79 files changed, 624 insertions, 297 deletions

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 9a0cc09e6653..e4a1490b42c2 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -260,7 +260,7 @@ static void armada_drm_vblank_off(struct armada_crtc *dcrtc)
          * Tell the DRM core that vblank IRQs aren't going to happen for
          * a while.  This cleans up any pending vblank events for us.
          */
-        drm_vblank_off(dev, dcrtc->num);
+        drm_crtc_vblank_off(&dcrtc->crtc);
 
         /* Handle any pending flip event. */
         spin_lock_irq(&dev->event_lock);
@@ -289,6 +289,8 @@ static void armada_drm_crtc_dpms(struct drm_crtc *crtc, int dpms)
                 armada_drm_crtc_update(dcrtc);
                 if (dpms_blanked(dpms))
                         armada_drm_vblank_off(dcrtc);
+                else
+                        drm_crtc_vblank_on(&dcrtc->crtc);
         }
 }
 
@@ -526,7 +528,7 @@ static int armada_drm_crtc_mode_set(struct drm_crtc *crtc,
         /* Wait for pending flips to complete */
         wait_event(dcrtc->frame_wait, !dcrtc->frame_work);
 
-        drm_vblank_pre_modeset(crtc->dev, dcrtc->num);
+        drm_crtc_vblank_off(crtc);
 
         crtc->mode = *adj;
 
@@ -617,7 +619,7 @@ static int armada_drm_crtc_mode_set(struct drm_crtc *crtc,
 
         armada_drm_crtc_update(dcrtc);
 
-        drm_vblank_post_modeset(crtc->dev, dcrtc->num);
+        drm_crtc_vblank_on(crtc);
         armada_drm_crtc_finish_fb(dcrtc, old_fb, dpms_blanked(dcrtc->dpms));
 
         return 0;
@@ -945,18 +947,15 @@ static int armada_drm_crtc_page_flip(struct drm_crtc *crtc,
         armada_reg_queue_end(work->regs, i);
 
         /*
-         * Hold the old framebuffer for the work - DRM appears to drop our
-         * reference to the old framebuffer in drm_mode_page_flip_ioctl().
+         * Ensure that we hold a reference on the new framebuffer.
+         * This has to match the behaviour in mode_set.
          */
-        drm_framebuffer_reference(work->old_fb);
+        drm_framebuffer_reference(fb);
 
         ret = armada_drm_crtc_queue_frame_work(dcrtc, work);
         if (ret) {
-                /*
-                 * Undo our reference above; DRM does not drop the reference
-                 * to this object on error, so that's okay.
-                 */
-                drm_framebuffer_unreference(work->old_fb);
+                /* Undo our reference above */
+                drm_framebuffer_unreference(fb);
                 kfree(work);
                 return ret;
         }
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index f672e6ad8afa..908e5316eac4 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -190,6 +190,7 @@ static int armada_drm_load(struct drm_device *dev, unsigned long flags)
         if (ret)
                 goto err_comp;
 
+        dev->irq_enabled = true;
         dev->vblank_disable_allowed = 1;
 
         ret = armada_fbdev_init(dev);
@@ -331,7 +332,7 @@ static struct drm_driver armada_drm_driver = {
         .desc                   = "Armada SoC DRM",
         .date                   = "20120730",
         .driver_features        = DRIVER_GEM | DRIVER_MODESET |
-                                  DRIVER_PRIME,
+                                  DRIVER_HAVE_IRQ | DRIVER_PRIME,
         .ioctls                 = armada_ioctls,
         .fops                   = &armada_drm_fops,
 };
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.c b/drivers/gpu/drm/cirrus/cirrus_drv.c
index e705335101a5..c2a1cba1e984 100644
--- a/drivers/gpu/drm/cirrus/cirrus_drv.c
+++ b/drivers/gpu/drm/cirrus/cirrus_drv.c
@@ -32,6 +32,8 @@ static struct drm_driver driver;
 static const struct pci_device_id pciidlist[] = {
         { PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5446, 0x1af4, 0x1100, 0,
           0, 0 },
+        { PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5446, PCI_VENDOR_ID_XEN,
+          0x0001, 0, 0, 0 },
         {0,}
 };
 
diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c
index cd50ece31601..6adb1e5cfb08 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_core.c
+++ b/drivers/gpu/drm/exynos/exynos_dp_core.c
@@ -1355,13 +1355,8 @@ static void exynos_dp_unbind(struct device *dev, struct device *master,
                                 void *data)
 {
         struct exynos_drm_display *display = dev_get_drvdata(dev);
-        struct exynos_dp_device *dp = display->ctx;
-        struct drm_encoder *encoder = dp->encoder;
 
         exynos_dp_dpms(display, DRM_MODE_DPMS_OFF);
-
-        exynos_dp_connector_destroy(&dp->connector);
-        encoder->funcs->destroy(encoder);
 }
 
 static const struct component_ops exynos_dp_ops = {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 8e38e9f8e542..45026e693225 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -71,13 +71,16 @@ static void exynos_drm_crtc_dpms(struct drm_crtc *crtc, int mode)
                                 !atomic_read(&exynos_crtc->pending_flip),
                                 HZ/20))
                         atomic_set(&exynos_crtc->pending_flip, 0);
-                drm_vblank_off(crtc->dev, exynos_crtc->pipe);
+                drm_crtc_vblank_off(crtc);
         }
 
         if (manager->ops->dpms)
                 manager->ops->dpms(manager, mode);
 
         exynos_crtc->dpms = mode;
+
+        if (mode == DRM_MODE_DPMS_ON)
+                drm_crtc_vblank_on(crtc);
 }
 
 static void exynos_drm_crtc_prepare(struct drm_crtc *crtc)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
index 96c87db388fb..3dc678ed9949 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
@@ -338,14 +338,10 @@ err_del_component:
 
 int exynos_dpi_remove(struct device *dev)
 {
-        struct drm_encoder *encoder = exynos_dpi_display.encoder;
         struct exynos_dpi *ctx = exynos_dpi_display.ctx;
 
         exynos_dpi_dpms(&exynos_dpi_display, DRM_MODE_DPMS_OFF);
 
-        exynos_dpi_connector_destroy(&ctx->connector);
-        encoder->funcs->destroy(encoder);
-
         if (ctx->panel)
                 drm_panel_detach(ctx->panel);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 443a2069858a..e5c4c6c8c967 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -87,16 +87,12 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 
                 plane = exynos_plane_init(dev, possible_crtcs,
                                           DRM_PLANE_TYPE_OVERLAY);
-                if (IS_ERR(plane))
-                        goto err_mode_config_cleanup;
-        }
-
-        /* init kms poll for handling hpd */
-        drm_kms_helper_poll_init(dev);
+                if (!IS_ERR(plane))
+                        continue;
 
-        ret = drm_vblank_init(dev, MAX_CRTC);
-        if (ret)
+                ret = PTR_ERR(plane);
                 goto err_mode_config_cleanup;
+        }
 
         /* setup possible_clones. */
         exynos_drm_encoder_setup(dev);
@@ -106,15 +102,16 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
         /* Try to bind all sub drivers. */
         ret = component_bind_all(dev->dev, dev);
         if (ret)
-                goto err_cleanup_vblank;
+                goto err_mode_config_cleanup;
 
-        /* Probe non kms sub drivers and virtual display driver. */
-        ret = exynos_drm_device_subdrv_probe(dev);
+        ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
         if (ret)
                 goto err_unbind_all;
 
-        /* force connectors detection */
-        drm_helper_hpd_irq_event(dev);
+        /* Probe non kms sub drivers and virtual display driver. */
+        ret = exynos_drm_device_subdrv_probe(dev);
+        if (ret)
+                goto err_cleanup_vblank;
 
         /*
          * enable drm irq mode.
@@ -133,12 +130,18 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
          */
         dev->vblank_disable_allowed = true;
 
+        /* init kms poll for handling hpd */
+        drm_kms_helper_poll_init(dev);
+
+        /* force connectors detection */
+        drm_helper_hpd_irq_event(dev);
+
         return 0;
 
-err_unbind_all:
-        component_unbind_all(dev->dev, dev);
 err_cleanup_vblank:
         drm_vblank_cleanup(dev);
+err_unbind_all:
+        component_unbind_all(dev->dev, dev);
 err_mode_config_cleanup:
         drm_mode_config_cleanup(dev);
         drm_release_iommu_mapping(dev);
@@ -155,8 +158,8 @@ static int exynos_drm_unload(struct drm_device *dev)
         exynos_drm_fbdev_fini(dev);
         drm_kms_helper_poll_fini(dev);
 
-        component_unbind_all(dev->dev, dev);
         drm_vblank_cleanup(dev);
+        component_unbind_all(dev->dev, dev);
         drm_mode_config_cleanup(dev);
         drm_release_iommu_mapping(dev);
 
@@ -191,8 +194,12 @@ static int exynos_drm_resume(struct drm_device *dev)
 
         drm_modeset_lock_all(dev);
         list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-                if (connector->funcs->dpms)
-                        connector->funcs->dpms(connector, connector->dpms);
+                if (connector->funcs->dpms) {
+                        int dpms = connector->dpms;
+
+                        connector->dpms = DRM_MODE_DPMS_OFF;
+                        connector->funcs->dpms(connector, dpms);
+                }
         }
         drm_modeset_unlock_all(dev);
 
@@ -488,6 +495,12 @@ static struct component_match *exynos_drm_match_add(struct device *dev)
 
         mutex_lock(&drm_component_lock);
 
+        /* Do not retry to probe if there is no any kms driver regitered. */
+        if (list_empty(&drm_component_list)) {
+                mutex_unlock(&drm_component_lock);
+                return ERR_PTR(-ENODEV);
+        }
+
         list_for_each_entry(cdev, &drm_component_list, list) {
                 /*
                  * Add components to master only in case that crtc and
@@ -578,10 +591,21 @@ static int exynos_drm_platform_probe(struct platform_device *pdev)
                 goto err_unregister_mixer_drv;
 #endif
 
+        match = exynos_drm_match_add(&pdev->dev);
+        if (IS_ERR(match)) {
+                ret = PTR_ERR(match);
+                goto err_unregister_hdmi_drv;
+        }
+
+        ret = component_master_add_with_match(&pdev->dev, &exynos_drm_ops,
+                                                match);
+        if (ret < 0)
+                goto err_unregister_hdmi_drv;
+
 #ifdef CONFIG_DRM_EXYNOS_G2D
         ret = platform_driver_register(&g2d_driver);
         if (ret < 0)
-                goto err_unregister_hdmi_drv;
+                goto err_del_component_master;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_FIMC
@@ -612,23 +636,9 @@ static int exynos_drm_platform_probe(struct platform_device *pdev)
                 goto err_unregister_ipp_drv;
 #endif
 
-        match = exynos_drm_match_add(&pdev->dev);
-        if (IS_ERR(match)) {
-                ret = PTR_ERR(match);
-                goto err_unregister_resources;
-        }
-
-        ret = component_master_add_with_match(&pdev->dev, &exynos_drm_ops,
-                                                match);
-        if (ret < 0)
-                goto err_unregister_resources;
-
         return ret;
 
-err_unregister_resources:
-
 #ifdef CONFIG_DRM_EXYNOS_IPP
-        exynos_platform_device_ipp_unregister();
 err_unregister_ipp_drv:
         platform_driver_unregister(&ipp_driver);
 err_unregister_gsc_drv:
@@ -651,9 +661,11 @@ err_unregister_g2d_drv:
 
 #ifdef CONFIG_DRM_EXYNOS_G2D
         platform_driver_unregister(&g2d_driver);
-err_unregister_hdmi_drv:
+err_del_component_master:
 #endif
+        component_master_del(&pdev->dev, &exynos_drm_ops);
 
+err_unregister_hdmi_drv:
 #ifdef CONFIG_DRM_EXYNOS_HDMI
         platform_driver_unregister(&hdmi_driver);
 err_unregister_mixer_drv:
@@ -734,6 +746,18 @@ static int exynos_drm_init(void)
 {
         int ret;
 
+        /*
+         * Register device object only in case of Exynos SoC.
+         *
+         * Below codes resolves temporarily infinite loop issue incurred
+         * by Exynos drm driver when using multi-platform kernel.
+         * So these codes will be replaced with more generic way later.
+         */
+        if (!of_machine_is_compatible("samsung,exynos3") &&
+                        !of_machine_is_compatible("samsung,exynos4") &&
+                        !of_machine_is_compatible("samsung,exynos5"))
+                return -ENODEV;
+
         exynos_drm_pdev = platform_device_register_simple("exynos-drm", -1,
                                                                 NULL, 0);
         if (IS_ERR(exynos_drm_pdev))
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index 24741d8758e8..acf7e9e39dcd 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -1660,13 +1660,9 @@ static void exynos_dsi_unbind(struct device *dev, struct device *master,
                                 void *data)
 {
         struct exynos_dsi *dsi = exynos_dsi_display.ctx;
-        struct drm_encoder *encoder = dsi->encoder;
 
         exynos_dsi_dpms(&exynos_dsi_display, DRM_MODE_DPMS_OFF);
 
