[PATCH 5.9 016/133] drm/i915: Mark ininitial fb obj as WT on eLLC machines to avoid rcu lockup during fbdev init

From: Greg Kroah-Hartman
Date: Mon Nov 09 2020 - 08:16:16 EST


From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>

commit 1664ffee760a5d98952318fdd9b198fae396d660 upstream.

Currently we leave the cache_level of the initial fb obj
set to NONE. This means on eLLC machines the first pin_to_display()
will try to switch it to WT which requires a vma unbind+bind.
If that happens during the fbdev initialization rcu does not
seem operational which causes the unbind to get stuck. To
most appearances this looks like a dead machine on boot.

Avoid the unbind by already marking the object cache_level
as WT when creating it. We still do an excplicit ggtt pin
which will rewrite the PTEs anyway, so they will match whatever
cache level we set.

Cc: <stable@xxxxxxxxxxxxxxx> # v5.7+
Suggested-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2381
Signed-off-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Link: https://patchwork.freedesktop.org/patch/msgid/20201007120329.17076-1-ville.syrjala@xxxxxxxxxxxxxxx
Link: https://patchwork.freedesktop.org/patch/msgid/20201015122138.30161-1-chris@xxxxxxxxxxxxxxxxxx
(cherry picked from commit d46b60a2e8d246f1f0faa38e52f4f5a73858c338)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
drivers/gpu/drm/i915/display/intel_display.c | 8 ++++++++
1 file changed, 8 insertions(+)

--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -3432,6 +3432,14 @@ initial_plane_vma(struct drm_i915_privat
if (IS_ERR(obj))
return NULL;

+ /*
+ * Mark it WT ahead of time to avoid changing the
+ * cache_level during fbdev initialization. The
+ * unbind there would get stuck waiting for rcu.
+ */
+ i915_gem_object_set_cache_coherency(obj, HAS_WT(i915) ?
+ I915_CACHE_WT : I915_CACHE_NONE);
+
switch (plane_config->tiling) {
case I915_TILING_NONE:
break;