41 files changed, 219 insertions, 132 deletions

diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index acf3a36c9ebc..32982da82694 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -68,15 +68,7 @@ void __armada_drm_queue_unref_work(struct drm_device *dev,
 {
         struct armada_private *priv = dev->dev_private;
 
-        /*
-         * Yes, we really must jump through these hoops just to store a
-         * _pointer_ to something into the kfifo.  This is utterly insane
-         * and idiotic, because it kfifo requires the _data_ pointed to by
-         * the pointer const, not the pointer itself.  Not only that, but
-         * you have to pass a pointer _to_ the pointer you want stored.
-         */
-        const struct drm_framebuffer *silly_api_alert = fb;
-        WARN_ON(!kfifo_put(&priv->fb_unref, &silly_api_alert));
+        WARN_ON(!kfifo_put(&priv->fb_unref, fb));
         schedule_work(&priv->fb_unref_work);
 }
 
diff --git a/drivers/gpu/drm/bochs/Kconfig b/drivers/gpu/drm/bochs/Kconfig
index c8fcf12019f0..5f8b0c2b9a44 100644
--- a/drivers/gpu/drm/bochs/Kconfig
+++ b/drivers/gpu/drm/bochs/Kconfig
@@ -2,6 +2,7 @@ config DRM_BOCHS
         tristate "DRM Support for bochs dispi vga interface (qemu stdvga)"
         depends on DRM && PCI
         select DRM_KMS_HELPER
+        select DRM_KMS_FB_HELPER
         select FB_SYS_FILLRECT
         select FB_SYS_COPYAREA
         select FB_SYS_IMAGEBLIT
diff --git a/drivers/gpu/drm/drm_cache.c b/drivers/gpu/drm/drm_cache.c
index bb8f58012189..534cb89b160d 100644
--- a/drivers/gpu/drm/drm_cache.c
+++ b/drivers/gpu/drm/drm_cache.c
@@ -32,6 +32,12 @@
 #include <drm/drmP.h>
 
 #if defined(CONFIG_X86)
+
+/*
+ * clflushopt is an unordered instruction which needs fencing with mfence or
+ * sfence to avoid ordering issues.  For drm_clflush_page this fencing happens
+ * in the caller.
+ */
 static void
 drm_clflush_page(struct page *page)
 {
@@ -44,7 +50,7 @@ drm_clflush_page(struct page *page)
 
         page_virtual = kmap_atomic(page);
         for (i = 0; i < PAGE_SIZE; i += size)
-                clflush(page_virtual + i);
+                clflushopt(page_virtual + i);
         kunmap_atomic(page_virtual);
 }
 
@@ -133,7 +139,7 @@ drm_clflush_virt_range(char *addr, unsigned long length)
                 mb();
                 for (; addr < end; addr += boot_cpu_data.x86_clflush_size)
                         clflush(addr);
-                clflush(end - 1);
+                clflushopt(end - 1);
                 mb();
                 return;
         }
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index 5736aaa7e86c..f7af69bcf3f4 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -468,8 +468,8 @@ void drm_pci_exit(struct drm_driver *driver, struct pci_driver *pdriver)
         } else {
                 list_for_each_entry_safe(dev, tmp, &driver->legacy_dev_list,
                                          legacy_dev_list) {
-                        drm_put_dev(dev);
                         list_del(&dev->legacy_dev_list);
+                        drm_put_dev(dev);
                 }
         }
         DRM_INFO("Module unloaded\n");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 215131ab1dd2..c204b4e3356e 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -172,20 +172,24 @@ static int exynos_drm_open(struct drm_device *dev, struct drm_file *file)
 
         ret = exynos_drm_subdrv_open(dev, file);
         if (ret)
-                goto out;
+                goto err_file_priv_free;
 
         anon_filp = anon_inode_getfile("exynos_gem", &exynos_drm_gem_fops,
                                         NULL, 0);
         if (IS_ERR(anon_filp)) {
                 ret = PTR_ERR(anon_filp);
-                goto out;
+                goto err_subdrv_close;
         }
 
         anon_filp->f_mode = FMODE_READ | FMODE_WRITE;
         file_priv->anon_filp = anon_filp;
 
         return ret;
-out:
+
+err_subdrv_close:
+        exynos_drm_subdrv_close(dev, file);
+
+err_file_priv_free:
         kfree(file_priv);
         file->driver_priv = NULL;
         return ret;
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 04f1f02c4019..ec7bb0fc71bc 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -403,7 +403,7 @@ MODULE_DEVICE_TABLE(pci, pciidlist);
 void intel_detect_pch(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
-        struct pci_dev *pch;
+        struct pci_dev *pch = NULL;
 
