78 files changed, 1224 insertions, 502 deletions

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 2baa6708e44c..654faa803dcb 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -2674,3 +2674,23 @@ out:
         mutex_unlock(&dev->mode_config.mutex);
         return ret;
 }
+
+void drm_mode_config_reset(struct drm_device *dev)
+{
+        struct drm_crtc *crtc;
+        struct drm_encoder *encoder;
+        struct drm_connector *connector;
+
+        list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+                if (crtc->funcs->reset)
+                        crtc->funcs->reset(crtc);
+
+        list_for_each_entry(encoder, &dev->mode_config.encoder_list, head)
+                if (encoder->funcs->reset)
+                        encoder->funcs->reset(encoder);
+
+        list_for_each_entry(connector, &dev->mode_config.connector_list, head)
+                if (connector->funcs->reset)
+                        connector->funcs->reset(connector);
+}
+EXPORT_SYMBOL(drm_mode_config_reset);
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 952b3d4fb2a6..92369655dca3 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -343,13 +343,12 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
         struct drm_encoder *encoder;
         bool ret = true;
 
-        adjusted_mode = drm_mode_duplicate(dev, mode);
-
         crtc->enabled = drm_helper_crtc_in_use(crtc);
-
         if (!crtc->enabled)
                 return true;
 
+        adjusted_mode = drm_mode_duplicate(dev, mode);
+
         saved_hwmode = crtc->hwmode;
         saved_mode = crtc->mode;
         saved_x = crtc->x;
@@ -437,10 +436,9 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
          */
         drm_calc_timestamping_constants(crtc);
 
-        /* XXX free adjustedmode */
-        drm_mode_destroy(dev, adjusted_mode);
         /* FIXME: add subpixel order */
 done:
+        drm_mode_destroy(dev, adjusted_mode);
         if (!ret) {
                 crtc->hwmode = saved_hwmode;
                 crtc->mode = saved_mode;
@@ -497,14 +495,17 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 
         crtc_funcs = set->crtc->helper_private;
 
+        if (!set->mode)
+                set->fb = NULL;
+
         if (set->fb) {
                 DRM_DEBUG_KMS("[CRTC:%d] [FB:%d] #connectors=%d (x y) (%i %i)\n",
                                 set->crtc->base.id, set->fb->base.id,
                                 (int)set->num_connectors, set->x, set->y);
         } else {
-                DRM_DEBUG_KMS("[CRTC:%d] [NOFB] #connectors=%d (x y) (%i %i)\n",
-                                set->crtc->base.id, (int)set->num_connectors,
-                                set->x, set->y);
+                DRM_DEBUG_KMS("[CRTC:%d] [NOFB]\n", set->crtc->base.id);
+                set->mode = NULL;
+                set->num_connectors = 0;
         }
 
         dev = set->crtc->dev;
@@ -649,8 +650,8 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
                 mode_changed = true;
 
         if (mode_changed) {
-                set->crtc->enabled = (set->mode != NULL);
-                if (set->mode != NULL) {
+                set->crtc->enabled = drm_helper_crtc_in_use(set->crtc);
+                if (set->crtc->enabled) {
                         DRM_DEBUG_KMS("attempting to set mode from"
                                         " userspace\n");
                         drm_mode_debug_printmodeline(set->mode);
@@ -665,6 +666,12 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
                                 ret = -EINVAL;
                                 goto fail;
                         }
+                        DRM_DEBUG_KMS("Setting connector DPMS state to on\n");
+                        for (i = 0; i < set->num_connectors; i++) {
+                                DRM_DEBUG_KMS("\t[CONNECTOR:%d:%s] set DPMS on\n", set->connectors[i]->base.id,
+                                              drm_get_connector_name(set->connectors[i]));
+                                set->connectors[i]->dpms = DRM_MODE_DPMS_ON;
+                        }
                 }
                 drm_helper_disable_unused_functions(dev);
         } else if (fb_changed) {
@@ -681,12 +688,6 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
                         goto fail;
                 }
         }
-        DRM_DEBUG_KMS("Setting connector DPMS state to on\n");
-        for (i = 0; i < set->num_connectors; i++) {
-                DRM_DEBUG_KMS("\t[CONNECTOR:%d:%s] set DPMS on\n", set->connectors[i]->base.id,
-                              drm_get_connector_name(set->connectors[i]));
-                set->connectors[i]->dpms = DRM_MODE_DPMS_ON;
-        }
 
         kfree(save_connectors);
         kfree(save_encoders);
diff --git a/drivers/gpu/drm/drm_info.c b/drivers/gpu/drm/drm_info.c
index 3cdbaf379bb5..be9a9c07d152 100644
--- a/drivers/gpu/drm/drm_info.c
+++ b/drivers/gpu/drm/drm_info.c
@@ -283,17 +283,18 @@ int drm_vma_info(struct seq_file *m, void *data)
 #endif
 
         mutex_lock(&dev->struct_mutex);
-        seq_printf(m, "vma use count: %d, high_memory = %p, 0x%08llx\n",
+        seq_printf(m, "vma use count: %d, high_memory = %pK, 0x%pK\n",
                    atomic_read(&dev->vma_count),
-                   high_memory, (u64)virt_to_phys(high_memory));
+                   high_memory, (void *)virt_to_phys(high_memory));
 
         list_for_each_entry(pt, &dev->vmalist, head) {
                 vma = pt->vma;
                 if (!vma)
                         continue;
                 seq_printf(m,
-                           "\n%5d 0x%08lx-0x%08lx %c%c%c%c%c%c 0x%08lx000",
-                           pt->pid, vma->vm_start, vma->vm_end,
+                           "\n%5d 0x%pK-0x%pK %c%c%c%c%c%c 0x%08lx000",
+                           pt->pid,
+                           (void *)vma->vm_start, (void *)vma->vm_end,
                            vma->vm_flags & VM_READ ? 'r' : '-',
                            vma->vm_flags & VM_WRITE ? 'w' : '-',
                            vma->vm_flags & VM_EXEC ? 'x' : '-',
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 0054e957203f..3dadfa2a8528 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -1250,7 +1250,7 @@ void drm_handle_vblank_events(struct drm_device *dev, int crtc)
  * Drivers should call this routine in their vblank interrupt handlers to
  * update the vblank counter and send any signals that may be pending.
  */
-void drm_handle_vblank(struct drm_device *dev, int crtc)
+bool drm_handle_vblank(struct drm_device *dev, int crtc)
 {
         u32 vblcount;
         s64 diff_ns;
@@ -1258,7 +1258,7 @@ void drm_handle_vblank(struct drm_device *dev, int crtc)
         unsigned long irqflags;
 
         if (!dev->num_crtcs)
-                return;
+                return false;
 
         /* Need timestamp lock to prevent concurrent execution with
          * vblank enable/disable, as this would cause inconsistent
@@ -1269,7 +1269,7 @@ void drm_handle_vblank(struct drm_device *dev, int crtc)
         /* Vblank irq handling disabled. Nothing to do. */
         if (!dev->vblank_enabled[crtc]) {
                 spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
-                return;
+                return false;
         }
 
         /* Fetch corresponding timestamp for this vblank interval from
@@ -1311,5 +1311,6 @@ void drm_handle_vblank(struct drm_device *dev, int crtc)
         drm_handle_vblank_events(dev, crtc);
 
         spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
+        return true;
 }
 EXPORT_SYMBOL(drm_handle_vblank);
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 66796bb82d3e..0ad533f06af9 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -46,6 +46,9 @@ module_param_named(fbpercrtc, i915_fbpercrtc, int, 0400);
 unsigned int i915_powersave = 1;
 module_param_named(powersave, i915_powersave, int, 0600);
 
+unsigned int i915_enable_rc6 = 0;
+module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600);
+
 unsigned int i915_lvds_downclock = 0;
 module_param_named(lvds_downclock, i915_lvds_downclock, int, 0400);
 
@@ -354,12 +357,13 @@ static int i915_drm_thaw(struct drm_device *dev)
                 error = i915_gem_init_ringbuffer(dev);
                 mutex_unlock(&dev->struct_mutex);
 
+                drm_mode_config_reset(dev);
                 drm_irq_install(dev);
 
                 /* Resume the modeset for every activated CRTC */
                 drm_helper_resume_force_mode(dev);
 
-                if (dev_priv->renderctx && dev_priv->pwrctx)
+                if (IS_IRONLAKE_M(dev))
                         ironlake_enable_rc6(dev);
         }
 
@@ -542,6 +546,7 @@ int i915_reset(struct drm_device *dev, u8 flags)
 
                 mutex_unlock(&dev->struct_mutex);
                 drm_irq_uninstall(dev);
+                drm_mode_config_reset(dev);
                 drm_irq_install(dev);
                 mutex_lock(&dev->struct_mutex);
         }
@@ -566,6 +571,14 @@ int i915_reset(struct drm_device *dev, u8 flags)
 static int __devinit
 i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
+        /* Only bind to function 0 of the device. Early generations
+         * used function 1 as a placeholder for multi-head. This causes
+         * us confusion instead, especially on the systems where both
+         * functions have the same PCI-ID!
+         */
+        if (PCI_FUNC(pdev->devfn))
+                return -ENODEV;
+
         return drm_get_pci_dev(pdev, ent, &driver);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index a0149c619cdd..65dfe81d0035 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -958,6 +958,7 @@ extern unsigned int i915_fbpercrtc;
 extern unsigned int i915_powersave;
 extern unsigned int i915_lvds_downclock;
 extern unsigned int i915_panel_use_ssc;
+extern unsigned int i915_enable_rc6;
 
 extern int i915_suspend(struct drm_device *dev, pm_message_t state);
 extern int i915_resume(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 062f353497e6..97f946dcc1aa 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1196,18 +1196,18 @@ irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS)
                                 intel_finish_page_flip_plane(dev, 1);
                 }
 
-                if (pipea_stats & vblank_status) {
+                if (pipea_stats & vblank_status &&
+                    drm_handle_vblank(dev, 0)) {
                         vblank++;
-                        drm_handle_vblank(dev, 0);
                         if (!dev_priv->flip_pending_is_done) {
                                 i915_pageflip_stall_check(dev, 0);
                                 intel_finish_page_flip(dev, 0);
                         }
                 }
 
-                if (pipeb_stats & vblank_status) {
+                if (pipeb_stats & vblank_status &&
+                    drm_handle_vblank(dev, 1)) {
                         vblank++;
-                        drm_handle_vblank(dev, 1);
                         if (!dev_priv->flip_pending_is_done) {
                                 i915_pageflip_stall_check(dev, 1);
                                 intel_finish_page_flip(dev, 1);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5cfc68940f17..15d94c63918c 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -174,7 +174,9 @@
  *   address/value pairs. Don't overdue it, though, x <= 2^4 must hold!
  */
 #define MI_LOAD_REGISTER_IMM(x) MI_INSTR(0x22, 2*x-1)
-#define MI_FLUSH_DW             MI_INSTR(0x26, 2) /* for GEN6 */
+#define MI_FLUSH_DW             MI_INSTR(0x26, 1) /* for GEN6 */
+#define   MI_INVALIDATE_TLB     (1<<18)
+#define   MI_INVALIDATE_BSD     (1<<7)
 #define MI_BATCH_BUFFER         MI_INSTR(0x30, 1)
 #define   MI_BATCH_NON_SECURE   (1)
 #define   MI_BATCH_NON_SECURE_I965 (1<<8)
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 17035b87ee46..8a77ff4a7237 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -535,6 +535,15 @@ static int intel_crt_set_property(struct drm_connector *connector,
         return 0;
 }
 
+static void intel_crt_reset(struct drm_connector *connector)
+{
+        struct drm_device *dev = connector->dev;
+        struct intel_crt *crt = intel_attached_crt(connector);
+
+        if (HAS_PCH_SPLIT(dev))
+                crt->force_hotplug_required = 1;
+}
+
 /*
  * Routines for controlling stuff on the analog port
  */
@@ -548,6 +557,7 @@ static const struct drm_encoder_helper_funcs intel_crt_helper_funcs = {
 };
 
 static const struct drm_connector_funcs intel_crt_connector_funcs = {
+        .reset = intel_crt_reset,
         .dpms = drm_helper_connector_dpms,
         .detect = intel_crt_detect,
         .fill_modes = drm_helper_probe_single_connector_modes,
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index d7f237deaaf0..3b006536b3d2 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5551,6 +5551,16 @@ cleanup_work:
         return ret;
 }
 
+static void intel_crtc_reset(struct drm_crtc *crtc)
+{
+        struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+
+        /* Reset flags back to the 'unknown' status so that they
+         * will be correctly set on the initial modeset.
+         */
+        intel_crtc->dpms_mode = -1;
+}
+
 static struct drm_crtc_helper_funcs intel_helper_funcs = {
         .dpms = intel_crtc_dpms,
         .mode_fixup = intel_crtc_mode_fixup,
@@ -5562,6 +5572,7 @@ static struct drm_crtc_helper_funcs intel_helper_funcs = {
 };
 
 static const struct drm_crtc_funcs intel_crtc_funcs = {
+        .reset = intel_crtc_reset,
         .cursor_set = intel_crtc_cursor_set,
         .cursor_move = intel_crtc_cursor_move,
         .gamma_set = intel_crtc_gamma_set,
@@ -5652,8 +5663,7 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
         dev_priv->plane_to_crtc_mapping[intel_crtc->plane] = &intel_crtc->base;
         dev_priv->pipe_to_crtc_mapping[intel_crtc->pipe] = &intel_crtc->base;
 
-        intel_crtc->cursor_addr = 0;
-        intel_crtc->dpms_mode = -1;
+        intel_crtc_reset(&intel_crtc->base);
         intel_crtc->active = true; /* force the pipe off on setup_init_config */
 
         if (HAS_PCH_SPLIT(dev)) {
@@ -6452,52 +6462,60 @@ void intel_enable_clock_gating(struct drm_device *dev)
         }
 }
 
-void intel_disable_clock_gating(struct drm_device *dev)
+static void ironlake_teardown_rc6(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
 
         if (dev_priv->renderctx) {
-                struct drm_i915_gem_object *obj = dev_priv->renderctx;
-
-                I915_WRITE(CCID, 0);
-                POSTING_READ(CCID);
-
-                i915_gem_object_unpin(obj);
-                drm_gem_object_unreference(&obj->base);
+                i915_gem_object_unpin(dev_priv->renderctx);
+                drm_gem_object_unreference(&dev_priv->renderctx->base);
                 dev_priv->renderctx = NULL;
         }
 
         if (dev_priv->pwrctx) {
-                struct drm_i915_gem_object *obj = dev_priv->pwrctx;
+                i915_gem_object_unpin(dev_priv->pwrctx);
+                drm_gem_object_unreference(&dev_priv->pwrctx->base);
+                dev_priv->pwrctx = NULL;
+        }
+}
+
+static void ironlake_disable_rc6(struct drm_device *dev)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+
+        if (I915_READ(PWRCTXA)) {
+                /* Wake the GPU, prevent RC6, then restore RSTDBYCTL */
+                I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) | RCX_SW_EXIT);
+                wait_for(((I915_READ(RSTDBYCTL) & RSX_STATUS_MASK) == RSX_STATUS_ON),
+                         50);
 
                 I915_WRITE(PWRCTXA, 0);
                 POSTING_READ(PWRCTXA);
 
-                i915_gem_object_unpin(obj);
-                drm_gem_object_unreference(&obj->base);
-                dev_priv->pwrctx = NULL;
+                I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
+                POSTING_READ(RSTDBYCTL);
         }
+
+        ironlake_disable_rc6(dev);
 }
 
-static void ironlake_disable_rc6(struct drm_device *dev)
+static int ironlake_setup_rc6(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
 
-        /* Wake the GPU, prevent RC6, then restore RSTDBYCTL */
-        I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) | RCX_SW_EXIT);
-        wait_for(((I915_READ(RSTDBYCTL) & RSX_STATUS_MASK) == RSX_STATUS_ON),
-                 10);
-        POSTING_READ(CCID);
-        I915_WRITE(PWRCTXA, 0);
-        POSTING_READ(PWRCTXA);
-        I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
-        POSTING_READ(RSTDBYCTL);
-        i915_gem_object_unpin(dev_priv->renderctx);
-        drm_gem_object_unreference(&dev_priv->renderctx->base);
-        dev_priv->renderctx = NULL;
-        i915_gem_object_unpin(dev_priv->pwrctx);
-        drm_gem_object_unreference(&dev_priv->pwrctx->base);
-        dev_priv->pwrctx = NULL;
+        if (dev_priv->renderctx == NULL)
+                dev_priv->renderctx = intel_alloc_context_page(dev);
+        if (!dev_priv->renderctx)
+                return -ENOMEM;
+
+        if (dev_priv->pwrctx == NULL)
+                dev_priv->pwrctx = intel_alloc_context_page(dev);
+        if (!dev_priv->pwrctx) {
+                ironlake_teardown_rc6(dev);
+                return -ENOMEM;
+        }
+
+        return 0;
 }
 
 void ironlake_enable_rc6(struct drm_device *dev)
@@ -6505,15 +6523,26 @@ void ironlake_enable_rc6(struct drm_device *dev)
         struct drm_i915_private *dev_priv = dev->dev_private;
         int ret;
 
+        /* rc6 disabled by default due to repeated reports of hanging during
+         * boot and resume.
+         */
+        if (!i915_enable_rc6)
+                return;
+
+        ret = ironlake_setup_rc6(dev);
+        if (ret)
+                return;
+
         /*
          * GPU can automatically power down the render unit if given a page
          * to save state.
          */
         ret = BEGIN_LP_RING(6);
         if (ret) {
-                ironlake_disable_rc6(dev);
+                ironlake_teardown_rc6(dev);
                 return;
         }
+
         OUT_RING(MI_SUSPEND_FLUSH | MI_SUSPEND_FLUSH_EN);
         OUT_RING(MI_SET_CONTEXT);
         OUT_RING(dev_priv->renderctx->gtt_offset |
@@ -6530,6 +6559,7 @@ void ironlake_enable_rc6(struct drm_device *dev)
         I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
 }
 
+
 /* Set up chip specific display functions */
 static void intel_init_display(struct drm_device *dev)
 {
@@ -6772,21 +6802,9 @@ void intel_modeset_init(struct drm_device *dev)
         if (IS_GEN6(dev))
                 gen6_enable_rps(dev_priv);
 
-        if (IS_IRONLAKE_M(dev)) {
-                dev_priv->renderctx = intel_alloc_context_page(dev);
-                if (!dev_priv->renderctx)
-                        goto skip_rc6;
-                dev_priv->pwrctx = intel_alloc_context_page(dev);
-                if (!dev_priv->pwrctx) {
-                        i915_gem_object_unpin(dev_priv->renderctx);
-                        drm_gem_object_unreference(&dev_priv->renderctx->base);
-                        dev_priv->renderctx = NULL;
-                        goto skip_rc6;
-                }
+        if (IS_IRONLAKE_M(dev))
                 ironlake_enable_rc6(dev);
-        }
 
-skip_rc6:
         INIT_WORK(&dev_priv->idle_work, intel_idle_update);
         setup_timer(&dev_priv->idle_timer, intel_gpu_idle_timer,
                     (unsigned long)dev);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 1f4242b682c8..51cb4e36997f 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1639,6 +1639,24 @@ static int intel_dp_get_modes(struct drm_connector *connector)
         return 0;
 }
 
+static bool
+intel_dp_detect_audio(struct drm_connector *connector)
+{
+        struct intel_dp *intel_dp = intel_attached_dp(connector);
+        struct edid *edid;
+        bool has_audio = false;
+
+        edid = drm_get_edid(connector, &intel_dp->adapter);
+        if (edid) {
+                has_audio = drm_detect_monitor_audio(edid);
+
+                connector->display_info.raw_edid = NULL;
+                kfree(edid);
+        }
+
+        return has_audio;
+}
+
 static int
 intel_dp_set_property(struct drm_connector *connector,
                       struct drm_property *property,
@@ -1652,17 +1670,23 @@ intel_dp_set_property(struct drm_connector *connector,
                 return ret;
 
         if (property == intel_dp->force_audio_property) {
-                if (val == intel_dp->force_audio)
+                int i = val;
+                bool has_audio;
+
+                if (i == intel_dp->force_audio)
                         return 0;
 
-                intel_dp->force_audio = val;
+                intel_dp->force_audio = i;
 
-                if (val > 0 && intel_dp->has_audio)
-                        return 0;
-                if (val < 0 && !intel_dp->has_audio)
+                if (i == 0)
+                        has_audio = intel_dp_detect_audio(connector);
+                else
+                        has_audio = i > 0;
+
+                if (has_audio == intel_dp->has_audio)
                         return 0;
 
-                intel_dp->has_audio = val > 0;
+                intel_dp->has_audio = has_audio;
                 goto done;
         }
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 74db2557d644..2c431049963c 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -298,7 +298,6 @@ extern void intel_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green,
 extern void intel_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
                                     u16 *blue, int regno);
 extern void intel_enable_clock_gating(struct drm_device *dev);
-extern void intel_disable_clock_gating(struct drm_device *dev);
 extern void ironlake_enable_drps(struct drm_device *dev);
 extern void ironlake_disable_drps(struct drm_device *dev);
 extern void gen6_enable_rps(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 0d0273e7b029..c635c9e357b9 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -251,6 +251,27 @@ static int intel_hdmi_get_modes(struct drm_connector *connector)
                                    &dev_priv->gmbus[intel_hdmi->ddc_bus].adapter);
 }
 
+static bool
+intel_hdmi_detect_audio(struct drm_connector *connector)
+{
+        struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
+        struct drm_i915_private *dev_priv = connector->dev->dev_private;
+        struct edid *edid;
+        bool has_audio = false;
+
+        edid = drm_get_edid(connector,
+                            &dev_priv->gmbus[intel_hdmi->ddc_bus].adapter);
+        if (edid) {
+                if (edid->input & DRM_EDID_INPUT_DIGITAL)
+                        has_audio = drm_detect_monitor_audio(edid);
+
+                connector->display_info.raw_edid = NULL;
+                kfree(edid);
+        }
+
+        return has_audio;
+}
+
 static int
 intel_hdmi_set_property(struct drm_connector *connector,
                       struct drm_property *property,
@@ -264,17 +285,23 @@ intel_hdmi_set_property(struct drm_connector *connector,
                 return ret;
 
         if (property == intel_hdmi->force_audio_property) {
-                if (val == intel_hdmi->force_audio)
+                int i = val;
+                bool has_audio;
+
+                if (i == intel_hdmi->force_audio)
                         return 0;
 
-                intel_hdmi->force_audio = val;
+                intel_hdmi->force_audio = i;
 
-                if (val > 0 && intel_hdmi->has_audio)
-                        return 0;
-                if (val < 0 && !intel_hdmi->has_audio)
+                if (i == 0)
+                        has_audio = intel_hdmi_detect_audio(connector);
+                else
+                        has_audio = i > 0;
+
+                if (has_audio == intel_hdmi->has_audio)
                         return 0;
 
-                intel_hdmi->has_audio = val > 0;
+                intel_hdmi->has_audio = has_audio;
                 goto done;
         }
 
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index ace8d5d30dd2..bcdba7bd5cfa 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -261,12 +261,6 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
                 return true;
         }
 
-        /* Make sure pre-965s set dither correctly */
-        if (INTEL_INFO(dev)->gen < 4) {
-                if (dev_priv->lvds_dither)
-                        pfit_control |= PANEL_8TO6_DITHER_ENABLE;
-        }
-
         /* Native modes don't need fitting */
         if (adjusted_mode->hdisplay == mode->hdisplay &&
             adjusted_mode->vdisplay == mode->vdisplay)
@@ -374,10 +368,16 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
         }
 
 out:
+        /* If not enabling scaling, be consistent and always use 0. */
         if ((pfit_control & PFIT_ENABLE) == 0) {
                 pfit_control = 0;
                 pfit_pgm_ratios = 0;
         }
+
+        /* Make sure pre-965 set dither correctly */
+        if (INTEL_INFO(dev)->gen < 4 && dev_priv->lvds_dither)
+                pfit_control |= PANEL_8TO6_DITHER_ENABLE;
+
         if (pfit_control != intel_lvds->pfit_control ||
             pfit_pgm_ratios != intel_lvds->pfit_pgm_ratios) {
                 intel_lvds->pfit_control = pfit_control;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 6218fa97aa1e..445f27efe677 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -1059,22 +1059,25 @@ static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring,
 }
 
 static int gen6_ring_flush(struct intel_ring_buffer *ring,
-                           u32 invalidate_domains,
-                           u32 flush_domains)
+                           u32 invalidate, u32 flush)
 {
+        uint32_t cmd;
         int ret;
 
-        if ((flush_domains & I915_GEM_DOMAIN_RENDER) == 0)
+        if (((invalidate | flush) & I915_GEM_GPU_DOMAINS) == 0)
                 return 0;
 
         ret = intel_ring_begin(ring, 4);
         if (ret)
                 return ret;
 
-        intel_ring_emit(ring, MI_FLUSH_DW);
-        intel_ring_emit(ring, 0);
+        cmd = MI_FLUSH_DW;
+        if (invalidate & I915_GEM_GPU_DOMAINS)
+                cmd |= MI_INVALIDATE_TLB | MI_INVALIDATE_BSD;
+        intel_ring_emit(ring, cmd);
         intel_ring_emit(ring, 0);
         intel_ring_emit(ring, 0);
+        intel_ring_emit(ring, MI_NOOP);
         intel_ring_advance(ring);
         return 0;
 }
@@ -1230,22 +1233,25 @@ static int blt_ring_begin(struct intel_ring_buffer *ring,
 }
 
 static int blt_ring_flush(struct intel_ring_buffer *ring,
-                           u32 invalidate_domains,
-                           u32 flush_domains)
+                          u32 invalidate, u32 flush)
 {
+        uint32_t cmd;
         int ret;
 
