Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "This is the main drm pull request for v4.2.

  I've one other new driver from freescale on my radar, it's been posted
  and reviewed, I'd just like to get someone to give it a last look, so
  maybe I'll send it or maybe I'll leave it.

  There is no major nouveau changes in here, Ben was working on
  something big, and we agreed it was a bit late, there wasn't anything
  else he considered urgent to merge.

  There might be another msm pull for some bits that are waiting on
  arm-soc, I'll see how we time it.

  This touches some "of" stuff, acks are in place except for the fixes
  to the build in various configs,t hat I just applied.

  Summary:

  New drivers:
      - virtio-gpu:
                KMS only pieces of driver for virtio-gpu in qemu.
                This is just the first part of this driver, enough to run
                unaccelerated userspace on. As qemu merges more we'll start
                adding the 3D features for the virgl 3d work.
      - amdgpu:
                a new driver from AMD to driver their newer GPUs. (VI+)
                It contains a new cleaner userspace API, and is a clean
                break from radeon moving forward, that AMD are going to
                concentrate on. It also contains a set of register headers
                auto generated from AMD internal database.

  core:
      - atomic modesetting API completed, enabled by default now.
      - Add support for mode_id blob to atomic ioctl to complete interface.
      - bunch of Displayport MST fixes
      - lots of misc fixes.

  panel:
      - new simple panels
      - fix some long-standing build issues with bridge drivers

  radeon:
      - VCE1 support
      - add a GPU reset counter for userspace
      - lots of fixes.

  amdkfd:
      - H/W debugger support module
      - static user-mode queues
      - support killing all the waves when a process terminates
      - use standard DECLARE_BITMAP

  i915:
      - Add Broxton support
      - S3, rotation support for Skylake
      - RPS booting tuning
      - CPT modeset sequence fixes
      - ns2501 dither support
      - enable cmd parser on haswell
      - cdclk handling fixes
      - gen8 dynamic pte allocation
      - lots of atomic conversion work

  exynos:
      - Add atomic modesetting support
      - Add iommu support
      - Consolidate drm driver initialization
      - and MIC, DECON and MIPI-DSI support for exynos5433

  omapdrm:
      - atomic modesetting support (fixes lots of things in rewrite)

  tegra:
      - DP aux transaction fixes
      - iommu support fix

  msm:
      - adreno a306 support
      - various dsi bits
      - various 64-bit fixes
      - NV12MT support

  rcar-du:
      - atomic and misc fixes

  sti:
      - fix HDMI timing complaince

  tilcdc:
      - use drm component API to access tda998x driver
      - fix module unloading

  qxl:
      - stability fixes"

* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (872 commits)
  drm/nouveau: Pause between setting gpu to D3hot and cutting the power
  drm/dp/mst: close deadlock in connector destruction.
  drm: Always enable atomic API
  drm/vgem: Set unique to "vgem"
  of: fix a build error to of_graph_get_endpoint_by_regs function
  drm/dp/mst: take lock around looking up the branch device on hpd irq
  drm/dp/mst: make sure mst_primary mstb is valid in work function
  of: add EXPORT_SYMBOL for of_graph_get_endpoint_by_regs
  ARM: dts: rename the clock of MIPI DSI 'pll_clk' to 'sclk_mipi'
  drm/atomic: Don't set crtc_state->enable manually
  drm/exynos: dsi: do not set TE GPIO direction by input
  drm/exynos: dsi: add support for MIC driver as a bridge
  drm/exynos: dsi: add support for Exynos5433
  drm/exynos: dsi: make use of array for clock access
  drm/exynos: dsi: make use of driver data for static values
  drm/exynos: dsi: add macros for register access
  drm/exynos: dsi: rename pll_clk to sclk_clk
  drm/exynos: mic: add MIC driver
  of: add helper for getting endpoint node of specific identifiers
  drm/exynos: add Exynos5433 decon driver
  ...

60 files changed, 9980 insertions, 4888 deletions

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index a69002e2257d..b7ddf48e1d75 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -2,8 +2,6 @@
 # Makefile for the drm device driver.  This driver provides support for the
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
-ccflags-y := -Iinclude/drm
-
 # Please keep these build lists sorted!
 
 # core driver code
@@ -12,7 +10,8 @@ i915-y := i915_drv.o \
           i915_suspend.o \
 	  i915_sysfs.o \
 	  intel_pm.o \
-	  intel_runtime_pm.o
+	  intel_runtime_pm.o \
+	  intel_csr.o
 
 i915-$(CONFIG_COMPAT)   += i915_ioc32.o
 i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o
diff --git a/drivers/gpu/drm/i915/dvo_ivch.c b/drivers/gpu/drm/i915/dvo_ivch.c
index 0f2587ff347c..89b08a896d20 100644
--- a/drivers/gpu/drm/i915/dvo_ivch.c
+++ b/drivers/gpu/drm/i915/dvo_ivch.c
@@ -23,6 +23,9 @@
  * Authors:
  *    Eric Anholt <eric@anholt.net>
  *
+ * Minor modifications (Dithering enable):
+ *    Thomas Richter <thor@math.tu-berlin.de>
+ *
  */
 
 #include "dvo.h"
@@ -59,6 +62,8 @@
 # define VR01_DVO_BYPASS_ENABLE		(1 << 1)
 /** Enables the DVO clock */
 # define VR01_DVO_ENABLE		(1 << 0)
+/** Enable dithering for 18bpp panels. Not documented. */
+# define VR01_DITHER_ENABLE             (1 << 4)
 
 /*
  * LCD Interface Format
@@ -74,6 +79,8 @@
 # define VR10_INTERFACE_2X18		(2 << 2)
 /** Enables 2x24-bit LVDS output */
 # define VR10_INTERFACE_2X24		(3 << 2)
+/** Mask that defines the depth of the pipeline */
+# define VR10_INTERFACE_DEPTH_MASK      (3 << 2)
 
 /*
  * VR20 LCD Horizontal Display Size
@@ -342,9 +349,15 @@ static void ivch_mode_set(struct intel_dvo_device *dvo,
 			  struct drm_display_mode *adjusted_mode)
 {
 	uint16_t vr40 = 0;
-	uint16_t vr01;
+	uint16_t vr01 = 0;
+	uint16_t vr10;
+
+	ivch_read(dvo, VR10, &vr10);
+	/* Enable dithering for 18 bpp pipelines */
+	vr10 &= VR10_INTERFACE_DEPTH_MASK;
+	if (vr10 == VR10_INTERFACE_2X18 || vr10 == VR10_INTERFACE_1X18)
+		vr01 = VR01_DITHER_ENABLE;
 
-	vr01 = 0;
 	vr40 = (VR40_STALL_ENABLE | VR40_VERTICAL_INTERP_ENABLE |
 		VR40_HORIZONTAL_INTERP_ENABLE);
 
@@ -353,7 +366,7 @@ static void ivch_mode_set(struct intel_dvo_device *dvo,
 		uint16_t x_ratio, y_ratio;
 
 		vr01 |= VR01_PANEL_FIT_ENABLE;
-		vr40 |= VR40_CLOCK_GATING_ENABLE;
+		vr40 |= VR40_CLOCK_GATING_ENABLE | VR40_ENHANCED_PANEL_FITTING;
 		x_ratio = (((mode->hdisplay - 1) << 16) /
 			   (adjusted_mode->hdisplay - 1)) >> 2;
 		y_ratio = (((mode->vdisplay - 1) << 16) /
@@ -380,6 +393,8 @@ static void ivch_dump_regs(struct intel_dvo_device *dvo)
 	DRM_DEBUG_KMS("VR00: 0x%04x\n", val);
 	ivch_read(dvo, VR01, &val);
 	DRM_DEBUG_KMS("VR01: 0x%04x\n", val);
+	ivch_read(dvo, VR10, &val);
+	DRM_DEBUG_KMS("VR10: 0x%04x\n", val);
 	ivch_read(dvo, VR30, &val);
 	DRM_DEBUG_KMS("VR30: 0x%04x\n", val);
 	ivch_read(dvo, VR40, &val);
diff --git a/drivers/gpu/drm/i915/dvo_ns2501.c b/drivers/gpu/drm/i915/dvo_ns2501.c
index 441630434d34..97ae8aa157e9 100644
--- a/drivers/gpu/drm/i915/dvo_ns2501.c
+++ b/drivers/gpu/drm/i915/dvo_ns2501.c
@@ -60,6 +60,130 @@
 
 #define NS2501_REGC 0x0c
 
+/*
+ * The following registers are not part of the official datasheet
+ * and are the result of reverse engineering.
+ */
+
+/*
+ * Register c0 controls how the DVO synchronizes with
+ * its input.
+ */
+#define NS2501_REGC0 0xc0
+#define NS2501_C0_ENABLE (1<<0)	/* enable the DVO sync in general */
+#define NS2501_C0_HSYNC (1<<1)	/* synchronize horizontal with input */
+#define NS2501_C0_VSYNC (1<<2)	/* synchronize vertical with input */
+#define NS2501_C0_RESET (1<<7)	/* reset the synchronization flip/flops */
+
+/*
+ * Register 41 is somehow related to the sync register and sync
+ * configuration. It should be 0x32 whenever regC0 is 0x05 (hsync off)
+ * and 0x00 otherwise.
+ */
+#define NS2501_REG41 0x41
+
+/*
+ * this register controls the dithering of the DVO
+ * One bit enables it, the other define the dithering depth.
+ * The higher the value, the lower the dithering depth.
+ */
+#define NS2501_F9_REG 0xf9
+#define NS2501_F9_ENABLE (1<<0)		/* if set, dithering is enabled */
+#define NS2501_F9_DITHER_MASK (0x7f<<1)	/* controls the dither depth */
+#define NS2501_F9_DITHER_SHIFT 1	/* shifts the dither mask */
+
+/*
+ * PLL configuration register. This is a pair of registers,
+ * one single byte register at 1B, and a pair at 1C,1D.
+ * These registers are counters/dividers.
+ */
+#define NS2501_REG1B 0x1b /* one byte PLL control register */
+#define NS2501_REG1C 0x1c /* low-part of the second register */
+#define NS2501_REG1D 0x1d /* high-part of the second register */
+
+/*
+ * Scaler control registers. Horizontal at b8,b9,
+ * vertical at 10,11. The scale factor is computed as
+ * 2^16/control-value. The low-byte comes first.
+ */
+#define NS2501_REG10 0x10 /* low-byte vertical scaler */
+#define NS2501_REG11 0x11 /* high-byte vertical scaler */
+#define NS2501_REGB8 0xb8 /* low-byte horizontal scaler */
+#define NS2501_REGB9 0xb9 /* high-byte horizontal scaler */
+
+/*
+ * Display window definition. This consists of four registers
+ * per dimension. One register pair defines the start of the
+ * display, one the end.
+ * As far as I understand, this defines the window within which
+ * the scaler samples the input.
+ */
+#define NS2501_REGC1 0xc1 /* low-byte horizontal display start */
+#define NS2501_REGC2 0xc2 /* high-byte horizontal display start */
+#define NS2501_REGC3 0xc3 /* low-byte horizontal display stop */
+#define NS2501_REGC4 0xc4 /* high-byte horizontal display stop */
+#define NS2501_REGC5 0xc5 /* low-byte vertical display start */
+#define NS2501_REGC6 0xc6 /* high-byte vertical display start */
+#define NS2501_REGC7 0xc7 /* low-byte vertical display stop */
+#define NS2501_REGC8 0xc8 /* high-byte vertical display stop */
+
+/*
+ * The following register pair seems to define the start of
+ * the vertical sync. If automatic syncing is enabled, and the
+ * register value defines a sync pulse that is later than the
+ * incoming sync, then the register value is ignored and the
+ * external hsync triggers the synchronization.
+ */
+#define NS2501_REG80 0x80 /* low-byte vsync-start */
+#define NS2501_REG81 0x81 /* high-byte vsync-start */
+
+/*
+ * The following register pair seems to define the total number
+ * of lines created at the output side of the scaler.
+ * This is again a low-high register pair.
+ */
+#define NS2501_REG82 0x82 /* output display height, low byte */
+#define NS2501_REG83 0x83 /* output display height, high byte */
+
+/*
+ * The following registers define the end of the front-porch
+ * in horizontal and vertical position and hence allow to shift
+ * the image left/right or up/down.
+ */
+#define NS2501_REG98 0x98 /* horizontal start of display + 256, low */
+#define NS2501_REG99 0x99 /* horizontal start of display + 256, high */
+#define NS2501_REG8E 0x8e /* vertical start of the display, low byte */
+#define NS2501_REG8F 0x8f /* vertical start of the display, high byte */
+
+/*
+ * The following register pair control the function of the
+ * backlight and the DVO output. To enable the corresponding
+ * function, the corresponding bit must be set in both registers.
+ */
+#define NS2501_REG34 0x34 /* DVO enable functions, first register */
+#define NS2501_REG35 0x35 /* DVO enable functions, second register */
+#define NS2501_34_ENABLE_OUTPUT (1<<0) /* enable DVO output */
+#define NS2501_34_ENABLE_BACKLIGHT (1<<1) /* enable backlight */
+
+/*
+ * Registers 9C and 9D define the vertical output offset
+ * of the visible region.
+ */
+#define NS2501_REG9C 0x9c
+#define NS2501_REG9D 0x9d
+
+/*
+ * The register 9F defines the dithering. This requires the
+ * scaler to be ON. Bit 0 enables dithering, the remaining
+ * bits control the depth of the dither. The higher the value,
+ * the LOWER the dithering amplitude. A good value seems to be
+ * 15 (total register value).
+ */
+#define NS2501_REGF9 0xf9
+#define NS2501_F9_ENABLE_DITHER (1<<0) /* enable dithering */
+#define NS2501_F9_DITHER_MASK (0x7f<<1) /* dither masking */
+#define NS2501_F9_DITHER_SHIFT 1	/* upshift of the dither mask */
+
 enum {
 	MODE_640x480,
 	MODE_800x600,
@@ -72,274 +196,178 @@ struct ns2501_reg {
 };
 
 /*
- * Magic values based on what the BIOS on
- * Fujitsu-Siemens Lifebook S6010 programs (1024x768 panel).
+ * The following structure keeps the complete configuration of
+ * the DVO, given a specific output configuration.
+ * This is pretty much guess-work from reverse-engineering, so
+ * read all this with a grain of salt.
+ */
+struct ns2501_configuration {
+	uint8_t sync;		/* configuration of the C0 register */
+	uint8_t conf;		/* configuration register 8 */
+	uint8_t syncb;		/* configuration register 41 */
+	uint8_t	dither;		/* configuration of the dithering */
+	uint8_t pll_a;		/* PLL configuration, register A, 1B */
+	uint16_t pll_b;		/* PLL configuration, register B, 1C/1D */
+	uint16_t hstart;	/* horizontal start, registers C1/C2 */
+	uint16_t hstop;		/* horizontal total, registers C3/C4 */
+	uint16_t vstart;	/* vertical start, registers C5/C6 */
+	uint16_t vstop;		/* vertical total, registers C7/C8 */
+	uint16_t vsync;         /* manual vertical sync start, 80/81 */
+	uint16_t vtotal;        /* number of lines generated, 82/83 */
+	uint16_t hpos;		/* horizontal position + 256, 98/99  */
+	uint16_t vpos;		/* vertical position, 8e/8f */
+	uint16_t voffs;		/* vertical output offset, 9c/9d */
+	uint16_t hscale;	/* horizontal scaling factor, b8/b9 */
+	uint16_t vscale;	/* vertical scaling factor, 10/11 */
+};
+
+/*
+ * DVO configuration values, partially based on what the BIOS
+ * of the Fujitsu Lifebook S6010 writes into registers,
+ * partially found by manual tweaking. These configurations assume
+ * a 1024x768 panel.
  */
-static const struct ns2501_reg regs_1024x768[][86] = {
+static const struct ns2501_configuration ns2501_modes[] = {
 	[MODE_640x480] = {
-		[0] = { .offset = 0x0a, .value = 0x81, },
-		[1] = { .offset = 0x18, .value = 0x07, },
-		[2] = { .offset = 0x19, .value = 0x00, },
-		[3] = { .offset = 0x1a, .value = 0x00, },
-		[4] = { .offset = 0x1b, .value = 0x11, },
-		[5] = { .offset = 0x1c, .value = 0x54, },
-		[6] = { .offset = 0x1d, .value = 0x03, },
-		[7] = { .offset = 0x1e, .value = 0x02, },
-		[8] = { .offset = 0xf3, .value = 0x90, },
-		[9] = { .offset = 0xf9, .value = 0x00, },
-		[10] = { .offset = 0xc1, .value = 0x90, },
-		[11] = { .offset = 0xc2, .value = 0x00, },
-		[12] = { .offset = 0xc3, .value = 0x0f, },
-		[13] = { .offset = 0xc4, .value = 0x03, },
-		[14] = { .offset = 0xc5, .value = 0x16, },
-		[15] = { .offset = 0xc6, .value = 0x00, },
-		[16] = { .offset = 0xc7, .value = 0x02, },
-		[17] = { .offset = 0xc8, .value = 0x02, },
-		[18] = { .offset = 0xf4, .value = 0x00, },
-		[19] = { .offset = 0x80, .value = 0xff, },
-		[20] = { .offset = 0x81, .value = 0x07, },
-		[21] = { .offset = 0x82, .value = 0x3d, },
-		[22] = { .offset = 0x83, .value = 0x05, },
-		[23] = { .offset = 0x94, .value = 0x00, },
-		[24] = { .offset = 0x95, .value = 0x00, },
-		[25] = { .offset = 0x96, .value = 0x05, },
-		[26] = { .offset = 0x97, .value = 0x00, },
-		[27] = { .offset = 0x9a, .value = 0x88, },
-		[28] = { .offset = 0x9b, .value = 0x00, },
-		[29] = { .offset = 0x98, .value = 0x00, },
-		[30] = { .offset = 0x99, .value = 0x00, },
-		[31] = { .offset = 0xf7, .value = 0x88, },
-		[32] = { .offset = 0xf8, .value = 0x0a, },
-		[33] = { .offset = 0x9c, .value = 0x24, },
-		[34] = { .offset = 0x9d, .value = 0x00, },
-		[35] = { .offset = 0x9e, .value = 0x25, },
-		[36] = { .offset = 0x9f, .value = 0x03, },
-		[37] = { .offset = 0xa0, .value = 0x28, },
-		[38] = { .offset = 0xa1, .value = 0x01, },
-		[39] = { .offset = 0xa2, .value = 0x28, },
-		[40] = { .offset = 0xa3, .value = 0x05, },
-		[41] = { .offset = 0xb6, .value = 0x09, },
-		[42] = { .offset = 0xb8, .value = 0x00, },
-		[43] = { .offset = 0xb9, .value = 0xa0, },
-		[44] = { .offset = 0xba, .value = 0x00, },
-		[45] = { .offset = 0xbb, .value = 0x20, },
-		[46] = { .offset = 0x10, .value = 0x00, },
-		[47] = { .offset = 0x11, .value = 0xa0, },
-		[48] = { .offset = 0x12, .value = 0x02, },
-		[49] = { .offset = 0x20, .value = 0x00, },
-		[50] = { .offset = 0x22, .value = 0x00, },
-		[51] = { .offset = 0x23, .value = 0x00, },
-		[52] = { .offset = 0x24, .value = 0x00, },
-		[53] = { .offset = 0x25, .value = 0x00, },
-		[54] = { .offset = 0x8c, .value = 0x10, },
-		[55] = { .offset = 0x8d, .value = 0x02, },
-		[56] = { .offset = 0x8e, .value = 0x10, },
-		[57] = { .offset = 0x8f, .value = 0x00, },
-		[58] = { .offset = 0x90, .value = 0xff, },
-		[59] = { .offset = 0x91, .value = 0x07, },
-		[60] = { .offset = 0x92, .value = 0xa0, },
-		[61] = { .offset = 0x93, .value = 0x02, },
-		[62] = { .offset = 0xa5, .value = 0x00, },
-		[63] = { .offset = 0xa6, .value = 0x00, },
-		[64] = { .offset = 0xa7, .value = 0x00, },
-		[65] = { .offset = 0xa8, .value = 0x00, },
-		[66] = { .offset = 0xa9, .value = 0x04, },
-		[67] = { .offset = 0xaa, .value = 0x70, },
-		[68] = { .offset = 0xab, .value = 0x4f, },
-		[69] = { .offset = 0xac, .value = 0x00, },
-		[70] = { .offset = 0xa4, .value = 0x84, },
-		[71] = { .offset = 0x7e, .value = 0x18, },
-		[72] = { .offset = 0x84, .value = 0x00, },
-		[73] = { .offset = 0x85, .value = 0x00, },
-		[74] = { .offset = 0x86, .value = 0x00, },
-		[75] = { .offset = 0x87, .value = 0x00, },
-		[76] = { .offset = 0x88, .value = 0x00, },
-		[77] = { .offset = 0x89, .value = 0x00, },
-		[78] = { .offset = 0x8a, .value = 0x00, },
-		[79] = { .offset = 0x8b, .value = 0x00, },
-		[80] = { .offset = 0x26, .value = 0x00, },
-		[81] = { .offset = 0x27, .value = 0x00, },
-		[82] = { .offset = 0xad, .value = 0x00, },
-		[83] = { .offset = 0x08, .value = 0x30, }, /* 0x31 */
-		[84] = { .offset = 0x41, .value = 0x00, },
-		[85] = { .offset = 0xc0, .value = 0x05, },
+		.sync	= NS2501_C0_ENABLE | NS2501_C0_VSYNC,
+		.conf	= NS2501_8_VEN | NS2501_8_HEN | NS2501_8_PD,
+		.syncb	= 0x32,
+		.dither	= 0x0f,
+		.pll_a	= 17,
+		.pll_b	= 852,
+		.hstart	= 144,
+		.hstop	= 783,
+		.vstart	= 22,
+		.vstop	= 514,
+		.vsync	= 2047, /* actually, ignored with this config */
+		.vtotal	= 1341,
+		.hpos	= 0,
+		.vpos	= 16,
+		.voffs	= 36,
+		.hscale	= 40960,
+		.vscale	= 40960
 	},
 	[MODE_800x600] = {
-		[0] = { .offset = 0x0a, .value = 0x81, },
-		[1] = { .offset = 0x18, .value = 0x07, },
-		[2] = { .offset = 0x19, .value = 0x00, },
-		[3] = { .offset = 0x1a, .value = 0x00, },
-		[4] = { .offset = 0x1b, .value = 0x19, },
-		[5] = { .offset = 0x1c, .value = 0x64, },
-		[6] = { .offset = 0x1d, .value = 0x02, },
-		[7] = { .offset = 0x1e, .value = 0x02, },
-		[8] = { .offset = 0xf3, .value = 0x90, },
-		[9] = { .offset = 0xf9, .value = 0x00, },
-		[10] = { .offset = 0xc1, .value = 0xd7, },
-		[11] = { .offset = 0xc2, .value = 0x00, },
-		[12] = { .offset = 0xc3, .value = 0xf8, },
-		[13] = { .offset = 0xc4, .value = 0x03, },
-		[14] = { .offset = 0xc5, .value = 0x1a, },
-		[15] = { .offset = 0xc6, .value = 0x00, },
-		[16] = { .offset = 0xc7, .value = 0x73, },
-		[17] = { .offset = 0xc8, .value = 0x02, },
-		[18] = { .offset = 0xf4, .value = 0x00, },
-		[19] = { .offset = 0x80, .value = 0x27, },
-		[20] = { .offset = 0x81, .value = 0x03, },
-		[21] = { .offset = 0x82, .value = 0x41, },
-		[22] = { .offset = 0x83, .value = 0x05, },
-		[23] = { .offset = 0x94, .value = 0x00, },
-		[24] = { .offset = 0x95, .value = 0x00, },
-		[25] = { .offset = 0x96, .value = 0x05, },
-		[26] = { .offset = 0x97, .value = 0x00, },
-		[27] = { .offset = 0x9a, .value = 0x88, },
-		[28] = { .offset = 0x9b, .value = 0x00, },
-		[29] = { .offset = 0x98, .value = 0x00, },
-		[30] = { .offset = 0x99, .value = 0x00, },
-		[31] = { .offset = 0xf7, .value = 0x88, },
-		[32] = { .offset = 0xf8, .value = 0x06, },
-		[33] = { .offset = 0x9c, .value = 0x23, },
-		[34] = { .offset = 0x9d, .value = 0x00, },
-		[35] = { .offset = 0x9e, .value = 0x25, },
-		[36] = { .offset = 0x9f, .value = 0x03, },
-		[37] = { .offset = 0xa0, .value = 0x28, },
-		[38] = { .offset = 0xa1, .value = 0x01, },
-		[39] = { .offset = 0xa2, .value = 0x28, },
-		[40] = { .offset = 0xa3, .value = 0x05, },
-		[41] = { .offset = 0xb6, .value = 0x09, },
-		[42] = { .offset = 0xb8, .value = 0x30, },
-		[43] = { .offset = 0xb9, .value = 0xc8, },
-		[44] = { .offset = 0xba, .value = 0x00, },
-		[45] = { .offset = 0xbb, .value = 0x20, },
-		[46] = { .offset = 0x10, .value = 0x20, },
-		[47] = { .offset = 0x11, .value = 0xc8, },
-		[48] = { .offset = 0x12, .value = 0x02, },
-		[49] = { .offset = 0x20, .value = 0x00, },
-		[50] = { .offset = 0x22, .value = 0x00, },
-		[51] = { .offset = 0x23, .value = 0x00, },
-		[52] = { .offset = 0x24, .value = 0x00, },
-		[53] = { .offset = 0x25, .value = 0x00, },
-		[54] = { .offset = 0x8c, .value = 0x10, },
-		[55] = { .offset = 0x8d, .value = 0x02, },
-		[56] = { .offset = 0x8e, .value = 0x04, },
-		[57] = { .offset = 0x8f, .value = 0x00, },
-		[58] = { .offset = 0x90, .value = 0xff, },
-		[59] = { .offset = 0x91, .value = 0x07, },
-		[60] = { .offset = 0x92, .value = 0xa0, },
-		[61] = { .offset = 0x93, .value = 0x02, },
-		[62] = { .offset = 0xa5, .value = 0x00, },
-		[63] = { .offset = 0xa6, .value = 0x00, },
-		[64] = { .offset = 0xa7, .value = 0x00, },
-		[65] = { .offset = 0xa8, .value = 0x00, },
-		[66] = { .offset = 0xa9, .value = 0x83, },
-		[67] = { .offset = 0xaa, .value = 0x40, },
-		[68] = { .offset = 0xab, .value = 0x32, },
-		[69] = { .offset = 0xac, .value = 0x00, },
-		[70] = { .offset = 0xa4, .value = 0x80, },
-		[71] = { .offset = 0x7e, .value = 0x18, },
-		[72] = { .offset = 0x84, .value = 0x00, },
-		[73] = { .offset = 0x85, .value = 0x00, },
-		[74] = { .offset = 0x86, .value = 0x00, },
-		[75] = { .offset = 0x87, .value = 0x00, },
-		[76] = { .offset = 0x88, .value = 0x00, },
-		[77] = { .offset = 0x89, .value = 0x00, },
-		[78] = { .offset = 0x8a, .value = 0x00, },
-		[79] = { .offset = 0x8b, .value = 0x00, },
-		[80] = { .offset = 0x26, .value = 0x00, },
-		[81] = { .offset = 0x27, .value = 0x00, },
-		[82] = { .offset = 0xad, .value = 0x00, },
-		[83] = { .offset = 0x08, .value = 0x30, }, /* 0x31 */
-		[84] = { .offset = 0x41, .value = 0x00, },
-		[85] = { .offset = 0xc0, .value = 0x07, },
+		.sync	= NS2501_C0_ENABLE |
+			  NS2501_C0_HSYNC | NS2501_C0_VSYNC,
+		.conf   = NS2501_8_VEN | NS2501_8_HEN | NS2501_8_PD,
+		.syncb	= 0x00,
+		.dither	= 0x0f,
+		.pll_a	= 25,
+		.pll_b	= 612,
+		.hstart	= 215,
+		.hstop	= 1016,
+		.vstart	= 26,
+		.vstop	= 627,
+		.vsync	= 807,
+		.vtotal	= 1341,
+		.hpos	= 0,
+		.vpos	= 4,
+		.voffs	= 35,
+		.hscale	= 51248,
+		.vscale	= 51232
 	},
 	[MODE_1024x768] = {
-		[0] = { .offset = 0x0a, .value = 0x81, },
-		[1] = { .offset = 0x18, .value = 0x07, },
-		[2] = { .offset = 0x19, .value = 0x00, },
-		[3] = { .offset = 0x1a, .value = 0x00, },
-		[4] = { .offset = 0x1b, .value = 0x11, },
-		[5] = { .offset = 0x1c, .value = 0x54, },
-		[6] = { .offset = 0x1d, .value = 0x03, },
-		[7] = { .offset = 0x1e, .value = 0x02, },
-		[8] = { .offset = 0xf3, .value = 0x90, },
-		[9] = { .offset = 0xf9, .value = 0x00, },
-		[10] = { .offset = 0xc1, .value = 0x90, },
-		[11] = { .offset = 0xc2, .value = 0x00, },
-		[12] = { .offset = 0xc3, .value = 0x0f, },
-		[13] = { .offset = 0xc4, .value = 0x03, },
-		[14] = { .offset = 0xc5, .value = 0x16, },
-		[15] = { .offset = 0xc6, .value = 0x00, },
-		[16] = { .offset = 0xc7, .value = 0x02, },
-		[17] = { .offset = 0xc8, .value = 0x02, },
-		[18] = { .offset = 0xf4, .value = 0x00, },
-		[19] = { .offset = 0x80, .value = 0xff, },
-		[20] = { .offset = 0x81, .value = 0x07, },
-		[21] = { .offset = 0x82, .value = 0x3d, },
-		[22] = { .offset = 0x83, .value = 0x05, },
-		[23] = { .offset = 0x94, .value = 0x00, },
-		[24] = { .offset = 0x95, .value = 0x00, },
-		[25] = { .offset = 0x96, .value = 0x05, },
-		[26] = { .offset = 0x97, .value = 0x00, },
-		[27] = { .offset = 0x9a, .value = 0x88, },
-		[28] = { .offset = 0x9b, .value = 0x00, },
-		[29] = { .offset = 0x98, .value = 0x00, },
-		[30] = { .offset = 0x99, .value = 0x00, },
-		[31] = { .offset = 0xf7, .value = 0x88, },
-		[32] = { .offset = 0xf8, .value = 0x0a, },
-		[33] = { .offset = 0x9c, .value = 0x24, },
-		[34] = { .offset = 0x9d, .value = 0x00, },
-		[35] = { .offset = 0x9e, .value = 0x25, },
-		[36] = { .offset = 0x9f, .value = 0x03, },
-		[37] = { .offset = 0xa0, .value = 0x28, },
-		[38] = { .offset = 0xa1, .value = 0x01, },
-		[39] = { .offset = 0xa2, .value = 0x28, },
-		[40] = { .offset = 0xa3, .value = 0x05, },
-		[41] = { .offset = 0xb6, .value = 0x09, },
-		[42] = { .offset = 0xb8, .value = 0x00, },
-		[43] = { .offset = 0xb9, .value = 0xa0, },
-		[44] = { .offset = 0xba, .value = 0x00, },
-		[45] = { .offset = 0xbb, .value = 0x20, },
-		[46] = { .offset = 0x10, .value = 0x00, },
-		[47] = { .offset = 0x11, .value = 0xa0, },
-		[48] = { .offset = 0x12, .value = 0x02, },
-		[49] = { .offset = 0x20, .value = 0x00, },
-		[50] = { .offset = 0x22, .value = 0x00, },
-		[51] = { .offset = 0x23, .value = 0x00, },
-		[52] = { .offset = 0x24, .value = 0x00, },
-		[53] = { .offset = 0x25, .value = 0x00, },
-		[54] = { .offset = 0x8c, .value = 0x10, },
-		[55] = { .offset = 0x8d, .value = 0x02, },
-		[56] = { .offset = 0x8e, .value = 0x10, },
-		[57] = { .offset = 0x8f, .value = 0x00, },
-		[58] = { .offset = 0x90, .value = 0xff, },
-		[59] = { .offset = 0x91, .value = 0x07, },
-		[60] = { .offset = 0x92, .value = 0xa0, },
-		[61] = { .offset = 0x93, .value = 0x02, },
-		[62] = { .offset = 0xa5, .value = 0x00, },
-		[63] = { .offset = 0xa6, .value = 0x00, },
-		[64] = { .offset = 0xa7, .value = 0x00, },
-		[65] = { .offset = 0xa8, .value = 0x00, },
-		[66] = { .offset = 0xa9, .value = 0x04, },
-		[67] = { .offset = 0xaa, .value = 0x70, },
-		[68] = { .offset = 0xab, .value = 0x4f, },
-		[69] = { .offset = 0xac, .value = 0x00, },
-		[70] = { .offset = 0xa4, .value = 0x84, },
-		[71] = { .offset = 0x7e, .value = 0x18, },
-		[72] = { .offset = 0x84, .value = 0x00, },
-		[73] = { .offset = 0x85, .value = 0x00, },
-		[74] = { .offset = 0x86, .value = 0x00, },
-		[75] = { .offset = 0x87, .value = 0x00, },
-		[76] = { .offset = 0x88, .value = 0x00, },
-		[77] = { .offset = 0x89, .value = 0x00, },
-		[78] = { .offset = 0x8a, .value = 0x00, },
-		[79] = { .offset = 0x8b, .value = 0x00, },
-		[80] = { .offset = 0x26, .value = 0x00, },
-		[81] = { .offset = 0x27, .value = 0x00, },
-		[82] = { .offset = 0xad, .value = 0x00, },
-		[83] = { .offset = 0x08, .value = 0x34, }, /* 0x35 */
-		[84] = { .offset = 0x41, .value = 0x00, },
-		[85] = { .offset = 0xc0, .value = 0x01, },
-	},
+		.sync	= NS2501_C0_ENABLE | NS2501_C0_VSYNC,
+		.conf   = NS2501_8_VEN | NS2501_8_HEN | NS2501_8_PD,
+		.syncb	= 0x32,
+		.dither	= 0x0f,
+		.pll_a	= 11,
+		.pll_b	= 1350,
+		.hstart	= 276,
+		.hstop	= 1299,
+		.vstart	= 15,
+		.vstop	= 1056,
+		.vsync	= 2047,
+		.vtotal	= 1341,
+		.hpos	= 0,
+		.vpos	= 7,
+		.voffs	= 27,
+		.hscale	= 65535,
+		.vscale	= 65535
+	}
+};
+
+/*
+ * Other configuration values left by the BIOS of the
+ * Fujitsu S6010 in the DVO control registers. Their
+ * value does not depend on the BIOS and their meaning
+ * is unknown.
+ */
+
+static const struct ns2501_reg mode_agnostic_values[] = {
+	/* 08 is mode specific */
+	[0] = { .offset = 0x0a, .value = 0x81, },
+	/* 10,11 are part of the mode specific configuration */
+	[1] = { .offset = 0x12, .value = 0x02, },
+	[2] = { .offset = 0x18, .value = 0x07, },
+	[3] = { .offset = 0x19, .value = 0x00, },
+	[4] = { .offset = 0x1a, .value = 0x00, }, /* PLL?, ignored */
+	/* 1b,1c,1d are part of the mode specific configuration */
+	[5] = { .offset = 0x1e, .value = 0x02, },
+	[6] = { .offset = 0x1f, .value = 0x40, },
+	[7] = { .offset = 0x20, .value = 0x00, },
+	[8] = { .offset = 0x21, .value = 0x00, },
+	[9] = { .offset = 0x22, .value = 0x00, },
+	[10] = { .offset = 0x23, .value = 0x00, },
+	[11] = { .offset = 0x24, .value = 0x00, },
+	[12] = { .offset = 0x25, .value = 0x00, },
+	[13] = { .offset = 0x26, .value = 0x00, },
+	[14] = { .offset = 0x27, .value = 0x00, },
+	[15] = { .offset = 0x7e, .value = 0x18, },
+	/* 80-84 are part of the mode-specific configuration */
+	[16] = { .offset = 0x84, .value = 0x00, },
+	[17] = { .offset = 0x85, .value = 0x00, },
+	[18] = { .offset = 0x86, .value = 0x00, },
+	[19] = { .offset = 0x87, .value = 0x00, },
+	[20] = { .offset = 0x88, .value = 0x00, },
+	[21] = { .offset = 0x89, .value = 0x00, },
+	[22] = { .offset = 0x8a, .value = 0x00, },
+	[23] = { .offset = 0x8b, .value = 0x00, },
+	[24] = { .offset = 0x8c, .value = 0x10, },
+	[25] = { .offset = 0x8d, .value = 0x02, },
+	/* 8e,8f are part of the mode-specific configuration */
+	[26] = { .offset = 0x90, .value = 0xff, },
+	[27] = { .offset = 0x91, .value = 0x07, },
+	[28] = { .offset = 0x92, .value = 0xa0, },
+	[29] = { .offset = 0x93, .value = 0x02, },
+	[30] = { .offset = 0x94, .value = 0x00, },
+	[31] = { .offset = 0x95, .value = 0x00, },
+	[32] = { .offset = 0x96, .value = 0x05, },
+	[33] = { .offset = 0x97, .value = 0x00, },
+	/* 98,99 are part of the mode-specific configuration */
+	[34] = { .offset = 0x9a, .value = 0x88, },
+	[35] = { .offset = 0x9b, .value = 0x00, },
+	/* 9c,9d are part of the mode-specific configuration */
+	[36] = { .offset = 0x9e, .value = 0x25, },
+	[37] = { .offset = 0x9f, .value = 0x03, },
+	[38] = { .offset = 0xa0, .value = 0x28, },
+	[39] = { .offset = 0xa1, .value = 0x01, },
+	[40] = { .offset = 0xa2, .value = 0x28, },
+	[41] = { .offset = 0xa3, .value = 0x05, },
+	/* register 0xa4 is mode specific, but 0x80..0x84 works always */
+	[42] = { .offset = 0xa4, .value = 0x84, },
+	[43] = { .offset = 0xa5, .value = 0x00, },
+	[44] = { .offset = 0xa6, .value = 0x00, },
+	[45] = { .offset = 0xa7, .value = 0x00, },
+	[46] = { .offset = 0xa8, .value = 0x00, },
+	/* 0xa9 to 0xab are mode specific, but have no visible effect */
+	[47] = { .offset = 0xa9, .value = 0x04, },
+	[48] = { .offset = 0xaa, .value = 0x70, },
+	[49] = { .offset = 0xab, .value = 0x4f, },
+	[50] = { .offset = 0xac, .value = 0x00, },
+	[51] = { .offset = 0xad, .value = 0x00, },
+	[52] = { .offset = 0xb6, .value = 0x09, },
+	[53] = { .offset = 0xb7, .value = 0x03, },
+	/* b8,b9 are part of the mode-specific configuration */
+	[54] = { .offset = 0xba, .value = 0x00, },
+	[55] = { .offset = 0xbb, .value = 0x20, },
+	[56] = { .offset = 0xf3, .value = 0x90, },
+	[57] = { .offset = 0xf4, .value = 0x00, },
+	[58] = { .offset = 0xf7, .value = 0x88, },
+	/* f8 is mode specific, but the value does not matter */
+	[59] = { .offset = 0xf8, .value = 0x0a, },
+	[60] = { .offset = 0xf9, .value = 0x00, }
 };
 
 static const struct ns2501_reg regs_init[] = {
@@ -350,25 +378,12 @@ static const struct ns2501_reg regs_init[] = {
 
 struct ns2501_priv {
 	bool quiet;
-	const struct ns2501_reg *regs;
+	const struct ns2501_configuration *conf;
 };
 
 #define NSPTR(d) ((NS2501Ptr)(d->DriverPrivate.ptr))
 
 /*
- * For reasons unclear to me, the ns2501 at least on the Fujitsu/Siemens
- * laptops does not react on the i2c bus unless
- * both the PLL is running and the display is configured in its native
- * resolution.
- * This function forces the DVO on, and stores the registers it touches.
- * Afterwards, registers are restored to regular values.
- *
- * This is pretty much a hack, though it works.
- * Without that, ns2501_readb and ns2501_writeb fail
- * when switching the resolution.
- */
-
-/*
 ** Read a register from the ns2501.
 ** Returns true if successful, false otherwise.
 ** If it returns false, it might be wise to enable the
@@ -534,6 +549,7 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 			    struct drm_display_mode *mode,
 			    struct drm_display_mode *adjusted_mode)
 {
+	const struct ns2501_configuration *conf;
 	struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv);
 	int mode_idx, i;
 
@@ -541,6 +557,36 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 	    ("set mode (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d).\n",
 	     mode->hdisplay, mode->htotal, mode->vdisplay, mode->vtotal);
 
+	DRM_DEBUG_KMS("Detailed requested mode settings are:\n"
+			"clock		: %d kHz\n"
+			"hdisplay	: %d\n"
+			"hblank start	: %d\n"
+			"hblank end	: %d\n"
+			"hsync start	: %d\n"
+			"hsync end	: %d\n"
+			"htotal		: %d\n"
+			"hskew		: %d\n"
+			"vdisplay	: %d\n"
+			"vblank start	: %d\n"
+			"hblank end	: %d\n"
+			"vsync start	: %d\n"
+			"vsync end	: %d\n"
+			"vtotal		: %d\n",
+			adjusted_mode->crtc_clock,
+			adjusted_mode->crtc_hdisplay,
+			adjusted_mode->crtc_hblank_start,
+			adjusted_mode->crtc_hblank_end,
+			adjusted_mode->crtc_hsync_start,
+			adjusted_mode->crtc_hsync_end,
+			adjusted_mode->crtc_htotal,
+			adjusted_mode->crtc_hskew,
+			adjusted_mode->crtc_vdisplay,
+			adjusted_mode->crtc_vblank_start,
+			adjusted_mode->crtc_vblank_end,
+			adjusted_mode->crtc_vsync_start,
+			adjusted_mode->crtc_vsync_end,
+			adjusted_mode->crtc_vtotal);
+
 	if (mode->hdisplay == 640 && mode->vdisplay == 480)
 		mode_idx = MODE_640x480;
 	else if (mode->hdisplay == 800 && mode->vdisplay == 600)
@@ -554,10 +600,44 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 	for (i = 0; i < ARRAY_SIZE(regs_init); i++)
 		ns2501_writeb(dvo, regs_init[i].offset, regs_init[i].value);
 
-	ns->regs = regs_1024x768[mode_idx];
-
-	for (i = 0; i < 84; i++)
-		ns2501_writeb(dvo, ns->regs[i].offset, ns->regs[i].value);
+	/* Write the mode-agnostic values */
+	for (i = 0; i < ARRAY_SIZE(mode_agnostic_values); i++)
+		ns2501_writeb(dvo, mode_agnostic_values[i].offset,
+				mode_agnostic_values[i].value);
+
+	/* Write now the mode-specific configuration */
+	conf = ns2501_modes + mode_idx;
+	ns->conf = conf;
+
+	ns2501_writeb(dvo, NS2501_REG8, conf->conf);
+	ns2501_writeb(dvo, NS2501_REG1B, conf->pll_a);
+	ns2501_writeb(dvo, NS2501_REG1C, conf->pll_b & 0xff);
+	ns2501_writeb(dvo, NS2501_REG1D, conf->pll_b >> 8);
+	ns2501_writeb(dvo, NS2501_REGC1, conf->hstart & 0xff);
+	ns2501_writeb(dvo, NS2501_REGC2, conf->hstart >> 8);
+	ns2501_writeb(dvo, NS2501_REGC3, conf->hstop & 0xff);
+	ns2501_writeb(dvo, NS2501_REGC4, conf->hstop >> 8);
+	ns2501_writeb(dvo, NS2501_REGC5, conf->vstart & 0xff);
+	ns2501_writeb(dvo, NS2501_REGC6, conf->vstart >> 8);
+	ns2501_writeb(dvo, NS2501_REGC7, conf->vstop & 0xff);
+	ns2501_writeb(dvo, NS2501_REGC8, conf->vstop >> 8);
+	ns2501_writeb(dvo, NS2501_REG80, conf->vsync & 0xff);
+	ns2501_writeb(dvo, NS2501_REG81, conf->vsync >> 8);
+	ns2501_writeb(dvo, NS2501_REG82, conf->vtotal & 0xff);
+	ns2501_writeb(dvo, NS2501_REG83, conf->vtotal >> 8);
+	ns2501_writeb(dvo, NS2501_REG98, conf->hpos & 0xff);
+	ns2501_writeb(dvo, NS2501_REG99, conf->hpos >> 8);
+	ns2501_writeb(dvo, NS2501_REG8E, conf->vpos & 0xff);
+	ns2501_writeb(dvo, NS2501_REG8F, conf->vpos >> 8);
+	ns2501_writeb(dvo, NS2501_REG9C, conf->voffs & 0xff);
+	ns2501_writeb(dvo, NS2501_REG9D, conf->voffs >> 8);
+	ns2501_writeb(dvo, NS2501_REGB8, conf->hscale & 0xff);
+	ns2501_writeb(dvo, NS2501_REGB9, conf->hscale >> 8);
+	ns2501_writeb(dvo, NS2501_REG10, conf->vscale & 0xff);
+	ns2501_writeb(dvo, NS2501_REG11, conf->vscale >> 8);
+	ns2501_writeb(dvo, NS2501_REGF9, conf->dither);
+	ns2501_writeb(dvo, NS2501_REG41, conf->syncb);
+	ns2501_writeb(dvo, NS2501_REGC0, conf->sync);
 }
 
 /* set the NS2501 power state */
@@ -579,34 +659,32 @@ static void ns2501_dpms(struct intel_dvo_device *dvo, bool enable)
 	DRM_DEBUG_KMS("Trying set the dpms of the DVO to %i\n", enable);
 
 	if (enable) {
-		if (WARN_ON(ns->regs[83].offset != 0x08 ||
-			    ns->regs[84].offset != 0x41 ||
-			    ns->regs[85].offset != 0xc0))
-			return;
-
-		ns2501_writeb(dvo, 0xc0, ns->regs[85].value | 0x08);
+		ns2501_writeb(dvo, NS2501_REGC0, ns->conf->sync | 0x08);
 
-		ns2501_writeb(dvo, 0x41, ns->regs[84].value);
+		ns2501_writeb(dvo, NS2501_REG41, ns->conf->syncb);
 
-		ns2501_writeb(dvo, 0x34, 0x01);
+		ns2501_writeb(dvo, NS2501_REG34, NS2501_34_ENABLE_OUTPUT);
 		msleep(15);
 
-		ns2501_writeb(dvo, 0x08, 0x35);
-		if (!(ns->regs[83].value & NS2501_8_BPAS))
-			ns2501_writeb(dvo, 0x08, 0x31);
+		ns2501_writeb(dvo, NS2501_REG8,
+				ns->conf->conf | NS2501_8_BPAS);
+		if (!(ns->conf->conf & NS2501_8_BPAS))
+			ns2501_writeb(dvo, NS2501_REG8, ns->conf->conf);
 		msleep(200);
 
-		ns2501_writeb(dvo, 0x34, 0x03);
+		ns2501_writeb(dvo, NS2501_REG34,
+			NS2501_34_ENABLE_OUTPUT | NS2501_34_ENABLE_BACKLIGHT);
 
-		ns2501_writeb(dvo, 0xc0, ns->regs[85].value);
+		ns2501_writeb(dvo, NS2501_REGC0, ns->conf->sync);
 	} else {
-		ns2501_writeb(dvo, 0x34, 0x01);
+		ns2501_writeb(dvo, NS2501_REG34, NS2501_34_ENABLE_OUTPUT);
 		msleep(200);
 
-		ns2501_writeb(dvo, 0x08, 0x34);
+		ns2501_writeb(dvo, NS2501_REG8, NS2501_8_VEN | NS2501_8_HEN |
+				NS2501_8_BPAS);
 		msleep(15);
 
-		ns2501_writeb(dvo, 0x34, 0x00);
+		ns2501_writeb(dvo, NS2501_REG34, 0x00);
 	}
 }
 
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 61ae8ff4eaed..306d9e4e5cf3 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -123,7 +123,7 @@ static const struct drm_i915_cmd_descriptor common_cmds[] = {
 	CMD(  MI_SEMAPHORE_MBOX,                SMI,   !F,  0xFF,   R  ),
 	CMD(  MI_STORE_DWORD_INDEX,             SMI,   !F,  0xFF,   R  ),
 	CMD(  MI_LOAD_REGISTER_IMM(1),          SMI,   !F,  0xFF,   W,
-	      .reg = { .offset = 1, .mask = 0x007FFFFC }               ),
+	      .reg = { .offset = 1, .mask = 0x007FFFFC, .step = 2 }    ),
 	CMD(  MI_STORE_REGISTER_MEM(1),         SMI,   !F,  0xFF,   W | B,
 	      .reg = { .offset = 1, .mask = 0x007FFFFC },
 	      .bits = {{
@@ -395,16 +395,38 @@ static const struct drm_i915_cmd_table hsw_blt_ring_cmds[] = {
 
 /*
  * Register whitelists, sorted by increasing register offset.
+ */
+
+/*
+ * An individual whitelist entry granting access to register addr.  If
+ * mask is non-zero the argument of immediate register writes will be
+ * AND-ed with mask, and the command will be rejected if the result
+ * doesn't match value.
+ *
+ * Registers with non-zero mask are only allowed to be written using
+ * LRI.
+ */
+struct drm_i915_reg_descriptor {
+	u32 addr;
+	u32 mask;
+	u32 value;
+};
+
+/* Convenience macro for adding 32-bit registers. */
+#define REG32(address, ...)                             \
+	{ .addr = address, __VA_ARGS__ }
+
+/*
+ * Convenience macro for adding 64-bit registers.
  *
  * Some registers that userspace accesses are 64 bits. The register
  * access commands only allow 32-bit accesses. Hence, we have to include
  * entries for both halves of the 64-bit registers.
  */
+#define REG64(addr)                                     \
+	REG32(addr), REG32(addr + sizeof(u32))
 
-/* Convenience macro for adding 64-bit registers */
-#define REG64(addr) (addr), (addr + sizeof(u32))
-
-static const u32 gen7_render_regs[] = {
+static const struct drm_i915_reg_descriptor gen7_render_regs[] = {
 	REG64(GPGPU_THREADS_DISPATCHED),
 	REG64(HS_INVOCATION_COUNT),
 	REG64(DS_INVOCATION_COUNT),
@@ -417,15 +439,15 @@ static const u32 gen7_render_regs[] = {
 	REG64(CL_PRIMITIVES_COUNT),
 	REG64(PS_INVOCATION_COUNT),
 	REG64(PS_DEPTH_COUNT),
-	OACONTROL, /* Only allowed for LRI and SRM. See below. */
+	REG32(OACONTROL), /* Only allowed for LRI and SRM. See below. */
 	REG64(MI_PREDICATE_SRC0),
 	REG64(MI_PREDICATE_SRC1),
-	GEN7_3DPRIM_END_OFFSET,
-	GEN7_3DPRIM_START_VERTEX,
-	GEN7_3DPRIM_VERTEX_COUNT,
-	GEN7_3DPRIM_INSTANCE_COUNT,
-	GEN7_3DPRIM_START_INSTANCE,
-	GEN7_3DPRIM_BASE_VERTEX,
+	REG32(GEN7_3DPRIM_END_OFFSET),
+	REG32(GEN7_3DPRIM_START_VERTEX),
+	REG32(GEN7_3DPRIM_VERTEX_COUNT),
+	REG32(GEN7_3DPRIM_INSTANCE_COUNT),
+	REG32(GEN7_3DPRIM_START_INSTANCE),
+	REG32(GEN7_3DPRIM_BASE_VERTEX),
 	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(0)),
 	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(1)),
 	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(2)),
@@ -434,33 +456,41 @@ static const u32 gen7_render_regs[] = {
 	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(1)),
 	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(2)),
 	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(3)),
-	GEN7_SO_WRITE_OFFSET(0),
-	GEN7_SO_WRITE_OFFSET(1),
-	GEN7_SO_WRITE_OFFSET(2),
-	GEN7_SO_WRITE_OFFSET(3),
-	GEN7_L3SQCREG1,
-	GEN7_L3CNTLREG2,
-	GEN7_L3CNTLREG3,
+	REG32(GEN7_SO_WRITE_OFFSET(0)),
+	REG32(GEN7_SO_WRITE_OFFSET(1)),
+	REG32(GEN7_SO_WRITE_OFFSET(2)),
+	REG32(GEN7_SO_WRITE_OFFSET(3)),
+	REG32(GEN7_L3SQCREG1),
+	REG32(GEN7_L3CNTLREG2),
+	REG32(GEN7_L3CNTLREG3),
+	REG32(HSW_SCRATCH1,
+	      .mask = ~HSW_SCRATCH1_L3_DATA_ATOMICS_DISABLE,
+	      .value = 0),
+	REG32(HSW_ROW_CHICKEN3,
+	      .mask = ~(HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE << 16 |
+                        HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE),
+	      .value = 0),
 };
 
-static const u32 gen7_blt_regs[] = {
-	BCS_SWCTRL,
+static const struct drm_i915_reg_descriptor gen7_blt_regs[] = {
+	REG32(BCS_SWCTRL),
 };
 
-static const u32 ivb_master_regs[] = {
-	FORCEWAKE_MT,
-	DERRMR,
-	GEN7_PIPE_DE_LOAD_SL(PIPE_A),
-	GEN7_PIPE_DE_LOAD_SL(PIPE_B),
-	GEN7_PIPE_DE_LOAD_SL(PIPE_C),
+static const struct drm_i915_reg_descriptor ivb_master_regs[] = {
+	REG32(FORCEWAKE_MT),
+	REG32(DERRMR),
+	REG32(GEN7_PIPE_DE_LOAD_SL(PIPE_A)),
+	REG32(GEN7_PIPE_DE_LOAD_SL(PIPE_B)),
+	REG32(GEN7_PIPE_DE_LOAD_SL(PIPE_C)),
 };
 
-static const u32 hsw_master_regs[] = {
-	FORCEWAKE_MT,
-	DERRMR,
+static const struct drm_i915_reg_descriptor hsw_master_regs[] = {
+	REG32(FORCEWAKE_MT),
+	REG32(DERRMR),
 };
 
 #undef REG64
+#undef REG32
 
 static u32 gen7_render_get_cmd_length_mask(u32 cmd_header)
 {
@@ -550,14 +580,16 @@ static bool validate_cmds_sorted(struct intel_engine_cs *ring,
 	return ret;
 }
 
-static bool check_sorted(int ring_id, const u32 *reg_table, int reg_count)
+static bool check_sorted(int ring_id,
+			 const struct drm_i915_reg_descriptor *reg_table,
+			 int reg_count)
 {
 	int i;
 	u32 previous = 0;
 	bool ret = true;
 
 	for (i = 0; i < reg_count; i++) {
-		u32 curr = reg_table[i];
+		u32 curr = reg_table[i].addr;
 
 		if (curr < previous) {
 			DRM_ERROR("CMD: table not sorted ring=%d entry=%d reg=0x%08X prev=0x%08X\n",
@@ -804,18 +836,20 @@ find_cmd(struct intel_engine_cs *ring,
 	return default_desc;
 }
 
-static bool valid_reg(const u32 *table, int count, u32 addr)
+static const struct drm_i915_reg_descriptor *
+find_reg(const struct drm_i915_reg_descriptor *table,
+	 int count, u32 addr)
 {
-	if (table && count != 0) {
+	if (table) {
 		int i;
 
 		for (i = 0; i < count; i++) {
-			if (table[i] == addr)
-				return true;
+			if (table[i].addr == addr)
+				return &table[i];
 		}
 	}
 
-	return false;
+	return NULL;
 }
 
 static u32 *vmap_batch(struct drm_i915_gem_object *obj,
@@ -869,6 +903,9 @@ static u32 *copy_batch(struct drm_i915_gem_object *dest_obj,
 	    batch_len + batch_start_offset > src_obj->base.size)
 		return ERR_PTR(-E2BIG);
 
+	if (WARN_ON(dest_obj->pages_pin_count == 0))
+		return ERR_PTR(-ENODEV);
+
 	ret = i915_gem_obj_prepare_shmem_read(src_obj, &needs_clflush);
 	if (ret) {
 		DRM_DEBUG_DRIVER("CMD: failed to prepare shadow batch\n");
@@ -882,13 +919,6 @@ static u32 *copy_batch(struct drm_i915_gem_object *dest_obj,
 		goto unpin_src;
 	}
 
-	ret = i915_gem_object_get_pages(dest_obj);
-	if (ret) {
-		DRM_DEBUG_DRIVER("CMD: Failed to get pages for shadow batch\n");
-		goto unmap_src;
-	}
-	i915_gem_object_pin_pages(dest_obj);
-
 	ret = i915_gem_object_set_to_cpu_domain(dest_obj, true);
 	if (ret) {
 		DRM_DEBUG_DRIVER("CMD: Failed to set shadow batch to CPU\n");
@@ -898,7 +928,6 @@ static u32 *copy_batch(struct drm_i915_gem_object *dest_obj,
 	dst = vmap_batch(dest_obj, 0, batch_len);
 	if (!dst) {
 		DRM_DEBUG_DRIVER("CMD: Failed to vmap shadow batch\n");
-		i915_gem_object_unpin_pages(dest_obj);
 		ret = -ENOMEM;
 		goto unmap_src;
 	}
@@ -939,7 +968,7 @@ bool i915_needs_cmd_parser(struct intel_engine_cs *ring)
 
 static bool check_cmd(const struct intel_engine_cs *ring,
 		      const struct drm_i915_cmd_descriptor *desc,
-		      const u32 *cmd,
+		      const u32 *cmd, u32 length,
 		      const bool is_master,
 		      bool *oacontrol_set)
 {
@@ -955,38 +984,70 @@ static bool check_cmd(const struct intel_engine_cs *ring,
 	}
 
 	if (desc->flags & CMD_DESC_REGISTER) {
-		u32 reg_addr = cmd[desc->reg.offset] & desc->reg.mask;
-
 		/*
-		 * OACONTROL requires some special handling for writes. We
-		 * want to make sure that any batch which enables OA also
-		 * disables it before the end of the batch. The goal is to
-		 * prevent one process from snooping on the perf data from
-		 * another process. To do that, we need to check the value
-		 * that will be written to the register. Hence, limit
-		 * OACONTROL writes to only MI_LOAD_REGISTER_IMM commands.
+		 * Get the distance between individual register offset
+		 * fields if the command can perform more than one
+		 * access at a time.
 		 */
-		if (reg_addr == OACONTROL) {
-			if (desc->cmd.value == MI_LOAD_REGISTER_MEM) {
-				DRM_DEBUG_DRIVER("CMD: Rejected LRM to OACONTROL\n");
+		const u32 step = desc->reg.step ? desc->reg.step : length;
+		u32 offset;
+
+		for (offset = desc->reg.offset; offset < length;
+		     offset += step) {
+			const u32 reg_addr = cmd[offset] & desc->reg.mask;
+			const struct drm_i915_reg_descriptor *reg =
+				find_reg(ring->reg_table, ring->reg_count,
+					 reg_addr);
+
+			if (!reg && is_master)
+				reg = find_reg(ring->master_reg_table,
+					       ring->master_reg_count,
+					       reg_addr);
+
+			if (!reg) {
+				DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
+						 reg_addr, *cmd, ring->id);
 				return false;
 			}
 
-			if (desc->cmd.value == MI_LOAD_REGISTER_IMM(1))
-				*oacontrol_set = (cmd[2] != 0);
-		}
+			/*
+			 * OACONTROL requires some special handling for
+			 * writes. We want to make sure that any batch which
+			 * enables OA also disables it before the end of the
+			 * batch. The goal is to prevent one process from
+			 * snooping on the perf data from another process. To do
+			 * that, we need to check the value that will be written
+			 * to the register. Hence, limit OACONTROL writes to
+			 * only MI_LOAD_REGISTER_IMM commands.
+			 */
+			if (reg_addr == OACONTROL) {
+				if (desc->cmd.value == MI_LOAD_REGISTER_MEM) {
+					DRM_DEBUG_DRIVER("CMD: Rejected LRM to OACONTROL\n");
+					return false;
+				}
+
+				if (desc->cmd.value == MI_LOAD_REGISTER_IMM(1))
+					*oacontrol_set = (cmd[offset + 1] != 0);
+			}
 
-		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);
-				return false;
+			/*
+			 * Check the value written to the register against the
+			 * allowed mask/value pair given in the whitelist entry.
+			 */
+			if (reg->mask) {
+				if (desc->cmd.value == MI_LOAD_REGISTER_MEM) {
+					DRM_DEBUG_DRIVER("CMD: Rejected LRM to masked register 0x%08X\n",
+							 reg_addr);
+					return false;
+				}
+
+				if (desc->cmd.value == MI_LOAD_REGISTER_IMM(1) &&
+				    (offset + 2 > length ||
+				     (cmd[offset + 1] & reg->mask) != reg->value)) {
+					DRM_DEBUG_DRIVER("CMD: Rejected LRI to masked register 0x%08X\n",
+							 reg_addr);
+					return false;
+				}
 			}
 		}
 	}
@@ -1110,7 +1171,8 @@ int i915_parse_cmds(struct intel_engine_cs *ring,
 			break;
 		}
 
-		if (!check_cmd(ring, desc, cmd, is_master, &oacontrol_set)) {
+		if (!check_cmd(ring, desc, cmd, length, is_master,
+			       &oacontrol_set)) {
 			ret = -EINVAL;
 			break;
 		}
@@ -1129,7 +1191,6 @@ int i915_parse_cmds(struct intel_engine_cs *ring,
 	}
 
 	vunmap(batch_base);
-	i915_gem_object_unpin_pages(shadow_batch_obj);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index dc55c51964ab..82bbe3f2a7e1 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -96,7 +96,7 @@ static int i915_capabilities(struct seq_file *m, void *data)
 
 static const char *get_pin_flag(struct drm_i915_gem_object *obj)
 {
-	if (i915_gem_obj_is_pinned(obj))
+	if (obj->pin_display)
 		return "p";
 	else
 		return " ";
@@ -120,18 +120,25 @@ static inline const char *get_global_flag(struct drm_i915_gem_object *obj)
 static void
 describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 {
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
+	struct intel_engine_cs *ring;
 	struct i915_vma *vma;
 	int pin_count = 0;
+	int i;
 
-	seq_printf(m, "%pK: %s%s%s %8zdKiB %02x %02x %x %x %x%s%s%s",
+	seq_printf(m, "%pK: %s%s%s%s %8zdKiB %02x %02x [ ",
 		   &obj->base,
+		   obj->active ? "*" : " ",
 		   get_pin_flag(obj),
 		   get_tiling_flag(obj),
 		   get_global_flag(obj),
 		   obj->base.size / 1024,
 		   obj->base.read_domains,
-		   obj->base.write_domain,
-		   i915_gem_request_get_seqno(obj->last_read_req),
+		   obj->base.write_domain);
+	for_each_ring(ring, dev_priv, i)
+		seq_printf(m, "%x ",
+				i915_gem_request_get_seqno(obj->last_read_req[i]));
+	seq_printf(m, "] %x %x%s%s%s",
 		   i915_gem_request_get_seqno(obj->last_write_req),
 		   i915_gem_request_get_seqno(obj->last_fenced_req),
 		   i915_cache_level_str(to_i915(obj->base.dev), obj->cache_level),
@@ -159,18 +166,18 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 	}
 	if (obj->stolen)
 		seq_printf(m, " (stolen: %08llx)", obj->stolen->start);
-	if (obj->pin_mappable || obj->fault_mappable) {
+	if (obj->pin_display || obj->fault_mappable) {
 		char s[3], *t = s;
-		if (obj->pin_mappable)
+		if (obj->pin_display)
 			*t++ = 'p';
 		if (obj->fault_mappable)
 			*t++ = 'f';
 		*t = '\0';
 		seq_printf(m, " (%s mappable)", s);
 	}
-	if (obj->last_read_req != NULL)
+	if (obj->last_write_req != NULL)
 		seq_printf(m, " (%s)",
-			   i915_gem_request_get_ring(obj->last_read_req)->name);
+			   i915_gem_request_get_ring(obj->last_write_req)->name);
 	if (obj->frontbuffer_bits)
 		seq_printf(m, " (frontbuffer: 0x%03x)", obj->frontbuffer_bits);
 }
@@ -361,31 +368,39 @@ static int per_file_stats(int id, void *ptr, void *data)
 	return 0;
 }
 
-#define print_file_stats(m, name, stats) \
-	seq_printf(m, "%s: %u objects, %zu bytes (%zu active, %zu inactive, %zu global, %zu shared, %zu unbound)\n", \
-		   name, \
-		   stats.count, \
-		   stats.total, \
-		   stats.active, \
-		   stats.inactive, \
-		   stats.global, \
-		   stats.shared, \
-		   stats.unbound)
+#define print_file_stats(m, name, stats) do { \
+	if (stats.count) \
+		seq_printf(m, "%s: %u objects, %zu bytes (%zu active, %zu inactive, %zu global, %zu shared, %zu unbound)\n", \
+			   name, \
+			   stats.count, \
+			   stats.total, \
+			   stats.active, \
+			   stats.inactive, \
+			   stats.global, \
+			   stats.shared, \
+			   stats.unbound); \
+} while (0)
 
 static void print_batch_pool_stats(struct seq_file *m,
 				   struct drm_i915_private *dev_priv)
 {
 	struct drm_i915_gem_object *obj;
 	struct file_stats stats;
+	struct intel_engine_cs *ring;
+	int i, j;
 
 	memset(&stats, 0, sizeof(stats));
 
-	list_for_each_entry(obj,
-			    &dev_priv->mm.batch_pool.cache_list,
-			    batch_pool_list)
-		per_file_stats(0, obj, &stats);
+	for_each_ring(ring, dev_priv, i) {
+		for (j = 0; j < ARRAY_SIZE(ring->batch_pool.cache_list); j++) {
+			list_for_each_entry(obj,
+					    &ring->batch_pool.cache_list[j],
+					    batch_pool_link)
+				per_file_stats(0, obj, &stats);
+		}
+	}
 
-	print_file_stats(m, "batch pool", stats);
+	print_file_stats(m, "[k]batch pool", stats);
 }
 
 #define count_vmas(list, member) do { \
@@ -449,7 +464,7 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 			size += i915_gem_obj_ggtt_size(obj);
 			++count;
 		}
-		if (obj->pin_mappable) {
+		if (obj->pin_display) {
 			mappable_size += i915_gem_obj_ggtt_size(obj);
 			++mappable_count;
 		}
@@ -471,8 +486,6 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 
 	seq_putc(m, '\n');
 	print_batch_pool_stats(m, dev_priv);
-
-	seq_putc(m, '\n');
 	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
 		struct file_stats stats;
 		struct task_struct *task;
@@ -613,24 +626,39 @@ static int i915_gem_batch_pool_info(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
-	int count = 0;
-	int ret;
+	struct intel_engine_cs *ring;
+	int total = 0;
+	int ret, i, j;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
 		return ret;
 
-	seq_puts(m, "cache:\n");
-	list_for_each_entry(obj,
-			    &dev_priv->mm.batch_pool.cache_list,
-			    batch_pool_list) {
-		seq_puts(m, "   ");
-		describe_obj(m, obj);
-		seq_putc(m, '\n');
-		count++;
+	for_each_ring(ring, dev_priv, i) {
+		for (j = 0; j < ARRAY_SIZE(ring->batch_pool.cache_list); j++) {
+			int count;
+
+			count = 0;
+			list_for_each_entry(obj,
+					    &ring->batch_pool.cache_list[j],
+					    batch_pool_link)
+				count++;
+			seq_printf(m, "%s cache[%d]: %d objects\n",
+				   ring->name, j, count);
+
+			list_for_each_entry(obj,
+					    &ring->batch_pool.cache_list[j],
+					    batch_pool_link) {
+				seq_puts(m, "   ");
+				describe_obj(m, obj);
+				seq_putc(m, '\n');
+			}
+
+			total += count;
+		}
 	}
 
-	seq_printf(m, "total: %d\n", count);
+	seq_printf(m, "total: %d\n", total);
 
 	mutex_unlock(&dev->struct_mutex);
 
@@ -643,31 +671,44 @@ static int i915_gem_request_info(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *ring;
-	struct drm_i915_gem_request *gem_request;
-	int ret, count, i;
+	struct drm_i915_gem_request *req;
+	int ret, any, i;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
 		return ret;
 
-	count = 0;
+	any = 0;
 	for_each_ring(ring, dev_priv, i) {
-		if (list_empty(&ring->request_list))
+		int count;
+
+		count = 0;
+		list_for_each_entry(req, &ring->request_list, list)
+			count++;
+		if (count == 0)
 			continue;
 
-		seq_printf(m, "%s requests:\n", ring->name);
-		list_for_each_entry(gem_request,
-				    &ring->request_list,
-				    list) {
-			seq_printf(m, "    %x @ %d\n",
-				   gem_request->seqno,
-				   (int) (jiffies - gem_request->emitted_jiffies));
+		seq_printf(m, "%s requests: %d\n", ring->name, count);
+		list_for_each_entry(req, &ring->request_list, list) {
+			struct task_struct *task;
+
+			rcu_read_lock();
+			task = NULL;
+			if (req->pid)
+				task = pid_task(req->pid, PIDTYPE_PID);
+			seq_printf(m, "    %x @ %d: %s [%d]\n",
+				   req->seqno,
+				   (int) (jiffies - req->emitted_jiffies),
+				   task ? task->comm : "<unknown>",
+				   task ? task->pid : -1);
+			rcu_read_unlock();
 		}
-		count++;
+
+		any++;
 	}
 	mutex_unlock(&dev->struct_mutex);
 
-	if (count == 0)
+	if (any == 0)
 		seq_puts(m, "No requests\n");
 
 	return 0;
@@ -1176,12 +1217,17 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 			   GEN6_CURBSYTAVG_MASK);
 		seq_printf(m, "RP PREV UP: %dus\n", rpprevup &
 			   GEN6_CURBSYTAVG_MASK);
+		seq_printf(m, "Up threshold: %d%%\n",
+			   dev_priv->rps.up_threshold);
+
 		seq_printf(m, "RP CUR DOWN EI: %dus\n", rpdownei &
 			   GEN6_CURIAVG_MASK);
 		seq_printf(m, "RP CUR DOWN: %dus\n", rpcurdown &
 			   GEN6_CURBSYTAVG_MASK);
 		seq_printf(m, "RP PREV DOWN: %dus\n", rpprevdown &
 			   GEN6_CURBSYTAVG_MASK);
+		seq_printf(m, "Down threshold: %d%%\n",
+			   dev_priv->rps.down_threshold);
 
 		max_freq = (rp_state_cap & 0xff0000) >> 16;
 		max_freq *= (IS_SKYLAKE(dev) ? GEN9_FREQ_SCALER : 1);
@@ -1197,12 +1243,21 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		max_freq *= (IS_SKYLAKE(dev) ? GEN9_FREQ_SCALER : 1);
 		seq_printf(m, "Max non-overclocked (RP0) frequency: %dMHz\n",
 			   intel_gpu_freq(dev_priv, max_freq));
-
 		seq_printf(m, "Max overclocked frequency: %dMHz\n",
 			   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq));
 
+		seq_printf(m, "Current freq: %d MHz\n",
+			   intel_gpu_freq(dev_priv, dev_priv->rps.cur_freq));
+		seq_printf(m, "Actual freq: %d MHz\n", cagf);
 		seq_printf(m, "Idle freq: %d MHz\n",
 			   intel_gpu_freq(dev_priv, dev_priv->rps.idle_freq));
+		seq_printf(m, "Min freq: %d MHz\n",
+			   intel_gpu_freq(dev_priv, dev_priv->rps.min_freq));
+		seq_printf(m, "Max freq: %d MHz\n",
+			   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq));
+		seq_printf(m,
+			   "efficient (RPe) frequency: %d MHz\n",
+			   intel_gpu_freq(dev_priv, dev_priv->rps.efficient_freq));
 	} else if (IS_VALLEYVIEW(dev)) {
 		u32 freq_sts;
 
@@ -1211,6 +1266,12 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		seq_printf(m, "PUNIT_REG_GPU_FREQ_STS: 0x%08x\n", freq_sts);
 		seq_printf(m, "DDR freq: %d MHz\n", dev_priv->mem_freq);
 
+		seq_printf(m, "actual GPU freq: %d MHz\n",
+			   intel_gpu_freq(dev_priv, (freq_sts >> 8) & 0xff));
+
+		seq_printf(m, "current GPU freq: %d MHz\n",
+			   intel_gpu_freq(dev_priv, dev_priv->rps.cur_freq));
+
 		seq_printf(m, "max GPU freq: %d MHz\n",
 			   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq));
 
@@ -1223,9 +1284,6 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		seq_printf(m,
 			   "efficient (RPe) frequency: %d MHz\n",
 			   intel_gpu_freq(dev_priv, dev_priv->rps.efficient_freq));
-
-		seq_printf(m, "current GPU freq: %d MHz\n",
-			   intel_gpu_freq(dev_priv, (freq_sts >> 8) & 0xff));
 		mutex_unlock(&dev_priv->rps.hw_lock);
 	} else {
 		seq_puts(m, "no P-state info available\n");
@@ -2156,8 +2214,6 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 	if (!ppgtt)
 		return;
 
-	seq_printf(m, "Page directories: %d\n", ppgtt->num_pd_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++) {
@@ -2229,6 +2285,60 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 	return 0;
 }
 
+static int count_irq_waiters(struct drm_i915_private *i915)
+{
+	struct intel_engine_cs *ring;
+	int count = 0;
+	int i;
+
+	for_each_ring(ring, i915, i)
+		count += ring->irq_refcount;
+
+	return count;
+}
+
+static int i915_rps_boost_info(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_file *file;
+
+	seq_printf(m, "RPS enabled? %d\n", dev_priv->rps.enabled);
+	seq_printf(m, "GPU busy? %d\n", dev_priv->mm.busy);
+	seq_printf(m, "CPU waiting? %d\n", count_irq_waiters(dev_priv));
+	seq_printf(m, "Frequency requested %d; min hard:%d, soft:%d; max soft:%d, hard:%d\n",
+		   intel_gpu_freq(dev_priv, dev_priv->rps.cur_freq),
+		   intel_gpu_freq(dev_priv, dev_priv->rps.min_freq),
+		   intel_gpu_freq(dev_priv, dev_priv->rps.min_freq_softlimit),
+		   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq_softlimit),
+		   intel_gpu_freq(dev_priv, dev_priv->rps.max_freq));
+	spin_lock(&dev_priv->rps.client_lock);
+	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
+		struct drm_i915_file_private *file_priv = file->driver_priv;
+		struct task_struct *task;
+
+		rcu_read_lock();
+		task = pid_task(file->pid, PIDTYPE_PID);
+		seq_printf(m, "%s [%d]: %d boosts%s\n",
+			   task ? task->comm : "<unknown>",
+			   task ? task->pid : -1,
+			   file_priv->rps.boosts,
+			   list_empty(&file_priv->rps.link) ? "" : ", active");
+		rcu_read_unlock();
+	}
+	seq_printf(m, "Semaphore boosts: %d%s\n",
+		   dev_priv->rps.semaphores.boosts,
+		   list_empty(&dev_priv->rps.semaphores.link) ? "" : ", active");
+	seq_printf(m, "MMIO flip boosts: %d%s\n",
+		   dev_priv->rps.mmioflips.boosts,
+		   list_empty(&dev_priv->rps.mmioflips.link) ? "" : ", active");
+	seq_printf(m, "Kernel boosts: %d\n", dev_priv->rps.boosts);
+	spin_unlock(&dev_priv->rps.client_lock);
+
+	return 0;
+}
+
 static int i915_llc(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = m->private;
@@ -2290,9 +2400,6 @@ static int i915_edp_psr_status(struct seq_file *m, void *data)
 		}
 	seq_puts(m, "\n");
 
-	seq_printf(m, "Link standby: %s\n",
-		   yesno((bool)dev_priv->psr.link_standby));
-
 	/* CHV PSR has no kind of performance counter */
 	if (HAS_DDI(dev)) {
 		psrperf = I915_READ(EDP_PSR_PERF_CNT(dev)) &
@@ -3529,8 +3636,7 @@ static void hsw_trans_edp_pipe_A_crc_wa(struct drm_device *dev)
 		intel_display_power_get(dev_priv,
 					POWER_DOMAIN_PIPE_PANEL_FITTER(PIPE_A));
 
-		dev_priv->display.crtc_disable(&crtc->base);
-		dev_priv->display.crtc_enable(&crtc->base);
+		intel_crtc_reset(crtc);
 	}
 	drm_modeset_unlock_all(dev);
 }
@@ -3551,8 +3657,7 @@ static void hsw_undo_trans_edp_pipe_A_crc_wa(struct drm_device *dev)
 	if (crtc->config->pch_pfit.force_thru) {
 		crtc->config->pch_pfit.force_thru = false;
 
-		dev_priv->display.crtc_disable(&crtc->base);
-		dev_priv->display.crtc_enable(&crtc->base);
+		intel_crtc_reset(crtc);
 
 		intel_display_power_put(dev_priv,
 					POWER_DOMAIN_PIPE_PANEL_FITTER(PIPE_A));
@@ -3869,6 +3974,212 @@ static const struct file_operations i915_display_crc_ctl_fops = {
 	.write = display_crc_ctl_write
 };
 
+static ssize_t i915_displayport_test_active_write(struct file *file,
+					    const char __user *ubuf,
+					    size_t len, loff_t *offp)
+{
+	char *input_buffer;
+	int status = 0;
+	struct seq_file *m;
+	struct drm_device *dev;
+	struct drm_connector *connector;
+	struct list_head *connector_list;
+	struct intel_dp *intel_dp;
+	int val = 0;
+
+	m = file->private_data;
+	if (!m) {
+		status = -ENODEV;
+		return status;
+	}
+	dev = m->private;
+
+	if (!dev) {
+		status = -ENODEV;
+		return status;
+	}
+	connector_list = &dev->mode_config.connector_list;
+
+	if (len == 0)
+		return 0;
+
+	input_buffer = kmalloc(len + 1, GFP_KERNEL);
+	if (!input_buffer)
+		return -ENOMEM;
+
+	if (copy_from_user(input_buffer, ubuf, len)) {
+		status = -EFAULT;
+		goto out;
+	}
+
+	input_buffer[len] = '\0';
+	DRM_DEBUG_DRIVER("Copied %d bytes from user\n", (unsigned int)len);
+
+	list_for_each_entry(connector, connector_list, head) {
+
+		if (connector->connector_type !=
+		    DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->connector_type ==
+		    DRM_MODE_CONNECTOR_DisplayPort &&
+		    connector->status == connector_status_connected &&
+		    connector->encoder != NULL) {
+			intel_dp = enc_to_intel_dp(connector->encoder);
+			status = kstrtoint(input_buffer, 10, &val);
+			if (status < 0)
+				goto out;
+			DRM_DEBUG_DRIVER("Got %d for test active\n", val);
+			/* To prevent erroneous activation of the compliance
+			 * testing code, only accept an actual value of 1 here
+			 */
+			if (val == 1)
+				intel_dp->compliance_test_active = 1;
+			else
+				intel_dp->compliance_test_active = 0;
+		}
+	}
+out:
+	kfree(input_buffer);
+	if (status < 0)
+		return status;
+
+	*offp += len;
+	return len;
+}
+
+static int i915_displayport_test_active_show(struct seq_file *m, void *data)
+{
+	struct drm_device *dev = m->private;
+	struct drm_connector *connector;
+	struct list_head *connector_list = &dev->mode_config.connector_list;
+	struct intel_dp *intel_dp;
+
+	if (!dev)
+		return -ENODEV;
+
+	list_for_each_entry(connector, connector_list, head) {
+
+		if (connector->connector_type !=
+		    DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->status == connector_status_connected &&
+		    connector->encoder != NULL) {
+			intel_dp = enc_to_intel_dp(connector->encoder);
+			if (intel_dp->compliance_test_active)
+				seq_puts(m, "1");
+			else
+				seq_puts(m, "0");
+		} else
+			seq_puts(m, "0");
+	}
+
+	return 0;
+}
+
+static int i915_displayport_test_active_open(struct inode *inode,
+				       struct file *file)
+{
+	struct drm_device *dev = inode->i_private;
+
+	return single_open(file, i915_displayport_test_active_show, dev);
+}
+
+static const struct file_operations i915_displayport_test_active_fops = {
+	.owner = THIS_MODULE,
+	.open = i915_displayport_test_active_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+	.write = i915_displayport_test_active_write
+};
+
+static int i915_displayport_test_data_show(struct seq_file *m, void *data)
+{
+	struct drm_device *dev = m->private;
+	struct drm_connector *connector;
+	struct list_head *connector_list = &dev->mode_config.connector_list;
+	struct intel_dp *intel_dp;
+
+	if (!dev)
+		return -ENODEV;
+
+	list_for_each_entry(connector, connector_list, head) {
+
+		if (connector->connector_type !=
+		    DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->status == connector_status_connected &&
+		    connector->encoder != NULL) {
+			intel_dp = enc_to_intel_dp(connector->encoder);
+			seq_printf(m, "%lx", intel_dp->compliance_test_data);
+		} else
+			seq_puts(m, "0");
+	}
+
+	return 0;
+}
+static int i915_displayport_test_data_open(struct inode *inode,
+				       struct file *file)
+{
+	struct drm_device *dev = inode->i_private;
+
+	return single_open(file, i915_displayport_test_data_show, dev);
+}
+
+static const struct file_operations i915_displayport_test_data_fops = {
+	.owner = THIS_MODULE,
+	.open = i915_displayport_test_data_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release
+};
+
+static int i915_displayport_test_type_show(struct seq_file *m, void *data)
+{
+	struct drm_device *dev = m->private;
+	struct drm_connector *connector;
+	struct list_head *connector_list = &dev->mode_config.connector_list;
+	struct intel_dp *intel_dp;
+
+	if (!dev)
+		return -ENODEV;
+
+	list_for_each_entry(connector, connector_list, head) {
+
+		if (connector->connector_type !=
+		    DRM_MODE_CONNECTOR_DisplayPort)
+			continue;
+
+		if (connector->status == connector_status_connected &&
+		    connector->encoder != NULL) {
+			intel_dp = enc_to_intel_dp(connector->encoder);
+			seq_printf(m, "%02lx", intel_dp->compliance_test_type);
+		} else
+			seq_puts(m, "0");
+	}
+
+	return 0;
+}
+
+static int i915_displayport_test_type_open(struct inode *inode,
+				       struct file *file)
+{
+	struct drm_device *dev = inode->i_private;
+
+	return single_open(file, i915_displayport_test_type_show, dev);
+}
+
+static const struct file_operations i915_displayport_test_type_fops = {
+	.owner = THIS_MODULE,
+	.open = i915_displayport_test_type_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release
+};
+
 static void wm_latency_show(struct seq_file *m, const uint16_t wm[8])
 {
 	struct drm_device *dev = m->private;
@@ -4473,12 +4784,116 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_cache_sharing_fops,
 			i915_cache_sharing_get, i915_cache_sharing_set,
 			"%llu\n");
 
+struct sseu_dev_status {
+	unsigned int slice_total;
+	unsigned int subslice_total;
+	unsigned int subslice_per_slice;
+	unsigned int eu_total;
+	unsigned int eu_per_subslice;
+};
+
+static void cherryview_sseu_device_status(struct drm_device *dev,
+					  struct sseu_dev_status *stat)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	const int ss_max = 2;
+	int ss;
+	u32 sig1[ss_max], sig2[ss_max];
+
+	sig1[0] = I915_READ(CHV_POWER_SS0_SIG1);
+	sig1[1] = I915_READ(CHV_POWER_SS1_SIG1);
+	sig2[0] = I915_READ(CHV_POWER_SS0_SIG2);
+	sig2[1] = I915_READ(CHV_POWER_SS1_SIG2);
+
+	for (ss = 0; ss < ss_max; ss++) {
+		unsigned int eu_cnt;
+
+		if (sig1[ss] & CHV_SS_PG_ENABLE)
+			/* skip disabled subslice */
+			continue;
+
+		stat->slice_total = 1;
+		stat->subslice_per_slice++;
+		eu_cnt = ((sig1[ss] & CHV_EU08_PG_ENABLE) ? 0 : 2) +
+			 ((sig1[ss] & CHV_EU19_PG_ENABLE) ? 0 : 2) +
+			 ((sig1[ss] & CHV_EU210_PG_ENABLE) ? 0 : 2) +
+			 ((sig2[ss] & CHV_EU311_PG_ENABLE) ? 0 : 2);
+		stat->eu_total += eu_cnt;
+		stat->eu_per_subslice = max(stat->eu_per_subslice, eu_cnt);
+	}
+	stat->subslice_total = stat->subslice_per_slice;
+}
+
+static void gen9_sseu_device_status(struct drm_device *dev,
+				    struct sseu_dev_status *stat)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int s_max = 3, ss_max = 4;
+	int s, ss;
+	u32 s_reg[s_max], eu_reg[2*s_max], eu_mask[2];
+
+	/* BXT has a single slice and at most 3 subslices. */
+	if (IS_BROXTON(dev)) {
+		s_max = 1;
+		ss_max = 3;
+	}
+
+	for (s = 0; s < s_max; s++) {
+		s_reg[s] = I915_READ(GEN9_SLICE_PGCTL_ACK(s));
+		eu_reg[2*s] = I915_READ(GEN9_SS01_EU_PGCTL_ACK(s));
+		eu_reg[2*s + 1] = I915_READ(GEN9_SS23_EU_PGCTL_ACK(s));
+	}
+
+	eu_mask[0] = GEN9_PGCTL_SSA_EU08_ACK |
+		     GEN9_PGCTL_SSA_EU19_ACK |
+		     GEN9_PGCTL_SSA_EU210_ACK |
+		     GEN9_PGCTL_SSA_EU311_ACK;
+	eu_mask[1] = GEN9_PGCTL_SSB_EU08_ACK |
+		     GEN9_PGCTL_SSB_EU19_ACK |
+		     GEN9_PGCTL_SSB_EU210_ACK |
+		     GEN9_PGCTL_SSB_EU311_ACK;
+
+	for (s = 0; s < s_max; s++) {
+		unsigned int ss_cnt = 0;
+
+		if ((s_reg[s] & GEN9_PGCTL_SLICE_ACK) == 0)
+			/* skip disabled slice */
+			continue;
+
+		stat->slice_total++;
+
+		if (IS_SKYLAKE(dev))
+			ss_cnt = INTEL_INFO(dev)->subslice_per_slice;
+
+		for (ss = 0; ss < ss_max; ss++) {
+			unsigned int eu_cnt;
+
+			if (IS_BROXTON(dev) &&
+			    !(s_reg[s] & (GEN9_PGCTL_SS_ACK(ss))))
+				/* skip disabled subslice */
+				continue;
+
+			if (IS_BROXTON(dev))
+				ss_cnt++;
+
+			eu_cnt = 2 * hweight32(eu_reg[2*s + ss/2] &
+					       eu_mask[ss%2]);
+			stat->eu_total += eu_cnt;
+			stat->eu_per_subslice = max(stat->eu_per_subslice,
+						    eu_cnt);
+		}
+
+		stat->subslice_total += ss_cnt;
+		stat->subslice_per_slice = max(stat->subslice_per_slice,
+					       ss_cnt);
+	}
+}
+
 static int i915_sseu_status(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned int s_tot = 0, ss_tot = 0, ss_per = 0, eu_tot = 0, eu_per = 0;
+	struct sseu_dev_status stat;
 
 	if ((INTEL_INFO(dev)->gen < 8) || IS_BROADWELL(dev))
 		return -ENODEV;
@@ -4502,79 +4917,22 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
 		   yesno(INTEL_INFO(dev)->has_eu_pg));
 
 	seq_puts(m, "SSEU Device Status\n");
+	memset(&stat, 0, sizeof(stat));
 	if (IS_CHERRYVIEW(dev)) {
-		const int ss_max = 2;
-		int ss;
-		u32 sig1[ss_max], sig2[ss_max];
-
-		sig1[0] = I915_READ(CHV_POWER_SS0_SIG1);
-		sig1[1] = I915_READ(CHV_POWER_SS1_SIG1);
-		sig2[0] = I915_READ(CHV_POWER_SS0_SIG2);
-		sig2[1] = I915_READ(CHV_POWER_SS1_SIG2);
-
-		for (ss = 0; ss < ss_max; ss++) {
-			unsigned int eu_cnt;
-
-			if (sig1[ss] & CHV_SS_PG_ENABLE)
-				/* skip disabled subslice */
-				continue;
-
-			s_tot = 1;
-			ss_per++;
-			eu_cnt = ((sig1[ss] & CHV_EU08_PG_ENABLE) ? 0 : 2) +
-				 ((sig1[ss] & CHV_EU19_PG_ENABLE) ? 0 : 2) +
-				 ((sig1[ss] & CHV_EU210_PG_ENABLE) ? 0 : 2) +
-				 ((sig2[ss] & CHV_EU311_PG_ENABLE) ? 0 : 2);
-			eu_tot += eu_cnt;
-			eu_per = max(eu_per, eu_cnt);
-		}
-		ss_tot = ss_per;
-	} else if (IS_SKYLAKE(dev)) {
-		const int s_max = 3, ss_max = 4;
-		int s, ss;
-		u32 s_reg[s_max], eu_reg[2*s_max], eu_mask[2];
-
-		s_reg[0] = I915_READ(GEN9_SLICE0_PGCTL_ACK);
-		s_reg[1] = I915_READ(GEN9_SLICE1_PGCTL_ACK);
-		s_reg[2] = I915_READ(GEN9_SLICE2_PGCTL_ACK);
-		eu_reg[0] = I915_READ(GEN9_SLICE0_SS01_EU_PGCTL_ACK);
-		eu_reg[1] = I915_READ(GEN9_SLICE0_SS23_EU_PGCTL_ACK);
-		eu_reg[2] = I915_READ(GEN9_SLICE1_SS01_EU_PGCTL_ACK);
-		eu_reg[3] = I915_READ(GEN9_SLICE1_SS23_EU_PGCTL_ACK);
-		eu_reg[4] = I915_READ(GEN9_SLICE2_SS01_EU_PGCTL_ACK);
-		eu_reg[5] = I915_READ(GEN9_SLICE2_SS23_EU_PGCTL_ACK);
-		eu_mask[0] = GEN9_PGCTL_SSA_EU08_ACK |
-			     GEN9_PGCTL_SSA_EU19_ACK |
-			     GEN9_PGCTL_SSA_EU210_ACK |
-			     GEN9_PGCTL_SSA_EU311_ACK;
-		eu_mask[1] = GEN9_PGCTL_SSB_EU08_ACK |
-			     GEN9_PGCTL_SSB_EU19_ACK |
-			     GEN9_PGCTL_SSB_EU210_ACK |
-			     GEN9_PGCTL_SSB_EU311_ACK;
-
-		for (s = 0; s < s_max; s++) {
-			if ((s_reg[s] & GEN9_PGCTL_SLICE_ACK) == 0)
-				/* skip disabled slice */
-				continue;
-
-			s_tot++;
-			ss_per = INTEL_INFO(dev)->subslice_per_slice;
-			ss_tot += ss_per;
-			for (ss = 0; ss < ss_max; ss++) {
-				unsigned int eu_cnt;
-
-				eu_cnt = 2 * hweight32(eu_reg[2*s + ss/2] &
-						       eu_mask[ss%2]);
-				eu_tot += eu_cnt;
-				eu_per = max(eu_per, eu_cnt);
-			}
-		}
+		cherryview_sseu_device_status(dev, &stat);
+	} else if (INTEL_INFO(dev)->gen >= 9) {
+		gen9_sseu_device_status(dev, &stat);
 	}
-	seq_printf(m, "  Enabled Slice Total: %u\n", s_tot);
-	seq_printf(m, "  Enabled Subslice Total: %u\n", ss_tot);
-	seq_printf(m, "  Enabled Subslice Per Slice: %u\n", ss_per);
-	seq_printf(m, "  Enabled EU Total: %u\n", eu_tot);
-	seq_printf(m, "  Enabled EU Per Subslice: %u\n", eu_per);
+	seq_printf(m, "  Enabled Slice Total: %u\n",
+		   stat.slice_total);
+	seq_printf(m, "  Enabled Subslice Total: %u\n",
+		   stat.subslice_total);
+	seq_printf(m, "  Enabled Subslice Per Slice: %u\n",
+		   stat.subslice_per_slice);
+	seq_printf(m, "  Enabled EU Total: %u\n",
+		   stat.eu_total);
+	seq_printf(m, "  Enabled EU Per Subslice: %u\n",
+		   stat.eu_per_subslice);
 
 	return 0;
 }
@@ -4694,6 +5052,7 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_ddb_info", i915_ddb_info, 0},
 	{"i915_sseu_status", i915_sseu_status, 0},
 	{"i915_drrs_status", i915_drrs_status, 0},
+	{"i915_rps_boost_info", i915_rps_boost_info, 0},
 };
 #define I915_DEBUGFS_ENTRIES ARRAY_SIZE(i915_debugfs_list)
 
@@ -4716,6 +5075,9 @@ static const struct i915_debugfs_files {
 	{"i915_spr_wm_latency", &i915_spr_wm_latency_fops},
 	{"i915_cur_wm_latency", &i915_cur_wm_latency_fops},
 	{"i915_fbc_false_color", &i915_fbc_fc_fops},
+	{"i915_dp_test_data", &i915_displayport_test_data_fops},
+	{"i915_dp_test_type", &i915_displayport_test_type_fops},
+	{"i915_dp_test_active", &i915_displayport_test_active_fops}
 };
 
 void intel_display_crc_init(struct drm_device *dev)
@@ -4783,3 +5145,102 @@ void i915_debugfs_cleanup(struct drm_minor *minor)
 		drm_debugfs_remove_files(info_list, 1, minor);
 	}
 }
+
+struct dpcd_block {
+	/* DPCD dump start address. */
+	unsigned int offset;
+	/* DPCD dump end address, inclusive. If unset, .size will be used. */
+	unsigned int end;
+	/* DPCD dump size. Used if .end is unset. If unset, defaults to 1. */
+	size_t size;
+	/* Only valid for eDP. */
+	bool edp;
+};
+
+static const struct dpcd_block i915_dpcd_debug[] = {
+	{ .offset = DP_DPCD_REV, .size = DP_RECEIVER_CAP_SIZE },
+	{ .offset = DP_PSR_SUPPORT, .end = DP_PSR_CAPS },
+	{ .offset = DP_DOWNSTREAM_PORT_0, .size = 16 },
+	{ .offset = DP_LINK_BW_SET, .end = DP_EDP_CONFIGURATION_SET },
+	{ .offset = DP_SINK_COUNT, .end = DP_ADJUST_REQUEST_LANE2_3 },
+	{ .offset = DP_SET_POWER },
+	{ .offset = DP_EDP_DPCD_REV },
+	{ .offset = DP_EDP_GENERAL_CAP_1, .end = DP_EDP_GENERAL_CAP_3 },
+	{ .offset = DP_EDP_DISPLAY_CONTROL_REGISTER, .end = DP_EDP_BACKLIGHT_FREQ_CAP_MAX_LSB },
+	{ .offset = DP_EDP_DBC_MINIMUM_BRIGHTNESS_SET, .end = DP_EDP_DBC_MAXIMUM_BRIGHTNESS_SET },
+};
+
+static int i915_dpcd_show(struct seq_file *m, void *data)
+{
+	struct drm_connector *connector = m->private;
+	struct intel_dp *intel_dp =
+		enc_to_intel_dp(&intel_attached_encoder(connector)->base);
+	uint8_t buf[16];
+	ssize_t err;
+	int i;
+
+	if (connector->status != connector_status_connected)
+		return -ENODEV;
+
+	for (i = 0; i < ARRAY_SIZE(i915_dpcd_debug); i++) {
+		const struct dpcd_block *b = &i915_dpcd_debug[i];
+		size_t size = b->end ? b->end - b->offset + 1 : (b->size ?: 1);
+
+		if (b->edp &&
+		    connector->connector_type != DRM_MODE_CONNECTOR_eDP)
+			continue;
+
+		/* low tech for now */
+		if (WARN_ON(size > sizeof(buf)))
+			continue;
+
+		err = drm_dp_dpcd_read(&intel_dp->aux, b->offset, buf, size);
+		if (err <= 0) {
+			DRM_ERROR("dpcd read (%zu bytes at %u) failed (%zd)\n",
+				  size, b->offset, err);
+			continue;
+		}
+
+		seq_printf(m, "%04x: %*ph\n", b->offset, (int) size, buf);
+	}
+
+	return 0;
+}
+
+static int i915_dpcd_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, i915_dpcd_show, inode->i_private);
+}
+
+static const struct file_operations i915_dpcd_fops = {
+	.owner = THIS_MODULE,
+	.open = i915_dpcd_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+/**
+ * i915_debugfs_connector_add - add i915 specific connector debugfs files
+ * @connector: pointer to a registered drm_connector
+ *
+ * Cleanup will be done by drm_connector_unregister() through a call to
+ * drm_debugfs_connector_remove().
+ *
+ * Returns 0 on success, negative error codes on error.
+ */
+int i915_debugfs_connector_add(struct drm_connector *connector)
+{
+	struct dentry *root = connector->debugfs_entry;
+
+	/* The connector must have been registered beforehands. */
+	if (!root)
+		return -ENODEV;
+
+	if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort ||
+	    connector->connector_type == DRM_MODE_CONNECTOR_eDP)
+		debugfs_create_file("i915_dpcd", S_IRUGO, root, connector,
+				    &i915_dpcd_fops);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 68e0c85a17cf..d2df321ba634 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -564,6 +564,140 @@ static void i915_dump_device_info(struct drm_i915_private *dev_priv)
 #undef SEP_COMMA
 }
 
+static void cherryview_sseu_info_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_device_info *info;
+	u32 fuse, eu_dis;
+
+	info = (struct intel_device_info *)&dev_priv->info;
+	fuse = I915_READ(CHV_FUSE_GT);
+
+	info->slice_total = 1;
+
+	if (!(fuse & CHV_FGT_DISABLE_SS0)) {
+		info->subslice_per_slice++;
+		eu_dis = fuse & (CHV_FGT_EU_DIS_SS0_R0_MASK |
+				 CHV_FGT_EU_DIS_SS0_R1_MASK);
+		info->eu_total += 8 - hweight32(eu_dis);
+	}
+
+	if (!(fuse & CHV_FGT_DISABLE_SS1)) {
+		info->subslice_per_slice++;
+		eu_dis = fuse & (CHV_FGT_EU_DIS_SS1_R0_MASK |
+				 CHV_FGT_EU_DIS_SS1_R1_MASK);
+		info->eu_total += 8 - hweight32(eu_dis);
+	}
+
+	info->subslice_total = info->subslice_per_slice;
+	/*
+	 * CHV expected to always have a uniform distribution of EU
+	 * across subslices.
+	*/
+	info->eu_per_subslice = info->subslice_total ?
+				info->eu_total / info->subslice_total :
+				0;
+	/*
+	 * CHV supports subslice power gating on devices with more than
+	 * one subslice, and supports EU power gating on devices with
+	 * more than one EU pair per subslice.
+	*/
+	info->has_slice_pg = 0;
+	info->has_subslice_pg = (info->subslice_total > 1);
+	info->has_eu_pg = (info->eu_per_subslice > 2);
+}
+
+static void gen9_sseu_info_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_device_info *info;
+	int s_max = 3, ss_max = 4, eu_max = 8;
+	int s, ss;
+	u32 fuse2, s_enable, ss_disable, eu_disable;
+	u8 eu_mask = 0xff;
+
+	/*
+	 * BXT has a single slice. BXT also has at most 6 EU per subslice,
+	 * and therefore only the lowest 6 bits of the 8-bit EU disable
+	 * fields are valid.
+	*/
+	if (IS_BROXTON(dev)) {
+		s_max = 1;
+		eu_max = 6;
+		eu_mask = 0x3f;
+	}
+
+	info = (struct intel_device_info *)&dev_priv->info;
+	fuse2 = I915_READ(GEN8_FUSE2);
+	s_enable = (fuse2 & GEN8_F2_S_ENA_MASK) >>
+		   GEN8_F2_S_ENA_SHIFT;
+	ss_disable = (fuse2 & GEN9_F2_SS_DIS_MASK) >>
+		     GEN9_F2_SS_DIS_SHIFT;
+
+	info->slice_total = hweight32(s_enable);
+	/*
+	 * The subslice disable field is global, i.e. it applies
+	 * to each of the enabled slices.
+	*/
+	info->subslice_per_slice = ss_max - hweight32(ss_disable);
+	info->subslice_total = info->slice_total *
+			       info->subslice_per_slice;
+
+	/*
+	 * Iterate through enabled slices and subslices to
+	 * count the total enabled EU.
+	*/
+	for (s = 0; s < s_max; s++) {
+		if (!(s_enable & (0x1 << s)))
+			/* skip disabled slice */
+			continue;
+
+		eu_disable = I915_READ(GEN9_EU_DISABLE(s));
+		for (ss = 0; ss < ss_max; ss++) {
+			int eu_per_ss;
+
+			if (ss_disable & (0x1 << ss))
+				/* skip disabled subslice */
+				continue;
+
+			eu_per_ss = eu_max - hweight8((eu_disable >> (ss*8)) &
+						      eu_mask);
+
+			/*
+			 * Record which subslice(s) has(have) 7 EUs. we
+			 * can tune the hash used to spread work among
+			 * subslices if they are unbalanced.
+			 */
+			if (eu_per_ss == 7)
+				info->subslice_7eu[s] |= 1 << ss;
+
+			info->eu_total += eu_per_ss;
+		}
+	}
+
+	/*
+	 * SKL is expected to always have a uniform distribution
+	 * of EU across subslices with the exception that any one
+	 * EU in any one subslice may be fused off for die
+	 * recovery. BXT is expected to be perfectly uniform in EU
+	 * distribution.
+	*/
+	info->eu_per_subslice = info->subslice_total ?
+				DIV_ROUND_UP(info->eu_total,
+					     info->subslice_total) : 0;
+	/*
+	 * SKL supports slice power gating on devices with more than
+	 * one slice, and supports EU power gating on devices with
+	 * more than one EU pair per subslice. BXT supports subslice
+	 * power gating on devices with more than one subslice, and
+	 * supports EU power gating on devices with more than one EU
+	 * pair per subslice.
+	*/
+	info->has_slice_pg = (IS_SKYLAKE(dev) && (info->slice_total > 1));
+	info->has_subslice_pg = (IS_BROXTON(dev) && (info->subslice_total > 1));
+	info->has_eu_pg = (info->eu_per_subslice > 2);
+}
+
 /*
  * Determine various intel_device_info fields at runtime.
  *
@@ -585,7 +719,11 @@ static void intel_device_info_runtime_init(struct drm_device *dev)
 
 	info = (struct intel_device_info *)&dev_priv->info;
 
-	if (IS_VALLEYVIEW(dev) || INTEL_INFO(dev)->gen == 9)
+	if (IS_BROXTON(dev)) {
+		info->num_sprites[PIPE_A] = 3;
+		info->num_sprites[PIPE_B] = 3;
+		info->num_sprites[PIPE_C] = 2;
+	} else if (IS_VALLEYVIEW(dev) || INTEL_INFO(dev)->gen == 9)
 		for_each_pipe(dev_priv, pipe)
 			info->num_sprites[pipe] = 2;
 	else
@@ -620,116 +758,11 @@ static void intel_device_info_runtime_init(struct drm_device *dev)
 	}
 
 	/* Initialize slice/subslice/EU info */
-	if (IS_CHERRYVIEW(dev)) {
-		u32 fuse, eu_dis;
-
-		fuse = I915_READ(CHV_FUSE_GT);
+	if (IS_CHERRYVIEW(dev))
+		cherryview_sseu_info_init(dev);
+	else if (INTEL_INFO(dev)->gen >= 9)
+		gen9_sseu_info_init(dev);
 
-		info->slice_total = 1;
-
-		if (!(fuse & CHV_FGT_DISABLE_SS0)) {
-			info->subslice_per_slice++;
-			eu_dis = fuse & (CHV_FGT_EU_DIS_SS0_R0_MASK |
-					 CHV_FGT_EU_DIS_SS0_R1_MASK);
-			info->eu_total += 8 - hweight32(eu_dis);
-		}
-
-		if (!(fuse & CHV_FGT_DISABLE_SS1)) {
-			info->subslice_per_slice++;
-			eu_dis = fuse & (CHV_FGT_EU_DIS_SS1_R0_MASK |
-					CHV_FGT_EU_DIS_SS1_R1_MASK);
-			info->eu_total += 8 - hweight32(eu_dis);
-		}
-
-		info->subslice_total = info->subslice_per_slice;
-		/*
-		 * CHV expected to always have a uniform distribution of EU
-		 * across subslices.
-		*/
-		info->eu_per_subslice = info->subslice_total ?
-					info->eu_total / info->subslice_total :
-					0;
-		/*
-		 * CHV supports subslice power gating on devices with more than
-		 * one subslice, and supports EU power gating on devices with
-		 * more than one EU pair per subslice.
-		*/
-		info->has_slice_pg = 0;
-		info->has_subslice_pg = (info->subslice_total > 1);
-		info->has_eu_pg = (info->eu_per_subslice > 2);
-	} else if (IS_SKYLAKE(dev)) {
-		const int s_max = 3, ss_max = 4, eu_max = 8;
-		int s, ss;
-		u32 fuse2, eu_disable[s_max], s_enable, ss_disable;
-
-		fuse2 = I915_READ(GEN8_FUSE2);
-		s_enable = (fuse2 & GEN8_F2_S_ENA_MASK) >>
-			   GEN8_F2_S_ENA_SHIFT;
-		ss_disable = (fuse2 & GEN9_F2_SS_DIS_MASK) >>
-			     GEN9_F2_SS_DIS_SHIFT;
-
-		eu_disable[0] = I915_READ(GEN8_EU_DISABLE0);
-		eu_disable[1] = I915_READ(GEN8_EU_DISABLE1);
-		eu_disable[2] = I915_READ(GEN8_EU_DISABLE2);
-
-		info->slice_total = hweight32(s_enable);
-		/*
-		 * The subslice disable field is global, i.e. it applies
-		 * to each of the enabled slices.
-		*/
-		info->subslice_per_slice = ss_max - hweight32(ss_disable);
-		info->subslice_total = info->slice_total *
-				       info->subslice_per_slice;
-
-		/*
-		 * Iterate through enabled slices and subslices to
-		 * count the total enabled EU.
-		*/
-		for (s = 0; s < s_max; s++) {
-			if (!(s_enable & (0x1 << s)))
-				/* skip disabled slice */
-				continue;
-
-			for (ss = 0; ss < ss_max; ss++) {
-				u32 n_disabled;
-
-				if (ss_disable & (0x1 << ss))
-					/* skip disabled subslice */
-					continue;
-
-				n_disabled = hweight8(eu_disable[s] >>
-						      (ss * eu_max));
-
-				/*
-				 * Record which subslice(s) has(have) 7 EUs. we
-				 * can tune the hash used to spread work among
-				 * subslices if they are unbalanced.
-				 */
-				if (eu_max - n_disabled == 7)
-					info->subslice_7eu[s] |= 1 << ss;
-
-				info->eu_total += eu_max - n_disabled;
-			}
-		}
-
-		/*
-		 * SKL is expected to always have a uniform distribution
-		 * of EU across subslices with the exception that any one
-		 * EU in any one subslice may be fused off for die
-		 * recovery.
-		*/
-		info->eu_per_subslice = info->subslice_total ?
-					DIV_ROUND_UP(info->eu_total,
-						     info->subslice_total) : 0;
-		/*
-		 * SKL supports slice power gating on devices with more than
-		 * one slice, and supports EU power gating on devices with
-		 * more than one EU pair per subslice.
-		*/
-		info->has_slice_pg = (info->slice_total > 1) ? 1 : 0;
-		info->has_subslice_pg = 0;
-		info->has_eu_pg = (info->eu_per_subslice > 2) ? 1 : 0;
-	}
 	DRM_DEBUG_DRIVER("slice total: %u\n", info->slice_total);
 	DRM_DEBUG_DRIVER("subslice total: %u\n", info->subslice_total);
 	DRM_DEBUG_DRIVER("subslice per slice: %u\n", info->subslice_per_slice);
@@ -781,8 +814,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	spin_lock_init(&dev_priv->uncore.lock);
 	spin_lock_init(&dev_priv->mm.object_stat_lock);
 	spin_lock_init(&dev_priv->mmio_flip_lock);
-	mutex_init(&dev_priv->dpio_lock);
+	mutex_init(&dev_priv->sb_lock);
 	mutex_init(&dev_priv->modeset_restore_lock);
+	mutex_init(&dev_priv->csr_lock);
 
 	intel_pm_setup(dev);
 