         /* In all current cases, num_pipes is equivalent to the PCH_NOP setting
          * (which really amounts to a PCH but no South Display).
@@ -424,12 +424,9 @@ void intel_detect_pch(struct drm_device *dev)
          * all the ISA bridge devices and check for the first match, instead
          * of only checking the first one.
          */
-        pch = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, NULL);
-        while (pch) {
-                struct pci_dev *curr = pch;
+        while ((pch = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, pch))) {
                 if (pch->vendor == PCI_VENDOR_ID_INTEL) {
-                        unsigned short id;
-                        id = pch->device & INTEL_PCH_DEVICE_ID_MASK;
+                        unsigned short id = pch->device & INTEL_PCH_DEVICE_ID_MASK;
                         dev_priv->pch_id = id;
 
                         if (id == INTEL_PCH_IBX_DEVICE_ID_TYPE) {
@@ -461,18 +458,16 @@ void intel_detect_pch(struct drm_device *dev)
                                 DRM_DEBUG_KMS("Found LynxPoint LP PCH\n");
                                 WARN_ON(!IS_HASWELL(dev));
                                 WARN_ON(!IS_ULT(dev));
-                        } else {
-                                goto check_next;
-                        }
-                        pci_dev_put(pch);
+                        } else
+                                continue;
+
                         break;
                 }
-check_next:
-                pch = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, curr);
-                pci_dev_put(curr);
         }
         if (!pch)
-                DRM_DEBUG_KMS("No PCH found?\n");
+                DRM_DEBUG_KMS("No PCH found.\n");
+
+        pci_dev_put(pch);
 }
 
 bool i915_semaphore_is_enabled(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 40a2b36b276b..d278be110805 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -842,7 +842,7 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev)
         dev_priv->gtt.base.clear_range(&dev_priv->gtt.base,
                                        dev_priv->gtt.base.start / PAGE_SIZE,
                                        dev_priv->gtt.base.total / PAGE_SIZE,
-                                       false);
+                                       true);
 }
 
 void i915_gem_restore_gtt_mappings(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 1a24e84f2315..28d24caa49f3 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -82,9 +82,22 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
         r = devm_request_mem_region(dev->dev, base, dev_priv->gtt.stolen_size,
                                     "Graphics Stolen Memory");
         if (r == NULL) {
-                DRM_ERROR("conflict detected with stolen region: [0x%08x - 0x%08x]\n",
-                          base, base + (uint32_t)dev_priv->gtt.stolen_size);
-                base = 0;
+                /*
+                 * One more attempt but this time requesting region from
+                 * base + 1, as we have seen that this resolves the region
+                 * conflict with the PCI Bus.
+                 * This is a BIOS w/a: Some BIOS wrap stolen in the root
+                 * PCI bus, but have an off-by-one error. Hence retry the
+                 * reservation starting from 1 instead of 0.
+                 */
+                r = devm_request_mem_region(dev->dev, base + 1,
+                                            dev_priv->gtt.stolen_size - 1,
+                                            "Graphics Stolen Memory");
+                if (r == NULL) {
+                        DRM_ERROR("conflict detected with stolen region: [0x%08x - 0x%08x]\n",
+                                  base, base + (uint32_t)dev_priv->gtt.stolen_size);
+                        base = 0;
+                }
         }
 
         return base;
@@ -201,6 +214,13 @@ int i915_gem_init_stolen(struct drm_device *dev)
         struct drm_i915_private *dev_priv = dev->dev_private;
         int bios_reserved = 0;
 
+#ifdef CONFIG_INTEL_IOMMU
+        if (intel_iommu_gfx_mapped) {
+                DRM_INFO("DMAR active, disabling use of stolen memory\n");
+                return 0;
+        }
+#endif
+
         if (dev_priv->gtt.stolen_size == 0)
                 return 0;
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 9fec71175571..d554169ac592 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -618,33 +618,25 @@ static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
 
 /* raw reads, only for fast reads of display block, no need for forcewake etc. */
 #define __raw_i915_read32(dev_priv__, reg__) readl((dev_priv__)->regs + (reg__))
-#define __raw_i915_read16(dev_priv__, reg__) readw((dev_priv__)->regs + (reg__))
 
 static bool ilk_pipe_in_vblank_locked(struct drm_device *dev, enum pipe pipe)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         uint32_t status;
-
-        if (INTEL_INFO(dev)->gen < 7) {
-                status = pipe == PIPE_A ?
-                        DE_PIPEA_VBLANK :
-                        DE_PIPEB_VBLANK;
+        int reg;
+
+        if (INTEL_INFO(dev)->gen >= 8) {
+                status = GEN8_PIPE_VBLANK;
+                reg = GEN8_DE_PIPE_ISR(pipe);
+        } else if (INTEL_INFO(dev)->gen >= 7) {
+                status = DE_PIPE_VBLANK_IVB(pipe);
+                reg = DEISR;
         } else {
-                switch (pipe) {
-                default:
-                case PIPE_A:
-                        status = DE_PIPEA_VBLANK_IVB;
-                        break;
-                case PIPE_B:
-                        status = DE_PIPEB_VBLANK_IVB;
-                        break;
-                case PIPE_C:
-                        status = DE_PIPEC_VBLANK_IVB;
-                        break;
-                }
+                status = DE_PIPE_VBLANK(pipe);
+                reg = DEISR;
         }
 
