74 files changed, 981 insertions, 440 deletions

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 6977a1ce9d98..f73ef4390db6 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -672,7 +672,7 @@ int drm_fb_helper_setcmap(struct fb_cmap *cmap, struct fb_info *info)
         struct drm_crtc_helper_funcs *crtc_funcs;
         u16 *red, *green, *blue, *transp;
         struct drm_crtc *crtc;
-        int i, rc = 0;
+        int i, j, rc = 0;
         int start;
 
         for (i = 0; i < fb_helper->crtc_count; i++) {
@@ -685,7 +685,7 @@ int drm_fb_helper_setcmap(struct fb_cmap *cmap, struct fb_info *info)
                 transp = cmap->transp;
                 start = cmap->start;
 
-                for (i = 0; i < cmap->len; i++) {
+                for (j = 0; j < cmap->len; j++) {
                         u16 hred, hgreen, hblue, htransp = 0xffff;
 
                         hred = *red++;
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 3dadfa2a8528..28d1d3c24d65 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -164,8 +164,10 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
          * available. In that case we can't account for this and just
          * hope for the best.
          */
-        if ((vblrc > 0) && (abs(diff_ns) > 1000000))
+        if ((vblrc > 0) && (abs64(diff_ns) > 1000000)) {
                 atomic_inc(&dev->_vblank_count[crtc]);
+                smp_mb__after_atomic_inc();
+        }
 
         /* Invalidate all timestamps while vblank irq's are off. */
         clear_vblank_timestamps(dev, crtc);
@@ -491,6 +493,12 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc)
         /* Dot clock in Hz: */
         dotclock = (u64) crtc->hwmode.clock * 1000;
 
+        /* Fields of interlaced scanout modes are only halve a frame duration.
+         * Double the dotclock to get halve the frame-/line-/pixelduration.
+         */
+        if (crtc->hwmode.flags & DRM_MODE_FLAG_INTERLACE)
+                dotclock *= 2;
+
         /* Valid dotclock? */
         if (dotclock > 0) {
                 /* Convert scanline length in pixels and video dot clock to
@@ -603,14 +611,6 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, int crtc,
                 return -EAGAIN;
         }
 
-        /* Don't know yet how to handle interlaced or
-         * double scan modes. Just no-op for now.
-         */
-        if (mode->flags & (DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLSCAN)) {
-                DRM_DEBUG("crtc %d: Noop due to unsupported mode.\n", crtc);
-                return -ENOTSUPP;
-        }
-
         /* Get current scanout position with system timestamp.
          * Repeat query up to DRM_TIMESTAMP_MAXRETRIES times
          * if single query takes longer than max_error nanoseconds.
@@ -858,10 +858,11 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
         if (rc) {
                 tslot = atomic_read(&dev->_vblank_count[crtc]) + diff;
                 vblanktimestamp(dev, crtc, tslot) = t_vblank;
-                smp_wmb();
         }
 
+        smp_mb__before_atomic_inc();
         atomic_add(diff, &dev->_vblank_count[crtc]);
+        smp_mb__after_atomic_inc();
 }
 
 /**
@@ -1011,7 +1012,8 @@ int drm_modeset_ctl(struct drm_device *dev, void *data,
                     struct drm_file *file_priv)
 {
         struct drm_modeset_ctl *modeset = data;
-        int crtc, ret = 0;
+        int ret = 0;
+        unsigned int crtc;
 
         /* If drm_vblank_init() hasn't been called yet, just no-op */
         if (!dev->num_crtcs)
@@ -1293,15 +1295,16 @@ bool drm_handle_vblank(struct drm_device *dev, int crtc)
          * e.g., due to spurious vblank interrupts. We need to
          * ignore those for accounting.
          */
-        if (abs(diff_ns) > DRM_REDUNDANT_VBLIRQ_THRESH_NS) {
+        if (abs64(diff_ns) > DRM_REDUNDANT_VBLIRQ_THRESH_NS) {
                 /* Store new timestamp in ringbuffer. */
                 vblanktimestamp(dev, crtc, vblcount + 1) = tvblank;
-                smp_wmb();
 
                 /* Increment cooked vblank count. This also atomically commits
                  * the timestamp computed above.
                  */
+                smp_mb__before_atomic_inc();
                 atomic_inc(&dev->_vblank_count[crtc]);
+                smp_mb__after_atomic_inc();
         } else {
                 DRM_DEBUG("crtc %d: Redundant vblirq ignored. diff_ns = %d\n",
                           crtc, (int) diff_ns);
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 3601466c5502..4ff9b6cc973f 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -865,7 +865,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
                 int max_freq;
 
                 /* RPSTAT1 is in the GT power well */
-                __gen6_force_wake_get(dev_priv);
+                __gen6_gt_force_wake_get(dev_priv);
 
                 seq_printf(m, "GT_PERF_STATUS: 0x%08x\n", gt_perf_status);
                 seq_printf(m, "RPSTAT1: 0x%08x\n", I915_READ(GEN6_RPSTAT1));
@@ -888,7 +888,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
                 seq_printf(m, "Max non-overclocked (RP0) frequency: %dMHz\n",
                            max_freq * 100);
 
-                __gen6_force_wake_put(dev_priv);
+                __gen6_gt_force_wake_put(dev_priv);
         } else {
                 seq_printf(m, "no P-state info available\n");
         }
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 17bd766f2081..e33d9be7df3b 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1895,6 +1895,17 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
         if (IS_GEN2(dev))
                 dma_set_coherent_mask(&dev->pdev->dev, DMA_BIT_MASK(30));
 
+        /* 965GM sometimes incorrectly writes to hardware status page (HWS)
+         * using 32bit addressing, overwriting memory if HWS is located
+         * above 4GB.
+         *
+         * The documentation also mentions an issue with undefined
+         * behaviour if any general state is accessed within a page above 4GB,
+         * which also needs to be handled carefully.
+         */
+        if (IS_BROADWATER(dev) || IS_CRESTLINE(dev))
+                dma_set_coherent_mask(&dev->pdev->dev, DMA_BIT_MASK(32));
+
         mmio_bar = IS_GEN2(dev) ? 1 : 0;
         dev_priv->regs = pci_iomap(dev->pdev, mmio_bar, 0);
         if (!dev_priv->regs) {
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index cfb56d0ff367..22ec066adae6 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -46,6 +46,12 @@ module_param_named(fbpercrtc, i915_fbpercrtc, int, 0400);
 unsigned int i915_powersave = 1;
 module_param_named(powersave, i915_powersave, int, 0600);
 
