Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "Intel and radeon fixes.

  Post KS/LC git requests from i915 and radeon stacked up.  They are all
  fixes along with some new pci ids for radeon, and one maintainers file
  entry.

   - i915: display fixes and irq fixes
   - radeon: pci ids, and misc gpuvm, dpm and hdp cache"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (29 commits)
  MAINTAINERS: Add entry for Renesas DRM drivers
  drm/radeon: add additional SI pci ids
  drm/radeon: add new bonaire pci ids
  drm/radeon: add new KV pci id
  Revert "drm/radeon: Use write-combined CPU mappings of ring buffers with PCIe"
  drm/radeon: fix active_cu mask on SI and CIK after re-init (v3)
  drm/radeon: fix active cu count for SI and CIK
  drm/radeon: re-enable selective GPUVM flushing
  drm/radeon: Sync ME and PFP after CP semaphore waits v4
  drm/radeon: fix display handling in radeon_gpu_reset
  drm/radeon: fix pm handling in radeon_gpu_reset
  drm/radeon: Only flush HDP cache for indirect buffers from userspace
  drm/radeon: properly document reloc priority mask
  drm/i915: don't try to retrain a DP link on an inactive CRTC
  drm/i915: make sure VDD is turned off during system suspend
  drm/i915: cancel hotplug and dig_port work during suspend and unload
  drm/i915: fix HPD IRQ reenable work cancelation
  drm/i915: take display port power domain in DP HPD handler
  drm/i915: Don't try to enable cursor from setplane when crtc is disabled
  drm/i915: Skip load detect when intel_crtc->new_enable==true
  ...

41 files changed, 309 insertions, 167 deletions

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index ec96f9a9724c..e27cdbe9d524 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -494,6 +494,36 @@ bool i915_semaphore_is_enabled(struct drm_device *dev)
 	return true;
 }
 
+void intel_hpd_cancel_work(struct drm_i915_private *dev_priv)
+{
+	spin_lock_irq(&dev_priv->irq_lock);
+
+	dev_priv->long_hpd_port_mask = 0;
+	dev_priv->short_hpd_port_mask = 0;
+	dev_priv->hpd_event_bits = 0;
+
+	spin_unlock_irq(&dev_priv->irq_lock);
+
+	cancel_work_sync(&dev_priv->dig_port_work);
+	cancel_work_sync(&dev_priv->hotplug_work);
+	cancel_delayed_work_sync(&dev_priv->hotplug_reenable_work);
+}
+
+static void intel_suspend_encoders(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct drm_encoder *encoder;
+
+	drm_modeset_lock_all(dev);
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
+		struct intel_encoder *intel_encoder = to_intel_encoder(encoder);
+
+		if (intel_encoder->suspend)
+			intel_encoder->suspend(intel_encoder);
+	}
+	drm_modeset_unlock_all(dev);
+}
+
 static int i915_drm_freeze(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -538,6 +568,9 @@ static int i915_drm_freeze(struct drm_device *dev)
 		flush_delayed_work(&dev_priv->rps.delayed_resume_work);
 
 		intel_runtime_pm_disable_interrupts(dev);
+		intel_hpd_cancel_work(dev_priv);
+
+		intel_suspend_encoders(dev_priv);
 
 		intel_suspend_gt_powersave(dev);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4412f6a4383b..7a830eac5ba3 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1458,7 +1458,7 @@ struct drm_i915_private {
 		} hpd_mark;
 	} hpd_stats[HPD_NUM_PINS];
 	u32 hpd_event_bits;
-	struct timer_list hotplug_reenable_timer;
+	struct delayed_work hotplug_reenable_work;
 
 	struct i915_fbc fbc;
 	struct i915_drrs drrs;
@@ -2178,6 +2178,7 @@ extern unsigned long i915_mch_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_gfx_val(struct drm_i915_private *dev_priv);
 extern void i915_update_gfx_val(struct drm_i915_private *dev_priv);
 int vlv_force_gfx_clock(struct drm_i915_private *dev_priv, bool on);
+void intel_hpd_cancel_work(struct drm_i915_private *dev_priv);
 
 extern void intel_console_resume(struct work_struct *work);
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 390ccc2a3096..0050ee9470f1 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1189,8 +1189,8 @@ static void i915_hotplug_work_func(struct work_struct *work)
 	  * some connectors */
 	if (hpd_disabled) {
 		drm_kms_helper_poll_enable(dev);
-		mod_timer(&dev_priv->hotplug_reenable_timer,
-			  jiffies + msecs_to_jiffies(I915_REENABLE_HOTPLUG_DELAY));
+		mod_delayed_work(system_wq, &dev_priv->hotplug_reenable_work,
+				 msecs_to_jiffies(I915_REENABLE_HOTPLUG_DELAY));
 	}
 
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
@@ -1213,11 +1213,6 @@ static void i915_hotplug_work_func(struct work_struct *work)
 		drm_kms_helper_hotplug_event(dev);
 }
 
-static void intel_hpd_irq_uninstall(struct drm_i915_private *dev_priv)
-{
-	del_timer_sync(&dev_priv->hotplug_reenable_timer);
-}
-
 static void ironlake_rps_change_irq_handler(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3892,8 +3887,6 @@ static void gen8_irq_uninstall(struct drm_device *dev)
 	if (!dev_priv)
 		return;
 
-	intel_hpd_irq_uninstall(dev_priv);
-
 	gen8_irq_reset(dev);
 }
 
@@ -3908,8 +3901,6 @@ static void valleyview_irq_uninstall(struct drm_device *dev)
 
 	I915_WRITE(VLV_MASTER_IER, 0);
 
-	intel_hpd_irq_uninstall(dev_priv);
-
 	for_each_pipe(pipe)
 		I915_WRITE(PIPESTAT(pipe), 0xffff);
 
@@ -3988,8 +3979,6 @@ static void ironlake_irq_uninstall(struct drm_device *dev)
 	if (!dev_priv)
 		return;
 
-	intel_hpd_irq_uninstall(dev_priv);
-
 	ironlake_irq_reset(dev);
 }
 
@@ -4360,8 +4349,6 @@ static void i915_irq_uninstall(struct drm_device * dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
 
-	intel_hpd_irq_uninstall(dev_priv);
-
 	if (I915_HAS_HOTPLUG(dev)) {
 		I915_WRITE(PORT_HOTPLUG_EN, 0);
 		I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
@@ -4598,8 +4585,6 @@ static void i965_irq_uninstall(struct drm_device * dev)
 	if (!dev_priv)
 		return;
 
-	intel_hpd_irq_uninstall(dev_priv);
-
 	I915_WRITE(PORT_HOTPLUG_EN, 0);
 	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 
@@ -4615,14 +4600,18 @@ static void i965_irq_uninstall(struct drm_device * dev)
 	I915_WRITE(IIR, I915_READ(IIR));
 }
 
-static void intel_hpd_irq_reenable(unsigned long data)
+static void intel_hpd_irq_reenable(struct work_struct *work)
 {
-	struct drm_i915_private *dev_priv = (struct drm_i915_private *)data;
+	struct drm_i915_private *dev_priv =
+		container_of(work, typeof(*dev_priv),
+			     hotplug_reenable_work.work);
 	struct drm_device *dev = dev_priv->dev;
 	struct drm_mode_config *mode_config = &dev->mode_config;
 	unsigned long irqflags;
 	int i;
 
