Re: [PATCH v2] PCI / PM: Allow runtime PM without callback functions

On Tue, 23 Oct 2018 14:45:52 +0300, Jarkko Nikula wrote:

Commit a9c8088c7988 ("i2c: i801: Don't restore config registers on runtime PM") nullified the runtime PM suspend/resume callback pointers while keeping the runtime PM enabled.

This causes that SMBus PCI device stays in D0 power state and sysfs /sys/bus/pci/devices/[SMBus PCI ID]/power/runtime_status shows "error" when the runtime PM framework attempts to autosuspend the device. This is due PCI bus runtime PM which checks for driver runtime PM callbacks and returns with -ENOSYS if they are not set.

Since i2c-i801.c don't need to do anything device specific beyond PCI device power state management Jean Delvare proposed if this can be fixed in the PCI subsystem core level rather than adding dummy runtime PM callback functions in the PCI drivers.

Change the pci_pm_runtime_suspend()/pci_pm_runtime_resume() semantics so that they allow change the PCI device power state during runtime PM transitions even if no runtime PM callback functions are defined.

This change fixes the runtime PM regression on i2c-i801.c.

It is not obvious why the code had hard requirements for the runtime PM callbacks. Test has been here since the code was introduced by the commit 6cbf82148ff2 ("PCI PM: Run-time callbacks for PCI bus type").

On the other hand similar change than this was done to generic runtime PM callbacks way back in the commit 05aa55dddb9e ("PM / Runtime: Lenient generic runtime pm callbacks").

Fixes: a9c8088c7988 ("i2c: i801: Don't restore config registers on runtime PM") Reported-by: Mika Westerberg mika.westerberg@linux.intel.com Cc: stable@vger.kernel.org # 4.18+ Signed-off-by: Jarkko Nikula jarkko.nikula@linux.intel.com

---

I Cc'ed stable since this fixes the regression on i2c-i801.c. But we probably want to get some test coverage first before applying into stable. Queueing for v4.20 sounds reasonable to me. v2: Previous version had a potential NULL dereference in WARN_ONCE() statement noted by Jean Delvare. Now covered by pm && pm->runtime_suspend test. Also handling of error code from pm->runtime_suspend() moved under the same code block where callback is called. v1: This is related to my i2c-i801.c fix thread back in June which I completely forgot till now: https://lkml.org/lkml/2018/6/27/642 Discussion back then was that it should be handled in the PCI PM instead of having dummy functions in the drivers. I wanted to respin with a patch.

---

drivers/pci/pci-driver.c | 27 ++++++++++++--------------- 1 file changed, 12 insertions(+), 15 deletions(-)

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index bef17c3fca67..33f3f475e5c6 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -1251,30 +1251,29 @@ static int pci_pm_runtime_suspend(struct device *dev) return 0; }

• if (!pm || !pm->runtime_suspend)

• return -ENOSYS;
  

• pci_dev->state_saved = false;
• error = pm->runtime_suspend(dev);
• if (error) {

• if (pm && pm->runtime_suspend) {

• error = pm->runtime_suspend(dev);
  

  /*
  • -EBUSY and -EAGAIN is used to request the runtime PM core
  • to schedule a new suspend, so log the event only with debug
  • log level.
  */

• if (error == -EBUSY || error == -EAGAIN)
  

• if (error == -EBUSY || error == -EAGAIN) {
  dev_dbg(dev, "can't suspend now (%pf returned %d)\n",
  	pm->runtime_suspend, error);
  

• else
  

• 	return error;
  

• } else if (error) {
  dev_err(dev, "can't suspend (%pf returned %d)\n",
  	pm->runtime_suspend, error);
  

• return error;
  

• 	return error;
  

• }
  

  }

pci_fixup_device(pci_fixup_suspend, pci_dev);

• if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0

• if (pm && pm->runtime_suspend

•    && !pci_dev->state_saved && pci_dev->current_state != PCI_D0
   && pci_dev->current_state != PCI_UNKNOWN) {
  

  WARN_ONCE(pci_dev->current_state != prev, "PCI PM: State of device not saved by %pF\n",

@@ -1292,7 +1291,7 @@ static int pci_pm_runtime_suspend(struct device *dev) static int pci_pm_runtime_resume(struct device *dev) {

• int rc;

• int rc = 0; struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;

@@ -1306,14 +1305,12 @@ static int pci_pm_runtime_resume(struct device *dev) if (!pci_dev->driver) return 0;

• if (!pm || !pm->runtime_resume)

• return -ENOSYS;
  

• pci_fixup_device(pci_fixup_resume_early, pci_dev); pci_enable_wake(pci_dev, PCI_D0, false); pci_fixup_device(pci_fixup_resume, pci_dev);

• rc = pm->runtime_resume(dev);

• if (pm && pm->runtime_resume)

• rc = pm->runtime_resume(dev);
  

pci_dev->runtime_d3cold = false;

Looks good to me.

Reviewed-by: Jean Delvare jdelvare@suse.de