Re: [PATCH 1/7] vfio/pci: Disable auto-enable of exclusive INTx IRQ

[Alex Williamson]

On Fri, 8 Mar 2024 07:23:21 +0000
"Tian, Kevin" <kevin.tian@xxxxxxxxx> wrote:

> > From: Alex Williamson <alex.williamson@xxxxxxxxxx>
> > Sent: Friday, March 8, 2024 4:24 AM
> > 
> > On Thu, 7 Mar 2024 08:39:16 +0000
> > "Tian, Kevin" <kevin.tian@xxxxxxxxx> wrote:
> >   
> > > > From: Alex Williamson <alex.williamson@xxxxxxxxxx>
> > > > Sent: Thursday, March 7, 2024 5:15 AM
> > > >
> > > > Currently for devices requiring masking at the irqchip for INTx, ie.
> > > > devices without DisINTx support, the IRQ is enabled in request_irq()
> > > > and subsequently disabled as necessary to align with the masked status
> > > > flag.  This presents a window where the interrupt could fire between
> > > > these events, resulting in the IRQ incrementing the disable depth twice.
> > > > This would be unrecoverable for a user since the masked flag prevents
> > > > nested enables through vfio.
> > > >
> > > > Instead, invert the logic using IRQF_NO_AUTOEN such that exclusive INTx
> > > > is never auto-enabled, then unmask as required.
> > > >
> > > > Fixes: 89e1f7d4c66d ("vfio: Add PCI device driver")
> > > > Signed-off-by: Alex Williamson <alex.williamson@xxxxxxxxxx>  
> > >
> > > CC stable?  
> > 
> > I've always found that having a Fixes: tag is sufficient to get picked
> > up for stable, so I typically don't do both.  If it helps out someone's
> > process I'd be happy to though.  Thanks,
> >   
> 
> According to "Documentation/process/submitting-patches.rst":
> 
>   Note: Attaching a Fixes: tag does not subvert the stable kernel rules
>   process nor the requirement to Cc: stable@xxxxxxxxxxxxxxx on all stable
>   patch candidates. For more information, please read
>   Documentation/process/stable-kernel-rules.rst.
> 
> Probably it's fine as long as the stable kernel maintainers don't complain. 😊

I think the stable maintainers are far more aggressive than the
documentation would suggest, but it doesn't hurt to include the Cc,
I'll add it next version.  Thanks,

Alex