@@ -828,9 +862,12 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 
 	intel_uncore_init(dev);
 
+	/* Load CSR Firmware for SKL */
+	intel_csr_ucode_init(dev);
+
 	ret = i915_gem_gtt_init(dev);
 	if (ret)
-		goto out_regs;
+		goto out_freecsr;
 
 	/* WARNING: Apparently we must kick fbdev drivers before vgacon,
 	 * otherwise the vga fbdev driver falls over. */
@@ -1000,14 +1037,19 @@ out_mtrrfree:
 	io_mapping_free(dev_priv->gtt.mappable);
 out_gtt:
 	i915_global_gtt_cleanup(dev);
-out_regs:
+out_freecsr:
+	intel_csr_ucode_fini(dev);
 	intel_uncore_fini(dev);
 	pci_iounmap(dev->pdev, dev_priv->regs);
 put_bridge:
 	pci_dev_put(dev_priv->bridge_dev);
 free_priv:
-	if (dev_priv->slab)
-		kmem_cache_destroy(dev_priv->slab);
+	if (dev_priv->requests)
+		kmem_cache_destroy(dev_priv->requests);
+	if (dev_priv->vmas)
+		kmem_cache_destroy(dev_priv->vmas);
+	if (dev_priv->objects)
+		kmem_cache_destroy(dev_priv->objects);
 	kfree(dev_priv);
 	return ret;
 }
@@ -1072,11 +1114,12 @@ int i915_driver_unload(struct drm_device *dev)
 
 	mutex_lock(&dev->struct_mutex);
 	i915_gem_cleanup_ringbuffer(dev);
-	i915_gem_batch_pool_fini(&dev_priv->mm.batch_pool);
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
 	i915_gem_cleanup_stolen(dev);
 
+	intel_csr_ucode_fini(dev);
+
 	intel_teardown_gmbus(dev);
 	intel_teardown_mchbar(dev);
 
@@ -1091,8 +1134,12 @@ int i915_driver_unload(struct drm_device *dev)
 	if (dev_priv->regs != NULL)
 		pci_iounmap(dev->pdev, dev_priv->regs);
 
-	if (dev_priv->slab)
-		kmem_cache_destroy(dev_priv->slab);
+	if (dev_priv->requests)
+		kmem_cache_destroy(dev_priv->requests);
+	if (dev_priv->vmas)
+		kmem_cache_destroy(dev_priv->vmas);
+	if (dev_priv->objects)
+		kmem_cache_destroy(dev_priv->objects);
 
 	pci_dev_put(dev_priv->bridge_dev);
 	kfree(dev_priv);
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index a19d2c71e205..884b4f9b81c4 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -381,6 +381,18 @@ static const struct intel_device_info intel_skylake_gt3_info = {
 	IVB_CURSOR_OFFSETS,
 };
 
+static const struct intel_device_info intel_broxton_info = {
+	.is_preliminary = 1,
+	.gen = 9,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
+	.num_pipes = 3,
+	.has_ddi = 1,
+	.has_fbc = 1,
+	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
+};
+
 /*
  * Make sure any device matches here are from most specific to most
  * general.  For example, since the Quanta match is based on the subsystem
@@ -420,7 +432,8 @@ static const struct intel_device_info intel_skylake_gt3_info = {
 	INTEL_CHV_IDS(&intel_cherryview_info),	\
 	INTEL_SKL_GT1_IDS(&intel_skylake_info),	\
 	INTEL_SKL_GT2_IDS(&intel_skylake_info),	\
-	INTEL_SKL_GT3_IDS(&intel_skylake_gt3_info)	\
+	INTEL_SKL_GT3_IDS(&intel_skylake_gt3_info),	\
+	INTEL_BXT_IDS(&intel_broxton_info)
 
 static const struct pci_device_id pciidlist[] = {		/* aka */
 	INTEL_PCI_IDS,
@@ -543,6 +556,26 @@ void intel_hpd_cancel_work(struct drm_i915_private *dev_priv)
 	cancel_delayed_work_sync(&dev_priv->hotplug_reenable_work);
 }
 
+void i915_firmware_load_error_print(const char *fw_path, int err)
+{
+	DRM_ERROR("failed to load firmware %s (%d)\n", fw_path, err);
+
+	/*
+	 * If the reason is not known assume -ENOENT since that's the most
+	 * usual failure mode.
+	 */
+	if (!err)
+		err = -ENOENT;
+
+	if (!(IS_BUILTIN(CONFIG_DRM_I915) && err == -ENOENT))
+		return;
+
+	DRM_ERROR(
+	  "The driver is built-in, so to load the firmware you need to\n"
+	  "include it either in the kernel (see CONFIG_EXTRA_FIRMWARE) or\n"
+	  "in your initrd/initramfs image.\n");
+}
+
 static void intel_suspend_encoders(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
@@ -561,6 +594,9 @@ static void intel_suspend_encoders(struct drm_i915_private *dev_priv)
 static int intel_suspend_complete(struct drm_i915_private *dev_priv);
 static int vlv_resume_prepare(struct drm_i915_private *dev_priv,
 			      bool rpm_resume);
+static int skl_resume_prepare(struct drm_i915_private *dev_priv);
+static int bxt_resume_prepare(struct drm_i915_private *dev_priv);
+
 
 static int i915_drm_suspend(struct drm_device *dev)
 {
@@ -776,11 +812,16 @@ static int i915_drm_resume_early(struct drm_device *dev)
 	if (IS_VALLEYVIEW(dev_priv))
 		ret = vlv_resume_prepare(dev_priv, false);
 	if (ret)
-		DRM_ERROR("Resume prepare failed: %d,Continuing resume\n", ret);
+		DRM_ERROR("Resume prepare failed: %d, continuing anyway\n",
+			  ret);
 
 	intel_uncore_early_sanitize(dev, true);
 
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+	if (IS_BROXTON(dev))
+		ret = bxt_resume_prepare(dev_priv);
+	else if (IS_SKYLAKE(dev_priv))
+		ret = skl_resume_prepare(dev_priv);
+	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
 		hsw_disable_pc8(dev_priv);
 
 	intel_uncore_sanitize(dev);
@@ -952,7 +993,7 @@ static int i915_pm_suspend_late(struct device *dev)
 	struct drm_device *drm_dev = dev_to_i915(dev)->dev;
 
 	/*
-	 * We have a suspedn ordering issue with the snd-hda driver also
+	 * We have a suspend ordering issue with the snd-hda driver also
 	 * requiring our device to be power up. Due to the lack of a
 	 * parent/child relationship we currently solve this with an late
 	 * suspend hook.
@@ -996,6 +1037,21 @@ static int i915_pm_resume(struct device *dev)
 	return i915_drm_resume(drm_dev);
 }
 
+static int skl_suspend_complete(struct drm_i915_private *dev_priv)
+{
+	/* Enabling DC6 is not a hard requirement to enter runtime D3 */
+
+	/*
+	 * This is to ensure that CSR isn't identified as loaded before
+	 * CSR-loading program is called during runtime-resume.
+	 */
+	intel_csr_load_status_set(dev_priv, FW_UNINITIALIZED);
+
+	skl_uninit_cdclk(dev_priv);
+
+	return 0;
+}
+
 static int hsw_suspend_complete(struct drm_i915_private *dev_priv)
 {
 	hsw_enable_pc8(dev_priv);
@@ -1003,6 +1059,48 @@ static int hsw_suspend_complete(struct drm_i915_private *dev_priv)
 	return 0;
 }
 
+static int bxt_suspend_complete(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	/* TODO: when DC5 support is added disable DC5 here. */
+
+	broxton_ddi_phy_uninit(dev);
+	broxton_uninit_cdclk(dev);
+	bxt_enable_dc9(dev_priv);
+
+	return 0;
+}
+
+static int bxt_resume_prepare(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	/* TODO: when CSR FW support is added make sure the FW is loaded */
+
+	bxt_disable_dc9(dev_priv);
+
+	/*
+	 * TODO: when DC5 support is added enable DC5 here if the CSR FW
+	 * is available.
+	 */
+	broxton_init_cdclk(dev);
+	broxton_ddi_phy_init(dev);
+	intel_prepare_ddi(dev);
+
+	return 0;
+}
+
+static int skl_resume_prepare(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	skl_init_cdclk(dev_priv);
+	intel_csr_load_program(dev);
+
+	return 0;
+}
+
 /*
  * Save all Gunit registers that may be lost after a D3 and a subsequent
  * S0i[R123] transition. The list of registers needing a save/restore is
@@ -1461,6 +1559,11 @@ static int intel_runtime_resume(struct device *device)
 
 	if (IS_GEN6(dev_priv))
 		intel_init_pch_refclk(dev);
+
+	if (IS_BROXTON(dev))
+		ret = bxt_resume_prepare(dev_priv);
+	else if (IS_SKYLAKE(dev))
+		ret = skl_resume_prepare(dev_priv);
 	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
 		hsw_disable_pc8(dev_priv);
 	else if (IS_VALLEYVIEW(dev_priv))
@@ -1490,12 +1593,15 @@ static int intel_runtime_resume(struct device *device)
  */
 static int intel_suspend_complete(struct drm_i915_private *dev_priv)
 {
-	struct drm_device *dev = dev_priv->dev;
 	int ret;
 
-	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
+	if (IS_BROXTON(dev_priv))
+		ret = bxt_suspend_complete(dev_priv);
+	else if (IS_SKYLAKE(dev_priv))
+		ret = skl_suspend_complete(dev_priv);
+	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
 		ret = hsw_suspend_complete(dev_priv);
-	else if (IS_VALLEYVIEW(dev))
+	else if (IS_VALLEYVIEW(dev_priv))
 		ret = vlv_suspend_complete(dev_priv);
 	else
 		ret = 0;
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 8ae6f7f06b3a..542fac628b28 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -56,7 +56,7 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20150327"
+#define DRIVER_DATE		"20150522"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
@@ -130,7 +130,7 @@ enum transcoder {
  *
  * This value doesn't count the cursor plane.
  */
-#define I915_MAX_PLANES	3
+#define I915_MAX_PLANES	4
 
 enum plane {
 	PLANE_A = 0,
@@ -238,6 +238,11 @@ enum hpd_pin {
 #define for_each_crtc(dev, crtc) \
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
 
+#define for_each_intel_plane(dev, intel_plane) \
+	list_for_each_entry(intel_plane,			\
+			    &dev->mode_config.plane_list,	\
+			    base.head)
+
 #define for_each_intel_crtc(dev, intel_crtc) \
 	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head)
 
@@ -251,7 +256,6 @@ enum hpd_pin {
 			    &dev->mode_config.connector_list,	\
 			    base.head)
 
-
 #define for_each_encoder_on_crtc(dev, __crtc, intel_encoder) \
 	list_for_each_entry((intel_encoder), &(dev)->mode_config.encoder_list, base.head) \
 		if ((intel_encoder)->base.crtc == (__crtc))
@@ -268,6 +272,30 @@ struct drm_i915_private;
 struct i915_mm_struct;
 struct i915_mmu_object;
 
+struct drm_i915_file_private {
+	struct drm_i915_private *dev_priv;
+	struct drm_file *file;
+
+	struct {
+		spinlock_t lock;
+		struct list_head request_list;
+/* 20ms is a fairly arbitrary limit (greater than the average frame time)
+ * chosen to prevent the CPU getting more than a frame ahead of the GPU
+ * (when using lax throttling for the frontbuffer). We also use it to
+ * offer free GPU waitboosts for severely congested workloads.
+ */
+#define DRM_I915_THROTTLE_JIFFIES msecs_to_jiffies(20)
+	} mm;
+	struct idr context_idr;
+
+	struct intel_rps_client {
+		struct list_head link;
+		unsigned boosts;
+	} rps;
+
+	struct intel_engine_cs *bsd_ring;
+};
+
 enum intel_dpll_id {
 	DPLL_ID_PRIVATE = -1, /* non-shared dpll in use */
 	/* real shared dpll ids must be >= 0 */
@@ -296,13 +324,16 @@ struct intel_dpll_hw_state {
 	/* skl */
 	/*
 	 * DPLL_CTRL1 has 6 bits for each each this DPLL. We store those in
-	 * lower part of crtl1 and they get shifted into position when writing
+	 * lower part of ctrl1 and they get shifted into position when writing
 	 * the register.  This allows us to easily compare the state to share
 	 * the DPLL.
 	 */
 	uint32_t ctrl1;
 	/* HDMI only, 0 when used for DP */
 	uint32_t cfgcr1, cfgcr2;
+
+	/* bxt */
+	uint32_t ebb0, pll0, pll1, pll2, pll3, pll6, pll8, pll10, pcsdw12;
 };
 
 struct intel_shared_dpll_config {
@@ -455,6 +486,7 @@ struct drm_i915_error_state {
 		u32 semaphore_seqno[I915_NUM_RINGS - 1];
 
 		/* Register state */
+		u32 start;
 		u32 tail;
 		u32 head;
 		u32 ctl;
@@ -500,7 +532,7 @@ struct drm_i915_error_state {
 	struct drm_i915_error_buffer {
 		u32 size;
 		u32 name;
-		u32 rseqno, wseqno;
+		u32 rseqno[I915_NUM_RINGS], wseqno;
 		u32 gtt_offset;
 		u32 read_domains;
 		u32 write_domain;
@@ -666,6 +698,22 @@ struct intel_uncore {
 #define for_each_fw_domain(domain__, dev_priv__, i__) \
 	for_each_fw_domain_mask(domain__, FORCEWAKE_ALL, dev_priv__, i__)
 
+enum csr_state {
+	FW_UNINITIALIZED = 0,
+	FW_LOADED,
+	FW_FAILED
+};
+
+struct intel_csr {
+	const char *fw_path;
+	__be32 *dmc_payload;
+	uint32_t dmc_fw_size;
+	uint32_t mmio_count;
+	uint32_t mmioaddr[8];
+	uint32_t mmiodata[8];
+	enum csr_state state;
+};
+
 #define DEV_INFO_FOR_EACH_FLAG(func, sep) \
 	func(is_mobile) sep \
 	func(is_i85x) sep \
@@ -766,7 +814,7 @@ struct i915_ctx_hang_stats {
  *	       context).
  * @hang_stats: information about the role of this context in possible GPU
  *		hangs.
- * @vm: virtual memory space used by this context.
+ * @ppgtt: virtual memory space used by this context.
  * @legacy_hw_ctx: render context backing object and whether it is correctly
  *                initialized (legacy ring submission mechanism only).
  * @link: link in the global list of contexts.
@@ -880,7 +928,8 @@ struct i915_psr {
 	bool active;
 	struct delayed_work work;
 	unsigned busy_frontbuffer_bits;
-	bool link_standby;
+	bool psr2_support;
+	bool aux_frame_sync;
 };
 
 enum intel_pch {
@@ -1034,18 +1083,30 @@ struct intel_gen6_power_mgmt {
 	u8 rp0_freq;		/* Non-overclocked max frequency. */
 	u32 cz_freq;
 
+	u8 up_threshold; /* Current %busy required to uplock */
+	u8 down_threshold; /* Current %busy required to downclock */
+
 	int last_adj;
 	enum { LOW_POWER, BETWEEN, HIGH_POWER } power;
 
+	spinlock_t client_lock;
+	struct list_head clients;
+	bool client_boost;
+
 	bool enabled;
 	struct delayed_work delayed_resume_work;
+	unsigned boosts;
+
+	struct intel_rps_client semaphores, mmioflips;
 
 	/* manual wa residency calculations */
 	struct intel_rps_ei up_ei, down_ei;
 
 	/*
 	 * Protects RPS/RC6 register access and PCU communication.
-	 * Must be taken after struct_mutex if nested.
+	 * Must be taken after struct_mutex if nested. Note that
+	 * this lock may be held for long periods of time when
+	 * talking to hw - so only take it when talking to hw!
 	 */
 	struct mutex hw_lock;
 };
@@ -1136,11 +1197,6 @@ struct intel_l3_parity {
 	int which_slice;
 };
 
-struct i915_gem_batch_pool {
-	struct drm_device *dev;
-	struct list_head cache_list;
-};
-
 struct i915_gem_mm {
 	/** Memory allocator for GTT stolen memory */
 	struct drm_mm stolen;
@@ -1154,13 +1210,6 @@ struct i915_gem_mm {
 	 */
 	struct list_head unbound_list;
 
-	/*
-	 * A pool of objects to use as shadow copies of client batch buffers
-	 * when the command parser is enabled. Prevents the client from
-	 * modifying the batch contents after software parsing.
-	 */
-	struct i915_gem_batch_pool batch_pool;
-
 	/** Usable portion of the GTT for GEM */
 	unsigned long stolen_base; /* limited to low memory (32-bit) */
 
@@ -1351,7 +1400,6 @@ struct intel_vbt_data {
 	bool edp_initialized;
 	bool edp_support;
 	int edp_bpp;
-	bool edp_low_vswing;
 	struct edp_power_seq edp_pps;
 
 	struct {
@@ -1451,7 +1499,8 @@ static inline bool skl_ddb_entry_equal(const struct skl_ddb_entry *e1,
 
 struct skl_ddb_allocation {
 	struct skl_ddb_entry pipe[I915_MAX_PIPES];
-	struct skl_ddb_entry plane[I915_MAX_PIPES][I915_MAX_PLANES];
+	struct skl_ddb_entry plane[I915_MAX_PIPES][I915_MAX_PLANES]; /* packed/uv */
+	struct skl_ddb_entry y_plane[I915_MAX_PIPES][I915_MAX_PLANES]; /* y-plane */
 	struct skl_ddb_entry cursor[I915_MAX_PIPES];
 };
 
@@ -1563,7 +1612,9 @@ struct i915_virtual_gpu {
 
 struct drm_i915_private {
 	struct drm_device *dev;
-	struct kmem_cache *slab;
+	struct kmem_cache *objects;
+	struct kmem_cache *vmas;
+	struct kmem_cache *requests;
 
 	const struct intel_device_info info;
 
@@ -1575,8 +1626,12 @@ struct drm_i915_private {
 
 	struct i915_virtual_gpu vgpu;
 
-	struct intel_gmbus gmbus[GMBUS_NUM_PORTS];
+	struct intel_csr csr;
+
+	/* Display CSR-related protection */
+	struct mutex csr_lock;
 
+	struct intel_gmbus gmbus[GMBUS_NUM_PINS];
 
 	/** gmbus_mutex protects against concurrent usage of the single hw gmbus
 	 * controller on different i2c buses. */
@@ -1611,8 +1666,8 @@ struct drm_i915_private {
 	/* To control wakeup latency, e.g. for irq-driven dp aux transfers. */
 	struct pm_qos_request pm_qos;
 
-	/* DPIO indirect register protection */
-	struct mutex dpio_lock;
+	/* Sideband mailbox protection */
+	struct mutex sb_lock;
 
 	/** Cached value of IMR to avoid reads in updating the bitfield */
 	union {
@@ -1661,7 +1716,8 @@ struct drm_i915_private {
 	int num_fence_regs; /* 8 on pre-965, 16 otherwise */
 
 	unsigned int fsb_freq, mem_freq, is_ddr3;
-	unsigned int vlv_cdclk_freq;
+	unsigned int skl_boot_cdclk;
+	unsigned int cdclk_freq;
 	unsigned int hpll_freq;
 
 	/**
@@ -1759,6 +1815,8 @@ struct drm_i915_private {
 
 	u32 fdi_rx_config;
 
+	u32 chv_phy_control;
+
 	u32 suspend_count;
 	struct i915_suspend_saved_registers regfile;
 	struct vlv_s0ix_state vlv_s0ix_state;
@@ -1815,19 +1873,19 @@ struct drm_i915_private {
 
 	/* Abstract the submission mechanism (legacy ringbuffer or execlists) away */
 	struct {
-		int (*do_execbuf)(struct drm_device *dev, struct drm_file *file,
-				  struct intel_engine_cs *ring,
-				  struct intel_context *ctx,
-				  struct drm_i915_gem_execbuffer2 *args,
-				  struct list_head *vmas,
-				  struct drm_i915_gem_object *batch_obj,
-				  u64 exec_start, u32 flags);
+		int (*execbuf_submit)(struct drm_device *dev, struct drm_file *file,
+				      struct intel_engine_cs *ring,
+				      struct intel_context *ctx,
+				      struct drm_i915_gem_execbuffer2 *args,
+				      struct list_head *vmas,
+				      struct drm_i915_gem_object *batch_obj,
+				      u64 exec_start, u32 flags);
 		int (*init_rings)(struct drm_device *dev);
 		void (*cleanup_ring)(struct intel_engine_cs *ring);
 		void (*stop_ring)(struct intel_engine_cs *ring);
 	} gt;
 
-	uint32_t request_uniq;
+	bool edp_low_vswing;
 
 	/*
 	 * NOTE: This is the dri1/ums dungeon, don't add stuff here. Your patch
@@ -1913,18 +1971,18 @@ struct drm_i915_gem_object {
 	struct drm_mm_node *stolen;
 	struct list_head global_list;
 
-	struct list_head ring_list;
+	struct list_head ring_list[I915_NUM_RINGS];
 	/** Used in execbuf to temporarily hold a ref */
 	struct list_head obj_exec_link;
 
-	struct list_head batch_pool_list;
+	struct list_head batch_pool_link;
 
 	/**
 	 * This is set if the object is on the active lists (has pending
 	 * rendering and so a non-zero seqno), and is not set if it i s on
 	 * inactive (ready to be unbound) list.
 	 */
-	unsigned int active:1;
+	unsigned int active:I915_NUM_RINGS;
 
 	/**
 	 * This is set if the object has been written to since last bound
@@ -1969,8 +2027,6 @@ struct drm_i915_gem_object {
 	 * accurate mappable working set.
 	 */
 	unsigned int fault_mappable:1;
-	unsigned int pin_mappable:1;
-	unsigned int pin_display:1;
 
 	/*
 	 * Is the object to be mapped as read-only to the GPU
@@ -1984,15 +2040,30 @@ struct drm_i915_gem_object {
 
 	unsigned int frontbuffer_bits:INTEL_FRONTBUFFER_BITS;
 
+	unsigned int pin_display;
+
 	struct sg_table *pages;
 	int pages_pin_count;
+	struct get_page {
+		struct scatterlist *sg;
+		int last;
+	} get_page;
 
 	/* prime dma-buf support */
 	void *dma_buf_vmapping;
 	int vmapping_count;
 
-	/** Breadcrumb of last rendering to the buffer. */
-	struct drm_i915_gem_request *last_read_req;
+	/** Breadcrumb of last rendering to the buffer.
+	 * There can only be one writer, but we allow for multiple readers.
+	 * If there is a writer that necessarily implies that all other
+	 * read requests are complete - but we may only be lazily clearing
+	 * the read requests. A read request is naturally the most recent
+	 * request on a ring, so we may have two different write and read
+	 * requests on one ring where the write request is older than the
+	 * read request. This allows for the CPU to read from an active
+	 * buffer by only waiting for the write to complete.
+	 * */
+	struct drm_i915_gem_request *last_read_req[I915_NUM_RINGS];
 	struct drm_i915_gem_request *last_write_req;
 	/** Breadcrumb of last fenced GPU access to the buffer. */
 	struct drm_i915_gem_request *last_fenced_req;
@@ -2046,6 +2117,7 @@ struct drm_i915_gem_request {
 	struct kref ref;
 
 	/** On Which ring this request was generated */
+	struct drm_i915_private *i915;
 	struct intel_engine_cs *ring;
 
 	/** GEM sequence number associated with this request. */
@@ -2093,8 +2165,6 @@ struct drm_i915_gem_request {
 	/** process identifier submitting this request */
 	struct pid *pid;
 
-	uint32_t uniq;
-
 	/**
 	 * The ELSP only accepts two elements at a time, so we queue
 	 * context/tail pairs on a given queue (ring->execlist_queue) until the
@@ -2116,6 +2186,8 @@ struct drm_i915_gem_request {
 
 };
 
+int i915_gem_request_alloc(struct intel_engine_cs *ring,
+			   struct intel_context *ctx);
 void i915_gem_request_free(struct kref *req_ref);
 
 static inline uint32_t
@@ -2130,10 +2202,12 @@ i915_gem_request_get_ring(struct drm_i915_gem_request *req)
 	return req ? req->ring : NULL;
 }
 
-static inline void
+static inline struct drm_i915_gem_request *
 i915_gem_request_reference(struct drm_i915_gem_request *req)
 {
-	kref_get(&req->ref);
+	if (req)
+		kref_get(&req->ref);
+	return req;
 }
 
 static inline void
@@ -2143,6 +2217,19 @@ i915_gem_request_unreference(struct drm_i915_gem_request *req)
 	kref_put(&req->ref, i915_gem_request_free);
 }
 
+static inline void
+i915_gem_request_unreference__unlocked(struct drm_i915_gem_request *req)
+{
+	struct drm_device *dev;
+
+	if (!req)
+		return;
+
+	dev = req->ring->dev;
+	if (kref_put_mutex(&req->ref, i915_gem_request_free, &dev->struct_mutex))
+		mutex_unlock(&dev->struct_mutex);
+}
+
 static inline void i915_gem_request_assign(struct drm_i915_gem_request **pdst,
 					   struct drm_i915_gem_request *src)
 {
@@ -2161,21 +2248,6 @@ static inline void i915_gem_request_assign(struct drm_i915_gem_request **pdst,
  * a later patch when the call to i915_seqno_passed() is obsoleted...
  */
 
-struct drm_i915_file_private {
-	struct drm_i915_private *dev_priv;
-	struct drm_file *file;
-
-	struct {
-		spinlock_t lock;
-		struct list_head request_list;
-		struct delayed_work idle_work;
-	} mm;
-	struct idr context_idr;
-
-	atomic_t rps_wait_boost;
-	struct  intel_engine_cs *bsd_ring;
-};
-
 /*
  * A command that requires special handling by the command parser.
  */
@@ -2228,10 +2300,15 @@ struct drm_i915_cmd_descriptor {
 	 * 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.
+	 *
+	 * A non-zero step value implies that the command may access multiple
+	 * registers in sequence (e.g. LRI), in that case step gives the
+	 * distance in dwords between individual offset fields.
 	 */
 	struct {
 		u32 offset;
 		u32 mask;
+		u32 step;
 	} reg;
 
 #define MAX_CMD_DESC_BITMASKS 3
@@ -2307,6 +2384,7 @@ struct drm_i915_cmd_table {
 #define IS_HASWELL(dev)	(INTEL_INFO(dev)->is_haswell)
 #define IS_BROADWELL(dev)	(!INTEL_INFO(dev)->is_valleyview && IS_GEN8(dev))
 #define IS_SKYLAKE(dev)	(INTEL_INFO(dev)->is_skylake)
+#define IS_BROXTON(dev)	(!INTEL_INFO(dev)->is_skylake && IS_GEN9(dev))
 #define IS_MOBILE(dev)		(INTEL_INFO(dev)->is_mobile)
 #define IS_HSW_EARLY_SDV(dev)	(IS_HASWELL(dev) && \
 				 (INTEL_DEVID(dev) & 0xFF00) == 0x0C00)
@@ -2330,6 +2408,11 @@ struct drm_i915_cmd_table {
 #define SKL_REVID_C0		(0x2)
 #define SKL_REVID_D0		(0x3)
 #define SKL_REVID_E0		(0x4)
+#define SKL_REVID_F0		(0x5)
+
+#define BXT_REVID_A0		(0x0)
+#define BXT_REVID_B0		(0x3)
+#define BXT_REVID_C0		(0x6)
 
 /*
  * The genX designation typically refers to the render engine, so render
@@ -2396,16 +2479,22 @@ struct drm_i915_cmd_table {
 
 #define HAS_IPS(dev)		(IS_HSW_ULT(dev) || IS_BROADWELL(dev))
 
+#define HAS_DP_MST(dev)		(IS_HASWELL(dev) || IS_BROADWELL(dev) || \
+				 INTEL_INFO(dev)->gen >= 9)
+
 #define HAS_DDI(dev)		(INTEL_INFO(dev)->has_ddi)
 #define HAS_FPGA_DBG_UNCLAIMED(dev)	(INTEL_INFO(dev)->has_fpga_dbg)
 #define HAS_PSR(dev)		(IS_HASWELL(dev) || IS_BROADWELL(dev) || \
 				 IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev) || \
 				 IS_SKYLAKE(dev))
 #define HAS_RUNTIME_PM(dev)	(IS_GEN6(dev) || IS_HASWELL(dev) || \
-				 IS_BROADWELL(dev) || IS_VALLEYVIEW(dev))
+				 IS_BROADWELL(dev) || IS_VALLEYVIEW(dev) || \
+				 IS_SKYLAKE(dev))
 #define HAS_RC6(dev)		(INTEL_INFO(dev)->gen >= 6)
 #define HAS_RC6p(dev)		(INTEL_INFO(dev)->gen == 6 || IS_IVYBRIDGE(dev))
 
+#define HAS_CSR(dev)	(IS_SKYLAKE(dev))
+
 #define INTEL_PCH_DEVICE_ID_MASK		0xff00
 #define INTEL_PCH_IBX_DEVICE_ID_TYPE		0x3b00
 #define INTEL_PCH_CPT_DEVICE_ID_TYPE		0x1c00
@@ -2471,6 +2560,7 @@ struct i915_params {
 	int mmio_debug;
 	bool verbose_state_checks;
 	bool nuclear_pageflip;
+	int edp_vswing;
 };
 extern struct i915_params i915 __read_mostly;
 
@@ -2496,6 +2586,7 @@ extern unsigned long i915_gfx_val(struct drm_i915_private *dev_priv);
 extern void i915_update_gfx_val(struct drm_i915_private *dev_priv);
 int vlv_force_gfx_clock(struct drm_i915_private *dev_priv, bool on);
 void intel_hpd_cancel_work(struct drm_i915_private *dev_priv);
+void i915_firmware_load_error_print(const char *fw_path, int err);
 
 /* i915_irq.c */
 void i915_queue_hangcheck(struct drm_device *dev);
@@ -2520,6 +2611,13 @@ void intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
 				enum forcewake_domains domains);
 void intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
 				enum forcewake_domains domains);
+/* Like above but the caller must manage the uncore.lock itself.
+ * Must be used with I915_READ_FW and friends.
+ */
+void intel_uncore_forcewake_get__locked(struct drm_i915_private *dev_priv,
+					enum forcewake_domains domains);
+void intel_uncore_forcewake_put__locked(struct drm_i915_private *dev_priv,
+					enum forcewake_domains domains);
 void assert_forcewakes_inactive(struct drm_i915_private *dev_priv);
 static inline bool intel_vgpu_active(struct drm_device *dev)
 {
@@ -2614,10 +2712,13 @@ void i915_init_vm(struct drm_i915_private *dev_priv,
 void i915_gem_free_object(struct drm_gem_object *obj);
 void i915_gem_vma_destroy(struct i915_vma *vma);
 
-#define PIN_MAPPABLE 0x1
-#define PIN_NONBLOCK 0x2
-#define PIN_GLOBAL 0x4
-#define PIN_OFFSET_BIAS 0x8
+/* Flags used by pin/bind&friends. */
+#define PIN_MAPPABLE	(1<<0)
+#define PIN_NONBLOCK	(1<<1)
+#define PIN_GLOBAL	(1<<2)
+#define PIN_OFFSET_BIAS	(1<<3)
+#define PIN_USER	(1<<4)
+#define PIN_UPDATE	(1<<5)
 #define PIN_OFFSET_MASK (~4095)
 int __must_check
 i915_gem_object_pin(struct drm_i915_gem_object *obj,
@@ -2641,15 +2742,32 @@ 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)
+
+static inline int __sg_page_count(struct scatterlist *sg)
 {
-	struct sg_page_iter sg_iter;
+	return sg->length >> PAGE_SHIFT;
+}
 
-	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, n)
-		return sg_page_iter_page(&sg_iter);
+static inline struct page *
+i915_gem_object_get_page(struct drm_i915_gem_object *obj, int n)
+{
+	if (WARN_ON(n >= obj->base.size >> PAGE_SHIFT))
+		return NULL;
+
+	if (n < obj->get_page.last) {
+		obj->get_page.sg = obj->pages->sgl;
+		obj->get_page.last = 0;
+	}
 
-	return NULL;
+	while (obj->get_page.last + __sg_page_count(obj->get_page.sg) <= n) {
+		obj->get_page.last += __sg_page_count(obj->get_page.sg++);
+		if (unlikely(sg_is_chain(obj->get_page.sg)))
+			obj->get_page.sg = sg_chain_ptr(obj->get_page.sg);
+	}
+
+	return nth_page(sg_page(obj->get_page.sg), n - obj->get_page.last);
 }
+
 static inline void i915_gem_object_pin_pages(struct drm_i915_gem_object *obj)
 {
 	BUG_ON(obj->pages == NULL);
@@ -2739,7 +2857,6 @@ static inline bool i915_stop_ring_allow_warn(struct drm_i915_private *dev_priv)
 
 void i915_gem_reset(struct drm_device *dev);
 bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, bool force);
-int __must_check i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_init(struct drm_device *dev);
 int i915_gem_init_rings(struct drm_device *dev);
 int __must_check i915_gem_init_hw(struct drm_device *dev);
@@ -2757,10 +2874,13 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 			unsigned reset_counter,
 			bool interruptible,
 			s64 *timeout,
-			struct drm_i915_file_private *file_priv);
+			struct intel_rps_client *rps);
 int __must_check i915_wait_request(struct drm_i915_gem_request *req);
 int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
 int __must_check
+i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
+			       bool readonly);
+int __must_check
 i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj,
 				  bool write);
 int __must_check
@@ -2993,8 +3113,10 @@ int i915_verify_lists(struct drm_device *dev);
 int i915_debugfs_init(struct drm_minor *minor);
 void i915_debugfs_cleanup(struct drm_minor *minor);
 #ifdef CONFIG_DEBUG_FS
+int i915_debugfs_connector_add(struct drm_connector *connector);
 void intel_display_crc_init(struct drm_device *dev);
 #else
+static inline int i915_debugfs_connector_add(struct drm_connector *connector) {}
 static inline void intel_display_crc_init(struct drm_device *dev) {}
 #endif
 
@@ -3021,13 +3143,6 @@ 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(struct drm_i915_private *i915, int type);
 
-/* i915_gem_batch_pool.c */
-void i915_gem_batch_pool_init(struct drm_device *dev,
-			      struct i915_gem_batch_pool *pool);
-void i915_gem_batch_pool_fini(struct i915_gem_batch_pool *pool);
-struct drm_i915_gem_object*
-i915_gem_batch_pool_get(struct i915_gem_batch_pool *pool, size_t size);
-
 /* i915_cmd_parser.c */
 int i915_cmd_parser_get_version(void);
 int i915_cmd_parser_init_ring(struct intel_engine_cs *ring);
@@ -3051,13 +3166,11 @@ void i915_teardown_sysfs(struct drm_device *dev_priv);
 /* intel_i2c.c */
 extern int intel_setup_gmbus(struct drm_device *dev);
 extern void intel_teardown_gmbus(struct drm_device *dev);
-static inline bool intel_gmbus_is_port_valid(unsigned port)
-{
-	return (port >= GMBUS_PORT_SSC && port <= GMBUS_PORT_DPD);
-}
+extern bool intel_gmbus_is_valid_pin(struct drm_i915_private *dev_priv,
+				     unsigned int pin);
 
-extern struct i2c_adapter *intel_gmbus_get_adapter(
-		struct drm_i915_private *dev_priv, unsigned port);
+extern struct i2c_adapter *
+intel_gmbus_get_adapter(struct drm_i915_private *dev_priv, unsigned int pin);
 extern void intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed);
 extern void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit);
 static inline bool intel_gmbus_is_forced_bit(struct i2c_adapter *adapter)
@@ -3203,6 +3316,17 @@ int intel_freq_opcode(struct drm_i915_private *dev_priv, int val);
 #define POSTING_READ(reg)	(void)I915_READ_NOTRACE(reg)
 #define POSTING_READ16(reg)	(void)I915_READ16_NOTRACE(reg)
 
+/* These are untraced mmio-accessors that are only valid to be used inside
+ * criticial sections inside IRQ handlers where forcewake is explicitly
+ * controlled.
+ * Think twice, and think again, before using these.
+ * Note: Should only be used between intel_uncore_forcewake_irqlock() and
+ * intel_uncore_forcewake_irqunlock().
+ */
+#define I915_READ_FW(reg__) readl(dev_priv->regs + (reg__))
+#define I915_WRITE_FW(reg__, val__) writel(val__, dev_priv->regs + (reg__))
+#define POSTING_READ_FW(reg__) (void)I915_READ_FW(reg__)
+
 /* "Broadcast RGB" property */
 #define INTEL_BROADCAST_RGB_AUTO 0
 #define INTEL_BROADCAST_RGB_FULL 1
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 2d0995e7afc3..248fd1ac7b3a 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -38,14 +38,14 @@
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 
+#define RQ_BUG_ON(expr)
+
 static void i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj);
 static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj);
-static __must_check int
-i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
-			       bool readonly);
 static void
-i915_gem_object_retire(struct drm_i915_gem_object *obj);
-
+i915_gem_object_retire__write(struct drm_i915_gem_object *obj);
+static void
+i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring);
 static void i915_gem_write_fence(struct drm_device *dev, int reg,
 				 struct drm_i915_gem_object *obj);
 static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj,
@@ -378,13 +378,13 @@ out:
 void *i915_gem_object_alloc(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	return kmem_cache_zalloc(dev_priv->slab, GFP_KERNEL);
+	return kmem_cache_zalloc(dev_priv->objects, GFP_KERNEL);
 }
 
 void i915_gem_object_free(struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
-	kmem_cache_free(dev_priv->slab, obj);
+	kmem_cache_free(dev_priv->objects, obj);
 }
 
 static int
@@ -518,8 +518,6 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
 		ret = i915_gem_object_wait_rendering(obj, true);
 		if (ret)
 			return ret;
-
-		i915_gem_object_retire(obj);
 	}
 
 	ret = i915_gem_object_get_pages(obj);
@@ -939,8 +937,6 @@ i915_gem_shmem_pwrite(struct drm_device *dev,
 		ret = i915_gem_object_wait_rendering(obj, false);
 		if (ret)
 			return ret;
-
-		i915_gem_object_retire(obj);
 	}
 	/* Same trick applies to invalidate partially written cachelines read
 	 * before writing. */
@@ -1181,12 +1177,27 @@ static bool missed_irq(struct drm_i915_private *dev_priv,
 	return test_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings);
 }
 
-static bool can_wait_boost(struct drm_i915_file_private *file_priv)
+static int __i915_spin_request(struct drm_i915_gem_request *req)
 {
-	if (file_priv == NULL)
-		return true;
+	unsigned long timeout;
 
-	return !atomic_xchg(&file_priv->rps_wait_boost, true);
+	if (i915_gem_request_get_ring(req)->irq_refcount)
+		return -EBUSY;
+
+	timeout = jiffies + 1;
+	while (!need_resched()) {
+		if (i915_gem_request_completed(req, true))
+			return 0;
+
+		if (time_after_eq(jiffies, timeout))
+			break;
+
+		cpu_relax_lowlatency();
+	}
+	if (i915_gem_request_completed(req, false))
+		return 0;
+
+	return -EAGAIN;
 }
 
 /**
@@ -1210,7 +1221,7 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 			unsigned reset_counter,
 			bool interruptible,
 			s64 *timeout,
-			struct drm_i915_file_private *file_priv)
+			struct intel_rps_client *rps)
 {
 	struct intel_engine_cs *ring = i915_gem_request_get_ring(req);
 	struct drm_device *dev = ring->dev;
@@ -1224,26 +1235,32 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 
 	WARN(!intel_irqs_enabled(dev_priv), "IRQs disabled");
 
+	if (list_empty(&req->list))
+		return 0;
+
 	if (i915_gem_request_completed(req, true))
 		return 0;
 
 	timeout_expire = timeout ?
 		jiffies + nsecs_to_jiffies_timeout((u64)*timeout) : 0;
 
-	if (INTEL_INFO(dev)->gen >= 6 && ring->id == RCS && can_wait_boost(file_priv)) {
-		gen6_rps_boost(dev_priv);
-		if (file_priv)
-			mod_delayed_work(dev_priv->wq,
-					 &file_priv->mm.idle_work,
-					 msecs_to_jiffies(100));
-	}
-
-	if (!irq_test_in_progress && WARN_ON(!ring->irq_get(ring)))
-		return -ENODEV;
+	if (INTEL_INFO(dev_priv)->gen >= 6)
+		gen6_rps_boost(dev_priv, rps, req->emitted_jiffies);
 
 	/* Record current time in case interrupted by signal, or wedged */
 	trace_i915_gem_request_wait_begin(req);
 	before = ktime_get_raw_ns();
+
+	/* Optimistic spin for the next jiffie before touching IRQs */
+	ret = __i915_spin_request(req);
+	if (ret == 0)
+		goto out;
+
+	if (!irq_test_in_progress && WARN_ON(!ring->irq_get(ring))) {
+		ret = -ENODEV;
+		goto out;
+	}
+
 	for (;;) {
 		struct timer_list timer;
 
@@ -1292,14 +1309,15 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 			destroy_timer_on_stack(&timer);
 		}
 	}
-	now = ktime_get_raw_ns();
-	trace_i915_gem_request_wait_end(req);
-
 	if (!irq_test_in_progress)
 		ring->irq_put(ring);
 
 	finish_wait(&ring->irq_queue, &wait);
 
+out:
+	now = ktime_get_raw_ns();
+	trace_i915_gem_request_wait_end(req);
+
 	if (timeout) {
 		s64 tres = *timeout - (now - before);
 
@@ -1319,6 +1337,63 @@ int __i915_wait_request(struct drm_i915_gem_request *req,
 	return ret;
 }
 
+static inline void
+i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
+{
+	struct drm_i915_file_private *file_priv = request->file_priv;
+
+	if (!file_priv)
+		return;
+
+	spin_lock(&file_priv->mm.lock);
+	list_del(&request->client_list);
+	request->file_priv = NULL;
+	spin_unlock(&file_priv->mm.lock);
+}
+
+static void i915_gem_request_retire(struct drm_i915_gem_request *request)
+{
+	trace_i915_gem_request_retire(request);
+
+	/* We know the GPU must have read the request to have
+	 * sent us the seqno + interrupt, so use the position
+	 * of tail of the request to update the last known position
+	 * of the GPU head.
+	 *
+	 * Note this requires that we are always called in request
+	 * completion order.
+	 */
+	request->ringbuf->last_retired_head = request->postfix;
+
+	list_del_init(&request->list);
+	i915_gem_request_remove_from_client(request);
+
+	put_pid(request->pid);
+
+	i915_gem_request_unreference(request);
+}
+
+static void
+__i915_gem_request_retire__upto(struct drm_i915_gem_request *req)
+{
+	struct intel_engine_cs *engine = req->ring;
+	struct drm_i915_gem_request *tmp;
+
+	lockdep_assert_held(&engine->dev->struct_mutex);
+
+	if (list_empty(&req->list))
+		return;
+
+	do {
+		tmp = list_first_entry(&engine->request_list,
+				       typeof(*tmp), list);
+
+		i915_gem_request_retire(tmp);
+	} while (tmp != req);
+
+	WARN_ON(i915_verify_lists(engine->dev));
+}
+
 /**
  * Waits for a request to be signaled, and cleans up the
  * request and object lists appropriately for that event.
@@ -1329,7 +1404,6 @@ i915_wait_request(struct drm_i915_gem_request *req)
 	struct drm_device *dev;
 	struct drm_i915_private *dev_priv;
 	bool interruptible;
-	unsigned reset_counter;
 	int ret;
 
 	BUG_ON(req == NULL);
@@ -1348,29 +1422,13 @@ i915_wait_request(struct drm_i915_gem_request *req)
 	if (ret)
 		return ret;
 
-	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
-	i915_gem_request_reference(req);
-	ret = __i915_wait_request(req, reset_counter,
+	ret = __i915_wait_request(req,
+				  atomic_read(&dev_priv->gpu_error.reset_counter),
 				  interruptible, NULL, NULL);
-	i915_gem_request_unreference(req);
-	return ret;
-}
-
-static int
-i915_gem_object_wait_rendering__tail(struct drm_i915_gem_object *obj)
-{
-	if (!obj->active)
-		return 0;
-
-	/* Manually manage the write flush as we may have not yet
-	 * retired the buffer.
-	 *
-	 * Note that the last_write_req is always the earlier of
-	 * the two (read/write) requests, so if we haved successfully waited,
-	 * we know we have passed the last write.
-	 */
-	i915_gem_request_assign(&obj->last_write_req, NULL);
+	if (ret)
+		return ret;
 
+	__i915_gem_request_retire__upto(req);
 	return 0;
 }
 
@@ -1378,22 +1436,56 @@ i915_gem_object_wait_rendering__tail(struct drm_i915_gem_object *obj)
  * Ensures that all rendering to the object has completed and the object is
  * safe to unbind from the GTT or access from the CPU.
  */
-static __must_check int
+int
 i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
 			       bool readonly)
 {
-	struct drm_i915_gem_request *req;
-	int ret;
+	int ret, i;
 
-	req = readonly ? obj->last_write_req : obj->last_read_req;
-	if (!req)
+	if (!obj->active)
 		return 0;
 
-	ret = i915_wait_request(req);
-	if (ret)
-		return ret;
+	if (readonly) {
+		if (obj->last_write_req != NULL) {
+			ret = i915_wait_request(obj->last_write_req);
+			if (ret)
+				return ret;
+
+			i = obj->last_write_req->ring->id;
+			if (obj->last_read_req[i] == obj->last_write_req)
+				i915_gem_object_retire__read(obj, i);
+			else
+				i915_gem_object_retire__write(obj);
+		}
+	} else {
+		for (i = 0; i < I915_NUM_RINGS; i++) {
+			if (obj->last_read_req[i] == NULL)
+				continue;
+
+			ret = i915_wait_request(obj->last_read_req[i]);
+			if (ret)
+				return ret;
+
+			i915_gem_object_retire__read(obj, i);
+		}
+		RQ_BUG_ON(obj->active);
+	}
+
+	return 0;
+}
+
+static void
+i915_gem_object_retire_request(struct drm_i915_gem_object *obj,
+			       struct drm_i915_gem_request *req)
+{
+	int ring = req->ring->id;
+
+	if (obj->last_read_req[ring] == req)
+		i915_gem_object_retire__read(obj, ring);
+	else if (obj->last_write_req == req)
+		i915_gem_object_retire__write(obj);
 
-	return i915_gem_object_wait_rendering__tail(obj);
+	__i915_gem_request_retire__upto(req);
 }
 
 /* A nonblocking variant of the above wait. This is a highly dangerous routine
@@ -1401,40 +1493,75 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
  */
 static __must_check int
 i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
-					    struct drm_i915_file_private *file_priv,
+					    struct intel_rps_client *rps,
 					    bool readonly)
 {
-	struct drm_i915_gem_request *req;
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_request *requests[I915_NUM_RINGS];
 	unsigned reset_counter;
-	int ret;
+	int ret, i, n = 0;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
 	BUG_ON(!dev_priv->mm.interruptible);
 
-	req = readonly ? obj->last_write_req : obj->last_read_req;
-	if (!req)
+	if (!obj->active)
 		return 0;
 
 	ret = i915_gem_check_wedge(&dev_priv->gpu_error, true);
 	if (ret)
 		return ret;
 
-	ret = i915_gem_check_olr(req);
-	if (ret)
-		return ret;
-
 	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
-	i915_gem_request_reference(req);
+
+	if (readonly) {
+		struct drm_i915_gem_request *req;
+
+		req = obj->last_write_req;
+		if (req == NULL)
+			return 0;
+
+		ret = i915_gem_check_olr(req);
+		if (ret)
+			goto err;
+
+		requests[n++] = i915_gem_request_reference(req);
+	} else {
+		for (i = 0; i < I915_NUM_RINGS; i++) {
+			struct drm_i915_gem_request *req;
+
+			req = obj->last_read_req[i];
+			if (req == NULL)
+				continue;
+
+			ret = i915_gem_check_olr(req);
+			if (ret)
+				goto err;
+
+			requests[n++] = i915_gem_request_reference(req);
+		}
+	}
+
 	mutex_unlock(&dev->struct_mutex);
-	ret = __i915_wait_request(req, reset_counter, true, NULL, file_priv);
+	for (i = 0; ret == 0 && i < n; i++)
+		ret = __i915_wait_request(requests[i], reset_counter, true,
+					  NULL, rps);
 	mutex_lock(&dev->struct_mutex);
-	i915_gem_request_unreference(req);
-	if (ret)
-		return ret;
 
-	return i915_gem_object_wait_rendering__tail(obj);
+err:
+	for (i = 0; i < n; i++) {
+		if (ret == 0)
+			i915_gem_object_retire_request(obj, requests[i]);
+		i915_gem_request_unreference(requests[i]);
+	}
+
+	return ret;
+}
+
+static struct intel_rps_client *to_rps_client(struct drm_file *file)
+{
+	struct drm_i915_file_private *fpriv = file->driver_priv;
+	return &fpriv->rps;
 }
 
 /**
@@ -1479,7 +1606,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	 * to catch cases where we are gazumped.
 	 */
 	ret = i915_gem_object_wait_rendering__nonblocking(obj,
-							  file->driver_priv,
+							  to_rps_client(file),
 							  !write_domain);
 	if (ret)
 		goto unref;
@@ -1616,6 +1743,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	struct drm_i915_gem_object *obj = to_intel_bo(vma->vm_private_data);
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct i915_ggtt_view view = i915_ggtt_view_normal;
 	pgoff_t page_offset;
 	unsigned long pfn;
 	int ret = 0;
@@ -1648,8 +1776,23 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 		goto unlock;
 	}
 
-	/* Now bind it into the GTT if needed */
-	ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_MAPPABLE);
+	/* Use a partial view if the object is bigger than the aperture. */
+	if (obj->base.size >= dev_priv->gtt.mappable_end &&
+	    obj->tiling_mode == I915_TILING_NONE) {
+		static const unsigned int chunk_size = 256; // 1 MiB
+
+		memset(&view, 0, sizeof(view));
+		view.type = I915_GGTT_VIEW_PARTIAL;
+		view.params.partial.offset = rounddown(page_offset, chunk_size);
+		view.params.partial.size =
+			min_t(unsigned int,
+			      chunk_size,
+			      (vma->vm_end - vma->vm_start)/PAGE_SIZE -
+			      view.params.partial.offset);
+	}
+
+	/* Now pin it into the GTT if needed */
+	ret = i915_gem_object_ggtt_pin(obj, &view, 0, PIN_MAPPABLE);
 	if (ret)
 		goto unlock;
 
@@ -1662,30 +1805,50 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 		goto unpin;
 
 	/* Finally, remap it using the new GTT offset */
-	pfn = dev_priv->gtt.mappable_base + i915_gem_obj_ggtt_offset(obj);
+	pfn = dev_priv->gtt.mappable_base +
+		i915_gem_obj_ggtt_offset_view(obj, &view);
 	pfn >>= PAGE_SHIFT;
 
-	if (!obj->fault_mappable) {
-		unsigned long size = min_t(unsigned long,
-					   vma->vm_end - vma->vm_start,
-					   obj->base.size);
-		int i;
+	if (unlikely(view.type == I915_GGTT_VIEW_PARTIAL)) {
+		/* Overriding existing pages in partial view does not cause
+		 * us any trouble as TLBs are still valid because the fault
+		 * is due to userspace losing part of the mapping or never
+		 * having accessed it before (at this partials' range).
+		 */
+		unsigned long base = vma->vm_start +
+				     (view.params.partial.offset << PAGE_SHIFT);
+		unsigned int i;
 
-		for (i = 0; i < size >> PAGE_SHIFT; i++) {
-			ret = vm_insert_pfn(vma,
-					    (unsigned long)vma->vm_start + i * PAGE_SIZE,
-					    pfn + i);
+		for (i = 0; i < view.params.partial.size; i++) {
+			ret = vm_insert_pfn(vma, base + i * PAGE_SIZE, pfn + i);
 			if (ret)
 				break;
 		}
 
 		obj->fault_mappable = true;
-	} else
-		ret = vm_insert_pfn(vma,
-				    (unsigned long)vmf->virtual_address,
-				    pfn + page_offset);
+	} else {
+		if (!obj->fault_mappable) {
+			unsigned long size = min_t(unsigned long,
+						   vma->vm_end - vma->vm_start,
+						   obj->base.size);
+			int i;
+
+			for (i = 0; i < size >> PAGE_SHIFT; i++) {
+				ret = vm_insert_pfn(vma,
+						    (unsigned long)vma->vm_start + i * PAGE_SIZE,
+						    pfn + i);
+				if (ret)
+					break;
+			}
+
+			obj->fault_mappable = true;
+		} else
+			ret = vm_insert_pfn(vma,
+					    (unsigned long)vmf->virtual_address,
+					    pfn + page_offset);
+	}
 unpin:
-	i915_gem_object_ggtt_unpin(obj);
+	i915_gem_object_ggtt_unpin_view(obj, &view);
 unlock:
 	mutex_unlock(&dev->struct_mutex);
 out:
@@ -1864,7 +2027,6 @@ i915_gem_mmap_gtt(struct drm_file *file,
 		  uint32_t handle,
 		  uint64_t *offset)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
 	int ret;
 
@@ -1878,11 +2040,6 @@ i915_gem_mmap_gtt(struct drm_file *file,
 		goto unlock;
 	}
 
-	if (obj->base.size > dev_priv->gtt.mappable_end) {
-		ret = -E2BIG;
-		goto out;
-	}
-
 	if (obj->madv != I915_MADV_WILLNEED) {
 		DRM_DEBUG("Attempting to mmap a purgeable buffer\n");
 		ret = -EFAULT;
@@ -2178,81 +2335,65 @@ i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 		return ret;
 
 	list_add_tail(&obj->global_list, &dev_priv->mm.unbound_list);
+
+	obj->get_page.sg = obj->pages->sgl;
+	obj->get_page.last = 0;
+
 	return 0;
 }
 
-static void
-i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
-			       struct intel_engine_cs *ring)
+void i915_vma_move_to_active(struct i915_vma *vma,
+			     struct intel_engine_cs *ring)
 {
-	struct drm_i915_gem_request *req;
-	struct intel_engine_cs *old_ring;
-
-	BUG_ON(ring == NULL);
-
-	req = intel_ring_get_request(ring);
-	old_ring = i915_gem_request_get_ring(obj->last_read_req);
-
-	if (old_ring != ring && obj->last_write_req) {
-		/* Keep the request relative to the current ring */
-		i915_gem_request_assign(&obj->last_write_req, req);
-	}
+	struct drm_i915_gem_object *obj = vma->obj;
 
 	/* Add a reference if we're newly entering the active list. */
-	if (!obj->active) {
+	if (obj->active == 0)
 		drm_gem_object_reference(&obj->base);
-		obj->active = 1;
-	}
+	obj->active |= intel_ring_flag(ring);
 
-	list_move_tail(&obj->ring_list, &ring->active_list);
+	list_move_tail(&obj->ring_list[ring->id], &ring->active_list);
+	i915_gem_request_assign(&obj->last_read_req[ring->id],
+				intel_ring_get_request(ring));
 
-	i915_gem_request_assign(&obj->last_read_req, req);
+	list_move_tail(&vma->mm_list, &vma->vm->active_list);
 }
 
-void i915_vma_move_to_active(struct i915_vma *vma,
-			     struct intel_engine_cs *ring)
+static void
+i915_gem_object_retire__write(struct drm_i915_gem_object *obj)
 {
-	list_move_tail(&vma->mm_list, &vma->vm->active_list);
-	return i915_gem_object_move_to_active(vma->obj, ring);
+	RQ_BUG_ON(obj->last_write_req == NULL);
+	RQ_BUG_ON(!(obj->active & intel_ring_flag(obj->last_write_req->ring)));
+
+	i915_gem_request_assign(&obj->last_write_req, NULL);
+	intel_fb_obj_flush(obj, true);
 }
 
 static void
-i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj)
+i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring)
 {
 	struct i915_vma *vma;
 
-	BUG_ON(obj->base.write_domain & ~I915_GEM_GPU_DOMAINS);
-	BUG_ON(!obj->active);
+	RQ_BUG_ON(obj->last_read_req[ring] == NULL);
+	RQ_BUG_ON(!(obj->active & (1 << ring)));
+
+	list_del_init(&obj->ring_list[ring]);
+	i915_gem_request_assign(&obj->last_read_req[ring], NULL);
+
+	if (obj->last_write_req && obj->last_write_req->ring->id == ring)
+		i915_gem_object_retire__write(obj);
+
+	obj->active &= ~(1 << ring);
+	if (obj->active)
+		return;
 
 	list_for_each_entry(vma, &obj->vma_list, vma_link) {
 		if (!list_empty(&vma->mm_list))
 			list_move_tail(&vma->mm_list, &vma->vm->inactive_list);
 	}
 
-	intel_fb_obj_flush(obj, true);
-
-	list_del_init(&obj->ring_list);
-
-	i915_gem_request_assign(&obj->last_read_req, NULL);
-	i915_gem_request_assign(&obj->last_write_req, NULL);
-	obj->base.write_domain = 0;
-
 	i915_gem_request_assign(&obj->last_fenced_req, NULL);
-
-	obj->active = 0;
 	drm_gem_object_unreference(&obj->base);
-
-	WARN_ON(i915_verify_lists(dev));
-}
-
-static void
-i915_gem_object_retire(struct drm_i915_gem_object *obj)
-{
-	if (obj->last_read_req == NULL)
-		return;
-
-	if (i915_gem_request_completed(obj->last_read_req, true))
-		i915_gem_object_move_to_inactive(obj);
 }
 
 static int
@@ -2421,7 +2562,6 @@ int __i915_add_request(struct intel_engine_cs *ring,
 
 	i915_queue_hangcheck(ring->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));
@@ -2430,20 +2570,6 @@ int __i915_add_request(struct intel_engine_cs *ring,
 	return 0;
 }
 
-static inline void
-i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
-{
-	struct drm_i915_file_private *file_priv = request->file_priv;
-
-	if (!file_priv)
-		return;
-
-	spin_lock(&file_priv->mm.lock);
-	list_del(&request->client_list);
-	request->file_priv = NULL;
-	spin_unlock(&file_priv->mm.lock);
-}
-
 static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
 				   const struct intel_context *ctx)
 {
@@ -2489,16 +2615,6 @@ static void i915_set_reset_status(struct drm_i915_private *dev_priv,
 	}
 }
 
-static void i915_gem_free_request(struct drm_i915_gem_request *request)
-{
-	list_del(&request->list);
-	i915_gem_request_remove_from_client(request);
-
-	put_pid(request->pid);
-
-	i915_gem_request_unreference(request);
-}
-
 void i915_gem_request_free(struct kref *req_ref)
 {
 	struct drm_i915_gem_request *req = container_of(req_ref,
@@ -2516,7 +2632,45 @@ void i915_gem_request_free(struct kref *req_ref)
 		i915_gem_context_unreference(ctx);
 	}
 
-	kfree(req);
+	kmem_cache_free(req->i915->requests, req);
+}
+
+int i915_gem_request_alloc(struct intel_engine_cs *ring,
+			   struct intel_context *ctx)
+{
+	struct drm_i915_private *dev_priv = to_i915(ring->dev);
+	struct drm_i915_gem_request *req;
+	int ret;
+
+	if (ring->outstanding_lazy_request)
+		return 0;
+
+	req = kmem_cache_zalloc(dev_priv->requests, GFP_KERNEL);
+	if (req == NULL)
+		return -ENOMEM;
+
+	kref_init(&req->ref);
+	req->i915 = dev_priv;
+
+	ret = i915_gem_get_seqno(ring->dev, &req->seqno);
+	if (ret)
+		goto err;
+
+	req->ring = ring;
+
+	if (i915.enable_execlists)
+		ret = intel_logical_ring_alloc_request_extras(req, ctx);
+	else
+		ret = intel_ring_alloc_request_extras(req);
+	if (ret)
+		goto err;
+
+	ring->outstanding_lazy_request = req;
+	return 0;
+
+err:
+	kmem_cache_free(dev_priv->requests, req);
+	return ret;
 }
 
 struct drm_i915_gem_request *
@@ -2561,9 +2715,9 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 
 		obj = list_first_entry(&ring->active_list,
 				       struct drm_i915_gem_object,
-				       ring_list);
+				       ring_list[ring->id]);
 
-		i915_gem_object_move_to_inactive(obj);
+		i915_gem_object_retire__read(obj, ring->id);
 	}
 
 	/*
@@ -2578,7 +2732,6 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 				struct drm_i915_gem_request,
 				execlist_link);
 		list_del(&submit_req->execlist_link);
-		intel_runtime_pm_put(dev_priv);
 
 		if (submit_req->ctx != ring->default_context)
 			intel_lr_context_unpin(ring, submit_req->ctx);
@@ -2600,7 +2753,7 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 					   struct drm_i915_gem_request,
 					   list);
 
-		i915_gem_free_request(request);
+		i915_gem_request_retire(request);
 	}
 
 	/* This may not have been flushed before the reset, so clean it now */
@@ -2648,6 +2801,8 @@ void i915_gem_reset(struct drm_device *dev)
 	i915_gem_context_reset(dev);
 
 	i915_gem_restore_fences(dev);
+
+	WARN_ON(i915_verify_lists(dev));
 }
 
 /**
@@ -2656,9 +2811,6 @@ void i915_gem_reset(struct drm_device *dev)
 void
 i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 {
-	if (list_empty(&ring->request_list))
-		return;
-
 	WARN_ON(i915_verify_lists(ring->dev));
 
 	/* Retire requests first as we use it above for the early return.
@@ -2676,16 +2828,7 @@ i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 		if (!i915_gem_request_completed(request, true))
 			break;
 
-		trace_i915_gem_request_retire(request);
-
-		/* We know the GPU must have read the request to have
-		 * sent us the seqno + interrupt, so use the position
-		 * of tail of the request to update the last known position
-		 * of the GPU head.
-		 */
-		request->ringbuf->last_retired_head = request->postfix;
-
-		i915_gem_free_request(request);
+		i915_gem_request_retire(request);
 	}
 
 	/* Move any buffers on the active list that are no longer referenced
@@ -2697,12 +2840,12 @@ i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 
 		obj = list_first_entry(&ring->active_list,
 				      struct drm_i915_gem_object,
-				      ring_list);
+				      ring_list[ring->id]);
 
-		if (!i915_gem_request_completed(obj->last_read_req, true))
+		if (!list_empty(&obj->last_read_req[ring->id]->list))
 			break;
 
-		i915_gem_object_move_to_inactive(obj);
+		i915_gem_object_retire__read(obj, ring->id);
 	}
 
 	if (unlikely(ring->trace_irq_req &&
@@ -2768,8 +2911,25 @@ i915_gem_idle_work_handler(struct work_struct *work)
 {
 	struct drm_i915_private *dev_priv =
 		container_of(work, typeof(*dev_priv), mm.idle_work.work);
+	struct drm_device *dev = dev_priv->dev;
+	struct intel_engine_cs *ring;
+	int i;
+
+	for_each_ring(ring, dev_priv, i)
+		if (!list_empty(&ring->request_list))
+			return;
+
+	intel_mark_idle(dev);
+
+	if (mutex_trylock(&dev->struct_mutex)) {
+		struct intel_engine_cs *ring;
+		int i;
 
-	intel_mark_idle(dev_priv->dev);
+		for_each_ring(ring, dev_priv, i)
+			i915_gem_batch_pool_fini(&ring->batch_pool);
+
+		mutex_unlock(&dev->struct_mutex);
+	}
 }
 
 /**
@@ -2780,17 +2940,30 @@ i915_gem_idle_work_handler(struct work_struct *work)
 static int
 i915_gem_object_flush_active(struct drm_i915_gem_object *obj)
 {
-	struct intel_engine_cs *ring;
-	int ret;
+	int ret, i;
+
+	if (!obj->active)
+		return 0;
+
+	for (i = 0; i < I915_NUM_RINGS; i++) {
+		struct drm_i915_gem_request *req;
+
+		req = obj->last_read_req[i];
+		if (req == NULL)
+			continue;
 
-	if (obj->active) {
-		ring = i915_gem_request_get_ring(obj->last_read_req);
+		if (list_empty(&req->list))
+			goto retire;
 
-		ret = i915_gem_check_olr(obj->last_read_req);
+		ret = i915_gem_check_olr(req);
 		if (ret)
 			return ret;
 
-		i915_gem_retire_requests_ring(ring);
+		if (i915_gem_request_completed(req, true)) {
+			__i915_gem_request_retire__upto(req);
+retire:
+			i915_gem_object_retire__read(obj, i);
+		}
 	}
 
 	return 0;
@@ -2824,9 +2997,10 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_wait *args = data;
 	struct drm_i915_gem_object *obj;
-	struct drm_i915_gem_request *req;
+	struct drm_i915_gem_request *req[I915_NUM_RINGS];
 	unsigned reset_counter;
-	int ret = 0;
+	int i, n = 0;
+	int ret;
 
 	if (args->flags != 0)
 		return -EINVAL;
@@ -2846,11 +3020,9 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	if (ret)
 		goto out;
 
-	if (!obj->active || !obj->last_read_req)
+	if (!obj->active)
 		goto out;
 
-	req = obj->last_read_req;
-
 	/* Do this after OLR check to make sure we make forward progress polling
 	 * on this IOCTL with a timeout == 0 (like busy ioctl)
 	 */
@@ -2861,15 +3033,23 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 
 	drm_gem_object_unreference(&obj->base);
 	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
-	i915_gem_request_reference(req);
-	mutex_unlock(&dev->struct_mutex);
 
-	ret = __i915_wait_request(req, reset_counter, true,
-				  args->timeout_ns > 0 ? &args->timeout_ns : NULL,
-				  file->driver_priv);
-	mutex_lock(&dev->struct_mutex);
-	i915_gem_request_unreference(req);
+	for (i = 0; i < I915_NUM_RINGS; i++) {
+		if (obj->last_read_req[i] == NULL)
+			continue;
+
+		req[n++] = i915_gem_request_reference(obj->last_read_req[i]);
+	}
+
 	mutex_unlock(&dev->struct_mutex);
+
+	for (i = 0; i < n; i++) {
+		if (ret == 0)
+			ret = __i915_wait_request(req[i], reset_counter, true,
+						  args->timeout_ns > 0 ? &args->timeout_ns : NULL,
+						  file->driver_priv);
+		i915_gem_request_unreference__unlocked(req[i]);
+	}
 	return ret;
 
 out:
@@ -2878,6 +3058,59 @@ out:
 	return ret;
 }
 
+static int
+__i915_gem_object_sync(struct drm_i915_gem_object *obj,
+		       struct intel_engine_cs *to,
+		       struct drm_i915_gem_request *req)
+{
+	struct intel_engine_cs *from;
+	int ret;
+
+	from = i915_gem_request_get_ring(req);
+	if (to == from)
+		return 0;
+
+	if (i915_gem_request_completed(req, true))
+		return 0;
+
+	ret = i915_gem_check_olr(req);
+	if (ret)
+		return ret;
+
+	if (!i915_semaphore_is_enabled(obj->base.dev)) {
+		struct drm_i915_private *i915 = to_i915(obj->base.dev);
+		ret = __i915_wait_request(req,
+					  atomic_read(&i915->gpu_error.reset_counter),
+					  i915->mm.interruptible,
+					  NULL,
+					  &i915->rps.semaphores);
+		if (ret)
+			return ret;
+
+		i915_gem_object_retire_request(obj, req);
+	} else {
+		int idx = intel_ring_sync_index(from, to);
+		u32 seqno = i915_gem_request_get_seqno(req);
+
+		if (seqno <= from->semaphore.sync_seqno[idx])
+			return 0;
+
+		trace_i915_gem_ring_sync_to(from, to, req);
+		ret = to->semaphore.sync_to(to, from, seqno);
+		if (ret)
+			return ret;
+
+		/* We use last_read_req because sync_to()
+		 * might have just caused seqno wrap under
+		 * the radar.
+		 */
+		from->semaphore.sync_seqno[idx] =
+			i915_gem_request_get_seqno(obj->last_read_req[from->id]);
+	}
+
+	return 0;
+}
+
 /**
  * i915_gem_object_sync - sync an object to a ring.
  *
@@ -2886,7 +3119,17 @@ out:
  *
  * This code is meant to abstract object synchronization with the GPU.
  * Calling with NULL implies synchronizing the object with the CPU
- * rather than a particular GPU ring.
+ * rather than a particular GPU ring. Conceptually we serialise writes
+ * between engines inside the GPU. We only allow on engine to write
+ * into a buffer at any time, but multiple readers. To ensure each has
+ * a coherent view of memory, we must:
+ *
+ * - If there is an outstanding write request to the object, the new
+ *   request must wait for it to complete (either CPU or in hw, requests
+ *   on the same ring will be naturally ordered).
+ *
+ * - If we are a write request (pending_write_domain is set), the new
+ *   request must wait for outstanding read requests to complete.
  *
  * Returns 0 if successful, else propagates up the lower layer error.
  */
@@ -2894,41 +3137,32 @@ int
 i915_gem_object_sync(struct drm_i915_gem_object *obj,
 		     struct intel_engine_cs *to)
 {
-	struct intel_engine_cs *from;
-	u32 seqno;
-	int ret, idx;
-
-	from = i915_gem_request_get_ring(obj->last_read_req);
-
-	if (from == NULL || to == from)
-		return 0;
-
-	if (to == NULL || !i915_semaphore_is_enabled(obj->base.dev))
-		return i915_gem_object_wait_rendering(obj, false);
+	const bool readonly = obj->base.pending_write_domain == 0;
+	struct drm_i915_gem_request *req[I915_NUM_RINGS];
+	int ret, i, n;
 
-	idx = intel_ring_sync_index(from, to);
-
-	seqno = i915_gem_request_get_seqno(obj->last_read_req);
-	/* Optimization: Avoid semaphore sync when we are sure we already
-	 * waited for an object with higher seqno */
-	if (seqno <= from->semaphore.sync_seqno[idx])
+	if (!obj->active)
 		return 0;
 
-	ret = i915_gem_check_olr(obj->last_read_req);
-	if (ret)
-		return ret;
+	if (to == NULL)
+		return i915_gem_object_wait_rendering(obj, readonly);
 
-	trace_i915_gem_ring_sync_to(from, to, obj->last_read_req);
-	ret = to->semaphore.sync_to(to, from, seqno);
-	if (!ret)
-		/* We use last_read_req because sync_to()
-		 * might have just caused seqno wrap under
-		 * the radar.
-		 */
-		from->semaphore.sync_seqno[idx] =
-				i915_gem_request_get_seqno(obj->last_read_req);
+	n = 0;
+	if (readonly) {
+		if (obj->last_write_req)
+			req[n++] = obj->last_write_req;
+	} else {
+		for (i = 0; i < I915_NUM_RINGS; i++)
+			if (obj->last_read_req[i])
+				req[n++] = obj->last_read_req[i];
+	}
+	for (i = 0; i < n; i++) {
+		ret = __i915_gem_object_sync(obj, to, req[i]);
+		if (ret)
+			return ret;
+	}
 
-	return ret;
+	return 0;
 }
 
 static void i915_gem_object_finish_gtt(struct drm_i915_gem_object *obj)
@@ -2974,7 +3208,7 @@ int i915_vma_unbind(struct i915_vma *vma)
 
 	BUG_ON(obj->pages == NULL);
 
-	ret = i915_gem_object_finish_gpu(obj);
+	ret = i915_gem_object_wait_rendering(obj, false);
 	if (ret)
 		return ret;
 	/* Continue on if we fail due to EIO, the GPU is hung so we
@@ -2994,7 +3228,8 @@ int i915_vma_unbind(struct i915_vma *vma)
 
 	trace_i915_vma_unbind(vma);
 
-	vma->unbind_vma(vma);
+	vma->vm->unbind_vma(vma);
+	vma->bound = 0;
 
 	list_del_init(&vma->mm_list);
 	if (i915_is_ggtt(vma->vm)) {
@@ -3013,10 +3248,6 @@ int i915_vma_unbind(struct i915_vma *vma)
 	/* Since the unbound list is global, only move to that list if
 	 * no more VMAs exist. */
 	if (list_empty(&obj->vma_list)) {
-		/* Throw away the active reference before
-		 * moving to the unbound list. */
-		i915_gem_object_retire(obj);
-
 		i915_gem_gtt_finish_object(obj);
 		list_move_tail(&obj->global_list, &dev_priv->mm.unbound_list);
 	}
@@ -3049,6 +3280,7 @@ int i915_gpu_idle(struct drm_device *dev)
 			return ret;
 	}
 
+	WARN_ON(i915_verify_lists(dev));
 	return 0;
 }
 
@@ -3423,7 +3655,8 @@ static bool i915_gem_valid_gtt_space(struct i915_vma *vma,
 }
 
 /**
- * Finds free space in the GTT aperture and binds the object there.
+ * Finds free space in the GTT aperture and binds the object or a view of it
+ * there.
  */
 static struct i915_vma *
 i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
@@ -3442,36 +3675,60 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 	struct i915_vma *vma;
 	int ret;
 
-	if(WARN_ON(i915_is_ggtt(vm) != !!ggtt_view))
-		return ERR_PTR(-EINVAL);
+	if (i915_is_ggtt(vm)) {
+		u32 view_size;
+
+		if (WARN_ON(!ggtt_view))
+			return ERR_PTR(-EINVAL);
 
-	fence_size = i915_gem_get_gtt_size(dev,
-					   obj->base.size,
-					   obj->tiling_mode);
-	fence_alignment = i915_gem_get_gtt_alignment(dev,
-						     obj->base.size,
-						     obj->tiling_mode, true);
-	unfenced_alignment =
-		i915_gem_get_gtt_alignment(dev,
-					   obj->base.size,
-					   obj->tiling_mode, false);
+		view_size = i915_ggtt_view_size(obj, ggtt_view);
+
+		fence_size = i915_gem_get_gtt_size(dev,
+						   view_size,
+						   obj->tiling_mode);
+		fence_alignment = i915_gem_get_gtt_alignment(dev,
+							     view_size,
+							     obj->tiling_mode,
+							     true);
+		unfenced_alignment = i915_gem_get_gtt_alignment(dev,
+								view_size,
+								obj->tiling_mode,
+								false);
+		size = flags & PIN_MAPPABLE ? fence_size : view_size;
+	} else {
+		fence_size = i915_gem_get_gtt_size(dev,
+						   obj->base.size,
+						   obj->tiling_mode);
+		fence_alignment = i915_gem_get_gtt_alignment(dev,
+							     obj->base.size,
+							     obj->tiling_mode,
+							     true);
+		unfenced_alignment =
+			i915_gem_get_gtt_alignment(dev,
+						   obj->base.size,
+						   obj->tiling_mode,
+						   false);
+		size = flags & PIN_MAPPABLE ? fence_size : obj->base.size;
+	}
 
 	if (alignment == 0)
 		alignment = flags & PIN_MAPPABLE ? fence_alignment :
 						unfenced_alignment;
 	if (flags & PIN_MAPPABLE && alignment & (fence_alignment - 1)) {
-		DRM_DEBUG("Invalid object alignment requested %u\n", alignment);
+		DRM_DEBUG("Invalid object (view type=%u) alignment requested %u\n",
+			  ggtt_view ? ggtt_view->type : 0,
+			  alignment);
 		return ERR_PTR(-EINVAL);
 	}
 
-	size = flags & PIN_MAPPABLE ? fence_size : obj->base.size;
-
-	/* If the object is bigger than the entire aperture, reject it early
-	 * before evicting everything in a vain attempt to find space.
+	/* If binding the object/GGTT view requires more space than the entire
+	 * aperture has, reject it early before evicting everything in a vain
+	 * attempt to find space.
 	 */
-	if (obj->base.size > end) {
-		DRM_DEBUG("Attempting to bind an object larger than the aperture: object=%zd > %s aperture=%lu\n",
-			  obj->base.size,
+	if (size > end) {
+		DRM_DEBUG("Attempting to bind an object (view type=%u) larger than the aperture: size=%u > %s aperture=%lu\n",
+			  ggtt_view ? ggtt_view->type : 0,
+			  size,
 			  flags & PIN_MAPPABLE ? "mappable" : "total",
 			  end);
 		return ERR_PTR(-E2BIG);
@@ -3515,20 +3772,8 @@ search_free:
 	if (ret)
 		goto err_remove_node;
 
-	/*  allocate before insert / bind */
-	if (vma->vm->allocate_va_range) {
-		trace_i915_va_alloc(vma->vm, vma->node.start, vma->node.size,
-				VM_TO_TRACE_NAME(vma->vm));
-		ret = vma->vm->allocate_va_range(vma->vm,
-						vma->node.start,
-						vma->node.size);
-		if (ret)
-			goto err_remove_node;
-	}
-
 	trace_i915_vma_bind(vma, flags);
-	ret = i915_vma_bind(vma, obj->cache_level,
-			    flags & PIN_GLOBAL ? GLOBAL_BIND : 0);
+	ret = i915_vma_bind(vma, obj->cache_level, flags);
 	if (ret)
 		goto err_finish_gtt;
 
@@ -3658,8 +3903,6 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 	if (ret)
 		return ret;
 
-	i915_gem_object_retire(obj);
-
 	/* Flush and acquire obj->pages so that we are coherent through
 	 * direct access in memory with previous cached writes through
 	 * shmemfs and that our cache domain tracking remains valid.
@@ -3735,7 +3978,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 	}
 
 	if (i915_gem_obj_bound_any(obj)) {
-		ret = i915_gem_object_finish_gpu(obj);
+		ret = i915_gem_object_wait_rendering(obj, false);
 		if (ret)
 			return ret;
 
@@ -3754,7 +3997,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 		list_for_each_entry(vma, &obj->vma_list, vma_link)
 			if (drm_mm_node_allocated(&vma->node)) {
 				ret = i915_vma_bind(vma, cache_level,
-						    vma->bound & GLOBAL_BIND);
+						    PIN_UPDATE);
 				if (ret)
 					return ret;
 			}
@@ -3779,17 +4022,10 @@ int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_i915_gem_caching *args = data;
 	struct drm_i915_gem_object *obj;
-	int ret;
-
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
 
 	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
-	if (&obj->base == NULL) {
-		ret = -ENOENT;
-		goto unlock;
-	}
+	if (&obj->base == NULL)
+		return -ENOENT;
 
 	switch (obj->cache_level) {
 	case I915_CACHE_LLC:
@@ -3806,10 +4042,8 @@ int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data,
 		break;
 	}
 
-	drm_gem_object_unreference(&obj->base);
-unlock:
-	mutex_unlock(&dev->struct_mutex);
-	return ret;
+	drm_gem_object_unreference_unlocked(&obj->base);
+	return 0;
 }
 
 int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
@@ -3852,24 +4086,6 @@ unlock:
 	return ret;
 }
 
-static bool is_pin_display(struct drm_i915_gem_object *obj)
-{
-	struct i915_vma *vma;
-
-	vma = i915_gem_obj_to_ggtt(obj);
-	if (!vma)
-		return false;
-
-	/* There are 2 sources that pin objects:
-	 *   1. The display engine (scanouts, sprites, cursors);
-	 *   2. Reservations for execbuffer;
-	 *
-	 * We can ignore reservations as we hold the struct_mutex and
-	 * are only called outside of the reservation path.
-	 */
-	return vma->pin_count;
-}
-
 /*
  * Prepare buffer for display plane (scanout, cursors, etc).
  * Can be called from an uninterruptible phase (modesetting) and allows
@@ -3882,20 +4098,16 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 				     const struct i915_ggtt_view *view)
 {
 	u32 old_read_domains, old_write_domain;
-	bool was_pin_display;
 	int ret;
 
-	if (pipelined != i915_gem_request_get_ring(obj->last_read_req)) {
-		ret = i915_gem_object_sync(obj, pipelined);
-		if (ret)
-			return ret;
-	}
+	ret = i915_gem_object_sync(obj, pipelined);
+	if (ret)
+		return ret;
 
 	/* Mark the pin_display early so that we account for the
 	 * display coherency whilst setting up the cache domains.
 	 */
-	was_pin_display = obj->pin_display;
-	obj->pin_display = true;
+	obj->pin_display++;
 
 	/* The display engine is not coherent with the LLC cache on gen6.  As
 	 * a result, we make sure that the pinning that is about to occur is
@@ -3939,8 +4151,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	return 0;
 
 err_unpin_display:
-	WARN_ON(was_pin_display != is_pin_display(obj));
-	obj->pin_display = was_pin_display;
+	obj->pin_display--;
 	return ret;
 }
 
@@ -3948,26 +4159,12 @@ void
 i915_gem_object_unpin_from_display_plane(struct drm_i915_gem_object *obj,
 					 const struct i915_ggtt_view *view)
 {
-	i915_gem_object_ggtt_unpin_view(obj, view);
-
-	obj->pin_display = is_pin_display(obj);
-}
-
-int
-i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj)
-{
-	int ret;
-
-	if ((obj->base.read_domains & I915_GEM_GPU_DOMAINS) == 0)
-		return 0;
+	if (WARN_ON(obj->pin_display == 0))
+		return;
 
-	ret = i915_gem_object_wait_rendering(obj, false);
-	if (ret)
-		return ret;
+	i915_gem_object_ggtt_unpin_view(obj, view);
 
-	/* Ensure that we invalidate the GPU's caches and TLBs. */
-	obj->base.read_domains &= ~I915_GEM_GPU_DOMAINS;
-	return 0;
+	obj->pin_display--;
 }
 
 /**
@@ -3989,7 +4186,6 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write)
 	if (ret)
 		return ret;
 
-	i915_gem_object_retire(obj);
 	i915_gem_object_flush_gtt_write_domain(obj);
 
 	old_write_domain = obj->base.write_domain;
@@ -4040,7 +4236,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
-	unsigned long recent_enough = jiffies - msecs_to_jiffies(20);
+	unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES;
 	struct drm_i915_gem_request *request, *target = NULL;
 	unsigned reset_counter;
 	int ret;
@@ -4072,9 +4268,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	if (ret == 0)
 		queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, 0);
 
-	mutex_lock(&dev->struct_mutex);
-	i915_gem_request_unreference(target);
-	mutex_unlock(&dev->struct_mutex);
+	i915_gem_request_unreference__unlocked(target);
 
 	return ret;
 }
@@ -4155,23 +4349,18 @@ i915_gem_object_do_pin(struct drm_i915_gem_object *obj,
 
 	bound = vma ? vma->bound : 0;
 	if (vma == NULL || !drm_mm_node_allocated(&vma->node)) {
-		/* In true PPGTT, bind has possibly changed PDEs, which
-		 * means we must do a context switch before the GPU can
-		 * accurately read some of the VMAs.
-		 */
 		vma = i915_gem_object_bind_to_vm(obj, vm, ggtt_view, alignment,
 						 flags);
 		if (IS_ERR(vma))
 			return PTR_ERR(vma);
-	}
-
-	if (flags & PIN_GLOBAL && !(vma->bound & GLOBAL_BIND)) {
-		ret = i915_vma_bind(vma, obj->cache_level, GLOBAL_BIND);
+	} else {
+		ret = i915_vma_bind(vma, obj->cache_level, flags);
 		if (ret)
 			return ret;
 	}
 
-	if ((bound ^ vma->bound) & GLOBAL_BIND) {
+	if (ggtt_view && ggtt_view->type == I915_GGTT_VIEW_NORMAL &&
+	    (bound ^ vma->bound) & GLOBAL_BIND) {
 		bool mappable, fenceable;
 		u32 fence_size, fence_alignment;
 
@@ -4190,14 +4379,11 @@ i915_gem_object_do_pin(struct drm_i915_gem_object *obj,
 			    dev_priv->gtt.mappable_end);
 
 		obj->map_and_fenceable = mappable && fenceable;
-	}
 
-	WARN_ON(flags & PIN_MAPPABLE && !obj->map_and_fenceable);
+		WARN_ON(flags & PIN_MAPPABLE && !obj->map_and_fenceable);
+	}
 
 	vma->pin_count++;
-	if (flags & PIN_MAPPABLE)
-		obj->pin_mappable |= true;
-
 	return 0;
 }
 
@@ -4235,8 +4421,7 @@ i915_gem_object_ggtt_unpin_view(struct drm_i915_gem_object *obj,
 	WARN_ON(vma->pin_count == 0);
 	WARN_ON(!i915_gem_obj_ggtt_bound_view(obj, view));
 
-	if (--vma->pin_count == 0 && view->type == I915_GGTT_VIEW_NORMAL)
-		obj->pin_mappable = false;
+	--vma->pin_count;
 }
 
 bool
@@ -4289,15 +4474,15 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
 	 * necessary flushes here.
 	 */
 	ret = i915_gem_object_flush_active(obj);
+	if (ret)
+		goto unref;
 
-	args->busy = obj->active;
-	if (obj->last_read_req) {
-		struct intel_engine_cs *ring;
-		BUILD_BUG_ON(I915_NUM_RINGS > 16);
-		ring = i915_gem_request_get_ring(obj->last_read_req);
-		args->busy |= intel_ring_flag(ring) << 16;
-	}
+	BUILD_BUG_ON(I915_NUM_RINGS > 16);
+	args->busy = obj->active << 16;
+	if (obj->last_write_req)
+		args->busy |= obj->last_write_req->ring->id;
 
+unref:
 	drm_gem_object_unreference(&obj->base);
 unlock:
 	mutex_unlock(&dev->struct_mutex);
@@ -4371,11 +4556,14 @@ unlock:
 void i915_gem_object_init(struct drm_i915_gem_object *obj,
 			  const struct drm_i915_gem_object_ops *ops)
 {
+	int i;
+
 	INIT_LIST_HEAD(&obj->global_list);
-	INIT_LIST_HEAD(&obj->ring_list);
+	for (i = 0; i < I915_NUM_RINGS; i++)
+		INIT_LIST_HEAD(&obj->ring_list[i]);
 	INIT_LIST_HEAD(&obj->obj_exec_link);
 	INIT_LIST_HEAD(&obj->vma_list);
-	INIT_LIST_HEAD(&obj->batch_pool_list);
+	INIT_LIST_HEAD(&obj->batch_pool_link);
 
 	obj->ops = ops;
 
@@ -4577,7 +4765,7 @@ void i915_gem_vma_destroy(struct i915_vma *vma)
 
 	list_del(&vma->vma_link);
 
-	kfree(vma);
+	kmem_cache_free(to_i915(vma->obj->base.dev)->vmas, vma);
 }
 
 static void
@@ -4864,12 +5052,12 @@ int i915_gem_init(struct drm_device *dev)
 	}
 
 	if (!i915.enable_execlists) {
-		dev_priv->gt.do_execbuf = i915_gem_ringbuffer_submission;
+		dev_priv->gt.execbuf_submit = i915_gem_ringbuffer_submission;
 		dev_priv->gt.init_rings = i915_gem_init_rings;
 		dev_priv->gt.cleanup_ring = intel_cleanup_ring_buffer;
 		dev_priv->gt.stop_ring = intel_stop_ring_buffer;
 	} else {
-		dev_priv->gt.do_execbuf = intel_execlists_submission;
+		dev_priv->gt.execbuf_submit = intel_execlists_submission;
 		dev_priv->gt.init_rings = intel_logical_rings_init;
 		dev_priv->gt.cleanup_ring = intel_logical_ring_cleanup;
 		dev_priv->gt.stop_ring = intel_logical_ring_stop;
@@ -4951,11 +5139,21 @@ i915_gem_load(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
 
-	dev_priv->slab =
+	dev_priv->objects =
 		kmem_cache_create("i915_gem_object",
 				  sizeof(struct drm_i915_gem_object), 0,
 				  SLAB_HWCACHE_ALIGN,
 				  NULL);
+	dev_priv->vmas =
+		kmem_cache_create("i915_gem_vma",
+				  sizeof(struct i915_vma), 0,
+				  SLAB_HWCACHE_ALIGN,
+				  NULL);
+	dev_priv->requests =
+		kmem_cache_create("i915_gem_request",
+				  sizeof(struct drm_i915_gem_request), 0,
+				  SLAB_HWCACHE_ALIGN,
+				  NULL);
 
 	INIT_LIST_HEAD(&dev_priv->vm_list);
 	i915_init_vm(dev_priv, &dev_priv->gtt.base);
@@ -4998,8 +5196,6 @@ i915_gem_load(struct drm_device *dev)
 
 	i915_gem_shrinker_init(dev_priv);
 
-	i915_gem_batch_pool_init(dev, &dev_priv->mm.batch_pool);
-
 	mutex_init(&dev_priv->fb_tracking.lock);
 }
 
@@ -5007,8 +5203,6 @@ void i915_gem_release(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
-	cancel_delayed_work_sync(&file_priv->mm.idle_work);
-
 	/* Clean up our request list when the client is going away, so that
 	 * later retire_requests won't dereference our soon-to-be-gone
 	 * file_priv.
@@ -5024,15 +5218,12 @@ void i915_gem_release(struct drm_device *dev, struct drm_file *file)
 		request->file_priv = NULL;
 	}
 	spin_unlock(&file_priv->mm.lock);
-}
 
-static void
-i915_gem_file_idle_work_handler(struct work_struct *work)
-{
-	struct drm_i915_file_private *file_priv =
-		container_of(work, typeof(*file_priv), mm.idle_work.work);
-
-	atomic_set(&file_priv->rps_wait_boost, false);
+	if (!list_empty(&file_priv->rps.link)) {
+		spin_lock(&to_i915(dev)->rps.client_lock);
+		list_del(&file_priv->rps.link);
+		spin_unlock(&to_i915(dev)->rps.client_lock);
+	}
 }
 
 int i915_gem_open(struct drm_device *dev, struct drm_file *file)
@@ -5049,11 +5240,10 @@ 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;
+	INIT_LIST_HEAD(&file_priv->rps.link);
 
 	spin_lock_init(&file_priv->mm.lock);
 	INIT_LIST_HEAD(&file_priv->mm.request_list);
-	INIT_DELAYED_WORK(&file_priv->mm.idle_work,
-			  i915_gem_file_idle_work_handler);
 
 	ret = i915_gem_context_open(dev, file);
 	if (ret)
@@ -5123,7 +5313,7 @@ i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
 		    i915_ggtt_view_equal(&vma->ggtt_view, view))
 			return vma->node.start;
 
-	WARN(1, "global vma for this object not found.\n");
+	WARN(1, "global vma for this object not found. (view=%u)\n", view->type);
 	return -1;
 }
 
@@ -5192,13 +5382,10 @@ unsigned long i915_gem_obj_size(struct drm_i915_gem_object *o,
 bool i915_gem_obj_is_pinned(struct drm_i915_gem_object *obj)
 {
 	struct i915_vma *vma;
-	list_for_each_entry(vma, &obj->vma_list, vma_link) {
-		if (i915_is_ggtt(vma->vm) &&
-		    vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
-			continue;
+	list_for_each_entry(vma, &obj->vma_list, vma_link)
 		if (vma->pin_count > 0)
 			return true;
-	}
+
 	return false;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gem_batch_pool.c b/drivers/gpu/drm/i915/i915_gem_batch_pool.c
index c690170a1c4f..7bf2f3f2968e 100644
--- a/drivers/gpu/drm/i915/i915_gem_batch_pool.c
+++ b/drivers/gpu/drm/i915/i915_gem_batch_pool.c
@@ -23,6 +23,7 @@
  */
 
 #include "i915_drv.h"
+#include "i915_gem_batch_pool.h"
 
 /**
  * DOC: batch pool
@@ -46,8 +47,12 @@
 void i915_gem_batch_pool_init(struct drm_device *dev,
 			      struct i915_gem_batch_pool *pool)
 {
+	int n;
+
 	pool->dev = dev;
-	INIT_LIST_HEAD(&pool->cache_list);
+
+	for (n = 0; n < ARRAY_SIZE(pool->cache_list); n++)
+		INIT_LIST_HEAD(&pool->cache_list[n]);
 }
 
 /**
@@ -58,33 +63,35 @@ void i915_gem_batch_pool_init(struct drm_device *dev,
  */
 void i915_gem_batch_pool_fini(struct i915_gem_batch_pool *pool)
 {
-	WARN_ON(!mutex_is_locked(&pool->dev->struct_mutex));
+	int n;
 
-	while (!list_empty(&pool->cache_list)) {
-		struct drm_i915_gem_object *obj =
-			list_first_entry(&pool->cache_list,
-					 struct drm_i915_gem_object,
-					 batch_pool_list);
+	WARN_ON(!mutex_is_locked(&pool->dev->struct_mutex));
 
-		WARN_ON(obj->active);
+	for (n = 0; n < ARRAY_SIZE(pool->cache_list); n++) {
+		while (!list_empty(&pool->cache_list[n])) {
+			struct drm_i915_gem_object *obj =
+				list_first_entry(&pool->cache_list[n],
+						 struct drm_i915_gem_object,
+						 batch_pool_link);
 
-		list_del_init(&obj->batch_pool_list);
-		drm_gem_object_unreference(&obj->base);
+			list_del(&obj->batch_pool_link);
+			drm_gem_object_unreference(&obj->base);
+		}
 	}
 }
 
 /**
- * i915_gem_batch_pool_get() - select a buffer from the pool
+ * i915_gem_batch_pool_get() - allocate a buffer from the pool
  * @pool: the batch buffer pool
  * @size: the minimum desired size of the returned buffer
  *
- * Finds or allocates a batch buffer in the pool with at least the requested
- * size. The caller is responsible for any domain, active/inactive, or
- * purgeability management for the returned buffer.
+ * Returns an inactive buffer from @pool with at least @size bytes,
+ * with the pages pinned. The caller must i915_gem_object_unpin_pages()
+ * on the returned object.
  *
  * Note: Callers must hold the struct_mutex
  *
- * Return: the selected batch buffer object
+ * Return: the buffer object or an error pointer
  */
 struct drm_i915_gem_object *
 i915_gem_batch_pool_get(struct i915_gem_batch_pool *pool,
@@ -92,46 +99,53 @@ i915_gem_batch_pool_get(struct i915_gem_batch_pool *pool,
 {
 	struct drm_i915_gem_object *obj = NULL;
 	struct drm_i915_gem_object *tmp, *next;
+	struct list_head *list;
+	int n;
 
 	WARN_ON(!mutex_is_locked(&pool->dev->struct_mutex));
 
-	list_for_each_entry_safe(tmp, next,
-			&pool->cache_list, batch_pool_list) {
-
+	/* Compute a power-of-two bucket, but throw everything greater than
+	 * 16KiB into the same bucket: i.e. the the buckets hold objects of
+	 * (1 page, 2 pages, 4 pages, 8+ pages).
+	 */
+	n = fls(size >> PAGE_SHIFT) - 1;
+	if (n >= ARRAY_SIZE(pool->cache_list))
+		n = ARRAY_SIZE(pool->cache_list) - 1;
+	list = &pool->cache_list[n];
+
+	list_for_each_entry_safe(tmp, next, list, batch_pool_link) {
+		/* The batches are strictly LRU ordered */
 		if (tmp->active)
-			continue;
+			break;
 
 		/* While we're looping, do some clean up */
 		if (tmp->madv == __I915_MADV_PURGED) {
-			list_del(&tmp->batch_pool_list);
+			list_del(&tmp->batch_pool_link);
 			drm_gem_object_unreference(&tmp->base);
 			continue;
 		}
 
-		/*
-		 * Select a buffer that is at least as big as needed
-		 * but not 'too much' bigger. A better way to do this
-		 * might be to bucket the pool objects based on size.
-		 */
-		if (tmp->base.size >= size &&
-		    tmp->base.size <= (2 * size)) {
+		if (tmp->base.size >= size) {
 			obj = tmp;
 			break;
 		}
 	}
 
-	if (!obj) {
+	if (obj == NULL) {
+		int ret;
+
 		obj = i915_gem_alloc_object(pool->dev, size);
-		if (!obj)
+		if (obj == NULL)
 			return ERR_PTR(-ENOMEM);
 
-		list_add_tail(&obj->batch_pool_list, &pool->cache_list);
-	}
-	else
-		/* Keep list in LRU order */
-		list_move_tail(&obj->batch_pool_list, &pool->cache_list);
+		ret = i915_gem_object_get_pages(obj);
+		if (ret)
+			return ERR_PTR(ret);
 
-	obj->madv = I915_MADV_WILLNEED;
+		obj->madv = I915_MADV_DONTNEED;
+	}
 
+	list_move_tail(&obj->batch_pool_link, list);
+	i915_gem_object_pin_pages(obj);
 	return obj;
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_batch_pool.h b/drivers/gpu/drm/i915/i915_gem_batch_pool.h
new file mode 100644
index 000000000000..848e90703eed
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_batch_pool.h
@@ -0,0 +1,42 @@
+/*
+ * Copyright © 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"),
+ * 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.
+ *
+ */
+
+#ifndef I915_GEM_BATCH_POOL_H
+#define I915_GEM_BATCH_POOL_H
+
+#include "i915_drv.h"
+
+struct i915_gem_batch_pool {
+	struct drm_device *dev;
+	struct list_head cache_list[4];
+};
+
+/* i915_gem_batch_pool.c */
+void i915_gem_batch_pool_init(struct drm_device *dev,
+			      struct i915_gem_batch_pool *pool);
+void i915_gem_batch_pool_fini(struct i915_gem_batch_pool *pool);
+struct drm_i915_gem_object*
+i915_gem_batch_pool_get(struct i915_gem_batch_pool *pool, size_t size);
+
+#endif /* I915_GEM_BATCH_POOL_H */
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index f3e84c44d009..8867818b1401 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -157,7 +157,9 @@ i915_gem_alloc_context_obj(struct drm_device *dev, size_t size)
 	struct drm_i915_gem_object *obj;
 	int ret;
 
-	obj = i915_gem_alloc_object(dev, size);
+	obj = i915_gem_object_create_stolen(dev, size);
+	if (obj == NULL)
+		obj = i915_gem_alloc_object(dev, size);
 	if (obj == NULL)
 		return ERR_PTR(-ENOMEM);
 
@@ -573,20 +575,12 @@ static inline bool should_skip_switch(struct intel_engine_cs *ring,
 				      struct intel_context *from,
 				      struct intel_context *to)
 {
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-
 	if (to->remap_slice)
 		return false;
 
-	if (to->ppgtt) {
-		if (from == to && !test_bit(ring->id,
-				&to->ppgtt->pd_dirty_rings))
-			return true;
-	} else if (dev_priv->mm.aliasing_ppgtt) {
-		if (from == to && !test_bit(ring->id,
-				&dev_priv->mm.aliasing_ppgtt->pd_dirty_rings))
-			return true;
-	}
+	if (to->ppgtt && from == to &&
+	    !(intel_ring_flag(ring) & to->ppgtt->pd_dirty_rings))
+		return true;
 
 	return false;
 }
@@ -636,7 +630,6 @@ static int do_switch(struct intel_engine_cs *ring,
 	struct intel_context *from = ring->last_context;
 	u32 hw_flags = 0;
 	bool uninitialized = false;
-	struct i915_vma *vma;
 	int ret, i;
 
 	if (from != NULL && ring == &dev_priv->ring[RCS]) {
@@ -673,7 +666,7 @@ static int do_switch(struct intel_engine_cs *ring,
 			goto unpin_out;
 
 		/* Doing a PD load always reloads the page dirs */
-		clear_bit(ring->id, &to->ppgtt->pd_dirty_rings);
+		to->ppgtt->pd_dirty_rings &= ~intel_ring_flag(ring);
 	}
 
 	if (ring != &dev_priv->ring[RCS]) {
@@ -694,16 +687,6 @@ static int do_switch(struct intel_engine_cs *ring,
 	if (ret)
 		goto unpin_out;
 
-	vma = i915_gem_obj_to_ggtt(to->legacy_hw_ctx.rcs_state);
-	if (!(vma->bound & GLOBAL_BIND)) {
-		ret = i915_vma_bind(vma,
-				    to->legacy_hw_ctx.rcs_state->cache_level,
-				    GLOBAL_BIND);
-		/* This shouldn't ever fail. */
-		if (WARN_ONCE(ret, "GGTT context bind failed!"))
-			goto unpin_out;
-	}
-
 	if (!to->legacy_hw_ctx.initialized) {
 		hw_flags |= MI_RESTORE_INHIBIT;
 		/* NB: If we inhibit the restore, the context is not allowed to
@@ -711,12 +694,14 @@ static int do_switch(struct intel_engine_cs *ring,
 		 * space. This means we must enforce that a page table load
 		 * occur when this occurs. */
 	} else if (to->ppgtt &&
-			test_and_clear_bit(ring->id, &to->ppgtt->pd_dirty_rings))
+		   (intel_ring_flag(ring) & to->ppgtt->pd_dirty_rings)) {
 		hw_flags |= MI_FORCE_RESTORE;
+		to->ppgtt->pd_dirty_rings &= ~intel_ring_flag(ring);
+	}
 
 	/* We should never emit switch_mm more than once */
 	WARN_ON(needs_pd_load_pre(ring, to) &&
-			needs_pd_load_post(ring, to, hw_flags));
+		needs_pd_load_post(ring, to, hw_flags));
 
 	ret = mi_set_context(ring, to, hw_flags);
 	if (ret)
@@ -768,8 +753,6 @@ static int do_switch(struct intel_engine_cs *ring,
 		 * swapped, but there is no way to do that yet.
 		 */
 		from->legacy_hw_ctx.rcs_state->dirty = 1;
-		BUG_ON(i915_gem_request_get_ring(
-			from->legacy_hw_ctx.rcs_state->last_read_req) != ring);
 
 		/* obj is kept alive until the next request by its active ref */
 		i915_gem_object_ggtt_unpin(from->legacy_hw_ctx.rcs_state);
diff --git a/drivers/gpu/drm/i915/i915_gem_debug.c b/drivers/gpu/drm/i915/i915_gem_debug.c
index f462d1b51d97..17299d04189f 100644
--- a/drivers/gpu/drm/i915/i915_gem_debug.c
+++ b/drivers/gpu/drm/i915/i915_gem_debug.c
@@ -34,82 +34,34 @@ int
 i915_verify_lists(struct drm_device *dev)
 {
 	static int warned;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_i915_gem_object *obj;
+	struct intel_engine_cs *ring;
 	int err = 0;
+	int i;
 
 	if (warned)
 		return 0;
 
-	list_for_each_entry(obj, &dev_priv->render_ring.active_list, list) {
-		if (obj->base.dev != dev ||
-		    !atomic_read(&obj->base.refcount.refcount)) {
-			DRM_ERROR("freed render active %p\n", obj);
-			err++;
-			break;
-		} else if (!obj->active ||
-			   (obj->base.read_domains & I915_GEM_GPU_DOMAINS) == 0) {
-			DRM_ERROR("invalid render active %p (a %d r %x)\n",
-				  obj,
-				  obj->active,
-				  obj->base.read_domains);
-			err++;
-		} else if (obj->base.write_domain && list_empty(&obj->gpu_write_list)) {
-			DRM_ERROR("invalid render active %p (w %x, gwl %d)\n",
-				  obj,
-				  obj->base.write_domain,
-				  !list_empty(&obj->gpu_write_list));
-			err++;
-		}
-	}
-
-	list_for_each_entry(obj, &dev_priv->mm.flushing_list, list) {
-		if (obj->base.dev != dev ||
-		    !atomic_read(&obj->base.refcount.refcount)) {
-			DRM_ERROR("freed flushing %p\n", obj);
-			err++;
-			break;
-		} else if (!obj->active ||
-			   (obj->base.write_domain & I915_GEM_GPU_DOMAINS) == 0 ||
-			   list_empty(&obj->gpu_write_list)) {
-			DRM_ERROR("invalid flushing %p (a %d w %x gwl %d)\n",
-				  obj,
-				  obj->active,
-				  obj->base.write_domain,
-				  !list_empty(&obj->gpu_write_list));
-			err++;
-		}
-	}
-
-	list_for_each_entry(obj, &dev_priv->mm.gpu_write_list, gpu_write_list) {
-		if (obj->base.dev != dev ||
-		    !atomic_read(&obj->base.refcount.refcount)) {
-			DRM_ERROR("freed gpu write %p\n", obj);
-			err++;
-			break;
-		} else if (!obj->active ||
-			   (obj->base.write_domain & I915_GEM_GPU_DOMAINS) == 0) {
-			DRM_ERROR("invalid gpu write %p (a %d w %x)\n",
-				  obj,
-				  obj->active,
-				  obj->base.write_domain);
-			err++;
-		}
-	}
-
-	list_for_each_entry(obj, &i915_gtt_vm->inactive_list, list) {
-		if (obj->base.dev != dev ||
-		    !atomic_read(&obj->base.refcount.refcount)) {
-			DRM_ERROR("freed inactive %p\n", obj);
-			err++;
-			break;
-		} else if (obj->pin_count || obj->active ||
-			   (obj->base.write_domain & I915_GEM_GPU_DOMAINS)) {
-			DRM_ERROR("invalid inactive %p (p %d a %d w %x)\n",
-				  obj,
-				  obj->pin_count, obj->active,
-				  obj->base.write_domain);
-			err++;
+	for_each_ring(ring, dev_priv, i) {
+		list_for_each_entry(obj, &ring->active_list, ring_list[ring->id]) {
+			if (obj->base.dev != dev ||
+			    !atomic_read(&obj->base.refcount.refcount)) {
+				DRM_ERROR("%s: freed active obj %p\n",
+					  ring->name, obj);
+				err++;
+				break;
+			} else if (!obj->active ||
+				   obj->last_read_req[ring->id] == NULL) {
+				DRM_ERROR("%s: invalid active obj %p\n",
+					  ring->name, obj);
+				err++;
+			} else if (obj->base.write_domain) {
+				DRM_ERROR("%s: invalid write obj %p (w %x)\n",
+					  ring->name,
+					  obj, obj->base.write_domain);
+				err++;
+			}
 		}
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index cc552a4c1f3b..a7fa14516cda 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -38,7 +38,6 @@
 #define  __EXEC_OBJECT_HAS_FENCE (1<<30)
 #define  __EXEC_OBJECT_NEEDS_MAP (1<<29)
 #define  __EXEC_OBJECT_NEEDS_BIAS (1<<28)
-#define  __EXEC_OBJECT_PURGEABLE (1<<27)
 
