From: Rafael J. Wysocki <[email protected]>
Updating the current_state field of struct pci_dev the way it is done
in pci_enable_device_flags() before calling do_pci_enable_device() may
not work. For example, if the given PCI device depends on an ACPI
power resource whose _STA method initially returns 0 ("off"), but the
config space of the PCI device is accessible and the power state
retrieved from the PCI_PM_CTRL register is D0, the current_state
field in the struct pci_dev representing that device will get out of
sync with the power.state of its ACPI companion object and that will
lead to power management issues going forward.
To avoid such issues, make pci_enable_device_flags() call
pci_update_current_state() which takes ACPI device power management
into account, if present, to retrieve the current power state of the
device.
Link: https://lore.kernel.org/lkml/[email protected]/
Reported-by: Maximilian Luz <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
---
Hi Maximilian,
Because commit 4514d991d992 ("PCI: PM: Do not read power state in
pci_enable_device_flags()"), the issue addressed by it is back, so
we need an alternative way to address it.
Can you please check if this patch makes that issue go away?
Thanks!
---
drivers/pci/pci.c | 6 +-----
1 file changed, 1 insertion(+), 5 deletions(-)
Index: linux-pm/drivers/pci/pci.c
===================================================================
--- linux-pm.orig/drivers/pci/pci.c
+++ linux-pm/drivers/pci/pci.c
@@ -1906,11 +1906,7 @@ static int pci_enable_device_flags(struc
* so that things like MSI message writing will behave as expected
* (e.g. if the device really is in D0 at enable time).
*/
- if (dev->pm_cap) {
- u16 pmcsr;
- pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
- dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
- }
+ pci_update_current_state(dev, dev->current_state);
if (atomic_inc_return(&dev->enable_cnt) > 1)
return 0; /* already enabled */
On 7/8/21 3:25 PM, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <[email protected]>
>
> Updating the current_state field of struct pci_dev the way it is done
> in pci_enable_device_flags() before calling do_pci_enable_device() may
> not work. For example, if the given PCI device depends on an ACPI
> power resource whose _STA method initially returns 0 ("off"), but the
> config space of the PCI device is accessible and the power state
> retrieved from the PCI_PM_CTRL register is D0, the current_state
> field in the struct pci_dev representing that device will get out of
> sync with the power.state of its ACPI companion object and that will
> lead to power management issues going forward.
>
> To avoid such issues, make pci_enable_device_flags() call
> pci_update_current_state() which takes ACPI device power management
> into account, if present, to retrieve the current power state of the
> device.
>
> Link: https://lore.kernel.org/lkml/[email protected]/
> Reported-by: Maximilian Luz <[email protected]>
> Signed-off-by: Rafael J. Wysocki <[email protected]>
> ---
>
> Hi Maximilian,
>
> Because commit 4514d991d992 ("PCI: PM: Do not read power state in
> pci_enable_device_flags()"), the issue addressed by it is back, so
> we need an alternative way to address it.
>
> Can you please check if this patch makes that issue go away?
Hi,
just tested this on v5.13 and it works, thanks! Feel free to add
Tested-by: Maximilian Luz <[email protected]>
Regards,
Max
On Thu, Jul 8, 2021 at 10:34 PM Maximilian Luz <[email protected]> wrote:
>
> On 7/8/21 3:25 PM, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <[email protected]>
> >
> > Updating the current_state field of struct pci_dev the way it is done
> > in pci_enable_device_flags() before calling do_pci_enable_device() may
> > not work. For example, if the given PCI device depends on an ACPI
> > power resource whose _STA method initially returns 0 ("off"), but the
> > config space of the PCI device is accessible and the power state
> > retrieved from the PCI_PM_CTRL register is D0, the current_state
> > field in the struct pci_dev representing that device will get out of
> > sync with the power.state of its ACPI companion object and that will
> > lead to power management issues going forward.
> >
> > To avoid such issues, make pci_enable_device_flags() call
> > pci_update_current_state() which takes ACPI device power management
> > into account, if present, to retrieve the current power state of the
> > device.
> >
> > Link: https://lore.kernel.org/lkml/[email protected]/
> > Reported-by: Maximilian Luz <[email protected]>
> > Signed-off-by: Rafael J. Wysocki <[email protected]>
> > ---
> >
> > Hi Maximilian,
> >
> > Because commit 4514d991d992 ("PCI: PM: Do not read power state in
> > pci_enable_device_flags()"), the issue addressed by it is back, so
> > we need an alternative way to address it.
> >
> > Can you please check if this patch makes that issue go away?
>
> Hi,
>
> just tested this on v5.13 and it works, thanks! Feel free to add
>
> Tested-by: Maximilian Luz <[email protected]>
Thank you!