-        if ((flush_domains & I915_GEM_DOMAIN_RENDER) == 0)
+        if (((invalidate | flush) & I915_GEM_DOMAIN_RENDER) == 0)
                 return 0;
 
         ret = blt_ring_begin(ring, 4);
         if (ret)
                 return ret;
 
-        intel_ring_emit(ring, MI_FLUSH_DW);
-        intel_ring_emit(ring, 0);
+        cmd = MI_FLUSH_DW;
+        if (invalidate & I915_GEM_DOMAIN_RENDER)
+                cmd |= MI_INVALIDATE_TLB;
+        intel_ring_emit(ring, cmd);
         intel_ring_emit(ring, 0);
         intel_ring_emit(ring, 0);
+        intel_ring_emit(ring, MI_NOOP);
         intel_ring_advance(ring);
         return 0;
 }
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 45cd37652a37..7c50cdce84f0 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -46,6 +46,7 @@
                          SDVO_TV_MASK)
 
 #define IS_TV(c)        (c->output_flag & SDVO_TV_MASK)
+#define IS_TMDS(c)      (c->output_flag & SDVO_TMDS_MASK)
 #define IS_LVDS(c)      (c->output_flag & SDVO_LVDS_MASK)
 #define IS_TV_OR_LVDS(c) (c->output_flag & (SDVO_TV_MASK | SDVO_LVDS_MASK))
 
@@ -473,20 +474,6 @@ static bool intel_sdvo_write_cmd(struct intel_sdvo *intel_sdvo, u8 cmd,
                 return false;
         }
 
-        i = 3;
-        while (status == SDVO_CMD_STATUS_PENDING && i--) {
-                if (!intel_sdvo_read_byte(intel_sdvo,
-                                          SDVO_I2C_CMD_STATUS,
-                                          &status))
-                        return false;
-        }
-        if (status != SDVO_CMD_STATUS_SUCCESS) {
-                DRM_DEBUG_KMS("command returns response %s [%d]\n",
-                              status <= SDVO_CMD_STATUS_SCALING_NOT_SUPP ? cmd_status_names[status] : "???",
-                              status);
-                return false;
-        }
-
         return true;
 }
 
@@ -497,6 +484,8 @@ static bool intel_sdvo_read_response(struct intel_sdvo *intel_sdvo,
         u8 status;
         int i;
 
+        DRM_DEBUG_KMS("%s: R: ", SDVO_NAME(intel_sdvo));
+
         /*
          * The documentation states that all commands will be
          * processed within 15µs, and that we need only poll
@@ -505,14 +494,19 @@ static bool intel_sdvo_read_response(struct intel_sdvo *intel_sdvo,
          *
          * Check 5 times in case the hardware failed to read the docs.
          */
-        do {
+        if (!intel_sdvo_read_byte(intel_sdvo,
+                                  SDVO_I2C_CMD_STATUS,
+                                  &status))
+                goto log_fail;
+
+        while (status == SDVO_CMD_STATUS_PENDING && retry--) {
+                udelay(15);
                 if (!intel_sdvo_read_byte(intel_sdvo,
                                           SDVO_I2C_CMD_STATUS,
                                           &status))
-                        return false;
-        } while (status == SDVO_CMD_STATUS_PENDING && --retry);
+                        goto log_fail;
+        }
 
-        DRM_DEBUG_KMS("%s: R: ", SDVO_NAME(intel_sdvo));
         if (status <= SDVO_CMD_STATUS_SCALING_NOT_SUPP)
                 DRM_LOG_KMS("(%s)", cmd_status_names[status]);
         else
@@ -533,7 +527,7 @@ static bool intel_sdvo_read_response(struct intel_sdvo *intel_sdvo,
         return true;
 
 log_fail:
-        DRM_LOG_KMS("\n");
+        DRM_LOG_KMS("... failed\n");
         return false;
 }
 
@@ -550,6 +544,7 @@ static int intel_sdvo_get_pixel_multiplier(struct drm_display_mode *mode)
 static bool intel_sdvo_set_control_bus_switch(struct intel_sdvo *intel_sdvo,
                                               u8 ddc_bus)
 {
+        /* This must be the immediately preceding write before the i2c xfer */
         return intel_sdvo_write_cmd(intel_sdvo,
                                     SDVO_CMD_SET_CONTROL_BUS_SWITCH,
                                     &ddc_bus, 1);
@@ -557,7 +552,10 @@ static bool intel_sdvo_set_control_bus_switch(struct intel_sdvo *intel_sdvo,
 
 static bool intel_sdvo_set_value(struct intel_sdvo *intel_sdvo, u8 cmd, const void *data, int len)
 {
-        return intel_sdvo_write_cmd(intel_sdvo, cmd, data, len);
+        if (!intel_sdvo_write_cmd(intel_sdvo, cmd, data, len))
+                return false;
+
+        return intel_sdvo_read_response(intel_sdvo, NULL, 0);
 }
 
 static bool
@@ -859,18 +857,21 @@ static bool intel_sdvo_set_avi_infoframe(struct intel_sdvo *intel_sdvo)
 
         intel_dip_infoframe_csum(&avi_if);
 
-        if (!intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_HBUF_INDEX,
+        if (!intel_sdvo_set_value(intel_sdvo,
+                                  SDVO_CMD_SET_HBUF_INDEX,
                                   set_buf_index, 2))
                 return false;
 
         for (i = 0; i < sizeof(avi_if); i += 8) {
-                if (!intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_HBUF_DATA,
+                if (!intel_sdvo_set_value(intel_sdvo,
+                                          SDVO_CMD_SET_HBUF_DATA,
                                           data, 8))
                         return false;
                 data++;
         }
 
-        return intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_HBUF_TXRATE,
+        return intel_sdvo_set_value(intel_sdvo,
+                                    SDVO_CMD_SET_HBUF_TXRATE,
                                     &tx_rate, 1);
 }
 
@@ -1359,7 +1360,8 @@ intel_sdvo_hdmi_sink_detect(struct drm_connector *connector)
                                 intel_sdvo->has_hdmi_monitor = drm_detect_hdmi_monitor(edid);
                                 intel_sdvo->has_hdmi_audio = drm_detect_monitor_audio(edid);
                         }
-                }
+                } else
+                        status = connector_status_disconnected;
                 connector->display_info.raw_edid = NULL;
                 kfree(edid);
         }
@@ -1407,10 +1409,25 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 
         if ((intel_sdvo_connector->output_flag & response) == 0)
                 ret = connector_status_disconnected;
-        else if (response & SDVO_TMDS_MASK)
+        else if (IS_TMDS(intel_sdvo_connector))
                 ret = intel_sdvo_hdmi_sink_detect(connector);
-        else
-                ret = connector_status_connected;
+        else {
+                struct edid *edid;
+
+                /* if we have an edid check it matches the connection */
+                edid = intel_sdvo_get_edid(connector);
+                if (edid == NULL)
+                        edid = intel_sdvo_get_analog_edid(connector);
+                if (edid != NULL) {
+                        if (edid->input & DRM_EDID_INPUT_DIGITAL)
+                                ret = connector_status_disconnected;
+                        else
+                                ret = connector_status_connected;
+                        connector->display_info.raw_edid = NULL;
+                        kfree(edid);
+                } else
+                        ret = connector_status_connected;
+        }
 
         /* May update encoder flag for like clock for SDVO TV, etc.*/
         if (ret == connector_status_connected) {
@@ -1446,10 +1463,15 @@ static void intel_sdvo_get_ddc_modes(struct drm_connector *connector)
                 edid = intel_sdvo_get_analog_edid(connector);
 
         if (edid != NULL) {
-                if (edid->input & DRM_EDID_INPUT_DIGITAL) {
+                struct intel_sdvo_connector *intel_sdvo_connector = to_intel_sdvo_connector(connector);
+                bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL);
+                bool connector_is_digital = !!IS_TMDS(intel_sdvo_connector);
+
+                if (connector_is_digital == monitor_is_digital) {
                         drm_mode_connector_update_edid_property(connector, edid);
                         drm_add_edid_modes(connector, edid);
                 }
+
                 connector->display_info.raw_edid = NULL;
                 kfree(edid);
         }
@@ -1668,6 +1690,22 @@ static void intel_sdvo_destroy(struct drm_connector *connector)
         kfree(connector);
 }
 
+static bool intel_sdvo_detect_hdmi_audio(struct drm_connector *connector)
+{
+        struct intel_sdvo *intel_sdvo = intel_attached_sdvo(connector);
+        struct edid *edid;
+        bool has_audio = false;
+
+        if (!intel_sdvo->is_hdmi)
+                return false;
+
+        edid = intel_sdvo_get_edid(connector);
+        if (edid != NULL && edid->input & DRM_EDID_INPUT_DIGITAL)
+                has_audio = drm_detect_monitor_audio(edid);
+
+        return has_audio;
+}
+
 static int
 intel_sdvo_set_property(struct drm_connector *connector,
                         struct drm_property *property,
@@ -1684,17 +1722,23 @@ intel_sdvo_set_property(struct drm_connector *connector,
                 return ret;
 
         if (property == intel_sdvo_connector->force_audio_property) {
-                if (val == intel_sdvo_connector->force_audio)
+                int i = val;
+                bool has_audio;
+
+                if (i == intel_sdvo_connector->force_audio)
                         return 0;
 
-                intel_sdvo_connector->force_audio = val;
+                intel_sdvo_connector->force_audio = i;
 
-                if (val > 0 && intel_sdvo->has_hdmi_audio)
-                        return 0;
-                if (val < 0 && !intel_sdvo->has_hdmi_audio)
+                if (i == 0)
+                        has_audio = intel_sdvo_detect_hdmi_audio(connector);
+                else
+                        has_audio = i > 0;
+
+                if (has_audio == intel_sdvo->has_hdmi_audio)
                         return 0;
 
-                intel_sdvo->has_hdmi_audio = val > 0;
+                intel_sdvo->has_hdmi_audio = has_audio;
                 goto done;
         }
 
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 93206e4eaa6f..fe4a53a50b83 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1234,7 +1234,8 @@ static const struct drm_display_mode reported_modes[] = {
  * \return false if TV is disconnected.
  */
 static int
-intel_tv_detect_type (struct intel_tv *intel_tv)
+intel_tv_detect_type (struct intel_tv *intel_tv,
+                      struct drm_connector *connector)
 {
         struct drm_encoder *encoder = &intel_tv->base.base;
         struct drm_device *dev = encoder->dev;
@@ -1245,11 +1246,13 @@ intel_tv_detect_type (struct intel_tv *intel_tv)
         int type;
 
         /* Disable TV interrupts around load detect or we'll recurse */
-        spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-        i915_disable_pipestat(dev_priv, 0,
-                              PIPE_HOTPLUG_INTERRUPT_ENABLE |
-                              PIPE_HOTPLUG_TV_INTERRUPT_ENABLE);
-        spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+        if (connector->polled & DRM_CONNECTOR_POLL_HPD) {
+                spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+                i915_disable_pipestat(dev_priv, 0,
+                                      PIPE_HOTPLUG_INTERRUPT_ENABLE |
+                                      PIPE_HOTPLUG_TV_INTERRUPT_ENABLE);
+                spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+        }
 
         save_tv_dac = tv_dac = I915_READ(TV_DAC);
         save_tv_ctl = tv_ctl = I915_READ(TV_CTL);
@@ -1302,11 +1305,13 @@ intel_tv_detect_type (struct intel_tv *intel_tv)
         I915_WRITE(TV_CTL, save_tv_ctl);
 
         /* Restore interrupt config */
-        spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-        i915_enable_pipestat(dev_priv, 0,
-                             PIPE_HOTPLUG_INTERRUPT_ENABLE |
-                             PIPE_HOTPLUG_TV_INTERRUPT_ENABLE);
-        spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+        if (connector->polled & DRM_CONNECTOR_POLL_HPD) {
+                spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+                i915_enable_pipestat(dev_priv, 0,
+                                     PIPE_HOTPLUG_INTERRUPT_ENABLE |
+                                     PIPE_HOTPLUG_TV_INTERRUPT_ENABLE);
+                spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+        }
 
         return type;
 }
@@ -1356,7 +1361,7 @@ intel_tv_detect(struct drm_connector *connector, bool force)
         drm_mode_set_crtcinfo(&mode, CRTC_INTERLACE_HALVE_V);
 
         if (intel_tv->base.base.crtc && intel_tv->base.base.crtc->enabled) {
-                type = intel_tv_detect_type(intel_tv);
+                type = intel_tv_detect_type(intel_tv, connector);
         } else if (force) {
                 struct drm_crtc *crtc;
                 int dpms_mode;
@@ -1364,7 +1369,7 @@ intel_tv_detect(struct drm_connector *connector, bool force)
                 crtc = intel_get_load_detect_pipe(&intel_tv->base, connector,
                                                   &mode, &dpms_mode);
                 if (crtc) {
-                        type = intel_tv_detect_type(intel_tv);
+                        type = intel_tv_detect_type(intel_tv, connector);
                         intel_release_load_detect_pipe(&intel_tv->base, connector,
                                                        dpms_mode);
                 } else
@@ -1658,6 +1663,18 @@ intel_tv_init(struct drm_device *dev)
         intel_encoder = &intel_tv->base;
         connector = &intel_connector->base;
 
+        /* The documentation, for the older chipsets at least, recommend
+         * using a polling method rather than hotplug detection for TVs.
+         * This is because in order to perform the hotplug detection, the PLLs
+         * for the TV must be kept alive increasing power drain and starving
+         * bandwidth from other encoders. Notably for instance, it causes
+         * pipe underruns on Crestline when this encoder is supposedly idle.
+         *
+         * More recent chipsets favour HDMI rather than integrated S-Video.
+         */
+        connector->polled =
+                DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
+
         drm_connector_init(dev, connector, &intel_tv_connector_funcs,
                            DRM_MODE_CONNECTOR_SVIDEO);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index 2aef5cd3acf5..6bdab891c64e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -6228,7 +6228,7 @@ parse_dcb15_entry(struct drm_device *dev, struct dcb_table *dcb,
                 entry->tvconf.has_component_output = false;
                 break;
         case OUTPUT_LVDS:
-                if ((conn & 0x00003f00) != 0x10)
+                if ((conn & 0x00003f00) >> 8 != 0x10)
                         entry->lvdsconf.use_straps_for_mode = true;
                 entry->lvdsconf.use_power_scripts = true;
                 break;
@@ -6310,6 +6310,9 @@ void merge_like_dcb_entries(struct drm_device *dev, struct dcb_table *dcb)
 static bool
 apply_dcb_encoder_quirks(struct drm_device *dev, int idx, u32 *conn, u32 *conf)
 {
+        struct drm_nouveau_private *dev_priv = dev->dev_private;
+        struct dcb_table *dcb = &dev_priv->vbios.dcb;
+
         /* Dell Precision M6300
          *   DCB entry 2: 02025312 00000010
          *   DCB entry 3: 02026312 00000020
@@ -6327,6 +6330,18 @@ apply_dcb_encoder_quirks(struct drm_device *dev, int idx, u32 *conn, u32 *conf)
                         return false;
         }
 
+        /* GeForce3 Ti 200
+         *
+         * DCB reports an LVDS output that should be TMDS:
+         *   DCB entry 1: f2005014 ffffffff
+         */
+        if (nv_match_device(dev, 0x0201, 0x1462, 0x8851)) {
+                if (*conn == 0xf2005014 && *conf == 0xffffffff) {
+                        fabricate_dcb_output(dcb, OUTPUT_TMDS, 1, 1, 1);
+                        return false;
+                }
+        }
+
         return true;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index a7fae26f4654..d38a4d9f9b0b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -128,6 +128,7 @@ nouveau_bo_new(struct drm_device *dev, struct nouveau_channel *chan,
                 }
         }
 
+        nvbo->bo.mem.num_pages = size >> PAGE_SHIFT;
         nouveau_bo_placement_set(nvbo, flags, 0);
 
         nvbo->channel = chan;
@@ -166,17 +167,17 @@ static void
 set_placement_range(struct nouveau_bo *nvbo, uint32_t type)
 {
         struct drm_nouveau_private *dev_priv = nouveau_bdev(nvbo->bo.bdev);
+        int vram_pages = dev_priv->vram_size >> PAGE_SHIFT;
 
         if (dev_priv->card_type == NV_10 &&
-            nvbo->tile_mode && (type & TTM_PL_FLAG_VRAM)) {
+            nvbo->tile_mode && (type & TTM_PL_FLAG_VRAM) &&
+            nvbo->bo.mem.num_pages < vram_pages / 2) {
                 /*
                  * Make sure that the color and depth buffers are handled
                  * by independent memory controller units. Up to a 9x
                  * speed up when alpha-blending and depth-test are enabled
                  * at the same time.
                  */
-                int vram_pages = dev_priv->vram_size >> PAGE_SHIFT;
-
                 if (nvbo->tile_flags & NOUVEAU_GEM_TILE_ZETA) {
                         nvbo->placement.fpfn = vram_pages / 2;
                         nvbo->placement.lpfn = ~0;
@@ -785,7 +786,7 @@ nouveau_bo_move_flipd(struct ttm_buffer_object *bo, bool evict, bool intr,
         if (ret)
                 goto out;
 
-        ret = ttm_bo_move_ttm(bo, evict, no_wait_reserve, no_wait_gpu, new_mem);
+        ret = ttm_bo_move_ttm(bo, true, no_wait_reserve, no_wait_gpu, new_mem);
 out:
         ttm_bo_mem_put(bo, &tmp_mem);
         return ret;
@@ -811,11 +812,11 @@ nouveau_bo_move_flips(struct ttm_buffer_object *bo, bool evict, bool intr,
         if (ret)
                 return ret;
 
-        ret = ttm_bo_move_ttm(bo, evict, no_wait_reserve, no_wait_gpu, &tmp_mem);
+        ret = ttm_bo_move_ttm(bo, true, no_wait_reserve, no_wait_gpu, &tmp_mem);
         if (ret)
                 goto out;
 
-        ret = nouveau_bo_move_m2mf(bo, evict, intr, no_wait_reserve, no_wait_gpu, new_mem);
+        ret = nouveau_bo_move_m2mf(bo, true, intr, no_wait_reserve, no_wait_gpu, new_mem);
         if (ret)
                 goto out;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index a21e00076839..390d82c3c4b0 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -507,6 +507,7 @@ nouveau_connector_native_mode(struct drm_connector *connector)
         int high_w = 0, high_h = 0, high_v = 0;
 
         list_for_each_entry(mode, &nv_connector->base.probed_modes, head) {
+                mode->vrefresh = drm_mode_vrefresh(mode);
                 if (helper->mode_valid(connector, mode) != MODE_OK ||
                     (mode->flags & DRM_MODE_FLAG_INTERLACE))
                         continue;
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.c b/drivers/gpu/drm/nouveau/nouveau_drv.c
index 13bb672a16f4..f658a04eecf9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.c
@@ -234,9 +234,9 @@ nouveau_pci_suspend(struct pci_dev *pdev, pm_message_t pm_state)
                 pci_set_power_state(pdev, PCI_D3hot);
         }
 
-        acquire_console_sem();
+        console_lock();
         nouveau_fbcon_set_suspend(dev, 1);
-        release_console_sem();
+        console_unlock();
         nouveau_fbcon_restore_accel(dev);
         return 0;
 
@@ -359,9 +359,9 @@ nouveau_pci_resume(struct pci_dev *pdev)
                 nv_crtc->lut.depth = 0;
         }
 
-        acquire_console_sem();
+        console_lock();
         nouveau_fbcon_set_suspend(dev, 0);
-        release_console_sem();
+        console_unlock();
 
         nouveau_fbcon_zfill_all(dev);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index 01bffc4412d2..9821fcacc3d2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -848,9 +848,6 @@ extern void nv10_mem_put_tile_region(struct drm_device *dev,
                                      struct nouveau_fence *fence);
 extern const struct ttm_mem_type_manager_func nouveau_vram_manager;
 
-/* nvc0_vram.c */
-extern const struct ttm_mem_type_manager_func nvc0_vram_manager;
-
 /* nouveau_notifier.c */
 extern int  nouveau_notifier_init_channel(struct nouveau_channel *);
 extern void nouveau_notifier_takedown_channel(struct nouveau_channel *);
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.c b/drivers/gpu/drm/nouveau/nouveau_pm.c
index fb846a3fef15..4399e2f34db4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_pm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_pm.c
@@ -443,7 +443,7 @@ nouveau_hwmon_fini(struct drm_device *dev)
         struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
 
         if (pm->hwmon) {
-                sysfs_remove_group(&pm->hwmon->kobj, &hwmon_attrgroup);
+                sysfs_remove_group(&dev->pdev->dev.kobj, &hwmon_attrgroup);
                 hwmon_device_unregister(pm->hwmon);
         }
 #endif
@@ -543,7 +543,7 @@ nouveau_pm_resume(struct drm_device *dev)
         struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
         struct nouveau_pm_level *perflvl;
 
-        if (pm->cur == &pm->boot)
+        if (!pm->cur || pm->cur == &pm->boot)
                 return;
 
         perflvl = pm->cur;
diff --git a/drivers/gpu/drm/nouveau/nouveau_temp.c b/drivers/gpu/drm/nouveau/nouveau_temp.c
index 7ecc4adc1e45..8d9968e1cba8 100644
--- a/drivers/gpu/drm/nouveau/nouveau_temp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_temp.c
@@ -265,8 +265,8 @@ nouveau_temp_probe_i2c(struct drm_device *dev)
         struct i2c_board_info info[] = {
                 { I2C_BOARD_INFO("w83l785ts", 0x2d) },
                 { I2C_BOARD_INFO("w83781d", 0x2d) },
-                { I2C_BOARD_INFO("f75375", 0x2e) },
                 { I2C_BOARD_INFO("adt7473", 0x2e) },
+                { I2C_BOARD_INFO("f75375", 0x2e) },
                 { I2C_BOARD_INFO("lm99", 0x4c) },
                 { }
         };
diff --git a/drivers/gpu/drm/nouveau/nv04_dfp.c b/drivers/gpu/drm/nouveau/nv04_dfp.c
index ef23550407b5..c82db37d9f41 100644
--- a/drivers/gpu/drm/nouveau/nv04_dfp.c
+++ b/drivers/gpu/drm/nouveau/nv04_dfp.c
@@ -342,8 +342,8 @@ static void nv04_dfp_mode_set(struct drm_encoder *encoder,
         if (nv_encoder->dcb->type == OUTPUT_LVDS) {
                 bool duallink, dummy;
 
-                nouveau_bios_parse_lvds_table(dev, nv_connector->native_mode->
-                                              clock, &duallink, &dummy);
+                nouveau_bios_parse_lvds_table(dev, output_mode->clock,
+                                              &duallink, &dummy);
                 if (duallink)
                         regp->fp_control |= (8 << 28);
         } else
@@ -518,8 +518,6 @@ static void nv04_lvds_dpms(struct drm_encoder *encoder, int mode)
                 return;
 
         if (nv_encoder->dcb->lvdsconf.use_power_scripts) {
-                struct nouveau_connector *nv_connector = nouveau_encoder_connector_get(nv_encoder);
-
                 /* when removing an output, crtc may not be set, but PANEL_OFF
                  * must still be run
                  */
@@ -527,12 +525,8 @@ static void nv04_lvds_dpms(struct drm_encoder *encoder, int mode)
                            nv04_dfp_get_bound_head(dev, nv_encoder->dcb);
 
                 if (mode == DRM_MODE_DPMS_ON) {
-                        if (!nv_connector->native_mode) {
-                                NV_ERROR(dev, "Not turning on LVDS without native mode\n");
-                                return;
-                        }
                         call_lvds_script(dev, nv_encoder->dcb, head,
-                                         LVDS_PANEL_ON, nv_connector->native_mode->clock);
+                                         LVDS_PANEL_ON, nv_encoder->mode.clock);
                 } else
                         /* pxclk of 0 is fine for PANEL_OFF, and for a
                          * disconnected LVDS encoder there is no native_mode
diff --git a/drivers/gpu/drm/nouveau/nv40_graph.c b/drivers/gpu/drm/nouveau/nv40_graph.c
index 8870d72388c8..18d30c2c1aa6 100644
--- a/drivers/gpu/drm/nouveau/nv40_graph.c
+++ b/drivers/gpu/drm/nouveau/nv40_graph.c
@@ -211,18 +211,32 @@ nv40_graph_set_tile_region(struct drm_device *dev, int i)
         struct nouveau_tile_reg *tile = &dev_priv->tile.reg[i];
 
         switch (dev_priv->chipset) {
+        case 0x40:
+        case 0x41: /* guess */
+        case 0x42:
+        case 0x43:
+        case 0x45: /* guess */
+        case 0x4e:
+                nv_wr32(dev, NV20_PGRAPH_TSIZE(i), tile->pitch);
+                nv_wr32(dev, NV20_PGRAPH_TLIMIT(i), tile->limit);
+                nv_wr32(dev, NV20_PGRAPH_TILE(i), tile->addr);
+                nv_wr32(dev, NV40_PGRAPH_TSIZE1(i), tile->pitch);
+                nv_wr32(dev, NV40_PGRAPH_TLIMIT1(i), tile->limit);
+                nv_wr32(dev, NV40_PGRAPH_TILE1(i), tile->addr);
+                break;
         case 0x44:
         case 0x4a:
-        case 0x4e:
                 nv_wr32(dev, NV20_PGRAPH_TSIZE(i), tile->pitch);
                 nv_wr32(dev, NV20_PGRAPH_TLIMIT(i), tile->limit);
                 nv_wr32(dev, NV20_PGRAPH_TILE(i), tile->addr);
                 break;
-
         case 0x46:
         case 0x47:
         case 0x49:
         case 0x4b:
+        case 0x4c:
+        case 0x67:
+        default:
                 nv_wr32(dev, NV47_PGRAPH_TSIZE(i), tile->pitch);
                 nv_wr32(dev, NV47_PGRAPH_TLIMIT(i), tile->limit);
                 nv_wr32(dev, NV47_PGRAPH_TILE(i), tile->addr);
@@ -230,15 +244,6 @@ nv40_graph_set_tile_region(struct drm_device *dev, int i)
                 nv_wr32(dev, NV40_PGRAPH_TLIMIT1(i), tile->limit);
                 nv_wr32(dev, NV40_PGRAPH_TILE1(i), tile->addr);
                 break;
-
-        default:
-                nv_wr32(dev, NV20_PGRAPH_TSIZE(i), tile->pitch);
-                nv_wr32(dev, NV20_PGRAPH_TLIMIT(i), tile->limit);
-                nv_wr32(dev, NV20_PGRAPH_TILE(i), tile->addr);
-                nv_wr32(dev, NV40_PGRAPH_TSIZE1(i), tile->pitch);
-                nv_wr32(dev, NV40_PGRAPH_TLIMIT1(i), tile->limit);
-                nv_wr32(dev, NV40_PGRAPH_TILE1(i), tile->addr);
-                break;
         }
 }
 