-        exynos_dsi_connector_destroy(&dsi->connector);
-        encoder->funcs->destroy(encoder);
-
         mipi_dsi_host_unregister(&dsi->dsi_host);
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
index df7a77d3eff8..6ff8599f6cbf 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
@@ -302,9 +302,12 @@ static void g2d_fini_cmdlist(struct g2d_data *g2d)
         struct exynos_drm_subdrv *subdrv = &g2d->subdrv;
 
         kfree(g2d->cmdlist_node);
-        dma_free_attrs(subdrv->drm_dev->dev, G2D_CMDLIST_POOL_SIZE,
-                        g2d->cmdlist_pool_virt,
-                        g2d->cmdlist_pool, &g2d->cmdlist_dma_attrs);
+
+        if (g2d->cmdlist_pool_virt && g2d->cmdlist_pool) {
+                dma_free_attrs(subdrv->drm_dev->dev, G2D_CMDLIST_POOL_SIZE,
+                                g2d->cmdlist_pool_virt,
+                                g2d->cmdlist_pool, &g2d->cmdlist_dma_attrs);
+        }
 }
 
 static struct g2d_cmdlist_node *g2d_get_cmdlist(struct g2d_data *g2d)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index d565207040a2..50faf913e574 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -630,7 +630,6 @@ static int vidi_remove(struct platform_device *pdev)
 {
         struct exynos_drm_manager *mgr = platform_get_drvdata(pdev);
         struct vidi_context *ctx = mgr->ctx;
-        struct drm_encoder *encoder = ctx->encoder;
 
         if (ctx->raw_edid != (struct edid *)fake_edid_info) {
                 kfree(ctx->raw_edid);
@@ -639,9 +638,6 @@ static int vidi_remove(struct platform_device *pdev)
                 return -EINVAL;
         }
 
-        encoder->funcs->destroy(encoder);
-        drm_connector_cleanup(&ctx->connector);
-
         return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 7910fb37d9bb..563a19e62eb2 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -2312,12 +2312,6 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 
 static void hdmi_unbind(struct device *dev, struct device *master, void *data)
 {
-        struct exynos_drm_display *display = get_hdmi_display(dev);
-        struct drm_encoder *encoder = display->encoder;
-        struct hdmi_context *hdata = display->ctx;
-
-        hdmi_connector_destroy(&hdata->connector);
-        encoder->funcs->destroy(encoder);
 }
 
 static const struct component_ops hdmi_component_ops = {
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 1403b01e8216..318ade9bb5af 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1670,15 +1670,17 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
                 goto out_regs;
 
         if (drm_core_check_feature(dev, DRIVER_MODESET)) {
-                ret = i915_kick_out_vgacon(dev_priv);
+                /* WARNING: Apparently we must kick fbdev drivers before vgacon,
+                 * otherwise the vga fbdev driver falls over. */
+                ret = i915_kick_out_firmware_fb(dev_priv);
                 if (ret) {
-                        DRM_ERROR("failed to remove conflicting VGA console\n");
+                        DRM_ERROR("failed to remove conflicting framebuffer drivers\n");
                         goto out_gtt;
                 }
 
-                ret = i915_kick_out_firmware_fb(dev_priv);
+                ret = i915_kick_out_vgacon(dev_priv);
                 if (ret) {
-                        DRM_ERROR("failed to remove conflicting framebuffer drivers\n");
+                        DRM_ERROR("failed to remove conflicting VGA console\n");
                         goto out_gtt;
                 }
         }
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 055d5e7fbf12..2318b4c7a8f8 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -986,6 +986,15 @@ static int i915_pm_freeze(struct device *dev)
         return i915_drm_freeze(drm_dev);
 }
 
+static int i915_pm_freeze_late(struct device *dev)
+{
+        struct pci_dev *pdev = to_pci_dev(dev);
+        struct drm_device *drm_dev = pci_get_drvdata(pdev);
+        struct drm_i915_private *dev_priv = drm_dev->dev_private;
+
+        return intel_suspend_complete(dev_priv);
+}
+
 static int i915_pm_thaw_early(struct device *dev)
 {
         struct pci_dev *pdev = to_pci_dev(dev);
@@ -1570,6 +1579,7 @@ static const struct dev_pm_ops i915_pm_ops = {
         .resume_early = i915_pm_resume_early,
         .resume = i915_pm_resume,
         .freeze = i915_pm_freeze,
+        .freeze_late = i915_pm_freeze_late,
         .thaw_early = i915_pm_thaw_early,
         .thaw = i915_pm_thaw,
         .poweroff = i915_pm_poweroff,
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index b672b843fd5e..728938f02341 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1902,6 +1902,22 @@ static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv)
               GEN8_PPAT(6, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(2)) |
               GEN8_PPAT(7, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(3));
 
+        if (!USES_PPGTT(dev_priv->dev))
+                /* Spec: "For GGTT, there is NO pat_sel[2:0] from the entry,
+                 * so RTL will always use the value corresponding to
+                 * pat_sel = 000".
+                 * So let's disable cache for GGTT to avoid screen corruptions.
+                 * MOCS still can be used though.
+                 * - System agent ggtt writes (i.e. cpu gtt mmaps) already work
+                 * before this patch, i.e. the same uncached + snooping access
+                 * like on gen6/7 seems to be in effect.
+                 * - So this just fixes blitter/render access. Again it looks
+                 * like it's not just uncached access, but uncached + snooping.
+                 * So we can still hold onto all our assumptions wrt cpu
+                 * clflushing on LLC machines.
+                 */
+                pat = GEN8_PPAT(0, GEN8_PPAT_UC);
+
         /* XXX: spec defines this as 2 distinct registers. It's unclear if a 64b
          * write would work. */
         I915_WRITE(GEN8_PRIVATE_PAT, pat);
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 2cefb597df6d..2b1eaa29ada4 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -364,22 +364,9 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
                  * has to also include the unfenced register the GPU uses
                  * whilst executing a fenced command for an untiled object.
                  */
-
-                obj->map_and_fenceable =
-                        !i915_gem_obj_ggtt_bound(obj) ||
-                        (i915_gem_obj_ggtt_offset(obj) +
-                         obj->base.size <= dev_priv->gtt.mappable_end &&
-                         i915_gem_object_fence_ok(obj, args->tiling_mode));
-
-                /* Rebind if we need a change of alignment */
-                if (!obj->map_and_fenceable) {
-                        u32 unfenced_align =
-                                i915_gem_get_gtt_alignment(dev, obj->base.size,
-                                                            args->tiling_mode,
-                                                            false);
-                        if (i915_gem_obj_ggtt_offset(obj) & (unfenced_align - 1))
-                                ret = i915_gem_object_ggtt_unbind(obj);
-                }
+                if (obj->map_and_fenceable &&
+                    !i915_gem_object_fence_ok(obj, args->tiling_mode))
+                        ret = i915_gem_object_ggtt_unbind(obj);
 
                 if (ret == 0) {
                         obj->fence_dirty =
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 3201986bf25e..f66392b6e287 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1711,7 +1711,7 @@ static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
 #define HPD_STORM_DETECT_PERIOD 1000
 #define HPD_STORM_THRESHOLD 5
 
-static int ilk_port_to_hotplug_shift(enum port port)
+static int pch_port_to_hotplug_shift(enum port port)
 {
         switch (port) {
         case PORT_A:
@@ -1727,7 +1727,7 @@ static int ilk_port_to_hotplug_shift(enum port port)
         }
 }
 
-static int g4x_port_to_hotplug_shift(enum port port)
+static int i915_port_to_hotplug_shift(enum port port)
 {
         switch (port) {
         case PORT_A:
@@ -1785,12 +1785,12 @@ static inline void intel_hpd_irq_handler(struct drm_device *dev,
                 if (port && dev_priv->hpd_irq_port[port]) {
                         bool long_hpd;
 
-                        if (IS_G4X(dev)) {
-                                dig_shift = g4x_port_to_hotplug_shift(port);
-                                long_hpd = (hotplug_trigger >> dig_shift) & PORTB_HOTPLUG_LONG_DETECT;
-                        } else {
-                                dig_shift = ilk_port_to_hotplug_shift(port);
+                        if (HAS_PCH_SPLIT(dev)) {
+                                dig_shift = pch_port_to_hotplug_shift(port);
                                 long_hpd = (dig_hotplug_reg >> dig_shift) & PORTB_HOTPLUG_LONG_DETECT;
+                        } else {
+                                dig_shift = i915_port_to_hotplug_shift(port);
+                                long_hpd = (hotplug_trigger >> dig_shift) & PORTB_HOTPLUG_LONG_DETECT;
                         }
 
                         DRM_DEBUG_DRIVER("digital hpd port %c - %s\n",
@@ -3458,12 +3458,13 @@ static void gen8_irq_reset(struct drm_device *dev)
 void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv)
 {
         unsigned long irqflags;
+        uint32_t extra_ier = GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN;
 
         spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
         GEN8_IRQ_INIT_NDX(DE_PIPE, PIPE_B, dev_priv->de_irq_mask[PIPE_B],
-                          ~dev_priv->de_irq_mask[PIPE_B]);
+                          ~dev_priv->de_irq_mask[PIPE_B] | extra_ier);
         GEN8_IRQ_INIT_NDX(DE_PIPE, PIPE_C, dev_priv->de_irq_mask[PIPE_C],
-                          ~dev_priv->de_irq_mask[PIPE_C]);
+                          ~dev_priv->de_irq_mask[PIPE_C] | extra_ier);
         spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 507370513f3d..9cb5c95d5898 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -73,9 +73,6 @@ static const uint32_t intel_cursor_formats[] = {
         DRM_FORMAT_ARGB8888,
 };
 
-#define DIV_ROUND_CLOSEST_ULL(ll, d)    \
-({ unsigned long long _tmp = (ll)+(d)/2; do_div(_tmp, d); _tmp; })
-
 static void intel_increase_pllclock(struct drm_device *dev,
                                     enum pipe pipe);
 static void intel_crtc_update_cursor(struct drm_crtc *crtc, bool on);
@@ -4328,7 +4325,6 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
                 ironlake_fdi_disable(crtc);
 
                 ironlake_disable_pch_transcoder(dev_priv, pipe);
-                intel_set_pch_fifo_underrun_reporting(dev, pipe, true);
 
                 if (HAS_PCH_CPT(dev)) {
                         /* disable TRANS_DP_CTL */
@@ -4392,7 +4388,6 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 
         if (intel_crtc->config.has_pch_encoder) {
                 lpt_disable_pch_transcoder(dev_priv);
-                intel_set_pch_fifo_underrun_reporting(dev, TRANSCODER_A, true);
                 intel_ddi_fdi_disable(crtc);
         }
 
@@ -4588,7 +4583,7 @@ static void vlv_update_cdclk(struct drm_device *dev)
          * BSpec erroneously claims we should aim for 4MHz, but
          * in fact 1MHz is the correct frequency.
          */
-        I915_WRITE(GMBUSFREQ_VLV, dev_priv->vlv_cdclk_freq);
+        I915_WRITE(GMBUSFREQ_VLV, DIV_ROUND_UP(dev_priv->vlv_cdclk_freq, 1000));
 }
 
 /* Adjust CDclk dividers to allow high res or save power if possible */
@@ -9411,6 +9406,10 @@ static bool page_flip_finished(struct intel_crtc *crtc)
         struct drm_device *dev = crtc->base.dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
 