+unsigned int i915_semaphores = 0;
+module_param_named(semaphores, i915_semaphores, 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);
 
@@ -251,7 +257,7 @@ void intel_detect_pch (struct drm_device *dev)
         }
 }
 
-void __gen6_force_wake_get(struct drm_i915_private *dev_priv)
+void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
 {
         int count;
 
@@ -267,12 +273,22 @@ void __gen6_force_wake_get(struct drm_i915_private *dev_priv)
                 udelay(10);
 }
 
-void __gen6_force_wake_put(struct drm_i915_private *dev_priv)
+void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
 {
         I915_WRITE_NOTRACE(FORCEWAKE, 0);
         POSTING_READ(FORCEWAKE);
 }
 
+void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
+{
+        int loop = 500;
+        u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+        while (fifo < 20 && loop--) {
+                udelay(10);
+                fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+        }
+}
+
 static int i915_drm_freeze(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
@@ -360,7 +376,7 @@ static int i915_drm_thaw(struct drm_device *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);
         }
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index a0149c619cdd..456f40484838 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -956,8 +956,10 @@ extern struct drm_ioctl_desc i915_ioctls[];
 extern int i915_max_ioctl;
 extern unsigned int i915_fbpercrtc;
 extern unsigned int i915_powersave;
+extern unsigned int i915_semaphores;
 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);
@@ -1176,6 +1178,9 @@ void i915_gem_detach_phys_object(struct drm_device *dev,
 void i915_gem_free_all_phys_object(struct drm_device *dev);
 void i915_gem_release(struct drm_device *dev, struct drm_file *file);
 
+uint32_t
+i915_gem_get_unfenced_gtt_alignment(struct drm_i915_gem_object *obj);
+
 /* i915_gem_gtt.c */
 void i915_gem_restore_gtt_mappings(struct drm_device *dev);
 int __must_check i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj);
@@ -1352,22 +1357,32 @@ __i915_write(64, q)
  * must be set to prevent GT core from power down and stale values being
  * returned.
  */
-void __gen6_force_wake_get(struct drm_i915_private *dev_priv);
-void __gen6_force_wake_put (struct drm_i915_private *dev_priv);
-static inline u32 i915_safe_read(struct drm_i915_private *dev_priv, u32 reg)
+void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv);
+void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv);
+void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv);
+
+static inline u32 i915_gt_read(struct drm_i915_private *dev_priv, u32 reg)
 {
         u32 val;
 
         if (dev_priv->info->gen >= 6) {
-                __gen6_force_wake_get(dev_priv);
+                __gen6_gt_force_wake_get(dev_priv);
                 val = I915_READ(reg);
-                __gen6_force_wake_put(dev_priv);
+                __gen6_gt_force_wake_put(dev_priv);
         } else
                 val = I915_READ(reg);
 
         return val;
 }
 
+static inline void i915_gt_write(struct drm_i915_private *dev_priv,
+                                u32 reg, u32 val)
+{
+        if (dev_priv->info->gen >= 6)
+                __gen6_gt_wait_for_fifo(dev_priv);
+        I915_WRITE(reg, val);
+}
+
 static inline void
 i915_write(struct drm_i915_private *dev_priv, u32 reg, u64 val, int len)
 {
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index cf4f74c7c6fb..36e66cc5225e 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1398,7 +1398,7 @@ i915_gem_get_gtt_alignment(struct drm_i915_gem_object *obj)
  * Return the required GTT alignment for an object, only taking into account
  * unfenced tiled surface requirements.
  */
-static uint32_t
+uint32_t
 i915_gem_get_unfenced_gtt_alignment(struct drm_i915_gem_object *obj)
 {
         struct drm_device *dev = obj->base.dev;
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index d2f445e825f2..50ab1614571c 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -772,8 +772,8 @@ i915_gem_execbuffer_sync_rings(struct drm_i915_gem_object *obj,
         if (from == NULL || to == from)
                 return 0;
 
-        /* XXX gpu semaphores are currently causing hard hangs on SNB mobile */
-        if (INTEL_INFO(obj->base.dev)->gen < 6 || IS_MOBILE(obj->base.dev))
+        /* XXX gpu semaphores are implicated in various hard hangs on SNB */
+        if (INTEL_INFO(obj->base.dev)->gen < 6 || !i915_semaphores)
                 return i915_gem_object_wait_rendering(obj, true);
 
         idx = intel_ring_sync_index(from, to);
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 22a32b9932c5..d64843e18df2 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -349,14 +349,27 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
                         (obj->gtt_offset + obj->base.size <= dev_priv->mm.gtt_mappable_end &&
                          i915_gem_object_fence_ok(obj, args->tiling_mode));
 
-                obj->tiling_changed = true;
-                obj->tiling_mode = args->tiling_mode;
-                obj->stride = args->stride;
+                /* Rebind if we need a change of alignment */
+                if (!obj->map_and_fenceable) {
+                        u32 unfenced_alignment =
+                                i915_gem_get_unfenced_gtt_alignment(obj);
+                        if (obj->gtt_offset & (unfenced_alignment - 1))
+                                ret = i915_gem_object_unbind(obj);
+                }
+
+                if (ret == 0) {
+                        obj->tiling_changed = true;
+                        obj->tiling_mode = args->tiling_mode;
+                        obj->stride = args->stride;
+                }
         }
+        /* we have to maintain this existing ABI... */
+        args->stride = obj->stride;
+        args->tiling_mode = obj->tiling_mode;
         drm_gem_object_unreference(&obj->base);
         mutex_unlock(&dev->struct_mutex);
 
-        return 0;
+        return ret;
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 97f946dcc1aa..8a9e08bf1cf7 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -316,6 +316,8 @@ static void i915_hotplug_work_func(struct work_struct *work)
         struct drm_mode_config *mode_config = &dev->mode_config;
         struct intel_encoder *encoder;
 
