The patch below does not apply to the 3.18-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@xxxxxxxxxxxxxxx>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 7f1241ed1a06b4846ad7a2a57eb088b757e58e16 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@xxxxxxxxxxxxxxx>
Date: Thu, 18 Dec 2014 11:44:06 +0200
Subject: [PATCH] drm/i915: Kill check_power_well() calls
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
pps_{lock,unlock}() call intel_display_power_{get,put}() outside
pps_mutes to avoid deadlocks with the power_domain mutex. In theory
during aux transfers we should usually have the relevant power domain
references already held by some higher level code, so this should not
result in much overhead (exception being userspace i2c-dev access).
However thanks to the check_power_well() calls in
intel_display_power_{get/put}() we end up doing a few Punit reads for
each aux transfer. Obviously doing this for each byte transferred via
i2c-over-aux is not a good idea.
I can't think of a good way to keep check_power_well() while eliminating
the overhead, so let's just remove check_power_well() entirely.
Fixes a driver init time regression introduced by:
commit 773538e86081d146e0020435d614f4b96996c1f9
Author: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
Date: Thu Sep 4 14:54:56 2014 +0300
drm/i915: Reset power sequencer pipe tracking when disp2d is off
Credit goes to Jani for figuring this out.
v2: Add the regression note in the commit message.
Cc: stable@xxxxxxxxxxxxxxx (v3.18+)
Cc: Egbert Eich <eich@xxxxxxx>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86201
Tested-by: Wendy Wang <wendy.wang@xxxxxxxxx>
Signed-off-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
Signed-off-by: Jani Nikula <jani.nikula@xxxxxxxxx>
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index f5a78d53e297..ac6da7102fbb 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -615,29 +615,6 @@ static void chv_pipe_power_well_disable(struct drm_i915_private *dev_priv,
vlv_power_sequencer_reset(dev_priv);
}
-static void check_power_well_state(struct drm_i915_private *dev_priv,
- struct i915_power_well *power_well)
-{
- bool enabled = power_well->ops->is_enabled(dev_priv, power_well);
-
- if (power_well->always_on || !i915.disable_power_well) {
- if (!enabled)
- goto mismatch;
-
- return;
- }
-
- if (enabled != (power_well->count > 0))
- goto mismatch;
-
- return;
-
-mismatch:
- WARN(1, "state mismatch for '%s' (always_on %d hw state %d use-count %d disable_power_well %d\n",
- power_well->name, power_well->always_on, enabled,
- power_well->count, i915.disable_power_well);
-}
-
/**
* intel_display_power_get - grab a power domain reference
* @dev_priv: i915 device instance
@@ -669,8 +646,6 @@ void intel_display_power_get(struct drm_i915_private *dev_priv,
power_well->ops->enable(dev_priv, power_well);
power_well->hw_enabled = true;
}
-
- check_power_well_state(dev_priv, power_well);
}
power_domains->domain_use_count[domain]++;
@@ -709,8 +684,6 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
power_well->hw_enabled = false;
power_well->ops->disable(dev_priv, power_well);
}
-
- check_power_well_state(dev_priv, power_well);
}
mutex_unlock(&power_domains->lock);
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html