Re: [PATCH] pci: Don't call resume callback for nearly bound devices

[Uwe Kleine-König]

On Tue, Nov 09, 2021 at 02:05:18PM -0600, Bjorn Helgaas wrote:
> On Tue, Nov 09, 2021 at 07:58:47PM +0100, Rafael J. Wysocki wrote:
> > On Tue, Nov 9, 2021 at 7:52 PM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
> > > On Tue, Nov 9, 2021 at 7:12 PM Bjorn Helgaas <helgaas@xxxxxxxxxx> wrote:
> > > > On Tue, Nov 09, 2021 at 06:18:18PM +0100, Rafael J. Wysocki wrote:
> > > > > On Tue, Nov 9, 2021 at 7:59 AM Uwe Kleine-König
> > > > > <u.kleine-koenig@xxxxxxxxxxxxxx> wrote:
> > > > > > On Mon, Nov 08, 2021 at 08:56:19PM -0600, Bjorn Helgaas wrote:
> > > > > > > [+cc Greg: new device_is_bound() use]
> > > > > >
> > > > > > ack, that's what I would have suggested now, too.
> > > > > >
> > > > > > > On Mon, Nov 08, 2021 at 10:22:26PM +0100, Uwe Kleine-König wrote:
> > > > > > > > pci_pm_runtime_resume() exits early when the device to resume isn't
> > > > > > > > bound yet:
> > > > > > > >
> > > > > > > >     if (!to_pci_driver(dev->driver))
> > > > > > > >             return 0;
> > > > > > > >
> > > > > > > > This however isn't true when the device currently probes and
> > > > > > > > local_pci_probe() calls pm_runtime_get_sync() because then the driver
> > > > > > > > core already setup dev->driver. As a result the driver's resume callback
> > > > > > > > is called before the driver's probe function is called and so more often
> > > > > > > > than not required driver data isn't setup yet.
> > > > > > > >
> > > > > > > > So replace the check for the device being unbound by a check that only
> > > > > > > > becomes true after .probe() succeeded.
> > > > > > >
> > > > > > > I like the fact that this patch is short and simple.
> > > > > > >
> > > > > > > But there are 30+ users of to_pci_driver().  This patch asserts that
> > > > > > > *one* of them, pci_pm_runtime_resume(), is special and needs to test
> > > > > > > device_is_bound() instead of using to_pci_driver().
> > > > > >
> > > > > > Maybe for the other locations using device_is_bound(&pdev->dev) instead
> > > > > > of to_pci_driver(pdev) != NULL would be nice, too?
> > > > > >
> > > > > > I have another doubt: device_is_bound() should (according to its
> > > > > > kernel-doc) be called with the device lock held. For the call stack that
> > > > > > is (maybe) fixed here, the lock is held (by __device_attach). We
> > > > > > probably should check if the lock is also held for the other calls of
> > > > > > pci_pm_runtime_resume().
> > > > > >
> > > > > > Hmm, the device lock is a mutex, the pm functions might be called in
> > > > > > atomic context, right?
> > > > > >
> > > > > > > It's special because the current PM implementation calls it via
> > > > > > > pm_runtime_get_sync() before the driver's .probe() method.  That
> > > > > > > connection is a little bit obscure and fragile.  What if the PM
> > > > > > > implementation changes?
> > > > > >
> > > > > > Maybe a saver bet would be to not use pm_runtime_get_sync() in
> > > > > > local_pci_probe()?
> > > > >
> > > > > Yes, in principle it might be replaced with pm_runtime_get_noresume().
> > > > >
> > > > > In theory, that may be problematic if a device is put into a low-power
> > > > > state on remove and then the driver is bound again to it.
> > > > >
> > > > > > I wonder if the same problem exists on remove, i.e. pci_device_remove()
> > > > > > calls pm_runtime_put_sync() after the driver's .remove() callback was
> > > > > > called.
> > > > >
> > > > > If it is called after ->remove() and before clearing the device's
> > > > > driver pointer, then yes.
> > > >
> > > > Yes, that is the case:
> > > >
> > > >   pci_device_remove
> > > >     if (drv->remove) {
> > > >       pm_runtime_get_sync
> > > >       drv->remove()                # <-- driver ->remove() method
> > > >       pm_runtime_put_noidle
> > > >     }
> > > >     ...
> > > >     pm_runtime_put_sync            # <-- after ->remove()
> > > >