+        DRM_DEBUG_KMS("running encoder hotplug functions\n");
+
         list_for_each_entry(encoder, &mode_config->encoder_list, base.head)
                 if (encoder->hot_plug)
                         encoder->hot_plug(encoder);
@@ -1649,9 +1651,7 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
         } else {
                 hotplug_mask = SDE_CRT_HOTPLUG | SDE_PORTB_HOTPLUG |
                                SDE_PORTC_HOTPLUG | SDE_PORTD_HOTPLUG;
-                hotplug_mask |= SDE_AUX_MASK | SDE_FDI_MASK | SDE_TRANS_MASK;
-                I915_WRITE(FDI_RXA_IMR, 0);
-                I915_WRITE(FDI_RXB_IMR, 0);
+                hotplug_mask |= SDE_AUX_MASK;
         }
 
         dev_priv->pch_irq_mask = ~hotplug_mask;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5cfc68940f17..2abe240dae58 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)
@@ -3269,6 +3271,8 @@
 #define  FORCEWAKE                              0xA18C
 #define  FORCEWAKE_ACK                          0x130090
 
+#define  GT_FIFO_FREE_ENTRIES                   0x120008
+
 #define GEN6_RPNSWREQ                           0xA008
 #define   GEN6_TURBO_DISABLE                    (1<<31)
 #define   GEN6_FREQUENCY(x)                     ((x)<<25)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 7e42aa586504..49fb54fd9a18 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1219,7 +1219,7 @@ static void sandybridge_blit_fbc_update(struct drm_device *dev)
         u32 blt_ecoskpd;
 
         /* Make sure blitter notifies FBC of writes */
-        __gen6_force_wake_get(dev_priv);
+        __gen6_gt_force_wake_get(dev_priv);
         blt_ecoskpd = I915_READ(GEN6_BLITTER_ECOSKPD);
         blt_ecoskpd |= GEN6_BLITTER_FBC_NOTIFY <<
                 GEN6_BLITTER_LOCK_SHIFT;
@@ -1230,7 +1230,7 @@ static void sandybridge_blit_fbc_update(struct drm_device *dev)
                          GEN6_BLITTER_LOCK_SHIFT);
         I915_WRITE(GEN6_BLITTER_ECOSKPD, blt_ecoskpd);
         POSTING_READ(GEN6_BLITTER_ECOSKPD);
-        __gen6_force_wake_put(dev_priv);
+        __gen6_gt_force_wake_put(dev_priv);
 }
 
 static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
@@ -1630,19 +1630,19 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
                 struct drm_i915_gem_object *obj = to_intel_framebuffer(old_fb)->obj;
 
                 wait_event(dev_priv->pending_flip_queue,
+                           atomic_read(&dev_priv->mm.wedged) ||
                            atomic_read(&obj->pending_flip) == 0);
 
                 /* Big Hammer, we also need to ensure that any pending
                  * MI_WAIT_FOR_EVENT inside a user batch buffer on the
                  * current scanout is retired before unpinning the old
                  * framebuffer.
+                 *
+                 * This should only fail upon a hung GPU, in which case we
+                 * can safely continue.
                  */
                 ret = i915_gem_object_flush_gpu(obj, false);
-                if (ret) {
-                        i915_gem_object_unpin(to_intel_framebuffer(crtc->fb)->obj);
-                        mutex_unlock(&dev->struct_mutex);
-                        return ret;
-                }
+                (void) ret;
         }
 
         ret = intel_pipe_set_base_atomic(crtc, crtc->fb, x, y,
@@ -2045,6 +2045,31 @@ static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
                    atomic_read(&obj->pending_flip) == 0);
 }
 
+static bool intel_crtc_driving_pch(struct drm_crtc *crtc)
+{
+        struct drm_device *dev = crtc->dev;
+        struct drm_mode_config *mode_config = &dev->mode_config;
+        struct intel_encoder *encoder;
+
+        /*
+         * If there's a non-PCH eDP on this crtc, it must be DP_A, and that
+         * must be driven by its own crtc; no sharing is possible.
+         */
+        list_for_each_entry(encoder, &mode_config->encoder_list, base.head) {
+                if (encoder->base.crtc != crtc)
+                        continue;
+
+                switch (encoder->type) {
+                case INTEL_OUTPUT_EDP:
+                        if (!intel_encoder_is_pch_edp(&encoder->base))
+                                return false;
+                        continue;
+                }
+        }
+
+        return true;
+}
+
 static void ironlake_crtc_enable(struct drm_crtc *crtc)
 {
         struct drm_device *dev = crtc->dev;
@@ -2053,6 +2078,7 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
         int pipe = intel_crtc->pipe;
         int plane = intel_crtc->plane;
         u32 reg, temp;
+        bool is_pch_port = false;
 
         if (intel_crtc->active)
                 return;
@@ -2066,7 +2092,56 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
                         I915_WRITE(PCH_LVDS, temp | LVDS_PORT_EN);
         }
 
-        ironlake_fdi_enable(crtc);
+        is_pch_port = intel_crtc_driving_pch(crtc);
+
+        if (is_pch_port)
+                ironlake_fdi_enable(crtc);
+        else {
+                /* disable CPU FDI tx and PCH FDI rx */
+                reg = FDI_TX_CTL(pipe);
+                temp = I915_READ(reg);
+                I915_WRITE(reg, temp & ~FDI_TX_ENABLE);
+                POSTING_READ(reg);
+
+                reg = FDI_RX_CTL(pipe);
+                temp = I915_READ(reg);
+                temp &= ~(0x7 << 16);
+                temp |= (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) << 11;
+                I915_WRITE(reg, temp & ~FDI_RX_ENABLE);
+
+                POSTING_READ(reg);
+                udelay(100);
+
+                /* Ironlake workaround, disable clock pointer after downing FDI */
+                if (HAS_PCH_IBX(dev))
+                        I915_WRITE(FDI_RX_CHICKEN(pipe),
+                                   I915_READ(FDI_RX_CHICKEN(pipe) &
+                                             ~FDI_RX_PHASE_SYNC_POINTER_ENABLE));
+
+                /* still set train pattern 1 */
+                reg = FDI_TX_CTL(pipe);
+                temp = I915_READ(reg);
+                temp &= ~FDI_LINK_TRAIN_NONE;
+                temp |= FDI_LINK_TRAIN_PATTERN_1;
+                I915_WRITE(reg, temp);
+
+                reg = FDI_RX_CTL(pipe);
+                temp = I915_READ(reg);
+                if (HAS_PCH_CPT(dev)) {
+                        temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
+                        temp |= FDI_LINK_TRAIN_PATTERN_1_CPT;
+                } else {
+                        temp &= ~FDI_LINK_TRAIN_NONE;
+                        temp |= FDI_LINK_TRAIN_PATTERN_1;
+                }
+                /* BPC in FDI rx is consistent with that in PIPECONF */
+                temp &= ~(0x07 << 16);
+                temp |= (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) << 11;
+                I915_WRITE(reg, temp);
+
+                POSTING_READ(reg);
+                udelay(100);
+        }
 
