Merge branch 'drm-intel-next' of git://git.freedesktop.org/git/drm-intel into drm-next

- fine-grained display power domains for byt (Imre)
- runtime pm prep patches for !hsw from Paulo
- WiZ hashing flag updates from Ville
- ppgtt setup cleanup and enabling of full 4G range on bdw (Ben)
- fixes from Jesse for the inherited intial config code
- gpu reset code improvements from Mika
- per-pipe num_planes refactoring from Damien
- stability fixes around bdw forcewake handling and other bdw w/a from Mika
  Ken
- and as usual a pile of smaller fixes all over

* 'drm-intel-next' of git://git.freedesktop.org/git/drm-intel: (107 commits)
  drm/i915: Go OCD on the Makefile
  drm/i915: Implement command buffer parsing logic
  drm/i915: Refactor shmem pread setup
  drm/i915: Avoid div by zero when pixel clock is large
  drm/i915: power domains: add vlv power wells
  drm/i915: factor out intel_set_cpu_fifo_underrun_reporting_nolock
  drm/i915: vlv: factor out valleyview_display_irq_install
  drm/i915: sanity check power well sw state against hw state
  drm/i915: factor out reset_vblank_counter
  drm/i915: sanitize PUNIT register macro definitions
  drm/i915: vlv: keep first level vblank IRQs masked
  drm/i915: check pipe power domain when reading its hw state
  drm/i915: check port power domain when reading the encoder hw state
  drm/i915: get port power domain in connector detect handlers
  drm/i915: add port power domains
  drm/i915: add noop power well handlers instead of NULL checking them
  drm/i915: split power well 'set' handler to separate enable/disable/sync_hw
  drm/i915: add init power domain to always-on power wells
  drm/i915: move power domain macros to intel_pm.c
  drm/i915: Disable full ppgtt by default
  ...

29 files changed, 2896 insertions, 1019 deletions

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index f33902ff2c22..b1445b73465b 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -3,58 +3,69 @@
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
 ccflags-y := -Iinclude/drm
-i915-y := i915_drv.o i915_dma.o i915_irq.o \
-	  i915_gpu_error.o \
+
+# Please keep these build lists sorted!
+
+# core driver code
+i915-y := i915_drv.o \
+	  i915_params.o \
           i915_suspend.o \
-	  i915_gem.o \
+	  i915_sysfs.o \
+	  intel_pm.o
+i915-$(CONFIG_COMPAT)   += i915_ioc32.o
+i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o
+
+# GEM code
+i915-y += i915_cmd_parser.o \
 	  i915_gem_context.o \
 	  i915_gem_debug.o \
+	  i915_gem_dmabuf.o \
 	  i915_gem_evict.o \
 	  i915_gem_execbuffer.o \
 	  i915_gem_gtt.o \
+	  i915_gem.o \
 	  i915_gem_stolen.o \
 	  i915_gem_tiling.o \
-	  i915_params.o \
-	  i915_sysfs.o \
+	  i915_gpu_error.o \
+	  i915_irq.o \
 	  i915_trace_points.o \
-	  i915_ums.o \
+	  intel_ringbuffer.o \
+	  intel_uncore.o
+
+# modesetting core code
+i915-y += intel_bios.o \
 	  intel_display.o \
-	  intel_crt.o \
-	  intel_lvds.o \
-	  intel_dsi.o \
-	  intel_dsi_cmd.o \
-	  intel_dsi_pll.o \
-	  intel_bios.o \
-	  intel_ddi.o \
-	  intel_dp.o \
-	  intel_hdmi.o \
-	  intel_sdvo.o \
 	  intel_modes.o \
-	  intel_panel.o \
-	  intel_pm.o \
-	  intel_i2c.o \
-	  intel_tv.o \
-	  intel_dvo.o \
-	  intel_ringbuffer.o \
 	  intel_overlay.o \
-	  intel_sprite.o \
 	  intel_sideband.o \
-	  intel_uncore.o \
+	  intel_sprite.o
+i915-$(CONFIG_ACPI)		+= intel_acpi.o intel_opregion.o
+i915-$(CONFIG_DRM_I915_FBDEV)	+= intel_fbdev.o
+
+# modesetting output/encoder code
+i915-y += dvo_ch7017.o \
 	  dvo_ch7xxx.o \
-	  dvo_ch7017.o \
 	  dvo_ivch.o \
-	  dvo_tfp410.o \
-	  dvo_sil164.o \
 	  dvo_ns2501.o \
-	  i915_gem_dmabuf.o
-
-i915-$(CONFIG_COMPAT)   += i915_ioc32.o
-
-i915-$(CONFIG_ACPI)	+= intel_acpi.o intel_opregion.o
-
-i915-$(CONFIG_DRM_I915_FBDEV) += intel_fbdev.o
+	  dvo_sil164.o \
+	  dvo_tfp410.o \
+	  intel_crt.o \
+	  intel_ddi.o \
+	  intel_dp.o \
+	  intel_dsi_cmd.o \
+	  intel_dsi.o \
+	  intel_dsi_pll.o \
+	  intel_dvo.o \
+	  intel_hdmi.o \
+	  intel_i2c.o \
+	  intel_lvds.o \
+	  intel_panel.o \
+	  intel_sdvo.o \
+	  intel_tv.o
 
-i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o
+# legacy horrors
+i915-y += i915_dma.o \
+	  i915_ums.o
 
 obj-$(CONFIG_DRM_I915)  += i915.o
 
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
new file mode 100644
index 000000000000..7a5756e9bb86
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -0,0 +1,485 @@
+/*
+ * Copyright © 2013 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *    Brad Volkin <bradley.d.volkin@intel.com>
+ *
+ */
+
+#include "i915_drv.h"
+
+/**
+ * DOC: i915 batch buffer command parser
+ *
+ * Motivation:
+ * Certain OpenGL features (e.g. transform feedback, performance monitoring)
+ * require userspace code to submit batches containing commands such as
+ * MI_LOAD_REGISTER_IMM to access various registers. Unfortunately, some
+ * generations of the hardware will noop these commands in "unsecure" batches
+ * (which includes all userspace batches submitted via i915) even though the
+ * commands may be safe and represent the intended programming model of the
+ * device.
+ *
+ * The software command parser is similar in operation to the command parsing
+ * done in hardware for unsecure batches. However, the software parser allows
+ * some operations that would be noop'd by hardware, if the parser determines
+ * the operation is safe, and submits the batch as "secure" to prevent hardware
+ * parsing.
+ *
+ * Threats:
+ * At a high level, the hardware (and software) checks attempt to prevent
+ * granting userspace undue privileges. There are three categories of privilege.
+ *
+ * First, commands which are explicitly defined as privileged or which should
+ * only be used by the kernel driver. The parser generally rejects such
+ * commands, though it may allow some from the drm master process.
+ *
+ * Second, commands which access registers. To support correct/enhanced
+ * userspace functionality, particularly certain OpenGL extensions, the parser
+ * provides a whitelist of registers which userspace may safely access (for both
+ * normal and drm master processes).
+ *
+ * Third, commands which access privileged memory (i.e. GGTT, HWS page, etc).
+ * The parser always rejects such commands.
+ *
+ * The majority of the problematic commands fall in the MI_* range, with only a
+ * few specific commands on each ring (e.g. PIPE_CONTROL and MI_FLUSH_DW).
+ *
+ * Implementation:
+ * Each ring maintains tables of commands and registers which the parser uses in
+ * scanning batch buffers submitted to that ring.
+ *
+ * Since the set of commands that the parser must check for is significantly
+ * smaller than the number of commands supported, the parser tables contain only
+ * those commands required by the parser. This generally works because command
+ * opcode ranges have standard command length encodings. So for commands that
+ * the parser does not need to check, it can easily skip them. This is
+ * implementated via a per-ring length decoding vfunc.
+ *
+ * Unfortunately, there are a number of commands that do not follow the standard
+ * length encoding for their opcode range, primarily amongst the MI_* commands.
+ * To handle this, the parser provides a way to define explicit "skip" entries
+ * in the per-ring command tables.
+ *
+ * Other command table entries map fairly directly to high level categories
+ * mentioned above: rejected, master-only, register whitelist. The parser
+ * implements a number of checks, including the privileged memory checks, via a
+ * general bitmasking mechanism.
+ */
+
+static u32 gen7_render_get_cmd_length_mask(u32 cmd_header)
+{
+	u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT;
+	u32 subclient =
+		(cmd_header & INSTR_SUBCLIENT_MASK) >> INSTR_SUBCLIENT_SHIFT;
+
+	if (client == INSTR_MI_CLIENT)
+		return 0x3F;
+	else if (client == INSTR_RC_CLIENT) {
+		if (subclient == INSTR_MEDIA_SUBCLIENT)
+			return 0xFFFF;
+		else
+			return 0xFF;
+	}
+
+	DRM_DEBUG_DRIVER("CMD: Abnormal rcs cmd length! 0x%08X\n", cmd_header);
+	return 0;
+}
+
+static u32 gen7_bsd_get_cmd_length_mask(u32 cmd_header)
+{
+	u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT;
+	u32 subclient =
+		(cmd_header & INSTR_SUBCLIENT_MASK) >> INSTR_SUBCLIENT_SHIFT;
+
+	if (client == INSTR_MI_CLIENT)
+		return 0x3F;
+	else if (client == INSTR_RC_CLIENT) {
+		if (subclient == INSTR_MEDIA_SUBCLIENT)
+			return 0xFFF;
+		else
+			return 0xFF;
+	}
+
+	DRM_DEBUG_DRIVER("CMD: Abnormal bsd cmd length! 0x%08X\n", cmd_header);
+	return 0;
+}
+
+static u32 gen7_blt_get_cmd_length_mask(u32 cmd_header)
+{
+	u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT;
+
+	if (client == INSTR_MI_CLIENT)
+		return 0x3F;
+	else if (client == INSTR_BC_CLIENT)
+		return 0xFF;
+
+	DRM_DEBUG_DRIVER("CMD: Abnormal blt cmd length! 0x%08X\n", cmd_header);
+	return 0;
+}
+
+static void validate_cmds_sorted(struct intel_ring_buffer *ring)
+{
+	int i;
+
+	if (!ring->cmd_tables || ring->cmd_table_count == 0)
+		return;
+
+	for (i = 0; i < ring->cmd_table_count; i++) {
+		const struct drm_i915_cmd_table *table = &ring->cmd_tables[i];
+		u32 previous = 0;
+		int j;
+
+		for (j = 0; j < table->count; j++) {
+			const struct drm_i915_cmd_descriptor *desc =
+				&table->table[i];
+			u32 curr = desc->cmd.value & desc->cmd.mask;
+
+			if (curr < previous)
+				DRM_ERROR("CMD: table not sorted ring=%d table=%d entry=%d cmd=0x%08X prev=0x%08X\n",
+					  ring->id, i, j, curr, previous);
+
+			previous = curr;
+		}
+	}
+}
+
+static void check_sorted(int ring_id, const u32 *reg_table, int reg_count)
+{
+	int i;
+	u32 previous = 0;
+
+	for (i = 0; i < reg_count; i++) {
+		u32 curr = reg_table[i];
+
+		if (curr < previous)
+			DRM_ERROR("CMD: table not sorted ring=%d entry=%d reg=0x%08X prev=0x%08X\n",
+				  ring_id, i, curr, previous);
+
+		previous = curr;
+	}
+}
+
+static void validate_regs_sorted(struct intel_ring_buffer *ring)
+{
+	check_sorted(ring->id, ring->reg_table, ring->reg_count);
+	check_sorted(ring->id, ring->master_reg_table, ring->master_reg_count);
+}
+
+/**
+ * i915_cmd_parser_init_ring() - set cmd parser related fields for a ringbuffer
+ * @ring: the ringbuffer to initialize
+ *
+ * Optionally initializes fields related to batch buffer command parsing in the
+ * struct intel_ring_buffer based on whether the platform requires software
+ * command parsing.
+ */
+void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring)
+{
+	if (!IS_GEN7(ring->dev))
+		return;
+
+	switch (ring->id) {
+	case RCS:
+		ring->get_cmd_length_mask = gen7_render_get_cmd_length_mask;
+		break;
+	case VCS:
+		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
+		break;
+	case BCS:
+		ring->get_cmd_length_mask = gen7_blt_get_cmd_length_mask;
+		break;
+	case VECS:
+		/* VECS can use the same length_mask function as VCS */
+		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
+		break;
+	default:
+		DRM_ERROR("CMD: cmd_parser_init with unknown ring: %d\n",
+			  ring->id);
+		BUG();
+	}
+
+	validate_cmds_sorted(ring);
+	validate_regs_sorted(ring);
+}
+
+static const struct drm_i915_cmd_descriptor*
+find_cmd_in_table(const struct drm_i915_cmd_table *table,
+		  u32 cmd_header)
+{
+	int i;
+
+	for (i = 0; i < table->count; i++) {
+		const struct drm_i915_cmd_descriptor *desc = &table->table[i];
+		u32 masked_cmd = desc->cmd.mask & cmd_header;
+		u32 masked_value = desc->cmd.value & desc->cmd.mask;
+
+		if (masked_cmd == masked_value)
+			return desc;
+	}
+
+	return NULL;
+}
+
+/*
+ * Returns a pointer to a descriptor for the command specified by cmd_header.
+ *
+ * The caller must supply space for a default descriptor via the default_desc
+ * parameter. If no descriptor for the specified command exists in the ring's
+ * command parser tables, this function fills in default_desc based on the
+ * ring's default length encoding and returns default_desc.
+ */
+static const struct drm_i915_cmd_descriptor*
+find_cmd(struct intel_ring_buffer *ring,
+	 u32 cmd_header,
+	 struct drm_i915_cmd_descriptor *default_desc)
+{
+	u32 mask;
+	int i;
+
+	for (i = 0; i < ring->cmd_table_count; i++) {
+		const struct drm_i915_cmd_descriptor *desc;
+
+		desc = find_cmd_in_table(&ring->cmd_tables[i], cmd_header);
+		if (desc)
+			return desc;
+	}
+
+	mask = ring->get_cmd_length_mask(cmd_header);
+	if (!mask)
+		return NULL;
+
+	BUG_ON(!default_desc);
+	default_desc->flags = CMD_DESC_SKIP;
+	default_desc->length.mask = mask;
+
+	return default_desc;
+}
+
+static bool valid_reg(const u32 *table, int count, u32 addr)
+{
+	if (table && count != 0) {
+		int i;
+
+		for (i = 0; i < count; i++) {
+			if (table[i] == addr)
+				return true;
+		}
+	}
+
+	return false;
+}
+
+static u32 *vmap_batch(struct drm_i915_gem_object *obj)
+{
+	int i;
+	void *addr = NULL;
+	struct sg_page_iter sg_iter;
+	struct page **pages;
+
+	pages = drm_malloc_ab(obj->base.size >> PAGE_SHIFT, sizeof(*pages));
+	if (pages == NULL) {
+		DRM_DEBUG_DRIVER("Failed to get space for pages\n");
+		goto finish;
+	}
+
+	i = 0;
+	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) {
+		pages[i] = sg_page_iter_page(&sg_iter);
+		i++;
+	}
+
+	addr = vmap(pages, i, 0, PAGE_KERNEL);
+	if (addr == NULL) {
+		DRM_DEBUG_DRIVER("Failed to vmap pages\n");
+		goto finish;
+	}
+
+finish:
+	if (pages)
+		drm_free_large(pages);
+	return (u32*)addr;
+}
+
+/**
+ * i915_needs_cmd_parser() - should a given ring use software command parsing?
+ * @ring: the ring in question
+ *
+ * Only certain platforms require software batch buffer command parsing, and
+ * only when enabled via module paramter.
+ *
+ * Return: true if the ring requires software command parsing
+ */
+bool i915_needs_cmd_parser(struct intel_ring_buffer *ring)
+{
+	/* No command tables indicates a platform without parsing */
+	if (!ring->cmd_tables)
+		return false;
+
+	return (i915.enable_cmd_parser == 1);
+}
+
+#define LENGTH_BIAS 2
+
+/**
+ * i915_parse_cmds() - parse a submitted batch buffer for privilege violations
+ * @ring: the ring on which the batch is to execute
+ * @batch_obj: the batch buffer in question
+ * @batch_start_offset: byte offset in the batch at which execution starts
+ * @is_master: is the submitting process the drm master?
+ *
+ * Parses the specified batch buffer looking for privilege violations as
+ * described in the overview.
+ *
+ * Return: non-zero if the parser finds violations or otherwise fails
+ */
+int i915_parse_cmds(struct intel_ring_buffer *ring,
+		    struct drm_i915_gem_object *batch_obj,
+		    u32 batch_start_offset,
+		    bool is_master)
+{
+	int ret = 0;
+	u32 *cmd, *batch_base, *batch_end;
+	struct drm_i915_cmd_descriptor default_desc = { 0 };
+	int needs_clflush = 0;
+
+	ret = i915_gem_obj_prepare_shmem_read(batch_obj, &needs_clflush);
+	if (ret) {
+		DRM_DEBUG_DRIVER("CMD: failed to prep read\n");
+		return ret;
+	}
+
+	batch_base = vmap_batch(batch_obj);
+	if (!batch_base) {
+		DRM_DEBUG_DRIVER("CMD: Failed to vmap batch\n");
+		i915_gem_object_unpin_pages(batch_obj);
+		return -ENOMEM;
+	}
+
+	if (needs_clflush)
+		drm_clflush_virt_range((char *)batch_base, batch_obj->base.size);
+
+	cmd = batch_base + (batch_start_offset / sizeof(*cmd));
+	batch_end = cmd + (batch_obj->base.size / sizeof(*batch_end));
+
+	while (cmd < batch_end) {
+		const struct drm_i915_cmd_descriptor *desc;
+		u32 length;
+
+		if (*cmd == MI_BATCH_BUFFER_END)
+			break;
+
+		desc = find_cmd(ring, *cmd, &default_desc);
+		if (!desc) {
+			DRM_DEBUG_DRIVER("CMD: Unrecognized command: 0x%08X\n",
+					 *cmd);
+			ret = -EINVAL;
+			break;
+		}
+
+		if (desc->flags & CMD_DESC_FIXED)
+			length = desc->length.fixed;
+		else
+			length = ((*cmd & desc->length.mask) + LENGTH_BIAS);
+
+		if ((batch_end - cmd) < length) {
+			DRM_DEBUG_DRIVER("CMD: Command length exceeds batch length: 0x%08X length=%d batchlen=%ld\n",
+					 *cmd,
+					 length,
+					 batch_end - cmd);
+			ret = -EINVAL;
+			break;
+		}
+
+		if (desc->flags & CMD_DESC_REJECT) {
+			DRM_DEBUG_DRIVER("CMD: Rejected command: 0x%08X\n", *cmd);
+			ret = -EINVAL;
+			break;
+		}
+
+		if ((desc->flags & CMD_DESC_MASTER) && !is_master) {
+			DRM_DEBUG_DRIVER("CMD: Rejected master-only command: 0x%08X\n",
+					 *cmd);
+			ret = -EINVAL;
+			break;
+		}
+
+		if (desc->flags & CMD_DESC_REGISTER) {
+			u32 reg_addr = cmd[desc->reg.offset] & desc->reg.mask;
+
+			if (!valid_reg(ring->reg_table,
+				       ring->reg_count, reg_addr)) {
+				if (!is_master ||
+				    !valid_reg(ring->master_reg_table,
+					       ring->master_reg_count,
+					       reg_addr)) {
+					DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
+							 reg_addr,
+							 *cmd,
+							 ring->id);
+					ret = -EINVAL;
+					break;
+				}
+			}
+		}
+
+		if (desc->flags & CMD_DESC_BITMASK) {
+			int i;
+
+			for (i = 0; i < MAX_CMD_DESC_BITMASKS; i++) {
+				u32 dword;
+
+				if (desc->bits[i].mask == 0)
+					break;
+
+				dword = cmd[desc->bits[i].offset] &
+					desc->bits[i].mask;
+
+				if (dword != desc->bits[i].expected) {
+					DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X for bitmask 0x%08X (exp=0x%08X act=0x%08X) (ring=%d)\n",
+							 *cmd,
+							 desc->bits[i].mask,
+							 desc->bits[i].expected,
+							 dword, ring->id);
+					ret = -EINVAL;
+					break;
+				}
+			}
+
+			if (ret)
+				break;
+		}
+
+		cmd += length;
+	}
+
+	if (cmd >= batch_end) {
+		DRM_DEBUG_DRIVER("CMD: Got to the end of the buffer w/o a BBE cmd!\n");
+		ret = -EINVAL;
+	}
+
+	vunmap(batch_base);
+
+	i915_gem_object_unpin_pages(batch_obj);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index d90a70744d93..a90d31c78643 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -602,7 +602,6 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 	intel_runtime_pm_get(dev_priv);
 
 	if (INTEL_INFO(dev)->gen >= 8) {
-		int i;
 		seq_printf(m, "Master Interrupt Control:\t%08x\n",
 			   I915_READ(GEN8_MASTER_IRQ));
 
@@ -615,16 +614,16 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 				   i, I915_READ(GEN8_GT_IER(i)));
 		}
 
-		for_each_pipe(i) {
+		for_each_pipe(pipe) {
 			seq_printf(m, "Pipe %c IMR:\t%08x\n",
-				   pipe_name(i),
-				   I915_READ(GEN8_DE_PIPE_IMR(i)));
+				   pipe_name(pipe),
+				   I915_READ(GEN8_DE_PIPE_IMR(pipe)));
 			seq_printf(m, "Pipe %c IIR:\t%08x\n",
-				   pipe_name(i),
-				   I915_READ(GEN8_DE_PIPE_IIR(i)));
+				   pipe_name(pipe),
+				   I915_READ(GEN8_DE_PIPE_IIR(pipe)));
 			seq_printf(m, "Pipe %c IER:\t%08x\n",
-				   pipe_name(i),
-				   I915_READ(GEN8_DE_PIPE_IER(i)));
+				   pipe_name(pipe),
+				   I915_READ(GEN8_DE_PIPE_IER(pipe)));
 		}
 
 		seq_printf(m, "Display Engine port interrupt mask:\t%08x\n",
@@ -1348,6 +1347,8 @@ static int i915_fbc_status(struct seq_file *m, void *unused)
 		return 0;
 	}
 
+	intel_runtime_pm_get(dev_priv);
+
 	if (intel_fbc_enabled(dev)) {
 		seq_puts(m, "FBC enabled\n");
 	} else {
@@ -1391,6 +1392,9 @@ static int i915_fbc_status(struct seq_file *m, void *unused)
 		}
 		seq_putc(m, '\n');
 	}
+
+	intel_runtime_pm_put(dev_priv);
+
 	return 0;
 }
 
@@ -1405,11 +1409,15 @@ static int i915_ips_status(struct seq_file *m, void *unused)
 		return 0;
 	}
 
+	intel_runtime_pm_get(dev_priv);
+
 	if (IS_BROADWELL(dev) || I915_READ(IPS_CTL) & IPS_ENABLE)
 		seq_puts(m, "enabled\n");
 	else
 		seq_puts(m, "disabled\n");
 
+	intel_runtime_pm_put(dev_priv);
+
 	return 0;
 }
 
@@ -1420,6 +1428,8 @@ static int i915_sr_status(struct seq_file *m, void *unused)
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	bool sr_enabled = false;
 
+	intel_runtime_pm_get(dev_priv);
+
 	if (HAS_PCH_SPLIT(dev))
 		sr_enabled = I915_READ(WM1_LP_ILK) & WM1_LP_SR_EN;
 	else if (IS_CRESTLINE(dev) || IS_I945G(dev) || IS_I945GM(dev))
@@ -1429,6 +1439,8 @@ static int i915_sr_status(struct seq_file *m, void *unused)
 	else if (IS_PINEVIEW(dev))
 		sr_enabled = I915_READ(DSPFW3) & PINEVIEW_SELF_REFRESH_EN;
 
+	intel_runtime_pm_put(dev_priv);
+
 	seq_printf(m, "self-refresh: %s\n",
 		   sr_enabled ? "enabled" : "disabled");
 
@@ -1468,7 +1480,7 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	int ret;
+	int ret = 0;
 	int gpu_freq, ia_freq;
 
 	if (!(IS_GEN6(dev) || IS_GEN7(dev))) {
@@ -1476,12 +1488,13 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused)
 		return 0;
 	}
 
+	intel_runtime_pm_get(dev_priv);
+
 	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
 
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
 	if (ret)
-		return ret;
-	intel_runtime_pm_get(dev_priv);
+		goto out;
 
 	seq_puts(m, "GPU freq (MHz)\tEffective CPU freq (MHz)\tEffective Ring freq (MHz)\n");
 
@@ -1498,10 +1511,11 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused)
 			   ((ia_freq >> 8) & 0xff) * 100);
 	}
 
-	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
-	return 0;
+out:
+	intel_runtime_pm_put(dev_priv);
+	return ret;
 }
 
 static int i915_gfxec(struct seq_file *m, void *unused)
@@ -1757,7 +1771,7 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 		return;
 
 	seq_printf(m, "Page directories: %d\n", ppgtt->num_pd_pages);
-	seq_printf(m, "Page tables: %d\n", ppgtt->num_pt_pages);
+	seq_printf(m, "Page tables: %d\n", ppgtt->num_pd_entries);
 	for_each_ring(ring, dev_priv, unused) {
 		seq_printf(m, "%s\n", ring->name);
 		for (i = 0; i < 4; i++) {
@@ -1972,12 +1986,16 @@ static int i915_energy_uJ(struct seq_file *m, void *data)
 	if (INTEL_INFO(dev)->gen < 6)
 		return -ENODEV;
 
+	intel_runtime_pm_get(dev_priv);
+
 	rdmsrl(MSR_RAPL_POWER_UNIT, power);
 	power = (power & 0x1f00) >> 8;
 	units = 1000000 / (1 << power); /* convert to uJ */
 	power = I915_READ(MCH_SECP_NRG_STTS);
 	power *= units;
 
+	intel_runtime_pm_put(dev_priv);
+
 	seq_printf(m, "%llu", (long long unsigned)power);
 
 	return 0;
@@ -1997,7 +2015,7 @@ static int i915_pc8_status(struct seq_file *m, void *unused)
 	mutex_lock(&dev_priv->pc8.lock);
 	seq_printf(m, "Requirements met: %s\n",
 		   yesno(dev_priv->pc8.requirements_met));
-	seq_printf(m, "GPU idle: %s\n", yesno(dev_priv->pc8.gpu_idle));
+	seq_printf(m, "GPU idle: %s\n", yesno(!dev_priv->mm.busy));
 	seq_printf(m, "Disable count: %d\n", dev_priv->pc8.disable_count);
 	seq_printf(m, "IRQs disabled: %s\n",
 		   yesno(dev_priv->pc8.irqs_disabled));
@@ -2030,6 +2048,28 @@ static const char *power_domain_str(enum intel_display_power_domain domain)
 		return "TRANSCODER_C";
 	case POWER_DOMAIN_TRANSCODER_EDP:
 		return "TRANSCODER_EDP";
+	case POWER_DOMAIN_PORT_DDI_A_2_LANES:
+		return "PORT_DDI_A_2_LANES";
+	case POWER_DOMAIN_PORT_DDI_A_4_LANES:
+		return "PORT_DDI_A_4_LANES";
+	case POWER_DOMAIN_PORT_DDI_B_2_LANES:
+		return "PORT_DDI_B_2_LANES";
+	case POWER_DOMAIN_PORT_DDI_B_4_LANES:
+		return "PORT_DDI_B_4_LANES";
+	case POWER_DOMAIN_PORT_DDI_C_2_LANES:
+		return "PORT_DDI_C_2_LANES";
+	case POWER_DOMAIN_PORT_DDI_C_4_LANES:
+		return "PORT_DDI_C_4_LANES";
+	case POWER_DOMAIN_PORT_DDI_D_2_LANES:
+		return "PORT_DDI_D_2_LANES";
+	case POWER_DOMAIN_PORT_DDI_D_4_LANES:
+		return "PORT_DDI_D_4_LANES";
+	case POWER_DOMAIN_PORT_DSI:
+		return "PORT_DSI";
+	case POWER_DOMAIN_PORT_CRT:
+		return "PORT_CRT";
+	case POWER_DOMAIN_PORT_OTHER:
+		return "PORT_OTHER";
 	case POWER_DOMAIN_VGA:
 		return "VGA";
 	case POWER_DOMAIN_AUDIO:
@@ -2180,6 +2220,7 @@ static void intel_connector_info(struct seq_file *m,
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct intel_encoder *intel_encoder = intel_connector->encoder;
+	struct drm_display_mode *mode;
 
 	seq_printf(m, "connector %d: type %s, status: %s\n",
 		   connector->base.id, drm_get_connector_name(connector),
@@ -2202,6 +2243,9 @@ static void intel_connector_info(struct seq_file *m,
 	else if (intel_encoder->type == INTEL_OUTPUT_LVDS)
 		intel_lvds_info(m, intel_connector);
 
+	seq_printf(m, "\tmodes:\n");
+	list_for_each_entry(mode, &connector->modes, head)
+		intel_seq_print_mode(m, 2, mode);
 }
 
 static int i915_display_info(struct seq_file *m, void *unused)
@@ -3167,9 +3211,8 @@ i915_wedged_set(void *data, u64 val)
 {
 	struct drm_device *dev = data;
 
-	DRM_INFO("Manually setting wedged to %llu\n", val);
-	i915_handle_error(dev, val);
-
+	i915_handle_error(dev, val,
+			  "Manually setting wedged to %llu", val);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 7688abc83fc0..e4d2b9f15ae2 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1321,12 +1321,12 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	if (ret)
 		goto cleanup_vga_switcheroo;
 
+	intel_power_domains_init_hw(dev_priv);
+
 	ret = drm_irq_install(dev);
 	if (ret)
 		goto cleanup_gem_stolen;
 
