[Intel-gfx] Hibernation test

[Imre Deak]

On ke, 2015-02-11 at 16:46 +0200, David Weinehall wrote:
> intel-gpu-tools currently has a bunch of tests for suspend,
> but currently none (that I could find) for hibernate.
> 
> Attached is a rudimentary patch to add said test.  It does so
> by repurposing the drv_suspend driver to handle both suspend
> and hibernate, since the difference is miniscule.
> 
> I decided to split the suspend/autoresume functions in
> igt_aux.c though, to be able to leave the igt_system_uspend_autoresume()
> function unchanged (the other option would be to
> introduce a boolean function argument and have that
> decide what parameters to pass to rtcwake).
> 
> The timeout passed to rtcwake probably needs tuning (it might
> even need to be dynamically adjusted, since the time hibernation takes
> varies wildly depending on the amount of non-cache memory in use).

I think in general we should try to keep the existing subtest names, but
I couldn't find any concrete problem changing them in this case. The
patch looks ok to me (please inline the patch next time):
Reviewed-by: Imre Deak <imre.deak at intel.com>