+	intel_runtime_pm_get(dev_priv);
+
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
 	for (i = (HPD_NONE + 1); i < HPD_NUM_PINS; i++) {
 		struct drm_connector *connector;
@@ -4648,6 +4637,8 @@ static void intel_hpd_irq_reenable(unsigned long data)
 	if (dev_priv->display.hpd_irq_setup)
 		dev_priv->display.hpd_irq_setup(dev);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
+	intel_runtime_pm_put(dev_priv);
 }
 
 void intel_irq_init(struct drm_device *dev)
@@ -4670,8 +4661,8 @@ void intel_irq_init(struct drm_device *dev)
 	setup_timer(&dev_priv->gpu_error.hangcheck_timer,
 		    i915_hangcheck_elapsed,
 		    (unsigned long) dev);
-	setup_timer(&dev_priv->hotplug_reenable_timer, intel_hpd_irq_reenable,
-		    (unsigned long) dev_priv);
+	INIT_DELAYED_WORK(&dev_priv->hotplug_reenable_work,
+			  intel_hpd_irq_reenable);
 
 	pm_qos_add_request(&dev_priv->pm_qos, PM_QOS_CPU_DMA_LATENCY, PM_QOS_DEFAULT_VALUE);
 
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 2efaf8e8d9c4..e8abfce40976 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -699,16 +699,21 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 		goto out;
 	}
 
+	drm_modeset_acquire_init(&ctx, 0);
+
 	/* for pre-945g platforms use load detect */
 	if (intel_get_load_detect_pipe(connector, NULL, &tmp, &ctx)) {
 		if (intel_crt_detect_ddc(connector))
 			status = connector_status_connected;
 		else
 			status = intel_crt_load_detect(crt);
-		intel_release_load_detect_pipe(connector, &tmp, &ctx);
+		intel_release_load_detect_pipe(connector, &tmp);
 	} else
 		status = connector_status_unknown;
 
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
 out:
 	intel_display_power_put(dev_priv, power_domain);
 	return status;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 018fb7222f60..d074d704f458 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -8462,8 +8462,6 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 		      connector->base.id, connector->name,
 		      encoder->base.id, encoder->name);
 
-	drm_modeset_acquire_init(ctx, 0);
-
 retry:
 	ret = drm_modeset_lock(&config->connection_mutex, ctx);
 	if (ret)
@@ -8502,10 +8500,14 @@ retry:
 		i++;
 		if (!(encoder->possible_crtcs & (1 << i)))
 			continue;
-		if (!possible_crtc->enabled) {
-			crtc = possible_crtc;
-			break;
-		}
+		if (possible_crtc->enabled)
+			continue;
+		/* This can occur when applying the pipe A quirk on resume. */
+		if (to_intel_crtc(possible_crtc)->new_enabled)
+			continue;
+
+		crtc = possible_crtc;
+		break;
 	}
 
 	/*
@@ -8574,15 +8576,11 @@ fail_unlock:
 		goto retry;
 	}
 
-	drm_modeset_drop_locks(ctx);
-	drm_modeset_acquire_fini(ctx);
-
 	return false;
 }
 
 void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old,
-				    struct drm_modeset_acquire_ctx *ctx)
+				    struct intel_load_detect_pipe *old)
 {
 	struct intel_encoder *intel_encoder =
 		intel_attached_encoder(connector);
@@ -8606,17 +8604,12 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 			drm_framebuffer_unreference(old->release_fb);
 		}
 
-		goto unlock;
 		return;
 	}
 
 	/* Switch crtc and encoder back off if necessary */
 	if (old->dpms_mode != DRM_MODE_DPMS_ON)
 		connector->funcs->dpms(connector, old->dpms_mode);
-
-unlock:
-	drm_modeset_drop_locks(ctx);
-	drm_modeset_acquire_fini(ctx);
 }
 
 static int i9xx_pll_refclk(struct drm_device *dev,
@@ -11700,8 +11693,8 @@ intel_cursor_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	};
 	const struct drm_rect clip = {
 		/* integer pixels */
-		.x2 = intel_crtc->config.pipe_src_w,
-		.y2 = intel_crtc->config.pipe_src_h,
+		.x2 = intel_crtc->active ? intel_crtc->config.pipe_src_w : 0,
+		.y2 = intel_crtc->active ? intel_crtc->config.pipe_src_h : 0,
 	};
 	bool visible;
 	int ret;
@@ -12659,7 +12652,7 @@ static void intel_enable_pipe_a(struct drm_device *dev)
 	struct intel_connector *connector;
 	struct drm_connector *crt = NULL;
 	struct intel_load_detect_pipe load_detect_temp;
-	struct drm_modeset_acquire_ctx ctx;
+	struct drm_modeset_acquire_ctx *ctx = dev->mode_config.acquire_ctx;
 
 	/* We can't just switch on the pipe A, we need to set things up with a
 	 * proper mode and output configuration. As a gross hack, enable pipe A
@@ -12676,10 +12669,8 @@ static void intel_enable_pipe_a(struct drm_device *dev)
 	if (!crt)
 		return;
 
-	if (intel_get_load_detect_pipe(crt, NULL, &load_detect_temp, &ctx))
-		intel_release_load_detect_pipe(crt, &load_detect_temp, &ctx);
-
-
+	if (intel_get_load_detect_pipe(crt, NULL, &load_detect_temp, ctx))
+		intel_release_load_detect_pipe(crt, &load_detect_temp);
 }
 
 static bool
@@ -13112,7 +13103,7 @@ void intel_modeset_cleanup(struct drm_device *dev)
 	 * experience fancy races otherwise.
 	 */
 	drm_irq_uninstall(dev);
-	cancel_work_sync(&dev_priv->hotplug_work);
+	intel_hpd_cancel_work(dev_priv);
 	dev_priv->pm._irqs_disabled = true;
 
 	/*
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index ee3942f0b068..67cfed6d911a 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3553,6 +3553,9 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 	if (WARN_ON(!intel_encoder->base.crtc))
 		return;
 
+	if (!to_intel_crtc(intel_encoder->base.crtc)->active)
+		return;
+
 	/* Try to read receiver status if the link appears to be up */
 	if (!intel_dp_get_link_status(intel_dp, link_status)) {
 		return;
@@ -4003,6 +4006,16 @@ void intel_dp_encoder_destroy(struct drm_encoder *encoder)
 	kfree(intel_dig_port);
 }
 
+static void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&intel_encoder->base);
+
+	if (!is_edp(intel_dp))
+		return;
+
+	edp_panel_vdd_off_sync(intel_dp);
+}
+
 static void intel_dp_encoder_reset(struct drm_encoder *encoder)
 {
 	intel_edp_panel_vdd_sanitize(to_intel_encoder(encoder));
@@ -4037,15 +4050,21 @@ bool
 intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
 {
 	struct intel_dp *intel_dp = &intel_dig_port->dp;
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
 	struct drm_device *dev = intel_dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret;
+	enum intel_display_power_domain power_domain;
+	bool ret = true;
+
 	if (intel_dig_port->base.type != INTEL_OUTPUT_EDP)
 		intel_dig_port->base.type = INTEL_OUTPUT_DISPLAYPORT;
 