-	intel_power_domains_init_hw(dev);
-
 	/* Important: The output setup functions called by modeset_init need
 	 * working irqs for e.g. gmbus and dp aux transfers. */
 	intel_modeset_init(dev);
@@ -1343,7 +1343,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	/* FIXME: do pre/post-mode set stuff in core KMS code */
 	dev->vblank_disable_allowed = true;
 	if (INTEL_INFO(dev)->num_pipes == 0) {
-		intel_display_power_put(dev, POWER_DOMAIN_VGA);
+		intel_display_power_put(dev_priv, POWER_DOMAIN_VGA);
 		return 0;
 	}
 
@@ -1381,7 +1381,7 @@ cleanup_gem:
 	WARN_ON(dev_priv->mm.aliasing_ppgtt);
 	drm_mm_takedown(&dev_priv->gtt.base.mm);
 cleanup_power:
-	intel_display_power_put(dev, POWER_DOMAIN_VGA);
+	intel_display_power_put(dev_priv, POWER_DOMAIN_VGA);
 	drm_irq_uninstall(dev);
 cleanup_gem_stolen:
 	i915_gem_cleanup_stolen(dev);
@@ -1480,12 +1480,16 @@ static void intel_device_info_runtime_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_device_info *info;
+	enum pipe pipe;
 
 	info = (struct intel_device_info *)&dev_priv->info;
 
-	info->num_sprites = 1;
 	if (IS_VALLEYVIEW(dev))
-		info->num_sprites = 2;
+		for_each_pipe(pipe)
+			info->num_sprites[pipe] = 2;
+	else
+		for_each_pipe(pipe)
+			info->num_sprites[pipe] = 1;
 
 	if (i915.disable_display) {
 		DRM_INFO("Display disabled (module parameter)\n");
@@ -1702,7 +1706,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 			goto out_gem_unload;
 	}
 
-	intel_power_domains_init(dev);
+	intel_power_domains_init(dev_priv);
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
 		ret = i915_load_modeset_init(dev);
@@ -1731,7 +1735,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	return 0;
 
 out_power_well:
-	intel_power_domains_remove(dev);
+	intel_power_domains_remove(dev_priv);
 	drm_vblank_cleanup(dev);
 out_gem_unload:
 	if (dev_priv->mm.inactive_shrinker.scan_objects)
@@ -1781,8 +1785,8 @@ int i915_driver_unload(struct drm_device *dev)
 	/* The i915.ko module is still not prepared to be loaded when
 	 * the power well is not enabled, so just enable it in case
 	 * we're going to unload/reload. */
-	intel_display_set_init_power(dev, true);
-	intel_power_domains_remove(dev);
+	intel_display_set_init_power(dev_priv, true);
+	intel_power_domains_remove(dev_priv);
 
 	i915_teardown_sysfs(dev);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 2d05d7ce4c29..6ac91fcdb4f8 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -265,6 +265,7 @@ static const struct intel_device_info intel_broadwell_d_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	.has_llc = 1,
 	.has_ddi = 1,
+	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
 };
 
@@ -274,6 +275,7 @@ static const struct intel_device_info intel_broadwell_m_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	.has_llc = 1,
 	.has_ddi = 1,
+	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
 };
 
@@ -401,15 +403,13 @@ bool i915_semaphore_is_enabled(struct drm_device *dev)
 	if (INTEL_INFO(dev)->gen < 6)
 		return false;
 
-	/* Until we get further testing... */
-	if (IS_GEN8(dev)) {
-		WARN_ON(!i915.preliminary_hw_support);
-		return false;
-	}
-
 	if (i915.semaphores >= 0)
 		return i915.semaphores;
 
+	/* Until we get further testing... */
+	if (IS_GEN8(dev))
+		return false;
+
 #ifdef CONFIG_INTEL_IOMMU
 	/* Enable semaphores on SNB when IO remapping is off */
 	if (INTEL_INFO(dev)->gen == 6 && intel_iommu_gfx_mapped)
@@ -434,7 +434,7 @@ static int i915_drm_freeze(struct drm_device *dev)
 	/* We do a lot of poking in a lot of registers, make sure they work
 	 * properly. */
 	hsw_disable_package_c8(dev_priv);
-	intel_display_set_init_power(dev, true);
+	intel_display_set_init_power(dev_priv, true);
 
 	drm_kms_helper_poll_disable(dev);
 
@@ -477,6 +477,8 @@ static int i915_drm_freeze(struct drm_device *dev)
 	intel_fbdev_set_suspend(dev, FBINFO_STATE_SUSPENDED);
 	console_unlock();
 
+	dev_priv->suspend_count++;
+
 	return 0;
 }
 
@@ -556,7 +558,7 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 		mutex_unlock(&dev->struct_mutex);
 	}
 
-	intel_power_domains_init_hw(dev);
+	intel_power_domains_init_hw(dev_priv);
 
 	i915_restore_state(dev);
 	intel_opregion_setup(dev);
@@ -847,6 +849,7 @@ static int i915_runtime_suspend(struct device *device)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	WARN_ON(!HAS_RUNTIME_PM(dev));
+	assert_force_wake_inactive(dev_priv);
 
 	DRM_DEBUG_KMS("Suspending device\n");
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 05cfcc163a72..2a319ba91a71 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -79,7 +79,7 @@ enum plane {
 };
 #define plane_name(p) ((p) + 'A')
 
-#define sprite_name(p, s) ((p) * INTEL_INFO(dev)->num_sprites + (s) + 'A')
+#define sprite_name(p, s) ((p) * INTEL_INFO(dev)->num_sprites[(p)] + (s) + 'A')
 
 enum port {
 	PORT_A = 0,
@@ -114,6 +114,17 @@ enum intel_display_power_domain {
 	POWER_DOMAIN_TRANSCODER_B,
 	POWER_DOMAIN_TRANSCODER_C,
 	POWER_DOMAIN_TRANSCODER_EDP,
+	POWER_DOMAIN_PORT_DDI_A_2_LANES,
+	POWER_DOMAIN_PORT_DDI_A_4_LANES,
+	POWER_DOMAIN_PORT_DDI_B_2_LANES,
+	POWER_DOMAIN_PORT_DDI_B_4_LANES,
+	POWER_DOMAIN_PORT_DDI_C_2_LANES,
+	POWER_DOMAIN_PORT_DDI_C_4_LANES,
+	POWER_DOMAIN_PORT_DDI_D_2_LANES,
+	POWER_DOMAIN_PORT_DDI_D_4_LANES,
+	POWER_DOMAIN_PORT_DSI,
+	POWER_DOMAIN_PORT_CRT,
+	POWER_DOMAIN_PORT_OTHER,
 	POWER_DOMAIN_VGA,
 	POWER_DOMAIN_AUDIO,
 	POWER_DOMAIN_INIT,
@@ -121,8 +132,6 @@ enum intel_display_power_domain {
 	POWER_DOMAIN_NUM,
 };
 
-#define POWER_DOMAIN_MASK (BIT(POWER_DOMAIN_NUM) - 1)
-
 #define POWER_DOMAIN_PIPE(pipe) ((pipe) + POWER_DOMAIN_PIPE_A)
 #define POWER_DOMAIN_PIPE_PANEL_FITTER(pipe) \
 		((pipe) + POWER_DOMAIN_PIPE_A_PANEL_FITTER)
@@ -130,14 +139,6 @@ enum intel_display_power_domain {
 	((tran) == TRANSCODER_EDP ? POWER_DOMAIN_TRANSCODER_EDP : \
 	 (tran) + POWER_DOMAIN_TRANSCODER_A)
 
-#define HSW_ALWAYS_ON_POWER_DOMAINS (		\
-	BIT(POWER_DOMAIN_PIPE_A) |		\
-	BIT(POWER_DOMAIN_TRANSCODER_EDP))
-#define BDW_ALWAYS_ON_POWER_DOMAINS (		\
-	BIT(POWER_DOMAIN_PIPE_A) |		\
-	BIT(POWER_DOMAIN_TRANSCODER_EDP) |	\
-	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER))
-
 enum hpd_pin {
 	HPD_NONE = 0,
 	HPD_PORT_A = HPD_NONE, /* PORT_A is internal */
@@ -159,6 +160,7 @@ enum hpd_pin {
 	 I915_GEM_DOMAIN_VERTEX)
 
 #define for_each_pipe(p) for ((p) = 0; (p) < INTEL_INFO(dev)->num_pipes; (p)++)
+#define for_each_sprite(p, s) for ((s) = 0; (s) < INTEL_INFO(dev)->num_sprites[(p)]; (s)++)
 
 #define for_each_encoder_on_crtc(dev, __crtc, intel_encoder) \
 	list_for_each_entry((intel_encoder), &(dev)->mode_config.encoder_list, base.head) \
@@ -303,6 +305,10 @@ struct drm_i915_error_state {
 	struct kref ref;
 	struct timeval time;
 
+	char error_msg[128];
+	u32 reset_count;
+	u32 suspend_count;
+
 	/* Generic register state */
 	u32 eir;
 	u32 pgtbl_er;
@@ -360,7 +366,7 @@ struct drm_i915_error_state {
 			int page_count;
 			u32 gtt_offset;
 			u32 *pages[0];
-		} *ringbuffer, *batchbuffer, *ctx, *hws_page;
+		} *ringbuffer, *batchbuffer, *wa_batchbuffer, *ctx, *hws_page;
 
 		struct drm_i915_error_request {
 			long jiffies;
@@ -375,6 +381,9 @@ struct drm_i915_error_state {
 				u32 pp_dir_base;
 			};
 		} vm_info;
+
+		pid_t pid;
+		char comm[TASK_COMM_LEN];
 	} ring[I915_NUM_RINGS];
 	struct drm_i915_error_buffer {
 		u32 size;
@@ -499,7 +508,7 @@ struct intel_uncore {
 	unsigned fw_rendercount;
 	unsigned fw_mediacount;
 
-	struct delayed_work force_wake_work;
+	struct timer_list force_wake_timer;
 };
 
 #define DEV_INFO_FOR_EACH_FLAG(func, sep) \
@@ -534,7 +543,7 @@ struct intel_uncore {
 struct intel_device_info {
 	u32 display_mmio_offset;
 	u8 num_pipes:3;
-	u8 num_sprites:2;
+	u8 num_sprites[I915_MAX_PIPES];
 	u8 gen;
 	u8 ring_mask; /* Rings supported by the HW */
 	DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG, SEP_SEMICOLON);
@@ -652,12 +661,12 @@ struct i915_address_space {
 				     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,
+			    uint64_t start,
+			    uint64_t length,
 			    bool use_scratch);
 	void (*insert_entries)(struct i915_address_space *vm,
 			       struct sg_table *st,
-			       unsigned int first_entry,
+			       uint64_t start,
 			       enum i915_cache_level cache_level);
 	void (*cleanup)(struct i915_address_space *vm);
 };
@@ -691,21 +700,21 @@ struct i915_gtt {
 };
 #define gtt_total_entries(gtt) ((gtt).base.total >> PAGE_SHIFT)
 
+#define GEN8_LEGACY_PDPS 4
 struct i915_hw_ppgtt {
 	struct i915_address_space base;
 	struct kref ref;
 	struct drm_mm_node node;
 	unsigned num_pd_entries;
+	unsigned num_pd_pages; /* gen8+ */
 	union {
 		struct page **pt_pages;
-		struct page *gen8_pt_pages;
+		struct page **gen8_pt_pages[GEN8_LEGACY_PDPS];
 	};
 	struct page *pd_pages;
-	int num_pd_pages;
-	int num_pt_pages;
 	union {
 		uint32_t pd_offset;
-		dma_addr_t pd_dma_addr[4];
+		dma_addr_t pd_dma_addr[GEN8_LEGACY_PDPS];
 	};
 	union {
 		dma_addr_t *pt_dma_addr;
@@ -1016,6 +1025,36 @@ struct intel_ilk_power_mgmt {
 	struct drm_i915_gem_object *renderctx;
 };
 
+struct drm_i915_private;
+struct i915_power_well;
+
+struct i915_power_well_ops {
+	/*
+	 * Synchronize the well's hw state to match the current sw state, for
+	 * example enable/disable it based on the current refcount. Called
+	 * during driver init and resume time, possibly after first calling
+	 * the enable/disable handlers.
+	 */
+	void (*sync_hw)(struct drm_i915_private *dev_priv,
+			struct i915_power_well *power_well);
+	/*
+	 * Enable the well and resources that depend on it (for example
+	 * interrupts located on the well). Called after the 0->1 refcount
+	 * transition.
+	 */
+	void (*enable)(struct drm_i915_private *dev_priv,
+		       struct i915_power_well *power_well);
+	/*
+	 * Disable the well and resources that depend on it. Called after
+	 * the 1->0 refcount transition.
+	 */
+	void (*disable)(struct drm_i915_private *dev_priv,
+			struct i915_power_well *power_well);
+	/* Returns the hw enabled state. */
+	bool (*is_enabled)(struct drm_i915_private *dev_priv,
+			   struct i915_power_well *power_well);
+};
+
 /* Power well structure for haswell */
 struct i915_power_well {
 	const char *name;
@@ -1023,11 +1062,8 @@ struct i915_power_well {
 	/* power well enable/disable usage count */
 	int count;
 	unsigned long domains;
-	void *data;
-	void (*set)(struct drm_device *dev, struct i915_power_well *power_well,
-		    bool enable);
-	bool (*is_enabled)(struct drm_device *dev,
-			   struct i915_power_well *power_well);
+	unsigned long data;
+	const struct i915_power_well_ops *ops;
 };
 
 struct i915_power_domains {
@@ -1124,6 +1160,14 @@ struct i915_gem_mm {
 	 */
 	bool interruptible;
 
+	/**
+	 * Is the GPU currently considered idle, or busy executing userspace
+	 * requests?  Whilst idle, we attempt to power down the hardware and
+	 * display clocks. In order to reduce the effect on performance, there
+	 * is a slight delay before we do so.
+	 */
+	bool busy;
+
 	/** Bit 6 swizzling required for X tiling */
 	uint32_t bit_6_swizzle_x;
 	/** Bit 6 swizzling required for Y tiling */
@@ -1313,11 +1357,10 @@ struct ilk_wm_values {
  * Ideally every piece of our code that needs PC8+ disabled would call
  * hsw_disable_package_c8, which would increment disable_count and prevent the
  * system from reaching PC8+. But we don't have a symmetric way to do this for
- * everything, so we have the requirements_met and gpu_idle variables. When we
- * switch requirements_met or gpu_idle to true we decrease disable_count, and
- * increase it in the opposite case. The requirements_met variable is true when
- * all the CRTCs, encoders and the power well are disabled. The gpu_idle
- * variable is true when the GPU is idle.
+ * everything, so we have the requirements_met variable. When we switch
+ * requirements_met to true we decrease disable_count, and increase it in the
+ * opposite case. The requirements_met variable is true when all the CRTCs,
+ * encoders and the power well are disabled.
  *
  * In addition to everything, we only actually enable PC8+ if disable_count
  * stays at zero for at least some seconds. This is implemented with the
@@ -1340,7 +1383,6 @@ struct ilk_wm_values {
  */
 struct i915_package_c8 {
 	bool requirements_met;
-	bool gpu_idle;
 	bool irqs_disabled;
 	/* Only true after the delayed work task actually enables it. */
 	bool enabled;
@@ -1427,6 +1469,8 @@ typedef struct drm_i915_private {
 	/* protects the irq masks */
 	spinlock_t irq_lock;
 
+	bool display_irqs_enabled;
+
 	/* To control wakeup latency, e.g. for irq-driven dp aux transfers. */
 	struct pm_qos_request pm_qos;
 
@@ -1594,6 +1638,8 @@ typedef struct drm_i915_private {
 	struct i915_dri1_state dri1;
 	/* Old ums support infrastructure, same warning applies. */
 	struct i915_ums_state ums;
+
+	u32 suspend_count;
 } drm_i915_private_t;
 
 static inline struct drm_i915_private *to_i915(const struct drm_device *dev)
@@ -1745,7 +1791,6 @@ struct drm_i915_gem_object {
 	/** for phy allocated objects */
 	struct drm_i915_gem_phys_object *phys_obj;
 };
-#define to_gem_object(obj) (&((struct drm_i915_gem_object *)(obj))->base)
 
 #define to_intel_bo(x) container_of(x, struct drm_i915_gem_object, base)
 
@@ -1791,6 +1836,7 @@ struct drm_i915_gem_request {
 
 struct drm_i915_file_private {
 	struct drm_i915_private *dev_priv;
+	struct drm_file *file;
 
 	struct {
 		spinlock_t lock;
@@ -1803,6 +1849,90 @@ struct drm_i915_file_private {
 	atomic_t rps_wait_boost;
 };
 
+/*
+ * A command that requires special handling by the command parser.
+ */
+struct drm_i915_cmd_descriptor {
+	/*
+	 * Flags describing how the command parser processes the command.
+	 *
+	 * CMD_DESC_FIXED: The command has a fixed length if this is set,
+	 *                 a length mask if not set
+	 * CMD_DESC_SKIP: The command is allowed but does not follow the
+	 *                standard length encoding for the opcode range in
+	 *                which it falls
+	 * CMD_DESC_REJECT: The command is never allowed
+	 * CMD_DESC_REGISTER: The command should be checked against the
+	 *                    register whitelist for the appropriate ring
+	 * CMD_DESC_MASTER: The command is allowed if the submitting process
+	 *                  is the DRM master
+	 */
+	u32 flags;
+#define CMD_DESC_FIXED    (1<<0)
+#define CMD_DESC_SKIP     (1<<1)
+#define CMD_DESC_REJECT   (1<<2)
+#define CMD_DESC_REGISTER (1<<3)
+#define CMD_DESC_BITMASK  (1<<4)
+#define CMD_DESC_MASTER   (1<<5)
+
+	/*
+	 * The command's unique identification bits and the bitmask to get them.
+	 * This isn't strictly the opcode field as defined in the spec and may
+	 * also include type, subtype, and/or subop fields.
+	 */
+	struct {
+		u32 value;
+		u32 mask;
+	} cmd;
+
+	/*
+	 * The command's length. The command is either fixed length (i.e. does
+	 * not include a length field) or has a length field mask. The flag
+	 * CMD_DESC_FIXED indicates a fixed length. Otherwise, the command has
+	 * a length mask. All command entries in a command table must include
+	 * length information.
+	 */
+	union {
+		u32 fixed;
+		u32 mask;
+	} length;
+
+	/*
+	 * Describes where to find a register address in the command to check
+	 * against the ring's register whitelist. Only valid if flags has the
+	 * CMD_DESC_REGISTER bit set.
+	 */
+	struct {
+		u32 offset;
+		u32 mask;
+	} reg;
+
+#define MAX_CMD_DESC_BITMASKS 3
+	/*
+	 * Describes command checks where a particular dword is masked and
+	 * compared against an expected value. If the command does not match
+	 * the expected value, the parser rejects it. Only valid if flags has
+	 * the CMD_DESC_BITMASK bit set. Only entries where mask is non-zero
+	 * are valid.
+	 */
+	struct {
+		u32 offset;
+		u32 mask;
+		u32 expected;
+	} bits[MAX_CMD_DESC_BITMASKS];
+};
+
+/*
+ * A table of commands requiring special handling by the command parser.
+ *
+ * Each ring has an array of tables. Each table consists of an array of command
+ * descriptors, which must be sorted with command opcodes in ascending order.
+ */
+struct drm_i915_cmd_table {
+	const struct drm_i915_cmd_descriptor *table;
+	int count;
+};
+
 #define INTEL_INFO(dev)	(&to_i915(dev)->info)
 
 #define IS_I830(dev)		((dev)->pdev->device == 0x3577)
@@ -1965,6 +2095,7 @@ struct i915_params {
 	int enable_pc8;
 	int pc8_timeout;
 	int invert_brightness;
+	int enable_cmd_parser;
 	/* leave bools at the end to not create holes */
 	bool enable_hangcheck;
 	bool fastboot;
@@ -2004,7 +2135,9 @@ extern void intel_console_resume(struct work_struct *work);
 
 /* i915_irq.c */
 void i915_queue_hangcheck(struct drm_device *dev);
-void i915_handle_error(struct drm_device *dev, bool wedged);
+__printf(3, 4)
+void i915_handle_error(struct drm_device *dev, bool wedged,
+		       const char *fmt, ...);
 
 void gen6_set_pm_mask(struct drm_i915_private *dev_priv, u32 pm_iir,
 							int new_delay);
@@ -2025,6 +2158,9 @@ void
 i915_disable_pipestat(drm_i915_private_t *dev_priv, enum pipe pipe,
 		      u32 status_mask);
 
+void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv);
+void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv);
+
 /* i915_gem.c */
 int i915_gem_init_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
@@ -2097,6 +2233,9 @@ void i915_gem_release_all_mmaps(struct drm_i915_private *dev_priv);
 void i915_gem_release_mmap(struct drm_i915_gem_object *obj);
 void i915_gem_lastclose(struct drm_device *dev);
 
+int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
+				    int *needs_clflush);
+
 int __must_check i915_gem_object_get_pages(struct drm_i915_gem_object *obj);
 static inline struct page *i915_gem_object_get_page(struct drm_i915_gem_object *obj, int n)
 {
@@ -2163,8 +2302,10 @@ i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
 	}
 }
 
+struct drm_i915_gem_request *
+i915_gem_find_active_request(struct intel_ring_buffer *ring);
+
 bool i915_gem_retire_requests(struct drm_device *dev);
-void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
 int __must_check i915_gem_check_wedge(struct i915_gpu_error *error,
 				      bool interruptible);
 static inline bool i915_reset_in_progress(struct i915_gpu_error *error)
@@ -2365,63 +2506,7 @@ static inline void i915_gem_chipset_flush(struct drm_device *dev)
 		intel_gtt_chipset_flush();
 }
 int i915_gem_init_ppgtt(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt);
-static inline bool intel_enable_ppgtt(struct drm_device *dev, bool full)
-{
-	if (i915.enable_ppgtt == 0 || !HAS_ALIASING_PPGTT(dev))
-		return false;
-
-	if (i915.enable_ppgtt == 1 && full)
-		return false;
-
-#ifdef CONFIG_INTEL_IOMMU
-	/* Disable ppgtt on SNB if VT-d is on. */
-	if (INTEL_INFO(dev)->gen == 6 && intel_iommu_gfx_mapped) {
-		DRM_INFO("Disabling PPGTT because VT-d is on\n");
-		return false;
-	}
-#endif
-
-	if (full)
-		return HAS_PPGTT(dev);
-	else
-		return HAS_ALIASING_PPGTT(dev);
-}
-
-static inline void ppgtt_release(struct kref *kref)
-{
-	struct i915_hw_ppgtt *ppgtt = container_of(kref, struct i915_hw_ppgtt, ref);
-	struct drm_device *dev = ppgtt->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_address_space *vm = &ppgtt->base;
-
-	if (ppgtt == dev_priv->mm.aliasing_ppgtt ||
-	    (list_empty(&vm->active_list) && list_empty(&vm->inactive_list))) {
-		ppgtt->base.cleanup(&ppgtt->base);
-		return;
-	}
-
-	/*
-	 * Make sure vmas are unbound before we take down the drm_mm
-	 *
-	 * FIXME: Proper refcounting should take care of this, this shouldn't be
-	 * needed at all.
-	 */
-	if (!list_empty(&vm->active_list)) {
-		struct i915_vma *vma;
-
-		list_for_each_entry(vma, &vm->active_list, mm_list)
-			if (WARN_ON(list_empty(&vma->vma_link) ||
-				    list_is_singular(&vma->vma_link)))
-				break;
-
-		i915_gem_evict_vm(&ppgtt->base, true);
-	} else {
-		i915_gem_retire_requests(dev);
-		i915_gem_evict_vm(&ppgtt->base, false);
-	}
-
-	ppgtt->base.cleanup(&ppgtt->base);
-}
+bool intel_enable_ppgtt(struct drm_device *dev, bool full);
 
 /* i915_gem_stolen.c */
 int i915_gem_init_stolen(struct drm_device *dev);
@@ -2478,7 +2563,8 @@ static inline void i915_error_state_buf_release(
 {
 	kfree(eb->buf);
 }
-void i915_capture_error_state(struct drm_device *dev);
+void i915_capture_error_state(struct drm_device *dev, bool wedge,
+			      const char *error_msg);
 void i915_error_state_get(struct drm_device *dev,
 			  struct i915_error_state_file_priv *error_priv);
 void i915_error_state_put(struct i915_error_state_file_priv *error_priv);
@@ -2487,6 +2573,14 @@ void i915_destroy_error_state(struct drm_device *dev);
 void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone);
 const char *i915_cache_level_str(int type);
 
+/* i915_cmd_parser.c */
+void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring);
+bool i915_needs_cmd_parser(struct intel_ring_buffer *ring);
+int i915_parse_cmds(struct intel_ring_buffer *ring,
+		    struct drm_i915_gem_object *batch_obj,
+		    u32 batch_start_offset,
+		    bool is_master);
+
 /* i915_suspend.c */
 extern int i915_save_state(struct drm_device *dev);
 extern int i915_restore_state(struct drm_device *dev);
@@ -2565,6 +2659,7 @@ extern int intel_modeset_vga_set_state(struct drm_device *dev, bool state);
 extern void intel_modeset_setup_hw_state(struct drm_device *dev,
 					 bool force_restore);
 extern void i915_redisable_vga(struct drm_device *dev);
+extern void i915_redisable_vga_power_on(struct drm_device *dev);
 extern bool intel_fbc_enabled(struct drm_device *dev);
 extern void intel_disable_fbc(struct drm_device *dev);
 extern bool ironlake_set_drps(struct drm_device *dev, u8 val);
@@ -2599,6 +2694,7 @@ extern void intel_display_print_error_state(struct drm_i915_error_state_buf *e,
  */
 void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv, int fw_engine);
 void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv, int fw_engine);
+void assert_force_wake_inactive(struct drm_i915_private *dev_priv);
 
 int sandybridge_pcode_read(struct drm_i915_private *dev_priv, u8 mbox, u32 *val);
 int sandybridge_pcode_write(struct drm_i915_private *dev_priv, u8 mbox, u32 val);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 3618bb0cda0a..177c20722656 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -61,6 +61,7 @@ static unsigned long i915_gem_inactive_scan(struct shrinker *shrinker,
 static unsigned long i915_gem_purge(struct drm_i915_private *dev_priv, long target);
 static unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
 static void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
+static void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
 
 static bool cpu_cache_is_coherent(struct drm_device *dev,
 				  enum i915_cache_level level)
@@ -326,6 +327,42 @@ __copy_from_user_swizzled(char *gpu_vaddr, int gpu_offset,
 	return 0;
 }
 
+/*
+ * Pins the specified object's pages and synchronizes the object with
+ * GPU accesses. Sets needs_clflush to non-zero if the caller should
+ * flush the object from the CPU cache.
+ */
+int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
+				    int *needs_clflush)
+{
+	int ret;
+
+	*needs_clflush = 0;
+
+	if (!obj->base.filp)
+		return -EINVAL;
+
+	if (!(obj->base.read_domains & I915_GEM_DOMAIN_CPU)) {
+		/* If we're not in the cpu read domain, set ourself into the gtt
+		 * read domain and manually flush cachelines (if required). This
+		 * optimizes for the case when the gpu will dirty the data
+		 * anyway again before the next pread happens. */
+		*needs_clflush = !cpu_cache_is_coherent(obj->base.dev,
+							obj->cache_level);
+		ret = i915_gem_object_wait_rendering(obj, true);
+		if (ret)
+			return ret;
+	}
+
+	ret = i915_gem_object_get_pages(obj);
+	if (ret)
+		return ret;
+
+	i915_gem_object_pin_pages(obj);
+
+	return ret;
+}
+
 /* Per-page copy function for the shmem pread fastpath.
  * Flushes invalid cachelines before reading the target if
  * needs_clflush is set. */
@@ -423,23 +460,10 @@ i915_gem_shmem_pread(struct drm_device *dev,
 
 	obj_do_bit17_swizzling = i915_gem_object_needs_bit17_swizzle(obj);
 
-	if (!(obj->base.read_domains & I915_GEM_DOMAIN_CPU)) {
-		/* If we're not in the cpu read domain, set ourself into the gtt
-		 * read domain and manually flush cachelines (if required). This
-		 * optimizes for the case when the gpu will dirty the data
-		 * anyway again before the next pread happens. */
-		needs_clflush = !cpu_cache_is_coherent(dev, obj->cache_level);
-		ret = i915_gem_object_wait_rendering(obj, true);
-		if (ret)
-			return ret;
-	}
-
-	ret = i915_gem_object_get_pages(obj);
+	ret = i915_gem_obj_prepare_shmem_read(obj, &needs_clflush);
 	if (ret)
 		return ret;
 
-	i915_gem_object_pin_pages(obj);
-
 	offset = args->offset;
 
 	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents,
@@ -2148,7 +2172,6 @@ int __i915_add_request(struct intel_ring_buffer *ring,
 	drm_i915_private_t *dev_priv = ring->dev->dev_private;
 	struct drm_i915_gem_request *request;
 	u32 request_ring_position, request_start;
-	int was_empty;
 	int ret;
 
 	request_start = intel_ring_get_tail(ring);
@@ -2199,7 +2222,6 @@ int __i915_add_request(struct intel_ring_buffer *ring,
 		i915_gem_context_reference(request->ctx);
 
 	request->emitted_jiffies = jiffies;
-	was_empty = list_empty(&ring->request_list);
 	list_add_tail(&request->list, &ring->request_list);
 	request->file_priv = NULL;
 
@@ -2220,13 +2242,11 @@ int __i915_add_request(struct intel_ring_buffer *ring,
 	if (!dev_priv->ums.mm_suspended) {
 		i915_queue_hangcheck(ring->dev);
 