         /* Enable panel fitting for LVDS */
         if (dev_priv->pch_pf_size &&
@@ -2100,6 +2175,10 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
                 intel_flush_display_plane(dev, plane);
         }
 
+        /* Skip the PCH stuff if possible */
+        if (!is_pch_port)
+                goto done;
+
         /* For PCH output, training FDI link */
         if (IS_GEN6(dev))
                 gen6_fdi_link_train(crtc);
@@ -2184,7 +2263,7 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
         I915_WRITE(reg, temp | TRANS_ENABLE);
         if (wait_for(I915_READ(reg) & TRANS_STATE_ENABLE, 100))
                 DRM_ERROR("failed to enable transcoder %d\n", pipe);
-
+done:
         intel_crtc_load_lut(crtc);
         intel_update_fbc(dev);
         intel_crtc_update_cursor(crtc, true);
@@ -5558,9 +5637,7 @@ static void intel_crtc_reset(struct drm_crtc *crtc)
         /* Reset flags back to the 'unknown' status so that they
          * will be correctly set on the initial modeset.
          */
-        intel_crtc->cursor_addr = 0;
         intel_crtc->dpms_mode = -1;
-        intel_crtc->active = true; /* force the pipe off on setup_init_config */
 }
 
 static struct drm_crtc_helper_funcs intel_helper_funcs = {
@@ -5666,6 +5743,7 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
         dev_priv->pipe_to_crtc_mapping[intel_crtc->pipe] = &intel_crtc->base;
 
         intel_crtc_reset(&intel_crtc->base);
+        intel_crtc->active = true; /* force the pipe off on setup_init_config */
 
         if (HAS_PCH_SPLIT(dev)) {
                 intel_helper_funcs.prepare = ironlake_crtc_prepare;
@@ -6204,7 +6282,7 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
          * userspace...
          */
         I915_WRITE(GEN6_RC_STATE, 0);
-        __gen6_force_wake_get(dev_priv);
+        __gen6_gt_force_wake_get(dev_priv);
 
         /* disable the counters and set deterministic thresholds */
         I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -6302,7 +6380,7 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
         /* enable all PM interrupts */
         I915_WRITE(GEN6_PMINTRMSK, 0);
 
-        __gen6_force_wake_put(dev_priv);
+        __gen6_gt_force_wake_put(dev_priv);
 }
 
 void intel_enable_clock_gating(struct drm_device *dev)
@@ -6463,52 +6541,60 @@ 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_teardown_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)
@@ -6516,15 +6602,26 @@ 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 |
@@ -6541,6 +6638,7 @@ 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)
 {
@@ -6783,21 +6881,9 @@ 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_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index c65992df458d..f8f86e57df22 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -208,7 +208,6 @@ u32 intel_panel_get_backlight(struct drm_device *dev)
                         val &= ~1;
                         pci_read_config_byte(dev->pdev, PCI_LBPC, &lbpc);
                         val *= lbpc;
-                        val >>= 1;
                 }
         }
 
@@ -235,11 +234,11 @@ void intel_panel_set_backlight(struct drm_device *dev, u32 level)
 
         if (is_backlight_combination_mode(dev)){
                 u32 max = intel_panel_get_max_backlight(dev);
-                u8 lpbc;
+                u8 lbpc;
 
-                lpbc = level * 0xfe / max + 1;
-                level /= lpbc;
-                pci_write_config_byte(dev->pdev, PCI_LBPC, lpbc);
+                lbpc = level * 0xfe / max + 1;
+                level /= lbpc;
+                pci_write_config_byte(dev->pdev, PCI_LBPC, lbpc);
         }
 
         tmp = I915_READ(BLC_PWM_CTL);
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_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 6d6fde85a636..34306865a5df 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -14,22 +14,23 @@ struct  intel_hw_status_page {
         struct          drm_i915_gem_object *obj;
 };
 
-#define I915_RING_READ(reg) i915_safe_read(dev_priv, reg)
+#define I915_RING_READ(reg) i915_gt_read(dev_priv, reg)
+#define I915_RING_WRITE(reg, val) i915_gt_write(dev_priv, reg, val)
 
 #define I915_READ_TAIL(ring) I915_RING_READ(RING_TAIL((ring)->mmio_base))
-#define I915_WRITE_TAIL(ring, val) I915_WRITE(RING_TAIL((ring)->mmio_base), val)
+#define I915_WRITE_TAIL(ring, val) I915_RING_WRITE(RING_TAIL((ring)->mmio_base), val)
 
 #define I915_READ_START(ring) I915_RING_READ(RING_START((ring)->mmio_base))
-#define I915_WRITE_START(ring, val) I915_WRITE(RING_START((ring)->mmio_base), val)
+#define I915_WRITE_START(ring, val) I915_RING_WRITE(RING_START((ring)->mmio_base), val)
 
 #define I915_READ_HEAD(ring)  I915_RING_READ(RING_HEAD((ring)->mmio_base))
-#define I915_WRITE_HEAD(ring, val) I915_WRITE(RING_HEAD((ring)->mmio_base), val)
+#define I915_WRITE_HEAD(ring, val) I915_RING_WRITE(RING_HEAD((ring)->mmio_base), val)
 
 #define I915_READ_CTL(ring) I915_RING_READ(RING_CTL((ring)->mmio_base))
-#define I915_WRITE_CTL(ring, val) I915_WRITE(RING_CTL((ring)->mmio_base), val)
+#define I915_WRITE_CTL(ring, val) I915_RING_WRITE(RING_CTL((ring)->mmio_base), val)
 
-#define I915_WRITE_IMR(ring, val) I915_WRITE(RING_IMR((ring)->mmio_base), val)
 #define I915_READ_IMR(ring) I915_RING_READ(RING_IMR((ring)->mmio_base))
+#define I915_WRITE_IMR(ring, val) I915_RING_WRITE(RING_IMR((ring)->mmio_base), val)
 
 #define I915_READ_NOPID(ring) I915_RING_READ(RING_NOPID((ring)->mmio_base))
 #define I915_READ_SYNC_0(ring) I915_RING_READ(RING_SYNC_0((ring)->mmio_base))
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 6a09c1413d60..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))
 