+        if (i915_reset_in_progress(&dev_priv->gpu_error) ||
+            crtc->reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
+                return true;
+
         /*
          * The relevant registers doen't exist on pre-ctg.
          * As the flip done interrupt doesn't trigger for mmio
@@ -12357,27 +12356,36 @@ static void intel_setup_outputs(struct drm_device *dev)
                 if (I915_READ(PCH_DP_D) & DP_DETECTED)
                         intel_dp_init(dev, PCH_DP_D, PORT_D);
         } else if (IS_VALLEYVIEW(dev)) {
-                if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIB) & SDVO_DETECTED) {
+                /*
+                 * The DP_DETECTED bit is the latched state of the DDC
+                 * SDA pin at boot. However since eDP doesn't require DDC
+                 * (no way to plug in a DP->HDMI dongle) the DDC pins for
+                 * eDP ports may have been muxed to an alternate function.
+                 * Thus we can't rely on the DP_DETECTED bit alone to detect
+                 * eDP ports. Consult the VBT as well as DP_DETECTED to
+                 * detect eDP ports.
+                 */
+                if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIB) & SDVO_DETECTED)
                         intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIB,
                                         PORT_B);
-                        if (I915_READ(VLV_DISPLAY_BASE + DP_B) & DP_DETECTED)
-                                intel_dp_init(dev, VLV_DISPLAY_BASE + DP_B, PORT_B);
-                }
+                if (I915_READ(VLV_DISPLAY_BASE + DP_B) & DP_DETECTED ||
+                    intel_dp_is_edp(dev, PORT_B))
+                        intel_dp_init(dev, VLV_DISPLAY_BASE + DP_B, PORT_B);
 
-                if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIC) & SDVO_DETECTED) {
+                if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIC) & SDVO_DETECTED)
                         intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIC,
                                         PORT_C);
-                        if (I915_READ(VLV_DISPLAY_BASE + DP_C) & DP_DETECTED)
-                                intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C, PORT_C);
-                }
+                if (I915_READ(VLV_DISPLAY_BASE + DP_C) & DP_DETECTED ||
+                    intel_dp_is_edp(dev, PORT_C))
+                        intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C, PORT_C);
 
                 if (IS_CHERRYVIEW(dev)) {
-                        if (I915_READ(VLV_DISPLAY_BASE + CHV_HDMID) & SDVO_DETECTED) {
+                        if (I915_READ(VLV_DISPLAY_BASE + CHV_HDMID) & SDVO_DETECTED)
                                 intel_hdmi_init(dev, VLV_DISPLAY_BASE + CHV_HDMID,
                                                 PORT_D);
-                                if (I915_READ(VLV_DISPLAY_BASE + DP_D) & DP_DETECTED)
-                                        intel_dp_init(dev, VLV_DISPLAY_BASE + DP_D, PORT_D);
-                        }
+                        /* eDP not supported on port D, so don't check VBT */
+                        if (I915_READ(VLV_DISPLAY_BASE + DP_D) & DP_DETECTED)
+                                intel_dp_init(dev, VLV_DISPLAY_BASE + DP_D, PORT_D);
                 }
 
                 intel_dsi_init(dev);
@@ -12879,6 +12887,9 @@ static struct intel_quirk intel_quirks[] = {
         /* Acer C720 Chromebook (Core i3 4005U) */
         { 0x0a16, 0x1025, 0x0a11, quirk_backlight_present },
 
+        /* Apple Macbook 2,1 (Core 2 T7400) */
+        { 0x27a2, 0x8086, 0x7270, quirk_backlight_present },
+
         /* Toshiba CB35 Chromebook (Celeron 2955U) */
         { 0x0a06, 0x1179, 0x0a88, quirk_backlight_present },
 
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index f6a3fdd5589e..4bcd91757321 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -2806,6 +2806,13 @@ intel_dp_dpcd_read_wake(struct drm_dp_aux *aux, unsigned int offset,
         ssize_t ret;
         int i;
 
+        /*
+         * Sometime we just get the same incorrect byte repeated
+         * over the entire buffer. Doing just one throw away read
+         * initially seems to "solve" it.
+         */
+        drm_dp_dpcd_read(aux, DP_DPCD_REV, buffer, 1);
+
         for (i = 0; i < 3; i++) {
                 ret = drm_dp_dpcd_read(aux, offset, buffer, size);
                 if (ret == size)
@@ -3724,9 +3731,10 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
                 }
         }
 
-        /* Training Pattern 3 support */
+        /* Training Pattern 3 support, both source and sink */
         if (intel_dp->dpcd[DP_DPCD_REV] >= 0x12 &&
-            intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_TPS3_SUPPORTED) {
+            intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_TPS3_SUPPORTED &&
+            (IS_HASWELL(dev_priv) || INTEL_INFO(dev_priv)->gen >= 8)) {
                 intel_dp->use_tps3 = true;
                 DRM_DEBUG_KMS("Displayport TPS3 supported\n");
         } else
@@ -4442,6 +4450,7 @@ static void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder)
          * vdd might still be enabled do to the delayed vdd off.
          * Make sure vdd is actually turned off here.
          */
+        cancel_delayed_work_sync(&intel_dp->panel_vdd_work);
         pps_lock(intel_dp);
         edp_panel_vdd_off_sync(intel_dp);
         pps_unlock(intel_dp);
@@ -4491,6 +4500,18 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
         if (intel_dig_port->base.type != INTEL_OUTPUT_EDP)
                 intel_dig_port->base.type = INTEL_OUTPUT_DISPLAYPORT;
 
+        if (long_hpd && intel_dig_port->base.type == INTEL_OUTPUT_EDP) {
+                /*
+                 * vdd off can generate a long pulse on eDP which
+                 * would require vdd on to handle it, and thus we
+                 * would end up in an endless cycle of
+                 * "vdd off -> long hpd -> vdd on -> detect -> vdd off -> ..."
+                 */
+                DRM_DEBUG_KMS("ignoring long hpd on eDP port %c\n",
+                              port_name(intel_dig_port->port));
+                return false;
+        }
+
         DRM_DEBUG_KMS("got hpd irq on port %c - %s\n",
                       port_name(intel_dig_port->port),
                       long_hpd ? "long" : "short");
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 07ce04683c30..ba715229a540 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -35,6 +35,9 @@
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_dp_mst_helper.h>
 
+#define DIV_ROUND_CLOSEST_ULL(ll, d)    \
+({ unsigned long long _tmp = (ll)+(d)/2; do_div(_tmp, d); _tmp; })
+
 /**
  * _wait_for - magic (register) wait macro
  *
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index a6bd1422e38f..c0bbf2172446 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -899,6 +899,17 @@ void intel_lvds_init(struct drm_device *dev)
         int pipe;
         u8 pin;
 
+        /*
+         * Unlock registers and just leave them unlocked. Do this before
+         * checking quirk lists to avoid bogus WARNINGs.
+         */
+        if (HAS_PCH_SPLIT(dev)) {
+                I915_WRITE(PCH_PP_CONTROL,
+                           I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS);
+        } else {
+                I915_WRITE(PP_CONTROL,
+                           I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS);
+        }
         if (!intel_lvds_supported(dev))
                 return;
 
@@ -1097,17 +1108,6 @@ out:
         lvds_encoder->a3_power = I915_READ(lvds_encoder->reg) &
                                  LVDS_A3_POWER_MASK;
 
-        /*
-         * Unlock registers and just
-         * leave them unlocked
-         */
-        if (HAS_PCH_SPLIT(dev)) {
-                I915_WRITE(PCH_PP_CONTROL,
-                           I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS);
-        } else {
-                I915_WRITE(PP_CONTROL,
-                           I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS);
-        }
         lvds_connector->lid_notifier.notifier_call = intel_lid_notify;
         if (acpi_lid_notifier_register(&lvds_connector->lid_notifier)) {
                 DRM_DEBUG_KMS("lid notifier registration failed\n");
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 18784470a760..41b3be217493 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -419,9 +419,8 @@ static uint32_t scale(uint32_t source_val,
         source_val = clamp(source_val, source_min, source_max);
 
         /* avoid overflows */
-        target_val = (uint64_t)(source_val - source_min) *
-                (target_max - target_min);
-        do_div(target_val, source_max - source_min);
+        target_val = DIV_ROUND_CLOSEST_ULL((uint64_t)(source_val - source_min) *
+                        (target_max - target_min), source_max - source_min);
         target_val += target_min;
 
         return target_val;
@@ -1099,12 +1098,25 @@ static u32 get_backlight_min_vbt(struct intel_connector *connector)
         struct drm_device *dev = connector->base.dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_panel *panel = &connector->panel;
+        int min;
 
         WARN_ON(panel->backlight.max == 0);
 
+        /*
+         * XXX: If the vbt value is 255, it makes min equal to max, which leads
+         * to problems. There are such machines out there. Either our
+         * interpretation is wrong or the vbt has bogus data. Or both. Safeguard
+         * against this by letting the minimum be at most (arbitrarily chosen)
+         * 25% of the max.
+         */
+        min = clamp_t(int, dev_priv->vbt.backlight.min_brightness, 0, 64);
+        if (min != dev_priv->vbt.backlight.min_brightness) {
+                DRM_DEBUG_KMS("clamping VBT min backlight %d/255 to %d/255\n",
+                              dev_priv->vbt.backlight.min_brightness, min);
+        }
+
         /* vbt value is a coefficient in range [0..255] */
-        return scale(dev_priv->vbt.backlight.min_brightness, 0, 255,
-                     0, panel->backlight.max);
+        return scale(min, 0, 255, 0, panel->backlight.max);
 }
 
 static int bdw_setup_backlight(struct intel_connector *connector)
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index c27b6140bfd1..ad2fd605f76b 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -5469,11 +5469,6 @@ static void gen6_init_clock_gating(struct drm_device *dev)
         I915_WRITE(_3D_CHICKEN,
                    _MASKED_BIT_ENABLE(_3D_CHICKEN_HIZ_PLANE_DISABLE_MSAA_4X_SNB));
 
-        /* WaSetupGtModeTdRowDispatch:snb */
-        if (IS_SNB_GT1(dev))
-                I915_WRITE(GEN6_GT_MODE,
-                           _MASKED_BIT_ENABLE(GEN6_TD_FOUR_ROW_DISPATCH_DISABLE));
-
         /* WaDisable_RenderCache_OperationalFlush:snb */
         I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
index cd05677ad4b7..72a40f95d048 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
@@ -218,7 +218,6 @@ nvc0_identify(struct nouveau_device *device)
                 device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
                 device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
                 device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
-                device->oclass[NVDEV_ENGINE_COPY1  ] = &nvc0_copy1_oclass;
                 device->oclass[NVDEV_ENGINE_DISP   ] =  nva3_disp_oclass;
                 device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
                 break;
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
index 5ae6a43893b5..1931057f9962 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
@@ -551,8 +551,8 @@ nv04_fifo_intr(struct nouveau_subdev *subdev)
                         }
 
                         if (status & 0x40000000) {
-                                nouveau_fifo_uevent(&priv->base);
                                 nv_wr32(priv, 0x002100, 0x40000000);
+                                nouveau_fifo_uevent(&priv->base);
                                 status &= ~0x40000000;
                         }
                 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
index 1fe1f8fbda0c..074d434c3077 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
@@ -740,6 +740,8 @@ nvc0_fifo_intr_engine_unit(struct nvc0_fifo_priv *priv, int engn)
         u32 inte = nv_rd32(priv, 0x002628);
         u32 unkn;
 
+        nv_wr32(priv, 0x0025a8 + (engn * 0x04), intr);
+
         for (unkn = 0; unkn < 8; unkn++) {
                 u32 ints = (intr >> (unkn * 0x04)) & inte;
                 if (ints & 0x1) {
@@ -751,8 +753,6 @@ nvc0_fifo_intr_engine_unit(struct nvc0_fifo_priv *priv, int engn)
                         nv_mask(priv, 0x002628, ints, 0);
                 }
         }
-
-        nv_wr32(priv, 0x0025a8 + (engn * 0x04), intr);
 }
 
 static void
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
index d2f0fd39c145..f8734eb74eaa 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
@@ -952,8 +952,8 @@ nve0_fifo_intr(struct nouveau_subdev *subdev)
         }
 
         if (stat & 0x80000000) {
-                nve0_fifo_intr_engine(priv);
                 nv_wr32(priv, 0x002100, 0x80000000);
+                nve0_fifo_intr_engine(priv);
                 stat &= ~0x80000000;
         }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv50.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv50.c
index 552fdbd45ebe..1d0e33fb5f61 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv50.c
@@ -113,6 +113,8 @@
 #define IS_NVA3F(x) (((x) > 0xa0 && (x) < 0xaa) || (x) == 0xaf)
 #define IS_NVAAF(x) ((x) >= 0xaa && (x) <= 0xac)
 