 	DRM_DEBUG_KMS("got hpd irq on port %d - %s\n", intel_dig_port->port,
 		      long_hpd ? "long" : "short");
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	if (long_hpd) {
 		if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
 			goto mst_fail;
@@ -4061,8 +4080,7 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
 
 	} else {
 		if (intel_dp->is_mst) {
-			ret = intel_dp_check_mst_status(intel_dp);
-			if (ret == -EINVAL)
+			if (intel_dp_check_mst_status(intel_dp) == -EINVAL)
 				goto mst_fail;
 		}
 
@@ -4076,7 +4094,8 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
 			drm_modeset_unlock(&dev->mode_config.connection_mutex);
 		}
 	}
-	return false;
+	ret = false;
+	goto put_power;
 mst_fail:
 	/* if we were in MST mode, and device is not there get out of MST mode */
 	if (intel_dp->is_mst) {
@@ -4084,7 +4103,10 @@ mst_fail:
 		intel_dp->is_mst = false;
 		drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr, intel_dp->is_mst);
 	}
-	return true;
+put_power:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 /* Return which DP Port should be selected for Transcoder DP control */
@@ -4722,6 +4744,7 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 	intel_encoder->disable = intel_disable_dp;
 	intel_encoder->get_hw_state = intel_dp_get_hw_state;
 	intel_encoder->get_config = intel_dp_get_config;
+	intel_encoder->suspend = intel_dp_encoder_suspend;
 	if (IS_CHERRYVIEW(dev)) {
 		intel_encoder->pre_pll_enable = chv_dp_pre_pll_enable;
 		intel_encoder->pre_enable = chv_pre_enable_dp;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 4b2664bd5b81..b8c8bbd8e5f9 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -153,6 +153,12 @@ struct intel_encoder {
 	 * be set correctly before calling this function. */
 	void (*get_config)(struct intel_encoder *,
 			   struct intel_crtc_config *pipe_config);
+	/*
+	 * Called during system suspend after all pending requests for the
+	 * encoder are flushed (for example for DP AUX transactions) and
+	 * device interrupts are disabled.
+	 */
+	void (*suspend)(struct intel_encoder *);
 	int crtc_mask;
 	enum hpd_pin hpd_pin;
 };
@@ -830,8 +836,7 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 				struct intel_load_detect_pipe *old,
 				struct drm_modeset_acquire_ctx *ctx);
 void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old,
-				    struct drm_modeset_acquire_ctx *ctx);
+				    struct intel_load_detect_pipe *old);
 int intel_pin_and_fence_fb_obj(struct drm_device *dev,
 			       struct drm_i915_gem_object *obj,
 			       struct intel_engine_cs *pipelined);
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index e211eef4b7e4..32186a656816 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1323,11 +1323,16 @@ intel_tv_detect(struct drm_connector *connector, bool force)
 		struct intel_load_detect_pipe tmp;
 		struct drm_modeset_acquire_ctx ctx;
 
+		drm_modeset_acquire_init(&ctx, 0);
+
 		if (intel_get_load_detect_pipe(connector, &mode, &tmp, &ctx)) {
 			type = intel_tv_detect_type(intel_tv, connector);
-			intel_release_load_detect_pipe(connector, &tmp, &ctx);
+			intel_release_load_detect_pipe(connector, &tmp);
 		} else
 			return connector_status_unknown;
+
+		drm_modeset_drop_locks(&ctx);
+		drm_modeset_acquire_fini(&ctx);
 	} else
 		return connector->status;
 
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index 0013ad0db9ef..f77b7135ee4c 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -76,7 +76,7 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
 	evergreen.o evergreen_cs.o evergreen_blit_shaders.o \
 	evergreen_hdmi.o radeon_trace_points.o ni.o cayman_blit_shaders.o \
 	atombios_encoders.o radeon_semaphore.o radeon_sa.o atombios_i2c.o si.o \
-	si_blit_shaders.o radeon_prime.o radeon_uvd.o cik.o cik_blit_shaders.o \
+	si_blit_shaders.o radeon_prime.o cik.o cik_blit_shaders.o \
 	r600_dpm.o rs780_dpm.o rv6xx_dpm.o rv770_dpm.o rv730_dpm.o rv740_dpm.o \
 	rv770_smc.o cypress_dpm.o btc_dpm.o sumo_dpm.o sumo_smc.o trinity_dpm.o \
 	trinity_smc.o ni_dpm.o si_smc.o si_dpm.o kv_smc.o kv_dpm.o ci_smc.o \
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 022561e28707..d416bb2ff48d 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -869,6 +869,9 @@ static int ci_set_thermal_temperature_range(struct radeon_device *rdev,
 	WREG32_SMC(CG_THERMAL_CTRL, tmp);
 #endif
 
+	rdev->pm.dpm.thermal.min_temp = low_temp;
+	rdev->pm.dpm.thermal.max_temp = high_temp;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index b625646bf3e2..79a5a5519bd6 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -3483,7 +3483,7 @@ static void cik_gpu_init(struct radeon_device *rdev)
 	u32 mc_shared_chmap, mc_arb_ramcfg;
 	u32 hdp_host_path_cntl;
 	u32 tmp;
-	int i, j, k;
+	int i, j;
 
 	switch (rdev->family) {
 	case CHIP_BONAIRE:
@@ -3544,6 +3544,7 @@ static void cik_gpu_init(struct radeon_device *rdev)
 			   (rdev->pdev->device == 0x130B) ||
 			   (rdev->pdev->device == 0x130E) ||
 			   (rdev->pdev->device == 0x1315) ||
+			   (rdev->pdev->device == 0x1318) ||
 			   (rdev->pdev->device == 0x131B)) {
 			rdev->config.cik.max_cu_per_sh = 4;
 			rdev->config.cik.max_backends_per_se = 1;
@@ -3672,12 +3673,11 @@ static void cik_gpu_init(struct radeon_device *rdev)
 		     rdev->config.cik.max_sh_per_se,
 		     rdev->config.cik.max_backends_per_se);
 
+	rdev->config.cik.active_cus = 0;
 	for (i = 0; i < rdev->config.cik.max_shader_engines; i++) {
 		for (j = 0; j < rdev->config.cik.max_sh_per_se; j++) {
-			for (k = 0; k < rdev->config.cik.max_cu_per_sh; k++) {
-				rdev->config.cik.active_cus +=
-					hweight32(cik_get_cu_active_bitmap(rdev, i, j));
-			}
+			rdev->config.cik.active_cus +=
+				hweight32(cik_get_cu_active_bitmap(rdev, i, j));
 		}
 	}
 
@@ -3801,7 +3801,7 @@ int cik_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	radeon_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
 	radeon_ring_write(ring, ((scratch - PACKET3_SET_UCONFIG_REG_START) >> 2));
 	radeon_ring_write(ring, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = RREG32(scratch);
@@ -3920,6 +3920,17 @@ void cik_fence_compute_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, 0);
 }
 
+/**
+ * cik_semaphore_ring_emit - emit a semaphore on the CP ring
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon ring buffer object
+ * @semaphore: radeon semaphore object
+ * @emit_wait: Is this a sempahore wait?
+ *
+ * Emits a semaphore signal/wait packet to the CP ring and prevents the PFP
+ * from running ahead of semaphore waits.
+ */
 bool cik_semaphore_ring_emit(struct radeon_device *rdev,
 			     struct radeon_ring *ring,
 			     struct radeon_semaphore *semaphore,
@@ -3932,6 +3943,12 @@ bool cik_semaphore_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xffff) | sel);
 