@@ -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 49e5e99917e2..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;
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index a7fae26f4654..a52184007f5f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -49,7 +49,10 @@ nouveau_bo_del_ttm(struct ttm_buffer_object *bo)
                 DRM_ERROR("bo %p still attached to GEM object\n", bo);
 
         nv10_mem_put_tile_region(dev, nvbo->tile, NULL);
-        nouveau_vm_put(&nvbo->vma);
+        if (nvbo->vma.node) {
+                nouveau_vm_unmap(&nvbo->vma);
+                nouveau_vm_put(&nvbo->vma);
+        }
         kfree(nvbo);
 }
 
@@ -128,6 +131,7 @@ 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 +170,17 @@ 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 +789,7 @@ 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 +815,11 @@ 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_dma.c b/drivers/gpu/drm/nouveau/nouveau_dma.c
index 65699bfaaaea..b368ed74aad7 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dma.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dma.c
@@ -83,7 +83,8 @@ nouveau_dma_init(struct nouveau_channel *chan)
                 return ret;
 
         /* NV_MEMORY_TO_MEMORY_FORMAT requires a notifier object */
-        ret = nouveau_notifier_alloc(chan, NvNotify0, 32, &chan->m2mf_ntfy);
+        ret = nouveau_notifier_alloc(chan, NvNotify0, 32, 0xfd0, 0x1000,
+                                     &chan->m2mf_ntfy);
         if (ret)
                 return ret;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index 9821fcacc3d2..982d70b12722 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -852,7 +852,8 @@ extern const struct ttm_mem_type_manager_func nouveau_vram_manager;
 extern int  nouveau_notifier_init_channel(struct nouveau_channel *);
 extern void nouveau_notifier_takedown_channel(struct nouveau_channel *);
 extern int  nouveau_notifier_alloc(struct nouveau_channel *, uint32_t handle,
-                                   int cout, uint32_t *offset);
+                                   int cout, uint32_t start, uint32_t end,
+                                   uint32_t *offset);
 extern int  nouveau_notifier_offset(struct nouveau_gpuobj *, uint32_t *);
 extern int  nouveau_ioctl_notifier_alloc(struct drm_device *, void *data,
                                          struct drm_file *);
diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.c b/drivers/gpu/drm/nouveau/nouveau_mem.c
index 26347b7cd872..b0fb9bdcddb7 100644
--- a/drivers/gpu/drm/nouveau/nouveau_mem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_mem.c
@@ -725,8 +725,10 @@ nouveau_vram_manager_new(struct ttm_mem_type_manager *man,
         ret = vram->get(dev, mem->num_pages << PAGE_SHIFT,
                         mem->page_alignment << PAGE_SHIFT, size_nc,
                         (nvbo->tile_flags >> 8) & 0xff, &node);
-        if (ret)
-                return ret;
+        if (ret) {
+                mem->mm_node = NULL;
+                return (ret == -ENOSPC) ? 0 : ret;
+        }
 
         node->page_shift = 12;
         if (nvbo->vma.node)
diff --git a/drivers/gpu/drm/nouveau/nouveau_mm.c b/drivers/gpu/drm/nouveau/nouveau_mm.c
index 8844b50c3e54..7609756b6faf 100644
--- a/drivers/gpu/drm/nouveau/nouveau_mm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_mm.c
@@ -123,7 +123,7 @@ nouveau_mm_get(struct nouveau_mm *rmm, int type, u32 size, u32 size_nc,
                 return 0;
         }
 
-        return -ENOMEM;
+        return -ENOSPC;
 }
 
 int
diff --git a/drivers/gpu/drm/nouveau/nouveau_notifier.c b/drivers/gpu/drm/nouveau/nouveau_notifier.c
index fe29d604b820..5ea167623a82 100644
--- a/drivers/gpu/drm/nouveau/nouveau_notifier.c
+++ b/drivers/gpu/drm/nouveau/nouveau_notifier.c
@@ -96,7 +96,8 @@ nouveau_notifier_gpuobj_dtor(struct drm_device *dev,
 
 int
 nouveau_notifier_alloc(struct nouveau_channel *chan, uint32_t handle,
-                       int size, uint32_t *b_offset)
+                       int size, uint32_t start, uint32_t end,
+                       uint32_t *b_offset)
 {
         struct drm_device *dev = chan->dev;
         struct nouveau_gpuobj *nobj = NULL;
@@ -104,9 +105,10 @@ nouveau_notifier_alloc(struct nouveau_channel *chan, uint32_t handle,
         uint32_t offset;
         int target, ret;
 
-        mem = drm_mm_search_free(&chan->notifier_heap, size, 0, 0);
+        mem = drm_mm_search_free_in_range(&chan->notifier_heap, size, 0,
+                                          start, end, 0);
         if (mem)
-                mem = drm_mm_get_block(mem, size, 0);
+                mem = drm_mm_get_block_range(mem, size, 0, start, end);
         if (!mem) {
                 NV_ERROR(dev, "Channel %d notifier block full\n", chan->id);
                 return -ENOMEM;
@@ -177,7 +179,8 @@ nouveau_ioctl_notifier_alloc(struct drm_device *dev, void *data,
         if (IS_ERR(chan))
                 return PTR_ERR(chan);
 
-        ret = nouveau_notifier_alloc(chan, na->handle, na->size, &na->offset);
+        ret = nouveau_notifier_alloc(chan, na->handle, na->size, 0, 0x1000,
+                                     &na->offset);
         nouveau_channel_put(&chan);
         return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.c b/drivers/gpu/drm/nouveau/nouveau_pm.c
index f05c0cddfeca..4399e2f34db4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_pm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_pm.c
@@ -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/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_instmem.c b/drivers/gpu/drm/nouveau/nv50_instmem.c
index ea0041810ae3..e57caa2a00e3 100644
--- a/drivers/gpu/drm/nouveau/nv50_instmem.c
+++ b/drivers/gpu/drm/nouveau/nv50_instmem.c
@@ -403,16 +403,24 @@ nv50_instmem_unmap(struct nouveau_gpuobj *gpuobj)
 void
 nv50_instmem_flush(struct drm_device *dev)
 {
+        struct drm_nouveau_private *dev_priv = dev->dev_private;
+
+        spin_lock(&dev_priv->ramin_lock);
         nv_wr32(dev, 0x00330c, 0x00000001);
         if (!nv_wait(dev, 0x00330c, 0x00000002, 0x00000000))
                 NV_ERROR(dev, "PRAMIN flush timeout\n");
+        spin_unlock(&dev_priv->ramin_lock);
 }
 