@@ -396,17 +401,20 @@ nv40_graph_init(struct drm_device *dev)
                 break;
         default:
                 switch (dev_priv->chipset) {
-                case 0x46:
-                case 0x47:
-                case 0x49:
-                case 0x4b:
-                        nv_wr32(dev, 0x400DF0, nv_rd32(dev, NV04_PFB_CFG0));
-                        nv_wr32(dev, 0x400DF4, nv_rd32(dev, NV04_PFB_CFG1));
-                        break;
-                default:
+                case 0x41:
+                case 0x42:
+                case 0x43:
+                case 0x45:
+                case 0x4e:
+                case 0x44:
+                case 0x4a:
                         nv_wr32(dev, 0x4009F0, nv_rd32(dev, NV04_PFB_CFG0));
                         nv_wr32(dev, 0x4009F4, nv_rd32(dev, NV04_PFB_CFG1));
                         break;
+                default:
+                        nv_wr32(dev, 0x400DF0, nv_rd32(dev, NV04_PFB_CFG0));
+                        nv_wr32(dev, 0x400DF4, nv_rd32(dev, NV04_PFB_CFG1));
+                        break;
                 }
                 nv_wr32(dev, 0x4069F0, nv_rd32(dev, NV04_PFB_CFG0));
                 nv_wr32(dev, 0x4069F4, nv_rd32(dev, NV04_PFB_CFG1));
diff --git a/drivers/gpu/drm/nouveau/nv50_evo.c b/drivers/gpu/drm/nouveau/nv50_evo.c
index 14e24e906ee8..0ea090f4244a 100644
--- a/drivers/gpu/drm/nouveau/nv50_evo.c
+++ b/drivers/gpu/drm/nouveau/nv50_evo.c
@@ -283,8 +283,7 @@ nv50_evo_create(struct drm_device *dev)
                         nv50_evo_channel_del(&dev_priv->evo);
                         return ret;
                 }
-        } else
-        if (dev_priv->chipset != 0x50) {
+        } else {
                 ret = nv50_evo_dmaobj_new(evo, 0x3d, NvEvoFB16, 0x70, 0x19,
                                           0, 0xffffffff, 0x00010000);
                 if (ret) {
diff --git a/drivers/gpu/drm/nouveau/nv50_graph.c b/drivers/gpu/drm/nouveau/nv50_graph.c
index 2d7ea75a09d4..37e21d2be95b 100644
--- a/drivers/gpu/drm/nouveau/nv50_graph.c
+++ b/drivers/gpu/drm/nouveau/nv50_graph.c
@@ -256,6 +256,7 @@ nv50_graph_destroy_context(struct nouveau_channel *chan)
         struct drm_device *dev = chan->dev;
         struct drm_nouveau_private *dev_priv = dev->dev_private;
         struct nouveau_pgraph_engine *pgraph = &dev_priv->engine.graph;
+        struct nouveau_fifo_engine *pfifo = &dev_priv->engine.fifo;
         int i, hdr = (dev_priv->chipset == 0x50) ? 0x200 : 0x20;
         unsigned long flags;
 
@@ -265,6 +266,7 @@ nv50_graph_destroy_context(struct nouveau_channel *chan)
                 return;
 
         spin_lock_irqsave(&dev_priv->context_switch_lock, flags);
+        pfifo->reassign(dev, false);
         pgraph->fifo_access(dev, false);
 
         if (pgraph->channel(dev) == chan)
@@ -275,6 +277,7 @@ nv50_graph_destroy_context(struct nouveau_channel *chan)
         dev_priv->engine.instmem.flush(dev);
 
         pgraph->fifo_access(dev, true);
+        pfifo->reassign(dev, true);
         spin_unlock_irqrestore(&dev_priv->context_switch_lock, flags);
 
         nouveau_gpuobj_ref(NULL, &chan->ramin_grctx);
diff --git a/drivers/gpu/drm/nouveau/nv50_vm.c b/drivers/gpu/drm/nouveau/nv50_vm.c
index 38e523e10995..459ff08241e5 100644
--- a/drivers/gpu/drm/nouveau/nv50_vm.c
+++ b/drivers/gpu/drm/nouveau/nv50_vm.c
@@ -45,11 +45,6 @@ nv50_vm_map_pgt(struct nouveau_gpuobj *pgd, u32 pde,
         }
 
         if (phys & 1) {
-                if (dev_priv->vram_sys_base) {
-                        phys += dev_priv->vram_sys_base;
-                        phys |= 0x30;
-                }
-
                 if (coverage <= 32 * 1024 * 1024)
                         phys |= 0x60;
                 else if (coverage <= 64 * 1024 * 1024)
diff --git a/drivers/gpu/drm/nouveau/nvc0_graph.c b/drivers/gpu/drm/nouveau/nvc0_graph.c
index e6ea7d83187f..eb18a7e89f5b 100644
--- a/drivers/gpu/drm/nouveau/nvc0_graph.c
+++ b/drivers/gpu/drm/nouveau/nvc0_graph.c
@@ -31,6 +31,7 @@
 #include "nvc0_graph.h"
 
 static void nvc0_graph_isr(struct drm_device *);
+static void nvc0_runk140_isr(struct drm_device *);
 static int  nvc0_graph_unload_context_to(struct drm_device *dev, u64 chan);
 
 void
@@ -281,6 +282,7 @@ nvc0_graph_destroy(struct drm_device *dev)
                 return;
 
         nouveau_irq_unregister(dev, 12);
+        nouveau_irq_unregister(dev, 25);
 
         nouveau_gpuobj_ref(NULL, &priv->unk4188b8);
         nouveau_gpuobj_ref(NULL, &priv->unk4188b4);
@@ -390,6 +392,7 @@ nvc0_graph_create(struct drm_device *dev)
         }
 
         nouveau_irq_register(dev, 12, nvc0_graph_isr);
+        nouveau_irq_register(dev, 25, nvc0_runk140_isr);
         NVOBJ_CLASS(dev, 0x902d, GR); /* 2D */
         NVOBJ_CLASS(dev, 0x9039, GR); /* M2MF */
         NVOBJ_CLASS(dev, 0x9097, GR); /* 3D */
@@ -512,8 +515,8 @@ nvc0_graph_init_gpc_1(struct drm_device *dev)
                         nv_wr32(dev, TP_UNIT(gpc, tp, 0x224), 0xc0000000);
                         nv_wr32(dev, TP_UNIT(gpc, tp, 0x48c), 0xc0000000);
                         nv_wr32(dev, TP_UNIT(gpc, tp, 0x084), 0xc0000000);
-                        nv_wr32(dev, TP_UNIT(gpc, tp, 0xe44), 0x001ffffe);
-                        nv_wr32(dev, TP_UNIT(gpc, tp, 0xe4c), 0x0000000f);
+                        nv_wr32(dev, TP_UNIT(gpc, tp, 0x644), 0x001ffffe);
+                        nv_wr32(dev, TP_UNIT(gpc, tp, 0x64c), 0x0000000f);
                 }
                 nv_wr32(dev, GPC_UNIT(gpc, 0x2c90), 0xffffffff);
                 nv_wr32(dev, GPC_UNIT(gpc, 0x2c94), 0xffffffff);
@@ -777,3 +780,19 @@ nvc0_graph_isr(struct drm_device *dev)
 
         nv_wr32(dev, 0x400500, 0x00010001);
 }
+
+static void
+nvc0_runk140_isr(struct drm_device *dev)
+{
+        u32 units = nv_rd32(dev, 0x00017c) & 0x1f;
+
+        while (units) {
+                u32 unit = ffs(units) - 1;
+                u32 reg = 0x140000 + unit * 0x2000;
+                u32 st0 = nv_mask(dev, reg + 0x1020, 0, 0);
+                u32 st1 = nv_mask(dev, reg + 0x1420, 0, 0);
+
+                NV_INFO(dev, "PRUNK140: %d 0x%08x 0x%08x\n", unit, st0, st1);
+                units &= ~(1 << unit);
+        }
+}
diff --git a/drivers/gpu/drm/nouveau/nvc0_grctx.c b/drivers/gpu/drm/nouveau/nvc0_grctx.c
index b9e68b2d30aa..f880ff776db8 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grctx.c
+++ b/drivers/gpu/drm/nouveau/nvc0_grctx.c
@@ -1830,7 +1830,7 @@ nvc0_grctx_generate(struct nouveau_channel *chan)
 
         for (tp = 0, id = 0; tp < 4; tp++) {
                 for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
-                        if (tp <= priv->tp_nr[gpc]) {
+                        if (tp < priv->tp_nr[gpc]) {
                                 nv_wr32(dev, TP_UNIT(gpc, tp, 0x698), id);
                                 nv_wr32(dev, TP_UNIT(gpc, tp, 0x4e8), id);
                                 nv_wr32(dev, GPC_UNIT(gpc, 0x0c10 + tp * 4), id);
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index d3ca17080df7..a4e5e53e0a62 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -48,29 +48,29 @@ static void atombios_overscan_setup(struct drm_crtc *crtc,
 
         switch (radeon_crtc->rmx_type) {
         case RMX_CENTER:
-                args.usOverscanTop = (adjusted_mode->crtc_vdisplay - mode->crtc_vdisplay) / 2;
-                args.usOverscanBottom = (adjusted_mode->crtc_vdisplay - mode->crtc_vdisplay) / 2;
-                args.usOverscanLeft = (adjusted_mode->crtc_hdisplay - mode->crtc_hdisplay) / 2;
-                args.usOverscanRight = (adjusted_mode->crtc_hdisplay - mode->crtc_hdisplay) / 2;
+                args.usOverscanTop = cpu_to_le16((adjusted_mode->crtc_vdisplay - mode->crtc_vdisplay) / 2);
+                args.usOverscanBottom = cpu_to_le16((adjusted_mode->crtc_vdisplay - mode->crtc_vdisplay) / 2);
+                args.usOverscanLeft = cpu_to_le16((adjusted_mode->crtc_hdisplay - mode->crtc_hdisplay) / 2);
+                args.usOverscanRight = cpu_to_le16((adjusted_mode->crtc_hdisplay - mode->crtc_hdisplay) / 2);
                 break;
         case RMX_ASPECT:
                 a1 = mode->crtc_vdisplay * adjusted_mode->crtc_hdisplay;
                 a2 = adjusted_mode->crtc_vdisplay * mode->crtc_hdisplay;
 
                 if (a1 > a2) {
-                        args.usOverscanLeft = (adjusted_mode->crtc_hdisplay - (a2 / mode->crtc_vdisplay)) / 2;
-                        args.usOverscanRight = (adjusted_mode->crtc_hdisplay - (a2 / mode->crtc_vdisplay)) / 2;
+                        args.usOverscanLeft = cpu_to_le16((adjusted_mode->crtc_hdisplay - (a2 / mode->crtc_vdisplay)) / 2);
+                        args.usOverscanRight = cpu_to_le16((adjusted_mode->crtc_hdisplay - (a2 / mode->crtc_vdisplay)) / 2);
                 } else if (a2 > a1) {
-                        args.usOverscanLeft = (adjusted_mode->crtc_vdisplay - (a1 / mode->crtc_hdisplay)) / 2;
-                        args.usOverscanRight = (adjusted_mode->crtc_vdisplay - (a1 / mode->crtc_hdisplay)) / 2;
+                        args.usOverscanLeft = cpu_to_le16((adjusted_mode->crtc_vdisplay - (a1 / mode->crtc_hdisplay)) / 2);
+                        args.usOverscanRight = cpu_to_le16((adjusted_mode->crtc_vdisplay - (a1 / mode->crtc_hdisplay)) / 2);
                 }
                 break;
         case RMX_FULL:
         default:
-                args.usOverscanRight = radeon_crtc->h_border;
-                args.usOverscanLeft = radeon_crtc->h_border;
-                args.usOverscanBottom = radeon_crtc->v_border;
-                args.usOverscanTop = radeon_crtc->v_border;
+                args.usOverscanRight = cpu_to_le16(radeon_crtc->h_border);
+                args.usOverscanLeft = cpu_to_le16(radeon_crtc->h_border);
+                args.usOverscanBottom = cpu_to_le16(radeon_crtc->v_border);
+                args.usOverscanTop = cpu_to_le16(radeon_crtc->v_border);
                 break;
         }
         atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
@@ -419,23 +419,23 @@ static void atombios_crtc_program_ss(struct drm_crtc *crtc,
         memset(&args, 0, sizeof(args));
 
         if (ASIC_IS_DCE5(rdev)) {
-                args.v3.usSpreadSpectrumAmountFrac = 0;
+                args.v3.usSpreadSpectrumAmountFrac = cpu_to_le16(0);
                 args.v3.ucSpreadSpectrumType = ss->type;
                 switch (pll_id) {
                 case ATOM_PPLL1:
                         args.v3.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V3_P1PLL;
-                        args.v3.usSpreadSpectrumAmount = ss->amount;
-                        args.v3.usSpreadSpectrumStep = ss->step;
+                        args.v3.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
+                        args.v3.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_PPLL2:
                         args.v3.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V3_P2PLL;
-                        args.v3.usSpreadSpectrumAmount = ss->amount;
-                        args.v3.usSpreadSpectrumStep = ss->step;
+                        args.v3.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
+                        args.v3.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_DCPLL:
                         args.v3.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V3_DCPLL;
-                        args.v3.usSpreadSpectrumAmount = 0;
-                        args.v3.usSpreadSpectrumStep = 0;
+                        args.v3.usSpreadSpectrumAmount = cpu_to_le16(0);
+                        args.v3.usSpreadSpectrumStep = cpu_to_le16(0);
                         break;
                 case ATOM_PPLL_INVALID:
                         return;
@@ -447,18 +447,18 @@ static void atombios_crtc_program_ss(struct drm_crtc *crtc,
                 switch (pll_id) {
                 case ATOM_PPLL1:
                         args.v2.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V2_P1PLL;
-                        args.v2.usSpreadSpectrumAmount = ss->amount;
-                        args.v2.usSpreadSpectrumStep = ss->step;
+                        args.v2.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
+                        args.v2.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_PPLL2:
                         args.v2.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V2_P2PLL;
-                        args.v2.usSpreadSpectrumAmount = ss->amount;
-                        args.v2.usSpreadSpectrumStep = ss->step;
+                        args.v2.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
+                        args.v2.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_DCPLL:
                         args.v2.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V2_DCPLL;
-                        args.v2.usSpreadSpectrumAmount = 0;
-                        args.v2.usSpreadSpectrumStep = 0;
+                        args.v2.usSpreadSpectrumAmount = cpu_to_le16(0);
+                        args.v2.usSpreadSpectrumStep = cpu_to_le16(0);
                         break;
                 case ATOM_PPLL_INVALID:
                         return;
@@ -538,7 +538,6 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
                         pll->flags |= RADEON_PLL_PREFER_HIGH_FB_DIV;
                 else
                         pll->flags |= RADEON_PLL_PREFER_LOW_REF_DIV;
-
         }
 
         list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
@@ -555,23 +554,28 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
                                         dp_clock = dig_connector->dp_clock;
                                 }
                         }
-#if 0 /* doesn't work properly on some laptops */
+
                         /* use recommended ref_div for ss */
                         if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) {
                                 if (ss_enabled) {
                                         if (ss->refdiv) {
+                                                pll->flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP;
                                                 pll->flags |= RADEON_PLL_USE_REF_DIV;
                                                 pll->reference_div = ss->refdiv;
+                                                if (ASIC_IS_AVIVO(rdev))
+                                                        pll->flags |= RADEON_PLL_USE_FRAC_FB_DIV;
                                         }
                                 }
                         }
-#endif
+
                         if (ASIC_IS_AVIVO(rdev)) {
                                 /* DVO wants 2x pixel clock if the DVO chip is in 12 bit mode */
                                 if (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1)
                                         adjusted_clock = mode->clock * 2;
                                 if (radeon_encoder->active_device & (ATOM_DEVICE_TV_SUPPORT))
                                         pll->flags |= RADEON_PLL_PREFER_CLOSEST_LOWER;
+                                if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT))
+                                        pll->flags |= RADEON_PLL_IS_LCD;
                         } else {
                                 if (encoder->encoder_type != DRM_MODE_ENCODER_DAC)
                                         pll->flags |= RADEON_PLL_NO_ODD_POST_DIV;
@@ -658,10 +662,12 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
                                                    index, (uint32_t *)&args);
                                 adjusted_clock = le32_to_cpu(args.v3.sOutput.ulDispPllFreq) * 10;
                                 if (args.v3.sOutput.ucRefDiv) {
+                                        pll->flags |= RADEON_PLL_USE_FRAC_FB_DIV;
                                         pll->flags |= RADEON_PLL_USE_REF_DIV;
                                         pll->reference_div = args.v3.sOutput.ucRefDiv;
                                 }
                                 if (args.v3.sOutput.ucPostDiv) {
+                                        pll->flags |= RADEON_PLL_USE_FRAC_FB_DIV;
                                         pll->flags |= RADEON_PLL_USE_POST_DIV;
                                         pll->post_div = args.v3.sOutput.ucPostDiv;
                                 }
@@ -715,14 +721,14 @@ static void atombios_crtc_set_dcpll(struct drm_crtc *crtc,
                          * SetPixelClock provides the dividers
                          */
                         args.v5.ucCRTC = ATOM_CRTC_INVALID;
-                        args.v5.usPixelClock = dispclk;
+                        args.v5.usPixelClock = cpu_to_le16(dispclk);
                         args.v5.ucPpll = ATOM_DCPLL;
                         break;
                 case 6:
                         /* if the default dcpll clock is specified,
                          * SetPixelClock provides the dividers
                          */
-                        args.v6.ulDispEngClkFreq = dispclk;
+                        args.v6.ulDispEngClkFreq = cpu_to_le32(dispclk);
                         args.v6.ucPpll = ATOM_DCPLL;
                         break;
                 default:
@@ -951,8 +957,12 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode
         /* adjust pixel clock as needed */
         adjusted_clock = atombios_adjust_pll(crtc, mode, pll, ss_enabled, &ss);
 
-        radeon_compute_pll(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div,
-                           &ref_div, &post_div);
+        if (ASIC_IS_AVIVO(rdev))
+                radeon_compute_pll_avivo(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div,
+                                         &ref_div, &post_div);
+        else
+                radeon_compute_pll_legacy(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div,
+                                          &ref_div, &post_div);
 
         atombios_crtc_program_ss(crtc, ATOM_DISABLE, radeon_crtc->pll_id, &ss);
 
@@ -981,9 +991,9 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode
         }
 }
 
-static int evergreen_crtc_do_set_base(struct drm_crtc *crtc,
-                                      struct drm_framebuffer *fb,
-                                      int x, int y, int atomic)
+static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
+                                 struct drm_framebuffer *fb,
+                                 int x, int y, int atomic)
 {
         struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
         struct drm_device *dev = crtc->dev;
@@ -994,6 +1004,7 @@ static int evergreen_crtc_do_set_base(struct drm_crtc *crtc,
         struct radeon_bo *rbo;
         uint64_t fb_location;
         uint32_t fb_format, fb_pitch_pixels, tiling_flags;
+        u32 fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_NONE);
         int r;
 
         /* no fb bound */
@@ -1045,11 +1056,17 @@ static int evergreen_crtc_do_set_base(struct drm_crtc *crtc,
         case 16:
                 fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_16BPP) |
                              EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_ARGB565));
+#ifdef __BIG_ENDIAN
+                fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN16);
+#endif
                 break;
         case 24:
         case 32:
                 fb_format = (EVERGREEN_GRPH_DEPTH(EVERGREEN_GRPH_DEPTH_32BPP) |
                              EVERGREEN_GRPH_FORMAT(EVERGREEN_GRPH_FORMAT_ARGB8888));
+#ifdef __BIG_ENDIAN
+                fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_8IN32);
+#endif
                 break;
         default:
                 DRM_ERROR("Unsupported screen depth %d\n",
@@ -1094,6 +1111,7 @@ static int evergreen_crtc_do_set_base(struct drm_crtc *crtc,
         WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
                (u32) fb_location & EVERGREEN_GRPH_SURFACE_ADDRESS_MASK);
         WREG32(EVERGREEN_GRPH_CONTROL + radeon_crtc->crtc_offset, fb_format);
+        WREG32(EVERGREEN_GRPH_SWAP_CONTROL + radeon_crtc->crtc_offset, fb_swap);
 
         WREG32(EVERGREEN_GRPH_SURFACE_OFFSET_X + radeon_crtc->crtc_offset, 0);
         WREG32(EVERGREEN_GRPH_SURFACE_OFFSET_Y + radeon_crtc->crtc_offset, 0);
@@ -1115,12 +1133,6 @@ static int evergreen_crtc_do_set_base(struct drm_crtc *crtc,
         WREG32(EVERGREEN_VIEWPORT_SIZE + radeon_crtc->crtc_offset,
                (crtc->mode.hdisplay << 16) | crtc->mode.vdisplay);
 
-        if (crtc->mode.flags & DRM_MODE_FLAG_INTERLACE)
-                WREG32(EVERGREEN_DATA_FORMAT + radeon_crtc->crtc_offset,
-                       EVERGREEN_INTERLEAVE_EN);
-        else
-                WREG32(EVERGREEN_DATA_FORMAT + radeon_crtc->crtc_offset, 0);
-
         if (!atomic && fb && fb != crtc->fb) {
                 radeon_fb = to_radeon_framebuffer(fb);
                 rbo = radeon_fb->obj->driver_private;
@@ -1150,6 +1162,7 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
         struct drm_framebuffer *target_fb;
         uint64_t fb_location;
         uint32_t fb_format, fb_pitch_pixels, tiling_flags;
+        u32 fb_swap = R600_D1GRPH_SWAP_ENDIAN_NONE;
         int r;
 
         /* no fb bound */
@@ -1203,12 +1216,18 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
                 fb_format =
                     AVIVO_D1GRPH_CONTROL_DEPTH_16BPP |
                     AVIVO_D1GRPH_CONTROL_16BPP_RGB565;
+#ifdef __BIG_ENDIAN
+                fb_swap = R600_D1GRPH_SWAP_ENDIAN_16BIT;
+#endif
                 break;
         case 24:
         case 32:
                 fb_format =
                     AVIVO_D1GRPH_CONTROL_DEPTH_32BPP |
                     AVIVO_D1GRPH_CONTROL_32BPP_ARGB8888;
+#ifdef __BIG_ENDIAN
+                fb_swap = R600_D1GRPH_SWAP_ENDIAN_32BIT;
+#endif
                 break;
         default:
                 DRM_ERROR("Unsupported screen depth %d\n",
@@ -1248,6 +1267,8 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
         WREG32(AVIVO_D1GRPH_SECONDARY_SURFACE_ADDRESS +
                radeon_crtc->crtc_offset, (u32) fb_location);
         WREG32(AVIVO_D1GRPH_CONTROL + radeon_crtc->crtc_offset, fb_format);
+        if (rdev->family >= CHIP_R600)
+                WREG32(R600_D1GRPH_SWAP_CONTROL + radeon_crtc->crtc_offset, fb_swap);
 
         WREG32(AVIVO_D1GRPH_SURFACE_OFFSET_X + radeon_crtc->crtc_offset, 0);
         WREG32(AVIVO_D1GRPH_SURFACE_OFFSET_Y + radeon_crtc->crtc_offset, 0);
@@ -1269,12 +1290,6 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
         WREG32(AVIVO_D1MODE_VIEWPORT_SIZE + radeon_crtc->crtc_offset,
                (crtc->mode.hdisplay << 16) | crtc->mode.vdisplay);
 