 #define BATCH_OFFSET_BIAS (256*1024)
 
@@ -225,12 +224,7 @@ i915_gem_execbuffer_unreserve_vma(struct i915_vma *vma)
 	if (entry->flags & __EXEC_OBJECT_HAS_PIN)
 		vma->pin_count--;
 
-	if (entry->flags & __EXEC_OBJECT_PURGEABLE)
-		obj->madv = I915_MADV_DONTNEED;
-
-	entry->flags &= ~(__EXEC_OBJECT_HAS_FENCE |
-			  __EXEC_OBJECT_HAS_PIN |
-			  __EXEC_OBJECT_PURGEABLE);
+	entry->flags &= ~(__EXEC_OBJECT_HAS_FENCE | __EXEC_OBJECT_HAS_PIN);
 }
 
 static void eb_destroy(struct eb_vmas *eb)
@@ -407,10 +401,9 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 	 * pipe_control writes because the gpu doesn't properly redirect them
 	 * through the ppgtt for non_secure batchbuffers. */
 	if (unlikely(IS_GEN6(dev) &&
-	    reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION &&
-	    !(target_vma->bound & GLOBAL_BIND))) {
+	    reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION)) {
 		ret = i915_vma_bind(target_vma, target_i915_obj->cache_level,
-				    GLOBAL_BIND);
+				    PIN_GLOBAL);
 		if (WARN_ONCE(ret, "Unexpected failure to bind target VMA!"))
 			return ret;
 	}
@@ -592,12 +585,13 @@ i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
 	uint64_t flags;
 	int ret;
 
-	flags = 0;
+	flags = PIN_USER;
+	if (entry->flags & EXEC_OBJECT_NEEDS_GTT)
+		flags |= PIN_GLOBAL;
+
 	if (!drm_mm_node_allocated(&vma->node)) {
 		if (entry->flags & __EXEC_OBJECT_NEEDS_MAP)
 			flags |= PIN_GLOBAL | PIN_MAPPABLE;
-		if (entry->flags & EXEC_OBJECT_NEEDS_GTT)
-			flags |= PIN_GLOBAL;
 		if (entry->flags & __EXEC_OBJECT_NEEDS_BIAS)
 			flags |= BATCH_OFFSET_BIAS | PIN_OFFSET_BIAS;
 	}
@@ -607,7 +601,7 @@ i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
 	    only_mappable_for_reloc(entry->flags))
 		ret = i915_gem_object_pin(obj, vma->vm,
 					  entry->alignment,
-					  flags & ~(PIN_GLOBAL | PIN_MAPPABLE));
+					  flags & ~PIN_MAPPABLE);
 	if (ret)
 		return ret;
 
@@ -896,6 +890,7 @@ static int
 i915_gem_execbuffer_move_to_gpu(struct intel_engine_cs *ring,
 				struct list_head *vmas)
 {
+	const unsigned other_rings = ~intel_ring_flag(ring);
 	struct i915_vma *vma;
 	uint32_t flush_domains = 0;
 	bool flush_chipset = false;
@@ -903,9 +898,12 @@ i915_gem_execbuffer_move_to_gpu(struct intel_engine_cs *ring,
 
 	list_for_each_entry(vma, vmas, exec_list) {
 		struct drm_i915_gem_object *obj = vma->obj;
-		ret = i915_gem_object_sync(obj, ring);
-		if (ret)
-			return ret;
+
+		if (obj->active & other_rings) {
+			ret = i915_gem_object_sync(obj, ring);
+			if (ret)
+				return ret;
+		}
 
 		if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
 			flush_chipset |= i915_gem_clflush_object(obj, false);
@@ -1143,12 +1141,11 @@ i915_gem_execbuffer_parse(struct intel_engine_cs *ring,
 			  u32 batch_len,
 			  bool is_master)
 {
-	struct drm_i915_private *dev_priv = to_i915(batch_obj->base.dev);
 	struct drm_i915_gem_object *shadow_batch_obj;
 	struct i915_vma *vma;
 	int ret;
 
-	shadow_batch_obj = i915_gem_batch_pool_get(&dev_priv->mm.batch_pool,
+	shadow_batch_obj = i915_gem_batch_pool_get(&ring->batch_pool,
 						   PAGE_ALIGN(batch_len));
 	if (IS_ERR(shadow_batch_obj))
 		return shadow_batch_obj;
@@ -1166,11 +1163,13 @@ i915_gem_execbuffer_parse(struct intel_engine_cs *ring,
 	if (ret)
 		goto err;
 
+	i915_gem_object_unpin_pages(shadow_batch_obj);
+
 	memset(shadow_exec_entry, 0, sizeof(*shadow_exec_entry));
 
 	vma = i915_gem_obj_to_ggtt(shadow_batch_obj);
 	vma->exec_entry = shadow_exec_entry;
-	vma->exec_entry->flags = __EXEC_OBJECT_PURGEABLE | __EXEC_OBJECT_HAS_PIN;
+	vma->exec_entry->flags = __EXEC_OBJECT_HAS_PIN;
 	drm_gem_object_reference(&shadow_batch_obj->base);
 	list_add_tail(&vma->exec_list, &eb->vmas);
 
@@ -1179,6 +1178,7 @@ i915_gem_execbuffer_parse(struct intel_engine_cs *ring,
 	return shadow_batch_obj;
 
 err:
+	i915_gem_object_unpin_pages(shadow_batch_obj);
 	if (ret == -EACCES) /* unhandled chained batch */
 		return batch_obj;
 	else
@@ -1252,12 +1252,8 @@ i915_gem_ringbuffer_submission(struct drm_device *dev, struct drm_file *file,
 	if (ret)
 		goto error;
 
-	if (ctx->ppgtt)
-		WARN(ctx->ppgtt->pd_dirty_rings & (1<<ring->id),
-			"%s didn't clear reload\n", ring->name);
-	else if (dev_priv->mm.aliasing_ppgtt)
-		WARN(dev_priv->mm.aliasing_ppgtt->pd_dirty_rings &
-			(1<<ring->id), "%s didn't clear reload\n", ring->name);
+	WARN(ctx->ppgtt && ctx->ppgtt->pd_dirty_rings & (1<<ring->id),
+	     "%s didn't clear reload\n", ring->name);
 
 	instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
 	instp_mask = I915_EXEC_CONSTANTS_MASK;
@@ -1549,33 +1545,39 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	}
 
 	if (i915_needs_cmd_parser(ring) && args->batch_len) {
-		batch_obj = i915_gem_execbuffer_parse(ring,
+		struct drm_i915_gem_object *parsed_batch_obj;
+
+		parsed_batch_obj = i915_gem_execbuffer_parse(ring,
 						      &shadow_exec_entry,
 						      eb,
 						      batch_obj,
 						      args->batch_start_offset,
 						      args->batch_len,
 						      file->is_master);
-		if (IS_ERR(batch_obj)) {
-			ret = PTR_ERR(batch_obj);
+		if (IS_ERR(parsed_batch_obj)) {
+			ret = PTR_ERR(parsed_batch_obj);
 			goto err;
 		}
 
 		/*
-		 * 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.
-		 *
-		 * FIXME: with aliasing ppgtt, buffers that should only
-		 * be in ggtt still end up in the aliasing ppgtt. remove
-		 * this check when that is fixed.
+		 * parsed_batch_obj == batch_obj means batch not fully parsed:
+		 * Accept, but don't promote to secure.
 		 */
-		if (USES_FULL_PPGTT(dev))
-			dispatch_flags |= I915_DISPATCH_SECURE;
 
-		exec_start = 0;
+		if (parsed_batch_obj != batch_obj) {
+			/*
+			 * Batch parsed and accepted:
+			 *
+			 * 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 on batches the
+			 * command parser has accepted.
+			 */
+			dispatch_flags |= I915_DISPATCH_SECURE;
+			exec_start = 0;
+			batch_obj = parsed_batch_obj;
+		}
 	}
 
 	batch_obj->base.pending_read_domains |= I915_GEM_DOMAIN_COMMAND;
@@ -1602,9 +1604,9 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	} else
 		exec_start += i915_gem_obj_offset(batch_obj, vm);
 
-	ret = dev_priv->gt.do_execbuf(dev, file, ring, ctx, args,
-				      &eb->vmas, batch_obj, exec_start,
-				      dispatch_flags);
+	ret = dev_priv->gt.execbuf_submit(dev, file, ring, ctx, args,
+					  &eb->vmas, batch_obj, exec_start,
+					  dispatch_flags);
 
 	/*
 	 * FIXME: We crucially rely upon the active tracking for the (ppgtt)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 0239fbff7bf7..619dad1b2386 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -92,14 +92,14 @@
  *
  */
 
+static int
+i915_get_ggtt_vma_pages(struct i915_vma *vma);
+
 const struct i915_ggtt_view i915_ggtt_view_normal;
 const struct i915_ggtt_view i915_ggtt_view_rotated = {
         .type = I915_GGTT_VIEW_ROTATED
 };
 
-static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv);
-static void chv_setup_private_ppat(struct drm_i915_private *dev_priv);
-
 static int sanitize_enable_ppgtt(struct drm_device *dev, int enable_ppgtt)
 {
 	bool has_aliasing_ppgtt;
@@ -146,14 +146,33 @@ static int sanitize_enable_ppgtt(struct drm_device *dev, int enable_ppgtt)
 		return has_aliasing_ppgtt ? 1 : 0;
 }
 
-static void ppgtt_bind_vma(struct i915_vma *vma,
-			   enum i915_cache_level cache_level,
-			   u32 flags);
-static void ppgtt_unbind_vma(struct i915_vma *vma);
+static int ppgtt_bind_vma(struct i915_vma *vma,
+			  enum i915_cache_level cache_level,
+			  u32 unused)
+{
+	u32 pte_flags = 0;
+
+	/* Currently applicable only to VLV */
+	if (vma->obj->gt_ro)
+		pte_flags |= PTE_READ_ONLY;
+
+	vma->vm->insert_entries(vma->vm, vma->obj->pages, vma->node.start,
+				cache_level, pte_flags);
 
-static inline gen8_pte_t gen8_pte_encode(dma_addr_t addr,
-					 enum i915_cache_level level,
-					 bool valid)
+	return 0;
+}
+
+static void ppgtt_unbind_vma(struct i915_vma *vma)
+{
+	vma->vm->clear_range(vma->vm,
+			     vma->node.start,
+			     vma->obj->base.size,
+			     true);
+}
+
+static gen8_pte_t gen8_pte_encode(dma_addr_t addr,
+				  enum i915_cache_level level,
+				  bool valid)
 {
 	gen8_pte_t pte = valid ? _PAGE_PRESENT | _PAGE_RW : 0;
 	pte |= addr;
@@ -173,9 +192,9 @@ static inline gen8_pte_t gen8_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
-static inline gen8_pde_t gen8_pde_encode(struct drm_device *dev,
-					  dma_addr_t addr,
-					  enum i915_cache_level level)
+static gen8_pde_t gen8_pde_encode(struct drm_device *dev,
+				  dma_addr_t addr,
+				  enum i915_cache_level level)
 {
 	gen8_pde_t pde = _PAGE_PRESENT | _PAGE_RW;
 	pde |= addr;
@@ -285,8 +304,8 @@ static gen6_pte_t iris_pte_encode(dma_addr_t addr,
 #define i915_dma_unmap_single(px, dev) \
 	__i915_dma_unmap_single((px)->daddr, dev)
 
-static inline void __i915_dma_unmap_single(dma_addr_t daddr,
-					struct drm_device *dev)
+static void __i915_dma_unmap_single(dma_addr_t daddr,
+				    struct drm_device *dev)
 {
 	struct device *device = &dev->pdev->dev;
 
@@ -307,9 +326,9 @@ static inline void __i915_dma_unmap_single(dma_addr_t daddr,
 #define i915_dma_map_single(px, dev) \
 	i915_dma_map_page_single((px)->page, (dev), &(px)->daddr)
 
-static inline int i915_dma_map_page_single(struct page *page,
-					   struct drm_device *dev,
-					   dma_addr_t *daddr)
+static int i915_dma_map_page_single(struct page *page,
+				    struct drm_device *dev,
+				    dma_addr_t *daddr)
 {
 	struct device *device = &dev->pdev->dev;
 
@@ -320,7 +339,7 @@ static inline int i915_dma_map_page_single(struct page *page,
 	return 0;
 }
 
-static void unmap_and_free_pt(struct i915_page_table_entry *pt,
+static void unmap_and_free_pt(struct i915_page_table *pt,
 			       struct drm_device *dev)
 {
 	if (WARN_ON(!pt->page))
@@ -332,9 +351,27 @@ static void unmap_and_free_pt(struct i915_page_table_entry *pt,
 	kfree(pt);
 }
 
-static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev)
+static void gen8_initialize_pt(struct i915_address_space *vm,
+			       struct i915_page_table *pt)
+{
+	gen8_pte_t *pt_vaddr, scratch_pte;
+	int i;
+
+	pt_vaddr = kmap_atomic(pt->page);
+	scratch_pte = gen8_pte_encode(vm->scratch.addr,
+				      I915_CACHE_LLC, true);
+
+	for (i = 0; i < GEN8_PTES; i++)
+		pt_vaddr[i] = scratch_pte;
+
+	if (!HAS_LLC(vm->dev))
+		drm_clflush_virt_range(pt_vaddr, PAGE_SIZE);
+	kunmap_atomic(pt_vaddr);
+}
+
+static struct i915_page_table *alloc_pt_single(struct drm_device *dev)
 {
-	struct i915_page_table_entry *pt;
+	struct i915_page_table *pt;
 	const size_t count = INTEL_INFO(dev)->gen >= 8 ?
 		GEN8_PTES : GEN6_PTES;
 	int ret = -ENOMEM;
@@ -369,78 +406,55 @@ fail_bitmap:
 	return ERR_PTR(ret);
 }
 
-/**
- * alloc_pt_range() - Allocate a multiple page tables
- * @pd:		The page directory which will have at least @count entries
- *		available to point to the allocated page tables.
- * @pde:	First page directory entry for which we are allocating.
- * @count:	Number of pages to allocate.
- * @dev:	DRM device.
- *
- * Allocates multiple page table pages and sets the appropriate entries in the
- * page table structure within the page directory. Function cleans up after
- * itself on any failures.
- *
- * Return: 0 if allocation succeeded.
- */
-static int alloc_pt_range(struct i915_page_directory_entry *pd, uint16_t pde, size_t count,
-			  struct drm_device *dev)
-{
-	int i, ret;
-
-	/* 512 is the max page tables per page_directory on any platform. */
-	if (WARN_ON(pde + count > I915_PDES))
-		return -EINVAL;
-
-	for (i = pde; i < pde + count; i++) {
-		struct i915_page_table_entry *pt = alloc_pt_single(dev);
-
-		if (IS_ERR(pt)) {
-			ret = PTR_ERR(pt);
-			goto err_out;
-		}
-		WARN(pd->page_table[i],
-		     "Leaking page directory entry %d (%p)\n",
-		     i, pd->page_table[i]);
-		pd->page_table[i] = pt;
-	}
-
-	return 0;
-
-err_out:
-	while (i-- > pde)
-		unmap_and_free_pt(pd->page_table[i], dev);
-	return ret;
-}
-
-static void unmap_and_free_pd(struct i915_page_directory_entry *pd)
+static void unmap_and_free_pd(struct i915_page_directory *pd,
+			      struct drm_device *dev)
 {
 	if (pd->page) {
+		i915_dma_unmap_single(pd, dev);
 		__free_page(pd->page);
+		kfree(pd->used_pdes);
 		kfree(pd);
 	}
 }
 
-static struct i915_page_directory_entry *alloc_pd_single(void)
+static struct i915_page_directory *alloc_pd_single(struct drm_device *dev)
 {
-	struct i915_page_directory_entry *pd;
+	struct i915_page_directory *pd;
+	int ret = -ENOMEM;
 
 	pd = kzalloc(sizeof(*pd), GFP_KERNEL);
 	if (!pd)
 		return ERR_PTR(-ENOMEM);
 
-	pd->page = alloc_page(GFP_KERNEL | __GFP_ZERO);
-	if (!pd->page) {
-		kfree(pd);
-		return ERR_PTR(-ENOMEM);
-	}
+	pd->used_pdes = kcalloc(BITS_TO_LONGS(I915_PDES),
+				sizeof(*pd->used_pdes), GFP_KERNEL);
+	if (!pd->used_pdes)
+		goto free_pd;
+
+	pd->page = alloc_page(GFP_KERNEL);
+	if (!pd->page)
+		goto free_bitmap;
+
+	ret = i915_dma_map_single(pd, dev);
+	if (ret)
+		goto free_page;
 
 	return pd;
+
+free_page:
+	__free_page(pd->page);
+free_bitmap:
+	kfree(pd->used_pdes);
+free_pd:
+	kfree(pd);
+
+	return ERR_PTR(ret);
 }
 
 /* Broadwell Page Directory Pointer Descriptors */
-static int gen8_write_pdp(struct intel_engine_cs *ring, unsigned entry,
-			   uint64_t val)
+static int gen8_write_pdp(struct intel_engine_cs *ring,
+			  unsigned entry,
+			  dma_addr_t addr)
 {
 	int ret;
 
@@ -452,10 +466,10 @@ static int gen8_write_pdp(struct intel_engine_cs *ring, unsigned entry,
 
 	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
 	intel_ring_emit(ring, GEN8_RING_PDP_UDW(ring, entry));
-	intel_ring_emit(ring, (u32)(val >> 32));
+	intel_ring_emit(ring, upper_32_bits(addr));
 	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
 	intel_ring_emit(ring, GEN8_RING_PDP_LDW(ring, entry));
-	intel_ring_emit(ring, (u32)(val));
+	intel_ring_emit(ring, lower_32_bits(addr));
 	intel_ring_advance(ring);
 
 	return 0;
@@ -466,12 +480,12 @@ static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
 {
 	int i, ret;
 
-	/* bit of a hack to find the actual last used pd */
-	int used_pd = ppgtt->num_pd_entries / I915_PDES;
-
-	for (i = used_pd - 1; i >= 0; i--) {
-		dma_addr_t addr = ppgtt->pdp.page_directory[i]->daddr;
-		ret = gen8_write_pdp(ring, i, addr);
+	for (i = GEN8_LEGACY_PDPES - 1; i >= 0; i--) {
+		struct i915_page_directory *pd = ppgtt->pdp.page_directory[i];
+		dma_addr_t pd_daddr = pd ? pd->daddr : ppgtt->scratch_pd->daddr;
+		/* The page directory might be NULL, but we need to clear out
+		 * whatever the previous context might have used. */
+		ret = gen8_write_pdp(ring, i, pd_daddr);
 		if (ret)
 			return ret;
 	}
@@ -497,8 +511,8 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 				      I915_CACHE_LLC, use_scratch);
 
 	while (num_entries) {
-		struct i915_page_directory_entry *pd;
-		struct i915_page_table_entry *pt;
+		struct i915_page_directory *pd;
+		struct i915_page_table *pt;
 		struct page *page_table;
 
 		if (WARN_ON(!ppgtt->pdp.page_directory[pdpe]))
@@ -559,8 +573,8 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 			break;
 
 		if (pt_vaddr == NULL) {
-			struct i915_page_directory_entry *pd = ppgtt->pdp.page_directory[pdpe];
-			struct i915_page_table_entry *pt = pd->page_table[pde];
+			struct i915_page_directory *pd = ppgtt->pdp.page_directory[pdpe];
+			struct i915_page_table *pt = pd->page_table[pde];
 			struct page *page_table = pt->page;
 
 			pt_vaddr = kmap_atomic(page_table);
@@ -588,14 +602,43 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 	}
 }
 
-static void gen8_free_page_tables(struct i915_page_directory_entry *pd, struct drm_device *dev)
+static void __gen8_do_map_pt(gen8_pde_t * const pde,
+			     struct i915_page_table *pt,
+			     struct drm_device *dev)
+{
+	gen8_pde_t entry =
+		gen8_pde_encode(dev, pt->daddr, I915_CACHE_LLC);
+	*pde = entry;
+}
+
+static void gen8_initialize_pd(struct i915_address_space *vm,
+			       struct i915_page_directory *pd)
+{
+	struct i915_hw_ppgtt *ppgtt =
+			container_of(vm, struct i915_hw_ppgtt, base);
+	gen8_pde_t *page_directory;
+	struct i915_page_table *pt;
+	int i;
+
+	page_directory = kmap_atomic(pd->page);
+	pt = ppgtt->scratch_pt;
+	for (i = 0; i < I915_PDES; i++)
+		/* Map the PDE to the page table */
+		__gen8_do_map_pt(page_directory + i, pt, vm->dev);
+
+	if (!HAS_LLC(vm->dev))
+		drm_clflush_virt_range(page_directory, PAGE_SIZE);
+	kunmap_atomic(page_directory);
+}
+
+static void gen8_free_page_tables(struct i915_page_directory *pd, struct drm_device *dev)
 {
 	int i;
 
 	if (!pd->page)
 		return;
 
-	for (i = 0; i < I915_PDES; i++) {
+	for_each_set_bit(i, pd->used_pdes, I915_PDES) {
 		if (WARN_ON(!pd->page_table[i]))
 			continue;
 
@@ -604,163 +647,287 @@ static void gen8_free_page_tables(struct i915_page_directory_entry *pd, struct d
 	}
 }
 
-static void gen8_ppgtt_free(struct i915_hw_ppgtt *ppgtt)
+static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 {
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
 	int i;
 
-	for (i = 0; i < ppgtt->num_pd_pages; i++) {
+	for_each_set_bit(i, ppgtt->pdp.used_pdpes, GEN8_LEGACY_PDPES) {
 		if (WARN_ON(!ppgtt->pdp.page_directory[i]))
 			continue;
 
 		gen8_free_page_tables(ppgtt->pdp.page_directory[i], ppgtt->base.dev);
-		unmap_and_free_pd(ppgtt->pdp.page_directory[i]);
+		unmap_and_free_pd(ppgtt->pdp.page_directory[i], ppgtt->base.dev);
 	}
+
+	unmap_and_free_pd(ppgtt->scratch_pd, ppgtt->base.dev);
+	unmap_and_free_pt(ppgtt->scratch_pt, ppgtt->base.dev);
 }
 
-static void gen8_ppgtt_unmap_pages(struct i915_hw_ppgtt *ppgtt)
+/**
+ * gen8_ppgtt_alloc_pagetabs() - Allocate page tables for VA range.
+ * @ppgtt:	Master ppgtt structure.
+ * @pd:		Page directory for this address range.
+ * @start:	Starting virtual address to begin allocations.
+ * @length	Size of the allocations.
+ * @new_pts:	Bitmap set by function with new allocations. Likely used by the
+ *		caller to free on error.
+ *
+ * Allocate the required number of page tables. Extremely similar to
+ * gen8_ppgtt_alloc_page_directories(). The main difference is here we are limited by
+ * the page directory boundary (instead of the page directory pointer). That
+ * boundary is 1GB virtual. Therefore, unlike gen8_ppgtt_alloc_page_directories(), it is
+ * possible, and likely that the caller will need to use multiple calls of this
+ * function to achieve the appropriate allocation.
+ *
+ * Return: 0 if success; negative error code otherwise.
+ */
+static int gen8_ppgtt_alloc_pagetabs(struct i915_hw_ppgtt *ppgtt,
+				     struct i915_page_directory *pd,
+				     uint64_t start,
+				     uint64_t length,
+				     unsigned long *new_pts)
 {
-	struct pci_dev *hwdev = ppgtt->base.dev->pdev;
-	int i, j;
-
-	for (i = 0; i < ppgtt->num_pd_pages; i++) {
-		/* TODO: In the future we'll support sparse mappings, so this
-		 * will have to change. */
-		if (!ppgtt->pdp.page_directory[i]->daddr)
+	struct drm_device *dev = ppgtt->base.dev;
+	struct i915_page_table *pt;
+	uint64_t temp;
+	uint32_t pde;
+
+	gen8_for_each_pde(pt, pd, start, length, temp, pde) {
+		/* Don't reallocate page tables */
+		if (pt) {
+			/* Scratch is never allocated this way */
+			WARN_ON(pt == ppgtt->scratch_pt);
 			continue;
+		}
 
-		pci_unmap_page(hwdev, ppgtt->pdp.page_directory[i]->daddr, PAGE_SIZE,
-			       PCI_DMA_BIDIRECTIONAL);
+		pt = alloc_pt_single(dev);
+		if (IS_ERR(pt))
+			goto unwind_out;
 
-		for (j = 0; j < I915_PDES; j++) {
-			struct i915_page_directory_entry *pd = ppgtt->pdp.page_directory[i];
-			struct i915_page_table_entry *pt;
-			dma_addr_t addr;
+		gen8_initialize_pt(&ppgtt->base, pt);
+		pd->page_table[pde] = pt;
+		set_bit(pde, new_pts);
+	}
 
-			if (WARN_ON(!pd->page_table[j]))
-				continue;
+	return 0;
 
-			pt = pd->page_table[j];
-			addr = pt->daddr;
+unwind_out:
+	for_each_set_bit(pde, new_pts, I915_PDES)
+		unmap_and_free_pt(pd->page_table[pde], dev);
 
-			if (addr)
-				pci_unmap_page(hwdev, addr, PAGE_SIZE,
-					       PCI_DMA_BIDIRECTIONAL);
-		}
-	}
+	return -ENOMEM;
 }
 
-static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
+/**
+ * gen8_ppgtt_alloc_page_directories() - Allocate page directories for VA range.
+ * @ppgtt:	Master ppgtt structure.
+ * @pdp:	Page directory pointer for this address range.
+ * @start:	Starting virtual address to begin allocations.
+ * @length	Size of the allocations.
+ * @new_pds	Bitmap set by function with new allocations. Likely used by the
+ *		caller to free on error.
+ *
+ * Allocate the required number of page directories starting at the pde index of
+ * @start, and ending at the pde index @start + @length. This function will skip
+ * over already allocated page directories within the range, and only allocate
+ * new ones, setting the appropriate pointer within the pdp as well as the
+ * correct position in the bitmap @new_pds.
+ *
+ * The function will only allocate the pages within the range for a give page
+ * directory pointer. In other words, if @start + @length straddles a virtually
+ * addressed PDP boundary (512GB for 4k pages), there will be more allocations
+ * required by the caller, This is not currently possible, and the BUG in the
+ * code will prevent it.
+ *
+ * Return: 0 if success; negative error code otherwise.
+ */
+static int gen8_ppgtt_alloc_page_directories(struct i915_hw_ppgtt *ppgtt,
+				     struct i915_page_directory_pointer *pdp,
+				     uint64_t start,
+				     uint64_t length,
+				     unsigned long *new_pds)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
+	struct drm_device *dev = ppgtt->base.dev;
+	struct i915_page_directory *pd;
+	uint64_t temp;
+	uint32_t pdpe;
 
-	gen8_ppgtt_unmap_pages(ppgtt);
-	gen8_ppgtt_free(ppgtt);
-}
+	WARN_ON(!bitmap_empty(new_pds, GEN8_LEGACY_PDPES));
 
-static int gen8_ppgtt_allocate_page_tables(struct i915_hw_ppgtt *ppgtt)
-{
-	int i, ret;
+	/* FIXME: upper bound must not overflow 32 bits  */
+	WARN_ON((start + length) > (1ULL << 32));
 
-	for (i = 0; i < ppgtt->num_pd_pages; i++) {
-		ret = alloc_pt_range(ppgtt->pdp.page_directory[i],
-				     0, I915_PDES, ppgtt->base.dev);
-		if (ret)
+	gen8_for_each_pdpe(pd, pdp, start, length, temp, pdpe) {
+		if (pd)
+			continue;
+
+		pd = alloc_pd_single(dev);
+		if (IS_ERR(pd))
 			goto unwind_out;
+
+		gen8_initialize_pd(&ppgtt->base, pd);
+		pdp->page_directory[pdpe] = pd;
+		set_bit(pdpe, new_pds);
 	}
 
 	return 0;
 
 unwind_out:
-	while (i--)
-		gen8_free_page_tables(ppgtt->pdp.page_directory[i], ppgtt->base.dev);
+	for_each_set_bit(pdpe, new_pds, GEN8_LEGACY_PDPES)
+		unmap_and_free_pd(pdp->page_directory[pdpe], dev);
 
 	return -ENOMEM;
 }
 
-static int gen8_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt,
-						const int max_pdp)
+static void
+free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long **new_pts)
 {
 	int i;
 
-	for (i = 0; i < max_pdp; i++) {
-		ppgtt->pdp.page_directory[i] = alloc_pd_single();
-		if (IS_ERR(ppgtt->pdp.page_directory[i]))
-			goto unwind_out;
+	for (i = 0; i < GEN8_LEGACY_PDPES; i++)
+		kfree(new_pts[i]);
+	kfree(new_pts);
+	kfree(new_pds);
+}
+
+/* Fills in the page directory bitmap, and the array of page tables bitmap. Both
+ * of these are based on the number of PDPEs in the system.
+ */
+static
+int __must_check alloc_gen8_temp_bitmaps(unsigned long **new_pds,
+					 unsigned long ***new_pts)
+{
+	int i;
+	unsigned long *pds;
+	unsigned long **pts;
+
+	pds = kcalloc(BITS_TO_LONGS(GEN8_LEGACY_PDPES), sizeof(unsigned long), GFP_KERNEL);
+	if (!pds)
+		return -ENOMEM;
+
+	pts = kcalloc(GEN8_LEGACY_PDPES, sizeof(unsigned long *), GFP_KERNEL);
+	if (!pts) {
+		kfree(pds);
+		return -ENOMEM;
 	}
 
-	ppgtt->num_pd_pages = max_pdp;
-	BUG_ON(ppgtt->num_pd_pages > GEN8_LEGACY_PDPES);
+	for (i = 0; i < GEN8_LEGACY_PDPES; i++) {
+		pts[i] = kcalloc(BITS_TO_LONGS(I915_PDES),
+				 sizeof(unsigned long), GFP_KERNEL);
+		if (!pts[i])
+			goto err_out;
+	}
 
-	return 0;
+	*new_pds = pds;
+	*new_pts = pts;
 
-unwind_out:
-	while (i--)
-		unmap_and_free_pd(ppgtt->pdp.page_directory[i]);
+	return 0;
 
+err_out:
+	free_gen8_temp_bitmaps(pds, pts);
 	return -ENOMEM;
 }
 
-static int gen8_ppgtt_alloc(struct i915_hw_ppgtt *ppgtt,
-			    const int max_pdp)
+static int gen8_alloc_va_range(struct i915_address_space *vm,
+			       uint64_t start,
+			       uint64_t length)
 {
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+	unsigned long *new_page_dirs, **new_page_tables;
+	struct i915_page_directory *pd;
+	const uint64_t orig_start = start;
+	const uint64_t orig_length = length;
+	uint64_t temp;
+	uint32_t pdpe;
 	int ret;
 
-	ret = gen8_ppgtt_allocate_page_directories(ppgtt, max_pdp);
+	/* Wrap is never okay since we can only represent 48b, and we don't
+	 * actually use the other side of the canonical address space.
+	 */
+	if (WARN_ON(start + length < start))
+		return -ERANGE;
+
+	ret = alloc_gen8_temp_bitmaps(&new_page_dirs, &new_page_tables);
 	if (ret)
 		return ret;
 
-	ret = gen8_ppgtt_allocate_page_tables(ppgtt);
-	if (ret)
-		goto err_out;
+	/* Do the allocations first so we can easily bail out */
+	ret = gen8_ppgtt_alloc_page_directories(ppgtt, &ppgtt->pdp, start, length,
+					new_page_dirs);
+	if (ret) {
+		free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
+		return ret;
+	}
 
-	ppgtt->num_pd_entries = max_pdp * I915_PDES;
+	/* For every page directory referenced, allocate page tables */
+	gen8_for_each_pdpe(pd, &ppgtt->pdp, start, length, temp, pdpe) {
+		ret = gen8_ppgtt_alloc_pagetabs(ppgtt, pd, start, length,
+						new_page_tables[pdpe]);
+		if (ret)
+			goto err_out;
+	}
 
-	return 0;
+	start = orig_start;
+	length = orig_length;
 
-err_out:
-	gen8_ppgtt_free(ppgtt);
-	return ret;
-}
+	/* Allocations have completed successfully, so set the bitmaps, and do
+	 * the mappings. */
+	gen8_for_each_pdpe(pd, &ppgtt->pdp, start, length, temp, pdpe) {
+		gen8_pde_t *const page_directory = kmap_atomic(pd->page);
+		struct i915_page_table *pt;
+		uint64_t pd_len = gen8_clamp_pd(start, length);
+		uint64_t pd_start = start;
+		uint32_t pde;
 
-static int gen8_ppgtt_setup_page_directories(struct i915_hw_ppgtt *ppgtt,
-					     const int pd)
-{
-	dma_addr_t pd_addr;
-	int ret;
+		/* Every pd should be allocated, we just did that above. */
+		WARN_ON(!pd);
 
-	pd_addr = pci_map_page(ppgtt->base.dev->pdev,
-			       ppgtt->pdp.page_directory[pd]->page, 0,
-			       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+		gen8_for_each_pde(pt, pd, pd_start, pd_len, temp, pde) {
+			/* Same reasoning as pd */
+			WARN_ON(!pt);
+			WARN_ON(!pd_len);
+			WARN_ON(!gen8_pte_count(pd_start, pd_len));
 
-	ret = pci_dma_mapping_error(ppgtt->base.dev->pdev, pd_addr);
-	if (ret)
-		return ret;
+			/* Set our used ptes within the page table */
+			bitmap_set(pt->used_ptes,
+				   gen8_pte_index(pd_start),
+				   gen8_pte_count(pd_start, pd_len));
 
-	ppgtt->pdp.page_directory[pd]->daddr = pd_addr;
+			/* Our pde is now pointing to the pagetable, pt */
+			set_bit(pde, pd->used_pdes);
 
-	return 0;
-}
+			/* Map the PDE to the page table */
+			__gen8_do_map_pt(page_directory + pde, pt, vm->dev);
 
-static int gen8_ppgtt_setup_page_tables(struct i915_hw_ppgtt *ppgtt,
-					const int pd,
-					const int pt)
-{
-	dma_addr_t pt_addr;
-	struct i915_page_directory_entry *pdir = ppgtt->pdp.page_directory[pd];
-	struct i915_page_table_entry *ptab = pdir->page_table[pt];
-	struct page *p = ptab->page;
-	int ret;
+			/* NB: We haven't yet mapped ptes to pages. At this
+			 * point we're still relying on insert_entries() */
+		}
 
-	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;
+		if (!HAS_LLC(vm->dev))
+			drm_clflush_virt_range(page_directory, PAGE_SIZE);
 
-	ptab->daddr = pt_addr;
+		kunmap_atomic(page_directory);
+
+		set_bit(pdpe, ppgtt->pdp.used_pdpes);
+	}
 
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
 	return 0;
+
+err_out:
+	while (pdpe--) {
+		for_each_set_bit(temp, new_page_tables[pdpe], I915_PDES)
+			unmap_and_free_pt(ppgtt->pdp.page_directory[pdpe]->page_table[temp], vm->dev);
+	}
+
+	for_each_set_bit(pdpe, new_page_dirs, GEN8_LEGACY_PDPES)
+		unmap_and_free_pd(ppgtt->pdp.page_directory[pdpe], vm->dev);
+
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
+	return ret;
 }
 
 /*
@@ -769,115 +936,57 @@ static int gen8_ppgtt_setup_page_tables(struct i915_hw_ppgtt *ppgtt,
  * 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)
+static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 {
-	const int max_pdp = DIV_ROUND_UP(size, 1 << 30);
-	const int min_pt_pages = I915_PDES * 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.
-	 * We allocate more than was asked so that we can point the unused parts
-	 * to valid entries that point to scratch page. Dynamic page tables
-	 * will fix this eventually.
-	 */
-	ret = gen8_ppgtt_alloc(ppgtt, GEN8_LEGACY_PDPES);
-	if (ret)
-		return ret;
+	ppgtt->scratch_pt = alloc_pt_single(ppgtt->base.dev);
+	if (IS_ERR(ppgtt->scratch_pt))
+		return PTR_ERR(ppgtt->scratch_pt);
 
-	/*
-	 * 2. Create DMA mappings for the page directories and page tables.
-	 */
-	for (i = 0; i < GEN8_LEGACY_PDPES; i++) {
-		ret = gen8_ppgtt_setup_page_directories(ppgtt, i);
-		if (ret)
-			goto bail;
+	ppgtt->scratch_pd = alloc_pd_single(ppgtt->base.dev);
+	if (IS_ERR(ppgtt->scratch_pd))
+		return PTR_ERR(ppgtt->scratch_pd);
 
-		for (j = 0; j < I915_PDES; j++) {
-			ret = gen8_ppgtt_setup_page_tables(ppgtt, i, j);
-			if (ret)
-				goto bail;
-		}
-	}
+	gen8_initialize_pt(&ppgtt->base, ppgtt->scratch_pt);
+	gen8_initialize_pd(&ppgtt->base, ppgtt->scratch_pd);
 
-	/*
-	 * 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.
-	 */
-	for (i = 0; i < GEN8_LEGACY_PDPES; i++) {
-		struct i915_page_directory_entry *pd = ppgtt->pdp.page_directory[i];
-		gen8_pde_t *pd_vaddr;
-		pd_vaddr = kmap_atomic(ppgtt->pdp.page_directory[i]->page);
-		for (j = 0; j < I915_PDES; j++) {
-			struct i915_page_table_entry *pt = pd->page_table[j];
-			dma_addr_t addr = pt->daddr;
-			pd_vaddr[j] = gen8_pde_encode(ppgtt->base.dev, addr,
-						      I915_CACHE_LLC);
-		}
-		if (!HAS_LLC(ppgtt->base.dev))
-			drm_clflush_virt_range(pd_vaddr, PAGE_SIZE);
-		kunmap_atomic(pd_vaddr);
-	}
-
-	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 = 1ULL << 32;
+	if (IS_ENABLED(CONFIG_X86_32))
+		/* While we have a proliferation of size_t variables
+		 * we cannot represent the full ppgtt size on 32bit,
+		 * so limit it to the same size as the GGTT (currently
+		 * 2GiB).
+		 */
+		ppgtt->base.total = to_i915(ppgtt->base.dev)->gtt.base.total;
+	ppgtt->base.cleanup = gen8_ppgtt_cleanup;
+	ppgtt->base.allocate_va_range = gen8_alloc_va_range;
+	ppgtt->base.insert_entries = gen8_ppgtt_insert_entries;
+	ppgtt->base.clear_range = gen8_ppgtt_clear_range;
+	ppgtt->base.unbind_vma = ppgtt_unbind_vma;
+	ppgtt->base.bind_vma = ppgtt_bind_vma;
 
-	/* This is the area that we advertise as usable for the caller */
-	ppgtt->base.total = max_pdp * I915_PDES * GEN8_PTES * PAGE_SIZE;
-
-	/* Set all ptes to a valid scratch page. Also above requested space */
-	ppgtt->base.clear_range(&ppgtt->base, 0,
-				ppgtt->num_pd_pages * GEN8_PTES * PAGE_SIZE,
-				true);
+	ppgtt->switch_mm = gen8_mm_switch;
 
-	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_pd_entries,
-			 (ppgtt->num_pd_entries - min_pt_pages) + size % (1<<30));
 	return 0;
-
-bail:
-	gen8_ppgtt_unmap_pages(ppgtt);
-	gen8_ppgtt_free(ppgtt);
-	return ret;
 }
 
 static void gen6_dump_ppgtt(struct i915_hw_ppgtt *ppgtt, struct seq_file *m)
 {
-	struct drm_i915_private *dev_priv = ppgtt->base.dev->dev_private;
 	struct i915_address_space *vm = &ppgtt->base;
-	gen6_pte_t __iomem *pd_addr;
+	struct i915_page_table *unused;
 	gen6_pte_t scratch_pte;
 	uint32_t pd_entry;
-	int pte, pde;
+	uint32_t  pte, pde, temp;
+	uint32_t start = ppgtt->base.start, length = ppgtt->base.total;
 
 	scratch_pte = vm->pte_encode(vm->scratch.addr, I915_CACHE_LLC, true, 0);
 
-	pd_addr = (gen6_pte_t __iomem *)dev_priv->gtt.gsm +
-		ppgtt->pd.pd_offset / sizeof(gen6_pte_t);
-
-	seq_printf(m, "  VM %p (pd_offset %x-%x):\n", vm,
-		   ppgtt->pd.pd_offset,
-		   ppgtt->pd.pd_offset + ppgtt->num_pd_entries);
-	for (pde = 0; pde < ppgtt->num_pd_entries; pde++) {
+	gen6_for_each_pde(unused, &ppgtt->pd, start, length, temp, pde) {
 		u32 expected;
 		gen6_pte_t *pt_vaddr;
 		dma_addr_t pt_addr = ppgtt->pd.page_table[pde]->daddr;
-		pd_entry = readl(pd_addr + pde);
+		pd_entry = readl(ppgtt->pd_addr + pde);
 		expected = (GEN6_PDE_ADDR_ENCODE(pt_addr) | GEN6_PDE_VALID);
 
 		if (pd_entry != expected)
@@ -914,8 +1023,8 @@ static void gen6_dump_ppgtt(struct i915_hw_ppgtt *ppgtt, struct seq_file *m)
 }
 
 /* Write pde (index) from the page directory @pd to the page table @pt */
-static void gen6_write_pde(struct i915_page_directory_entry *pd,
-			    const int pde, struct i915_page_table_entry *pt)
+static void gen6_write_pde(struct i915_page_directory *pd,
+			    const int pde, struct i915_page_table *pt)
 {
 	/* Caller needs to make sure the write completes if necessary */
 	struct i915_hw_ppgtt *ppgtt =
@@ -931,10 +1040,10 @@ static void gen6_write_pde(struct i915_page_directory_entry *pd,
 /* Write all the page tables found in the ppgtt structure to incrementing page
  * directories. */
 static void gen6_write_page_range(struct drm_i915_private *dev_priv,
-				  struct i915_page_directory_entry *pd,
+				  struct i915_page_directory *pd,
 				  uint32_t start, uint32_t length)
 {
-	struct i915_page_table_entry *pt;
+	struct i915_page_table *pt;
 	uint32_t pde, temp;
 
 	gen6_for_each_pde(pt, pd, start, length, temp, pde)
@@ -1162,14 +1271,14 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
  * are switching between contexts with the same LRCA, we also must do a force
  * restore.
  */
-static inline void mark_tlbs_dirty(struct i915_hw_ppgtt *ppgtt)
+static void mark_tlbs_dirty(struct i915_hw_ppgtt *ppgtt)
 {
 	/* If current vm != vm, */
 	ppgtt->pd_dirty_rings = INTEL_INFO(ppgtt->base.dev)->ring_mask;
 }
 
 static void gen6_initialize_pt(struct i915_address_space *vm,
-		struct i915_page_table_entry *pt)
+		struct i915_page_table *pt)
 {
 	gen6_pte_t *pt_vaddr, scratch_pte;
 	int i;
@@ -1195,7 +1304,7 @@ static int gen6_alloc_va_range(struct i915_address_space *vm,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_hw_ppgtt *ppgtt =
 				container_of(vm, struct i915_hw_ppgtt, base);
-	struct i915_page_table_entry *pt;
+	struct i915_page_table *pt;
 	const uint32_t start_save = start, length_save = length;
 	uint32_t pde, temp;
 	int ret;
@@ -1263,7 +1372,7 @@ static int gen6_alloc_va_range(struct i915_address_space *vm,
 
 unwind_out:
 	for_each_set_bit(pde, new_page_tables, I915_PDES) {
-		struct i915_page_table_entry *pt = ppgtt->pd.page_table[pde];
+		struct i915_page_table *pt = ppgtt->pd.page_table[pde];
 
 		ppgtt->pd.page_table[pde] = ppgtt->scratch_pt;
 		unmap_and_free_pt(pt, vm->dev);
@@ -1273,29 +1382,23 @@ unwind_out:
 	return ret;
 }
 
-static void gen6_ppgtt_free(struct i915_hw_ppgtt *ppgtt)
-{
-	int i;
-
-	for (i = 0; i < ppgtt->num_pd_entries; i++) {
-		struct i915_page_table_entry *pt = ppgtt->pd.page_table[i];
-
-		if (pt != ppgtt->scratch_pt)
-			unmap_and_free_pt(ppgtt->pd.page_table[i], ppgtt->base.dev);
-	}
-
-	unmap_and_free_pt(ppgtt->scratch_pt, ppgtt->base.dev);
-	unmap_and_free_pd(&ppgtt->pd);
-}
-
 static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_page_table *pt;
+	uint32_t pde;
+
 
 	drm_mm_remove_node(&ppgtt->node);
 
-	gen6_ppgtt_free(ppgtt);
+	gen6_for_all_pdes(pt, ppgtt, pde) {
+		if (pt != ppgtt->scratch_pt)
+			unmap_and_free_pt(pt, ppgtt->base.dev);
+	}
+
+	unmap_and_free_pt(ppgtt->scratch_pt, ppgtt->base.dev);
+	unmap_and_free_pd(&ppgtt->pd, ppgtt->base.dev);
 }
 
 static int gen6_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt)
@@ -1342,7 +1445,6 @@ alloc:
 	if (ppgtt->node.start < dev_priv->gtt.mappable_end)
 		DRM_DEBUG("Forced to use aperture for PDEs\n");
 
-	ppgtt->num_pd_entries = I915_PDES;
 	return 0;
 
 err_out:
@@ -1358,14 +1460,14 @@ static int gen6_ppgtt_alloc(struct i915_hw_ppgtt *ppgtt)
 static void gen6_scratch_va_range(struct i915_hw_ppgtt *ppgtt,
 				  uint64_t start, uint64_t length)
 {
-	struct i915_page_table_entry *unused;
+	struct i915_page_table *unused;
 	uint32_t pde, temp;
 
 	gen6_for_each_pde(unused, &ppgtt->pd, start, length, temp, pde)
 		ppgtt->pd.page_table[pde] = ppgtt->scratch_pt;
 }
 
-static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt, bool aliasing)
+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;
@@ -1388,23 +1490,14 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt, bool aliasing)
 	if (ret)
 		return ret;
 
-	if (aliasing) {
-		/* preallocate all pts */
-		ret = alloc_pt_range(&ppgtt->pd, 0, ppgtt->num_pd_entries,
-				ppgtt->base.dev);
-
-		if (ret) {
-			gen6_ppgtt_cleanup(&ppgtt->base);
-			return ret;
-		}
-	}
-
 	ppgtt->base.allocate_va_range = gen6_alloc_va_range;
 	ppgtt->base.clear_range = gen6_ppgtt_clear_range;
 	ppgtt->base.insert_entries = gen6_ppgtt_insert_entries;
+	ppgtt->base.unbind_vma = ppgtt_unbind_vma;
+	ppgtt->base.bind_vma = ppgtt_bind_vma;
 	ppgtt->base.cleanup = gen6_ppgtt_cleanup;
 	ppgtt->base.start = 0;
-	ppgtt->base.total = ppgtt->num_pd_entries * GEN6_PTES * PAGE_SIZE;
+	ppgtt->base.total = I915_PDES * GEN6_PTES * PAGE_SIZE;
 	ppgtt->debug_dump = gen6_dump_ppgtt;
 
 	ppgtt->pd.pd_offset =
@@ -1413,10 +1506,7 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt, bool aliasing)
 	ppgtt->pd_addr = (gen6_pte_t __iomem *)dev_priv->gtt.gsm +
 		ppgtt->pd.pd_offset / sizeof(gen6_pte_t);
 
-	if (aliasing)
-		ppgtt->base.clear_range(&ppgtt->base, 0, ppgtt->base.total, true);
-	else
-		gen6_scratch_va_range(ppgtt, 0, ppgtt->base.total);
+	gen6_scratch_va_range(ppgtt, 0, ppgtt->base.total);
 
 	gen6_write_page_range(dev_priv, &ppgtt->pd, 0, ppgtt->base.total);
 
@@ -1430,8 +1520,7 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt, bool aliasing)
 	return 0;
 }
 
-static int __hw_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt,
-		bool aliasing)
+static int __hw_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -1439,16 +1528,16 @@ static int __hw_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt,
 	ppgtt->base.scratch = dev_priv->gtt.base.scratch;
 
 	if (INTEL_INFO(dev)->gen < 8)
-		return gen6_ppgtt_init(ppgtt, aliasing);
+		return gen6_ppgtt_init(ppgtt);
 	else
-		return gen8_ppgtt_init(ppgtt, dev_priv->gtt.base.total);
+		return gen8_ppgtt_init(ppgtt);
 }
 int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret = 0;
 
-	ret = __hw_ppgtt_init(dev, ppgtt, false);
+	ret = __hw_ppgtt_init(dev, ppgtt);
 	if (ret == 0) {
 		kref_init(&ppgtt->ref);
 		drm_mm_init(&ppgtt->base.mm, ppgtt->base.start,
@@ -1535,32 +1624,11 @@ void  i915_ppgtt_release(struct kref *kref)
 	kfree(ppgtt);
 }
 
-static void
-ppgtt_bind_vma(struct i915_vma *vma,
-	       enum i915_cache_level cache_level,
-	       u32 flags)
-{
-	/* Currently applicable only to VLV */
-	if (vma->obj->gt_ro)
-		flags |= PTE_READ_ONLY;
-
-	vma->vm->insert_entries(vma->vm, vma->obj->pages, vma->node.start,
-				cache_level, flags);
-}
-
-static void ppgtt_unbind_vma(struct i915_vma *vma)
-{
-	vma->vm->clear_range(vma->vm,
-			     vma->node.start,
-			     vma->obj->base.size,
-			     true);
-}
-
 extern int intel_iommu_gfx_mapped;
 /* Certain Gen5 chipsets require require idling the GPU before
  * unmapping anything from the GTT when VT-d is enabled.
  */
-static inline bool needs_idle_maps(struct drm_device *dev)
+static bool needs_idle_maps(struct drm_device *dev)
 {
 #ifdef CONFIG_INTEL_IOMMU
 	/* Query intel_iommu to see if we need the workaround. Presumably that
@@ -1653,67 +1721,6 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev)
 	i915_ggtt_flush(dev_priv);
 }
 
-void i915_gem_restore_gtt_mappings(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_i915_gem_object *obj;
-	struct i915_address_space *vm;
-
-	i915_check_and_clear_faults(dev);
-
-	/* First fill our portion of the GTT with scratch pages */
-	dev_priv->gtt.base.clear_range(&dev_priv->gtt.base,
-				       dev_priv->gtt.base.start,
-				       dev_priv->gtt.base.total,
-				       true);
-
-	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		struct i915_vma *vma = i915_gem_obj_to_vma(obj,
-							   &dev_priv->gtt.base);
-		if (!vma)
-			continue;
-
-		i915_gem_clflush_object(obj, obj->pin_display);
-		/* The bind_vma code tries to be smart about tracking mappings.
-		 * Unfortunately above, we've just wiped out the mappings
-		 * without telling our object about it. So we need to fake it.
-		 *
-		 * Bind is not expected to fail since this is only called on
-		 * resume and assumption is all requirements exist already.
-		 */
-		vma->bound &= ~GLOBAL_BIND;
-		WARN_ON(i915_vma_bind(vma, obj->cache_level, GLOBAL_BIND));
-	}
-
-
-	if (INTEL_INFO(dev)->gen >= 8) {
-		if (IS_CHERRYVIEW(dev))
-			chv_setup_private_ppat(dev_priv);
-		else
-			bdw_setup_private_ppat(dev_priv);
-
-		return;
-	}
-
-	if (USES_PPGTT(dev)) {
-		list_for_each_entry(vm, &dev_priv->vm_list, global_link) {
-			/* TODO: Perhaps it shouldn't be gen6 specific */
-
-			struct i915_hw_ppgtt *ppgtt =
-					container_of(vm, struct i915_hw_ppgtt,
-						     base);
-
-			if (i915_is_ggtt(vm))
-				ppgtt = dev_priv->mm.aliasing_ppgtt;
-
-			gen6_write_page_range(dev_priv, &ppgtt->pd,
-					      0, ppgtt->base.total);
-		}
-	}
-
-	i915_ggtt_flush(dev_priv);
-}
-
 int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj)
 {
 	if (obj->has_dma_mapping)
@@ -1727,7 +1734,7 @@ int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj)
 	return 0;
 }
 
-static inline void gen8_set_pte(void __iomem *addr, gen8_pte_t pte)
+static void gen8_set_pte(void __iomem *addr, gen8_pte_t pte)
 {
 #ifdef writeq
 	writeq(pte, addr);
@@ -1872,18 +1879,16 @@ static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 	readl(gtt_base);
 }
 
-
-static void i915_ggtt_bind_vma(struct i915_vma *vma,
-			       enum i915_cache_level cache_level,
-			       u32 unused)
+static void i915_ggtt_insert_entries(struct i915_address_space *vm,
+				     struct sg_table *pages,
+				     uint64_t start,
+				     enum i915_cache_level cache_level, u32 unused)
 {
-	const unsigned long entry = vma->node.start >> PAGE_SHIFT;
 	unsigned int flags = (cache_level == I915_CACHE_NONE) ?
 		AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY;
 
-	BUG_ON(!i915_is_ggtt(vma->vm));
-	intel_gtt_insert_sg_entries(vma->ggtt_view.pages, entry, flags);
-	vma->bound = GLOBAL_BIND;
+	intel_gtt_insert_sg_entries(pages, start >> PAGE_SHIFT, flags);
+
 }
 
 static void i915_ggtt_clear_range(struct i915_address_space *vm,
@@ -1896,62 +1901,41 @@ static void i915_ggtt_clear_range(struct i915_address_space *vm,
 	intel_gtt_clear_range(first_entry, num_entries);
 }
 
-static void i915_ggtt_unbind_vma(struct i915_vma *vma)
-{
-	const unsigned int first = vma->node.start >> PAGE_SHIFT;
-	const unsigned int size = vma->obj->base.size >> PAGE_SHIFT;
-
-	BUG_ON(!i915_is_ggtt(vma->vm));
-	vma->bound = 0;
-	intel_gtt_clear_range(first, size);
-}
-
-static void ggtt_bind_vma(struct i915_vma *vma,
-			  enum i915_cache_level cache_level,
-			  u32 flags)
+static int ggtt_bind_vma(struct i915_vma *vma,
+			 enum i915_cache_level cache_level,
+			 u32 flags)
 {
 	struct drm_device *dev = vma->vm->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj = vma->obj;
 	struct sg_table *pages = obj->pages;
+	u32 pte_flags = 0;
+	int ret;
+
+	ret = i915_get_ggtt_vma_pages(vma);
+	if (ret)
+		return ret;
+	pages = vma->ggtt_view.pages;
 
 	/* Currently applicable only to VLV */
 	if (obj->gt_ro)
-		flags |= PTE_READ_ONLY;
+		pte_flags |= PTE_READ_ONLY;
 
-	if (i915_is_ggtt(vma->vm))
-		pages = vma->ggtt_view.pages;
 
-	/* 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
-	 * changed, set the global PTEs.
-	 *
-	 * If there is an aliasing PPGTT it is anecdotally faster, so use that
-	 * instead if none of the above hold true.
-	 *
-	 * NB: A global mapping should only be needed for special regions like
-	 * "gtt mappable", SNB errata, or if specified via special execbuf
-	 * flags. At all other times, the GPU will use the aliasing PPGTT.
-	 */
 	if (!dev_priv->mm.aliasing_ppgtt || flags & GLOBAL_BIND) {
-		if (!(vma->bound & GLOBAL_BIND) ||
-		    (cache_level != obj->cache_level)) {
-			vma->vm->insert_entries(vma->vm, pages,
-						vma->node.start,
-						cache_level, flags);
-			vma->bound |= GLOBAL_BIND;
-		}
+		vma->vm->insert_entries(vma->vm, pages,
+					vma->node.start,
+					cache_level, pte_flags);
 	}
 
-	if (dev_priv->mm.aliasing_ppgtt &&
-	    (!(vma->bound & LOCAL_BIND) ||
-	     (cache_level != obj->cache_level))) {
+	if (dev_priv->mm.aliasing_ppgtt && flags & LOCAL_BIND) {
 		struct i915_hw_ppgtt *appgtt = dev_priv->mm.aliasing_ppgtt;
 		appgtt->base.insert_entries(&appgtt->base, pages,
 					    vma->node.start,
-					    cache_level, flags);
-		vma->bound |= LOCAL_BIND;
+					    cache_level, pte_flags);
 	}
+
+	return 0;
 }
 
 static void ggtt_unbind_vma(struct i915_vma *vma)
@@ -1959,22 +1943,24 @@ 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 uint64_t size = min_t(uint64_t,
+				    obj->base.size,
+				    vma->node.size);
 
 	if (vma->bound & GLOBAL_BIND) {
 		vma->vm->clear_range(vma->vm,
 				     vma->node.start,
-				     obj->base.size,
+				     size,
 				     true);
-		vma->bound &= ~GLOBAL_BIND;
 	}
 
-	if (vma->bound & LOCAL_BIND) {
+	if (dev_priv->mm.aliasing_ppgtt && vma->bound & LOCAL_BIND) {
 		struct i915_hw_ppgtt *appgtt = dev_priv->mm.aliasing_ppgtt;
+
 		appgtt->base.clear_range(&appgtt->base,
 					 vma->node.start,
-					 obj->base.size,
+					 size,
 					 true);
-		vma->bound &= ~LOCAL_BIND;
 	}
 }
 
@@ -2083,12 +2069,27 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 		if (!ppgtt)
 			return -ENOMEM;
 
-		ret = __hw_ppgtt_init(dev, ppgtt, true);
+		ret = __hw_ppgtt_init(dev, ppgtt);
+		if (ret) {
+			ppgtt->base.cleanup(&ppgtt->base);
+			kfree(ppgtt);
+			return ret;
+		}
+
+		if (ppgtt->base.allocate_va_range)
+			ret = ppgtt->base.allocate_va_range(&ppgtt->base, 0,
+							    ppgtt->base.total);
 		if (ret) {
+			ppgtt->base.cleanup(&ppgtt->base);
 			kfree(ppgtt);
 			return ret;
 		}
 
+		ppgtt->base.clear_range(&ppgtt->base,
+					ppgtt->base.start,
+					ppgtt->base.total,
+					true);
+
 		dev_priv->mm.aliasing_ppgtt = ppgtt;
 	}
 
@@ -2164,14 +2165,14 @@ static void teardown_scratch_page(struct drm_device *dev)
 	__free_page(page);
 }
 
-static inline unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
+static unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
 {
 	snb_gmch_ctl >>= SNB_GMCH_GGMS_SHIFT;
 	snb_gmch_ctl &= SNB_GMCH_GGMS_MASK;
 	return snb_gmch_ctl << 20;
 }
 
-static inline unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
+static unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
 {
 	bdw_gmch_ctl >>= BDW_GMCH_GGMS_SHIFT;
 	bdw_gmch_ctl &= BDW_GMCH_GGMS_MASK;
@@ -2187,7 +2188,7 @@ static inline unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
 	return bdw_gmch_ctl << 20;
 }
 
-static inline unsigned int chv_get_total_gtt_size(u16 gmch_ctrl)
+static unsigned int chv_get_total_gtt_size(u16 gmch_ctrl)
 {
 	gmch_ctrl >>= SNB_GMCH_GGMS_SHIFT;
 	gmch_ctrl &= SNB_GMCH_GGMS_MASK;
@@ -2198,14 +2199,14 @@ static inline unsigned int chv_get_total_gtt_size(u16 gmch_ctrl)
 	return 0;
 }
 
-static inline size_t gen6_get_stolen_size(u16 snb_gmch_ctl)
+static size_t gen6_get_stolen_size(u16 snb_gmch_ctl)
 {
 	snb_gmch_ctl >>= SNB_GMCH_GMS_SHIFT;
 	snb_gmch_ctl &= SNB_GMCH_GMS_MASK;
 	return snb_gmch_ctl << 25; /* 32 MB units */
 }
 
-static inline size_t gen8_get_stolen_size(u16 bdw_gmch_ctl)
+static size_t gen8_get_stolen_size(u16 bdw_gmch_ctl)
 {
 	bdw_gmch_ctl >>= BDW_GMCH_GMS_SHIFT;
 	bdw_gmch_ctl &= BDW_GMCH_GMS_MASK;
@@ -2253,7 +2254,17 @@ static int ggtt_probe_common(struct drm_device *dev,
 	gtt_phys_addr = pci_resource_start(dev->pdev, 0) +
 		(pci_resource_len(dev->pdev, 0) / 2);
 
-	dev_priv->gtt.gsm = ioremap_wc(gtt_phys_addr, gtt_size);
+	/*
+	 * On BXT writes larger than 64 bit to the GTT pagetable range will be
+	 * dropped. For WC mappings in general we have 64 byte burst writes
+	 * when the WC buffer is flushed, so we can't use it, but have to
+	 * resort to an uncached mapping. The WC issue is easily caught by the
+	 * readback check when writing GTT PTE entries.
+	 */
+	if (IS_BROXTON(dev))
+		dev_priv->gtt.gsm = ioremap_nocache(gtt_phys_addr, gtt_size);
+	else
+		dev_priv->gtt.gsm = ioremap_wc(gtt_phys_addr, gtt_size);
 	if (!dev_priv->gtt.gsm) {
 		DRM_ERROR("Failed to map the gtt page table\n");
 		return -ENOMEM;
@@ -2375,7 +2386,7 @@ static int gen8_gmch_probe(struct drm_device *dev,
 
 	*gtt_total = (gtt_size / sizeof(gen8_pte_t)) << PAGE_SHIFT;
 
-	if (IS_CHERRYVIEW(dev))
+	if (IS_CHERRYVIEW(dev) || IS_BROXTON(dev))
 		chv_setup_private_ppat(dev_priv);
 	else
 		bdw_setup_private_ppat(dev_priv);
@@ -2384,6 +2395,8 @@ static int gen8_gmch_probe(struct drm_device *dev,
 
 	dev_priv->gtt.base.clear_range = gen8_ggtt_clear_range;
 	dev_priv->gtt.base.insert_entries = gen8_ggtt_insert_entries;
+	dev_priv->gtt.base.bind_vma = ggtt_bind_vma;
+	dev_priv->gtt.base.unbind_vma = ggtt_unbind_vma;
 
 	return ret;
 }
@@ -2424,6 +2437,8 @@ static int gen6_gmch_probe(struct drm_device *dev,
 
 	dev_priv->gtt.base.clear_range = gen6_ggtt_clear_range;
 	dev_priv->gtt.base.insert_entries = gen6_ggtt_insert_entries;
+	dev_priv->gtt.base.bind_vma = ggtt_bind_vma;
+	dev_priv->gtt.base.unbind_vma = ggtt_unbind_vma;
 
 	return ret;
 }
@@ -2455,7 +2470,10 @@ static int i915_gmch_probe(struct drm_device *dev,
 	intel_gtt_get(gtt_total, stolen, mappable_base, mappable_end);
 
 	dev_priv->gtt.do_idle_maps = needs_idle_maps(dev_priv->dev);
+	dev_priv->gtt.base.insert_entries = i915_ggtt_insert_entries;
 	dev_priv->gtt.base.clear_range = i915_ggtt_clear_range;
+	dev_priv->gtt.base.bind_vma = ggtt_bind_vma;
+	dev_priv->gtt.base.unbind_vma = ggtt_unbind_vma;
 
 	if (unlikely(dev_priv->gtt.do_idle_maps))
 		DRM_INFO("applying Ironlake quirks for intel_iommu\n");
@@ -2523,6 +2541,59 @@ int i915_gem_gtt_init(struct drm_device *dev)
 	return 0;
 }
 
+void i915_gem_restore_gtt_mappings(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_object *obj;
+	struct i915_address_space *vm;
+
+	i915_check_and_clear_faults(dev);
+
+	/* First fill our portion of the GTT with scratch pages */
+	dev_priv->gtt.base.clear_range(&dev_priv->gtt.base,
+				       dev_priv->gtt.base.start,
+				       dev_priv->gtt.base.total,
+				       true);
+
+	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
+		struct i915_vma *vma = i915_gem_obj_to_vma(obj,
+							   &dev_priv->gtt.base);
+		if (!vma)
+			continue;
+
+		i915_gem_clflush_object(obj, obj->pin_display);
+		WARN_ON(i915_vma_bind(vma, obj->cache_level, PIN_UPDATE));
+	}
+
+
+	if (INTEL_INFO(dev)->gen >= 8) {
+		if (IS_CHERRYVIEW(dev) || IS_BROXTON(dev))
+			chv_setup_private_ppat(dev_priv);
+		else
+			bdw_setup_private_ppat(dev_priv);
+
+		return;
+	}
+
+	if (USES_PPGTT(dev)) {
+		list_for_each_entry(vm, &dev_priv->vm_list, global_link) {
+			/* TODO: Perhaps it shouldn't be gen6 specific */
+
+			struct i915_hw_ppgtt *ppgtt =
+					container_of(vm, struct i915_hw_ppgtt,
+						     base);
+
+			if (i915_is_ggtt(vm))
+				ppgtt = dev_priv->mm.aliasing_ppgtt;
+
+			gen6_write_page_range(dev_priv, &ppgtt->pd,
+					      0, ppgtt->base.total);
+		}
+	}
+
+	i915_ggtt_flush(dev_priv);
+}
+
 static struct i915_vma *
 __i915_gem_vma_create(struct drm_i915_gem_object *obj,
 		      struct i915_address_space *vm,
@@ -2532,7 +2603,8 @@ __i915_gem_vma_create(struct drm_i915_gem_object *obj,
 
 	if (WARN_ON(i915_is_ggtt(vm) != !!ggtt_view))
 		return ERR_PTR(-EINVAL);
-	vma = kzalloc(sizeof(*vma), GFP_KERNEL);
+
+	vma = kmem_cache_zalloc(to_i915(obj->base.dev)->vmas, GFP_KERNEL);
 	if (vma == NULL)
 		return ERR_PTR(-ENOMEM);
 
@@ -2542,22 +2614,8 @@ __i915_gem_vma_create(struct drm_i915_gem_object *obj,
 	vma->vm = vm;
 	vma->obj = obj;
 
-	if (INTEL_INFO(vm->dev)->gen >= 6) {
-		if (i915_is_ggtt(vm)) {
-			vma->ggtt_view = *ggtt_view;
-
-			vma->unbind_vma = ggtt_unbind_vma;
-			vma->bind_vma = ggtt_bind_vma;
-		} else {
-			vma->unbind_vma = ppgtt_unbind_vma;
-			vma->bind_vma = ppgtt_bind_vma;
-		}
-	} else {
-		BUG_ON(!i915_is_ggtt(vm));
+	if (i915_is_ggtt(vm))
 		vma->ggtt_view = *ggtt_view;
-		vma->unbind_vma = i915_ggtt_unbind_vma;
-		vma->bind_vma = i915_ggtt_bind_vma;
-	}
 
 	list_add_tail(&vma->vma_link, &obj->vma_list);
 	if (!i915_is_ggtt(vm))
@@ -2702,7 +2760,48 @@ err_st_alloc:
 	return ERR_PTR(ret);
 }
 
-static inline int
+static struct sg_table *
+intel_partial_pages(const struct i915_ggtt_view *view,
+		    struct drm_i915_gem_object *obj)
+{
+	struct sg_table *st;
+	struct scatterlist *sg;
+	struct sg_page_iter obj_sg_iter;
+	int ret = -ENOMEM;
+
+	st = kmalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		goto err_st_alloc;
+
+	ret = sg_alloc_table(st, view->params.partial.size, GFP_KERNEL);
+	if (ret)
+		goto err_sg_alloc;
+
+	sg = st->sgl;
+	st->nents = 0;
+	for_each_sg_page(obj->pages->sgl, &obj_sg_iter, obj->pages->nents,
+		view->params.partial.offset)
+	{
+		if (st->nents >= view->params.partial.size)
+			break;
+
+		sg_set_page(sg, NULL, PAGE_SIZE, 0);
+		sg_dma_address(sg) = sg_page_iter_dma_address(&obj_sg_iter);
+		sg_dma_len(sg) = PAGE_SIZE;
+
+		sg = sg_next(sg);
+		st->nents++;
+	}
+
+	return st;
+
+err_sg_alloc:
+	kfree(st);
+err_st_alloc:
+	return ERR_PTR(ret);
+}
+
+static int
 i915_get_ggtt_vma_pages(struct i915_vma *vma)
 {
 	int ret = 0;
@@ -2715,6 +2814,9 @@ i915_get_ggtt_vma_pages(struct i915_vma *vma)
 	else if (vma->ggtt_view.type == I915_GGTT_VIEW_ROTATED)
 		vma->ggtt_view.pages =
 			intel_rotate_fb_obj_pages(&vma->ggtt_view, vma->obj);
+	else if (vma->ggtt_view.type == I915_GGTT_VIEW_PARTIAL)
+		vma->ggtt_view.pages =
+			intel_partial_pages(&vma->ggtt_view, vma->obj);
 	else
 		WARN_ONCE(1, "GGTT view %u not implemented!\n",
 			  vma->ggtt_view.type);
@@ -2746,14 +2848,66 @@ i915_get_ggtt_vma_pages(struct i915_vma *vma)
 int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 		  u32 flags)
 {
-	if (i915_is_ggtt(vma->vm)) {
-		int ret = i915_get_ggtt_vma_pages(vma);
+	int ret;
+	u32 bind_flags;
+
+	if (WARN_ON(flags == 0))
+		return -EINVAL;
+
+	bind_flags = 0;
+	if (flags & PIN_GLOBAL)
+		bind_flags |= GLOBAL_BIND;
+	if (flags & PIN_USER)
+		bind_flags |= LOCAL_BIND;
+
+	if (flags & PIN_UPDATE)
+		bind_flags |= vma->bound;
+	else
+		bind_flags &= ~vma->bound;
 
+	if (bind_flags == 0)
+		return 0;
+
+	if (vma->bound == 0 && vma->vm->allocate_va_range) {
+		trace_i915_va_alloc(vma->vm,
+				    vma->node.start,
+				    vma->node.size,
+				    VM_TO_TRACE_NAME(vma->vm));
+
+		ret = vma->vm->allocate_va_range(vma->vm,
+						 vma->node.start,
+						 vma->node.size);
 		if (ret)
 			return ret;
 	}
 
-	vma->bind_vma(vma, cache_level, flags);
+	ret = vma->vm->bind_vma(vma, cache_level, bind_flags);
+	if (ret)
+		return ret;
+
+	vma->bound |= bind_flags;
 
 	return 0;
 }
+
+/**
+ * i915_ggtt_view_size - Get the size of a GGTT view.
+ * @obj: Object the view is of.
+ * @view: The view in question.
+ *
+ * @return The size of the GGTT view in bytes.
+ */
+size_t
+i915_ggtt_view_size(struct drm_i915_gem_object *obj,
+		    const struct i915_ggtt_view *view)
+{
+	if (view->type == I915_GGTT_VIEW_NORMAL ||
+	    view->type == I915_GGTT_VIEW_ROTATED) {
+		return obj->base.size;
+	} else if (view->type == I915_GGTT_VIEW_PARTIAL) {
+		return view->params.partial.size << PAGE_SHIFT;
+	} else {
+		WARN_ONCE(1, "GGTT view %u not implemented!\n", view->type);
+		return obj->base.size;
+	}
+}
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index fc03c99317c9..0d46dd20bf71 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -117,7 +117,8 @@ typedef uint64_t gen8_pde_t;
 
 enum i915_ggtt_view_type {
 	I915_GGTT_VIEW_NORMAL = 0,
-	I915_GGTT_VIEW_ROTATED
+	I915_GGTT_VIEW_ROTATED,
+	I915_GGTT_VIEW_PARTIAL,
 };
 
 struct intel_rotation_info {
@@ -130,6 +131,13 @@ struct intel_rotation_info {
 struct i915_ggtt_view {
 	enum i915_ggtt_view_type type;
 
+	union {
+		struct {
+			unsigned long offset;
+			unsigned int size;
+		} partial;
+	} params;
+
 	struct sg_table *pages;
 
 	union {
@@ -158,7 +166,6 @@ struct i915_vma {
 	/** Flags and address space this VMA is bound to */
 #define GLOBAL_BIND	(1<<0)
 #define LOCAL_BIND	(1<<1)
-#define PTE_READ_ONLY	(1<<2)
 	unsigned int bound : 4;
 
 	/**
@@ -196,36 +203,30 @@ struct i915_vma {
 	 * bits with absolutely no headroom. So use 4 bits. */
 	unsigned int pin_count:4;
 #define DRM_I915_GEM_OBJECT_MAX_PIN_COUNT 0xf
-
-	/** Unmap an object from an address space. This usually consists of
-	 * setting the valid PTE entries to a reserved scratch page. */
-	void (*unbind_vma)(struct i915_vma *vma);
-	/* Map an object into an address space with the given cache flags. */
-	void (*bind_vma)(struct i915_vma *vma,
-			 enum i915_cache_level cache_level,
-			 u32 flags);
 };
 
-struct i915_page_table_entry {
+struct i915_page_table {
 	struct page *page;
 	dma_addr_t daddr;
 
 	unsigned long *used_ptes;
 };
 
-struct i915_page_directory_entry {
+struct i915_page_directory {
 	struct page *page; /* NULL for GEN6-GEN7 */
 	union {
 		uint32_t pd_offset;
 		dma_addr_t daddr;
 	};
 
-	struct i915_page_table_entry *page_table[I915_PDES]; /* PDEs */
+	unsigned long *used_pdes;
+	struct i915_page_table *page_table[I915_PDES]; /* PDEs */
 };
 
-struct i915_page_directory_pointer_entry {
+struct i915_page_directory_pointer {
 	/* struct page *page; */
-	struct i915_page_directory_entry *page_directory[GEN8_LEGACY_PDPES];
+	DECLARE_BITMAP(used_pdpes, GEN8_LEGACY_PDPES);
+	struct i915_page_directory *page_directory[GEN8_LEGACY_PDPES];
 };
 
 struct i915_address_space {
@@ -267,6 +268,8 @@ struct i915_address_space {
 	gen6_pte_t (*pte_encode)(dma_addr_t addr,
 				 enum i915_cache_level level,
 				 bool valid, u32 flags); /* Create a valid PTE */
+	/* flags for pte_encode */
+#define PTE_READ_ONLY	(1<<0)
 	int (*allocate_va_range)(struct i915_address_space *vm,
 				 uint64_t start,
 				 uint64_t length);
@@ -279,6 +282,13 @@ struct i915_address_space {
 			       uint64_t start,
 			       enum i915_cache_level cache_level, u32 flags);
 	void (*cleanup)(struct i915_address_space *vm);
+	/** Unmap an object from an address space. This usually consists of
+	 * setting the valid PTE entries to a reserved scratch page. */
+	void (*unbind_vma)(struct i915_vma *vma);
+	/* Map an object into an address space with the given cache flags. */
+	int (*bind_vma)(struct i915_vma *vma,
+			enum i915_cache_level cache_level,
+			u32 flags);
 };
 
 /* The Graphics Translation Table is the way in which GEN hardware translates a
@@ -314,14 +324,13 @@ struct i915_hw_ppgtt {
 	struct kref ref;
 	struct drm_mm_node node;
 	unsigned long pd_dirty_rings;
-	unsigned num_pd_entries;
-	unsigned num_pd_pages; /* gen8+ */
 	union {
-		struct i915_page_directory_pointer_entry pdp;
-		struct i915_page_directory_entry pd;
+		struct i915_page_directory_pointer pdp;
+		struct i915_page_directory pd;
 	};
 
-	struct i915_page_table_entry *scratch_pt;
+	struct i915_page_table *scratch_pt;
+	struct i915_page_directory *scratch_pd;
 
 	struct drm_i915_file_private *file_priv;
 
@@ -349,6 +358,11 @@ struct i915_hw_ppgtt {
 	     temp = min_t(unsigned, temp, length), \
 	     start += temp, length -= temp)
 
+#define gen6_for_all_pdes(pt, ppgtt, iter)  \
+	for (iter = 0;		\
+	     pt = ppgtt->pd.page_table[iter], iter < I915_PDES;	\
+	     iter++)
+
 static inline uint32_t i915_pte_index(uint64_t address, uint32_t pde_shift)
 {
 	const uint32_t mask = NUM_PTE(pde_shift) - 1;
@@ -397,6 +411,63 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
 	return i915_pde_index(addr, GEN6_PDE_SHIFT);
 }
 
+/* Equivalent to the gen6 version, For each pde iterates over every pde
+ * between from start until start + length. On gen8+ it simply iterates
+ * over every page directory entry in a page directory.
+ */
+#define gen8_for_each_pde(pt, pd, start, length, temp, iter)		\
+	for (iter = gen8_pde_index(start); \
+	     pt = (pd)->page_table[iter], length > 0 && iter < I915_PDES;	\
+	     iter++,				\
+	     temp = ALIGN(start+1, 1 << GEN8_PDE_SHIFT) - start,	\
+	     temp = min(temp, length),					\
+	     start += temp, length -= temp)
+
+#define gen8_for_each_pdpe(pd, pdp, start, length, temp, iter)		\
+	for (iter = gen8_pdpe_index(start);	\
+	     pd = (pdp)->page_directory[iter], length > 0 && iter < GEN8_LEGACY_PDPES;	\
+	     iter++,				\
+	     temp = ALIGN(start+1, 1 << GEN8_PDPE_SHIFT) - start,	\
+	     temp = min(temp, length),					\
+	     start += temp, length -= temp)
+
+/* Clamp length to the next page_directory boundary */
+static inline uint64_t gen8_clamp_pd(uint64_t start, uint64_t length)
+{
+	uint64_t next_pd = ALIGN(start + 1, 1 << GEN8_PDPE_SHIFT);
+
+	if (next_pd > (start + length))
+		return length;
+
+	return next_pd - start;
+}
+
+static inline uint32_t gen8_pte_index(uint64_t address)
+{
+	return i915_pte_index(address, GEN8_PDE_SHIFT);
+}
+
+static inline uint32_t gen8_pde_index(uint64_t address)
+{
+	return i915_pde_index(address, GEN8_PDE_SHIFT);
+}
+
+static inline uint32_t gen8_pdpe_index(uint64_t address)
+{
+	return (address >> GEN8_PDPE_SHIFT) & GEN8_PDPE_MASK;
+}
+
+static inline uint32_t gen8_pml4e_index(uint64_t address)
+{
+	WARN_ON(1); /* For 64B */
+	return 0;
+}
+
+static inline size_t gen8_pte_count(uint64_t address, uint64_t length)
+{
+	return i915_pte_count(address, length, GEN8_PDE_SHIFT);
+}
+
 int i915_gem_gtt_init(struct drm_device *dev);
 void i915_gem_init_global_gtt(struct drm_device *dev);
 void i915_global_gtt_cleanup(struct drm_device *dev);
@@ -432,7 +503,15 @@ i915_ggtt_view_equal(const struct i915_ggtt_view *a,
 	if (WARN_ON(!a || !b))
 		return false;
 
-	return a->type == b->type;
+	if (a->type != b->type)
+		return false;
+	if (a->type == I915_GGTT_VIEW_PARTIAL)
+		return !memcmp(&a->params, &b->params, sizeof(a->params));
+	return true;
 }
 
+size_t
+i915_ggtt_view_size(struct drm_i915_gem_object *obj,
+		    const struct i915_ggtt_view *view);
+
 #endif
diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index f7929e769250..f6ecbda2c604 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -184,9 +184,12 @@ static int num_vma_bound(struct drm_i915_gem_object *obj)
 	struct i915_vma *vma;
 	int count = 0;
 
-	list_for_each_entry(vma, &obj->vma_list, vma_link)
+	list_for_each_entry(vma, &obj->vma_list, vma_link) {
 		if (drm_mm_node_allocated(&vma->node))
 			count++;
+		if (vma->pin_count)
+			count++;
+	}
 
 	return count;
 }
@@ -210,8 +213,7 @@ i915_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 			count += obj->base.size >> PAGE_SHIFT;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		if (!i915_gem_obj_is_pinned(obj) &&
-		    obj->pages_pin_count == num_vma_bound(obj))
+		if (obj->pages_pin_count == num_vma_bound(obj))
 			count += obj->base.size >> PAGE_SHIFT;
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index f8da71682c96..348ed5abcdbf 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -209,7 +209,7 @@ static int i915_setup_compression(struct drm_device *dev, int size, int fb_cpp)
 
 	dev_priv->fbc.threshold = ret;
 
-	if (HAS_PCH_SPLIT(dev))
+	if (INTEL_INFO(dev_priv)->gen >= 5)
 		I915_WRITE(ILK_DPFC_CB_BASE, dev_priv->fbc.compressed_fb.start);
 	else if (IS_GM45(dev)) {
 		I915_WRITE(DPFC_CB_BASE, dev_priv->fbc.compressed_fb.start);
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 6377b22269ad..633bd1fcab69 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -336,7 +336,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 	}
 
 	mutex_lock(&dev->struct_mutex);
-	if (i915_gem_obj_is_pinned(obj) || obj->framebuffer_references) {
+	if (obj->pin_display || obj->framebuffer_references) {
 		ret = -EBUSY;
 		goto err;
 	}
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 1719078c763a..1f4e5a32a16e 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -22,8 +22,8 @@
  *
  */
 
-#include "drmP.h"
-#include "i915_drm.h"
+#include <drm/drmP.h>
+#include <drm/i915_drm.h>
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
@@ -219,11 +219,14 @@ i915_mmu_notifier_add(struct drm_device *dev,
 		      struct i915_mmu_object *mo)
 {
 	struct interval_tree_node *it;
-	int ret;
+	int ret = 0;
 
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
+	/* By this point we have already done a lot of expensive setup that
+	 * we do not want to repeat just because the caller (e.g. X) has a
+	 * signal pending (and partly because of that expensive setup, X
+	 * using an interrupt timer is likely to get stuck in an EINTR loop).
+	 */
+	mutex_lock(&dev->struct_mutex);
 
 	/* Make sure we drop the final active reference (and thereby
 	 * remove the objects from the interval tree) before we do
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 1d4e60df8883..6f4256918f76 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -192,15 +192,20 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 				struct drm_i915_error_buffer *err,
 				int count)
 {
+	int i;
+
 	err_printf(m, "  %s [%d]:\n", name, count);
 
 	while (count--) {
-		err_printf(m, "    %08x %8u %02x %02x %x %x",
+		err_printf(m, "    %08x %8u %02x %02x [ ",
 			   err->gtt_offset,
 			   err->size,
 			   err->read_domains,
-			   err->write_domain,
-			   err->rseqno, err->wseqno);
+			   err->write_domain);
+		for (i = 0; i < I915_NUM_RINGS; i++)
+			err_printf(m, "%02x ", err->rseqno[i]);
+
+		err_printf(m, "] %02x", err->wseqno);
 		err_puts(m, pin_flag(err->pinned));
 		err_puts(m, tiling_flag(err->tiling));
 		err_puts(m, dirty_flag(err->dirty));
@@ -251,10 +256,11 @@ static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
 		return;
 
 	err_printf(m, "%s command stream:\n", ring_str(ring_idx));
-	err_printf(m, "  HEAD: 0x%08x\n", ring->head);
-	err_printf(m, "  TAIL: 0x%08x\n", ring->tail);
-	err_printf(m, "  CTL: 0x%08x\n", ring->ctl);
-	err_printf(m, "  HWS: 0x%08x\n", ring->hws);
+	err_printf(m, "  START: 0x%08x\n", ring->start);
+	err_printf(m, "  HEAD:  0x%08x\n", ring->head);
+	err_printf(m, "  TAIL:  0x%08x\n", ring->tail);
+	err_printf(m, "  CTL:   0x%08x\n", ring->ctl);
+	err_printf(m, "  HWS:   0x%08x\n", ring->hws);
 	err_printf(m, "  ACTHD: 0x%08x %08x\n", (u32)(ring->acthd>>32), (u32)ring->acthd);
 	err_printf(m, "  IPEIR: 0x%08x\n", ring->ipeir);
 	err_printf(m, "  IPEHR: 0x%08x\n", ring->ipehr);
@@ -553,6 +559,7 @@ static void i915_error_state_free(struct kref *error_ref)
 
 	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
 		i915_error_object_free(error->ring[i].batchbuffer);
+		i915_error_object_free(error->ring[i].wa_batchbuffer);
 		i915_error_object_free(error->ring[i].ringbuffer);
 		i915_error_object_free(error->ring[i].hws_page);
 		i915_error_object_free(error->ring[i].ctx);
@@ -679,10 +686,12 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 		       struct i915_vma *vma)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
+	int i;
 
 	err->size = obj->base.size;
 	err->name = obj->base.name;
-	err->rseqno = i915_gem_request_get_seqno(obj->last_read_req);
+	for (i = 0; i < I915_NUM_RINGS; i++)
+		err->rseqno[i] = i915_gem_request_get_seqno(obj->last_read_req[i]);
 	err->wseqno = i915_gem_request_get_seqno(obj->last_write_req);
 	err->gtt_offset = vma->node.start;
 	err->read_domains = obj->base.read_domains;
@@ -695,8 +704,8 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 	err->dirty = obj->dirty;
 	err->purgeable = obj->madv != I915_MADV_WILLNEED;
 	err->userptr = obj->userptr.mm != NULL;
-	err->ring = obj->last_read_req ?
-			i915_gem_request_get_ring(obj->last_read_req)->id : -1;
+	err->ring = obj->last_write_req ?
+			i915_gem_request_get_ring(obj->last_write_req)->id : -1;
 	err->cache_level = obj->cache_level;
 }
 
@@ -883,6 +892,7 @@ static void i915_record_ring_state(struct drm_device *dev,
 	ering->instpm = I915_READ(RING_INSTPM(ring->mmio_base));
 	ering->seqno = ring->get_seqno(ring, false);
 	ering->acthd = intel_ring_get_active_head(ring);
+	ering->start = I915_READ_START(ring);
 	ering->head = I915_READ_HEAD(ring);
 	ering->tail = I915_READ_TAIL(ring);
 	ering->ctl = I915_READ_CTL(ring);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 6d494432b19f..e6bb72dca3ff 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -79,7 +79,7 @@ static const u32 hpd_status_g4x[HPD_NUM_PINS] = {
 	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS
 };
 
-static const u32 hpd_status_i915[HPD_NUM_PINS] = { /* i915 and valleyview are the same */
+static const u32 hpd_status_i915[HPD_NUM_PINS] = {
 	[HPD_CRT] = CRT_HOTPLUG_INT_STATUS,
 	[HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_I915,
 	[HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_I915,
@@ -88,6 +88,12 @@ static const u32 hpd_status_i915[HPD_NUM_PINS] = { /* i915 and valleyview are th
 	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS
 };
 
+/* BXT hpd list */
+static const u32 hpd_bxt[HPD_NUM_PINS] = {
+	[HPD_PORT_B] = BXT_DE_PORT_HP_DDIB,
+	[HPD_PORT_C] = BXT_DE_PORT_HP_DDIC
+};
+
 /* IIR can theoretically queue up two events. Be paranoid. */
 #define GEN8_IRQ_RESET_NDX(type, which) do { \
 	I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
@@ -985,8 +991,7 @@ static void ironlake_rps_change_irq_handler(struct drm_device *dev)
 	return;
 }
 
-static void notify_ring(struct drm_device *dev,
-			struct intel_engine_cs *ring)
+static void notify_ring(struct intel_engine_cs *ring)
 {
 	if (!intel_ring_initialized(ring))
 		return;
@@ -1049,7 +1054,7 @@ static u32 vlv_wa_c0_ei(struct drm_i915_private *dev_priv, u32 pm_iir)
 	if (pm_iir & GEN6_PM_RP_DOWN_EI_EXPIRED) {
 		if (!vlv_c0_above(dev_priv,
 				  &dev_priv->rps.down_ei, &now,
-				  VLV_RP_DOWN_EI_THRESHOLD))
+				  dev_priv->rps.down_threshold))
 			events |= GEN6_PM_RP_DOWN_THRESHOLD;
 		dev_priv->rps.down_ei = now;
 	}
@@ -1057,7 +1062,7 @@ static u32 vlv_wa_c0_ei(struct drm_i915_private *dev_priv, u32 pm_iir)
 	if (pm_iir & GEN6_PM_RP_UP_EI_EXPIRED) {
 		if (vlv_c0_above(dev_priv,
 				 &dev_priv->rps.up_ei, &now,
-				 VLV_RP_UP_EI_THRESHOLD))
+				 dev_priv->rps.up_threshold))
 			events |= GEN6_PM_RP_UP_THRESHOLD;
 		dev_priv->rps.up_ei = now;
 	}
@@ -1065,12 +1070,25 @@ static u32 vlv_wa_c0_ei(struct drm_i915_private *dev_priv, u32 pm_iir)
 	return events;
 }
 
+static bool any_waiters(struct drm_i915_private *dev_priv)
+{
+	struct intel_engine_cs *ring;
+	int i;
+
+	for_each_ring(ring, dev_priv, i)
+		if (ring->irq_refcount)
+			return true;
+
+	return false;
+}
+
 static void gen6_pm_rps_work(struct work_struct *work)
 {
 	struct drm_i915_private *dev_priv =
 		container_of(work, struct drm_i915_private, rps.work);
+	bool client_boost;
+	int new_delay, adj, min, max;
 	u32 pm_iir;
-	int new_delay, adj;
 
 	spin_lock_irq(&dev_priv->irq_lock);
 	/* Speed up work cancelation during disabling rps interrupts. */
@@ -1082,12 +1100,14 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	dev_priv->rps.pm_iir = 0;
 	/* Make sure not to corrupt PMIMR state used by ringbuffer on GEN6 */
 	gen6_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	client_boost = dev_priv->rps.client_boost;
+	dev_priv->rps.client_boost = false;
 	spin_unlock_irq(&dev_priv->irq_lock);
 
 	/* Make sure we didn't queue anything we're not going to process. */
 	WARN_ON(pm_iir & ~dev_priv->pm_rps_events);
 
-	if ((pm_iir & dev_priv->pm_rps_events) == 0)
+	if ((pm_iir & dev_priv->pm_rps_events) == 0 && !client_boost)
 		return;
 
 	mutex_lock(&dev_priv->rps.hw_lock);
@@ -1095,21 +1115,28 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	pm_iir |= vlv_wa_c0_ei(dev_priv, pm_iir);
 
 	adj = dev_priv->rps.last_adj;
-	if (pm_iir & GEN6_PM_RP_UP_THRESHOLD) {
+	new_delay = dev_priv->rps.cur_freq;
+	min = dev_priv->rps.min_freq_softlimit;
+	max = dev_priv->rps.max_freq_softlimit;
+
+	if (client_boost) {
+		new_delay = dev_priv->rps.max_freq_softlimit;
+		adj = 0;
+	} else if (pm_iir & GEN6_PM_RP_UP_THRESHOLD) {
 		if (adj > 0)
 			adj *= 2;
-		else {
-			/* CHV needs even encode values */
-			adj = IS_CHERRYVIEW(dev_priv->dev) ? 2 : 1;
-		}
-		new_delay = dev_priv->rps.cur_freq + adj;
-
+		else /* CHV needs even encode values */
+			adj = IS_CHERRYVIEW(dev_priv) ? 2 : 1;
 		/*
 		 * For better performance, jump directly
 		 * to RPe if we're below it.
 		 */
-		if (new_delay < dev_priv->rps.efficient_freq)
+		if (new_delay < dev_priv->rps.efficient_freq - adj) {
 			new_delay = dev_priv->rps.efficient_freq;
+			adj = 0;
+		}
+	} else if (any_waiters(dev_priv)) {
+		adj = 0;
 	} else if (pm_iir & GEN6_PM_RP_DOWN_TIMEOUT) {
 		if (dev_priv->rps.cur_freq > dev_priv->rps.efficient_freq)
 			new_delay = dev_priv->rps.efficient_freq;
@@ -1119,23 +1146,19 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	} else if (pm_iir & GEN6_PM_RP_DOWN_THRESHOLD) {
 		if (adj < 0)
 			adj *= 2;
-		else {
-			/* CHV needs even encode values */
-			adj = IS_CHERRYVIEW(dev_priv->dev) ? -2 : -1;
-		}
-		new_delay = dev_priv->rps.cur_freq + adj;
+		else /* CHV needs even encode values */
+			adj = IS_CHERRYVIEW(dev_priv) ? -2 : -1;
 	} else { /* unknown event */
-		new_delay = dev_priv->rps.cur_freq;
+		adj = 0;
 	}
 
+	dev_priv->rps.last_adj = adj;
+
 	/* sysfs frequency interfaces may have snuck in while servicing the
 	 * interrupt
 	 */
-	new_delay = clamp_t(int, new_delay,
-			    dev_priv->rps.min_freq_softlimit,
-			    dev_priv->rps.max_freq_softlimit);
-
-	dev_priv->rps.last_adj = new_delay - dev_priv->rps.cur_freq;
+	new_delay += adj;
+	new_delay = clamp_t(int, new_delay, min, max);
 
 	intel_set_rps(dev_priv->dev, new_delay);
 
@@ -1251,9 +1274,9 @@ static void ilk_gt_irq_handler(struct drm_device *dev,
 {
 	if (gt_iir &
 	    (GT_RENDER_USER_INTERRUPT | GT_RENDER_PIPECTL_NOTIFY_INTERRUPT))
-		notify_ring(dev, &dev_priv->ring[RCS]);
+		notify_ring(&dev_priv->ring[RCS]);
 	if (gt_iir & ILK_BSD_USER_INTERRUPT)
-		notify_ring(dev, &dev_priv->ring[VCS]);
+		notify_ring(&dev_priv->ring[VCS]);
 }
 
 static void snb_gt_irq_handler(struct drm_device *dev,
@@ -1263,11 +1286,11 @@ static void snb_gt_irq_handler(struct drm_device *dev,
 
 	if (gt_iir &
 	    (GT_RENDER_USER_INTERRUPT | GT_RENDER_PIPECTL_NOTIFY_INTERRUPT))
-		notify_ring(dev, &dev_priv->ring[RCS]);
+		notify_ring(&dev_priv->ring[RCS]);
 	if (gt_iir & GT_BSD_USER_INTERRUPT)
-		notify_ring(dev, &dev_priv->ring[VCS]);
+		notify_ring(&dev_priv->ring[VCS]);
 	if (gt_iir & GT_BLT_USER_INTERRUPT)
-		notify_ring(dev, &dev_priv->ring[BCS]);
+		notify_ring(&dev_priv->ring[BCS]);
 
 	if (gt_iir & (GT_BLT_CS_ERROR_INTERRUPT |
 		      GT_BSD_CS_ERROR_INTERRUPT |
@@ -1278,88 +1301,74 @@ static void snb_gt_irq_handler(struct drm_device *dev,
 		ivybridge_parity_error_irq_handler(dev, gt_iir);
 }
 
-static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
-				       struct drm_i915_private *dev_priv,
+static irqreturn_t gen8_gt_irq_handler(struct drm_i915_private *dev_priv,
 				       u32 master_ctl)
 {
-	struct intel_engine_cs *ring;
-	u32 rcs, bcs, vcs;
-	uint32_t tmp = 0;
 	irqreturn_t ret = IRQ_NONE;
 
 	if (master_ctl & (GEN8_GT_RCS_IRQ | GEN8_GT_BCS_IRQ)) {
-		tmp = I915_READ(GEN8_GT_IIR(0));
+		u32 tmp = I915_READ_FW(GEN8_GT_IIR(0));
 		if (tmp) {
-			I915_WRITE(GEN8_GT_IIR(0), tmp);
+			I915_WRITE_FW(GEN8_GT_IIR(0), tmp);
 			ret = IRQ_HANDLED;
 
-			rcs = tmp >> GEN8_RCS_IRQ_SHIFT;
-			ring = &dev_priv->ring[RCS];
-			if (rcs & GT_RENDER_USER_INTERRUPT)
-				notify_ring(dev, ring);
-			if (rcs & GT_CONTEXT_SWITCH_INTERRUPT)
-				intel_lrc_irq_handler(ring);
-
-			bcs = tmp >> GEN8_BCS_IRQ_SHIFT;
-			ring = &dev_priv->ring[BCS];
-			if (bcs & GT_RENDER_USER_INTERRUPT)
-				notify_ring(dev, ring);
-			if (bcs & GT_CONTEXT_SWITCH_INTERRUPT)
-				intel_lrc_irq_handler(ring);
+			if (tmp & (GT_CONTEXT_SWITCH_INTERRUPT << GEN8_RCS_IRQ_SHIFT))
+				intel_lrc_irq_handler(&dev_priv->ring[RCS]);
+			if (tmp & (GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT))
+				notify_ring(&dev_priv->ring[RCS]);
+
+			if (tmp & (GT_CONTEXT_SWITCH_INTERRUPT << GEN8_BCS_IRQ_SHIFT))
+				intel_lrc_irq_handler(&dev_priv->ring[BCS]);
+			if (tmp & (GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT))
+				notify_ring(&dev_priv->ring[BCS]);
 		} else
 			DRM_ERROR("The master control interrupt lied (GT0)!\n");
 	}
 
 	if (master_ctl & (GEN8_GT_VCS1_IRQ | GEN8_GT_VCS2_IRQ)) {
-		tmp = I915_READ(GEN8_GT_IIR(1));
+		u32 tmp = I915_READ_FW(GEN8_GT_IIR(1));
 		if (tmp) {
-			I915_WRITE(GEN8_GT_IIR(1), tmp);
+			I915_WRITE_FW(GEN8_GT_IIR(1), tmp);
 			ret = IRQ_HANDLED;
 
-			vcs = tmp >> GEN8_VCS1_IRQ_SHIFT;
-			ring = &dev_priv->ring[VCS];
-			if (vcs & GT_RENDER_USER_INTERRUPT)
-				notify_ring(dev, ring);
-			if (vcs & GT_CONTEXT_SWITCH_INTERRUPT)
-				intel_lrc_irq_handler(ring);
-
-			vcs = tmp >> GEN8_VCS2_IRQ_SHIFT;
-			ring = &dev_priv->ring[VCS2];
-			if (vcs & GT_RENDER_USER_INTERRUPT)
-				notify_ring(dev, ring);
-			if (vcs & GT_CONTEXT_SWITCH_INTERRUPT)
-				intel_lrc_irq_handler(ring);
-		} else
-			DRM_ERROR("The master control interrupt lied (GT1)!\n");
-	}
+			if (tmp & (GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VCS1_IRQ_SHIFT))
+				intel_lrc_irq_handler(&dev_priv->ring[VCS]);
+			if (tmp & (GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT))
+				notify_ring(&dev_priv->ring[VCS]);
 
-	if (master_ctl & GEN8_GT_PM_IRQ) {
-		tmp = I915_READ(GEN8_GT_IIR(2));
-		if (tmp & dev_priv->pm_rps_events) {
-			I915_WRITE(GEN8_GT_IIR(2),
-				   tmp & dev_priv->pm_rps_events);
-			ret = IRQ_HANDLED;
-			gen6_rps_irq_handler(dev_priv, tmp);
+			if (tmp & (GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VCS2_IRQ_SHIFT))
+				intel_lrc_irq_handler(&dev_priv->ring[VCS2]);
+			if (tmp & (GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT))
+				notify_ring(&dev_priv->ring[VCS2]);
 		} else
-			DRM_ERROR("The master control interrupt lied (PM)!\n");
+			DRM_ERROR("The master control interrupt lied (GT1)!\n");
 	}
 
 	if (master_ctl & GEN8_GT_VECS_IRQ) {
-		tmp = I915_READ(GEN8_GT_IIR(3));
+		u32 tmp = I915_READ_FW(GEN8_GT_IIR(3));
 		if (tmp) {
-			I915_WRITE(GEN8_GT_IIR(3), tmp);
+			I915_WRITE_FW(GEN8_GT_IIR(3), tmp);
 			ret = IRQ_HANDLED;
 
-			vcs = tmp >> GEN8_VECS_IRQ_SHIFT;
-			ring = &dev_priv->ring[VECS];
-			if (vcs & GT_RENDER_USER_INTERRUPT)
-				notify_ring(dev, ring);
-			if (vcs & GT_CONTEXT_SWITCH_INTERRUPT)
-				intel_lrc_irq_handler(ring);
+			if (tmp & (GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VECS_IRQ_SHIFT))
+				intel_lrc_irq_handler(&dev_priv->ring[VECS]);
+			if (tmp & (GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT))
+				notify_ring(&dev_priv->ring[VECS]);
 		} else
 			DRM_ERROR("The master control interrupt lied (GT3)!\n");
 	}
 
+	if (master_ctl & GEN8_GT_PM_IRQ) {
+		u32 tmp = I915_READ_FW(GEN8_GT_IIR(2));
+		if (tmp & dev_priv->pm_rps_events) {
+			I915_WRITE_FW(GEN8_GT_IIR(2),
+				      tmp & dev_priv->pm_rps_events);
+			ret = IRQ_HANDLED;
+			gen6_rps_irq_handler(dev_priv, tmp);
+		} else
+			DRM_ERROR("The master control interrupt lied (PM)!\n");
+	}
+
 	return ret;
 }
 
@@ -1398,7 +1407,7 @@ static int i915_port_to_hotplug_shift(enum port port)
 	}
 }
 
-static inline enum port get_port_from_pin(enum hpd_pin pin)
+static enum port get_port_from_pin(enum hpd_pin pin)
 {
 	switch (pin) {
 	case HPD_PORT_B:
@@ -1412,10 +1421,10 @@ static inline enum port get_port_from_pin(enum hpd_pin pin)
 	}
 }
 
-static inline void intel_hpd_irq_handler(struct drm_device *dev,
-					 u32 hotplug_trigger,
-					 u32 dig_hotplug_reg,
-					 const u32 hpd[HPD_NUM_PINS])
+static void intel_hpd_irq_handler(struct drm_device *dev,
+				  u32 hotplug_trigger,
+				  u32 dig_hotplug_reg,
+				  const u32 hpd[HPD_NUM_PINS])
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
@@ -1440,7 +1449,7 @@ static inline void intel_hpd_irq_handler(struct drm_device *dev,
 		if (port && dev_priv->hpd_irq_port[port]) {
 			bool long_hpd;
 
-			if (HAS_PCH_SPLIT(dev)) {
+			if (!HAS_GMCH_DISPLAY(dev_priv)) {
 				dig_shift = pch_port_to_hotplug_shift(port);
 				long_hpd = (dig_hotplug_reg >> dig_shift) & PORTB_HOTPLUG_LONG_DETECT;
 			} else {
@@ -1654,7 +1663,7 @@ static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
 
 	if (HAS_VEBOX(dev_priv->dev)) {
 		if (pm_iir & PM_VEBOX_USER_INTERRUPT)
-			notify_ring(dev_priv->dev, &dev_priv->ring[VECS]);
+			notify_ring(&dev_priv->ring[VECS]);
 
 		if (pm_iir & PM_VEBOX_CS_ERROR_INTERRUPT)
 			DRM_DEBUG("Command parser error, pm_iir 0x%08x\n", pm_iir);
@@ -1755,7 +1764,7 @@ static void i9xx_hpd_irq_handler(struct drm_device *dev)
 		 */
 		POSTING_READ(PORT_HOTPLUG_STAT);
 
-		if (IS_G4X(dev)) {
+		if (IS_G4X(dev) || IS_VALLEYVIEW(dev)) {
 			u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
 
 			intel_hpd_irq_handler(dev, hotplug_trigger, 0, hpd_status_g4x);
@@ -1848,7 +1857,7 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 			I915_WRITE(VLV_IIR, iir);
 		}
 
-		gen8_gt_irq_handler(dev, dev_priv, master_ctl);
+		gen8_gt_irq_handler(dev_priv, master_ctl);
 
 		/* Call regardless, as some status bits might not be
 		 * signalled in iir */
@@ -2164,6 +2173,38 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	return ret;
 }
 
+static void bxt_hpd_handler(struct drm_device *dev, uint32_t iir_status)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t hp_control;
+	uint32_t hp_trigger;
+
+	/* Get the status */
+	hp_trigger = iir_status & BXT_DE_PORT_HOTPLUG_MASK;
+	hp_control = I915_READ(BXT_HOTPLUG_CTL);
+
+	/* Hotplug not enabled ? */
+	if (!(hp_control & BXT_HOTPLUG_CTL_MASK)) {
+		DRM_ERROR("Interrupt when HPD disabled\n");
+		return;
+	}
+
+	DRM_DEBUG_DRIVER("hotplug event received, stat 0x%08x\n",
+		hp_control & BXT_HOTPLUG_CTL_MASK);
+
+	/* Check for HPD storm and schedule bottom half */
+	intel_hpd_irq_handler(dev, hp_trigger, hp_control, hpd_bxt);
+
+	/*
+	 * FIXME: Save the hot plug status for bottom half before
+	 * clearing the sticky status bits, else the status will be
+	 * lost.
+	 */
+
+	/* Clear sticky bits in hpd status */
+	I915_WRITE(BXT_HOTPLUG_CTL, hp_control);
+}
+
 static irqreturn_t gen8_irq_handler(int irq, void *arg)
 {
 	struct drm_device *dev = arg;
@@ -2181,17 +2222,16 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 		aux_mask |=  GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
 			GEN9_AUX_CHANNEL_D;
 