+#include <subdev/fb.h>
+
 /*
  * This code deals with PGRAPH contexts on NV50 family cards. Like NV40, it's
  * the GPU itself that does context-switching, but it needs a special
@@ -569,8 +571,12 @@ nv50_graph_construct_mmio(struct nouveau_grctx *ctx)
                         gr_def(ctx, 0x407d08, 0x00010040);
                 else if (device->chipset < 0xa0)
                         gr_def(ctx, 0x407d08, 0x00390040);
-                else
-                        gr_def(ctx, 0x407d08, 0x003d0040);
+                else {
+                        if (nouveau_fb(device)->ram->type != NV_MEM_TYPE_GDDR5)
+                                gr_def(ctx, 0x407d08, 0x003d0040);
+                        else
+                                gr_def(ctx, 0x407d08, 0x003c0040);
+                }
                 gr_def(ctx, 0x407d0c, 0x00000022);
         }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c b/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c
index a16024a74771..fde42e4d1b56 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c
@@ -27,6 +27,20 @@ struct gk20a_fb_priv {
 };
 
 static int
+gk20a_fb_init(struct nouveau_object *object)
+{
+        struct gk20a_fb_priv *priv = (void *)object;
+        int ret;
+
+        ret = nouveau_fb_init(&priv->base);
+        if (ret)
+                return ret;
+
+        nv_mask(priv, 0x100c80, 0x00000001, 0x00000000); /* 128KiB lpg */
+        return 0;
+}
+
+static int
 gk20a_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
              struct nouveau_oclass *oclass, void *data, u32 size,
              struct nouveau_object **pobject)
@@ -48,7 +62,7 @@ gk20a_fb_oclass = &(struct nouveau_fb_impl) {
         .base.ofuncs = &(struct nouveau_ofuncs) {
                 .ctor = gk20a_fb_ctor,
                 .dtor = _nouveau_fb_dtor,
-                .init = _nouveau_fb_init,
+                .init = gk20a_fb_init,
                 .fini = _nouveau_fb_fini,
         },
         .memtype = nvc0_fb_memtype_valid,
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index 589dbb582da2..fd3dbd59d73e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -400,15 +400,20 @@ nouveau_channel_new(struct nouveau_drm *drm, struct nvif_device *device,
                     struct nouveau_channel **pchan)
 {
         struct nouveau_cli *cli = (void *)nvif_client(&device->base);
+        bool super;
         int ret;
 
+        /* hack until fencenv50 is fixed, and agp access relaxed */
+        super = cli->base.super;
+        cli->base.super = true;
+
         ret = nouveau_channel_ind(drm, device, handle, arg0, pchan);
         if (ret) {
                 NV_PRINTK(debug, cli, "ib channel create, %d\n", ret);
                 ret = nouveau_channel_dma(drm, device, handle, pchan);
                 if (ret) {
                         NV_PRINTK(debug, cli, "dma channel create, %d\n", ret);
-                        return ret;
+                        goto done;
                 }
         }
 
@@ -416,8 +421,9 @@ nouveau_channel_new(struct nouveau_drm *drm, struct nvif_device *device,
         if (ret) {
                 NV_PRINTK(error, cli, "channel failed to initialise, %d\n", ret);
                 nouveau_channel_del(pchan);
-                return ret;
         }
 
-        return 0;
+done:
+        cli->base.super = super;
+        return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 57238076049f..62b97c4eef8d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -629,7 +629,6 @@ int nouveau_pmops_suspend(struct device *dev)
 
         pci_save_state(pdev);
         pci_disable_device(pdev);
-        pci_ignore_hotplug(pdev);
         pci_set_power_state(pdev, PCI_D3hot);
         return 0;
 }
@@ -933,6 +932,7 @@ static int nouveau_pmops_runtime_suspend(struct device *dev)
         ret = nouveau_do_suspend(drm_dev, true);
         pci_save_state(pdev);
         pci_disable_device(pdev);
+        pci_ignore_hotplug(pdev);
         pci_set_power_state(pdev, PCI_D3cold);
         drm_dev->switch_power_state = DRM_SWITCH_POWER_DYNAMIC_OFF;
         return ret;
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 515cd9aebb99..f32a434724e3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -52,20 +52,24 @@ nouveau_fctx(struct nouveau_fence *fence)
         return container_of(fence->base.lock, struct nouveau_fence_chan, lock);
 }
 
-static void
+static int
 nouveau_fence_signal(struct nouveau_fence *fence)
 {
+        int drop = 0;
+
         fence_signal_locked(&fence->base);
         list_del(&fence->head);
+        rcu_assign_pointer(fence->channel, NULL);
 
         if (test_bit(FENCE_FLAG_USER_BITS, &fence->base.flags)) {
                 struct nouveau_fence_chan *fctx = nouveau_fctx(fence);
 
                 if (!--fctx->notify_ref)
-                        nvif_notify_put(&fctx->notify);
+                        drop = 1;
         }
 
         fence_put(&fence->base);
+        return drop;
 }
 
 static struct nouveau_fence *
@@ -88,16 +92,23 @@ nouveau_fence_context_del(struct nouveau_fence_chan *fctx)
 {
         struct nouveau_fence *fence;
 
-        nvif_notify_fini(&fctx->notify);
-
         spin_lock_irq(&fctx->lock);
         while (!list_empty(&fctx->pending)) {
                 fence = list_entry(fctx->pending.next, typeof(*fence), head);
 
-                nouveau_fence_signal(fence);
-                fence->channel = NULL;
+                if (nouveau_fence_signal(fence))
+                        nvif_notify_put(&fctx->notify);
         }
         spin_unlock_irq(&fctx->lock);
+
+        nvif_notify_fini(&fctx->notify);
+        fctx->dead = 1;
+
+        /*
+         * Ensure that all accesses to fence->channel complete before freeing
+         * the channel.
+         */
+        synchronize_rcu();
 }
 
 static void
@@ -112,21 +123,23 @@ nouveau_fence_context_free(struct nouveau_fence_chan *fctx)
         kref_put(&fctx->fence_ref, nouveau_fence_context_put);
 }
 
-static void
+static int
 nouveau_fence_update(struct nouveau_channel *chan, struct nouveau_fence_chan *fctx)
 {
         struct nouveau_fence *fence;
-
+        int drop = 0;
         u32 seq = fctx->read(chan);
 
         while (!list_empty(&fctx->pending)) {
                 fence = list_entry(fctx->pending.next, typeof(*fence), head);
 
                 if ((int)(seq - fence->base.seqno) < 0)
-                        return;
+                        break;
 
-                nouveau_fence_signal(fence);
+                drop |= nouveau_fence_signal(fence);
         }
+
+        return drop;
 }
 
 static int
@@ -135,18 +148,21 @@ nouveau_fence_wait_uevent_handler(struct nvif_notify *notify)
         struct nouveau_fence_chan *fctx =
                 container_of(notify, typeof(*fctx), notify);
         unsigned long flags;
+        int ret = NVIF_NOTIFY_KEEP;
 
         spin_lock_irqsave(&fctx->lock, flags);
         if (!list_empty(&fctx->pending)) {
                 struct nouveau_fence *fence;
+                struct nouveau_channel *chan;
 
                 fence = list_entry(fctx->pending.next, typeof(*fence), head);
-                nouveau_fence_update(fence->channel, fctx);
+                chan = rcu_dereference_protected(fence->channel, lockdep_is_held(&fctx->lock));
+                if (nouveau_fence_update(fence->channel, fctx))
+                        ret = NVIF_NOTIFY_DROP;
         }
         spin_unlock_irqrestore(&fctx->lock, flags);
 
-        /* Always return keep here. NVIF refcount is handled with nouveau_fence_update */
-        return NVIF_NOTIFY_KEEP;
+        return ret;
 }
 
 void
@@ -262,7 +278,10 @@ nouveau_fence_emit(struct nouveau_fence *fence, struct nouveau_channel *chan)
         if (!ret) {
                 fence_get(&fence->base);
                 spin_lock_irq(&fctx->lock);
-                nouveau_fence_update(chan, fctx);
+
+                if (nouveau_fence_update(chan, fctx))
+                        nvif_notify_put(&fctx->notify);
+
                 list_add_tail(&fence->head, &fctx->pending);
                 spin_unlock_irq(&fctx->lock);
         }
@@ -276,13 +295,16 @@ nouveau_fence_done(struct nouveau_fence *fence)
         if (fence->base.ops == &nouveau_fence_ops_legacy ||
             fence->base.ops == &nouveau_fence_ops_uevent) {
                 struct nouveau_fence_chan *fctx = nouveau_fctx(fence);
+                struct nouveau_channel *chan;
                 unsigned long flags;
 
                 if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->base.flags))
                         return true;
 
                 spin_lock_irqsave(&fctx->lock, flags);
-                nouveau_fence_update(fence->channel, fctx);
+                chan = rcu_dereference_protected(fence->channel, lockdep_is_held(&fctx->lock));
+                if (chan && nouveau_fence_update(chan, fctx))
+                        nvif_notify_put(&fctx->notify);
                 spin_unlock_irqrestore(&fctx->lock, flags);
         }
         return fence_is_signaled(&fence->base);
@@ -387,12 +409,18 @@ nouveau_fence_sync(struct nouveau_bo *nvbo, struct nouveau_channel *chan, bool e
 
         if (fence && (!exclusive || !fobj || !fobj->shared_count)) {
                 struct nouveau_channel *prev = NULL;
+                bool must_wait = true;
 
                 f = nouveau_local_fence(fence, chan->drm);
-                if (f)
-                        prev = f->channel;
+                if (f) {
+                        rcu_read_lock();
+                        prev = rcu_dereference(f->channel);
+                        if (prev && (prev == chan || fctx->sync(f, prev, chan) == 0))
+                                must_wait = false;
+                        rcu_read_unlock();
+                }
 
-                if (!prev || (prev != chan && (ret = fctx->sync(f, prev, chan))))
+                if (must_wait)
                         ret = fence_wait(fence, intr);
 
                 return ret;
@@ -403,19 +431,22 @@ nouveau_fence_sync(struct nouveau_bo *nvbo, struct nouveau_channel *chan, bool e
 
         for (i = 0; i < fobj->shared_count && !ret; ++i) {
                 struct nouveau_channel *prev = NULL;
+                bool must_wait = true;
 
                 fence = rcu_dereference_protected(fobj->shared[i],
                                                 reservation_object_held(resv));
 
                 f = nouveau_local_fence(fence, chan->drm);
-                if (f)
-                        prev = f->channel;
+                if (f) {
+                        rcu_read_lock();
+                        prev = rcu_dereference(f->channel);
+                        if (prev && (prev == chan || fctx->sync(f, prev, chan) == 0))
+                                must_wait = false;
+                        rcu_read_unlock();
+                }
 
-                if (!prev || (prev != chan && (ret = fctx->sync(f, prev, chan))))
+                if (must_wait)
                         ret = fence_wait(fence, intr);
-
-                if (ret)
-                        break;
         }
 
         return ret;
@@ -463,7 +494,7 @@ static const char *nouveau_fence_get_timeline_name(struct fence *f)
         struct nouveau_fence *fence = from_fence(f);
         struct nouveau_fence_chan *fctx = nouveau_fctx(fence);
 
-        return fence->channel ? fctx->name : "dead channel";
+        return !fctx->dead ? fctx->name : "dead channel";
 }
 
 /*
@@ -476,9 +507,16 @@ static bool nouveau_fence_is_signaled(struct fence *f)
 {
         struct nouveau_fence *fence = from_fence(f);
         struct nouveau_fence_chan *fctx = nouveau_fctx(fence);
-        struct nouveau_channel *chan = fence->channel;
+        struct nouveau_channel *chan;
+        bool ret = false;
+
+        rcu_read_lock();
+        chan = rcu_dereference(fence->channel);
+        if (chan)
+                ret = (int)(fctx->read(chan) - fence->base.seqno) >= 0;
+        rcu_read_unlock();
 
-        return (int)(fctx->read(chan) - fence->base.seqno) >= 0;
+        return ret;
 }
 
 static bool nouveau_fence_no_signaling(struct fence *f)
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h
index 943b0b17b1fc..96e461c6f68f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.h
@@ -14,7 +14,7 @@ struct nouveau_fence {
 
         bool sysmem;
 
-        struct nouveau_channel *channel;
+        struct nouveau_channel __rcu *channel;
         unsigned long timeout;
 };
 