-		if (was_empty) {
-			cancel_delayed_work_sync(&dev_priv->mm.idle_work);
-			queue_delayed_work(dev_priv->wq,
-					   &dev_priv->mm.retire_work,
-					   round_jiffies_up_relative(HZ));
-			intel_mark_busy(dev_priv->dev);
-		}
+		cancel_delayed_work_sync(&dev_priv->mm.idle_work);
+		queue_delayed_work(dev_priv->wq,
+				   &dev_priv->mm.retire_work,
+				   round_jiffies_up_relative(HZ));
+		intel_mark_busy(dev_priv->dev);
 	}
 
 	if (out_seqno)
@@ -2259,14 +2279,13 @@ static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
 		return true;
 
 	if (elapsed <= DRM_I915_CTX_BAN_PERIOD) {
-		if (dev_priv->gpu_error.stop_rings == 0 &&
-		    i915_gem_context_is_default(ctx)) {
-			DRM_ERROR("gpu hanging too fast, banning!\n");
-		} else {
+		if (!i915_gem_context_is_default(ctx)) {
 			DRM_DEBUG("context hanging too fast, banning!\n");
+			return true;
+		} else if (dev_priv->gpu_error.stop_rings == 0) {
+			DRM_ERROR("gpu hanging too fast, banning!\n");
+			return true;
 		}
-
-		return true;
 	}
 
 	return false;
@@ -2303,11 +2322,13 @@ static void i915_gem_free_request(struct drm_i915_gem_request *request)
 	kfree(request);
 }
 
-static struct drm_i915_gem_request *
-i915_gem_find_first_non_complete(struct intel_ring_buffer *ring)
+struct drm_i915_gem_request *
+i915_gem_find_active_request(struct intel_ring_buffer *ring)
 {
 	struct drm_i915_gem_request *request;
-	const u32 completed_seqno = ring->get_seqno(ring, false);
+	u32 completed_seqno;
+
+	completed_seqno = ring->get_seqno(ring, false);
 
 	list_for_each_entry(request, &ring->request_list, list) {
 		if (i915_seqno_passed(completed_seqno, request->seqno))
@@ -2325,7 +2346,7 @@ static void i915_gem_reset_ring_status(struct drm_i915_private *dev_priv,
 	struct drm_i915_gem_request *request;
 	bool ring_hung;
 
-	request = i915_gem_find_first_non_complete(ring);
+	request = i915_gem_find_active_request(ring);
 
 	if (request == NULL)
 		return;
@@ -2417,7 +2438,7 @@ void i915_gem_reset(struct drm_device *dev)
 /**
  * This function clears the request list as sequence numbers are passed.
  */
-void
+static void
 i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
 {
 	uint32_t seqno;
@@ -2744,7 +2765,7 @@ int i915_vma_unbind(struct i915_vma *vma)
 
 	i915_gem_gtt_finish_object(obj);
 
-	list_del(&vma->mm_list);
+	list_del_init(&vma->mm_list);
 	/* Avoid an unnecessary call to unbind on rebind. */
 	if (i915_is_ggtt(vma->vm))
 		obj->map_and_fenceable = true;
@@ -4860,6 +4881,7 @@ int i915_gem_open(struct drm_device *dev, struct drm_file *file)
 
 	file->driver_priv = file_priv;
 	file_priv->dev_priv = dev->dev_private;
+	file_priv->file = file;
 
 	spin_lock_init(&file_priv->mm.lock);
 	INIT_LIST_HEAD(&file_priv->mm.request_list);
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index f8c21a6dd663..ce41cff84346 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -99,6 +99,50 @@
 static int do_switch(struct intel_ring_buffer *ring,
 		     struct i915_hw_context *to);
 
+static void do_ppgtt_cleanup(struct i915_hw_ppgtt *ppgtt)
+{
+	struct drm_device *dev = ppgtt->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct i915_address_space *vm = &ppgtt->base;
+
+	if (ppgtt == dev_priv->mm.aliasing_ppgtt ||
+	    (list_empty(&vm->active_list) && list_empty(&vm->inactive_list))) {
+		ppgtt->base.cleanup(&ppgtt->base);
+		return;
+	}
+
+	/*
+	 * Make sure vmas are unbound before we take down the drm_mm
+	 *
+	 * FIXME: Proper refcounting should take care of this, this shouldn't be
+	 * needed at all.
+	 */
+	if (!list_empty(&vm->active_list)) {
+		struct i915_vma *vma;
+
+		list_for_each_entry(vma, &vm->active_list, mm_list)
+			if (WARN_ON(list_empty(&vma->vma_link) ||
+				    list_is_singular(&vma->vma_link)))
+				break;
+
+		i915_gem_evict_vm(&ppgtt->base, true);
+	} else {
+		i915_gem_retire_requests(dev);
+		i915_gem_evict_vm(&ppgtt->base, false);
+	}
+
+	ppgtt->base.cleanup(&ppgtt->base);
+}
+
+static void ppgtt_release(struct kref *kref)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(kref, struct i915_hw_ppgtt, ref);
+
+	do_ppgtt_cleanup(ppgtt);
+	kfree(ppgtt);
+}
+
 static size_t get_context_alignment(struct drm_device *dev)
 {
 	if (IS_GEN6(dev))
@@ -714,7 +758,7 @@ unpin_out:
  * i915_switch_context() - perform a GPU context switch.
  * @ring: ring for which we'll execute the context switch
  * @file_priv: file_priv associated with the context, may be NULL
- * @id: context id number
+ * @to: the context to switch to
  *
  * The context life cycle is simple. The context refcount is incremented and
  * decremented by 1 and create and destroy. If the context is in use by the GPU,
@@ -746,9 +790,6 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 	struct i915_hw_context *ctx;
 	int ret;
 
-	if (!(dev->driver->driver_features & DRIVER_GEM))
-		return -ENODEV;
-
 	if (!HAS_HW_CONTEXTS(dev))
 		return -ENODEV;
 
@@ -775,9 +816,6 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 	struct i915_hw_context *ctx;
 	int ret;
 
-	if (!(dev->driver->driver_features & DRIVER_GEM))
-		return -ENODEV;
-
 	if (args->ctx_id == DEFAULT_CONTEXT_ID)
 		return -ENOENT;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index d7229ad2bd22..3851a1b1dc88 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1182,6 +1182,24 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	}
 	batch_obj->base.pending_read_domains |= I915_GEM_DOMAIN_COMMAND;
 
+	if (i915_needs_cmd_parser(ring)) {
+		ret = i915_parse_cmds(ring,
+				      batch_obj,
+				      args->batch_start_offset,
+				      file->is_master);
+		if (ret)
+			goto err;
+
+		/*
+		 * XXX: Actually do this when enabling batch copy...
+		 *
+		 * Set the DISPATCH_SECURE bit to remove the NON_SECURE bit
+		 * from MI_BATCH_BUFFER_START commands issued in the
+		 * dispatch_execbuffer implementations. We specifically don't
+		 * want that set when the command parser is enabled.
+		 */
+	}
+
 	/* snb/ivb/vlv conflate the "batch in ppgtt" bit with the "non-secure
 	 * batch" bit. Hence we need to pin secure batches into the global gtt.
 	 * hsw should have this fixed, but bdw mucks it up again. */
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 9673cffb07f3..63a6dc7a6bb6 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1,5 +1,6 @@
 /*
  * Copyright © 2010 Daniel Vetter
+ * Copyright © 2011-2014 Intel Corporation
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -29,6 +30,29 @@
 #include "i915_trace.h"
 #include "intel_drv.h"
 
+bool intel_enable_ppgtt(struct drm_device *dev, bool full)
+{
+	if (i915.enable_ppgtt == 0 || !HAS_ALIASING_PPGTT(dev))
+		return false;
+
+	if (i915.enable_ppgtt == 1 && full)
+		return false;
+
+#ifdef CONFIG_INTEL_IOMMU
+	/* Disable ppgtt on SNB if VT-d is on. */
+	if (INTEL_INFO(dev)->gen == 6 && intel_iommu_gfx_mapped) {
+		DRM_INFO("Disabling PPGTT because VT-d is on\n");
+		return false;
+	}
+#endif
+
+	/* Full ppgtt disabled by default for now due to issues. */
+	if (full)
+		return false; /* HAS_PPGTT(dev) */
+	else
+		return HAS_ALIASING_PPGTT(dev);
+}
+
 #define GEN6_PPGTT_PD_ENTRIES 512
 #define I915_PPGTT_PT_ENTRIES (PAGE_SIZE / sizeof(gen6_gtt_pte_t))
 typedef uint64_t gen8_gtt_pte_t;
@@ -64,7 +88,19 @@ typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
 
 #define GEN8_PTES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_gtt_pte_t))
 #define GEN8_PDES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_ppgtt_pde_t))
-#define GEN8_LEGACY_PDPS		4
+
+/* GEN8 legacy style addressis defined as a 3 level page table:
+ * 31:30 | 29:21 | 20:12 |  11:0
+ * PDPE  |  PDE  |  PTE  | offset
+ * The difference as compared to normal x86 3 level page table is the PDPEs are
+ * programmed via register.
+ */
+#define GEN8_PDPE_SHIFT			30
+#define GEN8_PDPE_MASK			0x3
+#define GEN8_PDE_SHIFT			21
+#define GEN8_PDE_MASK			0x1ff
+#define GEN8_PTE_SHIFT			12
+#define GEN8_PTE_MASK			0x1ff
 
 #define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
 #define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
@@ -254,84 +290,113 @@ static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
 }
 
 static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
-				   unsigned first_entry,
-				   unsigned num_entries,
+				   uint64_t start,
+				   uint64_t length,
 				   bool use_scratch)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	gen8_gtt_pte_t *pt_vaddr, scratch_pte;
-	unsigned act_pt = first_entry / GEN8_PTES_PER_PAGE;
-	unsigned first_pte = first_entry % GEN8_PTES_PER_PAGE;
+	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
+	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
+	unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
+	unsigned num_entries = length >> PAGE_SHIFT;
 	unsigned last_pte, i;
 
 	scratch_pte = gen8_pte_encode(ppgtt->base.scratch.addr,
 				      I915_CACHE_LLC, use_scratch);
 
 	while (num_entries) {
-		struct page *page_table = &ppgtt->gen8_pt_pages[act_pt];
+		struct page *page_table = ppgtt->gen8_pt_pages[pdpe][pde];
 
-		last_pte = first_pte + num_entries;
+		last_pte = pte + num_entries;
 		if (last_pte > GEN8_PTES_PER_PAGE)
 			last_pte = GEN8_PTES_PER_PAGE;
 
 		pt_vaddr = kmap_atomic(page_table);
 
-		for (i = first_pte; i < last_pte; i++)
+		for (i = pte; i < last_pte; i++) {
 			pt_vaddr[i] = scratch_pte;
+			num_entries--;
+		}
 
 		kunmap_atomic(pt_vaddr);
 
-		num_entries -= last_pte - first_pte;
-		first_pte = 0;
-		act_pt++;
+		pte = 0;
+		if (++pde == GEN8_PDES_PER_PAGE) {
+			pdpe++;
+			pde = 0;
+		}
 	}
 }
 
 static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 				      struct sg_table *pages,
-				      unsigned first_entry,
+				      uint64_t start,
 				      enum i915_cache_level cache_level)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	gen8_gtt_pte_t *pt_vaddr;
-	unsigned act_pt = first_entry / GEN8_PTES_PER_PAGE;
-	unsigned act_pte = first_entry % GEN8_PTES_PER_PAGE;
+	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
+	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
+	unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
 	struct sg_page_iter sg_iter;
 
 	pt_vaddr = NULL;
+
 	for_each_sg_page(pages->sgl, &sg_iter, pages->nents, 0) {
+		if (WARN_ON(pdpe >= GEN8_LEGACY_PDPS))
+			break;
+
 		if (pt_vaddr == NULL)
-			pt_vaddr = kmap_atomic(&ppgtt->gen8_pt_pages[act_pt]);
+			pt_vaddr = kmap_atomic(ppgtt->gen8_pt_pages[pdpe][pde]);
 
-		pt_vaddr[act_pte] =
+		pt_vaddr[pte] =
 			gen8_pte_encode(sg_page_iter_dma_address(&sg_iter),
 					cache_level, true);
-		if (++act_pte == GEN8_PTES_PER_PAGE) {
+		if (++pte == GEN8_PTES_PER_PAGE) {
 			kunmap_atomic(pt_vaddr);
 			pt_vaddr = NULL;
-			act_pt++;
-			act_pte = 0;
+			if (++pde == GEN8_PDES_PER_PAGE) {
+				pdpe++;
+				pde = 0;
+			}
+			pte = 0;
 		}
 	}
 	if (pt_vaddr)
 		kunmap_atomic(pt_vaddr);
 }
 
-static void gen8_ppgtt_free(struct i915_hw_ppgtt *ppgtt)
+static void gen8_free_page_tables(struct page **pt_pages)
 {
 	int i;
 
-	for (i = 0; i < ppgtt->num_pd_pages ; i++)
+	if (pt_pages == NULL)
+		return;
+
+	for (i = 0; i < GEN8_PDES_PER_PAGE; i++)
+		if (pt_pages[i])
+			__free_pages(pt_pages[i], 0);
+}
+
+static void gen8_ppgtt_free(const struct i915_hw_ppgtt *ppgtt)
+{
+	int i;
+
+	for (i = 0; i < ppgtt->num_pd_pages; i++) {
+		gen8_free_page_tables(ppgtt->gen8_pt_pages[i]);
+		kfree(ppgtt->gen8_pt_pages[i]);
 		kfree(ppgtt->gen8_pt_dma_addr[i]);
+	}
 
-	__free_pages(ppgtt->gen8_pt_pages, get_order(ppgtt->num_pt_pages << PAGE_SHIFT));
 	__free_pages(ppgtt->pd_pages, get_order(ppgtt->num_pd_pages << PAGE_SHIFT));
 }
 
 static void gen8_ppgtt_unmap_pages(struct i915_hw_ppgtt *ppgtt)
 {
+	struct pci_dev *hwdev = ppgtt->base.dev->pdev;
 	int i, j;
 
 	for (i = 0; i < ppgtt->num_pd_pages; i++) {
@@ -340,18 +405,14 @@ static void gen8_ppgtt_unmap_pages(struct i915_hw_ppgtt *ppgtt)
 		if (!ppgtt->pd_dma_addr[i])
 			continue;
 
-		pci_unmap_page(ppgtt->base.dev->pdev,
-			       ppgtt->pd_dma_addr[i],
-			       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+		pci_unmap_page(hwdev, ppgtt->pd_dma_addr[i], PAGE_SIZE,
+			       PCI_DMA_BIDIRECTIONAL);
 
 		for (j = 0; j < GEN8_PDES_PER_PAGE; j++) {
 			dma_addr_t addr = ppgtt->gen8_pt_dma_addr[i][j];
 			if (addr)
-				pci_unmap_page(ppgtt->base.dev->pdev,
-				       addr,
-				       PAGE_SIZE,
-				       PCI_DMA_BIDIRECTIONAL);
-
+				pci_unmap_page(hwdev, addr, PAGE_SIZE,
+					       PCI_DMA_BIDIRECTIONAL);
 		}
 	}
 }
@@ -368,88 +429,198 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 	gen8_ppgtt_free(ppgtt);
 }
 
-/**
- * GEN8 legacy ppgtt programming is accomplished through 4 PDP registers with a
- * net effect resembling a 2-level page table in normal x86 terms. Each PDP
- * represents 1GB of memory
- * 4 * 512 * 512 * 4096 = 4GB legacy 32b address space.
- *
- * TODO: Do something with the size parameter
- **/
-static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt, uint64_t size)
+static struct page **__gen8_alloc_page_tables(void)
 {
-	struct page *pt_pages;
-	int i, j, ret = -ENOMEM;
-	const int max_pdp = DIV_ROUND_UP(size, 1 << 30);
-	const int num_pt_pages = GEN8_PDES_PER_PAGE * max_pdp;
+	struct page **pt_pages;
+	int i;
 
-	if (size % (1<<30))
-		DRM_INFO("Pages will be wasted unless GTT size (%llu) is divisible by 1GB\n", size);
+	pt_pages = kcalloc(GEN8_PDES_PER_PAGE, sizeof(struct page *), GFP_KERNEL);
+	if (!pt_pages)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < GEN8_PDES_PER_PAGE; i++) {
+		pt_pages[i] = alloc_page(GFP_KERNEL);
+		if (!pt_pages[i])
+			goto bail;
+	}
 
-	/* FIXME: split allocation into smaller pieces. For now we only ever do
-	 * this once, but with full PPGTT, the multiple contiguous allocations
-	 * will be bad.
+	return pt_pages;
+
+bail:
+	gen8_free_page_tables(pt_pages);
+	kfree(pt_pages);
+	return ERR_PTR(-ENOMEM);
+}
+
+static int gen8_ppgtt_allocate_page_tables(struct i915_hw_ppgtt *ppgtt,
+					   const int max_pdp)
+{
+	struct page **pt_pages[GEN8_LEGACY_PDPS];
+	int i, ret;
+
+	for (i = 0; i < max_pdp; i++) {
+		pt_pages[i] = __gen8_alloc_page_tables();
+		if (IS_ERR(pt_pages[i])) {
+			ret = PTR_ERR(pt_pages[i]);
+			goto unwind_out;
+		}
+	}
+
+	/* NB: Avoid touching gen8_pt_pages until last to keep the allocation,
+	 * "atomic" - for cleanup purposes.
 	 */
+	for (i = 0; i < max_pdp; i++)
+		ppgtt->gen8_pt_pages[i] = pt_pages[i];
+
+	return 0;
+
+unwind_out:
+	while (i--) {
+		gen8_free_page_tables(pt_pages[i]);
+		kfree(pt_pages[i]);
+	}
+
+	return ret;
+}
+
+static int gen8_ppgtt_allocate_dma(struct i915_hw_ppgtt *ppgtt)
+{
+	int i;
+
+	for (i = 0; i < ppgtt->num_pd_pages; i++) {
+		ppgtt->gen8_pt_dma_addr[i] = kcalloc(GEN8_PDES_PER_PAGE,
+						     sizeof(dma_addr_t),
+						     GFP_KERNEL);
+		if (!ppgtt->gen8_pt_dma_addr[i])
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static int gen8_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt,
+						const int max_pdp)
+{
 	ppgtt->pd_pages = alloc_pages(GFP_KERNEL, get_order(max_pdp << PAGE_SHIFT));
 	if (!ppgtt->pd_pages)
 		return -ENOMEM;
 
-	pt_pages = alloc_pages(GFP_KERNEL, get_order(num_pt_pages << PAGE_SHIFT));
-	if (!pt_pages) {
+	ppgtt->num_pd_pages = 1 << get_order(max_pdp << PAGE_SHIFT);
+	BUG_ON(ppgtt->num_pd_pages > GEN8_LEGACY_PDPS);
+
+	return 0;
+}
+
+static int gen8_ppgtt_alloc(struct i915_hw_ppgtt *ppgtt,
+			    const int max_pdp)
+{
+	int ret;
+
+	ret = gen8_ppgtt_allocate_page_directories(ppgtt, max_pdp);
+	if (ret)
+		return ret;
+
+	ret = gen8_ppgtt_allocate_page_tables(ppgtt, max_pdp);
+	if (ret) {
 		__free_pages(ppgtt->pd_pages, get_order(max_pdp << PAGE_SHIFT));
-		return -ENOMEM;
+		return ret;
 	}
 
-	ppgtt->gen8_pt_pages = pt_pages;
-	ppgtt->num_pd_pages = 1 << get_order(max_pdp << PAGE_SHIFT);
-	ppgtt->num_pt_pages = 1 << get_order(num_pt_pages << PAGE_SHIFT);
 	ppgtt->num_pd_entries = max_pdp * GEN8_PDES_PER_PAGE;
-	ppgtt->enable = gen8_ppgtt_enable;
-	ppgtt->switch_mm = gen8_mm_switch;
-	ppgtt->base.clear_range = gen8_ppgtt_clear_range;
-	ppgtt->base.insert_entries = gen8_ppgtt_insert_entries;
-	ppgtt->base.cleanup = gen8_ppgtt_cleanup;
-	ppgtt->base.start = 0;
-	ppgtt->base.total = ppgtt->num_pt_pages * GEN8_PTES_PER_PAGE * PAGE_SIZE;
 
-	BUG_ON(ppgtt->num_pd_pages > GEN8_LEGACY_PDPS);
+	ret = gen8_ppgtt_allocate_dma(ppgtt);
+	if (ret)
+		gen8_ppgtt_free(ppgtt);
 
-	/*
-	 * - Create a mapping for the page directories.
-	 * - For each page directory:
-	 *      allocate space for page table mappings.
-	 *      map each page table
-	 */
-	for (i = 0; i < max_pdp; i++) {
-		dma_addr_t temp;
-		temp = pci_map_page(ppgtt->base.dev->pdev,
-				    &ppgtt->pd_pages[i], 0,
-				    PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
-		if (pci_dma_mapping_error(ppgtt->base.dev->pdev, temp))
-			goto err_out;
+	return ret;
+}
 
-		ppgtt->pd_dma_addr[i] = temp;
+static int gen8_ppgtt_setup_page_directories(struct i915_hw_ppgtt *ppgtt,
+					     const int pd)
+{
+	dma_addr_t pd_addr;
+	int ret;
 
-		ppgtt->gen8_pt_dma_addr[i] = kmalloc(sizeof(dma_addr_t) * GEN8_PDES_PER_PAGE, GFP_KERNEL);
-		if (!ppgtt->gen8_pt_dma_addr[i])
-			goto err_out;
+	pd_addr = pci_map_page(ppgtt->base.dev->pdev,
+			       &ppgtt->pd_pages[pd], 0,
+			       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
 
-		for (j = 0; j < GEN8_PDES_PER_PAGE; j++) {
-			struct page *p = &pt_pages[i * GEN8_PDES_PER_PAGE + j];
-			temp = pci_map_page(ppgtt->base.dev->pdev,
-					    p, 0, PAGE_SIZE,
-					    PCI_DMA_BIDIRECTIONAL);
+	ret = pci_dma_mapping_error(ppgtt->base.dev->pdev, pd_addr);
+	if (ret)
+		return ret;
+
+	ppgtt->pd_dma_addr[pd] = pd_addr;
+
+	return 0;
+}
 
-			if (pci_dma_mapping_error(ppgtt->base.dev->pdev, temp))
-				goto err_out;
+static int gen8_ppgtt_setup_page_tables(struct i915_hw_ppgtt *ppgtt,
+					const int pd,
+					const int pt)
+{
+	dma_addr_t pt_addr;
+	struct page *p;
+	int ret;
 
-			ppgtt->gen8_pt_dma_addr[i][j] = temp;
+	p = ppgtt->gen8_pt_pages[pd][pt];
+	pt_addr = pci_map_page(ppgtt->base.dev->pdev,
+			       p, 0, PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+	ret = pci_dma_mapping_error(ppgtt->base.dev->pdev, pt_addr);
+	if (ret)
+		return ret;
+
+	ppgtt->gen8_pt_dma_addr[pd][pt] = pt_addr;
+
+	return 0;
+}
+
+/**
+ * GEN8 legacy ppgtt programming is accomplished through a max 4 PDP registers
+ * with a net effect resembling a 2-level page table in normal x86 terms. Each
+ * PDP represents 1GB of memory 4 * 512 * 512 * 4096 = 4GB legacy 32b address
+ * space.
+ *
+ * FIXME: split allocation into smaller pieces. For now we only ever do this
+ * once, but with full PPGTT, the multiple contiguous allocations will be bad.
+ * TODO: Do something with the size parameter
+ */
+static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt, uint64_t size)
+{
+	const int max_pdp = DIV_ROUND_UP(size, 1 << 30);
+	const int min_pt_pages = GEN8_PDES_PER_PAGE * max_pdp;
+	int i, j, ret;
+
+	if (size % (1<<30))
+		DRM_INFO("Pages will be wasted unless GTT size (%llu) is divisible by 1GB\n", size);
+
+	/* 1. Do all our allocations for page directories and page tables. */
+	ret = gen8_ppgtt_alloc(ppgtt, max_pdp);
+	if (ret)
+		return ret;
+
+	/*
+	 * 2. Create DMA mappings for the page directories and page tables.
+	 */
+	for (i = 0; i < max_pdp; i++) {
+		ret = gen8_ppgtt_setup_page_directories(ppgtt, i);
+		if (ret)
+			goto bail;
+
+		for (j = 0; j < GEN8_PDES_PER_PAGE; j++) {
+			ret = gen8_ppgtt_setup_page_tables(ppgtt, i, j);
+			if (ret)
+				goto bail;
 		}
 	}
 
-	/* For now, the PPGTT helper functions all require that the PDEs are
+	/*
+	 * 3. Map all the page directory entires to point to the page tables
+	 * we've allocated.
+	 *
+	 * For now, the PPGTT helper functions all require that the PDEs are
 	 * plugged in correctly. So we do that now/here. For aliasing PPGTT, we
-	 * will never need to touch the PDEs again */
+	 * will never need to touch the PDEs again.
+	 */
 	for (i = 0; i < max_pdp; i++) {
 		gen8_ppgtt_pde_t *pd_vaddr;
 		pd_vaddr = kmap_atomic(&ppgtt->pd_pages[i]);
@@ -461,20 +632,26 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt, uint64_t size)
 		kunmap_atomic(pd_vaddr);
 	}
 
-	ppgtt->base.clear_range(&ppgtt->base, 0,
-				ppgtt->num_pd_entries * GEN8_PTES_PER_PAGE,
-				true);
+	ppgtt->enable = gen8_ppgtt_enable;
+	ppgtt->switch_mm = gen8_mm_switch;
+	ppgtt->base.clear_range = gen8_ppgtt_clear_range;
+	ppgtt->base.insert_entries = gen8_ppgtt_insert_entries;
+	ppgtt->base.cleanup = gen8_ppgtt_cleanup;
+	ppgtt->base.start = 0;
+	ppgtt->base.total = ppgtt->num_pd_entries * GEN8_PTES_PER_PAGE * PAGE_SIZE;
+
+	ppgtt->base.clear_range(&ppgtt->base, 0, ppgtt->base.total, true);
 
 	DRM_DEBUG_DRIVER("Allocated %d pages for page directories (%d wasted)\n",
 			 ppgtt->num_pd_pages, ppgtt->num_pd_pages - max_pdp);
 	DRM_DEBUG_DRIVER("Allocated %d pages for page tables (%lld wasted)\n",
-			 ppgtt->num_pt_pages,
-			 (ppgtt->num_pt_pages - num_pt_pages) +
-			 size % (1<<30));
+			 ppgtt->num_pd_entries,
+			 (ppgtt->num_pd_entries - min_pt_pages) + size % (1<<30));
 	return 0;
 
-err_out:
-	ppgtt->base.cleanup(&ppgtt->base);
+bail:
+	gen8_ppgtt_unmap_pages(ppgtt);
+	gen8_ppgtt_free(ppgtt);
 	return ret;
 }
 
@@ -776,13 +953,15 @@ static int gen6_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 
 /* PPGTT support for Sandybdrige/Gen6 and later */
 static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
-				   unsigned first_entry,
-				   unsigned num_entries,
+				   uint64_t start,
+				   uint64_t length,
 				   bool use_scratch)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	gen6_gtt_pte_t *pt_vaddr, scratch_pte;
+	unsigned first_entry = start >> PAGE_SHIFT;
+	unsigned num_entries = length >> PAGE_SHIFT;
 	unsigned act_pt = first_entry / I915_PPGTT_PT_ENTRIES;
 	unsigned first_pte = first_entry % I915_PPGTT_PT_ENTRIES;
 	unsigned last_pte, i;
@@ -809,12 +988,13 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 
 static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 				      struct sg_table *pages,
-				      unsigned first_entry,
+				      uint64_t start,
 				      enum i915_cache_level cache_level)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	gen6_gtt_pte_t *pt_vaddr;
+	unsigned first_entry = start >> PAGE_SHIFT;
 	unsigned act_pt = first_entry / I915_PPGTT_PT_ENTRIES;
 	unsigned act_pte = first_entry % I915_PPGTT_PT_ENTRIES;
 	struct sg_page_iter sg_iter;
@@ -838,38 +1018,49 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 		kunmap_atomic(pt_vaddr);
 }
 
-static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
+static void gen6_ppgtt_unmap_pages(struct i915_hw_ppgtt *ppgtt)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
 	int i;
 
-	list_del(&vm->global_link);
-	drm_mm_takedown(&ppgtt->base.mm);
-	drm_mm_remove_node(&ppgtt->node);
-
 	if (ppgtt->pt_dma_addr) {
 		for (i = 0; i < ppgtt->num_pd_entries; i++)
 			pci_unmap_page(ppgtt->base.dev->pdev,
 				       ppgtt->pt_dma_addr[i],
 				       4096, PCI_DMA_BIDIRECTIONAL);
 	}
+}
+
+static void gen6_ppgtt_free(struct i915_hw_ppgtt *ppgtt)
+{
+	int i;
 
 	kfree(ppgtt->pt_dma_addr);
 	for (i = 0; i < ppgtt->num_pd_entries; i++)
 		__free_page(ppgtt->pt_pages[i]);
 	kfree(ppgtt->pt_pages);
-	kfree(ppgtt);
 }
 
-static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
+static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+
+	list_del(&vm->global_link);
+	drm_mm_takedown(&ppgtt->base.mm);
+	drm_mm_remove_node(&ppgtt->node);
+
+	gen6_ppgtt_unmap_pages(ppgtt);
+	gen6_ppgtt_free(ppgtt);
+}
+
+static int gen6_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt)
 {
 #define GEN6_PD_ALIGN (PAGE_SIZE * 16)
 #define GEN6_PD_SIZE (GEN6_PPGTT_PD_ENTRIES * PAGE_SIZE)
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	bool retried = false;
-	int i, ret;
+	int ret;
 
 	/* PPGTT PDEs reside in the GGTT and consists of 512 entries. The
 	 * allocator works in address space sizes, so it's multiplied by page
@@ -896,42 +1087,60 @@ alloc:
 	if (ppgtt->node.start < dev_priv->gtt.mappable_end)
 		DRM_DEBUG("Forced to use aperture for PDEs\n");
 