+	if (emit_wait && ring->idx == RADEON_RING_TYPE_GFX_INDEX) {
+		/* Prevent the PFP from running ahead of the semaphore wait */
+		radeon_ring_write(ring, PACKET3(PACKET3_PFP_SYNC_ME, 0));
+		radeon_ring_write(ring, 0x0);
+	}
+
 	return true;
 }
 
@@ -4004,7 +4021,7 @@ int cik_copy_cpdma(struct radeon_device *rdev,
 		return r;
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_semaphore_free(rdev, &sem, *fence);
 
 	return r;
@@ -4103,7 +4120,7 @@ int cik_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	ib.ptr[1] = ((scratch - PACKET3_SET_UCONFIG_REG_START) >> 2);
 	ib.ptr[2] = 0xDEADBEEF;
 	ib.length_dw = 3;
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 		radeon_scratch_free(rdev, scratch);
 		radeon_ib_free(rdev, &ib);
@@ -4324,7 +4341,7 @@ static int cik_cp_gfx_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
 	radeon_ring_write(ring, 0x00000010); /* VGT_OUT_DEALLOC_CNTL */
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	return 0;
 }
@@ -5958,14 +5975,14 @@ void cik_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 
 	/* update SH_MEM_* regs */
 	radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
-	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
 				 WRITE_DATA_DST_SEL(0)));
 	radeon_ring_write(ring, SRBM_GFX_CNTL >> 2);
 	radeon_ring_write(ring, 0);
 	radeon_ring_write(ring, VMID(vm->id));
 
 	radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 6));
-	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
 				 WRITE_DATA_DST_SEL(0)));
 	radeon_ring_write(ring, SH_MEM_BASES >> 2);
 	radeon_ring_write(ring, 0);
@@ -5976,7 +5993,7 @@ void cik_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 	radeon_ring_write(ring, 0); /* SH_MEM_APE1_LIMIT */
 
 	radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
-	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
 				 WRITE_DATA_DST_SEL(0)));
 	radeon_ring_write(ring, SRBM_GFX_CNTL >> 2);
 	radeon_ring_write(ring, 0);
@@ -5987,7 +6004,7 @@ void cik_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 
 	/* bits 0-15 are the VM contexts0-15 */
 	radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
-	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
 				 WRITE_DATA_DST_SEL(0)));
 	radeon_ring_write(ring, VM_INVALIDATE_REQUEST >> 2);
 	radeon_ring_write(ring, 0);
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index bcf480510ac2..192278bc993c 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -596,7 +596,7 @@ int cik_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_semaphore_free(rdev, &sem, *fence);
 
 	return r;
@@ -638,7 +638,7 @@ int cik_sdma_ring_test(struct radeon_device *rdev,
 	radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr));
 	radeon_ring_write(ring, 1); /* number of DWs to follow */
 	radeon_ring_write(ring, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = readl(ptr);
@@ -695,7 +695,7 @@ int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	ib.ptr[4] = 0xDEADBEEF;
 	ib.length_dw = 5;
 
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 		radeon_ib_free(rdev, &ib);
 		DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 4fedd14e670a..dbca60c7d097 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -2869,7 +2869,7 @@ static int evergreen_cp_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
 	radeon_ring_write(ring, 0);
 	radeon_ring_write(ring, 0);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	cp_me = 0xff;
 	WREG32(CP_ME_CNTL, cp_me);
@@ -2912,7 +2912,7 @@ static int evergreen_cp_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
 	radeon_ring_write(ring, 0x00000010); /*  */
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/evergreen_dma.c b/drivers/gpu/drm/radeon/evergreen_dma.c
index 478caefe0fef..afaba388c36d 100644
--- a/drivers/gpu/drm/radeon/evergreen_dma.c
+++ b/drivers/gpu/drm/radeon/evergreen_dma.c
@@ -155,7 +155,7 @@ int evergreen_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_semaphore_free(rdev, &sem, *fence);
 
 	return r;
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
index 9ef8c38f2d66..8b58e11b64fa 100644
--- a/drivers/gpu/drm/radeon/kv_dpm.c
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -1438,14 +1438,14 @@ static int kv_update_uvd_dpm(struct radeon_device *rdev, bool gate)
 	return kv_enable_uvd_dpm(rdev, !gate);
 }
 
-static u8 kv_get_vce_boot_level(struct radeon_device *rdev)
+static u8 kv_get_vce_boot_level(struct radeon_device *rdev, u32 evclk)
 {
 	u8 i;
 	struct radeon_vce_clock_voltage_dependency_table *table =
 		&rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
 
 	for (i = 0; i < table->count; i++) {
-		if (table->entries[i].evclk >= 0) /* XXX */
+		if (table->entries[i].evclk >= evclk)
 			break;
 	}
 
@@ -1468,7 +1468,7 @@ static int kv_update_vce_dpm(struct radeon_device *rdev,
 		if (pi->caps_stable_p_state)
 			pi->vce_boot_level = table->count - 1;
 		else
-			pi->vce_boot_level = kv_get_vce_boot_level(rdev);
+			pi->vce_boot_level = kv_get_vce_boot_level(rdev, radeon_new_state->evclk);
 
 		ret = kv_copy_bytes_to_smc(rdev,
 					   pi->dpm_table_start +
@@ -2726,7 +2726,10 @@ int kv_dpm_init(struct radeon_device *rdev)
 	pi->caps_sclk_ds = true;
 	pi->enable_auto_thermal_throttling = true;
 	pi->disable_nb_ps3_in_battery = false;
-	pi->bapm_enable = true;
+	if (radeon_bapm == 0)
+		pi->bapm_enable = false;
+	else
+		pi->bapm_enable = true;
 	pi->voltage_drop_t = 0;
 	pi->caps_sclk_throttle_low_notification = false;
 	pi->caps_fps = false; /* true? */
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 327b85f7fd0d..ba89375f197f 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1505,7 +1505,7 @@ static int cayman_cp_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
 	radeon_ring_write(ring, 0);
 	radeon_ring_write(ring, 0);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	cayman_cp_enable(rdev, true);
 
@@ -1547,7 +1547,7 @@ static int cayman_cp_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
 	radeon_ring_write(ring, 0x00000010); /*  */
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	/* XXX init other rings */
 
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 04b5940b8923..4c5ec44ff328 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -925,7 +925,7 @@ int r100_copy_blit(struct radeon_device *rdev,
 	if (fence) {
 		r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
 	}
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	return r;
 }
 
