PCI / PM: Block races between runtime PM and system sleep
authorRafael J. Wysocki <rjw@sisk.pl>
Tue, 21 Jun 2011 21:47:15 +0000 (23:47 +0200)
committerRafael J. Wysocki <rjw@sisk.pl>
Tue, 21 Jun 2011 21:47:15 +0000 (23:47 +0200)
commita5f76d5eba157bf637beb2dd18026db2917c512e
tree0f23e14459b8324e965c6a82d8b74dde1b4018e4
parentca9c6890b598997165a7c85c001f382c910f12b0
PCI / PM: Block races between runtime PM and system sleep

After commit e8665002477f0278f84f898145b1f141ba26ee26
(PM: Allow pm_runtime_suspend() to succeed during system suspend) it
is possible that a device resumed by the pm_runtime_resume(dev) in
pci_pm_prepare() will be suspended immediately from a work item,
timer function or otherwise, defeating the very purpose of calling
pm_runtime_resume(dev) from there.  To prevent that from happening
it is necessary to increment the runtime PM usage counter of the
device by replacing pm_runtime_resume() with pm_runtime_get_sync().
Moreover, the incremented runtime PM usage counter has to be
decremented by the corresponding pci_pm_complete(), via
pm_runtime_put_sync().

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: stable@kernel.org
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
drivers/pci/pci-driver.c