On Thu, Jul 7, 2022 at 8:27 PM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
>
> On Thu, Jul 7, 2022 at 7:22 PM Rajvi Jingar
> <rajvi.jingar@xxxxxxxxxxxxxxx> wrote:
> >
> > Set ptm_enabled from 'struct pci_dev' to 0 in pci_ptm_disable() to
>
> I would set "Clear" here instead of "Set", but moreover ->
s/set/say/ sorry.
> > save the disabled PTM state for the device.
> >
> > Fixes: a697f072f5da ("PCI: Disable PTM during suspend to save power")
> > Signed-off-by: Rajvi Jingar <rajvi.jingar@xxxxxxxxxxxxxxx>
> > ---
> > drivers/pci/pcie/ptm.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/pci/pcie/ptm.c b/drivers/pci/pcie/ptm.c
> > index 368a254e3124..746e29779c27 100644
> > --- a/drivers/pci/pcie/ptm.c
> > +++ b/drivers/pci/pcie/ptm.c
> > @@ -44,6 +44,7 @@ void pci_disable_ptm(struct pci_dev *dev)
> > pci_read_config_word(dev, ptm + PCI_PTM_CTRL, &ctrl);
> > ctrl &= ~(PCI_PTM_CTRL_ENABLE | PCI_PTM_CTRL_ROOT);
> > pci_write_config_word(dev, ptm + PCI_PTM_CTRL, ctrl);
> > + dev->ptm_enabled = 0;
>
> -> I don't think that this is going to work without setting
> dev->ptm_enabled in pci_restore_ptm_state() if the restored state
> turns out to be "enabled".
And arguably, this function doesn't need to do anything if
dev->ptm_enabled is 0 already and this is the only check it needs to
do, because dev->ptm_enabled will only be set for PCIe devices that
have PTM.
> > }
> >
> > void pci_save_ptm_state(struct pci_dev *dev)
> > --