-        if (crtc->mode.flags & DRM_MODE_FLAG_INTERLACE)
-                WREG32(AVIVO_D1MODE_DATA_FORMAT + radeon_crtc->crtc_offset,
-                       AVIVO_D1MODE_INTERLEAVE_EN);
-        else
-                WREG32(AVIVO_D1MODE_DATA_FORMAT + radeon_crtc->crtc_offset, 0);
-
         if (!atomic && fb && fb != crtc->fb) {
                 radeon_fb = to_radeon_framebuffer(fb);
                 rbo = radeon_fb->obj->driver_private;
@@ -1298,7 +1313,7 @@ int atombios_crtc_set_base(struct drm_crtc *crtc, int x, int y,
         struct radeon_device *rdev = dev->dev_private;
 
         if (ASIC_IS_DCE4(rdev))
-                return evergreen_crtc_do_set_base(crtc, old_fb, x, y, 0);
+                return dce4_crtc_do_set_base(crtc, old_fb, x, y, 0);
         else if (ASIC_IS_AVIVO(rdev))
                 return avivo_crtc_do_set_base(crtc, old_fb, x, y, 0);
         else
@@ -1313,7 +1328,7 @@ int atombios_crtc_set_base_atomic(struct drm_crtc *crtc,
        struct radeon_device *rdev = dev->dev_private;
 
         if (ASIC_IS_DCE4(rdev))
-                return evergreen_crtc_do_set_base(crtc, fb, x, y, 1);
+                return dce4_crtc_do_set_base(crtc, fb, x, y, 1);
         else if (ASIC_IS_AVIVO(rdev))
                 return avivo_crtc_do_set_base(crtc, fb, x, y, 1);
         else
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 4e7778d44b8d..695de9a38506 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -187,9 +187,9 @@ static int dp_link_clock_for_mode_clock(u8 dpcd[DP_DPCD_SIZE], int mode_clock)
 int dp_mode_valid(u8 dpcd[DP_DPCD_SIZE], int mode_clock)
 {
         int lanes = dp_lanes_for_mode_clock(dpcd, mode_clock);
-        int bw = dp_lanes_for_mode_clock(dpcd, mode_clock);
+        int dp_clock = dp_link_clock_for_mode_clock(dpcd, mode_clock);
 
-        if ((lanes == 0) || (bw == 0))
+        if ((lanes == 0) || (dp_clock == 0))
                 return MODE_CLOCK_HIGH;
 
         return MODE_OK;
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 677af91b555c..d270b3ff896b 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -97,26 +97,29 @@ u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 }
 
 /* get temperature in millidegrees */
-u32 evergreen_get_temp(struct radeon_device *rdev)
+int evergreen_get_temp(struct radeon_device *rdev)
 {
         u32 temp = (RREG32(CG_MULT_THERMAL_STATUS) & ASIC_T_MASK) >>
                 ASIC_T_SHIFT;
         u32 actual_temp = 0;
 
-        if ((temp >> 10) & 1)
-                actual_temp = 0;
-        else if ((temp >> 9) & 1)
+        if (temp & 0x400)
+                actual_temp = -256;
+        else if (temp & 0x200)
                 actual_temp = 255;
-        else
-                actual_temp = (temp >> 1) & 0xff;
+        else if (temp & 0x100) {
+                actual_temp = temp & 0x1ff;
+                actual_temp |= ~0x1ff;
+        } else
+                actual_temp = temp & 0xff;
 
-        return actual_temp * 1000;
+        return (actual_temp * 1000) / 2;
 }
 
-u32 sumo_get_temp(struct radeon_device *rdev)
+int sumo_get_temp(struct radeon_device *rdev)
 {
         u32 temp = RREG32(CG_THERMAL_STATUS) & 0xff;
-        u32 actual_temp = (temp >> 1) & 0xff;
+        int actual_temp = temp - 49;
 
         return actual_temp * 1000;
 }
@@ -1182,6 +1185,22 @@ static void evergreen_mc_program(struct radeon_device *rdev)
 /*
  * CP.
  */
+void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
+{
+        /* set to DX10/11 mode */
+        radeon_ring_write(rdev, PACKET3(PACKET3_MODE_CONTROL, 0));
+        radeon_ring_write(rdev, 1);
+        /* FIXME: implement */
+        radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+        radeon_ring_write(rdev,
+#ifdef __BIG_ENDIAN
+                          (2 << 0) |
+#endif
+                          (ib->gpu_addr & 0xFFFFFFFC));
+        radeon_ring_write(rdev, upper_32_bits(ib->gpu_addr) & 0xFF);
+        radeon_ring_write(rdev, ib->length_dw);
+}
+
 
 static int evergreen_cp_load_microcode(struct radeon_device *rdev)
 {
@@ -1192,7 +1211,11 @@ static int evergreen_cp_load_microcode(struct radeon_device *rdev)
                 return -EINVAL;
 
         r700_cp_stop(rdev);
-        WREG32(CP_RB_CNTL, RB_NO_UPDATE | (15 << 8) | (3 << 0));
+        WREG32(CP_RB_CNTL,
+#ifdef __BIG_ENDIAN
+               BUF_SWAP_32BIT |
+#endif
+               RB_NO_UPDATE | RB_BLKSZ(15) | RB_BUFSZ(3));
 
         fw_data = (const __be32 *)rdev->pfp_fw->data;
         WREG32(CP_PFP_UCODE_ADDR, 0);
@@ -1233,7 +1256,7 @@ static int evergreen_cp_start(struct radeon_device *rdev)
         cp_me = 0xff;
         WREG32(CP_ME_CNTL, cp_me);
 
-        r = radeon_ring_lock(rdev, evergreen_default_size + 15);
+        r = radeon_ring_lock(rdev, evergreen_default_size + 19);
         if (r) {
                 DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
                 return r;
@@ -1266,6 +1289,11 @@ static int evergreen_cp_start(struct radeon_device *rdev)
         radeon_ring_write(rdev, 0xffffffff);
         radeon_ring_write(rdev, 0xffffffff);
 
+        radeon_ring_write(rdev, 0xc0026900);
+        radeon_ring_write(rdev, 0x00000316);
+        radeon_ring_write(rdev, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
+        radeon_ring_write(rdev, 0x00000010); /*  */
+
         radeon_ring_unlock_commit(rdev);
 
         return 0;
@@ -1306,7 +1334,11 @@ int evergreen_cp_resume(struct radeon_device *rdev)
         WREG32(CP_RB_WPTR, 0);
 
         /* set the wb address wether it's enabled or not */
-        WREG32(CP_RB_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC);
+        WREG32(CP_RB_RPTR_ADDR,
+#ifdef __BIG_ENDIAN
+               RB_RPTR_SWAP(2) |
+#endif
+               ((rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC));
         WREG32(CP_RB_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFF);
         WREG32(SCRATCH_ADDR, ((rdev->wb.gpu_addr + RADEON_WB_SCRATCH_OFFSET) >> 8) & 0xFFFFFFFF);
 
@@ -2072,6 +2104,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
         WREG32(VGT_CACHE_INVALIDATION, vgt_cache_invalidation);
 
         WREG32(VGT_GS_VERTEX_REUSE, 16);
+        WREG32(PA_SU_LINE_STIPPLE_VALUE, 0);
         WREG32(PA_SC_LINE_STIPPLE_STATE, 0);
 
         WREG32(VGT_VERTEX_REUSE_BLOCK_CNTL, 14);
@@ -2606,8 +2639,8 @@ restart_ih:
         while (rptr != wptr) {
                 /* wptr/rptr are in bytes! */
                 ring_index = rptr / 4;
-                src_id =  rdev->ih.ring[ring_index] & 0xff;
-                src_data = rdev->ih.ring[ring_index + 1] & 0xfffffff;
+                src_id =  le32_to_cpu(rdev->ih.ring[ring_index]) & 0xff;
+                src_data = le32_to_cpu(rdev->ih.ring[ring_index + 1]) & 0xfffffff;
 
                 switch (src_id) {
                 case 1: /* D1 vblank/vline */
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
index b758dc7f2f2c..2adfb03f479b 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
@@ -55,7 +55,7 @@ set_render_target(struct radeon_device *rdev, int format,
         if (h < 8)
                 h = 8;
 
-        cb_color_info = ((format << 2) | (1 << 24));
+        cb_color_info = ((format << 2) | (1 << 24) | (1 << 8));
         pitch = (w / 8) - 1;
         slice = ((w * h) / 64) - 1;
 
@@ -133,6 +133,9 @@ set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
 
         /* high addr, stride */
         sq_vtx_constant_word2 = ((upper_32_bits(gpu_addr) & 0xff) | (16 << 8));
+#ifdef __BIG_ENDIAN
+        sq_vtx_constant_word2 |= (2 << 30);
+#endif
         /* xyzw swizzles */
         sq_vtx_constant_word3 = (0 << 3) | (1 << 6) | (2 << 9) | (3 << 12);
 
@@ -173,7 +176,7 @@ set_tex_resource(struct radeon_device *rdev,
         sq_tex_resource_word0 = (1 << 0); /* 2D */
         sq_tex_resource_word0 |= ((((pitch >> 3) - 1) << 6) |
                                   ((w - 1) << 18));
-        sq_tex_resource_word1 = ((h - 1) << 0);
+        sq_tex_resource_word1 = ((h - 1) << 0) | (1 << 28);
         /* xyzw swizzles */
         sq_tex_resource_word4 = (0 << 16) | (1 << 19) | (2 << 22) | (3 << 25);
 
@@ -221,7 +224,11 @@ draw_auto(struct radeon_device *rdev)
         radeon_ring_write(rdev, DI_PT_RECTLIST);
 
         radeon_ring_write(rdev, PACKET3(PACKET3_INDEX_TYPE, 0));
-        radeon_ring_write(rdev, DI_INDEX_SIZE_16_BIT);
+        radeon_ring_write(rdev,
+#ifdef __BIG_ENDIAN
+                          (2 << 2) |
+#endif
+                          DI_INDEX_SIZE_16_BIT);
 
         radeon_ring_write(rdev, PACKET3(PACKET3_NUM_INSTANCES, 0));
         radeon_ring_write(rdev, 1);
@@ -232,7 +239,7 @@ draw_auto(struct radeon_device *rdev)
 
 }
 
-/* emits 30 */
+/* emits 36 */
 static void
 set_default_state(struct radeon_device *rdev)
 {
@@ -245,6 +252,8 @@ set_default_state(struct radeon_device *rdev)
         int num_hs_threads, num_ls_threads;
         int num_ps_stack_entries, num_vs_stack_entries, num_gs_stack_entries, num_es_stack_entries;
         int num_hs_stack_entries, num_ls_stack_entries;
+        u64 gpu_addr;
+        int dwords;
 
         switch (rdev->family) {
         case CHIP_CEDAR:
@@ -497,6 +506,18 @@ set_default_state(struct radeon_device *rdev)
         radeon_ring_write(rdev, 0x00000000);
         radeon_ring_write(rdev, 0x00000000);
 
+        /* set to DX10/11 mode */
+        radeon_ring_write(rdev, PACKET3(PACKET3_MODE_CONTROL, 0));
+        radeon_ring_write(rdev, 1);
+
+        /* emit an IB pointing at default state */
+        dwords = ALIGN(rdev->r600_blit.state_len, 0x10);
+        gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.state_offset;
+        radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+        radeon_ring_write(rdev, gpu_addr & 0xFFFFFFFC);
+        radeon_ring_write(rdev, upper_32_bits(gpu_addr) & 0xFF);
+        radeon_ring_write(rdev, dwords);
+
 }
 
 static inline uint32_t i2f(uint32_t input)
@@ -527,8 +548,10 @@ static inline uint32_t i2f(uint32_t input)
 int evergreen_blit_init(struct radeon_device *rdev)
 {
         u32 obj_size;
-        int r;
+        int i, r, dwords;
         void *ptr;
+        u32 packet2s[16];
+        int num_packet2s = 0;
 
         /* pin copy shader into vram if already initialized */
         if (rdev->r600_blit.shader_obj)
@@ -536,8 +559,17 @@ int evergreen_blit_init(struct radeon_device *rdev)
 
         mutex_init(&rdev->r600_blit.mutex);
         rdev->r600_blit.state_offset = 0;
-        rdev->r600_blit.state_len = 0;
-        obj_size = 0;
+
+        rdev->r600_blit.state_len = evergreen_default_size;
+
+        dwords = rdev->r600_blit.state_len;
+        while (dwords & 0xf) {
+                packet2s[num_packet2s++] = cpu_to_le32(PACKET2(0));
+                dwords++;
+        }
+
+        obj_size = dwords * 4;
+        obj_size = ALIGN(obj_size, 256);
 
         rdev->r600_blit.vs_offset = obj_size;
         obj_size += evergreen_vs_size * 4;
@@ -567,8 +599,16 @@ int evergreen_blit_init(struct radeon_device *rdev)
                 return r;
         }
 
-        memcpy(ptr + rdev->r600_blit.vs_offset, evergreen_vs, evergreen_vs_size * 4);
-        memcpy(ptr + rdev->r600_blit.ps_offset, evergreen_ps, evergreen_ps_size * 4);
+        memcpy_toio(ptr + rdev->r600_blit.state_offset,
+                    evergreen_default_state, rdev->r600_blit.state_len * 4);
+
+        if (num_packet2s)
+                memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
+                            packet2s, num_packet2s * 4);
+        for (i = 0; i < evergreen_vs_size; i++)
+                *(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(evergreen_vs[i]);
+        for (i = 0; i < evergreen_ps_size; i++)
+                *(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(evergreen_ps[i]);
         radeon_bo_kunmap(rdev->r600_blit.shader_obj);
         radeon_bo_unreserve(rdev->r600_blit.shader_obj);
 
@@ -652,7 +692,7 @@ int evergreen_blit_prepare_copy(struct radeon_device *rdev, int size_bytes)
         /* calculate number of loops correctly */
         ring_size = num_loops * dwords_per_loop;
         /* set default  + shaders */
-        ring_size += 46; /* shaders + def state */
+        ring_size += 52; /* shaders + def state */
         ring_size += 10; /* fence emit for VB IB */
         ring_size += 5; /* done copy */
         ring_size += 10; /* fence emit for done copy */
@@ -660,7 +700,7 @@ int evergreen_blit_prepare_copy(struct radeon_device *rdev, int size_bytes)
         if (r)
                 return r;
 
-        set_default_state(rdev); /* 30 */
+        set_default_state(rdev); /* 36 */
         set_shaders(rdev); /* 16 */
         return 0;
 }
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_shaders.c b/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
index ef1d28c07fbf..3a10399e0066 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
@@ -311,11 +311,19 @@ const u32 evergreen_vs[] =
         0x00000000,
         0x3c000000,
         0x67961001,
+#ifdef __BIG_ENDIAN
+        0x000a0000,
+#else
         0x00080000,
+#endif
         0x00000000,
         0x1c000000,
         0x67961000,
+#ifdef __BIG_ENDIAN
+        0x00020008,
+#else
         0x00000008,
+#endif
         0x00000000,
 };
 
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index 36d32d83d866..eb4acf4528ff 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -98,6 +98,7 @@
 #define         BUF_SWAP_32BIT                                  (2 << 16)
 #define CP_RB_RPTR                                      0x8700
 #define CP_RB_RPTR_ADDR                                 0xC10C
+#define         RB_RPTR_SWAP(x)                                 ((x) << 0)
 #define CP_RB_RPTR_ADDR_HI                              0xC110
 #define CP_RB_RPTR_WR                                   0xC108
 #define CP_RB_WPTR                                      0xC114
@@ -240,6 +241,7 @@
 #define         FORCE_EOV_MAX_CLK_CNT(x)                        ((x) << 0)
 #define         FORCE_EOV_MAX_REZ_CNT(x)                        ((x) << 16)
 #define PA_SC_LINE_STIPPLE                              0x28A0C
+#define PA_SU_LINE_STIPPLE_VALUE                        0x8A60
 #define PA_SC_LINE_STIPPLE_STATE                        0x8B10
 
 #define SCRATCH_REG0                                    0x8500
@@ -652,6 +654,7 @@
 #define PACKET3_DISPATCH_DIRECT                         0x15
 #define PACKET3_DISPATCH_INDIRECT                       0x16
 #define PACKET3_INDIRECT_BUFFER_END                     0x17
+#define PACKET3_MODE_CONTROL                            0x18
 #define PACKET3_SET_PREDICATION                         0x20
 #define PACKET3_REG_RMW                                 0x21
 #define PACKET3_COND_EXEC                               0x22
diff --git a/drivers/gpu/drm/radeon/mkregtable.c b/drivers/gpu/drm/radeon/mkregtable.c
index 607241c6a8a9..5a82b6b75849 100644
--- a/drivers/gpu/drm/radeon/mkregtable.c
+++ b/drivers/gpu/drm/radeon/mkregtable.c
@@ -673,8 +673,10 @@ static int parser_auth(struct table *t, const char *filename)
         last_reg = strtol(last_reg_s, NULL, 16);
 
         do {
-                if (fgets(buf, 1024, file) == NULL)
+                if (fgets(buf, 1024, file) == NULL) {
+                        fclose(file);
                         return -1;
+                }
                 len = strlen(buf);
                 if (ftell(file) == end)
                         done = 1;
@@ -685,6 +687,7 @@ static int parser_auth(struct table *t, const char *filename)
                                 fprintf(stderr,
                                         "Error matching regular expression %d in %s\n",
                                         r, filename);
+                                fclose(file);
                                 return -1;
                         } else {
                                 buf[match[0].rm_eo] = 0;
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 5968dde243e9..56deae5bf02e 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -1031,8 +1031,8 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size)
         WREG32(RADEON_CP_CSQ_MODE,
                REG_SET(RADEON_INDIRECT2_START, indirect2_start) |
                REG_SET(RADEON_INDIRECT1_START, indirect1_start));
-        WREG32(0x718, 0);
-        WREG32(0x744, 0x00004D4D);
+        WREG32(RADEON_CP_RB_WPTR_DELAY, 0);
+        WREG32(RADEON_CP_CSQ_MODE, 0x00004D4D);
         WREG32(RADEON_CP_CSQ_CNTL, RADEON_CSQ_PRIBM_INDBM);
         radeon_ring_start(rdev);
         r = radeon_ring_test(rdev);
@@ -1427,6 +1427,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->zb.robj = reloc->robj;
                 track->zb.offset = idx_value;
+                track->zb_dirty = true;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 break;
         case RADEON_RB3D_COLOROFFSET:
@@ -1439,6 +1440,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->cb[0].robj = reloc->robj;
                 track->cb[0].offset = idx_value;
+                track->cb_dirty = true;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 break;
         case RADEON_PP_TXOFFSET_0:
@@ -1454,6 +1456,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 }
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 track->textures[i].robj = reloc->robj;
+                track->tex_dirty = true;
                 break;
         case RADEON_PP_CUBIC_OFFSET_T0_0:
         case RADEON_PP_CUBIC_OFFSET_T0_1:
@@ -1471,6 +1474,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 track->textures[0].cube_info[i].offset = idx_value;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 track->textures[0].cube_info[i].robj = reloc->robj;
+                track->tex_dirty = true;
                 break;
         case RADEON_PP_CUBIC_OFFSET_T1_0:
         case RADEON_PP_CUBIC_OFFSET_T1_1:
@@ -1488,6 +1492,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 track->textures[1].cube_info[i].offset = idx_value;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 track->textures[1].cube_info[i].robj = reloc->robj;
+                track->tex_dirty = true;
                 break;
         case RADEON_PP_CUBIC_OFFSET_T2_0:
         case RADEON_PP_CUBIC_OFFSET_T2_1:
@@ -1505,9 +1510,12 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 track->textures[2].cube_info[i].offset = idx_value;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 track->textures[2].cube_info[i].robj = reloc->robj;
+                track->tex_dirty = true;
                 break;
         case RADEON_RE_WIDTH_HEIGHT:
                 track->maxy = ((idx_value >> 16) & 0x7FF);
+                track->cb_dirty = true;
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_COLORPITCH:
                 r = r100_cs_packet_next_reloc(p, &reloc);
@@ -1528,9 +1536,11 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 ib[idx] = tmp;
 
                 track->cb[0].pitch = idx_value & RADEON_COLORPITCH_MASK;
+                track->cb_dirty = true;
                 break;
         case RADEON_RB3D_DEPTHPITCH:
                 track->zb.pitch = idx_value & RADEON_DEPTHPITCH_MASK;
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_CNTL:
                 switch ((idx_value >> RADEON_RB3D_COLOR_FORMAT_SHIFT) & 0x1f) {
@@ -1555,6 +1565,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                         return -EINVAL;
                 }
                 track->z_enabled = !!(idx_value & RADEON_Z_ENABLE);
+                track->cb_dirty = true;
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_ZSTENCILCNTL:
                 switch (idx_value & 0xf) {
@@ -1572,6 +1584,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 default:
                         break;
                 }
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_ZPASS_ADDR:
                 r = r100_cs_packet_next_reloc(p, &reloc);
@@ -1588,6 +1601,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                         uint32_t temp = idx_value >> 4;
                         for (i = 0; i < track->num_texture; i++)
                                 track->textures[i].enabled = !!(temp & (1 << i));
+                        track->tex_dirty = true;
                 }
                 break;
         case RADEON_SE_VF_CNTL:
@@ -1602,12 +1616,14 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 i = (reg - RADEON_PP_TEX_SIZE_0) / 8;
                 track->textures[i].width = (idx_value & RADEON_TEX_USIZE_MASK) + 1;
                 track->textures[i].height = ((idx_value & RADEON_TEX_VSIZE_MASK) >> RADEON_TEX_VSIZE_SHIFT) + 1;
+                track->tex_dirty = true;
                 break;
         case RADEON_PP_TEX_PITCH_0:
         case RADEON_PP_TEX_PITCH_1:
         case RADEON_PP_TEX_PITCH_2:
                 i = (reg - RADEON_PP_TEX_PITCH_0) / 8;
                 track->textures[i].pitch = idx_value + 32;
+                track->tex_dirty = true;
                 break;
         case RADEON_PP_TXFILTER_0:
         case RADEON_PP_TXFILTER_1:
@@ -1621,6 +1637,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 tmp = (idx_value >> 27) & 0x7;
                 if (tmp == 2 || tmp == 6)
                         track->textures[i].roundup_h = false;
+                track->tex_dirty = true;
                 break;
         case RADEON_PP_TXFORMAT_0:
         case RADEON_PP_TXFORMAT_1:
@@ -1673,6 +1690,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->textures[i].cube_info[4].width = 1 << ((idx_value >> 16) & 0xf);
                 track->textures[i].cube_info[4].height = 1 << ((idx_value >> 20) & 0xf);
+                track->tex_dirty = true;
                 break;
         case RADEON_PP_CUBIC_FACES_0:
         case RADEON_PP_CUBIC_FACES_1:
@@ -1683,6 +1701,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
                         track->textures[i].cube_info[face].width = 1 << ((tmp >> (face * 8)) & 0xf);
                         track->textures[i].cube_info[face].height = 1 << ((tmp >> ((face * 8) + 4)) & 0xf);
                 }
+                track->tex_dirty = true;
                 break;
         default:
                 printk(KERN_ERR "Forbidden register 0x%04X in cs at %d\n",
@@ -2347,10 +2366,10 @@ void r100_vga_set_state(struct radeon_device *rdev, bool state)
 
         temp = RREG32(RADEON_CONFIG_CNTL);
         if (state == false) {
-                temp &= ~(1<<8);
-                temp |= (1<<9);
+                temp &= ~RADEON_CFG_VGA_RAM_EN;
+                temp |= RADEON_CFG_VGA_IO_DIS;
         } else {
-                temp &= ~(1<<9);
+                temp &= ~RADEON_CFG_VGA_IO_DIS;
         }
         WREG32(RADEON_CONFIG_CNTL, temp);
 }
@@ -3318,9 +3337,9 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
         unsigned long size;
         unsigned prim_walk;
         unsigned nverts;
-        unsigned num_cb = track->num_cb;
+        unsigned num_cb = track->cb_dirty ? track->num_cb : 0;
 
-        if (!track->zb_cb_clear && !track->color_channel_mask &&
+        if (num_cb && !track->zb_cb_clear && !track->color_channel_mask &&
             !track->blend_read_enable)
                 num_cb = 0;
 
@@ -3341,7 +3360,9 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
                         return -EINVAL;
                 }
         }
-        if (track->z_enabled) {
+        track->cb_dirty = false;
+
+        if (track->zb_dirty && track->z_enabled) {
                 if (track->zb.robj == NULL) {
                         DRM_ERROR("[drm] No buffer for z buffer !\n");
                         return -EINVAL;
@@ -3358,6 +3379,28 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
                         return -EINVAL;
                 }
         }
+        track->zb_dirty = false;
+
+        if (track->aa_dirty && track->aaresolve) {
+                if (track->aa.robj == NULL) {
+                        DRM_ERROR("[drm] No buffer for AA resolve buffer %d !\n", i);
+                        return -EINVAL;
+                }
+                /* I believe the format comes from colorbuffer0. */
+                size = track->aa.pitch * track->cb[0].cpp * track->maxy;
+                size += track->aa.offset;
+                if (size > radeon_bo_size(track->aa.robj)) {
+                        DRM_ERROR("[drm] Buffer too small for AA resolve buffer %d "
+                                  "(need %lu have %lu) !\n", i, size,
+                                  radeon_bo_size(track->aa.robj));
+                        DRM_ERROR("[drm] AA resolve buffer %d (%u %u %u %u)\n",
+                                  i, track->aa.pitch, track->cb[0].cpp,
+                                  track->aa.offset, track->maxy);
+                        return -EINVAL;
+                }
+        }
+        track->aa_dirty = false;
+
         prim_walk = (track->vap_vf_cntl >> 4) & 0x3;
         if (track->vap_vf_cntl & (1 << 14)) {
                 nverts = track->vap_alt_nverts;
@@ -3417,13 +3460,23 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
                           prim_walk);
                 return -EINVAL;
         }
-        return r100_cs_track_texture_check(rdev, track);
+
+        if (track->tex_dirty) {
+                track->tex_dirty = false;
+                return r100_cs_track_texture_check(rdev, track);
+        }
+        return 0;
 }
 
 void r100_cs_track_clear(struct radeon_device *rdev, struct r100_cs_track *track)
 {
         unsigned i, face;
 