@@ -47,7 +47,7 @@ struct nouveau_fence_chan {
         char name[32];
 
         struct nvif_notify notify;
-        int notify_ref;
+        int notify_ref, dead;
 };
 
 struct nouveau_fence_priv {
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index ae873d1a8d46..eb8b36714fa1 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -791,6 +791,22 @@ nv50_crtc_set_scale(struct nouveau_crtc *nv_crtc, bool update)
 }
 
 static int
+nv50_crtc_set_raster_vblank_dmi(struct nouveau_crtc *nv_crtc, u32 usec)
+{
+        struct nv50_mast *mast = nv50_mast(nv_crtc->base.dev);
+        u32 *push;
+
+        push = evo_wait(mast, 8);
+        if (!push)
+                return -ENOMEM;
+
+        evo_mthd(push, 0x0828 + (nv_crtc->index * 0x400), 1);
+        evo_data(push, usec);
+        evo_kick(push, mast);
+        return 0;
+}
+
+static int
 nv50_crtc_set_color_vibrance(struct nouveau_crtc *nv_crtc, bool update)
 {
         struct nv50_mast *mast = nv50_mast(nv_crtc->base.dev);
@@ -1104,14 +1120,14 @@ nv50_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *umode,
                         evo_mthd(push, 0x0804 + (nv_crtc->index * 0x400), 2);
                         evo_data(push, 0x00800000 | mode->clock);
                         evo_data(push, (ilace == 2) ? 2 : 0);
-                        evo_mthd(push, 0x0810 + (nv_crtc->index * 0x400), 8);
+                        evo_mthd(push, 0x0810 + (nv_crtc->index * 0x400), 6);
                         evo_data(push, 0x00000000);
                         evo_data(push, (vactive << 16) | hactive);
                         evo_data(push, ( vsynce << 16) | hsynce);
                         evo_data(push, (vblanke << 16) | hblanke);
                         evo_data(push, (vblanks << 16) | hblanks);
                         evo_data(push, (vblan2e << 16) | vblan2s);
-                        evo_data(push, vblankus);
+                        evo_mthd(push, 0x082c + (nv_crtc->index * 0x400), 1);
                         evo_data(push, 0x00000000);
                         evo_mthd(push, 0x0900 + (nv_crtc->index * 0x400), 2);
                         evo_data(push, 0x00000311);
@@ -1141,6 +1157,11 @@ nv50_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *umode,
         nv_connector = nouveau_crtc_connector_get(nv_crtc);
         nv50_crtc_set_dither(nv_crtc, false);
         nv50_crtc_set_scale(nv_crtc, false);
+
+        /* G94 only accepts this after setting scale */
+        if (nv50_vers(mast) < GF110_DISP_CORE_CHANNEL_DMA)
+                nv50_crtc_set_raster_vblank_dmi(nv_crtc, vblankus);
+
         nv50_crtc_set_color_vibrance(nv_crtc, false);
         nv50_crtc_set_image(nv_crtc, crtc->primary->fb, x, y, false);
         return 0;
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index af9e78546688..0d1396266857 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -572,7 +572,6 @@ static int qxl_crtc_mode_set(struct drm_crtc *crtc,
         struct qxl_framebuffer *qfb;
         struct qxl_bo *bo, *old_bo = NULL;
         struct qxl_crtc *qcrtc = to_qxl_crtc(crtc);
-        uint32_t width, height, base_offset;
         bool recreate_primary = false;
         int ret;
         int surf_id;
@@ -602,9 +601,10 @@ static int qxl_crtc_mode_set(struct drm_crtc *crtc,
         if (qcrtc->index == 0)
                 recreate_primary = true;
 
-        width = mode->hdisplay;
-        height = mode->vdisplay;
-        base_offset = 0;
+        if (bo->surf.stride * bo->surf.height > qdev->vram_size) {
+                DRM_ERROR("Mode doesn't fit in vram size (vgamem)");
+                return -EINVAL;
+        }
 
         ret = qxl_bo_reserve(bo, false);
         if (ret != 0)
@@ -618,10 +618,10 @@ static int qxl_crtc_mode_set(struct drm_crtc *crtc,
         if (recreate_primary) {
                 qxl_io_destroy_primary(qdev);
                 qxl_io_log(qdev,
-                           "recreate primary: %dx%d (was %dx%d,%d,%d)\n",
-                           width, height, bo->surf.width,
-                           bo->surf.height, bo->surf.stride, bo->surf.format);
-                qxl_io_create_primary(qdev, base_offset, bo);
+                           "recreate primary: %dx%d,%d,%d\n",
+                           bo->surf.width, bo->surf.height,
+                           bo->surf.stride, bo->surf.format);
+                qxl_io_create_primary(qdev, 0, bo);
                 bo->is_primary = true;
         }
 
diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c
index 15da7ef344a4..ec1593a6a561 100644
--- a/drivers/gpu/drm/radeon/atom.c
+++ b/drivers/gpu/drm/radeon/atom.c
@@ -1217,7 +1217,7 @@ free:
         return ret;
 }
 
-int atom_execute_table(struct atom_context *ctx, int index, uint32_t * params)
+int atom_execute_table_scratch_unlocked(struct atom_context *ctx, int index, uint32_t * params)
 {
         int r;
 
@@ -1238,6 +1238,15 @@ int atom_execute_table(struct atom_context *ctx, int index, uint32_t * params)
         return r;
 }
 
+int atom_execute_table(struct atom_context *ctx, int index, uint32_t * params)
+{
+        int r;
+        mutex_lock(&ctx->scratch_mutex);
+        r = atom_execute_table_scratch_unlocked(ctx, index, params);
+        mutex_unlock(&ctx->scratch_mutex);
+        return r;
+}
+
 static int atom_iio_len[] = { 1, 2, 3, 3, 3, 3, 4, 4, 4, 3 };
 
 static void atom_index_iio(struct atom_context *ctx, int base)
diff --git a/drivers/gpu/drm/radeon/atom.h b/drivers/gpu/drm/radeon/atom.h
index feba6b8d36b3..6d014ddb6b78 100644
--- a/drivers/gpu/drm/radeon/atom.h
+++ b/drivers/gpu/drm/radeon/atom.h
@@ -125,6 +125,7 @@ struct card_info {
 struct atom_context {
         struct card_info *card;
         struct mutex mutex;
+        struct mutex scratch_mutex;
         void *bios;
         uint32_t cmd_table, data_table;
         uint16_t *iio;
@@ -145,6 +146,7 @@ extern int atom_debug;
 
 struct atom_context *atom_parse(struct card_info *, void *);
 int atom_execute_table(struct atom_context *, int, uint32_t *);
+int atom_execute_table_scratch_unlocked(struct atom_context *, int, uint32_t *);
 int atom_asic_init(struct atom_context *);
 void atom_destroy(struct atom_context *);
 bool atom_parse_data_header(struct atom_context *ctx, int index, uint16_t *size,
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 95d5d4ab3335..11ba9d21b89b 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -100,6 +100,7 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
         memset(&args, 0, sizeof(args));
 
         mutex_lock(&chan->mutex);
+        mutex_lock(&rdev->mode_info.atom_context->scratch_mutex);
 
         base = (unsigned char *)(rdev->mode_info.atom_context->scratch + 1);
 
@@ -113,7 +114,7 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
         if (ASIC_IS_DCE4(rdev))
                 args.v2.ucHPD_ID = chan->rec.hpd;
 
-        atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+        atom_execute_table_scratch_unlocked(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 
         *ack = args.v1.ucReplyStatus;
 
@@ -147,6 +148,7 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
 
         r = recv_bytes;
 done:
+        mutex_unlock(&rdev->mode_info.atom_context->scratch_mutex);
         mutex_unlock(&chan->mutex);
 
         return r;
diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c
index 9c570fb15b8c..4157780585a0 100644
--- a/drivers/gpu/drm/radeon/atombios_i2c.c
+++ b/drivers/gpu/drm/radeon/atombios_i2c.c
@@ -48,6 +48,7 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
         memset(&args, 0, sizeof(args));
 
         mutex_lock(&chan->mutex);
+        mutex_lock(&rdev->mode_info.atom_context->scratch_mutex);
 
         base = (unsigned char *)rdev->mode_info.atom_context->scratch;
 
@@ -82,7 +83,7 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
         args.ucSlaveAddr = slave_addr << 1;
         args.ucLineNumber = chan->rec.i2c_id;
 
-        atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+        atom_execute_table_scratch_unlocked(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 
         /* error */
         if (args.ucStatus != HW_ASSISTED_I2C_STATUS_SUCCESS) {
@@ -95,6 +96,7 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
                 radeon_atom_copy_swap(buf, base, num, false);
 
 done:
+        mutex_unlock(&rdev->mode_info.atom_context->scratch_mutex);
         mutex_unlock(&chan->mutex);
 
         return r;
diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c
index 300d971187c4..0b2929de9f41 100644
--- a/drivers/gpu/drm/radeon/btc_dpm.c
+++ b/drivers/gpu/drm/radeon/btc_dpm.c
@@ -24,6 +24,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "btcd.h"
 #include "r600_dpm.h"
 #include "cypress_dpm.h"
@@ -1170,6 +1171,23 @@ static const struct radeon_blacklist_clocks btc_blacklist_clocks[] =
         { 25000, 30000, RADEON_SCLK_UP }
 };
 
+void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table,
+                                                     u32 *max_clock)
+{
+        u32 i, clock = 0;
+
+        if ((table == NULL) || (table->count == 0)) {
+                *max_clock = clock;
+                return;
+        }
+
+        for (i = 0; i < table->count; i++) {
+                if (clock < table->entries[i].clk)
+                        clock = table->entries[i].clk;
+        }
+        *max_clock = clock;
+}
+
 void btc_apply_voltage_dependency_rules(struct radeon_clock_voltage_dependency_table *table,
                                         u32 clock, u16 max_voltage, u16 *voltage)
 {
diff --git a/drivers/gpu/drm/radeon/btc_dpm.h b/drivers/gpu/drm/radeon/btc_dpm.h
index 1a15e0e41950..3b6f12b7760b 100644
--- a/drivers/gpu/drm/radeon/btc_dpm.h
+++ b/drivers/gpu/drm/radeon/btc_dpm.h
@@ -46,6 +46,8 @@ void btc_adjust_clock_combinations(struct radeon_device *rdev,
                                    struct rv7xx_pl *pl);
 void btc_apply_voltage_dependency_rules(struct radeon_clock_voltage_dependency_table *table,
                                         u32 clock, u16 max_voltage, u16 *voltage);
+void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table,
+                                                     u32 *max_clock);
 void btc_apply_voltage_delta_rules(struct radeon_device *rdev,
                                    u16 max_vddc, u16 max_vddci,
                                    u16 *vddc, u16 *vddci);
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index f5c8c0445a94..11a55e9dad7f 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -24,6 +24,7 @@
 #include <linux/firmware.h>
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "radeon_ucode.h"
 #include "cikd.h"
 #include "r600_dpm.h"
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 377afa504d2b..89c01fa6dd8e 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -4313,8 +4313,8 @@ static int cik_cp_gfx_start(struct radeon_device *rdev)
         /* init the CE partitions.  CE only used for gfx on CIK */
         radeon_ring_write(ring, PACKET3(PACKET3_SET_BASE, 2));
         radeon_ring_write(ring, PACKET3_BASE_INDEX(CE_PARTITION_BASE));
-        radeon_ring_write(ring, 0xc000);
-        radeon_ring_write(ring, 0xc000);
+        radeon_ring_write(ring, 0x8000);
+        radeon_ring_write(ring, 0x8000);
 
         /* setup clear context state */
         radeon_ring_write(ring, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
@@ -9447,6 +9447,9 @@ void dce8_bandwidth_update(struct radeon_device *rdev)
         u32 num_heads = 0, lb_size;
         int i;
 
+        if (!rdev->mode_info.mode_config_initialized)
+                return;
+
         radeon_update_display_priority(rdev);
 
         for (i = 0; i < rdev->num_crtc; i++) {
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index c77dad1a4576..d748963af08b 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -611,16 +611,19 @@ int cik_sdma_ring_test(struct radeon_device *rdev,
 {
         unsigned i;
         int r;
-        void __iomem *ptr = (void *)rdev->vram_scratch.ptr;
+        unsigned index;
         u32 tmp;
+        u64 gpu_addr;
 