-        return __raw_i915_read32(dev_priv, DEISR) & status;
+        return __raw_i915_read32(dev_priv, reg) & status;
 }
 
 static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
@@ -702,7 +694,28 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
                 else
                         position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3;
 
-                if (HAS_PCH_SPLIT(dev)) {
+                if (HAS_DDI(dev)) {
+                        /*
+                         * On HSW HDMI outputs there seems to be a 2 line
+                         * difference, whereas eDP has the normal 1 line
+                         * difference that earlier platforms have. External
+                         * DP is unknown. For now just check for the 2 line
+                         * difference case on all output types on HSW+.
+                         *
+                         * This might misinterpret the scanline counter being
+                         * one line too far along on eDP, but that's less
+                         * dangerous than the alternative since that would lead
+                         * the vblank timestamp code astray when it sees a
+                         * scanline count before vblank_start during a vblank
+                         * interrupt.
+                         */
+                        in_vbl = ilk_pipe_in_vblank_locked(dev, pipe);
+                        if ((in_vbl && (position == vbl_start - 2 ||
+                                        position == vbl_start - 1)) ||
+                            (!in_vbl && (position == vbl_end - 2 ||
+                                         position == vbl_end - 1)))
+                                position = (position + 2) % vtotal;
+                } else if (HAS_PCH_SPLIT(dev)) {
                         /*
                          * The scanline counter increments at the leading edge
                          * of hsync, ie. it completely misses the active portion
@@ -2769,10 +2782,9 @@ static void ibx_irq_postinstall(struct drm_device *dev)
                 return;
 
         if (HAS_PCH_IBX(dev)) {
-                mask = SDE_GMBUS | SDE_AUX_MASK | SDE_TRANSB_FIFO_UNDER |
-                       SDE_TRANSA_FIFO_UNDER | SDE_POISON;
+                mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
         } else {
-                mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT | SDE_ERROR_CPT;
+                mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
 
                 I915_WRITE(SERR_INT, I915_READ(SERR_INT));
         }
@@ -2832,20 +2844,19 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
                 display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE_IVB |
                                 DE_PCH_EVENT_IVB | DE_PLANEC_FLIP_DONE_IVB |
                                 DE_PLANEB_FLIP_DONE_IVB |
-                                DE_PLANEA_FLIP_DONE_IVB | DE_AUX_CHANNEL_A_IVB |
-                                DE_ERR_INT_IVB);
+                                DE_PLANEA_FLIP_DONE_IVB | DE_AUX_CHANNEL_A_IVB);
                 extra_mask = (DE_PIPEC_VBLANK_IVB | DE_PIPEB_VBLANK_IVB |
-                              DE_PIPEA_VBLANK_IVB);
+                              DE_PIPEA_VBLANK_IVB | DE_ERR_INT_IVB);
 
                 I915_WRITE(GEN7_ERR_INT, I915_READ(GEN7_ERR_INT));
         } else {
                 display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT |
                                 DE_PLANEA_FLIP_DONE | DE_PLANEB_FLIP_DONE |
                                 DE_AUX_CHANNEL_A |
-                                DE_PIPEB_FIFO_UNDERRUN | DE_PIPEA_FIFO_UNDERRUN |
                                 DE_PIPEB_CRC_DONE | DE_PIPEA_CRC_DONE |
                                 DE_POISON);
-                extra_mask = DE_PIPEA_VBLANK | DE_PIPEB_VBLANK | DE_PCU_EVENT;
+                extra_mask = DE_PIPEA_VBLANK | DE_PIPEB_VBLANK | DE_PCU_EVENT |
+                                DE_PIPEB_FIFO_UNDERRUN | DE_PIPEA_FIFO_UNDERRUN;
         }
 
         dev_priv->irq_mask = ~display_mask;
@@ -2961,9 +2972,9 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
         struct drm_device *dev = dev_priv->dev;
         uint32_t de_pipe_masked = GEN8_PIPE_FLIP_DONE |
                 GEN8_PIPE_CDCLK_CRC_DONE |
-                GEN8_PIPE_FIFO_UNDERRUN |
                 GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
-        uint32_t de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK;
+        uint32_t de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK |
+                GEN8_PIPE_FIFO_UNDERRUN;
         int pipe;
         dev_priv->de_irq_mask[PIPE_A] = ~de_pipe_masked;
         dev_priv->de_irq_mask[PIPE_B] = ~de_pipe_masked;
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index e06b9e017d6b..234ac5f7bc5a 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -1244,6 +1244,7 @@ static void intel_ddi_post_disable(struct intel_encoder *intel_encoder)
         if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
                 struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
                 intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_OFF);
+                ironlake_edp_panel_vdd_on(intel_dp);
                 ironlake_edp_panel_off(intel_dp);
         }
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 4c1672809493..9b8a7c7ea7fc 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1092,12 +1092,12 @@ static void assert_cursor(struct drm_i915_private *dev_priv,
         struct drm_device *dev = dev_priv->dev;
         bool cur_state;
 