@@ -958,7 +958,7 @@ void r100_ring_start(struct radeon_device *rdev, struct radeon_ring *ring)
 			  RADEON_ISYNC_ANY3D_IDLE2D |
 			  RADEON_ISYNC_WAIT_IDLEGUI |
 			  RADEON_ISYNC_CPSCRATCH_IDLEGUI);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 
@@ -3638,7 +3638,7 @@ int r100_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	}
 	radeon_ring_write(ring, PACKET0(scratch, 0));
 	radeon_ring_write(ring, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = RREG32(scratch);
 		if (tmp == 0xDEADBEEF) {
@@ -3700,7 +3700,7 @@ int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	ib.ptr[6] = PACKET2(0);
 	ib.ptr[7] = PACKET2(0);
 	ib.length_dw = 8;
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 		DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
 		goto free_ib;
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index 58f0473aa73f..67780374a652 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -121,7 +121,7 @@ int r200_copy_dma(struct radeon_device *rdev,
 	if (fence) {
 		r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
 	}
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 75b30338c226..1bc4704034ce 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -295,7 +295,7 @@ void r300_ring_start(struct radeon_device *rdev, struct radeon_ring *ring)
 	radeon_ring_write(ring,
 			  R300_GEOMETRY_ROUND_NEAREST |
 			  R300_COLOR_ROUND_NEAREST);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 static void r300_errata(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index 802b19220a21..2828605aef3f 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -219,7 +219,7 @@ static void r420_cp_errata_init(struct radeon_device *rdev)
 	radeon_ring_write(ring, PACKET0(R300_CP_RESYNC_ADDR, 1));
 	radeon_ring_write(ring, rdev->config.r300.resync_scratch);
 	radeon_ring_write(ring, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 static void r420_cp_errata_fini(struct radeon_device *rdev)
@@ -232,7 +232,7 @@ static void r420_cp_errata_fini(struct radeon_device *rdev)
 	radeon_ring_lock(rdev, ring, 8);
 	radeon_ring_write(ring, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
 	radeon_ring_write(ring, R300_RB3D_DC_FINISH);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_scratch_free(rdev, rdev->config.r300.resync_scratch);
 }
 
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index c70a504d96af..e8bf0ea2dade 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2547,7 +2547,7 @@ int r600_cp_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
 	radeon_ring_write(ring, 0);
 	radeon_ring_write(ring, 0);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	cp_me = 0xff;
 	WREG32(R_0086D8_CP_ME_CNTL, cp_me);
@@ -2683,7 +2683,7 @@ int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
 	radeon_ring_write(ring, ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2));
 	radeon_ring_write(ring, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = RREG32(scratch);
 		if (tmp == 0xDEADBEEF)
@@ -2753,6 +2753,17 @@ void r600_fence_ring_emit(struct radeon_device *rdev,
 	}
 }
 
+/**
+ * r600_semaphore_ring_emit - emit a semaphore on the CP ring
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon ring buffer object
+ * @semaphore: radeon semaphore object
+ * @emit_wait: Is this a sempahore wait?
+ *
+ * Emits a semaphore signal/wait packet to the CP ring and prevents the PFP
+ * from running ahead of semaphore waits.
+ */
 bool r600_semaphore_ring_emit(struct radeon_device *rdev,
 			      struct radeon_ring *ring,
 			      struct radeon_semaphore *semaphore,
@@ -2768,6 +2779,13 @@ bool r600_semaphore_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | sel);
 
+	/* PFP_SYNC_ME packet only exists on 7xx+ */
+	if (emit_wait && (rdev->family >= CHIP_RV770)) {
+		/* Prevent the PFP from running ahead of the semaphore wait */
+		radeon_ring_write(ring, PACKET3(PACKET3_PFP_SYNC_ME, 0));
+		radeon_ring_write(ring, 0x0);
+	}
+
 	return true;
 }
 
@@ -2845,7 +2863,7 @@ int r600_copy_cpdma(struct radeon_device *rdev,
 		return r;
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_semaphore_free(rdev, &sem, *fence);
 
 	return r;
@@ -3165,7 +3183,7 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	ib.ptr[1] = ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
 	ib.ptr[2] = 0xDEADBEEF;
 	ib.length_dw = 3;
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 		DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
 		goto free_ib;
diff --git a/drivers/gpu/drm/radeon/r600_dma.c b/drivers/gpu/drm/radeon/r600_dma.c
index 4969cef44a19..51fd98553eaf 100644
--- a/drivers/gpu/drm/radeon/r600_dma.c
+++ b/drivers/gpu/drm/radeon/r600_dma.c
@@ -261,7 +261,7 @@ int r600_dma_ring_test(struct radeon_device *rdev,
 	radeon_ring_write(ring, rdev->vram_scratch.gpu_addr & 0xfffffffc);
 	radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr) & 0xff);
 	radeon_ring_write(ring, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = readl(ptr);
@@ -368,7 +368,7 @@ int r600_dma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	ib.ptr[3] = 0xDEADBEEF;
 	ib.length_dw = 4;
 
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 		radeon_ib_free(rdev, &ib);
 		DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
@@ -493,7 +493,7 @@ int r600_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_semaphore_free(rdev, &sem, *fence);
 
 	return r;
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index f94e7a9afe75..0c4a7d8d93e0 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -1597,6 +1597,7 @@
 		 */
 #              define PACKET3_CP_DMA_CMD_SAIC      (1 << 28)
 #              define PACKET3_CP_DMA_CMD_DAIC      (1 << 29)
+#define	PACKET3_PFP_SYNC_ME				0x42 /* r7xx+ only */
 #define	PACKET3_SURFACE_SYNC				0x43
 #              define PACKET3_CB0_DEST_BASE_ENA    (1 << 6)
 #              define PACKET3_FULL_CACHE_ENA       (1 << 20) /* r7xx+ only */
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 9e1732eb402c..b281886f6f51 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -105,6 +105,7 @@ extern int radeon_vm_size;
 extern int radeon_vm_block_size;
 extern int radeon_deep_color;
 extern int radeon_use_pflipirq;
+extern int radeon_bapm;
 
 /*
  * Copy from radeon_drv.h so we don't have to include both and have conflicting
@@ -967,7 +968,7 @@ int radeon_ib_get(struct radeon_device *rdev, int ring,
 		  unsigned size);
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib);
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
-		       struct radeon_ib *const_ib);
+		       struct radeon_ib *const_ib, bool hdp_flush);
 int radeon_ib_pool_init(struct radeon_device *rdev);
 void radeon_ib_pool_fini(struct radeon_device *rdev);
 int radeon_ib_ring_tests(struct radeon_device *rdev);
@@ -977,8 +978,10 @@ bool radeon_ring_supports_scratch_reg(struct radeon_device *rdev,
 void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *cp);
 int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
 int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
-void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *cp);
-void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *cp);
+void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *cp,
+			bool hdp_flush);
+void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *cp,
+			       bool hdp_flush);
 void radeon_ring_undo(struct radeon_ring *ring);
 void radeon_ring_unlock_undo(struct radeon_device *rdev, struct radeon_ring *cp);
 int radeon_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index ee712c199b25..83f382e8e40e 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -132,7 +132,8 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 		 * the buffers used for read only, which doubles the range
 		 * to 0 to 31. 32 is reserved for the kernel driver.
 		 */
-		priority = (r->flags & 0xf) * 2 + !!r->write_domain;
+		priority = (r->flags & RADEON_RELOC_PRIO_MASK) * 2
+			   + !!r->write_domain;
 
