On Thu, 9 Jan 2014, Phillip Susi wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 1/9/2014 11:14 AM, Alan Stern wrote: > > That's true, but it isn't a problem. We know that requests with > > REQ_PM are sent only at certain, controlled times. In particular, > > the only time such a request would be sent while the disk is > > RPM_SUSPENDED is during a system resume. > > Yes, but if the disk and port were both already RPM_SUSPENDED when the > system was suspended, and so the port is still RPM_SUSPENDED when the > disk's system resume method is called, then it can't communicate with > the disk. The port should not be runtime suspended during system resume. It should always go to runtime active. > >> We need to put the device into one of the transitioning states > >> to block other IO, without allowing the port to suspend. > > > > No, we only need to make sure that the port doesn't go into runtime > > suspend while we carry out the "is the disk spinning" test. > > Well how else do you do that other than with the transition states? pm_runtime_get_noresume(dev->parent). Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
