Comme=
nt # 13
on bug 10715=
2
from Andrey Grodzovsky
(In reply to dwagner from comment #12)
> Indeed, I found my theory confirmed by many expe=
riments: If I use a script
> like
> > #!/bin/bash
> > cd /sys/class/drm/card0/device
> > echo manual >power_dpm_force_performance_level
> > # low
> > echo 0 >pp_dpm_mclk=20
> > echo 0 >pp_dpm_sclk
> > # medium
> > #echo 1 >pp_dpm_mclk=20
> > #echo 1 >pp_dpm_sclk
> > # high
> > #echo 1 >pp_dpm_mclk=20
> > #echo 6 >pp_dpm_sclk
> to enforce just any performance level, then the crashes do not occur a=
nymore
> - also with the "low frame rate video test".
>=20
> So it seems that the transition from one "dpm" performance l=
evel to another,
> with a certain probability, causes these crashes. And the more often t=
he
> transitions occur, the sooner one will experience them.
>=20
> The dynamic power management issue can now be pursued with the origina=
l bug
> report https://bugs.freedesktop.org/show_bug.=
cgi?id=3D102322 for the
> vm_update_mode=3D0 case - there is probably not much sense in keeping =
this bug
> report open just because errors also occur with wm_update_mode=3D3, ju=
st less
> often.
Agreed.