-	master_ctl = I915_READ(GEN8_MASTER_IRQ);
+	master_ctl = I915_READ_FW(GEN8_MASTER_IRQ);
 	master_ctl &= ~GEN8_MASTER_IRQ_CONTROL;
 	if (!master_ctl)
 		return IRQ_NONE;
 
-	I915_WRITE(GEN8_MASTER_IRQ, 0);
-	POSTING_READ(GEN8_MASTER_IRQ);
+	I915_WRITE_FW(GEN8_MASTER_IRQ, 0);
 
 	/* Find, clear, then process each source of interrupt */
 
-	ret = gen8_gt_irq_handler(dev, dev_priv, master_ctl);
+	ret = gen8_gt_irq_handler(dev_priv, master_ctl);
 
 	if (master_ctl & GEN8_DE_MISC_IRQ) {
 		tmp = I915_READ(GEN8_DE_MISC_IIR);
@@ -2210,12 +2250,27 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 	if (master_ctl & GEN8_DE_PORT_IRQ) {
 		tmp = I915_READ(GEN8_DE_PORT_IIR);
 		if (tmp) {
+			bool found = false;
+
 			I915_WRITE(GEN8_DE_PORT_IIR, tmp);
 			ret = IRQ_HANDLED;
 
-			if (tmp & aux_mask)
+			if (tmp & aux_mask) {
 				dp_aux_irq_handler(dev);
-			else
+				found = true;
+			}
+
+			if (IS_BROXTON(dev) && tmp & BXT_DE_PORT_HOTPLUG_MASK) {
+				bxt_hpd_handler(dev, tmp);
+				found = true;
+			}
+
+			if (IS_BROXTON(dev) && (tmp & BXT_DE_PORT_GMBUS)) {
+				gmbus_irq_handler(dev);
+				found = true;
+			}
+
+			if (!found)
 				DRM_ERROR("Unexpected DE Port interrupt\n");
 		}
 		else
@@ -2268,7 +2323,8 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 			DRM_ERROR("The master control interrupt lied (DE PIPE)!\n");
 	}
 
-	if (!HAS_PCH_NOP(dev) && master_ctl & GEN8_DE_PCH_IRQ) {
+	if (HAS_PCH_SPLIT(dev) && !HAS_PCH_NOP(dev) &&
+	    master_ctl & GEN8_DE_PCH_IRQ) {
 		/*
 		 * FIXME(BDW): Assume for now that the new interrupt handling
 		 * scheme also closed the SDE interrupt handling race we've seen
@@ -2284,8 +2340,8 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 
 	}
 
-	I915_WRITE(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
-	POSTING_READ(GEN8_MASTER_IRQ);
+	I915_WRITE_FW(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
+	POSTING_READ_FW(GEN8_MASTER_IRQ);
 
 	return ret;
 }
@@ -3104,7 +3160,8 @@ static void gen8_irq_reset(struct drm_device *dev)
 	GEN5_IRQ_RESET(GEN8_DE_MISC_);
 	GEN5_IRQ_RESET(GEN8_PCU_);
 
-	ibx_irq_reset(dev);
+	if (HAS_PCH_SPLIT(dev))
+		ibx_irq_reset(dev);
 }
 
 void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
@@ -3178,6 +3235,42 @@ static void ibx_hpd_irq_setup(struct drm_device *dev)
 	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
 }
 
+static void bxt_hpd_irq_setup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_encoder *intel_encoder;
+	u32 hotplug_port = 0;
+	u32 hotplug_ctrl;
+
+	/* Now, enable HPD */
+	for_each_intel_encoder(dev, intel_encoder) {
+		if (dev_priv->hpd_stats[intel_encoder->hpd_pin].hpd_mark
+				== HPD_ENABLED)
+			hotplug_port |= hpd_bxt[intel_encoder->hpd_pin];
+	}
+
+	/* Mask all HPD control bits */
+	hotplug_ctrl = I915_READ(BXT_HOTPLUG_CTL) & ~BXT_HOTPLUG_CTL_MASK;
+
+	/* Enable requested port in hotplug control */
+	/* TODO: implement (short) HPD support on port A */
+	WARN_ON_ONCE(hotplug_port & BXT_DE_PORT_HP_DDIA);
+	if (hotplug_port & BXT_DE_PORT_HP_DDIB)
+		hotplug_ctrl |= BXT_DDIB_HPD_ENABLE;
+	if (hotplug_port & BXT_DE_PORT_HP_DDIC)
+		hotplug_ctrl |= BXT_DDIC_HPD_ENABLE;
+	I915_WRITE(BXT_HOTPLUG_CTL, hotplug_ctrl);
+
+	/* Unmask DDI hotplug in IMR */
+	hotplug_ctrl = I915_READ(GEN8_DE_PORT_IMR) & ~hotplug_port;
+	I915_WRITE(GEN8_DE_PORT_IMR, hotplug_ctrl);
+
+	/* Enable DDI hotplug in IER */
+	hotplug_ctrl = I915_READ(GEN8_DE_PORT_IER) | hotplug_port;
+	I915_WRITE(GEN8_DE_PORT_IER, hotplug_ctrl);
+	POSTING_READ(GEN8_DE_PORT_IER);
+}
+
 static void ibx_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3448,13 +3541,16 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	uint32_t de_pipe_masked = GEN8_PIPE_CDCLK_CRC_DONE;
 	uint32_t de_pipe_enables;
 	int pipe;
-	u32 aux_en = GEN8_AUX_CHANNEL_A;
+	u32 de_port_en = GEN8_AUX_CHANNEL_A;
 
 	if (IS_GEN9(dev_priv)) {
 		de_pipe_masked |= GEN9_PIPE_PLANE1_FLIP_DONE |
 				  GEN9_DE_PIPE_IRQ_FAULT_ERRORS;
-		aux_en |= GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
+		de_port_en |= GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
 			GEN9_AUX_CHANNEL_D;
+
+		if (IS_BROXTON(dev_priv))
+			de_port_en |= BXT_DE_PORT_GMBUS;
 	} else
 		de_pipe_masked |= GEN8_PIPE_PRIMARY_FLIP_DONE |
 				  GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
@@ -3473,19 +3569,21 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 					  dev_priv->de_irq_mask[pipe],
 					  de_pipe_enables);
 
-	GEN5_IRQ_INIT(GEN8_DE_PORT_, ~aux_en, aux_en);
+	GEN5_IRQ_INIT(GEN8_DE_PORT_, ~de_port_en, de_port_en);
 }
 
 static int gen8_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	ibx_irq_pre_postinstall(dev);
+	if (HAS_PCH_SPLIT(dev))
+		ibx_irq_pre_postinstall(dev);
 
 	gen8_gt_irq_postinstall(dev_priv);
 	gen8_de_irq_postinstall(dev_priv);
 
-	ibx_irq_postinstall(dev);
+	if (HAS_PCH_SPLIT(dev))
+		ibx_irq_postinstall(dev);
 
 	I915_WRITE(GEN8_MASTER_IRQ, DE_MASTER_IRQ_CONTROL);
 	POSTING_READ(GEN8_MASTER_IRQ);
@@ -3694,7 +3792,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 		new_iir = I915_READ16(IIR); /* Flush posted writes */
 
 		if (iir & I915_USER_INTERRUPT)
-			notify_ring(dev, &dev_priv->ring[RCS]);
+			notify_ring(&dev_priv->ring[RCS]);
 
 		for_each_pipe(dev_priv, pipe) {
 			int plane = pipe;
@@ -3883,7 +3981,7 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 		new_iir = I915_READ(IIR); /* Flush posted writes */
 
 		if (iir & I915_USER_INTERRUPT)
-			notify_ring(dev, &dev_priv->ring[RCS]);
+			notify_ring(&dev_priv->ring[RCS]);
 
 		for_each_pipe(dev_priv, pipe) {
 			int plane = pipe;
@@ -4110,9 +4208,9 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		new_iir = I915_READ(IIR); /* Flush posted writes */
 
 		if (iir & I915_USER_INTERRUPT)
-			notify_ring(dev, &dev_priv->ring[RCS]);
+			notify_ring(&dev_priv->ring[RCS]);
 		if (iir & I915_BSD_USER_INTERRUPT)
-			notify_ring(dev, &dev_priv->ring[VCS]);
+			notify_ring(&dev_priv->ring[VCS]);
 
 		for_each_pipe(dev_priv, pipe) {
 			if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS &&
@@ -4294,7 +4392,10 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 		dev->driver->irq_uninstall = gen8_irq_uninstall;
 		dev->driver->enable_vblank = gen8_enable_vblank;
 		dev->driver->disable_vblank = gen8_disable_vblank;
-		dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
+		if (HAS_PCH_SPLIT(dev))
+			dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
+		else
+			dev_priv->display.hpd_irq_setup = bxt_hpd_irq_setup;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev->driver->irq_handler = ironlake_irq_handler;
 		dev->driver->irq_preinstall = ironlake_irq_reset;
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index bb64415a1c3e..8ac5a1b29ac0 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -53,6 +53,7 @@ struct i915_params i915 __read_mostly = {
 	.mmio_debug = 0,
 	.verbose_state_checks = 1,
 	.nuclear_pageflip = 0,
+	.edp_vswing = 0,
 };
 
 module_param_named(modeset, i915.modeset, int, 0400);
@@ -184,3 +185,10 @@ MODULE_PARM_DESC(verbose_state_checks,
 module_param_named_unsafe(nuclear_pageflip, i915.nuclear_pageflip, bool, 0600);
 MODULE_PARM_DESC(nuclear_pageflip,
 		 "Force atomic modeset functionality; only planes work for now (default: false).");
+
+/* WA to get away with the default setting in VBT for early platforms.Will be removed */
+module_param_named_unsafe(edp_vswing, i915.edp_vswing, int, 0400);
+MODULE_PARM_DESC(edp_vswing,
+		 "Ignore/Override vswing pre-emph table selection from VBT "
+		 "(0=use value from vbt [default], 1=low power swing(200mV),"
+		 "2=default swing(400mV))");
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 3aaed099e4fe..f5edb3504167 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -595,10 +595,6 @@ enum punit_power_well {
 	PUNIT_POWER_WELL_DPIO_RX0		= 10,
 	PUNIT_POWER_WELL_DPIO_RX1		= 11,
 	PUNIT_POWER_WELL_DPIO_CMN_D		= 12,
-	/* FIXME: guesswork below */
-	PUNIT_POWER_WELL_DPIO_TX_D_LANES_01	= 13,
-	PUNIT_POWER_WELL_DPIO_TX_D_LANES_23	= 14,
-	PUNIT_POWER_WELL_DPIO_RX2		= 15,
 
 	PUNIT_POWER_WELL_NUM,
 };
@@ -670,9 +666,13 @@ enum skl_disp_power_wells {
 #define   FB_FMAX_VMIN_FREQ_LO_SHIFT		27
 #define   FB_FMAX_VMIN_FREQ_LO_MASK		0xf8000000
 
+#define VLV_TURBO_SOC_OVERRIDE	0x04
+#define 	VLV_OVERRIDE_EN	1
+#define 	VLV_SOC_TDP_EN	(1 << 1)
+#define 	VLV_BIAS_CPU_125_SOC_875 (6 << 2)
+#define 	CHV_BIAS_CPU_50_SOC_50 (3 << 2)
+
 #define VLV_CZ_CLOCK_TO_MILLI_SEC		100000
-#define VLV_RP_UP_EI_THRESHOLD			90
-#define VLV_RP_DOWN_EI_THRESHOLD		70
 
 /* vlv2 north clock has */
 #define CCK_FUSE_REG				0x8
@@ -717,7 +717,7 @@ enum skl_disp_power_wells {
 /**
  * DOC: DPIO
  *
- * VLV and CHV have slightly peculiar display PHYs for driving DP/HDMI
+ * VLV, CHV and BXT have slightly peculiar display PHYs for driving DP/HDMI
  * ports. DPIO is the name given to such a display PHY. These PHYs
  * don't follow the standard programming model using direct MMIO
  * registers, and instead their registers must be accessed trough IOSF
@@ -748,7 +748,7 @@ enum skl_disp_power_wells {
  * controlled from the display controller side. No DPIO registers
  * need to be accessed during AUX communication,
  *
- * Generally the common lane corresponds to the pipe and
+ * Generally on VLV/CHV the common lane corresponds to the pipe and
  * the spline (PCS/TX) corresponds to the port.
  *
  * For dual channel PHY (VLV/CHV):
@@ -770,11 +770,17 @@ enum skl_disp_power_wells {
  *
  *  port D == PCS/TX CH0
  *
- * Note: digital port B is DDI0, digital port C is DDI1,
- * digital port D is DDI2
+ * On BXT the entire PHY channel corresponds to the port. That means
+ * the PLL is also now associated with the port rather than the pipe,
+ * and so the clock needs to be routed to the appropriate transcoder.
+ * Port A PLL is directly connected to transcoder EDP and port B/C
+ * PLLs can be routed to any transcoder A/B/C.
+ *
+ * Note: DDI0 is digital port B, DD1 is digital port C, and DDI2 is
+ * digital port D (CHV) or port A (BXT).
  */
 /*
- * Dual channel PHY (VLV/CHV)
+ * Dual channel PHY (VLV/CHV/BXT)
  * ---------------------------------
  * |      CH0      |      CH1      |
  * |  CMN/PLL/REF  |  CMN/PLL/REF  |
@@ -786,7 +792,7 @@ enum skl_disp_power_wells {
  * |     DDI0      |     DDI1      | DP/HDMI ports
  * ---------------------------------
  *
- * Single channel PHY (CHV)
+ * Single channel PHY (CHV/BXT)
  * -----------------
  * |      CH0      |
  * |  CMN/PLL/REF  |
@@ -951,6 +957,7 @@ enum skl_disp_power_wells {
 
 #define _VLV_PCS_DW11_CH0		0x822c
 #define _VLV_PCS_DW11_CH1		0x842c
+#define   DPIO_TX2_STAGGER_MASK(x)	((x)<<24)
 #define   DPIO_LANEDESKEW_STRAP_OVRD	(1<<3)
 #define   DPIO_LEFT_TXFIFO_RST_MASTER	(1<<1)
 #define   DPIO_RIGHT_TXFIFO_RST_MASTER	(1<<0)
@@ -963,8 +970,20 @@ enum skl_disp_power_wells {
 #define VLV_PCS01_DW11(ch) _PORT(ch, _VLV_PCS01_DW11_CH0, _VLV_PCS01_DW11_CH1)
 #define VLV_PCS23_DW11(ch) _PORT(ch, _VLV_PCS23_DW11_CH0, _VLV_PCS23_DW11_CH1)
 
+#define _VLV_PCS01_DW12_CH0		0x0230
+#define _VLV_PCS23_DW12_CH0		0x0430
+#define _VLV_PCS01_DW12_CH1		0x2630
+#define _VLV_PCS23_DW12_CH1		0x2830
+#define VLV_PCS01_DW12(ch) _PORT(ch, _VLV_PCS01_DW12_CH0, _VLV_PCS01_DW12_CH1)
+#define VLV_PCS23_DW12(ch) _PORT(ch, _VLV_PCS23_DW12_CH0, _VLV_PCS23_DW12_CH1)
+
 #define _VLV_PCS_DW12_CH0		0x8230
 #define _VLV_PCS_DW12_CH1		0x8430
+#define   DPIO_TX2_STAGGER_MULT(x)	((x)<<20)
+#define   DPIO_TX1_STAGGER_MULT(x)	((x)<<16)
+#define   DPIO_TX1_STAGGER_MASK(x)	((x)<<8)
+#define   DPIO_LANESTAGGER_STRAP_OVRD	(1<<6)
+#define   DPIO_LANESTAGGER_STRAP(x)	((x)<<0)
 #define VLV_PCS_DW12(ch) _PORT(ch, _VLV_PCS_DW12_CH0, _VLV_PCS_DW12_CH1)
 
 #define _VLV_PCS_DW14_CH0		0x8238
@@ -1119,6 +1138,245 @@ enum skl_disp_power_wells {
 #define   DPIO_FRC_LATENCY_SHFIT	8
 #define CHV_TX_DW14(ch, lane) _TXLANE(ch, lane, 0xb8)
 #define   DPIO_UPAR_SHIFT		30
+
+/* BXT PHY registers */
+#define _BXT_PHY(phy, a, b)		_PIPE((phy), (a), (b))
+
+#define BXT_P_CR_GT_DISP_PWRON		0x138090
+#define   GT_DISPLAY_POWER_ON(phy)	(1 << (phy))
+
+#define _PHY_CTL_FAMILY_EDP		0x64C80
+#define _PHY_CTL_FAMILY_DDI		0x64C90
+#define   COMMON_RESET_DIS		(1 << 31)
+#define BXT_PHY_CTL_FAMILY(phy)		_BXT_PHY((phy), _PHY_CTL_FAMILY_DDI, \
+							_PHY_CTL_FAMILY_EDP)
+
+/* BXT PHY PLL registers */
+#define _PORT_PLL_A			0x46074
+#define _PORT_PLL_B			0x46078
+#define _PORT_PLL_C			0x4607c
+#define   PORT_PLL_ENABLE		(1 << 31)
+#define   PORT_PLL_LOCK			(1 << 30)
+#define   PORT_PLL_REF_SEL		(1 << 27)
+#define BXT_PORT_PLL_ENABLE(port)	_PORT(port, _PORT_PLL_A, _PORT_PLL_B)
+
+#define _PORT_PLL_EBB_0_A		0x162034
+#define _PORT_PLL_EBB_0_B		0x6C034
+#define _PORT_PLL_EBB_0_C		0x6C340
+#define   PORT_PLL_P1_MASK		(0x07 << 13)
+#define   PORT_PLL_P1(x)		((x)  << 13)
+#define   PORT_PLL_P2_MASK		(0x1f << 8)
+#define   PORT_PLL_P2(x)		((x)  << 8)
+#define BXT_PORT_PLL_EBB_0(port)	_PORT3(port, _PORT_PLL_EBB_0_A, \
+						_PORT_PLL_EBB_0_B,	\
+						_PORT_PLL_EBB_0_C)
+
+#define _PORT_PLL_EBB_4_A		0x162038
+#define _PORT_PLL_EBB_4_B		0x6C038
+#define _PORT_PLL_EBB_4_C		0x6C344
+#define   PORT_PLL_10BIT_CLK_ENABLE	(1 << 13)
+#define   PORT_PLL_RECALIBRATE		(1 << 14)
+#define BXT_PORT_PLL_EBB_4(port)	_PORT3(port, _PORT_PLL_EBB_4_A, \
+						_PORT_PLL_EBB_4_B,	\
+						_PORT_PLL_EBB_4_C)
+
+#define _PORT_PLL_0_A			0x162100
+#define _PORT_PLL_0_B			0x6C100
+#define _PORT_PLL_0_C			0x6C380
+/* PORT_PLL_0_A */
+#define   PORT_PLL_M2_MASK		0xFF
+/* PORT_PLL_1_A */
+#define   PORT_PLL_N_MASK		(0x0F << 8)
+#define   PORT_PLL_N(x)			((x) << 8)
+/* PORT_PLL_2_A */
+#define   PORT_PLL_M2_FRAC_MASK		0x3FFFFF
+/* PORT_PLL_3_A */
+#define   PORT_PLL_M2_FRAC_ENABLE	(1 << 16)
+/* PORT_PLL_6_A */
+#define   PORT_PLL_PROP_COEFF_MASK	0xF
+#define   PORT_PLL_INT_COEFF_MASK	(0x1F << 8)
+#define   PORT_PLL_INT_COEFF(x)		((x)  << 8)
+#define   PORT_PLL_GAIN_CTL_MASK	(0x07 << 16)
+#define   PORT_PLL_GAIN_CTL(x)		((x)  << 16)
+/* PORT_PLL_8_A */
+#define   PORT_PLL_TARGET_CNT_MASK	0x3FF
+/* PORT_PLL_9_A */
+#define  PORT_PLL_LOCK_THRESHOLD_MASK	0xe
+/* PORT_PLL_10_A */
+#define  PORT_PLL_DCO_AMP_OVR_EN_H	(1<<27)
+#define  PORT_PLL_DCO_AMP_MASK		0x3c00
+#define  PORT_PLL_DCO_AMP(x)		(x<<10)
+#define _PORT_PLL_BASE(port)		_PORT3(port, _PORT_PLL_0_A,	\
+						_PORT_PLL_0_B,		\
+						_PORT_PLL_0_C)
+#define BXT_PORT_PLL(port, idx)		(_PORT_PLL_BASE(port) + (idx) * 4)
+
+/* BXT PHY common lane registers */
+#define _PORT_CL1CM_DW0_A		0x162000
+#define _PORT_CL1CM_DW0_BC		0x6C000
+#define   PHY_POWER_GOOD		(1 << 16)
+#define BXT_PORT_CL1CM_DW0(phy)		_BXT_PHY((phy), _PORT_CL1CM_DW0_BC, \
+							_PORT_CL1CM_DW0_A)
+
+#define _PORT_CL1CM_DW9_A		0x162024
+#define _PORT_CL1CM_DW9_BC		0x6C024
+#define   IREF0RC_OFFSET_SHIFT		8
+#define   IREF0RC_OFFSET_MASK		(0xFF << IREF0RC_OFFSET_SHIFT)
+#define BXT_PORT_CL1CM_DW9(phy)		_BXT_PHY((phy), _PORT_CL1CM_DW9_BC, \
+							_PORT_CL1CM_DW9_A)
+
+#define _PORT_CL1CM_DW10_A		0x162028
+#define _PORT_CL1CM_DW10_BC		0x6C028
+#define   IREF1RC_OFFSET_SHIFT		8
+#define   IREF1RC_OFFSET_MASK		(0xFF << IREF1RC_OFFSET_SHIFT)
+#define BXT_PORT_CL1CM_DW10(phy)	_BXT_PHY((phy), _PORT_CL1CM_DW10_BC, \
+							_PORT_CL1CM_DW10_A)
+
+#define _PORT_CL1CM_DW28_A		0x162070
+#define _PORT_CL1CM_DW28_BC		0x6C070
+#define   OCL1_POWER_DOWN_EN		(1 << 23)
+#define   DW28_OLDO_DYN_PWR_DOWN_EN	(1 << 22)
+#define   SUS_CLK_CONFIG		0x3
+#define BXT_PORT_CL1CM_DW28(phy)	_BXT_PHY((phy), _PORT_CL1CM_DW28_BC, \
+							_PORT_CL1CM_DW28_A)
+
+#define _PORT_CL1CM_DW30_A		0x162078
+#define _PORT_CL1CM_DW30_BC		0x6C078
+#define   OCL2_LDOFUSE_PWR_DIS		(1 << 6)
+#define BXT_PORT_CL1CM_DW30(phy)	_BXT_PHY((phy), _PORT_CL1CM_DW30_BC, \
+							_PORT_CL1CM_DW30_A)
+
+/* Defined for PHY0 only */
+#define BXT_PORT_CL2CM_DW6_BC		0x6C358
+#define   DW6_OLDO_DYN_PWR_DOWN_EN	(1 << 28)
+
+/* BXT PHY Ref registers */
+#define _PORT_REF_DW3_A			0x16218C
+#define _PORT_REF_DW3_BC		0x6C18C
+#define   GRC_DONE			(1 << 22)
+#define BXT_PORT_REF_DW3(phy)		_BXT_PHY((phy), _PORT_REF_DW3_BC, \
+							_PORT_REF_DW3_A)
+
+#define _PORT_REF_DW6_A			0x162198
+#define _PORT_REF_DW6_BC		0x6C198
+/*
+ * FIXME: BSpec/CHV ConfigDB disagrees on the following two fields, fix them
+ * after testing.
+ */
+#define   GRC_CODE_SHIFT		23
+#define   GRC_CODE_MASK			(0x1FF << GRC_CODE_SHIFT)
+#define   GRC_CODE_FAST_SHIFT		16
+#define   GRC_CODE_FAST_MASK		(0x7F << GRC_CODE_FAST_SHIFT)
+#define   GRC_CODE_SLOW_SHIFT		8
+#define   GRC_CODE_SLOW_MASK		(0xFF << GRC_CODE_SLOW_SHIFT)
+#define   GRC_CODE_NOM_MASK		0xFF
+#define BXT_PORT_REF_DW6(phy)		_BXT_PHY((phy), _PORT_REF_DW6_BC,	\
+						      _PORT_REF_DW6_A)
+
+#define _PORT_REF_DW8_A			0x1621A0
+#define _PORT_REF_DW8_BC		0x6C1A0
+#define   GRC_DIS			(1 << 15)
+#define   GRC_RDY_OVRD			(1 << 1)
+#define BXT_PORT_REF_DW8(phy)		_BXT_PHY((phy), _PORT_REF_DW8_BC,	\
+						      _PORT_REF_DW8_A)
+
+/* BXT PHY PCS registers */
+#define _PORT_PCS_DW10_LN01_A		0x162428
+#define _PORT_PCS_DW10_LN01_B		0x6C428
+#define _PORT_PCS_DW10_LN01_C		0x6C828
+#define _PORT_PCS_DW10_GRP_A		0x162C28
+#define _PORT_PCS_DW10_GRP_B		0x6CC28
+#define _PORT_PCS_DW10_GRP_C		0x6CE28
+#define BXT_PORT_PCS_DW10_LN01(port)	_PORT3(port, _PORT_PCS_DW10_LN01_A, \
+						     _PORT_PCS_DW10_LN01_B, \
+						     _PORT_PCS_DW10_LN01_C)
+#define BXT_PORT_PCS_DW10_GRP(port)	_PORT3(port, _PORT_PCS_DW10_GRP_A,  \
+						     _PORT_PCS_DW10_GRP_B,  \
+						     _PORT_PCS_DW10_GRP_C)
+#define   TX2_SWING_CALC_INIT		(1 << 31)
+#define   TX1_SWING_CALC_INIT		(1 << 30)
+
+#define _PORT_PCS_DW12_LN01_A		0x162430
+#define _PORT_PCS_DW12_LN01_B		0x6C430
+#define _PORT_PCS_DW12_LN01_C		0x6C830
+#define _PORT_PCS_DW12_LN23_A		0x162630
+#define _PORT_PCS_DW12_LN23_B		0x6C630
+#define _PORT_PCS_DW12_LN23_C		0x6CA30
+#define _PORT_PCS_DW12_GRP_A		0x162c30
+#define _PORT_PCS_DW12_GRP_B		0x6CC30
+#define _PORT_PCS_DW12_GRP_C		0x6CE30
+#define   LANESTAGGER_STRAP_OVRD	(1 << 6)
+#define   LANE_STAGGER_MASK		0x1F
+#define BXT_PORT_PCS_DW12_LN01(port)	_PORT3(port, _PORT_PCS_DW12_LN01_A, \
+						     _PORT_PCS_DW12_LN01_B, \
+						     _PORT_PCS_DW12_LN01_C)
+#define BXT_PORT_PCS_DW12_LN23(port)	_PORT3(port, _PORT_PCS_DW12_LN23_A, \
+						     _PORT_PCS_DW12_LN23_B, \
+						     _PORT_PCS_DW12_LN23_C)
+#define BXT_PORT_PCS_DW12_GRP(port)	_PORT3(port, _PORT_PCS_DW12_GRP_A, \
+						     _PORT_PCS_DW12_GRP_B, \
+						     _PORT_PCS_DW12_GRP_C)
+
+/* BXT PHY TX registers */
+#define _BXT_LANE_OFFSET(lane)           (((lane) >> 1) * 0x200 +	\
+					  ((lane) & 1) * 0x80)
+
+#define _PORT_TX_DW2_LN0_A		0x162508
+#define _PORT_TX_DW2_LN0_B		0x6C508
+#define _PORT_TX_DW2_LN0_C		0x6C908
+#define _PORT_TX_DW2_GRP_A		0x162D08
+#define _PORT_TX_DW2_GRP_B		0x6CD08
+#define _PORT_TX_DW2_GRP_C		0x6CF08
+#define BXT_PORT_TX_DW2_GRP(port)	_PORT3(port, _PORT_TX_DW2_GRP_A,  \
+						     _PORT_TX_DW2_GRP_B,  \
+						     _PORT_TX_DW2_GRP_C)
+#define BXT_PORT_TX_DW2_LN0(port)	_PORT3(port, _PORT_TX_DW2_LN0_A,  \
+						     _PORT_TX_DW2_LN0_B,  \
+						     _PORT_TX_DW2_LN0_C)
+#define   MARGIN_000_SHIFT		16
+#define   MARGIN_000			(0xFF << MARGIN_000_SHIFT)
+#define   UNIQ_TRANS_SCALE_SHIFT	8
+#define   UNIQ_TRANS_SCALE		(0xFF << UNIQ_TRANS_SCALE_SHIFT)
+
+#define _PORT_TX_DW3_LN0_A		0x16250C
+#define _PORT_TX_DW3_LN0_B		0x6C50C
+#define _PORT_TX_DW3_LN0_C		0x6C90C
+#define _PORT_TX_DW3_GRP_A		0x162D0C
+#define _PORT_TX_DW3_GRP_B		0x6CD0C
+#define _PORT_TX_DW3_GRP_C		0x6CF0C
+#define BXT_PORT_TX_DW3_GRP(port)	_PORT3(port, _PORT_TX_DW3_GRP_A,  \
+						     _PORT_TX_DW3_GRP_B,  \
+						     _PORT_TX_DW3_GRP_C)
+#define BXT_PORT_TX_DW3_LN0(port)	_PORT3(port, _PORT_TX_DW3_LN0_A,  \
+						     _PORT_TX_DW3_LN0_B,  \
+						     _PORT_TX_DW3_LN0_C)
+#define   UNIQE_TRANGE_EN_METHOD	(1 << 27)
+
+#define _PORT_TX_DW4_LN0_A		0x162510
+#define _PORT_TX_DW4_LN0_B		0x6C510
+#define _PORT_TX_DW4_LN0_C		0x6C910
+#define _PORT_TX_DW4_GRP_A		0x162D10
+#define _PORT_TX_DW4_GRP_B		0x6CD10
+#define _PORT_TX_DW4_GRP_C		0x6CF10
+#define BXT_PORT_TX_DW4_LN0(port)	_PORT3(port, _PORT_TX_DW4_LN0_A,  \
+						     _PORT_TX_DW4_LN0_B,  \
+						     _PORT_TX_DW4_LN0_C)
+#define BXT_PORT_TX_DW4_GRP(port)	_PORT3(port, _PORT_TX_DW4_GRP_A,  \
+						     _PORT_TX_DW4_GRP_B,  \
+						     _PORT_TX_DW4_GRP_C)
+#define   DEEMPH_SHIFT			24
+#define   DE_EMPHASIS			(0xFF << DEEMPH_SHIFT)
+
+#define _PORT_TX_DW14_LN0_A		0x162538
+#define _PORT_TX_DW14_LN0_B		0x6C538
+#define _PORT_TX_DW14_LN0_C		0x6C938
+#define   LATENCY_OPTIM_SHIFT		30
+#define   LATENCY_OPTIM			(1 << LATENCY_OPTIM_SHIFT)
+#define BXT_PORT_TX_DW14_LN(port, lane)	(_PORT3((port), _PORT_TX_DW14_LN0_A,   \
+							_PORT_TX_DW14_LN0_B,   \
+							_PORT_TX_DW14_LN0_C) + \
+					 _BXT_LANE_OFFSET(lane))
+
 /*
  * Fence registers
  */
@@ -1150,6 +1408,7 @@ enum skl_disp_power_wells {
 /* control register for cpu gtt access */
 #define TILECTL				0x101000
 #define   TILECTL_SWZCTL			(1 << 0)
+#define   TILECTL_TLBPF			(1 << 1)
 #define   TILECTL_TLB_PREFETCH_DIS	(1 << 2)
 #define   TILECTL_BACKSNOOP_DIS		(1 << 3)
 
@@ -1198,6 +1457,8 @@ enum skl_disp_power_wells {
 #define RING_HWS_PGA(base)	((base)+0x80)
 #define RING_HWS_PGA_GEN6(base)	((base)+0x2080)
 
+#define HSW_GTT_CACHE_EN	0x4024
+#define   GTT_CACHE_EN_ALL	0xF0007FFF
 #define GEN7_WR_WATERMARK	0x4028
 #define GEN7_GFX_PRIO_CTRL	0x402C
 #define ARB_MODE		0x4030
@@ -1554,9 +1815,7 @@ enum skl_disp_power_wells {
 #define   GEN9_F2_SS_DIS_SHIFT		20
 #define   GEN9_F2_SS_DIS_MASK		(0xf << GEN9_F2_SS_DIS_SHIFT)
 
-#define GEN8_EU_DISABLE0		0x9134
-#define GEN8_EU_DISABLE1		0x9138
-#define GEN8_EU_DISABLE2		0x913c
+#define GEN9_EU_DISABLE(slice)		(0x9134 + (slice)*0x4)
 
 #define GEN6_BSD_SLEEP_PSMI_CONTROL	0x12050
 #define   GEN6_BSD_SLEEP_MSG_DISABLE	(1 << 0)
@@ -1788,16 +2047,19 @@ enum skl_disp_power_wells {
 #define   GMBUS_RATE_400KHZ	(2<<8) /* reserved on Pineview */
 #define   GMBUS_RATE_1MHZ	(3<<8) /* reserved on Pineview */
 #define   GMBUS_HOLD_EXT	(1<<7) /* 300ns hold time, rsvd on Pineview */
-#define   GMBUS_PORT_DISABLED	0
-#define   GMBUS_PORT_SSC	1
-#define   GMBUS_PORT_VGADDC	2
-#define   GMBUS_PORT_PANEL	3
-#define   GMBUS_PORT_DPD_CHV	3 /* HDMID_CHV */
-#define   GMBUS_PORT_DPC	4 /* HDMIC */
-#define   GMBUS_PORT_DPB	5 /* SDVO, HDMIB */
-#define   GMBUS_PORT_DPD	6 /* HDMID */
-#define   GMBUS_PORT_RESERVED	7 /* 7 reserved */
-#define   GMBUS_NUM_PORTS	(GMBUS_PORT_DPD - GMBUS_PORT_SSC + 1)
+#define   GMBUS_PIN_DISABLED	0
+#define   GMBUS_PIN_SSC		1
+#define   GMBUS_PIN_VGADDC	2
+#define   GMBUS_PIN_PANEL	3
+#define   GMBUS_PIN_DPD_CHV	3 /* HDMID_CHV */
+#define   GMBUS_PIN_DPC		4 /* HDMIC */
+#define   GMBUS_PIN_DPB		5 /* SDVO, HDMIB */
+#define   GMBUS_PIN_DPD		6 /* HDMID */
+#define   GMBUS_PIN_RESERVED	7 /* 7 reserved */
+#define   GMBUS_PIN_1_BXT	1
+#define   GMBUS_PIN_2_BXT	2
+#define   GMBUS_PIN_3_BXT	3
+#define   GMBUS_NUM_PINS	7 /* including 0 */
 #define GMBUS1			0x5104 /* command/status */
 #define   GMBUS_SW_CLR_INT	(1<<31)
 #define   GMBUS_SW_RDY		(1<<30)
@@ -1879,7 +2141,14 @@ enum skl_disp_power_wells {
 #define DPIO_PHY_STATUS			(VLV_DISPLAY_BASE + 0x6240)
 #define   DPLL_PORTD_READY_MASK		(0xf)
 #define DISPLAY_PHY_CONTROL (VLV_DISPLAY_BASE + 0x60100)
-#define   PHY_COM_LANE_RESET_DEASSERT(phy) (1 << (phy))
+#define   PHY_LDO_DELAY_0NS			0x0
+#define   PHY_LDO_DELAY_200NS			0x1
+#define   PHY_LDO_DELAY_600NS			0x2
+#define   PHY_LDO_SEQ_DELAY(delay, phy)		((delay) << (2*(phy)+23))
+#define   PHY_CH_SU_PSR				0x1
+#define   PHY_CH_DEEP_PSR			0x7
+#define   PHY_CH_POWER_MODE(mode, phy, ch)	((mode) << (6*(phy)+3*(ch)+2))
+#define   PHY_COM_LANE_RESET_DEASSERT(phy)	(1 << (phy))
 #define DISPLAY_PHY_STATUS (VLV_DISPLAY_BASE + 0x60104)
 #define   PHY_POWERGOOD(phy)	(((phy) == DPIO_PHY0) ? (1<<31) : (1<<30))
 
@@ -2689,7 +2958,6 @@ enum skl_disp_power_wells {
 #define EDP_PSR_CTL(dev)			(EDP_PSR_BASE(dev) + 0)
 #define   EDP_PSR_ENABLE			(1<<31)
 #define   BDW_PSR_SINGLE_FRAME			(1<<30)
-#define   EDP_PSR_LINK_DISABLE			(0<<27)
 #define   EDP_PSR_LINK_STANDBY			(1<<27)
 #define   EDP_PSR_MIN_LINK_ENTRY_TIME_MASK	(3<<25)
 #define   EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES	(0<<25)
@@ -2749,6 +3017,20 @@ enum skl_disp_power_wells {
 #define   EDP_PSR_DEBUG_MASK_MEMUP	(1<<26)
 #define   EDP_PSR_DEBUG_MASK_HPD	(1<<25)
 
+#define EDP_PSR2_CTL			0x6f900
+#define   EDP_PSR2_ENABLE		(1<<31)
+#define   EDP_SU_TRACK_ENABLE		(1<<30)
+#define   EDP_MAX_SU_DISABLE_TIME(t)	((t)<<20)
+#define   EDP_MAX_SU_DISABLE_TIME_MASK	(0x1f<<20)
+#define   EDP_PSR2_TP2_TIME_500		(0<<8)
+#define   EDP_PSR2_TP2_TIME_100		(1<<8)
+#define   EDP_PSR2_TP2_TIME_2500	(2<<8)
+#define   EDP_PSR2_TP2_TIME_50		(3<<8)
+#define   EDP_PSR2_TP2_TIME_MASK	(3<<8)
+#define   EDP_PSR2_FRAME_BEFORE_SU_SHIFT 4
+#define   EDP_PSR2_FRAME_BEFORE_SU_MASK	(0xf<<4)
+#define   EDP_PSR2_IDLE_MASK		0xf
+
 /* VGA port control */
 #define ADPA			0x61100
 #define PCH_ADPA                0xe1100
@@ -3228,6 +3510,18 @@ enum skl_disp_power_wells {
 #define UTIL_PIN_CTL		0x48400
 #define   UTIL_PIN_ENABLE	(1 << 31)
 
+/* BXT backlight register definition. */
+#define BXT_BLC_PWM_CTL1			0xC8250
+#define   BXT_BLC_PWM_ENABLE			(1 << 31)
+#define   BXT_BLC_PWM_POLARITY			(1 << 29)
+#define BXT_BLC_PWM_FREQ1			0xC8254
+#define BXT_BLC_PWM_DUTY1			0xC8258
+
+#define BXT_BLC_PWM_CTL2			0xC8350
+#define BXT_BLC_PWM_FREQ2			0xC8354
+#define BXT_BLC_PWM_DUTY2			0xC8358
+
+
 #define PCH_GTC_CTL		0xe7000
 #define   PCH_GTC_ENABLE	(1 << 31)
 
@@ -4855,7 +5149,9 @@ enum skl_disp_power_wells {
 #define   PLANE_CTL_ALPHA_HW_PREMULTIPLY	(  3 << 4)
 #define   PLANE_CTL_ROTATE_MASK			0x3
 #define   PLANE_CTL_ROTATE_0			0x0
+#define   PLANE_CTL_ROTATE_90			0x1
 #define   PLANE_CTL_ROTATE_180			0x2
+#define   PLANE_CTL_ROTATE_270			0x3
 #define _PLANE_STRIDE_1_A			0x70188
 #define _PLANE_STRIDE_2_A			0x70288
 #define _PLANE_STRIDE_3_A			0x70388
@@ -4879,6 +5175,8 @@ enum skl_disp_power_wells {
 #define _PLANE_KEYMAX_2_A			0x702a0
 #define _PLANE_BUF_CFG_1_A			0x7027c
 #define _PLANE_BUF_CFG_2_A			0x7037c
+#define _PLANE_NV12_BUF_CFG_1_A		0x70278
+#define _PLANE_NV12_BUF_CFG_2_A		0x70378
 
 #define _PLANE_CTL_1_B				0x71180
 #define _PLANE_CTL_2_B				0x71280
@@ -4965,6 +5263,15 @@ enum skl_disp_power_wells {
 #define PLANE_BUF_CFG(pipe, plane)	\
 	_PLANE(plane, _PLANE_BUF_CFG_1(pipe), _PLANE_BUF_CFG_2(pipe))
 
+#define _PLANE_NV12_BUF_CFG_1_B		0x71278
+#define _PLANE_NV12_BUF_CFG_2_B		0x71378
+#define _PLANE_NV12_BUF_CFG_1(pipe)	\
+	_PIPE(pipe, _PLANE_NV12_BUF_CFG_1_A, _PLANE_NV12_BUF_CFG_1_B)
+#define _PLANE_NV12_BUF_CFG_2(pipe)	\
+	_PIPE(pipe, _PLANE_NV12_BUF_CFG_2_A, _PLANE_NV12_BUF_CFG_2_B)
+#define PLANE_NV12_BUF_CFG(pipe, plane)	\
+	_PLANE(plane, _PLANE_NV12_BUF_CFG_1(pipe), _PLANE_NV12_BUF_CFG_2(pipe))
+
 /* SKL new cursor registers */
 #define _CUR_BUF_CFG_A				0x7017c
 #define _CUR_BUF_CFG_B				0x7117c
@@ -5097,6 +5404,121 @@ enum skl_disp_power_wells {
 #define PS_WIN_SZ(pipe)		_PIPE(pipe, _PSA_WIN_SZ, _PSB_WIN_SZ)
 #define PS_WIN_POS(pipe)	_PIPE(pipe, _PSA_WIN_POS, _PSB_WIN_POS)
 
+/*
+ * Skylake scalers
+ */
+#define _PS_1A_CTRL      0x68180
+#define _PS_2A_CTRL      0x68280
+#define _PS_1B_CTRL      0x68980
+#define _PS_2B_CTRL      0x68A80
+#define _PS_1C_CTRL      0x69180
+#define PS_SCALER_EN        (1 << 31)
+#define PS_SCALER_MODE_MASK (3 << 28)
+#define PS_SCALER_MODE_DYN  (0 << 28)
+#define PS_SCALER_MODE_HQ  (1 << 28)
+#define PS_PLANE_SEL_MASK  (7 << 25)
+#define PS_PLANE_SEL(plane) ((plane + 1) << 25)
+#define PS_FILTER_MASK         (3 << 23)
+#define PS_FILTER_MEDIUM       (0 << 23)
+#define PS_FILTER_EDGE_ENHANCE (2 << 23)
+#define PS_FILTER_BILINEAR     (3 << 23)
+#define PS_VERT3TAP            (1 << 21)
+#define PS_VERT_INT_INVERT_FIELD1 (0 << 20)
+#define PS_VERT_INT_INVERT_FIELD0 (1 << 20)
+#define PS_PWRUP_PROGRESS         (1 << 17)
+#define PS_V_FILTER_BYPASS        (1 << 8)
+#define PS_VADAPT_EN              (1 << 7)
+#define PS_VADAPT_MODE_MASK        (3 << 5)
+#define PS_VADAPT_MODE_LEAST_ADAPT (0 << 5)
+#define PS_VADAPT_MODE_MOD_ADAPT   (1 << 5)
+#define PS_VADAPT_MODE_MOST_ADAPT  (3 << 5)
+
+#define _PS_PWR_GATE_1A     0x68160
+#define _PS_PWR_GATE_2A     0x68260
+#define _PS_PWR_GATE_1B     0x68960
+#define _PS_PWR_GATE_2B     0x68A60
+#define _PS_PWR_GATE_1C     0x69160
+#define PS_PWR_GATE_DIS_OVERRIDE       (1 << 31)
+#define PS_PWR_GATE_SETTLING_TIME_32   (0 << 3)
+#define PS_PWR_GATE_SETTLING_TIME_64   (1 << 3)
+#define PS_PWR_GATE_SETTLING_TIME_96   (2 << 3)
+#define PS_PWR_GATE_SETTLING_TIME_128  (3 << 3)
+#define PS_PWR_GATE_SLPEN_8             0
+#define PS_PWR_GATE_SLPEN_16            1
+#define PS_PWR_GATE_SLPEN_24            2
+#define PS_PWR_GATE_SLPEN_32            3
+
+#define _PS_WIN_POS_1A      0x68170
+#define _PS_WIN_POS_2A      0x68270
+#define _PS_WIN_POS_1B      0x68970
+#define _PS_WIN_POS_2B      0x68A70
+#define _PS_WIN_POS_1C      0x69170
+
+#define _PS_WIN_SZ_1A       0x68174
+#define _PS_WIN_SZ_2A       0x68274
+#define _PS_WIN_SZ_1B       0x68974
+#define _PS_WIN_SZ_2B       0x68A74
+#define _PS_WIN_SZ_1C       0x69174
+
+#define _PS_VSCALE_1A       0x68184
+#define _PS_VSCALE_2A       0x68284
+#define _PS_VSCALE_1B       0x68984
+#define _PS_VSCALE_2B       0x68A84
+#define _PS_VSCALE_1C       0x69184
+
+#define _PS_HSCALE_1A       0x68190
+#define _PS_HSCALE_2A       0x68290
+#define _PS_HSCALE_1B       0x68990
+#define _PS_HSCALE_2B       0x68A90
+#define _PS_HSCALE_1C       0x69190
+
+#define _PS_VPHASE_1A       0x68188
+#define _PS_VPHASE_2A       0x68288
+#define _PS_VPHASE_1B       0x68988
+#define _PS_VPHASE_2B       0x68A88
+#define _PS_VPHASE_1C       0x69188
+
+#define _PS_HPHASE_1A       0x68194
+#define _PS_HPHASE_2A       0x68294
+#define _PS_HPHASE_1B       0x68994
+#define _PS_HPHASE_2B       0x68A94
+#define _PS_HPHASE_1C       0x69194
+
+#define _PS_ECC_STAT_1A     0x681D0
+#define _PS_ECC_STAT_2A     0x682D0
+#define _PS_ECC_STAT_1B     0x689D0
+#define _PS_ECC_STAT_2B     0x68AD0
+#define _PS_ECC_STAT_1C     0x691D0
+
+#define _ID(id, a, b) ((a) + (id)*((b)-(a)))
+#define SKL_PS_CTRL(pipe, id) _PIPE(pipe,        \
+			_ID(id, _PS_1A_CTRL, _PS_2A_CTRL),       \
+			_ID(id, _PS_1B_CTRL, _PS_2B_CTRL))
+#define SKL_PS_PWR_GATE(pipe, id) _PIPE(pipe,    \
+			_ID(id, _PS_PWR_GATE_1A, _PS_PWR_GATE_2A), \
+			_ID(id, _PS_PWR_GATE_1B, _PS_PWR_GATE_2B))
+#define SKL_PS_WIN_POS(pipe, id) _PIPE(pipe,     \
+			_ID(id, _PS_WIN_POS_1A, _PS_WIN_POS_2A), \
+			_ID(id, _PS_WIN_POS_1B, _PS_WIN_POS_2B))
+#define SKL_PS_WIN_SZ(pipe, id)  _PIPE(pipe,     \
+			_ID(id, _PS_WIN_SZ_1A, _PS_WIN_SZ_2A),   \
+			_ID(id, _PS_WIN_SZ_1B, _PS_WIN_SZ_2B))
+#define SKL_PS_VSCALE(pipe, id)  _PIPE(pipe,     \
+			_ID(id, _PS_VSCALE_1A, _PS_VSCALE_2A),   \
+			_ID(id, _PS_VSCALE_1B, _PS_VSCALE_2B))
+#define SKL_PS_HSCALE(pipe, id)  _PIPE(pipe,     \
+			_ID(id, _PS_HSCALE_1A, _PS_HSCALE_2A),   \
+			_ID(id, _PS_HSCALE_1B, _PS_HSCALE_2B))
+#define SKL_PS_VPHASE(pipe, id)  _PIPE(pipe,     \
+			_ID(id, _PS_VPHASE_1A, _PS_VPHASE_2A),   \
+			_ID(id, _PS_VPHASE_1B, _PS_VPHASE_2B))
+#define SKL_PS_HPHASE(pipe, id)  _PIPE(pipe,     \
+			_ID(id, _PS_HPHASE_1A, _PS_HPHASE_2A),   \
+			_ID(id, _PS_HPHASE_1B, _PS_HPHASE_2B))
+#define SKL_PS_ECC_STAT(pipe, id)  _PIPE(pipe,     \
+			_ID(id, _PS_ECC_STAT_1A, _PS_ECC_STAT_2A),   \
+			_ID(id, _PS_ECC_STAT_1B, _PS_ECC_STAT_2B)
+
 /* legacy palette */
 #define _LGC_PALETTE_A           0x4a000
 #define _LGC_PALETTE_B           0x4a800
@@ -5219,9 +5641,11 @@ enum skl_disp_power_wells {
 #define  GEN8_PIPE_VSYNC		(1 << 1)
 #define  GEN8_PIPE_VBLANK		(1 << 0)
 #define  GEN9_PIPE_CURSOR_FAULT		(1 << 11)
+#define  GEN9_PIPE_PLANE4_FAULT		(1 << 10)
 #define  GEN9_PIPE_PLANE3_FAULT		(1 << 9)
 #define  GEN9_PIPE_PLANE2_FAULT		(1 << 8)
 #define  GEN9_PIPE_PLANE1_FAULT		(1 << 7)
+#define  GEN9_PIPE_PLANE4_FLIP_DONE	(1 << 6)
 #define  GEN9_PIPE_PLANE3_FLIP_DONE	(1 << 5)
 #define  GEN9_PIPE_PLANE2_FLIP_DONE	(1 << 4)
 #define  GEN9_PIPE_PLANE1_FLIP_DONE	(1 << 3)
@@ -5232,6 +5656,7 @@ enum skl_disp_power_wells {
 	 GEN8_PIPE_PRIMARY_FAULT)
 #define GEN9_DE_PIPE_IRQ_FAULT_ERRORS \
 	(GEN9_PIPE_CURSOR_FAULT | \
+	 GEN9_PIPE_PLANE4_FAULT | \
 	 GEN9_PIPE_PLANE3_FAULT | \
 	 GEN9_PIPE_PLANE2_FAULT | \
 	 GEN9_PIPE_PLANE1_FAULT)
@@ -5240,10 +5665,17 @@ enum skl_disp_power_wells {
 #define GEN8_DE_PORT_IMR 0x44444
 #define GEN8_DE_PORT_IIR 0x44448
 #define GEN8_DE_PORT_IER 0x4444c
-#define  GEN8_PORT_DP_A_HOTPLUG		(1 << 3)
 #define  GEN9_AUX_CHANNEL_D		(1 << 27)
 #define  GEN9_AUX_CHANNEL_C		(1 << 26)
 #define  GEN9_AUX_CHANNEL_B		(1 << 25)
+#define  BXT_DE_PORT_HP_DDIC		(1 << 5)
+#define  BXT_DE_PORT_HP_DDIB		(1 << 4)
+#define  BXT_DE_PORT_HP_DDIA		(1 << 3)
+#define  BXT_DE_PORT_HOTPLUG_MASK	(BXT_DE_PORT_HP_DDIA | \
+					 BXT_DE_PORT_HP_DDIB | \
+					 BXT_DE_PORT_HP_DDIC)
+#define  GEN8_PORT_DP_A_HOTPLUG		(1 << 3)
+#define  BXT_DE_PORT_GMBUS		(1 << 1)
 #define  GEN8_AUX_CHANNEL_A		(1 << 0)
 
 #define GEN8_DE_MISC_ISR 0x44460
@@ -5257,6 +5689,21 @@ enum skl_disp_power_wells {
 #define GEN8_PCU_IIR 0x444e8
 #define GEN8_PCU_IER 0x444ec
 
+/* BXT hotplug control */
+#define BXT_HOTPLUG_CTL			0xC4030
+#define   BXT_DDIA_HPD_ENABLE		(1 << 28)
+#define   BXT_DDIA_HPD_STATUS		(3 << 24)
+#define   BXT_DDIC_HPD_ENABLE		(1 << 12)
+#define   BXT_DDIC_HPD_STATUS		(3 << 8)
+#define   BXT_DDIB_HPD_ENABLE		(1 << 4)
+#define   BXT_DDIB_HPD_STATUS		(3 << 0)
+#define   BXT_HOTPLUG_CTL_MASK		(BXT_DDIA_HPD_ENABLE | \
+					 BXT_DDIB_HPD_ENABLE | \
+					 BXT_DDIC_HPD_ENABLE)
+#define   BXT_HPD_STATUS_MASK		(BXT_DDIA_HPD_STATUS | \
+					 BXT_DDIB_HPD_STATUS | \
+					 BXT_DDIC_HPD_STATUS)
+
 #define ILK_DISPLAY_CHICKEN2	0x42004
 /* Required on all Ironlake and Sandybridge according to the B-Spec. */
 #define  ILK_ELPIN_409_SELECT	(1 << 25)
@@ -5297,13 +5744,16 @@ enum skl_disp_power_wells {
 #define  DISP_FBC_WM_DIS		(1<<15)
 #define DISP_ARB_CTL2	0x45004
 #define  DISP_DATA_PARTITION_5_6	(1<<6)
+#define DBUF_CTL	0x45008
+#define  DBUF_POWER_REQUEST		(1<<31)
+#define  DBUF_POWER_STATE		(1<<30)
 #define GEN7_MSG_CTL	0x45010
 #define  WAIT_FOR_PCH_RESET_ACK		(1<<1)
 #define  WAIT_FOR_PCH_FLR_ACK		(1<<0)
 #define HSW_NDE_RSTWRN_OPT	0x46408
 #define  RESET_PCH_HANDSHAKE_ENABLE	(1<<4)
 
-#define FF_SLICE_CS_CHICKEN2			0x02e4
+#define FF_SLICE_CS_CHICKEN2			0x20e4
 #define  GEN9_TSG_BARRIER_ACK_DISABLE		(1<<8)
 
 /* GEN7 chicken */
@@ -5323,6 +5773,9 @@ enum skl_disp_power_wells {
 #define GEN7_L3SQCREG1				0xB010
 #define  VLV_B0_WA_L3SQCREG1_VALUE		0x00D30000
 
+#define GEN8_L3SQCREG1				0xB100
+#define  BDW_WA_L3SQCREG1_DEFAULT		0x784000
+
 #define GEN7_L3CNTLREG1				0xB01C
 #define  GEN7_WA_FOR_GEN7_L3_CONTROL			0x3C47FF8C
 #define  GEN7_L3AGDIS				(1<<19)
@@ -5340,12 +5793,17 @@ enum skl_disp_power_wells {
 
 /* GEN8 chicken */
 #define HDC_CHICKEN0				0x7300
+#define  HDC_FORCE_CSR_NON_COHERENT_OVR_DISABLE	(1<<15)
 #define  HDC_FENCE_DEST_SLM_DISABLE		(1<<14)
 #define  HDC_DONOT_FETCH_MEM_WHEN_MASKED	(1<<11)
 #define  HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT	(1<<5)
 #define  HDC_FORCE_NON_COHERENT			(1<<4)
 #define  HDC_BARRIER_PERFORMANCE_DISABLE	(1<<10)
 
+/* GEN9 chicken */
+#define SLICE_ECO_CHICKEN0			0x7308
+#define   PIXEL_MASK_CAMMING_DISABLE		(1 << 14)
+
 /* WaCatErrorRejectionIssue */
 #define GEN7_SQ_CHICKEN_MBCUNIT_CONFIG		0x9030
 #define  GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB	(1<<11)
@@ -5984,6 +6442,7 @@ enum skl_disp_power_wells {
 #define  TRANS_DP_PORT_SEL_D	(2<<29)
 #define  TRANS_DP_PORT_SEL_NONE	(3<<29)
 #define  TRANS_DP_PORT_SEL_MASK	(3<<29)
+#define  TRANS_DP_PIPE_TO_PORT(val)	((((val) & TRANS_DP_PORT_SEL_MASK) >> 29) + PORT_B)
 #define  TRANS_DP_AUDIO_ONLY	(1<<26)
 #define  TRANS_DP_ENH_FRAMING	(1<<18)
 #define  TRANS_DP_8BPC		(0<<9)
@@ -6088,6 +6547,7 @@ enum skl_disp_power_wells {
 # define GEN6_CSUNIT_CLOCK_GATE_DISABLE			(1 << 7)
 
 #define GEN6_UCGCTL2				0x9404
+# define GEN6_VFUNIT_CLOCK_GATE_DISABLE			(1 << 31)
 # define GEN7_VDSUNIT_CLOCK_GATE_DISABLE		(1 << 30)
 # define GEN7_TDLUNIT_CLOCK_GATE_DISABLE		(1 << 22)
 # define GEN6_RCZUNIT_CLOCK_GATE_DISABLE		(1 << 13)
@@ -6106,6 +6566,7 @@ enum skl_disp_power_wells {
 #define GEN8_UCGCTL6				0x9430
 #define   GEN8_GAPSUNIT_CLOCK_GATE_DISABLE	(1<<24)
 #define   GEN8_SDEUNIT_CLOCK_GATE_DISABLE	(1<<14)
+#define   GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ (1<<28)
 
 #define GEN6_GFXPAUSE				0xA000
 #define GEN6_RPNSWREQ				0xA008
@@ -6185,6 +6646,8 @@ enum skl_disp_power_wells {
 #define GEN9_MEDIA_PG_IDLE_HYSTERESIS		0xA0C4
 #define GEN9_RENDER_PG_IDLE_HYSTERESIS		0xA0C8
 #define GEN9_PG_ENABLE				0xA210
+#define GEN9_RENDER_PG_ENABLE			(1<<0)
+#define GEN9_MEDIA_PG_ENABLE			(1<<1)
 
 #define VLV_CHICKEN_3				(VLV_DISPLAY_BASE + 0x7040C)
 #define  PIXEL_OVERLAP_CNT_MASK			(3 << 30)
@@ -6230,15 +6693,24 @@ enum skl_disp_power_wells {
 
 #define GEN6_PCODE_MAILBOX			0x138124
 #define   GEN6_PCODE_READY			(1<<31)
-#define   GEN6_READ_OC_PARAMS			0xc
-#define   GEN6_PCODE_WRITE_MIN_FREQ_TABLE	0x8
-#define   GEN6_PCODE_READ_MIN_FREQ_TABLE	0x9
 #define	  GEN6_PCODE_WRITE_RC6VIDS		0x4
 #define	  GEN6_PCODE_READ_RC6VIDS		0x5
+#define     GEN6_ENCODE_RC6_VID(mv)		(((mv) - 245) / 5)
+#define     GEN6_DECODE_RC6_VID(vids)		(((vids) * 5) + 245)
+#define   GEN9_PCODE_READ_MEM_LATENCY		0x6
+#define     GEN9_MEM_LATENCY_LEVEL_MASK		0xFF
+#define     GEN9_MEM_LATENCY_LEVEL_1_5_SHIFT	8
+#define     GEN9_MEM_LATENCY_LEVEL_2_6_SHIFT	16
+#define     GEN9_MEM_LATENCY_LEVEL_3_7_SHIFT	24
+#define   SKL_PCODE_CDCLK_CONTROL		0x7
+#define     SKL_CDCLK_PREPARE_FOR_CHANGE	0x3
+#define     SKL_CDCLK_READY_FOR_CHANGE		0x1
+#define   GEN6_PCODE_WRITE_MIN_FREQ_TABLE	0x8
+#define   GEN6_PCODE_READ_MIN_FREQ_TABLE	0x9
+#define   GEN6_READ_OC_PARAMS			0xc
 #define   GEN6_PCODE_READ_D_COMP		0x10
 #define   GEN6_PCODE_WRITE_D_COMP		0x11
-#define   GEN6_ENCODE_RC6_VID(mv)		(((mv) - 245) / 5)
-#define   GEN6_DECODE_RC6_VID(vids)		(((vids) * 5) + 245)
+#define   HSW_PCODE_DE_WRITE_FREQ_REQ		0x17
 #define   DISPLAY_IPS_CONTROL			0x19
 #define	  HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL	0x1A
 #define GEN6_PCODE_DATA				0x138128
@@ -6246,12 +6718,6 @@ enum skl_disp_power_wells {
 #define   GEN6_PCODE_FREQ_RING_RATIO_SHIFT	16
 #define GEN6_PCODE_DATA1			0x13812C
 
-#define   GEN9_PCODE_READ_MEM_LATENCY		0x6
-#define   GEN9_MEM_LATENCY_LEVEL_MASK		0xFF
-#define   GEN9_MEM_LATENCY_LEVEL_1_5_SHIFT	8
-#define   GEN9_MEM_LATENCY_LEVEL_2_6_SHIFT	16
-#define   GEN9_MEM_LATENCY_LEVEL_3_7_SHIFT	24
-
 #define GEN6_GT_CORE_STATUS		0x138060
 #define   GEN6_CORE_CPD_STATE_MASK	(7<<4)
 #define   GEN6_RCn_MASK			7
@@ -6271,17 +6737,12 @@ enum skl_disp_power_wells {
 #define CHV_POWER_SS1_SIG2		0xa72c
 #define   CHV_EU311_PG_ENABLE		(1<<1)
 
-#define GEN9_SLICE0_PGCTL_ACK		0x804c
-#define GEN9_SLICE1_PGCTL_ACK		0x8050
-#define GEN9_SLICE2_PGCTL_ACK		0x8054
+#define GEN9_SLICE_PGCTL_ACK(slice)	(0x804c + (slice)*0x4)
 #define   GEN9_PGCTL_SLICE_ACK		(1 << 0)
+#define   GEN9_PGCTL_SS_ACK(subslice)	(1 << (2 + (subslice)*2))
 
-#define GEN9_SLICE0_SS01_EU_PGCTL_ACK	0x805c
-#define GEN9_SLICE0_SS23_EU_PGCTL_ACK	0x8060
-#define GEN9_SLICE1_SS01_EU_PGCTL_ACK	0x8064
-#define GEN9_SLICE1_SS23_EU_PGCTL_ACK	0x8068
-#define GEN9_SLICE2_SS01_EU_PGCTL_ACK	0x806c
-#define GEN9_SLICE2_SS23_EU_PGCTL_ACK	0x8070
+#define GEN9_SS01_EU_PGCTL_ACK(slice)	(0x805c + (slice)*0x8)
+#define GEN9_SS23_EU_PGCTL_ACK(slice)	(0x8060 + (slice)*0x8)
 #define   GEN9_PGCTL_SSA_EU08_ACK	(1 << 0)
 #define   GEN9_PGCTL_SSA_EU19_ACK	(1 << 2)
 #define   GEN9_PGCTL_SSA_EU210_ACK	(1 << 4)
@@ -6317,6 +6778,7 @@ enum skl_disp_power_wells {
 #define GEN7_HALF_SLICE_CHICKEN1_GT2	0xf100
 #define   GEN7_MAX_PS_THREAD_DEP		(8<<12)
 #define   GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE	(1<<10)
+#define   GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE	(1<<4)
 #define   GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE	(1<<3)
 
 #define GEN9_HALF_SLICE_CHICKEN5	0xe188
@@ -6718,6 +7180,13 @@ enum skl_disp_power_wells {
 #define  CDCLK_FREQ_675_617		(3<<26)
 #define  CDCLK_FREQ_DECIMAL_MASK	(0x7ff)
 
+#define  BXT_CDCLK_CD2X_DIV_SEL_MASK	(3<<22)
+#define  BXT_CDCLK_CD2X_DIV_SEL_1	(0<<22)
+#define  BXT_CDCLK_CD2X_DIV_SEL_1_5	(1<<22)
+#define  BXT_CDCLK_CD2X_DIV_SEL_2	(2<<22)
+#define  BXT_CDCLK_CD2X_DIV_SEL_4	(3<<22)
+#define  BXT_CDCLK_SSA_PRECHARGE_ENABLE	(1<<16)
+
 /* LCPLL_CTL */
 #define LCPLL1_CTL		0x46010
 #define LCPLL2_CTL		0x46014
@@ -6727,16 +7196,16 @@ enum skl_disp_power_wells {
 #define DPLL_CTRL1		0x6C058
 #define  DPLL_CTRL1_HDMI_MODE(id)		(1<<((id)*6+5))
 #define  DPLL_CTRL1_SSC(id)			(1<<((id)*6+4))
-#define  DPLL_CRTL1_LINK_RATE_MASK(id)		(7<<((id)*6+1))
-#define  DPLL_CRTL1_LINK_RATE_SHIFT(id)		((id)*6+1)
-#define  DPLL_CRTL1_LINK_RATE(linkrate, id)	((linkrate)<<((id)*6+1))
+#define  DPLL_CTRL1_LINK_RATE_MASK(id)		(7<<((id)*6+1))
+#define  DPLL_CTRL1_LINK_RATE_SHIFT(id)		((id)*6+1)
+#define  DPLL_CTRL1_LINK_RATE(linkrate, id)	((linkrate)<<((id)*6+1))
 #define  DPLL_CTRL1_OVERRIDE(id)		(1<<((id)*6))
-#define  DPLL_CRTL1_LINK_RATE_2700		0
-#define  DPLL_CRTL1_LINK_RATE_1350		1
-#define  DPLL_CRTL1_LINK_RATE_810		2
-#define  DPLL_CRTL1_LINK_RATE_1620		3
-#define  DPLL_CRTL1_LINK_RATE_1080		4
-#define  DPLL_CRTL1_LINK_RATE_2160		5
+#define  DPLL_CTRL1_LINK_RATE_2700		0
+#define  DPLL_CTRL1_LINK_RATE_1350		1
+#define  DPLL_CTRL1_LINK_RATE_810		2
+#define  DPLL_CTRL1_LINK_RATE_1620		3
+#define  DPLL_CTRL1_LINK_RATE_1080		4
+#define  DPLL_CTRL1_LINK_RATE_2160		5
 
 /* DPLL control2 */
 #define DPLL_CTRL2				0x6C05C
@@ -6782,6 +7251,31 @@ enum skl_disp_power_wells {
 #define GET_CFG_CR1_REG(id) (DPLL1_CFGCR1 + (id - SKL_DPLL1) * 8)
 #define GET_CFG_CR2_REG(id) (DPLL1_CFGCR2 + (id - SKL_DPLL1) * 8)
 
+/* BXT display engine PLL */
+#define BXT_DE_PLL_CTL			0x6d000
+#define   BXT_DE_PLL_RATIO(x)		(x)	/* {60,65,100} * 19.2MHz */
+#define   BXT_DE_PLL_RATIO_MASK		0xff
+
+#define BXT_DE_PLL_ENABLE		0x46070
+#define   BXT_DE_PLL_PLL_ENABLE		(1 << 31)
+#define   BXT_DE_PLL_LOCK		(1 << 30)
+
+/* GEN9 DC */
+#define DC_STATE_EN			0x45504
+#define  DC_STATE_EN_UPTO_DC5		(1<<0)
+#define  DC_STATE_EN_DC9		(1<<3)
+
+/*
+* SKL DC
+*/
+#define  DC_STATE_EN			0x45504
+#define  DC_STATE_EN_UPTO_DC5		(1<<0)
+#define  DC_STATE_EN_UPTO_DC6		(2<<0)
+#define  DC_STATE_EN_UPTO_DC5_DC6_MASK   0x3
+
+#define  DC_STATE_DEBUG                  0x45520
+#define  DC_STATE_DEBUG_MASK_MEMORY_UP	(1<<1)
+
 /* Please see hsw_read_dcomp() and hsw_write_dcomp() before using this register,
  * since on HSW we can't write to it using I915_WRITE. */
 #define D_COMP_HSW			(MCHBAR_MIRROR_BASE_SNB + 0x5F0C)
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index 5fda6c70b423..497cba5deb1e 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -220,7 +220,7 @@ DEFINE_EVENT(i915_page_table_entry, i915_page_table_entry_alloc,
 
 DECLARE_EVENT_CLASS(i915_page_table_entry_update,
 	TP_PROTO(struct i915_address_space *vm, u32 pde,
-		 struct i915_page_table_entry *pt, u32 first, u32 count, u32 bits),
+		 struct i915_page_table *pt, u32 first, u32 count, u32 bits),
 	TP_ARGS(vm, pde, pt, first, count, bits),
 
 	TP_STRUCT__entry(
@@ -250,7 +250,7 @@ DECLARE_EVENT_CLASS(i915_page_table_entry_update,
 
 DEFINE_EVENT(i915_page_table_entry_update, i915_page_table_entry_map,
 	TP_PROTO(struct i915_address_space *vm, u32 pde,
-		 struct i915_page_table_entry *pt, u32 first, u32 count, u32 bits),
+		 struct i915_page_table *pt, u32 first, u32 count, u32 bits),
 	TP_ARGS(vm, pde, pt, first, count, bits)
 );
 
@@ -504,7 +504,6 @@ DECLARE_EVENT_CLASS(i915_gem_request,
 	    TP_STRUCT__entry(
 			     __field(u32, dev)
 			     __field(u32, ring)
-			     __field(u32, uniq)
 			     __field(u32, seqno)
 			     ),
 
@@ -513,13 +512,11 @@ DECLARE_EVENT_CLASS(i915_gem_request,
 						i915_gem_request_get_ring(req);
 			   __entry->dev = ring->dev->primary->index;
 			   __entry->ring = ring->id;
-			   __entry->uniq = req ? req->uniq : 0;
 			   __entry->seqno = i915_gem_request_get_seqno(req);
 			   ),
 
-	    TP_printk("dev=%u, ring=%u, uniq=%u, seqno=%u",
-		      __entry->dev, __entry->ring, __entry->uniq,
-		      __entry->seqno)
+	    TP_printk("dev=%u, ring=%u, seqno=%u",
+		      __entry->dev, __entry->ring, __entry->seqno)
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_add,
@@ -564,7 +561,6 @@ TRACE_EVENT(i915_gem_request_wait_begin,
 	    TP_STRUCT__entry(
 			     __field(u32, dev)
 			     __field(u32, ring)
-			     __field(u32, uniq)
 			     __field(u32, seqno)
 			     __field(bool, blocking)
 			     ),
@@ -580,14 +576,13 @@ TRACE_EVENT(i915_gem_request_wait_begin,
 						i915_gem_request_get_ring(req);
 			   __entry->dev = ring->dev->primary->index;
 			   __entry->ring = ring->id;
-			   __entry->uniq = req ? req->uniq : 0;
 			   __entry->seqno = i915_gem_request_get_seqno(req);
 			   __entry->blocking =
 				     mutex_is_locked(&ring->dev->struct_mutex);
 			   ),
 
-	    TP_printk("dev=%u, ring=%u, uniq=%u, seqno=%u, blocking=%s",
-		      __entry->dev, __entry->ring, __entry->uniq,
+	    TP_printk("dev=%u, ring=%u, seqno=%u, blocking=%s",
+		      __entry->dev, __entry->ring,
 		      __entry->seqno, __entry->blocking ?  "yes (NB)" : "no")
 );
 
@@ -596,33 +591,6 @@ DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_end,
 	    TP_ARGS(req)
 );
 
-DECLARE_EVENT_CLASS(i915_ring,
-	    TP_PROTO(struct intel_engine_cs *ring),
-	    TP_ARGS(ring),
-
-	    TP_STRUCT__entry(
-			     __field(u32, dev)
-			     __field(u32, ring)
-			     ),
-
-	    TP_fast_assign(
-			   __entry->dev = ring->dev->primary->index;
-			   __entry->ring = ring->id;
-			   ),
-
-	    TP_printk("dev=%u, ring=%u", __entry->dev, __entry->ring)
-);
-
-DEFINE_EVENT(i915_ring, i915_ring_wait_begin,
-	    TP_PROTO(struct intel_engine_cs *ring),
-	    TP_ARGS(ring)
-);
-
-DEFINE_EVENT(i915_ring, i915_ring_wait_end,
-	    TP_PROTO(struct intel_engine_cs *ring),
-	    TP_ARGS(ring)
-);
-
 TRACE_EVENT(i915_flip_request,
 	    TP_PROTO(int plane, struct drm_i915_gem_object *obj),
 
diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 3903b90fb64e..7ed8033aae60 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -48,6 +48,8 @@ int intel_atomic_check(struct drm_device *dev,
 	int ncrtcs = dev->mode_config.num_crtc;
 	int nconnectors = dev->mode_config.num_connector;
 	enum pipe nuclear_pipe = INVALID_PIPE;
+	struct intel_crtc *nuclear_crtc = NULL;
+	struct intel_crtc_state *crtc_state = NULL;
 	int ret;
 	int i;
 	bool not_nuclear = false;
@@ -76,8 +78,14 @@ int intel_atomic_check(struct drm_device *dev,
 	state->allow_modeset = false;
 	for (i = 0; i < ncrtcs; i++) {
 		struct intel_crtc *crtc = to_intel_crtc(state->crtcs[i]);
+		if (crtc)
+			memset(&crtc->atomic, 0, sizeof(crtc->atomic));
 		if (crtc && crtc->pipe != nuclear_pipe)
 			not_nuclear = true;
+		if (crtc && crtc->pipe == nuclear_pipe) {
+			nuclear_crtc = crtc;
+			crtc_state = to_intel_crtc_state(state->crtc_states[i]);
+		}
 	}
 	for (i = 0; i < nconnectors; i++)
 		if (state->connectors[i] != NULL)
@@ -92,6 +100,11 @@ int intel_atomic_check(struct drm_device *dev,
 	if (ret)
 		return ret;
 
+	/* FIXME: move to crtc atomic check function once it is ready */
+	ret = intel_atomic_setup_scalers(dev, nuclear_crtc, crtc_state);
+	if (ret)
+		return ret;
+
 	return ret;
 }
 
@@ -155,6 +168,21 @@ int intel_atomic_commit(struct drm_device *dev,
 		swap(state->plane_states[i], plane->state);
 		plane->state->state = NULL;
 	}
+
+	/* swap crtc_scaler_state */
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
+		struct drm_crtc *crtc = state->crtcs[i];
+		if (!crtc) {
+			continue;
+		}
+
+		to_intel_crtc(crtc)->config->scaler_state =
+			to_intel_crtc_state(state->crtc_states[i])->scaler_state;
+
+		if (INTEL_INFO(dev)->gen >= 9)
+			skl_detach_scalers(to_intel_crtc(crtc));
+	}
+
 	drm_atomic_helper_commit_planes(dev, state);
 	drm_atomic_helper_wait_for_vblanks(dev, state);
 	drm_atomic_helper_cleanup_planes(dev, state);
@@ -222,8 +250,12 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
 		crtc_state = kmemdup(intel_crtc->config,
 				     sizeof(*intel_crtc->config), GFP_KERNEL);
 
-	if (crtc_state)
-		crtc_state->base.crtc = crtc;
+	if (!crtc_state)
+		return NULL;
+
+	__drm_atomic_helper_crtc_duplicate_state(crtc, &crtc_state->base);
+
+	crtc_state->base.crtc = crtc;
 
 	return &crtc_state->base;
 }
@@ -241,3 +273,151 @@ intel_crtc_destroy_state(struct drm_crtc *crtc,
 {
 	drm_atomic_helper_crtc_destroy_state(crtc, state);
 }
+
+/**
+ * intel_atomic_setup_scalers() - setup scalers for crtc per staged requests
+ * @dev: DRM device
+ * @crtc: intel crtc
+ * @crtc_state: incoming crtc_state to validate and setup scalers
+ *
+ * This function sets up scalers based on staged scaling requests for
+ * a @crtc and its planes. It is called from crtc level check path. If request
+ * is a supportable request, it attaches scalers to requested planes and crtc.
+ *
+ * This function takes into account the current scaler(s) in use by any planes
+ * not being part of this atomic state
+ *
+ *  Returns:
+ *         0 - scalers were setup succesfully
+ *         error code - otherwise
+ */
+int intel_atomic_setup_scalers(struct drm_device *dev,
+	struct intel_crtc *intel_crtc,
+	struct intel_crtc_state *crtc_state)
+{
+	struct drm_plane *plane = NULL;
+	struct intel_plane *intel_plane;
+	struct intel_plane_state *plane_state = NULL;
+	struct intel_crtc_scaler_state *scaler_state;
+	struct drm_atomic_state *drm_state;
+	int num_scalers_need;
+	int i, j;
+
+	if (INTEL_INFO(dev)->gen < 9 || !intel_crtc || !crtc_state)
+		return 0;
+
+	scaler_state = &crtc_state->scaler_state;
+	drm_state = crtc_state->base.state;
+
+	num_scalers_need = hweight32(scaler_state->scaler_users);
+	DRM_DEBUG_KMS("crtc_state = %p need = %d avail = %d scaler_users = 0x%x\n",
+		crtc_state, num_scalers_need, intel_crtc->num_scalers,
+		scaler_state->scaler_users);
+
+	/*
+	 * High level flow:
+	 * - staged scaler requests are already in scaler_state->scaler_users
+	 * - check whether staged scaling requests can be supported
+	 * - add planes using scalers that aren't in current transaction
+	 * - assign scalers to requested users
+	 * - as part of plane commit, scalers will be committed
+	 *   (i.e., either attached or detached) to respective planes in hw
+	 * - as part of crtc_commit, scaler will be either attached or detached
+	 *   to crtc in hw
+	 */
+
+	/* fail if required scalers > available scalers */
+	if (num_scalers_need > intel_crtc->num_scalers){
+		DRM_DEBUG_KMS("Too many scaling requests %d > %d\n",
+			num_scalers_need, intel_crtc->num_scalers);
+		return -EINVAL;
+	}
+
+	/* walkthrough scaler_users bits and start assigning scalers */
+	for (i = 0; i < sizeof(scaler_state->scaler_users) * 8; i++) {
+		int *scaler_id;
+
+		/* skip if scaler not required */
+		if (!(scaler_state->scaler_users & (1 << i)))
+			continue;
+
+		if (i == SKL_CRTC_INDEX) {
+			/* panel fitter case: assign as a crtc scaler */
+			scaler_id = &scaler_state->scaler_id;
+		} else {
+			if (!drm_state)
+				continue;
+
+			/* plane scaler case: assign as a plane scaler */
+			/* find the plane that set the bit as scaler_user */
+			plane = drm_state->planes[i];
+
+			/*
+			 * to enable/disable hq mode, add planes that are using scaler
+			 * into this transaction
+			 */
+			if (!plane) {
+				struct drm_plane_state *state;
+				plane = drm_plane_from_index(dev, i);
+				state = drm_atomic_get_plane_state(drm_state, plane);
+				if (IS_ERR(state)) {
+					DRM_DEBUG_KMS("Failed to add [PLANE:%d] to drm_state\n",
+						plane->base.id);
+					return PTR_ERR(state);
+				}
+			}
+
+			intel_plane = to_intel_plane(plane);
+
+			/* plane on different crtc cannot be a scaler user of this crtc */
+			if (WARN_ON(intel_plane->pipe != intel_crtc->pipe)) {
+				continue;
+			}
+
+			plane_state = to_intel_plane_state(drm_state->plane_states[i]);
+			scaler_id = &plane_state->scaler_id;
+		}
+
+		if (*scaler_id < 0) {
+			/* find a free scaler */
+			for (j = 0; j < intel_crtc->num_scalers; j++) {
+				if (!scaler_state->scalers[j].in_use) {
+					scaler_state->scalers[j].in_use = 1;
+					*scaler_id = scaler_state->scalers[j].id;
+					DRM_DEBUG_KMS("Attached scaler id %u.%u to %s:%d\n",
+						intel_crtc->pipe,
+						i == SKL_CRTC_INDEX ? scaler_state->scaler_id :
+							plane_state->scaler_id,
+						i == SKL_CRTC_INDEX ? "CRTC" : "PLANE",
+						i == SKL_CRTC_INDEX ?  intel_crtc->base.base.id :
+						plane->base.id);
+					break;
+				}
+			}
+		}
+
+		if (WARN_ON(*scaler_id < 0)) {
+			DRM_DEBUG_KMS("Cannot find scaler for %s:%d\n",
+				i == SKL_CRTC_INDEX ? "CRTC" : "PLANE",
+				i == SKL_CRTC_INDEX ? intel_crtc->base.base.id:plane->base.id);
+			continue;
+		}
+
+		/* set scaler mode */
+		if (num_scalers_need == 1 && intel_crtc->pipe != PIPE_C) {
+			/*
+			 * when only 1 scaler is in use on either pipe A or B,
+			 * scaler 0 operates in high quality (HQ) mode.
+			 * In this case use scaler 0 to take advantage of HQ mode
+			 */
+			*scaler_id = 0;
+			scaler_state->scalers[0].in_use = 1;
+			scaler_state->scalers[0].mode = PS_SCALER_MODE_HQ;
+			scaler_state->scalers[1].in_use = 0;
+		} else {
+			scaler_state->scalers[*scaler_id].mode = PS_SCALER_MODE_DYN;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c
index 976b89156570..86ba4b2c3a65 100644
--- a/drivers/gpu/drm/i915/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/intel_atomic_plane.c
@@ -85,8 +85,8 @@ intel_plane_duplicate_state(struct drm_plane *plane)
 		return NULL;
 
 	state = &intel_state->base;
-	if (state->fb)
-		drm_framebuffer_reference(state->fb);
+
+	__drm_atomic_helper_plane_duplicate_state(plane, state);
 
 	return state;
 }
@@ -111,6 +111,7 @@ static int intel_plane_atomic_check(struct drm_plane *plane,
 {
 	struct drm_crtc *crtc = state->crtc;
 	struct intel_crtc *intel_crtc;
+	struct intel_crtc_state *crtc_state;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
 	struct intel_plane_state *intel_state = to_intel_plane_state(state);
 
@@ -126,6 +127,17 @@ static int intel_plane_atomic_check(struct drm_plane *plane,
 	if (!crtc)
 		return 0;
 
+	/* FIXME: temporary hack necessary while we still use the plane update
+	 * helper. */
+	if (state->state) {
+		crtc_state =
+			intel_atomic_get_crtc_state(state->state, intel_crtc);
+		if (IS_ERR(crtc_state))
+			return PTR_ERR(crtc_state);
+	} else {
+		crtc_state = intel_crtc->config;
+	}
+
 	/*
 	 * The original src/dest coordinates are stored in state->base, but
 	 * we want to keep another copy internal to our driver that we can
@@ -144,9 +156,9 @@ static int intel_plane_atomic_check(struct drm_plane *plane,
 	intel_state->clip.x1 = 0;
 	intel_state->clip.y1 = 0;
 	intel_state->clip.x2 =
-		intel_crtc->active ? intel_crtc->config->pipe_src_w : 0;
+		crtc_state->base.active ? crtc_state->pipe_src_w : 0;
 	intel_state->clip.y2 =
-		intel_crtc->active ? intel_crtc->config->pipe_src_h : 0;
+		crtc_state->base.active ? crtc_state->pipe_src_h : 0;
 
 	/*
 	 * Disabling a plane is always okay; we just need to update
@@ -162,6 +174,30 @@ static int intel_plane_atomic_check(struct drm_plane *plane,
 			(1 << drm_plane_index(plane));
 	}
 
+	if (state->fb && intel_rotation_90_or_270(state->rotation)) {
+		if (!(state->fb->modifier[0] == I915_FORMAT_MOD_Y_TILED ||
+			state->fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED)) {
+			DRM_DEBUG_KMS("Y/Yf tiling required for 90/270!\n");
+			return -EINVAL;
+		}
+
+		/*
+		 * 90/270 is not allowed with RGB64 16:16:16:16,
+		 * RGB 16-bit 5:6:5, and Indexed 8-bit.
+		 * TBD: Add RGB64 case once its added in supported format list.
+		 */
+		switch (state->fb->pixel_format) {
+		case DRM_FORMAT_C8:
+		case DRM_FORMAT_RGB565:
+			DRM_DEBUG_KMS("Unsupported pixel format %s for 90/270!\n",
+					drm_get_format_name(state->fb->pixel_format));
+			return -EINVAL;
+
+		default:
+			break;
+		}
+	}
+
 	return intel_plane->check_plane(plane, intel_state);
 }
 
@@ -172,10 +208,6 @@ static void intel_plane_atomic_update(struct drm_plane *plane,
 	struct intel_plane_state *intel_state =
 		to_intel_plane_state(plane->state);
 
-	/* Don't disable an already disabled plane */
-	if (!plane->state->fb && !old_state->fb)
-		return;
-
 	intel_plane->commit_plane(plane, intel_state);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index ef342571ae6a..3da9b8409f20 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -28,7 +28,6 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_edid.h>
-#include "intel_drv.h"
 #include "i915_drv.h"
 
 /**
@@ -270,6 +269,9 @@ static void ilk_audio_codec_disable(struct intel_encoder *encoder)
 	DRM_DEBUG_KMS("Disable audio codec on port %c, pipe %c\n",
 		      port_name(port), pipe_name(pipe));
 
+	if (WARN_ON(port == PORT_A))
+		return;
+
 	if (HAS_PCH_IBX(dev_priv->dev)) {
 		aud_config = IBX_AUD_CFG(pipe);
 		aud_cntrl_st2 = IBX_AUD_CNTL_ST2;
@@ -291,12 +293,7 @@ static void ilk_audio_codec_disable(struct intel_encoder *encoder)
 		tmp |= AUD_CONFIG_N_VALUE_INDEX;
 	I915_WRITE(aud_config, tmp);
 
-	if (WARN_ON(!port)) {
-		eldv = IBX_ELD_VALID(PORT_B) | IBX_ELD_VALID(PORT_C) |
-			IBX_ELD_VALID(PORT_D);
-	} else {
-		eldv = IBX_ELD_VALID(port);
-	}
+	eldv = IBX_ELD_VALID(port);
 
 	/* Invalidate ELD */
 	tmp = I915_READ(aud_cntrl_st2);
@@ -326,6 +323,9 @@ static void ilk_audio_codec_enable(struct drm_connector *connector,
 	DRM_DEBUG_KMS("Enable audio codec on port %c, pipe %c, %u bytes ELD\n",
 		      port_name(port), pipe_name(pipe), drm_eld_size(eld));
 
+	if (WARN_ON(port == PORT_A))
+		return;
+
 	/*
 	 * FIXME: We're supposed to wait for vblank here, but we have vblanks
 	 * disabled during the mode set. The proper fix would be to push the
@@ -350,12 +350,7 @@ static void ilk_audio_codec_enable(struct drm_connector *connector,
 		aud_cntrl_st2 = CPT_AUD_CNTRL_ST2;
 	}
 
-	if (WARN_ON(!port)) {
-		eldv = IBX_ELD_VALID(PORT_B) | IBX_ELD_VALID(PORT_C) |
-			IBX_ELD_VALID(PORT_D);
-	} else {
-		eldv = IBX_ELD_VALID(port);
-	}
+	eldv = IBX_ELD_VALID(port);
 
 	/* Invalidate ELD */
 	tmp = I915_READ(aud_cntrl_st2);
@@ -511,7 +506,8 @@ static int i915_audio_component_get_cdclk_freq(struct device *dev)
 		return -ENODEV;
 
 	intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO);
-	ret = intel_ddi_get_cdclk_freq(dev_priv);
+	ret = dev_priv->display.get_display_clock_speed(dev_priv->dev);
+
 	intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO);
 
 	return ret;
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index c684085cb56a..198fc3c3291b 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -36,10 +36,11 @@
 
 static int panel_type;
 
-static void *
-find_section(struct bdb_header *bdb, int section_id)
+static const void *
+find_section(const void *_bdb, int section_id)
 {
-	u8 *base = (u8 *)bdb;
+	const struct bdb_header *bdb = _bdb;
+	const u8 *base = _bdb;
 	int index = 0;
 	u16 total, current_size;
 	u8 current_id;
@@ -53,7 +54,7 @@ find_section(struct bdb_header *bdb, int section_id)
 		current_id = *(base + index);
 		index++;
 
-		current_size = *((u16 *)(base + index));
+		current_size = *((const u16 *)(base + index));
 		index += 2;
 
 		if (index + current_size > total)
@@ -69,7 +70,7 @@ find_section(struct bdb_header *bdb, int section_id)
 }
 
 static u16
-get_blocksize(void *p)
+get_blocksize(const void *p)
 {
 	u16 *block_ptr, block_size;
 
@@ -204,7 +205,7 @@ get_lvds_fp_timing(const struct bdb_header *bdb,
 /* Try to find integrated panel data */
 static void
 parse_lfp_panel_data(struct drm_i915_private *dev_priv,
-			    struct bdb_header *bdb)
+		     const struct bdb_header *bdb)
 {
 	const struct bdb_lvds_options *lvds_options;
 	const struct bdb_lvds_lfp_data *lvds_lfp_data;
@@ -310,7 +311,8 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 }
 
 static void
-parse_lfp_backlight(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
+parse_lfp_backlight(struct drm_i915_private *dev_priv,
+		    const struct bdb_header *bdb)
 {
 	const struct bdb_lfp_backlight_data *backlight_data;
 	const struct bdb_lfp_backlight_data_entry *entry;
@@ -348,9 +350,9 @@ parse_lfp_backlight(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 /* Try to find sdvo panel data */
 static void
 parse_sdvo_panel_data(struct drm_i915_private *dev_priv,
-		      struct bdb_header *bdb)
+		      const struct bdb_header *bdb)
 {
-	struct lvds_dvo_timing *dvo_timing;
+	const struct lvds_dvo_timing *dvo_timing;
 	struct drm_display_mode *panel_fixed_mode;
 	int index;
 
@@ -361,7 +363,7 @@ parse_sdvo_panel_data(struct drm_i915_private *dev_priv,
 	}
 
 	if (index == -1) {
-		struct bdb_sdvo_lvds_options *sdvo_lvds_options;
+		const struct bdb_sdvo_lvds_options *sdvo_lvds_options;
 
 		sdvo_lvds_options = find_section(bdb, BDB_SDVO_LVDS_OPTIONS);
 		if (!sdvo_lvds_options)
@@ -402,10 +404,10 @@ static int intel_bios_ssc_frequency(struct drm_device *dev,
 
 static void
 parse_general_features(struct drm_i915_private *dev_priv,
-		       struct bdb_header *bdb)
+		       const struct bdb_header *bdb)
 {
 	struct drm_device *dev = dev_priv->dev;
-	struct bdb_general_features *general;
+	const struct bdb_general_features *general;
 
 	general = find_section(bdb, BDB_GENERAL_FEATURES);
 	if (general) {
@@ -428,9 +430,9 @@ parse_general_features(struct drm_i915_private *dev_priv,
 
 static void
 parse_general_definitions(struct drm_i915_private *dev_priv,
-			  struct bdb_header *bdb)
+			  const struct bdb_header *bdb)
 {
-	struct bdb_general_definitions *general;
+	const struct bdb_general_definitions *general;
 
 	general = find_section(bdb, BDB_GENERAL_DEFINITIONS);
 	if (general) {
@@ -438,7 +440,7 @@ parse_general_definitions(struct drm_i915_private *dev_priv,
 		if (block_size >= sizeof(*general)) {
 			int bus_pin = general->crt_ddc_gmbus_pin;
 			DRM_DEBUG_KMS("crt_ddc_bus_pin: %d\n", bus_pin);
-			if (intel_gmbus_is_port_valid(bus_pin))
+			if (intel_gmbus_is_valid_pin(dev_priv, bus_pin))
 				dev_priv->vbt.crt_ddc_pin = bus_pin;
 		} else {
 			DRM_DEBUG_KMS("BDB_GD too small (%d). Invalid.\n",
@@ -447,13 +449,19 @@ parse_general_definitions(struct drm_i915_private *dev_priv,
 	}
 }
 
+static const union child_device_config *
+child_device_ptr(const struct bdb_general_definitions *p_defs, int i)
+{
+	return (const void *) &p_defs->devices[i * p_defs->child_dev_size];
+}
+
 static void
 parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
-			  struct bdb_header *bdb)
+			  const struct bdb_header *bdb)
 {
 	struct sdvo_device_mapping *p_mapping;
-	struct bdb_general_definitions *p_defs;
-	union child_device_config *p_child;
+	const struct bdb_general_definitions *p_defs;
+	const union child_device_config *p_child;
 	int i, child_device_num, count;
 	u16	block_size;
 
@@ -476,10 +484,10 @@ parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
 	block_size = get_blocksize(p_defs);
 	/* get the number of child device */
 	child_device_num = (block_size - sizeof(*p_defs)) /
-				sizeof(*p_child);
+		p_defs->child_dev_size;
 	count = 0;
 	for (i = 0; i < child_device_num; i++) {
-		p_child = &(p_defs->devices[i]);
+		p_child = child_device_ptr(p_defs, i);
 		if (!p_child->old.device_type) {
 			/* skip the device block if device type is invalid */
 			continue;
@@ -539,9 +547,9 @@ parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
 
 static void
 parse_driver_features(struct drm_i915_private *dev_priv,
-		       struct bdb_header *bdb)
+		      const struct bdb_header *bdb)
 {
-	struct bdb_driver_features *driver;
+	const struct bdb_driver_features *driver;
 
 	driver = find_section(bdb, BDB_DRIVER_FEATURES);
 	if (!driver)
@@ -565,11 +573,11 @@ parse_driver_features(struct drm_i915_private *dev_priv,
 }
 
 static void
-parse_edp(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
+parse_edp(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 {
-	struct bdb_edp *edp;
-	struct edp_power_seq *edp_pps;
-	struct edp_link_params *edp_link_params;
+	const struct bdb_edp *edp;
+	const struct edp_power_seq *edp_pps;
+	const struct edp_link_params *edp_link_params;
 
 	edp = find_section(bdb, BDB_EDP);
 	if (!edp) {
@@ -666,16 +674,21 @@ parse_edp(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 	if (bdb->version >= 173) {
 		uint8_t vswing;
 
-		vswing = (edp->edp_vswing_preemph >> (panel_type * 4)) & 0xF;
-		dev_priv->vbt.edp_low_vswing = vswing == 0;
+		/* Don't read from VBT if module parameter has valid value*/
+		if (i915.edp_vswing) {
+			dev_priv->edp_low_vswing = i915.edp_vswing == 1;
+		} else {
+			vswing = (edp->edp_vswing_preemph >> (panel_type * 4)) & 0xF;
+			dev_priv->edp_low_vswing = vswing == 0;
+		}
 	}
 }
 
 static void
-parse_psr(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
+parse_psr(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 {
-	struct bdb_psr *psr;
-	struct psr_table *psr_table;
+	const struct bdb_psr *psr;
+	const struct psr_table *psr_table;
 
 	psr = find_section(bdb, BDB_PSR);
 	if (!psr) {
@@ -783,13 +796,14 @@ static u8 *goto_next_sequence(u8 *data, int *size)
 }
 
 static void
-parse_mipi(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
+parse_mipi(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 {
-	struct bdb_mipi_config *start;
-	struct bdb_mipi_sequence *sequence;
-	struct mipi_config *config;
-	struct mipi_pps_data *pps;
-	u8 *data, *seq_data;
+	const struct bdb_mipi_config *start;
+	const struct bdb_mipi_sequence *sequence;
+	const struct mipi_config *config;
+	const struct mipi_pps_data *pps;
+	u8 *data;
+	const u8 *seq_data;
 	int i, panel_id, seq_size;
 	u16 block_size;
 
@@ -933,7 +947,7 @@ err:
 }
 
 static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
-			   struct bdb_header *bdb)
+			   const struct bdb_header *bdb)
 {
 	union child_device_config *it, *child = NULL;
 	struct ddi_vbt_port_info *info = &dev_priv->vbt.ddi_port_info[port];
@@ -1035,7 +1049,7 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
 }
 
 static void parse_ddi_ports(struct drm_i915_private *dev_priv,
-			    struct bdb_header *bdb)
+			    const struct bdb_header *bdb)
 {
 	struct drm_device *dev = dev_priv->dev;
 	enum port port;
@@ -1055,10 +1069,11 @@ static void parse_ddi_ports(struct drm_i915_private *dev_priv,
 
 static void
 parse_device_mapping(struct drm_i915_private *dev_priv,
-		       struct bdb_header *bdb)
+		     const struct bdb_header *bdb)
 {
-	struct bdb_general_definitions *p_defs;
-	union child_device_config *p_child, *child_dev_ptr;
+	const struct bdb_general_definitions *p_defs;
+	const union child_device_config *p_child;
+	union child_device_config *child_dev_ptr;
 	int i, child_device_num, count;
 	u16	block_size;
 
@@ -1067,25 +1082,19 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 		DRM_DEBUG_KMS("No general definition block is found, no devices defined.\n");
 		return;
 	}
-	/* judge whether the size of child device meets the requirements.
-	 * If the child device size obtained from general definition block
-	 * is different with sizeof(struct child_device_config), skip the
-	 * parsing of sdvo device info
-	 */
-	if (p_defs->child_dev_size != sizeof(*p_child)) {
-		/* different child dev size . Ignore it */
-		DRM_DEBUG_KMS("different child size is found. Invalid.\n");
+	if (p_defs->child_dev_size < sizeof(*p_child)) {
+		DRM_ERROR("General definiton block child device size is too small.\n");
 		return;
 	}
 	/* get the block size of general definitions */
 	block_size = get_blocksize(p_defs);
 	/* get the number of child device */
 	child_device_num = (block_size - sizeof(*p_defs)) /
-				sizeof(*p_child);
+				p_defs->child_dev_size;
 	count = 0;
 	/* get the number of child device that is present */
 	for (i = 0; i < child_device_num; i++) {
-		p_child = &(p_defs->devices[i]);
+		p_child = child_device_ptr(p_defs, i);
 		if (!p_child->common.device_type) {
 			/* skip the device block if device type is invalid */
 			continue;
@@ -1105,7 +1114,7 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 	dev_priv->vbt.child_dev_num = count;
 	count = 0;
 	for (i = 0; i < child_device_num; i++) {
-		p_child = &(p_defs->devices[i]);
+		p_child = child_device_ptr(p_defs, i);
 		if (!p_child->common.device_type) {
 			/* skip the device block if device type is invalid */
 			continue;
@@ -1121,8 +1130,7 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 
 		child_dev_ptr = dev_priv->vbt.child_dev + count;
 		count++;
-		memcpy((void *)child_dev_ptr, (void *)p_child,
-					sizeof(*p_child));
+		memcpy(child_dev_ptr, p_child, sizeof(*p_child));
 	}
 	return;
 }
@@ -1133,7 +1141,7 @@ init_vbt_defaults(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	enum port port;
 
-	dev_priv->vbt.crt_ddc_pin = GMBUS_PORT_VGADDC;
+	dev_priv->vbt.crt_ddc_pin = GMBUS_PIN_VGADDC;
 
 	/* Default to having backlight */
 	dev_priv->vbt.backlight.present = true;
@@ -1191,19 +1199,22 @@ static const struct dmi_system_id intel_no_opregion_vbt[] = {
 	{ }
 };
 
-static struct bdb_header *validate_vbt(char *base, size_t size,
-				       struct vbt_header *vbt,
-				       const char *source)
+static const struct bdb_header *validate_vbt(const void __iomem *_base,
+					     size_t size,
+					     const void __iomem *_vbt,
+					     const char *source)
 {
-	size_t offset;
-	struct bdb_header *bdb;
-
-	if (vbt == NULL) {
-		DRM_DEBUG_DRIVER("VBT signature missing\n");
-		return NULL;
-	}
+	/*
+	 * This is the one place where we explicitly discard the address space
+	 * (__iomem) of the BIOS/VBT. (And this will cause a sparse complaint.)
+	 * From now on everything is based on 'base', and treated as regular
+	 * memory.
+	 */
+	const void *base = (const void *) _base;
+	size_t offset = _vbt - _base;
+	const struct vbt_header *vbt = base + offset;
+	const struct bdb_header *bdb;
 
-	offset = (char *)vbt - base;
 	if (offset + sizeof(struct vbt_header) > size) {
 		DRM_DEBUG_DRIVER("VBT header incomplete\n");
 		return NULL;
@@ -1220,7 +1231,7 @@ static struct bdb_header *validate_vbt(char *base, size_t size,
 		return NULL;
 	}
 
-	bdb = (struct bdb_header *)(base + offset);
+	bdb = base + offset;
 	if (offset + bdb->bdb_size > size) {
 		DRM_DEBUG_DRIVER("BDB incomplete\n");
 		return NULL;
@@ -1231,6 +1242,22 @@ static struct bdb_header *validate_vbt(char *base, size_t size,
 	return bdb;
 }
 
+static const struct bdb_header *find_vbt(void __iomem *bios, size_t size)
+{
+	const struct bdb_header *bdb = NULL;
+	size_t i;
+
+	/* Scour memory looking for the VBT signature. */
+	for (i = 0; i + 4 < size; i++) {
+		if (ioread32(bios + i) == *((const u32 *) "$VBT")) {
+			bdb = validate_vbt(bios, size, bios + i, "PCI ROM");
+			break;
+		}
+	}
+
+	return bdb;
+}
+
 /**
  * intel_parse_bios - find VBT and initialize settings from the BIOS
  * @dev: DRM device
@@ -1245,7 +1272,7 @@ intel_parse_bios(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct pci_dev *pdev = dev->pdev;
-	struct bdb_header *bdb = NULL;
+	const struct bdb_header *bdb = NULL;
 	u8 __iomem *bios = NULL;
 
 	if (HAS_PCH_NOP(dev))
@@ -1255,27 +1282,17 @@ intel_parse_bios(struct drm_device *dev)
 
 	/* XXX Should this validation be moved to intel_opregion.c? */
 	if (!dmi_check_system(intel_no_opregion_vbt) && dev_priv->opregion.vbt)
-		bdb = validate_vbt((char *)dev_priv->opregion.header, OPREGION_SIZE,
-				   (struct vbt_header *)dev_priv->opregion.vbt,
-				   "OpRegion");
+		bdb = validate_vbt(dev_priv->opregion.header, OPREGION_SIZE,
+				   dev_priv->opregion.vbt, "OpRegion");
 
 	if (bdb == NULL) {
-		size_t i, size;
+		size_t size;
 
 		bios = pci_map_rom(pdev, &size);
 		if (!bios)
 			return -1;
 
-		/* Scour memory looking for the VBT signature */
-		for (i = 0; i + 4 < size; i++) {
-			if (memcmp(bios + i, "$VBT", 4) == 0) {
-				bdb = validate_vbt(bios, size,
-						   (struct vbt_header *)(bios + i),
-						   "PCI ROM");
-				break;
-			}
-		}
-
+		bdb = find_vbt(bios, size);
 		if (!bdb) {
 			pci_unmap_rom(pdev, bios);
 			return -1;
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index 6afd5be33367..af0b47652752 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -277,9 +277,9 @@ struct bdb_general_definitions {
 	 * And the device num is related with the size of general definition
 	 * block. It is obtained by using the following formula:
 	 * number = (block_size - sizeof(bdb_general_definitions))/
-	 *	     sizeof(child_device_config);
+	 *	     defs->child_dev_size;
 	 */
-	union child_device_config devices[0];
+	uint8_t devices[0];
 } __packed;
 
 /* Mask for DRRS / Panel Channel / SSC / BLT control bits extraction */
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 515d7123785d..521af2c069cb 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -207,6 +207,14 @@ static void intel_disable_crt(struct intel_encoder *encoder)
 	intel_crt_set_dpms(encoder, DRM_MODE_DPMS_OFF);
 }
 
+static void pch_disable_crt(struct intel_encoder *encoder)
+{
+}
+
+static void pch_post_disable_crt(struct intel_encoder *encoder)
+{
+	intel_disable_crt(encoder);
+}
 
 static void hsw_crt_post_disable(struct intel_encoder *encoder)
 {
@@ -747,7 +755,7 @@ static int intel_crt_get_modes(struct drm_connector *connector)
 		goto out;
 