-	ppgtt->base.pte_encode = dev_priv->gtt.base.pte_encode;
 	ppgtt->num_pd_entries = GEN6_PPGTT_PD_ENTRIES;
-	if (IS_GEN6(dev)) {
-		ppgtt->enable = gen6_ppgtt_enable;
-		ppgtt->switch_mm = gen6_mm_switch;
-	} else if (IS_HASWELL(dev)) {
-		ppgtt->enable = gen7_ppgtt_enable;
-		ppgtt->switch_mm = hsw_mm_switch;
-	} else if (IS_GEN7(dev)) {
-		ppgtt->enable = gen7_ppgtt_enable;
-		ppgtt->switch_mm = gen7_mm_switch;
-	} else
-		BUG();
-	ppgtt->base.clear_range = gen6_ppgtt_clear_range;
-	ppgtt->base.insert_entries = gen6_ppgtt_insert_entries;
-	ppgtt->base.cleanup = gen6_ppgtt_cleanup;
-	ppgtt->base.scratch = dev_priv->gtt.base.scratch;
-	ppgtt->base.start = 0;
-	ppgtt->base.total = GEN6_PPGTT_PD_ENTRIES * I915_PPGTT_PT_ENTRIES * PAGE_SIZE;
+	return ret;
+}
+
+static int gen6_ppgtt_allocate_page_tables(struct i915_hw_ppgtt *ppgtt)
+{
+	int i;
+
 	ppgtt->pt_pages = kcalloc(ppgtt->num_pd_entries, sizeof(struct page *),
 				  GFP_KERNEL);
-	if (!ppgtt->pt_pages) {
-		drm_mm_remove_node(&ppgtt->node);
+
+	if (!ppgtt->pt_pages)
 		return -ENOMEM;
-	}
 
 	for (i = 0; i < ppgtt->num_pd_entries; i++) {
 		ppgtt->pt_pages[i] = alloc_page(GFP_KERNEL);
-		if (!ppgtt->pt_pages[i])
-			goto err_pt_alloc;
+		if (!ppgtt->pt_pages[i]) {
+			gen6_ppgtt_free(ppgtt);
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+static int gen6_ppgtt_alloc(struct i915_hw_ppgtt *ppgtt)
+{
+	int ret;
+
+	ret = gen6_ppgtt_allocate_page_directories(ppgtt);
+	if (ret)
+		return ret;
+
+	ret = gen6_ppgtt_allocate_page_tables(ppgtt);
+	if (ret) {
+		drm_mm_remove_node(&ppgtt->node);
+		return ret;
 	}
 
 	ppgtt->pt_dma_addr = kcalloc(ppgtt->num_pd_entries, sizeof(dma_addr_t),
 				     GFP_KERNEL);
-	if (!ppgtt->pt_dma_addr)
-		goto err_pt_alloc;
+	if (!ppgtt->pt_dma_addr) {
+		drm_mm_remove_node(&ppgtt->node);
+		gen6_ppgtt_free(ppgtt);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static int gen6_ppgtt_setup_page_tables(struct i915_hw_ppgtt *ppgtt)
+{
+	struct drm_device *dev = ppgtt->base.dev;
+	int i;
 
 	for (i = 0; i < ppgtt->num_pd_entries; i++) {
 		dma_addr_t pt_addr;
@@ -940,41 +1149,63 @@ alloc:
 				       PCI_DMA_BIDIRECTIONAL);
 
 		if (pci_dma_mapping_error(dev->pdev, pt_addr)) {
-			ret = -EIO;
-			goto err_pd_pin;
-
+			gen6_ppgtt_unmap_pages(ppgtt);
+			return -EIO;
 		}
+
 		ppgtt->pt_dma_addr[i] = pt_addr;
 	}
 
-	ppgtt->base.clear_range(&ppgtt->base, 0,
-				ppgtt->num_pd_entries * I915_PPGTT_PT_ENTRIES, true);
+	return 0;
+}
+
+static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
+{
+	struct drm_device *dev = ppgtt->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	ppgtt->base.pte_encode = dev_priv->gtt.base.pte_encode;
+	if (IS_GEN6(dev)) {
+		ppgtt->enable = gen6_ppgtt_enable;
+		ppgtt->switch_mm = gen6_mm_switch;
+	} else if (IS_HASWELL(dev)) {
+		ppgtt->enable = gen7_ppgtt_enable;
+		ppgtt->switch_mm = hsw_mm_switch;
+	} else if (IS_GEN7(dev)) {
+		ppgtt->enable = gen7_ppgtt_enable;
+		ppgtt->switch_mm = gen7_mm_switch;
+	} else
+		BUG();
+
+	ret = gen6_ppgtt_alloc(ppgtt);
+	if (ret)
+		return ret;
+
+	ret = gen6_ppgtt_setup_page_tables(ppgtt);
+	if (ret) {
+		gen6_ppgtt_free(ppgtt);
+		return ret;
+	}
+
+	ppgtt->base.clear_range = gen6_ppgtt_clear_range;
+	ppgtt->base.insert_entries = gen6_ppgtt_insert_entries;
+	ppgtt->base.cleanup = gen6_ppgtt_cleanup;
+	ppgtt->base.scratch = dev_priv->gtt.base.scratch;
+	ppgtt->base.start = 0;
+	ppgtt->base.total = GEN6_PPGTT_PD_ENTRIES * I915_PPGTT_PT_ENTRIES * PAGE_SIZE;
 	ppgtt->debug_dump = gen6_dump_ppgtt;
 
-	DRM_DEBUG_DRIVER("Allocated pde space (%ldM) at GTT entry: %lx\n",
-			 ppgtt->node.size >> 20,
-			 ppgtt->node.start / PAGE_SIZE);
 	ppgtt->pd_offset =
 		ppgtt->node.start / PAGE_SIZE * sizeof(gen6_gtt_pte_t);
 
-	return 0;
+	ppgtt->base.clear_range(&ppgtt->base, 0, ppgtt->base.total, true);
 
-err_pd_pin:
-	if (ppgtt->pt_dma_addr) {
-		for (i--; i >= 0; i--)
-			pci_unmap_page(dev->pdev, ppgtt->pt_dma_addr[i],
-				       4096, PCI_DMA_BIDIRECTIONAL);
-	}
-err_pt_alloc:
-	kfree(ppgtt->pt_dma_addr);
-	for (i = 0; i < ppgtt->num_pd_entries; i++) {
-		if (ppgtt->pt_pages[i])
-			__free_page(ppgtt->pt_pages[i]);
-	}
-	kfree(ppgtt->pt_pages);
-	drm_mm_remove_node(&ppgtt->node);
+	DRM_DEBUG_DRIVER("Allocated pde space (%ldM) at GTT entry: %lx\n",
+			 ppgtt->node.size >> 20,
+			 ppgtt->node.start / PAGE_SIZE);
 
-	return ret;
+	return 0;
 }
 
 int i915_gem_init_ppgtt(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
@@ -1012,20 +1243,17 @@ ppgtt_bind_vma(struct i915_vma *vma,
 	       enum i915_cache_level cache_level,
 	       u32 flags)
 {
-	const unsigned long entry = vma->node.start >> PAGE_SHIFT;
-
 	WARN_ON(flags);
 
-	vma->vm->insert_entries(vma->vm, vma->obj->pages, entry, cache_level);
+	vma->vm->insert_entries(vma->vm, vma->obj->pages, vma->node.start,
+				cache_level);
 }
 
 static void ppgtt_unbind_vma(struct i915_vma *vma)
 {
-	const unsigned long entry = vma->node.start >> PAGE_SHIFT;
-
 	vma->vm->clear_range(vma->vm,
-			     entry,
-			     vma->obj->base.size >> PAGE_SHIFT,
+			     vma->node.start,
+			     vma->obj->base.size,
 			     true);
 }
 
@@ -1109,8 +1337,8 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev)
 	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,
+				       dev_priv->gtt.base.start,
+				       dev_priv->gtt.base.total,
 				       false);
 }
 
@@ -1124,8 +1352,8 @@ void i915_gem_restore_gtt_mappings(struct drm_device *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.start,
+				       dev_priv->gtt.base.total,
 				       true);
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
@@ -1186,10 +1414,11 @@ static inline void gen8_set_pte(void __iomem *addr, gen8_gtt_pte_t pte)
 
 static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct sg_table *st,
-				     unsigned int first_entry,
+				     uint64_t start,
 				     enum i915_cache_level level)
 {
 	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	unsigned first_entry = start >> PAGE_SHIFT;
 	gen8_gtt_pte_t __iomem *gtt_entries =
 		(gen8_gtt_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
 	int i = 0;
@@ -1231,10 +1460,11 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
  */
 static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct sg_table *st,
-				     unsigned int first_entry,
+				     uint64_t start,
 				     enum i915_cache_level level)
 {
 	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	unsigned first_entry = start >> PAGE_SHIFT;
 	gen6_gtt_pte_t __iomem *gtt_entries =
 		(gen6_gtt_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
 	int i = 0;
@@ -1266,11 +1496,13 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 }
 
 static void gen8_ggtt_clear_range(struct i915_address_space *vm,
-				  unsigned int first_entry,
-				  unsigned int num_entries,
+				  uint64_t start,
+				  uint64_t length,
 				  bool use_scratch)
 {
 	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	unsigned first_entry = start >> PAGE_SHIFT;
+	unsigned num_entries = length >> PAGE_SHIFT;
 	gen8_gtt_pte_t scratch_pte, __iomem *gtt_base =
 		(gen8_gtt_pte_t __iomem *) dev_priv->gtt.gsm + first_entry;
 	const int max_entries = gtt_total_entries(dev_priv->gtt) - first_entry;
@@ -1290,11 +1522,13 @@ static void gen8_ggtt_clear_range(struct i915_address_space *vm,
 }
 
 static void gen6_ggtt_clear_range(struct i915_address_space *vm,
-				  unsigned int first_entry,
-				  unsigned int num_entries,
+				  uint64_t start,
+				  uint64_t length,
 				  bool use_scratch)
 {
 	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	unsigned first_entry = start >> PAGE_SHIFT;
+	unsigned num_entries = length >> PAGE_SHIFT;
 	gen6_gtt_pte_t scratch_pte, __iomem *gtt_base =
 		(gen6_gtt_pte_t __iomem *) dev_priv->gtt.gsm + first_entry;
 	const int max_entries = gtt_total_entries(dev_priv->gtt) - first_entry;
@@ -1327,10 +1561,12 @@ static void i915_ggtt_bind_vma(struct i915_vma *vma,
 }
 
 static void i915_ggtt_clear_range(struct i915_address_space *vm,
-				  unsigned int first_entry,
-				  unsigned int num_entries,
+				  uint64_t start,
+				  uint64_t length,
 				  bool unused)
 {
+	unsigned first_entry = start >> PAGE_SHIFT;
+	unsigned num_entries = length >> PAGE_SHIFT;
 	intel_gtt_clear_range(first_entry, num_entries);
 }
 
@@ -1351,7 +1587,6 @@ static void ggtt_bind_vma(struct i915_vma *vma,
 	struct drm_device *dev = vma->vm->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj = vma->obj;
-	const unsigned long entry = vma->node.start >> PAGE_SHIFT;
 
 	/* If there is no aliasing PPGTT, or the caller needs a global mapping,
 	 * or we have a global mapping already but the cacheability flags have
@@ -1367,7 +1602,8 @@ static void ggtt_bind_vma(struct i915_vma *vma,
 	if (!dev_priv->mm.aliasing_ppgtt || flags & GLOBAL_BIND) {
 		if (!obj->has_global_gtt_mapping ||
 		    (cache_level != obj->cache_level)) {
-			vma->vm->insert_entries(vma->vm, obj->pages, entry,
+			vma->vm->insert_entries(vma->vm, obj->pages,
+						vma->node.start,
 						cache_level);
 			obj->has_global_gtt_mapping = 1;
 		}
@@ -1378,7 +1614,9 @@ static void ggtt_bind_vma(struct i915_vma *vma,
 	     (cache_level != obj->cache_level))) {
 		struct i915_hw_ppgtt *appgtt = dev_priv->mm.aliasing_ppgtt;
 		appgtt->base.insert_entries(&appgtt->base,
-					    vma->obj->pages, entry, cache_level);
+					    vma->obj->pages,
+					    vma->node.start,
+					    cache_level);
 		vma->obj->has_aliasing_ppgtt_mapping = 1;
 	}
 }
@@ -1388,11 +1626,11 @@ static void ggtt_unbind_vma(struct i915_vma *vma)
 	struct drm_device *dev = vma->vm->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj = vma->obj;
-	const unsigned long entry = vma->node.start >> PAGE_SHIFT;
 
 	if (obj->has_global_gtt_mapping) {
-		vma->vm->clear_range(vma->vm, entry,
-				     vma->obj->base.size >> PAGE_SHIFT,
+		vma->vm->clear_range(vma->vm,
+				     vma->node.start,
+				     obj->base.size,
 				     true);
 		obj->has_global_gtt_mapping = 0;
 	}
@@ -1400,8 +1638,8 @@ static void ggtt_unbind_vma(struct i915_vma *vma)
 	if (obj->has_aliasing_ppgtt_mapping) {
 		struct i915_hw_ppgtt *appgtt = dev_priv->mm.aliasing_ppgtt;
 		appgtt->base.clear_range(&appgtt->base,
-					 entry,
-					 obj->base.size >> PAGE_SHIFT,
+					 vma->node.start,
+					 obj->base.size,
 					 true);
 		obj->has_aliasing_ppgtt_mapping = 0;
 	}
@@ -1486,14 +1724,14 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,
 
 	/* Clear any non-preallocated blocks */
 	drm_mm_for_each_hole(entry, &ggtt_vm->mm, hole_start, hole_end) {
-		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, true);
+		ggtt_vm->clear_range(ggtt_vm, hole_start,
+				     hole_end - hole_start, true);
 	}
 
 	/* And finally clear the reserved guard page */
-	ggtt_vm->clear_range(ggtt_vm, end / PAGE_SIZE - 1, 1, true);
+	ggtt_vm->clear_range(ggtt_vm, end - PAGE_SIZE, PAGE_SIZE, true);
 }
 
 void i915_gem_init_global_gtt(struct drm_device *dev)
@@ -1558,11 +1796,6 @@ static inline unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
 	bdw_gmch_ctl &= BDW_GMCH_GGMS_MASK;
 	if (bdw_gmch_ctl)
 		bdw_gmch_ctl = 1 << bdw_gmch_ctl;
-	if (bdw_gmch_ctl > 4) {
-		WARN_ON(!i915.preliminary_hw_support);
-		return 4<<20;
-	}
-
 	return bdw_gmch_ctl << 20;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 000b3694f349..144a5e2bc7ef 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -304,22 +304,54 @@ void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...)
 	va_end(args);
 }
 
+static void print_error_obj(struct drm_i915_error_state_buf *m,
+			    struct drm_i915_error_object *obj)
+{
+	int page, offset, elt;
+
+	for (page = offset = 0; page < obj->page_count; page++) {
+		for (elt = 0; elt < PAGE_SIZE/4; elt++) {
+			err_printf(m, "%08x :  %08x\n", offset,
+				   obj->pages[page][elt]);
+			offset += 4;
+		}
+	}
+}
+
 int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			    const struct i915_error_state_file_priv *error_priv)
 {
 	struct drm_device *dev = error_priv->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct drm_i915_error_state *error = error_priv->error;
-	int i, j, page, offset, elt;
+	int i, j, offset, elt;
+	int max_hangcheck_score;
 
 	if (!error) {
 		err_printf(m, "no error state collected\n");
 		goto out;
 	}
 
+	err_printf(m, "%s\n", error->error_msg);
 	err_printf(m, "Time: %ld s %ld us\n", error->time.tv_sec,
 		   error->time.tv_usec);
 	err_printf(m, "Kernel: " UTS_RELEASE "\n");
+	max_hangcheck_score = 0;
+	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
+		if (error->ring[i].hangcheck_score > max_hangcheck_score)
+			max_hangcheck_score = error->ring[i].hangcheck_score;
+	}
+	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
+		if (error->ring[i].hangcheck_score == max_hangcheck_score &&
+		    error->ring[i].pid != -1) {
+			err_printf(m, "Active process (on ring %s): %s [%d]\n",
+				   ring_str(i),
+				   error->ring[i].comm,
+				   error->ring[i].pid);
+		}
+	}
+	err_printf(m, "Reset count: %u\n", error->reset_count);
+	err_printf(m, "Suspend count: %u\n", error->suspend_count);
 	err_printf(m, "PCI ID: 0x%04x\n", dev->pdev->device);
 	err_printf(m, "EIR: 0x%08x\n", error->eir);
 	err_printf(m, "IER: 0x%08x\n", error->ier);
@@ -362,18 +394,23 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
 		struct drm_i915_error_object *obj;
 
-		if ((obj = error->ring[i].batchbuffer)) {
-			err_printf(m, "%s --- gtt_offset = 0x%08x\n",
-				   dev_priv->ring[i].name,
+		obj = error->ring[i].batchbuffer;
+		if (obj) {
+			err_puts(m, dev_priv->ring[i].name);
+			if (error->ring[i].pid != -1)
+				err_printf(m, " (submitted by %s [%d])",
+					   error->ring[i].comm,
+					   error->ring[i].pid);
+			err_printf(m, " --- gtt_offset = 0x%08x\n",
 				   obj->gtt_offset);
-			offset = 0;
-			for (page = 0; page < obj->page_count; page++) {
-				for (elt = 0; elt < PAGE_SIZE/4; elt++) {
-					err_printf(m, "%08x :  %08x\n", offset,
-						   obj->pages[page][elt]);
-					offset += 4;
-				}
-			}
+			print_error_obj(m, obj);
+		}
+
+		obj = error->ring[i].wa_batchbuffer;
+		if (obj) {
+			err_printf(m, "%s (w/a) --- gtt_offset = 0x%08x\n",
+				   dev_priv->ring[i].name, obj->gtt_offset);
+			print_error_obj(m, obj);
 		}
 
 		if (error->ring[i].num_requests) {
@@ -392,15 +429,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			err_printf(m, "%s --- ringbuffer = 0x%08x\n",
 				   dev_priv->ring[i].name,
 				   obj->gtt_offset);
-			offset = 0;
-			for (page = 0; page < obj->page_count; page++) {
-				for (elt = 0; elt < PAGE_SIZE/4; elt++) {
-					err_printf(m, "%08x :  %08x\n",
-						   offset,
-						   obj->pages[page][elt]);
-					offset += 4;
-				}
-			}
+			print_error_obj(m, obj);
 		}
 
 		if ((obj = error->ring[i].hws_page)) {
@@ -666,7 +695,8 @@ static u32 capture_pinned_bo(struct drm_i915_error_buffer *err,
  * It's only a small step better than a random number in its current form.
  */
 static uint32_t i915_error_generate_code(struct drm_i915_private *dev_priv,
-					 struct drm_i915_error_state *error)
+					 struct drm_i915_error_state *error,
+					 int *ring_id)
 {
 	uint32_t error_code = 0;
 	int i;
@@ -676,9 +706,14 @@ static uint32_t i915_error_generate_code(struct drm_i915_private *dev_priv,
 	 * synchronization commands which almost always appear in the case
 	 * strictly a client bug. Use instdone to differentiate those some.
 	 */
-	for (i = 0; i < I915_NUM_RINGS; i++)
-		if (error->ring[i].hangcheck_action == HANGCHECK_HUNG)
+	for (i = 0; i < I915_NUM_RINGS; i++) {
+		if (error->ring[i].hangcheck_action == HANGCHECK_HUNG) {
+			if (ring_id)
+				*ring_id = i;
+
 			return error->ring[i].ipehr ^ error->ring[i].instdone;
+		}
+	}
 
 	return error_code;
 }
@@ -716,87 +751,6 @@ static void i915_gem_record_fences(struct drm_device *dev,
 	}
 }
 
-/* This assumes all batchbuffers are executed from the PPGTT. It might have to
- * change in the future. */
-static bool is_active_vm(struct i915_address_space *vm,
-			 struct intel_ring_buffer *ring)
-{
-	struct drm_device *dev = vm->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_ppgtt *ppgtt;
-
-	if (INTEL_INFO(dev)->gen < 7)
-		return i915_is_ggtt(vm);
-
-	/* FIXME: This ignores that the global gtt vm is also on this list. */
-	ppgtt = container_of(vm, struct i915_hw_ppgtt, base);
-
-	if (INTEL_INFO(dev)->gen >= 8) {
-		u64 pdp0 = (u64)I915_READ(GEN8_RING_PDP_UDW(ring, 0)) << 32;
-		pdp0 |=  I915_READ(GEN8_RING_PDP_LDW(ring, 0));
-		return pdp0 == ppgtt->pd_dma_addr[0];
-	} else {
-		u32 pp_db;
-		pp_db = I915_READ(RING_PP_DIR_BASE(ring));
-		return (pp_db >> 10) == ppgtt->pd_offset;
-	}
-}
-
-static struct drm_i915_error_object *
-i915_error_first_batchbuffer(struct drm_i915_private *dev_priv,
-			     struct intel_ring_buffer *ring)
-{
-	struct i915_address_space *vm;
-	struct i915_vma *vma;
-	struct drm_i915_gem_object *obj;
-	bool found_active = false;
-	u32 seqno;
-
-	if (!ring->get_seqno)
-		return NULL;
-
-	if (HAS_BROKEN_CS_TLB(dev_priv->dev)) {
-		u32 acthd = I915_READ(ACTHD);
-
-		if (WARN_ON(ring->id != RCS))
-			return NULL;
-
-		obj = ring->scratch.obj;
-		if (obj != NULL &&
-		    acthd >= i915_gem_obj_ggtt_offset(obj) &&
-		    acthd < i915_gem_obj_ggtt_offset(obj) + obj->base.size)
-			return i915_error_ggtt_object_create(dev_priv, obj);
-	}
-
-	seqno = ring->get_seqno(ring, false);
-	list_for_each_entry(vm, &dev_priv->vm_list, global_link) {
-		if (!is_active_vm(vm, ring))
-			continue;
-
-		found_active = true;
-
-		list_for_each_entry(vma, &vm->active_list, mm_list) {
-			obj = vma->obj;
-			if (obj->ring != ring)
-				continue;
-
-			if (i915_seqno_passed(seqno, obj->last_read_seqno))
-				continue;
-
-			if ((obj->base.read_domains & I915_GEM_DOMAIN_COMMAND) == 0)
-				continue;
-
-			/* We need to copy these to an anonymous buffer as the simplest
-			 * method to avoid being overwritten by userspace.
-			 */
-			return i915_error_object_create(dev_priv, obj, vm);
-		}
-	}
-
-	WARN_ON(!found_active);
-	return NULL;
-}
-
 static void i915_record_ring_state(struct drm_device *dev,
 				   struct intel_ring_buffer *ring,
 				   struct drm_i915_error_ring *ering)
@@ -945,8 +899,39 @@ static void i915_gem_record_rings(struct drm_device *dev,
 
 		i915_record_ring_state(dev, ring, &error->ring[i]);
 
-		error->ring[i].batchbuffer =
-			i915_error_first_batchbuffer(dev_priv, ring);
+		error->ring[i].pid = -1;
+		request = i915_gem_find_active_request(ring);
+		if (request) {
+			/* We need to copy these to an anonymous buffer
+			 * as the simplest method to avoid being overwritten
+			 * by userspace.
+			 */
+			error->ring[i].batchbuffer =
+				i915_error_object_create(dev_priv,
+							 request->batch_obj,
+							 request->ctx ?
+							 request->ctx->vm :
+							 &dev_priv->gtt.base);
+
+			if (HAS_BROKEN_CS_TLB(dev_priv->dev) &&
+			    ring->scratch.obj)
+				error->ring[i].wa_batchbuffer =
+					i915_error_ggtt_object_create(dev_priv,
+							     ring->scratch.obj);
+
+			if (request->file_priv) {
+				struct task_struct *task;
+
+				rcu_read_lock();
+				task = pid_task(request->file_priv->file->pid,
+						PIDTYPE_PID);
+				if (task) {
+					strcpy(error->ring[i].comm, task->comm);
+					error->ring[i].pid = task->pid;
+				}
+				rcu_read_unlock();
+			}
+		}
 
 		error->ring[i].ringbuffer =
 			i915_error_ggtt_object_create(dev_priv, ring->obj);
@@ -1113,6 +1098,40 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 	i915_get_extra_instdone(dev, error->extra_instdone);
 }
 
+static void i915_error_capture_msg(struct drm_device *dev,
+				   struct drm_i915_error_state *error,
+				   bool wedged,
+				   const char *error_msg)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 ecode;
+	int ring_id = -1, len;
+
+	ecode = i915_error_generate_code(dev_priv, error, &ring_id);
+
+	len = scnprintf(error->error_msg, sizeof(error->error_msg),
+			"GPU HANG: ecode %d:0x%08x", ring_id, ecode);
+
+	if (ring_id != -1 && error->ring[ring_id].pid != -1)
+		len += scnprintf(error->error_msg + len,
+				 sizeof(error->error_msg) - len,
+				 ", in %s [%d]",
+				 error->ring[ring_id].comm,
+				 error->ring[ring_id].pid);
+
+	scnprintf(error->error_msg + len, sizeof(error->error_msg) - len,
+		  ", reason: %s, action: %s",
+		  error_msg,
+		  wedged ? "reset" : "continue");
+}
+
+static void i915_capture_gen_state(struct drm_i915_private *dev_priv,
+				   struct drm_i915_error_state *error)
+{
+	error->reset_count = i915_reset_count(&dev_priv->gpu_error);
+	error->suspend_count = dev_priv->suspend_count;
+}
+
 /**
  * i915_capture_error_state - capture an error record for later analysis
  * @dev: drm device
@@ -1122,19 +1141,13 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
  * out a structure which becomes available in debugfs for user level tools
  * to pick up.
  */
-void i915_capture_error_state(struct drm_device *dev)
+void i915_capture_error_state(struct drm_device *dev, bool wedged,
+			      const char *error_msg)
 {
 	static bool warned;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_error_state *error;
 	unsigned long flags;
-	uint32_t ecode;
-
-	spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
-	error = dev_priv->gpu_error.first_error;
-	spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
-	if (error)
-		return;
 
 	/* Account for pipe specific data like PIPE*STAT */
 	error = kzalloc(sizeof(*error), GFP_ATOMIC);
@@ -1143,30 +1156,22 @@ void i915_capture_error_state(struct drm_device *dev)
 		return;
 	}
 
-	DRM_INFO("GPU crash dump saved to /sys/class/drm/card%d/error\n",
-		 dev->primary->index);
 	kref_init(&error->ref);
 
+	i915_capture_gen_state(dev_priv, error);
 	i915_capture_reg_state(dev_priv, error);
 	i915_gem_capture_buffers(dev_priv, error);
 	i915_gem_record_fences(dev, error);
 	i915_gem_record_rings(dev, error);
-	ecode = i915_error_generate_code(dev_priv, error);
-
-	if (!warned) {
-		DRM_INFO("GPU HANG [%x]\n", ecode);
-		DRM_INFO("GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n");
-		DRM_INFO("Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel\n");
-		DRM_INFO("drm/i915 developers can then reassign to the right component if it's not a kernel issue.\n");
-		DRM_INFO("The gpu crash dump is required to analyze gpu hangs, so please always attach it.\n");
-		warned = true;
-	}
 
 	do_gettimeofday(&error->time);
 
 	error->overlay = intel_overlay_capture_error_state(dev);
 	error->display = intel_display_capture_error_state(dev);
 
+	i915_error_capture_msg(dev, error, wedged, error_msg);
+	DRM_INFO("%s\n", error->error_msg);
+
 	spin_lock_irqsave(&dev_priv->gpu_error.lock, flags);
 	if (dev_priv->gpu_error.first_error == NULL) {
 		dev_priv->gpu_error.first_error = error;
@@ -1174,8 +1179,19 @@ void i915_capture_error_state(struct drm_device *dev)
 	}
 	spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags);
 
-	if (error)
+	if (error) {
 		i915_error_state_free(&error->ref);
+		return;
+	}
+
+	if (!warned) {
+		DRM_INFO("GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n");
+		DRM_INFO("Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel\n");
+		DRM_INFO("drm/i915 developers can then reassign to the right component if it's not a kernel issue.\n");
+		DRM_INFO("The gpu crash dump is required to analyze gpu hangs, so please always attach it.\n");
+		DRM_INFO("GPU crash dump saved to /sys/class/drm/card%d/error\n", dev->primary->index);
+		warned = true;
+	}
 }
 
 void i915_error_state_get(struct drm_device *dev,
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index f68aee31e565..be2713f12e08 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -387,16 +387,15 @@ static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
  *
  * Returns the previous state of underrun reporting.
  */
-bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
-					   enum pipe pipe, bool enable)
+bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
+					     enum pipe pipe, bool enable)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	unsigned long flags;
 	bool ret;
 
-	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+	assert_spin_locked(&dev_priv->irq_lock);
 
 	ret = !intel_crtc->cpu_fifo_underrun_disabled;
 
@@ -415,7 +414,20 @@ bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
 		broadwell_set_fifo_underrun_reporting(dev, pipe, enable);
 
 done:
+	return ret;
+}
+
+bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
+					   enum pipe pipe, bool enable)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long flags;
+	bool ret;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+	ret = __intel_set_cpu_fifo_underrun_reporting(dev, pipe, enable);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+
 	return ret;
 }
 
@@ -482,7 +494,7 @@ done:
 }
 
 
-void
+static void
 __i915_enable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 		       u32 enable_mask, u32 status_mask)
 {
@@ -506,7 +518,7 @@ __i915_enable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 	POSTING_READ(reg);
 }
 
