Re: Totally broken PCI PM calls

[David Brownell <david-b@pacbell.net>]

On Sunday 10 October 2004 9:24 pm, Paul Mackerras wrote:
> Linus Torvalds writes:
> 
> > And they are unbroken again (well, at least they work for me again).  
> > Partly by the PM_ renumbering under discussion.
> 
> Interesting.  I find that with suspend-to-ram, USB keyboards don't
> work after resume, and that the system will hang on resume if you
> remove a USB device during sleep.

A "hang" sounds like the pmcore bug I reported about a year ago...

It's rather foolish of the PM core to use the same semaphore to
protect system-wide suspend/resume operations that it uses to
for mutual exclusion on the device add/remove (which suspend
and resume callbacks did happily in 2.4) ... since it's routine to
unplug peripherals on suspended systems!

Alternatively, if you're combininging USB_SUSPEND with any
system-wide suspend operation, you're asking for trouble;
the PM core is just not ready for that.  (In fact I've wondered
if maybe 2.6.9 shouldn't discourage that combination more
actively...)

But a keyboard-specific issue might be improved with the
HID patch I posted last week, teaching that driver how to
handle suspend() and resume() callbacks.

- Dave