 	/* Try to probe digital port for output in DVI-I -> VGA mode. */
-	i2c = intel_gmbus_get_adapter(dev_priv, GMBUS_PORT_DPB);
+	i2c = intel_gmbus_get_adapter(dev_priv, GMBUS_PIN_DPB);
 	ret = intel_crt_ddc_get_modes(connector, i2c);
 
 out:
@@ -888,7 +896,12 @@ void intel_crt_init(struct drm_device *dev)
 		crt->adpa_reg = ADPA;
 
 	crt->base.compute_config = intel_crt_compute_config;
-	crt->base.disable = intel_disable_crt;
+	if (HAS_PCH_SPLIT(dev) && !HAS_DDI(dev)) {
+		crt->base.disable = pch_disable_crt;
+		crt->base.post_disable = pch_post_disable_crt;
+	} else {
+		crt->base.disable = intel_disable_crt;
+	}
 	crt->base.enable = intel_enable_crt;
 	if (I915_HAS_HOTPLUG(dev))
 		crt->base.hpd_pin = HPD_CRT;
diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
new file mode 100644
index 000000000000..bcb41e61877d
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -0,0 +1,467 @@
+/*
+ * Copyright © 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"),
+ * 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.
+ *
+ */
+#include <linux/firmware.h>
+#include "i915_drv.h"
+#include "i915_reg.h"
+
+/**
+ * DOC: csr support for dmc
+ *
+ * Display Context Save and Restore (CSR) firmware support added from gen9
+ * onwards to drive newly added DMC (Display microcontroller) in display
+ * engine to save and restore the state of display engine when it enter into
+ * low-power state and comes back to normal.
+ *
+ * Firmware loading status will be one of the below states: FW_UNINITIALIZED,
+ * FW_LOADED, FW_FAILED.
+ *
+ * Once the firmware is written into the registers status will be moved from
+ * FW_UNINITIALIZED to FW_LOADED and for any erroneous condition status will
+ * be moved to FW_FAILED.
+ */
+
+#define I915_CSR_SKL "i915/skl_dmc_ver1.bin"
+
+MODULE_FIRMWARE(I915_CSR_SKL);
+
+/*
+* SKL CSR registers for DC5 and DC6
+*/
+#define CSR_PROGRAM_BASE		0x80000
+#define CSR_SSP_BASE_ADDR_GEN9		0x00002FC0
+#define CSR_HTP_ADDR_SKL		0x00500034
+#define CSR_SSP_BASE			0x8F074
+#define CSR_HTP_SKL			0x8F004
+#define CSR_LAST_WRITE			0x8F034
+#define CSR_LAST_WRITE_VALUE		0xc003b400
+/* MMIO address range for CSR program (0x80000 - 0x82FFF) */
+#define CSR_MAX_FW_SIZE			0x2FFF
+#define CSR_DEFAULT_FW_OFFSET		0xFFFFFFFF
+#define CSR_MMIO_START_RANGE	0x80000
+#define CSR_MMIO_END_RANGE		0x8FFFF
+
+struct intel_css_header {
+	/* 0x09 for DMC */
+	uint32_t module_type;
+
+	/* Includes the DMC specific header in dwords */
+	uint32_t header_len;
+
+	/* always value would be 0x10000 */
+	uint32_t header_ver;
+
+	/* Not used */
+	uint32_t module_id;
+
+	/* Not used */
+	uint32_t module_vendor;
+
+	/* in YYYYMMDD format */
+	uint32_t date;
+
+	/* Size in dwords (CSS_Headerlen + PackageHeaderLen + dmc FWsLen)/4 */
+	uint32_t size;
+
+	/* Not used */
+	uint32_t key_size;
+
+	/* Not used */
+	uint32_t modulus_size;
+
+	/* Not used */
+	uint32_t exponent_size;
+
+	/* Not used */
+	uint32_t reserved1[12];
+
+	/* Major Minor */
+	uint32_t version;
+
+	/* Not used */
+	uint32_t reserved2[8];
+
+	/* Not used */
+	uint32_t kernel_header_info;
+} __packed;
+
+struct intel_fw_info {
+	uint16_t reserved1;
+
+	/* Stepping (A, B, C, ..., *). * is a wildcard */
+	char stepping;
+
+	/* Sub-stepping (0, 1, ..., *). * is a wildcard */
+	char substepping;
+
+	uint32_t offset;
+	uint32_t reserved2;
+} __packed;
+
+struct intel_package_header {
+	/* DMC container header length in dwords */
+	unsigned char header_len;
+
+	/* always value would be 0x01 */
+	unsigned char header_ver;
+
+	unsigned char reserved[10];
+
+	/* Number of valid entries in the FWInfo array below */
+	uint32_t num_entries;
+
+	struct intel_fw_info fw_info[20];
+} __packed;
+
+struct intel_dmc_header {
+	/* always value would be 0x40403E3E */
+	uint32_t signature;
+
+	/* DMC binary header length */
+	unsigned char header_len;
+
+	/* 0x01 */
+	unsigned char header_ver;
+
+	/* Reserved */
+	uint16_t dmcc_ver;
+
+	/* Major, Minor */
+	uint32_t	project;
+
+	/* Firmware program size (excluding header) in dwords */
+	uint32_t	fw_size;
+
+	/* Major Minor version */
+	uint32_t fw_version;
+
+	/* Number of valid MMIO cycles present. */
+	uint32_t mmio_count;
+
+	/* MMIO address */
+	uint32_t mmioaddr[8];
+
+	/* MMIO data */
+	uint32_t mmiodata[8];
+
+	/* FW filename  */
+	unsigned char dfile[32];
+
+	uint32_t reserved1[2];
+} __packed;
+
+struct stepping_info {
+	char stepping;
+	char substepping;
+};
+
+static const struct stepping_info skl_stepping_info[] = {
+		{'A', '0'}, {'B', '0'}, {'C', '0'},
+		{'D', '0'}, {'E', '0'}, {'F', '0'},
+		{'G', '0'}, {'H', '0'}, {'I', '0'}
+};
+
+static char intel_get_stepping(struct drm_device *dev)
+{
+	if (IS_SKYLAKE(dev) && (dev->pdev->revision <
+			ARRAY_SIZE(skl_stepping_info)))
+		return skl_stepping_info[dev->pdev->revision].stepping;
+	else
+		return -ENODATA;
+}
+
+static char intel_get_substepping(struct drm_device *dev)
+{
+	if (IS_SKYLAKE(dev) && (dev->pdev->revision <
+			ARRAY_SIZE(skl_stepping_info)))
+		return skl_stepping_info[dev->pdev->revision].substepping;
+	else
+		return -ENODATA;
+}
+
+/**
+ * intel_csr_load_status_get() - to get firmware loading status.
+ * @dev_priv: i915 device.
+ *
+ * This function helps to get the firmware loading status.
+ *
+ * Return: Firmware loading status.
+ */
+enum csr_state intel_csr_load_status_get(struct drm_i915_private *dev_priv)
+{
+	enum csr_state state;
+
+	mutex_lock(&dev_priv->csr_lock);
+	state = dev_priv->csr.state;
+	mutex_unlock(&dev_priv->csr_lock);
+
+	return state;
+}
+
+/**
+ * intel_csr_load_status_set() - help to set firmware loading status.
+ * @dev_priv: i915 device.
+ * @state: enumeration of firmware loading status.
+ *
+ * Set the firmware loading status.
+ */
+void intel_csr_load_status_set(struct drm_i915_private *dev_priv,
+			enum csr_state state)
+{
+	mutex_lock(&dev_priv->csr_lock);
+	dev_priv->csr.state = state;
+	mutex_unlock(&dev_priv->csr_lock);
+}
+
+/**
+ * intel_csr_load_program() - write the firmware from memory to register.
+ * @dev: drm device.
+ *
+ * CSR firmware is read from a .bin file and kept in internal memory one time.
+ * Everytime display comes back from low power state this function is called to
+ * copy the firmware from internal memory to registers.
+ */
+void intel_csr_load_program(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	__be32 *payload = dev_priv->csr.dmc_payload;
+	uint32_t i, fw_size;
+
+	if (!IS_GEN9(dev)) {
+		DRM_ERROR("No CSR support available for this platform\n");
+		return;
+	}
+
+	mutex_lock(&dev_priv->csr_lock);
+	fw_size = dev_priv->csr.dmc_fw_size;
+	for (i = 0; i < fw_size; i++)
+		I915_WRITE(CSR_PROGRAM_BASE + i * 4,
+			(u32 __force)payload[i]);
+
+	for (i = 0; i < dev_priv->csr.mmio_count; i++) {
+		I915_WRITE(dev_priv->csr.mmioaddr[i],
+			dev_priv->csr.mmiodata[i]);
+	}
+
+	dev_priv->csr.state = FW_LOADED;
+	mutex_unlock(&dev_priv->csr_lock);
+}
+
+static void finish_csr_load(const struct firmware *fw, void *context)
+{
+	struct drm_i915_private *dev_priv = context;
+	struct drm_device *dev = dev_priv->dev;
+	struct intel_css_header *css_header;
+	struct intel_package_header *package_header;
+	struct intel_dmc_header *dmc_header;
+	struct intel_csr *csr = &dev_priv->csr;
+	char stepping = intel_get_stepping(dev);
+	char substepping = intel_get_substepping(dev);
+	uint32_t dmc_offset = CSR_DEFAULT_FW_OFFSET, readcount = 0, nbytes;
+	uint32_t i;
+	__be32 *dmc_payload;
+	bool fw_loaded = false;
+
+	if (!fw) {
+		i915_firmware_load_error_print(csr->fw_path, 0);
+		goto out;
+	}
+
+	if ((stepping == -ENODATA) || (substepping == -ENODATA)) {
+		DRM_ERROR("Unknown stepping info, firmware loading failed\n");
+		goto out;
+	}
+
+	/* Extract CSS Header information*/
+	css_header = (struct intel_css_header *)fw->data;
+	if (sizeof(struct intel_css_header) !=
+		(css_header->header_len * 4)) {
+		DRM_ERROR("Firmware has wrong CSS header length %u bytes\n",
+			(css_header->header_len * 4));
+		goto out;
+	}
+	readcount += sizeof(struct intel_css_header);
+
+	/* Extract Package Header information*/
+	package_header = (struct intel_package_header *)
+					&fw->data[readcount];
+	if (sizeof(struct intel_package_header) !=
+		(package_header->header_len * 4)) {
+		DRM_ERROR("Firmware has wrong package header length %u bytes\n",
+			(package_header->header_len * 4));
+		goto out;
+	}
+	readcount += sizeof(struct intel_package_header);
+
+	/* Search for dmc_offset to find firware binary. */
+	for (i = 0; i < package_header->num_entries; i++) {
+		if (package_header->fw_info[i].substepping == '*' &&
+			stepping == package_header->fw_info[i].stepping) {
+			dmc_offset = package_header->fw_info[i].offset;
+			break;
+		} else if (stepping == package_header->fw_info[i].stepping &&
+			substepping == package_header->fw_info[i].substepping) {
+			dmc_offset = package_header->fw_info[i].offset;
+			break;
+		} else if (package_header->fw_info[i].stepping == '*' &&
+			package_header->fw_info[i].substepping == '*')
+			dmc_offset = package_header->fw_info[i].offset;
+	}
+	if (dmc_offset == CSR_DEFAULT_FW_OFFSET) {
+		DRM_ERROR("Firmware not supported for %c stepping\n", stepping);
+		goto out;
+	}
+	readcount += dmc_offset;
+
+	/* Extract dmc_header information. */
+	dmc_header = (struct intel_dmc_header *)&fw->data[readcount];
+	if (sizeof(struct intel_dmc_header) != (dmc_header->header_len)) {
+		DRM_ERROR("Firmware has wrong dmc header length %u bytes\n",
+				(dmc_header->header_len));
+		goto out;
+	}
+	readcount += sizeof(struct intel_dmc_header);
+
+	/* Cache the dmc header info. */
+	if (dmc_header->mmio_count > ARRAY_SIZE(csr->mmioaddr)) {
+		DRM_ERROR("Firmware has wrong mmio count %u\n",
+						dmc_header->mmio_count);
+		goto out;
+	}
+	csr->mmio_count = dmc_header->mmio_count;
+	for (i = 0; i < dmc_header->mmio_count; i++) {
+		if (dmc_header->mmioaddr[i] < CSR_MMIO_START_RANGE &&
+			dmc_header->mmioaddr[i] > CSR_MMIO_END_RANGE) {
+			DRM_ERROR(" Firmware has wrong mmio address 0x%x\n",
+						dmc_header->mmioaddr[i]);
+			goto out;
+		}
+		csr->mmioaddr[i] = dmc_header->mmioaddr[i];
+		csr->mmiodata[i] = dmc_header->mmiodata[i];
+	}
+
+	/* fw_size is in dwords, so multiplied by 4 to convert into bytes. */
+	nbytes = dmc_header->fw_size * 4;
+	if (nbytes > CSR_MAX_FW_SIZE) {
+		DRM_ERROR("CSR firmware too big (%u) bytes\n", nbytes);
+		goto out;
+	}
+	csr->dmc_fw_size = dmc_header->fw_size;
+
+	csr->dmc_payload = kmalloc(nbytes, GFP_KERNEL);
+	if (!csr->dmc_payload) {
+		DRM_ERROR("Memory allocation failed for dmc payload\n");
+		goto out;
+	}
+
+	dmc_payload = csr->dmc_payload;
+	for (i = 0; i < dmc_header->fw_size; i++) {
+		uint32_t *tmp = (u32 *)&fw->data[readcount + i * 4];
+		/*
+		 * The firmware payload is an array of 32 bit words stored in
+		 * little-endian format in the firmware image and programmed
+		 * as 32 bit big-endian format to memory.
+		 */
+		dmc_payload[i] = cpu_to_be32(*tmp);
+	}
+
+	/* load csr program during system boot, as needed for DC states */
+	intel_csr_load_program(dev);
+	fw_loaded = true;
+
+out:
+	if (fw_loaded)
+		intel_runtime_pm_put(dev_priv);
+	else
+		intel_csr_load_status_set(dev_priv, FW_FAILED);
+
+	release_firmware(fw);
+}
+
+/**
+ * intel_csr_ucode_init() - initialize the firmware loading.
+ * @dev: drm device.
+ *
+ * This function is called at the time of loading the display driver to read
+ * firmware from a .bin file and copied into a internal memory.
+ */
+void intel_csr_ucode_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_csr *csr = &dev_priv->csr;
+	int ret;
+
+	if (!HAS_CSR(dev))
+		return;
+
+	if (IS_SKYLAKE(dev))
+		csr->fw_path = I915_CSR_SKL;
+	else {
+		DRM_ERROR("Unexpected: no known CSR firmware for platform\n");
+		intel_csr_load_status_set(dev_priv, FW_FAILED);
+		return;
+	}
+
+	/*
+	 * Obtain a runtime pm reference, until CSR is loaded,
+	 * to avoid entering runtime-suspend.
+	 */
+	intel_runtime_pm_get(dev_priv);
+
+	/* CSR supported for platform, load firmware */
+	ret = request_firmware_nowait(THIS_MODULE, true, csr->fw_path,
+				&dev_priv->dev->pdev->dev,
+				GFP_KERNEL, dev_priv,
+				finish_csr_load);
+	if (ret) {
+		i915_firmware_load_error_print(csr->fw_path, ret);
+		intel_csr_load_status_set(dev_priv, FW_FAILED);
+	}
+}
+
+/**
+ * intel_csr_ucode_fini() - unload the CSR firmware.
+ * @dev: drm device.
+ *
+ * Firmmware unloading includes freeing the internal momory and reset the
+ * firmware loading status.
+ */
+void intel_csr_ucode_fini(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!HAS_CSR(dev))
+		return;
+
+	intel_csr_load_status_set(dev_priv, FW_FAILED);
+	kfree(dev_priv->csr.dmc_payload);
+}
+
+void assert_csr_loaded(struct drm_i915_private *dev_priv)
+{
+	WARN((intel_csr_load_status_get(dev_priv) != FW_LOADED), "CSR is not loaded.\n");
+	WARN(!I915_READ(CSR_PROGRAM_BASE),
+				"CSR program storage start is NULL\n");
+	WARN(!I915_READ(CSR_SSP_BASE), "CSR SSP Base Not fine\n");
+	WARN(!I915_READ(CSR_HTP_SKL), "CSR HTP Not fine\n");
+}
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 3eb0efc2dd0d..cacb07b7a8f1 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -155,33 +155,100 @@ static const struct ddi_buf_trans skl_ddi_translations_edp[] = {
 
 
 static const struct ddi_buf_trans skl_ddi_translations_hdmi[] = {
-					/* Idx	NT mV   T mV    db  */
-	{ 0x00004014, 0x00000087 },	/* 0:	800	1000	2   */
+	{ 0x00000018, 0x000000ac },
+	{ 0x00005012, 0x0000009d },
+	{ 0x00007011, 0x00000088 },
+	{ 0x00000018, 0x000000a1 },
+	{ 0x00000018, 0x00000098 },
+	{ 0x00004013, 0x00000088 },
+	{ 0x00006012, 0x00000087 },
+	{ 0x00000018, 0x000000df },
+	{ 0x00003015, 0x00000087 },
+	{ 0x00003015, 0x000000c7 },
+	{ 0x00000018, 0x000000c7 },
 };
 
-enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
+struct bxt_ddi_buf_trans {
+	u32 margin;	/* swing value */
+	u32 scale;	/* scale value */
+	u32 enable;	/* scale enable */
+	u32 deemphasis;
+	bool default_index; /* true if the entry represents default value */
+};
+
+/* BSpec does not define separate vswing/pre-emphasis values for eDP.
+ * Using DP values for eDP as well.
+ */
+static const struct bxt_ddi_buf_trans bxt_ddi_translations_dp[] = {
+					/* Idx	NT mV diff	db  */
+	{ 52,  0,    0, 128, true  },	/* 0:	400		0   */
+	{ 78,  0,    0, 85,  false },	/* 1:	400		3.5 */
+	{ 104, 0,    0, 64,  false },	/* 2:	400		6   */
+	{ 154, 0,    0, 43,  false },	/* 3:	400		9.5 */
+	{ 77,  0,    0, 128, false },	/* 4:	600		0   */
+	{ 116, 0,    0, 85,  false },	/* 5:	600		3.5 */
+	{ 154, 0,    0, 64,  false },	/* 6:	600		6   */
+	{ 102, 0,    0, 128, false },	/* 7:	800		0   */
+	{ 154, 0,    0, 85,  false },	/* 8:	800		3.5 */
+	{ 154, 0x9A, 1, 128, false },  /* 9:	1200		0   */
+};
+
+/* BSpec has 2 recommended values - entries 0 and 8.
+ * Using the entry with higher vswing.
+ */
+static const struct bxt_ddi_buf_trans bxt_ddi_translations_hdmi[] = {
+					/* Idx	NT mV diff	db  */
+	{ 52,  0,    0, 128, false },	/* 0:	400		0   */
+	{ 52,  0,    0, 85,  false },	/* 1:	400		3.5 */
+	{ 52,  0,    0, 64,  false },	/* 2:	400		6   */
+	{ 42,  0,    0, 43,  false },	/* 3:	400		9.5 */
+	{ 77,  0,    0, 128, false },	/* 4:	600		0   */
+	{ 77,  0,    0, 85,  false },	/* 5:	600		3.5 */
+	{ 77,  0,    0, 64,  false },	/* 6:	600		6   */
+	{ 102, 0,    0, 128, false },	/* 7:	800		0   */
+	{ 102, 0,    0, 85,  false },	/* 8:	800		3.5 */
+	{ 154, 0x9A, 1, 128, true },	/* 9:	1200		0   */
+};
+
+static void ddi_get_encoder_port(struct intel_encoder *intel_encoder,
+				 struct intel_digital_port **dig_port,
+				 enum port *port)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
 	int type = intel_encoder->type;
 
 	if (type == INTEL_OUTPUT_DP_MST) {
-		struct intel_digital_port *intel_dig_port = enc_to_mst(encoder)->primary;
-		return intel_dig_port->port;
+		*dig_port = enc_to_mst(encoder)->primary;
+		*port = (*dig_port)->port;
 	} else if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP ||
 	    type == INTEL_OUTPUT_HDMI || type == INTEL_OUTPUT_UNKNOWN) {
-		struct intel_digital_port *intel_dig_port =
-			enc_to_dig_port(encoder);
-		return intel_dig_port->port;
-
+		*dig_port = enc_to_dig_port(encoder);
+		*port = (*dig_port)->port;
 	} else if (type == INTEL_OUTPUT_ANALOG) {
-		return PORT_E;
-
+		*dig_port = NULL;
+		*port = PORT_E;
 	} else {
 		DRM_ERROR("Invalid DDI encoder type %d\n", type);
 		BUG();
 	}
 }
 
+enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
+{
+	struct intel_digital_port *dig_port;
+	enum port port;
+
+	ddi_get_encoder_port(intel_encoder, &dig_port, &port);
+
+	return port;
+}
+
+static bool
+intel_dig_port_supports_hdmi(const struct intel_digital_port *intel_dig_port)
+{
+	return intel_dig_port->hdmi.hdmi_reg;
+}
+
 /*
  * Starting with Haswell, DDI port buffers must be programmed with correct
  * values in advance. The buffer values are different for FDI and DP modes,
@@ -189,7 +256,8 @@ enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
  * in either FDI or DP modes only, as HDMI connections will work with both
  * of those
  */
-static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port)
+static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port,
+				      bool supports_hdmi)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 reg;
@@ -202,11 +270,19 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port)
 	const struct ddi_buf_trans *ddi_translations_hdmi;
 	const struct ddi_buf_trans *ddi_translations;
 
-	if (IS_SKYLAKE(dev)) {
+	if (IS_BROXTON(dev)) {
+		if (!supports_hdmi)
+			return;
+
+		/* Vswing programming for HDMI */
+		bxt_ddi_vswing_sequence(dev, hdmi_level, port,
+					INTEL_OUTPUT_HDMI);
+		return;
+	} else if (IS_SKYLAKE(dev)) {
 		ddi_translations_fdi = NULL;
 		ddi_translations_dp = skl_ddi_translations_dp;
 		n_dp_entries = ARRAY_SIZE(skl_ddi_translations_dp);
-		if (dev_priv->vbt.edp_low_vswing) {
+		if (dev_priv->edp_low_vswing) {
 			ddi_translations_edp = skl_ddi_translations_edp;
 			n_edp_entries = ARRAY_SIZE(skl_ddi_translations_edp);
 		} else {
@@ -214,16 +290,9 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port)
 			n_edp_entries = ARRAY_SIZE(skl_ddi_translations_dp);
 		}
 
-		/*
-		 * On SKL, the recommendation from the hw team is to always use
-		 * a certain type of level shifter (and thus the corresponding
-		 * 800mV+2dB entry). Given that's the only validated entry, we
-		 * override what is in the VBT, at least until further notice.
-		 */
-		hdmi_level = 0;
 		ddi_translations_hdmi = skl_ddi_translations_hdmi;
 		n_hdmi_entries = ARRAY_SIZE(skl_ddi_translations_hdmi);
-		hdmi_default_entry = 0;
+		hdmi_default_entry = 7;
 	} else if (IS_BROADWELL(dev)) {
 		ddi_translations_fdi = bdw_ddi_translations_fdi;
 		ddi_translations_dp = bdw_ddi_translations_dp;
@@ -290,6 +359,9 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port)
 		reg += 4;
 	}
 
+	if (!supports_hdmi)
+		return;
+
 	/* Choose a good default if VBT is badly populated */
 	if (hdmi_level == HDMI_LEVEL_SHIFT_UNKNOWN ||
 	    hdmi_level >= n_hdmi_entries)
@@ -307,13 +379,28 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port)
  */
 void intel_prepare_ddi(struct drm_device *dev)
 {
-	int port;
+	struct intel_encoder *intel_encoder;
+	bool visited[I915_MAX_PORTS] = { 0, };
 
 	if (!HAS_DDI(dev))
 		return;
 
-	for (port = PORT_A; port <= PORT_E; port++)
-		intel_prepare_ddi_buffers(dev, port);
+	for_each_intel_encoder(dev, intel_encoder) {
+		struct intel_digital_port *intel_dig_port;
+		enum port port;
+		bool supports_hdmi;
+
+		ddi_get_encoder_port(intel_encoder, &intel_dig_port, &port);
+
+		if (visited[port])
+			continue;
+
+		supports_hdmi = intel_dig_port &&
+				intel_dig_port_supports_hdmi(intel_dig_port);
+
+		intel_prepare_ddi_buffers(dev, port, supports_hdmi);
+		visited[port] = true;
+	}
 }
 
 static void intel_wait_ddi_buf_idle(struct drm_i915_private *dev_priv,
@@ -322,7 +409,7 @@ static void intel_wait_ddi_buf_idle(struct drm_i915_private *dev_priv,
 	uint32_t reg = DDI_BUF_CTL(port);
 	int i;
 
-	for (i = 0; i < 8; i++) {
+	for (i = 0; i < 16; i++) {
 		udelay(1);
 		if (I915_READ(reg) & DDI_BUF_IS_IDLE)
 			return;
@@ -491,23 +578,24 @@ intel_ddi_get_crtc_encoder(struct drm_crtc *crtc)
 	return ret;
 }
 
-static struct intel_encoder *
+struct intel_encoder *
 intel_ddi_get_crtc_new_encoder(struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct intel_encoder *ret = NULL;
 	struct drm_atomic_state *state;
+	struct drm_connector *connector;
+	struct drm_connector_state *connector_state;
 	int num_encoders = 0;
 	int i;
 
 	state = crtc_state->base.state;
 
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i] ||
-		    state->connector_states[i]->crtc != crtc_state->base.crtc)
+	for_each_connector_in_state(state, connector, connector_state, i) {
+		if (connector_state->crtc != crtc_state->base.crtc)
 			continue;
 
-		ret = to_intel_encoder(state->connector_states[i]->best_encoder);
+		ret = to_intel_encoder(connector_state->best_encoder);
 		num_encoders++;
 	}
 
@@ -783,26 +871,26 @@ static void skl_ddi_clock_get(struct intel_encoder *encoder,
 	if (dpll_ctl1 & DPLL_CTRL1_HDMI_MODE(dpll)) {
 		link_clock = skl_calc_wrpll_link(dev_priv, dpll);
 	} else {
-		link_clock = dpll_ctl1 & DPLL_CRTL1_LINK_RATE_MASK(dpll);
-		link_clock >>= DPLL_CRTL1_LINK_RATE_SHIFT(dpll);
+		link_clock = dpll_ctl1 & DPLL_CTRL1_LINK_RATE_MASK(dpll);
+		link_clock >>= DPLL_CTRL1_LINK_RATE_SHIFT(dpll);
 
 		switch (link_clock) {
-		case DPLL_CRTL1_LINK_RATE_810:
+		case DPLL_CTRL1_LINK_RATE_810:
 			link_clock = 81000;
 			break;
-		case DPLL_CRTL1_LINK_RATE_1080:
+		case DPLL_CTRL1_LINK_RATE_1080:
 			link_clock = 108000;
 			break;
-		case DPLL_CRTL1_LINK_RATE_1350:
+		case DPLL_CTRL1_LINK_RATE_1350:
 			link_clock = 135000;
 			break;
-		case DPLL_CRTL1_LINK_RATE_1620:
+		case DPLL_CTRL1_LINK_RATE_1620:
 			link_clock = 162000;
 			break;
-		case DPLL_CRTL1_LINK_RATE_2160:
+		case DPLL_CTRL1_LINK_RATE_2160:
 			link_clock = 216000;
 			break;
-		case DPLL_CRTL1_LINK_RATE_2700:
+		case DPLL_CTRL1_LINK_RATE_2700:
 			link_clock = 270000;
 			break;
 		default:
@@ -878,6 +966,32 @@ static void hsw_ddi_clock_get(struct intel_encoder *encoder,
 		pipe_config->base.adjusted_mode.crtc_clock = pipe_config->port_clock;
 }
 
+static int bxt_calc_pll_link(struct drm_i915_private *dev_priv,
+				enum intel_dpll_id dpll)
+{
+	/* FIXME formula not available in bspec */
+	return 0;
+}
+
+static void bxt_ddi_clock_get(struct intel_encoder *encoder,
+				struct intel_crtc_state *pipe_config)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	enum port port = intel_ddi_get_encoder_port(encoder);
+	uint32_t dpll = port;
+
+	pipe_config->port_clock =
+		bxt_calc_pll_link(dev_priv, dpll);
+
+	if (pipe_config->has_dp_encoder)
+		pipe_config->base.adjusted_mode.crtc_clock =
+			intel_dotclock_calculate(pipe_config->port_clock,
+							&pipe_config->dp_m_n);
+	else
+		pipe_config->base.adjusted_mode.crtc_clock =
+							pipe_config->port_clock;
+}
+
 void intel_ddi_clock_get(struct intel_encoder *encoder,
 			 struct intel_crtc_state *pipe_config)
 {
@@ -885,8 +999,10 @@ void intel_ddi_clock_get(struct intel_encoder *encoder,
 
 	if (INTEL_INFO(dev)->gen <= 8)
 		hsw_ddi_clock_get(encoder, pipe_config);
-	else
+	else if (IS_SKYLAKE(dev))
 		skl_ddi_clock_get(encoder, pipe_config);
+	else if (IS_BROXTON(dev))
+		bxt_ddi_clock_get(encoder, pipe_config);
 }
 
 static void
@@ -971,6 +1087,9 @@ hsw_ddi_pll_select(struct intel_crtc *intel_crtc,
 		      WRPLL_DIVIDER_REFERENCE(r2) | WRPLL_DIVIDER_FEEDBACK(n2) |
 		      WRPLL_DIVIDER_POST(p);
 
+		memset(&crtc_state->dpll_hw_state, 0,
+		       sizeof(crtc_state->dpll_hw_state));
+
 		crtc_state->dpll_hw_state.wrpll = val;
 
 		pll = intel_get_shared_dpll(intel_crtc, crtc_state);
@@ -1073,69 +1192,69 @@ found:
 	if (min_dco_index > 2) {
 		WARN(1, "No valid values found for the given pixel clock\n");
 	} else {
-		 wrpll_params->central_freq = dco_central_freq[min_dco_index];
+		wrpll_params->central_freq = dco_central_freq[min_dco_index];
 
-		 switch (dco_central_freq[min_dco_index]) {
-		 case 9600000000ULL:
+		switch (dco_central_freq[min_dco_index]) {
+		case 9600000000ULL:
 			wrpll_params->central_freq = 0;
 			break;
-		 case 9000000000ULL:
+		case 9000000000ULL:
 			wrpll_params->central_freq = 1;
 			break;
-		 case 8400000000ULL:
+		case 8400000000ULL:
 			wrpll_params->central_freq = 3;
-		 }
+		}
 
-		 switch (candidate_p0[min_dco_index]) {
-		 case 1:
+		switch (candidate_p0[min_dco_index]) {
+		case 1:
 			wrpll_params->pdiv = 0;
 			break;
-		 case 2:
+		case 2:
 			wrpll_params->pdiv = 1;
 			break;
-		 case 3:
+		case 3:
 			wrpll_params->pdiv = 2;
 			break;
-		 case 7:
+		case 7:
 			wrpll_params->pdiv = 4;
 			break;
-		 default:
+		default:
 			WARN(1, "Incorrect PDiv\n");
-		 }
+		}
 
-		 switch (candidate_p2[min_dco_index]) {
-		 case 5:
+		switch (candidate_p2[min_dco_index]) {
+		case 5:
 			wrpll_params->kdiv = 0;
 			break;
-		 case 2:
+		case 2:
 			wrpll_params->kdiv = 1;
 			break;
-		 case 3:
+		case 3:
 			wrpll_params->kdiv = 2;
 			break;
-		 case 1:
+		case 1:
 			wrpll_params->kdiv = 3;
 			break;
-		 default:
+		default:
 			WARN(1, "Incorrect KDiv\n");
-		 }
+		}
 
-		 wrpll_params->qdiv_ratio = candidate_p1[min_dco_index];
-		 wrpll_params->qdiv_mode =
+		wrpll_params->qdiv_ratio = candidate_p1[min_dco_index];
+		wrpll_params->qdiv_mode =
 			(wrpll_params->qdiv_ratio == 1) ? 0 : 1;
 
-		 dco_freq = candidate_p0[min_dco_index] *
-			 candidate_p1[min_dco_index] *
-			 candidate_p2[min_dco_index] * afe_clock;
+		dco_freq = candidate_p0[min_dco_index] *
+			candidate_p1[min_dco_index] *
+			candidate_p2[min_dco_index] * afe_clock;
 
 		/*
-		* Intermediate values are in Hz.
-		* Divide by MHz to match bsepc
-		*/
-		 wrpll_params->dco_integer = div_u64(dco_freq, (24 * MHz(1)));
-		 wrpll_params->dco_fraction =
-			 div_u64(((div_u64(dco_freq, 24) -
-				   wrpll_params->dco_integer * MHz(1)) * 0x8000), MHz(1));
+		 * Intermediate values are in Hz.
+		 * Divide by MHz to match bsepc
+		 */
+		wrpll_params->dco_integer = div_u64(dco_freq, (24 * MHz(1)));
+		wrpll_params->dco_fraction =
+			div_u64(((div_u64(dco_freq, 24) -
+				  wrpll_params->dco_integer * MHz(1)) * 0x8000), MHz(1));
 
 	}
 }
@@ -1179,13 +1298,13 @@ skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 
 		switch (intel_dp->link_bw) {
 		case DP_LINK_BW_1_62:
-			ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_810, 0);
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810, 0);
 			break;
 		case DP_LINK_BW_2_7:
-			ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_1350, 0);
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1350, 0);
 			break;
 		case DP_LINK_BW_5_4:
-			ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_2700, 0);
+			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2700, 0);
 			break;
 		}
 
@@ -1193,6 +1312,9 @@ skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 	} else /* eDP */
 		return true;
 
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
 	crtc_state->dpll_hw_state.ctrl1 = ctrl1;
 	crtc_state->dpll_hw_state.cfgcr1 = cfgcr1;
 	crtc_state->dpll_hw_state.cfgcr2 = cfgcr2;
@@ -1210,6 +1332,161 @@ skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 	return true;
 }
 
+/* bxt clock parameters */
+struct bxt_clk_div {
+	uint32_t p1;
+	uint32_t p2;
+	uint32_t m2_int;
+	uint32_t m2_frac;
+	bool m2_frac_en;
+	uint32_t n;
+};
+
+/* pre-calculated values for DP linkrates */
+static struct bxt_clk_div bxt_dp_clk_val[7] = {
+	/* 162 */ {4, 2, 32, 1677722, 1, 1},
+	/* 270 */ {4, 1, 27,       0, 0, 1},
+	/* 540 */ {2, 1, 27,       0, 0, 1},
+	/* 216 */ {3, 2, 32, 1677722, 1, 1},
+	/* 243 */ {4, 1, 24, 1258291, 1, 1},
+	/* 324 */ {4, 1, 32, 1677722, 1, 1},
+	/* 432 */ {3, 1, 32, 1677722, 1, 1}
+};
+
+static bool
+bxt_ddi_pll_select(struct intel_crtc *intel_crtc,
+		   struct intel_crtc_state *crtc_state,
+		   struct intel_encoder *intel_encoder,
+		   int clock)
+{
+	struct intel_shared_dpll *pll;
+	struct bxt_clk_div clk_div = {0};
+	int vco = 0;
+	uint32_t prop_coef, int_coef, gain_ctl, targ_cnt;
+	uint32_t dcoampovr_en_h, dco_amp, lanestagger;
+
+	if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
+		intel_clock_t best_clock;
+
+		/* Calculate HDMI div */
+		/*
+		 * FIXME: tie the following calculation into
+		 * i9xx_crtc_compute_clock
+		 */
+		if (!bxt_find_best_dpll(crtc_state, clock, &best_clock)) {
+			DRM_DEBUG_DRIVER("no PLL dividers found for clock %d pipe %c\n",
+					 clock, pipe_name(intel_crtc->pipe));
+			return false;
+		}
+
+		clk_div.p1 = best_clock.p1;
+		clk_div.p2 = best_clock.p2;
+		WARN_ON(best_clock.m1 != 2);
+		clk_div.n = best_clock.n;
+		clk_div.m2_int = best_clock.m2 >> 22;
+		clk_div.m2_frac = best_clock.m2 & ((1 << 22) - 1);
+		clk_div.m2_frac_en = clk_div.m2_frac != 0;
+
+		vco = best_clock.vco;
+	} else if (intel_encoder->type == INTEL_OUTPUT_DISPLAYPORT ||
+			intel_encoder->type == INTEL_OUTPUT_EDP) {
+		struct drm_encoder *encoder = &intel_encoder->base;
+		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+		switch (intel_dp->link_bw) {
+		case DP_LINK_BW_1_62:
+			clk_div = bxt_dp_clk_val[0];
+			break;
+		case DP_LINK_BW_2_7:
+			clk_div = bxt_dp_clk_val[1];
+			break;
+		case DP_LINK_BW_5_4:
+			clk_div = bxt_dp_clk_val[2];
+			break;
+		default:
+			clk_div = bxt_dp_clk_val[0];
+			DRM_ERROR("Unknown link rate\n");
+		}
+		vco = clock * 10 / 2 * clk_div.p1 * clk_div.p2;
+	}
+
+	dco_amp = 15;
+	dcoampovr_en_h = 0;
+	if (vco >= 6200000 && vco <= 6480000) {
+		prop_coef = 4;
+		int_coef = 9;
+		gain_ctl = 3;
+		targ_cnt = 8;
+	} else if ((vco > 5400000 && vco < 6200000) ||
+			(vco >= 4800000 && vco < 5400000)) {
+		prop_coef = 5;
+		int_coef = 11;
+		gain_ctl = 3;
+		targ_cnt = 9;
+		if (vco >= 4800000 && vco < 5400000)
+			dcoampovr_en_h = 1;
+	} else if (vco == 5400000) {
+		prop_coef = 3;
+		int_coef = 8;
+		gain_ctl = 1;
+		targ_cnt = 9;
+	} else {
+		DRM_ERROR("Invalid VCO\n");
+		return false;
+	}
+
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	if (clock > 270000)
+		lanestagger = 0x18;
+	else if (clock > 135000)
+		lanestagger = 0x0d;
+	else if (clock > 67000)
+		lanestagger = 0x07;
+	else if (clock > 33000)
+		lanestagger = 0x04;
+	else
+		lanestagger = 0x02;
+
+	crtc_state->dpll_hw_state.ebb0 =
+		PORT_PLL_P1(clk_div.p1) | PORT_PLL_P2(clk_div.p2);
+	crtc_state->dpll_hw_state.pll0 = clk_div.m2_int;
+	crtc_state->dpll_hw_state.pll1 = PORT_PLL_N(clk_div.n);
+	crtc_state->dpll_hw_state.pll2 = clk_div.m2_frac;
+
+	if (clk_div.m2_frac_en)
+		crtc_state->dpll_hw_state.pll3 =
+			PORT_PLL_M2_FRAC_ENABLE;
+
+	crtc_state->dpll_hw_state.pll6 =
+		prop_coef | PORT_PLL_INT_COEFF(int_coef);
+	crtc_state->dpll_hw_state.pll6 |=
+		PORT_PLL_GAIN_CTL(gain_ctl);
+
+	crtc_state->dpll_hw_state.pll8 = targ_cnt;
+
+	if (dcoampovr_en_h)
+		crtc_state->dpll_hw_state.pll10 = PORT_PLL_DCO_AMP_OVR_EN_H;
+
+	crtc_state->dpll_hw_state.pll10 |= PORT_PLL_DCO_AMP(dco_amp);
+
+	crtc_state->dpll_hw_state.pcsdw12 =
+		LANESTAGGER_STRAP_OVRD | lanestagger;
+
+	pll = intel_get_shared_dpll(intel_crtc, crtc_state);
+	if (pll == NULL) {
+		DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
+			pipe_name(intel_crtc->pipe));
+		return false;
+	}
+
+	/* shared DPLL id 0 is DPLL A */
+	crtc_state->ddi_pll_sel = pll->id;
+
+	return true;
+}
+
 /*
  * Tries to find a *shared* PLL for the CRTC and store it in
  * intel_crtc->ddi_pll_sel.
@@ -1228,6 +1505,9 @@ bool intel_ddi_pll_select(struct intel_crtc *intel_crtc,
 	if (IS_SKYLAKE(dev))
 		return skl_ddi_pll_select(intel_crtc, crtc_state,
 					  intel_encoder, clock);
+	else if (IS_BROXTON(dev))
+		return bxt_ddi_pll_select(intel_crtc, crtc_state,
+					  intel_encoder, clock);
 	else
 		return hsw_ddi_pll_select(intel_crtc, crtc_state,
 					  intel_encoder, clock);
@@ -1519,6 +1799,67 @@ void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc)
 			   TRANS_CLK_SEL_DISABLED);
 }
 
+void bxt_ddi_vswing_sequence(struct drm_device *dev, u32 level,
+			     enum port port, int type)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	const struct bxt_ddi_buf_trans *ddi_translations;
+	u32 n_entries, i;
+	uint32_t val;
+
+	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
+		n_entries = ARRAY_SIZE(bxt_ddi_translations_dp);
+		ddi_translations = bxt_ddi_translations_dp;
+	} else if (type == INTEL_OUTPUT_HDMI) {
+		n_entries = ARRAY_SIZE(bxt_ddi_translations_hdmi);
+		ddi_translations = bxt_ddi_translations_hdmi;
+	} else {
+		DRM_DEBUG_KMS("Vswing programming not done for encoder %d\n",
+				type);
+		return;
+	}
+
+	/* Check if default value has to be used */
+	if (level >= n_entries ||
+	    (type == INTEL_OUTPUT_HDMI && level == HDMI_LEVEL_SHIFT_UNKNOWN)) {
+		for (i = 0; i < n_entries; i++) {
+			if (ddi_translations[i].default_index) {
+				level = i;
+				break;
+			}
+		}
+	}
+
+	/*
+	 * While we write to the group register to program all lanes at once we
+	 * can read only lane registers and we pick lanes 0/1 for that.
+	 */
+	val = I915_READ(BXT_PORT_PCS_DW10_LN01(port));
+	val &= ~(TX2_SWING_CALC_INIT | TX1_SWING_CALC_INIT);
+	I915_WRITE(BXT_PORT_PCS_DW10_GRP(port), val);
+
+	val = I915_READ(BXT_PORT_TX_DW2_LN0(port));
+	val &= ~(MARGIN_000 | UNIQ_TRANS_SCALE);
+	val |= ddi_translations[level].margin << MARGIN_000_SHIFT |
+	       ddi_translations[level].scale << UNIQ_TRANS_SCALE_SHIFT;
+	I915_WRITE(BXT_PORT_TX_DW2_GRP(port), val);
+
+	val = I915_READ(BXT_PORT_TX_DW3_LN0(port));
+	val &= ~UNIQE_TRANGE_EN_METHOD;
+	if (ddi_translations[level].enable)
+		val |= UNIQE_TRANGE_EN_METHOD;
+	I915_WRITE(BXT_PORT_TX_DW3_GRP(port), val);
+
+	val = I915_READ(BXT_PORT_TX_DW4_LN0(port));
+	val &= ~DE_EMPHASIS;
+	val |= ddi_translations[level].deemphasis << DEEMPH_SHIFT;
+	I915_WRITE(BXT_PORT_TX_DW4_GRP(port), val);
+
+	val = I915_READ(BXT_PORT_PCS_DW10_LN01(port));
+	val |= TX2_SWING_CALC_INIT | TX1_SWING_CALC_INIT;
+	I915_WRITE(BXT_PORT_PCS_DW10_GRP(port), val);
+}
+
 static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
@@ -1527,6 +1868,7 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 	struct intel_crtc *crtc = to_intel_crtc(encoder->crtc);
 	enum port port = intel_ddi_get_encoder_port(intel_encoder);
 	int type = intel_encoder->type;
+	int hdmi_level;
 
 	if (type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
@@ -1548,7 +1890,7 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 
 			val &= ~(DPLL_CTRL1_HDMI_MODE(dpll) |
 				 DPLL_CTRL1_SSC(dpll) |
-				 DPLL_CRTL1_LINK_RATE_MASK(dpll));
+				 DPLL_CTRL1_LINK_RATE_MASK(dpll));
 			val |= crtc->config->dpll_hw_state.ctrl1 << (dpll * 6);
 
 			I915_WRITE(DPLL_CTRL1, val);
@@ -1565,7 +1907,7 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 
 		I915_WRITE(DPLL_CTRL2, val);
 
-	} else {
+	} else if (INTEL_INFO(dev)->gen < 9) {
 		WARN_ON(crtc->config->ddi_pll_sel == PORT_CLK_SEL_NONE);
 		I915_WRITE(PORT_CLK_SEL(port), crtc->config->ddi_pll_sel);
 	}
@@ -1583,6 +1925,12 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 	} else if (type == INTEL_OUTPUT_HDMI) {
 		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 
+		if (IS_BROXTON(dev)) {
+			hdmi_level = dev_priv->vbt.
+				ddi_port_info[port].hdmi_level_shift;
+			bxt_ddi_vswing_sequence(dev, hdmi_level, port,
+					INTEL_OUTPUT_HDMI);
+		}
 		intel_hdmi->set_infoframes(encoder,
 					   crtc->config->has_hdmi_sink,
 					   &crtc->config->base.adjusted_mode);
@@ -1624,7 +1972,7 @@ static void intel_ddi_post_disable(struct intel_encoder *intel_encoder)
 	if (IS_SKYLAKE(dev))
 		I915_WRITE(DPLL_CTRL2, (I915_READ(DPLL_CTRL2) |
 					DPLL_CTRL2_DDI_CLK_OFF(port)));
-	else
+	else if (INTEL_INFO(dev)->gen < 9)
 		I915_WRITE(PORT_CLK_SEL(port), PORT_CLK_SEL_NONE);
 }
 
@@ -1689,105 +2037,6 @@ static void intel_disable_ddi(struct intel_encoder *intel_encoder)
 	}
 }
 
-static int skl_get_cdclk_freq(struct drm_i915_private *dev_priv)
-{
-	uint32_t lcpll1 = I915_READ(LCPLL1_CTL);
-	uint32_t cdctl = I915_READ(CDCLK_CTL);
-	uint32_t linkrate;
-
-	if (!(lcpll1 & LCPLL_PLL_ENABLE)) {
-		WARN(1, "LCPLL1 not enabled\n");
-		return 24000; /* 24MHz is the cd freq with NSSC ref */
-	}
-
-	if ((cdctl & CDCLK_FREQ_SEL_MASK) == CDCLK_FREQ_540)
-		return 540000;
-
-	linkrate = (I915_READ(DPLL_CTRL1) &
-		    DPLL_CRTL1_LINK_RATE_MASK(SKL_DPLL0)) >> 1;
-
-	if (linkrate == DPLL_CRTL1_LINK_RATE_2160 ||
-	    linkrate == DPLL_CRTL1_LINK_RATE_1080) {
-		/* vco 8640 */
-		switch (cdctl & CDCLK_FREQ_SEL_MASK) {
-		case CDCLK_FREQ_450_432:
-			return 432000;
-		case CDCLK_FREQ_337_308:
-			return 308570;
-		case CDCLK_FREQ_675_617:
-			return 617140;
-		default:
-			WARN(1, "Unknown cd freq selection\n");
-		}
-	} else {
-		/* vco 8100 */
-		switch (cdctl & CDCLK_FREQ_SEL_MASK) {
-		case CDCLK_FREQ_450_432:
-			return 450000;
-		case CDCLK_FREQ_337_308:
-			return 337500;
-		case CDCLK_FREQ_675_617:
-			return 675000;
-		default:
-			WARN(1, "Unknown cd freq selection\n");
-		}
-	}
-
-	/* error case, do as if DPLL0 isn't enabled */
-	return 24000;
-}
-
-static int bdw_get_cdclk_freq(struct drm_i915_private *dev_priv)
-{
-	uint32_t lcpll = I915_READ(LCPLL_CTL);
-	uint32_t freq = lcpll & LCPLL_CLK_FREQ_MASK;
-
-	if (lcpll & LCPLL_CD_SOURCE_FCLK)
-		return 800000;
-	else if (I915_READ(FUSE_STRAP) & HSW_CDCLK_LIMIT)
-		return 450000;
-	else if (freq == LCPLL_CLK_FREQ_450)
-		return 450000;
-	else if (freq == LCPLL_CLK_FREQ_54O_BDW)
-		return 540000;
-	else if (freq == LCPLL_CLK_FREQ_337_5_BDW)
-		return 337500;
-	else
-		return 675000;
-}
-
-static int hsw_get_cdclk_freq(struct drm_i915_private *dev_priv)
-{
-	struct drm_device *dev = dev_priv->dev;
-	uint32_t lcpll = I915_READ(LCPLL_CTL);
-	uint32_t freq = lcpll & LCPLL_CLK_FREQ_MASK;
-
-	if (lcpll & LCPLL_CD_SOURCE_FCLK)
-		return 800000;
-	else if (I915_READ(FUSE_STRAP) & HSW_CDCLK_LIMIT)
-		return 450000;
-	else if (freq == LCPLL_CLK_FREQ_450)
-		return 450000;
-	else if (IS_HSW_ULT(dev))
-		return 337500;
-	else
-		return 540000;
-}
-
-int intel_ddi_get_cdclk_freq(struct drm_i915_private *dev_priv)
-{
-	struct drm_device *dev = dev_priv->dev;
-
-	if (IS_SKYLAKE(dev))
-		return skl_get_cdclk_freq(dev_priv);
-
-	if (IS_BROADWELL(dev))
-		return bdw_get_cdclk_freq(dev_priv);
-
-	/* Haswell */
-	return hsw_get_cdclk_freq(dev_priv);
-}
-
 static void hsw_ddi_pll_enable(struct drm_i915_private *dev_priv,
 			       struct intel_shared_dpll *pll)
 {
@@ -1887,7 +2136,7 @@ static void skl_ddi_pll_enable(struct drm_i915_private *dev_priv,
 	val = I915_READ(DPLL_CTRL1);
 
 	val &= ~(DPLL_CTRL1_HDMI_MODE(dpll) | DPLL_CTRL1_SSC(dpll) |
-		 DPLL_CRTL1_LINK_RATE_MASK(dpll));
+		 DPLL_CTRL1_LINK_RATE_MASK(dpll));
 	val |= pll->config.hw_state.ctrl1 << (dpll * 6);
 
 	I915_WRITE(DPLL_CTRL1, val);
@@ -1963,22 +2212,325 @@ static void skl_shared_dplls_init(struct drm_i915_private *dev_priv)
 	}
 }
 
+static void broxton_phy_init(struct drm_i915_private *dev_priv,
+			     enum dpio_phy phy)
+{
+	enum port port;
+	uint32_t val;
+
+	val = I915_READ(BXT_P_CR_GT_DISP_PWRON);
+	val |= GT_DISPLAY_POWER_ON(phy);
+	I915_WRITE(BXT_P_CR_GT_DISP_PWRON, val);
+
+	/* Considering 10ms timeout until BSpec is updated */
+	if (wait_for(I915_READ(BXT_PORT_CL1CM_DW0(phy)) & PHY_POWER_GOOD, 10))
+		DRM_ERROR("timeout during PHY%d power on\n", phy);
+
+	for (port =  (phy == DPIO_PHY0 ? PORT_B : PORT_A);
+	     port <= (phy == DPIO_PHY0 ? PORT_C : PORT_A); port++) {
+		int lane;
+
+		for (lane = 0; lane < 4; lane++) {
+			val = I915_READ(BXT_PORT_TX_DW14_LN(port, lane));
+			/*
+			 * Note that on CHV this flag is called UPAR, but has
+			 * the same function.
+			 */
+			val &= ~LATENCY_OPTIM;
+			if (lane != 1)
+				val |= LATENCY_OPTIM;
+
+			I915_WRITE(BXT_PORT_TX_DW14_LN(port, lane), val);
+		}
+	}
+
+	/* Program PLL Rcomp code offset */
+	val = I915_READ(BXT_PORT_CL1CM_DW9(phy));
+	val &= ~IREF0RC_OFFSET_MASK;
+	val |= 0xE4 << IREF0RC_OFFSET_SHIFT;
+	I915_WRITE(BXT_PORT_CL1CM_DW9(phy), val);
+
+	val = I915_READ(BXT_PORT_CL1CM_DW10(phy));
+	val &= ~IREF1RC_OFFSET_MASK;
+	val |= 0xE4 << IREF1RC_OFFSET_SHIFT;
+	I915_WRITE(BXT_PORT_CL1CM_DW10(phy), val);
+
+	/* Program power gating */
+	val = I915_READ(BXT_PORT_CL1CM_DW28(phy));
+	val |= OCL1_POWER_DOWN_EN | DW28_OLDO_DYN_PWR_DOWN_EN |
+		SUS_CLK_CONFIG;
+	I915_WRITE(BXT_PORT_CL1CM_DW28(phy), val);
+
+	if (phy == DPIO_PHY0) {
+		val = I915_READ(BXT_PORT_CL2CM_DW6_BC);
+		val |= DW6_OLDO_DYN_PWR_DOWN_EN;
+		I915_WRITE(BXT_PORT_CL2CM_DW6_BC, val);
+	}
+
+	val = I915_READ(BXT_PORT_CL1CM_DW30(phy));
+	val &= ~OCL2_LDOFUSE_PWR_DIS;
+	/*
+	 * On PHY1 disable power on the second channel, since no port is
+	 * connected there. On PHY0 both channels have a port, so leave it
+	 * enabled.
+	 * TODO: port C is only connected on BXT-P, so on BXT0/1 we should
+	 * power down the second channel on PHY0 as well.
+	 */
+	if (phy == DPIO_PHY1)
+		val |= OCL2_LDOFUSE_PWR_DIS;
+	I915_WRITE(BXT_PORT_CL1CM_DW30(phy), val);
+
+	if (phy == DPIO_PHY0) {
+		uint32_t grc_code;
+		/*
+		 * PHY0 isn't connected to an RCOMP resistor so copy over
+		 * the corresponding calibrated value from PHY1, and disable
+		 * the automatic calibration on PHY0.
+		 */
+		if (wait_for(I915_READ(BXT_PORT_REF_DW3(DPIO_PHY1)) & GRC_DONE,
+			     10))
+			DRM_ERROR("timeout waiting for PHY1 GRC\n");
+
+		val = I915_READ(BXT_PORT_REF_DW6(DPIO_PHY1));
+		val = (val & GRC_CODE_MASK) >> GRC_CODE_SHIFT;
+		grc_code = val << GRC_CODE_FAST_SHIFT |
+			   val << GRC_CODE_SLOW_SHIFT |
+			   val;
+		I915_WRITE(BXT_PORT_REF_DW6(DPIO_PHY0), grc_code);
+
+		val = I915_READ(BXT_PORT_REF_DW8(DPIO_PHY0));
+		val |= GRC_DIS | GRC_RDY_OVRD;
+		I915_WRITE(BXT_PORT_REF_DW8(DPIO_PHY0), val);
+	}
+
+	val = I915_READ(BXT_PHY_CTL_FAMILY(phy));
+	val |= COMMON_RESET_DIS;
+	I915_WRITE(BXT_PHY_CTL_FAMILY(phy), val);
+}
+
+void broxton_ddi_phy_init(struct drm_device *dev)
+{
+	/* Enable PHY1 first since it provides Rcomp for PHY0 */
+	broxton_phy_init(dev->dev_private, DPIO_PHY1);
+	broxton_phy_init(dev->dev_private, DPIO_PHY0);
+}
+
+static void broxton_phy_uninit(struct drm_i915_private *dev_priv,
+			       enum dpio_phy phy)
+{
+	uint32_t val;
+
+	val = I915_READ(BXT_PHY_CTL_FAMILY(phy));
+	val &= ~COMMON_RESET_DIS;
+	I915_WRITE(BXT_PHY_CTL_FAMILY(phy), val);
+}
+
+void broxton_ddi_phy_uninit(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	broxton_phy_uninit(dev_priv, DPIO_PHY1);
+	broxton_phy_uninit(dev_priv, DPIO_PHY0);
+
+	/* FIXME: do this in broxton_phy_uninit per phy */
+	I915_WRITE(BXT_P_CR_GT_DISP_PWRON, 0);
+}
+
+static const char * const bxt_ddi_pll_names[] = {
+	"PORT PLL A",
+	"PORT PLL B",
+	"PORT PLL C",
+};
+
+static void bxt_ddi_pll_enable(struct drm_i915_private *dev_priv,
+				struct intel_shared_dpll *pll)
+{
+	uint32_t temp;
+	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
+
+	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	temp &= ~PORT_PLL_REF_SEL;
+	/* Non-SSC reference */
+	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
+
+	/* Disable 10 bit clock */
+	temp = I915_READ(BXT_PORT_PLL_EBB_4(port));
+	temp &= ~PORT_PLL_10BIT_CLK_ENABLE;
+	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
+
+	/* Write P1 & P2 */
+	temp = I915_READ(BXT_PORT_PLL_EBB_0(port));
+	temp &= ~(PORT_PLL_P1_MASK | PORT_PLL_P2_MASK);
+	temp |= pll->config.hw_state.ebb0;
+	I915_WRITE(BXT_PORT_PLL_EBB_0(port), temp);
+
+	/* Write M2 integer */
+	temp = I915_READ(BXT_PORT_PLL(port, 0));
+	temp &= ~PORT_PLL_M2_MASK;
+	temp |= pll->config.hw_state.pll0;
+	I915_WRITE(BXT_PORT_PLL(port, 0), temp);
+
+	/* Write N */
+	temp = I915_READ(BXT_PORT_PLL(port, 1));
+	temp &= ~PORT_PLL_N_MASK;
+	temp |= pll->config.hw_state.pll1;
+	I915_WRITE(BXT_PORT_PLL(port, 1), temp);
+
+	/* Write M2 fraction */
+	temp = I915_READ(BXT_PORT_PLL(port, 2));
+	temp &= ~PORT_PLL_M2_FRAC_MASK;
+	temp |= pll->config.hw_state.pll2;
+	I915_WRITE(BXT_PORT_PLL(port, 2), temp);
+
+	/* Write M2 fraction enable */
+	temp = I915_READ(BXT_PORT_PLL(port, 3));
+	temp &= ~PORT_PLL_M2_FRAC_ENABLE;
+	temp |= pll->config.hw_state.pll3;
+	I915_WRITE(BXT_PORT_PLL(port, 3), temp);
+
+	/* Write coeff */
+	temp = I915_READ(BXT_PORT_PLL(port, 6));
+	temp &= ~PORT_PLL_PROP_COEFF_MASK;
+	temp &= ~PORT_PLL_INT_COEFF_MASK;
+	temp &= ~PORT_PLL_GAIN_CTL_MASK;
+	temp |= pll->config.hw_state.pll6;
+	I915_WRITE(BXT_PORT_PLL(port, 6), temp);
+
+	/* Write calibration val */
+	temp = I915_READ(BXT_PORT_PLL(port, 8));
+	temp &= ~PORT_PLL_TARGET_CNT_MASK;
+	temp |= pll->config.hw_state.pll8;
+	I915_WRITE(BXT_PORT_PLL(port, 8), temp);
+
+	temp = I915_READ(BXT_PORT_PLL(port, 9));
+	temp &= ~PORT_PLL_LOCK_THRESHOLD_MASK;
+	temp |= (5 << 1);
+	I915_WRITE(BXT_PORT_PLL(port, 9), temp);
+
+	temp = I915_READ(BXT_PORT_PLL(port, 10));
+	temp &= ~PORT_PLL_DCO_AMP_OVR_EN_H;
+	temp &= ~PORT_PLL_DCO_AMP_MASK;
+	temp |= pll->config.hw_state.pll10;
+	I915_WRITE(BXT_PORT_PLL(port, 10), temp);
+
+	/* Recalibrate with new settings */
+	temp = I915_READ(BXT_PORT_PLL_EBB_4(port));
+	temp |= PORT_PLL_RECALIBRATE;
+	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
+	/* Enable 10 bit clock */
+	temp |= PORT_PLL_10BIT_CLK_ENABLE;
+	I915_WRITE(BXT_PORT_PLL_EBB_4(port), temp);
+
+	/* Enable PLL */
+	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	temp |= PORT_PLL_ENABLE;
+	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
+	POSTING_READ(BXT_PORT_PLL_ENABLE(port));
+
+	if (wait_for_atomic_us((I915_READ(BXT_PORT_PLL_ENABLE(port)) &
+			PORT_PLL_LOCK), 200))
+		DRM_ERROR("PLL %d not locked\n", port);
+
+	/*
+	 * While we write to the group register to program all lanes at once we
+	 * can read only lane registers and we pick lanes 0/1 for that.
+	 */
+	temp = I915_READ(BXT_PORT_PCS_DW12_LN01(port));
+	temp &= ~LANE_STAGGER_MASK;
+	temp &= ~LANESTAGGER_STRAP_OVRD;
+	temp |= pll->config.hw_state.pcsdw12;
+	I915_WRITE(BXT_PORT_PCS_DW12_GRP(port), temp);
+}
+
+static void bxt_ddi_pll_disable(struct drm_i915_private *dev_priv,
+					struct intel_shared_dpll *pll)
+{
+	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
+	uint32_t temp;
+
+	temp = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	temp &= ~PORT_PLL_ENABLE;
+	I915_WRITE(BXT_PORT_PLL_ENABLE(port), temp);
+	POSTING_READ(BXT_PORT_PLL_ENABLE(port));
+}
+
+static bool bxt_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
+					struct intel_shared_dpll *pll,
+					struct intel_dpll_hw_state *hw_state)
+{
+	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
+	uint32_t val;
+
+	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	val = I915_READ(BXT_PORT_PLL_ENABLE(port));
+	if (!(val & PORT_PLL_ENABLE))
+		return false;
+
+	hw_state->ebb0 = I915_READ(BXT_PORT_PLL_EBB_0(port));
+	hw_state->pll0 = I915_READ(BXT_PORT_PLL(port, 0));
+	hw_state->pll1 = I915_READ(BXT_PORT_PLL(port, 1));
+	hw_state->pll2 = I915_READ(BXT_PORT_PLL(port, 2));
+	hw_state->pll3 = I915_READ(BXT_PORT_PLL(port, 3));
+	hw_state->pll6 = I915_READ(BXT_PORT_PLL(port, 6));
+	hw_state->pll8 = I915_READ(BXT_PORT_PLL(port, 8));
+	hw_state->pll10 = I915_READ(BXT_PORT_PLL(port, 10));
+	/*
+	 * While we write to the group register to program all lanes at once we
+	 * can read only lane registers. We configure all lanes the same way, so
+	 * here just read out lanes 0/1 and output a note if lanes 2/3 differ.
+	 */
+	hw_state->pcsdw12 = I915_READ(BXT_PORT_PCS_DW12_LN01(port));
+	if (I915_READ(BXT_PORT_PCS_DW12_LN23(port) != hw_state->pcsdw12))
+		DRM_DEBUG_DRIVER("lane stagger config different for lane 01 (%08x) and 23 (%08x)\n",
+				 hw_state->pcsdw12,
+				 I915_READ(BXT_PORT_PCS_DW12_LN23(port)));
+
+	return true;
+}
+
+static void bxt_shared_dplls_init(struct drm_i915_private *dev_priv)
+{
+	int i;
+
+	dev_priv->num_shared_dpll = 3;
+
+	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
+		dev_priv->shared_dplls[i].id = i;
+		dev_priv->shared_dplls[i].name = bxt_ddi_pll_names[i];
+		dev_priv->shared_dplls[i].disable = bxt_ddi_pll_disable;
+		dev_priv->shared_dplls[i].enable = bxt_ddi_pll_enable;
+		dev_priv->shared_dplls[i].get_hw_state =
+			bxt_ddi_pll_get_hw_state;
+	}
+}
+
 void intel_ddi_pll_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t val = I915_READ(LCPLL_CTL);
+	int cdclk_freq;
 
 	if (IS_SKYLAKE(dev))
 		skl_shared_dplls_init(dev_priv);
+	else if (IS_BROXTON(dev))
+		bxt_shared_dplls_init(dev_priv);
 	else
 		hsw_shared_dplls_init(dev_priv);
 
-	DRM_DEBUG_KMS("CDCLK running at %dKHz\n",
-		      intel_ddi_get_cdclk_freq(dev_priv));
+	cdclk_freq = dev_priv->display.get_display_clock_speed(dev);
+	DRM_DEBUG_KMS("CDCLK running at %dKHz\n", cdclk_freq);
 
 	if (IS_SKYLAKE(dev)) {
+		dev_priv->skl_boot_cdclk = cdclk_freq;
 		if (!(I915_READ(LCPLL1_CTL) & LCPLL_PLL_ENABLE))
 			DRM_ERROR("LCPLL1 is disabled\n");
+		else
+			intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
+	} else if (IS_BROXTON(dev)) {
+		broxton_init_cdclk(dev);
+		broxton_ddi_phy_init(dev);
 	} else {
 		/*
 		 * The LCPLL register should be turned on by the BIOS. For now
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index d0f3cbc87474..dcb1d25d6f05 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -45,29 +45,33 @@
 #include <drm/drm_rect.h>
 #include <linux/dma_remapping.h>
 
-/* Primary plane formats supported by all gen */
-#define COMMON_PRIMARY_FORMATS \
-	DRM_FORMAT_C8, \
-	DRM_FORMAT_RGB565, \
-	DRM_FORMAT_XRGB8888, \
-	DRM_FORMAT_ARGB8888
-
 /* Primary plane formats for gen <= 3 */
-static const uint32_t intel_primary_formats_gen2[] = {
-	COMMON_PRIMARY_FORMATS,
+static const uint32_t i8xx_primary_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
 	DRM_FORMAT_XRGB1555,
-	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_XRGB8888,
 };
 
 /* Primary plane formats for gen >= 4 */
-static const uint32_t intel_primary_formats_gen4[] = {
-	COMMON_PRIMARY_FORMATS, \
+static const uint32_t i965_primary_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_XRGB2101010,
+	DRM_FORMAT_XBGR2101010,
+};
+
+static const uint32_t skl_primary_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
 	DRM_FORMAT_ABGR8888,
 	DRM_FORMAT_XRGB2101010,
-	DRM_FORMAT_ARGB2101010,
 	DRM_FORMAT_XBGR2101010,
-	DRM_FORMAT_ABGR2101010,
 };
 
 /* Cursor formats */
@@ -82,9 +86,9 @@ static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 static void ironlake_pch_clock_get(struct intel_crtc *crtc,
 				   struct intel_crtc_state *pipe_config);
 
-static int intel_set_mode(struct drm_crtc *crtc, struct drm_display_mode *mode,
-			  int x, int y, struct drm_framebuffer *old_fb,
-			  struct drm_atomic_state *state);
+static int intel_set_mode(struct drm_crtc *crtc,
+			  struct drm_atomic_state *state,
+			  bool force_restore);
 static int intel_framebuffer_init(struct drm_device *dev,
 				  struct intel_framebuffer *ifb,
 				  struct drm_mode_fb_cmd2 *mode_cmd,
@@ -103,6 +107,12 @@ static void chv_prepare_pll(struct intel_crtc *crtc,
 			    const struct intel_crtc_state *pipe_config);
 static void intel_begin_crtc_commit(struct drm_crtc *crtc);
 static void intel_finish_crtc_commit(struct drm_crtc *crtc);
+static void skl_init_scalers(struct drm_device *dev, struct intel_crtc *intel_crtc,
+	struct intel_crtc_state *crtc_state);
+static int i9xx_get_refclk(const struct intel_crtc_state *crtc_state,
+			   int num_connectors);
+static void intel_crtc_enable_planes(struct drm_crtc *crtc);
+static void intel_crtc_disable_planes(struct drm_crtc *crtc);
 
 static struct intel_encoder *intel_find_encoder(struct intel_connector *connector, int pipe)
 {
@@ -400,6 +410,18 @@ static const intel_limit_t intel_limits_chv = {
 	.p2 = {	.p2_slow = 1, .p2_fast = 14 },
 };
 
+static const intel_limit_t intel_limits_bxt = {
+	/* FIXME: find real dot limits */
+	.dot = { .min = 0, .max = INT_MAX },
+	.vco = { .min = 4800000, .max = 6480000 },
+	.n = { .min = 1, .max = 1 },
+	.m1 = { .min = 2, .max = 2 },
+	/* FIXME: find real m2 limits */
+	.m2 = { .min = 2 << 22, .max = 255 << 22 },
+	.p1 = { .min = 2, .max = 4 },
+	.p2 = { .p2_slow = 1, .p2_fast = 20 },
+};
+
 static void vlv_clock(int refclk, intel_clock_t *clock)
 {
 	clock->m = clock->m1 * clock->m2;
@@ -435,15 +457,12 @@ static bool intel_pipe_will_have_type(const struct intel_crtc_state *crtc_state,
 				      int type)
 {
 	struct drm_atomic_state *state = crtc_state->base.state;
+	struct drm_connector *connector;
 	struct drm_connector_state *connector_state;
 	struct intel_encoder *encoder;
 	int i, num_connectors = 0;
 
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i])
-			continue;
-
-		connector_state = state->connector_states[i];
+	for_each_connector_in_state(state, connector, connector_state, i) {
 		if (connector_state->crtc != crtc_state->base.crtc)
 			continue;
 
@@ -511,7 +530,9 @@ intel_limit(struct intel_crtc_state *crtc_state, int refclk)
 	struct drm_device *dev = crtc_state->base.crtc->dev;
 	const intel_limit_t *limit;
 
-	if (HAS_PCH_SPLIT(dev))
+	if (IS_BROXTON(dev))
+		limit = &intel_limits_bxt;
+	else if (HAS_PCH_SPLIT(dev))
 		limit = intel_ironlake_limit(crtc_state, refclk);
 	else if (IS_G4X(dev)) {
 		limit = intel_g4x_limit(crtc_state);
@@ -596,11 +617,11 @@ static bool intel_PLL_is_valid(struct drm_device *dev,
 	if (clock->m1  < limit->m1.min  || limit->m1.max  < clock->m1)
 		INTELPllInvalid("m1 out of range\n");
 
-	if (!IS_PINEVIEW(dev) && !IS_VALLEYVIEW(dev))
+	if (!IS_PINEVIEW(dev) && !IS_VALLEYVIEW(dev) && !IS_BROXTON(dev))
 		if (clock->m1 <= clock->m2)
 			INTELPllInvalid("m1 <= m2\n");
 
-	if (!IS_VALLEYVIEW(dev)) {
+	if (!IS_VALLEYVIEW(dev) && !IS_BROXTON(dev)) {
 		if (clock->p < limit->p.min || limit->p.max < clock->p)
 			INTELPllInvalid("p out of range\n");
 		if (clock->m < limit->m.min || limit->m.max < clock->m)
@@ -953,6 +974,15 @@ chv_find_best_dpll(const intel_limit_t *limit,
 	return found;
 }
 
+bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state, int target_clock,
+			intel_clock_t *best_clock)
+{
+	int refclk = i9xx_get_refclk(crtc_state, 0);
+
+	return chv_find_best_dpll(intel_limit(crtc_state, refclk), crtc_state,
+				  target_clock, refclk, NULL, best_clock);
+}
+
 bool intel_crtc_active(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -1111,9 +1141,9 @@ static void assert_dsi_pll(struct drm_i915_private *dev_priv, bool state)
 	u32 val;
 	bool cur_state;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	val = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	cur_state = val & DSI_PLL_VCO_EN;
 	I915_STATE_WARN(cur_state != state,
@@ -1632,13 +1662,15 @@ static void chv_enable_pll(struct intel_crtc *crtc,
 
 	BUG_ON(!IS_CHERRYVIEW(dev_priv->dev));
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* Enable back the 10bit clock to display controller */
 	tmp = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port));
 	tmp |= DPIO_DCLKP_EN;
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port), tmp);
 
+	mutex_unlock(&dev_priv->sb_lock);
+
 	/*
 	 * Need to wait > 100ns between dclkp clock enable bit and PLL enable.
 	 */
@@ -1654,8 +1686,6 @@ static void chv_enable_pll(struct intel_crtc *crtc,
 	/* not sure when this should be written */
 	I915_WRITE(DPLL_MD(pipe), pipe_config->dpll_hw_state.dpll_md);
 	POSTING_READ(DPLL_MD(pipe));
-
-	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 static int intel_num_dvo_pipes(struct drm_device *dev)
@@ -1797,7 +1827,7 @@ static void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 	I915_WRITE(DPLL(pipe), val);
 	POSTING_READ(DPLL(pipe));
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* Disable 10bit clock to display controller */
 	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port));
@@ -1815,11 +1845,12 @@ static void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW1_CH1, val);
 	}
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
-		struct intel_digital_port *dport)
+			 struct intel_digital_port *dport,
+			 unsigned int expected_mask)
 {
 	u32 port_mask;
 	int dpll_reg;
@@ -1832,6 +1863,7 @@ void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 	case PORT_C:
 		port_mask = DPLL_PORTC_READY_MASK;
 		dpll_reg = DPLL(0);
+		expected_mask <<= 4;
 		break;
 	case PORT_D:
 		port_mask = DPLL_PORTD_READY_MASK;
@@ -1841,9 +1873,9 @@ void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 		BUG();
 	}
 
-	if (wait_for((I915_READ(dpll_reg) & port_mask) == 0, 1000))
-		WARN(1, "timed out waiting for port %c ready: 0x%08x\n",
-		     port_name(dport->port), I915_READ(dpll_reg));
+	if (wait_for((I915_READ(dpll_reg) & port_mask) == expected_mask, 1000))
+		WARN(1, "timed out waiting for port %c ready: got 0x%x, expected 0x%x\n",
+		     port_name(dport->port), I915_READ(dpll_reg) & port_mask, expected_mask);
 }
 
 static void intel_prepare_shared_dpll(struct intel_crtc *crtc)
@@ -2104,7 +2136,7 @@ static void intel_enable_pipe(struct intel_crtc *crtc)
 	 * a plane.  On ILK+ the pipe PLLs are integrated, so we don't
 	 * need the check.
 	 */
-	if (!HAS_PCH_SPLIT(dev_priv->dev))
+	if (HAS_GMCH_DISPLAY(dev_priv->dev))
 		if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI))
 			assert_dsi_pll_enabled(dev_priv);
 		else
@@ -2179,20 +2211,6 @@ static void intel_disable_pipe(struct intel_crtc *crtc)
 		intel_wait_for_pipe_off(crtc);
 }
 
-/*
- * Plane regs are double buffered, going from enabled->disabled needs a
- * trigger in order to latch.  The display address reg provides this.
- */
-void intel_flush_primary_plane(struct drm_i915_private *dev_priv,
-			       enum plane plane)
-{
-	struct drm_device *dev = dev_priv->dev;
-	u32 reg = INTEL_INFO(dev)->gen >= 4 ? DSPSURF(plane) : DSPADDR(plane);
-
-	I915_WRITE(reg, I915_READ(reg));
-	POSTING_READ(reg);
-}
-
 /**
  * intel_enable_primary_hw_plane - enable the primary plane on a given pipe
  * @plane:  plane to be enabled
@@ -2209,45 +2227,7 @@ static void intel_enable_primary_hw_plane(struct drm_plane *plane,
 
 	/* If the pipe isn't enabled, we can't pump pixels and may hang */
 	assert_pipe_enabled(dev_priv, intel_crtc->pipe);
-
-	if (intel_crtc->primary_enabled)
-		return;
-
-	intel_crtc->primary_enabled = true;
-
-	dev_priv->display.update_primary_plane(crtc, plane->fb,
-					       crtc->x, crtc->y);
-
-	/*
-	 * BDW signals flip done immediately if the plane
-	 * is disabled, even if the plane enable is already
-	 * armed to occur at the next vblank :(
-	 */
-	if (IS_BROADWELL(dev))
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-}
-
-/**
- * intel_disable_primary_hw_plane - disable the primary hardware plane
- * @plane: plane to be disabled
- * @crtc: crtc for the plane
- *
- * Disable @plane on @crtc, making sure that the pipe is running first.
- */
-static void intel_disable_primary_hw_plane(struct drm_plane *plane,
-					   struct drm_crtc *crtc)
-{
-	struct drm_device *dev = plane->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-
-	if (WARN_ON(!intel_crtc->active))
-		return;
-
-	if (!intel_crtc->primary_enabled)
-		return;
-
-	intel_crtc->primary_enabled = false;
+	to_intel_plane_state(plane->state)->visible = true;
 
 	dev_priv->display.update_primary_plane(crtc, plane->fb,
 					       crtc->x, crtc->y);
@@ -2338,13 +2318,6 @@ intel_fill_fb_ggtt_view(struct i915_ggtt_view *view, struct drm_framebuffer *fb,
 	info->pitch = fb->pitches[0];
 	info->fb_modifier = fb->modifier[0];
 
-	if (!(info->fb_modifier == I915_FORMAT_MOD_Y_TILED ||
-	      info->fb_modifier == I915_FORMAT_MOD_Yf_TILED)) {
-		DRM_DEBUG_KMS(
-			      "Y or Yf tiling is needed for 90/270 rotation!\n");
-		return -EINVAL;
-	}
-
 	return 0;
 }
 
@@ -2667,6 +2640,8 @@ static void i9xx_update_primary_plane(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);
+	struct drm_plane *primary = crtc->primary;
+	bool visible = to_intel_plane_state(primary->state)->visible;
 	struct drm_i915_gem_object *obj;
 	int plane = intel_crtc->plane;
 	unsigned long linear_offset;
@@ -2674,7 +2649,7 @@ static void i9xx_update_primary_plane(struct drm_crtc *crtc,
 	u32 reg = DSPCNTR(plane);
 	int pixel_size;
 
-	if (!intel_crtc->primary_enabled) {
+	if (!visible || !fb) {
 		I915_WRITE(reg, 0);
 		if (INTEL_INFO(dev)->gen >= 4)
 			I915_WRITE(DSPSURF(plane), 0);
@@ -2718,26 +2693,21 @@ static void i9xx_update_primary_plane(struct drm_crtc *crtc,
 		dspcntr |= DISPPLANE_8BPP;
 		break;
 	case DRM_FORMAT_XRGB1555:
-	case DRM_FORMAT_ARGB1555:
 		dspcntr |= DISPPLANE_BGRX555;
 		break;
 	case DRM_FORMAT_RGB565:
 		dspcntr |= DISPPLANE_BGRX565;
 		break;
 	case DRM_FORMAT_XRGB8888:
-	case DRM_FORMAT_ARGB8888:
 		dspcntr |= DISPPLANE_BGRX888;
 		break;
 	case DRM_FORMAT_XBGR8888:
-	case DRM_FORMAT_ABGR8888:
 		dspcntr |= DISPPLANE_RGBX888;
 		break;
 	case DRM_FORMAT_XRGB2101010:
-	case DRM_FORMAT_ARGB2101010:
 		dspcntr |= DISPPLANE_BGRX101010;
 		break;
 	case DRM_FORMAT_XBGR2101010:
-	case DRM_FORMAT_ABGR2101010:
 		dspcntr |= DISPPLANE_RGBX101010;
 		break;
 	default:
@@ -2796,6 +2766,8 @@ static void ironlake_update_primary_plane(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);
+	struct drm_plane *primary = crtc->primary;
+	bool visible = to_intel_plane_state(primary->state)->visible;
 	struct drm_i915_gem_object *obj;
 	int plane = intel_crtc->plane;
 	unsigned long linear_offset;
@@ -2803,7 +2775,7 @@ static void ironlake_update_primary_plane(struct drm_crtc *crtc,
 	u32 reg = DSPCNTR(plane);
 	int pixel_size;
 
-	if (!intel_crtc->primary_enabled) {
+	if (!visible || !fb) {
 		I915_WRITE(reg, 0);
 		I915_WRITE(DSPSURF(plane), 0);
 		POSTING_READ(reg);
@@ -2831,19 +2803,15 @@ static void ironlake_update_primary_plane(struct drm_crtc *crtc,
 		dspcntr |= DISPPLANE_BGRX565;
 		break;
 	case DRM_FORMAT_XRGB8888:
-	case DRM_FORMAT_ARGB8888:
 		dspcntr |= DISPPLANE_BGRX888;
 		break;
 	case DRM_FORMAT_XBGR8888:
-	case DRM_FORMAT_ABGR8888:
 		dspcntr |= DISPPLANE_RGBX888;
 		break;
 	case DRM_FORMAT_XRGB2101010:
-	case DRM_FORMAT_ARGB2101010:
 		dspcntr |= DISPPLANE_BGRX101010;
 		break;
 	case DRM_FORMAT_XBGR2101010:
-	case DRM_FORMAT_ABGR2101010:
 		dspcntr |= DISPPLANE_RGBX101010;
 		break;
 	default:
@@ -2936,92 +2904,221 @@ unsigned long intel_plane_obj_offset(struct intel_plane *intel_plane,
 	return i915_gem_obj_ggtt_offset_view(obj, view);
 }
 
-static void skylake_update_primary_plane(struct drm_crtc *crtc,
-					 struct drm_framebuffer *fb,
-					 int x, int y)
+/*
+ * This function detaches (aka. unbinds) unused scalers in hardware
+ */
+void skl_detach_scalers(struct intel_crtc *intel_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);
-	struct drm_i915_gem_object *obj;
-	int pipe = intel_crtc->pipe;
-	u32 plane_ctl, stride_div;
-	unsigned long surf_addr;
+	struct drm_device *dev;
+	struct drm_i915_private *dev_priv;
+	struct intel_crtc_scaler_state *scaler_state;
+	int i;
 
-	if (!intel_crtc->primary_enabled) {
-		I915_WRITE(PLANE_CTL(pipe, 0), 0);
-		I915_WRITE(PLANE_SURF(pipe, 0), 0);
-		POSTING_READ(PLANE_CTL(pipe, 0));
+	if (!intel_crtc || !intel_crtc->config)
 		return;
-	}
 
-	plane_ctl = PLANE_CTL_ENABLE |
-		    PLANE_CTL_PIPE_GAMMA_ENABLE |
-		    PLANE_CTL_PIPE_CSC_ENABLE;
+	dev = intel_crtc->base.dev;
+	dev_priv = dev->dev_private;
+	scaler_state = &intel_crtc->config->scaler_state;
 
-	switch (fb->pixel_format) {
+	/* loop through and disable scalers that aren't in use */
+	for (i = 0; i < intel_crtc->num_scalers; i++) {
+		if (!scaler_state->scalers[i].in_use) {
+			I915_WRITE(SKL_PS_CTRL(intel_crtc->pipe, i), 0);
+			I915_WRITE(SKL_PS_WIN_POS(intel_crtc->pipe, i), 0);
+			I915_WRITE(SKL_PS_WIN_SZ(intel_crtc->pipe, i), 0);
+			DRM_DEBUG_KMS("CRTC:%d Disabled scaler id %u.%u\n",
+				intel_crtc->base.base.id, intel_crtc->pipe, i);
+		}
+	}
+}
+
+u32 skl_plane_ctl_format(uint32_t pixel_format)
+{
+	switch (pixel_format) {
+	case DRM_FORMAT_C8:
+		return PLANE_CTL_FORMAT_INDEXED;
 	case DRM_FORMAT_RGB565:
-		plane_ctl |= PLANE_CTL_FORMAT_RGB_565;
-		break;
-	case DRM_FORMAT_XRGB8888:
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888;
-		break;
-	case DRM_FORMAT_ARGB8888:
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888;
-		plane_ctl |= PLANE_CTL_ALPHA_SW_PREMULTIPLY;
-		break;
+		return PLANE_CTL_FORMAT_RGB_565;
 	case DRM_FORMAT_XBGR8888:
-		plane_ctl |= PLANE_CTL_ORDER_RGBX;
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888;
-		break;
+		return PLANE_CTL_FORMAT_XRGB_8888 | PLANE_CTL_ORDER_RGBX;
+	case DRM_FORMAT_XRGB8888:
+		return PLANE_CTL_FORMAT_XRGB_8888;
+	/*
+	 * XXX: For ARBG/ABGR formats we default to expecting scanout buffers
+	 * to be already pre-multiplied. We need to add a knob (or a different
+	 * DRM_FORMAT) for user-space to configure that.
+	 */
 	case DRM_FORMAT_ABGR8888:
-		plane_ctl |= PLANE_CTL_ORDER_RGBX;
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888;
-		plane_ctl |= PLANE_CTL_ALPHA_SW_PREMULTIPLY;
-		break;
+		return PLANE_CTL_FORMAT_XRGB_8888 | PLANE_CTL_ORDER_RGBX |
+			PLANE_CTL_ALPHA_SW_PREMULTIPLY;
+	case DRM_FORMAT_ARGB8888:
+		return PLANE_CTL_FORMAT_XRGB_8888 |
+			PLANE_CTL_ALPHA_SW_PREMULTIPLY;
 	case DRM_FORMAT_XRGB2101010:
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_2101010;
-		break;
+		return PLANE_CTL_FORMAT_XRGB_2101010;
 	case DRM_FORMAT_XBGR2101010:
-		plane_ctl |= PLANE_CTL_ORDER_RGBX;
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_2101010;
-		break;
+		return PLANE_CTL_ORDER_RGBX | PLANE_CTL_FORMAT_XRGB_2101010;
+	case DRM_FORMAT_YUYV:
+		return PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_YUYV;
+	case DRM_FORMAT_YVYU:
+		return PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_YVYU;
+	case DRM_FORMAT_UYVY:
+		return PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_UYVY;
+	case DRM_FORMAT_VYUY:
+		return PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_VYUY;
 	default:
-		BUG();
+		MISSING_CASE(pixel_format);
 	}
 
-	switch (fb->modifier[0]) {
+	return 0;
+}
+
+u32 skl_plane_ctl_tiling(uint64_t fb_modifier)
+{
+	switch (fb_modifier) {
 	case DRM_FORMAT_MOD_NONE:
 		break;
 	case I915_FORMAT_MOD_X_TILED:
-		plane_ctl |= PLANE_CTL_TILED_X;
-		break;
+		return PLANE_CTL_TILED_X;
 	case I915_FORMAT_MOD_Y_TILED:
-		plane_ctl |= PLANE_CTL_TILED_Y;
-		break;
+		return PLANE_CTL_TILED_Y;
 	case I915_FORMAT_MOD_Yf_TILED:
-		plane_ctl |= PLANE_CTL_TILED_YF;
+		return PLANE_CTL_TILED_YF;
+	default:
+		MISSING_CASE(fb_modifier);
+	}
+
+	return 0;
+}
+
+u32 skl_plane_ctl_rotation(unsigned int rotation)
+{
+	switch (rotation) {
+	case BIT(DRM_ROTATE_0):
 		break;
+	/*
+	 * DRM_ROTATE_ is counter clockwise to stay compatible with Xrandr
+	 * while i915 HW rotation is clockwise, thats why this swapping.
+	 */
+	case BIT(DRM_ROTATE_90):
+		return PLANE_CTL_ROTATE_270;
+	case BIT(DRM_ROTATE_180):
+		return PLANE_CTL_ROTATE_180;
+	case BIT(DRM_ROTATE_270):
+		return PLANE_CTL_ROTATE_90;
 	default:
-		MISSING_CASE(fb->modifier[0]);
+		MISSING_CASE(rotation);
 	}
 
+	return 0;
+}
+
+static void skylake_update_primary_plane(struct drm_crtc *crtc,
+					 struct drm_framebuffer *fb,
+					 int x, int y)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct drm_plane *plane = crtc->primary;
+	bool visible = to_intel_plane_state(plane->state)->visible;
+	struct drm_i915_gem_object *obj;
+	int pipe = intel_crtc->pipe;
+	u32 plane_ctl, stride_div, stride;
+	u32 tile_height, plane_offset, plane_size;
+	unsigned int rotation;
+	int x_offset, y_offset;
+	unsigned long surf_addr;
+	struct intel_crtc_state *crtc_state = intel_crtc->config;
+	struct intel_plane_state *plane_state;
+	int src_x = 0, src_y = 0, src_w = 0, src_h = 0;
+	int dst_x = 0, dst_y = 0, dst_w = 0, dst_h = 0;
+	int scaler_id = -1;
+
+	plane_state = to_intel_plane_state(plane->state);
+
+	if (!visible || !fb) {
+		I915_WRITE(PLANE_CTL(pipe, 0), 0);
+		I915_WRITE(PLANE_SURF(pipe, 0), 0);
+		POSTING_READ(PLANE_CTL(pipe, 0));
+		return;
+	}
+
+	plane_ctl = PLANE_CTL_ENABLE |
+		    PLANE_CTL_PIPE_GAMMA_ENABLE |
+		    PLANE_CTL_PIPE_CSC_ENABLE;
+
+	plane_ctl |= skl_plane_ctl_format(fb->pixel_format);
+	plane_ctl |= skl_plane_ctl_tiling(fb->modifier[0]);
 	plane_ctl |= PLANE_CTL_PLANE_GAMMA_DISABLE;
-	if (crtc->primary->state->rotation == BIT(DRM_ROTATE_180))
-		plane_ctl |= PLANE_CTL_ROTATE_180;
+
+	rotation = plane->state->rotation;
+	plane_ctl |= skl_plane_ctl_rotation(rotation);
 
 	obj = intel_fb_obj(fb);
 	stride_div = intel_fb_stride_alignment(dev, fb->modifier[0],
 					       fb->pixel_format);
-	surf_addr = intel_plane_obj_offset(to_intel_plane(crtc->primary), obj);
+	surf_addr = intel_plane_obj_offset(to_intel_plane(plane), obj);
+
+	/*
+	 * FIXME: intel_plane_state->src, dst aren't set when transitional
+	 * update_plane helpers are called from legacy paths.
+	 * Once full atomic crtc is available, below check can be avoided.
+	 */
+	if (drm_rect_width(&plane_state->src)) {
+		scaler_id = plane_state->scaler_id;
+		src_x = plane_state->src.x1 >> 16;
+		src_y = plane_state->src.y1 >> 16;
+		src_w = drm_rect_width(&plane_state->src) >> 16;
+		src_h = drm_rect_height(&plane_state->src) >> 16;
+		dst_x = plane_state->dst.x1;
+		dst_y = plane_state->dst.y1;
+		dst_w = drm_rect_width(&plane_state->dst);
+		dst_h = drm_rect_height(&plane_state->dst);
+
+		WARN_ON(x != src_x || y != src_y);
+	} else {
+		src_w = intel_crtc->config->pipe_src_w;
+		src_h = intel_crtc->config->pipe_src_h;
+	}
+
+	if (intel_rotation_90_or_270(rotation)) {
+		/* stride = Surface height in tiles */
+		tile_height = intel_tile_height(dev, fb->pixel_format,
+						fb->modifier[0]);
+		stride = DIV_ROUND_UP(fb->height, tile_height);
+		x_offset = stride * tile_height - y - src_h;
+		y_offset = x;
+		plane_size = (src_w - 1) << 16 | (src_h - 1);
+	} else {
+		stride = fb->pitches[0] / stride_div;
+		x_offset = x;
+		y_offset = y;
+		plane_size = (src_h - 1) << 16 | (src_w - 1);
+	}
+	plane_offset = y_offset << 16 | x_offset;
 
 	I915_WRITE(PLANE_CTL(pipe, 0), plane_ctl);
-	I915_WRITE(PLANE_POS(pipe, 0), 0);
-	I915_WRITE(PLANE_OFFSET(pipe, 0), (y << 16) | x);
-	I915_WRITE(PLANE_SIZE(pipe, 0),
-		   (intel_crtc->config->pipe_src_h - 1) << 16 |
-		   (intel_crtc->config->pipe_src_w - 1));
-	I915_WRITE(PLANE_STRIDE(pipe, 0), fb->pitches[0] / stride_div);
+	I915_WRITE(PLANE_OFFSET(pipe, 0), plane_offset);
+	I915_WRITE(PLANE_SIZE(pipe, 0), plane_size);
+	I915_WRITE(PLANE_STRIDE(pipe, 0), stride);
+
+	if (scaler_id >= 0) {
+		uint32_t ps_ctrl = 0;
+
+		WARN_ON(!dst_w || !dst_h);
+		ps_ctrl = PS_SCALER_EN | PS_PLANE_SEL(0) |
+			crtc_state->scaler_state.scalers[scaler_id].mode;
+		I915_WRITE(SKL_PS_CTRL(pipe, scaler_id), ps_ctrl);
+		I915_WRITE(SKL_PS_PWR_GATE(pipe, scaler_id), 0);
+		I915_WRITE(SKL_PS_WIN_POS(pipe, scaler_id), (dst_x << 16) | dst_y);
+		I915_WRITE(SKL_PS_WIN_SZ(pipe, scaler_id), (dst_w << 16) | dst_h);
+		I915_WRITE(PLANE_POS(pipe, 0), 0);
+	} else {
+		I915_WRITE(PLANE_POS(pipe, 0), (dst_y << 16) | dst_x);
+	}
+
 	I915_WRITE(PLANE_SURF(pipe, 0), surf_addr);
 
 	POSTING_READ(PLANE_SURF(pipe, 0));
@@ -3079,6 +3176,19 @@ static void intel_update_primary_planes(struct drm_device *dev)
 	}
 }
 
+void intel_crtc_reset(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+
+	if (!crtc->active)
+		return;
+
+	intel_crtc_disable_planes(&crtc->base);
+	dev_priv->display.crtc_disable(&crtc->base);
+	dev_priv->display.crtc_enable(&crtc->base);
+	intel_crtc_enable_planes(&crtc->base);
+}
+
 void intel_prepare_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -3099,8 +3209,11 @@ void intel_prepare_reset(struct drm_device *dev)
 	 * g33 docs say we should at least disable all the planes.
 	 */
 	for_each_intel_crtc(dev, crtc) {
-		if (crtc->active)
-			dev_priv->display.crtc_disable(&crtc->base);
+		if (!crtc->active)
+			continue;
+
+		intel_crtc_disable_planes(&crtc->base);
+		dev_priv->display.crtc_disable(&crtc->base);
 	}
 }
 
@@ -3152,27 +3265,30 @@ void intel_finish_reset(struct drm_device *dev)
 	drm_modeset_unlock_all(dev);
 }
 
-static int
+static void
 intel_finish_fb(struct drm_framebuffer *old_fb)
 {
 	struct drm_i915_gem_object *obj = intel_fb_obj(old_fb);
-	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 	bool was_interruptible = dev_priv->mm.interruptible;
 	int ret;
 
 	/* Big Hammer, we also need to ensure that any pending
 	 * MI_WAIT_FOR_EVENT inside a user batch buffer on the
 	 * current scanout is retired before unpinning the old
-	 * framebuffer.
+	 * framebuffer. Note that we rely on userspace rendering
+	 * into the buffer attached to the pipe they are waiting
+	 * on. If not, userspace generates a GPU hang with IPEHR
+	 * point to the MI_WAIT_FOR_EVENT.
 	 *
 	 * This should only fail upon a hung GPU, in which case we
 	 * can safely continue.
 	 */
 	dev_priv->mm.interruptible = false;
-	ret = i915_gem_object_finish_gpu(obj);
+	ret = i915_gem_object_wait_rendering(obj, true);
 	dev_priv->mm.interruptible = was_interruptible;
 
-	return ret;
+	WARN_ON(ret);
 }
 
 static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
@@ -3824,7 +3940,7 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 	u32 divsel, phaseinc, auxdiv, phasedir = 0;
 	u32 temp;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* It is necessary to ungate the pixclk gate prior to programming
 	 * the divisors, and gate it back when it is done.
@@ -3901,7 +4017,7 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 
 	I915_WRITE(PIXCLK_GATE, PIXCLK_GATE_UNGATE);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void ironlake_pch_transcoder_set_timings(struct intel_crtc *crtc,
@@ -4039,8 +4155,7 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 		temp &= ~(TRANS_DP_PORT_SEL_MASK |
 			  TRANS_DP_SYNC_MASK |
 			  TRANS_DP_BPC_MASK);
-		temp |= (TRANS_DP_OUTPUT_ENABLE |
-			 TRANS_DP_ENH_FRAMING);
+		temp |= TRANS_DP_OUTPUT_ENABLE;
 		temp |= bpc << 9; /* same format but at 11:9 */
 
 		if (crtc->mode.flags & DRM_MODE_FLAG_PHSYNC)
@@ -4126,6 +4241,26 @@ struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc,
 		goto found;
 	}
 
+	if (IS_BROXTON(dev_priv->dev)) {
+		/* PLL is attached to port in bxt */
+		struct intel_encoder *encoder;
+		struct intel_digital_port *intel_dig_port;
+
+		encoder = intel_ddi_get_crtc_new_encoder(crtc_state);
+		if (WARN_ON(!encoder))
+			return NULL;
+
+		intel_dig_port = enc_to_dig_port(&encoder->base);
+		/* 1:1 mapping between ports and PLLs */
+		i = (enum intel_dpll_id)intel_dig_port->port;
+		pll = &dev_priv->shared_dplls[i];
+		DRM_DEBUG_KMS("CRTC:%d using pre-allocated %s\n",
+			crtc->base.base.id, pll->name);
+		WARN_ON(pll->new_config->crtc_mask);
+
+		goto found;
+	}
+
 	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
 		pll = &dev_priv->shared_dplls[i];
 
@@ -4251,16 +4386,180 @@ static void cpt_verify_modeset(struct drm_device *dev, int pipe)
 	}
 }
 
-static void skylake_pfit_enable(struct intel_crtc *crtc)
+/**
+ * skl_update_scaler_users - Stages update to crtc's scaler state
+ * @intel_crtc: crtc
+ * @crtc_state: crtc_state
+ * @plane: plane (NULL indicates crtc is requesting update)
+ * @plane_state: plane's state
+ * @force_detach: request unconditional detachment of scaler
+ *
+ * This function updates scaler state for requested plane or crtc.
+ * To request scaler usage update for a plane, caller shall pass plane pointer.
+ * To request scaler usage update for crtc, caller shall pass plane pointer
+ * as NULL.
+ *
+ * Return
+ *     0 - scaler_usage updated successfully
+ *    error - requested scaling cannot be supported or other error condition
+ */
+int
+skl_update_scaler_users(
+	struct intel_crtc *intel_crtc, struct intel_crtc_state *crtc_state,
+	struct intel_plane *intel_plane, struct intel_plane_state *plane_state,
+	int force_detach)
+{
+	int need_scaling;
+	int idx;
+	int src_w, src_h, dst_w, dst_h;
+	int *scaler_id;
+	struct drm_framebuffer *fb;
+	struct intel_crtc_scaler_state *scaler_state;
+	unsigned int rotation;
+
+	if (!intel_crtc || !crtc_state)
+		return 0;
+
+	scaler_state = &crtc_state->scaler_state;
+
+	idx = intel_plane ? drm_plane_index(&intel_plane->base) : SKL_CRTC_INDEX;
+	fb = intel_plane ? plane_state->base.fb : NULL;
+
+	if (intel_plane) {
+		src_w = drm_rect_width(&plane_state->src) >> 16;
+		src_h = drm_rect_height(&plane_state->src) >> 16;
+		dst_w = drm_rect_width(&plane_state->dst);
+		dst_h = drm_rect_height(&plane_state->dst);
+		scaler_id = &plane_state->scaler_id;
+		rotation = plane_state->base.rotation;
+	} else {
+		struct drm_display_mode *adjusted_mode =
+			&crtc_state->base.adjusted_mode;
+		src_w = crtc_state->pipe_src_w;
+		src_h = crtc_state->pipe_src_h;
+		dst_w = adjusted_mode->hdisplay;
+		dst_h = adjusted_mode->vdisplay;
+		scaler_id = &scaler_state->scaler_id;
+		rotation = DRM_ROTATE_0;
+	}
+
+	need_scaling = intel_rotation_90_or_270(rotation) ?
+		(src_h != dst_w || src_w != dst_h):
+		(src_w != dst_w || src_h != dst_h);
+
+	/*
+	 * if plane is being disabled or scaler is no more required or force detach
+	 *  - free scaler binded to this plane/crtc
+	 *  - in order to do this, update crtc->scaler_usage
+	 *
+	 * Here scaler state in crtc_state is set free so that
+	 * scaler can be assigned to other user. Actual register
+	 * update to free the scaler is done in plane/panel-fit programming.
+	 * For this purpose crtc/plane_state->scaler_id isn't reset here.
+	 */
+	if (force_detach || !need_scaling || (intel_plane &&
+		(!fb || !plane_state->visible))) {
+		if (*scaler_id >= 0) {
+			scaler_state->scaler_users &= ~(1 << idx);
+			scaler_state->scalers[*scaler_id].in_use = 0;
+
+			DRM_DEBUG_KMS("Staged freeing scaler id %d.%d from %s:%d "
+				"crtc_state = %p scaler_users = 0x%x\n",
+				intel_crtc->pipe, *scaler_id, intel_plane ? "PLANE" : "CRTC",
+				intel_plane ? intel_plane->base.base.id :
+				intel_crtc->base.base.id, crtc_state,
+				scaler_state->scaler_users);
+			*scaler_id = -1;
+		}
+		return 0;
+	}
+
+	/* range checks */
+	if (src_w < SKL_MIN_SRC_W || src_h < SKL_MIN_SRC_H ||
+		dst_w < SKL_MIN_DST_W || dst_h < SKL_MIN_DST_H ||
+
+		src_w > SKL_MAX_SRC_W || src_h > SKL_MAX_SRC_H ||
+		dst_w > SKL_MAX_DST_W || dst_h > SKL_MAX_DST_H) {
+		DRM_DEBUG_KMS("%s:%d scaler_user index %u.%u: src %ux%u dst %ux%u "
+			"size is out of scaler range\n",
+			intel_plane ? "PLANE" : "CRTC",
+			intel_plane ? intel_plane->base.base.id : intel_crtc->base.base.id,
+			intel_crtc->pipe, idx, src_w, src_h, dst_w, dst_h);
+		return -EINVAL;
+	}
+
+	/* check colorkey */
+	if (WARN_ON(intel_plane &&
+		intel_plane->ckey.flags != I915_SET_COLORKEY_NONE)) {
+		DRM_DEBUG_KMS("PLANE:%d scaling %ux%u->%ux%u not allowed with colorkey",
+			intel_plane->base.base.id, src_w, src_h, dst_w, dst_h);
+		return -EINVAL;
+	}
+
+	/* Check src format */
+	if (intel_plane) {
+		switch (fb->pixel_format) {
+		case DRM_FORMAT_RGB565:
+		case DRM_FORMAT_XBGR8888:
+		case DRM_FORMAT_XRGB8888:
+		case DRM_FORMAT_ABGR8888:
+		case DRM_FORMAT_ARGB8888:
+		case DRM_FORMAT_XRGB2101010:
+		case DRM_FORMAT_XBGR2101010:
+		case DRM_FORMAT_YUYV:
+		case DRM_FORMAT_YVYU:
+		case DRM_FORMAT_UYVY:
+		case DRM_FORMAT_VYUY:
+			break;
+		default:
+			DRM_DEBUG_KMS("PLANE:%d FB:%d unsupported scaling format 0x%x\n",
+				intel_plane->base.base.id, fb->base.id, fb->pixel_format);
+			return -EINVAL;
+		}
+	}
+
+	/* mark this plane as a scaler user in crtc_state */
+	scaler_state->scaler_users |= (1 << idx);
+	DRM_DEBUG_KMS("%s:%d staged scaling request for %ux%u->%ux%u "
+		"crtc_state = %p scaler_users = 0x%x\n",
+		intel_plane ? "PLANE" : "CRTC",
+		intel_plane ? intel_plane->base.base.id : intel_crtc->base.base.id,
+		src_w, src_h, dst_w, dst_h, crtc_state, scaler_state->scaler_users);
+	return 0;
+}
+
+static void skylake_pfit_update(struct intel_crtc *crtc, int enable)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe = crtc->pipe;
+	struct intel_crtc_scaler_state *scaler_state =
+		&crtc->config->scaler_state;
+
+	DRM_DEBUG_KMS("for crtc_state = %p\n", crtc->config);
+
+	/* To update pfit, first update scaler state */
+	skl_update_scaler_users(crtc, crtc->config, NULL, NULL, !enable);
+	intel_atomic_setup_scalers(crtc->base.dev, crtc, crtc->config);
+	skl_detach_scalers(crtc);
+	if (!enable)
+		return;
 
 	if (crtc->config->pch_pfit.enabled) {
-		I915_WRITE(PS_CTL(pipe), PS_ENABLE);
-		I915_WRITE(PS_WIN_POS(pipe), crtc->config->pch_pfit.pos);
-		I915_WRITE(PS_WIN_SZ(pipe), crtc->config->pch_pfit.size);
+		int id;
+
+		if (WARN_ON(crtc->config->scaler_state.scaler_id < 0)) {
+			DRM_ERROR("Requesting pfit without getting a scaler first\n");
+			return;
+		}
+
+		id = scaler_state->scaler_id;
+		I915_WRITE(SKL_PS_CTRL(pipe, id), PS_SCALER_EN |
+			PS_FILTER_MEDIUM | scaler_state->scalers[id].mode);
+		I915_WRITE(SKL_PS_WIN_POS(pipe, id), crtc->config->pch_pfit.pos);
+		I915_WRITE(SKL_PS_WIN_SZ(pipe, id), crtc->config->pch_pfit.size);
+
+		DRM_DEBUG_KMS("for crtc_state = %p scaler_id = %d\n", crtc->config, id);
 	}
 }
 
@@ -4299,38 +4598,6 @@ static void intel_enable_sprite_planes(struct drm_crtc *crtc)
 	}
 }
 
-/*
- * Disable a plane internally without actually modifying the plane's state.
- * This will allow us to easily restore the plane later by just reprogramming
- * its state.
- */
-static void disable_plane_internal(struct drm_plane *plane)
-{
-	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct drm_plane_state *state =
-		plane->funcs->atomic_duplicate_state(plane);
-	struct intel_plane_state *intel_state = to_intel_plane_state(state);
-
-	intel_state->visible = false;
-	intel_plane->commit_plane(plane, intel_state);
-
-	intel_plane_destroy_state(plane, state);
-}
-
-static void intel_disable_sprite_planes(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
-	struct drm_plane *plane;
-	struct intel_plane *intel_plane;
-
-	drm_for_each_legacy_plane(plane, &dev->mode_config.plane_list) {
-		intel_plane = to_intel_plane(plane);
-		if (plane->fb && intel_plane->pipe == pipe)
-			disable_plane_internal(plane);
-	}
-}
-
 void hsw_enable_ips(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
@@ -4404,7 +4671,7 @@ static void intel_crtc_load_lut(struct drm_crtc *crtc)
 	if (!crtc->state->enable || !intel_crtc->active)
 		return;
 
-	if (!HAS_PCH_SPLIT(dev_priv->dev)) {
+	if (HAS_GMCH_DISPLAY(dev_priv->dev)) {
 		if (intel_pipe_has_type(intel_crtc, INTEL_OUTPUT_DSI))
 			assert_dsi_pll_enabled(dev_priv);
 		else
@@ -4436,9 +4703,9 @@ static void intel_crtc_load_lut(struct drm_crtc *crtc)
 		hsw_enable_ips(intel_crtc);
 }
 
-static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
+static void intel_crtc_dpms_overlay_disable(struct intel_crtc *intel_crtc)
 {
-	if (!enable && intel_crtc->overlay) {
+	if (intel_crtc->overlay) {
 		struct drm_device *dev = intel_crtc->base.dev;
 		struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -4454,17 +4721,38 @@ static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
 	 */
 }
 
-static void intel_crtc_enable_planes(struct drm_crtc *crtc)
+/**
+ * intel_post_enable_primary - Perform operations after enabling primary plane
+ * @crtc: the CRTC whose primary plane was just enabled
+ *
+ * Performs potentially sleeping operations that must be done after the primary
+ * plane is enabled, such as updating FBC and IPS.  Note that this may be
+ * called due to an explicit primary plane update, or due to an implicit
+ * re-enable that is caused when a sprite plane is updated to no longer
+ * completely hide the primary plane.
+ */
+static void
+intel_post_enable_primary(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);
 	int pipe = intel_crtc->pipe;
 
-	intel_enable_primary_hw_plane(crtc->primary, crtc);
-	intel_enable_sprite_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
-	intel_crtc_dpms_overlay(intel_crtc, true);
+	/*
+	 * BDW signals flip done immediately if the plane
+	 * is disabled, even if the plane enable is already
+	 * armed to occur at the next vblank :(
+	 */
+	if (IS_BROADWELL(dev))
+		intel_wait_for_vblank(dev, pipe);
 
+	/*
+	 * FIXME IPS should be fine as long as one plane is
+	 * enabled, but in practice it seems to have problems
+	 * when going from primary only to sprite only and vice
+	 * versa.
+	 */
 	hsw_enable_ips(intel_crtc);
 
 	mutex_lock(&dev->struct_mutex);
@@ -4472,9 +4760,89 @@ static void intel_crtc_enable_planes(struct drm_crtc *crtc)
 	mutex_unlock(&dev->struct_mutex);
 