On Mon, Jul 12, 2021 at 5:30 PM Rafael J. Wysocki <[email protected]> wrote:
>
> On Fri, Jul 9, 2021 at 1:21 PM Rafael J. Wysocki <[email protected]> wrote:
> >
> > On Thu, Jul 8, 2021 at 10:34 PM Maximilian Luz <[email protected]> wrote:
> > >
> > > On 7/8/21 3:25 PM, Rafael J. Wysocki wrote:
> > > > From: Rafael J. Wysocki <[email protected]>
> > > >
> > > > Updating the current_state field of struct pci_dev the way it is done
> > > > in pci_enable_device_flags() before calling do_pci_enable_device() may
> > > > not work. For example, if the given PCI device depends on an ACPI
> > > > power resource whose _STA method initially returns 0 ("off"), but the
> > > > config space of the PCI device is accessible and the power state
> > > > retrieved from the PCI_PM_CTRL register is D0, the current_state
> > > > field in the struct pci_dev representing that device will get out of
> > > > sync with the power.state of its ACPI companion object and that will
> > > > lead to power management issues going forward.
> > > >
> > > > To avoid such issues, make pci_enable_device_flags() call
> > > > pci_update_current_state() which takes ACPI device power management
> > > > into account, if present, to retrieve the current power state of the
> > > > device.
> > > >
> > > > Link: https://lore.kernel.org/lkml/[email protected]/
> > > > Reported-by: Maximilian Luz <[email protected]>
> > > > Signed-off-by: Rafael J. Wysocki <[email protected]>
> > > > ---
> > > >
> > > > Hi Maximilian,
> > > >
> > > > Because commit 4514d991d992 ("PCI: PM: Do not read power state in
> > > > pci_enable_device_flags()"), the issue addressed by it is back, so
> > > > we need an alternative way to address it.
> > > >
> > > > Can you please check if this patch makes that issue go away?
> > >
> > > Hi,
> > >
> > > just tested this on v5.13 and it works, thanks! Feel free to add
> > >
> > > Tested-by: Maximilian Luz <[email protected]>
Gmail decided to drop Salvatore's address from the To list for me,
sorry about that.
> Michael, Salvatore, since this alternative to commit 4514d991d992
> works for Maximilian, I'm going to queue it up for 5.15 (barring any
> objections from Bjorn), so I would appreciate it if you could test it
> and let me know whether or not there were any problems with it.
>
> The patch can be downloaded from
> https://patchwork.kernel.org/project/linux-pm/patch/4327888.LvFx2qVVIh@kreacher
>
> Thanks!
On Fri, Jul 9, 2021 at 1:21 PM Rafael J. Wysocki <[email protected]> wrote:
>
> On Thu, Jul 8, 2021 at 10:34 PM Maximilian Luz <[email protected]> wrote:
> >
> > On 7/8/21 3:25 PM, Rafael J. Wysocki wrote:
> > > From: Rafael J. Wysocki <[email protected]>
> > >
> > > Updating the current_state field of struct pci_dev the way it is done
> > > in pci_enable_device_flags() before calling do_pci_enable_device() may
> > > not work. For example, if the given PCI device depends on an ACPI
> > > power resource whose _STA method initially returns 0 ("off"), but the
> > > config space of the PCI device is accessible and the power state
> > > retrieved from the PCI_PM_CTRL register is D0, the current_state
> > > field in the struct pci_dev representing that device will get out of
> > > sync with the power.state of its ACPI companion object and that will
> > > lead to power management issues going forward.
> > >
> > > To avoid such issues, make pci_enable_device_flags() call
> > > pci_update_current_state() which takes ACPI device power management
> > > into account, if present, to retrieve the current power state of the
> > > device.
> > >
> > > Link: https://lore.kernel.org/lkml/[email protected]/
> > > Reported-by: Maximilian Luz <[email protected]>
> > > Signed-off-by: Rafael J. Wysocki <[email protected]>
> > > ---
> > >
> > > Hi Maximilian,
> > >
> > > Because commit 4514d991d992 ("PCI: PM: Do not read power state in
> > > pci_enable_device_flags()"), the issue addressed by it is back, so
> > > we need an alternative way to address it.
> > >
> > > Can you please check if this patch makes that issue go away?
> >
> > Hi,
> >
> > just tested this on v5.13 and it works, thanks! Feel free to add
> >
> > Tested-by: Maximilian Luz <[email protected]>
Michael, Salvatore, since this alternative to commit 4514d991d992
works for Maximilian, I'm going to queue it up for 5.15 (barring any
objections from Bjorn), so I would appreciate it if you could test it
and let me know whether or not there were any problems with it.
The patch can be downloaded from
https://patchwork.kernel.org/project/linux-pm/patch/4327888.LvFx2qVVIh@kreacher
Thanks!