-void
+static void
 __i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 		        u32 enable_mask, u32 status_mask)
 {
@@ -1296,8 +1308,8 @@ static void snb_gt_irq_handler(struct drm_device *dev,
 	if (gt_iir & (GT_BLT_CS_ERROR_INTERRUPT |
 		      GT_BSD_CS_ERROR_INTERRUPT |
 		      GT_RENDER_CS_MASTER_ERROR_INTERRUPT)) {
-		DRM_ERROR("GT error interrupt 0x%08x\n", gt_iir);
-		i915_handle_error(dev, false);
+		i915_handle_error(dev, false, "GT error interrupt 0x%08x",
+				  gt_iir);
 	}
 
 	if (gt_iir & GT_PARITY_ERROR(dev))
@@ -1544,8 +1556,9 @@ static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
 			notify_ring(dev_priv->dev, &dev_priv->ring[VECS]);
 
 		if (pm_iir & PM_VEBOX_CS_ERROR_INTERRUPT) {
-			DRM_ERROR("VEBOX CS error interrupt 0x%08x\n", pm_iir);
-			i915_handle_error(dev_priv->dev, false);
+			i915_handle_error(dev_priv->dev, false,
+					  "VEBOX CS error interrupt 0x%08x",
+					  pm_iir);
 		}
 	}
 }
@@ -1865,7 +1878,7 @@ static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	enum pipe i;
+	enum pipe pipe;
 
 	if (de_iir & DE_ERR_INT_IVB)
 		ivb_err_int_handler(dev);
@@ -1876,14 +1889,14 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 	if (de_iir & DE_GSE_IVB)
 		intel_opregion_asle_intr(dev);
 
-	for_each_pipe(i) {
-		if (de_iir & (DE_PIPE_VBLANK_IVB(i)))
-			drm_handle_vblank(dev, i);
+	for_each_pipe(pipe) {
+		if (de_iir & (DE_PIPE_VBLANK_IVB(pipe)))
+			drm_handle_vblank(dev, pipe);
 
 		/* plane/pipes map 1:1 on ilk+ */
-		if (de_iir & DE_PLANE_FLIP_DONE_IVB(i)) {
-			intel_prepare_page_flip(dev, i);
-			intel_finish_page_flip_plane(dev, i);
+		if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe)) {
+			intel_prepare_page_flip(dev, pipe);
+			intel_finish_page_flip_plane(dev, pipe);
 		}
 	}
 
@@ -2277,11 +2290,18 @@ static void i915_report_and_clear_eir(struct drm_device *dev)
  * so userspace knows something bad happened (should trigger collection
  * of a ring dump etc.).
  */
-void i915_handle_error(struct drm_device *dev, bool wedged)
+void i915_handle_error(struct drm_device *dev, bool wedged,
+		       const char *fmt, ...)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	va_list args;
+	char error_msg[80];
+
+	va_start(args, fmt);
+	vscnprintf(error_msg, sizeof(error_msg), fmt, args);
+	va_end(args);
 
-	i915_capture_error_state(dev);
+	i915_capture_error_state(dev, wedged, error_msg);
 	i915_report_and_clear_eir(dev);
 
 	if (wedged) {
@@ -2584,9 +2604,9 @@ ring_stuck(struct intel_ring_buffer *ring, u32 acthd)
 	 */
 	tmp = I915_READ_CTL(ring);
 	if (tmp & RING_WAIT) {
-		DRM_ERROR("Kicking stuck wait on %s\n",
-			  ring->name);
-		i915_handle_error(dev, false);
+		i915_handle_error(dev, false,
+				  "Kicking stuck wait on %s",
+				  ring->name);
 		I915_WRITE_CTL(ring, tmp);
 		return HANGCHECK_KICK;
 	}
@@ -2596,9 +2616,9 @@ ring_stuck(struct intel_ring_buffer *ring, u32 acthd)
 		default:
 			return HANGCHECK_HUNG;
 		case 1:
-			DRM_ERROR("Kicking stuck semaphore on %s\n",
-				  ring->name);
-			i915_handle_error(dev, false);
+			i915_handle_error(dev, false,
+					  "Kicking stuck semaphore on %s",
+					  ring->name);
 			I915_WRITE_CTL(ring, tmp);
 			return HANGCHECK_KICK;
 		case 0:
@@ -2720,7 +2740,7 @@ static void i915_hangcheck_elapsed(unsigned long data)
 	}
 
 	if (rings_hung)
-		return i915_handle_error(dev, true);
+		return i915_handle_error(dev, true, "Ring hung");
 
 	if (busy_count)
 		/* Reset timer case chip hangs without another request
@@ -3016,44 +3036,113 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
+static void valleyview_display_irqs_install(struct drm_i915_private *dev_priv)
+{
+	u32 pipestat_mask;
+	u32 iir_mask;
+
+	pipestat_mask = PIPESTAT_INT_STATUS_MASK |
+			PIPE_FIFO_UNDERRUN_STATUS;
+
+	I915_WRITE(PIPESTAT(PIPE_A), pipestat_mask);
+	I915_WRITE(PIPESTAT(PIPE_B), pipestat_mask);
+	POSTING_READ(PIPESTAT(PIPE_A));
+
+	pipestat_mask = PLANE_FLIP_DONE_INT_STATUS_VLV |
+			PIPE_CRC_DONE_INTERRUPT_STATUS;
+
+	i915_enable_pipestat(dev_priv, PIPE_A, pipestat_mask |
+					       PIPE_GMBUS_INTERRUPT_STATUS);
+	i915_enable_pipestat(dev_priv, PIPE_B, pipestat_mask);
+
+	iir_mask = I915_DISPLAY_PORT_INTERRUPT |
+		   I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
+		   I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
+	dev_priv->irq_mask &= ~iir_mask;
+
+	I915_WRITE(VLV_IIR, iir_mask);
+	I915_WRITE(VLV_IIR, iir_mask);
+	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
+	I915_WRITE(VLV_IER, ~dev_priv->irq_mask);
+	POSTING_READ(VLV_IER);
+}
+
+static void valleyview_display_irqs_uninstall(struct drm_i915_private *dev_priv)
+{
+	u32 pipestat_mask;
+	u32 iir_mask;
+
+	iir_mask = I915_DISPLAY_PORT_INTERRUPT |
+		   I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
+		   I915_DISPLAY_PIPE_A_EVENT_INTERRUPT;
+
+	dev_priv->irq_mask |= iir_mask;
+	I915_WRITE(VLV_IER, ~dev_priv->irq_mask);
+	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
+	I915_WRITE(VLV_IIR, iir_mask);
+	I915_WRITE(VLV_IIR, iir_mask);
+	POSTING_READ(VLV_IIR);
+
+	pipestat_mask = PLANE_FLIP_DONE_INT_STATUS_VLV |
+			PIPE_CRC_DONE_INTERRUPT_STATUS;
+
+	i915_disable_pipestat(dev_priv, PIPE_A, pipestat_mask |
+					        PIPE_GMBUS_INTERRUPT_STATUS);
+	i915_disable_pipestat(dev_priv, PIPE_B, pipestat_mask);
+
+	pipestat_mask = PIPESTAT_INT_STATUS_MASK |
+			PIPE_FIFO_UNDERRUN_STATUS;
+	I915_WRITE(PIPESTAT(PIPE_A), pipestat_mask);
+	I915_WRITE(PIPESTAT(PIPE_B), pipestat_mask);
+	POSTING_READ(PIPESTAT(PIPE_A));
+}
+
+void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv)
+{
+	assert_spin_locked(&dev_priv->irq_lock);
+
+	if (dev_priv->display_irqs_enabled)
+		return;
+
+	dev_priv->display_irqs_enabled = true;
+
+	if (dev_priv->dev->irq_enabled)
+		valleyview_display_irqs_install(dev_priv);
+}
+
+void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv)
+{
+	assert_spin_locked(&dev_priv->irq_lock);
+
+	if (!dev_priv->display_irqs_enabled)
+		return;
+
+	dev_priv->display_irqs_enabled = false;
+
+	if (dev_priv->dev->irq_enabled)
+		valleyview_display_irqs_uninstall(dev_priv);
+}
+
 static int valleyview_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
-	u32 enable_mask;
-	u32 pipestat_enable = PLANE_FLIP_DONE_INT_STATUS_VLV |
-		PIPE_CRC_DONE_INTERRUPT_STATUS;
 	unsigned long irqflags;
 
-	enable_mask = I915_DISPLAY_PORT_INTERRUPT;
-	enable_mask |= I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
-		I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT |
-		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
-		I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
-
-	/*
-	 *Leave vblank interrupts masked initially.  enable/disable will
-	 * toggle them based on usage.
-	 */
-	dev_priv->irq_mask = (~enable_mask) |
-		I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT |
-		I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
+	dev_priv->irq_mask = ~0;
 
 	I915_WRITE(PORT_HOTPLUG_EN, 0);
 	POSTING_READ(PORT_HOTPLUG_EN);
 
 	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
-	I915_WRITE(VLV_IER, enable_mask);
+	I915_WRITE(VLV_IER, ~dev_priv->irq_mask);
 	I915_WRITE(VLV_IIR, 0xffffffff);
-	I915_WRITE(PIPESTAT(0), 0xffff);
-	I915_WRITE(PIPESTAT(1), 0xffff);
 	POSTING_READ(VLV_IER);
 
 	/* Interrupt setup is already guaranteed to be single-threaded, this is
 	 * just to make the assert_spin_locked check happy. */
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_enable_pipestat(dev_priv, PIPE_A, pipestat_enable);
-	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
-	i915_enable_pipestat(dev_priv, PIPE_B, pipestat_enable);
+	if (dev_priv->display_irqs_enabled)
+		valleyview_display_irqs_install(dev_priv);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
 	I915_WRITE(VLV_IIR, 0xffffffff);
@@ -3184,6 +3273,7 @@ static void gen8_irq_uninstall(struct drm_device *dev)
 static void valleyview_irq_uninstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+	unsigned long irqflags;
 	int pipe;
 
 	if (!dev_priv)
@@ -3197,8 +3287,14 @@ static void valleyview_irq_uninstall(struct drm_device *dev)
 	I915_WRITE(HWSTAM, 0xffffffff);
 	I915_WRITE(PORT_HOTPLUG_EN, 0);
 	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
-	for_each_pipe(pipe)
-		I915_WRITE(PIPESTAT(pipe), 0xffff);
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	if (dev_priv->display_irqs_enabled)
+		valleyview_display_irqs_uninstall(dev_priv);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
+	dev_priv->irq_mask = 0;
+
 	I915_WRITE(VLV_IIR, 0xffffffff);
 	I915_WRITE(VLV_IMR, 0xffffffff);
 	I915_WRITE(VLV_IER, 0x0);
@@ -3337,7 +3433,9 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 		 */
 		spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
 		if (iir & I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT)
-			i915_handle_error(dev, false);
+			i915_handle_error(dev, false,
+					  "Command parser error, iir 0x%08x",
+					  iir);
 
 		for_each_pipe(pipe) {
 			int reg = PIPESTAT(pipe);
@@ -3519,7 +3617,9 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 		 */
 		spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
 		if (iir & I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT)
-			i915_handle_error(dev, false);
+			i915_handle_error(dev, false,
+					  "Command parser error, iir 0x%08x",
+					  iir);
 
 		for_each_pipe(pipe) {
 			int reg = PIPESTAT(pipe);
@@ -3756,7 +3856,9 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		 */
 		spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
 		if (iir & I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT)
-			i915_handle_error(dev, false);
+			i915_handle_error(dev, false,
+					  "Command parser error, iir 0x%08x",
+					  iir);
 
 		for_each_pipe(pipe) {
 			int reg = PIPESTAT(pipe);
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index 3b482585c5ae..a66ffb652bee 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -48,6 +48,7 @@ struct i915_params i915 __read_mostly = {
 	.reset = true,
 	.invert_brightness = 0,
 	.disable_display = 0,
+	.enable_cmd_parser = 0,
 };
 
 module_param_named(modeset, i915.modeset, int, 0400);
@@ -157,3 +158,7 @@ MODULE_PARM_DESC(invert_brightness,
 
 module_param_named(disable_display, i915.disable_display, bool, 0600);
 MODULE_PARM_DESC(disable_display, "Disable display (default: false)");
+
+module_param_named(enable_cmd_parser, i915.enable_cmd_parser, int, 0600);
+MODULE_PARM_DESC(enable_cmd_parser,
+		 "Enable command parsing (1=enabled, 0=disabled [default])");
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 2f564ce37d2c..146609ab42bb 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -175,6 +175,18 @@
 #define VGA_CR_DATA_CGA 0x3d5
 
 /*
+ * Instruction field definitions used by the command parser
+ */
+#define INSTR_CLIENT_SHIFT      29
+#define INSTR_CLIENT_MASK       0xE0000000
+#define   INSTR_MI_CLIENT       0x0
+#define   INSTR_BC_CLIENT       0x2
+#define   INSTR_RC_CLIENT       0x3
+#define INSTR_SUBCLIENT_SHIFT   27
+#define INSTR_SUBCLIENT_MASK    0x18000000
+#define   INSTR_MEDIA_SUBCLIENT 0x2
+
+/*
  * Memory interface instructions used by the kernel
  */
 #define MI_INSTR(opcode, flags) (((opcode) << 23) | (flags))
@@ -377,14 +389,30 @@
 #define   DSPFREQSTAT_MASK			(0x3 << DSPFREQSTAT_SHIFT)
 #define   DSPFREQGUAR_SHIFT			14
 #define   DSPFREQGUAR_MASK			(0x3 << DSPFREQGUAR_SHIFT)
+
+/* See the PUNIT HAS v0.8 for the below bits */
+enum punit_power_well {
+	PUNIT_POWER_WELL_RENDER			= 0,
+	PUNIT_POWER_WELL_MEDIA			= 1,
+	PUNIT_POWER_WELL_DISP2D			= 3,
+	PUNIT_POWER_WELL_DPIO_CMN_BC		= 5,
+	PUNIT_POWER_WELL_DPIO_TX_B_LANES_01	= 6,
+	PUNIT_POWER_WELL_DPIO_TX_B_LANES_23	= 7,
+	PUNIT_POWER_WELL_DPIO_TX_C_LANES_01	= 8,
+	PUNIT_POWER_WELL_DPIO_TX_C_LANES_23	= 9,
+	PUNIT_POWER_WELL_DPIO_RX0		= 10,
+	PUNIT_POWER_WELL_DPIO_RX1		= 11,
+
+	PUNIT_POWER_WELL_NUM,
+};
+
 #define PUNIT_REG_PWRGT_CTRL			0x60
 #define PUNIT_REG_PWRGT_STATUS			0x61
-#define	  PUNIT_CLK_GATE			1
-#define	  PUNIT_PWR_RESET			2
-#define	  PUNIT_PWR_GATE			3
-#define	  RENDER_PWRGT				(PUNIT_PWR_GATE << 0)
-#define	  MEDIA_PWRGT				(PUNIT_PWR_GATE << 2)
-#define	  DISP2D_PWRGT				(PUNIT_PWR_GATE << 6)
+#define   PUNIT_PWRGT_MASK(power_well)		(3 << ((power_well) * 2))
+#define   PUNIT_PWRGT_PWR_ON(power_well)	(0 << ((power_well) * 2))
+#define   PUNIT_PWRGT_CLK_GATE(power_well)	(1 << ((power_well) * 2))
+#define   PUNIT_PWRGT_RESET(power_well)		(2 << ((power_well) * 2))
+#define   PUNIT_PWRGT_PWR_GATE(power_well)	(3 << ((power_well) * 2))
 
 #define PUNIT_REG_GPU_LFM			0xd3
 #define PUNIT_REG_GPU_FREQ_REQ			0xd4
@@ -798,7 +826,12 @@
 # define ASYNC_FLIP_PERF_DISABLE			(1 << 14)
 
 #define GEN6_GT_MODE	0x20d0
-#define   GEN6_GT_MODE_HI				(1 << 9)
+#define GEN7_GT_MODE	0x7008
+#define   GEN6_WIZ_HASHING(hi, lo)			(((hi) << 9) | ((lo) << 7))
+#define   GEN6_WIZ_HASHING_8x8				GEN6_WIZ_HASHING(0, 0)
+#define   GEN6_WIZ_HASHING_8x4				GEN6_WIZ_HASHING(0, 1)
+#define   GEN6_WIZ_HASHING_16x4				GEN6_WIZ_HASHING(1, 0)
+#define   GEN6_WIZ_HASHING_MASK				(GEN6_WIZ_HASHING(1, 1) << 16)
 #define   GEN6_TD_FOUR_ROW_DISPATCH_DISABLE		(1 << 5)
 
 #define GFX_MODE	0x02520
@@ -944,6 +977,9 @@
 #define   GEN6_BLITTER_LOCK_SHIFT			16
 #define   GEN6_BLITTER_FBC_NOTIFY			(1<<3)
 
+#define GEN6_RC_SLEEP_PSMI_CONTROL	0x2050
+#define   GEN8_RC_SEMA_IDLE_MSG_DISABLE	(1 << 12)
+
 #define GEN6_BSD_SLEEP_PSMI_CONTROL	0x12050
 #define   GEN6_BSD_SLEEP_MSG_DISABLE	(1 << 0)
 #define   GEN6_BSD_SLEEP_FLUSH_DISABLE	(1 << 2)
@@ -1121,13 +1157,6 @@
 #define   FBC_REND_NUKE		(1<<2)
 #define   FBC_REND_CACHE_CLEAN	(1<<1)
 
-#define _HSW_PIPE_SLICE_CHICKEN_1_A	0x420B0
-#define _HSW_PIPE_SLICE_CHICKEN_1_B	0x420B4
-#define   HSW_BYPASS_FBC_QUEUE		(1<<22)
-#define HSW_PIPE_SLICE_CHICKEN_1(pipe) _PIPE(pipe, + \
-					     _HSW_PIPE_SLICE_CHICKEN_1_A, + \
-					     _HSW_PIPE_SLICE_CHICKEN_1_B)
-
 /*
  * GPIO regs
  */
@@ -4140,7 +4169,8 @@
 
 #define _CHICKEN_PIPESL_1_A	0x420b0
 #define _CHICKEN_PIPESL_1_B	0x420b4
-#define  DPRS_MASK_VBLANK_SRD	(1 << 0)
+#define  HSW_FBCQ_DIS			(1 << 22)
+#define  BDW_DPRS_MASK_VBLANK_SRD	(1 << 0)
 #define CHICKEN_PIPESL_1(pipe) _PIPE(pipe, _CHICKEN_PIPESL_1_A, _CHICKEN_PIPESL_1_B)
 
 #define DISP_ARB_CTL	0x45000
@@ -4164,7 +4194,7 @@
 #define  VLV_B0_WA_L3SQCREG1_VALUE		0x00D30000
 
 #define GEN7_L3CNTLREG1				0xB01C
-#define  GEN7_WA_FOR_GEN7_L3_CONTROL			0x3C4FFF8C
+#define  GEN7_WA_FOR_GEN7_L3_CONTROL			0x3C47FF8C
 #define  GEN7_L3AGDIS				(1<<19)
 
 #define GEN7_L3_CHICKEN_MODE_REGISTER		0xB030
@@ -4898,6 +4928,9 @@
 #define GEN7_UCGCTL4				0x940c
 #define  GEN7_L3BANK2X_CLOCK_GATE_DISABLE	(1<<25)
 
+#define GEN8_UCGCTL6				0x9430
+#define   GEN8_SDEUNIT_CLOCK_GATE_DISABLE	(1<<14)
+
 #define GEN6_RPNSWREQ				0xA008
 #define   GEN6_TURBO_DISABLE			(1<<31)
 #define   GEN6_FREQUENCY(x)			((x)<<25)
@@ -5043,6 +5076,10 @@
 #define   GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE	(1<<10)
 #define   GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE	(1<<3)
 
+#define GEN8_ROW_CHICKEN		0xe4f0
+#define   PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE	(1<<8)
+#define   STALL_DOP_GATING_DISABLE		(1<<5)
+
 #define GEN7_ROW_CHICKEN2		0xe4f4
 #define GEN7_ROW_CHICKEN2_GT2		0xf4f4
 #define   DOP_CLOCK_GATING_DISABLE	(1<<0)
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index 86b95ca413d1..4867f4cc0938 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -599,14 +599,14 @@ parse_mipi(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 {
 	struct bdb_mipi *mipi;
 
-	mipi = find_section(bdb, BDB_MIPI);
+	mipi = find_section(bdb, BDB_MIPI_CONFIG);
 	if (!mipi) {
 		DRM_DEBUG_KMS("No MIPI BDB found");
 		return;
 	}
 
 	/* XXX: add more info */
-	dev_priv->vbt.dsi.panel_id = mipi->panel_id;
+	dev_priv->vbt.dsi.panel_id = MIPI_DSI_GENERIC_PANEL_ID;
 }
 
 static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index 282de5e9f39d..83b7629e4367 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -104,7 +104,8 @@ struct vbios_data {
 #define BDB_LVDS_LFP_DATA	 42
 #define BDB_LVDS_BACKLIGHT	 43
 #define BDB_LVDS_POWER		 44
-#define BDB_MIPI		 50
+#define BDB_MIPI_CONFIG		 52
+#define BDB_MIPI_SEQUENCE	 53
 #define BDB_SKIP		254 /* VBIOS private block, ignore */
 
 struct bdb_general_features {
@@ -711,44 +712,159 @@ int intel_parse_bios(struct drm_device *dev);
 #define DVO_PORT_DPD	9
 #define DVO_PORT_DPA	10
 
-/* MIPI DSI panel info */
-struct bdb_mipi {
-	u16 panel_id;
-	u16 bridge_revision;
-
-	/* General params */
-	u32 dithering:1;
-	u32 bpp_pixel_format:1;
-	u32 rsvd1:1;
-	u32 dphy_valid:1;
-	u32 resvd2:28;
+/* Block 52 contains MIPI Panel info
+ * 6 such enteries will there. Index into correct
+ * entery is based on the panel_index in #40 LFP
+ */
+#define MAX_MIPI_CONFIGURATIONS	6
 
-	u16 port_info;
-	u16 rsvd3:2;
-	u16 num_lanes:2;
-	u16 rsvd4:12;
+#define MIPI_DSI_UNDEFINED_PANEL_ID	0
+#define MIPI_DSI_GENERIC_PANEL_ID	1
 
-	/* DSI config */
-	u16 virt_ch_num:2;
-	u16 vtm:2;
-	u16 rsvd5:12;
+struct mipi_config {
+	u16 panel_id;
 
-	u32 dsi_clock;
+	/* General Params */
+	u32 enable_dithering:1;
+	u32 rsvd1:1;
+	u32 is_bridge:1;
+
+	u32 panel_arch_type:2;
+	u32 is_cmd_mode:1;
+
+#define NON_BURST_SYNC_PULSE	0x1
+#define NON_BURST_SYNC_EVENTS	0x2
+#define BURST_MODE		0x3
+	u32 video_transfer_mode:2;
+
+	u32 cabc_supported:1;
+	u32 pwm_blc:1;
+
+	/* Bit 13:10 */
+#define PIXEL_FORMAT_RGB565			0x1
+#define PIXEL_FORMAT_RGB666			0x2
+#define PIXEL_FORMAT_RGB666_LOOSELY_PACKED	0x3
+#define PIXEL_FORMAT_RGB888			0x4
+	u32 videomode_color_format:4;
+
+	/* Bit 15:14 */
+#define ENABLE_ROTATION_0	0x0
+#define ENABLE_ROTATION_90	0x1
+#define ENABLE_ROTATION_180	0x2
+#define ENABLE_ROTATION_270	0x3
+	u32 rotation:2;
+	u32 bta_enabled:1;
+	u32 rsvd2:15;
+
+	/* 2 byte Port Description */
+#define DUAL_LINK_NOT_SUPPORTED	0
+#define DUAL_LINK_FRONT_BACK	1
+#define DUAL_LINK_PIXEL_ALT	2
+	u16 dual_link:2;
+	u16 lane_cnt:2;
+	u16 rsvd3:12;
+
+	u16 rsvd4;
+
+	u8 rsvd5[5];
+	u32 dsi_ddr_clk;
 	u32 bridge_ref_clk;
-	u16 rsvd_pwr;
 
-	/* Dphy Params */
-	u32 prepare_cnt:5;
-	u32 rsvd6:3;
+#define  BYTE_CLK_SEL_20MHZ		0
+#define  BYTE_CLK_SEL_10MHZ		1
+#define  BYTE_CLK_SEL_5MHZ		2
+	u8 byte_clk_sel:2;
+
+	u8 rsvd6:6;
+
+	/* DPHY Flags */
+	u16 dphy_param_valid:1;
+	u16 eot_pkt_disabled:1;
+	u16 enable_clk_stop:1;
+	u16 rsvd7:13;
+
+	u32 hs_tx_timeout;
+	u32 lp_rx_timeout;
+	u32 turn_around_timeout;
+	u32 device_reset_timer;
+	u32 master_init_timer;
+	u32 dbi_bw_timer;
+	u32 lp_byte_clk_val;
+
+	/*  4 byte Dphy Params */
+	u32 prepare_cnt:6;
+	u32 rsvd8:2;
 	u32 clk_zero_cnt:8;
 	u32 trail_cnt:5;
-	u32 rsvd7:3;
+	u32 rsvd9:3;
 	u32 exit_zero_cnt:6;
-	u32 rsvd8:2;
+	u32 rsvd10:2;
 
-	u32 hl_switch_cnt;
-	u32 lp_byte_clk;
 	u32 clk_lane_switch_cnt;
+	u32 hl_switch_cnt;
+
+	u32 rsvd11[6];
+
+	/* timings based on dphy spec */
+	u8 tclk_miss;
+	u8 tclk_post;
+	u8 rsvd12;
+	u8 tclk_pre;
+	u8 tclk_prepare;
+	u8 tclk_settle;
+	u8 tclk_term_enable;
+	u8 tclk_trail;
+	u16 tclk_prepare_clkzero;
+	u8 rsvd13;
+	u8 td_term_enable;
+	u8 teot;
+	u8 ths_exit;
+	u8 ths_prepare;
+	u16 ths_prepare_hszero;
+	u8 rsvd14;
+	u8 ths_settle;
+	u8 ths_skip;
+	u8 ths_trail;
+	u8 tinit;
+	u8 tlpx;
+	u8 rsvd15[3];
+
+	/* GPIOs */
+	u8 panel_enable;
+	u8 bl_enable;
+	u8 pwm_enable;
+	u8 reset_r_n;
+	u8 pwr_down_r;
+	u8 stdby_r_n;
+
 } __packed;
 
+/* Block 52 contains MIPI configuration block
+ * 6 * bdb_mipi_config, followed by 6 pps data
+ * block below
+ *
+ * all delays has a unit of 100us
+ */
+struct mipi_pps_data {
+	u16 panel_on_delay;
+	u16 bl_enable_delay;
+	u16 bl_disable_delay;
+	u16 panel_off_delay;
+	u16 panel_power_cycle_delay;
+};
+
+struct bdb_mipi_config {
+	struct mipi_config config[MAX_MIPI_CONFIGURATIONS];
+	struct mipi_pps_data pps[MAX_MIPI_CONFIGURATIONS];
+};
+
+/* Block 53 contains MIPI sequences as needed by the panel
+ * for enabling it. This block can be variable in size and
+ * can be maximum of 6 blocks
+ */
+struct bdb_mipi_sequence {
+	u8 version;
+	u8 data[0];
+};
+
 #endif /* _I830_BIOS_H_ */
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 9864aa1ccbe8..4ef6d69c078d 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -68,8 +68,13 @@ static bool intel_crt_get_hw_state(struct intel_encoder *encoder,
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crt *crt = intel_encoder_to_crt(encoder);
+	enum intel_display_power_domain power_domain;
 	u32 tmp;
 
+	power_domain = intel_display_port_power_domain(encoder);
+	if (!intel_display_power_enabled(dev_priv, power_domain))
+		return false;
+
 	tmp = I915_READ(crt->adpa_reg);
 
 	if (!(tmp & ADPA_DAC_ENABLE))
@@ -262,6 +267,10 @@ static bool intel_crt_compute_config(struct intel_encoder *encoder,
 	if (HAS_PCH_LPT(dev))
 		pipe_config->pipe_bpp = 24;
 
+	/* FDI must always be 2.7 GHz */
+	if (HAS_DDI(dev))
+		pipe_config->port_clock = 135000 * 2;
+
 	return true;
 }
 
@@ -630,14 +639,22 @@ static enum drm_connector_status
 intel_crt_detect(struct drm_connector *connector, bool force)
 {
 	struct drm_device *dev = connector->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crt *crt = intel_attached_crt(connector);
+	struct intel_encoder *intel_encoder = &crt->base;
+	enum intel_display_power_domain power_domain;
 	enum drm_connector_status status;
 	struct intel_load_detect_pipe tmp;
 
+	intel_runtime_pm_get(dev_priv);
+
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] force=%d\n",
 		      connector->base.id, drm_get_connector_name(connector),
 		      force);
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	if (I915_HAS_HOTPLUG(dev)) {
 		/* We can not rely on the HPD pin always being correctly wired
 		 * up, for example many KVM do not pass it through, and so
@@ -645,23 +662,30 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 		 */
 		if (intel_crt_detect_hotplug(connector)) {
 			DRM_DEBUG_KMS("CRT detected via hotplug\n");
-			return connector_status_connected;
+			status = connector_status_connected;
+			goto out;
 		} else
 			DRM_DEBUG_KMS("CRT not detected via hotplug\n");
 	}
 
-	if (intel_crt_detect_ddc(connector))
-		return connector_status_connected;
+	if (intel_crt_detect_ddc(connector)) {
+		status = connector_status_connected;
+		goto out;
+	}
 
 	/* Load detection is broken on HPD capable machines. Whoever wants a
 	 * broken monitor (without edid) to work behind a broken kvm (that fails
 	 * to have the right resistors for HP detection) needs to fix this up.
 	 * For now just bail out. */
-	if (I915_HAS_HOTPLUG(dev))
-		return connector_status_disconnected;
+	if (I915_HAS_HOTPLUG(dev)) {
+		status = connector_status_disconnected;
+		goto out;
+	}
 
-	if (!force)
-		return connector->status;
+	if (!force) {
+		status = connector->status;
+		goto out;
+	}
 
 	/* for pre-945g platforms use load detect */
 	if (intel_get_load_detect_pipe(connector, NULL, &tmp)) {
@@ -673,6 +697,10 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 	} else
 		status = connector_status_unknown;
 