+        track->cb_dirty = true;
+        track->zb_dirty = true;
+        track->tex_dirty = true;
+        track->aa_dirty = true;
+
         if (rdev->family < CHIP_R300) {
                 track->num_cb = 1;
                 if (rdev->family <= CHIP_RS200)
@@ -3437,6 +3490,8 @@ void r100_cs_track_clear(struct radeon_device *rdev, struct r100_cs_track *track
                 track->num_texture = 16;
                 track->maxy = 4096;
                 track->separate_cube = 0;
+                track->aaresolve = true;
+                track->aa.robj = NULL;
         }
 
         for (i = 0; i < track->num_cb; i++) {
diff --git a/drivers/gpu/drm/radeon/r100_track.h b/drivers/gpu/drm/radeon/r100_track.h
index af65600e6564..2fef9de7f363 100644
--- a/drivers/gpu/drm/radeon/r100_track.h
+++ b/drivers/gpu/drm/radeon/r100_track.h
@@ -52,14 +52,7 @@ struct r100_cs_track_texture {
         unsigned                compress_format;
 };
 
-struct r100_cs_track_limits {
-        unsigned num_cb;
-        unsigned num_texture;
-        unsigned max_levels;
-};
-
 struct r100_cs_track {
-        struct radeon_device *rdev;
         unsigned                        num_cb;
         unsigned                        num_texture;
         unsigned                        maxy;
@@ -73,11 +66,17 @@ struct r100_cs_track {
         struct r100_cs_track_array      arrays[11];
         struct r100_cs_track_cb         cb[R300_MAX_CB];
         struct r100_cs_track_cb         zb;
+        struct r100_cs_track_cb         aa;
         struct r100_cs_track_texture    textures[R300_TRACK_MAX_TEXTURE];
         bool                            z_enabled;
         bool                            separate_cube;
         bool                            zb_cb_clear;
         bool                            blend_read_enable;
+        bool                            cb_dirty;
+        bool                            zb_dirty;
+        bool                            tex_dirty;
+        bool                            aa_dirty;
+        bool                            aaresolve;
 };
 
 int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track);
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index d2408c395619..f24058300413 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -184,6 +184,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->zb.robj = reloc->robj;
                 track->zb.offset = idx_value;
+                track->zb_dirty = true;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 break;
         case RADEON_RB3D_COLOROFFSET:
@@ -196,6 +197,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->cb[0].robj = reloc->robj;
                 track->cb[0].offset = idx_value;
+                track->cb_dirty = true;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 break;
         case R200_PP_TXOFFSET_0:
@@ -214,6 +216,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 }
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 track->textures[i].robj = reloc->robj;
+                track->tex_dirty = true;
                 break;
         case R200_PP_CUBIC_OFFSET_F1_0:
         case R200_PP_CUBIC_OFFSET_F2_0:
@@ -257,9 +260,12 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 track->textures[i].cube_info[face - 1].offset = idx_value;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 track->textures[i].cube_info[face - 1].robj = reloc->robj;
+                track->tex_dirty = true;
                 break;
         case RADEON_RE_WIDTH_HEIGHT:
                 track->maxy = ((idx_value >> 16) & 0x7FF);
+                track->cb_dirty = true;
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_COLORPITCH:
                 r = r100_cs_packet_next_reloc(p, &reloc);
@@ -280,9 +286,11 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 ib[idx] = tmp;
 
                 track->cb[0].pitch = idx_value & RADEON_COLORPITCH_MASK;
+                track->cb_dirty = true;
                 break;
         case RADEON_RB3D_DEPTHPITCH:
                 track->zb.pitch = idx_value & RADEON_DEPTHPITCH_MASK;
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_CNTL:
                 switch ((idx_value >> RADEON_RB3D_COLOR_FORMAT_SHIFT) & 0x1f) {
@@ -312,6 +320,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 }
 
                 track->z_enabled = !!(idx_value & RADEON_Z_ENABLE);
+                track->cb_dirty = true;
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_ZSTENCILCNTL:
                 switch (idx_value & 0xf) {
@@ -329,6 +339,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 default:
                         break;
                 }
+                track->zb_dirty = true;
                 break;
         case RADEON_RB3D_ZPASS_ADDR:
                 r = r100_cs_packet_next_reloc(p, &reloc);
@@ -345,6 +356,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                         uint32_t temp = idx_value >> 4;
                         for (i = 0; i < track->num_texture; i++)
                                 track->textures[i].enabled = !!(temp & (1 << i));
+                        track->tex_dirty = true;
                 }
                 break;
         case RADEON_SE_VF_CNTL:
@@ -369,6 +381,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 i = (reg - R200_PP_TXSIZE_0) / 32;
                 track->textures[i].width = (idx_value & RADEON_TEX_USIZE_MASK) + 1;
                 track->textures[i].height = ((idx_value & RADEON_TEX_VSIZE_MASK) >> RADEON_TEX_VSIZE_SHIFT) + 1;
+                track->tex_dirty = true;
                 break;
         case R200_PP_TXPITCH_0:
         case R200_PP_TXPITCH_1:
@@ -378,6 +391,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
         case R200_PP_TXPITCH_5:
                 i = (reg - R200_PP_TXPITCH_0) / 32;
                 track->textures[i].pitch = idx_value + 32;
+                track->tex_dirty = true;
                 break;
         case R200_PP_TXFILTER_0:
         case R200_PP_TXFILTER_1:
@@ -394,6 +408,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 tmp = (idx_value >> 27) & 0x7;
                 if (tmp == 2 || tmp == 6)
                         track->textures[i].roundup_h = false;
+                track->tex_dirty = true;
                 break;
         case R200_PP_TXMULTI_CTL_0:
         case R200_PP_TXMULTI_CTL_1:
@@ -432,6 +447,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                         track->textures[i].tex_coord_type = 1;
                         break;
                 }
+                track->tex_dirty = true;
                 break;
         case R200_PP_TXFORMAT_0:
         case R200_PP_TXFORMAT_1:
@@ -488,6 +504,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->textures[i].cube_info[4].width = 1 << ((idx_value >> 16) & 0xf);
                 track->textures[i].cube_info[4].height = 1 << ((idx_value >> 20) & 0xf);
+                track->tex_dirty = true;
                 break;
         case R200_PP_CUBIC_FACES_0:
         case R200_PP_CUBIC_FACES_1:
@@ -501,6 +518,7 @@ int r200_packet0_check(struct radeon_cs_parser *p,
                         track->textures[i].cube_info[face].width = 1 << ((tmp >> (face * 8)) & 0xf);
                         track->textures[i].cube_info[face].height = 1 << ((tmp >> ((face * 8) + 4)) & 0xf);
                 }
+                track->tex_dirty = true;
                 break;
         default:
                 printk(KERN_ERR "Forbidden register 0x%04X in cs at %d\n",
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index cf862ca580bf..069efa8c8ecf 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -69,6 +69,9 @@ void rv370_pcie_gart_tlb_flush(struct radeon_device *rdev)
         mb();
 }
 
+#define R300_PTE_WRITEABLE (1 << 2)
+#define R300_PTE_READABLE  (1 << 3)
+
 int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
 {
         void __iomem *ptr = (void *)rdev->gart.table.vram.ptr;
@@ -78,7 +81,7 @@ int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
         }
         addr = (lower_32_bits(addr) >> 8) |
                ((upper_32_bits(addr) & 0xff) << 24) |
-               0xc;
+               R300_PTE_WRITEABLE | R300_PTE_READABLE;
         /* on x86 we want this to be CPU endian, on powerpc
          * on powerpc without HW swappers, it'll get swapped on way
          * into VRAM - so no need for cpu_to_le32 on VRAM tables */
@@ -135,7 +138,7 @@ int rv370_pcie_gart_enable(struct radeon_device *rdev)
         WREG32_PCIE(RADEON_PCIE_TX_DISCARD_RD_ADDR_LO, rdev->mc.vram_start);
         WREG32_PCIE(RADEON_PCIE_TX_DISCARD_RD_ADDR_HI, 0);
         /* Clear error */
-        WREG32_PCIE(0x18, 0);
+        WREG32_PCIE(RADEON_PCIE_TX_GART_ERROR, 0);
         tmp = RREG32_PCIE(RADEON_PCIE_TX_GART_CNTL);
         tmp |= RADEON_PCIE_TX_GART_EN;
         tmp |= RADEON_PCIE_TX_GART_UNMAPPED_ACCESS_DISCARD;
@@ -664,6 +667,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->cb[i].robj = reloc->robj;
                 track->cb[i].offset = idx_value;
+                track->cb_dirty = true;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 break;
         case R300_ZB_DEPTHOFFSET:
@@ -676,6 +680,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 }
                 track->zb.robj = reloc->robj;
                 track->zb.offset = idx_value;
+                track->zb_dirty = true;
                 ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
                 break;
         case R300_TX_OFFSET_0:
@@ -714,6 +719,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 tmp |= tile_flags;
                 ib[idx] = tmp;
                 track->textures[i].robj = reloc->robj;
+                track->tex_dirty = true;
                 break;
         /* Tracked registers */
         case 0x2084:
@@ -740,6 +746,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 if (p->rdev->family < CHIP_RV515) {
                         track->maxy -= 1440;
                 }
+                track->cb_dirty = true;
+                track->zb_dirty = true;
                 break;
         case 0x4E00:
                 /* RB3D_CCTL */
@@ -749,6 +757,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                         return -EINVAL;
                 }
                 track->num_cb = ((idx_value >> 5) & 0x3) + 1;
+                track->cb_dirty = true;
                 break;
         case 0x4E38:
         case 0x4E3C:
@@ -811,6 +820,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                                   ((idx_value >> 21) & 0xF));
                         return -EINVAL;
                 }
+                track->cb_dirty = true;
                 break;
         case 0x4F00:
                 /* ZB_CNTL */
@@ -819,6 +829,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 } else {
                         track->z_enabled = false;
                 }
+                track->zb_dirty = true;
                 break;
         case 0x4F10:
                 /* ZB_FORMAT */
@@ -835,6 +846,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                                   (idx_value & 0xF));
                         return -EINVAL;
                 }
+                track->zb_dirty = true;
                 break;
         case 0x4F24:
                 /* ZB_DEPTHPITCH */
@@ -858,14 +870,17 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 ib[idx] = tmp;
 
                 track->zb.pitch = idx_value & 0x3FFC;
+                track->zb_dirty = true;
                 break;
         case 0x4104:
+                /* TX_ENABLE */
                 for (i = 0; i < 16; i++) {
                         bool enabled;
 
                         enabled = !!(idx_value & (1 << i));
                         track->textures[i].enabled = enabled;
                 }
+                track->tex_dirty = true;
                 break;
         case 0x44C0:
         case 0x44C4:
@@ -895,6 +910,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                         track->textures[i].compress_format = R100_TRACK_COMP_NONE;
                         break;
                 case R300_TX_FORMAT_X16:
+                case R300_TX_FORMAT_FL_I16:
                 case R300_TX_FORMAT_Y8X8:
                 case R300_TX_FORMAT_Z5Y6X5:
                 case R300_TX_FORMAT_Z6Y5X5:
@@ -907,6 +923,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                         track->textures[i].compress_format = R100_TRACK_COMP_NONE;
                         break;
                 case R300_TX_FORMAT_Y16X16:
+                case R300_TX_FORMAT_FL_I16A16:
                 case R300_TX_FORMAT_Z11Y11X10:
                 case R300_TX_FORMAT_Z10Y11X11:
                 case R300_TX_FORMAT_W8Z8Y8X8:
@@ -948,8 +965,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                         DRM_ERROR("Invalid texture format %u\n",
                                   (idx_value & 0x1F));
                         return -EINVAL;
-                        break;
                 }
+                track->tex_dirty = true;
                 break;
         case 0x4400:
         case 0x4404:
@@ -977,6 +994,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 if (tmp == 2 || tmp == 4 || tmp == 6) {
                         track->textures[i].roundup_h = false;
                 }
+                track->tex_dirty = true;
                 break;
         case 0x4500:
         case 0x4504:
@@ -1014,6 +1032,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                         DRM_ERROR("Forbidden bit TXFORMAT_MSB\n");
                         return -EINVAL;
                 }
+                track->tex_dirty = true;
                 break;
         case 0x4480:
         case 0x4484:
@@ -1043,6 +1062,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
                 track->textures[i].use_pitch = !!tmp;
                 tmp = (idx_value >> 22) & 0xF;
                 track->textures[i].txdepth = tmp;
+                track->tex_dirty = true;
                 break;
         case R300_ZB_ZPASS_ADDR:
                 r = r100_cs_packet_next_reloc(p, &reloc);
@@ -1057,6 +1077,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
         case 0x4e0c:
                 /* RB3D_COLOR_CHANNEL_MASK */
                 track->color_channel_mask = idx_value;
+                track->cb_dirty = true;
                 break;
         case 0x43a4:
                 /* SC_HYPERZ_EN */
@@ -1070,6 +1091,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
         case 0x4f1c:
                 /* ZB_BW_CNTL */
                 track->zb_cb_clear = !!(idx_value & (1 << 5));
+                track->cb_dirty = true;
+                track->zb_dirty = true;
                 if (p->rdev->hyperz_filp != p->filp) {
                         if (idx_value & (R300_HIZ_ENABLE |
                                          R300_RD_COMP_ENABLE |
@@ -1081,8 +1104,28 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
         case 0x4e04:
                 /* RB3D_BLENDCNTL */
                 track->blend_read_enable = !!(idx_value & (1 << 2));
+                track->cb_dirty = true;
+                break;
+        case R300_RB3D_AARESOLVE_OFFSET:
+                r = r100_cs_packet_next_reloc(p, &reloc);
+                if (r) {
+                        DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
+                                  idx, reg);
+                        r100_cs_dump_packet(p, pkt);
+                        return r;
+                }
+                track->aa.robj = reloc->robj;
+                track->aa.offset = idx_value;
+                track->aa_dirty = true;
+                ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset);
+                break;
+        case R300_RB3D_AARESOLVE_PITCH:
+                track->aa.pitch = idx_value & 0x3FFE;
+                track->aa_dirty = true;
                 break;
-        case 0x4f28: /* ZB_DEPTHCLEARVALUE */
+        case R300_RB3D_AARESOLVE_CTL:
+                track->aaresolve = idx_value & 0x1;
+                track->aa_dirty = true;
                 break;
         case 0x4f30: /* ZB_MASK_OFFSET */
         case 0x4f34: /* ZB_ZMASK_PITCH */
diff --git a/drivers/gpu/drm/radeon/r300_reg.h b/drivers/gpu/drm/radeon/r300_reg.h
index 1a0d5362cd79..f0bce399c9f3 100644
--- a/drivers/gpu/drm/radeon/r300_reg.h
+++ b/drivers/gpu/drm/radeon/r300_reg.h
@@ -1371,6 +1371,8 @@
 #define R300_RB3D_COLORPITCH2               0x4E40 /* GUESS */
 #define R300_RB3D_COLORPITCH3               0x4E44 /* GUESS */
 
+#define R300_RB3D_AARESOLVE_OFFSET          0x4E80
+#define R300_RB3D_AARESOLVE_PITCH           0x4E84
 #define R300_RB3D_AARESOLVE_CTL             0x4E88
 /* gap */
 
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index c387346f93a9..0b59ed7c7d2c 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -96,7 +96,7 @@ void r420_pipes_init(struct radeon_device *rdev)
                        "programming pipes. Bad things might happen.\n");
         }
         /* get max number of pipes */
-        gb_pipe_select = RREG32(0x402C);
+        gb_pipe_select = RREG32(R400_GB_PIPE_SELECT);
         num_pipes = ((gb_pipe_select >> 12) & 3) + 1;
 
         /* SE chips have 1 pipe */
diff --git a/drivers/gpu/drm/radeon/r520.c b/drivers/gpu/drm/radeon/r520.c
index 3c8677f9e385..2ce80d976568 100644
--- a/drivers/gpu/drm/radeon/r520.c
+++ b/drivers/gpu/drm/radeon/r520.c
@@ -79,8 +79,8 @@ static void r520_gpu_init(struct radeon_device *rdev)
                 WREG32(0x4128, 0xFF);
         }
         r420_pipes_init(rdev);
-        gb_pipe_select = RREG32(0x402C);
-        tmp = RREG32(0x170C);
+        gb_pipe_select = RREG32(R400_GB_PIPE_SELECT);
+        tmp = RREG32(R300_DST_PIPE_CONFIG);
         pipe_select_current = (tmp >> 2) & 3;
         tmp = (1 << pipe_select_current) |
               (((gb_pipe_select >> 8) & 0xF) << 4);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 1e10e3e2ba2a..de88624d5f87 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -97,12 +97,16 @@ void r600_irq_disable(struct radeon_device *rdev);
 static void r600_pcie_gen2_enable(struct radeon_device *rdev);
 
 /* get temperature in millidegrees */
-u32 rv6xx_get_temp(struct radeon_device *rdev)
+int rv6xx_get_temp(struct radeon_device *rdev)
 {
         u32 temp = (RREG32(CG_THERMAL_STATUS) & ASIC_T_MASK) >>
                 ASIC_T_SHIFT;
+        int actual_temp = temp & 0xff;
 
-        return temp * 1000;
+        if (temp & 0x100)
+                actual_temp -= 256;
+
+        return actual_temp * 1000;
 }
 
 void r600_pm_get_dynpm_state(struct radeon_device *rdev)
@@ -2101,7 +2105,11 @@ static int r600_cp_load_microcode(struct radeon_device *rdev)
 
         r600_cp_stop(rdev);
 
-        WREG32(CP_RB_CNTL, RB_NO_UPDATE | RB_BLKSZ(15) | RB_BUFSZ(3));
+        WREG32(CP_RB_CNTL,
+#ifdef __BIG_ENDIAN
+               BUF_SWAP_32BIT |
+#endif
+               RB_NO_UPDATE | RB_BLKSZ(15) | RB_BUFSZ(3));
 
         /* Reset cp */
         WREG32(GRBM_SOFT_RESET, SOFT_RESET_CP);
@@ -2188,7 +2196,11 @@ int r600_cp_resume(struct radeon_device *rdev)
         WREG32(CP_RB_WPTR, 0);
 
         /* set the wb address whether it's enabled or not */
-        WREG32(CP_RB_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC);
+        WREG32(CP_RB_RPTR_ADDR,
+#ifdef __BIG_ENDIAN
+               RB_RPTR_SWAP(2) |
+#endif
+               ((rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC));
         WREG32(CP_RB_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFF);
         WREG32(SCRATCH_ADDR, ((rdev->wb.gpu_addr + RADEON_WB_SCRATCH_OFFSET) >> 8) & 0xFFFFFFFF);
 
@@ -2624,7 +2636,11 @@ void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
         /* FIXME: implement */
         radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
-        radeon_ring_write(rdev, ib->gpu_addr & 0xFFFFFFFC);
+        radeon_ring_write(rdev,
+#ifdef __BIG_ENDIAN
+                          (2 << 0) |
+#endif
+                          (ib->gpu_addr & 0xFFFFFFFC));
         radeon_ring_write(rdev, upper_32_bits(ib->gpu_addr) & 0xFF);
         radeon_ring_write(rdev, ib->length_dw);
 }
@@ -3293,8 +3309,8 @@ restart_ih:
         while (rptr != wptr) {
                 /* wptr/rptr are in bytes! */
                 ring_index = rptr / 4;
-                src_id =  rdev->ih.ring[ring_index] & 0xff;
-                src_data = rdev->ih.ring[ring_index + 1] & 0xfffffff;
+                src_id = le32_to_cpu(rdev->ih.ring[ring_index]) & 0xff;
+                src_data = le32_to_cpu(rdev->ih.ring[ring_index + 1]) & 0xfffffff;
 
                 switch (src_id) {
                 case 1: /* D1 vblank/vline */
diff --git a/drivers/gpu/drm/radeon/r600_blit.c b/drivers/gpu/drm/radeon/r600_blit.c
index ca5c29f70779..7f1043448d25 100644
--- a/drivers/gpu/drm/radeon/r600_blit.c
+++ b/drivers/gpu/drm/radeon/r600_blit.c
@@ -137,9 +137,9 @@ set_shaders(struct drm_device *dev)
         ps = (u32 *) ((char *)dev->agp_buffer_map->handle + dev_priv->blit_vb->offset + 256);
 
         for (i = 0; i < r6xx_vs_size; i++)
-                vs[i] = r6xx_vs[i];
+                vs[i] = cpu_to_le32(r6xx_vs[i]);
         for (i = 0; i < r6xx_ps_size; i++)
-                ps[i] = r6xx_ps[i];
+                ps[i] = cpu_to_le32(r6xx_ps[i]);
 
         dev_priv->blit_vb->used = 512;
 
@@ -192,6 +192,9 @@ set_vtx_resource(drm_radeon_private_t *dev_priv, u64 gpu_addr)
         DRM_DEBUG("\n");
 
         sq_vtx_constant_word2 = (((gpu_addr >> 32) & 0xff) | (16 << 8));
+#ifdef __BIG_ENDIAN
+        sq_vtx_constant_word2 |= (2 << 30);
+#endif
 
         BEGIN_RING(9);
         OUT_RING(CP_PACKET3(R600_IT_SET_RESOURCE, 7));
@@ -291,7 +294,11 @@ draw_auto(drm_radeon_private_t *dev_priv)
         OUT_RING(DI_PT_RECTLIST);
 
         OUT_RING(CP_PACKET3(R600_IT_INDEX_TYPE, 0));
+#ifdef __BIG_ENDIAN
+        OUT_RING((2 << 2) | DI_INDEX_SIZE_16_BIT);
+#else
         OUT_RING(DI_INDEX_SIZE_16_BIT);
+#endif
 
         OUT_RING(CP_PACKET3(R600_IT_NUM_INSTANCES, 0));
         OUT_RING(1);
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
index 86e5aa07f0db..41f7aafc97c4 100644
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ b/drivers/gpu/drm/radeon/r600_blit_kms.c
@@ -54,7 +54,7 @@ set_render_target(struct radeon_device *rdev, int format,
         if (h < 8)
                 h = 8;
 
-        cb_color_info = ((format << 2) | (1 << 27));
+        cb_color_info = ((format << 2) | (1 << 27) | (1 << 8));
         pitch = (w / 8) - 1;
         slice = ((w * h) / 64) - 1;
 
@@ -165,6 +165,9 @@ set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
         u32 sq_vtx_constant_word2;
 
         sq_vtx_constant_word2 = ((upper_32_bits(gpu_addr) & 0xff) | (16 << 8));
+#ifdef __BIG_ENDIAN
+        sq_vtx_constant_word2 |= (2 << 30);
+#endif
 
         radeon_ring_write(rdev, PACKET3(PACKET3_SET_RESOURCE, 7));
         radeon_ring_write(rdev, 0x460);
@@ -199,7 +202,7 @@ set_tex_resource(struct radeon_device *rdev,
         if (h < 1)
                 h = 1;
 
-        sq_tex_resource_word0 = (1 << 0);
+        sq_tex_resource_word0 = (1 << 0) | (1 << 3);
         sq_tex_resource_word0 |= ((((pitch >> 3) - 1) << 8) |
                                   ((w - 1) << 19));
 
@@ -253,7 +256,11 @@ draw_auto(struct radeon_device *rdev)
         radeon_ring_write(rdev, DI_PT_RECTLIST);
 
         radeon_ring_write(rdev, PACKET3(PACKET3_INDEX_TYPE, 0));
-        radeon_ring_write(rdev, DI_INDEX_SIZE_16_BIT);
+        radeon_ring_write(rdev,
+#ifdef __BIG_ENDIAN
+                          (2 << 2) |
+#endif
+                          DI_INDEX_SIZE_16_BIT);
 
         radeon_ring_write(rdev, PACKET3(PACKET3_NUM_INSTANCES, 0));
         radeon_ring_write(rdev, 1);
@@ -424,7 +431,11 @@ set_default_state(struct radeon_device *rdev)
         dwords = ALIGN(rdev->r600_blit.state_len, 0x10);
         gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.state_offset;
         radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
-        radeon_ring_write(rdev, gpu_addr & 0xFFFFFFFC);
+        radeon_ring_write(rdev,
+#ifdef __BIG_ENDIAN
+                          (2 << 0) |
+#endif
+                          (gpu_addr & 0xFFFFFFFC));
         radeon_ring_write(rdev, upper_32_bits(gpu_addr) & 0xFF);
         radeon_ring_write(rdev, dwords);
 
@@ -467,7 +478,7 @@ static inline uint32_t i2f(uint32_t input)
 int r600_blit_init(struct radeon_device *rdev)
 {
         u32 obj_size;
-        int r, dwords;
+        int i, r, dwords;
         void *ptr;
         u32 packet2s[16];
         int num_packet2s = 0;
@@ -486,7 +497,7 @@ int r600_blit_init(struct radeon_device *rdev)
 
         dwords = rdev->r600_blit.state_len;
         while (dwords & 0xf) {
-                packet2s[num_packet2s++] = PACKET2(0);
+                packet2s[num_packet2s++] = cpu_to_le32(PACKET2(0));
                 dwords++;
         }
 
