drm/i915/selftests: Mark live_forcewake_ops as unreliable

A couple of machines in the farm show quite frequent errors in the
powerwells not being released. Either there is an external agent
interferring with the powerwells, or the powerwell doesn't quite behave
as we anticipate -- either way, the test is not reliable enough to be
enabled by default in CI. It has served its immediate purpose in
providing coverage as we made tweaks to forcewake, so keep it available
for future testing.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110210
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190407192649.14750-1-chris@chris-wilson.co.uk

diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c
index ee0bc91..e0d7ebe 100644 (file)
--- a/drivers/gpu/drm/i915/selftests/intel_uncore.c
+++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c
@@ -155,6 +155,17 @@ static int live_forcewake_ops(void *arg)
                return 0;
        }
 
+       /*
+        * Not quite as reliable across the gen as one would hope.
+        *
+        * Either our theory of operation is incorrect, or there remain
+        * external parties interfering with the powerwells.
+        *
+        * https://bugs.freedesktop.org/show_bug.cgi?id=110210
+        */
+       if (!IS_ENABLED(CONFIG_DRM_I915_SELFTEST_BROKEN))
+               return 0;
+
        /* We have to pick carefully to get the exact behaviour we need */
        for (r = registers; r->name; r++)
                if (r->platforms & INTEL_INFO(i915)->gen_mask)