Re: [Bug #12667] Badness at kernel/time/timekeeping.c:98 in pmud (timekeeping_suspended)

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Sun, 2009-02-22 at 20:31 +0100, Thomas Gleixner wrote:
> We had problems in the past where we just returned frozen time and the
> calling code got surprised when the time jumped 5 hours ahead just a
> few microseconds later.
> 
> What I find more fishy is the fact that the lid switch needs to be a
> sysdev. It's a simple input event, which causes the user space code to
> trigger the suspend sequence when the lid is shut.
> 
Well, the PMU is a sysdev for lots of good reasons.. then it -also-
happens to provide us with the lid switch event.

I'm a bit surprised by the explanation about the code that is surprised
to see the time jump. IE. Code -will- see the time jump anyway. IE. I
don't see how making gtod blow instead of returning a frozen time will
make any difference whatsoever for code who get the time some time
before suspend and then some time after resume and see a 5 hours gap...

Cheers,
Ben.