Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

Conflicts:
	drivers/net/usb/qmi_wwan.c
	include/net/dst.h

Trivial merge conflicts, both were overlapping changes.

Signed-off-by: David S. Miller <davem@davemloft.net>

9 files changed, 123 insertions, 80 deletions

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index c27a21034a5e..d5c784d48671 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1290,12 +1290,9 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	 * then we do not take part in VGA arbitration and the
 	 * vga_client_register() fails with -ENODEV.
 	 */
-	if (!HAS_PCH_SPLIT(dev)) {
-		ret = vga_client_register(dev->pdev, dev, NULL,
-					  i915_vga_set_decode);
-		if (ret && ret != -ENODEV)
-			goto out;
-	}
+	ret = vga_client_register(dev->pdev, dev, NULL, i915_vga_set_decode);
+	if (ret && ret != -ENODEV)
+		goto out;
 
 	intel_register_dsm_handler();
 
@@ -1351,12 +1348,6 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	 */
 	intel_fbdev_initial_config(dev);
 
-	/*
-	 * Must do this after fbcon init so that
-	 * vgacon_save_screen() works during the handover.
-	 */
-	i915_disable_vga_mem(dev);
-
 	/* Only enable hotplug handling once the fbdev is fully set up. */
 	dev_priv->enable_hotplug_processing = true;
 
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 69d8ed5416c3..2ad27880cd04 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -505,6 +505,8 @@ static int i915_drm_freeze(struct drm_device *dev)
 		intel_modeset_suspend_hw(dev);
 	}
 
+	i915_gem_suspend_gtt_mappings(dev);
+
 	i915_save_state(dev);
 
 	intel_opregion_fini(dev);
@@ -648,7 +650,8 @@ static int i915_drm_thaw(struct drm_device *dev)
 		mutex_lock(&dev->struct_mutex);
 		i915_gem_restore_gtt_mappings(dev);
 		mutex_unlock(&dev->struct_mutex);
-	}
+	} else if (drm_core_check_feature(dev, DRIVER_MODESET))
+		i915_check_and_clear_faults(dev);
 
 	__i915_drm_thaw(dev);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 35874b3a86dc..ab0f2c0a440c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -497,10 +497,12 @@ struct i915_address_space {
 
 	/* FIXME: Need a more generic return type */
 	gen6_gtt_pte_t (*pte_encode)(dma_addr_t addr,
-				     enum i915_cache_level level);
+				     enum i915_cache_level level,
+				     bool valid); /* Create a valid PTE */
 	void (*clear_range)(struct i915_address_space *vm,
 			    unsigned int first_entry,
-			    unsigned int num_entries);
+			    unsigned int num_entries,
+			    bool use_scratch);
 	void (*insert_entries)(struct i915_address_space *vm,
 			       struct sg_table *st,
 			       unsigned int first_entry,
@@ -2065,6 +2067,8 @@ void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt,
 void i915_ppgtt_unbind_object(struct i915_hw_ppgtt *ppgtt,
 			      struct drm_i915_gem_object *obj);
 
+void i915_check_and_clear_faults(struct drm_device *dev);
+void i915_gem_suspend_gtt_mappings(struct drm_device *dev);
 void i915_gem_restore_gtt_mappings(struct drm_device *dev);
 int __must_check i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj);
 void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj,
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 212f6d8c35ec..1f7b4caefb6e 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -58,9 +58,10 @@
 #define HSW_WT_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0x6)
 
 static gen6_gtt_pte_t snb_pte_encode(dma_addr_t addr,
-				     enum i915_cache_level level)
+				     enum i915_cache_level level,
+				     bool valid)
 {
-	gen6_gtt_pte_t pte = GEN6_PTE_VALID;
+	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= GEN6_PTE_ADDR_ENCODE(addr);
 
 	switch (level) {
@@ -79,9 +80,10 @@ static gen6_gtt_pte_t snb_pte_encode(dma_addr_t addr,
 }
 
 static gen6_gtt_pte_t ivb_pte_encode(dma_addr_t addr,
-				     enum i915_cache_level level)
+				     enum i915_cache_level level,
+				     bool valid)
 {
-	gen6_gtt_pte_t pte = GEN6_PTE_VALID;
+	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= GEN6_PTE_ADDR_ENCODE(addr);
 
 	switch (level) {
@@ -105,9 +107,10 @@ static gen6_gtt_pte_t ivb_pte_encode(dma_addr_t addr,
 #define BYT_PTE_SNOOPED_BY_CPU_CACHES	(1 << 2)
 
 static gen6_gtt_pte_t byt_pte_encode(dma_addr_t addr,
-				     enum i915_cache_level level)
+				     enum i915_cache_level level,
+				     bool valid)
 {
-	gen6_gtt_pte_t pte = GEN6_PTE_VALID;
+	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= GEN6_PTE_ADDR_ENCODE(addr);
 