-        if (!ptr) {
-                DRM_ERROR("invalid vram scratch pointer\n");
-                return -EINVAL;
-        }
+        if (ring->idx == R600_RING_TYPE_DMA_INDEX)
+                index = R600_WB_DMA_RING_TEST_OFFSET;
+        else
+                index = CAYMAN_WB_DMA1_RING_TEST_OFFSET;
+
+        gpu_addr = rdev->wb.gpu_addr + index;
 
         tmp = 0xCAFEDEAD;
-        writel(tmp, ptr);
+        rdev->wb.wb[index/4] = cpu_to_le32(tmp);
 
         r = radeon_ring_lock(rdev, ring, 5);
         if (r) {
@@ -628,14 +631,14 @@ int cik_sdma_ring_test(struct radeon_device *rdev,
                 return r;
         }
         radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0));
-        radeon_ring_write(ring, rdev->vram_scratch.gpu_addr & 0xfffffffc);
-        radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr));
+        radeon_ring_write(ring, lower_32_bits(gpu_addr));
+        radeon_ring_write(ring, upper_32_bits(gpu_addr));
         radeon_ring_write(ring, 1); /* number of DWs to follow */
         radeon_ring_write(ring, 0xDEADBEEF);
         radeon_ring_unlock_commit(rdev, ring, false);
 
         for (i = 0; i < rdev->usec_timeout; i++) {
-                tmp = readl(ptr);
+                tmp = le32_to_cpu(rdev->wb.wb[index/4]);
                 if (tmp == 0xDEADBEEF)
                         break;
                 DRM_UDELAY(1);
@@ -664,17 +667,20 @@ int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         struct radeon_ib ib;
         unsigned i;
+        unsigned index;
         int r;
-        void __iomem *ptr = (void *)rdev->vram_scratch.ptr;
         u32 tmp = 0;
+        u64 gpu_addr;
 
-        if (!ptr) {
-                DRM_ERROR("invalid vram scratch pointer\n");
-                return -EINVAL;
-        }
+        if (ring->idx == R600_RING_TYPE_DMA_INDEX)
+                index = R600_WB_DMA_RING_TEST_OFFSET;
+        else
+                index = CAYMAN_WB_DMA1_RING_TEST_OFFSET;
+
+        gpu_addr = rdev->wb.gpu_addr + index;
 
         tmp = 0xCAFEDEAD;
-        writel(tmp, ptr);
+        rdev->wb.wb[index/4] = cpu_to_le32(tmp);
 
         r = radeon_ib_get(rdev, ring->idx, &ib, NULL, 256);
         if (r) {
@@ -683,8 +689,8 @@ int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
         }
 
         ib.ptr[0] = SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
-        ib.ptr[1] = rdev->vram_scratch.gpu_addr & 0xfffffffc;
-        ib.ptr[2] = upper_32_bits(rdev->vram_scratch.gpu_addr);
+        ib.ptr[1] = lower_32_bits(gpu_addr);
+        ib.ptr[2] = upper_32_bits(gpu_addr);
         ib.ptr[3] = 1;
         ib.ptr[4] = 0xDEADBEEF;
         ib.length_dw = 5;
@@ -701,7 +707,7 @@ int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
                 return r;
         }
         for (i = 0; i < rdev->usec_timeout; i++) {
-                tmp = readl(ptr);
+                tmp = le32_to_cpu(rdev->wb.wb[index/4]);
                 if (tmp == 0xDEADBEEF)
                         break;
                 DRM_UDELAY(1);
diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c
index 47d31e915758..9aad0327e4d1 100644
--- a/drivers/gpu/drm/radeon/cypress_dpm.c
+++ b/drivers/gpu/drm/radeon/cypress_dpm.c
@@ -24,6 +24,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "evergreend.h"
 #include "r600_dpm.h"
 #include "cypress_dpm.h"
diff --git a/drivers/gpu/drm/radeon/dce3_1_afmt.c b/drivers/gpu/drm/radeon/dce3_1_afmt.c
index 950af153f30e..2fe8cfc966d9 100644
--- a/drivers/gpu/drm/radeon/dce3_1_afmt.c
+++ b/drivers/gpu/drm/radeon/dce3_1_afmt.c
@@ -32,7 +32,7 @@ static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder)
         struct drm_connector *connector;
         struct radeon_connector *radeon_connector = NULL;
         u32 tmp;
-        u8 *sadb;
+        u8 *sadb = NULL;
         int sad_count;
 
         list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
@@ -49,8 +49,8 @@ static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder)
 
         sad_count = drm_edid_to_speaker_allocation(radeon_connector->edid, &sadb);
         if (sad_count < 0) {
-                DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
-                return;
+                DRM_DEBUG("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
+                sad_count = 0;
         }
 
         /* program the speaker allocation */
diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c
index c0bbf68dbc27..f312edf4d50e 100644
--- a/drivers/gpu/drm/radeon/dce6_afmt.c
+++ b/drivers/gpu/drm/radeon/dce6_afmt.c
@@ -155,7 +155,7 @@ void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder)
         struct drm_connector *connector;
         struct radeon_connector *radeon_connector = NULL;
         u32 offset, tmp;
-        u8 *sadb;
+        u8 *sadb = NULL;
         int sad_count;
 
         if (!dig || !dig->afmt || !dig->afmt->pin)
@@ -176,9 +176,9 @@ void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder)
         }
 
         sad_count = drm_edid_to_speaker_allocation(radeon_connector_edid(connector), &sadb);
-        if (sad_count <= 0) {
-                DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
-                return;
+        if (sad_count < 0) {
+                DRM_DEBUG("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
+                sad_count = 0;
         }
 
         /* program the speaker allocation */
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index a31f1ca40c6a..85995b4e3338 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -2345,6 +2345,9 @@ void evergreen_bandwidth_update(struct radeon_device *rdev)
         u32 num_heads = 0, lb_size;
         int i;
 
+        if (!rdev->mode_info.mode_config_initialized)
+                return;
+
         radeon_update_display_priority(rdev);
 
         for (i = 0; i < rdev->num_crtc; i++) {
@@ -2552,6 +2555,7 @@ void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *sav
                                         WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 1);
                                         tmp |= EVERGREEN_CRTC_BLANK_DATA_EN;
                                         WREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i], tmp);
+                                        WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 0);
                                 }
                         } else {
                                 tmp = RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]);
@@ -3005,7 +3009,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
         u32 vgt_cache_invalidation;
         u32 hdp_host_path_cntl, tmp;
         u32 disabled_rb_mask;
-        int i, j, num_shader_engines, ps_thread_count;
+        int i, j, ps_thread_count;
 
         switch (rdev->family) {
         case CHIP_CYPRESS:
@@ -3303,8 +3307,6 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
         rdev->config.evergreen.tile_config |=
                 ((gb_addr_config & 0x30000000) >> 28) << 12;
 
-        num_shader_engines = (gb_addr_config & NUM_SHADER_ENGINES(3) >> 12) + 1;
-
         if ((rdev->family >= CHIP_CEDAR) && (rdev->family <= CHIP_HEMLOCK)) {
                 u32 efuse_straps_4;
                 u32 efuse_straps_3;
diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c
index 2514d659b1ba..53abd9b17a50 100644
--- a/drivers/gpu/drm/radeon/evergreen_hdmi.c
+++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c
@@ -133,7 +133,7 @@ static void dce4_afmt_write_speaker_allocation(struct drm_encoder *encoder)
         struct drm_connector *connector;
         struct radeon_connector *radeon_connector = NULL;
         u32 tmp;
-        u8 *sadb;
+        u8 *sadb = NULL;
         int sad_count;
 
         list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
@@ -149,9 +149,9 @@ static void dce4_afmt_write_speaker_allocation(struct drm_encoder *encoder)
         }
 
         sad_count = drm_edid_to_speaker_allocation(radeon_connector_edid(connector), &sadb);
-        if (sad_count <= 0) {
-                DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
-                return;
+        if (sad_count < 0) {
+                DRM_DEBUG("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
+                sad_count = 0;
         }
 
         /* program the speaker allocation */
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
index 1dd976f447fa..9b42001295ba 100644
--- a/drivers/gpu/drm/radeon/kv_dpm.c
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -2725,7 +2725,11 @@ int kv_dpm_init(struct radeon_device *rdev)
 
         pi->sram_end = SMC_RAM_END;
 
-        pi->enable_nb_dpm = true;
+        /* Enabling nb dpm on an asrock system prevents dpm from working */
+        if (rdev->pdev->subsystem_vendor == 0x1849)
+                pi->enable_nb_dpm = false;
+        else
+                pi->enable_nb_dpm = true;
 
         pi->caps_power_containment = true;
         pi->caps_cac = true;
@@ -2740,10 +2744,19 @@ 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;
-        if (radeon_bapm == 0)
+        if (radeon_bapm == -1) {
+                /* There are stability issues reported on with
+                 * bapm enabled on an asrock system.
+                 */
+                if (rdev->pdev->subsystem_vendor == 0x1849)
+                        pi->bapm_enable = false;
+                else
+                        pi->bapm_enable = true;
+        } else if (radeon_bapm == 0) {
                 pi->bapm_enable = false;
-        else
+        } 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_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index 715b181c6243..6d2f16cf2c1c 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -23,6 +23,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "nid.h"
 #include "r600_dpm.h"
 #include "ni_dpm.h"
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 10f8be0ee173..b53b31a7b76f 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -3207,6 +3207,9 @@ void r100_bandwidth_update(struct radeon_device *rdev)
         uint32_t pixel_bytes1 = 0;
         uint32_t pixel_bytes2 = 0;
 
+        if (!rdev->mode_info.mode_config_initialized)
+                return;
+
         radeon_update_display_priority(rdev);
 
         if (rdev->mode_info.crtcs[0]->base.enabled) {
diff --git a/drivers/gpu/drm/radeon/r600_dma.c b/drivers/gpu/drm/radeon/r600_dma.c
index 100189ec5fa8..cf0df45d455e 100644
--- a/drivers/gpu/drm/radeon/r600_dma.c
+++ b/drivers/gpu/drm/radeon/r600_dma.c
@@ -232,16 +232,19 @@ int r600_dma_ring_test(struct radeon_device *rdev,
 {
         unsigned i;
         int r;
-        void __iomem *ptr = (void *)rdev->vram_scratch.ptr;
+        unsigned index;
         u32 tmp;
+        u64 gpu_addr;
 
-        if (!ptr) {
-                DRM_ERROR("invalid vram scratch pointer\n");
-                return -EINVAL;
-        }
+        if (ring->idx == R600_RING_TYPE_DMA_INDEX)
+                index = R600_WB_DMA_RING_TEST_OFFSET;
+        else
+                index = CAYMAN_WB_DMA1_RING_TEST_OFFSET;
+
+        gpu_addr = rdev->wb.gpu_addr + index;
 
         tmp = 0xCAFEDEAD;
-        writel(tmp, ptr);
+        rdev->wb.wb[index/4] = cpu_to_le32(tmp);
 
         r = radeon_ring_lock(rdev, ring, 4);
         if (r) {
@@ -249,13 +252,13 @@ int r600_dma_ring_test(struct radeon_device *rdev,
                 return r;
         }
         radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_WRITE, 0, 0, 1));
-        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, lower_32_bits(gpu_addr));
+        radeon_ring_write(ring, upper_32_bits(gpu_addr) & 0xff);
         radeon_ring_write(ring, 0xDEADBEEF);
         radeon_ring_unlock_commit(rdev, ring, false);
 
         for (i = 0; i < rdev->usec_timeout; i++) {
-                tmp = readl(ptr);
+                tmp = le32_to_cpu(rdev->wb.wb[index/4]);
                 if (tmp == 0xDEADBEEF)
                         break;
                 DRM_UDELAY(1);
@@ -335,17 +338,17 @@ int r600_dma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         struct radeon_ib ib;
         unsigned i;
+        unsigned index;
         int r;
-        void __iomem *ptr = (void *)rdev->vram_scratch.ptr;
         u32 tmp = 0;
+        u64 gpu_addr;
 
-        if (!ptr) {
-                DRM_ERROR("invalid vram scratch pointer\n");
-                return -EINVAL;
-        }
+        if (ring->idx == R600_RING_TYPE_DMA_INDEX)
+                index = R600_WB_DMA_RING_TEST_OFFSET;
+        else
+                index = CAYMAN_WB_DMA1_RING_TEST_OFFSET;
 
-        tmp = 0xCAFEDEAD;
-        writel(tmp, ptr);
+        gpu_addr = rdev->wb.gpu_addr + index;
 
         r = radeon_ib_get(rdev, ring->idx, &ib, NULL, 256);
         if (r) {
@@ -354,8 +357,8 @@ int r600_dma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
         }
 
         ib.ptr[0] = DMA_PACKET(DMA_PACKET_WRITE, 0, 0, 1);
-        ib.ptr[1] = rdev->vram_scratch.gpu_addr & 0xfffffffc;
-        ib.ptr[2] = upper_32_bits(rdev->vram_scratch.gpu_addr) & 0xff;
+        ib.ptr[1] = lower_32_bits(gpu_addr);
+        ib.ptr[2] = upper_32_bits(gpu_addr) & 0xff;
         ib.ptr[3] = 0xDEADBEEF;
         ib.length_dw = 4;
 