 void
 nv84_instmem_flush(struct drm_device *dev)
 {
+        struct drm_nouveau_private *dev_priv = dev->dev_private;
+
+        spin_lock(&dev_priv->ramin_lock);
         nv_wr32(dev, 0x070000, 0x00000001);
         if (!nv_wait(dev, 0x070000, 0x00000002, 0x00000000))
                 NV_ERROR(dev, "PRAMIN flush timeout\n");
+        spin_unlock(&dev_priv->ramin_lock);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nv50_vm.c b/drivers/gpu/drm/nouveau/nv50_vm.c
index 459ff08241e5..6144156f255a 100644
--- a/drivers/gpu/drm/nouveau/nv50_vm.c
+++ b/drivers/gpu/drm/nouveau/nv50_vm.c
@@ -169,7 +169,11 @@ nv50_vm_flush(struct nouveau_vm *vm)
 void
 nv50_vm_flush_engine(struct drm_device *dev, int engine)
 {
+        struct drm_nouveau_private *dev_priv = dev->dev_private;
+
+        spin_lock(&dev_priv->ramin_lock);
         nv_wr32(dev, 0x100c80, (engine << 16) | 1);
         if (!nv_wait(dev, 0x100c80, 0x00000001, 0x00000000))
                 NV_ERROR(dev, "vm flush timeout: engine %d\n", engine);
+        spin_unlock(&dev_priv->ramin_lock);
 }
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index b1537000a104..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) {
@@ -555,29 +554,28 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
                                         dp_clock = dig_connector->dp_clock;
                                 }
                         }
-/* this might work properly with the new pll algo */
-#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;
-                                /* rv515 needs more testing with this option */
-                                if (rdev->family != CHIP_RV515) {
-                                        if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT))
-                                                pll->flags |= RADEON_PLL_IS_LCD;
-                                }
+                                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;
@@ -664,10 +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;
                                 }
@@ -721,14 +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:
@@ -957,11 +957,7 @@ 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);
 
-        /* rv515 seems happier with the old algo */
-        if (rdev->family == CHIP_RV515)
-                radeon_compute_pll_legacy(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div,
-                                          &ref_div, &post_div);
-        else if (ASIC_IS_AVIVO(rdev))
+        if (ASIC_IS_AVIVO(rdev))
                 radeon_compute_pll_avivo(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div,
                                          &ref_div, &post_div);
         else
@@ -995,9 +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;
@@ -1137,12 +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;
@@ -1300,12 +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;
@@ -1329,7 +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
@@ -1344,7 +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/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index ffdc8332b76e..6140ea1de45a 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1192,7 +1192,11 @@ void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
         radeon_ring_write(rdev, 1);
         /* 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);
 }
@@ -1207,7 +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);
@@ -1326,7 +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);
 
@@ -2182,7 +2194,6 @@ int evergreen_mc_init(struct radeon_device *rdev)
                 rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
         }
         rdev->mc.visible_vram_size = rdev->mc.aper_size;
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
         r700_vram_gtt_location(rdev, &rdev->mc);
         radeon_update_bandwidth_info(rdev);
 
@@ -2627,8 +2638,8 @@ 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 */
@@ -2922,7 +2933,7 @@ static int evergreen_startup(struct radeon_device *rdev)
         /* XXX: ontario has problems blitting to gart at the moment */
         if (rdev->family == CHIP_PALM) {
                 rdev->asic->copy = NULL;
-                rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
+                radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
         }
 
         /* allocate wb buffer */
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
index a1ba4b3053d0..2be698e78ff2 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);
@@ -541,7 +548,7 @@ static inline uint32_t i2f(uint32_t input)
 int evergreen_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;
@@ -557,7 +564,7 @@ int evergreen_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++;
         }
 
@@ -598,8 +605,10 @@ int evergreen_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, evergreen_vs, evergreen_vs_size * 4);
-        memcpy(ptr + rdev->r600_blit.ps_offset, evergreen_ps, evergreen_ps_size * 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);
 
@@ -614,7 +623,7 @@ done:
                 dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
                 return r;
         }
-        rdev->mc.active_vram_size = rdev->mc.real_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
         return 0;
 }
 
@@ -622,7 +631,7 @@ void evergreen_blit_fini(struct radeon_device *rdev)
 {
         int r;
 
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
         if (rdev->r600_blit.shader_obj == NULL)
                 return;
         /* If we can't reserve the bo, unref should be enough to destroy
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 afec1aca2a73..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
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 5f15820efe12..e372f9e1e5ce 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -70,23 +70,6 @@ MODULE_FIRMWARE(FIRMWARE_R520);
 
 void r100_pre_page_flip(struct radeon_device *rdev, int crtc)
 {
-        struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc];
-        u32 tmp;
-
-        /* make sure flip is at vb rather than hb */
-        tmp = RREG32(RADEON_CRTC_OFFSET_CNTL + radeon_crtc->crtc_offset);
-        tmp &= ~RADEON_CRTC_OFFSET_FLIP_CNTL;
-        /* make sure pending bit is asserted */
-        tmp |= RADEON_CRTC_GUI_TRIG_OFFSET_LEFT_EN;
-        WREG32(RADEON_CRTC_OFFSET_CNTL + radeon_crtc->crtc_offset, tmp);
-
-        /* set pageflip to happen as late as possible in the vblank interval.
-         * same field for crtc1/2
-         */
-        tmp = RREG32(RADEON_CRTC_GEN_CNTL);
-        tmp &= ~RADEON_CRTC_VSTAT_MODE_MASK;
-        WREG32(RADEON_CRTC_GEN_CNTL, tmp);
-
         /* enable the pflip int */
         radeon_irq_kms_pflip_irq_get(rdev, crtc);
 }
@@ -1041,7 +1024,7 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size)
                 return r;
         }
         rdev->cp.ready = true;
-        rdev->mc.active_vram_size = rdev->mc.real_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
         return 0;
 }
 