+out:
+	intel_display_power_put(dev_priv, power_domain);
+	intel_runtime_pm_put(dev_priv);
+
 	return status;
 }
 
@@ -686,17 +714,28 @@ static int intel_crt_get_modes(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crt *crt = intel_attached_crt(connector);
+	struct intel_encoder *intel_encoder = &crt->base;
+	enum intel_display_power_domain power_domain;
 	int ret;
 	struct i2c_adapter *i2c;
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	i2c = intel_gmbus_get_adapter(dev_priv, dev_priv->vbt.crt_ddc_pin);
 	ret = intel_crt_ddc_get_modes(connector, i2c);
 	if (ret || !IS_G4X(dev))
-		return ret;
+		goto out;
 
 	/* Try to probe digital port for output in DVI-I -> VGA mode. */
 	i2c = intel_gmbus_get_adapter(dev_priv, GMBUS_PORT_DPB);
-	return intel_crt_ddc_get_modes(connector, i2c);
+	ret = intel_crt_ddc_get_modes(connector, i2c);
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static int intel_crt_set_property(struct drm_connector *connector,
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 2643d3b8b67d..e2665e09d5df 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -1145,9 +1145,14 @@ bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum port port = intel_ddi_get_encoder_port(encoder);
+	enum intel_display_power_domain power_domain;
 	u32 tmp;
 	int i;
 
+	power_domain = intel_display_port_power_domain(encoder);
+	if (!intel_display_power_enabled(dev_priv, power_domain))
+		return false;
+
 	tmp = I915_READ(DDI_BUF_CTL(port));
 
 	if (!(tmp & DDI_BUF_CTL_ENABLE))
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index f19e6ea36dc4..11746e28243a 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1122,7 +1122,7 @@ void assert_pipe(struct drm_i915_private *dev_priv,
 	if (pipe == PIPE_A && dev_priv->quirks & QUIRK_PIPEA_FORCE)
 		state = true;
 
-	if (!intel_display_power_enabled(dev_priv->dev,
+	if (!intel_display_power_enabled(dev_priv,
 				POWER_DOMAIN_TRANSCODER(cpu_transcoder))) {
 		cur_state = false;
 	} else {
@@ -1188,16 +1188,16 @@ static void assert_sprites_disabled(struct drm_i915_private *dev_priv,
 				    enum pipe pipe)
 {
 	struct drm_device *dev = dev_priv->dev;
-	int reg, i;
+	int reg, sprite;
 	u32 val;
 
 	if (IS_VALLEYVIEW(dev)) {
-		for (i = 0; i < INTEL_INFO(dev)->num_sprites; i++) {
-			reg = SPCNTR(pipe, i);
+		for_each_sprite(pipe, sprite) {
+			reg = SPCNTR(pipe, sprite);
 			val = I915_READ(reg);
 			WARN((val & SP_ENABLE),
 			     "sprite %c assertion failure, should be off on pipe %c but is still active\n",
-			     sprite_name(pipe, i), pipe_name(pipe));
+			     sprite_name(pipe, sprite), pipe_name(pipe));
 		}
 	} else if (INTEL_INFO(dev)->gen >= 7) {
 		reg = SPRCTL(pipe);
@@ -2321,6 +2321,25 @@ intel_finish_fb(struct drm_framebuffer *old_fb)
 	return ret;
 }
 
+static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	unsigned long flags;
+	bool pending;
+
+	if (i915_reset_in_progress(&dev_priv->gpu_error) ||
+	    intel_crtc->reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
+		return false;
+
+	spin_lock_irqsave(&dev->event_lock, flags);
+	pending = to_intel_crtc(crtc)->unpin_work != NULL;
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+
+	return pending;
+}
+
 static int
 intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		    struct drm_framebuffer *fb)
@@ -2331,6 +2350,11 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 	struct drm_framebuffer *old_fb;
 	int ret;
 
+	if (intel_crtc_has_pending_flip(crtc)) {
+		DRM_ERROR("pipe is still busy with an old pageflip\n");
+		return -EBUSY;
+	}
+
 	/* no fb bound */
 	if (!fb) {
 		DRM_ERROR("No FB bound\n");
@@ -2956,25 +2980,6 @@ static void ironlake_fdi_disable(struct drm_crtc *crtc)
 	udelay(100);
 }
 
-static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	unsigned long flags;
-	bool pending;
-
-	if (i915_reset_in_progress(&dev_priv->gpu_error) ||
-	    intel_crtc->reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
-		return false;
-
-	spin_lock_irqsave(&dev->event_lock, flags);
-	pending = to_intel_crtc(crtc)->unpin_work != NULL;
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-
-	return pending;
-}
-
 bool intel_has_pending_fb_unpin(struct drm_device *dev)
 {
 	struct intel_crtc *crtc;
@@ -3953,6 +3958,117 @@ static void i9xx_pfit_enable(struct intel_crtc *crtc)
 	I915_WRITE(BCLRPAT(crtc->pipe), 0);
 }
 
+#define for_each_power_domain(domain, mask)				\
+	for ((domain) = 0; (domain) < POWER_DOMAIN_NUM; (domain)++)	\
+		if ((1 << (domain)) & (mask))
+
+enum intel_display_power_domain
+intel_display_port_power_domain(struct intel_encoder *intel_encoder)
+{
+	struct drm_device *dev = intel_encoder->base.dev;
+	struct intel_digital_port *intel_dig_port;
+
+	switch (intel_encoder->type) {
+	case INTEL_OUTPUT_UNKNOWN:
+		/* Only DDI platforms should ever use this output type */
+		WARN_ON_ONCE(!HAS_DDI(dev));
+	case INTEL_OUTPUT_DISPLAYPORT:
+	case INTEL_OUTPUT_HDMI:
+	case INTEL_OUTPUT_EDP:
+		intel_dig_port = enc_to_dig_port(&intel_encoder->base);
+		switch (intel_dig_port->port) {
+		case PORT_A:
+			return POWER_DOMAIN_PORT_DDI_A_4_LANES;
+		case PORT_B:
+			return POWER_DOMAIN_PORT_DDI_B_4_LANES;
+		case PORT_C:
+			return POWER_DOMAIN_PORT_DDI_C_4_LANES;
+		case PORT_D:
+			return POWER_DOMAIN_PORT_DDI_D_4_LANES;
+		default:
+			WARN_ON_ONCE(1);
+			return POWER_DOMAIN_PORT_OTHER;
+		}
+	case INTEL_OUTPUT_ANALOG:
+		return POWER_DOMAIN_PORT_CRT;
+	case INTEL_OUTPUT_DSI:
+		return POWER_DOMAIN_PORT_DSI;
+	default:
+		return POWER_DOMAIN_PORT_OTHER;
+	}
+}
+
+static unsigned long get_crtc_power_domains(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct intel_encoder *intel_encoder;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	bool pfit_enabled = intel_crtc->config.pch_pfit.enabled;
+	unsigned long mask;
+	enum transcoder transcoder;
+
+	transcoder = intel_pipe_to_cpu_transcoder(dev->dev_private, pipe);
+
+	mask = BIT(POWER_DOMAIN_PIPE(pipe));
+	mask |= BIT(POWER_DOMAIN_TRANSCODER(transcoder));
+	if (pfit_enabled)
+		mask |= BIT(POWER_DOMAIN_PIPE_PANEL_FITTER(pipe));
+
+	for_each_encoder_on_crtc(dev, crtc, intel_encoder)
+		mask |= BIT(intel_display_port_power_domain(intel_encoder));
+
+	return mask;
+}
+
+void intel_display_set_init_power(struct drm_i915_private *dev_priv,
+				  bool enable)
+{
+	if (dev_priv->power_domains.init_power_on == enable)
+		return;
+
+	if (enable)
+		intel_display_power_get(dev_priv, POWER_DOMAIN_INIT);
+	else
+		intel_display_power_put(dev_priv, POWER_DOMAIN_INIT);
+
+	dev_priv->power_domains.init_power_on = enable;
+}
+
+static void modeset_update_crtc_power_domains(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long pipe_domains[I915_MAX_PIPES] = { 0, };
+	struct intel_crtc *crtc;
+
+	/*
+	 * First get all needed power domains, then put all unneeded, to avoid
+	 * any unnecessary toggling of the power wells.
+	 */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+		enum intel_display_power_domain domain;
+
+		if (!crtc->base.enabled)
+			continue;
+
+		pipe_domains[crtc->pipe] = get_crtc_power_domains(&crtc->base);
+
+		for_each_power_domain(domain, pipe_domains[crtc->pipe])
+			intel_display_power_get(dev_priv, domain);
+	}
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+		enum intel_display_power_domain domain;
+
+		for_each_power_domain(domain, crtc->enabled_power_domains)
+			intel_display_power_put(dev_priv, domain);
+
+		crtc->enabled_power_domains = pipe_domains[crtc->pipe];
+	}
+
+	intel_display_set_init_power(dev_priv, false);
+}
+
 int valleyview_get_vco(struct drm_i915_private *dev_priv)
 {
 	int hpll_freq, vco_freq[] = { 800, 1600, 2000, 2400 };
@@ -4113,6 +4229,7 @@ static void valleyview_modeset_global_resources(struct drm_device *dev)
 
 	if (req_cdclk != cur_cdclk)
 		valleyview_set_cdclk(dev, req_cdclk);
+	modeset_update_crtc_power_domains(dev);
 }
 
 static void valleyview_crtc_enable(struct drm_crtc *crtc)
@@ -5495,6 +5612,10 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t tmp;
 
+	if (!intel_display_power_enabled(dev_priv,
+					 POWER_DOMAIN_PIPE(crtc->pipe)))
+		return false;
+
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
 	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
 
@@ -6156,7 +6277,7 @@ int ironlake_get_lanes_required(int target_clock, int link_bw, int bpp)
 	 * is 2.5%; use 5% for safety's sake.
 	 */
 	u32 bps = target_clock * bpp * 21 / 20;
-	return bps / (link_bw * 8) + 1;
+	return DIV_ROUND_UP(bps, link_bw * 8);
 }
 
 static bool ironlake_needs_fb_cb_tune(struct dpll *dpll, int factor)
@@ -6812,105 +6933,9 @@ done:
 	mutex_unlock(&dev_priv->pc8.lock);
 }
 
-static void hsw_package_c8_gpu_idle(struct drm_i915_private *dev_priv)
-{
-	if (!HAS_PC8(dev_priv->dev))
-		return;
-
-	mutex_lock(&dev_priv->pc8.lock);
-	if (!dev_priv->pc8.gpu_idle) {
-		dev_priv->pc8.gpu_idle = true;
-		__hsw_enable_package_c8(dev_priv);
-	}
-	mutex_unlock(&dev_priv->pc8.lock);
-}
-
-static void hsw_package_c8_gpu_busy(struct drm_i915_private *dev_priv)
-{
-	if (!HAS_PC8(dev_priv->dev))
-		return;
-
-	mutex_lock(&dev_priv->pc8.lock);
-	if (dev_priv->pc8.gpu_idle) {
-		dev_priv->pc8.gpu_idle = false;
-		__hsw_disable_package_c8(dev_priv);
-	}
-	mutex_unlock(&dev_priv->pc8.lock);
-}
-
-#define for_each_power_domain(domain, mask)				\
-	for ((domain) = 0; (domain) < POWER_DOMAIN_NUM; (domain)++)	\
-		if ((1 << (domain)) & (mask))
-
-static unsigned long get_pipe_power_domains(struct drm_device *dev,
-					    enum pipe pipe, bool pfit_enabled)
-{
-	unsigned long mask;
-	enum transcoder transcoder;
-
-	transcoder = intel_pipe_to_cpu_transcoder(dev->dev_private, pipe);
-
-	mask = BIT(POWER_DOMAIN_PIPE(pipe));
-	mask |= BIT(POWER_DOMAIN_TRANSCODER(transcoder));
-	if (pfit_enabled)
-		mask |= BIT(POWER_DOMAIN_PIPE_PANEL_FITTER(pipe));
-
-	return mask;
-}
-
-void intel_display_set_init_power(struct drm_device *dev, bool enable)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (dev_priv->power_domains.init_power_on == enable)
-		return;
-
-	if (enable)
-		intel_display_power_get(dev, POWER_DOMAIN_INIT);
-	else
-		intel_display_power_put(dev, POWER_DOMAIN_INIT);
-
-	dev_priv->power_domains.init_power_on = enable;
-}
-
-static void modeset_update_power_wells(struct drm_device *dev)
-{
-	unsigned long pipe_domains[I915_MAX_PIPES] = { 0, };
-	struct intel_crtc *crtc;
-
-	/*
-	 * First get all needed power domains, then put all unneeded, to avoid
-	 * any unnecessary toggling of the power wells.
-	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
-		enum intel_display_power_domain domain;
-
-		if (!crtc->base.enabled)
-			continue;
-
-		pipe_domains[crtc->pipe] = get_pipe_power_domains(dev,
-						crtc->pipe,
-						crtc->config.pch_pfit.enabled);
-
-		for_each_power_domain(domain, pipe_domains[crtc->pipe])
-			intel_display_power_get(dev, domain);
-	}
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
-		enum intel_display_power_domain domain;
-
-		for_each_power_domain(domain, crtc->enabled_power_domains)
-			intel_display_power_put(dev, domain);
-
-		crtc->enabled_power_domains = pipe_domains[crtc->pipe];
-	}
-
-	intel_display_set_init_power(dev, false);
-}
-
 static void haswell_modeset_global_resources(struct drm_device *dev)
 {
-	modeset_update_power_wells(dev);
+	modeset_update_crtc_power_domains(dev);
 	hsw_update_package_c8(dev);
 }
 
@@ -6961,6 +6986,10 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	enum intel_display_power_domain pfit_domain;
 	uint32_t tmp;
 
+	if (!intel_display_power_enabled(dev_priv,
+					 POWER_DOMAIN_PIPE(crtc->pipe)))
+		return false;
+
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
 	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
 
@@ -6986,7 +7015,7 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 			pipe_config->cpu_transcoder = TRANSCODER_EDP;
 	}
 
-	if (!intel_display_power_enabled(dev,
+	if (!intel_display_power_enabled(dev_priv,
 			POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder)))
 		return false;
 
@@ -7014,7 +7043,7 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	intel_get_pipe_timings(crtc, pipe_config);
 
 	pfit_domain = POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe);
-	if (intel_display_power_enabled(dev, pfit_domain))
+	if (intel_display_power_enabled(dev_priv, pfit_domain))
 		ironlake_get_pfit_config(crtc, pipe_config);
 
 	if (IS_HASWELL(dev))
@@ -7549,7 +7578,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 		return -ENOENT;
 
 	if (obj->base.size < width * height * 4) {
-		DRM_ERROR("buffer is to small\n");
+		DRM_DEBUG_KMS("buffer is to small\n");
 		ret = -ENOMEM;
 		goto fail;
 	}
@@ -7560,7 +7589,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 		unsigned alignment;
 
 		if (obj->tiling_mode) {
-			DRM_ERROR("cursor cannot be tiled\n");
+			DRM_DEBUG_KMS("cursor cannot be tiled\n");
 			ret = -EINVAL;
 			goto fail_locked;
 		}
@@ -7576,13 +7605,13 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 
 		ret = i915_gem_object_pin_to_display_plane(obj, alignment, NULL);
 		if (ret) {
-			DRM_ERROR("failed to move cursor bo into the GTT\n");
+			DRM_DEBUG_KMS("failed to move cursor bo into the GTT\n");
 			goto fail_locked;
 		}
 
 		ret = i915_gem_object_put_fence(obj);
 		if (ret) {
-			DRM_ERROR("failed to release fence for cursor");
+			DRM_DEBUG_KMS("failed to release fence for cursor");
 			goto fail_unpin;
 		}
 
@@ -7593,7 +7622,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 						  (intel_crtc->pipe == 0) ? I915_GEM_PHYS_CURSOR_0 : I915_GEM_PHYS_CURSOR_1,
 						  align);
 		if (ret) {
-			DRM_ERROR("failed to attach phys object\n");
+			DRM_DEBUG_KMS("failed to attach phys object\n");
 			goto fail_locked;
 		}
 		addr = obj->phys_obj->handle->busaddr;
@@ -7692,7 +7721,7 @@ err:
 	return ERR_PTR(ret);
 }
 
-struct drm_framebuffer *
+static struct drm_framebuffer *
 intel_framebuffer_create(struct drm_device *dev,
 			 struct drm_mode_fb_cmd2 *mode_cmd,
 			 struct drm_i915_gem_object *obj)
@@ -8192,8 +8221,12 @@ void intel_mark_busy(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	hsw_package_c8_gpu_busy(dev_priv);
+	if (dev_priv->mm.busy)
+		return;
+
+	hsw_disable_package_c8(dev_priv);
 	i915_update_gfx_val(dev_priv);
+	dev_priv->mm.busy = true;
 }
 
 void intel_mark_idle(struct drm_device *dev)
@@ -8201,10 +8234,13 @@ void intel_mark_idle(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc;
 
-	hsw_package_c8_gpu_idle(dev_priv);
+	if (!dev_priv->mm.busy)
+		return;
+
+	dev_priv->mm.busy = false;
 
 	if (!i915.powersave)
-		return;
+		goto out;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		if (!crtc->fb)
@@ -8215,6 +8251,9 @@ void intel_mark_idle(struct drm_device *dev)
 
 	if (INTEL_INFO(dev)->gen >= 6)
 		gen6_rps_idle(dev->dev_private);
+
+out:
+	hsw_enable_package_c8(dev_priv);
 }
 
 void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
@@ -8678,6 +8717,9 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	     fb->pitches[0] != crtc->fb->pitches[0]))
 		return -EINVAL;
 
+	if (i915_terminally_wedged(&dev_priv->gpu_error))
+		goto out_hang;
+
 	work = kzalloc(sizeof(*work), GFP_KERNEL);
 	if (work == NULL)
 		return -ENOMEM;
@@ -8752,6 +8794,13 @@ cleanup:
 free_work:
 	kfree(work);
 
+	if (ret == -EIO) {
+out_hang:
+		intel_crtc_wait_for_pending_flips(crtc);
+		ret = intel_pipe_set_base(crtc, crtc->x, crtc->y, fb);
+		if (ret == 0 && event)
+			drm_send_vblank_event(dev, intel_crtc->pipe, event);
+	}
 	return ret;
 }
 
@@ -10555,10 +10604,10 @@ static const struct drm_framebuffer_funcs intel_fb_funcs = {
 	.create_handle = intel_user_framebuffer_create_handle,
 };
 
-int intel_framebuffer_init(struct drm_device *dev,
-			   struct intel_framebuffer *intel_fb,
-			   struct drm_mode_fb_cmd2 *mode_cmd,
-			   struct drm_i915_gem_object *obj)
+static int intel_framebuffer_init(struct drm_device *dev,
+				  struct intel_framebuffer *intel_fb,
+				  struct drm_mode_fb_cmd2 *mode_cmd,
+				  struct drm_i915_gem_object *obj)
 {
 	int aligned_height;
 	int pitch_limit;
@@ -10996,7 +11045,8 @@ void intel_modeset_suspend_hw(struct drm_device *dev)
 void intel_modeset_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int i, j, ret;
+	int sprite, ret;
+	enum pipe pipe;
 
 	drm_mode_config_init(dev);
 
@@ -11033,13 +11083,13 @@ void intel_modeset_init(struct drm_device *dev)
 		      INTEL_INFO(dev)->num_pipes,
 		      INTEL_INFO(dev)->num_pipes > 1 ? "s" : "");
 
-	for_each_pipe(i) {
-		intel_crtc_init(dev, i);
-		for (j = 0; j < INTEL_INFO(dev)->num_sprites; j++) {
-			ret = intel_plane_init(dev, i, j);
+	for_each_pipe(pipe) {
+		intel_crtc_init(dev, pipe);
+		for_each_sprite(pipe, sprite) {
+			ret = intel_plane_init(dev, pipe, sprite);
 			if (ret)
 				DRM_DEBUG_KMS("pipe %c sprite %c init failed: %d\n",
-					      pipe_name(i), sprite_name(i, j), ret);
+					      pipe_name(pipe), sprite_name(pipe, sprite), ret);
 		}
 	}
 
@@ -11056,7 +11106,9 @@ void intel_modeset_init(struct drm_device *dev)
 	/* Just in case the BIOS is doing something questionable. */
 	intel_disable_fbc(dev);
 
+	mutex_lock(&dev->mode_config.mutex);
 	intel_modeset_setup_hw_state(dev, false);
+	mutex_unlock(&dev->mode_config.mutex);
 }
 
 static void
@@ -11239,11 +11291,21 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 	 * the crtc fixup. */
 }
 
-void i915_redisable_vga(struct drm_device *dev)
+void i915_redisable_vga_power_on(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 vga_reg = i915_vgacntrl_reg(dev);
 
+	if (!(I915_READ(vga_reg) & VGA_DISP_DISABLE)) {
+		DRM_DEBUG_KMS("Something enabled VGA plane, disabling it\n");
+		i915_disable_vga(dev);
+	}
+}
+
+void i915_redisable_vga(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
 	/* This function can be called both from intel_modeset_setup_hw_state or
 	 * at a very early point in our resume sequence, where the power well
 	 * structures are not yet restored. Since this function is at a very
@@ -11251,14 +11313,10 @@ void i915_redisable_vga(struct drm_device *dev)
 	 * level, just check if the power well is enabled instead of trying to
 	 * follow the "don't touch the power well if we don't need it" policy
 	 * the rest of the driver uses. */
-	if ((IS_HASWELL(dev) || IS_BROADWELL(dev)) &&
-	    (I915_READ(HSW_PWR_WELL_DRIVER) & HSW_PWR_WELL_STATE_ENABLED) == 0)
+	if (!intel_display_power_enabled(dev_priv, POWER_DOMAIN_VGA))
 		return;
 
-	if (!(I915_READ(vga_reg) & VGA_DISP_DISABLE)) {
-		DRM_DEBUG_KMS("Something enabled VGA plane, disabling it\n");
-		i915_disable_vga(dev);
-	}
+	i915_redisable_vga_power_on(dev);
 }
 
 static void intel_modeset_readout_hw_state(struct drm_device *dev)
@@ -11602,7 +11660,8 @@ intel_display_capture_error_state(struct drm_device *dev)
 
 	for_each_pipe(i) {
 		error->pipe[i].power_domain_on =
-			intel_display_power_enabled_sw(dev, POWER_DOMAIN_PIPE(i));
+			intel_display_power_enabled_sw(dev_priv,
+						       POWER_DOMAIN_PIPE(i));
 		if (!error->pipe[i].power_domain_on)
 			continue;
 
@@ -11640,7 +11699,7 @@ intel_display_capture_error_state(struct drm_device *dev)
 		enum transcoder cpu_transcoder = transcoders[i];
 
 		error->transcoder[i].power_domain_on =
-			intel_display_power_enabled_sw(dev,
+			intel_display_power_enabled_sw(dev_priv,
 				POWER_DOMAIN_TRANSCODER(cpu_transcoder));
 		if (!error->transcoder[i].power_domain_on)
 			continue;
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index c512d78af271..49d12d341ab2 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -916,8 +916,8 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 		mode_rate = intel_dp_link_required(adjusted_mode->crtc_clock,
 						   bpp);
 
-		for (clock = 0; clock <= max_clock; clock++) {
-			for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
+		for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
+			for (clock = 0; clock <= max_clock; clock++) {
 				link_clock = drm_dp_bw_code_to_link_rate(bws[clock]);
 				link_avail = intel_dp_max_data_rate(link_clock,
 								    lane_count);
@@ -1333,7 +1333,8 @@ void intel_edp_panel_off(struct intel_dp *intel_dp)
 	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 | PANEL_POWER_RESET | EDP_FORCE_VDD |
+		EDP_BLC_ENABLE);
 
 	pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 
@@ -1485,7 +1486,14 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder,
 	enum port port = dp_to_dig_port(intel_dp)->port;
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 tmp = I915_READ(intel_dp->output_reg);
+	enum intel_display_power_domain power_domain;
+	u32 tmp;
+
+	power_domain = intel_display_port_power_domain(encoder);
+	if (!intel_display_power_enabled(dev_priv, power_domain))
+		return false;
+
+	tmp = I915_READ(intel_dp->output_reg);
 
 	if (!(tmp & DP_PORT_EN))
 		return false;
@@ -1868,9 +1876,11 @@ 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. */
+	edp_panel_vdd_on(intel_dp);
 	intel_edp_backlight_off(intel_dp);
 	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_OFF);
 	intel_edp_panel_off(intel_dp);
+	edp_panel_vdd_off(intel_dp, true);
 
 	/* cpu edp my only be disable _after_ the cpu pipe/plane is disabled. */
 	if (!(port == PORT_A || IS_VALLEYVIEW(dev)))
@@ -3224,10 +3234,14 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	struct drm_device *dev = connector->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum drm_connector_status status;
+	enum intel_display_power_domain power_domain;
 	struct edid *edid = NULL;
 
 	intel_runtime_pm_get(dev_priv);
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, drm_get_connector_name(connector));
 
@@ -3258,21 +3272,32 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	status = connector_status_connected;
 
 out:
+	intel_display_power_put(dev_priv, power_domain);
+
 	intel_runtime_pm_put(dev_priv);
+
 	return status;
 }
 
 static int intel_dp_get_modes(struct drm_connector *connector)
 {
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct drm_device *dev = connector->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	int ret;
 
 	/* We should parse the EDID data and find out if it has an audio sink
 	 */
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	ret = intel_dp_get_edid_modes(connector, &intel_dp->adapter);
+	intel_display_power_put(dev_priv, power_domain);
 	if (ret)
 		return ret;
 
@@ -3293,15 +3318,25 @@ static bool
 intel_dp_detect_audio(struct drm_connector *connector)
 {
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
+	struct drm_device *dev = connector->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	struct edid *edid;
 	bool has_audio = false;
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	edid = intel_dp_get_edid(connector, &intel_dp->adapter);
 	if (edid) {
 		has_audio = drm_detect_monitor_audio(edid);
 		kfree(edid);
 	}
 
+	intel_display_power_put(dev_priv, power_domain);
+
 	return has_audio;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index a4ffc021c317..9c7090590776 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -609,6 +609,8 @@ hdmi_to_dig_port(struct intel_hdmi *intel_hdmi)
 /* i915_irq.c */
 bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
 					   enum pipe pipe, bool enable);
+bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
+					     enum pipe pipe, bool enable);
 bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
 					   enum transcoder pch_transcoder,
 					   bool enable);
@@ -732,7 +734,9 @@ ironlake_check_encoder_dotclock(const struct intel_crtc_config *pipe_config,
 bool intel_crtc_active(struct drm_crtc *crtc);
 void hsw_enable_ips(struct intel_crtc *crtc);
 void hsw_disable_ips(struct intel_crtc *crtc);
-void intel_display_set_init_power(struct drm_device *dev, bool enable);
+void intel_display_set_init_power(struct drm_i915_private *dev, bool enable);
+enum intel_display_power_domain
+intel_display_port_power_domain(struct intel_encoder *intel_encoder);
 int valleyview_get_vco(struct drm_i915_private *dev_priv);
 void intel_mode_from_pipe_config(struct drm_display_mode *mode,
 				 struct intel_crtc_config *pipe_config);
@@ -871,18 +875,17 @@ bool intel_fbc_enabled(struct drm_device *dev);
 void intel_update_fbc(struct drm_device *dev);
 void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
 void intel_gpu_ips_teardown(void);
-int intel_power_domains_init(struct drm_device *dev);
-void intel_power_domains_remove(struct drm_device *dev);
-bool intel_display_power_enabled(struct drm_device *dev,
+int intel_power_domains_init(struct drm_i915_private *);
+void intel_power_domains_remove(struct drm_i915_private *);
+bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
 				 enum intel_display_power_domain domain);
-bool intel_display_power_enabled_sw(struct drm_device *dev,
+bool intel_display_power_enabled_sw(struct drm_i915_private *dev_priv,
 				    enum intel_display_power_domain domain);
-void intel_display_power_get(struct drm_device *dev,
+void intel_display_power_get(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain);
-void intel_display_power_put(struct drm_device *dev,
+void intel_display_power_put(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain);
-void intel_power_domains_init_hw(struct drm_device *dev);
-void intel_set_power_well(struct drm_device *dev, bool enable);
+void intel_power_domains_init_hw(struct drm_i915_private *dev_priv);
 void intel_enable_gt_powersave(struct drm_device *dev);
 void intel_disable_gt_powersave(struct drm_device *dev);
 void ironlake_teardown_rc6(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 3ee1db1407b0..cf7322e95278 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -243,11 +243,16 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 				   enum pipe *pipe)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	u32 port, func;
 	enum pipe p;
 
 	DRM_DEBUG_KMS("\n");
 