-        if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev))
-                cur_state = I915_READ(CURCNTR_IVB(pipe)) & CURSOR_MODE;
-        else if (IS_845G(dev) || IS_I865G(dev))
+        if (IS_845G(dev) || IS_I865G(dev))
                 cur_state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
-        else
+        else if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev))
                 cur_state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
+        else
+                cur_state = I915_READ(CURCNTR_IVB(pipe)) & CURSOR_MODE;
 
         WARN(cur_state != state,
              "cursor on pipe %c assertion failure (expected %s, current %s)\n",
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 57552eb386b0..2688f6d64bb9 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1249,17 +1249,24 @@ void ironlake_edp_panel_off(struct intel_dp *intel_dp)
 
         DRM_DEBUG_KMS("Turn eDP power off\n");
 
+        WARN(!intel_dp->want_panel_vdd, "Need VDD to turn off panel\n");
+
         pp = ironlake_get_pp_control(intel_dp);
         /* We need to switch off panel power _and_ force vdd, for otherwise some
          * panels get very unhappy and cease to work. */
-        pp &= ~(POWER_TARGET_ON | PANEL_POWER_RESET | EDP_BLC_ENABLE);
+        pp &= ~(POWER_TARGET_ON | EDP_FORCE_VDD | PANEL_POWER_RESET | EDP_BLC_ENABLE);
 
         pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 
         I915_WRITE(pp_ctrl_reg, pp);
         POSTING_READ(pp_ctrl_reg);
 
+        intel_dp->want_panel_vdd = false;
+
         ironlake_wait_panel_off(intel_dp);
+
+        /* We got a reference when we enabled the VDD. */
+        intel_runtime_pm_put(dev_priv);
 }
 
 void ironlake_edp_backlight_on(struct intel_dp *intel_dp)
@@ -1639,7 +1646,7 @@ static void intel_edp_psr_enable_source(struct intel_dp *intel_dp)
                 val |= EDP_PSR_LINK_DISABLE;
 
         I915_WRITE(EDP_PSR_CTL(dev), val |
-                   IS_BROADWELL(dev) ? 0 : link_entry_time |
+                   (IS_BROADWELL(dev) ? 0 : link_entry_time) |
                    max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT |
                    idle_frames << EDP_PSR_IDLE_FRAME_SHIFT |
                    EDP_PSR_ENABLE);
@@ -1784,6 +1791,7 @@ static void intel_disable_dp(struct intel_encoder *encoder)
 
         /* Make sure the panel is off before trying to change the mode. But also
          * ensure that we have vdd while we switch off the panel. */
+        ironlake_edp_panel_vdd_on(intel_dp);
         ironlake_edp_backlight_off(intel_dp);
         intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_OFF);
         ironlake_edp_panel_off(intel_dp);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 6db0d9d17f47..ee3181ebcc92 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -845,7 +845,7 @@ static int hdmi_portclock_limit(struct intel_hdmi *hdmi)
 {
         struct drm_device *dev = intel_hdmi_to_dev(hdmi);
 
-        if (IS_G4X(dev))
+        if (!hdmi->has_hdmi_sink || IS_G4X(dev))
                 return 165000;
         else if (IS_HASWELL(dev) || INTEL_INFO(dev)->gen >= 8)
                 return 300000;
@@ -899,8 +899,8 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
          * outputs. We also need to check that the higher clock still fits
          * within limits.
          */
-        if (pipe_config->pipe_bpp > 8*3 && clock_12bpc <= portclock_limit
-            && HAS_PCH_SPLIT(dev)) {
+        if (pipe_config->pipe_bpp > 8*3 && intel_hdmi->has_hdmi_sink &&
+            clock_12bpc <= portclock_limit && HAS_PCH_SPLIT(dev)) {
                 DRM_DEBUG_KMS("picking bpc to 12 for HDMI output\n");
                 desired_bpp = 12*3;
 
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 350de359123a..079ea38f14d9 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -698,7 +698,7 @@ static void i9xx_enable_backlight(struct intel_connector *connector)
                 freq /= 0xff;
 
         ctl = freq << 17;
-        if (IS_GEN2(dev) && panel->backlight.combination_mode)
+        if (panel->backlight.combination_mode)
                 ctl |= BLM_LEGACY_MODE;
         if (IS_PINEVIEW(dev) && panel->backlight.active_low_pwm)
                 ctl |= BLM_POLARITY_PNV;
@@ -979,7 +979,7 @@ static int i9xx_setup_backlight(struct intel_connector *connector)
 
         ctl = I915_READ(BLC_PWM_CTL);
 