@@ -371,7 +374,7 @@ int r600_dma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
                 return r;
         }
         for (i = 0; i < rdev->usec_timeout; i++) {
-                tmp = readl(ptr);
+                tmp = le32_to_cpu(rdev->wb.wb[index/4]);
                 if (tmp == 0xDEADBEEF)
                         break;
                 DRM_UDELAY(1);
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index 9c61b74ef441..b5c73df8e202 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -24,6 +24,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "r600d.h"
 #include "r600_dpm.h"
 #include "atom.h"
@@ -1255,7 +1256,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
                                         (mode_info->atom_context->bios + data_offset +
                                          le16_to_cpu(ext_hdr->usPowerTuneTableOffset));
                                 rdev->pm.dpm.dyn_state.cac_tdp_table->maximum_power_delivery_limit =
-                                        ppt->usMaximumPowerDeliveryLimit;
+                                        le16_to_cpu(ppt->usMaximumPowerDeliveryLimit);
                                 pt = &ppt->power_tune_table;
                         } else {
                                 ATOM_PPLIB_POWERTUNE_Table *ppt = (ATOM_PPLIB_POWERTUNE_Table *)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index f7c4b226a284..a9717b3fbf1b 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1133,6 +1133,8 @@ struct radeon_wb {
 #define R600_WB_EVENT_OFFSET     3072
 #define CIK_WB_CP1_WPTR_OFFSET     3328
 #define CIK_WB_CP2_WPTR_OFFSET     3584
+#define R600_WB_DMA_RING_TEST_OFFSET 3588
+#define CAYMAN_WB_DMA1_RING_TEST_OFFSET 3592
 
 /**
  * struct radeon_pm - power management datas
diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c
index 6a03624fadaa..63ccb8fa799c 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -658,12 +658,10 @@ bool radeon_get_bios(struct radeon_device *rdev)
                 r = igp_read_bios_from_vram(rdev);
         if (r == false)
                 r = radeon_read_bios(rdev);
-        if (r == false) {
+        if (r == false)
                 r = radeon_read_disabled_bios(rdev);
-        }
-        if (r == false) {
+        if (r == false)
                 r = radeon_read_platform_bios(rdev);
-        }
         if (r == false || rdev->bios == NULL) {
                 DRM_ERROR("Unable to locate a BIOS ROM\n");
                 rdev->bios = NULL;
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 300c4b3d4669..26baa9c05f6c 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -322,6 +322,12 @@ static void radeon_connector_get_edid(struct drm_connector *connector)
         }
 
         if (!radeon_connector->edid) {
+                /* don't fetch the edid from the vbios if ddc fails and runpm is
+                 * enabled so we report disconnected.
+                 */
+                if ((rdev->flags & RADEON_IS_PX) && (radeon_runtime_pm != 0))
+                        return;
+
                 if (rdev->is_atom_bios) {
                         /* some laptops provide a hardcoded edid in rom for LCDs */
                         if (((connector->connector_type == DRM_MODE_CONNECTOR_LVDS) ||
@@ -826,6 +832,8 @@ static int radeon_lvds_mode_valid(struct drm_connector *connector,
 static enum drm_connector_status
 radeon_lvds_detect(struct drm_connector *connector, bool force)
 {
+        struct drm_device *dev = connector->dev;
+        struct radeon_device *rdev = dev->dev_private;
         struct radeon_connector *radeon_connector = to_radeon_connector(connector);
         struct drm_encoder *encoder = radeon_best_single_encoder(connector);
         enum drm_connector_status ret = connector_status_disconnected;
@@ -842,7 +850,11 @@ radeon_lvds_detect(struct drm_connector *connector, bool force)
                 /* check if panel is valid */
                 if (native_mode->hdisplay >= 320 && native_mode->vdisplay >= 240)
                         ret = connector_status_connected;
-
+                /* don't fetch the edid from the vbios if ddc fails and runpm is
+                 * enabled so we report disconnected.
+                 */
+                if ((rdev->flags & RADEON_IS_PX) && (radeon_runtime_pm != 0))
+                        ret = connector_status_disconnected;
         }
 
         /* check for edid as well */
@@ -1589,6 +1601,11 @@ radeon_dp_detect(struct drm_connector *connector, bool force)
                         /* check if panel is valid */
                         if (native_mode->hdisplay >= 320 && native_mode->vdisplay >= 240)
                                 ret = connector_status_connected;
+                        /* don't fetch the edid from the vbios if ddc fails and runpm is
+                         * enabled so we report disconnected.
+                         */
+                        if ((rdev->flags & RADEON_IS_PX) && (radeon_runtime_pm != 0))
+                                ret = connector_status_disconnected;
                 }
                 /* eDP is always DP */
                 radeon_dig_connector->dp_sink_type = CONNECTOR_OBJECT_ID_DISPLAYPORT;
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 1c893447d7cd..6f377de099f9 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -251,22 +251,19 @@ static int radeon_cs_get_ring(struct radeon_cs_parser *p, u32 ring, s32 priority
 
 static int radeon_cs_sync_rings(struct radeon_cs_parser *p)
 {
-        int i, r = 0;
+        struct radeon_cs_reloc *reloc;
+        int r;
 
-        for (i = 0; i < p->nrelocs; i++) {
+        list_for_each_entry(reloc, &p->validated, tv.head) {
                 struct reservation_object *resv;
 
-                if (!p->relocs[i].robj)
-                        continue;
-
-                resv = p->relocs[i].robj->tbo.resv;
+                resv = reloc->robj->tbo.resv;
                 r = radeon_semaphore_sync_resv(p->rdev, p->ib.semaphore, resv,
-                                               p->relocs[i].tv.shared);
-
+                                               reloc->tv.shared);
                 if (r)
-                        break;
+                        return r;
         }
-        return r;
+        return 0;
 }
 
 /* XXX: note that this is called from the legacy UMS CS ioctl as well */
@@ -450,7 +447,7 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo
         kfree(parser->track);
         kfree(parser->relocs);
         kfree(parser->relocs_ptr);
-        kfree(parser->vm_bos);
+        drm_free_large(parser->vm_bos);
         for (i = 0; i < parser->nchunks; i++)
                 drm_free_large(parser->chunks[i].kdata);
         kfree(parser->chunks);
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index f41cc1538e48..995a8b1770dd 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -952,6 +952,7 @@ int radeon_atombios_init(struct radeon_device *rdev)
         }
 
         mutex_init(&rdev->mode_info.atom_context->mutex);
+        mutex_init(&rdev->mode_info.atom_context->scratch_mutex);
         radeon_atom_initialize_bios_scratch_regs(rdev->ddev);
         atom_allocate_fb_scratch(rdev->mode_info.atom_context);
         return 0;
@@ -1130,7 +1131,7 @@ static void radeon_check_arguments(struct radeon_device *rdev)
         if (radeon_vm_block_size == -1) {
 
                 /* Total bits covered by PD + PTs */
-                unsigned bits = ilog2(radeon_vm_size) + 17;
+                unsigned bits = ilog2(radeon_vm_size) + 18;
 
                 /* Make sure the PD is 4K in size up to 8GB address space.
                    Above that split equal between PD and PTs */
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index 9a19e52cc655..6b670b0bc47b 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -179,6 +179,9 @@ static void radeon_encoder_add_backlight(struct radeon_encoder *radeon_encoder,
                     (rdev->pdev->subsystem_vendor == 0x1734) &&
                     (rdev->pdev->subsystem_device == 0x1107))
                         use_bl = false;
+                /* disable native backlight control on older asics */
+                else if (rdev->family < CHIP_R600)
+                        use_bl = false;
                 else
                         use_bl = true;
         }
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 7784911d78ef..00fc59762e0d 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -185,6 +185,16 @@ static bool radeon_msi_ok(struct radeon_device *rdev)
         if (rdev->flags & RADEON_IS_AGP)
                 return false;
 
+        /*
+         * Older chips have a HW limitation, they can only generate 40 bits
+         * of address for "64-bit" MSIs which breaks on some platforms, notably
+         * IBM POWER servers, so we limit them
+         */
+        if (rdev->family < CHIP_BONAIRE) {
+                dev_info(rdev->dev, "radeon: MSI limited to 32-bit\n");
+                rdev->pdev->no_64bit_msi = 1;
+        }
+
         /* force MSI on */
         if (radeon_msi == 1)
                 return true;
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 8309b11e674d..03586763ee86 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -795,6 +795,8 @@ int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
 
         /* Get associated drm_crtc: */
         drmcrtc = &rdev->mode_info.crtcs[crtc]->base;
+        if (!drmcrtc)
+                return -EINVAL;
 
         /* Helper routine in DRM core does all the work: */
         return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error,
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 99a960a4f302..4c0d786d5c7a 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -213,6 +213,13 @@ int radeon_bo_create(struct radeon_device *rdev,
         if (!(rdev->flags & RADEON_IS_PCIE))
                 bo->flags &= ~(RADEON_GEM_GTT_WC | RADEON_GEM_GTT_UC);
 
+#ifdef CONFIG_X86_32
+        /* XXX: Write-combined CPU mappings of GTT seem broken on 32-bit
+         * See https://bugs.freedesktop.org/show_bug.cgi?id=84627
+         */
+        bo->flags &= ~RADEON_GEM_GTT_WC;
+#endif
+
         radeon_ttm_placement_from_domain(bo, domain);
         /* Kernel allocation are uninterruptible */
         down_read(&rdev->pm.mclk_lock);
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 3d17af34afa7..2456f69efd23 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -314,7 +314,7 @@ unsigned radeon_ring_backup(struct radeon_device *rdev, struct radeon_ring *ring
         }
 
         /* and then save the content of the ring */
-        *data = kmalloc_array(size, sizeof(uint32_t), GFP_KERNEL);
+        *data = drm_malloc_ab(size, sizeof(uint32_t));
         if (!*data) {
                 mutex_unlock(&rdev->ring_lock);
                 return 0;
@@ -356,7 +356,7 @@ int radeon_ring_restore(struct radeon_device *rdev, struct radeon_ring *ring,
         }
 
         radeon_ring_unlock_commit(rdev, ring, false);
-        kfree(data);
+        drm_free_large(data);
         return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index 4532cc76a0a6..dfde266529e2 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -132,8 +132,8 @@ struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev,
         struct radeon_cs_reloc *list;
         unsigned i, idx;
 
-        list = kmalloc_array(vm->max_pde_used + 2,
-                             sizeof(struct radeon_cs_reloc), GFP_KERNEL);
+        list = drm_malloc_ab(vm->max_pde_used + 2,
+                             sizeof(struct radeon_cs_reloc));
         if (!list)
                 return NULL;
 
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 5f6db4629aaa..9acb1c3c005b 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -879,6 +879,9 @@ void rs600_bandwidth_update(struct radeon_device *rdev)
         u32 d1mode_priority_a_cnt, d2mode_priority_a_cnt;
         /* FIXME: implement full support */
 
+        if (!rdev->mode_info.mode_config_initialized)
+                return;
+
         radeon_update_display_priority(rdev);
 
         if (rdev->mode_info.crtcs[0]->base.enabled)
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 3462b64369bf..0a2d36e81108 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -579,6 +579,9 @@ void rs690_bandwidth_update(struct radeon_device *rdev)
         u32 d1mode_priority_a_cnt, d1mode_priority_b_cnt;
         u32 d2mode_priority_a_cnt, d2mode_priority_b_cnt;
 
+        if (!rdev->mode_info.mode_config_initialized)
+                return;
+
         radeon_update_display_priority(rdev);
 
         if (rdev->mode_info.crtcs[0]->base.enabled)
diff --git a/drivers/gpu/drm/radeon/rs780_dpm.c b/drivers/gpu/drm/radeon/rs780_dpm.c
index 02f7710de470..9031f4b69824 100644
--- a/drivers/gpu/drm/radeon/rs780_dpm.c
+++ b/drivers/gpu/drm/radeon/rs780_dpm.c
@@ -24,6 +24,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "rs780d.h"
 #include "r600_dpm.h"
 #include "rs780_dpm.h"
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 8a477bf1fdb3..c55d653aaf5f 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -1277,6 +1277,9 @@ void rv515_bandwidth_update(struct radeon_device *rdev)
         struct drm_display_mode *mode0 = NULL;
         struct drm_display_mode *mode1 = NULL;
 