> > > > So pm_runtime_put_sync() is called after drv->remove(), and it may
> > > > call drv->pm->runtime_idle().  I think the driver may not expect this.
> > > >
> > > > > If this is turned into pm_runtime_put_noidle(), all should work.
> > > >
> > > > pci_device_remove() already calls pm_runtime_put_noidle() immediately
> > > > after calling the driver ->remove() method.
> > > >
> > > > Are you saying we should do this, which means pci_device_remove()
> > > > would call pm_runtime_put_noidle() twice?
> > >
> > > Well, they are both needed to keep the PM-runtime reference counting in balance.
> > >
> > > This still has an issue, though, because user space would be able to
> > > trigger a runtime suspend via sysfs after we've dropped the last
> > > reference to the device in pci_device_remove().
> > >
> > > So instead, we can drop the pm_runtime_get_sync() and
> > > pm_runtime_put_sync() from local_pci_probe() and pci_device_remove(),
> > > respectively, and add pm_runtine_get_noresume() to pci_pm_init(),
> > > which will prevent PM-runtime from touching the device until it has a
> > > driver that supports PM-runtime.
> > >
> > > We'll lose the theoretical ability to put unbound devices into D3 this
> > > way, but we learned some time ago that this isn't safe in all cases
> > > anyway.
> > 
> > IOW, something like this (untested and most likely white-space-damaged).
> 
> Thanks!  I applied this manually to for-linus in hopes of making the
> the next linux-next build.
> 
> Please send any testing reports and corrections to the patch and
> commit log!
> 
> commit dd414877b58b ("PCI/PM: Prevent runtime PM until claimed by a driver that supports it")
> Author: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> Date:   Tue Nov 9 13:36:09 2021 -0600
> 
>     PCI/PM: Prevent runtime PM until claimed by a driver that supports it
>     
>     Previously we had a path that could call a driver's ->runtime_resume()
>     method before calling the driver's ->probe() method, which is a problem
>     because ->runtime_resume() often relies on initialization done in
>     ->probe():
>     
>       local_pci_probe
>         pm_runtime_get_sync
>           ...
>             pci_pm_runtime_resume
>               if (!pci_dev->driver)
>                 return 0;                          <-- early exit
>               dev->driver->pm->runtime_resume();   <-- driver ->runtime_resume()
>         pci_dev->driver = pci_drv;
>         pci_drv->probe()                           <-- driver ->probe()
>     
>     Prior to 2a4d9408c9e8 ("PCI: Use to_pci_driver() instead of
>     pci_dev->driver"), we took the early exit, which avoided the problem.  But
>     2a4d9408c9e8 removed pci_dev->driver (since it's redundant with
>     device->driver), so we no longer take the early exit, which leads to havoc
>     in ->runtime_resume().
>     
>     Similarly, we could call the driver's ->runtime_idle() method after its
>     ->remove() method.
>     
>     Avoid the problem by dropping the pm_runtime_get_sync() and
>     pm_runtime_put_sync() from local_pci_probe() and pci_device_remove(),
>     respectively.
>     
>     Add pm_runtime_get_noresume(), which uses no driver PM callbacks, to the
>     pci_pm_init() enumeration path.  This will prevent PM-runtime from touching
>     the device until it has a driver that supports PM-runtime.
>     
>     Link: https://lore.kernel.org/r/CAJZ5v0impb8uscbp8LUTBMExfMoGz=cPrTWhSGh0GF_SANNKPQ@xxxxxxxxxxxxxx
>     Fixes: 2a4d9408c9e8 ("PCI: Use to_pci_driver() instead of pci_dev->driver")
>     Reported-by: Robert Święcki <robert@xxxxxxxxxxx>
>     Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
>     Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>

I like this, this feels better than my initial suggestion using
device_is_bound().

Acked-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>

Thanks
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |

Attachment: signature.asc
Description: PGP signature