 	/* Mark the page as writeable.  Other platforms don't have a
@@ -122,9 +125,10 @@ static gen6_gtt_pte_t byt_pte_encode(dma_addr_t addr,
 }
 
 static gen6_gtt_pte_t hsw_pte_encode(dma_addr_t addr,
-				     enum i915_cache_level level)
+				     enum i915_cache_level level,
+				     bool valid)
 {
-	gen6_gtt_pte_t pte = GEN6_PTE_VALID;
+	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= HSW_PTE_ADDR_ENCODE(addr);
 
 	if (level != I915_CACHE_NONE)
@@ -134,9 +138,10 @@ static gen6_gtt_pte_t hsw_pte_encode(dma_addr_t addr,
 }
 
 static gen6_gtt_pte_t iris_pte_encode(dma_addr_t addr,
-				      enum i915_cache_level level)
+				      enum i915_cache_level level,
+				      bool valid)
 {
-	gen6_gtt_pte_t pte = GEN6_PTE_VALID;
+	gen6_gtt_pte_t pte = valid ? GEN6_PTE_VALID : 0;
 	pte |= HSW_PTE_ADDR_ENCODE(addr);
 
 	switch (level) {
@@ -236,7 +241,8 @@ static int gen6_ppgtt_enable(struct drm_device *dev)
 /* PPGTT support for Sandybdrige/Gen6 and later */
 static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 				   unsigned first_entry,
-				   unsigned num_entries)
+				   unsigned num_entries,
+				   bool use_scratch)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
@@ -245,7 +251,7 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 	unsigned first_pte = first_entry % I915_PPGTT_PT_ENTRIES;
 	unsigned last_pte, i;
 
-	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC);
+	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, true);
 
 	while (num_entries) {
 		last_pte = first_pte + num_entries;
@@ -282,7 +288,7 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 		dma_addr_t page_addr;
 
 		page_addr = sg_page_iter_dma_address(&sg_iter);
-		pt_vaddr[act_pte] = vm->pte_encode(page_addr, cache_level);
+		pt_vaddr[act_pte] = vm->pte_encode(page_addr, cache_level, true);
 		if (++act_pte == I915_PPGTT_PT_ENTRIES) {
 			kunmap_atomic(pt_vaddr);
 			act_pt++;
@@ -367,7 +373,7 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 	}
 
 	ppgtt->base.clear_range(&ppgtt->base, 0,
-				ppgtt->num_pd_entries * I915_PPGTT_PT_ENTRIES);
+				ppgtt->num_pd_entries * I915_PPGTT_PT_ENTRIES, true);
 
 	ppgtt->pd_offset = first_pd_entry_in_global_pt * sizeof(gen6_gtt_pte_t);
 
@@ -444,7 +450,8 @@ void i915_ppgtt_unbind_object(struct i915_hw_ppgtt *ppgtt,
 {
 	ppgtt->base.clear_range(&ppgtt->base,
 				i915_gem_obj_ggtt_offset(obj) >> PAGE_SHIFT,
-				obj->base.size >> PAGE_SHIFT);
+				obj->base.size >> PAGE_SHIFT,
+				true);
 }
 
 extern int intel_iommu_gfx_mapped;
@@ -485,15 +492,65 @@ static void undo_idling(struct drm_i915_private *dev_priv, bool interruptible)
 		dev_priv->mm.interruptible = interruptible;
 }
 
+void i915_check_and_clear_faults(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_ring_buffer *ring;
+	int i;
+
+	if (INTEL_INFO(dev)->gen < 6)
+		return;
+
+	for_each_ring(ring, dev_priv, i) {
+		u32 fault_reg;
+		fault_reg = I915_READ(RING_FAULT_REG(ring));
+		if (fault_reg & RING_FAULT_VALID) {
+			DRM_DEBUG_DRIVER("Unexpected fault\n"
+					 "\tAddr: 0x%08lx\\n"
+					 "\tAddress space: %s\n"
+					 "\tSource ID: %d\n"
+					 "\tType: %d\n",
+					 fault_reg & PAGE_MASK,
+					 fault_reg & RING_FAULT_GTTSEL_MASK ? "GGTT" : "PPGTT",
+					 RING_FAULT_SRCID(fault_reg),
+					 RING_FAULT_FAULT_TYPE(fault_reg));
+			I915_WRITE(RING_FAULT_REG(ring),
+				   fault_reg & ~RING_FAULT_VALID);
+		}
+	}
+	POSTING_READ(RING_FAULT_REG(&dev_priv->ring[RCS]));
+}
+
+void i915_gem_suspend_gtt_mappings(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* Don't bother messing with faults pre GEN6 as we have little
+	 * documentation supporting that it's a good idea.
+	 */
+	if (INTEL_INFO(dev)->gen < 6)
+		return;
+
+	i915_check_and_clear_faults(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);
+}
+
 void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
 
