> -----Original Message----- > From: Rafael J. Wysocki [mailto:r...@sisk.pl] > Sent: Friday, March 01, 2013 10:22 AM > To: Liu, Chuansheng > Cc: Li, Fei; gre...@linuxfoundation.org; Lan, Tianyu; > st...@rowland.harvard.edu; sarah.a.sh...@linux.intel.com; > linux-usb@vger.kernel.org; linux-ker...@vger.kernel.org > Subject: Re: [PATCH 4/5 V2] usb: call pm_runtime_put_sync in > pm_runtime_get_sync failed case > > On Friday, March 01, 2013 02:07:54 AM Liu, Chuansheng wrote: > > > > > -----Original Message----- > > > From: Li, Fei > > > Sent: Thursday, February 28, 2013 5:06 PM > > > To: gre...@linuxfoundation.org; Lan, Tianyu; st...@rowland.harvard.edu; > > > sarah.a.sh...@linux.intel.com > > > Cc: r...@sisk.pl; linux-usb@vger.kernel.org; linux-ker...@vger.kernel.org; > Liu, > > > Chuansheng; Li, Fei > > > Subject: [PATCH 4/5 V2] usb: call pm_runtime_put_sync in > > > pm_runtime_get_sync failed case > > > > > > > > > Even in failed case of pm_runtime_get_sync, the usage_count > > > is incremented. In order to keep the usage_count with correct > > > value and runtime power management to behave correctly, call > > > pm_runtime_put(_sync) in such case. > > > > > > Signed-off-by Liu Chuansheng <chuansheng....@intel.com> > > > Signed-off-by: Li Fei <fei...@intel.com> > > > --- > > > drivers/usb/core/hub.c | 3 ++- > > > 1 files changed, 2 insertions(+), 1 deletions(-) > > > > > > diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c > > > index 5480352..f72dede 100644 > > > --- a/drivers/usb/core/hub.c > > > +++ b/drivers/usb/core/hub.c > > > @@ -3148,12 +3148,13 @@ int usb_port_resume(struct usb_device > *udev, > > > pm_message_t msg) > > > > > > if (port_dev->did_runtime_put) { > > > status = pm_runtime_get_sync(&port_dev->dev); > > > - port_dev->did_runtime_put = false; > > > if (status < 0) { > > > dev_dbg(&udev->dev, "can't resume usb port, > status %d\n", > > > status); > > > + pm_runtime_put_sync(&port_dev->dev); > > Rechecked the usb similar codes, in usb_autoresume_device() and > usb_autopm_get_interface(), > > when pm_runtime_get_sync() failed, the paired pm_runtime_put_sync() will > be called. > > Alan and Rafael, is it reasonable to consider this cleanup patch also? > > Thanks. > > You can very well use pm_runtime_put_noidle() here too. Then, it will > be kind of clear what it's for. Thanks. Your advice really express we want to do. Will update the patch soon.
> > > > > > return status; > > > } > > > + port_dev->did_runtime_put = false; > > > } > > > > > > /* Skip the initial Clear-Suspend step for a remote wakeup */ > > > -- > > > 1.7.4.1 > > Thanks, > Rafael > > > -- > I speak only for myself. > Rafael J. Wysocki, Intel Open Source Technology Center.