 		/* the first reloc of an UVD job is the msg and that must be in
 		   VRAM, also but everything into VRAM on AGP cards to avoid
@@ -450,7 +451,7 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev,
 		radeon_vce_note_usage(rdev);
 
 	radeon_cs_sync_rings(parser);
-	r = radeon_ib_schedule(rdev, &parser->ib, NULL);
+	r = radeon_ib_schedule(rdev, &parser->ib, NULL, true);
 	if (r) {
 		DRM_ERROR("Failed to schedule IB !\n");
 	}
@@ -541,9 +542,9 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
 
 	if ((rdev->family >= CHIP_TAHITI) &&
 	    (parser->chunk_const_ib_idx != -1)) {
-		r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib);
+		r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib, true);
 	} else {
-		r = radeon_ib_schedule(rdev, &parser->ib, NULL);
+		r = radeon_ib_schedule(rdev, &parser->ib, NULL, true);
 	}
 
 out:
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index c8ea050c8fa4..6a219bcee66d 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1680,8 +1680,8 @@ int radeon_gpu_reset(struct radeon_device *rdev)
 	radeon_save_bios_scratch_regs(rdev);
 	/* block TTM */
 	resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
-	radeon_pm_suspend(rdev);
 	radeon_suspend(rdev);
+	radeon_hpd_fini(rdev);
 
 	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
 		ring_sizes[i] = radeon_ring_backup(rdev, &rdev->ring[i],
@@ -1726,9 +1726,39 @@ retry:
 		}
 	}
 
-	radeon_pm_resume(rdev);
+	if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled) {
+		/* do dpm late init */
+		r = radeon_pm_late_init(rdev);
+		if (r) {
+			rdev->pm.dpm_enabled = false;
+			DRM_ERROR("radeon_pm_late_init failed, disabling dpm\n");
+		}
+	} else {
+		/* resume old pm late */
+		radeon_pm_resume(rdev);
+	}
+
+	/* init dig PHYs, disp eng pll */
+	if (rdev->is_atom_bios) {
+		radeon_atom_encoder_init(rdev);
+		radeon_atom_disp_eng_pll_init(rdev);
+		/* turn on the BL */
+		if (rdev->mode_info.bl_encoder) {
+			u8 bl_level = radeon_get_backlight_level(rdev,
+								 rdev->mode_info.bl_encoder);
+			radeon_set_backlight_level(rdev, rdev->mode_info.bl_encoder,
+						   bl_level);
+		}
+	}
+	/* reset hpd state */
+	radeon_hpd_init(rdev);
+
 	drm_helper_resume_force_mode(rdev->ddev);
 
+	/* set the power state here in case we are a PX system or headless */
+	if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled)
+		radeon_pm_compute_clocks(rdev);
+
 	ttm_bo_unlock_delayed_workqueue(&rdev->mman.bdev, resched);
 	if (r) {
 		/* bad news, how to tell it to userspace ? */
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 092d067f93e1..8df888908833 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -180,6 +180,7 @@ int radeon_vm_size = 8;
 int radeon_vm_block_size = -1;
 int radeon_deep_color = 0;
 int radeon_use_pflipirq = 2;
+int radeon_bapm = -1;
 
 MODULE_PARM_DESC(no_wb, "Disable AGP writeback for scratch registers");
 module_param_named(no_wb, radeon_no_wb, int, 0444);
@@ -259,6 +260,9 @@ module_param_named(deep_color, radeon_deep_color, int, 0444);
 MODULE_PARM_DESC(use_pflipirq, "Pflip irqs for pageflip completion (0 = disable, 1 = as fallback, 2 = exclusive (default))");
 module_param_named(use_pflipirq, radeon_use_pflipirq, int, 0444);
 
+MODULE_PARM_DESC(bapm, "BAPM support (1 = enable, 0 = disable, -1 = auto)");
+module_param_named(bapm, radeon_bapm, int, 0444);
+
 static struct pci_device_id pciidlist[] = {
 	radeon_PCI_IDS
 };
diff --git a/drivers/gpu/drm/radeon/radeon_ib.c b/drivers/gpu/drm/radeon/radeon_ib.c
index 65b0c213488d..5bf2c0a05827 100644
--- a/drivers/gpu/drm/radeon/radeon_ib.c
+++ b/drivers/gpu/drm/radeon/radeon_ib.c
@@ -107,6 +107,7 @@ void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib)
  * @rdev: radeon_device pointer
  * @ib: IB object to schedule
  * @const_ib: Const IB to schedule (SI only)
+ * @hdp_flush: Whether or not to perform an HDP cache flush
  *
  * Schedule an IB on the associated ring (all asics).
  * Returns 0 on success, error on failure.
@@ -122,7 +123,7 @@ void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib)
  * to SI there was just a DE IB.
  */
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
-		       struct radeon_ib *const_ib)
+		       struct radeon_ib *const_ib, bool hdp_flush)
 {
 	struct radeon_ring *ring = &rdev->ring[ib->ring];
 	int r = 0;
@@ -176,7 +177,7 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
 	if (ib->vm)
 		radeon_vm_fence(rdev, ib->vm, ib->fence);
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, hdp_flush);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 23314be49480..164898b0010c 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -460,10 +460,6 @@ static ssize_t radeon_get_dpm_state(struct device *dev,
 	struct radeon_device *rdev = ddev->dev_private;
 	enum radeon_pm_state_type pm = rdev->pm.dpm.user_state;
 
-	if  ((rdev->flags & RADEON_IS_PX) &&
-	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
-		return snprintf(buf, PAGE_SIZE, "off\n");
-
 	return snprintf(buf, PAGE_SIZE, "%s\n",
 			(pm == POWER_STATE_TYPE_BATTERY) ? "battery" :
 			(pm == POWER_STATE_TYPE_BALANCED) ? "balanced" : "performance");
@@ -477,11 +473,6 @@ static ssize_t radeon_set_dpm_state(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct radeon_device *rdev = ddev->dev_private;
 
-	/* Can't set dpm state when the card is off */
-	if  ((rdev->flags & RADEON_IS_PX) &&
-	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
-		return -EINVAL;
-
 	mutex_lock(&rdev->pm.mutex);
 	if (strncmp("battery", buf, strlen("battery")) == 0)
 		rdev->pm.dpm.user_state = POWER_STATE_TYPE_BATTERY;
@@ -495,7 +486,12 @@ static ssize_t radeon_set_dpm_state(struct device *dev,
 		goto fail;
 	}
 	mutex_unlock(&rdev->pm.mutex);
-	radeon_pm_compute_clocks(rdev);
+
+	/* Can't set dpm state when the card is off */
+	if (!(rdev->flags & RADEON_IS_PX) ||
+	    (ddev->switch_power_state == DRM_SWITCH_POWER_ON))
+		radeon_pm_compute_clocks(rdev);
+
 fail:
 	return count;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 5b4e0cf231a0..d65607902537 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -177,16 +177,18 @@ int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *ring, unsig
  *
  * @rdev: radeon_device pointer
  * @ring: radeon_ring structure holding ring information
+ * @hdp_flush: Whether or not to perform an HDP cache flush
  *
  * Update the wptr (write pointer) to tell the GPU to
  * execute new commands on the ring buffer (all asics).
  */
-void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
+void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring,
+			bool hdp_flush)
 {
 	/* If we are emitting the HDP flush via the ring buffer, we need to
 	 * do it before padding.
 	 */
-	if (rdev->asic->ring[ring->idx]->hdp_flush)
+	if (hdp_flush && rdev->asic->ring[ring->idx]->hdp_flush)
 		rdev->asic->ring[ring->idx]->hdp_flush(rdev, ring);
 	/* We pad to match fetch size */
 	while (ring->wptr & ring->align_mask) {
@@ -196,7 +198,7 @@ void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
 	/* If we are emitting the HDP flush via MMIO, we need to do it after
 	 * all CPU writes to VRAM finished.
 	 */
-	if (rdev->asic->mmio_hdp_flush)
+	if (hdp_flush && rdev->asic->mmio_hdp_flush)
 		rdev->asic->mmio_hdp_flush(rdev);
 	radeon_ring_set_wptr(rdev, ring);
 }
@@ -207,12 +209,14 @@ void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
  *
  * @rdev: radeon_device pointer
  * @ring: radeon_ring structure holding ring information
+ * @hdp_flush: Whether or not to perform an HDP cache flush
  *
  * Call radeon_ring_commit() then unlock the ring (all asics).
  */
-void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *ring)
+void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *ring,
+			       bool hdp_flush)
 {
-	radeon_ring_commit(rdev, ring);
+	radeon_ring_commit(rdev, ring, hdp_flush);
 	mutex_unlock(&rdev->ring_lock);
 }
 