@@ -529,8 +540,10 @@ int r600_blit_init(struct radeon_device *rdev)
         if (num_packet2s)
                 memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
                             packet2s, num_packet2s * 4);
-        memcpy(ptr + rdev->r600_blit.vs_offset, r6xx_vs, r6xx_vs_size * 4);
-        memcpy(ptr + rdev->r600_blit.ps_offset, r6xx_ps, r6xx_ps_size * 4);
+        for (i = 0; i < r6xx_vs_size; i++)
+                *(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(r6xx_vs[i]);
+        for (i = 0; i < r6xx_ps_size; i++)
+                *(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(r6xx_ps[i]);
         radeon_bo_kunmap(rdev->r600_blit.shader_obj);
         radeon_bo_unreserve(rdev->r600_blit.shader_obj);
 
diff --git a/drivers/gpu/drm/radeon/r600_blit_shaders.c b/drivers/gpu/drm/radeon/r600_blit_shaders.c
index e8151c1d55b2..2d1f6c5ee2a7 100644
--- a/drivers/gpu/drm/radeon/r600_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/r600_blit_shaders.c
@@ -684,7 +684,11 @@ const u32 r6xx_vs[] =
         0x00000000,
         0x3c000000,
         0x68cd1000,
+#ifdef __BIG_ENDIAN
+        0x000a0000,
+#else
         0x00080000,
+#endif
         0x00000000,
 };
 
diff --git a/drivers/gpu/drm/radeon/r600_cp.c b/drivers/gpu/drm/radeon/r600_cp.c
index 4f4cd8b286d5..c3ab959bdc7c 100644
--- a/drivers/gpu/drm/radeon/r600_cp.c
+++ b/drivers/gpu/drm/radeon/r600_cp.c
@@ -396,6 +396,9 @@ static void r600_cp_load_microcode(drm_radeon_private_t *dev_priv)
         r600_do_cp_stop(dev_priv);
 
         RADEON_WRITE(R600_CP_RB_CNTL,
+#ifdef __BIG_ENDIAN
+                     R600_BUF_SWAP_32BIT |
+#endif
                      R600_RB_NO_UPDATE |
                      R600_RB_BLKSZ(15) |
                      R600_RB_BUFSZ(3));
@@ -486,9 +489,12 @@ static void r700_cp_load_microcode(drm_radeon_private_t *dev_priv)
         r600_do_cp_stop(dev_priv);
 
         RADEON_WRITE(R600_CP_RB_CNTL,
+#ifdef __BIG_ENDIAN
+                     R600_BUF_SWAP_32BIT |
+#endif
                      R600_RB_NO_UPDATE |
-                     (15 << 8) |
-                     (3 << 0));
+                     R600_RB_BLKSZ(15) |
+                     R600_RB_BUFSZ(3));
 
         RADEON_WRITE(R600_GRBM_SOFT_RESET, R600_SOFT_RESET_CP);
         RADEON_READ(R600_GRBM_SOFT_RESET);
@@ -550,8 +556,12 @@ static void r600_test_writeback(drm_radeon_private_t *dev_priv)
 
         if (!dev_priv->writeback_works) {
                 /* Disable writeback to avoid unnecessary bus master transfer */
-                RADEON_WRITE(R600_CP_RB_CNTL, RADEON_READ(R600_CP_RB_CNTL) |
-                             RADEON_RB_NO_UPDATE);
+                RADEON_WRITE(R600_CP_RB_CNTL,
+#ifdef __BIG_ENDIAN
+                             R600_BUF_SWAP_32BIT |
+#endif
+                             RADEON_READ(R600_CP_RB_CNTL) |
+                             R600_RB_NO_UPDATE);
                 RADEON_WRITE(R600_SCRATCH_UMSK, 0);
         }
 }
@@ -575,7 +585,11 @@ int r600_do_engine_reset(struct drm_device *dev)
 
         RADEON_WRITE(R600_CP_RB_WPTR_DELAY, 0);
         cp_rb_cntl = RADEON_READ(R600_CP_RB_CNTL);
-        RADEON_WRITE(R600_CP_RB_CNTL, R600_RB_RPTR_WR_ENA);
+        RADEON_WRITE(R600_CP_RB_CNTL,
+#ifdef __BIG_ENDIAN
+                     R600_BUF_SWAP_32BIT |
+#endif
+                     R600_RB_RPTR_WR_ENA);
 
         RADEON_WRITE(R600_CP_RB_RPTR_WR, cp_ptr);
         RADEON_WRITE(R600_CP_RB_WPTR, cp_ptr);
@@ -1838,7 +1852,10 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
                         + dev_priv->gart_vm_start;
         }
         RADEON_WRITE(R600_CP_RB_RPTR_ADDR,
-                     rptr_addr & 0xffffffff);
+#ifdef __BIG_ENDIAN
+                     (2 << 0) |
+#endif
+                     (rptr_addr & 0xfffffffc));
         RADEON_WRITE(R600_CP_RB_RPTR_ADDR_HI,
                      upper_32_bits(rptr_addr));
 
@@ -1889,7 +1906,7 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
         {
                 u64 scratch_addr;
 
-                scratch_addr = RADEON_READ(R600_CP_RB_RPTR_ADDR);
+                scratch_addr = RADEON_READ(R600_CP_RB_RPTR_ADDR) & 0xFFFFFFFC;
                 scratch_addr |= ((u64)RADEON_READ(R600_CP_RB_RPTR_ADDR_HI)) << 32;
                 scratch_addr += R600_SCRATCH_REG_OFFSET;
                 scratch_addr >>= 8;
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 7831e0890210..153095fba62f 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -295,17 +295,18 @@ static inline int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
         }
 
         if (!IS_ALIGNED(pitch, pitch_align)) {
-                dev_warn(p->dev, "%s:%d cb pitch (%d) invalid\n",
-                         __func__, __LINE__, pitch);
+                dev_warn(p->dev, "%s:%d cb pitch (%d, 0x%x, %d) invalid\n",
+                         __func__, __LINE__, pitch, pitch_align, array_mode);
                 return -EINVAL;
         }
         if (!IS_ALIGNED(height, height_align)) {
-                dev_warn(p->dev, "%s:%d cb height (%d) invalid\n",
-                         __func__, __LINE__, height);
+                dev_warn(p->dev, "%s:%d cb height (%d, 0x%x, %d) invalid\n",
+                         __func__, __LINE__, height, height_align, array_mode);
                 return -EINVAL;
         }
         if (!IS_ALIGNED(base_offset, base_align)) {
-                dev_warn(p->dev, "%s offset[%d] 0x%llx not aligned\n", __func__, i, base_offset);
+                dev_warn(p->dev, "%s offset[%d] 0x%llx 0x%llx, %d not aligned\n", __func__, i,
+                         base_offset, base_align, array_mode);
                 return -EINVAL;
         }
 
@@ -320,7 +321,10 @@ static inline int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
                          * broken userspace.
                          */
                 } else {
-                        dev_warn(p->dev, "%s offset[%d] %d %d %lu too big\n", __func__, i, track->cb_color_bo_offset[i], tmp, radeon_bo_size(track->cb_color_bo[i]));
+                        dev_warn(p->dev, "%s offset[%d] %d %d %d %lu too big\n", __func__, i,
+                                 array_mode,
+                                 track->cb_color_bo_offset[i], tmp,
+                                 radeon_bo_size(track->cb_color_bo[i]));
                         return -EINVAL;
                 }
         }
@@ -455,17 +459,18 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
                         }
 
                         if (!IS_ALIGNED(pitch, pitch_align)) {
-                                dev_warn(p->dev, "%s:%d db pitch (%d) invalid\n",
-                                         __func__, __LINE__, pitch);
+                                dev_warn(p->dev, "%s:%d db pitch (%d, 0x%x, %d) invalid\n",
+                                         __func__, __LINE__, pitch, pitch_align, array_mode);
                                 return -EINVAL;
                         }
                         if (!IS_ALIGNED(height, height_align)) {
-                                dev_warn(p->dev, "%s:%d db height (%d) invalid\n",
-                                         __func__, __LINE__, height);
+                                dev_warn(p->dev, "%s:%d db height (%d, 0x%x, %d) invalid\n",
+                                         __func__, __LINE__, height, height_align, array_mode);
                                 return -EINVAL;
                         }
                         if (!IS_ALIGNED(base_offset, base_align)) {
-                                dev_warn(p->dev, "%s offset[%d] 0x%llx not aligned\n", __func__, i, base_offset);
+                                dev_warn(p->dev, "%s offset[%d] 0x%llx, 0x%llx, %d not aligned\n", __func__, i,
+                                         base_offset, base_align, array_mode);
                                 return -EINVAL;
                         }
 
@@ -473,9 +478,10 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
                         nviews = G_028004_SLICE_MAX(track->db_depth_view) + 1;
                         tmp = ntiles * bpe * 64 * nviews;
                         if ((tmp + track->db_offset) > radeon_bo_size(track->db_bo)) {
-                                dev_warn(p->dev, "z/stencil buffer too small (0x%08X %d %d %d -> %u have %lu)\n",
-                                                track->db_depth_size, ntiles, nviews, bpe, tmp + track->db_offset,
-                                                radeon_bo_size(track->db_bo));
+                                dev_warn(p->dev, "z/stencil buffer (%d) too small (0x%08X %d %d %d -> %u have %lu)\n",
+                                         array_mode,
+                                         track->db_depth_size, ntiles, nviews, bpe, tmp + track->db_offset,
+                                         radeon_bo_size(track->db_bo));
                                 return -EINVAL;
                         }
                 }
@@ -1227,18 +1233,18 @@ static inline int r600_check_texture_resource(struct radeon_cs_parser *p,  u32 i
         /* XXX check height as well... */
 
         if (!IS_ALIGNED(pitch, pitch_align)) {
-                dev_warn(p->dev, "%s:%d tex pitch (%d) invalid\n",
-                         __func__, __LINE__, pitch);
+                dev_warn(p->dev, "%s:%d tex pitch (%d, 0x%x, %d) invalid\n",
+                         __func__, __LINE__, pitch, pitch_align, G_038000_TILE_MODE(word0));
                 return -EINVAL;
         }
         if (!IS_ALIGNED(base_offset, base_align)) {
-                dev_warn(p->dev, "%s:%d tex base offset (0x%llx) invalid\n",
-                         __func__, __LINE__, base_offset);
+                dev_warn(p->dev, "%s:%d tex base offset (0x%llx, 0x%llx, %d) invalid\n",
+                         __func__, __LINE__, base_offset, base_align, G_038000_TILE_MODE(word0));
                 return -EINVAL;
         }
         if (!IS_ALIGNED(mip_offset, base_align)) {
-                dev_warn(p->dev, "%s:%d tex mip offset (0x%llx) invalid\n",
-                         __func__, __LINE__, mip_offset);
+                dev_warn(p->dev, "%s:%d tex mip offset (0x%llx, 0x%llx, %d) invalid\n",
+                         __func__, __LINE__, mip_offset, base_align, G_038000_TILE_MODE(word0));
                 return -EINVAL;
         }
 
diff --git a/drivers/gpu/drm/radeon/r600_reg.h b/drivers/gpu/drm/radeon/r600_reg.h
index 33cda016b083..f869897c7456 100644
--- a/drivers/gpu/drm/radeon/r600_reg.h
+++ b/drivers/gpu/drm/radeon/r600_reg.h
@@ -81,7 +81,11 @@
 #define R600_MEDIUM_VID_LOWER_GPIO_CNTL                            0x720
 #define R600_LOW_VID_LOWER_GPIO_CNTL                               0x724
 
-
+#define R600_D1GRPH_SWAP_CONTROL                               0x610C
+#       define R600_D1GRPH_SWAP_ENDIAN_NONE                    (0 << 0)
+#       define R600_D1GRPH_SWAP_ENDIAN_16BIT                   (1 << 0)
+#       define R600_D1GRPH_SWAP_ENDIAN_32BIT                   (2 << 0)
+#       define R600_D1GRPH_SWAP_ENDIAN_64BIT                   (3 << 0)
 
 #define R600_HDP_NONSURFACE_BASE                                0x2c04
 
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index a5d898b4bad2..04bac0bbd3ec 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -154,13 +154,14 @@
 #define         ROQ_IB2_START(x)                                ((x) << 8)
 #define CP_RB_BASE                                      0xC100
 #define CP_RB_CNTL                                      0xC104
-#define         RB_BUFSZ(x)                                     ((x)<<0)
-#define         RB_BLKSZ(x)                                     ((x)<<8)
-#define         RB_NO_UPDATE                                    (1<<27)
-#define         RB_RPTR_WR_ENA                                  (1<<31)
+#define         RB_BUFSZ(x)                                     ((x) << 0)
+#define         RB_BLKSZ(x)                                     ((x) << 8)
+#define         RB_NO_UPDATE                                    (1 << 27)
+#define         RB_RPTR_WR_ENA                                  (1 << 31)
 #define         BUF_SWAP_32BIT                                  (2 << 16)
 #define CP_RB_RPTR                                      0x8700
 #define CP_RB_RPTR_ADDR                                 0xC10C
+#define         RB_RPTR_SWAP(x)                                 ((x) << 0)
 #define CP_RB_RPTR_ADDR_HI                              0xC110
 #define CP_RB_RPTR_WR                                   0xC108
 #define CP_RB_WPTR                                      0xC114
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 71d2a554bbe6..56c48b67ef3d 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -179,10 +179,10 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev);
 void radeon_atombios_get_power_modes(struct radeon_device *rdev);
 void radeon_atom_set_voltage(struct radeon_device *rdev, u16 level);
 void rs690_pm_info(struct radeon_device *rdev);
-extern u32 rv6xx_get_temp(struct radeon_device *rdev);
-extern u32 rv770_get_temp(struct radeon_device *rdev);
-extern u32 evergreen_get_temp(struct radeon_device *rdev);
-extern u32 sumo_get_temp(struct radeon_device *rdev);
+extern int rv6xx_get_temp(struct radeon_device *rdev);
+extern int rv770_get_temp(struct radeon_device *rdev);
+extern int evergreen_get_temp(struct radeon_device *rdev);
+extern int sumo_get_temp(struct radeon_device *rdev);
 
 /*
  * Fences.
@@ -812,8 +812,7 @@ struct radeon_pm {
         fixed20_12              sclk;
         fixed20_12              mclk;
         fixed20_12              needed_bandwidth;
-        /* XXX: use a define for num power modes */
-        struct radeon_power_state power_state[8];
+        struct radeon_power_state *power_state;
         /* number of valid power states */
         int                     num_power_states;
         int                     current_power_state_index;
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 3a1b16186224..e75d63b8e21d 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -759,7 +759,7 @@ static struct radeon_asic evergreen_asic = {
         .gart_tlb_flush = &evergreen_pcie_gart_tlb_flush,
         .gart_set_page = &rs600_gart_set_page,
         .ring_test = &r600_ring_test,
-        .ring_ib_execute = &r600_ring_ib_execute,
+        .ring_ib_execute = &evergreen_ring_ib_execute,
         .irq_set = &evergreen_irq_set,
         .irq_process = &evergreen_irq_process,
         .get_vblank_counter = &evergreen_get_vblank_counter,
@@ -805,7 +805,7 @@ static struct radeon_asic sumo_asic = {
         .gart_tlb_flush = &evergreen_pcie_gart_tlb_flush,
         .gart_set_page = &rs600_gart_set_page,
         .ring_test = &r600_ring_test,
-        .ring_ib_execute = &r600_ring_ib_execute,
+        .ring_ib_execute = &evergreen_ring_ib_execute,
         .irq_set = &evergreen_irq_set,
         .irq_process = &evergreen_irq_process,
         .get_vblank_counter = &evergreen_get_vblank_counter,
@@ -848,7 +848,7 @@ static struct radeon_asic btc_asic = {
         .gart_tlb_flush = &evergreen_pcie_gart_tlb_flush,
         .gart_set_page = &rs600_gart_set_page,
         .ring_test = &r600_ring_test,
-        .ring_ib_execute = &r600_ring_ib_execute,
+        .ring_ib_execute = &evergreen_ring_ib_execute,
         .irq_set = &evergreen_irq_set,
         .irq_process = &evergreen_irq_process,
         .get_vblank_counter = &evergreen_get_vblank_counter,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index e01f07718539..c59bd98a2029 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -355,6 +355,7 @@ int evergreen_resume(struct radeon_device *rdev);
 bool evergreen_gpu_is_lockup(struct radeon_device *rdev);
 int evergreen_asic_reset(struct radeon_device *rdev);
 void evergreen_bandwidth_update(struct radeon_device *rdev);
+void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
 int evergreen_copy_blit(struct radeon_device *rdev,
                         uint64_t src_offset, uint64_t dst_offset,
                         unsigned num_pages, struct radeon_fence *fence);
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 52777902bbcc..02d5c415f499 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -88,7 +88,7 @@ static inline struct radeon_i2c_bus_rec radeon_lookup_i2c_gpio(struct radeon_dev
                         /* some evergreen boards have bad data for this entry */
                         if (ASIC_IS_DCE4(rdev)) {
                                 if ((i == 7) &&
-                                    (gpio->usClkMaskRegisterIndex == 0x1936) &&
+                                    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
                                     (gpio->sucI2cId.ucAccess == 0)) {
                                         gpio->sucI2cId.ucAccess = 0x97;
                                         gpio->ucDataMaskShift = 8;
@@ -101,7 +101,7 @@ static inline struct radeon_i2c_bus_rec radeon_lookup_i2c_gpio(struct radeon_dev
                         /* some DCE3 boards have bad data for this entry */
                         if (ASIC_IS_DCE3(rdev)) {
                                 if ((i == 4) &&
-                                    (gpio->usClkMaskRegisterIndex == 0x1fda) &&
+                                    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
                                     (gpio->sucI2cId.ucAccess == 0x94))
                                         gpio->sucI2cId.ucAccess = 0x14;
                         }
@@ -172,7 +172,7 @@ void radeon_atombios_i2c_init(struct radeon_device *rdev)
                         /* some evergreen boards have bad data for this entry */
                         if (ASIC_IS_DCE4(rdev)) {
                                 if ((i == 7) &&
-                                    (gpio->usClkMaskRegisterIndex == 0x1936) &&
+                                    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
                                     (gpio->sucI2cId.ucAccess == 0)) {
                                         gpio->sucI2cId.ucAccess = 0x97;
                                         gpio->ucDataMaskShift = 8;
@@ -185,7 +185,7 @@ void radeon_atombios_i2c_init(struct radeon_device *rdev)
                         /* some DCE3 boards have bad data for this entry */
                         if (ASIC_IS_DCE3(rdev)) {
                                 if ((i == 4) &&
-                                    (gpio->usClkMaskRegisterIndex == 0x1fda) &&
+                                    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
                                     (gpio->sucI2cId.ucAccess == 0x94))
                                         gpio->sucI2cId.ucAccess = 0x14;
                         }
@@ -252,7 +252,7 @@ static inline struct radeon_gpio_rec radeon_lookup_gpio(struct radeon_device *rd
                         pin = &gpio_info->asGPIO_Pin[i];
                         if (id == pin->ucGPIO_ID) {
                                 gpio.id = pin->ucGPIO_ID;
-                                gpio.reg = pin->usGpioPin_AIndex * 4;
+                                gpio.reg = le16_to_cpu(pin->usGpioPin_AIndex) * 4;
                                 gpio.mask = (1 << pin->ucGpioPinBitShift);
                                 gpio.valid = true;
                                 break;
@@ -1163,16 +1163,6 @@ bool radeon_atom_get_clock_info(struct drm_device *dev)
                                 p1pll->pll_out_min = 64800;
                         else
                                 p1pll->pll_out_min = 20000;
-                } else if (p1pll->pll_out_min > 64800) {
-                        /* Limiting the pll output range is a good thing generally as
-                         * it limits the number of possible pll combinations for a given
-                         * frequency presumably to the ones that work best on each card.
-                         * However, certain duallink DVI monitors seem to like
-                         * pll combinations that would be limited by this at least on
-                         * pre-DCE 3.0 r6xx hardware.  This might need to be adjusted per
-                         * family.
-                         */
-                        p1pll->pll_out_min = 64800;
                 }
 
                 p1pll->pll_in_min =
@@ -1284,11 +1274,11 @@ bool radeon_atombios_sideport_present(struct radeon_device *rdev)
                                       data_offset);
                 switch (crev) {
                 case 1:
-                        if (igp_info->info.ulBootUpMemoryClock)
+                        if (le32_to_cpu(igp_info->info.ulBootUpMemoryClock))
                                 return true;
                         break;
                 case 2:
-                        if (igp_info->info_2.ulBootUpSidePortClock)
+                        if (le32_to_cpu(igp_info->info_2.ulBootUpSidePortClock))
                                 return true;
                         break;
                 default:
@@ -1452,7 +1442,7 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
 
                         for (i = 0; i < num_indices; i++) {
                                 if ((ss_info->info.asSpreadSpectrum[i].ucClockIndication == id) &&
-                                    (clock <= ss_info->info.asSpreadSpectrum[i].ulTargetClockRange)) {
+                                    (clock <= le32_to_cpu(ss_info->info.asSpreadSpectrum[i].ulTargetClockRange))) {
                                         ss->percentage =
                                                 le16_to_cpu(ss_info->info.asSpreadSpectrum[i].usSpreadSpectrumPercentage);
                                         ss->type = ss_info->info.asSpreadSpectrum[i].ucSpreadSpectrumMode;
@@ -1466,7 +1456,7 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
                                 sizeof(ATOM_ASIC_SS_ASSIGNMENT_V2);
                         for (i = 0; i < num_indices; i++) {
                                 if ((ss_info->info_2.asSpreadSpectrum[i].ucClockIndication == id) &&
-                                    (clock <= ss_info->info_2.asSpreadSpectrum[i].ulTargetClockRange)) {
+                                    (clock <= le32_to_cpu(ss_info->info_2.asSpreadSpectrum[i].ulTargetClockRange))) {
                                         ss->percentage =
                                                 le16_to_cpu(ss_info->info_2.asSpreadSpectrum[i].usSpreadSpectrumPercentage);
                                         ss->type = ss_info->info_2.asSpreadSpectrum[i].ucSpreadSpectrumMode;
@@ -1480,7 +1470,7 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
                                 sizeof(ATOM_ASIC_SS_ASSIGNMENT_V3);
                         for (i = 0; i < num_indices; i++) {
                                 if ((ss_info->info_3.asSpreadSpectrum[i].ucClockIndication == id) &&
-                                    (clock <= ss_info->info_3.asSpreadSpectrum[i].ulTargetClockRange)) {
+                                    (clock <= le32_to_cpu(ss_info->info_3.asSpreadSpectrum[i].ulTargetClockRange))) {
                                         ss->percentage =
                                                 le16_to_cpu(ss_info->info_3.asSpreadSpectrum[i].usSpreadSpectrumPercentage);
                                         ss->type = ss_info->info_3.asSpreadSpectrum[i].ucSpreadSpectrumMode;
@@ -1563,8 +1553,8 @@ struct radeon_encoder_atom_dig *radeon_atombios_get_lvds_info(struct
                 if (misc & ATOM_DOUBLE_CLOCK_MODE)
                         lvds->native_mode.flags |= DRM_MODE_FLAG_DBLSCAN;
 
-                lvds->native_mode.width_mm = lvds_info->info.sLCDTiming.usImageHSize;
-                lvds->native_mode.height_mm = lvds_info->info.sLCDTiming.usImageVSize;
+                lvds->native_mode.width_mm = le16_to_cpu(lvds_info->info.sLCDTiming.usImageHSize);
+                lvds->native_mode.height_mm = le16_to_cpu(lvds_info->info.sLCDTiming.usImageVSize);
 
                 /* set crtc values */
                 drm_mode_set_crtcinfo(&lvds->native_mode, CRTC_INTERLACE_HALVE_V);
@@ -1579,13 +1569,13 @@ struct radeon_encoder_atom_dig *radeon_atombios_get_lvds_info(struct
                         lvds->linkb = false;
 