+	power_domain = intel_display_port_power_domain(encoder);
+	if (!intel_display_power_enabled(dev_priv, power_domain))
+		return false;
+
 	/* XXX: this only works for one DSI output */
 	for (p = PIPE_A; p <= PIPE_B; p++) {
 		port = I915_READ(MIPI_PORT_CTRL(p));
@@ -488,8 +493,19 @@ static enum drm_connector_status
 intel_dsi_detect(struct drm_connector *connector, bool force)
 {
 	struct intel_dsi *intel_dsi = intel_attached_dsi(connector);
+	struct intel_encoder *intel_encoder = &intel_dsi->base;
+	enum intel_display_power_domain power_domain;
+	enum drm_connector_status connector_status;
+	struct drm_i915_private *dev_priv = intel_encoder->base.dev->dev_private;
+
 	DRM_DEBUG_KMS("\n");
-	return intel_dsi->dev.dev_ops->detect(&intel_dsi->dev);
+	power_domain = intel_display_port_power_domain(intel_encoder);
+
+	intel_display_power_get(dev_priv, power_domain);
+	connector_status = intel_dsi->dev.dev_ops->detect(&intel_dsi->dev);
+	intel_display_power_put(dev_priv, power_domain);
+
+	return connector_status;
 }
 
 static int intel_dsi_get_modes(struct drm_connector *connector)
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 19be4bfbcc59..6b5beed28d70 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -289,7 +289,27 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 	struct drm_device *dev = fb_helper->dev;
 	int i, j;
 	bool *save_enabled;
-	bool any_enabled = false;
+	bool fallback = true;
+	int num_connectors_enabled = 0;
+	int num_connectors_detected = 0;
+
+	/*
+	 * If the user specified any force options, just bail here
+	 * and use that config.
+	 */
+	for (i = 0; i < fb_helper->connector_count; i++) {
+		struct drm_fb_helper_connector *fb_conn;
+		struct drm_connector *connector;
+
+		fb_conn = fb_helper->connector_info[i];
+		connector = fb_conn->connector;
+
+		if (!enabled[i])
+			continue;
+
+		if (connector->force != DRM_FORCE_UNSPECIFIED)
+			return false;
+	}
 
 	save_enabled = kcalloc(dev->mode_config.num_connector, sizeof(bool),
 			       GFP_KERNEL);
@@ -306,6 +326,10 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 
 		fb_conn = fb_helper->connector_info[i];
 		connector = fb_conn->connector;
+
+		if (connector->status == connector_status_connected)
+			num_connectors_detected++;
+
 		if (!enabled[i]) {
 			DRM_DEBUG_KMS("connector %d not enabled, skipping\n",
 				      connector->base.id);
@@ -320,6 +344,8 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			continue;
 		}
 
+		num_connectors_enabled++;
+
 		new_crtc = intel_fb_helper_crtc(fb_helper, encoder->crtc);
 
 		/*
@@ -329,7 +355,8 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 		 */
 		for (j = 0; j < fb_helper->connector_count; j++) {
 			if (crtcs[j] == new_crtc) {
-				any_enabled = false;
+				DRM_DEBUG_KMS("fallback: cloned configuration\n");
+				fallback = true;
 				goto out;
 			}
 		}
@@ -372,11 +399,25 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			      encoder->crtc->base.id,
 			      modes[i]->name);
 
-		any_enabled = true;
+		fallback = false;
+	}
+
+	/*
+	 * If the BIOS didn't enable everything it could, fall back to have the
+	 * same user experiencing of lighting up as much as possible like the
+	 * fbdev helper library.
+	 */
+	if (num_connectors_enabled != num_connectors_detected &&
+	    num_connectors_enabled < INTEL_INFO(dev)->num_pipes) {
+		DRM_DEBUG_KMS("fallback: Not all outputs enabled\n");
+		DRM_DEBUG_KMS("Enabled: %i, detected: %i\n", num_connectors_enabled,
+			      num_connectors_detected);
+		fallback = true;
 	}
 
 out:
-	if (!any_enabled) {
+	if (fallback) {
+		DRM_DEBUG_KMS("Not using firmware configuration\n");
 		memcpy(enabled, save_enabled, dev->mode_config.num_connector);
 		kfree(save_enabled);
 		return false;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 98d68ab04de4..f410cc03e08a 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -667,8 +667,13 @@ static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
+	enum intel_display_power_domain power_domain;
 	u32 tmp;
 
+	power_domain = intel_display_port_power_domain(encoder);
+	if (!intel_display_power_enabled(dev_priv, power_domain))
+		return false;
+
 	tmp = I915_READ(intel_hdmi->hdmi_reg);
 
 	if (!(tmp & SDVO_ENABLE))
@@ -909,11 +914,15 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	struct intel_encoder *intel_encoder = &intel_dig_port->base;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct edid *edid;
+	enum intel_display_power_domain power_domain;
 	enum drm_connector_status status = connector_status_disconnected;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, drm_get_connector_name(connector));
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	intel_hdmi->has_hdmi_sink = false;
 	intel_hdmi->has_audio = false;
 	intel_hdmi->rgb_quant_range_selectable = false;
@@ -941,31 +950,48 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 		intel_encoder->type = INTEL_OUTPUT_HDMI;
 	}
 
+	intel_display_power_put(dev_priv, power_domain);
+
 	return status;
 }
 
 static int intel_hdmi_get_modes(struct drm_connector *connector)
 {
-	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
+	struct intel_encoder *intel_encoder = intel_attached_encoder(connector);
+	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&intel_encoder->base);
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
+	enum intel_display_power_domain power_domain;
+	int ret;
 
 	/* We should parse the EDID data and find out if it's an HDMI sink so
 	 * we can send audio to it.
 	 */
 
-	return intel_ddc_get_modes(connector,
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
+	ret = intel_ddc_get_modes(connector,
 				   intel_gmbus_get_adapter(dev_priv,
 							   intel_hdmi->ddc_bus));
+
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static bool
 intel_hdmi_detect_audio(struct drm_connector *connector)
 {
-	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
+	struct intel_encoder *intel_encoder = intel_attached_encoder(connector);
+	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&intel_encoder->base);
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	struct edid *edid;
 	bool has_audio = false;
 
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
+
 	edid = drm_get_edid(connector,
 			    intel_gmbus_get_adapter(dev_priv,
 						    intel_hdmi->ddc_bus));
@@ -975,6 +1001,8 @@ intel_hdmi_detect_audio(struct drm_connector *connector)
 		kfree(edid);
 	}
 
+	intel_display_power_put(dev_priv, power_domain);
+
 	return has_audio;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index ac519cb46f22..312961a8472e 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -1076,7 +1076,7 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 	mutex_lock(&dev->struct_mutex);
 
 	if (new_bo->tiling_mode) {
-		DRM_ERROR("buffer used for overlay image can not be tiled\n");
+		DRM_DEBUG_KMS("buffer used for overlay image can not be tiled\n");
 		ret = -EINVAL;
 		goto out_unlock;
 	}
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index a6b877a4a916..fc96e611a8a7 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -294,11 +294,14 @@ static void gen7_enable_fbc(struct drm_crtc *crtc)
 
 	if (IS_IVYBRIDGE(dev)) {
 		/* WaFbcAsynchFlipDisableFbcQueue:ivb */
-		I915_WRITE(ILK_DISPLAY_CHICKEN1, ILK_FBCQ_DIS);
+		I915_WRITE(ILK_DISPLAY_CHICKEN1,
+			   I915_READ(ILK_DISPLAY_CHICKEN1) |
+			   ILK_FBCQ_DIS);
 	} else {
-		/* WaFbcAsynchFlipDisableFbcQueue:hsw */
-		I915_WRITE(HSW_PIPE_SLICE_CHICKEN_1(intel_crtc->pipe),
-			   HSW_BYPASS_FBC_QUEUE);
+		/* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */
+		I915_WRITE(CHICKEN_PIPESL_1(intel_crtc->pipe),
+			   I915_READ(CHICKEN_PIPESL_1(intel_crtc->pipe)) |
+			   HSW_FBCQ_DIS);
 	}
 
 	I915_WRITE(SNB_DPFC_CTL_SA,
@@ -540,7 +543,7 @@ void intel_update_fbc(struct drm_device *dev)
 			DRM_DEBUG_KMS("mode too large for compression, disabling\n");
 		goto out_disable;
 	}
-	if ((INTEL_INFO(dev)->gen < 4 || IS_HASWELL(dev)) &&
+	if ((INTEL_INFO(dev)->gen < 4 || HAS_DDI(dev)) &&
 	    intel_crtc->plane != PLANE_A) {
 		if (set_no_fbc_reason(dev_priv, FBC_BAD_PLANE))
 			DRM_DEBUG_KMS("plane not A, disabling compression\n");
@@ -1131,7 +1134,7 @@ static bool g4x_compute_wm0(struct drm_device *dev,
 		*plane_wm = display->max_wm;
 
 	/* Use the large buffer method to calculate cursor watermark */
-	line_time_us = ((htotal * 1000) / clock);
+	line_time_us = max(htotal * 1000 / clock, 1);
 	line_count = (cursor_latency_ns / line_time_us + 1000) / 1000;
 	entries = line_count * 64 * pixel_size;
 	tlb_miss = cursor->fifo_size*cursor->cacheline_size - hdisplay * 8;
@@ -1207,7 +1210,7 @@ static bool g4x_compute_srwm(struct drm_device *dev,
 	hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 	pixel_size = crtc->fb->bits_per_pixel / 8;
 
-	line_time_us = (htotal * 1000) / clock;
+	line_time_us = max(htotal * 1000 / clock, 1);
 	line_count = (latency_ns / line_time_us + 1000) / 1000;
 	line_size = hdisplay * pixel_size;
 
@@ -1440,7 +1443,7 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 		unsigned long line_time_us;
 		int entries;
 
-		line_time_us = ((htotal * 1000) / clock);
+		line_time_us = max(htotal * 1000 / clock, 1);
 
 		/* Use ns/us then divide to preserve precision */
 		entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) *
@@ -1566,7 +1569,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 		unsigned long line_time_us;
 		int entries;
 
-		line_time_us = (htotal * 1000) / clock;
+		line_time_us = max(htotal * 1000 / clock, 1);
 
 		/* Use ns/us then divide to preserve precision */
 		entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) *
@@ -4661,6 +4664,17 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 		I915_WRITE(GEN6_GT_MODE,
 			   _MASKED_BIT_ENABLE(GEN6_TD_FOUR_ROW_DISPATCH_DISABLE));
 
+	/*
+	 * BSpec recoomends 8x4 when MSAA is used,
+	 * however in practice 16x4 seems fastest.
+	 *
+	 * Note that PS/WM thread counts depend on the WIZ hashing
+	 * disable bit, which we don't touch here, but it's good
+	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
+	 */
+	I915_WRITE(GEN6_GT_MODE,
+		   GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4);
+
 	ilk_init_lp_watermarks(dev);
 
 	I915_WRITE(CACHE_MODE_0,
@@ -4688,9 +4702,9 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 		   GEN6_RCPBUNIT_CLOCK_GATE_DISABLE |
 		   GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
 
-	/* Bspec says we need to always set all mask bits. */
-	I915_WRITE(_3D_CHICKEN3, (0xFFFF << 16) |
-		   _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL);
+	/* WaStripsFansDisableFastClipPerformanceFix:snb */
+	I915_WRITE(_3D_CHICKEN3,
+		   _MASKED_BIT_ENABLE(_3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL));
 
 	/*
 	 * Bspec says:
@@ -4724,11 +4738,6 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 
 	g4x_disable_trickle_feed(dev);
 
-	/* The default value should be 0x200 according to docs, but the two
-	 * platforms I checked have a 0 for this. (Maybe BIOS overrides?) */
-	I915_WRITE(GEN6_GT_MODE, _MASKED_BIT_DISABLE(0xffff));
-	I915_WRITE(GEN6_GT_MODE, _MASKED_BIT_ENABLE(GEN6_GT_MODE_HI));
-
 	cpt_init_clock_gating(dev);
 
 	gen6_check_mch_setup(dev);
@@ -4786,7 +4795,7 @@ static void lpt_suspend_hw(struct drm_device *dev)
 static void gen8_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	enum pipe i;
+	enum pipe pipe;
 
 	I915_WRITE(WM3_LP_ILK, 0);
 	I915_WRITE(WM2_LP_ILK, 0);
@@ -4795,6 +4804,15 @@ static void gen8_init_clock_gating(struct drm_device *dev)
 	/* FIXME(BDW): Check all the w/a, some might only apply to
 	 * pre-production hw. */
 
+	/* WaDisablePartialInstShootdown:bdw */
+	I915_WRITE(GEN8_ROW_CHICKEN,
+		   _MASKED_BIT_ENABLE(PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE));
+
+	/* WaDisableThreadStallDopClockGating:bdw */
+	/* FIXME: Unclear whether we really need this on production bdw. */
+	I915_WRITE(GEN8_ROW_CHICKEN,
+		   _MASKED_BIT_ENABLE(STALL_DOP_GATING_DISABLE));
+
 	/*
 	 * This GEN8_CENTROID_PIXEL_OPT_DIS W/A is only needed for
 	 * pre-production hardware
@@ -4822,10 +4840,10 @@ static void gen8_init_clock_gating(struct drm_device *dev)
 		   I915_READ(CHICKEN_PAR1_1) | DPA_MASK_VBLANK_SRD);
 
 	/* WaPsrDPRSUnmaskVBlankInSRD:bdw */
-	for_each_pipe(i) {
-		I915_WRITE(CHICKEN_PIPESL_1(i),
-			   I915_READ(CHICKEN_PIPESL_1(i) |
-				     DPRS_MASK_VBLANK_SRD));
+	for_each_pipe(pipe) {
+		I915_WRITE(CHICKEN_PIPESL_1(pipe),
+			   I915_READ(CHICKEN_PIPESL_1(pipe)) |
+			   BDW_DPRS_MASK_VBLANK_SRD);
 	}
 
 	/* Use Force Non-Coherent whenever executing a 3D context. This is a
@@ -4841,6 +4859,24 @@ static void gen8_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN7_FF_THREAD_MODE,
 		   I915_READ(GEN7_FF_THREAD_MODE) &
 		   ~(GEN8_FF_DS_REF_CNT_FFME | GEN7_FF_VS_REF_CNT_FFME));
+
+	/*
+	 * BSpec recommends 8x4 when MSAA is used,
+	 * however in practice 16x4 seems fastest.
+	 *
+	 * Note that PS/WM thread counts depend on the WIZ hashing
+	 * disable bit, which we don't touch here, but it's good
+	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
+	 */
+	I915_WRITE(GEN7_GT_MODE,
+		   GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4);
+
+	I915_WRITE(GEN6_RC_SLEEP_PSMI_CONTROL,
+		   _MASKED_BIT_ENABLE(GEN8_RC_SEMA_IDLE_MSG_DISABLE));
+
+	/* WaDisableSDEUnitClockGating:bdw */
+	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
+		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
 }
 
 static void haswell_init_clock_gating(struct drm_device *dev)
@@ -4871,6 +4907,17 @@ static void haswell_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(CACHE_MODE_1,
 		   _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE));
 
+	/*
+	 * BSpec recommends 8x4 when MSAA is used,
+	 * however in practice 16x4 seems fastest.
+	 *
+	 * Note that PS/WM thread counts depend on the WIZ hashing
+	 * disable bit, which we don't touch here, but it's good
+	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
+	 */
+	I915_WRITE(GEN7_GT_MODE,
+		   GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4);
+
 	/* WaSwitchSolVfFArbitrationPriority:hsw */
 	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | HSW_ECOCHK_ARB_PRIO_SOL);
 
@@ -4944,14 +4991,27 @@ static void ivybridge_init_clock_gating(struct drm_device *dev)
 
 	gen7_setup_fixed_func_scheduler(dev_priv);
 
-	/* enable HiZ Raw Stall Optimization */
-	I915_WRITE(CACHE_MODE_0_GEN7,
-		   _MASKED_BIT_DISABLE(HIZ_RAW_STALL_OPT_DISABLE));
+	if (0) { /* causes HiZ corruption on ivb:gt1 */
+		/* enable HiZ Raw Stall Optimization */
+		I915_WRITE(CACHE_MODE_0_GEN7,
+			   _MASKED_BIT_DISABLE(HIZ_RAW_STALL_OPT_DISABLE));
+	}
 
 	/* WaDisable4x2SubspanOptimization:ivb */
 	I915_WRITE(CACHE_MODE_1,
 		   _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE));
 
+	/*
+	 * BSpec recommends 8x4 when MSAA is used,
+	 * however in practice 16x4 seems fastest.
+	 *
+	 * Note that PS/WM thread counts depend on the WIZ hashing
+	 * disable bit, which we don't touch here, but it's good
+	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
+	 */
+	I915_WRITE(GEN7_GT_MODE,
+		   GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4);
+
 	snpcr = I915_READ(GEN6_MBCUNIT_SNPCR);
 	snpcr &= ~GEN6_MBC_SNPCR_MASK;
 	snpcr |= GEN6_MBC_SNPCR_MED;
@@ -5004,9 +5064,6 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 		   _MASKED_BIT_ENABLE(GEN7_MAX_PS_THREAD_DEP |
 				      GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE));
 
-	/* WaDisableL3CacheAging:vlv */
-	I915_WRITE(GEN7_L3CNTLREG1, I915_READ(GEN7_L3CNTLREG1) | GEN7_L3AGDIS);
-
 	/* WaForceL3Serialization:vlv */
 	I915_WRITE(GEN7_L3SQCREG4, I915_READ(GEN7_L3SQCREG4) &
 		   ~L3SQ_URB_READ_CAM_MATCH_DISABLE);
@@ -5167,19 +5224,16 @@ void intel_suspend_hw(struct drm_device *dev)
  * enable it, so check if it's enabled and also check if we've requested it to
  * be enabled.
  */
-static bool hsw_power_well_enabled(struct drm_device *dev,
+static bool hsw_power_well_enabled(struct drm_i915_private *dev_priv,
 				   struct i915_power_well *power_well)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
 	return I915_READ(HSW_PWR_WELL_DRIVER) ==
 		     (HSW_PWR_WELL_ENABLE_REQUEST | HSW_PWR_WELL_STATE_ENABLED);
 }
 
-bool intel_display_power_enabled_sw(struct drm_device *dev,
+bool intel_display_power_enabled_sw(struct drm_i915_private *dev_priv,
 				    enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains;
 
 	power_domains = &dev_priv->power_domains;
@@ -5187,10 +5241,9 @@ bool intel_display_power_enabled_sw(struct drm_device *dev,
 	return power_domains->domain_use_count[domain];
 }
 
-bool intel_display_power_enabled(struct drm_device *dev,
+bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
 				 enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains;
 	struct i915_power_well *power_well;
 	bool is_enabled;
@@ -5205,7 +5258,7 @@ bool intel_display_power_enabled(struct drm_device *dev,
 		if (power_well->always_on)
 			continue;
 
-		if (!power_well->is_enabled(dev, power_well)) {
+		if (!power_well->ops->is_enabled(dev_priv, power_well)) {
 			is_enabled = false;
 			break;
 		}
@@ -5215,6 +5268,12 @@ bool intel_display_power_enabled(struct drm_device *dev,
 	return is_enabled;
 }
 
+/*
+ * Starting with Haswell, we have a "Power Down Well" that can be turned off
+ * when not needed anymore. We have 4 registers that can request the power well
+ * to be enabled, and it will only be disabled if none of the registers is
+ * requesting it to be enabled.
+ */
 static void hsw_power_well_post_enable(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
@@ -5251,10 +5310,17 @@ static void hsw_power_well_post_enable(struct drm_i915_private *dev_priv)
 	}
 }
 
+static void reset_vblank_counter(struct drm_device *dev, enum pipe pipe)
+{
+	assert_spin_locked(&dev->vbl_lock);
+
+	dev->vblank[pipe].last = 0;
+}
+
 static void hsw_power_well_post_disable(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
-	enum pipe p;
+	enum pipe pipe;
 	unsigned long irqflags;
 
 	/*
@@ -5265,16 +5331,15 @@ static void hsw_power_well_post_disable(struct drm_i915_private *dev_priv)
 	 * FIXME: Should we do this in general in drm_vblank_post_modeset?
 	 */
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-	for_each_pipe(p)
-		if (p != PIPE_A)
-			dev->vblank[p].last = 0;
+	for_each_pipe(pipe)
+		if (pipe != PIPE_A)
+			reset_vblank_counter(dev, pipe);
 	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 }
 
-static void hsw_set_power_well(struct drm_device *dev,
+static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 			       struct i915_power_well *power_well, bool enable)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	bool is_enabled, enable_requested;
 	uint32_t tmp;
 
@@ -5308,35 +5373,204 @@ static void hsw_set_power_well(struct drm_device *dev,
 	}
 }
 
-static void __intel_power_well_get(struct drm_device *dev,
+static void hsw_power_well_sync_hw(struct drm_i915_private *dev_priv,
 				   struct i915_power_well *power_well)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	hsw_set_power_well(dev_priv, power_well, power_well->count > 0);
 
-	if (!power_well->count++ && power_well->set) {
-		hsw_disable_package_c8(dev_priv);
-		power_well->set(dev, power_well, true);
-	}
+	/*
+	 * We're taking over the BIOS, so clear any requests made by it since
+	 * the driver is in charge now.
+	 */
+	if (I915_READ(HSW_PWR_WELL_BIOS) & HSW_PWR_WELL_ENABLE_REQUEST)
+		I915_WRITE(HSW_PWR_WELL_BIOS, 0);
 }
 
-static void __intel_power_well_put(struct drm_device *dev,
+static void hsw_power_well_enable(struct drm_i915_private *dev_priv,
+				  struct i915_power_well *power_well)
+{
+	hsw_disable_package_c8(dev_priv);
+	hsw_set_power_well(dev_priv, power_well, true);
+}
+
+static void hsw_power_well_disable(struct drm_i915_private *dev_priv,
 				   struct i915_power_well *power_well)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	hsw_set_power_well(dev_priv, power_well, false);
+	hsw_enable_package_c8(dev_priv);
+}
 
-	WARN_ON(!power_well->count);
+static void i9xx_always_on_power_well_noop(struct drm_i915_private *dev_priv,
+					   struct i915_power_well *power_well)
+{
+}
 
-	if (!--power_well->count && power_well->set &&
-	    i915.disable_power_well) {
-		power_well->set(dev, power_well, false);
-		hsw_enable_package_c8(dev_priv);
+static bool i9xx_always_on_power_well_enabled(struct drm_i915_private *dev_priv,
+					     struct i915_power_well *power_well)
+{
+	return true;
+}
+
+static void vlv_set_power_well(struct drm_i915_private *dev_priv,
+			       struct i915_power_well *power_well, bool enable)
+{
+	enum punit_power_well power_well_id = power_well->data;
+	u32 mask;
+	u32 state;
+	u32 ctrl;
+
+	mask = PUNIT_PWRGT_MASK(power_well_id);
+	state = enable ? PUNIT_PWRGT_PWR_ON(power_well_id) :
+			 PUNIT_PWRGT_PWR_GATE(power_well_id);
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+
+#define COND \
+	((vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS) & mask) == state)
+
+	if (COND)
+		goto out;
+
+	ctrl = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_CTRL);
+	ctrl &= ~mask;
+	ctrl |= state;
+	vlv_punit_write(dev_priv, PUNIT_REG_PWRGT_CTRL, ctrl);
+
+	if (wait_for(COND, 100))
+		DRM_ERROR("timout setting power well state %08x (%08x)\n",
+			  state,
+			  vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_CTRL));
+
+#undef COND
+
+out:
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
+static void vlv_power_well_sync_hw(struct drm_i915_private *dev_priv,
+				   struct i915_power_well *power_well)
+{
+	vlv_set_power_well(dev_priv, power_well, power_well->count > 0);
+}
+
+static void vlv_power_well_enable(struct drm_i915_private *dev_priv,
+				  struct i915_power_well *power_well)
+{
+	vlv_set_power_well(dev_priv, power_well, true);
+}
+
+static void vlv_power_well_disable(struct drm_i915_private *dev_priv,
+				   struct i915_power_well *power_well)
+{
+	vlv_set_power_well(dev_priv, power_well, false);
+}
+
+static bool vlv_power_well_enabled(struct drm_i915_private *dev_priv,
+				   struct i915_power_well *power_well)
+{
+	int power_well_id = power_well->data;
+	bool enabled = false;
+	u32 mask;
+	u32 state;
+	u32 ctrl;
+
+	mask = PUNIT_PWRGT_MASK(power_well_id);
+	ctrl = PUNIT_PWRGT_PWR_ON(power_well_id);
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+
+	state = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS) & mask;
+	/*
+	 * We only ever set the power-on and power-gate states, anything
+	 * else is unexpected.
+	 */
+	WARN_ON(state != PUNIT_PWRGT_PWR_ON(power_well_id) &&
+		state != PUNIT_PWRGT_PWR_GATE(power_well_id));
+	if (state == ctrl)
+		enabled = true;
+
+	/*
+	 * A transient state at this point would mean some unexpected party
+	 * is poking at the power controls too.
+	 */
+	ctrl = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_CTRL) & mask;
+	WARN_ON(ctrl != state);
+
+	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	return enabled;
+}
+
+static void vlv_display_power_well_enable(struct drm_i915_private *dev_priv,
+					  struct i915_power_well *power_well)
+{
+	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DISP2D);
+
+	vlv_set_power_well(dev_priv, power_well, true);
+
+	spin_lock_irq(&dev_priv->irq_lock);
+	valleyview_enable_display_irqs(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
+
+	/*
+	 * During driver initialization we need to defer enabling hotplug
+	 * processing until fbdev is set up.
+	 */
+	if (dev_priv->enable_hotplug_processing)
+		intel_hpd_init(dev_priv->dev);
+
+	i915_redisable_vga_power_on(dev_priv->dev);
+}
+
+static void vlv_display_power_well_disable(struct drm_i915_private *dev_priv,
+					   struct i915_power_well *power_well)
+{
+	struct drm_device *dev = dev_priv->dev;
+	enum pipe pipe;
+
+	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DISP2D);
+
+	spin_lock_irq(&dev_priv->irq_lock);
+	for_each_pipe(pipe)
+		__intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
+
+	valleyview_disable_display_irqs(dev_priv);
+	spin_unlock_irq(&dev_priv->irq_lock);
+
+	spin_lock_irq(&dev->vbl_lock);
+	for_each_pipe(pipe)
+		reset_vblank_counter(dev, pipe);
+	spin_unlock_irq(&dev->vbl_lock);
+
+	vlv_set_power_well(dev_priv, power_well, false);
+}
+
+static void check_power_well_state(struct drm_i915_private *dev_priv,
+				   struct i915_power_well *power_well)
+{
+	bool enabled = power_well->ops->is_enabled(dev_priv, power_well);
+
+	if (power_well->always_on || !i915.disable_power_well) {
+		if (!enabled)
+			goto mismatch;
+
+		return;
 	}
+
+	if (enabled != (power_well->count > 0))
+		goto mismatch;
+
+	return;
+
+mismatch:
+	WARN(1, "state mismatch for '%s' (always_on %d hw state %d use-count %d disable_power_well %d\n",
+		  power_well->name, power_well->always_on, enabled,
+		  power_well->count, i915.disable_power_well);
 }
 
-void intel_display_power_get(struct drm_device *dev,
+void intel_display_power_get(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains;
 	struct i915_power_well *power_well;
 	int i;
@@ -5345,18 +5579,23 @@ void intel_display_power_get(struct drm_device *dev,
 
 	mutex_lock(&power_domains->lock);
 
-	for_each_power_well(i, power_well, BIT(domain), power_domains)
-		__intel_power_well_get(dev, power_well);
+	for_each_power_well(i, power_well, BIT(domain), power_domains) {
+		if (!power_well->count++) {
+			DRM_DEBUG_KMS("enabling %s\n", power_well->name);
+			power_well->ops->enable(dev_priv, power_well);
+		}
+
+		check_power_well_state(dev_priv, power_well);
+	}
 
 	power_domains->domain_use_count[domain]++;
 
 	mutex_unlock(&power_domains->lock);
 }
 
-void intel_display_power_put(struct drm_device *dev,
+void intel_display_power_put(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains;
 	struct i915_power_well *power_well;
 	int i;
@@ -5368,8 +5607,16 @@ void intel_display_power_put(struct drm_device *dev,
 	WARN_ON(!power_domains->domain_use_count[domain]);
 	power_domains->domain_use_count[domain]--;
 
-	for_each_power_well_rev(i, power_well, BIT(domain), power_domains)
-		__intel_power_well_put(dev, power_well);
+	for_each_power_well_rev(i, power_well, BIT(domain), power_domains) {
+		WARN_ON(!power_well->count);
+
+		if (!--power_well->count && i915.disable_power_well) {
+			DRM_DEBUG_KMS("disabling %s\n", power_well->name);
+			power_well->ops->disable(dev_priv, power_well);
+		}
+
+		check_power_well_state(dev_priv, power_well);
+	}
 
 	mutex_unlock(&power_domains->lock);
 }
@@ -5386,7 +5633,7 @@ void i915_request_power_well(void)
 
 	dev_priv = container_of(hsw_pwr, struct drm_i915_private,
 				power_domains);
-	intel_display_power_get(dev_priv->dev, POWER_DOMAIN_AUDIO);
+	intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO);
 }
 EXPORT_SYMBOL_GPL(i915_request_power_well);
 
@@ -5400,29 +5647,99 @@ void i915_release_power_well(void)
 
 	dev_priv = container_of(hsw_pwr, struct drm_i915_private,
 				power_domains);
-	intel_display_power_put(dev_priv->dev, POWER_DOMAIN_AUDIO);
+	intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO);
 }
 EXPORT_SYMBOL_GPL(i915_release_power_well);
 