@@ -1059,7 +1042,7 @@ void r100_cp_fini(struct radeon_device *rdev)
 void r100_cp_disable(struct radeon_device *rdev)
 {
         /* Disable ring */
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
         rdev->cp.ready = false;
         WREG32(RADEON_CP_CSQ_MODE, 0);
         WREG32(RADEON_CP_CSQ_CNTL, 0);
@@ -1427,6 +1410,7 @@ 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 +1423,7 @@ 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 +1439,7 @@ 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 +1457,7 @@ 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 +1475,7 @@ 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 +1493,12 @@ 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 +1519,11 @@ 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 +1548,8 @@ 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 +1567,7 @@ 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 +1584,7 @@ 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 +1599,14 @@ 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 +1620,7 @@ 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 +1673,7 @@ 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 +1684,7 @@ 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",
@@ -2310,7 +2312,6 @@ void r100_vram_init_sizes(struct radeon_device *rdev)
         /* FIXME we don't use the second aperture yet when we could use it */
         if (rdev->mc.visible_vram_size > rdev->mc.aper_size)
                 rdev->mc.visible_vram_size = rdev->mc.aper_size;
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
         config_aper_size = RREG32(RADEON_CONFIG_APER_SIZE);
         if (rdev->flags & RADEON_IS_IGP) {
                 uint32_t tom;
@@ -3318,9 +3319,9 @@ 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 +3342,9 @@ 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 +3361,28 @@ 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 +3442,23 @@ 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 +3472,8 @@ 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 = false;
+                track->aa.robj = NULL;
         }
 
         for (i = 0; i < track->num_cb; i++) {
@@ -3746,8 +3783,6 @@ static int r100_startup(struct radeon_device *rdev)
         r100_mc_program(rdev);
         /* Resume clock */
         r100_clock_startup(rdev);
-        /* Initialize GPU configuration (# pipes, ...) */
-//      r100_gpu_init(rdev);
         /* Initialize GART (initialize after TTM so we can allocate
          * memory through TTM but finalize after TTM) */
         r100_enable_bm(rdev);
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 55fe5ba7def3..069efa8c8ecf 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -667,6 +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:
@@ -679,6 +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:
@@ -717,6 +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:
@@ -743,6 +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 */
@@ -752,6 +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:
@@ -814,6 +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 */
@@ -822,6 +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 */
@@ -838,6 +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 */
@@ -861,14 +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:
@@ -898,6 +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:
@@ -910,6 +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:
@@ -951,8 +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:
@@ -980,6 +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:
@@ -1017,6 +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:
@@ -1046,6 +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);
@@ -1060,6 +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 */
@@ -1073,6 +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 |
@@ -1084,8 +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/r600.c b/drivers/gpu/drm/radeon/r600.c
index 650672a0f5ad..9b3fad23b76c 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1255,7 +1255,6 @@ int r600_mc_init(struct radeon_device *rdev)
         rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE);
         rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE);
         rdev->mc.visible_vram_size = rdev->mc.aper_size;
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
         r600_vram_gtt_location(rdev, &rdev->mc);
 
         if (rdev->flags & RADEON_IS_IGP) {
@@ -1937,7 +1936,7 @@ void r600_pciep_wreg(struct radeon_device *rdev, u32 reg, u32 v)
  */
 void r600_cp_stop(struct radeon_device *rdev)
 {
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
         WREG32(R_0086D8_CP_ME_CNTL, S_0086D8_CP_ME_HALT(1));
         WREG32(SCRATCH_UMSK, 0);
 }
@@ -2105,7 +2104,11 @@ 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);
@@ -2192,7 +2195,11 @@ 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);
 
@@ -2628,7 +2635,11 @@ 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);
 }
@@ -3297,8 +3308,8 @@ 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..df68d91e8190 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);
 
@@ -545,7 +558,7 @@ done:
                 dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
                 return r;
         }
-        rdev->mc.active_vram_size = rdev->mc.real_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
         return 0;
 }
 
@@ -553,7 +566,7 @@ void r600_blit_fini(struct radeon_device *rdev)
 {
         int r;
 
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
         if (rdev->r600_blit.shader_obj == NULL)
                 return;
         /* If we can't reserve the bo, unref should be enough to destroy
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/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 56c48b67ef3d..6b3429495118 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -345,7 +345,6 @@ struct radeon_mc {
          * about vram size near mc fb location */
         u64                     mc_vram_size;
         u64                     visible_vram_size;
-        u64                     active_vram_size;
         u64                     gtt_size;
         u64                     gtt_start;
         u64                     gtt_end;
@@ -1448,6 +1447,7 @@ extern void radeon_vram_location(struct radeon_device *rdev, struct radeon_mc *m
 extern void radeon_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 extern int radeon_resume_kms(struct drm_device *dev);
 extern int radeon_suspend_kms(struct drm_device *dev, pm_message_t state);
+extern void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size);
 
 /* r600, rv610, rv630, rv620, rv635, rv670, rs780, rs880 */
 extern bool r600_card_posted(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index e75d63b8e21d..793c5e6026ad 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -834,6 +834,9 @@ static struct radeon_asic sumo_asic = {
         .pm_finish = &evergreen_pm_finish,
         .pm_init_profile = &rs780_pm_init_profile,
         .pm_get_dynpm_state = &r600_pm_get_dynpm_state,
+        .pre_page_flip = &evergreen_pre_page_flip,
+        .page_flip = &evergreen_page_flip,
+        .post_page_flip = &evergreen_post_page_flip,
 };
 
 static struct radeon_asic btc_asic = {
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 5c1cc7ad9a15..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;
@@ -1274,11 +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:
@@ -1442,7 +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;
@@ -1456,7 +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;
@@ -1470,7 +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;
@@ -1553,8 +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);
@@ -1569,13 +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:
@@ -2189,7 +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;
@@ -2284,7 +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;
@@ -2295,7 +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) {
@@ -2408,13 +2408,13 @@ 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)
@@ -2533,7 +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)
@@ -2542,7 +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,
@@ -2551,7 +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);
 }
@@ -2565,7 +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);
 }
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index d27ef74590cd..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);
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 0d478932b1a9..4954e2d6ffa2 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -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 2eff98cfd728..3e7e7f9eb781 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -793,6 +793,11 @@ static void avivo_get_fb_div(struct radeon_pll *pll,
         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,
@@ -826,6 +831,11 @@ static u32 avivo_get_post_div(struct radeon_pll *pll,
                         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;
 }
 
@@ -961,7 +971,7 @@ void radeon_compute_pll_legacy(struct radeon_pll *pll,
                 max_fractional_feed_div = pll->max_frac_feedback_div;
         }
 