@@ -372,7 +376,7 @@ int radeon_ring_restore(struct radeon_device *rdev, struct radeon_ring *ring,
 		radeon_ring_write(ring, data[i]);
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	kfree(data);
 	return 0;
 }
@@ -400,9 +404,7 @@ int radeon_ring_init(struct radeon_device *rdev, struct radeon_ring *ring, unsig
 	/* Allocate ring buffer */
 	if (ring->ring_obj == NULL) {
 		r = radeon_bo_create(rdev, ring->ring_size, PAGE_SIZE, true,
-				     RADEON_GEM_DOMAIN_GTT,
-				     (rdev->flags & RADEON_IS_PCIE) ?
-				     RADEON_GEM_GTT_WC : 0,
+				     RADEON_GEM_DOMAIN_GTT, 0,
 				     NULL, &ring->ring_obj);
 		if (r) {
 			dev_err(rdev->dev, "(%d) ring create failed\n", r);
diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c
index dbd6bcde92de..56d9fd66d8ae 100644
--- a/drivers/gpu/drm/radeon/radeon_semaphore.c
+++ b/drivers/gpu/drm/radeon/radeon_semaphore.c
@@ -179,7 +179,7 @@ int radeon_semaphore_sync_rings(struct radeon_device *rdev,
 			continue;
 		}
 
-		radeon_ring_commit(rdev, &rdev->ring[i]);
+		radeon_ring_commit(rdev, &rdev->ring[i], false);
 		radeon_fence_note_sync(fence, ring);
 
 		semaphore->gpu_addr += 8;
diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c
index 5adf4207453d..17bc3dced9f1 100644
--- a/drivers/gpu/drm/radeon/radeon_test.c
+++ b/drivers/gpu/drm/radeon/radeon_test.c
@@ -288,7 +288,7 @@ static int radeon_test_create_and_emit_fence(struct radeon_device *rdev,
 			return r;
 		}
 		radeon_fence_emit(rdev, fence, ring->idx);
-		radeon_ring_unlock_commit(rdev, ring);
+		radeon_ring_unlock_commit(rdev, ring, false);
 	}
 	return 0;
 }
@@ -313,7 +313,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringA);
+	radeon_ring_unlock_commit(rdev, ringA, false);
 
 	r = radeon_test_create_and_emit_fence(rdev, ringA, &fence1);
 	if (r)
@@ -325,7 +325,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringA);
+	radeon_ring_unlock_commit(rdev, ringA, false);
 
 	r = radeon_test_create_and_emit_fence(rdev, ringA, &fence2);
 	if (r)
@@ -344,7 +344,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_signal(rdev, ringB->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringB);
+	radeon_ring_unlock_commit(rdev, ringB, false);
 
 	r = radeon_fence_wait(fence1, false);
 	if (r) {
@@ -365,7 +365,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_signal(rdev, ringB->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringB);
+	radeon_ring_unlock_commit(rdev, ringB, false);
 
 	r = radeon_fence_wait(fence2, false);
 	if (r) {
@@ -408,7 +408,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringA);
+	radeon_ring_unlock_commit(rdev, ringA, false);
 
 	r = radeon_test_create_and_emit_fence(rdev, ringA, &fenceA);
 	if (r)
@@ -420,7 +420,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_wait(rdev, ringB->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringB);
+	radeon_ring_unlock_commit(rdev, ringB, false);
 	r = radeon_test_create_and_emit_fence(rdev, ringB, &fenceB);
 	if (r)
 		goto out_cleanup;
@@ -442,7 +442,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringC);
+	radeon_ring_unlock_commit(rdev, ringC, false);
 
 	for (i = 0; i < 30; ++i) {
 		mdelay(100);
@@ -468,7 +468,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
-	radeon_ring_unlock_commit(rdev, ringC);
+	radeon_ring_unlock_commit(rdev, ringC, false);
 
 	mdelay(1000);
 
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 6bf55ec85b62..341848a14376 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -646,7 +646,7 @@ static int radeon_uvd_send_msg(struct radeon_device *rdev,
 		ib.ptr[i] = PACKET2(0);
 	ib.length_dw = 16;
 
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r)
 		goto err;
 	ttm_eu_fence_buffer_objects(&ticket, &head, ib.fence);
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c
index f9b70a43aa52..c7190aadbd89 100644
--- a/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/drivers/gpu/drm/radeon/radeon_vce.c
@@ -368,7 +368,7 @@ int radeon_vce_get_create_msg(struct radeon_device *rdev, int ring,
 	for (i = ib.length_dw; i < ib_size_dw; ++i)
 		ib.ptr[i] = 0x0;
 
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 	        DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
 	}
@@ -425,7 +425,7 @@ int radeon_vce_get_destroy_msg(struct radeon_device *rdev, int ring,
 	for (i = ib.length_dw; i < ib_size_dw; ++i)
 		ib.ptr[i] = 0x0;
 
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 	        DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
 	}
@@ -715,7 +715,7 @@ int radeon_vce_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 		return r;
 	}
 	radeon_ring_write(ring, VCE_CMD_END);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	for (i = 0; i < rdev->usec_timeout; i++) {
 	        if (vce_v1_0_get_rptr(rdev, ring) != rptr)
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index ccae4d9dc3de..088ffdc2f577 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -420,7 +420,7 @@ static int radeon_vm_clear_bo(struct radeon_device *rdev,
 	radeon_asic_vm_pad_ib(rdev, &ib);
 	WARN_ON(ib.length_dw > 64);
 
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r)
                 goto error;
 