+#define POWER_DOMAIN_MASK (BIT(POWER_DOMAIN_NUM) - 1)
+
+#define HSW_ALWAYS_ON_POWER_DOMAINS (			\
+	BIT(POWER_DOMAIN_PIPE_A) |			\
+	BIT(POWER_DOMAIN_TRANSCODER_EDP) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_A_2_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_A_4_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_B_2_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_B_4_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_C_2_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_C_4_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_D_2_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_D_4_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_CRT) |			\
+	BIT(POWER_DOMAIN_INIT))
+#define HSW_DISPLAY_POWER_DOMAINS (				\
+	(POWER_DOMAIN_MASK & ~HSW_ALWAYS_ON_POWER_DOMAINS) |	\
+	BIT(POWER_DOMAIN_INIT))
+
+#define BDW_ALWAYS_ON_POWER_DOMAINS (			\
+	HSW_ALWAYS_ON_POWER_DOMAINS |			\
+	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER))
+#define BDW_DISPLAY_POWER_DOMAINS (				\
+	(POWER_DOMAIN_MASK & ~BDW_ALWAYS_ON_POWER_DOMAINS) |	\
+	BIT(POWER_DOMAIN_INIT))
+
+#define VLV_ALWAYS_ON_POWER_DOMAINS	BIT(POWER_DOMAIN_INIT)
+#define VLV_DISPLAY_POWER_DOMAINS	POWER_DOMAIN_MASK
+
+#define VLV_DPIO_CMN_BC_POWER_DOMAINS (		\
+	BIT(POWER_DOMAIN_PORT_DDI_B_2_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_B_4_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_C_2_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_C_4_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_CRT) |		\
+	BIT(POWER_DOMAIN_INIT))
+
+#define VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS (	\
+	BIT(POWER_DOMAIN_PORT_DDI_B_2_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_B_4_LANES) |	\
+	BIT(POWER_DOMAIN_INIT))
+
+#define VLV_DPIO_TX_B_LANES_23_POWER_DOMAINS (	\
+	BIT(POWER_DOMAIN_PORT_DDI_B_4_LANES) |	\
+	BIT(POWER_DOMAIN_INIT))
+
+#define VLV_DPIO_TX_C_LANES_01_POWER_DOMAINS (	\
+	BIT(POWER_DOMAIN_PORT_DDI_C_2_LANES) |	\
+	BIT(POWER_DOMAIN_PORT_DDI_C_4_LANES) |	\
+	BIT(POWER_DOMAIN_INIT))
+
+#define VLV_DPIO_TX_C_LANES_23_POWER_DOMAINS (	\
+	BIT(POWER_DOMAIN_PORT_DDI_C_4_LANES) |	\
+	BIT(POWER_DOMAIN_INIT))
+
+static const struct i915_power_well_ops i9xx_always_on_power_well_ops = {
+	.sync_hw = i9xx_always_on_power_well_noop,
+	.enable = i9xx_always_on_power_well_noop,
+	.disable = i9xx_always_on_power_well_noop,
+	.is_enabled = i9xx_always_on_power_well_enabled,
+};
+
 static struct i915_power_well i9xx_always_on_power_well[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
 		.domains = POWER_DOMAIN_MASK,
+		.ops = &i9xx_always_on_power_well_ops,
 	},
 };
 
+static const struct i915_power_well_ops hsw_power_well_ops = {
+	.sync_hw = hsw_power_well_sync_hw,
+	.enable = hsw_power_well_enable,
+	.disable = hsw_power_well_disable,
+	.is_enabled = hsw_power_well_enabled,
+};
+
 static struct i915_power_well hsw_power_wells[] = {
 	{
 		.name = "always-on",
 		.always_on = 1,
 		.domains = HSW_ALWAYS_ON_POWER_DOMAINS,
+		.ops = &i9xx_always_on_power_well_ops,
 	},
 	{
 		.name = "display",
-		.domains = POWER_DOMAIN_MASK & ~HSW_ALWAYS_ON_POWER_DOMAINS,
-		.is_enabled = hsw_power_well_enabled,
-		.set = hsw_set_power_well,
+		.domains = HSW_DISPLAY_POWER_DOMAINS,
+		.ops = &hsw_power_well_ops,
 	},
 };
 
@@ -5431,12 +5748,83 @@ static struct i915_power_well bdw_power_wells[] = {
 		.name = "always-on",
 		.always_on = 1,
 		.domains = BDW_ALWAYS_ON_POWER_DOMAINS,
+		.ops = &i9xx_always_on_power_well_ops,
 	},
 	{
 		.name = "display",
-		.domains = POWER_DOMAIN_MASK & ~BDW_ALWAYS_ON_POWER_DOMAINS,
-		.is_enabled = hsw_power_well_enabled,
-		.set = hsw_set_power_well,
+		.domains = BDW_DISPLAY_POWER_DOMAINS,
+		.ops = &hsw_power_well_ops,
+	},
+};
+
+static const struct i915_power_well_ops vlv_display_power_well_ops = {
+	.sync_hw = vlv_power_well_sync_hw,
+	.enable = vlv_display_power_well_enable,
+	.disable = vlv_display_power_well_disable,
+	.is_enabled = vlv_power_well_enabled,
+};
+
+static const struct i915_power_well_ops vlv_dpio_power_well_ops = {
+	.sync_hw = vlv_power_well_sync_hw,
+	.enable = vlv_power_well_enable,
+	.disable = vlv_power_well_disable,
+	.is_enabled = vlv_power_well_enabled,
+};
+
+static struct i915_power_well vlv_power_wells[] = {
+	{
+		.name = "always-on",
+		.always_on = 1,
+		.domains = VLV_ALWAYS_ON_POWER_DOMAINS,
+		.ops = &i9xx_always_on_power_well_ops,
+	},
+	{
+		.name = "display",
+		.domains = VLV_DISPLAY_POWER_DOMAINS,
+		.data = PUNIT_POWER_WELL_DISP2D,
+		.ops = &vlv_display_power_well_ops,
+	},
+	{
+		.name = "dpio-common",
+		.domains = VLV_DPIO_CMN_BC_POWER_DOMAINS,
+		.data = PUNIT_POWER_WELL_DPIO_CMN_BC,
+		.ops = &vlv_dpio_power_well_ops,
+	},
+	{
+		.name = "dpio-tx-b-01",
+		.domains = VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_B_LANES_23_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_23_POWER_DOMAINS,
+		.ops = &vlv_dpio_power_well_ops,
+		.data = PUNIT_POWER_WELL_DPIO_TX_B_LANES_01,
+	},
+	{
+		.name = "dpio-tx-b-23",
+		.domains = VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_B_LANES_23_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_23_POWER_DOMAINS,
+		.ops = &vlv_dpio_power_well_ops,
+		.data = PUNIT_POWER_WELL_DPIO_TX_B_LANES_23,
+	},
+	{
+		.name = "dpio-tx-c-01",
+		.domains = VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_B_LANES_23_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_23_POWER_DOMAINS,
+		.ops = &vlv_dpio_power_well_ops,
+		.data = PUNIT_POWER_WELL_DPIO_TX_C_LANES_01,
+	},
+	{
+		.name = "dpio-tx-c-23",
+		.domains = VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_B_LANES_23_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_01_POWER_DOMAINS |
+			   VLV_DPIO_TX_C_LANES_23_POWER_DOMAINS,
+		.ops = &vlv_dpio_power_well_ops,
+		.data = PUNIT_POWER_WELL_DPIO_TX_C_LANES_23,
 	},
 };
 
@@ -5445,9 +5833,8 @@ static struct i915_power_well bdw_power_wells[] = {
 	(power_domains)->power_well_count = ARRAY_SIZE(__power_wells);	\
 })
 
-int intel_power_domains_init(struct drm_device *dev)
+int intel_power_domains_init(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
 
 	mutex_init(&power_domains->lock);
@@ -5456,12 +5843,14 @@ int intel_power_domains_init(struct drm_device *dev)
 	 * The enabling order will be from lower to higher indexed wells,
 	 * the disabling order is reversed.
 	 */
-	if (IS_HASWELL(dev)) {
+	if (IS_HASWELL(dev_priv->dev)) {
 		set_power_wells(power_domains, hsw_power_wells);
 		hsw_pwr = power_domains;
-	} else if (IS_BROADWELL(dev)) {
+	} else if (IS_BROADWELL(dev_priv->dev)) {
 		set_power_wells(power_domains, bdw_power_wells);
 		hsw_pwr = power_domains;
+	} else if (IS_VALLEYVIEW(dev_priv->dev)) {
+		set_power_wells(power_domains, vlv_power_wells);
 	} else {
 		set_power_wells(power_domains, i9xx_always_on_power_well);
 	}
@@ -5469,47 +5858,28 @@ int intel_power_domains_init(struct drm_device *dev)
 	return 0;
 }
 
-void intel_power_domains_remove(struct drm_device *dev)
+void intel_power_domains_remove(struct drm_i915_private *dev_priv)
 {
 	hsw_pwr = NULL;
 }
 
-static void intel_power_domains_resume(struct drm_device *dev)
+static void intel_power_domains_resume(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
 	struct i915_power_well *power_well;
 	int i;
 
 	mutex_lock(&power_domains->lock);
-	for_each_power_well(i, power_well, POWER_DOMAIN_MASK, power_domains) {
-		if (power_well->set)
-			power_well->set(dev, power_well, power_well->count > 0);
-	}
+	for_each_power_well(i, power_well, POWER_DOMAIN_MASK, power_domains)
+		power_well->ops->sync_hw(dev_priv, power_well);
 	mutex_unlock(&power_domains->lock);
 }
 
-/*
- * Starting with Haswell, we have a "Power Down Well" that can be turned off
- * when not needed anymore. We have 4 registers that can request the power well
- * to be enabled, and it will only be disabled if none of the registers is
- * requesting it to be enabled.
- */
-void intel_power_domains_init_hw(struct drm_device *dev)
+void intel_power_domains_init_hw(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
 	/* For now, we need the power well to be always enabled. */
-	intel_display_set_init_power(dev, true);
-	intel_power_domains_resume(dev);
-
-	if (!(IS_HASWELL(dev) || IS_BROADWELL(dev)))
-		return;
-
-	/* We're taking over the BIOS, so clear any requests made by it since
-	 * the driver is in charge now. */
-	if (I915_READ(HSW_PWR_WELL_BIOS) & HSW_PWR_WELL_ENABLE_REQUEST)
-		I915_WRITE(HSW_PWR_WELL_BIOS, 0);
+	intel_display_set_init_power(dev_priv, true);
+	intel_power_domains_resume(dev_priv);
 }
 
 /* Disables PC8 so we can use the GMBUS and DP AUX interrupts. */
@@ -5786,10 +6156,9 @@ void intel_pm_setup(struct drm_device *dev)
 
 	mutex_init(&dev_priv->pc8.lock);
 	dev_priv->pc8.requirements_met = false;
-	dev_priv->pc8.gpu_idle = false;
 	dev_priv->pc8.irqs_disabled = false;
 	dev_priv->pc8.enabled = false;
-	dev_priv->pc8.disable_count = 2; /* requirements_met + gpu_idle */
+	dev_priv->pc8.disable_count = 1; /* requirements_met */
 	INIT_DELAYED_WORK(&dev_priv->pc8.enable_work, hsw_enable_pc8_work);
 	INIT_DELAYED_WORK(&dev_priv->rps.delayed_resume_work,
 			  intel_gen6_powersave_work);
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index b340c7587629..859092130c18 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -571,7 +571,7 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 	 * to use MI_WAIT_FOR_EVENT within the CS. It should already be
 	 * programmed to '1' on all products.
 	 *
-	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv
+	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv,bdw
 	 */
 	if (INTEL_INFO(dev)->gen >= 6)
 		I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(ASYNC_FLIP_PERF_DISABLE));
@@ -1388,6 +1388,8 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	if (IS_I830(ring->dev) || IS_845G(ring->dev))
 		ring->effective_size -= 128;
 
+	i915_cmd_parser_init_ring(ring);
+
 	return 0;
 
 err_unmap:
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 08b91c6ac70a..09af92099c1b 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -164,6 +164,38 @@ struct  intel_ring_buffer {
 		u32 gtt_offset;
 		volatile u32 *cpu_page;
 	} scratch;
+
+	/*
+	 * Tables of commands the command parser needs to know about
+	 * for this ring.
+	 */
+	const struct drm_i915_cmd_table *cmd_tables;
+	int cmd_table_count;
+
+	/*
+	 * Table of registers allowed in commands that read/write registers.
+	 */
+	const u32 *reg_table;
+	int reg_count;
+
+	/*
+	 * Table of registers allowed in commands that read/write registers, but
+	 * only from the DRM master.
+	 */
+	const u32 *master_reg_table;
+	int master_reg_count;
+
+	/*
+	 * Returns the bitmask for the length field of the specified command.
+	 * Return 0 for an unrecognized/invalid command.
+	 *
+	 * If the command parser finds an entry for a command in the ring's
+	 * cmd_tables, it gets the command's length based on the table entry.
+	 * If not, it calls this function to determine the per-ring length field
+	 * encoding for the command (i.e. certain opcode ranges use certain bits
+	 * to encode the command length in the header).
+	 */
+	u32 (*get_cmd_length_mask)(u32 cmd_header);
 };
 
 static inline bool
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index c62841404c82..7861d97600e1 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -40,6 +40,12 @@
 
 #define __raw_posting_read(dev_priv__, reg__) (void)__raw_i915_read32(dev_priv__, reg__)
 
+static void
+assert_device_not_suspended(struct drm_i915_private *dev_priv)
+{
+	WARN(HAS_RUNTIME_PM(dev_priv->dev) && dev_priv->pm.suspended,
+	     "Device suspended\n");
+}
 
 static void __gen6_gt_wait_for_thread_c0(struct drm_i915_private *dev_priv)
 {
@@ -83,14 +89,14 @@ static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv,
 	__gen6_gt_wait_for_thread_c0(dev_priv);
 }
 
-static void __gen6_gt_force_wake_mt_reset(struct drm_i915_private *dev_priv)
+static void __gen7_gt_force_wake_mt_reset(struct drm_i915_private *dev_priv)
 {
 	__raw_i915_write32(dev_priv, FORCEWAKE_MT, _MASKED_BIT_DISABLE(0xffff));
 	/* something from same cacheline, but !FORCEWAKE_MT */
 	__raw_posting_read(dev_priv, ECOBUS);
 }
 
-static void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv,
+static void __gen7_gt_force_wake_mt_get(struct drm_i915_private *dev_priv,
 							int fw_engine)
 {
 	u32 forcewake_ack;
@@ -136,14 +142,16 @@ static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv,
 	gen6_gt_check_fifodbg(dev_priv);
 }
 
-static void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv,
+static void __gen7_gt_force_wake_mt_put(struct drm_i915_private *dev_priv,
 							int fw_engine)
 {
 	__raw_i915_write32(dev_priv, FORCEWAKE_MT,
 			   _MASKED_BIT_DISABLE(FORCEWAKE_KERNEL));
 	/* something from same cacheline, but !FORCEWAKE_MT */
 	__raw_posting_read(dev_priv, ECOBUS);
-	gen6_gt_check_fifodbg(dev_priv);
+
+	if (IS_GEN7(dev_priv->dev))
+		gen6_gt_check_fifodbg(dev_priv);
 }
 
 static int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
@@ -251,16 +259,16 @@ void vlv_force_wake_get(struct drm_i915_private *dev_priv,
 	unsigned long irqflags;
 
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-	if (FORCEWAKE_RENDER & fw_engine) {
-		if (dev_priv->uncore.fw_rendercount++ == 0)
-			dev_priv->uncore.funcs.force_wake_get(dev_priv,
-							FORCEWAKE_RENDER);
-	}
-	if (FORCEWAKE_MEDIA & fw_engine) {
-		if (dev_priv->uncore.fw_mediacount++ == 0)
-			dev_priv->uncore.funcs.force_wake_get(dev_priv,
-							FORCEWAKE_MEDIA);
-	}
+
+	if (fw_engine & FORCEWAKE_RENDER &&
+	    dev_priv->uncore.fw_rendercount++ != 0)
+		fw_engine &= ~FORCEWAKE_RENDER;
+	if (fw_engine & FORCEWAKE_MEDIA &&
+	    dev_priv->uncore.fw_mediacount++ != 0)
+		fw_engine &= ~FORCEWAKE_MEDIA;
+
+	if (fw_engine)
+		dev_priv->uncore.funcs.force_wake_get(dev_priv, fw_engine);
 
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
@@ -272,46 +280,45 @@ void vlv_force_wake_put(struct drm_i915_private *dev_priv,
 
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
 
-	if (FORCEWAKE_RENDER & fw_engine) {
-		WARN_ON(dev_priv->uncore.fw_rendercount == 0);
-		if (--dev_priv->uncore.fw_rendercount == 0)
-			dev_priv->uncore.funcs.force_wake_put(dev_priv,
-							FORCEWAKE_RENDER);
-	}
+	if (fw_engine & FORCEWAKE_RENDER &&
+	    --dev_priv->uncore.fw_rendercount != 0)
+		fw_engine &= ~FORCEWAKE_RENDER;
+	if (fw_engine & FORCEWAKE_MEDIA &&
+	    --dev_priv->uncore.fw_mediacount != 0)
+		fw_engine &= ~FORCEWAKE_MEDIA;
 
-	if (FORCEWAKE_MEDIA & fw_engine) {
-		WARN_ON(dev_priv->uncore.fw_mediacount == 0);
-		if (--dev_priv->uncore.fw_mediacount == 0)
-			dev_priv->uncore.funcs.force_wake_put(dev_priv,
-							FORCEWAKE_MEDIA);
-	}
+	if (fw_engine)
+		dev_priv->uncore.funcs.force_wake_put(dev_priv, fw_engine);
 
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
-static void gen6_force_wake_work(struct work_struct *work)
+static void gen6_force_wake_timer(unsigned long arg)
 {
-	struct drm_i915_private *dev_priv =
-		container_of(work, typeof(*dev_priv), uncore.force_wake_work.work);
+	struct drm_i915_private *dev_priv = (void *)arg;
 	unsigned long irqflags;
 
+	assert_device_not_suspended(dev_priv);
+
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
 	if (--dev_priv->uncore.forcewake_count == 0)
 		dev_priv->uncore.funcs.force_wake_put(dev_priv, FORCEWAKE_ALL);
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+
+	intel_runtime_pm_put(dev_priv);
 }
 
 static void intel_uncore_forcewake_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_VALLEYVIEW(dev))
 		vlv_force_wake_reset(dev_priv);
-	} else if (INTEL_INFO(dev)->gen >= 6) {
+	else if (IS_GEN6(dev) || IS_GEN7(dev))
 		__gen6_gt_force_wake_reset(dev_priv);
-		if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev))
-			__gen6_gt_force_wake_mt_reset(dev_priv);
-	}
+
+	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_GEN8(dev))
+		__gen7_gt_force_wake_mt_reset(dev_priv);
 }
 
 void intel_uncore_early_sanitize(struct drm_device *dev)
@@ -354,7 +361,9 @@ void intel_uncore_sanitize(struct drm_device *dev)
 		mutex_lock(&dev_priv->rps.hw_lock);
 		reg_val = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS);
 
-		if (reg_val & (RENDER_PWRGT | MEDIA_PWRGT | DISP2D_PWRGT))
+		if (reg_val & (PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_RENDER) |
+			       PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_MEDIA) |
+			       PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_DISP2D)))
 			vlv_punit_write(dev_priv, PUNIT_REG_PWRGT_CTRL, 0x0);
 
 		mutex_unlock(&dev_priv->rps.hw_lock);
@@ -393,25 +402,38 @@ void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv, int fw_engine)
 void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv, int fw_engine)
 {
 	unsigned long irqflags;
+	bool delayed = false;
 
 	if (!dev_priv->uncore.funcs.force_wake_put)
 		return;
 
 	/* Redirect to VLV specific routine */
-	if (IS_VALLEYVIEW(dev_priv->dev))
-		return vlv_force_wake_put(dev_priv, fw_engine);
+	if (IS_VALLEYVIEW(dev_priv->dev)) {
+		vlv_force_wake_put(dev_priv, fw_engine);
+		goto out;
+	}
 
 
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
 	if (--dev_priv->uncore.forcewake_count == 0) {
 		dev_priv->uncore.forcewake_count++;
-		mod_delayed_work(dev_priv->wq,
-				 &dev_priv->uncore.force_wake_work,
-				 1);
+		delayed = true;
+		mod_timer_pinned(&dev_priv->uncore.force_wake_timer,
+				 jiffies + 1);
 	}
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 
-	intel_runtime_pm_put(dev_priv);
+out:
+	if (!delayed)
+		intel_runtime_pm_put(dev_priv);
+}
+
+void assert_force_wake_inactive(struct drm_i915_private *dev_priv)
+{
+	if (!dev_priv->uncore.funcs.force_wake_get)
+		return;
+
+	WARN_ON(dev_priv->uncore.forcewake_count > 0);
 }
 
 /* We give fast paths for the really cool registers */
@@ -446,16 +468,10 @@ hsw_unclaimed_reg_check(struct drm_i915_private *dev_priv, u32 reg)
 	}
 }
 
-static void
-assert_device_not_suspended(struct drm_i915_private *dev_priv)
-{
-	WARN(HAS_RUNTIME_PM(dev_priv->dev) && dev_priv->pm.suspended,
-	     "Device suspended\n");
-}
-
 #define REG_READ_HEADER(x) \
 	unsigned long irqflags; \
 	u##x val = 0; \
+	assert_device_not_suspended(dev_priv); \
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags)
 
 #define REG_READ_FOOTER \
@@ -484,17 +500,15 @@ gen5_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 static u##x \
 gen6_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 	REG_READ_HEADER(x); \
-	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
-		if (dev_priv->uncore.forcewake_count == 0) \
-			dev_priv->uncore.funcs.force_wake_get(dev_priv, \
-							FORCEWAKE_ALL); \
-		val = __raw_i915_read##x(dev_priv, reg); \
-		if (dev_priv->uncore.forcewake_count == 0) \
-			dev_priv->uncore.funcs.force_wake_put(dev_priv, \
-							FORCEWAKE_ALL); \
-	} else { \
-		val = __raw_i915_read##x(dev_priv, reg); \
+	if (dev_priv->uncore.forcewake_count == 0 && \
+	    NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
+		dev_priv->uncore.funcs.force_wake_get(dev_priv, \
+						      FORCEWAKE_ALL); \
+		dev_priv->uncore.forcewake_count++; \
+		mod_timer_pinned(&dev_priv->uncore.force_wake_timer, \
+				 jiffies + 1); \
 	} \
+	val = __raw_i915_read##x(dev_priv, reg); \
 	REG_READ_FOOTER; \
 }
 
@@ -502,27 +516,19 @@ gen6_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 static u##x \
 vlv_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 	unsigned fwengine = 0; \
-	unsigned *fwcount; \
 	REG_READ_HEADER(x); \
-	if (FORCEWAKE_VLV_RENDER_RANGE_OFFSET(reg)) {   \
-		fwengine = FORCEWAKE_RENDER;            \
-		fwcount = &dev_priv->uncore.fw_rendercount;    \
-	}                                               \
-	else if (FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(reg)) {       \
-		fwengine = FORCEWAKE_MEDIA;             \
-		fwcount = &dev_priv->uncore.fw_mediacount;     \
+	if (FORCEWAKE_VLV_RENDER_RANGE_OFFSET(reg)) { \
+		if (dev_priv->uncore.fw_rendercount == 0) \
+			fwengine = FORCEWAKE_RENDER; \
+	} else if (FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(reg)) { \
+		if (dev_priv->uncore.fw_mediacount == 0) \
+			fwengine = FORCEWAKE_MEDIA; \
 	}  \
-	if (fwengine != 0) {		\
-		if ((*fwcount)++ == 0) \
-			(dev_priv)->uncore.funcs.force_wake_get(dev_priv, \
-								fwengine); \
-		val = __raw_i915_read##x(dev_priv, reg); \
-		if (--(*fwcount) == 0) \
-			(dev_priv)->uncore.funcs.force_wake_put(dev_priv, \
-							fwengine); \
-	} else { \
-		val = __raw_i915_read##x(dev_priv, reg); \
-	} \
+	if (fwengine) \
+		dev_priv->uncore.funcs.force_wake_get(dev_priv, fwengine); \
+	val = __raw_i915_read##x(dev_priv, reg); \
+	if (fwengine) \
+		dev_priv->uncore.funcs.force_wake_put(dev_priv, fwengine); \
 	REG_READ_FOOTER; \
 }
 
@@ -554,6 +560,7 @@ __gen4_read(64)
 #define REG_WRITE_HEADER \
 	unsigned long irqflags; \
 	trace_i915_reg_rw(true, reg, val, sizeof(val), trace); \
+	assert_device_not_suspended(dev_priv); \
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags)
 
 #define REG_WRITE_FOOTER \
@@ -584,7 +591,6 @@ gen6_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace
 	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
 		__fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \
 	} \
-	assert_device_not_suspended(dev_priv); \
 	__raw_i915_write##x(dev_priv, reg, val); \
 	if (unlikely(__fifo_ret)) { \
 		gen6_gt_check_fifodbg(dev_priv); \
@@ -600,7 +606,6 @@ hsw_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace)
 	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
 		__fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \
 	} \
-	assert_device_not_suspended(dev_priv); \
 	hsw_unclaimed_reg_clear(dev_priv, reg); \
 	__raw_i915_write##x(dev_priv, reg, val); \
 	if (unlikely(__fifo_ret)) { \
@@ -634,16 +639,17 @@ static bool is_gen8_shadowed(struct drm_i915_private *dev_priv, u32 reg)
 #define __gen8_write(x) \
 static void \
 gen8_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
-	bool __needs_put = reg < 0x40000 && !is_gen8_shadowed(dev_priv, reg); \
 	REG_WRITE_HEADER; \
-	if (__needs_put) { \
-		dev_priv->uncore.funcs.force_wake_get(dev_priv, \
-							FORCEWAKE_ALL); \
-	} \
-	__raw_i915_write##x(dev_priv, reg, val); \
-	if (__needs_put) { \
-		dev_priv->uncore.funcs.force_wake_put(dev_priv, \
-							FORCEWAKE_ALL); \
+	if (reg < 0x40000 && !is_gen8_shadowed(dev_priv, reg)) { \
+		if (dev_priv->uncore.forcewake_count == 0) \
+			dev_priv->uncore.funcs.force_wake_get(dev_priv,	\
+							      FORCEWAKE_ALL); \
+		__raw_i915_write##x(dev_priv, reg, val); \
+		if (dev_priv->uncore.forcewake_count == 0) \
+			dev_priv->uncore.funcs.force_wake_put(dev_priv, \
+							      FORCEWAKE_ALL); \
+	} else { \
+		__raw_i915_write##x(dev_priv, reg, val); \
 	} \
 	REG_WRITE_FOOTER; \
 }
@@ -681,15 +687,15 @@ void intel_uncore_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	INIT_DELAYED_WORK(&dev_priv->uncore.force_wake_work,
-			  gen6_force_wake_work);
+	setup_timer(&dev_priv->uncore.force_wake_timer,
+		    gen6_force_wake_timer, (unsigned long)dev_priv);
 
 	if (IS_VALLEYVIEW(dev)) {
 		dev_priv->uncore.funcs.force_wake_get = __vlv_force_wake_get;
 		dev_priv->uncore.funcs.force_wake_put = __vlv_force_wake_put;
 	} else if (IS_HASWELL(dev) || IS_GEN8(dev)) {
-		dev_priv->uncore.funcs.force_wake_get = __gen6_gt_force_wake_mt_get;
-		dev_priv->uncore.funcs.force_wake_put = __gen6_gt_force_wake_mt_put;
+		dev_priv->uncore.funcs.force_wake_get = __gen7_gt_force_wake_mt_get;
+		dev_priv->uncore.funcs.force_wake_put = __gen7_gt_force_wake_mt_put;
 	} else if (IS_IVYBRIDGE(dev)) {
 		u32 ecobus;
 
@@ -703,16 +709,16 @@ void intel_uncore_init(struct drm_device *dev)
 		 * forcewake being disabled.
 		 */
 		mutex_lock(&dev->struct_mutex);
-		__gen6_gt_force_wake_mt_get(dev_priv, FORCEWAKE_ALL);
+		__gen7_gt_force_wake_mt_get(dev_priv, FORCEWAKE_ALL);
 		ecobus = __raw_i915_read32(dev_priv, ECOBUS);
-		__gen6_gt_force_wake_mt_put(dev_priv, FORCEWAKE_ALL);
+		__gen7_gt_force_wake_mt_put(dev_priv, FORCEWAKE_ALL);
 		mutex_unlock(&dev->struct_mutex);
 
 		if (ecobus & FORCEWAKE_MT_ENABLE) {
 			dev_priv->uncore.funcs.force_wake_get =
-				__gen6_gt_force_wake_mt_get;
+				__gen7_gt_force_wake_mt_get;
 			dev_priv->uncore.funcs.force_wake_put =
-				__gen6_gt_force_wake_mt_put;
+				__gen7_gt_force_wake_mt_put;
 		} else {
 			DRM_INFO("No MT forcewake available on Ivybridge, this can result in issues\n");
 			DRM_INFO("when using vblank-synced partial screen updates.\n");
@@ -794,10 +800,11 @@ void intel_uncore_fini(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	flush_delayed_work(&dev_priv->uncore.force_wake_work);
+	del_timer_sync(&dev_priv->uncore.force_wake_timer);
 
 	/* Paranoia: make sure we have disabled everything before we exit. */
 	intel_uncore_sanitize(dev);
+	intel_uncore_forcewake_reset(dev);
 }
 
 static const struct register_whitelist {
@@ -947,6 +954,7 @@ static int gen6_do_reset(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int	ret;
 	unsigned long irqflags;
+	u32 fw_engine = 0;
 
 	/* Hold uncore.lock across reset to prevent any register access
 	 * with forcewake not set correctly
@@ -966,14 +974,25 @@ static int gen6_do_reset(struct drm_device *dev)
 
 	intel_uncore_forcewake_reset(dev);
 
-	/* If reset with a user forcewake, try to restore, otherwise turn it off */
-	if (dev_priv->uncore.forcewake_count)
-		dev_priv->uncore.funcs.force_wake_get(dev_priv, FORCEWAKE_ALL);
-	else
-		dev_priv->uncore.funcs.force_wake_put(dev_priv, FORCEWAKE_ALL);
+	/* If reset with a user forcewake, try to restore */
+	if (IS_VALLEYVIEW(dev)) {
+		if (dev_priv->uncore.fw_rendercount)
+			fw_engine |= FORCEWAKE_RENDER;
+
+		if (dev_priv->uncore.fw_mediacount)
+			fw_engine |= FORCEWAKE_MEDIA;
+	} else {
+		if (dev_priv->uncore.forcewake_count)
+			fw_engine = FORCEWAKE_ALL;
+	}
 
-	/* Restore fifo count */
-	dev_priv->uncore.fifo_count = __raw_i915_read32(dev_priv, GTFIFOCTL) & GT_FIFO_FREE_ENTRIES_MASK;
+	if (fw_engine)
+		dev_priv->uncore.funcs.force_wake_get(dev_priv, fw_engine);
+
+	if (IS_GEN6(dev) || IS_GEN7(dev))
+		dev_priv->uncore.fifo_count =
+			__raw_i915_read32(dev_priv, GTFIFOCTL) &
+			GT_FIFO_FREE_ENTRIES_MASK;
 
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 	return ret;