+	i915_check_and_clear_faults(dev);
+
 	/* First fill our portion of the GTT with scratch pages */
 	dev_priv->gtt.base.clear_range(&dev_priv->gtt.base,
 				       dev_priv->gtt.base.start / PAGE_SIZE,
-				       dev_priv->gtt.base.total / PAGE_SIZE);
+				       dev_priv->gtt.base.total / PAGE_SIZE,
+				       true);
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
 		i915_gem_clflush_object(obj, obj->pin_display);
@@ -536,7 +593,7 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 
 	for_each_sg_page(st->sgl, &sg_iter, st->nents, 0) {
 		addr = sg_page_iter_dma_address(&sg_iter);
-		iowrite32(vm->pte_encode(addr, level), &gtt_entries[i]);
+		iowrite32(vm->pte_encode(addr, level, true), &gtt_entries[i]);
 		i++;
 	}
 
@@ -548,7 +605,7 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 	 */
 	if (i != 0)
 		WARN_ON(readl(&gtt_entries[i-1]) !=
-			vm->pte_encode(addr, level));
+			vm->pte_encode(addr, level, true));
 
 	/* This next bit makes the above posting read even more important. We
 	 * want to flush the TLBs only after we're certain all the PTE updates
@@ -560,7 +617,8 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 
 static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 				  unsigned int first_entry,
-				  unsigned int num_entries)
+				  unsigned int num_entries,
+				  bool use_scratch)
 {
 	struct drm_i915_private *dev_priv = vm->dev->dev_private;
 	gen6_gtt_pte_t scratch_pte, __iomem *gtt_base =
@@ -573,7 +631,8 @@ static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 		 first_entry, num_entries, max_entries))
 		num_entries = max_entries;
 
-	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC);
+	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, use_scratch);
+
 	for (i = 0; i < num_entries; i++)
 		iowrite32(scratch_pte, &gtt_base[i]);
 	readl(gtt_base);
@@ -594,7 +653,8 @@ static void i915_ggtt_insert_entries(struct i915_address_space *vm,
 
 static void i915_ggtt_clear_range(struct i915_address_space *vm,
 				  unsigned int first_entry,
-				  unsigned int num_entries)
+				  unsigned int num_entries,
+				  bool unused)
 {
 	intel_gtt_clear_range(first_entry, num_entries);
 }
@@ -622,7 +682,8 @@ void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj)
 
 	dev_priv->gtt.base.clear_range(&dev_priv->gtt.base,
 				       entry,
-				       obj->base.size >> PAGE_SHIFT);
+				       obj->base.size >> PAGE_SHIFT,
+				       true);
 
 	obj->has_global_gtt_mapping = 0;
 }
@@ -709,11 +770,11 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,
 		const unsigned long count = (hole_end - hole_start) / PAGE_SIZE;
 		DRM_DEBUG_KMS("clearing unused GTT space: [%lx, %lx]\n",
 			      hole_start, hole_end);
-		ggtt_vm->clear_range(ggtt_vm, hole_start / PAGE_SIZE, count);
+		ggtt_vm->clear_range(ggtt_vm, hole_start / PAGE_SIZE, count, true);
 	}
 
 	/* And finally clear the reserved guard page */
-	ggtt_vm->clear_range(ggtt_vm, end / PAGE_SIZE - 1, 1);
+	ggtt_vm->clear_range(ggtt_vm, end / PAGE_SIZE - 1, 1, true);
 }
 
 static bool
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index c159e1a6810f..ef9b35479f01 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -604,6 +604,10 @@
 #define   ARB_MODE_SWIZZLE_IVB	(1<<5)
 #define RENDER_HWS_PGA_GEN7	(0x04080)
 #define RING_FAULT_REG(ring)	(0x4094 + 0x100*(ring)->id)
+#define   RING_FAULT_GTTSEL_MASK (1<<11)
+#define   RING_FAULT_SRCID(x)	((x >> 3) & 0xff)
+#define   RING_FAULT_FAULT_TYPE(x) ((x >> 1) & 0x3)
+#define   RING_FAULT_VALID	(1<<0)
 #define DONE_REG		0x40b0
 #define BSD_HWS_PGA_GEN7	(0x04180)
 #define BLT_HWS_PGA_GEN7	(0x04280)
@@ -3881,6 +3885,9 @@
 #define GEN7_SQ_CHICKEN_MBCUNIT_CONFIG		0x9030
 #define  GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB	(1<<11)
 
