Hi! > > Well, imagine battery-backed usb display. You might want to write "I'm > > suspended to RAM" on the display when machine goes suspend-to-RAM. > > That could always be done by a userspace program as part of the overall > system sleep procedure. In fact, this would be more appropriate than > having a device driver do it... Yes, it was only an example ;-). > > If the driver wants to implement suspend as disconnect... well, why > > not. But for example usb disk, you probably want it to be left mounted > > during suspend but unplug should probably trigger emergency unmount. > > Yes, certainly. David's original point was that if VBUS power isn't > available during system sleep, then sleep is indistinguishable from > disconnect as far as USB is concerned. So a sleep transition would > already be enough to trigger an emergency unmount. In practice, of > course, the hotplug scripts should be set up so that they unmount such > drives cleanly at the start of the sleep transition. I do not think you can umount for sleep... what if you have files open on it? It may be true that electrically disconnect is same as sleep for USB, but you want to umount the disk in disconnect case and park it but leave it mounted for the sleep case. ...so leaving flexibility to the drivers seems like good thing to me. Pavel -- People were complaining that M$ turns users into beta-testers... ...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!