+        if (!rdev->mode_info.mode_config_initialized)
+                return;
+
         radeon_update_display_priority(rdev);
 
         if (rdev->mode_info.crtcs[0]->base.enabled)
diff --git a/drivers/gpu/drm/radeon/rv6xx_dpm.c b/drivers/gpu/drm/radeon/rv6xx_dpm.c
index e7045b085715..6a5c233361e9 100644
--- a/drivers/gpu/drm/radeon/rv6xx_dpm.c
+++ b/drivers/gpu/drm/radeon/rv6xx_dpm.c
@@ -24,6 +24,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "rv6xxd.h"
 #include "r600_dpm.h"
 #include "rv6xx_dpm.h"
diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c
index 3c76e1dcdf04..755a8f96fe46 100644
--- a/drivers/gpu/drm/radeon/rv770_dpm.c
+++ b/drivers/gpu/drm/radeon/rv770_dpm.c
@@ -24,6 +24,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "rv770d.h"
 #include "r600_dpm.h"
 #include "rv770_dpm.h"
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index eeea5b6a1775..7d5083dc4acb 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -2384,6 +2384,9 @@ void dce6_bandwidth_update(struct radeon_device *rdev)
         u32 num_heads = 0, lb_size;
         int i;
 
+        if (!rdev->mode_info.mode_config_initialized)
+                return;
+
         radeon_update_display_priority(rdev);
 
         for (i = 0; i < rdev->num_crtc; i++) {
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 9e4d5d7d348f..676e6c2ba90a 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -23,6 +23,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "sid.h"
 #include "r600_dpm.h"
 #include "si_dpm.h"
@@ -2916,6 +2917,7 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
         bool disable_sclk_switching = false;
         u32 mclk, sclk;
         u16 vddc, vddci;
+        u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
         int i;
 
         if ((rdev->pm.dpm.new_active_crtc_count > 1) ||
@@ -2949,6 +2951,29 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
                 }
         }
 
+        /* limit clocks to max supported clocks based on voltage dependency tables */
+        btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
+                                                        &max_sclk_vddc);
+        btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
+                                                        &max_mclk_vddci);
+        btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
+                                                        &max_mclk_vddc);
+
+        for (i = 0; i < ps->performance_level_count; i++) {
+                if (max_sclk_vddc) {
+                        if (ps->performance_levels[i].sclk > max_sclk_vddc)
+                                ps->performance_levels[i].sclk = max_sclk_vddc;
+                }
+                if (max_mclk_vddci) {
+                        if (ps->performance_levels[i].mclk > max_mclk_vddci)
+                                ps->performance_levels[i].mclk = max_mclk_vddci;
+                }
+                if (max_mclk_vddc) {
+                        if (ps->performance_levels[i].mclk > max_mclk_vddc)
+                                ps->performance_levels[i].mclk = max_mclk_vddc;
+                }
+        }
+
         /* XXX validate the min clocks required for display */
 
         if (disable_mclk_switching) {
@@ -6231,7 +6256,7 @@ static void si_parse_pplib_clock_info(struct radeon_device *rdev,
         if ((rps->class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) &&
             index == 0) {
                 /* XXX disable for A0 tahiti */
-                si_pi->ulv.supported = true;
+                si_pi->ulv.supported = false;
                 si_pi->ulv.pl = *pl;
                 si_pi->ulv.one_pcie_lane_in_ulv = false;
                 si_pi->ulv.volt_change_delay = SISLANDS_ULVVOLTAGECHANGEDELAY_DFLT;
diff --git a/drivers/gpu/drm/radeon/sumo_dpm.c b/drivers/gpu/drm/radeon/sumo_dpm.c
index 3f0e8d7b8dbe..1f8a8833e1be 100644
--- a/drivers/gpu/drm/radeon/sumo_dpm.c
+++ b/drivers/gpu/drm/radeon/sumo_dpm.c
@@ -23,6 +23,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "sumod.h"
 #include "r600_dpm.h"
 #include "cypress_dpm.h"
diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c
index 57f780053b3e..b4ec5c4e7969 100644
--- a/drivers/gpu/drm/radeon/trinity_dpm.c
+++ b/drivers/gpu/drm/radeon/trinity_dpm.c
@@ -23,6 +23,7 @@
 
 #include "drmP.h"
 #include "radeon.h"
+#include "radeon_asic.h"
 #include "trinityd.h"
 #include "r600_dpm.h"
 #include "trinity_dpm.h"
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index 6553fd238685..054a79f143ae 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -736,7 +736,6 @@ static const struct drm_crtc_funcs tegra_crtc_funcs = {
 
 static void tegra_crtc_disable(struct drm_crtc *crtc)
 {
-        struct tegra_dc *dc = to_tegra_dc(crtc);
         struct drm_device *drm = crtc->dev;
         struct drm_plane *plane;
 
@@ -752,7 +751,7 @@ static void tegra_crtc_disable(struct drm_crtc *crtc)
                 }
         }
 
-        drm_vblank_off(drm, dc->pipe);
+        drm_crtc_vblank_off(crtc);
 }
 
 static bool tegra_crtc_mode_fixup(struct drm_crtc *crtc,
@@ -841,8 +840,6 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
         u32 value;
         int err;
 
-        drm_vblank_pre_modeset(crtc->dev, dc->pipe);
-
         err = tegra_crtc_setup_clk(crtc, mode);
         if (err) {
                 dev_err(dc->dev, "failed to setup clock for CRTC: %d\n", err);
@@ -896,6 +893,8 @@ static void tegra_crtc_prepare(struct drm_crtc *crtc)
         unsigned int syncpt;
         unsigned long value;
 
+        drm_crtc_vblank_off(crtc);
+
         /* hardware initialization */
         reset_control_deassert(dc->rst);
         usleep_range(10000, 20000);
@@ -943,7 +942,7 @@ static void tegra_crtc_commit(struct drm_crtc *crtc)
         value = GENERAL_ACT_REQ | WIN_A_ACT_REQ;
         tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL);
 
-        drm_vblank_post_modeset(crtc->dev, dc->pipe);
+        drm_crtc_vblank_on(crtc);
 }
 
 static void tegra_crtc_load_lut(struct drm_crtc *crtc)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 8f5cec67c47d..d395b0bef73b 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -709,6 +709,7 @@ out:
 
 static int ttm_mem_evict_first(struct ttm_bo_device *bdev,
                                 uint32_t mem_type,
+                                const struct ttm_place *place,
                                 bool interruptible,
                                 bool no_wait_gpu)
 {
@@ -720,8 +721,21 @@ static int ttm_mem_evict_first(struct ttm_bo_device *bdev,
         spin_lock(&glob->lru_lock);
         list_for_each_entry(bo, &man->lru, lru) {
                 ret = __ttm_bo_reserve(bo, false, true, false, NULL);
-                if (!ret)
+                if (!ret) {
+                        if (place && (place->fpfn || place->lpfn)) {
+                                /* Don't evict this BO if it's outside of the
+                                 * requested placement range
+                                 */
+                                if (place->fpfn >= (bo->mem.start + bo->mem.size) ||
+                                    (place->lpfn && place->lpfn <= bo->mem.start)) {
+                                        __ttm_bo_unreserve(bo);
+                                        ret = -EBUSY;
+                                        continue;
+                                }
+                        }
+
                         break;
+                }
         }
 
         if (ret) {
@@ -782,7 +796,7 @@ static int ttm_bo_mem_force_space(struct ttm_buffer_object *bo,
                         return ret;
                 if (mem->mm_node)
                         break;
-                ret = ttm_mem_evict_first(bdev, mem_type,
+                ret = ttm_mem_evict_first(bdev, mem_type, place,
                                           interruptible, no_wait_gpu);
                 if (unlikely(ret != 0))
                         return ret;
@@ -994,9 +1008,9 @@ static bool ttm_bo_mem_compat(struct ttm_placement *placement,
 
         for (i = 0; i < placement->num_placement; i++) {
                 const struct ttm_place *heap = &placement->placement[i];
-                if (mem->mm_node && heap->lpfn != 0 &&
+                if (mem->mm_node &&
                     (mem->start < heap->fpfn ||
-                     mem->start + mem->num_pages > heap->lpfn))
+                     (heap->lpfn != 0 && (mem->start + mem->num_pages) > heap->lpfn)))
                         continue;
 
                 *new_flags = heap->flags;
@@ -1007,9 +1021,9 @@ static bool ttm_bo_mem_compat(struct ttm_placement *placement,
 
         for (i = 0; i < placement->num_busy_placement; i++) {
                 const struct ttm_place *heap = &placement->busy_placement[i];
-                if (mem->mm_node && heap->lpfn != 0 &&
+                if (mem->mm_node &&
                     (mem->start < heap->fpfn ||
-                     mem->start + mem->num_pages > heap->lpfn))
+                     (heap->lpfn != 0 && (mem->start + mem->num_pages) > heap->lpfn)))
                         continue;
 
                 *new_flags = heap->flags;
@@ -1233,7 +1247,7 @@ static int ttm_bo_force_list_clean(struct ttm_bo_device *bdev,
         spin_lock(&glob->lru_lock);
         while (!list_empty(&man->lru)) {
                 spin_unlock(&glob->lru_lock);
-                ret = ttm_mem_evict_first(bdev, mem_type, false, false);
+                ret = ttm_mem_evict_first(bdev, mem_type, NULL, false, false);
                 if (ret) {
                         if (allow_errors) {
                                 return ret;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c
index bfeb4b1f2acf..21e9b7f8dad0 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c
@@ -246,7 +246,8 @@ int vmw_cmdbuf_res_remove(struct vmw_cmdbuf_res_manager *man,
         struct drm_hash_item *hash;
         int ret;
 
-        ret = drm_ht_find_item(&man->resources, user_key, &hash);
+        ret = drm_ht_find_item(&man->resources, user_key | (res_type << 24),
+                               &hash);
         if (likely(ret != 0))
                 return -EINVAL;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 7197af157313..25f3c250fd98 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -688,7 +688,11 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
                 goto out_err0;
         }
 
-        if (unlikely(dev_priv->prim_bb_mem < dev_priv->vram_size))
+        /*
+         * Limit back buffer size to VRAM size.  Remove this once
+         * screen targets are implemented.
+         */
+        if (dev_priv->prim_bb_mem > dev_priv->vram_size)
                 dev_priv->prim_bb_mem = dev_priv->vram_size;
 
         mutex_unlock(&dev_priv->hw_mutex);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index d2bc2b03d4c6..941a7bc0b791 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -187,7 +187,7 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
          * can do this since the caller in the drm core doesn't check anything
          * which is protected by any looks.
          */
-        drm_modeset_unlock(&crtc->mutex);
+        drm_modeset_unlock_crtc(crtc);
         drm_modeset_lock_all(dev_priv->dev);
 
         /* A lot of the code assumes this */
@@ -252,7 +252,7 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
         ret = 0;
 out:
         drm_modeset_unlock_all(dev_priv->dev);
-        drm_modeset_lock(&crtc->mutex, NULL);
+        drm_modeset_lock_crtc(crtc);
 
         return ret;
 }
@@ -273,7 +273,7 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
          * can do this since the caller in the drm core doesn't check anything
          * which is protected by any looks.
          */
-        drm_modeset_unlock(&crtc->mutex);
+        drm_modeset_unlock_crtc(crtc);
         drm_modeset_lock_all(dev_priv->dev);
 
         vmw_cursor_update_position(dev_priv, shown,
@@ -281,7 +281,7 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
                                    du->cursor_y + du->hotspot_y);
 
         drm_modeset_unlock_all(dev_priv->dev);
-        drm_modeset_lock(&crtc->mutex, NULL);
+        drm_modeset_lock_crtc(crtc);
 
         return 0;
 }
@@ -1950,6 +1950,14 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector,
                 DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC)
         };
         int i;
+        u32 assumed_bpp = 2;
+
+        /*
+         * If using screen objects, then assume 32-bpp because that's what the
+         * SVGA device is assuming
+         */
+        if (dev_priv->sou_priv)
+                assumed_bpp = 4;
 
         /* Add preferred mode */
         {
@@ -1960,8 +1968,9 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector,
                 mode->vdisplay = du->pref_height;
                 vmw_guess_mode_timing(mode);
 
-                if (vmw_kms_validate_mode_vram(dev_priv, mode->hdisplay * 2,
-                                               mode->vdisplay)) {
+                if (vmw_kms_validate_mode_vram(dev_priv,
+                                                mode->hdisplay * assumed_bpp,
+                                                mode->vdisplay)) {
                         drm_mode_probed_add(connector, mode);
                 } else {
                         drm_mode_destroy(dev, mode);
@@ -1983,7 +1992,8 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector,
                     bmode->vdisplay > max_height)
                         continue;
 
-                if (!vmw_kms_validate_mode_vram(dev_priv, bmode->hdisplay * 2,
+                if (!vmw_kms_validate_mode_vram(dev_priv,
+                                                bmode->hdisplay * assumed_bpp,
                                                 bmode->vdisplay))
                         continue;