Re: [PATCH v2] PCI: Fix potential deadlock in pcim_intx()

[Philipp Stanner]

On Wed, 2024-09-11 at 09:27 -0500, Bjorn Helgaas wrote:
> On Thu, Sep 05, 2024 at 09:25:57AM +0200, Philipp Stanner wrote:
> > commit 25216afc9db5 ("PCI: Add managed pcim_intx()") moved the
> > allocation step for pci_intx()'s device resource from
> > pcim_enable_device() to pcim_intx(). As before,
> > pcim_enable_device()
> > sets pci_dev.is_managed to true; and it is never set to false
> > again.
> > 
> > Due to the lifecycle of a struct pci_dev, it can happen that a
> > second
> > driver obtains the same pci_dev after a first driver ran.
> > If one driver uses pcim_enable_device() and the other doesn't,
> > this causes the other driver to run into managed pcim_intx(), which
> > will
> > try to allocate when called for the first time.
> > 
> > Allocations might sleep, so calling pci_intx() while holding
> > spinlocks
> > becomes then invalid, which causes lockdep warnings and could cause
> > deadlocks:
> > 
> > ========================================================
> > WARNING: possible irq lock inversion dependency detected
> > 6.11.0-rc6+ #59 Tainted: G        W
> > --------------------------------------------------------
> > CPU 0/KVM/1537 just changed the state of lock:
> > ffffa0f0cff965f0 (&vdev->irqlock){-...}-{2:2}, at:
> > vfio_intx_handler+0x21/0xd0 [vfio_pci_core] but this lock took
> > another,
> > HARDIRQ-unsafe lock in the past: (fs_reclaim){+.+.}-{0:0}
> > 
> > and interrupts could create inverse lock ordering between them.
> > 
> > other info that might help us debug this:
> >  Possible interrupt unsafe locking scenario:
> > 
> >        CPU0                    CPU1
> >        ----                    ----
> >   lock(fs_reclaim);
> >                                local_irq_disable();
> >                                lock(&vdev->irqlock);
> >                                lock(fs_reclaim);
> >   <Interrupt>
> >     lock(&vdev->irqlock);
> > 
> >  *** DEADLOCK ***
> > 
> > Have pcim_enable_device()'s release function,
> > pcim_disable_device(), set
> > pci_dev.is_managed to false so that subsequent drivers using the
> > same
> > struct pci_dev do implicitly run into managed code.

Oops, that should obviously be "do *not* run into managed code."

Mea culpa. Maybe you can ammend that, Bjorn?

> > 
> > Fixes: 25216afc9db5 ("PCI: Add managed pcim_intx()")
> > Reported-by: Alex Williamson <alex.williamson@xxxxxxxxxx>
> > Closes:
> > https://lore.kernel.org/all/20240903094431.63551744.alex.williamson@xxxxxxxxxx/
> > Suggested-by: Alex Williamson <alex.williamson@xxxxxxxxxx>
> > Signed-off-by: Philipp Stanner <pstanner@xxxxxxxxxx>
> > Tested-by: Alex Williamson <alex.williamson@xxxxxxxxxx>
> > ---
> > @Bjorn:
> > This problem was introduced in the v6.11 merge window. So one might
> > consider getting it into mainline before v6.11.0 gets tagged.
> 
> Applied with Damien's Reviewed-by to pci/for-linus for v6.11, thanks.

thx!

P.

> 
> > ---
> >  drivers/pci/devres.c | 2 ++
> >  1 file changed, 2 insertions(+)
> > 
> > diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c
> > index 3780a9f9ec00..c7affbbf73ab 100644
> > --- a/drivers/pci/devres.c
> > +++ b/drivers/pci/devres.c
> > @@ -483,6 +483,8 @@ static void pcim_disable_device(void *pdev_raw)
> >  
> >  	if (!pdev->pinned)
> >  		pci_disable_device(pdev);
> > +
> > +	pdev->is_managed = false;
> >  }
> >  
> >  /**
> > -- 
> > 2.46.0
> > 
>