 	/*
+	 * Gen2 reports pipe underruns whenever all planes are disabled.
+	 * So don't enable underrun reporting before at least some planes
+	 * are enabled.
+	 * FIXME: Need to fix the logic to work when we turn off all planes
+	 * but leave the pipe running.
+	 */
+	if (IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
+
+	/* Underruns don't raise interrupts, so check manually. */
+	if (HAS_GMCH_DISPLAY(dev))
+		i9xx_check_fifo_underruns(dev_priv);
+}
+
+/**
+ * intel_pre_disable_primary - Perform operations before disabling primary plane
+ * @crtc: the CRTC whose primary plane is to be disabled
+ *
+ * Performs potentially sleeping operations that must be done before the
+ * primary plane is disabled, such as updating FBC and IPS.  Note that this may
+ * be called due to an explicit primary plane update, or due to an implicit
+ * disable that is caused when a sprite plane completely hides the primary
+ * plane.
+ */
+static void
+intel_pre_disable_primary(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);
+	int pipe = intel_crtc->pipe;
+
+	/*
+	 * Gen2 reports pipe underruns whenever all planes are disabled.
+	 * So diasble underrun reporting before all the planes get disabled.
+	 * FIXME: Need to fix the logic to work when we turn off all planes
+	 * but leave the pipe running.
+	 */
+	if (IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
+
+	/*
+	 * Vblank time updates from the shadow to live plane control register
+	 * are blocked if the memory self-refresh mode is active at that
+	 * moment. So to make sure the plane gets truly disabled, disable
+	 * first the self-refresh mode. The self-refresh enable bit in turn
+	 * will be checked/applied by the HW only at the next frame start
+	 * event which is after the vblank start event, so we need to have a
+	 * wait-for-vblank between disabling the plane and the pipe.
+	 */
+	if (HAS_GMCH_DISPLAY(dev))
+		intel_set_memory_cxsr(dev_priv, false);
+
+	mutex_lock(&dev->struct_mutex);
+	if (dev_priv->fbc.crtc == intel_crtc)
+		intel_fbc_disable(dev);
+	mutex_unlock(&dev->struct_mutex);
+
+	/*
+	 * FIXME IPS should be fine as long as one plane is
+	 * enabled, but in practice it seems to have problems
+	 * when going from primary only to sprite only and vice
+	 * versa.
+	 */
+	hsw_disable_ips(intel_crtc);
+}
+
+static void intel_crtc_enable_planes(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+
+	intel_enable_primary_hw_plane(crtc->primary, crtc);
+	intel_enable_sprite_planes(crtc);
+	intel_crtc_update_cursor(crtc, true);
+
+	intel_post_enable_primary(crtc);
+
+	/*
 	 * FIXME: Once we grow proper nuclear flip support out of this we need
 	 * to compute the mask of flip planes precisely. For the time being
-	 * consider this a flip from a NULL plane.
+	 * consider this a flip to a NULL plane.
 	 */
 	intel_frontbuffer_flip(dev, INTEL_FRONTBUFFER_ALL_MASK(pipe));
 }
@@ -4482,21 +4850,23 @@ static void intel_crtc_enable_planes(struct drm_crtc *crtc)
 static void intel_crtc_disable_planes(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);
+	struct intel_plane *intel_plane;
 	int pipe = intel_crtc->pipe;
 
 	intel_crtc_wait_for_pending_flips(crtc);
 
-	if (dev_priv->fbc.crtc == intel_crtc)
-		intel_fbc_disable(dev);
+	intel_pre_disable_primary(crtc);
 
-	hsw_disable_ips(intel_crtc);
+	intel_crtc_dpms_overlay_disable(intel_crtc);
+	for_each_intel_plane(dev, intel_plane) {
+		if (intel_plane->pipe == pipe) {
+			struct drm_crtc *from = intel_plane->base.crtc;
 
-	intel_crtc_dpms_overlay(intel_crtc, false);
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_sprite_planes(crtc);
-	intel_disable_primary_hw_plane(crtc->primary, crtc);
+			intel_plane->disable_plane(&intel_plane->base,
+						   from ?: crtc, true);
+		}
+	}
 
 	/*
 	 * FIXME: Once we grow proper nuclear flip support out of this we need
@@ -4575,8 +4945,6 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 
 	if (HAS_PCH_CPT(dev))
 		cpt_verify_modeset(dev, intel_crtc->pipe);
-
-	intel_crtc_enable_planes(crtc);
 }
 
 /* IPS only exists on ULT machines and is tied to pipe A. */
@@ -4664,10 +5032,12 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 
 	intel_ddi_enable_pipe_clock(intel_crtc);
 
-	if (IS_SKYLAKE(dev))
-		skylake_pfit_enable(intel_crtc);
-	else
+	if (INTEL_INFO(dev)->gen == 9)
+		skylake_pfit_update(intel_crtc, 1);
+	else if (INTEL_INFO(dev)->gen < 9)
 		ironlake_pfit_enable(intel_crtc);
+	else
+		MISSING_CASE(INTEL_INFO(dev)->gen);
 
 	/*
 	 * On ILK+ LUT must be loaded before the pipe is running but with
@@ -4698,22 +5068,6 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	/* If we change the relative order between pipe/planes enabling, we need
 	 * to change the workaround. */
 	haswell_mode_set_planes_workaround(intel_crtc);
-	intel_crtc_enable_planes(crtc);
-}
-
-static void skylake_pfit_disable(struct intel_crtc *crtc)
-{
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe = crtc->pipe;
-
-	/* To avoid upsetting the power well on haswell only disable the pfit if
-	 * it's in use. The hw state code will make sure we get this right. */
-	if (crtc->config->pch_pfit.enabled) {
-		I915_WRITE(PS_CTL(pipe), 0);
-		I915_WRITE(PS_WIN_POS(pipe), 0);
-		I915_WRITE(PS_WIN_SZ(pipe), 0);
-	}
 }
 
 static void ironlake_pfit_disable(struct intel_crtc *crtc)
@@ -4743,8 +5097,6 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 	if (!intel_crtc->active)
 		return;
 
-	intel_crtc_disable_planes(crtc);
-
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
 
@@ -4758,13 +5110,14 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 
 	ironlake_pfit_disable(intel_crtc);
 
+	if (intel_crtc->config->has_pch_encoder)
+		ironlake_fdi_disable(crtc);
+
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->post_disable)
 			encoder->post_disable(encoder);
 
 	if (intel_crtc->config->has_pch_encoder) {
-		ironlake_fdi_disable(crtc);
-
 		ironlake_disable_pch_transcoder(dev_priv, pipe);
 
 		if (HAS_PCH_CPT(dev)) {
@@ -4807,8 +5160,6 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	if (!intel_crtc->active)
 		return;
 
-	intel_crtc_disable_planes(crtc);
-
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		intel_opregion_notify_encoder(encoder, false);
 		encoder->disable(encoder);
@@ -4827,10 +5178,12 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 
 	intel_ddi_disable_transcoder_func(dev_priv, cpu_transcoder);
 
-	if (IS_SKYLAKE(dev))
-		skylake_pfit_disable(intel_crtc);
-	else
+	if (INTEL_INFO(dev)->gen == 9)
+		skylake_pfit_update(intel_crtc, 0);
+	else if (INTEL_INFO(dev)->gen < 9)
 		ironlake_pfit_disable(intel_crtc);
+	else
+		MISSING_CASE(INTEL_INFO(dev)->gen);
 
 	intel_ddi_disable_pipe_clock(intel_crtc);
 
@@ -4994,16 +5347,403 @@ static void modeset_update_crtc_power_domains(struct drm_atomic_state *state)
 	intel_display_set_init_power(dev_priv, false);
 }
 
+void broxton_set_cdclk(struct drm_device *dev, int frequency)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t divider;
+	uint32_t ratio;
+	uint32_t current_freq;
+	int ret;
+
+	/* frequency = 19.2MHz * ratio / 2 / div{1,1.5,2,4} */
+	switch (frequency) {
+	case 144000:
+		divider = BXT_CDCLK_CD2X_DIV_SEL_4;
+		ratio = BXT_DE_PLL_RATIO(60);
+		break;
+	case 288000:
+		divider = BXT_CDCLK_CD2X_DIV_SEL_2;
+		ratio = BXT_DE_PLL_RATIO(60);
+		break;
+	case 384000:
+		divider = BXT_CDCLK_CD2X_DIV_SEL_1_5;
+		ratio = BXT_DE_PLL_RATIO(60);
+		break;
+	case 576000:
+		divider = BXT_CDCLK_CD2X_DIV_SEL_1;
+		ratio = BXT_DE_PLL_RATIO(60);
+		break;
+	case 624000:
+		divider = BXT_CDCLK_CD2X_DIV_SEL_1;
+		ratio = BXT_DE_PLL_RATIO(65);
+		break;
+	case 19200:
+		/*
+		 * Bypass frequency with DE PLL disabled. Init ratio, divider
+		 * to suppress GCC warning.
+		 */
+		ratio = 0;
+		divider = 0;
+		break;
+	default:
+		DRM_ERROR("unsupported CDCLK freq %d", frequency);
+
+		return;
+	}
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+	/* Inform power controller of upcoming frequency change */
+	ret = sandybridge_pcode_write(dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ,
+				      0x80000000);
+	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	if (ret) {
+		DRM_ERROR("PCode CDCLK freq change notify failed (err %d, freq %d)\n",
+			  ret, frequency);
+		return;
+	}
+
+	current_freq = I915_READ(CDCLK_CTL) & CDCLK_FREQ_DECIMAL_MASK;
+	/* convert from .1 fixpoint MHz with -1MHz offset to kHz */
+	current_freq = current_freq * 500 + 1000;
+
+	/*
+	 * DE PLL has to be disabled when
+	 * - setting to 19.2MHz (bypass, PLL isn't used)
+	 * - before setting to 624MHz (PLL needs toggling)
+	 * - before setting to any frequency from 624MHz (PLL needs toggling)
+	 */
+	if (frequency == 19200 || frequency == 624000 ||
+	    current_freq == 624000) {
+		I915_WRITE(BXT_DE_PLL_ENABLE, ~BXT_DE_PLL_PLL_ENABLE);
+		/* Timeout 200us */
+		if (wait_for(!(I915_READ(BXT_DE_PLL_ENABLE) & BXT_DE_PLL_LOCK),
+			     1))
+			DRM_ERROR("timout waiting for DE PLL unlock\n");
+	}
+
+	if (frequency != 19200) {
+		uint32_t val;
+
+		val = I915_READ(BXT_DE_PLL_CTL);
+		val &= ~BXT_DE_PLL_RATIO_MASK;
+		val |= ratio;
+		I915_WRITE(BXT_DE_PLL_CTL, val);
+
+		I915_WRITE(BXT_DE_PLL_ENABLE, BXT_DE_PLL_PLL_ENABLE);
+		/* Timeout 200us */
+		if (wait_for(I915_READ(BXT_DE_PLL_ENABLE) & BXT_DE_PLL_LOCK, 1))
+			DRM_ERROR("timeout waiting for DE PLL lock\n");
+
+		val = I915_READ(CDCLK_CTL);
+		val &= ~BXT_CDCLK_CD2X_DIV_SEL_MASK;
+		val |= divider;
+		/*
+		 * Disable SSA Precharge when CD clock frequency < 500 MHz,
+		 * enable otherwise.
+		 */
+		val &= ~BXT_CDCLK_SSA_PRECHARGE_ENABLE;
+		if (frequency >= 500000)
+			val |= BXT_CDCLK_SSA_PRECHARGE_ENABLE;
+
+		val &= ~CDCLK_FREQ_DECIMAL_MASK;
+		/* convert from kHz to .1 fixpoint MHz with -1MHz offset */
+		val |= (frequency - 1000) / 500;
+		I915_WRITE(CDCLK_CTL, val);
+	}
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+	ret = sandybridge_pcode_write(dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ,
+				      DIV_ROUND_UP(frequency, 25000));
+	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	if (ret) {
+		DRM_ERROR("PCode CDCLK freq set failed, (err %d, freq %d)\n",
+			  ret, frequency);
+		return;
+	}
+
+	dev_priv->cdclk_freq = frequency;
+}
+
+void broxton_init_cdclk(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t val;
+
+	/*
+	 * NDE_RSTWRN_OPT RST PCH Handshake En must always be 0b on BXT
+	 * or else the reset will hang because there is no PCH to respond.
+	 * Move the handshake programming to initialization sequence.
+	 * Previously was left up to BIOS.
+	 */
+	val = I915_READ(HSW_NDE_RSTWRN_OPT);
+	val &= ~RESET_PCH_HANDSHAKE_ENABLE;
+	I915_WRITE(HSW_NDE_RSTWRN_OPT, val);
+
+	/* Enable PG1 for cdclk */
+	intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
+
+	/* check if cd clock is enabled */
+	if (I915_READ(BXT_DE_PLL_ENABLE) & BXT_DE_PLL_PLL_ENABLE) {
+		DRM_DEBUG_KMS("Display already initialized\n");
+		return;
+	}
+
+	/*
+	 * FIXME:
+	 * - The initial CDCLK needs to be read from VBT.
+	 *   Need to make this change after VBT has changes for BXT.
+	 * - check if setting the max (or any) cdclk freq is really necessary
+	 *   here, it belongs to modeset time
+	 */
+	broxton_set_cdclk(dev, 624000);
+
+	I915_WRITE(DBUF_CTL, I915_READ(DBUF_CTL) | DBUF_POWER_REQUEST);
+	POSTING_READ(DBUF_CTL);
+
+	udelay(10);
+
+	if (!(I915_READ(DBUF_CTL) & DBUF_POWER_STATE))
+		DRM_ERROR("DBuf power enable timeout!\n");
+}
+
+void broxton_uninit_cdclk(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(DBUF_CTL, I915_READ(DBUF_CTL) & ~DBUF_POWER_REQUEST);
+	POSTING_READ(DBUF_CTL);
+
+	udelay(10);
+
+	if (I915_READ(DBUF_CTL) & DBUF_POWER_STATE)
+		DRM_ERROR("DBuf power disable timeout!\n");
+
+	/* Set minimum (bypass) frequency, in effect turning off the DE PLL */
+	broxton_set_cdclk(dev, 19200);
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+}
+
+static const struct skl_cdclk_entry {
+	unsigned int freq;
+	unsigned int vco;
+} skl_cdclk_frequencies[] = {
+	{ .freq = 308570, .vco = 8640 },
+	{ .freq = 337500, .vco = 8100 },
+	{ .freq = 432000, .vco = 8640 },
+	{ .freq = 450000, .vco = 8100 },
+	{ .freq = 540000, .vco = 8100 },
+	{ .freq = 617140, .vco = 8640 },
+	{ .freq = 675000, .vco = 8100 },
+};
+
+static unsigned int skl_cdclk_decimal(unsigned int freq)
+{
+	return (freq - 1000) / 500;
+}
+
+static unsigned int skl_cdclk_get_vco(unsigned int freq)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(skl_cdclk_frequencies); i++) {
+		const struct skl_cdclk_entry *e = &skl_cdclk_frequencies[i];
+
+		if (e->freq == freq)
+			return e->vco;
+	}
+
+	return 8100;
+}
+
+static void
+skl_dpll0_enable(struct drm_i915_private *dev_priv, unsigned int required_vco)
+{
+	unsigned int min_freq;
+	u32 val;
+
+	/* select the minimum CDCLK before enabling DPLL 0 */
+	val = I915_READ(CDCLK_CTL);
+	val &= ~CDCLK_FREQ_SEL_MASK | ~CDCLK_FREQ_DECIMAL_MASK;
+	val |= CDCLK_FREQ_337_308;
+
+	if (required_vco == 8640)
+		min_freq = 308570;
+	else
+		min_freq = 337500;
+
+	val = CDCLK_FREQ_337_308 | skl_cdclk_decimal(min_freq);
+
+	I915_WRITE(CDCLK_CTL, val);
+	POSTING_READ(CDCLK_CTL);
+
+	/*
+	 * We always enable DPLL0 with the lowest link rate possible, but still
+	 * taking into account the VCO required to operate the eDP panel at the
+	 * desired frequency. The usual DP link rates operate with a VCO of
+	 * 8100 while the eDP 1.4 alternate link rates need a VCO of 8640.
+	 * The modeset code is responsible for the selection of the exact link
+	 * rate later on, with the constraint of choosing a frequency that
+	 * works with required_vco.
+	 */
+	val = I915_READ(DPLL_CTRL1);
+
+	val &= ~(DPLL_CTRL1_HDMI_MODE(SKL_DPLL0) | DPLL_CTRL1_SSC(SKL_DPLL0) |
+		 DPLL_CTRL1_LINK_RATE_MASK(SKL_DPLL0));
+	val |= DPLL_CTRL1_OVERRIDE(SKL_DPLL0);
+	if (required_vco == 8640)
+		val |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1080,
+					    SKL_DPLL0);
+	else
+		val |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810,
+					    SKL_DPLL0);
+
+	I915_WRITE(DPLL_CTRL1, val);
+	POSTING_READ(DPLL_CTRL1);
+
+	I915_WRITE(LCPLL1_CTL, I915_READ(LCPLL1_CTL) | LCPLL_PLL_ENABLE);
+
+	if (wait_for(I915_READ(LCPLL1_CTL) & LCPLL_PLL_LOCK, 5))
+		DRM_ERROR("DPLL0 not locked\n");
+}
+
+static bool skl_cdclk_pcu_ready(struct drm_i915_private *dev_priv)
+{
+	int ret;
+	u32 val;
+
+	/* inform PCU we want to change CDCLK */
+	val = SKL_CDCLK_PREPARE_FOR_CHANGE;
+	mutex_lock(&dev_priv->rps.hw_lock);
+	ret = sandybridge_pcode_read(dev_priv, SKL_PCODE_CDCLK_CONTROL, &val);
+	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	return ret == 0 && (val & SKL_CDCLK_READY_FOR_CHANGE);
+}
+
+static bool skl_cdclk_wait_for_pcu_ready(struct drm_i915_private *dev_priv)
+{
+	unsigned int i;
+
+	for (i = 0; i < 15; i++) {
+		if (skl_cdclk_pcu_ready(dev_priv))
+			return true;
+		udelay(10);
+	}
+
+	return false;
+}
+
+static void skl_set_cdclk(struct drm_i915_private *dev_priv, unsigned int freq)
+{
+	u32 freq_select, pcu_ack;
+
+	DRM_DEBUG_DRIVER("Changing CDCLK to %dKHz\n", freq);
+
+	if (!skl_cdclk_wait_for_pcu_ready(dev_priv)) {
+		DRM_ERROR("failed to inform PCU about cdclk change\n");
+		return;
+	}
+
+	/* set CDCLK_CTL */
+	switch(freq) {
+	case 450000:
+	case 432000:
+		freq_select = CDCLK_FREQ_450_432;
+		pcu_ack = 1;
+		break;
+	case 540000:
+		freq_select = CDCLK_FREQ_540;
+		pcu_ack = 2;
+		break;
+	case 308570:
+	case 337500:
+	default:
+		freq_select = CDCLK_FREQ_337_308;
+		pcu_ack = 0;
+		break;
+	case 617140:
+	case 675000:
+		freq_select = CDCLK_FREQ_675_617;
+		pcu_ack = 3;
+		break;
+	}
+
+	I915_WRITE(CDCLK_CTL, freq_select | skl_cdclk_decimal(freq));
+	POSTING_READ(CDCLK_CTL);
+
+	/* inform PCU of the change */
+	mutex_lock(&dev_priv->rps.hw_lock);
+	sandybridge_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, pcu_ack);
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
+void skl_uninit_cdclk(struct drm_i915_private *dev_priv)
+{
+	/* disable DBUF power */
+	I915_WRITE(DBUF_CTL, I915_READ(DBUF_CTL) & ~DBUF_POWER_REQUEST);
+	POSTING_READ(DBUF_CTL);
+
+	udelay(10);
+
+	if (I915_READ(DBUF_CTL) & DBUF_POWER_STATE)
+		DRM_ERROR("DBuf power disable timeout\n");
+
+	/* disable DPLL0 */
+	I915_WRITE(LCPLL1_CTL, I915_READ(LCPLL1_CTL) & ~LCPLL_PLL_ENABLE);
+	if (wait_for(!(I915_READ(LCPLL1_CTL) & LCPLL_PLL_LOCK), 1))
+		DRM_ERROR("Couldn't disable DPLL0\n");
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+}
+
+void skl_init_cdclk(struct drm_i915_private *dev_priv)
+{
+	u32 val;
+	unsigned int required_vco;
+
+	/* enable PCH reset handshake */
+	val = I915_READ(HSW_NDE_RSTWRN_OPT);
+	I915_WRITE(HSW_NDE_RSTWRN_OPT, val | RESET_PCH_HANDSHAKE_ENABLE);
+
+	/* enable PG1 and Misc I/O */
+	intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
+
+	/* DPLL0 already enabed !? */
+	if (I915_READ(LCPLL1_CTL) & LCPLL_PLL_ENABLE) {
+		DRM_DEBUG_DRIVER("DPLL0 already running\n");
+		return;
+	}
+
+	/* enable DPLL0 */
+	required_vco = skl_cdclk_get_vco(dev_priv->skl_boot_cdclk);
+	skl_dpll0_enable(dev_priv, required_vco);
+
+	/* set CDCLK to the frequency the BIOS chose */
+	skl_set_cdclk(dev_priv, dev_priv->skl_boot_cdclk);
+
+	/* enable DBUF power */
+	I915_WRITE(DBUF_CTL, I915_READ(DBUF_CTL) | DBUF_POWER_REQUEST);
+	POSTING_READ(DBUF_CTL);
+
+	udelay(10);
+
+	if (!(I915_READ(DBUF_CTL) & DBUF_POWER_STATE))
+		DRM_ERROR("DBuf power enable timeout\n");
+}
+
 /* returns HPLL frequency in kHz */
 static int valleyview_get_vco(struct drm_i915_private *dev_priv)
 {
 	int hpll_freq, vco_freq[] = { 800, 1600, 2000, 2400 };
 
 	/* Obtain SKU information */
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	hpll_freq = vlv_cck_read(dev_priv, CCK_FUSE_REG) &
 		CCK_FUSE_HPLL_FREQ_MASK;
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	return vco_freq[hpll_freq] * 1000;
 }
@@ -5012,16 +5752,16 @@ static void vlv_update_cdclk(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	dev_priv->vlv_cdclk_freq = dev_priv->display.get_display_clock_speed(dev);
+	dev_priv->cdclk_freq = dev_priv->display.get_display_clock_speed(dev);
 	DRM_DEBUG_DRIVER("Current CD clock rate: %d kHz\n",
-			 dev_priv->vlv_cdclk_freq);
+			 dev_priv->cdclk_freq);
 
 	/*
 	 * Program the gmbus_freq based on the cdclk frequency.
 	 * BSpec erroneously claims we should aim for 4MHz, but
 	 * in fact 1MHz is the correct frequency.
 	 */
-	I915_WRITE(GMBUSFREQ_VLV, DIV_ROUND_UP(dev_priv->vlv_cdclk_freq, 1000));
+	I915_WRITE(GMBUSFREQ_VLV, DIV_ROUND_UP(dev_priv->cdclk_freq, 1000));
 }
 
 /* Adjust CDclk dividers to allow high res or save power if possible */
@@ -5030,7 +5770,8 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, cmd;
 
-	WARN_ON(dev_priv->display.get_display_clock_speed(dev) != dev_priv->vlv_cdclk_freq);
+	WARN_ON(dev_priv->display.get_display_clock_speed(dev)
+					!= dev_priv->cdclk_freq);
 
 	if (cdclk >= 320000) /* jump to highest voltage for 400MHz too */
 		cmd = 2;
@@ -5051,12 +5792,13 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	}
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
+	mutex_lock(&dev_priv->sb_lock);
+
 	if (cdclk == 400000) {
 		u32 divider;
 
 		divider = DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, cdclk) - 1;
 
-		mutex_lock(&dev_priv->dpio_lock);
 		/* adjust cdclk divider */
 		val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
 		val &= ~DISPLAY_FREQUENCY_VALUES;
@@ -5067,10 +5809,8 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 			      DISPLAY_FREQUENCY_STATUS) == (divider << DISPLAY_FREQUENCY_STATUS_SHIFT),
 			     50))
 			DRM_ERROR("timed out waiting for CDclk change\n");
-		mutex_unlock(&dev_priv->dpio_lock);
 	}
 
-	mutex_lock(&dev_priv->dpio_lock);
 	/* adjust self-refresh exit latency value */
 	val = vlv_bunit_read(dev_priv, BUNIT_REG_BISOC);
 	val &= ~0x7f;
@@ -5084,7 +5824,8 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	else
 		val |= 3000 / 250; /* 3.0 usec */
 	vlv_bunit_write(dev_priv, BUNIT_REG_BISOC, val);
-	mutex_unlock(&dev_priv->dpio_lock);
+
+	mutex_unlock(&dev_priv->sb_lock);
 
 	vlv_update_cdclk(dev);
 }
@@ -5094,7 +5835,8 @@ static void cherryview_set_cdclk(struct drm_device *dev, int cdclk)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, cmd;
 
-	WARN_ON(dev_priv->display.get_display_clock_speed(dev) != dev_priv->vlv_cdclk_freq);
+	WARN_ON(dev_priv->display.get_display_clock_speed(dev)
+						!= dev_priv->cdclk_freq);
 
 	switch (cdclk) {
 	case 333333:
@@ -5159,37 +5901,89 @@ static int valleyview_calc_cdclk(struct drm_i915_private *dev_priv,
 		return 200000;
 }
 
-/* compute the max pixel clock for new configuration */
-static int intel_mode_max_pixclk(struct drm_i915_private *dev_priv)
+static int broxton_calc_cdclk(struct drm_i915_private *dev_priv,
+			      int max_pixclk)
+{
+	/*
+	 * FIXME:
+	 * - remove the guardband, it's not needed on BXT
+	 * - set 19.2MHz bypass frequency if there are no active pipes
+	 */
+	if (max_pixclk > 576000*9/10)
+		return 624000;
+	else if (max_pixclk > 384000*9/10)
+		return 576000;
+	else if (max_pixclk > 288000*9/10)
+		return 384000;
+	else if (max_pixclk > 144000*9/10)
+		return 288000;
+	else
+		return 144000;
+}
+
+/* Compute the max pixel clock for new configuration. Uses atomic state if
+ * that's non-NULL, look at current state otherwise. */
+static int intel_mode_max_pixclk(struct drm_device *dev,
+				 struct drm_atomic_state *state)
 {
-	struct drm_device *dev = dev_priv->dev;
 	struct intel_crtc *intel_crtc;
+	struct intel_crtc_state *crtc_state;
 	int max_pixclk = 0;
 
 	for_each_intel_crtc(dev, intel_crtc) {
-		if (intel_crtc->new_enabled)
-			max_pixclk = max(max_pixclk,
-					 intel_crtc->new_config->base.adjusted_mode.crtc_clock);
+		if (state)
+			crtc_state =
+				intel_atomic_get_crtc_state(state, intel_crtc);
+		else
+			crtc_state = intel_crtc->config;
+		if (IS_ERR(crtc_state))
+			return PTR_ERR(crtc_state);
+
+		if (!crtc_state->base.enable)
+			continue;
+
+		max_pixclk = max(max_pixclk,
+				 crtc_state->base.adjusted_mode.crtc_clock);
 	}
 
 	return max_pixclk;
 }
 
-static void valleyview_modeset_global_pipes(struct drm_device *dev,
-					    unsigned *prepare_pipes)
+static int valleyview_modeset_global_pipes(struct drm_atomic_state *state)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc;
-	int max_pixclk = intel_mode_max_pixclk(dev_priv);
+	struct drm_i915_private *dev_priv = to_i915(state->dev);
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *crtc_state;
+	int max_pixclk = intel_mode_max_pixclk(state->dev, state);
+	int cdclk, i;
 
-	if (valleyview_calc_cdclk(dev_priv, max_pixclk) ==
-	    dev_priv->vlv_cdclk_freq)
-		return;
+	if (max_pixclk < 0)
+		return max_pixclk;
+
+	if (IS_VALLEYVIEW(dev_priv))
+		cdclk = valleyview_calc_cdclk(dev_priv, max_pixclk);
+	else
+		cdclk = broxton_calc_cdclk(dev_priv, max_pixclk);
+
+	if (cdclk == dev_priv->cdclk_freq)
+		return 0;
+
+	/* add all active pipes to the state */
+	for_each_crtc(state->dev, crtc) {
+		if (!crtc->state->enable)
+			continue;
+
+		crtc_state = drm_atomic_get_crtc_state(state, crtc);
+		if (IS_ERR(crtc_state))
+			return PTR_ERR(crtc_state);
+	}
 
 	/* disable/enable all currently active pipes while we change cdclk */
-	for_each_intel_crtc(dev, intel_crtc)
-		if (intel_crtc->base.state->enable)
-			*prepare_pipes |= (1 << intel_crtc->pipe);
+	for_each_crtc_in_state(state, crtc, crtc_state, i)
+		if (crtc_state->enable)
+			crtc_state->mode_changed = true;
+
+	return 0;
 }
 
 static void vlv_program_pfi_credits(struct drm_i915_private *dev_priv)
@@ -5201,7 +5995,7 @@ static void vlv_program_pfi_credits(struct drm_i915_private *dev_priv)
 	else
 		default_credits = PFI_CREDIT(8);
 
-	if (DIV_ROUND_CLOSEST(dev_priv->vlv_cdclk_freq, 1000) >= dev_priv->rps.cz_freq) {
+	if (DIV_ROUND_CLOSEST(dev_priv->cdclk_freq, 1000) >= dev_priv->rps.cz_freq) {
 		/* CHV suggested value is 31 or 63 */
 		if (IS_CHERRYVIEW(dev_priv))
 			credits = PFI_CREDIT_31;
@@ -5228,14 +6022,21 @@ static void vlv_program_pfi_credits(struct drm_i915_private *dev_priv)
 	WARN_ON(I915_READ(GCI_CONTROL) & PFI_CREDIT_RESEND);
 }
 
-static void valleyview_modeset_global_resources(struct drm_atomic_state *state)
+static void valleyview_modeset_global_resources(struct drm_atomic_state *old_state)
 {
-	struct drm_device *dev = state->dev;
+	struct drm_device *dev = old_state->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int max_pixclk = intel_mode_max_pixclk(dev_priv);
-	int req_cdclk = valleyview_calc_cdclk(dev_priv, max_pixclk);
+	int max_pixclk = intel_mode_max_pixclk(dev, NULL);
+	int req_cdclk;
+
+	/* The path in intel_mode_max_pixclk() with a NULL atomic state should
+	 * never fail. */
+	if (WARN_ON(max_pixclk < 0))
+		return;
 
-	if (req_cdclk != dev_priv->vlv_cdclk_freq) {
+	req_cdclk = valleyview_calc_cdclk(dev_priv, max_pixclk);
+
+	if (req_cdclk != dev_priv->cdclk_freq) {
 		/*
 		 * FIXME: We can end up here with all power domains off, yet
 		 * with a CDCLK frequency other than the minimum. To account
@@ -5326,11 +6127,6 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
-
-	intel_crtc_enable_planes(crtc);
-
-	/* Underruns don't raise interrupts, so check manually. */
-	i9xx_check_fifo_underruns(dev_priv);
 }
 
 static void i9xx_set_pll_dividers(struct intel_crtc *crtc)
@@ -5387,21 +6183,6 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
-
-	intel_crtc_enable_planes(crtc);
-
-	/*
-	 * Gen2 reports pipe underruns whenever all planes are disabled.
-	 * So don't enable underrun reporting before at least some planes
-	 * are enabled.
-	 * FIXME: Need to fix the logic to work when we turn off all planes
-	 * but leave the pipe running.
-	 */
-	if (IS_GEN2(dev))
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
-
-	/* Underruns don't raise interrupts, so check manually. */
-	i9xx_check_fifo_underruns(dev_priv);
 }
 
 static void i9xx_pfit_disable(struct intel_crtc *crtc)
@@ -5431,27 +6212,6 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 		return;
 
 	/*
-	 * Gen2 reports pipe underruns whenever all planes are disabled.
-	 * So diasble underrun reporting before all the planes get disabled.
-	 * FIXME: Need to fix the logic to work when we turn off all planes
-	 * but leave the pipe running.
-	 */
-	if (IS_GEN2(dev))
-		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
-
-	/*
-	 * Vblank time updates from the shadow to live plane control register
-	 * are blocked if the memory self-refresh mode is active at that
-	 * moment. So to make sure the plane gets truly disabled, disable
-	 * first the self-refresh mode. The self-refresh enable bit in turn
-	 * will be checked/applied by the HW only at the next frame start
-	 * event which is after the vblank start event, so we need to have a
-	 * wait-for-vblank between disabling the plane and the pipe.
-	 */
-	intel_set_memory_cxsr(dev_priv, false);
-	intel_crtc_disable_planes(crtc);
-
-	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
 	 * We also need to wait on all gmch platforms because of the
@@ -5514,9 +6274,11 @@ void intel_crtc_control(struct drm_crtc *crtc, bool enable)
 			intel_crtc->enabled_power_domains = domains;
 
 			dev_priv->display.crtc_enable(crtc);
+			intel_crtc_enable_planes(crtc);
 		}
 	} else {
 		if (intel_crtc->active) {
+			intel_crtc_disable_planes(crtc);
 			dev_priv->display.crtc_disable(crtc);
 
 			domains = intel_crtc->enabled_power_domains;
@@ -5540,6 +6302,8 @@ void intel_crtc_update_dpms(struct drm_crtc *crtc)
 		enable |= intel_encoder->connectors_active;
 
 	intel_crtc_control(crtc, enable);
+
+	crtc->state->active = enable;
 }
 
 static void intel_crtc_disable(struct drm_crtc *crtc)
@@ -5551,10 +6315,11 @@ static void intel_crtc_disable(struct drm_crtc *crtc)
 	/* crtc should still be enabled when we disable it. */
 	WARN_ON(!crtc->state->enable);
 
+	intel_crtc_disable_planes(crtc);
 	dev_priv->display.crtc_disable(crtc);
 	dev_priv->display.off(crtc);
 
-	crtc->primary->funcs->disable_plane(crtc->primary);
+	drm_plane_helper_disable(crtc->primary);
 
 	/* Update computed state. */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
@@ -5695,65 +6460,80 @@ bool intel_connector_get_hw_state(struct intel_connector *connector)
 	return encoder->get_hw_state(encoder, &pipe);
 }
 
-static int pipe_required_fdi_lanes(struct drm_device *dev, enum pipe pipe)
+static int pipe_required_fdi_lanes(struct intel_crtc_state *crtc_state)
 {
-	struct intel_crtc *crtc =
-		to_intel_crtc(intel_get_crtc_for_pipe(dev, pipe));
-
-	if (crtc->base.state->enable &&
-	    crtc->config->has_pch_encoder)
-		return crtc->config->fdi_lanes;
+	if (crtc_state->base.enable && crtc_state->has_pch_encoder)
+		return crtc_state->fdi_lanes;
 
 	return 0;
 }
 
-static bool ironlake_check_fdi_lanes(struct drm_device *dev, enum pipe pipe,
+static int ironlake_check_fdi_lanes(struct drm_device *dev, enum pipe pipe,
 				     struct intel_crtc_state *pipe_config)
 {
+	struct drm_atomic_state *state = pipe_config->base.state;
+	struct intel_crtc *other_crtc;
+	struct intel_crtc_state *other_crtc_state;
+
 	DRM_DEBUG_KMS("checking fdi config on pipe %c, lanes %i\n",
 		      pipe_name(pipe), pipe_config->fdi_lanes);
 	if (pipe_config->fdi_lanes > 4) {
 		DRM_DEBUG_KMS("invalid fdi lane config on pipe %c: %i lanes\n",
 			      pipe_name(pipe), pipe_config->fdi_lanes);
-		return false;
+		return -EINVAL;
 	}
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
 		if (pipe_config->fdi_lanes > 2) {
 			DRM_DEBUG_KMS("only 2 lanes on haswell, required: %i lanes\n",
 				      pipe_config->fdi_lanes);
-			return false;
+			return -EINVAL;
 		} else {
-			return true;
+			return 0;
 		}
 	}
 
 	if (INTEL_INFO(dev)->num_pipes == 2)
-		return true;
+		return 0;
 
 	/* Ivybridge 3 pipe is really complicated */
 	switch (pipe) {
 	case PIPE_A:
-		return true;
+		return 0;
 	case PIPE_B:
-		if (pipe_config->fdi_lanes > 2 &&
-		    pipe_required_fdi_lanes(dev, PIPE_C) > 0) {
+		if (pipe_config->fdi_lanes <= 2)
+			return 0;
+
+		other_crtc = to_intel_crtc(intel_get_crtc_for_pipe(dev, PIPE_C));
+		other_crtc_state =
+			intel_atomic_get_crtc_state(state, other_crtc);
+		if (IS_ERR(other_crtc_state))
+			return PTR_ERR(other_crtc_state);
+
+		if (pipe_required_fdi_lanes(other_crtc_state) > 0) {
 			DRM_DEBUG_KMS("invalid shared fdi lane config on pipe %c: %i lanes\n",
 				      pipe_name(pipe), pipe_config->fdi_lanes);
-			return false;
+			return -EINVAL;
 		}
-		return true;
+		return 0;
 	case PIPE_C:
 		if (pipe_config->fdi_lanes > 2) {
 			DRM_DEBUG_KMS("only 2 lanes on pipe %c: required %i lanes\n",
 				      pipe_name(pipe), pipe_config->fdi_lanes);
-			return false;
+			return -EINVAL;
 		}
-		if (pipe_required_fdi_lanes(dev, PIPE_B) > 2) {
+
+		other_crtc = to_intel_crtc(intel_get_crtc_for_pipe(dev, PIPE_B));
+		other_crtc_state =
+			intel_atomic_get_crtc_state(state, other_crtc);
+		if (IS_ERR(other_crtc_state))
+			return PTR_ERR(other_crtc_state);
+
+		if (pipe_required_fdi_lanes(other_crtc_state) > 2) {
 			DRM_DEBUG_KMS("fdi link B uses too many lanes to enable link C\n");
-			return false;
+			return -EINVAL;
 		}
-		return true;
+		return 0;
 	default:
 		BUG();
 	}
@@ -5765,8 +6545,8 @@ static int ironlake_fdi_compute_config(struct intel_crtc *intel_crtc,
 {
 	struct drm_device *dev = intel_crtc->base.dev;
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
-	int lane, link_bw, fdi_dotclock;
-	bool setup_ok, needs_recompute = false;
+	int lane, link_bw, fdi_dotclock, ret;
+	bool needs_recompute = false;
 
 retry:
 	/* FDI is a binary signal running at ~2.7GHz, encoding
@@ -5788,9 +6568,9 @@ retry:
 	intel_link_compute_m_n(pipe_config->pipe_bpp, lane, fdi_dotclock,
 			       link_bw, &pipe_config->fdi_m_n);
 
-	setup_ok = ironlake_check_fdi_lanes(intel_crtc->base.dev,
-					    intel_crtc->pipe, pipe_config);
-	if (!setup_ok && pipe_config->pipe_bpp > 6*3) {
+	ret = ironlake_check_fdi_lanes(intel_crtc->base.dev,
+				       intel_crtc->pipe, pipe_config);
+	if (ret == -EINVAL && pipe_config->pipe_bpp > 6*3) {
 		pipe_config->pipe_bpp -= 2*3;
 		DRM_DEBUG_KMS("fdi link bw constraint, reducing pipe bpp to %i\n",
 			      pipe_config->pipe_bpp);
@@ -5803,7 +6583,7 @@ retry:
 	if (needs_recompute)
 		return RETRY;
 
-	return setup_ok ? 0 : -EINVAL;
+	return ret;
 }
 
 static void hsw_compute_ips_config(struct intel_crtc *crtc,
@@ -5820,6 +6600,7 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
+	int ret;
 
 	/* FIXME should check pixel clock limits on all platforms */
 	if (INTEL_INFO(dev)->gen < 4) {
@@ -5860,21 +6641,107 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 		adjusted_mode->hsync_start == adjusted_mode->hdisplay)
 		return -EINVAL;
 
-	if ((IS_G4X(dev) || IS_VALLEYVIEW(dev)) && pipe_config->pipe_bpp > 10*3) {
-		pipe_config->pipe_bpp = 10*3; /* 12bpc is gen5+ */
-	} else if (INTEL_INFO(dev)->gen <= 4 && pipe_config->pipe_bpp > 8*3) {
-		/* only a 8bpc pipe, with 6bpc dither through the panel fitter
-		 * for lvds. */
-		pipe_config->pipe_bpp = 8*3;
-	}
-
 	if (HAS_IPS(dev))
 		hsw_compute_ips_config(crtc, pipe_config);
 
 	if (pipe_config->has_pch_encoder)
 		return ironlake_fdi_compute_config(crtc, pipe_config);
 
-	return 0;
+	/* FIXME: remove below call once atomic mode set is place and all crtc
+	 * related checks called from atomic_crtc_check function */
+	ret = 0;
+	DRM_DEBUG_KMS("intel_crtc = %p drm_state (pipe_config->base.state) = %p\n",
+		crtc, pipe_config->base.state);
+	ret = intel_atomic_setup_scalers(dev, crtc, pipe_config);
+
+	return ret;
+}
+
+static int skylake_get_display_clock_speed(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	uint32_t lcpll1 = I915_READ(LCPLL1_CTL);
+	uint32_t cdctl = I915_READ(CDCLK_CTL);
+	uint32_t linkrate;
+
+	if (!(lcpll1 & LCPLL_PLL_ENABLE)) {
+		WARN(1, "LCPLL1 not enabled\n");
+		return 24000; /* 24MHz is the cd freq with NSSC ref */
+	}
+
+	if ((cdctl & CDCLK_FREQ_SEL_MASK) == CDCLK_FREQ_540)
+		return 540000;
+
+	linkrate = (I915_READ(DPLL_CTRL1) &
+		    DPLL_CTRL1_LINK_RATE_MASK(SKL_DPLL0)) >> 1;
+
+	if (linkrate == DPLL_CTRL1_LINK_RATE_2160 ||
+	    linkrate == DPLL_CTRL1_LINK_RATE_1080) {
+		/* vco 8640 */
+		switch (cdctl & CDCLK_FREQ_SEL_MASK) {
+		case CDCLK_FREQ_450_432:
+			return 432000;
+		case CDCLK_FREQ_337_308:
+			return 308570;
+		case CDCLK_FREQ_675_617:
+			return 617140;
+		default:
+			WARN(1, "Unknown cd freq selection\n");
+		}
+	} else {
+		/* vco 8100 */
+		switch (cdctl & CDCLK_FREQ_SEL_MASK) {
+		case CDCLK_FREQ_450_432:
+			return 450000;
+		case CDCLK_FREQ_337_308:
+			return 337500;
+		case CDCLK_FREQ_675_617:
+			return 675000;
+		default:
+			WARN(1, "Unknown cd freq selection\n");
+		}
+	}
+
+	/* error case, do as if DPLL0 isn't enabled */
+	return 24000;
+}
+
+static int broadwell_get_display_clock_speed(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t lcpll = I915_READ(LCPLL_CTL);
+	uint32_t freq = lcpll & LCPLL_CLK_FREQ_MASK;
+
+	if (lcpll & LCPLL_CD_SOURCE_FCLK)
+		return 800000;
+	else if (I915_READ(FUSE_STRAP) & HSW_CDCLK_LIMIT)
+		return 450000;
+	else if (freq == LCPLL_CLK_FREQ_450)
+		return 450000;
+	else if (freq == LCPLL_CLK_FREQ_54O_BDW)
+		return 540000;
+	else if (freq == LCPLL_CLK_FREQ_337_5_BDW)
+		return 337500;
+	else
+		return 675000;
+}
+
+static int haswell_get_display_clock_speed(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t lcpll = I915_READ(LCPLL_CTL);
+	uint32_t freq = lcpll & LCPLL_CLK_FREQ_MASK;
+
+	if (lcpll & LCPLL_CD_SOURCE_FCLK)
+		return 800000;
+	else if (I915_READ(FUSE_STRAP) & HSW_CDCLK_LIMIT)
+		return 450000;
+	else if (freq == LCPLL_CLK_FREQ_450)
+		return 450000;
+	else if (IS_HSW_ULT(dev))
+		return 337500;
+	else
+		return 540000;
 }
 
 static int valleyview_get_display_clock_speed(struct drm_device *dev)
@@ -5886,9 +6753,9 @@ static int valleyview_get_display_clock_speed(struct drm_device *dev)
 	if (dev_priv->hpll_freq == 0)
 		dev_priv->hpll_freq = valleyview_get_vco(dev_priv);
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	divider = val & DISPLAY_FREQUENCY_VALUES;
 
@@ -5899,6 +6766,11 @@ static int valleyview_get_display_clock_speed(struct drm_device *dev)
 	return DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, divider + 1);
 }
 
+static int ilk_get_display_clock_speed(struct drm_device *dev)
+{
+	return 450000;
+}
+
 static int i945_get_display_clock_speed(struct drm_device *dev)
 {
 	return 400000;
@@ -5906,7 +6778,7 @@ static int i945_get_display_clock_speed(struct drm_device *dev)
 
 static int i915_get_display_clock_speed(struct drm_device *dev)
 {
-	return 333000;
+	return 333333;
 }
 
 static int i9xx_misc_get_display_clock_speed(struct drm_device *dev)
@@ -5922,19 +6794,19 @@ static int pnv_get_display_clock_speed(struct drm_device *dev)
 
 	switch (gcfgc & GC_DISPLAY_CLOCK_MASK) {
 	case GC_DISPLAY_CLOCK_267_MHZ_PNV:
-		return 267000;
+		return 266667;
 	case GC_DISPLAY_CLOCK_333_MHZ_PNV:
-		return 333000;
+		return 333333;
 	case GC_DISPLAY_CLOCK_444_MHZ_PNV:
-		return 444000;
+		return 444444;
 	case GC_DISPLAY_CLOCK_200_MHZ_PNV:
 		return 200000;
 	default:
 		DRM_ERROR("Unknown pnv display core clock 0x%04x\n", gcfgc);
 	case GC_DISPLAY_CLOCK_133_MHZ_PNV:
-		return 133000;
+		return 133333;
 	case GC_DISPLAY_CLOCK_167_MHZ_PNV:
-		return 167000;
+		return 166667;
 	}
 }
 
@@ -5945,11 +6817,11 @@ static int i915gm_get_display_clock_speed(struct drm_device *dev)
 	pci_read_config_word(dev->pdev, GCFGC, &gcfgc);
 
 	if (gcfgc & GC_LOW_FREQUENCY_ENABLE)
-		return 133000;
+		return 133333;
 	else {
 		switch (gcfgc & GC_DISPLAY_CLOCK_MASK) {
 		case GC_DISPLAY_CLOCK_333_MHZ:
-			return 333000;
+			return 333333;
 		default:
 		case GC_DISPLAY_CLOCK_190_200_MHZ:
 			return 190000;
@@ -5959,7 +6831,7 @@ static int i915gm_get_display_clock_speed(struct drm_device *dev)
 
 static int i865_get_display_clock_speed(struct drm_device *dev)
 {
-	return 266000;
+	return 266667;
 }
 
 static int i855_get_display_clock_speed(struct drm_device *dev)
@@ -5975,7 +6847,7 @@ static int i855_get_display_clock_speed(struct drm_device *dev)
 	case GC_CLOCK_166_250:
 		return 250000;
 	case GC_CLOCK_100_133:
-		return 133000;
+		return 133333;
 	}
 
 	/* Shouldn't happen */
@@ -5984,7 +6856,7 @@ static int i855_get_display_clock_speed(struct drm_device *dev)
 
 static int i830_get_display_clock_speed(struct drm_device *dev)
 {
-	return 133000;
+	return 133333;
 }
 
 static void
@@ -6037,7 +6909,7 @@ static int i9xx_get_refclk(const struct intel_crtc_state *crtc_state,
 
 	WARN_ON(!crtc_state->base.state);
 
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_VALLEYVIEW(dev) || IS_BROXTON(dev)) {
 		refclk = 100000;
 	} else if (intel_pipe_will_have_type(crtc_state, INTEL_OUTPUT_LVDS) &&
 	    intel_panel_use_ssc(dev_priv) && num_connectors < 2) {
@@ -6225,7 +7097,7 @@ static void vlv_prepare_pll(struct intel_crtc *crtc,
 	u32 bestn, bestm1, bestm2, bestp1, bestp2;
 	u32 coreclk, reg_val;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	bestn = pipe_config->dpll.n;
 	bestm1 = pipe_config->dpll.m1;
@@ -6303,7 +7175,7 @@ static void vlv_prepare_pll(struct intel_crtc *crtc,
 	vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW7(pipe), coreclk);
 
 	vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW11(pipe), 0x87871000);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void chv_update_pll(struct intel_crtc *crtc,
@@ -6348,7 +7220,7 @@ static void chv_prepare_pll(struct intel_crtc *crtc,
 	I915_WRITE(dpll_reg,
 		   pipe_config->dpll_hw_state.dpll & ~DPLL_VCO_ENABLE);
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* p1 and p2 divider */
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW13(port),
@@ -6421,7 +7293,7 @@ static void chv_prepare_pll(struct intel_crtc *crtc,
 			vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port)) |
 			DPIO_AFC_RECAL);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 /**
@@ -6792,14 +7664,14 @@ static int i9xx_crtc_compute_clock(struct intel_crtc *crtc,
 	struct intel_encoder *encoder;
 	const intel_limit_t *limit;
 	struct drm_atomic_state *state = crtc_state->base.state;
+	struct drm_connector *connector;
 	struct drm_connector_state *connector_state;
 	int i;
 
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i])
-			continue;
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
 
-		connector_state = state->connector_states[i];
+	for_each_connector_in_state(state, connector, connector_state, i) {
 		if (connector_state->crtc != &crtc->base)
 			continue;
 
@@ -6922,9 +7794,9 @@ static void vlv_crtc_clock_get(struct intel_crtc *crtc,
 	if (!(pipe_config->dpll_hw_state.dpll & DPLL_VCO_ENABLE))
 		return;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	mdiv = vlv_dpio_read(dev_priv, pipe, VLV_PLL_DW3(pipe));
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	clock.m1 = (mdiv >> DPIO_M1DIV_SHIFT) & 7;
 	clock.m2 = mdiv & DPIO_M2DIV_MASK;
@@ -7018,12 +7890,12 @@ static void chv_crtc_clock_get(struct intel_crtc *crtc,
 	u32 cmn_dw13, pll_dw0, pll_dw1, pll_dw2;
 	int refclk = 100000;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	cmn_dw13 = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW13(port));
 	pll_dw0 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW0(port));
 	pll_dw1 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW1(port));
 	pll_dw2 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW2(port));
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	clock.m1 = (pll_dw1 & 0x7) == DPIO_CHV_M1_DIV_BY_2 ? 2 : 0;
 	clock.m2 = ((pll_dw0 & 0xff) << 22) | (pll_dw2 & 0x3fffff);
@@ -7389,7 +8261,7 @@ static void lpt_enable_clkout_dp(struct drm_device *dev, bool with_spread,
 		 with_fdi, "LP PCH doesn't have FDI\n"))
 		with_fdi = false;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	tmp = intel_sbi_read(dev_priv, SBI_SSCCTL, SBI_ICLK);
 	tmp &= ~SBI_SSCCTL_DISABLE;
@@ -7415,7 +8287,7 @@ static void lpt_enable_clkout_dp(struct drm_device *dev, bool with_spread,
 	tmp |= SBI_GEN0_CFG_BUFFENABLE_DISABLE;
 	intel_sbi_write(dev_priv, reg, tmp, SBI_ICLK);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 /* Sequence to disable CLKOUT_DP */
@@ -7424,7 +8296,7 @@ static void lpt_disable_clkout_dp(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t reg, tmp;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	reg = (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) ?
 	       SBI_GEN0 : SBI_DBUFF0;
@@ -7443,7 +8315,7 @@ static void lpt_disable_clkout_dp(struct drm_device *dev)
 		intel_sbi_write(dev_priv, SBI_SSCCTL, tmp, SBI_ICLK);
 	}
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void lpt_init_pch_refclk(struct drm_device *dev)
@@ -7483,16 +8355,13 @@ static int ironlake_get_refclk(struct intel_crtc_state *crtc_state)
 	struct drm_device *dev = crtc_state->base.crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_atomic_state *state = crtc_state->base.state;
+	struct drm_connector *connector;
 	struct drm_connector_state *connector_state;
 	struct intel_encoder *encoder;
 	int num_connectors = 0, i;
 	bool is_lvds = false;
 
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i])
-			continue;
-
-		connector_state = state->connector_states[i];
+	for_each_connector_in_state(state, connector, connector_state, i) {
 		if (connector_state->crtc != crtc_state->base.crtc)
 			continue;
 
@@ -7746,17 +8615,14 @@ static uint32_t ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_atomic_state *state = crtc_state->base.state;
+	struct drm_connector *connector;
 	struct drm_connector_state *connector_state;
 	struct intel_encoder *encoder;
 	uint32_t dpll;
 	int factor, num_connectors = 0, i;
 	bool is_lvds = false, is_sdvo = false;
 
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i])
-			continue;
-
-		connector_state = state->connector_states[i];
+	for_each_connector_in_state(state, connector, connector_state, i) {
 		if (connector_state->crtc != crtc_state->base.crtc)
 			continue;
 
@@ -7846,6 +8712,9 @@ static int ironlake_crtc_compute_clock(struct intel_crtc *crtc,
 	bool is_lvds = false;
 	struct intel_shared_dpll *pll;
 
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
 	is_lvds = intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS);
 
 	WARN(!(HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)),
@@ -7980,14 +8849,28 @@ static void skylake_get_pfit_config(struct intel_crtc *crtc,
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t tmp;
+	struct intel_crtc_scaler_state *scaler_state = &pipe_config->scaler_state;
+	uint32_t ps_ctrl = 0;
+	int id = -1;
+	int i;
 
-	tmp = I915_READ(PS_CTL(crtc->pipe));
+	/* find scaler attached to this pipe */
+	for (i = 0; i < crtc->num_scalers; i++) {
+		ps_ctrl = I915_READ(SKL_PS_CTRL(crtc->pipe, i));
+		if (ps_ctrl & PS_SCALER_EN && !(ps_ctrl & PS_PLANE_SEL_MASK)) {
+			id = i;
+			pipe_config->pch_pfit.enabled = true;
+			pipe_config->pch_pfit.pos = I915_READ(SKL_PS_WIN_POS(crtc->pipe, i));
+			pipe_config->pch_pfit.size = I915_READ(SKL_PS_WIN_SZ(crtc->pipe, i));
+			break;
+		}
+	}
 
-	if (tmp & PS_ENABLE) {
-		pipe_config->pch_pfit.enabled = true;
-		pipe_config->pch_pfit.pos = I915_READ(PS_WIN_POS(crtc->pipe));
-		pipe_config->pch_pfit.size = I915_READ(PS_WIN_SZ(crtc->pipe));
+	scaler_state->scaler_id = id;
+	if (id >= 0) {
+		scaler_state->scaler_users |= (1 << SKL_CRTC_INDEX);
+	} else {
+		scaler_state->scaler_users &= ~(1 << SKL_CRTC_INDEX);
 	}
 }
 
@@ -8472,6 +9355,23 @@ void hsw_disable_pc8(struct drm_i915_private *dev_priv)
 	intel_prepare_ddi(dev);
 }
 
+static void broxton_modeset_global_resources(struct drm_atomic_state *old_state)
+{
+	struct drm_device *dev = old_state->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int max_pixclk = intel_mode_max_pixclk(dev, NULL);
+	int req_cdclk;
+
+	/* see the comment in valleyview_modeset_global_resources */
+	if (WARN_ON(max_pixclk < 0))
+		return;
+
+	req_cdclk = broxton_calc_cdclk(dev_priv, max_pixclk);
+
+	if (req_cdclk != dev_priv->cdclk_freq)
+		broxton_set_cdclk(dev, req_cdclk);
+}
+
 static int haswell_crtc_compute_clock(struct intel_crtc *crtc,
 				      struct intel_crtc_state *crtc_state)
 {
@@ -8483,6 +9383,28 @@ static int haswell_crtc_compute_clock(struct intel_crtc *crtc,
 	return 0;
 }
 
+static void bxt_get_ddi_pll(struct drm_i915_private *dev_priv,
+				enum port port,
+				struct intel_crtc_state *pipe_config)
+{
+	switch (port) {
+	case PORT_A:
+		pipe_config->ddi_pll_sel = SKL_DPLL0;
+		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL1;
+		break;
+	case PORT_B:
+		pipe_config->ddi_pll_sel = SKL_DPLL1;
+		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL2;
+		break;
+	case PORT_C:
+		pipe_config->ddi_pll_sel = SKL_DPLL2;
+		pipe_config->shared_dpll = DPLL_ID_SKL_DPLL3;
+		break;
+	default:
+		DRM_ERROR("Incorrect port type\n");
+	}
+}
+
 static void skylake_get_ddi_pll(struct drm_i915_private *dev_priv,
 				enum port port,
 				struct intel_crtc_state *pipe_config)
@@ -8545,6 +9467,8 @@ static void haswell_get_ddi_port_state(struct intel_crtc *crtc,
 
 	if (IS_SKYLAKE(dev))
 		skylake_get_ddi_pll(dev_priv, port, pipe_config);
+	else if (IS_BROXTON(dev))
+		bxt_get_ddi_pll(dev_priv, port, pipe_config);
 	else
 		haswell_get_ddi_pll(dev_priv, port, pipe_config);
 
@@ -8621,12 +9545,24 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 
 	intel_get_pipe_timings(crtc, pipe_config);
 
+	if (INTEL_INFO(dev)->gen >= 9) {
+		skl_init_scalers(dev, crtc, pipe_config);
+	}
+
 	pfit_domain = POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe);
+
+	if (INTEL_INFO(dev)->gen >= 9) {
+		pipe_config->scaler_state.scaler_id = -1;
+		pipe_config->scaler_state.scaler_users &= ~(1 << SKL_CRTC_INDEX);
+	}
+
 	if (intel_display_power_is_enabled(dev_priv, pfit_domain)) {
-		if (IS_SKYLAKE(dev))
+		if (INTEL_INFO(dev)->gen == 9)
 			skylake_get_pfit_config(crtc, pipe_config);
-		else
+		else if (INTEL_INFO(dev)->gen < 9)
 			ironlake_get_pfit_config(crtc, pipe_config);
+		else
+			MISSING_CASE(INTEL_INFO(dev)->gen);
 	}
 
 	if (IS_HASWELL(dev))
@@ -8978,6 +9914,41 @@ mode_fits_in_fbdev(struct drm_device *dev,
 #endif
 }
 
+static int intel_modeset_setup_plane_state(struct drm_atomic_state *state,
+					   struct drm_crtc *crtc,
+					   struct drm_display_mode *mode,
+					   struct drm_framebuffer *fb,
+					   int x, int y)
+{
+	struct drm_plane_state *plane_state;
+	int hdisplay, vdisplay;
+	int ret;
+
+	plane_state = drm_atomic_get_plane_state(state, crtc->primary);
+	if (IS_ERR(plane_state))
+		return PTR_ERR(plane_state);
+
+	if (mode)
+		drm_crtc_get_hv_timing(mode, &hdisplay, &vdisplay);
+	else
+		hdisplay = vdisplay = 0;
+
+	ret = drm_atomic_set_crtc_for_plane(plane_state, fb ? crtc : NULL);
+	if (ret)
+		return ret;
+	drm_atomic_set_fb_for_plane(plane_state, fb);
+	plane_state->crtc_x = 0;
+	plane_state->crtc_y = 0;
+	plane_state->crtc_w = hdisplay;
+	plane_state->crtc_h = vdisplay;
+	plane_state->src_x = x << 16;
+	plane_state->src_y = y << 16;
+	plane_state->src_w = hdisplay << 16;
+	plane_state->src_h = vdisplay << 16;
+
+	return 0;
+}
+
 bool intel_get_load_detect_pipe(struct drm_connector *connector,
 				struct drm_display_mode *mode,
 				struct intel_load_detect_pipe *old,
@@ -8994,6 +9965,7 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	struct drm_mode_config *config = &dev->mode_config;
 	struct drm_atomic_state *state = NULL;
 	struct drm_connector_state *connector_state;
+	struct intel_crtc_state *crtc_state;
 	int ret, i = -1;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
@@ -9070,7 +10042,6 @@ retry:
 
 	intel_crtc = to_intel_crtc(crtc);
 	intel_crtc->new_enabled = true;
-	intel_crtc->new_config = intel_crtc->config;
 	old->dpms_mode = connector->dpms;
 	old->load_detect_temp = true;
 	old->release_fb = NULL;
@@ -9090,6 +10061,14 @@ retry:
 	connector_state->crtc = crtc;
 	connector_state->best_encoder = &intel_encoder->base;
 
+	crtc_state = intel_atomic_get_crtc_state(state, intel_crtc);
+	if (IS_ERR(crtc_state)) {
+		ret = PTR_ERR(crtc_state);
+		goto fail;
+	}
+
+	crtc_state->base.active = crtc_state->base.enable = true;
+
 	if (!mode)
 		mode = &load_detect_mode;
 
@@ -9112,7 +10091,13 @@ retry:
 		goto fail;
 	}
 
-	if (intel_set_mode(crtc, mode, 0, 0, fb, state)) {
+	ret = intel_modeset_setup_plane_state(state, crtc, mode, fb, 0, 0);
+	if (ret)
+		goto fail;
+
+	drm_mode_copy(&crtc_state->base.mode, mode);
+
+	if (intel_set_mode(crtc, state, true)) {
 		DRM_DEBUG_KMS("failed to set mode on load-detect pipe\n");
 		if (old->release_fb)
 			old->release_fb->funcs->destroy(old->release_fb);
@@ -9126,15 +10111,9 @@ retry:
 
  fail:
 	intel_crtc->new_enabled = crtc->state->enable;
-	if (intel_crtc->new_enabled)
-		intel_crtc->new_config = intel_crtc->config;
-	else
-		intel_crtc->new_config = NULL;
 fail_unlock:
-	if (state) {
-		drm_atomic_state_free(state);
-		state = NULL;
-	}
+	drm_atomic_state_free(state);
+	state = NULL;
 
 	if (ret == -EDEADLK) {
 		drm_modeset_backoff(ctx);
@@ -9156,6 +10135,8 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_atomic_state *state;
 	struct drm_connector_state *connector_state;
+	struct intel_crtc_state *crtc_state;
+	int ret;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
 		      connector->base.id, connector->name,
@@ -9172,17 +10153,27 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 		if (IS_ERR(connector_state))
 			goto fail;
 
+		crtc_state = intel_atomic_get_crtc_state(state, intel_crtc);
+		if (IS_ERR(crtc_state))
+			goto fail;
+
 		to_intel_connector(connector)->new_encoder = NULL;
 		intel_encoder->new_crtc = NULL;
 		intel_crtc->new_enabled = false;
-		intel_crtc->new_config = NULL;
 
 		connector_state->best_encoder = NULL;
 		connector_state->crtc = NULL;
 
-		intel_set_mode(crtc, NULL, 0, 0, NULL, state);
+		crtc_state->base.enable = crtc_state->base.active = false;
 
-		drm_atomic_state_free(state);
+		ret = intel_modeset_setup_plane_state(state, crtc, NULL, NULL,
+						      0, 0);
+		if (ret)
+			goto fail;
+
+		ret = intel_set_mode(crtc, state, true);
+		if (ret)
+			goto fail;
 
 		if (old->release_fb) {
 			drm_framebuffer_unregister_private(old->release_fb);
@@ -9466,14 +10457,6 @@ void intel_mark_idle(struct drm_device *dev)
 	intel_runtime_pm_put(dev_priv);
 }
 
-static void intel_crtc_set_state(struct intel_crtc *crtc,
-				 struct intel_crtc_state *crtc_state)
-{
-	kfree(crtc->config);
-	crtc->config = crtc_state;
-	crtc->base.state = &crtc_state->base;
-}
-
 static void intel_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -9490,7 +10473,6 @@ static void intel_crtc_destroy(struct drm_crtc *crtc)
 		kfree(work);
 	}
 
-	intel_crtc_set_state(intel_crtc, NULL);
 	drm_crtc_cleanup(crtc);
 
 	kfree(intel_crtc);
@@ -9907,7 +10889,7 @@ static bool use_mmio_flip(struct intel_engine_cs *ring,
 	else if (i915.enable_execlists)
 		return true;
 	else
-		return ring != i915_gem_request_get_ring(obj->last_read_req);
+		return ring != i915_gem_request_get_ring(obj->last_write_req);
 }
 
 static void skl_do_mmio_flip(struct intel_crtc *intel_crtc)
@@ -9915,23 +10897,34 @@ static void skl_do_mmio_flip(struct intel_crtc *intel_crtc)
 	struct drm_device *dev = intel_crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_framebuffer *fb = intel_crtc->base.primary->fb;
-	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
-	struct drm_i915_gem_object *obj = intel_fb->obj;
 	const enum pipe pipe = intel_crtc->pipe;
 	u32 ctl, stride;
 
 	ctl = I915_READ(PLANE_CTL(pipe, 0));
 	ctl &= ~PLANE_CTL_TILED_MASK;
-	if (obj->tiling_mode == I915_TILING_X)
+	switch (fb->modifier[0]) {
+	case DRM_FORMAT_MOD_NONE:
+		break;
+	case I915_FORMAT_MOD_X_TILED:
 		ctl |= PLANE_CTL_TILED_X;
+		break;
+	case I915_FORMAT_MOD_Y_TILED:
+		ctl |= PLANE_CTL_TILED_Y;
+		break;
+	case I915_FORMAT_MOD_Yf_TILED:
+		ctl |= PLANE_CTL_TILED_YF;
+		break;
+	default:
+		MISSING_CASE(fb->modifier[0]);
+	}
 
 	/*
 	 * The stride is either expressed as a multiple of 64 bytes chunks for
 	 * linear buffers or in number of tiles for tiled buffers.
 	 */
-	stride = fb->pitches[0] >> 6;
-	if (obj->tiling_mode == I915_TILING_X)
-		stride = fb->pitches[0] >> 9; /* X tiles are 512 bytes wide */
+	stride = fb->pitches[0] /
+		 intel_fb_stride_alignment(dev, fb->modifier[0],
+					   fb->pixel_format);
 
 	/*
 	 * Both PLANE_CTL and PLANE_STRIDE are not updated on vblank but on
@@ -9996,22 +10989,19 @@ static void intel_do_mmio_flip(struct intel_crtc *intel_crtc)
 
 static void intel_mmio_flip_work_func(struct work_struct *work)
 {
-	struct intel_crtc *crtc =
-		container_of(work, struct intel_crtc, mmio_flip.work);
-	struct intel_mmio_flip *mmio_flip;
+	struct intel_mmio_flip *mmio_flip =
+		container_of(work, struct intel_mmio_flip, work);
 
-	mmio_flip = &crtc->mmio_flip;
 	if (mmio_flip->req)
 		WARN_ON(__i915_wait_request(mmio_flip->req,
-					    crtc->reset_counter,
-					    false, NULL, NULL) != 0);
+					    mmio_flip->crtc->reset_counter,
+					    false, NULL,
+					    &mmio_flip->i915->rps.mmioflips));
 
-	intel_do_mmio_flip(crtc);
-	if (mmio_flip->req) {
-		mutex_lock(&crtc->base.dev->struct_mutex);
-		i915_gem_request_assign(&mmio_flip->req, NULL);
-		mutex_unlock(&crtc->base.dev->struct_mutex);
-	}
+	intel_do_mmio_flip(mmio_flip->crtc);
+
+	i915_gem_request_unreference__unlocked(mmio_flip->req);
+	kfree(mmio_flip);
 }
 
 static int intel_queue_mmio_flip(struct drm_device *dev,
@@ -10021,12 +11011,18 @@ static int intel_queue_mmio_flip(struct drm_device *dev,
 				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_mmio_flip *mmio_flip;
+
+	mmio_flip = kmalloc(sizeof(*mmio_flip), GFP_KERNEL);
+	if (mmio_flip == NULL)
+		return -ENOMEM;
 
-	i915_gem_request_assign(&intel_crtc->mmio_flip.req,
-				obj->last_write_req);
+	mmio_flip->i915 = to_i915(dev);
+	mmio_flip->req = i915_gem_request_reference(obj->last_write_req);
+	mmio_flip->crtc = to_intel_crtc(crtc);
 
-	schedule_work(&intel_crtc->mmio_flip.work);
+	INIT_WORK(&mmio_flip->work, intel_mmio_flip_work_func);
+	schedule_work(&mmio_flip->work);
 
 	return 0;
 }
@@ -10085,6 +11081,7 @@ void intel_check_page_flip(struct drm_device *dev, int pipe)
 	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);
+	struct intel_unpin_work *work;
 
 	WARN_ON(!in_interrupt());
 
@@ -10092,12 +11089,16 @@ void intel_check_page_flip(struct drm_device *dev, int pipe)
 		return;
 
 	spin_lock(&dev->event_lock);
-	if (intel_crtc->unpin_work && __intel_pageflip_stall_check(dev, crtc)) {
+	work = intel_crtc->unpin_work;
+	if (work != NULL && __intel_pageflip_stall_check(dev, crtc)) {
 		WARN_ONCE(1, "Kicking stuck page flip: queued at %d, now %d\n",
-			 intel_crtc->unpin_work->flip_queued_vblank,
-			 drm_vblank_count(dev, pipe));
+			 work->flip_queued_vblank, drm_vblank_count(dev, pipe));
 		page_flip_completed(intel_crtc);
+		work = NULL;
 	}
+	if (work != NULL &&
+	    drm_vblank_count(dev, pipe) - work->flip_queued_vblank > 1)
+		intel_queue_rps_boost_for_request(dev, work->flip_queued_req);
 	spin_unlock(&dev->event_lock);
 }
 
@@ -10115,6 +11116,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	enum pipe pipe = intel_crtc->pipe;
 	struct intel_unpin_work *work;
 	struct intel_engine_cs *ring;
+	bool mmio_flip;
 	int ret;
 
 	/*
@@ -10205,22 +11207,30 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	} else if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
 		ring = &dev_priv->ring[BCS];
 	} else if (INTEL_INFO(dev)->gen >= 7) {
-		ring = i915_gem_request_get_ring(obj->last_read_req);
+		ring = i915_gem_request_get_ring(obj->last_write_req);
 		if (ring == NULL || ring->id != RCS)
 			ring = &dev_priv->ring[BCS];
 	} else {
 		ring = &dev_priv->ring[RCS];
 	}
 
+	mmio_flip = use_mmio_flip(ring, obj);
+
+	/* When using CS flips, we want to emit semaphores between rings.
+	 * However, when using mmio flips we will create a task to do the
+	 * synchronisation, so all we want here is to pin the framebuffer
+	 * into the display plane and skip any waits.
+	 */
 	ret = intel_pin_and_fence_fb_obj(crtc->primary, fb,
-					 crtc->primary->state, ring);
+					 crtc->primary->state,
+					 mmio_flip ? i915_gem_request_get_ring(obj->last_write_req) : ring);
 	if (ret)
 		goto cleanup_pending;
 
 	work->gtt_offset = intel_plane_obj_offset(to_intel_plane(primary), obj)
 						  + intel_crtc->dspaddr_offset;
 
-	if (use_mmio_flip(ring, obj)) {
+	if (mmio_flip) {
 		ret = intel_queue_mmio_flip(dev, crtc, fb, obj, ring,
 					    page_flip_flags);
 		if (ret)
@@ -10229,6 +11239,12 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 		i915_gem_request_assign(&work->flip_queued_req,
 					obj->last_write_req);
 	} else {
+		if (obj->last_write_req) {
+			ret = i915_gem_check_olr(obj->last_write_req);
+			if (ret)
+				goto cleanup_unpin;
+		}
+
 		ret = dev_priv->display.queue_flip(dev, crtc, fb, obj, ring,
 						   page_flip_flags);
 		if (ret)
@@ -10284,7 +11300,7 @@ out_hang:
 	return ret;
 }
 
-static struct drm_crtc_helper_funcs intel_helper_funcs = {
+static const struct drm_crtc_helper_funcs intel_helper_funcs = {
 	.mode_set_base_atomic = intel_pipe_set_base_atomic,
 	.load_lut = intel_crtc_load_lut,
 	.atomic_begin = intel_begin_crtc_commit,
@@ -10315,11 +11331,6 @@ static void intel_modeset_update_staged_output_state(struct drm_device *dev)
 
 	for_each_intel_crtc(dev, crtc) {
 		crtc->new_enabled = crtc->base.state->enable;
-
-		if (crtc->new_enabled)
-			crtc->new_config = crtc->config;
-		else
-			crtc->new_config = NULL;
 	}
 }
 
@@ -10344,31 +11355,37 @@ static void intel_modeset_update_connector_atomic_state(struct drm_device *dev)
 	}
 }
 
-/**
- * intel_modeset_commit_output_state
- *
- * This function copies the stage display pipe configuration to the real one.
+/* Fixup legacy state after an atomic state swap.
  */
-static void intel_modeset_commit_output_state(struct drm_device *dev)
+static void intel_modeset_fixup_state(struct drm_atomic_state *state)
 {
 	struct intel_crtc *crtc;
 	struct intel_encoder *encoder;
 	struct intel_connector *connector;
 
-	for_each_intel_connector(dev, connector) {
-		connector->base.encoder = &connector->new_encoder->base;
+	for_each_intel_connector(state->dev, connector) {
+		connector->base.encoder = connector->base.state->best_encoder;
+		if (connector->base.encoder)
+			connector->base.encoder->crtc =
+				connector->base.state->crtc;
 	}
 
-	for_each_intel_encoder(dev, encoder) {
-		encoder->base.crtc = &encoder->new_crtc->base;
-	}
+	/* Update crtc of disabled encoders */
+	for_each_intel_encoder(state->dev, encoder) {
+		int num_connectors = 0;
 
-	for_each_intel_crtc(dev, crtc) {
-		crtc->base.state->enable = crtc->new_enabled;
-		crtc->base.enabled = crtc->new_enabled;
+		for_each_intel_connector(state->dev, connector)
+			if (connector->base.encoder == &encoder->base)
+				num_connectors++;
+
+		if (num_connectors == 0)
+			encoder->base.crtc = NULL;
 	}
 
-	intel_modeset_update_connector_atomic_state(dev);
+	for_each_intel_crtc(state->dev, crtc) {
+		crtc->base.enabled = crtc->base.state->enable;
+		crtc->config = to_intel_crtc_state(crtc->base.state);
+	}
 }
 
 static void
@@ -10399,64 +11416,33 @@ connected_sink_compute_bpp(struct intel_connector *connector,
 
 static int
 compute_baseline_pipe_bpp(struct intel_crtc *crtc,
-			  struct drm_framebuffer *fb,
 			  struct intel_crtc_state *pipe_config)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_atomic_state *state;
-	struct intel_connector *connector;
+	struct drm_connector *connector;
+	struct drm_connector_state *connector_state;
 	int bpp, i;
 
-	switch (fb->pixel_format) {
-	case DRM_FORMAT_C8:
-		bpp = 8*3; /* since we go through a colormap */
-		break;
-	case DRM_FORMAT_XRGB1555:
-	case DRM_FORMAT_ARGB1555:
-		/* checked in intel_framebuffer_init already */
-		if (WARN_ON(INTEL_INFO(dev)->gen > 3))
-			return -EINVAL;
-	case DRM_FORMAT_RGB565:
-		bpp = 6*3; /* min is 18bpp */
-		break;
-	case DRM_FORMAT_XBGR8888:
-	case DRM_FORMAT_ABGR8888:
-		/* checked in intel_framebuffer_init already */
-		if (WARN_ON(INTEL_INFO(dev)->gen < 4))
-			return -EINVAL;
-	case DRM_FORMAT_XRGB8888:
-	case DRM_FORMAT_ARGB8888:
-		bpp = 8*3;
-		break;
-	case DRM_FORMAT_XRGB2101010:
-	case DRM_FORMAT_ARGB2101010:
-	case DRM_FORMAT_XBGR2101010:
-	case DRM_FORMAT_ABGR2101010:
-		/* checked in intel_framebuffer_init already */
-		if (WARN_ON(INTEL_INFO(dev)->gen < 4))
-			return -EINVAL;
+	if ((IS_G4X(dev) || IS_VALLEYVIEW(dev)))
 		bpp = 10*3;
-		break;
-	/* TODO: gen4+ supports 16 bpc floating point, too. */
-	default:
-		DRM_DEBUG_KMS("unsupported depth\n");
-		return -EINVAL;
-	}
+	else if (INTEL_INFO(dev)->gen >= 5)
+		bpp = 12*3;
+	else
+		bpp = 8*3;
+
 
 	pipe_config->pipe_bpp = bpp;
 
 	state = pipe_config->base.state;
 
 	/* Clamp display bpp to EDID value */
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i])
-			continue;
-
-		connector = to_intel_connector(state->connectors[i]);
-		if (state->connector_states[i]->crtc != &crtc->base)
+	for_each_connector_in_state(state, connector, connector_state, i) {
+		if (connector_state->crtc != &crtc->base)
 			continue;
 
-		connected_sink_compute_bpp(connector, pipe_config);
+		connected_sink_compute_bpp(to_intel_connector(connector),
+					   pipe_config);
 	}
 
 	return bpp;
@@ -10477,8 +11463,14 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 				   struct intel_crtc_state *pipe_config,
 				   const char *context)
 {
-	DRM_DEBUG_KMS("[CRTC:%d]%s config for pipe %c\n", crtc->base.base.id,
-		      context, pipe_name(crtc->pipe));
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_plane *plane;
+	struct intel_plane *intel_plane;
+	struct intel_plane_state *state;
+	struct drm_framebuffer *fb;
+
+	DRM_DEBUG_KMS("[CRTC:%d]%s config %p for pipe %c\n", crtc->base.base.id,
+		      context, pipe_config, pipe_name(crtc->pipe));
 
 	DRM_DEBUG_KMS("cpu_transcoder: %c\n", transcoder_name(pipe_config->cpu_transcoder));
 	DRM_DEBUG_KMS("pipe bpp: %i, dithering: %i\n",
@@ -10515,6 +11507,10 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 	DRM_DEBUG_KMS("port clock: %d\n", pipe_config->port_clock);
 	DRM_DEBUG_KMS("pipe src size: %dx%d\n",
 		      pipe_config->pipe_src_w, pipe_config->pipe_src_h);
+	DRM_DEBUG_KMS("num_scalers: %d, scaler_users: 0x%x, scaler_id: %d\n",
+		      crtc->num_scalers,
+		      pipe_config->scaler_state.scaler_users,
+		      pipe_config->scaler_state.scaler_id);
 	DRM_DEBUG_KMS("gmch pfit: control: 0x%08x, ratios: 0x%08x, lvds border: 0x%08x\n",
 		      pipe_config->gmch_pfit.control,
 		      pipe_config->gmch_pfit.pgm_ratios,
@@ -10525,6 +11521,73 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 		      pipe_config->pch_pfit.enabled ? "enabled" : "disabled");
 	DRM_DEBUG_KMS("ips: %i\n", pipe_config->ips_enabled);
 	DRM_DEBUG_KMS("double wide: %i\n", pipe_config->double_wide);
+
+	if (IS_BROXTON(dev)) {
+		DRM_DEBUG_KMS("ddi_pll_sel: %u; dpll_hw_state: ebb0: 0x%x, "
+			      "pll0: 0x%x, pll1: 0x%x, pll2: 0x%x, pll3: 0x%x, "
+			      "pll6: 0x%x, pll8: 0x%x, pcsdw12: 0x%x\n",
+			      pipe_config->ddi_pll_sel,
+			      pipe_config->dpll_hw_state.ebb0,
+			      pipe_config->dpll_hw_state.pll0,
+			      pipe_config->dpll_hw_state.pll1,
+			      pipe_config->dpll_hw_state.pll2,
+			      pipe_config->dpll_hw_state.pll3,
+			      pipe_config->dpll_hw_state.pll6,
+			      pipe_config->dpll_hw_state.pll8,
+			      pipe_config->dpll_hw_state.pcsdw12);
+	} else if (IS_SKYLAKE(dev)) {
+		DRM_DEBUG_KMS("ddi_pll_sel: %u; dpll_hw_state: "
+			      "ctrl1: 0x%x, cfgcr1: 0x%x, cfgcr2: 0x%x\n",
+			      pipe_config->ddi_pll_sel,
+			      pipe_config->dpll_hw_state.ctrl1,
+			      pipe_config->dpll_hw_state.cfgcr1,
+			      pipe_config->dpll_hw_state.cfgcr2);
+	} else if (HAS_DDI(dev)) {
+		DRM_DEBUG_KMS("ddi_pll_sel: %u; dpll_hw_state: wrpll: 0x%x\n",
+			      pipe_config->ddi_pll_sel,
+			      pipe_config->dpll_hw_state.wrpll);
+	} else {
+		DRM_DEBUG_KMS("dpll_hw_state: dpll: 0x%x, dpll_md: 0x%x, "
+			      "fp0: 0x%x, fp1: 0x%x\n",
+			      pipe_config->dpll_hw_state.dpll,
+			      pipe_config->dpll_hw_state.dpll_md,
+			      pipe_config->dpll_hw_state.fp0,
+			      pipe_config->dpll_hw_state.fp1);
+	}
+
+	DRM_DEBUG_KMS("planes on this crtc\n");
+	list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
+		intel_plane = to_intel_plane(plane);
+		if (intel_plane->pipe != crtc->pipe)
+			continue;
+
+		state = to_intel_plane_state(plane->state);
+		fb = state->base.fb;
+		if (!fb) {
+			DRM_DEBUG_KMS("%s PLANE:%d plane: %u.%u idx: %d "
+				"disabled, scaler_id = %d\n",
+				plane->type == DRM_PLANE_TYPE_CURSOR ? "CURSOR" : "STANDARD",
+				plane->base.id, intel_plane->pipe,
+				(crtc->base.primary == plane) ? 0 : intel_plane->plane + 1,
+				drm_plane_index(plane), state->scaler_id);
+			continue;
+		}
+
+		DRM_DEBUG_KMS("%s PLANE:%d plane: %u.%u idx: %d enabled",
+			plane->type == DRM_PLANE_TYPE_CURSOR ? "CURSOR" : "STANDARD",
+			plane->base.id, intel_plane->pipe,
+			crtc->base.primary == plane ? 0 : intel_plane->plane + 1,
+			drm_plane_index(plane));
+		DRM_DEBUG_KMS("\tFB:%d, fb = %ux%u format = 0x%x",
+			fb->base.id, fb->width, fb->height, fb->pixel_format);
+		DRM_DEBUG_KMS("\tscaler:%d src (%u, %u) %ux%u dst (%u, %u) %ux%u\n",
+			state->scaler_id,
+			state->src.x1 >> 16, state->src.y1 >> 16,
+			drm_rect_width(&state->src) >> 16,
+			drm_rect_height(&state->src) >> 16,
+			state->dst.x1, state->dst.y1,
+			drm_rect_width(&state->dst), drm_rect_height(&state->dst));
+	}
 }
 
 static bool encoders_cloneable(const struct intel_encoder *a,
@@ -10535,16 +11598,21 @@ static bool encoders_cloneable(const struct intel_encoder *a,
 			  b->cloneable & (1 << a->type));
 }
 
-static bool check_single_encoder_cloning(struct intel_crtc *crtc,
+static bool check_single_encoder_cloning(struct drm_atomic_state *state,
+					 struct intel_crtc *crtc,
 					 struct intel_encoder *encoder)
 {
-	struct drm_device *dev = crtc->base.dev;
 	struct intel_encoder *source_encoder;
+	struct drm_connector *connector;
+	struct drm_connector_state *connector_state;
+	int i;
 
-	for_each_intel_encoder(dev, source_encoder) {
-		if (source_encoder->new_crtc != crtc)
+	for_each_connector_in_state(state, connector, connector_state, i) {
+		if (connector_state->crtc != &crtc->base)
 			continue;
 
+		source_encoder =
+			to_intel_encoder(connector_state->best_encoder);
 		if (!encoders_cloneable(encoder, source_encoder))
 			return false;
 	}
@@ -10552,39 +11620,47 @@ static bool check_single_encoder_cloning(struct intel_crtc *crtc,
 	return true;
 }
 
-static bool check_encoder_cloning(struct intel_crtc *crtc)
+static bool check_encoder_cloning(struct drm_atomic_state *state,
+				  struct intel_crtc *crtc)
 {
-	struct drm_device *dev = crtc->base.dev;
 	struct intel_encoder *encoder;
+	struct drm_connector *connector;
+	struct drm_connector_state *connector_state;
+	int i;
 
-	for_each_intel_encoder(dev, encoder) {
-		if (encoder->new_crtc != crtc)
+	for_each_connector_in_state(state, connector, connector_state, i) {
+		if (connector_state->crtc != &crtc->base)
 			continue;
 
-		if (!check_single_encoder_cloning(crtc, encoder))
+		encoder = to_intel_encoder(connector_state->best_encoder);
+		if (!check_single_encoder_cloning(state, crtc, encoder))
 			return false;
 	}
 
 	return true;
 }
 
-static bool check_digital_port_conflicts(struct drm_device *dev)
+static bool check_digital_port_conflicts(struct drm_atomic_state *state)
 {
-	struct intel_connector *connector;
+	struct drm_device *dev = state->dev;
+	struct intel_encoder *encoder;
+	struct drm_connector *connector;
+	struct drm_connector_state *connector_state;
 	unsigned int used_ports = 0;
+	int i;
 
 	/*
 	 * Walk the connector list instead of the encoder
 	 * list to detect the problem on ddi platforms
 	 * where there's just one encoder per digital port.
 	 */
-	for_each_intel_connector(dev, connector) {
-		struct intel_encoder *encoder = connector->new_encoder;
-
-		if (!encoder)
+	for_each_connector_in_state(state, connector, connector_state, i) {
+		if (!connector_state->best_encoder)
 			continue;
 
-		WARN_ON(!encoder->new_crtc);
+		encoder = to_intel_encoder(connector_state->best_encoder);
+
+		WARN_ON(!connector_state->crtc);
 
 		switch (encoder->type) {
 			unsigned int port_mask;
@@ -10613,51 +11689,57 @@ static void
 clear_intel_crtc_state(struct intel_crtc_state *crtc_state)
 {
 	struct drm_crtc_state tmp_state;
+	struct intel_crtc_scaler_state scaler_state;
+	struct intel_dpll_hw_state dpll_hw_state;
+	enum intel_dpll_id shared_dpll;
+	uint32_t ddi_pll_sel;
+
+	/* FIXME: before the switch to atomic started, a new pipe_config was
+	 * kzalloc'd. Code that depends on any field being zero should be
+	 * fixed, so that the crtc_state can be safely duplicated. For now,
+	 * only fields that are know to not cause problems are preserved. */
 
-	/* Clear only the intel specific part of the crtc state */
 	tmp_state = crtc_state->base;
+	scaler_state = crtc_state->scaler_state;
+	shared_dpll = crtc_state->shared_dpll;
+	dpll_hw_state = crtc_state->dpll_hw_state;
+	ddi_pll_sel = crtc_state->ddi_pll_sel;
+
 	memset(crtc_state, 0, sizeof *crtc_state);
+
 	crtc_state->base = tmp_state;
+	crtc_state->scaler_state = scaler_state;
+	crtc_state->shared_dpll = shared_dpll;
+	crtc_state->dpll_hw_state = dpll_hw_state;
+	crtc_state->ddi_pll_sel = ddi_pll_sel;
 }
 
-static struct intel_crtc_state *
+static int
 intel_modeset_pipe_config(struct drm_crtc *crtc,
-			  struct drm_framebuffer *fb,
-			  struct drm_display_mode *mode,
-			  struct drm_atomic_state *state)
+			  struct drm_atomic_state *state,
+			  struct intel_crtc_state *pipe_config)
 {
-	struct drm_device *dev = crtc->dev;
 	struct intel_encoder *encoder;
-	struct intel_connector *connector;
+	struct drm_connector *connector;
 	struct drm_connector_state *connector_state;
-	struct intel_crtc_state *pipe_config;
-	int plane_bpp, ret = -EINVAL;
+	int base_bpp, ret = -EINVAL;
 	int i;
 	bool retry = true;
 
-	if (!check_encoder_cloning(to_intel_crtc(crtc))) {
+	if (!check_encoder_cloning(state, to_intel_crtc(crtc))) {
 		DRM_DEBUG_KMS("rejecting invalid cloning configuration\n");
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 	}
 
-	if (!check_digital_port_conflicts(dev)) {
+	if (!check_digital_port_conflicts(state)) {
 		DRM_DEBUG_KMS("rejecting conflicting digital port configuration\n");
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 	}
 
-	pipe_config = intel_atomic_get_crtc_state(state, to_intel_crtc(crtc));
-	if (IS_ERR(pipe_config))
-		return pipe_config;
-
 	clear_intel_crtc_state(pipe_config);
 
-	pipe_config->base.crtc = crtc;
-	drm_mode_copy(&pipe_config->base.adjusted_mode, mode);
-	drm_mode_copy(&pipe_config->base.mode, mode);
-
 	pipe_config->cpu_transcoder =
 		(enum transcoder) to_intel_crtc(crtc)->pipe;
-	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
 
 	/*
 	 * Sanitize sync polarity flags based on requested ones. If neither
@@ -10676,9 +11758,9 @@ intel_modeset_pipe_config(struct drm_crtc *crtc,
 	 * plane pixel format and any sink constraints into account. Returns the
 	 * source plane bpp so that dithering can be selected on mismatches
 	 * after encoders and crtc also have had their say. */
-	plane_bpp = compute_baseline_pipe_bpp(to_intel_crtc(crtc),
-					      fb, pipe_config);
-	if (plane_bpp < 0)
+	base_bpp = compute_baseline_pipe_bpp(to_intel_crtc(crtc),
+					     pipe_config);
+	if (base_bpp < 0)
 		goto fail;
 
 	/*
@@ -10706,12 +11788,7 @@ encoder_retry:
 	 * adjust it according to limitations or connector properties, and also
 	 * a chance to reject the mode entirely.
 	 */
-	for (i = 0; i < state->num_connector; i++) {
-		connector = to_intel_connector(state->connectors[i]);
-		if (!connector)
-			continue;
-
-		connector_state = state->connector_states[i];
+	for_each_connector_in_state(state, connector, connector_state, i) {
 		if (connector_state->crtc != crtc)
 			continue;
 
@@ -10746,101 +11823,13 @@ encoder_retry:
 		goto encoder_retry;
 	}
 
-	pipe_config->dither = pipe_config->pipe_bpp != plane_bpp;
+	pipe_config->dither = pipe_config->pipe_bpp != base_bpp;
 	DRM_DEBUG_KMS("plane bpp: %i, pipe bpp: %i, dithering: %i\n",
-		      plane_bpp, pipe_config->pipe_bpp, pipe_config->dither);
+		      base_bpp, pipe_config->pipe_bpp, pipe_config->dither);
 
-	return pipe_config;
+	return 0;
 fail:
-	return ERR_PTR(ret);
-}
-
-/* Computes which crtcs are affected and sets the relevant bits in the mask. For
- * simplicity we use the crtc's pipe number (because it's easier to obtain). */
-static void
-intel_modeset_affected_pipes(struct drm_crtc *crtc, unsigned *modeset_pipes,
-			     unsigned *prepare_pipes, unsigned *disable_pipes)
-{
-	struct intel_crtc *intel_crtc;
-	struct drm_device *dev = crtc->dev;
-	struct intel_encoder *encoder;
-	struct intel_connector *connector;
-	struct drm_crtc *tmp_crtc;
-
-	*disable_pipes = *modeset_pipes = *prepare_pipes = 0;
-
-	/* Check which crtcs have changed outputs connected to them, these need
-	 * to be part of the prepare_pipes mask. We don't (yet) support global
-	 * modeset across multiple crtcs, so modeset_pipes will only have one
-	 * bit set at most. */
-	for_each_intel_connector(dev, connector) {
-		if (connector->base.encoder == &connector->new_encoder->base)
-			continue;
-
-		if (connector->base.encoder) {
-			tmp_crtc = connector->base.encoder->crtc;
-
-			*prepare_pipes |= 1 << to_intel_crtc(tmp_crtc)->pipe;
-		}
-
-		if (connector->new_encoder)
-			*prepare_pipes |=
-				1 << connector->new_encoder->new_crtc->pipe;
-	}
-
-	for_each_intel_encoder(dev, encoder) {
-		if (encoder->base.crtc == &encoder->new_crtc->base)
-			continue;
-
-		if (encoder->base.crtc) {
-			tmp_crtc = encoder->base.crtc;
-
-			*prepare_pipes |= 1 << to_intel_crtc(tmp_crtc)->pipe;
-		}
-
-		if (encoder->new_crtc)
-			*prepare_pipes |= 1 << encoder->new_crtc->pipe;
-	}
-
-	/* Check for pipes that will be enabled/disabled ... */
-	for_each_intel_crtc(dev, intel_crtc) {
-		if (intel_crtc->base.state->enable == intel_crtc->new_enabled)
-			continue;
-
-		if (!intel_crtc->new_enabled)
-			*disable_pipes |= 1 << intel_crtc->pipe;
-		else
-			*prepare_pipes |= 1 << intel_crtc->pipe;
-	}
-
-
-	/* set_mode is also used to update properties on life display pipes. */
-	intel_crtc = to_intel_crtc(crtc);
-	if (intel_crtc->new_enabled)
-		*prepare_pipes |= 1 << intel_crtc->pipe;
-
-	/*
-	 * For simplicity do a full modeset on any pipe where the output routing
-	 * changed. We could be more clever, but that would require us to be
-	 * more careful with calling the relevant encoder->mode_set functions.
-	 */
-	if (*prepare_pipes)
-		*modeset_pipes = *prepare_pipes;
-
-	/* ... and mask these out. */
-	*modeset_pipes &= ~(*disable_pipes);
-	*prepare_pipes &= ~(*disable_pipes);
-
-	/*
-	 * HACK: We don't (yet) fully support global modesets. intel_set_config
-	 * obies this rule, but the modeset restore mode of
-	 * intel_modeset_setup_hw_state does not.
-	 */
-	*modeset_pipes &= 1 << intel_crtc->pipe;
-	*prepare_pipes &= 1 << intel_crtc->pipe;
-
-	DRM_DEBUG_KMS("set mode pipe masks: modeset: %x, prepare: %x, disable: %x\n",
-		      *modeset_pipes, *prepare_pipes, *disable_pipes);
+	return ret;
 }
 
 static bool intel_crtc_in_use(struct drm_crtc *crtc)
@@ -10855,13 +11844,22 @@ static bool intel_crtc_in_use(struct drm_crtc *crtc)
 	return false;
 }
 
+static bool
+needs_modeset(struct drm_crtc_state *state)
+{
+	return state->mode_changed || state->active_changed;
+}
+
 static void
-intel_modeset_update_state(struct drm_device *dev, unsigned prepare_pipes)
+intel_modeset_update_state(struct drm_atomic_state *state)
 {
+	struct drm_device *dev = state->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_encoder *intel_encoder;
-	struct intel_crtc *intel_crtc;
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *crtc_state;
 	struct drm_connector *connector;
+	int i;
 
 	intel_shared_dpll_commit(dev_priv);
 
@@ -10869,39 +11867,47 @@ intel_modeset_update_state(struct drm_device *dev, unsigned prepare_pipes)
 		if (!intel_encoder->base.crtc)
 			continue;
 
-		intel_crtc = to_intel_crtc(intel_encoder->base.crtc);
+		for_each_crtc_in_state(state, crtc, crtc_state, i) {
+			if (crtc != intel_encoder->base.crtc)
+				continue;
 
-		if (prepare_pipes & (1 << intel_crtc->pipe))
-			intel_encoder->connectors_active = false;
+			if (crtc_state->enable && needs_modeset(crtc_state))
+				intel_encoder->connectors_active = false;
+
+			break;
+		}
 	}
 
-	intel_modeset_commit_output_state(dev);
+	drm_atomic_helper_swap_state(state->dev, state);
+	intel_modeset_fixup_state(state);
 
 	/* Double check state. */
-	for_each_intel_crtc(dev, intel_crtc) {
-		WARN_ON(intel_crtc->base.state->enable != intel_crtc_in_use(&intel_crtc->base));
-		WARN_ON(intel_crtc->new_config &&
-			intel_crtc->new_config != intel_crtc->config);
-		WARN_ON(intel_crtc->base.state->enable != !!intel_crtc->new_config);
+	for_each_crtc(dev, crtc) {
+		WARN_ON(crtc->state->enable != intel_crtc_in_use(crtc));
 	}
 
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		if (!connector->encoder || !connector->encoder->crtc)
 			continue;
 
-		intel_crtc = to_intel_crtc(connector->encoder->crtc);
+		for_each_crtc_in_state(state, crtc, crtc_state, i) {
+			if (crtc != connector->encoder->crtc)
+				continue;
 
-		if (prepare_pipes & (1 << intel_crtc->pipe)) {
-			struct drm_property *dpms_property =
-				dev->mode_config.dpms_property;
+			if (crtc->state->enable && needs_modeset(crtc->state)) {
+				struct drm_property *dpms_property =
+					dev->mode_config.dpms_property;
 
-			connector->dpms = DRM_MODE_DPMS_ON;
-			drm_object_property_set_value(&connector->base,
-							 dpms_property,
-							 DRM_MODE_DPMS_ON);
+				connector->dpms = DRM_MODE_DPMS_ON;
+				drm_object_property_set_value(&connector->base,
+								 dpms_property,
+								 DRM_MODE_DPMS_ON);
 
-			intel_encoder = to_intel_encoder(connector->encoder);
-			intel_encoder->connectors_active = true;
+				intel_encoder = to_intel_encoder(connector->encoder);
+				intel_encoder->connectors_active = true;
+			}
+
+			break;
 		}
 	}
 
@@ -11087,6 +12093,8 @@ intel_pipe_config_compare(struct drm_device *dev,
 		PIPE_CONF_CHECK_I(pch_pfit.size);
 	}
 
+	PIPE_CONF_CHECK_I(scaler_state.scaler_id);
+
 	/* BDW+ don't expose a synchronous way to read the state */
 	if (IS_HASWELL(dev))
 		PIPE_CONF_CHECK_I(ips_enabled);
@@ -11428,32 +12436,18 @@ static void update_scanline_offset(struct intel_crtc *crtc)
 
 static struct intel_crtc_state *
 intel_modeset_compute_config(struct drm_crtc *crtc,
-			     struct drm_display_mode *mode,
-			     struct drm_framebuffer *fb,
-			     struct drm_atomic_state *state,
-			     unsigned *modeset_pipes,
-			     unsigned *prepare_pipes,
-			     unsigned *disable_pipes)
+			     struct drm_atomic_state *state)
 {
-	struct drm_device *dev = crtc->dev;
-	struct intel_crtc_state *pipe_config = NULL;
-	struct intel_crtc *intel_crtc;
+	struct intel_crtc_state *pipe_config;
 	int ret = 0;
 
 	ret = drm_atomic_add_affected_connectors(state, crtc);
 	if (ret)
 		return ERR_PTR(ret);
 
-	intel_modeset_affected_pipes(crtc, modeset_pipes,
-				     prepare_pipes, disable_pipes);
-
-	for_each_intel_crtc_masked(dev, *disable_pipes, intel_crtc) {
-		pipe_config = intel_atomic_get_crtc_state(state, intel_crtc);
-		if (IS_ERR(pipe_config))
-			return pipe_config;
-
-		pipe_config->base.enable = false;
-	}
+	ret = drm_atomic_helper_check_modeset(state->dev, state);
+	if (ret)
+		return ERR_PTR(ret);
 
 	/*
 	 * Note this needs changes when we start tracking multiple modes
@@ -11461,43 +12455,76 @@ intel_modeset_compute_config(struct drm_crtc *crtc,
 	 * (i.e. one pipe_config for each crtc) rather than just the one
 	 * for this crtc.
 	 */
-	for_each_intel_crtc_masked(dev, *modeset_pipes, intel_crtc) {
-		/* FIXME: For now we still expect modeset_pipes has at most
-		 * one bit set. */
-		if (WARN_ON(&intel_crtc->base != crtc))
-			continue;
+	pipe_config = intel_atomic_get_crtc_state(state, to_intel_crtc(crtc));
+	if (IS_ERR(pipe_config))
+		return pipe_config;
+
+	if (!pipe_config->base.enable)
+		return pipe_config;
 
-		pipe_config = intel_modeset_pipe_config(crtc, fb, mode, state);
-		if (IS_ERR(pipe_config))
-			return pipe_config;
+	ret = intel_modeset_pipe_config(crtc, state, pipe_config);
+	if (ret)
+		return ERR_PTR(ret);
 
-		intel_dump_pipe_config(to_intel_crtc(crtc), pipe_config,
-				       "[modeset]");
-	}
+	/* Check things that can only be changed through modeset */
+	if (pipe_config->has_audio !=
+	    to_intel_crtc(crtc)->config->has_audio)
+		pipe_config->base.mode_changed = true;
 
-	return intel_atomic_get_crtc_state(state, to_intel_crtc(crtc));;
+	/*
+	 * Note we have an issue here with infoframes: current code
+	 * only updates them on the full mode set path per hw
+	 * requirements.  So here we should be checking for any
+	 * required changes and forcing a mode set.
+	 */
+
+	intel_dump_pipe_config(to_intel_crtc(crtc), pipe_config,"[modeset]");
+
+	ret = drm_atomic_helper_check_planes(state->dev, state);
+	if (ret)
+		return ERR_PTR(ret);
+
+	return pipe_config;
 }
 
-static int __intel_set_mode_setup_plls(struct drm_device *dev,
-				       unsigned modeset_pipes,
-				       unsigned disable_pipes)
+static int __intel_set_mode_setup_plls(struct drm_atomic_state *state)
 {
+	struct drm_device *dev = state->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	unsigned clear_pipes = modeset_pipes | disable_pipes;
+	unsigned clear_pipes = 0;
 	struct intel_crtc *intel_crtc;
+	struct intel_crtc_state *intel_crtc_state;
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *crtc_state;
 	int ret = 0;
+	int i;
 
 	if (!dev_priv->display.crtc_compute_clock)
 		return 0;
 
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		intel_crtc = to_intel_crtc(crtc);
+		intel_crtc_state = to_intel_crtc_state(crtc_state);
+
+		if (needs_modeset(crtc_state)) {
+			clear_pipes |= 1 << intel_crtc->pipe;
+			intel_crtc_state->shared_dpll = DPLL_ID_PRIVATE;
+		}
+	}
+
 	ret = intel_shared_dpll_start_config(dev_priv, clear_pipes);
 	if (ret)
 		goto done;
 
-	for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) {
-		struct intel_crtc_state *state = intel_crtc->new_config;
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		if (!needs_modeset(crtc_state) || !crtc_state->enable)
+			continue;
+
+		intel_crtc = to_intel_crtc(crtc);
+		intel_crtc_state = to_intel_crtc_state(crtc_state);
+
 		ret = dev_priv->display.crtc_compute_clock(intel_crtc,
-							   state);
+							   intel_crtc_state);
 		if (ret) {
 			intel_shared_dpll_abort_config(dev_priv);
 			goto done;
@@ -11508,35 +12535,11 @@ done:
 	return ret;
 }
 
-static int __intel_set_mode(struct drm_crtc *crtc,
-			    struct drm_display_mode *mode,
-			    int x, int y, struct drm_framebuffer *fb,
-			    struct intel_crtc_state *pipe_config,
-			    unsigned modeset_pipes,
-			    unsigned prepare_pipes,
-			    unsigned disable_pipes)
+/* Code that should eventually be part of atomic_check() */
+static int __intel_set_mode_checks(struct drm_atomic_state *state)
 {
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_display_mode *saved_mode;
-	struct intel_crtc_state *crtc_state_copy = NULL;
-	struct intel_crtc *intel_crtc;
-	int ret = 0;
-
-	saved_mode = kmalloc(sizeof(*saved_mode), GFP_KERNEL);
-	if (!saved_mode)
-		return -ENOMEM;
-
-	crtc_state_copy = kmalloc(sizeof(*crtc_state_copy), GFP_KERNEL);
-	if (!crtc_state_copy) {
-		ret = -ENOMEM;
-		goto done;
-	}
-
-	*saved_mode = crtc->mode;
-
-	if (modeset_pipes)
-		to_intel_crtc(crtc)->new_config = pipe_config;
+	struct drm_device *dev = state->dev;
+	int ret;
 
 	/*
 	 * See if the config requires any additional preparation, e.g.
@@ -11545,23 +12548,48 @@ static int __intel_set_mode(struct drm_crtc *crtc,
 	 * mode set on this crtc.  For other crtcs we need to use the
 	 * adjusted_mode bits in the crtc directly.
 	 */
-	if (IS_VALLEYVIEW(dev)) {
-		valleyview_modeset_global_pipes(dev, &prepare_pipes);
-
-		/* may have added more to prepare_pipes than we should */
-		prepare_pipes &= ~disable_pipes;
+	if (IS_VALLEYVIEW(dev) || IS_BROXTON(dev)) {
+		ret = valleyview_modeset_global_pipes(state);
+		if (ret)
+			return ret;
 	}
 
-	ret = __intel_set_mode_setup_plls(dev, modeset_pipes, disable_pipes);
+	ret = __intel_set_mode_setup_plls(state);
 	if (ret)
-		goto done;
+		return ret;
+
+	return 0;
+}
+
+static int __intel_set_mode(struct drm_crtc *modeset_crtc,
+			    struct intel_crtc_state *pipe_config)
+{
+	struct drm_device *dev = modeset_crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_atomic_state *state = pipe_config->base.state;
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *crtc_state;
+	int ret = 0;
+	int i;
 
-	for_each_intel_crtc_masked(dev, disable_pipes, intel_crtc)
-		intel_crtc_disable(&intel_crtc->base);
+	ret = __intel_set_mode_checks(state);
+	if (ret < 0)
+		return ret;
 
-	for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc) {
-		if (intel_crtc->base.state->enable)
-			dev_priv->display.crtc_disable(&intel_crtc->base);
+	ret = drm_atomic_helper_prepare_planes(dev, state);
+	if (ret)
+		return ret;
+
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		if (!needs_modeset(crtc_state))
+			continue;
+
+		if (!crtc_state->enable) {
+			intel_crtc_disable(crtc);
+		} else if (crtc->state->enable) {
+			intel_crtc_disable_planes(crtc);
+			dev_priv->display.crtc_disable(crtc);
+		}
 	}
 
 	/* crtc->mode is already used by the ->mode_set callbacks, hence we need
@@ -11571,115 +12599,79 @@ static int __intel_set_mode(struct drm_crtc *crtc,
 	 * pipes; here we assume a single modeset_pipe and only track the
 	 * single crtc and mode.
 	 */
-	if (modeset_pipes) {
-		crtc->mode = *mode;
-		/* mode_set/enable/disable functions rely on a correct pipe
-		 * config. */
-		intel_crtc_set_state(to_intel_crtc(crtc), pipe_config);
+	if (pipe_config->base.enable && needs_modeset(&pipe_config->base)) {
+		modeset_crtc->mode = pipe_config->base.mode;
 