-        if (IS_GEN2(dev))
+        if (IS_GEN2(dev) || IS_I915GM(dev) || IS_I945GM(dev))
                 panel->backlight.combination_mode = ctl & BLM_LEGACY_MODE;
 
         if (IS_PINEVIEW(dev))
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d77cc81900f9..e1fc35a72656 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3493,6 +3493,8 @@ static void valleyview_setup_pctx(struct drm_device *dev)
         u32 pcbr;
         int pctx_size = 24*1024;
 
+        WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
         pcbr = I915_READ(VLV_PCBR);
         if (pcbr) {
                 /* BIOS set it up already, grab the pre-alloc'd space */
@@ -3542,8 +3544,6 @@ static void valleyview_enable_rps(struct drm_device *dev)
                 I915_WRITE(GTFIFODBG, gtfifodbg);
         }
 
-        valleyview_setup_pctx(dev);
-
         /* If VLV, Forcewake all wells, else re-direct to regular path */
         gen6_gt_force_wake_get(dev_priv, FORCEWAKE_ALL);
 
@@ -4395,6 +4395,8 @@ void intel_enable_gt_powersave(struct drm_device *dev)
                 ironlake_enable_rc6(dev);
                 intel_init_emon(dev);
         } else if (IS_GEN6(dev) || IS_GEN7(dev)) {
+                if (IS_VALLEYVIEW(dev))
+                        valleyview_setup_pctx(dev);
                 /*
                  * PCU communication is slow and this doesn't need to be
                  * done at any specific time, so do this out of our fast path
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 89c484d8ac26..4ee702ac8907 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -866,13 +866,16 @@ static int nouveau_pmops_runtime_suspend(struct device *dev)
         struct drm_device *drm_dev = pci_get_drvdata(pdev);
         int ret;
 
-        if (nouveau_runtime_pm == 0)
-                return -EINVAL;
+        if (nouveau_runtime_pm == 0) {
+                pm_runtime_forbid(dev);
+                return -EBUSY;
+        }
 
         /* are we optimus enabled? */
         if (nouveau_runtime_pm == -1 && !nouveau_is_optimus() && !nouveau_is_v1_dsm()) {
                 DRM_DEBUG_DRIVER("failing to power off - not optimus\n");
-                return -EINVAL;
+                pm_runtime_forbid(dev);
+                return -EBUSY;
         }
 
         nv_debug_level(SILENT);
@@ -923,12 +926,15 @@ static int nouveau_pmops_runtime_idle(struct device *dev)
         struct nouveau_drm *drm = nouveau_drm(drm_dev);
         struct drm_crtc *crtc;
 
-        if (nouveau_runtime_pm == 0)
+        if (nouveau_runtime_pm == 0) {
+                pm_runtime_forbid(dev);
                 return -EBUSY;
+        }
 
         /* are we optimus enabled? */
         if (nouveau_runtime_pm == -1 && !nouveau_is_optimus() && !nouveau_is_v1_dsm()) {
                 DRM_DEBUG_DRIVER("failing to power off - not optimus\n");
+                pm_runtime_forbid(dev);
                 return -EBUSY;
         }
 
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index 2cec2ab02f80..607dc14d195e 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -1314,7 +1314,7 @@ atombios_dig_transmitter_setup(struct drm_encoder *encoder, int action, uint8_t
                         }
                         if (is_dp)
                                 args.v5.ucLaneNum = dp_lane_count;
-                        else if (radeon_encoder->pixel_clock > 165000)
+                        else if (radeon_dig_monitor_is_duallink(encoder, radeon_encoder->pixel_clock))
                                 args.v5.ucLaneNum = 8;
                         else
                                 args.v5.ucLaneNum = 4;
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index e6419ca7cd37..bbb17841a9e5 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -3046,7 +3046,7 @@ static u32 cik_create_bitmask(u32 bit_width)
 }
 
 /**
- * cik_select_se_sh - select which SE, SH to address
+ * cik_get_rb_disabled - computes the mask of disabled RBs
  *
  * @rdev: radeon_device pointer
  * @max_rb_num: max RBs (render backends) for the asic
@@ -4134,8 +4134,11 @@ static void cik_cp_compute_enable(struct radeon_device *rdev, bool enable)
 {
         if (enable)
                 WREG32(CP_MEC_CNTL, 0);
-        else
+        else {
                 WREG32(CP_MEC_CNTL, (MEC_ME1_HALT | MEC_ME2_HALT));
+                rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX].ready = false;
+                rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX].ready = false;
+        }
         udelay(50);
 }
 
@@ -7902,7 +7905,8 @@ int cik_resume(struct radeon_device *rdev)
         /* init golden registers */
         cik_init_golden_registers(rdev);
 
-        radeon_pm_resume(rdev);
+        if (rdev->pm.pm_method == PM_METHOD_DPM)
+                radeon_pm_resume(rdev);
 
         rdev->accel_working = true;
         r = cik_startup(rdev);
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index 1ecb3f1070e3..94626ea90fa5 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -264,6 +264,8 @@ static void cik_sdma_gfx_stop(struct radeon_device *rdev)
                 WREG32(SDMA0_GFX_RB_CNTL + reg_offset, rb_cntl);
                 WREG32(SDMA0_GFX_IB_CNTL + reg_offset, 0);
         }
+        rdev->ring[R600_RING_TYPE_DMA_INDEX].ready = false;
+        rdev->ring[CAYMAN_RING_TYPE_DMA1_INDEX].ready = false;
 }
 
