[PATCH 5.1 107/121] drm/i915: Dont clobber M/N values during fastset check

From: Greg Kroah-Hartman
Date: Mon Jun 24 2019 - 06:22:29 EST


From: Ville SyrjÃlà <ville.syrjala@xxxxxxxxxxxxxxx>

commit 475df5d0f3eb2d031e4505f84d8fba75baaf2e80 upstream.

We're now calling intel_pipe_config_compare(..., true) uncoditionally
which means we're always going clobber the calculated M/N values with
the old values if the fuzzy M/N check passes. That causes problems
because the fuzzy check allows for a huge difference in the values.

I'm actually tempted to just make the M/N checks exact, but that might
prevent fastboot from kicking in when people want it. So for now let's
overwrite the computed values with the old values only if decide to skip
the modeset.

v2: Copy has_drrs along with M/N M2/N2 values

Cc: stable@xxxxxxxxxxxxxxx
Cc: Blubberbub@xxxxxxxxxxxxxx
Cc: Maarten Lankhorst <maarten.lankhorst@xxxxxxxxxxxxxxx>
Cc: Hans de Goede <hdegoede@xxxxxxxxxx>
Tested-by: Blubberbub@xxxxxxxxxxxxxx
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110782
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110675
Fixes: d19f958db23c ("drm/i915: Enable fastset for non-boot modesets.")
Signed-off-by: Ville SyrjÃlà <ville.syrjala@xxxxxxxxxxxxxxx>
Link: https://patchwork.freedesktop.org/patch/msgid/20190612172423.25231-1-ville.syrjala@xxxxxxxxxxxxxxx
Reviewed-by: Imre Deak <imre.deak@xxxxxxxxx>
(cherry picked from commit f0521558a2a89d58a08745e225025d338572e60a)
Signed-off-by: Jani Nikula <jani.nikula@xxxxxxxxx>
Link: https://patchwork.freedesktop.org/patch/msgid/20190619120929.4057-1-ville.syrjala@xxxxxxxxxxxxxxx
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
drivers/gpu/drm/i915/intel_display.c | 38 ++++++++++++++++++++++++++---------
1 file changed, 29 insertions(+), 9 deletions(-)

--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11820,9 +11820,6 @@ intel_compare_link_m_n(const struct inte
m2_n2->gmch_m, m2_n2->gmch_n, !adjust) &&
intel_compare_m_n(m_n->link_m, m_n->link_n,
m2_n2->link_m, m2_n2->link_n, !adjust)) {
- if (adjust)
- *m2_n2 = *m_n;
-
return true;
}

@@ -12855,6 +12852,33 @@ static int calc_watermark_data(struct in
return 0;
}

+static void intel_crtc_check_fastset(struct intel_crtc_state *old_crtc_state,
+ struct intel_crtc_state *new_crtc_state)
+{
+ struct drm_i915_private *dev_priv =
+ to_i915(new_crtc_state->base.crtc->dev);
+
+ if (!intel_pipe_config_compare(dev_priv, old_crtc_state,
+ new_crtc_state, true))
+ return;
+
+ new_crtc_state->base.mode_changed = false;
+ new_crtc_state->update_pipe = true;
+
+ /*
+ * If we're not doing the full modeset we want to
+ * keep the current M/N values as they may be
+ * sufficiently different to the computed values
+ * to cause problems.
+ *
+ * FIXME: should really copy more fuzzy state here
+ */
+ new_crtc_state->fdi_m_n = old_crtc_state->fdi_m_n;
+ new_crtc_state->dp_m_n = old_crtc_state->dp_m_n;
+ new_crtc_state->dp_m2_n2 = old_crtc_state->dp_m2_n2;
+ new_crtc_state->has_drrs = old_crtc_state->has_drrs;
+}
+
/**
* intel_atomic_check - validate state object
* @dev: drm device
@@ -12903,12 +12927,8 @@ static int intel_atomic_check(struct drm
return ret;
}

- if (intel_pipe_config_compare(dev_priv,
- to_intel_crtc_state(old_crtc_state),
- pipe_config, true)) {
- crtc_state->mode_changed = false;
- pipe_config->update_pipe = true;
- }
+ intel_crtc_check_fastset(to_intel_crtc_state(old_crtc_state),
+ pipe_config);

if (needs_modeset(crtc_state))
any_ms = true;