Commen=
t # 3
on bug 10910=
2
from Gert vd Kraats
I am using intel i915 gen 3 with 16 fence registers.
When displaying the variable total_fences
at function drm_intel_gem_check_aperture_space at
libdrm-2.4.95/intel/intel_bufmgr_gem.c,
it is seen that the deadlock-abort is occurring at total_fences =3D=3D 16,
where the previous display shows total_fences =3D=3D 14,
so there is an increment of 2 fences in stead of the usual increment of 1.
Apparently the deadlock occurs at the "freeing fences phase", tha=
t is started
if total_fences is too high.
I decided to lower the available_fences from 14 to 13 (see proposed patch).
This solved the problem.
After this patch I still saw sometimes total_fences =3D=3D 16,
where the previous display shows total_fences =3D=3D 13, so there is an inc=
rement
of 3 fences. This did not cause a deadlock at the "freeing fences phas=
e".