 /**
@@ -291,6 +293,11 @@ void cik_sdma_enable(struct radeon_device *rdev, bool enable)
         u32 me_cntl, reg_offset;
         int i;
 
+        if (enable == false) {
+                cik_sdma_gfx_stop(rdev);
+                cik_sdma_rlc_stop(rdev);
+        }
+
         for (i = 0; i < 2; i++) {
                 if (i == 0)
                         reg_offset = SDMA0_REGISTER_OFFSET;
@@ -420,10 +427,6 @@ static int cik_sdma_load_microcode(struct radeon_device *rdev)
         if (!rdev->sdma_fw)
                 return -EINVAL;
 
-        /* stop the gfx rings and rlc compute queues */
-        cik_sdma_gfx_stop(rdev);
-        cik_sdma_rlc_stop(rdev);
-
         /* halt the MEs */
         cik_sdma_enable(rdev, false);
 
@@ -492,9 +495,6 @@ int cik_sdma_resume(struct radeon_device *rdev)
  */
 void cik_sdma_fini(struct radeon_device *rdev)
 {
-        /* stop the gfx rings and rlc compute queues */
-        cik_sdma_gfx_stop(rdev);
-        cik_sdma_rlc_stop(rdev);
         /* halt the MEs */
         cik_sdma_enable(rdev, false);
         radeon_ring_fini(rdev, &rdev->ring[R600_RING_TYPE_DMA_INDEX]);
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 8a2c010b7dc5..27b0ff16082e 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -5299,7 +5299,8 @@ int evergreen_resume(struct radeon_device *rdev)
         /* init golden registers */
         evergreen_init_golden_registers(rdev);
 
-        radeon_pm_resume(rdev);
+        if (rdev->pm.pm_method == PM_METHOD_DPM)
+                radeon_pm_resume(rdev);
 
         rdev->accel_working = true;
         r = evergreen_startup(rdev);
diff --git a/drivers/gpu/drm/radeon/evergreen_smc.h b/drivers/gpu/drm/radeon/evergreen_smc.h
index 76ada8cfe902..3a03ba37d043 100644
--- a/drivers/gpu/drm/radeon/evergreen_smc.h
+++ b/drivers/gpu/drm/radeon/evergreen_smc.h
@@ -57,7 +57,7 @@ typedef struct SMC_Evergreen_MCRegisters SMC_Evergreen_MCRegisters;
 
 #define EVERGREEN_SMC_FIRMWARE_HEADER_LOCATION 0x100
 
-#define EVERGREEN_SMC_FIRMWARE_HEADER_softRegisters   0x0
+#define EVERGREEN_SMC_FIRMWARE_HEADER_softRegisters   0x8
 #define EVERGREEN_SMC_FIRMWARE_HEADER_stateTable      0xC
 #define EVERGREEN_SMC_FIRMWARE_HEADER_mcRegisterTable 0x20
 
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index ea932ac66fc6..bf6300cfd62d 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -2105,7 +2105,8 @@ int cayman_resume(struct radeon_device *rdev)
         /* init golden registers */
         ni_init_golden_registers(rdev);
 
-        radeon_pm_resume(rdev);
+        if (rdev->pm.pm_method == PM_METHOD_DPM)
+                radeon_pm_resume(rdev);
 
         rdev->accel_working = true;
         r = cayman_startup(rdev);
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index ef024ce3f7cc..3cc78bb66042 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -3942,8 +3942,6 @@ int r100_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r = r100_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 7c63ef840e86..0b658b34b33a 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -1430,8 +1430,6 @@ int r300_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r = r300_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index 3768aab2710b..802b19220a21 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -325,8 +325,6 @@ int r420_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r = r420_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/r520.c b/drivers/gpu/drm/radeon/r520.c
index e209eb75024f..98d6053c36c6 100644
--- a/drivers/gpu/drm/radeon/r520.c
+++ b/drivers/gpu/drm/radeon/r520.c
@@ -240,8 +240,6 @@ int r520_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r = r520_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index cdbc4171fe73..647ef4079217 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2968,7 +2968,8 @@ int r600_resume(struct radeon_device *rdev)
         /* post card */
         atom_asic_init(rdev->mode_info.atom_context);
 
-        radeon_pm_resume(rdev);
+        if (rdev->pm.pm_method == PM_METHOD_DPM)
+                radeon_pm_resume(rdev);
 
         rdev->accel_working = true;
         r = r600_startup(rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index b012cbbc3ed5..044bc98fb459 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1521,13 +1521,16 @@ int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
         if (r)
                 DRM_ERROR("ib ring test failed (%d).\n", r);
 