 		/*
 		 * Calculate and store various constants which
 		 * are later needed by vblank and swap-completion
 		 * timestamping. They are derived from true hwmode.
 		 */
-		drm_calc_timestamping_constants(crtc,
+		drm_calc_timestamping_constants(modeset_crtc,
 						&pipe_config->base.adjusted_mode);
 	}
 
 	/* Only after disabling all output pipelines that will be changed can we
 	 * update the the output configuration. */
-	intel_modeset_update_state(dev, prepare_pipes);
+	intel_modeset_update_state(state);
 
-	modeset_update_crtc_power_domains(pipe_config->base.state);
+	/* The state has been swaped above, so state actually contains the
+	 * old state now. */
 
-	/* Set up the DPLL and any encoders state that needs to adjust or depend
-	 * on the DPLL.
-	 */
-	for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) {
-		struct drm_plane *primary = intel_crtc->base.primary;
-		int vdisplay, hdisplay;
+	modeset_update_crtc_power_domains(state);
 
-		drm_crtc_get_hv_timing(mode, &hdisplay, &vdisplay);
-		ret = primary->funcs->update_plane(primary, &intel_crtc->base,
-						   fb, 0, 0,
-						   hdisplay, vdisplay,
-						   x << 16, y << 16,
-						   hdisplay << 16, vdisplay << 16);
-	}
+	drm_atomic_helper_commit_planes(dev, state);
 
 	/* Now enable the clocks, plane, pipe, and connectors that we set up. */
-	for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc) {
-		update_scanline_offset(intel_crtc);
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		if (!needs_modeset(crtc->state) || !crtc->state->enable)
+			continue;
+
+		update_scanline_offset(to_intel_crtc(crtc));
 
-		dev_priv->display.crtc_enable(&intel_crtc->base);
+		dev_priv->display.crtc_enable(crtc);
+		intel_crtc_enable_planes(crtc);
 	}
 
 	/* FIXME: add subpixel order */
-done:
-	if (ret && crtc->state->enable)
-		crtc->mode = *saved_mode;
-
-	if (ret == 0 && pipe_config) {
-		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
-		/* The pipe_config will be freed with the atomic state, so
-		 * make a copy. */
-		memcpy(crtc_state_copy, intel_crtc->config,
-		       sizeof *crtc_state_copy);
-		intel_crtc->config = crtc_state_copy;
-		intel_crtc->base.state = &crtc_state_copy->base;
+	drm_atomic_helper_cleanup_planes(dev, state);
 
-		if (modeset_pipes)
-			intel_crtc->new_config = intel_crtc->config;
-	} else {
-		kfree(crtc_state_copy);
-	}
+	drm_atomic_state_free(state);
 
-	kfree(saved_mode);
-	return ret;
+	return 0;
 }
 
-static int intel_set_mode_pipes(struct drm_crtc *crtc,
-				struct drm_display_mode *mode,
-				int x, int y, struct drm_framebuffer *fb,
-				struct intel_crtc_state *pipe_config,
-				unsigned modeset_pipes,
-				unsigned prepare_pipes,
-				unsigned disable_pipes)
+static int intel_set_mode_with_config(struct drm_crtc *crtc,
+				      struct intel_crtc_state *pipe_config,
+				      bool force_restore)
 {
 	int ret;
 
-	ret = __intel_set_mode(crtc, mode, x, y, fb, pipe_config, modeset_pipes,
-			       prepare_pipes, disable_pipes);
+	ret = __intel_set_mode(crtc, pipe_config);
 
-	if (ret == 0)
+	if (ret == 0 && force_restore) {
+		intel_modeset_update_staged_output_state(crtc->dev);
 		intel_modeset_check_state(crtc->dev);
+	}
 
 	return ret;
 }
 
 static int intel_set_mode(struct drm_crtc *crtc,
-			  struct drm_display_mode *mode,
-			  int x, int y, struct drm_framebuffer *fb,
-			  struct drm_atomic_state *state)
+			  struct drm_atomic_state *state,
+			  bool force_restore)
 {
 	struct intel_crtc_state *pipe_config;
-	unsigned modeset_pipes, prepare_pipes, disable_pipes;
 	int ret = 0;
 
-	pipe_config = intel_modeset_compute_config(crtc, mode, fb, state,
-						   &modeset_pipes,
-						   &prepare_pipes,
-						   &disable_pipes);
-
+	pipe_config = intel_modeset_compute_config(crtc, state);
 	if (IS_ERR(pipe_config)) {
 		ret = PTR_ERR(pipe_config);
 		goto out;
 	}
 
-	ret = intel_set_mode_pipes(crtc, mode, x, y, fb, pipe_config,
-				   modeset_pipes, prepare_pipes,
-				   disable_pipes);
+	ret = intel_set_mode_with_config(crtc, pipe_config, force_restore);
 	if (ret)
 		goto out;
 
@@ -11694,6 +12686,8 @@ void intel_crtc_restore_mode(struct drm_crtc *crtc)
 	struct intel_encoder *encoder;
 	struct intel_connector *connector;
 	struct drm_connector_state *connector_state;
+	struct intel_crtc_state *crtc_state;
+	int ret;
 
 	state = drm_atomic_state_alloc(dev);
 	if (!state) {
@@ -11731,186 +12725,52 @@ void intel_crtc_restore_mode(struct drm_crtc *crtc)
 		}
 	}
 
-	intel_set_mode(crtc, &crtc->mode, crtc->x, crtc->y, crtc->primary->fb,
-		       state);
-
-	drm_atomic_state_free(state);
-}
-
-#undef for_each_intel_crtc_masked
-
-static void intel_set_config_free(struct intel_set_config *config)
-{
-	if (!config)
+	crtc_state = intel_atomic_get_crtc_state(state, to_intel_crtc(crtc));
+	if (IS_ERR(crtc_state)) {
+		DRM_DEBUG_KMS("Failed to add [CRTC:%d] to state: %ld\n",
+			      crtc->base.id, PTR_ERR(crtc_state));
+		drm_atomic_state_free(state);
 		return;
-
-	kfree(config->save_connector_encoders);
-	kfree(config->save_encoder_crtcs);
-	kfree(config->save_crtc_enabled);
-	kfree(config);
-}
-
-static int intel_set_config_save_state(struct drm_device *dev,
-				       struct intel_set_config *config)
-{
-	struct drm_crtc *crtc;
-	struct drm_encoder *encoder;
-	struct drm_connector *connector;
-	int count;
-
-	config->save_crtc_enabled =
-		kcalloc(dev->mode_config.num_crtc,
-			sizeof(bool), GFP_KERNEL);
-	if (!config->save_crtc_enabled)
-		return -ENOMEM;
-
-	config->save_encoder_crtcs =
-		kcalloc(dev->mode_config.num_encoder,
-			sizeof(struct drm_crtc *), GFP_KERNEL);
-	if (!config->save_encoder_crtcs)
-		return -ENOMEM;
-
-	config->save_connector_encoders =
-		kcalloc(dev->mode_config.num_connector,
-			sizeof(struct drm_encoder *), GFP_KERNEL);
-	if (!config->save_connector_encoders)
-		return -ENOMEM;
-
-	/* Copy data. Note that driver private data is not affected.
-	 * Should anything bad happen only the expected state is
-	 * restored, not the drivers personal bookkeeping.
-	 */
-	count = 0;
-	for_each_crtc(dev, crtc) {
-		config->save_crtc_enabled[count++] = crtc->state->enable;
 	}
 
-	count = 0;
-	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
-		config->save_encoder_crtcs[count++] = encoder->crtc;
-	}
+	crtc_state->base.active = crtc_state->base.enable =
+		to_intel_crtc(crtc)->new_enabled;
 
-	count = 0;
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		config->save_connector_encoders[count++] = connector->encoder;
-	}
-
-	return 0;
-}
+	drm_mode_copy(&crtc_state->base.mode, &crtc->mode);
 
-static void intel_set_config_restore_state(struct drm_device *dev,
-					   struct intel_set_config *config)
-{
-	struct intel_crtc *crtc;
-	struct intel_encoder *encoder;
-	struct intel_connector *connector;
-	int count;
+	intel_modeset_setup_plane_state(state, crtc, &crtc->mode,
+					crtc->primary->fb, crtc->x, crtc->y);
 
-	count = 0;
-	for_each_intel_crtc(dev, crtc) {
-		crtc->new_enabled = config->save_crtc_enabled[count++];
-
-		if (crtc->new_enabled)
-			crtc->new_config = crtc->config;
-		else
-			crtc->new_config = NULL;
-	}
-
-	count = 0;
-	for_each_intel_encoder(dev, encoder) {
-		encoder->new_crtc =
-			to_intel_crtc(config->save_encoder_crtcs[count++]);
-	}
-
-	count = 0;
-	for_each_intel_connector(dev, connector) {
-		connector->new_encoder =
-			to_intel_encoder(config->save_connector_encoders[count++]);
-	}
+	ret = intel_set_mode(crtc, state, false);
+	if (ret)
+		drm_atomic_state_free(state);
 }
 
-static bool
-is_crtc_connector_off(struct drm_mode_set *set)
-{
-	int i;
-
-	if (set->num_connectors == 0)
-		return false;
+#undef for_each_intel_crtc_masked
 
-	if (WARN_ON(set->connectors == NULL))
-		return false;
+static bool intel_connector_in_mode_set(struct intel_connector *connector,
+					struct drm_mode_set *set)
+{
+	int ro;
 
-	for (i = 0; i < set->num_connectors; i++)
-		if (set->connectors[i]->encoder &&
-		    set->connectors[i]->encoder->crtc == set->crtc &&
-		    set->connectors[i]->dpms != DRM_MODE_DPMS_ON)
+	for (ro = 0; ro < set->num_connectors; ro++)
+		if (set->connectors[ro] == &connector->base)
 			return true;
 
 	return false;
 }
 
-static void
-intel_set_config_compute_mode_changes(struct drm_mode_set *set,
-				      struct intel_set_config *config)
-{
-
-	/* We should be able to check here if the fb has the same properties
-	 * and then just flip_or_move it */
-	if (is_crtc_connector_off(set)) {
-		config->mode_changed = true;
-	} else if (set->crtc->primary->fb != set->fb) {
-		/*
-		 * If we have no fb, we can only flip as long as the crtc is
-		 * active, otherwise we need a full mode set.  The crtc may
-		 * be active if we've only disabled the primary plane, or
-		 * in fastboot situations.
-		 */
-		if (set->crtc->primary->fb == NULL) {
-			struct intel_crtc *intel_crtc =
-				to_intel_crtc(set->crtc);
-
-			if (intel_crtc->active) {
-				DRM_DEBUG_KMS("crtc has no fb, will flip\n");
-				config->fb_changed = true;
-			} else {
-				DRM_DEBUG_KMS("inactive crtc, full mode set\n");
-				config->mode_changed = true;
-			}
-		} else if (set->fb == NULL) {
-			config->mode_changed = true;
-		} else if (set->fb->pixel_format !=
-			   set->crtc->primary->fb->pixel_format) {
-			config->mode_changed = true;
-		} else {
-			config->fb_changed = true;
-		}
-	}
-
-	if (set->fb && (set->x != set->crtc->x || set->y != set->crtc->y))
-		config->fb_changed = true;
-
-	if (set->mode && !drm_mode_equal(set->mode, &set->crtc->mode)) {
-		DRM_DEBUG_KMS("modes are different, full mode set\n");
-		drm_mode_debug_printmodeline(&set->crtc->mode);
-		drm_mode_debug_printmodeline(set->mode);
-		config->mode_changed = true;
-	}
-
-	DRM_DEBUG_KMS("computed changes for [CRTC:%d], mode_changed=%d, fb_changed=%d\n",
-			set->crtc->base.id, config->mode_changed, config->fb_changed);
-}
-
 static int
 intel_modeset_stage_output_state(struct drm_device *dev,
 				 struct drm_mode_set *set,
-				 struct intel_set_config *config,
 				 struct drm_atomic_state *state)
 {
 	struct intel_connector *connector;
+	struct drm_connector *drm_connector;
 	struct drm_connector_state *connector_state;
-	struct intel_encoder *encoder;
-	struct intel_crtc *crtc;
-	int ro;
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *crtc_state;
+	int i, ret;
 
 	/* The upper layers ensure that we either disable a crtc or have a list
 	 * of connectors. For paranoia, double-check this. */
@@ -11918,169 +12778,129 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 	WARN_ON(set->fb && (set->num_connectors == 0));
 
 	for_each_intel_connector(dev, connector) {
-		/* Otherwise traverse passed in connector list and get encoders
-		 * for them. */
-		for (ro = 0; ro < set->num_connectors; ro++) {
-			if (set->connectors[ro] == &connector->base) {
-				connector->new_encoder = intel_find_encoder(connector, to_intel_crtc(set->crtc)->pipe);
-				break;
-			}
+		bool in_mode_set = intel_connector_in_mode_set(connector, set);
+
+		if (!in_mode_set && connector->base.state->crtc != set->crtc)
+			continue;
+
+		connector_state =
+			drm_atomic_get_connector_state(state, &connector->base);
+		if (IS_ERR(connector_state))
+			return PTR_ERR(connector_state);
+
+		if (in_mode_set) {
+			int pipe = to_intel_crtc(set->crtc)->pipe;
+			connector_state->best_encoder =
+				&intel_find_encoder(connector, pipe)->base;
 		}
 
+		if (connector->base.state->crtc != set->crtc)
+			continue;
+
 		/* If we disable the crtc, disable all its connectors. Also, if
 		 * the connector is on the changing crtc but not on the new
 		 * connector list, disable it. */
-		if ((!set->fb || ro == set->num_connectors) &&
-		    connector->base.encoder &&
-		    connector->base.encoder->crtc == set->crtc) {
-			connector->new_encoder = NULL;
+		if (!set->fb || !in_mode_set) {
+			connector_state->best_encoder = NULL;
 
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [NOCRTC]\n",
 				connector->base.base.id,
 				connector->base.name);
 		}
-
-
-		if (&connector->new_encoder->base != connector->base.encoder) {
-			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] encoder changed, full mode switch\n",
-				      connector->base.base.id,
-				      connector->base.name);
-			config->mode_changed = true;
-		}
 	}
 	/* connector->new_encoder is now updated for all connectors. */
 
-	/* Update crtc of enabled connectors. */
-	for_each_intel_connector(dev, connector) {
-		struct drm_crtc *new_crtc;
+	for_each_connector_in_state(state, drm_connector, connector_state, i) {
+		connector = to_intel_connector(drm_connector);
+
+		if (!connector_state->best_encoder) {
+			ret = drm_atomic_set_crtc_for_connector(connector_state,
+								NULL);
+			if (ret)
+				return ret;
 
-		if (!connector->new_encoder)
 			continue;
+		}
 
-		new_crtc = connector->new_encoder->base.crtc;
+		if (intel_connector_in_mode_set(connector, set)) {
+			struct drm_crtc *crtc = connector->base.state->crtc;
 
-		for (ro = 0; ro < set->num_connectors; ro++) {
-			if (set->connectors[ro] == &connector->base)
-				new_crtc = set->crtc;
+			/* If this connector was in a previous crtc, add it
+			 * to the state. We might need to disable it. */
+			if (crtc) {
+				crtc_state =
+					drm_atomic_get_crtc_state(state, crtc);
+				if (IS_ERR(crtc_state))
+					return PTR_ERR(crtc_state);
+			}
+
+			ret = drm_atomic_set_crtc_for_connector(connector_state,
+								set->crtc);
+			if (ret)
+				return ret;
 		}
 
 		/* Make sure the new CRTC will work with the encoder */
-		if (!drm_encoder_crtc_ok(&connector->new_encoder->base,
-					 new_crtc)) {
+		if (!drm_encoder_crtc_ok(connector_state->best_encoder,
+					 connector_state->crtc)) {
 			return -EINVAL;
 		}
-		connector->new_encoder->new_crtc = to_intel_crtc(new_crtc);
-
-		connector_state =
-			drm_atomic_get_connector_state(state, &connector->base);
-		if (IS_ERR(connector_state))
-			return PTR_ERR(connector_state);
-
-		connector_state->crtc = new_crtc;
-		connector_state->best_encoder = &connector->new_encoder->base;
 
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d]\n",
 			connector->base.base.id,
 			connector->base.name,
-			new_crtc->base.id);
+			connector_state->crtc->base.id);
+
+		if (connector_state->best_encoder != &connector->encoder->base)
+			connector->encoder =
+				to_intel_encoder(connector_state->best_encoder);
 	}
 
-	/* Check for any encoders that needs to be disabled. */
-	for_each_intel_encoder(dev, encoder) {
-		int num_connectors = 0;
-		for_each_intel_connector(dev, connector) {
-			if (connector->new_encoder == encoder) {
-				WARN_ON(!connector->new_encoder->new_crtc);
-				num_connectors++;
-			}
-		}
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		bool has_connectors;
 
-		if (num_connectors == 0)
-			encoder->new_crtc = NULL;
-		else if (num_connectors > 1)
-			return -EINVAL;
+		ret = drm_atomic_add_affected_connectors(state, crtc);
+		if (ret)
+			return ret;
 
-		/* Only now check for crtc changes so we don't miss encoders
-		 * that will be disabled. */
-		if (&encoder->new_crtc->base != encoder->base.crtc) {
-			DRM_DEBUG_KMS("[ENCODER:%d:%s] crtc changed, full mode switch\n",
-				      encoder->base.base.id,
-				      encoder->base.name);
-			config->mode_changed = true;
-		}
+		has_connectors = !!drm_atomic_connectors_for_crtc(state, crtc);
+		if (has_connectors != crtc_state->enable)
+			crtc_state->enable =
+			crtc_state->active = has_connectors;
 	}
-	/* Now we've also updated encoder->new_crtc for all encoders. */
-	for_each_intel_connector(dev, connector) {
-		connector_state =
-			drm_atomic_get_connector_state(state, &connector->base);
-		if (IS_ERR(connector_state))
-			return PTR_ERR(connector_state);
 
-		if (connector->new_encoder) {
-			if (connector->new_encoder != connector->encoder)
-				connector->encoder = connector->new_encoder;
-		} else {
-			connector_state->crtc = NULL;
-		}
-	}
-	for_each_intel_crtc(dev, crtc) {
-		crtc->new_enabled = false;
+	ret = intel_modeset_setup_plane_state(state, set->crtc, set->mode,
+					      set->fb, set->x, set->y);
+	if (ret)
+		return ret;
 
-		for_each_intel_encoder(dev, encoder) {
-			if (encoder->new_crtc == crtc) {
-				crtc->new_enabled = true;
-				break;
-			}
-		}
+	crtc_state = drm_atomic_get_crtc_state(state, set->crtc);
+	if (IS_ERR(crtc_state))
+		return PTR_ERR(crtc_state);
 
-		if (crtc->new_enabled != crtc->base.state->enable) {
-			DRM_DEBUG_KMS("[CRTC:%d] %sabled, full mode switch\n",
-				      crtc->base.base.id,
-				      crtc->new_enabled ? "en" : "dis");
-			config->mode_changed = true;
-		}
+	if (set->mode)
+		drm_mode_copy(&crtc_state->mode, set->mode);
 
-		if (crtc->new_enabled)
-			crtc->new_config = crtc->config;
-		else
-			crtc->new_config = NULL;
-	}
+	if (set->num_connectors)
+		crtc_state->active = true;
 
 	return 0;
 }
 
-static void disable_crtc_nofb(struct intel_crtc *crtc)
+static bool primary_plane_visible(struct drm_crtc *crtc)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct intel_encoder *encoder;
-	struct intel_connector *connector;
-
-	DRM_DEBUG_KMS("Trying to restore without FB -> disabling pipe %c\n",
-		      pipe_name(crtc->pipe));
-
-	for_each_intel_connector(dev, connector) {
-		if (connector->new_encoder &&
-		    connector->new_encoder->new_crtc == crtc)
-			connector->new_encoder = NULL;
-	}
+	struct intel_plane_state *plane_state =
+		to_intel_plane_state(crtc->primary->state);
 
-	for_each_intel_encoder(dev, encoder) {
-		if (encoder->new_crtc == crtc)
-			encoder->new_crtc = NULL;
-	}
-
-	crtc->new_enabled = false;
-	crtc->new_config = NULL;
+	return plane_state->visible;
 }
 
 static int intel_crtc_set_config(struct drm_mode_set *set)
 {
 	struct drm_device *dev;
-	struct drm_mode_set save_set;
 	struct drm_atomic_state *state = NULL;
-	struct intel_set_config *config;
 	struct intel_crtc_state *pipe_config;
-	unsigned modeset_pipes, prepare_pipes, disable_pipes;
+	bool primary_plane_was_visible;
 	int ret;
 
 	BUG_ON(!set);
@@ -12101,85 +12921,42 @@ static int intel_crtc_set_config(struct drm_mode_set *set)
 
 	dev = set->crtc->dev;
 
-	ret = -ENOMEM;
-	config = kzalloc(sizeof(*config), GFP_KERNEL);
-	if (!config)
-		goto out_config;
-
-	ret = intel_set_config_save_state(dev, config);
-	if (ret)
-		goto out_config;
-
-	save_set.crtc = set->crtc;
-	save_set.mode = &set->crtc->mode;
-	save_set.x = set->crtc->x;
-	save_set.y = set->crtc->y;
-	save_set.fb = set->crtc->primary->fb;
-
-	/* Compute whether we need a full modeset, only an fb base update or no
-	 * change at all. In the future we might also check whether only the
-	 * mode changed, e.g. for LVDS where we only change the panel fitter in
-	 * such cases. */
-	intel_set_config_compute_mode_changes(set, config);
-
 	state = drm_atomic_state_alloc(dev);
-	if (!state) {
-		ret = -ENOMEM;
-		goto out_config;
-	}
+	if (!state)
+		return -ENOMEM;
 
 	state->acquire_ctx = dev->mode_config.acquire_ctx;
 
-	ret = intel_modeset_stage_output_state(dev, set, config, state);
+	ret = intel_modeset_stage_output_state(dev, set, state);
 	if (ret)
-		goto fail;
+		goto out;
 
-	pipe_config = intel_modeset_compute_config(set->crtc, set->mode,
-						   set->fb, state,
-						   &modeset_pipes,
-						   &prepare_pipes,
-						   &disable_pipes);
+	pipe_config = intel_modeset_compute_config(set->crtc, state);
 	if (IS_ERR(pipe_config)) {
 		ret = PTR_ERR(pipe_config);
-		goto fail;
-	} else if (pipe_config) {
-		if (pipe_config->has_audio !=
-		    to_intel_crtc(set->crtc)->config->has_audio)
-			config->mode_changed = true;
-
-		/*
-		 * Note we have an issue here with infoframes: current code
-		 * only updates them on the full mode set path per hw
-		 * requirements.  So here we should be checking for any
-		 * required changes and forcing a mode set.
-		 */
+		goto out;
 	}
 
 	intel_update_pipe_size(to_intel_crtc(set->crtc));
 
-	if (config->mode_changed) {
-		ret = intel_set_mode_pipes(set->crtc, set->mode,
-					   set->x, set->y, set->fb, pipe_config,
-					   modeset_pipes, prepare_pipes,
-					   disable_pipes);
-	} else if (config->fb_changed) {
-		struct intel_crtc *intel_crtc = to_intel_crtc(set->crtc);
-		struct drm_plane *primary = set->crtc->primary;
-		int vdisplay, hdisplay;
+	primary_plane_was_visible = primary_plane_visible(set->crtc);
+
+	ret = intel_set_mode_with_config(set->crtc, pipe_config, true);
 
-		drm_crtc_get_hv_timing(set->mode, &hdisplay, &vdisplay);
-		ret = primary->funcs->update_plane(primary, set->crtc, set->fb,
-						   0, 0, hdisplay, vdisplay,
-						   set->x << 16, set->y << 16,
-						   hdisplay << 16, vdisplay << 16);
+	if (ret == 0 &&
+	    pipe_config->base.enable &&
+	    pipe_config->base.planes_changed &&
+	    !needs_modeset(&pipe_config->base)) {
+		struct intel_crtc *intel_crtc = to_intel_crtc(set->crtc);
 
 		/*
 		 * We need to make sure the primary plane is re-enabled if it
 		 * has previously been turned off.
 		 */
-		if (!intel_crtc->primary_enabled && ret == 0) {
+		if (ret == 0 && !primary_plane_was_visible &&
+		    primary_plane_visible(set->crtc)) {
 			WARN_ON(!intel_crtc->active);
-			intel_enable_primary_hw_plane(set->crtc->primary, set->crtc);
+			intel_post_enable_primary(set->crtc);
 		}
 
 		/*
@@ -12197,33 +12974,11 @@ static int intel_crtc_set_config(struct drm_mode_set *set)
 	if (ret) {
 		DRM_DEBUG_KMS("failed to set mode on [CRTC:%d], err = %d\n",
 			      set->crtc->base.id, ret);
-fail:
-		intel_set_config_restore_state(dev, config);
-
-		drm_atomic_state_clear(state);
-
-		/*
-		 * HACK: if the pipe was on, but we didn't have a framebuffer,
-		 * force the pipe off to avoid oopsing in the modeset code
-		 * due to fb==NULL. This should only happen during boot since
-		 * we don't yet reconstruct the FB from the hardware state.
-		 */
-		if (to_intel_crtc(save_set.crtc)->new_enabled && !save_set.fb)
-			disable_crtc_nofb(to_intel_crtc(save_set.crtc));
-
-		/* Try to restore the config */
-		if (config->mode_changed &&
-		    intel_set_mode(save_set.crtc, save_set.mode,
-				   save_set.x, save_set.y, save_set.fb,
-				   state))
-			DRM_ERROR("failed to restore config after modeset failure\n");
 	}
 
-out_config:
-	if (state)
+out:
+	if (ret)
 		drm_atomic_state_free(state);
-
-	intel_set_config_free(config);
 	return ret;
 }
 
@@ -12444,6 +13199,36 @@ intel_cleanup_plane_fb(struct drm_plane *plane,
 	}
 }
 
+int
+skl_max_scale(struct intel_crtc *intel_crtc, struct intel_crtc_state *crtc_state)
+{
+	int max_scale;
+	struct drm_device *dev;
+	struct drm_i915_private *dev_priv;
+	int crtc_clock, cdclk;
+
+	if (!intel_crtc || !crtc_state)
+		return DRM_PLANE_HELPER_NO_SCALING;
+
+	dev = intel_crtc->base.dev;
+	dev_priv = dev->dev_private;
+	crtc_clock = crtc_state->base.adjusted_mode.crtc_clock;
+	cdclk = dev_priv->display.get_display_clock_speed(dev);
+
+	if (!crtc_clock || !cdclk)
+		return DRM_PLANE_HELPER_NO_SCALING;
+
+	/*
+	 * skl max scale is lower of:
+	 *    close to 3 but not 3, -1 is for that purpose
+	 *            or
+	 *    cdclk/crtc_clock
+	 */
+	max_scale = min((1 << 16) * 3 - 1, (1 << 8) * ((cdclk << 8) / crtc_clock));
+
+	return max_scale;
+}
+
 static int
 intel_check_primary_plane(struct drm_plane *plane,
 			  struct intel_plane_state *state)
@@ -12452,24 +13237,43 @@ intel_check_primary_plane(struct drm_plane *plane,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc = state->base.crtc;
 	struct intel_crtc *intel_crtc;
+	struct intel_crtc_state *crtc_state;
 	struct drm_framebuffer *fb = state->base.fb;
 	struct drm_rect *dest = &state->dst;
 	struct drm_rect *src = &state->src;
 	const struct drm_rect *clip = &state->clip;
+	bool can_position = false;
+	int max_scale = DRM_PLANE_HELPER_NO_SCALING;
+	int min_scale = DRM_PLANE_HELPER_NO_SCALING;
 	int ret;
 
 	crtc = crtc ? crtc : plane->crtc;
 	intel_crtc = to_intel_crtc(crtc);
+	crtc_state = state->base.state ?
+		intel_atomic_get_crtc_state(state->base.state, intel_crtc) : NULL;
+
+	if (INTEL_INFO(dev)->gen >= 9) {
+		/* use scaler when colorkey is not required */
+		if (to_intel_plane(plane)->ckey.flags == I915_SET_COLORKEY_NONE) {
+			min_scale = 1;
+			max_scale = skl_max_scale(intel_crtc, crtc_state);
+		}
+		can_position = true;
+	}
 
 	ret = drm_plane_helper_check_update(plane, crtc, fb,
 					    src, dest, clip,
-					    DRM_PLANE_HELPER_NO_SCALING,
-					    DRM_PLANE_HELPER_NO_SCALING,
-					    false, true, &state->visible);
+					    min_scale,
+					    max_scale,
+					    can_position, true,
+					    &state->visible);
 	if (ret)
 		return ret;
 
 	if (intel_crtc->active) {
+		struct intel_plane_state *old_state =
+			to_intel_plane_state(plane->state);
+
 		intel_crtc->atomic.wait_for_flips = true;
 
 		/*
@@ -12482,20 +13286,20 @@ intel_check_primary_plane(struct drm_plane *plane,
 		 * one is done too late. We eventually need to unify
 		 * this.
 		 */
-		if (intel_crtc->primary_enabled &&
+		if (state->visible &&
 		    INTEL_INFO(dev)->gen <= 4 && !IS_G4X(dev) &&
 		    dev_priv->fbc.crtc == intel_crtc &&
 		    state->base.rotation != BIT(DRM_ROTATE_0)) {
 			intel_crtc->atomic.disable_fbc = true;
 		}
 
-		if (state->visible) {
+		if (state->visible && !old_state->visible) {
 			/*
 			 * BDW signals flip done immediately if the plane
 			 * is disabled, even if the plane enable is already
 			 * armed to occur at the next vblank :(
 			 */
-			if (IS_BROADWELL(dev) && !intel_crtc->primary_enabled)
+			if (IS_BROADWELL(dev))
 				intel_crtc->atomic.wait_vblank = true;
 		}
 
@@ -12508,6 +13312,13 @@ intel_check_primary_plane(struct drm_plane *plane,
 			intel_crtc->atomic.update_wm = true;
 	}
 
+	if (INTEL_INFO(dev)->gen >= 9) {
+		ret = skl_update_scaler_users(intel_crtc, crtc_state,
+			to_intel_plane(plane), state, 0);
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 
@@ -12530,27 +13341,26 @@ intel_commit_primary_plane(struct drm_plane *plane,
 	crtc->y = src->y1 >> 16;
 
 	if (intel_crtc->active) {
-		if (state->visible) {
+		if (state->visible)
 			/* FIXME: kill this fastboot hack */
 			intel_update_pipe_size(intel_crtc);
 
-			intel_crtc->primary_enabled = true;
-
-			dev_priv->display.update_primary_plane(crtc, plane->fb,
-					crtc->x, crtc->y);
-		} else {
-			/*
-			 * If clipping results in a non-visible primary plane,
-			 * we'll disable the primary plane.  Note that this is
-			 * a bit different than what happens if userspace
-			 * explicitly disables the plane by passing fb=0
-			 * because plane->fb still gets set and pinned.
-			 */
-			intel_disable_primary_hw_plane(plane, crtc);
-		}
+		dev_priv->display.update_primary_plane(crtc, plane->fb,
+						       crtc->x, crtc->y);
 	}
 }
 
+static void
+intel_disable_primary_plane(struct drm_plane *plane,
+			    struct drm_crtc *crtc,
+			    bool force)
+{
+	struct drm_device *dev = plane->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->display.update_primary_plane(crtc, NULL, 0, 0);
+}
+
 static void intel_begin_crtc_commit(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -12655,8 +13465,8 @@ void intel_plane_destroy(struct drm_plane *plane)
 }
 
 const struct drm_plane_funcs intel_plane_funcs = {
-	.update_plane = drm_plane_helper_update,
-	.disable_plane = drm_plane_helper_disable,
+	.update_plane = drm_atomic_helper_update_plane,
+	.disable_plane = drm_atomic_helper_disable_plane,
 	.destroy = intel_plane_destroy,
 	.set_property = drm_atomic_helper_plane_set_property,
 	.atomic_get_property = intel_plane_atomic_get_property,
@@ -12687,19 +13497,28 @@ static struct drm_plane *intel_primary_plane_create(struct drm_device *dev,
 
 	primary->can_scale = false;
 	primary->max_downscale = 1;
+	if (INTEL_INFO(dev)->gen >= 9) {
+		primary->can_scale = true;
+		state->scaler_id = -1;
+	}
 	primary->pipe = pipe;
 	primary->plane = pipe;
 	primary->check_plane = intel_check_primary_plane;
 	primary->commit_plane = intel_commit_primary_plane;
+	primary->disable_plane = intel_disable_primary_plane;
+	primary->ckey.flags = I915_SET_COLORKEY_NONE;
 	if (HAS_FBC(dev) && INTEL_INFO(dev)->gen < 4)
 		primary->plane = !pipe;
 
-	if (INTEL_INFO(dev)->gen <= 3) {
-		intel_primary_formats = intel_primary_formats_gen2;
-		num_formats = ARRAY_SIZE(intel_primary_formats_gen2);
+	if (INTEL_INFO(dev)->gen >= 9) {
+		intel_primary_formats = skl_primary_formats;
+		num_formats = ARRAY_SIZE(skl_primary_formats);
+	} else if (INTEL_INFO(dev)->gen >= 4) {
+		intel_primary_formats = i965_primary_formats;
+		num_formats = ARRAY_SIZE(i965_primary_formats);
 	} else {
-		intel_primary_formats = intel_primary_formats_gen4;
-		num_formats = ARRAY_SIZE(intel_primary_formats_gen4);
+		intel_primary_formats = i8xx_primary_formats;
+		num_formats = ARRAY_SIZE(i8xx_primary_formats);
 	}
 
 	drm_universal_plane_init(dev, &primary->base, 0,
@@ -12707,23 +13526,32 @@ static struct drm_plane *intel_primary_plane_create(struct drm_device *dev,
 				 intel_primary_formats, num_formats,
 				 DRM_PLANE_TYPE_PRIMARY);
 
-	if (INTEL_INFO(dev)->gen >= 4) {
-		if (!dev->mode_config.rotation_property)
-			dev->mode_config.rotation_property =
-				drm_mode_create_rotation_property(dev,
-							BIT(DRM_ROTATE_0) |
-							BIT(DRM_ROTATE_180));
-		if (dev->mode_config.rotation_property)
-			drm_object_attach_property(&primary->base.base,
-				dev->mode_config.rotation_property,
-				state->base.rotation);
-	}
+	if (INTEL_INFO(dev)->gen >= 4)
+		intel_create_rotation_property(dev, primary);
 
 	drm_plane_helper_add(&primary->base, &intel_plane_helper_funcs);
 
 	return &primary->base;
 }
 
+void intel_create_rotation_property(struct drm_device *dev, struct intel_plane *plane)
+{
+	if (!dev->mode_config.rotation_property) {
+		unsigned long flags = BIT(DRM_ROTATE_0) |
+			BIT(DRM_ROTATE_180);
+
+		if (INTEL_INFO(dev)->gen >= 9)
+			flags |= BIT(DRM_ROTATE_90) | BIT(DRM_ROTATE_270);
+
+		dev->mode_config.rotation_property =
+			drm_mode_create_rotation_property(dev, flags);
+	}
+	if (dev->mode_config.rotation_property)
+		drm_object_attach_property(&plane->base.base,
+				dev->mode_config.rotation_property,
+				plane->base.state->rotation);
+}
+
 static int
 intel_check_cursor_plane(struct drm_plane *plane,
 			 struct intel_plane_state *state)
@@ -12786,6 +13614,22 @@ finish:
 }
 
 static void
+intel_disable_cursor_plane(struct drm_plane *plane,
+			   struct drm_crtc *crtc,
+			   bool force)
+{
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+
+	if (!force) {
+		plane->fb = NULL;
+		intel_crtc->cursor_bo = NULL;
+		intel_crtc->cursor_addr = 0;
+	}
+
+	intel_crtc_update_cursor(crtc, false);
+}
+
+static void
 intel_commit_cursor_plane(struct drm_plane *plane,
 			  struct intel_plane_state *state)
 {
@@ -12843,6 +13687,7 @@ static struct drm_plane *intel_cursor_plane_create(struct drm_device *dev,
 	cursor->plane = pipe;
 	cursor->check_plane = intel_check_cursor_plane;
 	cursor->commit_plane = intel_commit_cursor_plane;
+	cursor->disable_plane = intel_disable_cursor_plane;
 
 	drm_universal_plane_init(dev, &cursor->base, 0,
 				 &intel_plane_funcs,
@@ -12862,11 +13707,32 @@ static struct drm_plane *intel_cursor_plane_create(struct drm_device *dev,
 				state->base.rotation);
 	}
 
+	if (INTEL_INFO(dev)->gen >=9)
+		state->scaler_id = -1;
+
 	drm_plane_helper_add(&cursor->base, &intel_plane_helper_funcs);
 
 	return &cursor->base;
 }
 
+static void skl_init_scalers(struct drm_device *dev, struct intel_crtc *intel_crtc,
+	struct intel_crtc_state *crtc_state)
+{
+	int i;
+	struct intel_scaler *intel_scaler;
+	struct intel_crtc_scaler_state *scaler_state = &crtc_state->scaler_state;
+
+	for (i = 0; i < intel_crtc->num_scalers; i++) {
+		intel_scaler = &scaler_state->scalers[i];
+		intel_scaler->in_use = 0;
+		intel_scaler->id = i;
+
+		intel_scaler->mode = PS_SCALER_MODE_DYN;
+	}
+
+	scaler_state->scaler_id = -1;
+}
+
 static void intel_crtc_init(struct drm_device *dev, int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -12883,9 +13749,20 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 	crtc_state = kzalloc(sizeof(*crtc_state), GFP_KERNEL);
 	if (!crtc_state)
 		goto fail;
-	intel_crtc_set_state(intel_crtc, crtc_state);
+	intel_crtc->config = crtc_state;
+	intel_crtc->base.state = &crtc_state->base;
 	crtc_state->base.crtc = &intel_crtc->base;
 
+	/* initialize shared scalers */
+	if (INTEL_INFO(dev)->gen >= 9) {
+		if (pipe == PIPE_C)
+			intel_crtc->num_scalers = 1;
+		else
+			intel_crtc->num_scalers = SKL_NUM_SCALERS;
+
+		skl_init_scalers(dev, intel_crtc, crtc_state);
+	}
+
 	primary = intel_primary_plane_create(dev, pipe);
 	if (!primary)
 		goto fail;
@@ -12926,8 +13803,6 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 	dev_priv->plane_to_crtc_mapping[intel_crtc->plane] = &intel_crtc->base;
 	dev_priv->pipe_to_crtc_mapping[intel_crtc->pipe] = &intel_crtc->base;
 
-	INIT_WORK(&intel_crtc->mmio_flip.work, intel_mmio_flip_work_func);
-
 	drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs);
 
 	WARN_ON(drm_crtc_index(&intel_crtc->base) != intel_crtc->pipe);
@@ -13038,7 +13913,16 @@ static void intel_setup_outputs(struct drm_device *dev)
 	if (intel_crt_present(dev))
 		intel_crt_init(dev);
 
-	if (HAS_DDI(dev)) {
+	if (IS_BROXTON(dev)) {
+		/*
+		 * FIXME: Broxton doesn't support port detection via the
+		 * DDI_BUF_CTL_A or SFUSE_STRAP registers, find another way to
+		 * detect the ports.
+		 */
+		intel_ddi_init(dev, PORT_A);
+		intel_ddi_init(dev, PORT_B);
+		intel_ddi_init(dev, PORT_C);
+	} else if (HAS_DDI(dev)) {
 		int found;
 
 		/*
@@ -13315,25 +14199,35 @@ static int intel_framebuffer_init(struct drm_device *dev,
 	case DRM_FORMAT_ARGB8888:
 		break;
 	case DRM_FORMAT_XRGB1555:
-	case DRM_FORMAT_ARGB1555:
 		if (INTEL_INFO(dev)->gen > 3) {
 			DRM_DEBUG("unsupported pixel format: %s\n",
 				  drm_get_format_name(mode_cmd->pixel_format));
 			return -EINVAL;
 		}
 		break;
-	case DRM_FORMAT_XBGR8888:
 	case DRM_FORMAT_ABGR8888:
+		if (!IS_VALLEYVIEW(dev) && INTEL_INFO(dev)->gen < 9) {
+			DRM_DEBUG("unsupported pixel format: %s\n",
+				  drm_get_format_name(mode_cmd->pixel_format));
+			return -EINVAL;
+		}
+		break;
+	case DRM_FORMAT_XBGR8888:
 	case DRM_FORMAT_XRGB2101010:
-	case DRM_FORMAT_ARGB2101010:
 	case DRM_FORMAT_XBGR2101010:
-	case DRM_FORMAT_ABGR2101010:
 		if (INTEL_INFO(dev)->gen < 4) {
 			DRM_DEBUG("unsupported pixel format: %s\n",
 				  drm_get_format_name(mode_cmd->pixel_format));
 			return -EINVAL;
 		}
 		break;
+	case DRM_FORMAT_ABGR2101010:
+		if (!IS_VALLEYVIEW(dev)) {
+			DRM_DEBUG("unsupported pixel format: %s\n",
+				  drm_get_format_name(mode_cmd->pixel_format));
+			return -EINVAL;
+		}
+		break;
 	case DRM_FORMAT_YUYV:
 	case DRM_FORMAT_UYVY:
 	case DRM_FORMAT_YVYU:
@@ -13474,10 +14368,23 @@ static void intel_init_display(struct drm_device *dev)
 	}
 
 	/* Returns the core display clock speed */
-	if (IS_VALLEYVIEW(dev))
+	if (IS_SKYLAKE(dev))
+		dev_priv->display.get_display_clock_speed =
+			skylake_get_display_clock_speed;
+	else if (IS_BROADWELL(dev))
+		dev_priv->display.get_display_clock_speed =
+			broadwell_get_display_clock_speed;
+	else if (IS_HASWELL(dev))
+		dev_priv->display.get_display_clock_speed =
+			haswell_get_display_clock_speed;
+	else if (IS_VALLEYVIEW(dev))
 		dev_priv->display.get_display_clock_speed =
 			valleyview_get_display_clock_speed;
-	else if (IS_I945G(dev) || (IS_G33(dev) && !IS_PINEVIEW_M(dev)))
+	else if (IS_GEN5(dev))
+		dev_priv->display.get_display_clock_speed =
+			ilk_get_display_clock_speed;
+	else if (IS_I945G(dev) || IS_BROADWATER(dev) ||
+		 IS_GEN6(dev) || IS_IVYBRIDGE(dev) || (IS_G33(dev) && !IS_PINEVIEW_M(dev)))
 		dev_priv->display.get_display_clock_speed =
 			i945_get_display_clock_speed;
 	else if (IS_I915G(dev))
@@ -13514,6 +14421,9 @@ static void intel_init_display(struct drm_device *dev)
 	} else if (IS_VALLEYVIEW(dev)) {
 		dev_priv->display.modeset_global_resources =
 			valleyview_modeset_global_resources;
+	} else if (IS_BROXTON(dev)) {
+		dev_priv->display.modeset_global_resources =
+			broxton_modeset_global_resources;
 	}
 
 	switch (INTEL_INFO(dev)->gen) {
@@ -13922,8 +14832,9 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 		 * Temporarily change the plane mapping and disable everything
 		 * ...  */
 		plane = crtc->plane;
+		to_intel_plane_state(crtc->base.primary->state)->visible = true;
 		crtc->plane = !plane;
-		crtc->primary_enabled = true;
+		intel_crtc_disable_planes(&crtc->base);
 		dev_priv->display.crtc_disable(&crtc->base);
 		crtc->plane = plane;
 
@@ -13945,6 +14856,7 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 
 		WARN_ON(crtc->active);
 		crtc->base.state->enable = false;
+		crtc->base.state->active = false;
 		crtc->base.enabled = false;
 	}
 
@@ -13973,6 +14885,7 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 			      crtc->active ? "enabled" : "disabled");
 
 		crtc->base.state->enable = crtc->active;
+		crtc->base.state->active = crtc->active;
 		crtc->base.enabled = crtc->active;
 
 		/* Because we only establish the connector -> encoder ->
@@ -14100,6 +15013,9 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 	int i;
 
 	for_each_intel_crtc(dev, crtc) {
+		struct drm_plane *primary = crtc->base.primary;
+		struct intel_plane_state *plane_state;
+
 		memset(crtc->config, 0, sizeof(*crtc->config));
 
 		crtc->config->quirks |= PIPE_CONFIG_QUIRK_INHERITED_MODE;
@@ -14108,8 +15024,11 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 								 crtc->config);
 
 		crtc->base.state->enable = crtc->active;
+		crtc->base.state->active = crtc->active;
 		crtc->base.enabled = crtc->active;
-		crtc->primary_enabled = primary_get_hw_state(crtc);
+
+		plane_state = to_intel_plane_state(primary->state);
+		plane_state->visible = primary_get_hw_state(crtc);
 
 		DRM_DEBUG_KMS("[CRTC:%d] hw state readout: %s\n",
 			      crtc->base.base.id,
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index d714a4b5711e..76afc62373d7 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -41,6 +41,12 @@
 
 #define DP_LINK_CHECK_TIMEOUT	(10 * 1000)
 
+/* Compliance test status bits  */
+#define INTEL_DP_RESOLUTION_SHIFT_MASK	0
+#define INTEL_DP_RESOLUTION_PREFERRED	(1 << INTEL_DP_RESOLUTION_SHIFT_MASK)
+#define INTEL_DP_RESOLUTION_STANDARD	(2 << INTEL_DP_RESOLUTION_SHIFT_MASK)
+#define INTEL_DP_RESOLUTION_FAILSAFE	(3 << INTEL_DP_RESOLUTION_SHIFT_MASK)
+
 struct dp_link_dpll {
 	int link_bw;
 	struct dpll dpll;
@@ -84,8 +90,8 @@ static const struct dp_link_dpll chv_dpll[] = {
 	{ DP_LINK_BW_5_4,	/* m2_int = 27, m2_fraction = 0 */
 		{ .p1 = 2, .p2 = 1, .n = 1, .m1 = 2, .m2 = 0x6c00000 } }
 };
-/* Skylake supports following rates */
-static const int gen9_rates[] = { 162000, 216000, 270000,
+
+static const int skl_rates[] = { 162000, 216000, 270000,
 				  324000, 432000, 540000 };
 static const int chv_rates[] = { 162000, 202500, 210000, 216000,
 				 243000, 270000, 324000, 405000,
@@ -696,15 +702,13 @@ static uint32_t ilk_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (index)
 		return 0;
 
 	if (intel_dig_port->port == PORT_A) {
-		if (IS_GEN6(dev) || IS_GEN7(dev))
-			return 200; /* SNB & IVB eDP input clock at 400Mhz */
-		else
-			return 225; /* eDP input clock at 450Mhz */
+		return DIV_ROUND_UP(dev_priv->display.get_display_clock_speed(dev), 2000);
 	} else {
 		return DIV_ROUND_UP(intel_pch_rawclk(dev), 2);
 	}
@@ -719,7 +723,7 @@ static uint32_t hsw_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 	if (intel_dig_port->port == PORT_A) {
 		if (index)
 			return 0;
-		return DIV_ROUND_CLOSEST(intel_ddi_get_cdclk_freq(dev_priv), 2000);
+		return DIV_ROUND_CLOSEST(dev_priv->display.get_display_clock_speed(dev), 2000);
 	} else if (dev_priv->pch_id == INTEL_PCH_LPT_DEVICE_ID_TYPE) {
 		/* Workaround for non-ULT HSW */
 		switch (index) {
@@ -876,9 +880,18 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 				   DP_AUX_CH_CTL_TIME_OUT_ERROR |
 				   DP_AUX_CH_CTL_RECEIVE_ERROR);
 
-			if (status & (DP_AUX_CH_CTL_TIME_OUT_ERROR |
-				      DP_AUX_CH_CTL_RECEIVE_ERROR))
+			if (status & DP_AUX_CH_CTL_TIME_OUT_ERROR)
+				continue;
+
+			/* DP CTS 1.2 Core Rev 1.1, 4.2.1.1 & 4.2.1.2
+			 *   400us delay required for errors and timeouts
+			 *   Timeout errors from the HW already meet this
+			 *   requirement so skip to next iteration
+			 */
+			if (status & DP_AUX_CH_CTL_RECEIVE_ERROR) {
+				usleep_range(400, 500);
 				continue;
+			}
 			if (status & DP_AUX_CH_CTL_DONE)
 				goto done;
 		}
@@ -1083,6 +1096,9 @@ skl_edp_set_pll_config(struct intel_crtc_state *pipe_config, int link_clock)
 {
 	u32 ctrl1;
 
+	memset(&pipe_config->dpll_hw_state, 0,
+	       sizeof(pipe_config->dpll_hw_state));
+
 	pipe_config->ddi_pll_sel = SKL_DPLL0;
 	pipe_config->dpll_hw_state.cfgcr1 = 0;
 	pipe_config->dpll_hw_state.cfgcr2 = 0;
@@ -1090,30 +1106,30 @@ skl_edp_set_pll_config(struct intel_crtc_state *pipe_config, int link_clock)
 	ctrl1 = DPLL_CTRL1_OVERRIDE(SKL_DPLL0);
 	switch (link_clock / 2) {
 	case 81000:
-		ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_810,
+		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810,
 					      SKL_DPLL0);
 		break;
 	case 135000:
-		ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_1350,
+		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1350,
 					      SKL_DPLL0);
 		break;
 	case 270000:
-		ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_2700,
+		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2700,
 					      SKL_DPLL0);
 		break;
 	case 162000:
-		ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_1620,
+		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1620,
 					      SKL_DPLL0);
 		break;
 	/* TBD: For DP link rates 2.16 GHz and 4.32 GHz, VCO is 8640 which
 	results in CDCLK change. Need to handle the change of CDCLK by
 	disabling pipes and re-enabling them */
 	case 108000:
-		ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_1080,
+		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1080,
 					      SKL_DPLL0);
 		break;
 	case 216000:
-		ctrl1 |= DPLL_CRTL1_LINK_RATE(DPLL_CRTL1_LINK_RATE_2160,
+		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2160,
 					      SKL_DPLL0);
 		break;
 
@@ -1153,9 +1169,9 @@ intel_dp_sink_rates(struct intel_dp *intel_dp, const int **sink_rates)
 static int
 intel_dp_source_rates(struct drm_device *dev, const int **source_rates)
 {
-	if (INTEL_INFO(dev)->gen >= 9) {
-		*source_rates = gen9_rates;
-		return ARRAY_SIZE(gen9_rates);
+	if (IS_SKYLAKE(dev)) {
+		*source_rates = skl_rates;
+		return ARRAY_SIZE(skl_rates);
 	} else if (IS_CHERRYVIEW(dev)) {
 		*source_rates = chv_rates;
 		return ARRAY_SIZE(chv_rates);
@@ -1252,7 +1268,7 @@ static void snprintf_int_array(char *str, size_t len,
 	str[0] = '\0';
 
 	for (i = 0; i < nelem; i++) {
-		int r = snprintf(str, len, "%d,", array[i]);
+		int r = snprintf(str, len, "%s%d", i ? ", " : "", array[i]);
 		if (r >= len)
 			return;
 		str += r;
@@ -1352,6 +1368,14 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	if (is_edp(intel_dp) && intel_connector->panel.fixed_mode) {
 		intel_fixed_panel_mode(intel_connector->panel.fixed_mode,
 				       adjusted_mode);
+
+		if (INTEL_INFO(dev)->gen >= 9) {
+			int ret;
+			ret = skl_update_scaler_users(intel_crtc, pipe_config, NULL, NULL, 0);
+			if (ret)
+				return ret;
+		}
+
 		if (!HAS_PCH_SPLIT(dev))
 			intel_gmch_panel_fitting(intel_crtc, pipe_config,
 						 intel_connector->panel.fitting_mode);
@@ -1464,6 +1488,8 @@ found:
 
 	if (IS_SKYLAKE(dev) && is_edp(intel_dp))
 		skl_edp_set_pll_config(pipe_config, common_rates[clock]);
+	else if (IS_BROXTON(dev))
+		/* handled in ddi */;
 	else if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		hsw_dp_set_ddi_pll_sel(pipe_config, intel_dp->link_bw);
 	else
@@ -1543,7 +1569,7 @@ static void intel_dp_prepare(struct intel_encoder *encoder)
 
 	/* Split out the IBX/CPU vs CPT settings */
 
-	if (port == PORT_A && IS_GEN7(dev) && !IS_VALLEYVIEW(dev)) {
+	if (IS_GEN7(dev) && port == PORT_A) {
 		if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
 			intel_dp->DP |= DP_SYNC_HS_HIGH;
 		if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
@@ -1554,7 +1580,18 @@ static void intel_dp_prepare(struct intel_encoder *encoder)
 			intel_dp->DP |= DP_ENHANCED_FRAMING;
 
 		intel_dp->DP |= crtc->pipe << 29;
-	} else if (!HAS_PCH_CPT(dev) || port == PORT_A) {
+	} else if (HAS_PCH_CPT(dev) && port != PORT_A) {
+		u32 trans_dp;
+
+		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
+
+		trans_dp = I915_READ(TRANS_DP_CTL(crtc->pipe));
+		if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
+			trans_dp |= TRANS_DP_ENH_FRAMING;
+		else
+			trans_dp &= ~TRANS_DP_ENH_FRAMING;
+		I915_WRITE(TRANS_DP_CTL(crtc->pipe), trans_dp);
+	} else {
 		if (!HAS_PCH_SPLIT(dev) && !IS_VALLEYVIEW(dev))
 			intel_dp->DP |= intel_dp->color_range;
 
@@ -1567,14 +1604,10 @@ static void intel_dp_prepare(struct intel_encoder *encoder)
 		if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 			intel_dp->DP |= DP_ENHANCED_FRAMING;
 
-		if (!IS_CHERRYVIEW(dev)) {
-			if (crtc->pipe == 1)
-				intel_dp->DP |= DP_PIPEB_SELECT;
-		} else {
+		if (IS_CHERRYVIEW(dev))
 			intel_dp->DP |= DP_PIPE_SELECT_CHV(crtc->pipe);
-		}
-	} else {
-		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
+		else if (crtc->pipe == PIPE_B)
+			intel_dp->DP |= DP_PIPEB_SELECT;
 	}
 }
 
@@ -2158,41 +2191,25 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder,
 	if (!(tmp & DP_PORT_EN))
 		return false;
 
-	if (port == PORT_A && IS_GEN7(dev) && !IS_VALLEYVIEW(dev)) {
+	if (IS_GEN7(dev) && port == PORT_A) {
 		*pipe = PORT_TO_PIPE_CPT(tmp);
-	} else if (IS_CHERRYVIEW(dev)) {
-		*pipe = DP_PORT_TO_PIPE_CHV(tmp);
-	} else if (!HAS_PCH_CPT(dev) || port == PORT_A) {
-		*pipe = PORT_TO_PIPE(tmp);
-	} else {
-		u32 trans_sel;
-		u32 trans_dp;
-		int i;
+	} else if (HAS_PCH_CPT(dev) && port != PORT_A) {
+		enum pipe p;
 
-		switch (intel_dp->output_reg) {
-		case PCH_DP_B:
-			trans_sel = TRANS_DP_PORT_SEL_B;
-			break;
-		case PCH_DP_C:
-			trans_sel = TRANS_DP_PORT_SEL_C;
-			break;
-		case PCH_DP_D:
-			trans_sel = TRANS_DP_PORT_SEL_D;
-			break;
-		default:
-			return true;
-		}
-
-		for_each_pipe(dev_priv, i) {
-			trans_dp = I915_READ(TRANS_DP_CTL(i));
-			if ((trans_dp & TRANS_DP_PORT_SEL_MASK) == trans_sel) {
-				*pipe = i;
+		for_each_pipe(dev_priv, p) {
+			u32 trans_dp = I915_READ(TRANS_DP_CTL(p));
+			if (TRANS_DP_PIPE_TO_PORT(trans_dp) == port) {
+				*pipe = p;
 				return true;
 			}
 		}
 
 		DRM_DEBUG_KMS("No pipe for dp port 0x%x found\n",
 			      intel_dp->output_reg);
+	} else if (IS_CHERRYVIEW(dev)) {
+		*pipe = DP_PORT_TO_PIPE_CHV(tmp);
+	} else {
+		*pipe = PORT_TO_PIPE(tmp);
 	}
 
 	return true;
@@ -2213,24 +2230,24 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 
 	pipe_config->has_audio = tmp & DP_AUDIO_OUTPUT_ENABLE && port != PORT_A;
 
-	if ((port == PORT_A) || !HAS_PCH_CPT(dev)) {
-		if (tmp & DP_SYNC_HS_HIGH)
+	if (HAS_PCH_CPT(dev) && port != PORT_A) {
+		tmp = I915_READ(TRANS_DP_CTL(crtc->pipe));
+		if (tmp & TRANS_DP_HSYNC_ACTIVE_HIGH)
 			flags |= DRM_MODE_FLAG_PHSYNC;
 		else
 			flags |= DRM_MODE_FLAG_NHSYNC;
 
-		if (tmp & DP_SYNC_VS_HIGH)
+		if (tmp & TRANS_DP_VSYNC_ACTIVE_HIGH)
 			flags |= DRM_MODE_FLAG_PVSYNC;
 		else
 			flags |= DRM_MODE_FLAG_NVSYNC;
 	} else {
-		tmp = I915_READ(TRANS_DP_CTL(crtc->pipe));
-		if (tmp & TRANS_DP_HSYNC_ACTIVE_HIGH)
+		if (tmp & DP_SYNC_HS_HIGH)
 			flags |= DRM_MODE_FLAG_PHSYNC;
 		else
 			flags |= DRM_MODE_FLAG_NHSYNC;
 
-		if (tmp & TRANS_DP_VSYNC_ACTIVE_HIGH)
+		if (tmp & DP_SYNC_VS_HIGH)
 			flags |= DRM_MODE_FLAG_PVSYNC;
 		else
 			flags |= DRM_MODE_FLAG_NVSYNC;
@@ -2337,7 +2354,7 @@ static void chv_post_disable_dp(struct intel_encoder *encoder)
 
 	intel_dp_link_down(intel_dp);
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* Propagate soft reset to data lane reset */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
@@ -2356,7 +2373,7 @@ static void chv_post_disable_dp(struct intel_encoder *encoder)
 	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void
@@ -2395,7 +2412,8 @@ _intel_dp_set_link_train(struct intel_dp *intel_dp,
 		}
 		I915_WRITE(DP_TP_CTL(port), temp);
 
-	} else if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || port != PORT_A)) {
+	} else if ((IS_GEN7(dev) && port == PORT_A) ||
+		   (HAS_PCH_CPT(dev) && port != PORT_A)) {
 		*DP &= ~DP_LINK_TRAIN_MASK_CPT;
 
 		switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) {
@@ -2473,6 +2491,7 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	uint32_t dp_reg = I915_READ(intel_dp->output_reg);
+	unsigned int lane_mask = 0x0;
 
 	if (WARN_ON(dp_reg & DP_PORT_EN))
 		return;
@@ -2491,7 +2510,8 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 	pps_unlock(intel_dp);
 
 	if (IS_VALLEYVIEW(dev))
-		vlv_wait_port_ready(dev_priv, dp_to_dig_port(intel_dp));
+		vlv_wait_port_ready(dev_priv, dp_to_dig_port(intel_dp),
+				    lane_mask);
 
 	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
 	intel_dp_start_link_train(intel_dp);
@@ -2650,7 +2670,7 @@ static void vlv_pre_enable_dp(struct intel_encoder *encoder)
 	int pipe = intel_crtc->pipe;
 	u32 val;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW8(port));
 	val = 0;
@@ -2663,7 +2683,7 @@ static void vlv_pre_enable_dp(struct intel_encoder *encoder)
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW14(port), 0x00760018);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW23(port), 0x00400888);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	intel_enable_dp(encoder);
 }
@@ -2681,7 +2701,7 @@ static void vlv_dp_pre_pll_enable(struct intel_encoder *encoder)
 	intel_dp_prepare(encoder);
 
 	/* Program Tx lane resets to default */
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW0(port),
 			 DPIO_PCS_TX_LANE2_RESET |
 			 DPIO_PCS_TX_LANE1_RESET);
@@ -2695,7 +2715,7 @@ static void vlv_dp_pre_pll_enable(struct intel_encoder *encoder)
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW12(port), 0x00750f00);
 	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW11(port), 0x00001500);
 	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW14(port), 0x40400000);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void chv_pre_enable_dp(struct intel_encoder *encoder)
@@ -2708,10 +2728,10 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 		to_intel_crtc(encoder->base.crtc);
 	enum dpio_channel ch = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
-	int data, i;
+	int data, i, stagger;
 	u32 val;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* allow hardware to manage TX FIFO reset source */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW11(ch));
@@ -2748,9 +2768,40 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 	}
 
 	/* Data lane stagger programming */
-	/* FIXME: Fix up value only after power analysis */
+	if (intel_crtc->config->port_clock > 270000)
+		stagger = 0x18;
+	else if (intel_crtc->config->port_clock > 135000)
+		stagger = 0xd;
+	else if (intel_crtc->config->port_clock > 67500)
+		stagger = 0x7;
+	else if (intel_crtc->config->port_clock > 33750)
+		stagger = 0x4;
+	else
+		stagger = 0x2;
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW11(ch));
+	val |= DPIO_TX2_STAGGER_MASK(0x1f);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW11(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW11(ch));
+	val |= DPIO_TX2_STAGGER_MASK(0x1f);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW12(ch),
+		       DPIO_LANESTAGGER_STRAP(stagger) |
+		       DPIO_LANESTAGGER_STRAP_OVRD |
+		       DPIO_TX1_STAGGER_MASK(0x1f) |
+		       DPIO_TX1_STAGGER_MULT(6) |
+		       DPIO_TX2_STAGGER_MULT(0));
+
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW12(ch),
+		       DPIO_LANESTAGGER_STRAP(stagger) |
+		       DPIO_LANESTAGGER_STRAP_OVRD |
+		       DPIO_TX1_STAGGER_MASK(0x1f) |
+		       DPIO_TX1_STAGGER_MULT(7) |
+		       DPIO_TX2_STAGGER_MULT(5));
+
+	mutex_unlock(&dev_priv->sb_lock);
 
 	intel_enable_dp(encoder);
 }
@@ -2768,7 +2819,7 @@ static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
 
 	intel_dp_prepare(encoder);
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* program left/right clock distribution */
 	if (pipe != PIPE_B) {
@@ -2818,7 +2869,7 @@ static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
 		val |= CHV_CMN_USEDCLKCHANNEL;
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW19(ch), val);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 /*
@@ -2873,8 +2924,10 @@ intel_dp_voltage_max(struct intel_dp *intel_dp)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum port port = dp_to_dig_port(intel_dp)->port;
 
-	if (INTEL_INFO(dev)->gen >= 9) {
-		if (dev_priv->vbt.edp_low_vswing && port == PORT_A)
+	if (IS_BROXTON(dev))
+		return DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
+	else if (INTEL_INFO(dev)->gen >= 9) {
+		if (dev_priv->edp_low_vswing && port == PORT_A)
 			return DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
 		return DP_TRAIN_VOLTAGE_SWING_LEVEL_2;
 	} else if (IS_VALLEYVIEW(dev))
@@ -2955,7 +3008,7 @@ intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, uint8_t voltage_swing)
 	}
 }
 
-static uint32_t intel_vlv_signal_levels(struct intel_dp *intel_dp)
+static uint32_t vlv_signal_levels(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3041,7 +3094,7 @@ static uint32_t intel_vlv_signal_levels(struct intel_dp *intel_dp)
 		return 0;
 	}
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW5(port), 0x00000000);
 	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW4(port), demph_reg_value);
 	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW2(port),
@@ -3050,12 +3103,12 @@ static uint32_t intel_vlv_signal_levels(struct intel_dp *intel_dp)
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW11(port), 0x00030000);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW9(port), preemph_reg_value);
 	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW5(port), 0x80000000);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	return 0;
 }
 
-static uint32_t intel_chv_signal_levels(struct intel_dp *intel_dp)
+static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3137,7 +3190,7 @@ static uint32_t intel_chv_signal_levels(struct intel_dp *intel_dp)
 		return 0;
 	}
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* Clear calc init */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
@@ -3224,7 +3277,7 @@ static uint32_t intel_chv_signal_levels(struct intel_dp *intel_dp)
 	val |= DPIO_LRC_BYPASS;
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	return 0;
 }
@@ -3262,7 +3315,7 @@ intel_get_adjust_train(struct intel_dp *intel_dp,
 }
 
 static uint32_t
-intel_gen4_signal_levels(uint8_t train_set)
+gen4_signal_levels(uint8_t train_set)
 {
 	uint32_t	signal_levels = 0;
 
@@ -3301,7 +3354,7 @@ intel_gen4_signal_levels(uint8_t train_set)
 
 /* Gen6's DP voltage swing and pre-emphasis control */
 static uint32_t
-intel_gen6_edp_signal_levels(uint8_t train_set)
+gen6_edp_signal_levels(uint8_t train_set)
 {
 	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
 					 DP_TRAIN_PRE_EMPHASIS_MASK);
@@ -3329,7 +3382,7 @@ intel_gen6_edp_signal_levels(uint8_t train_set)
 
 /* Gen7's DP voltage swing and pre-emphasis control */
 static uint32_t
-intel_gen7_edp_signal_levels(uint8_t train_set)
+gen7_edp_signal_levels(uint8_t train_set)
 {
 	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
 					 DP_TRAIN_PRE_EMPHASIS_MASK);
@@ -3360,7 +3413,7 @@ intel_gen7_edp_signal_levels(uint8_t train_set)
 
 /* Gen7.5's (HSW) DP voltage swing and pre-emphasis control */
 static uint32_t
-intel_hsw_signal_levels(uint8_t train_set)
+hsw_signal_levels(uint8_t train_set)
 {
 	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
 					 DP_TRAIN_PRE_EMPHASIS_MASK);
@@ -3395,6 +3448,55 @@ intel_hsw_signal_levels(uint8_t train_set)
 	}
 }
 
+static void bxt_signal_levels(struct intel_dp *intel_dp)
+{
+	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
+	enum port port = dport->port;
+	struct drm_device *dev = dport->base.base.dev;
+	struct intel_encoder *encoder = &dport->base;
+	uint8_t train_set = intel_dp->train_set[0];
+	uint32_t level = 0;
+
+	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
+					 DP_TRAIN_PRE_EMPHASIS_MASK);
+	switch (signal_levels) {
+	default:
+		DRM_DEBUG_KMS("Unsupported voltage swing/pre-emph level\n");
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_0 | DP_TRAIN_PRE_EMPH_LEVEL_0:
+		level = 0;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_0 | DP_TRAIN_PRE_EMPH_LEVEL_1:
+		level = 1;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_0 | DP_TRAIN_PRE_EMPH_LEVEL_2:
+		level = 2;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_0 | DP_TRAIN_PRE_EMPH_LEVEL_3:
+		level = 3;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_1 | DP_TRAIN_PRE_EMPH_LEVEL_0:
+		level = 4;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_1 | DP_TRAIN_PRE_EMPH_LEVEL_1:
+		level = 5;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_1 | DP_TRAIN_PRE_EMPH_LEVEL_2:
+		level = 6;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_2 | DP_TRAIN_PRE_EMPH_LEVEL_0:
+		level = 7;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_2 | DP_TRAIN_PRE_EMPH_LEVEL_1:
+		level = 8;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_LEVEL_3 | DP_TRAIN_PRE_EMPH_LEVEL_0:
+		level = 9;
+		break;
+	}
+
+	bxt_ddi_vswing_sequence(dev, level, port, encoder->type);
+}
+
 /* Properly updates "DP" with the correct signal levels. */
 static void
 intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
@@ -3405,27 +3507,38 @@ intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
 	uint32_t signal_levels, mask;
 	uint8_t train_set = intel_dp->train_set[0];
 
-	if (IS_HASWELL(dev) || IS_BROADWELL(dev) || INTEL_INFO(dev)->gen >= 9) {
-		signal_levels = intel_hsw_signal_levels(train_set);
+	if (IS_BROXTON(dev)) {
+		signal_levels = 0;
+		bxt_signal_levels(intel_dp);
+		mask = 0;
+	} else if (HAS_DDI(dev)) {
+		signal_levels = hsw_signal_levels(train_set);
 		mask = DDI_BUF_EMP_MASK;
 	} else if (IS_CHERRYVIEW(dev)) {
-		signal_levels = intel_chv_signal_levels(intel_dp);
+		signal_levels = chv_signal_levels(intel_dp);
 		mask = 0;
 	} else if (IS_VALLEYVIEW(dev)) {
-		signal_levels = intel_vlv_signal_levels(intel_dp);
+		signal_levels = vlv_signal_levels(intel_dp);
 		mask = 0;
 	} else if (IS_GEN7(dev) && port == PORT_A) {
-		signal_levels = intel_gen7_edp_signal_levels(train_set);
+		signal_levels = gen7_edp_signal_levels(train_set);
 		mask = EDP_LINK_TRAIN_VOL_EMP_MASK_IVB;
 	} else if (IS_GEN6(dev) && port == PORT_A) {
-		signal_levels = intel_gen6_edp_signal_levels(train_set);
+		signal_levels = gen6_edp_signal_levels(train_set);
 		mask = EDP_LINK_TRAIN_VOL_EMP_MASK_SNB;
 	} else {
-		signal_levels = intel_gen4_signal_levels(train_set);
+		signal_levels = gen4_signal_levels(train_set);
 		mask = DP_VOLTAGE_MASK | DP_PRE_EMPHASIS_MASK;
 	}
 
-	DRM_DEBUG_KMS("Using signal levels %08x\n", signal_levels);
+	if (mask)
+		DRM_DEBUG_KMS("Using signal levels %08x\n", signal_levels);
+
+	DRM_DEBUG_KMS("Using vswing level %d\n",
+		train_set & DP_TRAIN_VOLTAGE_SWING_MASK);
+	DRM_DEBUG_KMS("Using pre-emphasis level %d\n",
+		(train_set & DP_TRAIN_PRE_EMPHASIS_MASK) >>
+			DP_TRAIN_PRE_EMPHASIS_SHIFT);
 
 	*DP = (*DP & ~mask) | signal_levels;
 }
@@ -3467,7 +3580,8 @@ static bool
 intel_dp_reset_link_train(struct intel_dp *intel_dp, uint32_t *DP,
 			uint8_t dp_train_pat)
 {
-	memset(intel_dp->train_set, 0, sizeof(intel_dp->train_set));
+	if (!intel_dp->train_set_valid)
+		memset(intel_dp->train_set, 0, sizeof(intel_dp->train_set));
 	intel_dp_set_signal_levels(intel_dp, DP);
 	return intel_dp_set_link_train(intel_dp, DP, dp_train_pat);
 }
@@ -3580,6 +3694,23 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 			break;
 		}
 
+		/*
+		 * if we used previously trained voltage and pre-emphasis values
+		 * and we don't get clock recovery, reset link training values
+		 */
+		if (intel_dp->train_set_valid) {
+			DRM_DEBUG_KMS("clock recovery not ok, reset");
+			/* clear the flag as we are not reusing train set */
+			intel_dp->train_set_valid = false;
+			if (!intel_dp_reset_link_train(intel_dp, &DP,
+						       DP_TRAINING_PATTERN_1 |
+						       DP_LINK_SCRAMBLING_DISABLE)) {
+				DRM_ERROR("failed to enable link training\n");
+				return;
+			}
+			continue;
+		}
+
 		/* Check to see if we've tried the max voltage */
 		for (i = 0; i < intel_dp->lane_count; i++)
 			if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
@@ -3657,6 +3788,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 
 		/* Make sure clock is still ok */
 		if (!drm_dp_clock_recovery_ok(link_status, intel_dp->lane_count)) {
+			intel_dp->train_set_valid = false;
 			intel_dp_start_link_train(intel_dp);
 			intel_dp_set_link_train(intel_dp, &DP,
 						training_pattern |
@@ -3672,6 +3804,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 
 		/* Try 5 times, then try clock recovery if that fails */
 		if (tries > 5) {
+			intel_dp->train_set_valid = false;
 			intel_dp_start_link_train(intel_dp);
 			intel_dp_set_link_train(intel_dp, &DP,
 						training_pattern |
@@ -3693,9 +3826,10 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 
 	intel_dp->DP = DP;
 
-	if (channel_eq)
+	if (channel_eq) {
+		intel_dp->train_set_valid = true;
 		DRM_DEBUG_KMS("Channel EQ done. DP Training successful\n");
-
+	}
 }
 
 void intel_dp_stop_link_train(struct intel_dp *intel_dp)
@@ -3708,6 +3842,7 @@ static void
 intel_dp_link_down(struct intel_dp *intel_dp)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_crtc *crtc = to_intel_crtc(intel_dig_port->base.base.crtc);
 	enum port port = intel_dig_port->port;
 	struct drm_device *dev = intel_dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3721,36 +3856,41 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 
 	DRM_DEBUG_KMS("\n");
 
-	if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || port != PORT_A)) {
+	if ((IS_GEN7(dev) && port == PORT_A) ||
+	    (HAS_PCH_CPT(dev) && port != PORT_A)) {
 		DP &= ~DP_LINK_TRAIN_MASK_CPT;
-		I915_WRITE(intel_dp->output_reg, DP | DP_LINK_TRAIN_PAT_IDLE_CPT);
+		DP |= DP_LINK_TRAIN_PAT_IDLE_CPT;
 	} else {
 		if (IS_CHERRYVIEW(dev))
 			DP &= ~DP_LINK_TRAIN_MASK_CHV;
 		else
 			DP &= ~DP_LINK_TRAIN_MASK;
-		I915_WRITE(intel_dp->output_reg, DP | DP_LINK_TRAIN_PAT_IDLE);
+		DP |= DP_LINK_TRAIN_PAT_IDLE;
 	}
+	I915_WRITE(intel_dp->output_reg, DP);
 	POSTING_READ(intel_dp->output_reg);
 
-	if (HAS_PCH_IBX(dev) &&
-	    I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) {
-		/* Hardware workaround: leaving our transcoder select
-		 * set to transcoder B while it's off will prevent the
-		 * corresponding HDMI output on transcoder A.
-		 *
-		 * Combine this with another hardware workaround:
-		 * transcoder select bit can only be cleared while the
-		 * port is enabled.
-		 */
-		DP &= ~DP_PIPEB_SELECT;
+	DP &= ~(DP_PORT_EN | DP_AUDIO_OUTPUT_ENABLE);
+	I915_WRITE(intel_dp->output_reg, DP);
+	POSTING_READ(intel_dp->output_reg);
+
+	/*
+	 * HW workaround for IBX, we need to move the port
+	 * to transcoder A after disabling it to allow the
+	 * matching HDMI port to be enabled on transcoder A.
+	 */
+	if (HAS_PCH_IBX(dev) && crtc->pipe == PIPE_B && port != PORT_A) {
+		/* always enable with pattern 1 (as per spec) */
+		DP &= ~(DP_PIPEB_SELECT | DP_LINK_TRAIN_MASK);
+		DP |= DP_PORT_EN | DP_LINK_TRAIN_PAT_1;
+		I915_WRITE(intel_dp->output_reg, DP);
+		POSTING_READ(intel_dp->output_reg);
+
+		DP &= ~DP_PORT_EN;
 		I915_WRITE(intel_dp->output_reg, DP);
 		POSTING_READ(intel_dp->output_reg);
 	}
 
-	DP &= ~DP_AUDIO_OUTPUT_ENABLE;
-	I915_WRITE(intel_dp->output_reg, DP & ~DP_PORT_EN);
-	POSTING_READ(intel_dp->output_reg);
 	msleep(intel_dp->panel_power_down_delay);
 }
 
@@ -3781,6 +3921,21 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 			dev_priv->psr.sink_support = true;
 			DRM_DEBUG_KMS("Detected EDP PSR Panel.\n");
 		}
+
+		if (INTEL_INFO(dev)->gen >= 9 &&
+			(intel_dp->psr_dpcd[0] & DP_PSR2_IS_SUPPORTED)) {
+			uint8_t frame_sync_cap;
+
+			dev_priv->psr.sink_support = true;
+			intel_dp_dpcd_read_wake(&intel_dp->aux,
+					DP_SINK_DEVICE_AUX_FRAME_SYNC_CAP,
+					&frame_sync_cap, 1);
+			dev_priv->psr.aux_frame_sync = frame_sync_cap ? true : false;
+			/* PSR2 needs frame sync as well */
+			dev_priv->psr.psr2_support = dev_priv->psr.aux_frame_sync;
+			DRM_DEBUG_KMS("PSR2 %s on sink",
+				dev_priv->psr.psr2_support ? "supported" : "not supported");
+		}
 	}
 
 	/* Training Pattern 3 support, both source and sink */
@@ -3885,46 +4040,70 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc)
 	u8 buf;
 	int test_crc_count;
 	int attempts = 6;
+	int ret = 0;
 
-	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0)
-		return -EIO;
+	hsw_disable_ips(intel_crtc);
 
-	if (!(buf & DP_TEST_CRC_SUPPORTED))
-		return -ENOTTY;
+	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0) {
+		ret = -EIO;
+		goto out;
+	}
+
+	if (!(buf & DP_TEST_CRC_SUPPORTED)) {
+		ret = -ENOTTY;
+		goto out;
+	}
 
-	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0)
-		return -EIO;
+	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0) {
+		ret = -EIO;
+		goto out;
+	}
 
 	if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TEST_SINK,
-				buf | DP_TEST_SINK_START) < 0)
-		return -EIO;
+				buf | DP_TEST_SINK_START) < 0) {
+		ret = -EIO;
+		goto out;
+	}
+
+	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0) {
+		ret = -EIO;
+		goto out;
+	}
 
-	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0)
-		return -EIO;
 	test_crc_count = buf & DP_TEST_COUNT_MASK;
 
 	do {
 		if (drm_dp_dpcd_readb(&intel_dp->aux,
-				      DP_TEST_SINK_MISC, &buf) < 0)
-			return -EIO;
+				      DP_TEST_SINK_MISC, &buf) < 0) {
+			ret = -EIO;
+			goto out;
+		}
 		intel_wait_for_vblank(dev, intel_crtc->pipe);
 	} while (--attempts && (buf & DP_TEST_COUNT_MASK) == test_crc_count);
 
 	if (attempts == 0) {
 		DRM_DEBUG_KMS("Panel is unable to calculate CRC after 6 vblanks\n");
-		return -ETIMEDOUT;
+		ret = -ETIMEDOUT;
+		goto out;
 	}
 
-	if (drm_dp_dpcd_read(&intel_dp->aux, DP_TEST_CRC_R_CR, crc, 6) < 0)
-		return -EIO;
+	if (drm_dp_dpcd_read(&intel_dp->aux, DP_TEST_CRC_R_CR, crc, 6) < 0) {
+		ret = -EIO;
+		goto out;
+	}
 
-	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0)
-		return -EIO;
+	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0) {
+		ret = -EIO;
+		goto out;
+	}
 	if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TEST_SINK,
-			       buf & ~DP_TEST_SINK_START) < 0)
-		return -EIO;
-
-	return 0;
+			       buf & ~DP_TEST_SINK_START) < 0) {
+		ret = -EIO;
+		goto out;
+	}
+out:
+	hsw_enable_ips(intel_crtc);
+	return ret;
 }
 
 static bool
@@ -3949,11 +4128,114 @@ intel_dp_get_sink_irq_esi(struct intel_dp *intel_dp, u8 *sink_irq_vector)
 	return true;
 }
 
-static void
-intel_dp_handle_test_request(struct intel_dp *intel_dp)
+static uint8_t intel_dp_autotest_link_training(struct intel_dp *intel_dp)
+{
+	uint8_t test_result = DP_TEST_ACK;
+	return test_result;
+}
+
+static uint8_t intel_dp_autotest_video_pattern(struct intel_dp *intel_dp)
+{
+	uint8_t test_result = DP_TEST_NAK;
+	return test_result;
+}
+
+static uint8_t intel_dp_autotest_edid(struct intel_dp *intel_dp)
+{
+	uint8_t test_result = DP_TEST_NAK;
+	struct intel_connector *intel_connector = intel_dp->attached_connector;
+	struct drm_connector *connector = &intel_connector->base;
+
+	if (intel_connector->detect_edid == NULL ||
+	    connector->edid_corrupt ||
+	    intel_dp->aux.i2c_defer_count > 6) {
+		/* Check EDID read for NACKs, DEFERs and corruption
+		 * (DP CTS 1.2 Core r1.1)
+		 *    4.2.2.4 : Failed EDID read, I2C_NAK
+		 *    4.2.2.5 : Failed EDID read, I2C_DEFER
+		 *    4.2.2.6 : EDID corruption detected
+		 * Use failsafe mode for all cases
+		 */
+		if (intel_dp->aux.i2c_nack_count > 0 ||
+			intel_dp->aux.i2c_defer_count > 0)
+			DRM_DEBUG_KMS("EDID read had %d NACKs, %d DEFERs\n",
+				      intel_dp->aux.i2c_nack_count,
+				      intel_dp->aux.i2c_defer_count);
+		intel_dp->compliance_test_data = INTEL_DP_RESOLUTION_FAILSAFE;
+	} else {
+		if (!drm_dp_dpcd_write(&intel_dp->aux,
+					DP_TEST_EDID_CHECKSUM,
+					&intel_connector->detect_edid->checksum,
+					1))
+			DRM_DEBUG_KMS("Failed to write EDID checksum\n");
+
+		test_result = DP_TEST_ACK | DP_TEST_EDID_CHECKSUM_WRITE;
+		intel_dp->compliance_test_data = INTEL_DP_RESOLUTION_STANDARD;
+	}
+
+	/* Set test active flag here so userspace doesn't interrupt things */
+	intel_dp->compliance_test_active = 1;
+
+	return test_result;
+}
+
+static uint8_t intel_dp_autotest_phy_pattern(struct intel_dp *intel_dp)
+{
+	uint8_t test_result = DP_TEST_NAK;
+	return test_result;
+}
+
+static void intel_dp_handle_test_request(struct intel_dp *intel_dp)
 {
-	/* NAK by default */
-	drm_dp_dpcd_writeb(&intel_dp->aux, DP_TEST_RESPONSE, DP_TEST_NAK);
+	uint8_t response = DP_TEST_NAK;
+	uint8_t rxdata = 0;
+	int status = 0;
+
+	intel_dp->compliance_test_active = 0;
+	intel_dp->compliance_test_type = 0;
+	intel_dp->compliance_test_data = 0;
+
+	intel_dp->aux.i2c_nack_count = 0;
+	intel_dp->aux.i2c_defer_count = 0;
+
+	status = drm_dp_dpcd_read(&intel_dp->aux, DP_TEST_REQUEST, &rxdata, 1);
+	if (status <= 0) {
+		DRM_DEBUG_KMS("Could not read test request from sink\n");
+		goto update_status;
+	}
+
+	switch (rxdata) {
+	case DP_TEST_LINK_TRAINING:
+		DRM_DEBUG_KMS("LINK_TRAINING test requested\n");
+		intel_dp->compliance_test_type = DP_TEST_LINK_TRAINING;
+		response = intel_dp_autotest_link_training(intel_dp);
+		break;
+	case DP_TEST_LINK_VIDEO_PATTERN:
+		DRM_DEBUG_KMS("TEST_PATTERN test requested\n");
+		intel_dp->compliance_test_type = DP_TEST_LINK_VIDEO_PATTERN;
+		response = intel_dp_autotest_video_pattern(intel_dp);
+		break;
+	case DP_TEST_LINK_EDID_READ:
+		DRM_DEBUG_KMS("EDID test requested\n");
+		intel_dp->compliance_test_type = DP_TEST_LINK_EDID_READ;
+		response = intel_dp_autotest_edid(intel_dp);
+		break;
+	case DP_TEST_LINK_PHY_TEST_PATTERN:
+		DRM_DEBUG_KMS("PHY_PATTERN test requested\n");
+		intel_dp->compliance_test_type = DP_TEST_LINK_PHY_TEST_PATTERN;
+		response = intel_dp_autotest_phy_pattern(intel_dp);
+		break;
+	default:
+		DRM_DEBUG_KMS("Invalid test request '%02x'\n", rxdata);
+		break;
+	}
+
+update_status:
+	status = drm_dp_dpcd_write(&intel_dp->aux,
+				   DP_TEST_RESPONSE,
+				   &response, 1);
+	if (status <= 0)
+		DRM_DEBUG_KMS("Could not write test response to sink\n");
 }
 
 static int
@@ -4059,7 +4341,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 				   sink_irq_vector);
 
 		if (sink_irq_vector & DP_AUTOMATED_TEST_REQUEST)
-			intel_dp_handle_test_request(intel_dp);
+			DRM_DEBUG_DRIVER("Test request in short pulse not handled\n");
 		if (sink_irq_vector & (DP_CP_IRQ | DP_SINK_SPECIFIC_IRQ))
 			DRM_DEBUG_DRIVER("CP or sink specific irq unhandled\n");
 	}
@@ -4289,6 +4571,7 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status status;
 	enum intel_display_power_domain power_domain;
 	bool ret;
+	u8 sink_irq_vector;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
@@ -4331,6 +4614,20 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 		intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
 	status = connector_status_connected;
 
+	/* Try to read the source of the interrupt */
+	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
+	    intel_dp_get_sink_irq(intel_dp, &sink_irq_vector)) {
+		/* Clear interrupt source */
+		drm_dp_dpcd_writeb(&intel_dp->aux,
+				   DP_DEVICE_SERVICE_IRQ_VECTOR,
+				   sink_irq_vector);
+
+		if (sink_irq_vector & DP_AUTOMATED_TEST_REQUEST)
+			intel_dp_handle_test_request(intel_dp);
+		if (sink_irq_vector & (DP_CP_IRQ | DP_SINK_SPECIFIC_IRQ))
+			DRM_DEBUG_DRIVER("CP or sink specific irq unhandled\n");
+	}
+
 out:
 	intel_dp_power_put(intel_dp, power_domain);
 	return status;
@@ -4661,6 +4958,8 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
 	intel_display_power_get(dev_priv, power_domain);
 
 	if (long_hpd) {
+		/* indicate that we need to restart link training */
+		intel_dp->train_set_valid = false;
 
 		if (HAS_PCH_SPLIT(dev)) {
 			if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
@@ -5539,12 +5838,10 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 	intel_dp_aux_init(intel_dp, intel_connector);
 
 	/* init MST on ports that can support it */
-	if (IS_HASWELL(dev) || IS_BROADWELL(dev) || INTEL_INFO(dev)->gen >= 9) {
-		if (port == PORT_B || port == PORT_C || port == PORT_D) {
-			intel_dp_mst_encoder_init(intel_dig_port,
-						  intel_connector->base.base.id);
-		}
-	}
+	if (HAS_DP_MST(dev) &&
+	    (port == PORT_B || port == PORT_C || port == PORT_D))
+		intel_dp_mst_encoder_init(intel_dig_port,
+					  intel_connector->base.base.id);
 
 	if (!intel_edp_init_connector(intel_dp, intel_connector)) {
 		drm_dp_aux_unregister(&intel_dp->aux);
@@ -5574,6 +5871,8 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 		I915_WRITE(PEG_BAND_GAP_DATA, (temp & ~0xf) | 0xd);
 	}
 
+	i915_debugfs_connector_add(connector);
+
 	return true;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index 5cb47482d29f..6e4cc5334f47 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -40,7 +40,9 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	int bpp, i;
 	int lane_count, slots, rate;
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
-	struct intel_connector *found = NULL;
+	struct drm_connector *drm_connector;
+	struct intel_connector *connector, *found = NULL;
+	struct drm_connector_state *connector_state;
 	int mst_pbn;
 
 	pipe_config->dp_encoder_is_mst = true;
@@ -70,12 +72,11 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 
 	state = pipe_config->base.state;
 
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i])
-			continue;
+	for_each_connector_in_state(state, drm_connector, connector_state, i) {
+		connector = to_intel_connector(drm_connector);
 
-		if (state->connector_states[i]->best_encoder == &encoder->base) {
-			found = to_intel_connector(state->connectors[i]);
+		if (connector_state->best_encoder == &encoder->base) {
+			found = connector;
 			break;
 		}
 	}
@@ -150,14 +151,14 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder)
 	enum port port = intel_dig_port->port;
 	int ret;
 	uint32_t temp;
-	struct intel_connector *found = NULL, *intel_connector;
+	struct intel_connector *found = NULL, *connector;
 	int slots;
 	struct drm_crtc *crtc = encoder->base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
-	for_each_intel_connector(dev, intel_connector) {
-		if (intel_connector->new_encoder == encoder) {
-			found = intel_connector;
+	for_each_intel_connector(dev, connector) {
+		if (connector->base.state->best_encoder == &encoder->base) {
+			found = connector;
 			break;
 		}
 	}
@@ -173,8 +174,10 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder)
 	if (intel_dp->active_mst_links == 0) {
 		enum port port = intel_ddi_get_encoder_port(encoder);
 
-		I915_WRITE(PORT_CLK_SEL(port),
-			   intel_crtc->config->ddi_pll_sel);
+		/* FIXME: add support for SKL */
+		if (INTEL_INFO(dev)->gen < 9)
+			I915_WRITE(PORT_CLK_SEL(port),
+				   intel_crtc->config->ddi_pll_sel);
 
 		intel_ddi_init_dp_buf_reg(&intel_dig_port->base);
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 897f17db08af..2afb31a46275 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -249,10 +249,24 @@ struct intel_plane_state {
 	bool visible;
 
 	/*
-	 * used only for sprite planes to determine when to implicitly
-	 * enable/disable the primary plane
+	 * scaler_id
+	 *    = -1 : not using a scaler
+	 *    >=  0 : using a scalers
+	 *
+	 * plane requiring a scaler:
+	 *   - During check_plane, its bit is set in
+	 *     crtc_state->scaler_state.scaler_users by calling helper function
+	 *     update_scaler_users.
+	 *   - scaler_id indicates the scaler it got assigned.
+	 *
+	 * plane doesn't require a scaler:
+	 *   - this can happen when scaling is no more required or plane simply
+	 *     got disabled.
+	 *   - During check_plane, corresponding bit is reset in
+	 *     crtc_state->scaler_state.scaler_users by calling helper function
+	 *     update_scaler_users.
 	 */
-	bool hides_primary;
+	int scaler_id;
 };
 
 struct intel_initial_plane_config {
@@ -262,6 +276,49 @@ struct intel_initial_plane_config {
 	u32 base;
 };
 
+#define SKL_MIN_SRC_W 8
+#define SKL_MAX_SRC_W 4096
+#define SKL_MIN_SRC_H 8
+#define SKL_MAX_SRC_H 4096
+#define SKL_MIN_DST_W 8
+#define SKL_MAX_DST_W 4096
+#define SKL_MIN_DST_H 8
+#define SKL_MAX_DST_H 4096
+
+struct intel_scaler {
+	int id;
+	int in_use;
+	uint32_t mode;
+};
+
+struct intel_crtc_scaler_state {
+#define SKL_NUM_SCALERS 2
+	struct intel_scaler scalers[SKL_NUM_SCALERS];
+
+	/*
+	 * scaler_users: keeps track of users requesting scalers on this crtc.
+	 *
+	 *     If a bit is set, a user is using a scaler.
+	 *     Here user can be a plane or crtc as defined below:
+	 *       bits 0-30 - plane (bit position is index from drm_plane_index)
+	 *       bit 31    - crtc
+	 *
+	 * Instead of creating a new index to cover planes and crtc, using
+	 * existing drm_plane_index for planes which is well less than 31
+	 * planes and bit 31 for crtc. This should be fine to cover all
+	 * our platforms.
+	 *
+	 * intel_atomic_setup_scalers will setup available scalers to users
+	 * requesting scalers. It will gracefully fail if request exceeds
+	 * avilability.
+	 */
+#define SKL_CRTC_INDEX 31
+	unsigned scaler_users;
+
+	/* scaler used by crtc for panel fitting purpose */
+	int scaler_id;
+};
+
 struct intel_crtc_state {
 	struct drm_crtc_state base;
 
@@ -388,6 +445,8 @@ struct intel_crtc_state {
 
 	bool dp_encoder_is_mst;
 	int pbn;
+
+	struct intel_crtc_scaler_state scaler_state;
 };
 
 struct intel_pipe_wm {
@@ -400,8 +459,10 @@ struct intel_pipe_wm {
 };
 
 struct intel_mmio_flip {
-	struct drm_i915_gem_request *req;
 	struct work_struct work;
+	struct drm_i915_private *i915;
+	struct drm_i915_gem_request *req;
+	struct intel_crtc *crtc;
 };
 
 struct skl_pipe_wm {
@@ -448,7 +509,6 @@ struct intel_crtc {
 	 */
 	bool active;
 	unsigned long enabled_power_domains;
-	bool primary_enabled; /* is the primary plane (partially) visible? */
 	bool lowfreq_avail;
 	struct intel_overlay *overlay;
 	struct intel_unpin_work *unpin_work;
@@ -468,7 +528,6 @@ struct intel_crtc {
 
 	struct intel_initial_plane_config plane_config;
 	struct intel_crtc_state *config;
-	struct intel_crtc_state *new_config;
 	bool new_enabled;
 
 	/* reset counter value when the last flip was submitted */
@@ -487,15 +546,25 @@ struct intel_crtc {
 	} wm;
 
 	int scanline_offset;
-	struct intel_mmio_flip mmio_flip;
 
 	struct intel_crtc_atomic_commit atomic;
+
+	/* scalers available on this crtc */
+	int num_scalers;
 };
 
 struct intel_plane_wm_parameters {
 	uint32_t horiz_pixels;
 	uint32_t vert_pixels;
+	/*
+	 *   For packed pixel formats:
+	 *     bytes_per_pixel - holds bytes per pixel
+	 *   For planar pixel formats:
+	 *     bytes_per_pixel - holds bytes per pixel for uv-plane
+	 *     y_bytes_per_pixel - holds bytes per pixel for y-plane
+	 */
 	uint8_t bytes_per_pixel;
+	uint8_t y_bytes_per_pixel;
 	bool enabled;
 	bool scaled;
 	u64 tiling;
@@ -533,7 +602,7 @@ struct intel_plane {
 			     uint32_t x, uint32_t y,
 			     uint32_t src_w, uint32_t src_h);
 	void (*disable_plane)(struct drm_plane *plane,
-			      struct drm_crtc *crtc);
+			      struct drm_crtc *crtc, bool force);
 	int (*check_plane)(struct drm_plane *plane,
 			   struct intel_plane_state *state);
 	void (*commit_plane)(struct drm_plane *plane,
@@ -669,6 +738,12 @@ struct intel_dp {
 				     bool has_aux_irq,
 				     int send_bytes,
 				     uint32_t aux_clock_divider);
+	bool train_set_valid;
+
+	/* Displayport compliance testing */
+	unsigned long compliance_test_type;
+	unsigned long compliance_test_data;
+	bool compliance_test_active;
 };
 
 struct intel_digital_port {
@@ -747,15 +822,6 @@ struct intel_unpin_work {
 	bool enable_stall_check;
 };
 
-struct intel_set_config {
-	struct drm_encoder **save_connector_encoders;
-	struct drm_crtc **save_encoder_crtcs;
-	bool *save_crtc_enabled;
-
-	bool fb_changed;
-	bool mode_changed;
-};
-
 struct intel_load_detect_pipe {
 	struct drm_framebuffer *release_fb;
 	bool load_detect_temp;
@@ -852,7 +918,6 @@ void hsw_fdi_link_train(struct drm_crtc *crtc);
 void intel_ddi_init(struct drm_device *dev, enum port port);
 enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder);
 bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe);
-int intel_ddi_get_cdclk_freq(struct drm_i915_private *dev_priv);
 void intel_ddi_pll_init(struct drm_device *dev);
 void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc);
 void intel_ddi_disable_transcoder_func(struct drm_i915_private *dev_priv,
@@ -867,11 +932,15 @@ bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector);
 void intel_ddi_fdi_disable(struct drm_crtc *crtc);
 void intel_ddi_get_config(struct intel_encoder *encoder,
 			  struct intel_crtc_state *pipe_config);
+struct intel_encoder *
+intel_ddi_get_crtc_new_encoder(struct intel_crtc_state *crtc_state);
 
 void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder);
 void intel_ddi_clock_get(struct intel_encoder *encoder,
 			 struct intel_crtc_state *pipe_config);
 void intel_ddi_set_vc_payload_alloc(struct drm_crtc *crtc, bool state);
+void bxt_ddi_vswing_sequence(struct drm_device *dev, u32 level,
+				enum port port, int type);
 
 /* intel_frontbuffer.c */
 void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
@@ -925,6 +994,7 @@ void intel_mark_busy(struct drm_device *dev);
 void intel_mark_idle(struct drm_device *dev);
 void intel_crtc_restore_mode(struct drm_crtc *crtc);
 void intel_crtc_control(struct drm_crtc *crtc, bool enable);
+void intel_crtc_reset(struct intel_crtc *crtc);
 void intel_crtc_update_dpms(struct drm_crtc *crtc);
 void intel_encoder_destroy(struct drm_encoder *encoder);
 int intel_connector_init(struct intel_connector *);
@@ -952,7 +1022,8 @@ intel_wait_for_vblank(struct drm_device *dev, int pipe)
 }
 int ironlake_get_lanes_required(int target_clock, int link_bw, int bpp);
 void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
-			 struct intel_digital_port *dport);
+			 struct intel_digital_port *dport,
+			 unsigned int expected_mask);
 bool intel_get_load_detect_pipe(struct drm_connector *connector,
 				struct drm_display_mode *mode,
 				struct intel_load_detect_pipe *old,
@@ -997,6 +1068,9 @@ intel_rotation_90_or_270(unsigned int rotation)
 	return rotation & (BIT(DRM_ROTATE_90) | BIT(DRM_ROTATE_270));
 }
 
+void intel_create_rotation_property(struct drm_device *dev,
+					struct intel_plane *plane);
+
 bool intel_wm_need_update(struct drm_plane *plane,
 			  struct drm_plane_state *state);
 
@@ -1037,6 +1111,15 @@ void intel_prepare_reset(struct drm_device *dev);
 void intel_finish_reset(struct drm_device *dev);
 void hsw_enable_pc8(struct drm_i915_private *dev_priv);
 void hsw_disable_pc8(struct drm_i915_private *dev_priv);
+void broxton_init_cdclk(struct drm_device *dev);
+void broxton_uninit_cdclk(struct drm_device *dev);
+void broxton_set_cdclk(struct drm_device *dev, int frequency);
+void broxton_ddi_phy_init(struct drm_device *dev);
+void broxton_ddi_phy_uninit(struct drm_device *dev);
+void bxt_enable_dc9(struct drm_i915_private *dev_priv);
+void bxt_disable_dc9(struct drm_i915_private *dev_priv);
+void skl_init_cdclk(struct drm_i915_private *dev_priv);
+void skl_uninit_cdclk(struct drm_i915_private *dev_priv);
 void intel_dp_get_m_n(struct intel_crtc *crtc,
 		      struct intel_crtc_state *pipe_config);
 void intel_dp_set_m_n(struct intel_crtc *crtc, enum link_m_n_set m_n);
@@ -1044,6 +1127,8 @@ int intel_dotclock_calculate(int link_freq, const struct intel_link_m_n *m_n);
 void
 ironlake_check_encoder_dotclock(const struct intel_crtc_state *pipe_config,
 				int dotclock);
+bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state, int target_clock,
+			intel_clock_t *best_clock);
 bool intel_crtc_active(struct drm_crtc *crtc);
 void hsw_enable_ips(struct intel_crtc *crtc);
 void hsw_disable_ips(struct intel_crtc *crtc);
@@ -1053,9 +1138,26 @@ void intel_mode_from_pipe_config(struct drm_display_mode *mode,
 				 struct intel_crtc_state *pipe_config);
 void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc);
 void intel_modeset_preclose(struct drm_device *dev, struct drm_file *file);
+void skl_detach_scalers(struct intel_crtc *intel_crtc);
+int skl_update_scaler_users(struct intel_crtc *intel_crtc,
+	struct intel_crtc_state *crtc_state, struct intel_plane *intel_plane,
+	struct intel_plane_state *plane_state, int force_detach);
+int skl_max_scale(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state);
 
 unsigned long intel_plane_obj_offset(struct intel_plane *intel_plane,
 				     struct drm_i915_gem_object *obj);
+u32 skl_plane_ctl_format(uint32_t pixel_format);
+u32 skl_plane_ctl_tiling(uint64_t fb_modifier);
+u32 skl_plane_ctl_rotation(unsigned int rotation);
+
+/* intel_csr.c */
+void intel_csr_ucode_init(struct drm_device *dev);
+enum csr_state intel_csr_load_status_get(struct drm_i915_private *dev_priv);
+void intel_csr_load_status_set(struct drm_i915_private *dev_priv,
+					enum csr_state state);
+void intel_csr_load_program(struct drm_device *dev);
+void intel_csr_ucode_fini(struct drm_device *dev);
+void assert_csr_loaded(struct drm_i915_private *dev_priv);
 
 /* intel_dp.c */
 void intel_dp_init(struct drm_device *dev, int output_reg, enum port port);
@@ -1215,6 +1317,7 @@ void intel_psr_invalidate(struct drm_device *dev,
 void intel_psr_flush(struct drm_device *dev,
 			 unsigned frontbuffer_bits);
 void intel_psr_init(struct drm_device *dev);
+void intel_psr_single_frame_update(struct drm_device *dev);
 
 /* intel_runtime_pm.c */
 int intel_power_domains_init(struct drm_i915_private *);
@@ -1263,7 +1366,11 @@ void gen6_update_ring_freq(struct drm_device *dev);
 void gen6_rps_busy(struct drm_i915_private *dev_priv);
 void gen6_rps_reset_ei(struct drm_i915_private *dev_priv);
 void gen6_rps_idle(struct drm_i915_private *dev_priv);
-void gen6_rps_boost(struct drm_i915_private *dev_priv);
+void gen6_rps_boost(struct drm_i915_private *dev_priv,
+		    struct intel_rps_client *rps,
+		    unsigned long submitted);
+void intel_queue_rps_boost_for_request(struct drm_device *dev,
+				       struct drm_i915_gem_request *req);
 void ilk_wm_get_hw_state(struct drm_device *dev);
 void skl_wm_get_hw_state(struct drm_device *dev);
 void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
@@ -1276,16 +1383,12 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob);
 
 /* intel_sprite.c */
 int intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane);
-void intel_flush_primary_plane(struct drm_i915_private *dev_priv,
-			       enum plane plane);
 int intel_plane_restore(struct drm_plane *plane);
 int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
 			      struct drm_file *file_priv);
 bool intel_pipe_update_start(struct intel_crtc *crtc,
 			     uint32_t *start_vbl_count);
 void intel_pipe_update_end(struct intel_crtc *crtc, u32 start_vbl_count);
-void intel_post_enable_primary(struct drm_crtc *crtc);
-void intel_pre_disable_primary(struct drm_crtc *crtc);
 
 /* intel_tv.c */
 void intel_tv_init(struct drm_device *dev);
@@ -1310,10 +1413,13 @@ intel_atomic_get_crtc_state(struct drm_atomic_state *state,
 	struct drm_crtc_state *crtc_state;
 	crtc_state = drm_atomic_get_crtc_state(state, &crtc->base);
 	if (IS_ERR(crtc_state))
-		return ERR_PTR(PTR_ERR(crtc_state));
+		return ERR_CAST(crtc_state);
 
 	return to_intel_crtc_state(crtc_state);
 }
+int intel_atomic_setup_scalers(struct drm_device *dev,
+	struct intel_crtc *intel_crtc,
+	struct intel_crtc_state *crtc_state);
 
 /* intel_atomic_plane.c */
 struct intel_plane_state *intel_create_plane_state(struct drm_plane *plane);
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 51966426addf..b5a5558ecd63 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -239,7 +239,7 @@ static int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd, bool hs,
 
 static void band_gap_reset(struct drm_i915_private *dev_priv)
 {
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	vlv_flisdsi_write(dev_priv, 0x08, 0x0001);
 	vlv_flisdsi_write(dev_priv, 0x0F, 0x0005);
@@ -248,7 +248,7 @@ static void band_gap_reset(struct drm_i915_private *dev_priv)
 	vlv_flisdsi_write(dev_priv, 0x0F, 0x0000);
 	vlv_flisdsi_write(dev_priv, 0x08, 0x0000);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static inline bool is_vid_mode(struct intel_dsi *intel_dsi)
@@ -346,11 +346,11 @@ static void intel_dsi_device_ready(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	/* program rcomp for compliance, reduce from 50 ohms to 45 ohms
 	 * needed everytime after power gate */
 	vlv_flisdsi_write(dev_priv, 0x04, 0x0004);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	/* bandgap reset is needed after everytime we do power gate */
 	band_gap_reset(dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
index d2cd8d5b27a1..a5e99ac305da 100644
--- a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
+++ b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
@@ -212,7 +212,7 @@ static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
 	function = gtable[gpio].function_reg;
 	pad = gtable[gpio].pad_reg;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	if (!gtable[gpio].init) {
 		/* program the function */
 		/* FIXME: remove constant below */
@@ -224,7 +224,7 @@ static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
 
 	/* pull up/down */
 	vlv_gpio_nc_write(dev_priv, pad, val);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	return data;
 }
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index 3622d0bafdf8..d20cf37b6901 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -162,59 +162,41 @@ static u32 dsi_clk_from_pclk(u32 pclk, int pixel_format, int lane_count)
 
 #endif
 
-static int dsi_calc_mnp(u32 dsi_clk, struct dsi_mnp *dsi_mnp)
+static int dsi_calc_mnp(int target_dsi_clk, struct dsi_mnp *dsi_mnp)
 {
-	u32 m, n, p;
-	u32 ref_clk;
-	u32 error;
-	u32 tmp_error;
-	int target_dsi_clk;
-	int calc_dsi_clk;
-	u32 calc_m;
-	u32 calc_p;
+	unsigned int calc_m = 0, calc_p = 0;
+	unsigned int m, n = 1, p;
+	int ref_clk = 25000;
+	int delta = target_dsi_clk;
 	u32 m_seed;
 
-	/* dsi_clk is expected in KHZ */
-	if (dsi_clk < 300000 || dsi_clk > 1150000) {
+	/* target_dsi_clk is expected in kHz */
+	if (target_dsi_clk < 300000 || target_dsi_clk > 1150000) {
 		DRM_ERROR("DSI CLK Out of Range\n");
 		return -ECHRNG;
 	}
 