                 /* parse the lcd record table */
-                if (lvds_info->info.usModePatchTableOffset) {
+                if (le16_to_cpu(lvds_info->info.usModePatchTableOffset)) {
                         ATOM_FAKE_EDID_PATCH_RECORD *fake_edid_record;
                         ATOM_PANEL_RESOLUTION_PATCH_RECORD *panel_res_record;
                         bool bad_record = false;
                         u8 *record = (u8 *)(mode_info->atom_context->bios +
                                             data_offset +
-                                            lvds_info->info.usModePatchTableOffset);
+                                            le16_to_cpu(lvds_info->info.usModePatchTableOffset));
                         while (*record != ATOM_RECORD_END_TYPE) {
                                 switch (*record) {
                                 case LCD_MODE_PATCH_RECORD_MODE_TYPE:
@@ -1987,6 +1977,9 @@ static int radeon_atombios_parse_power_table_1_3(struct radeon_device *rdev)
         num_modes = power_info->info.ucNumOfPowerModeEntries;
         if (num_modes > ATOM_MAX_NUMBEROF_POWER_BLOCK)
                 num_modes = ATOM_MAX_NUMBEROF_POWER_BLOCK;
+        rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state) * num_modes, GFP_KERNEL);
+        if (!rdev->pm.power_state)
+                return state_index;
         /* last mode is usually default, array is low to high */
         for (i = 0; i < num_modes; i++) {
                 rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
@@ -2196,7 +2189,7 @@ static u16 radeon_atombios_get_default_vddc(struct radeon_device *rdev)
                 firmware_info =
                         (union firmware_info *)(mode_info->atom_context->bios +
                                                 data_offset);
-                vddc = firmware_info->info_14.usBootUpVDDCVoltage;
+                vddc = le16_to_cpu(firmware_info->info_14.usBootUpVDDCVoltage);
         }
 
         return vddc;
@@ -2291,7 +2284,7 @@ static bool radeon_atombios_parse_pplib_clock_info(struct radeon_device *rdev,
                 rdev->pm.power_state[state_index].clock_info[mode_index].voltage.type =
                         VOLTAGE_SW;
                 rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage =
-                        clock_info->evergreen.usVDDC;
+                        le16_to_cpu(clock_info->evergreen.usVDDC);
         } else {
                 sclk = le16_to_cpu(clock_info->r600.usEngineClockLow);
                 sclk |= clock_info->r600.ucEngineClockHigh << 16;
@@ -2302,7 +2295,7 @@ static bool radeon_atombios_parse_pplib_clock_info(struct radeon_device *rdev,
                 rdev->pm.power_state[state_index].clock_info[mode_index].voltage.type =
                         VOLTAGE_SW;
                 rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage =
-                        clock_info->r600.usVDDC;
+                        le16_to_cpu(clock_info->r600.usVDDC);
         }
 
         if (rdev->flags & RADEON_IS_IGP) {
@@ -2338,6 +2331,10 @@ static int radeon_atombios_parse_power_table_4_5(struct radeon_device *rdev)
         power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
 
         radeon_atombios_add_pplib_thermal_controller(rdev, &power_info->pplib.sThermalController);
+        rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state) *
+                                       power_info->pplib.ucNumStates, GFP_KERNEL);
+        if (!rdev->pm.power_state)
+                return state_index;
         /* first mode is usually default, followed by low to high */
         for (i = 0; i < power_info->pplib.ucNumStates; i++) {
                 mode_index = 0;
@@ -2411,13 +2408,17 @@ static int radeon_atombios_parse_power_table_6(struct radeon_device *rdev)
         radeon_atombios_add_pplib_thermal_controller(rdev, &power_info->pplib.sThermalController);
         state_array = (struct StateArray *)
                 (mode_info->atom_context->bios + data_offset +
-                 power_info->pplib.usStateArrayOffset);
+                 le16_to_cpu(power_info->pplib.usStateArrayOffset));
         clock_info_array = (struct ClockInfoArray *)
                 (mode_info->atom_context->bios + data_offset +
-                 power_info->pplib.usClockInfoArrayOffset);
+                 le16_to_cpu(power_info->pplib.usClockInfoArrayOffset));
         non_clock_info_array = (struct NonClockInfoArray *)
                 (mode_info->atom_context->bios + data_offset +
-                 power_info->pplib.usNonClockInfoArrayOffset);
+                 le16_to_cpu(power_info->pplib.usNonClockInfoArrayOffset));
+        rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state) *
+                                       state_array->ucNumEntries, GFP_KERNEL);
+        if (!rdev->pm.power_state)
+                return state_index;
         for (i = 0; i < state_array->ucNumEntries; i++) {
                 mode_index = 0;
                 power_state = (union pplib_power_state *)&state_array->states[i];
@@ -2491,19 +2492,22 @@ void radeon_atombios_get_power_modes(struct radeon_device *rdev)
                         break;
                 }
         } else {
-                /* add the default mode */
-                rdev->pm.power_state[state_index].type =
-                        POWER_STATE_TYPE_DEFAULT;
-                rdev->pm.power_state[state_index].num_clock_modes = 1;
-                rdev->pm.power_state[state_index].clock_info[0].mclk = rdev->clock.default_mclk;
-                rdev->pm.power_state[state_index].clock_info[0].sclk = rdev->clock.default_sclk;
-                rdev->pm.power_state[state_index].default_clock_mode =
-                        &rdev->pm.power_state[state_index].clock_info[0];
-                rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
-                rdev->pm.power_state[state_index].pcie_lanes = 16;
-                rdev->pm.default_power_state_index = state_index;
-                rdev->pm.power_state[state_index].flags = 0;
-                state_index++;
+                rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state), GFP_KERNEL);
+                if (rdev->pm.power_state) {
+                        /* add the default mode */
+                        rdev->pm.power_state[state_index].type =
+                                POWER_STATE_TYPE_DEFAULT;
+                        rdev->pm.power_state[state_index].num_clock_modes = 1;
+                        rdev->pm.power_state[state_index].clock_info[0].mclk = rdev->clock.default_mclk;
+                        rdev->pm.power_state[state_index].clock_info[0].sclk = rdev->clock.default_sclk;
+                        rdev->pm.power_state[state_index].default_clock_mode =
+                                &rdev->pm.power_state[state_index].clock_info[0];
+                        rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
+                        rdev->pm.power_state[state_index].pcie_lanes = 16;
+                        rdev->pm.default_power_state_index = state_index;
+                        rdev->pm.power_state[state_index].flags = 0;
+                        state_index++;
+                }
         }
 
         rdev->pm.num_power_states = state_index;
@@ -2529,7 +2533,7 @@ uint32_t radeon_atom_get_engine_clock(struct radeon_device *rdev)
         int index = GetIndexIntoMasterTable(COMMAND, GetEngineClock);
 
         atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
-        return args.ulReturnEngineClock;
+        return le32_to_cpu(args.ulReturnEngineClock);
 }
 
 uint32_t radeon_atom_get_memory_clock(struct radeon_device *rdev)
@@ -2538,7 +2542,7 @@ uint32_t radeon_atom_get_memory_clock(struct radeon_device *rdev)
         int index = GetIndexIntoMasterTable(COMMAND, GetMemoryClock);
 
         atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
-        return args.ulReturnMemoryClock;
+        return le32_to_cpu(args.ulReturnMemoryClock);
 }
 
 void radeon_atom_set_engine_clock(struct radeon_device *rdev,
@@ -2547,7 +2551,7 @@ void radeon_atom_set_engine_clock(struct radeon_device *rdev,
         SET_ENGINE_CLOCK_PS_ALLOCATION args;
         int index = GetIndexIntoMasterTable(COMMAND, SetEngineClock);
 
-        args.ulTargetEngineClock = eng_clock;   /* 10 khz */
+        args.ulTargetEngineClock = cpu_to_le32(eng_clock);      /* 10 khz */
 
         atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 }
@@ -2561,7 +2565,7 @@ void radeon_atom_set_memory_clock(struct radeon_device *rdev,
         if (rdev->flags & RADEON_IS_IGP)
                 return;
 
-        args.ulTargetMemoryClock = mem_clock;   /* 10 khz */
+        args.ulTargetMemoryClock = cpu_to_le32(mem_clock);      /* 10 khz */
 
         atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 }
@@ -2619,7 +2623,7 @@ void radeon_atom_initialize_bios_scratch_regs(struct drm_device *dev)
         bios_2_scratch &= ~ATOM_S2_VRI_BRIGHT_ENABLE;
 
         /* tell the bios not to handle mode switching */
-        bios_6_scratch |= (ATOM_S6_ACC_BLOCK_DISPLAY_SWITCH | ATOM_S6_ACC_MODE);
+        bios_6_scratch |= ATOM_S6_ACC_BLOCK_DISPLAY_SWITCH;
 
         if (rdev->family >= CHIP_R600) {
                 WREG32(R600_BIOS_2_SCRATCH, bios_2_scratch);
@@ -2670,10 +2674,13 @@ void radeon_atom_output_lock(struct drm_encoder *encoder, bool lock)
         else
                 bios_6_scratch = RREG32(RADEON_BIOS_6_SCRATCH);
 
-        if (lock)
+        if (lock) {
                 bios_6_scratch |= ATOM_S6_CRITICAL_STATE;
-        else
+                bios_6_scratch &= ~ATOM_S6_ACC_MODE;
+        } else {
                 bios_6_scratch &= ~ATOM_S6_CRITICAL_STATE;
+                bios_6_scratch |= ATOM_S6_ACC_MODE;
+        }
 
         if (rdev->family >= CHIP_R600)
                 WREG32(R600_BIOS_6_SCRATCH, bios_6_scratch);
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index 591fcae8f224..cf7c8d5b4ec2 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -1504,6 +1504,11 @@ bool radeon_get_legacy_connector_info_from_table(struct drm_device *dev)
                            (rdev->pdev->subsystem_device == 0x4a48)) {
                         /* Mac X800 */
                         rdev->mode_info.connector_table = CT_MAC_X800;
+                } else if ((rdev->pdev->device == 0x4150) &&
+                           (rdev->pdev->subsystem_vendor == 0x1002) &&
+                           (rdev->pdev->subsystem_device == 0x4150)) {
+                        /* Mac G5 9600 */
+                        rdev->mode_info.connector_table = CT_MAC_G5_9600;
                 } else
 #endif /* CONFIG_PPC_PMAC */
 #ifdef CONFIG_PPC64
@@ -2022,6 +2027,48 @@ bool radeon_get_legacy_connector_info_from_table(struct drm_device *dev)
                                             CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_I,
                                             &hpd);
                 break;
+        case CT_MAC_G5_9600:
+                DRM_INFO("Connector Table: %d (mac g5 9600)\n",
+                         rdev->mode_info.connector_table);
+                /* DVI - tv dac, dvo */
+                ddc_i2c = combios_setup_i2c_bus(rdev, DDC_DVI, 0, 0);
+                hpd.hpd = RADEON_HPD_1; /* ??? */
+                radeon_add_legacy_encoder(dev,
+                                          radeon_get_encoder_enum(dev,
+                                                                  ATOM_DEVICE_DFP2_SUPPORT,
+                                                                  0),
+                                          ATOM_DEVICE_DFP2_SUPPORT);
+                radeon_add_legacy_encoder(dev,
+                                          radeon_get_encoder_enum(dev,
+                                                                  ATOM_DEVICE_CRT2_SUPPORT,
+                                                                  2),
+                                          ATOM_DEVICE_CRT2_SUPPORT);
+                radeon_add_legacy_connector(dev, 0,
+                                            ATOM_DEVICE_DFP2_SUPPORT |
+                                            ATOM_DEVICE_CRT2_SUPPORT,
+                                            DRM_MODE_CONNECTOR_DVII, &ddc_i2c,
+                                            CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_I,
+                                            &hpd);
+                /* ADC - primary dac, internal tmds */
+                ddc_i2c = combios_setup_i2c_bus(rdev, DDC_VGA, 0, 0);
+                hpd.hpd = RADEON_HPD_2; /* ??? */
+                radeon_add_legacy_encoder(dev,
+                                          radeon_get_encoder_enum(dev,
+                                                                  ATOM_DEVICE_DFP1_SUPPORT,
+                                                                  0),
+                                          ATOM_DEVICE_DFP1_SUPPORT);
+                radeon_add_legacy_encoder(dev,
+                                          radeon_get_encoder_enum(dev,
+                                                                  ATOM_DEVICE_CRT1_SUPPORT,
+                                                                  1),
+                                          ATOM_DEVICE_CRT1_SUPPORT);
+                radeon_add_legacy_connector(dev, 1,
+                                            ATOM_DEVICE_DFP1_SUPPORT |
+                                            ATOM_DEVICE_CRT1_SUPPORT,
+                                            DRM_MODE_CONNECTOR_DVII, &ddc_i2c,
+                                            CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_I,
+                                            &hpd);
+                break;
         default:
                 DRM_INFO("Connector table: %d (invalid)\n",
                          rdev->mode_info.connector_table);
@@ -2442,6 +2489,17 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev)
 
         rdev->pm.default_power_state_index = -1;
 
+        /* allocate 2 power states */
+        rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state) * 2, GFP_KERNEL);
+        if (!rdev->pm.power_state) {
+                rdev->pm.default_power_state_index = state_index;
+                rdev->pm.num_power_states = 0;
+
+                rdev->pm.current_power_state_index = rdev->pm.default_power_state_index;
+                rdev->pm.current_clock_mode_index = 0;
+                return;
+        }
+
         if (rdev->flags & RADEON_IS_MOBILITY) {
                 offset = combios_get_table_offset(dev, COMBIOS_POWERPLAY_INFO_TABLE);
                 if (offset) {
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 26091d602b84..4954e2d6ffa2 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -891,9 +891,9 @@ int radeon_suspend_kms(struct drm_device *dev, pm_message_t state)
                 pci_disable_device(dev->pdev);
                 pci_set_power_state(dev->pdev, PCI_D3hot);
         }
-        acquire_console_sem();
+        console_lock();
         radeon_fbdev_set_suspend(rdev, 1);
-        release_console_sem();
+        console_unlock();
         return 0;
 }
 
@@ -905,11 +905,11 @@ int radeon_resume_kms(struct drm_device *dev)
         if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
                 return 0;
 
-        acquire_console_sem();
+        console_lock();
         pci_set_power_state(dev->pdev, PCI_D0);
         pci_restore_state(dev->pdev);
         if (pci_enable_device(dev->pdev)) {
-                release_console_sem();
+                console_unlock();
                 return -1;
         }
         pci_set_master(dev->pdev);
@@ -920,7 +920,7 @@ int radeon_resume_kms(struct drm_device *dev)
         radeon_restore_bios_scratch_regs(rdev);
 
         radeon_fbdev_set_suspend(rdev, 0);
-        release_console_sem();
+        console_unlock();
 
         /* reset hpd state */
         radeon_hpd_init(rdev);
@@ -936,8 +936,11 @@ int radeon_resume_kms(struct drm_device *dev)
 int radeon_gpu_reset(struct radeon_device *rdev)
 {
         int r;
+        int resched;
 
         radeon_save_bios_scratch_regs(rdev);
+        /* block TTM */
+        resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
         radeon_suspend(rdev);
 
         r = radeon_asic_reset(rdev);
@@ -946,6 +949,7 @@ int radeon_gpu_reset(struct radeon_device *rdev)
                 radeon_resume(rdev);
                 radeon_restore_bios_scratch_regs(rdev);
                 drm_helper_resume_force_mode(rdev->ddev);
+                ttm_bo_unlock_delayed_workqueue(&rdev->mman.bdev, resched);
                 return 0;
         }
         /* bad news, how to tell it to userspace ? */
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index d26dabf878d9..0e657095de7c 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -780,6 +780,125 @@ static int radeon_ddc_dump(struct drm_connector *connector)
         return ret;
 }
 
+/* avivo */
+static void avivo_get_fb_div(struct radeon_pll *pll,
+                             u32 target_clock,
+                             u32 post_div,
+                             u32 ref_div,
+                             u32 *fb_div,
+                             u32 *frac_fb_div)
+{
+        u32 tmp = post_div * ref_div;
+
+        tmp *= target_clock;
+        *fb_div = tmp / pll->reference_freq;
+        *frac_fb_div = tmp % pll->reference_freq;
+
+        if (*fb_div > pll->max_feedback_div)
+                *fb_div = pll->max_feedback_div;
+        else if (*fb_div < pll->min_feedback_div)
+                *fb_div = pll->min_feedback_div;
+}
+
+static u32 avivo_get_post_div(struct radeon_pll *pll,
+                              u32 target_clock)
+{
+        u32 vco, post_div, tmp;
+
+        if (pll->flags & RADEON_PLL_USE_POST_DIV)
+                return pll->post_div;
+
+        if (pll->flags & RADEON_PLL_PREFER_MINM_OVER_MAXP) {
+                if (pll->flags & RADEON_PLL_IS_LCD)
+                        vco = pll->lcd_pll_out_min;
+                else
+                        vco = pll->pll_out_min;
+        } else {
+                if (pll->flags & RADEON_PLL_IS_LCD)
+                        vco = pll->lcd_pll_out_max;
+                else
+                        vco = pll->pll_out_max;
+        }
+
+        post_div = vco / target_clock;
+        tmp = vco % target_clock;
+
+        if (pll->flags & RADEON_PLL_PREFER_MINM_OVER_MAXP) {
+                if (tmp)
+                        post_div++;
+        } else {
+                if (!tmp)
+                        post_div--;
+        }
+
+        if (post_div > pll->max_post_div)
+                post_div = pll->max_post_div;
+        else if (post_div < pll->min_post_div)
+                post_div = pll->min_post_div;
+
+        return post_div;
+}
+
+#define MAX_TOLERANCE 10
+
+void radeon_compute_pll_avivo(struct radeon_pll *pll,
+                              u32 freq,
+                              u32 *dot_clock_p,
+                              u32 *fb_div_p,
+                              u32 *frac_fb_div_p,
+                              u32 *ref_div_p,
+                              u32 *post_div_p)
+{
+        u32 target_clock = freq / 10;
+        u32 post_div = avivo_get_post_div(pll, target_clock);
+        u32 ref_div = pll->min_ref_div;
+        u32 fb_div = 0, frac_fb_div = 0, tmp;
+
+        if (pll->flags & RADEON_PLL_USE_REF_DIV)
+                ref_div = pll->reference_div;
+
+        if (pll->flags & RADEON_PLL_USE_FRAC_FB_DIV) {
+                avivo_get_fb_div(pll, target_clock, post_div, ref_div, &fb_div, &frac_fb_div);
+                frac_fb_div = (100 * frac_fb_div) / pll->reference_freq;
+                if (frac_fb_div >= 5) {
+                        frac_fb_div -= 5;
+                        frac_fb_div = frac_fb_div / 10;
+                        frac_fb_div++;
+                }
+                if (frac_fb_div >= 10) {
+                        fb_div++;
+                        frac_fb_div = 0;
+                }
+        } else {
+                while (ref_div <= pll->max_ref_div) {
+                        avivo_get_fb_div(pll, target_clock, post_div, ref_div,
+                                         &fb_div, &frac_fb_div);
+                        if (frac_fb_div >= (pll->reference_freq / 2))
+                                fb_div++;
+                        frac_fb_div = 0;
+                        tmp = (pll->reference_freq * fb_div) / (post_div * ref_div);
+                        tmp = (tmp * 10000) / target_clock;
+
+                        if (tmp > (10000 + MAX_TOLERANCE))
+                                ref_div++;
+                        else if (tmp >= (10000 - MAX_TOLERANCE))
+                                break;
+                        else
+                                ref_div++;
+                }
+        }
+
+        *dot_clock_p = ((pll->reference_freq * fb_div * 10) + (pll->reference_freq * frac_fb_div)) /
+                (ref_div * post_div * 10);
+        *fb_div_p = fb_div;
+        *frac_fb_div_p = frac_fb_div;
+        *ref_div_p = ref_div;
+        *post_div_p = post_div;
+        DRM_DEBUG_KMS("%d, pll dividers - fb: %d.%d ref: %d, post %d\n",
+                      *dot_clock_p, fb_div, frac_fb_div, ref_div, post_div);
+}
+
+/* pre-avivo */
 static inline uint32_t radeon_div(uint64_t n, uint32_t d)
 {
         uint64_t mod;
@@ -790,13 +909,13 @@ static inline uint32_t radeon_div(uint64_t n, uint32_t d)
         return n;
 }
 
-void radeon_compute_pll(struct radeon_pll *pll,
-                        uint64_t freq,
-                        uint32_t *dot_clock_p,
-                        uint32_t *fb_div_p,
-                        uint32_t *frac_fb_div_p,
-                        uint32_t *ref_div_p,
-                        uint32_t *post_div_p)
+void radeon_compute_pll_legacy(struct radeon_pll *pll,
+                               uint64_t freq,
+                               uint32_t *dot_clock_p,
+                               uint32_t *fb_div_p,
+                               uint32_t *frac_fb_div_p,
+                               uint32_t *ref_div_p,
+                               uint32_t *post_div_p)
 {
         uint32_t min_ref_div = pll->min_ref_div;
         uint32_t max_ref_div = pll->max_ref_div;
@@ -826,6 +945,9 @@ void radeon_compute_pll(struct radeon_pll *pll,
                 pll_out_max = pll->pll_out_max;
         }
 
+        if (pll_out_min > 64800)
+                pll_out_min = 64800;
+
         if (pll->flags & RADEON_PLL_USE_REF_DIV)
                 min_ref_div = max_ref_div = pll->reference_div;
         else {
@@ -849,7 +971,7 @@ void radeon_compute_pll(struct radeon_pll *pll,
                 max_fractional_feed_div = pll->max_frac_feedback_div;
         }
 
-        for (post_div = max_post_div; post_div >= min_post_div; --post_div) {
+        for (post_div = min_post_div; post_div <= max_post_div; ++post_div) {
                 uint32_t ref_div;
 
                 if ((pll->flags & RADEON_PLL_NO_ODD_POST_DIV) && (post_div & 1))
@@ -965,6 +1087,10 @@ void radeon_compute_pll(struct radeon_pll *pll,
         *frac_fb_div_p = best_frac_feedback_div;
         *ref_div_p = best_ref_div;
         *post_div_p = best_post_div;
+        DRM_DEBUG_KMS("%d %d, pll dividers - fb: %d.%d ref: %d, post %d\n",
+                      freq, best_freq / 1000, best_feedback_div, best_frac_feedback_div,
+                      best_ref_div, best_post_div);
+
 }
 
 static void radeon_user_framebuffer_destroy(struct drm_framebuffer *fb)
diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h
index 448eba89d1e6..5cba46b9779a 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.h
+++ b/drivers/gpu/drm/radeon/radeon_drv.h
@@ -1524,6 +1524,7 @@ extern u32 radeon_get_scratch(drm_radeon_private_t *dev_priv, int index);
 #define R600_CP_RB_CNTL                                        0xc104
 #       define R600_RB_BUFSZ(x)                                ((x) << 0)
 #       define R600_RB_BLKSZ(x)                                ((x) << 8)
+#       define R600_BUF_SWAP_32BIT                             (2 << 16)
 #       define R600_RB_NO_UPDATE                               (1 << 27)
 #       define R600_RB_RPTR_WR_ENA                             (1 << 31)
 #define R600_CP_RB_RPTR_WR                                     0xc108
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index 8fd184286c0b..b4274883227f 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -641,7 +641,7 @@ atombios_get_encoder_mode(struct drm_encoder *encoder)
         switch (connector->connector_type) {
         case DRM_MODE_CONNECTOR_DVII:
         case DRM_MODE_CONNECTOR_HDMIB: /* HDMI-B is basically DL-DVI; analog works fine */
-                if (drm_detect_monitor_audio(radeon_connector->edid)) {
+                if (drm_detect_monitor_audio(radeon_connector->edid) && radeon_audio) {
                         /* fix me */
                         if (ASIC_IS_DCE4(rdev))
                                 return ATOM_ENCODER_MODE_DVI;
@@ -655,7 +655,7 @@ atombios_get_encoder_mode(struct drm_encoder *encoder)
         case DRM_MODE_CONNECTOR_DVID:
         case DRM_MODE_CONNECTOR_HDMIA:
         default:
-                if (drm_detect_monitor_audio(radeon_connector->edid)) {
+                if (drm_detect_monitor_audio(radeon_connector->edid) && radeon_audio) {
                         /* fix me */
                         if (ASIC_IS_DCE4(rdev))
                                 return ATOM_ENCODER_MODE_DVI;
@@ -673,7 +673,7 @@ atombios_get_encoder_mode(struct drm_encoder *encoder)
                 if ((dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT) ||
                     (dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP))
                         return ATOM_ENCODER_MODE_DP;
-                else if (drm_detect_monitor_audio(radeon_connector->edid)) {
+                else if (drm_detect_monitor_audio(radeon_connector->edid) && radeon_audio) {
                         /* fix me */
                         if (ASIC_IS_DCE4(rdev))
                                 return ATOM_ENCODER_MODE_DVI;
@@ -910,7 +910,7 @@ atombios_dig_transmitter_setup(struct drm_encoder *encoder, int action, uint8_t
 
         args.v1.ucAction = action;
         if (action == ATOM_TRANSMITTER_ACTION_INIT) {
-                args.v1.usInitInfo = connector_object_id;
+                args.v1.usInitInfo = cpu_to_le16(connector_object_id);
         } else if (action == ATOM_TRANSMITTER_ACTION_SETUP_VSEMPH) {
                 args.v1.asMode.ucLaneSel = lane_num;
                 args.v1.asMode.ucLaneSet = lane_set;
@@ -1063,7 +1063,7 @@ atombios_set_edp_panel_power(struct drm_connector *connector, int action)
         if (!ASIC_IS_DCE4(rdev))
                 return;
 
-        if ((action != ATOM_TRANSMITTER_ACTION_POWER_ON) ||
+        if ((action != ATOM_TRANSMITTER_ACTION_POWER_ON) &&
             (action != ATOM_TRANSMITTER_ACTION_POWER_OFF))
                 return;
 
@@ -1140,7 +1140,7 @@ atombios_external_encoder_setup(struct drm_encoder *encoder,
                 case 3:
                         args.v3.sExtEncoder.ucAction = action;
                         if (action == EXTERNAL_ENCODER_ACTION_V3_ENCODER_INIT)
-                                args.v3.sExtEncoder.usConnectorId = connector_object_id;
+                                args.v3.sExtEncoder.usConnectorId = cpu_to_le16(connector_object_id);
                         else
                                 args.v3.sExtEncoder.usPixelClock = cpu_to_le16(radeon_encoder->pixel_clock / 10);
                         args.v3.sExtEncoder.ucEncoderMode = atombios_get_encoder_mode(encoder);
@@ -1570,11 +1570,21 @@ atombios_apply_encoder_quirks(struct drm_encoder *encoder,
         }
 