-        if (rdev->pm.dpm_enabled) {
+        if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled) {
                 /* do dpm late init */
                 r = radeon_pm_late_init(rdev);
                 if (r) {
                         rdev->pm.dpm_enabled = false;
                         DRM_ERROR("radeon_pm_late_init failed, disabling dpm\n");
                 }
+        } else {
+                /* resume old pm late */
+                radeon_pm_resume(rdev);
         }
 
         radeon_restore_bios_scratch_regs(rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 84a1bbb75f91..f633c2782170 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -403,11 +403,15 @@ static int radeon_pmops_runtime_suspend(struct device *dev)
         struct drm_device *drm_dev = pci_get_drvdata(pdev);
         int ret;
 
-        if (radeon_runtime_pm == 0)
-                return -EINVAL;
+        if (radeon_runtime_pm == 0) {
+                pm_runtime_forbid(dev);
+                return -EBUSY;
+        }
 
-        if (radeon_runtime_pm == -1 && !radeon_is_px())
-                return -EINVAL;
+        if (radeon_runtime_pm == -1 && !radeon_is_px()) {
+                pm_runtime_forbid(dev);
+                return -EBUSY;
+        }
 
         drm_dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
         drm_kms_helper_poll_disable(drm_dev);
@@ -456,12 +460,15 @@ static int radeon_pmops_runtime_idle(struct device *dev)
         struct drm_device *drm_dev = pci_get_drvdata(pdev);
         struct drm_crtc *crtc;
 
-        if (radeon_runtime_pm == 0)
+        if (radeon_runtime_pm == 0) {
+                pm_runtime_forbid(dev);
                 return -EBUSY;
+        }
 
         /* are we PX enabled? */
         if (radeon_runtime_pm == -1 && !radeon_is_px()) {
                 DRM_DEBUG_DRIVER("failing to power off - not px\n");
+                pm_runtime_forbid(dev);
                 return -EBUSY;
         }
 
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 2aecd6dc2610..66ed3ea71440 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -33,6 +33,13 @@
 #include <linux/vga_switcheroo.h>
 #include <linux/slab.h>
 #include <linux/pm_runtime.h>
+
+#if defined(CONFIG_VGA_SWITCHEROO)
+bool radeon_is_px(void);
+#else
+static inline bool radeon_is_px(void) { return false; }
+#endif
+
 /**
  * radeon_driver_unload_kms - Main unload function for KMS.
  *
@@ -130,7 +137,8 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags)
                                 "Error during ACPI methods call\n");
         }
 
-        if (radeon_runtime_pm != 0) {
+        if ((radeon_runtime_pm == 1) ||
+            ((radeon_runtime_pm == -1) && radeon_is_px())) {
                 pm_runtime_use_autosuspend(dev->dev);
                 pm_runtime_set_autosuspend_delay(dev->dev, 5000);
                 pm_runtime_set_active(dev->dev);
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 77f5b0c3edb8..040a2a10ea17 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -714,6 +714,9 @@ int radeon_ttm_init(struct radeon_device *rdev)
                 DRM_ERROR("Failed initializing VRAM heap.\n");
                 return r;
         }
+        /* Change the size here instead of the init above so only lpfn is affected */
+        radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
+
         r = radeon_bo_create(rdev, 256 * 1024, PAGE_SIZE, true,
                              RADEON_GEM_DOMAIN_VRAM,
                              NULL, &rdev->stollen_vga_memory);
@@ -935,7 +938,7 @@ static ssize_t radeon_ttm_gtt_read(struct file *f, char __user *buf,
         while (size) {
                 loff_t p = *pos / PAGE_SIZE;
                 unsigned off = *pos & ~PAGE_MASK;
-                ssize_t cur_size = min(size, PAGE_SIZE - off);
+                size_t cur_size = min_t(size_t, size, PAGE_SIZE - off);
                 struct page *page;
                 void *ptr;
 
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index b5c2369cda2f..130d5cc50d43 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -474,8 +474,6 @@ int rs400_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r = rs400_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index fdcde7693032..72d3616de08e 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -1048,8 +1048,6 @@ int rs600_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r = rs600_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 35950738bd5e..3462b64369bf 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -756,8 +756,6 @@ int rs690_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r = rs690_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 98e8138ff779..237dd29d9f1c 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -586,8 +586,6 @@ int rv515_resume(struct radeon_device *rdev)
         /* Initialize surface registers */
         radeon_surface_init(rdev);
 
-        radeon_pm_resume(rdev);
-
         rdev->accel_working = true;
         r =  rv515_startup(rdev);
         if (r) {
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 4e37a42305d8..fef310773aad 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -1811,7 +1811,8 @@ int rv770_resume(struct radeon_device *rdev)
         /* init golden registers */
         rv770_init_golden_registers(rdev);
 