@@ -483,6 +483,10 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 			/* add a clone of the bo_va to clear the old address */
 			struct radeon_bo_va *tmp;
 			tmp = kzalloc(sizeof(struct radeon_bo_va), GFP_KERNEL);
+			if (!tmp) {
+				mutex_unlock(&vm->mutex);
+				return -ENOMEM;
+			}
 			tmp->it.start = bo_va->it.start;
 			tmp->it.last = bo_va->it.last;
 			tmp->vm = vm;
@@ -693,7 +697,7 @@ int radeon_vm_update_page_directory(struct radeon_device *rdev,
 		radeon_semaphore_sync_to(ib.semaphore, pd->tbo.sync_obj);
 		radeon_semaphore_sync_to(ib.semaphore, vm->last_id_use);
 		WARN_ON(ib.length_dw > ndw);
-		r = radeon_ib_schedule(rdev, &ib, NULL);
+		r = radeon_ib_schedule(rdev, &ib, NULL, false);
 		if (r) {
 			radeon_ib_free(rdev, &ib);
 			return r;
@@ -957,7 +961,7 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
 	WARN_ON(ib.length_dw > ndw);
 
 	radeon_semaphore_sync_to(ib.semaphore, vm->fence);
-	r = radeon_ib_schedule(rdev, &ib, NULL);
+	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
 		radeon_ib_free(rdev, &ib);
 		return r;
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 3e21e869015f..8a477bf1fdb3 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -124,7 +124,7 @@ void rv515_ring_start(struct radeon_device *rdev, struct radeon_ring *ring)
 	radeon_ring_write(ring, GEOMETRY_ROUND_NEAREST | COLOR_ROUND_NEAREST);
 	radeon_ring_write(ring, PACKET0(0x20C8, 0));
 	radeon_ring_write(ring, 0);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 int rv515_mc_wait_for_idle(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/rv770_dma.c b/drivers/gpu/drm/radeon/rv770_dma.c
index bbf2e076ee45..74426ac2bb5c 100644
--- a/drivers/gpu/drm/radeon/rv770_dma.c
+++ b/drivers/gpu/drm/radeon/rv770_dma.c
@@ -90,7 +90,7 @@ int rv770_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_semaphore_free(rdev, &sem, *fence);
 
 	return r;
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 011779bd2b3d..a1274a31405c 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -3057,7 +3057,7 @@ static void si_gpu_init(struct radeon_device *rdev)
 	u32 sx_debug_1;
 	u32 hdp_host_path_cntl;
 	u32 tmp;
-	int i, j, k;
+	int i, j;
 
 	switch (rdev->family) {
 	case CHIP_TAHITI:
@@ -3255,12 +3255,11 @@ static void si_gpu_init(struct radeon_device *rdev)
 		     rdev->config.si.max_sh_per_se,
 		     rdev->config.si.max_cu_per_sh);
 
+	rdev->config.si.active_cus = 0;
 	for (i = 0; i < rdev->config.si.max_shader_engines; i++) {
 		for (j = 0; j < rdev->config.si.max_sh_per_se; j++) {
-			for (k = 0; k < rdev->config.si.max_cu_per_sh; k++) {
-				rdev->config.si.active_cus +=
-					hweight32(si_get_cu_active_bitmap(rdev, i, j));
-			}
+			rdev->config.si.active_cus +=
+				hweight32(si_get_cu_active_bitmap(rdev, i, j));
 		}
 	}
 
@@ -3541,7 +3540,7 @@ static int si_cp_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, PACKET3_BASE_INDEX(CE_PARTITION_BASE));
 	radeon_ring_write(ring, 0xc000);
 	radeon_ring_write(ring, 0xe000);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	si_cp_enable(rdev, true);
 
@@ -3570,7 +3569,7 @@ static int si_cp_start(struct radeon_device *rdev)
 	radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
 	radeon_ring_write(ring, 0x00000010); /* VGT_OUT_DEALLOC_CNTL */
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 	for (i = RADEON_RING_TYPE_GFX_INDEX; i <= CAYMAN_RING_TYPE_CP2_INDEX; ++i) {
 		ring = &rdev->ring[i];
@@ -3580,7 +3579,7 @@ static int si_cp_start(struct radeon_device *rdev)
 		radeon_ring_write(ring, PACKET3_COMPUTE(PACKET3_CLEAR_STATE, 0));
 		radeon_ring_write(ring, 0);
 
-		radeon_ring_unlock_commit(rdev, ring);
+		radeon_ring_unlock_commit(rdev, ring, false);
 	}
 
 	return 0;
@@ -5028,7 +5027,7 @@ void si_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 
 	/* flush hdp cache */
 	radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
-	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(1) |
 				 WRITE_DATA_DST_SEL(0)));
 	radeon_ring_write(ring, HDP_MEM_COHERENCY_FLUSH_CNTL >> 2);
 	radeon_ring_write(ring, 0);
@@ -5036,7 +5035,7 @@ void si_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 
 	/* bits 0-15 are the VM contexts0-15 */
 	radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
-	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+	radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(1) |
 				 WRITE_DATA_DST_SEL(0)));
 	radeon_ring_write(ring, VM_INVALIDATE_REQUEST >> 2);
 	radeon_ring_write(ring, 0);
diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c
index 716505129450..7c22baaf94db 100644
--- a/drivers/gpu/drm/radeon/si_dma.c
+++ b/drivers/gpu/drm/radeon/si_dma.c
@@ -275,7 +275,7 @@ int si_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	radeon_semaphore_free(rdev, &sem, *fence);
 
 	return r;
diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c
index 32e50be9c4ac..57f780053b3e 100644
--- a/drivers/gpu/drm/radeon/trinity_dpm.c
+++ b/drivers/gpu/drm/radeon/trinity_dpm.c
@@ -1874,16 +1874,22 @@ int trinity_dpm_init(struct radeon_device *rdev)
 	for (i = 0; i < SUMO_MAX_HARDWARE_POWERLEVELS; i++)
 		pi->at[i] = TRINITY_AT_DFLT;
 
-	/* There are stability issues reported on with
-	 * bapm enabled when switching between AC and battery
-	 * power.  At the same time, some MSI boards hang
-	 * if it's not enabled and dpm is enabled.  Just enable
-	 * it for MSI boards right now.
-	 */
-	if (rdev->pdev->subsystem_vendor == 0x1462)
-		pi->enable_bapm = true;
-	else
+	if (radeon_bapm == -1) {
+		/* There are stability issues reported on with
+		 * bapm enabled when switching between AC and battery
+		 * power.  At the same time, some MSI boards hang
+		 * if it's not enabled and dpm is enabled.  Just enable
+		 * it for MSI boards right now.
+		 */
+		if (rdev->pdev->subsystem_vendor == 0x1462)
+			pi->enable_bapm = true;
+		else
+			pi->enable_bapm = false;
+	} else if (radeon_bapm == 0) {
 		pi->enable_bapm = false;
+	} else {
+		pi->enable_bapm = true;
+	}
 	pi->enable_nbps_policy = true;
 	pi->enable_sclk_ds = true;
 	pi->enable_gfx_power_gating = true;
diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c
index be42c8125203..cda391347286 100644
--- a/drivers/gpu/drm/radeon/uvd_v1_0.c
+++ b/drivers/gpu/drm/radeon/uvd_v1_0.c
@@ -124,7 +124,7 @@ int uvd_v1_0_init(struct radeon_device *rdev)
 	radeon_ring_write(ring, PACKET0(UVD_SEMA_CNTL, 0));
 	radeon_ring_write(ring, 3);
 
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 
 done:
 	/* lower clocks again */
@@ -331,7 +331,7 @@ int uvd_v1_0_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	}
 	radeon_ring_write(ring, PACKET0(UVD_CONTEXT_ID, 0));
 	radeon_ring_write(ring, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev, ring);
+	radeon_ring_unlock_commit(rdev, ring, false);
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = RREG32(UVD_CONTEXT_ID);
 		if (tmp == 0xDEADBEEF)