         /* set scaler clears this on some chips */
-        /* XXX check DCE4 */
-        if (!(radeon_encoder->active_device & (ATOM_DEVICE_TV_SUPPORT))) {
-                if (ASIC_IS_AVIVO(rdev) && (mode->flags & DRM_MODE_FLAG_INTERLACE))
-                        WREG32(AVIVO_D1MODE_DATA_FORMAT + radeon_crtc->crtc_offset,
-                               AVIVO_D1MODE_INTERLEAVE_EN);
+        if (ASIC_IS_AVIVO(rdev) &&
+            (!(radeon_encoder->active_device & (ATOM_DEVICE_TV_SUPPORT)))) {
+                if (ASIC_IS_DCE4(rdev)) {
+                        if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+                                WREG32(EVERGREEN_DATA_FORMAT + radeon_crtc->crtc_offset,
+                                       EVERGREEN_INTERLEAVE_EN);
+                        else
+                                WREG32(EVERGREEN_DATA_FORMAT + radeon_crtc->crtc_offset, 0);
+                } else {
+                        if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+                                WREG32(AVIVO_D1MODE_DATA_FORMAT + radeon_crtc->crtc_offset,
+                                       AVIVO_D1MODE_INTERLEAVE_EN);
+                        else
+                                WREG32(AVIVO_D1MODE_DATA_FORMAT + radeon_crtc->crtc_offset, 0);
+                }
         }
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 98321298cffd..8387d32caaa7 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -247,6 +247,8 @@ void radeon_driver_preclose_kms(struct drm_device *dev,
         struct radeon_device *rdev = dev->dev_private;
         if (rdev->hyperz_filp == file_priv)
                 rdev->hyperz_filp = NULL;
+        if (rdev->cmask_filp == file_priv)
+                rdev->cmask_filp = NULL;
 }
 
 /*
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index ace2e6384d40..cf0638c3b7c7 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -778,9 +778,9 @@ static void radeon_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode)
         DRM_DEBUG_KMS("\n");
 
         if (!use_bios_divs) {
-                radeon_compute_pll(pll, mode->clock,
-                                   &freq, &feedback_div, &frac_fb_div,
-                                   &reference_div, &post_divider);
+                radeon_compute_pll_legacy(pll, mode->clock,
+                                          &freq, &feedback_div, &frac_fb_div,
+                                          &reference_div, &post_divider);
 
                 for (post_div = &post_divs[0]; post_div->divider; ++post_div) {
                         if (post_div->divider == post_divider)
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 12bdeab91c86..a670caaee29e 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -149,6 +149,7 @@ struct radeon_tmds_pll {
 #define RADEON_PLL_PREFER_CLOSEST_LOWER (1 << 11)
 #define RADEON_PLL_USE_POST_DIV         (1 << 12)
 #define RADEON_PLL_IS_LCD               (1 << 13)
+#define RADEON_PLL_PREFER_MINM_OVER_MAXP (1 << 14)
 
 struct radeon_pll {
         /* reference frequency */
@@ -208,6 +209,7 @@ enum radeon_connector_table {
         CT_EMAC,
         CT_RN50_POWER,
         CT_MAC_X800,
+        CT_MAC_G5_9600,
 };
 
 enum radeon_dvo_chip {
@@ -510,13 +512,21 @@ extern bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
                                              struct radeon_atom_ss *ss,
                                              int id, u32 clock);
 
-extern void radeon_compute_pll(struct radeon_pll *pll,
-                               uint64_t freq,
-                               uint32_t *dot_clock_p,
-                               uint32_t *fb_div_p,
-                               uint32_t *frac_fb_div_p,
-                               uint32_t *ref_div_p,
-                               uint32_t *post_div_p);
+extern void radeon_compute_pll_legacy(struct radeon_pll *pll,
+                                      uint64_t freq,
+                                      uint32_t *dot_clock_p,
+                                      uint32_t *fb_div_p,
+                                      uint32_t *frac_fb_div_p,
+                                      uint32_t *ref_div_p,
+                                      uint32_t *post_div_p);
+
+extern void radeon_compute_pll_avivo(struct radeon_pll *pll,
+                                     u32 freq,
+                                     u32 *dot_clock_p,
+                                     u32 *fb_div_p,
+                                     u32 *frac_fb_div_p,
+                                     u32 *ref_div_p,
+                                     u32 *post_div_p);
 
 extern void radeon_setup_encoder_clones(struct drm_device *dev);
 
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 3b1b2bf9cdd5..2aed03bde4b2 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -430,7 +430,7 @@ static ssize_t radeon_hwmon_show_temp(struct device *dev,
 {
         struct drm_device *ddev = pci_get_drvdata(to_pci_dev(dev));
         struct radeon_device *rdev = ddev->dev_private;
-        u32 temp;
+        int temp;
 
         switch (rdev->pm.int_thermal_type) {
         case THERMAL_TYPE_RV6XX:
@@ -646,6 +646,9 @@ void radeon_pm_fini(struct radeon_device *rdev)
 #endif
         }
 
+        if (rdev->pm.power_state)
+                kfree(rdev->pm.power_state);
+
         radeon_hwmon_fini(rdev);
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_reg.h b/drivers/gpu/drm/radeon/radeon_reg.h
index 3cd4dace57c7..ec93a75369e6 100644
--- a/drivers/gpu/drm/radeon/radeon_reg.h
+++ b/drivers/gpu/drm/radeon/radeon_reg.h
@@ -375,6 +375,8 @@
 #define RADEON_CONFIG_APER_SIZE             0x0108
 #define RADEON_CONFIG_BONDS                 0x00e8
 #define RADEON_CONFIG_CNTL                  0x00e0
+#       define RADEON_CFG_VGA_RAM_EN        (1 << 8)
+#       define RADEON_CFG_VGA_IO_DIS        (1 << 9)
 #       define RADEON_CFG_ATI_REV_A11       (0   << 16)
 #       define RADEON_CFG_ATI_REV_A12       (1   << 16)
 #       define RADEON_CFG_ATI_REV_A13       (2   << 16)
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 1272e4b6a1d4..e5b2cf10cbf4 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -787,9 +787,9 @@ static int radeon_ttm_debugfs_init(struct radeon_device *rdev)
                 radeon_mem_types_list[i].show = &radeon_mm_dump_table;
                 radeon_mem_types_list[i].driver_features = 0;
                 if (i == 0)
-                        radeon_mem_types_list[i].data = &rdev->mman.bdev.man[TTM_PL_VRAM].priv;
+                        radeon_mem_types_list[i].data = rdev->mman.bdev.man[TTM_PL_VRAM].priv;
                 else
-                        radeon_mem_types_list[i].data = &rdev->mman.bdev.man[TTM_PL_TT].priv;
+                        radeon_mem_types_list[i].data = rdev->mman.bdev.man[TTM_PL_TT].priv;
 
         }
         /* Add ttm page pool to debugfs */
diff --git a/drivers/gpu/drm/radeon/reg_srcs/r300 b/drivers/gpu/drm/radeon/reg_srcs/r300
index b506ec1cab4b..e8a1786b6426 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/r300
+++ b/drivers/gpu/drm/radeon/reg_srcs/r300
@@ -683,9 +683,7 @@ r300 0x4f60
 0x4DF4 US_ALU_CONST_G_31
 0x4DF8 US_ALU_CONST_B_31
 0x4DFC US_ALU_CONST_A_31
-0x4E04 RB3D_BLENDCNTL_R3
 0x4E08 RB3D_ABLENDCNTL_R3
-0x4E0C RB3D_COLOR_CHANNEL_MASK
 0x4E10 RB3D_CONSTANT_COLOR
 0x4E14 RB3D_COLOR_CLEAR_VALUE
 0x4E18 RB3D_ROPCNTL_R3
@@ -706,13 +704,11 @@ r300 0x4f60
 0x4E74 RB3D_CMASK_WRINDEX
 0x4E78 RB3D_CMASK_DWORD
 0x4E7C RB3D_CMASK_RDINDEX
-0x4E80 RB3D_AARESOLVE_OFFSET
-0x4E84 RB3D_AARESOLVE_PITCH
-0x4E88 RB3D_AARESOLVE_CTL
 0x4EA0 RB3D_DISCARD_SRC_PIXEL_LTE_THRESHOLD
 0x4EA4 RB3D_DISCARD_SRC_PIXEL_GTE_THRESHOLD
 0x4F04 ZB_ZSTENCILCNTL
 0x4F08 ZB_STENCILREFMASK
 0x4F14 ZB_ZTOP
 0x4F18 ZB_ZCACHE_CTLSTAT
+0x4F28 ZB_DEPTHCLEARVALUE
 0x4F58 ZB_ZPASS_DATA
diff --git a/drivers/gpu/drm/radeon/reg_srcs/r420 b/drivers/gpu/drm/radeon/reg_srcs/r420
index 8c1214c2390f..722074e21e2f 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/r420
+++ b/drivers/gpu/drm/radeon/reg_srcs/r420
@@ -130,7 +130,6 @@ r420 0x4f60
 0x401C GB_SELECT
 0x4020 GB_AA_CONFIG
 0x4024 GB_FIFO_SIZE
-0x4028 GB_Z_PEQ_CONFIG
 0x4100 TX_INVALTAGS
 0x4200 GA_POINT_S0
 0x4204 GA_POINT_T0
@@ -750,9 +749,7 @@ r420 0x4f60
 0x4DF4 US_ALU_CONST_G_31
 0x4DF8 US_ALU_CONST_B_31
 0x4DFC US_ALU_CONST_A_31
-0x4E04 RB3D_BLENDCNTL_R3
 0x4E08 RB3D_ABLENDCNTL_R3
-0x4E0C RB3D_COLOR_CHANNEL_MASK
 0x4E10 RB3D_CONSTANT_COLOR
 0x4E14 RB3D_COLOR_CLEAR_VALUE
 0x4E18 RB3D_ROPCNTL_R3
@@ -773,13 +770,11 @@ r420 0x4f60
 0x4E74 RB3D_CMASK_WRINDEX
 0x4E78 RB3D_CMASK_DWORD
 0x4E7C RB3D_CMASK_RDINDEX
-0x4E80 RB3D_AARESOLVE_OFFSET
-0x4E84 RB3D_AARESOLVE_PITCH
-0x4E88 RB3D_AARESOLVE_CTL
 0x4EA0 RB3D_DISCARD_SRC_PIXEL_LTE_THRESHOLD
 0x4EA4 RB3D_DISCARD_SRC_PIXEL_GTE_THRESHOLD
 0x4F04 ZB_ZSTENCILCNTL
 0x4F08 ZB_STENCILREFMASK
 0x4F14 ZB_ZTOP
 0x4F18 ZB_ZCACHE_CTLSTAT
+0x4F28 ZB_DEPTHCLEARVALUE
 0x4F58 ZB_ZPASS_DATA
diff --git a/drivers/gpu/drm/radeon/reg_srcs/rs600 b/drivers/gpu/drm/radeon/reg_srcs/rs600
index 0828d80396f2..d9f62866bbc1 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/rs600
+++ b/drivers/gpu/drm/radeon/reg_srcs/rs600
@@ -749,9 +749,7 @@ rs600 0x6d40
 0x4DF4 US_ALU_CONST_G_31
 0x4DF8 US_ALU_CONST_B_31
 0x4DFC US_ALU_CONST_A_31
-0x4E04 RB3D_BLENDCNTL_R3
 0x4E08 RB3D_ABLENDCNTL_R3
-0x4E0C RB3D_COLOR_CHANNEL_MASK
 0x4E10 RB3D_CONSTANT_COLOR
 0x4E14 RB3D_COLOR_CLEAR_VALUE
 0x4E18 RB3D_ROPCNTL_R3
@@ -772,13 +770,11 @@ rs600 0x6d40
 0x4E74 RB3D_CMASK_WRINDEX
 0x4E78 RB3D_CMASK_DWORD
 0x4E7C RB3D_CMASK_RDINDEX
-0x4E80 RB3D_AARESOLVE_OFFSET
-0x4E84 RB3D_AARESOLVE_PITCH
-0x4E88 RB3D_AARESOLVE_CTL
 0x4EA0 RB3D_DISCARD_SRC_PIXEL_LTE_THRESHOLD
 0x4EA4 RB3D_DISCARD_SRC_PIXEL_GTE_THRESHOLD
 0x4F04 ZB_ZSTENCILCNTL
 0x4F08 ZB_STENCILREFMASK
 0x4F14 ZB_ZTOP
 0x4F18 ZB_ZCACHE_CTLSTAT
+0x4F28 ZB_DEPTHCLEARVALUE
 0x4F58 ZB_ZPASS_DATA
diff --git a/drivers/gpu/drm/radeon/reg_srcs/rv515 b/drivers/gpu/drm/radeon/reg_srcs/rv515
index ef422bbacfc1..911a8fbd32bb 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/rv515
+++ b/drivers/gpu/drm/radeon/reg_srcs/rv515
@@ -164,7 +164,6 @@ rv515 0x6d40
 0x401C GB_SELECT
 0x4020 GB_AA_CONFIG
 0x4024 GB_FIFO_SIZE
-0x4028 GB_Z_PEQ_CONFIG
 0x4100 TX_INVALTAGS
 0x4114 SU_TEX_WRAP_PS3
 0x4118 PS3_ENABLE
@@ -461,9 +460,7 @@ rv515 0x6d40
 0x4DF4 US_ALU_CONST_G_31
 0x4DF8 US_ALU_CONST_B_31
 0x4DFC US_ALU_CONST_A_31
-0x4E04 RB3D_BLENDCNTL_R3
 0x4E08 RB3D_ABLENDCNTL_R3
-0x4E0C RB3D_COLOR_CHANNEL_MASK
 0x4E10 RB3D_CONSTANT_COLOR
 0x4E14 RB3D_COLOR_CLEAR_VALUE
 0x4E18 RB3D_ROPCNTL_R3
@@ -484,9 +481,6 @@ rv515 0x6d40
 0x4E74 RB3D_CMASK_WRINDEX
 0x4E78 RB3D_CMASK_DWORD
 0x4E7C RB3D_CMASK_RDINDEX
-0x4E80 RB3D_AARESOLVE_OFFSET
-0x4E84 RB3D_AARESOLVE_PITCH
-0x4E88 RB3D_AARESOLVE_CTL
 0x4EA0 RB3D_DISCARD_SRC_PIXEL_LTE_THRESHOLD
 0x4EA4 RB3D_DISCARD_SRC_PIXEL_GTE_THRESHOLD
 0x4EF8 RB3D_CONSTANT_COLOR_AR
@@ -496,4 +490,5 @@ rv515 0x6d40
 0x4F14 ZB_ZTOP
 0x4F18 ZB_ZCACHE_CTLSTAT
 0x4F58 ZB_ZPASS_DATA
+0x4F28 ZB_DEPTHCLEARVALUE
 0x4FD4 ZB_STENCILREFMASK_BF
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index 5512e4e5e636..c76283d9eb3d 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -203,6 +203,9 @@ void rs400_gart_fini(struct radeon_device *rdev)
         radeon_gart_table_ram_free(rdev);
 }
 
+#define RS400_PTE_WRITEABLE (1 << 2)
+#define RS400_PTE_READABLE  (1 << 3)
+
 int rs400_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
 {
         uint32_t entry;
@@ -213,7 +216,7 @@ int rs400_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
 
         entry = (lower_32_bits(addr) & PAGE_MASK) |
                 ((upper_32_bits(addr) & 0xff) << 4) |
-                0xc;
+                RS400_PTE_WRITEABLE | RS400_PTE_READABLE;
         entry = cpu_to_le32(entry);
         rdev->gart.table.ram.ptr[i] = entry;
         return 0;
@@ -226,8 +229,8 @@ int rs400_mc_wait_for_idle(struct radeon_device *rdev)
 
         for (i = 0; i < rdev->usec_timeout; i++) {
                 /* read MC_STATUS */
-                tmp = RREG32(0x0150);
-                if (tmp & (1 << 2)) {
+                tmp = RREG32(RADEON_MC_STATUS);
+                if (tmp & RADEON_MC_IDLE) {
                         return 0;
                 }
                 DRM_UDELAY(1);
@@ -241,7 +244,7 @@ void rs400_gpu_init(struct radeon_device *rdev)
         r420_pipes_init(rdev);
         if (rs400_mc_wait_for_idle(rdev)) {
                 printk(KERN_WARNING "rs400: Failed to wait MC idle while "
-                       "programming pipes. Bad things might happen. %08x\n", RREG32(0x150));
+                       "programming pipes. Bad things might happen. %08x\n", RREG32(RADEON_MC_STATUS));
         }
 }
 
@@ -300,9 +303,9 @@ static int rs400_debugfs_gart_info(struct seq_file *m, void *data)
                 seq_printf(m, "MCCFG_AGP_BASE_2 0x%08x\n", tmp);
                 tmp = RREG32_MC(RS690_MCCFG_AGP_LOCATION);
                 seq_printf(m, "MCCFG_AGP_LOCATION 0x%08x\n", tmp);
-                tmp = RREG32_MC(0x100);
+                tmp = RREG32_MC(RS690_MCCFG_FB_LOCATION);
                 seq_printf(m, "MCCFG_FB_LOCATION 0x%08x\n", tmp);
-                tmp = RREG32(0x134);
+                tmp = RREG32(RS690_HDP_FB_LOCATION);
                 seq_printf(m, "HDP_FB_LOCATION 0x%08x\n", tmp);
         } else {
                 tmp = RREG32(RADEON_AGP_BASE);
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 0137d3e3728d..6638c8e4c81b 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -77,9 +77,9 @@ void rs690_pm_info(struct radeon_device *rdev)
                 switch (crev) {
                 case 1:
                         tmp.full = dfixed_const(100);
-                        rdev->pm.igp_sideport_mclk.full = dfixed_const(info->info.ulBootUpMemoryClock);
+                        rdev->pm.igp_sideport_mclk.full = dfixed_const(le32_to_cpu(info->info.ulBootUpMemoryClock));
                         rdev->pm.igp_sideport_mclk.full = dfixed_div(rdev->pm.igp_sideport_mclk, tmp);
-                        if (info->info.usK8MemoryClock)
+                        if (le16_to_cpu(info->info.usK8MemoryClock))
                                 rdev->pm.igp_system_mclk.full = dfixed_const(le16_to_cpu(info->info.usK8MemoryClock));
                         else if (rdev->clock.default_mclk) {
                                 rdev->pm.igp_system_mclk.full = dfixed_const(rdev->clock.default_mclk);
@@ -91,16 +91,16 @@ void rs690_pm_info(struct radeon_device *rdev)
                         break;
                 case 2:
                         tmp.full = dfixed_const(100);
-                        rdev->pm.igp_sideport_mclk.full = dfixed_const(info->info_v2.ulBootUpSidePortClock);
+                        rdev->pm.igp_sideport_mclk.full = dfixed_const(le32_to_cpu(info->info_v2.ulBootUpSidePortClock));
                         rdev->pm.igp_sideport_mclk.full = dfixed_div(rdev->pm.igp_sideport_mclk, tmp);
-                        if (info->info_v2.ulBootUpUMAClock)
-                                rdev->pm.igp_system_mclk.full = dfixed_const(info->info_v2.ulBootUpUMAClock);
+                        if (le32_to_cpu(info->info_v2.ulBootUpUMAClock))
+                                rdev->pm.igp_system_mclk.full = dfixed_const(le32_to_cpu(info->info_v2.ulBootUpUMAClock));
                         else if (rdev->clock.default_mclk)
                                 rdev->pm.igp_system_mclk.full = dfixed_const(rdev->clock.default_mclk);
                         else
                                 rdev->pm.igp_system_mclk.full = dfixed_const(66700);
                         rdev->pm.igp_system_mclk.full = dfixed_div(rdev->pm.igp_system_mclk, tmp);
-                        rdev->pm.igp_ht_link_clk.full = dfixed_const(info->info_v2.ulHTLinkFreq);
+                        rdev->pm.igp_ht_link_clk.full = dfixed_const(le32_to_cpu(info->info_v2.ulHTLinkFreq));
                         rdev->pm.igp_ht_link_clk.full = dfixed_div(rdev->pm.igp_ht_link_clk, tmp);
                         rdev->pm.igp_ht_link_width.full = dfixed_const(le16_to_cpu(info->info_v2.usMinHTLinkWidth));
                         break;
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 5d569f41f4ae..64b57af93714 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -69,13 +69,13 @@ void rv515_ring_start(struct radeon_device *rdev)
                           ISYNC_CPSCRATCH_IDLEGUI);
         radeon_ring_write(rdev, PACKET0(WAIT_UNTIL, 0));
         radeon_ring_write(rdev, WAIT_2D_IDLECLEAN | WAIT_3D_IDLECLEAN);
-        radeon_ring_write(rdev, PACKET0(0x170C, 0));
-        radeon_ring_write(rdev, 1 << 31);
+        radeon_ring_write(rdev, PACKET0(R300_DST_PIPE_CONFIG, 0));
+        radeon_ring_write(rdev, R300_PIPE_AUTO_CONFIG);
         radeon_ring_write(rdev, PACKET0(GB_SELECT, 0));
         radeon_ring_write(rdev, 0);
         radeon_ring_write(rdev, PACKET0(GB_ENABLE, 0));
         radeon_ring_write(rdev, 0);
-        radeon_ring_write(rdev, PACKET0(0x42C8, 0));
+        radeon_ring_write(rdev, PACKET0(R500_SU_REG_DEST, 0));
         radeon_ring_write(rdev, (1 << rdev->num_gb_pipes) - 1);
         radeon_ring_write(rdev, PACKET0(VAP_INDEX_OFFSET, 0));
         radeon_ring_write(rdev, 0);
@@ -153,8 +153,8 @@ void rv515_gpu_init(struct radeon_device *rdev)
         }
         rv515_vga_render_disable(rdev);
         r420_pipes_init(rdev);
-        gb_pipe_select = RREG32(0x402C);
-        tmp = RREG32(0x170C);
+        gb_pipe_select = RREG32(R400_GB_PIPE_SELECT);
+        tmp = RREG32(R300_DST_PIPE_CONFIG);
         pipe_select_current = (tmp >> 2) & 3;
         tmp = (1 << pipe_select_current) |
               (((gb_pipe_select >> 8) & 0xF) << 4);
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 491dc9000655..d8ba67690656 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -78,18 +78,23 @@ u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 }
 
 /* get temperature in millidegrees */
-u32 rv770_get_temp(struct radeon_device *rdev)
+int rv770_get_temp(struct radeon_device *rdev)
 {
         u32 temp = (RREG32(CG_MULT_THERMAL_STATUS) & ASIC_T_MASK) >>
                 ASIC_T_SHIFT;
-        u32 actual_temp = 0;
-
-        if ((temp >> 9) & 1)
-                actual_temp = 0;
-        else
-                actual_temp = (temp >> 1) & 0xff;
-
-        return actual_temp * 1000;
+        int actual_temp;
+
+        if (temp & 0x400)
+                actual_temp = -256;
+        else if (temp & 0x200)
+                actual_temp = 255;
+        else if (temp & 0x100) {
+                actual_temp = temp & 0x1ff;
+                actual_temp |= ~0x1ff;
+        } else
+                actual_temp = temp & 0xff;
+
+        return (actual_temp * 1000) / 2;
 }
 
 void rv770_pm_misc(struct radeon_device *rdev)
@@ -316,7 +321,11 @@ static int rv770_cp_load_microcode(struct radeon_device *rdev)
                 return -EINVAL;
 
         r700_cp_stop(rdev);
-        WREG32(CP_RB_CNTL, RB_NO_UPDATE | (15 << 8) | (3 << 0));
+        WREG32(CP_RB_CNTL,
+#ifdef __BIG_ENDIAN
+               BUF_SWAP_32BIT |
+#endif
+               RB_NO_UPDATE | RB_BLKSZ(15) | RB_BUFSZ(3));
 
         /* Reset cp */
         WREG32(GRBM_SOFT_RESET, SOFT_RESET_CP);
diff --git a/drivers/gpu/drm/radeon/rv770d.h b/drivers/gpu/drm/radeon/rv770d.h
index abc8cf5a3672..79fa588e9ed5 100644
--- a/drivers/gpu/drm/radeon/rv770d.h
+++ b/drivers/gpu/drm/radeon/rv770d.h
@@ -76,10 +76,10 @@
 #define         ROQ_IB1_START(x)                                ((x) << 0)
 #define         ROQ_IB2_START(x)                                ((x) << 8)
 #define CP_RB_CNTL                                      0xC104
-#define         RB_BUFSZ(x)                                     ((x)<<0)
-#define         RB_BLKSZ(x)                                     ((x)<<8)
-#define         RB_NO_UPDATE                                    (1<<27)
-#define         RB_RPTR_WR_ENA                                  (1<<31)
+#define         RB_BUFSZ(x)                                     ((x) << 0)
+#define         RB_BLKSZ(x)                                     ((x) << 8)
+#define         RB_NO_UPDATE                                    (1 << 27)
+#define         RB_RPTR_WR_ENA                                  (1 << 31)
 #define         BUF_SWAP_32BIT                                  (2 << 16)
 #define CP_RB_RPTR                                      0x8700
 #define CP_RB_RPTR_ADDR                                 0xC10C
diff --git a/drivers/gpu/stub/Kconfig b/drivers/gpu/stub/Kconfig
index 09aea5f1556d..70e60a4bb678 100644
--- a/drivers/gpu/stub/Kconfig
+++ b/drivers/gpu/stub/Kconfig
@@ -1,11 +1,13 @@
 config STUB_POULSBO
         tristate "Intel GMA500 Stub Driver"
         depends on PCI
+        depends on NET # for THERMAL
         # Poulsbo stub depends on ACPI_VIDEO when ACPI is enabled
         # but for select to work, need to select ACPI_VIDEO's dependencies, ick
         select BACKLIGHT_CLASS_DEVICE if ACPI
         select INPUT if ACPI
         select ACPI_VIDEO if ACPI
+        select THERMAL if ACPI
         help
           Choose this option if you have a system that has Intel GMA500
           (Poulsbo) integrated graphics. If M is selected, the module will