+#define HSW_SCRATCH1				0xb038
+#define  HSW_SCRATCH1_L3_DATA_ATOMICS_DISABLE	(1<<27)
+
 #define HSW_FUSE_STRAP		0x42014
 #define  HSW_CDCLK_LIMIT	(1 << 24)
 
@@ -4276,7 +4283,9 @@
 #define FDI_RX_CHICKEN(pipe) _PIPE(pipe, _FDI_RXA_CHICKEN, _FDI_RXB_CHICKEN)
 
 #define SOUTH_DSPCLK_GATE_D	0xc2020
+#define  PCH_DPLUNIT_CLOCK_GATE_DISABLE (1<<30)
 #define  PCH_DPLSUNIT_CLOCK_GATE_DISABLE (1<<29)
+#define  PCH_CPUNIT_CLOCK_GATE_DISABLE (1<<14)
 #define  PCH_LP_PARTITION_LEVEL_DISABLE  (1<<12)
 
 /* CPU: FDI_TX */
@@ -4728,6 +4737,9 @@
 #define GEN7_ROW_CHICKEN2_GT2		0xf4f4
 #define   DOP_CLOCK_GATING_DISABLE	(1<<0)
 
+#define HSW_ROW_CHICKEN3		0xe49c
+#define  HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE    (1 << 6)
+
 #define G4X_AUD_VID_DID			(dev_priv->info->display_mmio_offset + 0x62020)
 #define INTEL_AUDIO_DEVCL		0x808629FB
 #define INTEL_AUDIO_DEVBLC		0x80862801
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index e5822e79f912..581fb4b2f766 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -3941,8 +3941,6 @@ static void intel_connector_check_state(struct intel_connector *connector)
  * consider. */
 void intel_connector_dpms(struct drm_connector *connector, int mode)
 {
-	struct intel_encoder *encoder = intel_attached_encoder(connector);
-
 	/* All the simple cases only support two dpms states. */
 	if (mode != DRM_MODE_DPMS_ON)
 		mode = DRM_MODE_DPMS_OFF;
@@ -3953,10 +3951,8 @@ void intel_connector_dpms(struct drm_connector *connector, int mode)
 	connector->dpms = mode;
 
 	/* Only need to change hw state when actually enabled */
-	if (encoder->base.crtc)
-		intel_encoder_dpms(encoder, mode);
-	else
-		WARN_ON(encoder->connectors_active != false);
+	if (connector->encoder)
+		intel_encoder_dpms(to_intel_encoder(connector->encoder), mode);
 
 	intel_modeset_check_state(connector->dev);
 }
@@ -10049,33 +10045,6 @@ static void i915_disable_vga(struct drm_device *dev)
 	POSTING_READ(vga_reg);
 }
 
-static void i915_enable_vga_mem(struct drm_device *dev)
-{
-	/* Enable VGA memory on Intel HD */
-	if (HAS_PCH_SPLIT(dev)) {
-		vga_get_uninterruptible(dev->pdev, VGA_RSRC_LEGACY_IO);
-		outb(inb(VGA_MSR_READ) | VGA_MSR_MEM_EN, VGA_MSR_WRITE);
-		vga_set_legacy_decoding(dev->pdev, VGA_RSRC_LEGACY_IO |
-						   VGA_RSRC_LEGACY_MEM |
-						   VGA_RSRC_NORMAL_IO |
-						   VGA_RSRC_NORMAL_MEM);
-		vga_put(dev->pdev, VGA_RSRC_LEGACY_IO);
-	}
-}
-
-void i915_disable_vga_mem(struct drm_device *dev)
-{
-	/* Disable VGA memory on Intel HD */
-	if (HAS_PCH_SPLIT(dev)) {
-		vga_get_uninterruptible(dev->pdev, VGA_RSRC_LEGACY_IO);
-		outb(inb(VGA_MSR_READ) & ~VGA_MSR_MEM_EN, VGA_MSR_WRITE);
-		vga_set_legacy_decoding(dev->pdev, VGA_RSRC_LEGACY_IO |
-						   VGA_RSRC_NORMAL_IO |
-						   VGA_RSRC_NORMAL_MEM);
-		vga_put(dev->pdev, VGA_RSRC_LEGACY_IO);
-	}
-}
-
 void intel_modeset_init_hw(struct drm_device *dev)
 {
 	intel_init_power_well(dev);
@@ -10354,7 +10323,6 @@ void i915_redisable_vga(struct drm_device *dev)
 	if (I915_READ(vga_reg) != VGA_DISP_DISABLE) {
 		DRM_DEBUG_KMS("Something enabled VGA plane, disabling it\n");
 		i915_disable_vga(dev);
-		i915_disable_vga_mem(dev);
 	}
 }
 
