Bug 107704

Note debugfs-read is bug 105588. The module-reload spam was suppressed by 

commit e0c3033a57d85c0d2eb33af0451afa16edc79f10 (upstream/master, origin/master, origin/HEAD)
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Aug 17 11:56:38 2018 +0100

    igt/pm_rpm: Reload the module with full mmio debugging
    
    Our unclaimed mmio access debugging is lazy, doing cheap checks
    periodically and only if they fail do a full check around every mmio
    access. When testing for runtime pm, enable the full mmio debugging from
    the initial load.
    
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
    Cc: Imre Deak <imre.deak@intel.com>
    Reviewed-by: Imre Deak <imre.deak@intel.com>

in an attempt to find the culprit, we demonstrated it wasn't us!

(In reply to Chris Wilson from comment #1)
> Note debugfs-read is bug 105588. 

OK, moved BYT out of this bug. Thanks!

> The module-reload spam was suppressed by 
> 
> commit e0c3033a57d85c0d2eb33af0451afa16edc79f10 (upstream/master,
> origin/master, origin/HEAD)
> Author: Chris Wilson <chris@chris-wilson.co.uk>
> Date:   Fri Aug 17 11:56:38 2018 +0100
> 
>     igt/pm_rpm: Reload the module with full mmio debugging
>     
>     Our unclaimed mmio access debugging is lazy, doing cheap checks
>     periodically and only if they fail do a full check around every mmio
>     access. When testing for runtime pm, enable the full mmio debugging from
>     the initial load.
>     
>     Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>     Cc: Imre Deak <imre.deak@intel.com>
>     Reviewed-by: Imre Deak <imre.deak@intel.com>
> 
> in an attempt to find the culprit, we demonstrated it wasn't us!

This commit did not fix all the issues:

 - https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4767/fi-bsw-cyan/igt@pm_rpm@module-reload.html

 - https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4767/fi-bsw-kefka/igt@pm_rpm@module-reload.html

- https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4767/fi-bsw-n3050/igt@pm_rpm@module-reload.html

[  466.740822] [drm:intel_runtime_suspend [i915]] *ERROR* Unclaimed access detected prior to suspending

Re-opening.

It did "fix" all the ones in BAT, afaict. DRM_4767 is the run before this landed in IGT_4627,

(In reply to Chris Wilson from comment #3)
> It did "fix" all the ones in BAT, afaict. DRM_4767 is the run before this
> landed in IGT_4627,

The date was very misleading then ;) I guess it seated on the ML for waaayyy too long. Sorry for the noise.