-        for (post_div = min_post_div; post_div <= max_post_div; ++post_div) {
+        for (post_div = max_post_div; post_div >= min_post_div; --post_div) {
                 uint32_t ref_div;
 
                 if ((pll->flags & RADEON_PLL_NO_ODD_POST_DIV) && (post_div & 1))
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 d4a542247618..b4274883227f 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -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;
@@ -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_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index 66324b5bb5ba..cc44bdfec80f 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -113,11 +113,14 @@ static int radeonfb_create_pinned_object(struct radeon_fbdev *rfbdev,
         u32 tiling_flags = 0;
         int ret;
         int aligned_size, size;
+        int height = mode_cmd->height;
 
         /* need to align pitch with crtc limits */
         mode_cmd->pitch = radeon_align_pitch(rdev, mode_cmd->width, mode_cmd->bpp, fb_tiled) * ((mode_cmd->bpp + 1) / 8);
 
-        size = mode_cmd->pitch * mode_cmd->height;
+        if (rdev->family >= CHIP_R600)
+                height = ALIGN(mode_cmd->height, 8);
+        size = mode_cmd->pitch * height;
         aligned_size = ALIGN(size, PAGE_SIZE);
         ret = radeon_gem_object_create(rdev, aligned_size, 0,
                                        RADEON_GEM_DOMAIN_VRAM,
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index df95eb83dac6..1fe95dfe48c9 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -156,9 +156,12 @@ int radeon_gem_info_ioctl(struct drm_device *dev, void *data,
 {
         struct radeon_device *rdev = dev->dev_private;
         struct drm_radeon_gem_info *args = data;
+        struct ttm_mem_type_manager *man;
+
+        man = &rdev->mman.bdev.man[TTM_PL_VRAM];
 
         args->vram_size = rdev->mc.real_vram_size;
-        args->vram_visible = rdev->mc.real_vram_size;
+        args->vram_visible = (u64)man->size << PAGE_SHIFT;
         if (rdev->stollen_vga_memory)
                 args->vram_visible -= radeon_bo_size(rdev->stollen_vga_memory);
         args->vram_visible -= radeon_fbdev_total_size(rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index cf0638c3b7c7..78968b738e88 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -443,7 +443,7 @@ int radeon_crtc_do_set_base(struct drm_crtc *crtc,
                        (target_fb->bits_per_pixel * 8));
         crtc_pitch |= crtc_pitch << 16;
 
-
+        crtc_offset_cntl |= RADEON_CRTC_GUI_TRIG_OFFSET_LEFT_EN;
         if (tiling_flags & RADEON_TILING_MACRO) {
                 if (ASIC_IS_R300(rdev))
                         crtc_offset_cntl |= (R300_CRTC_X_Y_MODE_EN |
@@ -502,6 +502,7 @@ int radeon_crtc_do_set_base(struct drm_crtc *crtc,
         gen_cntl_val = RREG32(gen_cntl_reg);
         gen_cntl_val &= ~(0xf << 8);
         gen_cntl_val |= (format << 8);
+        gen_cntl_val &= ~RADEON_CRTC_VSTAT_MODE_MASK;
         WREG32(gen_cntl_reg, gen_cntl_val);
 
         crtc_offset = (u32)base;
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 6794cdf91f28..a670caaee29e 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -209,6 +209,7 @@ enum radeon_connector_table {
         CT_EMAC,
         CT_RN50_POWER,
         CT_MAC_X800,
+        CT_MAC_G5_9600,
 };
 
 enum radeon_dvo_chip {
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 1272e4b6a1d4..8389b4c63d12 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -589,6 +589,20 @@ void radeon_ttm_fini(struct radeon_device *rdev)
         DRM_INFO("radeon: ttm finalized\n");
 }
 
+/* this should only be called at bootup or when userspace
+ * isn't running */
+void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size)
+{
+        struct ttm_mem_type_manager *man;
+
+        if (!rdev->mman.initialized)
+                return;
+
+        man = &rdev->mman.bdev.man[TTM_PL_VRAM];
+        /* this just adjusts TTM size idea, which sets lpfn to the correct value */
+        man->size = size >> PAGE_SHIFT;
+}
+
 static struct vm_operations_struct radeon_ttm_vm_ops;
 static const struct vm_operations_struct *ttm_vm_ops = NULL;
 
@@ -787,9 +801,9 @@ 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/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 5afe294ed51f..8af4679db23e 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -751,7 +751,6 @@ void rs600_mc_init(struct radeon_device *rdev)
         rdev->mc.real_vram_size = RREG32(RADEON_CONFIG_MEMSIZE);
         rdev->mc.mc_vram_size = rdev->mc.real_vram_size;
         rdev->mc.visible_vram_size = rdev->mc.aper_size;
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
         rdev->mc.igp_sideport_enabled = radeon_atombios_sideport_present(rdev);
         base = RREG32_MC(R_000004_MC_FB_LOCATION);
         base = G_000004_MC_FB_START(base) << 16;
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 0137d3e3728d..66c949b7c18c 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;
@@ -157,7 +157,6 @@ void rs690_mc_init(struct radeon_device *rdev)
         rdev->mc.aper_base = pci_resource_start(rdev->pdev, 0);
         rdev->mc.aper_size = pci_resource_len(rdev->pdev, 0);
         rdev->mc.visible_vram_size = rdev->mc.aper_size;
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
         base = RREG32_MC(R_000100_MCCFG_FB_LOCATION);
         base = G_000100_MC_FB_START(base) << 16;
         rdev->mc.igp_sideport_enabled = radeon_atombios_sideport_present(rdev);
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 2211a323db41..714ad45757d0 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -307,7 +307,7 @@ static void rv770_mc_program(struct radeon_device *rdev)
  */
 void r700_cp_stop(struct radeon_device *rdev)
 {
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
         WREG32(CP_ME_CNTL, (CP_ME_HALT | CP_PFP_HALT));
         WREG32(SCRATCH_UMSK, 0);
 }
@@ -321,7 +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);
@@ -1119,7 +1123,6 @@ int rv770_mc_init(struct radeon_device *rdev)
         rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE);
         rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE);
         rdev->mc.visible_vram_size = rdev->mc.aper_size;
-        rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
         r700_vram_gtt_location(rdev, &rdev->mc);
         radeon_update_bandwidth_info(rdev);
 
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