@@ -10568,8 +10536,6 @@ void intel_modeset_cleanup(struct drm_device *dev)
 
 	intel_disable_fbc(dev);
 
-	i915_enable_vga_mem(dev);
-
 	intel_disable_gt_powersave(dev);
 
 	ironlake_teardown_rc6(dev);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 79c14e298ba6..2c555f91bfae 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1467,7 +1467,7 @@ static void intel_edp_psr_setup(struct intel_dp *intel_dp)
 
 	/* Avoid continuous PSR exit by masking memup and hpd */
 	I915_WRITE(EDP_PSR_DEBUG_CTL, EDP_PSR_DEBUG_MASK_MEMUP |
-		   EDP_PSR_DEBUG_MASK_HPD);
+		   EDP_PSR_DEBUG_MASK_HPD | EDP_PSR_DEBUG_MASK_LPSP);
 
 	intel_dp->psr_setup_done = true;
 }
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 28cae80495e2..9b7b68fd5d47 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -793,6 +793,5 @@ extern void hsw_pc8_disable_interrupts(struct drm_device *dev);
 extern void hsw_pc8_restore_interrupts(struct drm_device *dev);
 extern void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv);
 extern void intel_aux_display_runtime_put(struct drm_i915_private *dev_priv);
-extern void i915_disable_vga_mem(struct drm_device *dev);
 
 #endif /* __INTEL_DRV_H__ */
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index dd176b7296c1..26c2ea3e985c 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3864,8 +3864,6 @@ static void valleyview_enable_rps(struct drm_device *dev)
 				      dev_priv->rps.rpe_delay),
 			 dev_priv->rps.rpe_delay);
 
-	INIT_DELAYED_WORK(&dev_priv->rps.vlv_work, vlv_rps_timer_work);
-
 	valleyview_set_rps(dev_priv->dev, dev_priv->rps.rpe_delay);
 
 	gen6_enable_rps_interrupts(dev);
@@ -4761,7 +4759,9 @@ static void cpt_init_clock_gating(struct drm_device *dev)
 	 * gating for the panel power sequencer or it will fail to
 	 * start up when no ports are active.
 	 */
-	I915_WRITE(SOUTH_DSPCLK_GATE_D, PCH_DPLSUNIT_CLOCK_GATE_DISABLE);
+	I915_WRITE(SOUTH_DSPCLK_GATE_D, PCH_DPLSUNIT_CLOCK_GATE_DISABLE |
+		   PCH_DPLUNIT_CLOCK_GATE_DISABLE |
+		   PCH_CPUNIT_CLOCK_GATE_DISABLE);
 	I915_WRITE(SOUTH_CHICKEN2, I915_READ(SOUTH_CHICKEN2) |
 		   DPLS_EDP_PPS_FIX_DIS);
 	/* The below fixes the weird display corruption, a few pixels shifted
@@ -4955,6 +4955,11 @@ static void haswell_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN7_L3_CHICKEN_MODE_REGISTER,
 			GEN7_WA_L3_CHICKEN_MODE);
 
+	/* L3 caching of data atomics doesn't work -- disable it. */
+	I915_WRITE(HSW_SCRATCH1, HSW_SCRATCH1_L3_DATA_ATOMICS_DISABLE);
+	I915_WRITE(HSW_ROW_CHICKEN3,
+		   _MASKED_BIT_ENABLE(HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE));
+
 	/* This is required by WaCatErrorRejectionIssue:hsw */
 	I915_WRITE(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG,
 			I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) |
@@ -5681,5 +5686,7 @@ void intel_pm_init(struct drm_device *dev)
 
 	INIT_DELAYED_WORK(&dev_priv->rps.delayed_resume_work,
 			  intel_gen6_powersave_work);
+
+	INIT_DELAYED_WORK(&dev_priv->rps.vlv_work, vlv_rps_timer_work);
 }