owns: after its mapping was disposed of and the number reused, or
> > from a release that races gpiod_free_irqs() and
> > gpiochip_irqchip_remove(). free_irq() then warns "Trying to free
> > already-free IRQ" or uses a disposed descriptor.
> > 
> > Call the notifier once the chip pointer is cleared and the SRCU grace
> > period has elapsed, when no ioctl can request IRQs any more, and free
> > the remaining IRQs after it. Have the release paths free their IRQs
> > before unregistering from the notifier, under the lock its callback
> > takes (config_mutex, or a new mutex for v1 line events). Herve Codina
> > first proposed calling the notifier earlier.
> > 
> > IRQs that in-kernel users leave requested are now freed after the chip
> > pointer is cleared, when gpiod_*() calls on the chip fail.
> > 
> > Reproduced with a test-only chip on the hierarchical irqchip: both races
> > warn in all 30 cycles; with this change, none does.  
> 
> > Link: https://lore.kernel.org/all/20240227113426.253232-1-herve.codina@xxxxxxxxxxx/  
> 
> No. Either make it Closes tag (if appropriate) or add a reference like
> 
> Link: $URL [1]
> 
> and mention [1] in the text, otherwise it makes Linus T. grumpy. I think
> I already told that.
> 

IHMO, "Link:" instead of "Closes:" according to this reply in the series
pointed out [2].

[2]: https://lore.kernel.org/all/20240923150522.66b418d1@xxxxxxxxxxx/

The history was:
 - [1] (the link in this patch) proposed a fix
 - It has been rejected and a improved one has been done [2]
 - [2] has been accepted and led to the commit ec8b6f55b981 ("gpio: free irqs
   that are still requested when the chip is being removed"). This commit
   is already mentioned as 'Fixes:" in this patch.

So, IMHO, in this patch:
  ...
  takes (config_mutex, or a new mutex for v1 line events). Herve Codina
  first proposed calling the notifier earlier [1].
  ...
  Link: https://lore.kernel.org/all/20240227113426.253232-1-herve.codina@xxxxxxxxxxx/ [1]
  Fixes: ec8b6f55b981 ("gpio: free irqs that are still requested when the chip is being removed")
  ...

Best regards,
Hervé