-        radeon_pm_resume(rdev);
+        if (rdev->pm.pm_method == PM_METHOD_DPM)
+                radeon_pm_resume(rdev);
 
         rdev->accel_working = true;
         r = rv770_startup(rdev);
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 83578324e5d1..9a124d0608b3 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -6618,7 +6618,8 @@ int si_resume(struct radeon_device *rdev)
         /* init golden registers */
         si_init_golden_registers(rdev);
 
-        radeon_pm_resume(rdev);
+        if (rdev->pm.pm_method == PM_METHOD_DPM)
+                radeon_pm_resume(rdev);
 
         rdev->accel_working = true;
         r = si_startup(rdev);
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index a06651309388..214b7992a3aa 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -351,9 +351,11 @@ static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo,
 
 moved:
         if (bo->evicted) {
-                ret = bdev->driver->invalidate_caches(bdev, bo->mem.placement);
-                if (ret)
-                        pr_err("Can not flush read caches\n");
+                if (bdev->driver->invalidate_caches) {
+                        ret = bdev->driver->invalidate_caches(bdev, bo->mem.placement);
+                        if (ret)
+                                pr_err("Can not flush read caches\n");
+                }
                 bo->evicted = false;
         }
 
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index 801231c9ae48..0ce48e5a9cb4 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -339,11 +339,13 @@ int ttm_bo_mmap(struct file *filp, struct vm_area_struct *vma,
         vma->vm_private_data = bo;
 
         /*
-         * PFNMAP is faster than MIXEDMAP due to reduced page
-         * administration. So use MIXEDMAP only if private VMA, where
-         * we need to support COW.
+         * We'd like to use VM_PFNMAP on shared mappings, where
+         * (vma->vm_flags & VM_SHARED) != 0, for performance reasons,
+         * but for some reason VM_PFNMAP + x86 PAT + write-combine is very
+         * bad for performance. Until that has been sorted out, use
+         * VM_MIXEDMAP on all mappings. See freedesktop.org bug #75719
          */
-        vma->vm_flags |= (vma->vm_flags & VM_SHARED) ? VM_PFNMAP : VM_MIXEDMAP;
+        vma->vm_flags |= VM_MIXEDMAP;
         vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP;
         return 0;
 out_unref:
@@ -359,7 +361,7 @@ int ttm_fbdev_mmap(struct vm_area_struct *vma, struct ttm_buffer_object *bo)
 
         vma->vm_ops = &ttm_bo_vm_ops;
         vma->vm_private_data = ttm_bo_reference(bo);
-        vma->vm_flags |= (vma->vm_flags & VM_SHARED) ? VM_PFNMAP : VM_MIXEDMAP;
+        vma->vm_flags |= VM_MIXEDMAP;
         vma->vm_flags |= VM_IO | VM_DONTEXPAND;
         return 0;
 }
diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c
index 8d67b943ac05..0394811251bd 100644
--- a/drivers/gpu/drm/udl/udl_gem.c
+++ b/drivers/gpu/drm/udl/udl_gem.c
@@ -177,8 +177,10 @@ void udl_gem_free_object(struct drm_gem_object *gem_obj)
         if (obj->vmapping)
                 udl_gem_vunmap(obj);
 
-        if (gem_obj->import_attach)
+        if (gem_obj->import_attach) {
                 drm_prime_gem_destroy(gem_obj, obj->sg);
+                put_device(gem_obj->dev->dev);
+        }
 
         if (obj->pages)
                 udl_gem_put_pages(obj);
@@ -256,9 +258,12 @@ struct drm_gem_object *udl_gem_prime_import(struct drm_device *dev,
         int ret;
 
         /* need to attach */
+        get_device(dev->dev);
         attach = dma_buf_attach(dma_buf, dev->dev);
-        if (IS_ERR(attach))
+        if (IS_ERR(attach)) {
+                put_device(dev->dev);
                 return ERR_CAST(attach);
+        }
 
         get_dma_buf(dma_buf);
 
@@ -282,6 +287,6 @@ fail_unmap:
 fail_detach:
         dma_buf_detach(dma_buf, attach);
         dma_buf_put(dma_buf);
-
+        put_device(dev->dev);
         return ERR_PTR(ret);
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
index 82468d902915..e7af580ab977 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
@@ -830,6 +830,24 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data,
         if (unlikely(ret != 0))
                 goto out_unlock;
 
+        /*
+         * A gb-aware client referencing a shared surface will
+         * expect a backup buffer to be present.
+         */
+        if (dev_priv->has_mob && req->shareable) {
+                uint32_t backup_handle;
+
+                ret = vmw_user_dmabuf_alloc(dev_priv, tfile,
+                                            res->backup_size,
+                                            true,
+                                            &backup_handle,
+                                            &res->backup);
+                if (unlikely(ret != 0)) {
+                        vmw_resource_unreference(&res);
+                        goto out_unlock;
+                }
+        }
+
         tmp = vmw_resource_reference(&srf->res);
         ret = ttm_prime_object_init(tfile, res->backup_size, &user_srf->prime,
                                     req->shareable, VMW_RES_SURFACE,