-	ref_clk = 25000;
-	target_dsi_clk = dsi_clk;
-	error = 0xFFFFFFFF;
-	tmp_error = 0xFFFFFFFF;
-	calc_m = 0;
-	calc_p = 0;
-
-	for (m = 62; m <= 92; m++) {
-		for (p = 2; p <= 6; p++) {
-			/* Find the optimal m and p divisors
-			   with minimal error +/- the required clock */
-			calc_dsi_clk = (m * ref_clk) / p;
-			if (calc_dsi_clk == target_dsi_clk) {
-				calc_m = m;
-				calc_p = p;
-				error = 0;
-				break;
-			} else
-				tmp_error = abs(target_dsi_clk - calc_dsi_clk);
-
-			if (tmp_error < error) {
-				error = tmp_error;
+	for (m = 62; m <= 92 && delta; m++) {
+		for (p = 2; p <= 6 && delta; p++) {
+			/*
+			 * Find the optimal m and p divisors with minimal delta
+			 * +/- the required clock
+			 */
+			int calc_dsi_clk = (m * ref_clk) / (p * n);
+			int d = abs(target_dsi_clk - calc_dsi_clk);
+			if (d < delta) {
+				delta = d;
 				calc_m = m;
 				calc_p = p;
 			}
 		}
-
-		if (error == 0)
-			break;
 	}
 
+	/* register has log2(N1), this works fine for powers of two */
+	n = ffs(n) - 1;
 	m_seed = lfsr_converts[calc_m - 62];
-	n = 1;
 	dsi_mnp->dsi_pll_ctrl = 1 << (DSI_PLL_P1_POST_DIV_SHIFT + calc_p - 2);
-	dsi_mnp->dsi_pll_div = (n - 1) << DSI_PLL_N1_DIV_SHIFT |
+	dsi_mnp->dsi_pll_div = n << DSI_PLL_N1_DIV_SHIFT |
 		m_seed << DSI_PLL_M1_DIV_SHIFT;
 
 	return 0;
@@ -262,7 +244,7 @@ void vlv_enable_dsi_pll(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	vlv_configure_dsi_pll(encoder);
 
@@ -276,11 +258,11 @@ void vlv_enable_dsi_pll(struct intel_encoder *encoder)
 	if (wait_for(vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL) &
 						DSI_PLL_LOCK, 20)) {
 
-		mutex_unlock(&dev_priv->dpio_lock);
+		mutex_unlock(&dev_priv->sb_lock);
 		DRM_ERROR("DSI PLL lock failed\n");
 		return;
 	}
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	DRM_DEBUG_KMS("DSI PLL locked\n");
 }
@@ -292,14 +274,14 @@ void vlv_disable_dsi_pll(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	tmp = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
 	tmp &= ~DSI_PLL_VCO_EN;
 	tmp |= DSI_PLL_LDO_GATE;
 	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, tmp);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void assert_bpp_mismatch(int pixel_format, int pipe_bpp)
@@ -331,21 +313,25 @@ u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	u32 dsi_clock, pclk;
 	u32 pll_ctl, pll_div;
-	u32 m = 0, p = 0;
+	u32 m = 0, p = 0, n;
 	int refclk = 25000;
 	int i;
 
 	DRM_DEBUG_KMS("\n");
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	pll_ctl = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
 	pll_div = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_DIVIDER);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	/* mask out other bits and extract the P1 divisor */
 	pll_ctl &= DSI_PLL_P1_POST_DIV_MASK;
 	pll_ctl = pll_ctl >> (DSI_PLL_P1_POST_DIV_SHIFT - 2);
 
+	/* N1 divisor */
+	n = (pll_div & DSI_PLL_N1_DIV_MASK) >> DSI_PLL_N1_DIV_SHIFT;
+	n = 1 << n; /* register has log2(N1) */
+
 	/* mask out the other bits and extract the M1 divisor */
 	pll_div &= DSI_PLL_M1_DIV_MASK;
 	pll_div = pll_div >> DSI_PLL_M1_DIV_SHIFT;
@@ -373,7 +359,7 @@ u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
 
 	m = i + 62;
 
-	dsi_clock = (m * refclk) / p;
+	dsi_clock = (m * refclk) / (p * n);
 
 	/* pixel_format and pipe_bpp should agree */
 	assert_bpp_mismatch(intel_dsi->pixel_format, pipe_bpp);
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 770040ff486e..ece5bd754f85 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -80,7 +80,7 @@ static const struct intel_dvo_device intel_dvo_devices[] = {
 		.name = "ch7017",
 		.dvo_reg = DVOC,
 		.slave_addr = 0x75,
-		.gpio = GMBUS_PORT_DPB,
+		.gpio = GMBUS_PIN_DPB,
 		.dev_ops = &ch7017_ops,
 	},
 	{
@@ -364,7 +364,7 @@ static int intel_dvo_get_modes(struct drm_connector *connector)
 	 * that's not the case.
 	 */
 	intel_ddc_get_modes(connector,
-			    intel_gmbus_get_adapter(dev_priv, GMBUS_PORT_DPC));
+			    intel_gmbus_get_adapter(dev_priv, GMBUS_PIN_DPC));
 	if (!list_empty(&connector->probed_modes))
 		return 1;
 
@@ -495,17 +495,19 @@ void intel_dvo_init(struct drm_device *dev)
 		struct i2c_adapter *i2c;
 		int gpio;
 		bool dvoinit;
+		enum pipe pipe;
+		uint32_t dpll[I915_MAX_PIPES];
 
 		/* Allow the I2C driver info to specify the GPIO to be used in
 		 * special cases, but otherwise default to what's defined
 		 * in the spec.
 		 */
-		if (intel_gmbus_is_port_valid(dvo->gpio))
+		if (intel_gmbus_is_valid_pin(dev_priv, dvo->gpio))
 			gpio = dvo->gpio;
 		else if (dvo->type == INTEL_DVO_CHIP_LVDS)
-			gpio = GMBUS_PORT_SSC;
+			gpio = GMBUS_PIN_SSC;
 		else
-			gpio = GMBUS_PORT_DPB;
+			gpio = GMBUS_PIN_DPB;
 
 		/* Set up the I2C bus necessary for the chip we're probing.
 		 * It appears that everything is on GPIOE except for panels
@@ -520,8 +522,23 @@ void intel_dvo_init(struct drm_device *dev)
 		 */
 		intel_gmbus_force_bit(i2c, true);
 
+		/* ns2501 requires the DVO 2x clock before it will
+		 * respond to i2c accesses, so make sure we have
+		 * have the clock enabled before we attempt to
+		 * initialize the device.
+		 */
+		for_each_pipe(dev_priv, pipe) {
+			dpll[pipe] = I915_READ(DPLL(pipe));
+			I915_WRITE(DPLL(pipe), dpll[pipe] | DPLL_DVO_2X_MODE);
+		}
+
 		dvoinit = dvo->dev_ops->init(&intel_dvo->dev, i2c);
 
+		/* restore the DVO 2x clock state to original */
+		for_each_pipe(dev_priv, pipe) {
+			I915_WRITE(DPLL(pipe), dpll[pipe]);
+		}
+
 		intel_gmbus_force_bit(i2c, false);
 
 		if (!dvoinit)
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 4165ce0644f7..6abb83432d4d 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -457,7 +457,7 @@ static struct drm_crtc *intel_fbc_find_crtc(struct drm_i915_private *dev_priv)
 		tmp_crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 
 		if (intel_crtc_active(tmp_crtc) &&
-		    to_intel_crtc(tmp_crtc)->primary_enabled) {
+		    to_intel_plane_state(tmp_crtc->primary->state)->visible) {
 			if (one_pipe_only && crtc) {
 				if (set_no_fbc_reason(dev_priv, FBC_MULTIPLE_PIPES))
 					DRM_DEBUG_KMS("more than one pipe active, disabling compression\n");
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 4e7e7da2e03b..6372cfc7d053 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -96,6 +96,32 @@ static int intel_fbdev_blank(int blank, struct fb_info *info)
 	return ret;
 }
 
+static int intel_fbdev_pan_display(struct fb_var_screeninfo *var,
+				   struct fb_info *info)
+{
+	struct drm_fb_helper *fb_helper = info->par;
+	struct intel_fbdev *ifbdev =
+		container_of(fb_helper, struct intel_fbdev, helper);
+
+	int ret;
+	ret = drm_fb_helper_pan_display(var, info);
+
+	if (ret == 0) {
+		/*
+		 * FIXME: fbdev presumes that all callbacks also work from
+		 * atomic contexts and relies on that for emergency oops
+		 * printing. KMS totally doesn't do that and the locking here is
+		 * by far not the only place this goes wrong.  Ignore this for
+		 * now until we solve this for real.
+		 */
+		mutex_lock(&fb_helper->dev->struct_mutex);
+		intel_fb_obj_invalidate(ifbdev->fb->obj, NULL, ORIGIN_GTT);
+		mutex_unlock(&fb_helper->dev->struct_mutex);
+	}
+
+	return ret;
+}
+
 static struct fb_ops intelfb_ops = {
 	.owner = THIS_MODULE,
 	.fb_check_var = drm_fb_helper_check_var,
@@ -103,7 +129,7 @@ static struct fb_ops intelfb_ops = {
 	.fb_fillrect = cfb_fillrect,
 	.fb_copyarea = cfb_copyarea,
 	.fb_imageblit = cfb_imageblit,
-	.fb_pan_display = drm_fb_helper_pan_display,
+	.fb_pan_display = intel_fbdev_pan_display,
 	.fb_blank = intel_fbdev_blank,
 	.fb_setcmap = drm_fb_helper_setcmap,
 	.fb_debug_enter = drm_fb_helper_debug_enter,
diff --git a/drivers/gpu/drm/i915/intel_frontbuffer.c b/drivers/gpu/drm/i915/intel_frontbuffer.c
index a20cffb78c0f..57095f54c1f2 100644
--- a/drivers/gpu/drm/i915/intel_frontbuffer.c
+++ b/drivers/gpu/drm/i915/intel_frontbuffer.c
@@ -243,6 +243,8 @@ void intel_frontbuffer_flip_prepare(struct drm_device *dev,
 	/* Remove stale busy bits due to the old buffer. */
 	dev_priv->fb_tracking.busy_bits &= ~frontbuffer_bits;
 	mutex_unlock(&dev_priv->fb_tracking.lock);
+
+	intel_psr_single_frame_update(dev);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index bfbe07b6ddce..e97731aab6dc 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -223,10 +223,14 @@ static bool ibx_infoframe_enabled(struct drm_encoder *encoder)
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
 	int reg = TVIDEO_DIP_CTL(intel_crtc->pipe);
 	u32 val = I915_READ(reg);
 
-	return val & VIDEO_DIP_ENABLE;
+	if (VIDEO_DIP_PORT(intel_dig_port->port) == (val & VIDEO_DIP_PORT_MASK))
+		return val & VIDEO_DIP_ENABLE;
+
+	return false;
 }
 
 static void cpt_write_infoframe(struct drm_encoder *encoder,
@@ -324,10 +328,14 @@ static bool vlv_infoframe_enabled(struct drm_encoder *encoder)
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
 	int reg = VLV_TVIDEO_DIP_CTL(intel_crtc->pipe);
 	u32 val = I915_READ(reg);
 
-	return val & VIDEO_DIP_ENABLE;
+	if (VIDEO_DIP_PORT(intel_dig_port->port) == (val & VIDEO_DIP_PORT_MASK))
+		return val & VIDEO_DIP_ENABLE;
+
+	return false;
 }
 
 static void hsw_write_infoframe(struct drm_encoder *encoder,
@@ -865,59 +873,59 @@ static void intel_disable_hdmi(struct intel_encoder *encoder)
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	u32 temp;
-	u32 enable_bits = SDVO_ENABLE | SDVO_AUDIO_ENABLE;
-
-	if (crtc->config->has_audio)
-		intel_audio_codec_disable(encoder);
 
 	temp = I915_READ(intel_hdmi->hdmi_reg);
 
-	/* HW workaround for IBX, we need to move the port to transcoder A
-	 * before disabling it. */
-	if (HAS_PCH_IBX(dev)) {
-		struct drm_crtc *crtc = encoder->base.crtc;
-		int pipe = crtc ? to_intel_crtc(crtc)->pipe : -1;
-
-		if (temp & SDVO_PIPE_B_SELECT) {
-			temp &= ~SDVO_PIPE_B_SELECT;
-			I915_WRITE(intel_hdmi->hdmi_reg, temp);
-			POSTING_READ(intel_hdmi->hdmi_reg);
-
-			/* Again we need to write this twice. */
-			I915_WRITE(intel_hdmi->hdmi_reg, temp);
-			POSTING_READ(intel_hdmi->hdmi_reg);
-
-			/* Transcoder selection bits only update
-			 * effectively on vblank. */
-			if (crtc)
-				intel_wait_for_vblank(dev, pipe);
-			else
-				msleep(50);
-		}
-	}
-
-	/* HW workaround, need to toggle enable bit off and on for 12bpc, but
-	 * we do this anyway which shows more stable in testing.
-	 */
-	if (HAS_PCH_SPLIT(dev)) {
-		I915_WRITE(intel_hdmi->hdmi_reg, temp & ~SDVO_ENABLE);
-		POSTING_READ(intel_hdmi->hdmi_reg);
-	}
-
-	temp &= ~enable_bits;
-
+	temp &= ~(SDVO_ENABLE | SDVO_AUDIO_ENABLE);
 	I915_WRITE(intel_hdmi->hdmi_reg, temp);
 	POSTING_READ(intel_hdmi->hdmi_reg);
 
-	/* HW workaround, need to write this twice for issue that may result
-	 * in first write getting masked.
+	/*
+	 * HW workaround for IBX, we need to move the port
+	 * to transcoder A after disabling it to allow the
+	 * matching DP port to be enabled on transcoder A.
 	 */
-	if (HAS_PCH_SPLIT(dev)) {
+	if (HAS_PCH_IBX(dev) && crtc->pipe == PIPE_B) {
+		temp &= ~SDVO_PIPE_B_SELECT;
+		temp |= SDVO_ENABLE;
+		/*
+		 * HW workaround, need to write this twice for issue
+		 * that may result in first write getting masked.
+		 */
+		I915_WRITE(intel_hdmi->hdmi_reg, temp);
+		POSTING_READ(intel_hdmi->hdmi_reg);
+		I915_WRITE(intel_hdmi->hdmi_reg, temp);
+		POSTING_READ(intel_hdmi->hdmi_reg);
+
+		temp &= ~SDVO_ENABLE;
 		I915_WRITE(intel_hdmi->hdmi_reg, temp);
 		POSTING_READ(intel_hdmi->hdmi_reg);
 	}
 }
 
+static void g4x_disable_hdmi(struct intel_encoder *encoder)
+{
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
+
+	if (crtc->config->has_audio)
+		intel_audio_codec_disable(encoder);
+
+	intel_disable_hdmi(encoder);
+}
+
+static void pch_disable_hdmi(struct intel_encoder *encoder)
+{
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
+
+	if (crtc->config->has_audio)
+		intel_audio_codec_disable(encoder);
+}
+
+static void pch_post_disable_hdmi(struct intel_encoder *encoder)
+{
+	intel_disable_hdmi(encoder);
+}
+
 static int hdmi_portclock_limit(struct intel_hdmi *hdmi, bool respect_dvi_limit)
 {
 	struct drm_device *dev = intel_hdmi_to_dev(hdmi);
@@ -956,6 +964,7 @@ static bool hdmi_12bpc_possible(struct intel_crtc_state *crtc_state)
 	struct drm_device *dev = crtc_state->base.crtc->dev;
 	struct drm_atomic_state *state;
 	struct intel_encoder *encoder;
+	struct drm_connector *connector;
 	struct drm_connector_state *connector_state;
 	int count = 0, count_hdmi = 0;
 	int i;
@@ -965,11 +974,7 @@ static bool hdmi_12bpc_possible(struct intel_crtc_state *crtc_state)
 
 	state = crtc_state->base.state;
 
-	for (i = 0; i < state->num_connector; i++) {
-		if (!state->connectors[i])
-			continue;
-
-		connector_state = state->connector_states[i];
+	for_each_connector_in_state(state, connector, connector_state, i) {
 		if (connector_state->crtc != crtc_state->base.crtc)
 			continue;
 
@@ -1031,7 +1036,8 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 	 */
 	if (pipe_config->pipe_bpp > 8*3 && pipe_config->has_hdmi_sink &&
 	    clock_12bpc <= portclock_limit &&
-	    hdmi_12bpc_possible(pipe_config)) {
+	    hdmi_12bpc_possible(pipe_config) &&
+	    0 /* FIXME 12bpc support totally broken */) {
 		DRM_DEBUG_KMS("picking bpc to 12 for HDMI output\n");
 		desired_bpp = 12*3;
 
@@ -1288,7 +1294,7 @@ static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 	u32 val;
 
 	/* Enable clock channels for this port */
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW8(port));
 	val = 0;
 	if (pipe)
@@ -1311,7 +1317,7 @@ static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 	/* Program lane clock */
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW14(port), 0x00760018);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW23(port), 0x00400888);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	intel_hdmi->set_infoframes(&encoder->base,
 				   intel_crtc->config->has_hdmi_sink,
@@ -1319,7 +1325,7 @@ static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 
 	intel_enable_hdmi(encoder);
 
-	vlv_wait_port_ready(dev_priv, dport);
+	vlv_wait_port_ready(dev_priv, dport, 0x0);
 }
 
 static void vlv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
@@ -1335,7 +1341,7 @@ static void vlv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 	intel_hdmi_prepare(encoder);
 
 	/* Program Tx lane resets to default */
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW0(port),
 			 DPIO_PCS_TX_LANE2_RESET |
 			 DPIO_PCS_TX_LANE1_RESET);
@@ -1352,7 +1358,7 @@ static void vlv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW9(port), 0x00002000);
 	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW5(port), DPIO_TX_OCALINIT_EN);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
@@ -1368,7 +1374,7 @@ static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 
 	intel_hdmi_prepare(encoder);
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* program left/right clock distribution */
 	if (pipe != PIPE_B) {
@@ -1418,7 +1424,7 @@ static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 		val |= CHV_CMN_USEDCLKCHANNEL;
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW19(ch), val);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
@@ -1431,10 +1437,10 @@ static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
 	int pipe = intel_crtc->pipe;
 
 	/* Reset lanes to avoid HDMI flicker (VLV w/a) */
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW0(port), 0x00000000);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW1(port), 0x00e00060);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void chv_hdmi_post_disable(struct intel_encoder *encoder)
@@ -1448,7 +1454,7 @@ static void chv_hdmi_post_disable(struct intel_encoder *encoder)
 	enum pipe pipe = intel_crtc->pipe;
 	u32 val;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* Propagate soft reset to data lane reset */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
@@ -1467,7 +1473,7 @@ static void chv_hdmi_post_disable(struct intel_encoder *encoder)
 	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
@@ -1482,10 +1488,10 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 		&intel_crtc->config->base.adjusted_mode;
 	enum dpio_channel ch = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
-	int data, i;
+	int data, i, stagger;
 	u32 val;
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 
 	/* allow hardware to manage TX FIFO reset source */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW11(ch));
@@ -1522,7 +1528,38 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 	}
 
 	/* Data lane stagger programming */
-	/* FIXME: Fix up value only after power analysis */
+	if (intel_crtc->config->port_clock > 270000)
+		stagger = 0x18;
+	else if (intel_crtc->config->port_clock > 135000)
+		stagger = 0xd;
+	else if (intel_crtc->config->port_clock > 67500)
+		stagger = 0x7;
+	else if (intel_crtc->config->port_clock > 33750)
+		stagger = 0x4;
+	else
+		stagger = 0x2;
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW11(ch));
+	val |= DPIO_TX2_STAGGER_MASK(0x1f);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW11(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW11(ch));
+	val |= DPIO_TX2_STAGGER_MASK(0x1f);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
+
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW12(ch),
+		       DPIO_LANESTAGGER_STRAP(stagger) |
+		       DPIO_LANESTAGGER_STRAP_OVRD |
+		       DPIO_TX1_STAGGER_MASK(0x1f) |
+		       DPIO_TX1_STAGGER_MULT(6) |
+		       DPIO_TX2_STAGGER_MULT(0));
+
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW12(ch),
+		       DPIO_LANESTAGGER_STRAP(stagger) |
+		       DPIO_LANESTAGGER_STRAP_OVRD |
+		       DPIO_TX1_STAGGER_MASK(0x1f) |
+		       DPIO_TX1_STAGGER_MULT(7) |
+		       DPIO_TX2_STAGGER_MULT(5));
 
 	/* Clear calc init */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
@@ -1597,7 +1634,7 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 	val |= DPIO_LRC_BYPASS;
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
 
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	intel_hdmi->set_infoframes(&encoder->base,
 				   intel_crtc->config->has_hdmi_sink,
@@ -1605,7 +1642,7 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 
 	intel_enable_hdmi(encoder);
 
-	vlv_wait_port_ready(dev_priv, dport);
+	vlv_wait_port_ready(dev_priv, dport, 0x0);
 }
 
 static void intel_hdmi_destroy(struct drm_connector *connector)
@@ -1676,18 +1713,26 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 
 	switch (port) {
 	case PORT_B:
-		intel_hdmi->ddc_bus = GMBUS_PORT_DPB;
+		if (IS_BROXTON(dev_priv))
+			intel_hdmi->ddc_bus = GMBUS_PIN_1_BXT;
+		else
+			intel_hdmi->ddc_bus = GMBUS_PIN_DPB;
 		intel_encoder->hpd_pin = HPD_PORT_B;
 		break;
 	case PORT_C:
-		intel_hdmi->ddc_bus = GMBUS_PORT_DPC;
+		if (IS_BROXTON(dev_priv))
+			intel_hdmi->ddc_bus = GMBUS_PIN_2_BXT;
+		else
+			intel_hdmi->ddc_bus = GMBUS_PIN_DPC;
 		intel_encoder->hpd_pin = HPD_PORT_C;
 		break;
 	case PORT_D:
-		if (IS_CHERRYVIEW(dev))
-			intel_hdmi->ddc_bus = GMBUS_PORT_DPD_CHV;
+		if (WARN_ON(IS_BROXTON(dev_priv)))
+			intel_hdmi->ddc_bus = GMBUS_PIN_DISABLED;
+		else if (IS_CHERRYVIEW(dev_priv))
+			intel_hdmi->ddc_bus = GMBUS_PIN_DPD_CHV;
 		else
-			intel_hdmi->ddc_bus = GMBUS_PORT_DPD;
+			intel_hdmi->ddc_bus = GMBUS_PIN_DPD;
 		intel_encoder->hpd_pin = HPD_PORT_D;
 		break;
 	case PORT_A:
@@ -1762,7 +1807,12 @@ void intel_hdmi_init(struct drm_device *dev, int hdmi_reg, enum port port)
 			 DRM_MODE_ENCODER_TMDS);
 
 	intel_encoder->compute_config = intel_hdmi_compute_config;
-	intel_encoder->disable = intel_disable_hdmi;
+	if (HAS_PCH_SPLIT(dev)) {
+		intel_encoder->disable = pch_disable_hdmi;
+		intel_encoder->post_disable = pch_post_disable_hdmi;
+	} else {
+		intel_encoder->disable = g4x_disable_hdmi;
+	}
 	intel_encoder->get_hw_state = intel_hdmi_get_hw_state;
 	intel_encoder->get_config = intel_hdmi_get_config;
 	if (IS_CHERRYVIEW(dev)) {
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index ae628001fd97..a64f26c670af 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -34,20 +34,71 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
-struct gmbus_port {
+struct gmbus_pin {
 	const char *name;
 	int reg;
 };
 
-static const struct gmbus_port gmbus_ports[] = {
-	{ "ssc", GPIOB },
-	{ "vga", GPIOA },
-	{ "panel", GPIOC },
-	{ "dpc", GPIOD },
-	{ "dpb", GPIOE },
-	{ "dpd", GPIOF },
+/* Map gmbus pin pairs to names and registers. */
+static const struct gmbus_pin gmbus_pins[] = {
+	[GMBUS_PIN_SSC] = { "ssc", GPIOB },
+	[GMBUS_PIN_VGADDC] = { "vga", GPIOA },
+	[GMBUS_PIN_PANEL] = { "panel", GPIOC },
+	[GMBUS_PIN_DPC] = { "dpc", GPIOD },
+	[GMBUS_PIN_DPB] = { "dpb", GPIOE },
+	[GMBUS_PIN_DPD] = { "dpd", GPIOF },
 };
 
+static const struct gmbus_pin gmbus_pins_bdw[] = {
+	[GMBUS_PIN_VGADDC] = { "vga", GPIOA },
+	[GMBUS_PIN_DPC] = { "dpc", GPIOD },
+	[GMBUS_PIN_DPB] = { "dpb", GPIOE },
+	[GMBUS_PIN_DPD] = { "dpd", GPIOF },
+};
+
+static const struct gmbus_pin gmbus_pins_skl[] = {
+	[GMBUS_PIN_DPC] = { "dpc", GPIOD },
+	[GMBUS_PIN_DPB] = { "dpb", GPIOE },
+	[GMBUS_PIN_DPD] = { "dpd", GPIOF },
+};
+
+static const struct gmbus_pin gmbus_pins_bxt[] = {
+	[GMBUS_PIN_1_BXT] = { "dpb", PCH_GPIOB },
+	[GMBUS_PIN_2_BXT] = { "dpc", PCH_GPIOC },
+	[GMBUS_PIN_3_BXT] = { "misc", PCH_GPIOD },
+};
+
+/* pin is expected to be valid */
+static const struct gmbus_pin *get_gmbus_pin(struct drm_i915_private *dev_priv,
+					     unsigned int pin)
+{
+	if (IS_BROXTON(dev_priv))
+		return &gmbus_pins_bxt[pin];
+	else if (IS_SKYLAKE(dev_priv))
+		return &gmbus_pins_skl[pin];
+	else if (IS_BROADWELL(dev_priv))
+		return &gmbus_pins_bdw[pin];
+	else
+		return &gmbus_pins[pin];
+}
+
+bool intel_gmbus_is_valid_pin(struct drm_i915_private *dev_priv,
+			      unsigned int pin)
+{
+	unsigned int size;
+
+	if (IS_BROXTON(dev_priv))
+		size = ARRAY_SIZE(gmbus_pins_bxt);
+	else if (IS_SKYLAKE(dev_priv))
+		size = ARRAY_SIZE(gmbus_pins_skl);
+	else if (IS_BROADWELL(dev_priv))
+		size = ARRAY_SIZE(gmbus_pins_bdw);
+	else
+		size = ARRAY_SIZE(gmbus_pins);
+
+	return pin < size && get_gmbus_pin(dev_priv, pin)->reg;
+}
+
 /* Intel GPIO access functions */
 
 #define I2C_RISEFALL_TIME 10
@@ -182,15 +233,15 @@ intel_gpio_post_xfer(struct i2c_adapter *adapter)
 }
 
 static void
-intel_gpio_setup(struct intel_gmbus *bus, u32 pin)
+intel_gpio_setup(struct intel_gmbus *bus, unsigned int pin)
 {
 	struct drm_i915_private *dev_priv = bus->dev_priv;
 	struct i2c_algo_bit_data *algo;
 
 	algo = &bus->bit_algo;
 
-	/* -1 to map pin pair to gmbus index */
-	bus->gpio_reg = dev_priv->gpio_mmio_base + gmbus_ports[pin - 1].reg;
+	bus->gpio_reg = dev_priv->gpio_mmio_base +
+		get_gmbus_pin(dev_priv, pin)->reg;
 
 	bus->adapter.algo_data = algo;
 	algo->setsda = set_data;
@@ -577,7 +628,9 @@ static const struct i2c_algorithm gmbus_algorithm = {
 int intel_setup_gmbus(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret, i;
+	struct intel_gmbus *bus;
+	unsigned int pin;
+	int ret;
 
 	if (HAS_PCH_NOP(dev))
 		return 0;
@@ -591,16 +644,18 @@ int intel_setup_gmbus(struct drm_device *dev)
 	mutex_init(&dev_priv->gmbus_mutex);
 	init_waitqueue_head(&dev_priv->gmbus_wait_queue);
 
-	for (i = 0; i < GMBUS_NUM_PORTS; i++) {
-		struct intel_gmbus *bus = &dev_priv->gmbus[i];
-		u32 port = i + 1; /* +1 to map gmbus index to pin pair */
+	for (pin = 0; pin < ARRAY_SIZE(dev_priv->gmbus); pin++) {
+		if (!intel_gmbus_is_valid_pin(dev_priv, pin))
+			continue;
+
+		bus = &dev_priv->gmbus[pin];
 
 		bus->adapter.owner = THIS_MODULE;
 		bus->adapter.class = I2C_CLASS_DDC;
 		snprintf(bus->adapter.name,
 			 sizeof(bus->adapter.name),
 			 "i915 gmbus %s",
-			 gmbus_ports[i].name);
+			 get_gmbus_pin(dev_priv, pin)->name);
 
 		bus->adapter.dev.parent = &dev->pdev->dev;
 		bus->dev_priv = dev_priv;
@@ -608,13 +663,13 @@ int intel_setup_gmbus(struct drm_device *dev)
 		bus->adapter.algo = &gmbus_algorithm;
 
 		/* By default use a conservative clock rate */
-		bus->reg0 = port | GMBUS_RATE_100KHZ;
+		bus->reg0 = pin | GMBUS_RATE_100KHZ;
 
 		/* gmbus seems to be broken on i830 */
 		if (IS_I830(dev))
 			bus->force_bit = 1;
 
-		intel_gpio_setup(bus, port);
+		intel_gpio_setup(bus, pin);
 
 		ret = i2c_add_adapter(&bus->adapter);
 		if (ret)
@@ -626,20 +681,23 @@ int intel_setup_gmbus(struct drm_device *dev)
 	return 0;
 
 err:
-	while (--i) {
-		struct intel_gmbus *bus = &dev_priv->gmbus[i];
+	while (--pin) {
+		if (!intel_gmbus_is_valid_pin(dev_priv, pin))
+			continue;
+
+		bus = &dev_priv->gmbus[pin];
 		i2c_del_adapter(&bus->adapter);
 	}
 	return ret;
 }
 
 struct i2c_adapter *intel_gmbus_get_adapter(struct drm_i915_private *dev_priv,
-					    unsigned port)
+					    unsigned int pin)
 {
-	WARN_ON(!intel_gmbus_is_port_valid(port));
-	/* -1 to map pin pair to gmbus index */
-	return (intel_gmbus_is_port_valid(port)) ?
-		&dev_priv->gmbus[port - 1].adapter : NULL;
+	if (WARN_ON(!intel_gmbus_is_valid_pin(dev_priv, pin)))
+		return NULL;
+
+	return &dev_priv->gmbus[pin].adapter;
 }
 
 void intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed)
@@ -662,10 +720,14 @@ void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit)
 void intel_teardown_gmbus(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int i;
+	struct intel_gmbus *bus;
+	unsigned int pin;
+
+	for (pin = 0; pin < ARRAY_SIZE(dev_priv->gmbus); pin++) {
+		if (!intel_gmbus_is_valid_pin(dev_priv, pin))
+			continue;
 
-	for (i = 0; i < GMBUS_NUM_PORTS; i++) {
-		struct intel_gmbus *bus = &dev_priv->gmbus[i];
+		bus = &dev_priv->gmbus[pin];
 		i2c_del_adapter(&bus->adapter);
 	}
 }
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 424e62197787..9b74ffae5f5a 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -188,6 +188,15 @@
 #define GEN8_CTX_FORCE_RESTORE (1<<2)
 #define GEN8_CTX_L3LLC_COHERENT (1<<5)
 #define GEN8_CTX_PRIVILEGE (1<<8)
+
+#define ASSIGN_CTX_PDP(ppgtt, reg_state, n) { \
+	const u64 _addr = test_bit(n, ppgtt->pdp.used_pdpes) ? \
+		ppgtt->pdp.page_directory[n]->daddr : \
+		ppgtt->scratch_pd->daddr; \
+	reg_state[CTX_PDP ## n ## _UDW+1] = upper_32_bits(_addr); \
+	reg_state[CTX_PDP ## n ## _LDW+1] = lower_32_bits(_addr); \
+}
+
 enum {
 	ADVANCED_CONTEXT = 0,
 	LEGACY_CONTEXT,
@@ -265,7 +274,8 @@ static uint64_t execlists_ctx_descriptor(struct intel_engine_cs *ring,
 
 	desc = GEN8_CTX_VALID;
 	desc |= LEGACY_CONTEXT << GEN8_CTX_MODE_SHIFT;
-	desc |= GEN8_CTX_L3LLC_COHERENT;
+	if (IS_GEN8(ctx_obj->base.dev))
+		desc |= GEN8_CTX_L3LLC_COHERENT;
 	desc |= GEN8_CTX_PRIVILEGE;
 	desc |= lrca;
 	desc |= (u64)intel_execlists_ctx_id(ctx_obj) << GEN8_CTX_ID_SHIFT;
@@ -305,21 +315,24 @@ static void execlists_elsp_write(struct intel_engine_cs *ring,
 	desc[3] = (u32)(temp >> 32);
 	desc[2] = (u32)temp;
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
-	I915_WRITE(RING_ELSP(ring), desc[1]);
-	I915_WRITE(RING_ELSP(ring), desc[0]);
-	I915_WRITE(RING_ELSP(ring), desc[3]);
+	spin_lock(&dev_priv->uncore.lock);
+	intel_uncore_forcewake_get__locked(dev_priv, FORCEWAKE_ALL);
+	I915_WRITE_FW(RING_ELSP(ring), desc[1]);
+	I915_WRITE_FW(RING_ELSP(ring), desc[0]);
+	I915_WRITE_FW(RING_ELSP(ring), desc[3]);
 
 	/* The context is automatically loaded after the following */
-	I915_WRITE(RING_ELSP(ring), desc[2]);
+	I915_WRITE_FW(RING_ELSP(ring), desc[2]);
 
 	/* ELSP is a wo register, so use another nearby reg for posting instead */
-	POSTING_READ(RING_EXECLIST_STATUS(ring));
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	POSTING_READ_FW(RING_EXECLIST_STATUS(ring));
+	intel_uncore_forcewake_put__locked(dev_priv, FORCEWAKE_ALL);
+	spin_unlock(&dev_priv->uncore.lock);
 }
 
 static int execlists_update_context(struct drm_i915_gem_object *ctx_obj,
 				    struct drm_i915_gem_object *ring_obj,
+				    struct i915_hw_ppgtt *ppgtt,
 				    u32 tail)
 {
 	struct page *page;
@@ -331,6 +344,16 @@ static int execlists_update_context(struct drm_i915_gem_object *ctx_obj,
 	reg_state[CTX_RING_TAIL+1] = tail;
 	reg_state[CTX_RING_BUFFER_START+1] = i915_gem_obj_ggtt_offset(ring_obj);
 
+	/* True PPGTT with dynamic page allocation: update PDP registers and
+	 * point the unallocated PDPs to the scratch page
+	 */
+	if (ppgtt) {
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 0);
+	}
+
 	kunmap_atomic(reg_state);
 
 	return 0;
@@ -349,7 +372,7 @@ static void execlists_submit_contexts(struct intel_engine_cs *ring,
 	WARN_ON(!i915_gem_obj_is_pinned(ctx_obj0));
 	WARN_ON(!i915_gem_obj_is_pinned(ringbuf0->obj));
 
-	execlists_update_context(ctx_obj0, ringbuf0->obj, tail0);
+	execlists_update_context(ctx_obj0, ringbuf0->obj, to0->ppgtt, tail0);
 
 	if (to1) {
 		ringbuf1 = to1->engine[ring->id].ringbuf;
@@ -358,7 +381,7 @@ static void execlists_submit_contexts(struct intel_engine_cs *ring,
 		WARN_ON(!i915_gem_obj_is_pinned(ctx_obj1));
 		WARN_ON(!i915_gem_obj_is_pinned(ringbuf1->obj));
 
-		execlists_update_context(ctx_obj1, ringbuf1->obj, tail1);
+		execlists_update_context(ctx_obj1, ringbuf1->obj, to1->ppgtt, tail1);
 	}
 
 	execlists_elsp_write(ring, ctx_obj0, ctx_obj1);
@@ -371,6 +394,12 @@ static void execlists_context_unqueue(struct intel_engine_cs *ring)
 
 	assert_spin_locked(&ring->execlist_lock);
 
+	/*
+	 * If irqs are not active generate a warning as batches that finish
+	 * without the irqs may get lost and a GPU Hang may occur.
+	 */
+	WARN_ON(!intel_irqs_enabled(ring->dev->dev_private));
+
 	if (list_empty(&ring->execlist_queue))
 		return;
 
@@ -398,7 +427,7 @@ static void execlists_context_unqueue(struct intel_engine_cs *ring)
 		 * WaIdleLiteRestore: make sure we never cause a lite
 		 * restore with HEAD==TAIL
 		 */
-		if (req0 && req0->elsp_submitted) {
+		if (req0->elsp_submitted) {
 			/*
 			 * Apply the wa NOOPS to prevent ring:HEAD == req:TAIL
 			 * as we resubmit the request. See gen8_emit_request()
@@ -520,8 +549,6 @@ static int execlists_context_queue(struct intel_engine_cs *ring,
 				   struct drm_i915_gem_request *request)
 {
 	struct drm_i915_gem_request *cursor;
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	unsigned long flags;
 	int num_elements = 0;
 
 	if (to != ring->default_context)
@@ -538,7 +565,6 @@ static int execlists_context_queue(struct intel_engine_cs *ring,
 		request->ring = ring;
 		request->ctx = to;
 		kref_init(&request->ref);
-		request->uniq = dev_priv->request_uniq++;
 		i915_gem_context_reference(request->ctx);
 	} else {
 		i915_gem_request_reference(request);
@@ -546,9 +572,7 @@ static int execlists_context_queue(struct intel_engine_cs *ring,
 	}
 	request->tail = tail;
 
-	intel_runtime_pm_get(dev_priv);
-
-	spin_lock_irqsave(&ring->execlist_lock, flags);
+	spin_lock_irq(&ring->execlist_lock);
 
 	list_for_each_entry(cursor, &ring->execlist_queue, execlist_link)
 		if (++num_elements > 2)
@@ -574,7 +598,7 @@ static int execlists_context_queue(struct intel_engine_cs *ring,
 	if (num_elements == 0)
 		execlists_context_unqueue(ring);
 
-	spin_unlock_irqrestore(&ring->execlist_lock, flags);
+	spin_unlock_irq(&ring->execlist_lock);
 
 	return 0;
 }
@@ -604,6 +628,7 @@ static int execlists_move_to_gpu(struct intel_ringbuffer *ringbuf,
 				 struct list_head *vmas)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
+	const unsigned other_rings = ~intel_ring_flag(ring);
 	struct i915_vma *vma;
 	uint32_t flush_domains = 0;
 	bool flush_chipset = false;
@@ -612,9 +637,11 @@ static int execlists_move_to_gpu(struct intel_ringbuffer *ringbuf,
 	list_for_each_entry(vma, vmas, exec_list) {
 		struct drm_i915_gem_object *obj = vma->obj;
 
-		ret = i915_gem_object_sync(obj, ring);
-		if (ret)
-			return ret;
+		if (obj->active & other_rings) {
+			ret = i915_gem_object_sync(obj, ring);
+			if (ret)
+				return ret;
+		}
 
 		if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
 			flush_chipset |= i915_gem_clflush_object(obj, false);
@@ -631,6 +658,170 @@ static int execlists_move_to_gpu(struct intel_ringbuffer *ringbuf,
 	return logical_ring_invalidate_all_caches(ringbuf, ctx);
 }
 
+int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request,
+					    struct intel_context *ctx)
+{
+	int ret;
+
+	if (ctx != request->ring->default_context) {
+		ret = intel_lr_context_pin(request->ring, ctx);
+		if (ret)
+			return ret;
+	}
+
+	request->ringbuf = ctx->engine[request->ring->id].ringbuf;
+	request->ctx     = ctx;
+	i915_gem_context_reference(request->ctx);
+
+	return 0;
+}
+
+static int logical_ring_wait_for_space(struct intel_ringbuffer *ringbuf,
+				       struct intel_context *ctx,
+				       int bytes)
+{
+	struct intel_engine_cs *ring = ringbuf->ring;
+	struct drm_i915_gem_request *request;
+	unsigned space;
+	int ret;
+
+	if (intel_ring_space(ringbuf) >= bytes)
+		return 0;
+
+	list_for_each_entry(request, &ring->request_list, list) {
+		/*
+		 * The request queue is per-engine, so can contain requests
+		 * from multiple ringbuffers. Here, we must ignore any that
+		 * aren't from the ringbuffer we're considering.
+		 */
+		if (request->ringbuf != ringbuf)
+			continue;
+
+		/* Would completion of this request free enough space? */
+		space = __intel_ring_space(request->postfix, ringbuf->tail,
+					   ringbuf->size);
+		if (space >= bytes)
+			break;
+	}
+
+	if (WARN_ON(&request->list == &ring->request_list))
+		return -ENOSPC;
+
+	ret = i915_wait_request(request);
+	if (ret)
+		return ret;
+
+	ringbuf->space = space;
+	return 0;
+}
+
+/*
+ * intel_logical_ring_advance_and_submit() - advance the tail and submit the workload
+ * @ringbuf: Logical Ringbuffer to advance.
+ *
+ * The tail is updated in our logical ringbuffer struct, not in the actual context. What
+ * really happens during submission is that the context and current tail will be placed
+ * on a queue waiting for the ELSP to be ready to accept a new context submission. At that
+ * point, the tail *inside* the context is updated and the ELSP written to.
+ */
+static void
+intel_logical_ring_advance_and_submit(struct intel_ringbuffer *ringbuf,
+				      struct intel_context *ctx,
+				      struct drm_i915_gem_request *request)
+{
+	struct intel_engine_cs *ring = ringbuf->ring;
+
+	intel_logical_ring_advance(ringbuf);
+
+	if (intel_ring_stopped(ring))
+		return;
+
+	execlists_context_queue(ring, ctx, ringbuf->tail, request);
+}
+
+static int logical_ring_wrap_buffer(struct intel_ringbuffer *ringbuf,
+				    struct intel_context *ctx)
+{
+	uint32_t __iomem *virt;
+	int rem = ringbuf->size - ringbuf->tail;
+
+	if (ringbuf->space < rem) {
+		int ret = logical_ring_wait_for_space(ringbuf, ctx, rem);
+
+		if (ret)
+			return ret;
+	}
+
+	virt = ringbuf->virtual_start + ringbuf->tail;
+	rem /= 4;
+	while (rem--)
+		iowrite32(MI_NOOP, virt++);
+
+	ringbuf->tail = 0;
+	intel_ring_update_space(ringbuf);
+
+	return 0;
+}
+
+static int logical_ring_prepare(struct intel_ringbuffer *ringbuf,
+				struct intel_context *ctx, int bytes)
+{
+	int ret;
+
+	if (unlikely(ringbuf->tail + bytes > ringbuf->effective_size)) {
+		ret = logical_ring_wrap_buffer(ringbuf, ctx);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	if (unlikely(ringbuf->space < bytes)) {
+		ret = logical_ring_wait_for_space(ringbuf, ctx, bytes);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	return 0;
+}
+
+/**
+ * intel_logical_ring_begin() - prepare the logical ringbuffer to accept some commands
+ *
+ * @ringbuf: Logical ringbuffer.
+ * @num_dwords: number of DWORDs that we plan to write to the ringbuffer.
+ *
+ * The ringbuffer might not be ready to accept the commands right away (maybe it needs to
+ * be wrapped, or wait a bit for the tail to be updated). This function takes care of that
+ * and also preallocates a request (every workload submission is still mediated through
+ * requests, same as it did with legacy ringbuffer submission).
+ *
+ * Return: non-zero if the ringbuffer is not ready to be written to.
+ */
+static int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf,
+				    struct intel_context *ctx, int num_dwords)
+{
+	struct intel_engine_cs *ring = ringbuf->ring;
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	ret = i915_gem_check_wedge(&dev_priv->gpu_error,
+				   dev_priv->mm.interruptible);
+	if (ret)
+		return ret;
+
+	ret = logical_ring_prepare(ringbuf, ctx, num_dwords * sizeof(uint32_t));
+	if (ret)
+		return ret;
+
+	/* Preallocate the olr before touching the ring */
+	ret = i915_gem_request_alloc(ring, ctx);
+	if (ret)
+		return ret;
+
+	ringbuf->space -= num_dwords * sizeof(uint32_t);
+	return 0;
+}
+
 /**
  * execlists_submission() - submit a batchbuffer for execution, Execlists style
  * @dev: DRM device.
@@ -742,8 +933,6 @@ int intel_execlists_submission(struct drm_device *dev, struct drm_file *file,
 void intel_execlists_retire_requests(struct intel_engine_cs *ring)
 {
 	struct drm_i915_gem_request *req, *tmp;
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	unsigned long flags;
 	struct list_head retired_list;
 
 	WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
@@ -751,9 +940,9 @@ void intel_execlists_retire_requests(struct intel_engine_cs *ring)
 		return;
 
 	INIT_LIST_HEAD(&retired_list);
-	spin_lock_irqsave(&ring->execlist_lock, flags);
+	spin_lock_irq(&ring->execlist_lock);
 	list_replace_init(&ring->execlist_retired_req_list, &retired_list);
-	spin_unlock_irqrestore(&ring->execlist_lock, flags);
+	spin_unlock_irq(&ring->execlist_lock);
 
 	list_for_each_entry_safe(req, tmp, &retired_list, execlist_link) {
 		struct intel_context *ctx = req->ctx;
@@ -762,7 +951,6 @@ void intel_execlists_retire_requests(struct intel_engine_cs *ring)
 
 		if (ctx_obj && (ctx != ring->default_context))
 			intel_lr_context_unpin(ring, ctx);
-		intel_runtime_pm_put(dev_priv);
 		list_del(&req->execlist_link);
 		i915_gem_request_unreference(req);
 	}
@@ -807,30 +995,6 @@ int logical_ring_flush_all_caches(struct intel_ringbuffer *ringbuf,
 	return 0;
 }
 
-/*
- * intel_logical_ring_advance_and_submit() - advance the tail and submit the workload
- * @ringbuf: Logical Ringbuffer to advance.
- *
- * The tail is updated in our logical ringbuffer struct, not in the actual context. What
- * really happens during submission is that the context and current tail will be placed
- * on a queue waiting for the ELSP to be ready to accept a new context submission. At that
- * point, the tail *inside* the context is updated and the ELSP written to.
- */
-static void
-intel_logical_ring_advance_and_submit(struct intel_ringbuffer *ringbuf,
-				      struct intel_context *ctx,
-				      struct drm_i915_gem_request *request)
-{
-	struct intel_engine_cs *ring = ringbuf->ring;
-
-	intel_logical_ring_advance(ringbuf);
-
-	if (intel_ring_stopped(ring))
-		return;
-
-	execlists_context_queue(ring, ctx, ringbuf->tail, request);
-}
-
 static int intel_lr_context_pin(struct intel_engine_cs *ring,
 		struct intel_context *ctx)
 {
@@ -875,219 +1039,6 @@ void intel_lr_context_unpin(struct intel_engine_cs *ring,
 	}
 }
 
-static int logical_ring_alloc_request(struct intel_engine_cs *ring,
-				      struct intel_context *ctx)
-{
-	struct drm_i915_gem_request *request;
-	struct drm_i915_private *dev_private = ring->dev->dev_private;
-	int ret;
-
-	if (ring->outstanding_lazy_request)
-		return 0;
-
-	request = kzalloc(sizeof(*request), GFP_KERNEL);
-	if (request == NULL)
-		return -ENOMEM;
-
-	if (ctx != ring->default_context) {
-		ret = intel_lr_context_pin(ring, ctx);
-		if (ret) {
-			kfree(request);
-			return ret;
-		}
-	}
-
-	kref_init(&request->ref);
-	request->ring = ring;
-	request->uniq = dev_private->request_uniq++;
-
-	ret = i915_gem_get_seqno(ring->dev, &request->seqno);
-	if (ret) {
-		intel_lr_context_unpin(ring, ctx);
-		kfree(request);
-		return ret;
-	}
-
-	request->ctx = ctx;
-	i915_gem_context_reference(request->ctx);
-	request->ringbuf = ctx->engine[ring->id].ringbuf;
-
-	ring->outstanding_lazy_request = request;
-	return 0;
-}
-
-static int logical_ring_wait_request(struct intel_ringbuffer *ringbuf,
-				     int bytes)
-{
-	struct intel_engine_cs *ring = ringbuf->ring;
-	struct drm_i915_gem_request *request;
-	int ret;
-
-	if (intel_ring_space(ringbuf) >= bytes)
-		return 0;
-
-	list_for_each_entry(request, &ring->request_list, list) {
-		/*
-		 * The request queue is per-engine, so can contain requests
-		 * from multiple ringbuffers. Here, we must ignore any that
-		 * aren't from the ringbuffer we're considering.
-		 */
-		struct intel_context *ctx = request->ctx;
-		if (ctx->engine[ring->id].ringbuf != ringbuf)
-			continue;
-
-		/* Would completion of this request free enough space? */
-		if (__intel_ring_space(request->tail, ringbuf->tail,
-				       ringbuf->size) >= bytes) {
-			break;
-		}
-	}
-
-	if (&request->list == &ring->request_list)
-		return -ENOSPC;
-
-	ret = i915_wait_request(request);
-	if (ret)
-		return ret;
-
-	i915_gem_retire_requests_ring(ring);
-
-	return intel_ring_space(ringbuf) >= bytes ? 0 : -ENOSPC;
-}
-
-static int logical_ring_wait_for_space(struct intel_ringbuffer *ringbuf,
-				       struct intel_context *ctx,
-				       int bytes)
-{
-	struct intel_engine_cs *ring = ringbuf->ring;
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long end;
-	int ret;
-
-	ret = logical_ring_wait_request(ringbuf, bytes);
-	if (ret != -ENOSPC)
-		return ret;
-
-	/* Force the context submission in case we have been skipping it */
-	intel_logical_ring_advance_and_submit(ringbuf, ctx, NULL);
-
-	/* With GEM the hangcheck timer should kick us out of the loop,
-	 * leaving it early runs the risk of corrupting GEM state (due
-	 * to running on almost untested codepaths). But on resume
-	 * timers don't work yet, so prevent a complete hang in that
-	 * case by choosing an insanely large timeout. */
-	end = jiffies + 60 * HZ;
-
-	ret = 0;
-	do {
-		if (intel_ring_space(ringbuf) >= bytes)
-			break;
-
-		msleep(1);
-
-		if (dev_priv->mm.interruptible && signal_pending(current)) {
-			ret = -ERESTARTSYS;
-			break;
-		}
-
-		ret = i915_gem_check_wedge(&dev_priv->gpu_error,
-					   dev_priv->mm.interruptible);
-		if (ret)
-			break;
-
-		if (time_after(jiffies, end)) {
-			ret = -EBUSY;
-			break;
-		}
-	} while (1);
-
-	return ret;
-}
-
-static int logical_ring_wrap_buffer(struct intel_ringbuffer *ringbuf,
-				    struct intel_context *ctx)
-{
-	uint32_t __iomem *virt;
-	int rem = ringbuf->size - ringbuf->tail;
-
-	if (ringbuf->space < rem) {
-		int ret = logical_ring_wait_for_space(ringbuf, ctx, rem);
-
-		if (ret)
-			return ret;
-	}
-
-	virt = ringbuf->virtual_start + ringbuf->tail;
-	rem /= 4;
-	while (rem--)
-		iowrite32(MI_NOOP, virt++);
-
-	ringbuf->tail = 0;
-	intel_ring_update_space(ringbuf);
-
-	return 0;
-}
-
-static int logical_ring_prepare(struct intel_ringbuffer *ringbuf,
-				struct intel_context *ctx, int bytes)
-{
-	int ret;
-
-	if (unlikely(ringbuf->tail + bytes > ringbuf->effective_size)) {
-		ret = logical_ring_wrap_buffer(ringbuf, ctx);
-		if (unlikely(ret))
-			return ret;
-	}
-
-	if (unlikely(ringbuf->space < bytes)) {
-		ret = logical_ring_wait_for_space(ringbuf, ctx, bytes);
-		if (unlikely(ret))
-			return ret;
-	}
-
-	return 0;
-}
-
-/**
- * intel_logical_ring_begin() - prepare the logical ringbuffer to accept some commands
- *
- * @ringbuf: Logical ringbuffer.
- * @num_dwords: number of DWORDs that we plan to write to the ringbuffer.
- *
- * The ringbuffer might not be ready to accept the commands right away (maybe it needs to
- * be wrapped, or wait a bit for the tail to be updated). This function takes care of that
- * and also preallocates a request (every workload submission is still mediated through
- * requests, same as it did with legacy ringbuffer submission).
- *
- * Return: non-zero if the ringbuffer is not ready to be written to.
- */
-int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf,
-			     struct intel_context *ctx, int num_dwords)
-{
-	struct intel_engine_cs *ring = ringbuf->ring;
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret;
-
-	ret = i915_gem_check_wedge(&dev_priv->gpu_error,
-				   dev_priv->mm.interruptible);
-	if (ret)
-		return ret;
-
-	ret = logical_ring_prepare(ringbuf, ctx, num_dwords * sizeof(uint32_t));
-	if (ret)
-		return ret;
-
-	/* Preallocate the olr before touching the ring */
-	ret = logical_ring_alloc_request(ring, ctx);
-	if (ret)
-		return ret;
-
-	ringbuf->space -= num_dwords * sizeof(uint32_t);
-	return 0;
-}
-
 static int intel_logical_ring_workarounds_emit(struct intel_engine_cs *ring,
 					       struct intel_context *ctx)
 {
@@ -1288,6 +1239,7 @@ static int gen8_emit_flush_render(struct intel_ringbuffer *ringbuf,
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
 	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
+	bool vf_flush_wa;
 	u32 flags = 0;
 	int ret;
 
@@ -1309,10 +1261,26 @@ static int gen8_emit_flush_render(struct intel_ringbuffer *ringbuf,
 		flags |= PIPE_CONTROL_GLOBAL_GTT_IVB;
 	}
 
-	ret = intel_logical_ring_begin(ringbuf, ctx, 6);
+	/*
+	 * On GEN9+ Before VF_CACHE_INVALIDATE we need to emit a NULL pipe
+	 * control.
+	 */
+	vf_flush_wa = INTEL_INFO(ring->dev)->gen >= 9 &&
+		      flags & PIPE_CONTROL_VF_CACHE_INVALIDATE;
+
+	ret = intel_logical_ring_begin(ringbuf, ctx, vf_flush_wa ? 12 : 6);
 	if (ret)
 		return ret;
 
+	if (vf_flush_wa) {
+		intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
+		intel_logical_ring_emit(ringbuf, 0);
+		intel_logical_ring_emit(ringbuf, 0);
+		intel_logical_ring_emit(ringbuf, 0);
+		intel_logical_ring_emit(ringbuf, 0);
+		intel_logical_ring_emit(ringbuf, 0);
+	}
+
 	intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
 	intel_logical_ring_emit(ringbuf, flags);
 	intel_logical_ring_emit(ringbuf, scratch_addr);
@@ -1443,6 +1411,7 @@ void intel_logical_ring_cleanup(struct intel_engine_cs *ring)
 		ring->cleanup(ring);
 
 	i915_cmd_parser_fini_ring(ring);
+	i915_gem_batch_pool_fini(&ring->batch_pool);
 
 	if (ring->status_page.obj) {
 		kunmap(sg_page(ring->status_page.obj->pages->sgl));
@@ -1460,6 +1429,7 @@ static int logical_ring_init(struct drm_device *dev, struct intel_engine_cs *rin
 	ring->dev = dev;
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
+	i915_gem_batch_pool_init(dev, &ring->batch_pool);
 	init_waitqueue_head(&ring->irq_queue);
 
 	INIT_LIST_HEAD(&ring->execlist_queue);
@@ -1812,14 +1782,14 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 	reg_state[CTX_PDP1_LDW] = GEN8_RING_PDP_LDW(ring, 1);
 	reg_state[CTX_PDP0_UDW] = GEN8_RING_PDP_UDW(ring, 0);
 	reg_state[CTX_PDP0_LDW] = GEN8_RING_PDP_LDW(ring, 0);
-	reg_state[CTX_PDP3_UDW+1] = upper_32_bits(ppgtt->pdp.page_directory[3]->daddr);
-	reg_state[CTX_PDP3_LDW+1] = lower_32_bits(ppgtt->pdp.page_directory[3]->daddr);
-	reg_state[CTX_PDP2_UDW+1] = upper_32_bits(ppgtt->pdp.page_directory[2]->daddr);
-	reg_state[CTX_PDP2_LDW+1] = lower_32_bits(ppgtt->pdp.page_directory[2]->daddr);
-	reg_state[CTX_PDP1_UDW+1] = upper_32_bits(ppgtt->pdp.page_directory[1]->daddr);
-	reg_state[CTX_PDP1_LDW+1] = lower_32_bits(ppgtt->pdp.page_directory[1]->daddr);
-	reg_state[CTX_PDP0_UDW+1] = upper_32_bits(ppgtt->pdp.page_directory[0]->daddr);
-	reg_state[CTX_PDP0_LDW+1] = lower_32_bits(ppgtt->pdp.page_directory[0]->daddr);
+
+	/* With dynamic page allocation, PDPs may not be allocated at this point,
+	 * Point the unallocated PDPs to the scratch page
+	 */
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
+	ASSIGN_CTX_PDP(ppgtt, reg_state, 0);
 	if (ring->id == RCS) {
 		reg_state[CTX_LRI_HEADER_2] = MI_LOAD_REGISTER_IMM(1);
 		reg_state[CTX_R_PWR_CLK_STATE] = GEN8_R_PWR_CLK_STATE;
@@ -1936,11 +1906,10 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 
 	context_size = round_up(get_lr_context_size(ring), 4096);
 
-	ctx_obj = i915_gem_alloc_context_obj(dev, context_size);
-	if (IS_ERR(ctx_obj)) {
-		ret = PTR_ERR(ctx_obj);
-		DRM_DEBUG_DRIVER("Alloc LRC backing obj failed: %d\n", ret);
-		return ret;
+	ctx_obj = i915_gem_alloc_object(dev, context_size);
+	if (!ctx_obj) {
+		DRM_DEBUG_DRIVER("Alloc LRC backing obj failed.\n");
+		return -ENOMEM;
 	}
 
 	if (is_global_default_ctx) {
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index adb731e49c57..04d3a6d8b207 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -36,6 +36,8 @@
 #define RING_CONTEXT_STATUS_PTR(ring)	((ring)->mmio_base+0x3a0)
 
 /* Logical Rings */
+int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request,
+					    struct intel_context *ctx);
 void intel_logical_ring_stop(struct intel_engine_cs *ring);
 void intel_logical_ring_cleanup(struct intel_engine_cs *ring);
 int intel_logical_rings_init(struct drm_device *dev);
@@ -63,9 +65,6 @@ static inline void intel_logical_ring_emit(struct intel_ringbuffer *ringbuf,
 	iowrite32(data, ringbuf->virtual_start + ringbuf->tail);
 	ringbuf->tail += 4;
 }
-int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf,
-			     struct intel_context *ctx,
-			     int num_dwords);
 
 /* Logical Ring Contexts */
 void intel_lr_context_free(struct intel_context *ctx);
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index fbcc7dff0d63..161ab26f81fb 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -781,7 +781,7 @@ static bool lvds_is_present_in_vbt(struct drm_device *dev,
 		    child->device_type != DEVICE_TYPE_LFP)
 			continue;
 
-		if (intel_gmbus_is_port_valid(child->i2c_pin))
+		if (intel_gmbus_is_valid_pin(dev_priv, child->i2c_pin))
 			*i2c_pin = child->i2c_pin;
 
 		/* However, we cannot trust the BIOS writers to populate
@@ -942,7 +942,7 @@ void intel_lvds_init(struct drm_device *dev)
 	if (dmi_check_system(intel_no_lvds))
 		return;
 
-	pin = GMBUS_PORT_PANEL;
+	pin = GMBUS_PIN_PANEL;
 	if (!lvds_is_present_in_vbt(dev, &pin)) {
 		DRM_DEBUG_KMS("LVDS is not present in VBT\n");
 		return;
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index dd92122ed95c..25c8ec697da1 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -172,10 +172,11 @@ struct intel_overlay {
 	struct intel_crtc *crtc;
 	struct drm_i915_gem_object *vid_bo;
 	struct drm_i915_gem_object *old_vid_bo;
-	int active;
-	int pfit_active;
+	bool active;
+	bool pfit_active;
 	u32 pfit_vscale_ratio; /* shifted-point number, (1<<12) == 1.0 */
-	u32 color_key;
+	u32 color_key:24;
+	u32 color_key_enabled:1;
 	u32 brightness, contrast, saturation;
 	u32 old_xscale, old_yscale;
 	/* register access */
@@ -216,7 +217,7 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
 	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	BUG_ON(overlay->last_flip_req);
+	WARN_ON(overlay->last_flip_req);
 	i915_gem_request_assign(&overlay->last_flip_req,
 					     ring->outstanding_lazy_request);
 	ret = i915_add_request(ring);
@@ -227,7 +228,6 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
 	ret = i915_wait_request(overlay->last_flip_req);
 	if (ret)
 		return ret;
-	i915_gem_retire_requests(dev);
 
 	i915_gem_request_assign(&overlay->last_flip_req, NULL);
 	return 0;
@@ -241,15 +241,15 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	BUG_ON(overlay->active);
-	overlay->active = 1;
-
+	WARN_ON(overlay->active);
 	WARN_ON(IS_I830(dev) && !(dev_priv->quirks & QUIRK_PIPEA_FORCE));
 
 	ret = intel_ring_begin(ring, 4);
 	if (ret)
 		return ret;
 
+	overlay->active = true;
+
 	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_ON);
 	intel_ring_emit(ring, overlay->flip_addr | OFC_UPDATE);
 	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
@@ -270,7 +270,7 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 	u32 tmp;
 	int ret;
 
-	BUG_ON(!overlay->active);
+	WARN_ON(!overlay->active);
 
 	if (load_polyphase_filter)
 		flip_addr |= OFC_UPDATE;
@@ -309,7 +309,8 @@ static void intel_overlay_off_tail(struct intel_overlay *overlay)
 	struct drm_i915_gem_object *obj = overlay->vid_bo;
 
 	/* never have the overlay hw on without showing a frame */
-	BUG_ON(!overlay->vid_bo);
+	if (WARN_ON(!obj))
+		return;
 
 	i915_gem_object_ggtt_unpin(obj);
 	drm_gem_object_unreference(&obj->base);
@@ -317,7 +318,7 @@ static void intel_overlay_off_tail(struct intel_overlay *overlay)
 
 	overlay->crtc->overlay = NULL;
 	overlay->crtc = NULL;
-	overlay->active = 0;
+	overlay->active = false;
 }
 
 /* overlay needs to be disabled in OCMD reg */
@@ -329,7 +330,7 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 	u32 flip_addr = overlay->flip_addr;
 	int ret;
 
-	BUG_ON(!overlay->active);
+	WARN_ON(!overlay->active);
 
 	/* According to intel docs the overlay hw may hang (when switching
 	 * off) without loading the filter coeffs. It is however unclear whether
@@ -374,7 +375,6 @@ static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay)
 	ret = i915_wait_request(overlay->last_flip_req);
 	if (ret)
 		return ret;
-	i915_gem_retire_requests(overlay->dev);
 
 	if (overlay->flip_tail)
 		overlay->flip_tail(overlay);
@@ -629,31 +629,36 @@ static void update_colorkey(struct intel_overlay *overlay,
 			    struct overlay_registers __iomem *regs)
 {
 	u32 key = overlay->color_key;
+	u32 flags;
+
+	flags = 0;
+	if (overlay->color_key_enabled)
+		flags |= DST_KEY_ENABLE;
 
 	switch (overlay->crtc->base.primary->fb->bits_per_pixel) {
 	case 8:
-		iowrite32(0, &regs->DCLRKV);
-		iowrite32(CLK_RGB8I_MASK | DST_KEY_ENABLE, &regs->DCLRKM);
+		key = 0;
+		flags |= CLK_RGB8I_MASK;
 		break;
 
 	case 16:
 		if (overlay->crtc->base.primary->fb->depth == 15) {
-			iowrite32(RGB15_TO_COLORKEY(key), &regs->DCLRKV);
-			iowrite32(CLK_RGB15_MASK | DST_KEY_ENABLE,
-				  &regs->DCLRKM);
+			key = RGB15_TO_COLORKEY(key);
+			flags |= CLK_RGB15_MASK;
 		} else {
-			iowrite32(RGB16_TO_COLORKEY(key), &regs->DCLRKV);
-			iowrite32(CLK_RGB16_MASK | DST_KEY_ENABLE,
-				  &regs->DCLRKM);
+			key = RGB16_TO_COLORKEY(key);
+			flags |= CLK_RGB16_MASK;
 		}
 		break;
 
 	case 24:
 	case 32:
-		iowrite32(key, &regs->DCLRKV);
-		iowrite32(CLK_RGB24_MASK | DST_KEY_ENABLE, &regs->DCLRKM);
+		flags |= CLK_RGB24_MASK;
 		break;
 	}
+
+	iowrite32(key, &regs->DCLRKV);
+	iowrite32(flags, &regs->DCLRKM);
 }
 
 static u32 overlay_cmd_reg(struct put_image_params *params)
@@ -712,9 +717,8 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
 	u32 swidth, swidthsw, sheight, ostride;
 	enum pipe pipe = overlay->crtc->pipe;
 
-	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-	BUG_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
-	BUG_ON(!overlay);
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	ret = intel_overlay_release_old_vid(overlay);
 	if (ret != 0)
@@ -824,8 +828,8 @@ int intel_overlay_switch_off(struct intel_overlay *overlay)
 	struct drm_device *dev = overlay->dev;
 	int ret;
 
-	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-	BUG_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	ret = intel_overlay_recover_from_interrupt(overlay);
 	if (ret != 0)
@@ -1131,10 +1135,10 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 		/* line too wide, i.e. one-line-mode */
 		if (mode->hdisplay > 1024 &&
 		    intel_panel_fitter_pipe(dev) == crtc->pipe) {
-			overlay->pfit_active = 1;
+			overlay->pfit_active = true;
 			update_pfit_vscale_ratio(overlay);
 		} else
-			overlay->pfit_active = 0;
+			overlay->pfit_active = false;
 	}
 
 	ret = check_overlay_dst(overlay, put_image_rec);
@@ -1329,6 +1333,7 @@ int intel_overlay_attrs(struct drm_device *dev, void *data,
 			I915_WRITE(OGAMC5, attrs->gamma5);
 		}
 	}
+	overlay->color_key_enabled = (attrs->flags & I915_OVERLAY_DISABLE_DEST_COLORKEY) == 0;
 
 	ret = 0;
 out_unlock:
@@ -1392,6 +1397,7 @@ void intel_setup_overlay(struct drm_device *dev)
 
 	/* init all values */
 	overlay->color_key = 0x0101fe;
+	overlay->color_key_enabled = true;
 	overlay->brightness = -19;
 	overlay->contrast = 75;
 	overlay->saturation = 146;
@@ -1432,7 +1438,7 @@ void intel_cleanup_overlay(struct drm_device *dev)
 	/* The bo's should be free'd by the generic code already.
 	 * Furthermore modesetting teardown happens beforehand so the
 	 * hardware should be off already */
-	BUG_ON(dev_priv->overlay->active);
+	WARN_ON(dev_priv->overlay->active);
 
 	drm_gem_object_unreference_unlocked(&dev_priv->overlay->reg_bo->base);
 	kfree(dev_priv->overlay);
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 08532d4ffe0a..7d83527f95f7 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -536,6 +536,14 @@ static u32 vlv_get_backlight(struct intel_connector *connector)
 	return _vlv_get_backlight(dev, pipe);
 }
 
+static u32 bxt_get_backlight(struct intel_connector *connector)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	return I915_READ(BXT_BLC_PWM_DUTY1);
+}
+
 static u32 intel_panel_get_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
@@ -616,6 +624,14 @@ static void vlv_set_backlight(struct intel_connector *connector, u32 level)
 	I915_WRITE(VLV_BLC_PWM_CTL(pipe), tmp | level);
 }
 
+static void bxt_set_backlight(struct intel_connector *connector, u32 level)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(BXT_BLC_PWM_DUTY1, level);
+}
+
 static void
 intel_panel_actually_set_backlight(struct intel_connector *connector, u32 level)
 {
@@ -741,6 +757,18 @@ static void vlv_disable_backlight(struct intel_connector *connector)
 	I915_WRITE(VLV_BLC_PWM_CTL2(pipe), tmp & ~BLM_PWM_ENABLE);
 }
 
+static void bxt_disable_backlight(struct intel_connector *connector)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 tmp;
+
+	intel_panel_actually_set_backlight(connector, 0);
+
+	tmp = I915_READ(BXT_BLC_PWM_CTL1);
+	I915_WRITE(BXT_BLC_PWM_CTL1, tmp & ~BXT_BLC_PWM_ENABLE);
+}
+
 void intel_panel_disable_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
@@ -947,6 +975,33 @@ static void vlv_enable_backlight(struct intel_connector *connector)
 	I915_WRITE(VLV_BLC_PWM_CTL2(pipe), ctl2 | BLM_PWM_ENABLE);
 }
 
+static void bxt_enable_backlight(struct intel_connector *connector)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_panel *panel = &connector->panel;
+	u32 pwm_ctl;
+
+	pwm_ctl = I915_READ(BXT_BLC_PWM_CTL1);
+	if (pwm_ctl & BXT_BLC_PWM_ENABLE) {
+		DRM_DEBUG_KMS("backlight already enabled\n");
+		pwm_ctl &= ~BXT_BLC_PWM_ENABLE;
+		I915_WRITE(BXT_BLC_PWM_CTL1, pwm_ctl);
+	}
+
+	I915_WRITE(BXT_BLC_PWM_FREQ1, panel->backlight.max);
+
+	intel_panel_actually_set_backlight(connector, panel->backlight.level);
+
+	pwm_ctl = 0;
+	if (panel->backlight.active_low_pwm)
+		pwm_ctl |= BXT_BLC_PWM_POLARITY;
+
+	I915_WRITE(BXT_BLC_PWM_CTL1, pwm_ctl);
+	POSTING_READ(BXT_BLC_PWM_CTL1);
+	I915_WRITE(BXT_BLC_PWM_CTL1, pwm_ctl | BXT_BLC_PWM_ENABLE);
+}
+
 void intel_panel_enable_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
@@ -1299,6 +1354,30 @@ static int vlv_setup_backlight(struct intel_connector *connector, enum pipe pipe
 	return 0;
 }
 
+static int
+bxt_setup_backlight(struct intel_connector *connector, enum pipe unused)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_panel *panel = &connector->panel;
+	u32 pwm_ctl, val;
+
+	pwm_ctl = I915_READ(BXT_BLC_PWM_CTL1);
+	panel->backlight.active_low_pwm = pwm_ctl & BXT_BLC_PWM_POLARITY;
+
+	panel->backlight.max = I915_READ(BXT_BLC_PWM_FREQ1);
+	if (!panel->backlight.max)
+		return -ENODEV;
+
+	val = bxt_get_backlight(connector);
+	panel->backlight.level = intel_panel_compute_brightness(connector, val);
+
+	panel->backlight.enabled = (pwm_ctl & BXT_BLC_PWM_ENABLE) &&
+		panel->backlight.level != 0;
+
+	return 0;
+}
+
 int intel_panel_setup_backlight(struct drm_connector *connector, enum pipe pipe)
 {
 	struct drm_device *dev = connector->dev;
@@ -1350,7 +1429,13 @@ void intel_panel_init_backlight_funcs(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (IS_BROADWELL(dev) || (INTEL_INFO(dev)->gen >= 9)) {
+	if (IS_BROXTON(dev)) {
+		dev_priv->display.setup_backlight = bxt_setup_backlight;
+		dev_priv->display.enable_backlight = bxt_enable_backlight;
+		dev_priv->display.disable_backlight = bxt_disable_backlight;
+		dev_priv->display.set_backlight = bxt_set_backlight;
+		dev_priv->display.get_backlight = bxt_get_backlight;
+	} else if (IS_BROADWELL(dev) || IS_SKYLAKE(dev)) {
 		dev_priv->display.setup_backlight = bdw_setup_backlight;
 		dev_priv->display.enable_backlight = bdw_enable_backlight;
 		dev_priv->display.disable_backlight = pch_disable_backlight;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 555b896d2bda..eadc15cddbeb 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -67,7 +67,7 @@ static void skl_init_clock_gating(struct drm_device *dev)
 
 	gen9_init_clock_gating(dev);
 
-	if (INTEL_REVID(dev) == SKL_REVID_A0) {
+	if (INTEL_REVID(dev) <= SKL_REVID_B0) {
 		/*
 		 * WaDisableSDEUnitClockGating:skl
 		 * WaSetGAPSunitClckGateDisable:skl
@@ -75,6 +75,10 @@ static void skl_init_clock_gating(struct drm_device *dev)
 		I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 			   GEN8_GAPSUNIT_CLOCK_GATE_DISABLE |
 			   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
+
+		/* WaDisableVFUnitClockGating:skl */
+		I915_WRITE(GEN6_UCGCTL2, I915_READ(GEN6_UCGCTL2) |
+			   GEN6_VFUNIT_CLOCK_GATE_DISABLE);
 	}
 
 	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
@@ -84,8 +88,7 @@ static void skl_init_clock_gating(struct drm_device *dev)
 
 		/* WaDisableChickenBitTSGBarrierAckForFFSliceCS:skl */
 		I915_WRITE(FF_SLICE_CS_CHICKEN2,
-			   I915_READ(FF_SLICE_CS_CHICKEN2) |
-			   GEN9_TSG_BARRIER_ACK_DISABLE);
+			   _MASKED_BIT_ENABLE(GEN9_TSG_BARRIER_ACK_DISABLE));
 	}
 
 	if (INTEL_REVID(dev) <= SKL_REVID_E0)
@@ -94,6 +97,26 @@ static void skl_init_clock_gating(struct drm_device *dev)
 			   GEN8_LQSC_RO_PERF_DIS);
 }
 
+static void bxt_init_clock_gating(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	gen9_init_clock_gating(dev);
+
+	/*
+	 * FIXME:
+	 * GEN8_SDEUNIT_CLOCK_GATE_DISABLE applies on A0 only.
+	 * GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ applies on 3x6 GT SKUs only.
+	 */
+	 /* WaDisableSDEUnitClockGating:bxt */
+	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
+		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE |
+		   GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ);
+
+	/* FIXME: apply on A0 only */
+	I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
+}
+
 static void i915_pineview_get_mem_freq(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1792,7 +1815,7 @@ hsw_compute_linetime_wm(struct drm_device *dev, struct drm_crtc *crtc)
 	linetime = DIV_ROUND_CLOSEST(mode->crtc_htotal * 1000 * 8,
 				     mode->crtc_clock);
 	ips_linetime = DIV_ROUND_CLOSEST(mode->crtc_htotal * 1000 * 8,
-					 intel_ddi_get_cdclk_freq(dev_priv));
+					 dev_priv->display.get_display_clock_speed(dev_priv->dev));
 
 	return PIPE_WM_LINETIME_IPS_LINETIME(ips_linetime) |
 	       PIPE_WM_LINETIME_TIME(linetime);
@@ -1923,7 +1946,7 @@ static void intel_fixup_cur_wm_latency(struct drm_device *dev, uint16_t wm[5])
 int ilk_wm_max_level(const struct drm_device *dev)
 {
 	/* how many WM levels are we expecting */
-	if (IS_GEN9(dev))
+	if (INTEL_INFO(dev)->gen >= 9)
 		return 7;
 	else if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		return 4;
@@ -2536,6 +2559,7 @@ static bool ilk_disable_lp_wm(struct drm_device *dev)
  */
 
 #define SKL_DDB_SIZE		896	/* in blocks */
+#define BXT_DDB_SIZE		512
 
 static void
 skl_ddb_get_pipe_allocation_limits(struct drm_device *dev,
@@ -2554,7 +2578,10 @@ skl_ddb_get_pipe_allocation_limits(struct drm_device *dev,
 		return;
 	}
 
-	ddb_size = SKL_DDB_SIZE;
+	if (IS_BROXTON(dev))
+		ddb_size = BXT_DDB_SIZE;
+	else
+		ddb_size = SKL_DDB_SIZE;
 
 	ddb_size -= 4; /* 4 blocks for bypass path allocation */
 
@@ -2610,8 +2637,18 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 }
 
 static unsigned int
-skl_plane_relative_data_rate(const struct intel_plane_wm_parameters *p)
+skl_plane_relative_data_rate(const struct intel_plane_wm_parameters *p, int y)
 {
+
+	/* for planar format */
+	if (p->y_bytes_per_pixel) {
+		if (y)  /* y-plane data rate */
+			return p->horiz_pixels * p->vert_pixels * p->y_bytes_per_pixel;
+		else    /* uv-plane data rate */
+			return (p->horiz_pixels/2) * (p->vert_pixels/2) * p->bytes_per_pixel;
+	}
+
+	/* for packed formats */
 	return p->horiz_pixels * p->vert_pixels * p->bytes_per_pixel;
 }
 
@@ -2634,7 +2671,10 @@ skl_get_total_relative_data_rate(struct intel_crtc *intel_crtc,
 		if (!p->enabled)
 			continue;
 
-		total_data_rate += skl_plane_relative_data_rate(p);
+		total_data_rate += skl_plane_relative_data_rate(p, 0); /* packed/uv */
+		if (p->y_bytes_per_pixel) {
+			total_data_rate += skl_plane_relative_data_rate(p, 1); /* y-plane */
+		}
 	}
 
 	return total_data_rate;
@@ -2653,6 +2693,7 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 	struct skl_ddb_entry *alloc = &ddb->pipe[pipe];
 	uint16_t alloc_size, start, cursor_blocks;
 	uint16_t minimum[I915_MAX_PLANES];
+	uint16_t y_minimum[I915_MAX_PLANES];
 	unsigned int total_data_rate;
 	int plane;
 
@@ -2681,6 +2722,8 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 
 		minimum[plane] = 8;
 		alloc_size -= minimum[plane];
+		y_minimum[plane] = p->y_bytes_per_pixel ? 8 : 0;
+		alloc_size -= y_minimum[plane];
 	}
 
 	/*
@@ -2694,16 +2737,17 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 	start = alloc->start;
 	for (plane = 0; plane < intel_num_planes(intel_crtc); plane++) {
 		const struct intel_plane_wm_parameters *p;
-		unsigned int data_rate;
-		uint16_t plane_blocks;
+		unsigned int data_rate, y_data_rate;
+		uint16_t plane_blocks, y_plane_blocks = 0;
 
 		p = &params->plane[plane];
 		if (!p->enabled)
 			continue;
 
-		data_rate = skl_plane_relative_data_rate(p);
+		data_rate = skl_plane_relative_data_rate(p, 0);
 
 		/*
+		 * allocation for (packed formats) or (uv-plane part of planar format):
 		 * promote the expression to 64 bits to avoid overflowing, the
 		 * result is < available as data_rate / total_data_rate < 1
 		 */
@@ -2715,6 +2759,22 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 		ddb->plane[pipe][plane].end = start + plane_blocks;
 
 		start += plane_blocks;
+
+		/*
+		 * allocation for y_plane part of planar format:
+		 */
+		if (p->y_bytes_per_pixel) {
+			y_data_rate = skl_plane_relative_data_rate(p, 1);
+			y_plane_blocks = y_minimum[plane];
+			y_plane_blocks += div_u64((uint64_t)alloc_size * y_data_rate,
+						total_data_rate);
+
+			ddb->y_plane[pipe][plane].start = start;
+			ddb->y_plane[pipe][plane].end = start + y_plane_blocks;
+
+			start += y_plane_blocks;
+		}
+
 	}
 
 }
@@ -2827,13 +2887,18 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc,
 		p->pixel_rate = skl_pipe_pixel_rate(intel_crtc->config);
 
 		fb = crtc->primary->state->fb;
+		/* For planar: Bpp is for uv plane, y_Bpp is for y plane */
 		if (fb) {
 			p->plane[0].enabled = true;
-			p->plane[0].bytes_per_pixel = fb->bits_per_pixel / 8;
+			p->plane[0].bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
+				drm_format_plane_cpp(fb->pixel_format, 1) : fb->bits_per_pixel / 8;
+			p->plane[0].y_bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
+				drm_format_plane_cpp(fb->pixel_format, 0) : 0;
 			p->plane[0].tiling = fb->modifier[0];
 		} else {
 			p->plane[0].enabled = false;
 			p->plane[0].bytes_per_pixel = 0;
+			p->plane[0].y_bytes_per_pixel = 0;
 			p->plane[0].tiling = DRM_FORMAT_MOD_NONE;
 		}
 		p->plane[0].horiz_pixels = intel_crtc->config->pipe_src_w;
@@ -2841,6 +2906,7 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc,
 		p->plane[0].rotation = crtc->primary->state->rotation;
 
 		fb = crtc->cursor->state->fb;
+		p->cursor.y_bytes_per_pixel = 0;
 		if (fb) {
 			p->cursor.enabled = true;
 			p->cursor.bytes_per_pixel = fb->bits_per_pixel / 8;
@@ -2876,22 +2942,25 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 	uint32_t plane_bytes_per_line, plane_blocks_per_line;
 	uint32_t res_blocks, res_lines;
 	uint32_t selected_result;
+	uint8_t bytes_per_pixel;
 
 	if (latency == 0 || !p->active || !p_params->enabled)
 		return false;
 
+	bytes_per_pixel = p_params->y_bytes_per_pixel ?
+		p_params->y_bytes_per_pixel :
+		p_params->bytes_per_pixel;
 	method1 = skl_wm_method1(p->pixel_rate,
-				 p_params->bytes_per_pixel,
+				 bytes_per_pixel,
 				 latency);
 	method2 = skl_wm_method2(p->pixel_rate,
 				 p->pipe_htotal,
 				 p_params->horiz_pixels,
-				 p_params->bytes_per_pixel,
+				 bytes_per_pixel,
 				 p_params->tiling,
 				 latency);
 
-	plane_bytes_per_line = p_params->horiz_pixels *
-					p_params->bytes_per_pixel;
+	plane_bytes_per_line = p_params->horiz_pixels * bytes_per_pixel;
 	plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 
 	if (p_params->tiling == I915_FORMAT_MOD_Y_TILED ||
@@ -3108,10 +3177,14 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv,
 				   new->plane_trans[pipe][i]);
 		I915_WRITE(CUR_WM_TRANS(pipe), new->cursor_trans[pipe]);
 
-		for (i = 0; i < intel_num_planes(crtc); i++)
+		for (i = 0; i < intel_num_planes(crtc); i++) {
 			skl_ddb_entry_write(dev_priv,
 					    PLANE_BUF_CFG(pipe, i),
 					    &new->ddb.plane[pipe][i]);
+			skl_ddb_entry_write(dev_priv,
+					    PLANE_NV12_BUF_CFG(pipe, i),
+					    &new->ddb.y_plane[pipe][i]);
+		}
 
 		skl_ddb_entry_write(dev_priv, CUR_BUF_CFG(pipe),
 				    &new->ddb.cursor[pipe]);
@@ -3176,7 +3249,7 @@ static void skl_flush_wm_values(struct drm_i915_private *dev_priv,
 {
 	struct drm_device *dev = dev_priv->dev;
 	struct skl_ddb_allocation *cur_ddb, *new_ddb;
-	bool reallocated[I915_MAX_PIPES] = {false, false, false};
+	bool reallocated[I915_MAX_PIPES] = {};
 	struct intel_crtc *crtc;
 	enum pipe pipe;
 
@@ -3269,6 +3342,7 @@ static bool skl_update_pipe_wm(struct drm_crtc *crtc,
 		return false;
 
 	intel_crtc->wm.skl_active = *pipe_wm;
+
 	return true;
 }
 
@@ -3362,8 +3436,16 @@ skl_update_sprite_wm(struct drm_plane *plane, struct drm_crtc *crtc,
 	intel_plane->wm.scaled = scaled;
 	intel_plane->wm.horiz_pixels = sprite_width;
 	intel_plane->wm.vert_pixels = sprite_height;
-	intel_plane->wm.bytes_per_pixel = pixel_size;
 	intel_plane->wm.tiling = DRM_FORMAT_MOD_NONE;
+
+	/* For planar: Bpp is for UV plane, y_Bpp is for Y plane */
+	intel_plane->wm.bytes_per_pixel =
+		(fb && fb->pixel_format == DRM_FORMAT_NV12) ?
+		drm_format_plane_cpp(plane->state->fb->pixel_format, 1) : pixel_size;
+	intel_plane->wm.y_bytes_per_pixel =
+		(fb && fb->pixel_format == DRM_FORMAT_NV12) ?
+		drm_format_plane_cpp(plane->state->fb->pixel_format, 0) : 0;
+
 	/*
 	 * Framebuffer can be NULL on plane disable, but it does not
 	 * matter for watermarks if we assume no tiling in that case.
@@ -3928,6 +4010,8 @@ static void gen6_set_rps_thresholds(struct drm_i915_private *dev_priv, u8 val)
 		    GEN6_RP_DOWN_IDLE_AVG);
 
 	dev_priv->rps.power = new_power;
+	dev_priv->rps.up_threshold = threshold_up;
+	dev_priv->rps.down_threshold = threshold_down;
 	dev_priv->rps.last_adj = 0;
 }
 
@@ -3999,8 +4083,11 @@ static void valleyview_set_rps(struct drm_device *dev, u8 val)
 		      "Odd GPU freq value\n"))
 		val &= ~1;
 
-	if (val != dev_priv->rps.cur_freq)
+	if (val != dev_priv->rps.cur_freq) {
 		vlv_punit_write(dev_priv, PUNIT_REG_GPU_FREQ_REQ, val);
+		if (!IS_CHERRYVIEW(dev_priv))
+			gen6_set_rps_thresholds(dev_priv, val);
+	}
 
 	I915_WRITE(GEN6_PMINTRMSK, gen6_rps_pm_mask(dev_priv, val));
 
@@ -4008,50 +4095,25 @@ static void valleyview_set_rps(struct drm_device *dev, u8 val)
 	trace_intel_gpu_freq_change(intel_gpu_freq(dev_priv, val));
 }
 
-/* vlv_set_rps_idle: Set the frequency to Rpn if Gfx clocks are down
+/* vlv_set_rps_idle: Set the frequency to idle, if Gfx clocks are down
  *
  * * If Gfx is Idle, then
- * 1. Mask Turbo interrupts
- * 2. Bring up Gfx clock
- * 3. Change the freq to Rpn and wait till P-Unit updates freq
- * 4. Clear the Force GFX CLK ON bit so that Gfx can down
- * 5. Unmask Turbo interrupts
+ * 1. Forcewake Media well.
+ * 2. Request idle freq.
+ * 3. Release Forcewake of Media well.
 */
 static void vlv_set_rps_idle(struct drm_i915_private *dev_priv)
 {
-	struct drm_device *dev = dev_priv->dev;
 	u32 val = dev_priv->rps.idle_freq;
 
-	/* CHV and latest VLV don't need to force the gfx clock */
-	if (IS_CHERRYVIEW(dev) || dev->pdev->revision >= 0xd) {
-		valleyview_set_rps(dev_priv->dev, val);
-		return;
-	}
-
-	/*
-	 * When we are idle.  Drop to min voltage state.
-	 */
-
 	if (dev_priv->rps.cur_freq <= val)
 		return;
 
-	/* Mask turbo interrupt so that they will not come in between */
-	I915_WRITE(GEN6_PMINTRMSK,
-		   gen6_sanitize_rps_pm_mask(dev_priv, ~0));
-
-	vlv_force_gfx_clock(dev_priv, true);
-
-	dev_priv->rps.cur_freq = val;
-
-	vlv_punit_write(dev_priv, PUNIT_REG_GPU_FREQ_REQ, val);
-
-	if (wait_for(((vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS))
-				& GENFREQSTATUS) == 0, 100))
-		DRM_ERROR("timed out waiting for Punit\n");
-
-	vlv_force_gfx_clock(dev_priv, false);
-
-	I915_WRITE(GEN6_PMINTRMSK, gen6_rps_pm_mask(dev_priv, val));
+	/* Wake up the media well, as that takes a lot less
+	 * power than the Render well. */
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_MEDIA);
+	valleyview_set_rps(dev_priv->dev, val);
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_MEDIA);
 }
 
 void gen6_rps_busy(struct drm_i915_private *dev_priv)
@@ -4080,21 +4142,47 @@ void gen6_rps_idle(struct drm_i915_private *dev_priv)
 		I915_WRITE(GEN6_PMINTRMSK, 0xffffffff);
 	}
 	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	spin_lock(&dev_priv->rps.client_lock);
+	while (!list_empty(&dev_priv->rps.clients))
+		list_del_init(dev_priv->rps.clients.next);
+	spin_unlock(&dev_priv->rps.client_lock);
 }
 
-void gen6_rps_boost(struct drm_i915_private *dev_priv)
+void gen6_rps_boost(struct drm_i915_private *dev_priv,
+		    struct intel_rps_client *rps,
+		    unsigned long submitted)
 {
-	u32 val;
+	/* This is intentionally racy! We peek at the state here, then
+	 * validate inside the RPS worker.
+	 */
+	if (!(dev_priv->mm.busy &&
+	      dev_priv->rps.enabled &&
+	      dev_priv->rps.cur_freq < dev_priv->rps.max_freq_softlimit))
+		return;
 
-	mutex_lock(&dev_priv->rps.hw_lock);
-	val = dev_priv->rps.max_freq_softlimit;
-	if (dev_priv->rps.enabled &&
-	    dev_priv->mm.busy &&
-	    dev_priv->rps.cur_freq < val) {
-		intel_set_rps(dev_priv->dev, val);
-		dev_priv->rps.last_adj = 0;
+	/* Force a RPS boost (and don't count it against the client) if
+	 * the GPU is severely congested.
+	 */
+	if (rps && time_after(jiffies, submitted + DRM_I915_THROTTLE_JIFFIES))
+		rps = NULL;
+
+	spin_lock(&dev_priv->rps.client_lock);
+	if (rps == NULL || list_empty(&rps->link)) {
+		spin_lock_irq(&dev_priv->irq_lock);
+		if (dev_priv->rps.interrupts_enabled) {
+			dev_priv->rps.client_boost = true;
+			queue_work(dev_priv->wq, &dev_priv->rps.work);
+		}
+		spin_unlock_irq(&dev_priv->irq_lock);
+
+		if (rps != NULL) {
+			list_add(&rps->link, &dev_priv->rps.clients);
+			rps->boosts++;
+		} else
+			dev_priv->rps.boosts++;
 	}
-	mutex_unlock(&dev_priv->rps.hw_lock);
+	spin_unlock(&dev_priv->rps.client_lock);
 }
 
 void intel_set_rps(struct drm_device *dev, u8 val)
@@ -4248,8 +4336,8 @@ static void gen6_init_rps_frequencies(struct drm_device *dev)
 	if (dev_priv->rps.min_freq_softlimit == 0) {
 		if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 			dev_priv->rps.min_freq_softlimit =
-				/* max(RPe, 450 MHz) */
-				max(dev_priv->rps.efficient_freq, (u8) 9);
+				max_t(int, dev_priv->rps.efficient_freq,
+				      intel_freq_opcode(dev_priv, 450));
 		else
 			dev_priv->rps.min_freq_softlimit =
 				dev_priv->rps.min_freq;
@@ -4323,8 +4411,13 @@ static void gen9_enable_rc6(struct drm_device *dev)
 				   GEN6_RC_CTL_EI_MODE(1) |
 				   rc6_mask);
 
-	/* 3b: Enable Coarse Power Gating only when RC6 is enabled */
-	I915_WRITE(GEN9_PG_ENABLE, (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ? 3 : 0);
+	/*
+	 * 3b: Enable Coarse Power Gating only when RC6 is enabled.
+	 * WaDisableRenderPowerGating:skl,bxt - Render PG need to be disabled with RC6.
+	 */
+	I915_WRITE(GEN9_PG_ENABLE, (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ?
+			GEN9_MEDIA_PG_ENABLE : 0);
+
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 
@@ -4663,24 +4756,6 @@ static int cherryview_rps_guar_freq(struct drm_i915_private *dev_priv)
 	return rp1;
 }
 
-static int cherryview_rps_min_freq(struct drm_i915_private *dev_priv)
-{
-	struct drm_device *dev = dev_priv->dev;
-	u32 val, rpn;
-
-	if (dev->pdev->revision >= 0x20) {
-		val = vlv_punit_read(dev_priv, FB_GFX_FMIN_AT_VMIN_FUSE);
-		rpn = ((val >> FB_GFX_FMIN_AT_VMIN_FUSE_SHIFT) &
-		       FB_GFX_FREQ_FUSE_MASK);
-	} else { /* For pre-production hardware */
-		val = vlv_punit_read(dev_priv, PUNIT_GPU_STATUS_REG);
-		rpn = ((val >> PUNIT_GPU_STATIS_GFX_MIN_FREQ_SHIFT) &
-		       PUNIT_GPU_STATUS_GFX_MIN_FREQ_MASK);
-	}
-
-	return rpn;
-}
-
 static int valleyview_rps_guar_freq(struct drm_i915_private *dev_priv)
 {
 	u32 val, rp1;
@@ -4887,9 +4962,9 @@ static void cherryview_init_gt_powersave(struct drm_device *dev)
 
 	mutex_lock(&dev_priv->rps.hw_lock);
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	val = vlv_cck_read(dev_priv, CCK_FUSE_REG);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	switch ((val >> 2) & 0x7) {
 	case 0:
@@ -4932,7 +5007,8 @@ static void cherryview_init_gt_powersave(struct drm_device *dev)
 			 intel_gpu_freq(dev_priv, dev_priv->rps.rp1_freq),
 			 dev_priv->rps.rp1_freq);
 
-	dev_priv->rps.min_freq = cherryview_rps_min_freq(dev_priv);
+	/* PUnit validated range is only [RPe, RP0] */
+	dev_priv->rps.min_freq = dev_priv->rps.efficient_freq;
 	DRM_DEBUG_DRIVER("min GPU freq: %d MHz (%u)\n",
 			 intel_gpu_freq(dev_priv, dev_priv->rps.min_freq),
 			 dev_priv->rps.min_freq);
@@ -4994,8 +5070,8 @@ static void cherryview_enable_rps(struct drm_device *dev)
 		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
 	I915_WRITE(GEN6_RC_SLEEP, 0);
 
-	/* TO threshold set to 1750 us ( 0x557 * 1.28 us) */
-	I915_WRITE(GEN6_RC6_THRESHOLD, 0x557);
+	/* TO threshold set to 500 us ( 0x186 * 1.28 us) */
+	I915_WRITE(GEN6_RC6_THRESHOLD, 0x186);
 
 	/* allows RC6 residency counter to work */
 	I915_WRITE(VLV_COUNTER_CONTROL,
@@ -5030,6 +5106,12 @@ static void cherryview_enable_rps(struct drm_device *dev)
 		   GEN6_RP_UP_BUSY_AVG |
 		   GEN6_RP_DOWN_IDLE_AVG);
 
+	/* Setting Fixed Bias */
+	val = VLV_OVERRIDE_EN |
+		  VLV_SOC_TDP_EN |
+		  CHV_BIAS_CPU_50_SOC_50;
+	vlv_punit_write(dev_priv, VLV_TURBO_SOC_OVERRIDE, val);
+
 	val = vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS);
 
 	/* RPS code assumes GPLL is used */
@@ -5114,6 +5196,12 @@ static void valleyview_enable_rps(struct drm_device *dev)
 
 	I915_WRITE(GEN6_RC_CONTROL, rc6_mode);
 
+	/* Setting Fixed Bias */
+	val = VLV_OVERRIDE_EN |
+		  VLV_SOC_TDP_EN |
+		  VLV_BIAS_CPU_125_SOC_875;
+	vlv_punit_write(dev_priv, VLV_TURBO_SOC_OVERRIDE, val);
+
 	val = vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS);
 
 	/* RPS code assumes GPLL is used */
@@ -5796,13 +5884,15 @@ static void ibx_init_clock_gating(struct drm_device *dev)
 static void g4x_disable_trickle_feed(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe;
+	enum pipe pipe;
 
 	for_each_pipe(dev_priv, pipe) {
 		I915_WRITE(DSPCNTR(pipe),
 			   I915_READ(DSPCNTR(pipe)) |
 			   DISPPLANE_TRICKLE_FEED_DISABLE);
-		intel_flush_primary_plane(dev_priv, pipe);
+
+		I915_WRITE(DSPSURF(pipe), I915_READ(DSPSURF(pipe)));
+		POSTING_READ(DSPSURF(pipe));
 	}
 }
 
@@ -6092,10 +6182,9 @@ static void broadwell_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe;
+	uint32_t misccpctl;
 
-	I915_WRITE(WM3_LP_ILK, 0);
-	I915_WRITE(WM2_LP_ILK, 0);
-	I915_WRITE(WM1_LP_ILK, 0);
+	ilk_init_lp_watermarks(dev);
 
 	/* WaSwitchSolVfFArbitrationPriority:bdw */
 	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | HSW_ECOCHK_ARB_PRIO_SOL);
@@ -6124,6 +6213,22 @@ static void broadwell_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
 
+	/*
+	 * WaProgramL3SqcReg1Default:bdw
+	 * WaTempDisableDOPClkGating:bdw
+	 */
+	misccpctl = I915_READ(GEN7_MISCCPCTL);
+	I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE);
+	I915_WRITE(GEN8_L3SQCREG1, BDW_WA_L3SQCREG1_DEFAULT);
+	I915_WRITE(GEN7_MISCCPCTL, misccpctl);
+
+	/*
+	 * WaGttCachingOffByDefault:bdw
+	 * GTT cache may not work with big pages, so if those
+	 * are ever enabled GTT cache may need to be disabled.
+	 */
+	I915_WRITE(HSW_GTT_CACHE_EN, GTT_CACHE_EN_ALL);
+
 	lpt_init_clock_gating(dev);
 }
 
@@ -6399,6 +6504,12 @@ static void cherryview_init_clock_gating(struct drm_device *dev)
 	/* WaDisableSDEUnitClockGating:chv */
 	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
+
+	/*
+	 * GTT cache may not work with big pages, so if those
+	 * are ever enabled GTT cache may need to be disabled.
+	 */
+	I915_WRITE(HSW_GTT_CACHE_EN, GTT_CACHE_EN_ALL);
 }
 
 static void g4x_init_clock_gating(struct drm_device *dev)
@@ -6542,7 +6653,12 @@ void intel_init_pm(struct drm_device *dev)
 	if (INTEL_INFO(dev)->gen >= 9) {
 		skl_setup_wm_latency(dev);
 
-		dev_priv->display.init_clock_gating = skl_init_clock_gating;
+		if (IS_BROXTON(dev))
+			dev_priv->display.init_clock_gating =
+				bxt_init_clock_gating;
+		else if (IS_SKYLAKE(dev))
+			dev_priv->display.init_clock_gating =
+				skl_init_clock_gating;
 		dev_priv->display.update_wm = skl_update_wm;
 		dev_priv->display.update_sprite_wm = skl_update_sprite_wm;
 	} else if (HAS_PCH_SPLIT(dev)) {
@@ -6760,14 +6876,58 @@ int intel_freq_opcode(struct drm_i915_private *dev_priv, int val)
 		return val / GT_FREQUENCY_MULTIPLIER;
 }
 
+struct request_boost {
+	struct work_struct work;
+	struct drm_i915_gem_request *req;
+};
+
+static void __intel_rps_boost_work(struct work_struct *work)
+{
+	struct request_boost *boost = container_of(work, struct request_boost, work);
+	struct drm_i915_gem_request *req = boost->req;
+
+	if (!i915_gem_request_completed(req, true))
+		gen6_rps_boost(to_i915(req->ring->dev), NULL,
+			       req->emitted_jiffies);
+
+	i915_gem_request_unreference__unlocked(req);
+	kfree(boost);
+}
+
+void intel_queue_rps_boost_for_request(struct drm_device *dev,
+				       struct drm_i915_gem_request *req)
+{
+	struct request_boost *boost;
+
+	if (req == NULL || INTEL_INFO(dev)->gen < 6)
+		return;
+
+	if (i915_gem_request_completed(req, true))
+		return;
+
+	boost = kmalloc(sizeof(*boost), GFP_ATOMIC);
+	if (boost == NULL)
+		return;
+
+	i915_gem_request_reference(req);
+	boost->req = req;
+
+	INIT_WORK(&boost->work, __intel_rps_boost_work);
+	queue_work(to_i915(dev)->wq, &boost->work);
+}
+
 void intel_pm_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	mutex_init(&dev_priv->rps.hw_lock);
+	spin_lock_init(&dev_priv->rps.client_lock);
 
 	INIT_DELAYED_WORK(&dev_priv->rps.delayed_resume_work,
 			  intel_gen6_powersave_work);
+	INIT_LIST_HEAD(&dev_priv->rps.clients);
+	INIT_LIST_HEAD(&dev_priv->rps.semaphores.link);
+	INIT_LIST_HEAD(&dev_priv->rps.mmioflips.link);
 
 	dev_priv->pm.suspended = false;
 }
diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c
index a8f9348259ae..5ee0fa57ed19 100644
--- a/drivers/gpu/drm/i915/intel_psr.c
+++ b/drivers/gpu/drm/i915/intel_psr.c
@@ -117,6 +117,19 @@ static void vlv_psr_setup_vsc(struct intel_dp *intel_dp)
 	I915_WRITE(VLV_VSCSDP(pipe), val);
 }
 
+static void skl_psr_setup_su_vsc(struct intel_dp *intel_dp)
+{
+	struct edp_vsc_psr psr_vsc;
+
+	/* Prepare VSC Header for SU as per EDP 1.4 spec, Table 6.11 */
+	memset(&psr_vsc, 0, sizeof(psr_vsc));
+	psr_vsc.sdp_header.HB0 = 0;
+	psr_vsc.sdp_header.HB1 = 0x7;
+	psr_vsc.sdp_header.HB2 = 0x3;
+	psr_vsc.sdp_header.HB3 = 0xb;
+	intel_psr_write_vsc(intel_dp, &psr_vsc);
+}
+
 static void hsw_psr_setup_vsc(struct intel_dp *intel_dp)
 {
 	struct edp_vsc_psr psr_vsc;
@@ -133,7 +146,7 @@ static void hsw_psr_setup_vsc(struct intel_dp *intel_dp)
 static void vlv_psr_enable_sink(struct intel_dp *intel_dp)
 {
 	drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG,
-			   DP_PSR_ENABLE);
+			   DP_PSR_ENABLE | DP_PSR_MAIN_LINK_ACTIVE);
 }
 
 static void hsw_psr_enable_sink(struct intel_dp *intel_dp)
@@ -157,13 +170,14 @@ static void hsw_psr_enable_sink(struct intel_dp *intel_dp)
 
 	aux_clock_divider = intel_dp->get_aux_clock_divider(intel_dp, 0);
 
-	/* Enable PSR in sink */
-	if (dev_priv->psr.link_standby)
-		drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG,
-				   DP_PSR_ENABLE | DP_PSR_MAIN_LINK_ACTIVE);
-	else
-		drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG,
-				   DP_PSR_ENABLE & ~DP_PSR_MAIN_LINK_ACTIVE);
+	drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG,
+			   DP_PSR_ENABLE & ~DP_PSR_MAIN_LINK_ACTIVE);
+
+	/* Enable AUX frame sync at sink */
+	if (dev_priv->psr.aux_frame_sync)
+		drm_dp_dpcd_writeb(&intel_dp->aux,
+				DP_SINK_DEVICE_AUX_FRAME_SYNC_CONF,
+				DP_AUX_FRAME_SYNC_ENABLE);
 
 	aux_data_reg = (INTEL_INFO(dev)->gen >= 9) ?
 				DPA_AUX_CH_DATA1 : EDP_PSR_AUX_DATA1(dev);
@@ -183,8 +197,10 @@ static void hsw_psr_enable_sink(struct intel_dp *intel_dp)
 		val |= DP_AUX_CH_CTL_TIME_OUT_1600us;
 		val &= ~DP_AUX_CH_CTL_MESSAGE_SIZE_MASK;
 		val |= (sizeof(aux_msg) << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT);
-		/* Use hardcoded data values for PSR */
+		/* Use hardcoded data values for PSR, frame sync and GTC */
 		val &= ~DP_AUX_CH_CTL_PSR_DATA_AUX_REG_SKL;
+		val &= ~DP_AUX_CH_CTL_FS_DATA_AUX_REG_SKL;
+		val &= ~DP_AUX_CH_CTL_GTC_DATA_AUX_REG_SKL;
 		I915_WRITE(aux_ctl_reg, val);
 	} else {
 		I915_WRITE(aux_ctl_reg,
@@ -193,6 +209,8 @@ static void hsw_psr_enable_sink(struct intel_dp *intel_dp)
 		   (precharge << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
 		   (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT));
 	}
+
+	drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG, DP_PSR_ENABLE);
 }
 
 static void vlv_psr_enable_source(struct intel_dp *intel_dp)
@@ -232,6 +250,7 @@ static void hsw_psr_enable_source(struct intel_dp *intel_dp)
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+
 	uint32_t max_sleep_time = 0x1f;
 	/* Lately it was identified that depending on panel idle frame count
 	 * calculated at HW can be off by 1. So let's use what came
@@ -242,19 +261,25 @@ static void hsw_psr_enable_source(struct intel_dp *intel_dp)
 	uint32_t val = 0x0;
 	const uint32_t link_entry_time = EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
 
-	if (dev_priv->psr.link_standby) {
-		val |= EDP_PSR_LINK_STANDBY;
+	if (intel_dp->psr_dpcd[1] & DP_PSR_NO_TRAIN_ON_EXIT) {
+		/* It doesn't mean we shouldn't send TPS patters, so let's
+		   send the minimal TP1 possible and skip TP2. */
+		val |= EDP_PSR_TP1_TIME_100us;
 		val |= EDP_PSR_TP2_TP3_TIME_0us;
-		val |= EDP_PSR_TP1_TIME_0us;
 		val |= EDP_PSR_SKIP_AUX_EXIT;
-	} else
-		val |= EDP_PSR_LINK_DISABLE;
+		/* Sink should be able to train with the 5 or 6 idle patterns */
+		idle_frames += 4;
+	}
 
 	I915_WRITE(EDP_PSR_CTL(dev), val |
 		   (IS_BROADWELL(dev) ? 0 : link_entry_time) |
 		   max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT |
 		   idle_frames << EDP_PSR_IDLE_FRAME_SHIFT |
 		   EDP_PSR_ENABLE);
+
+	if (dev_priv->psr.psr2_support)
+		I915_WRITE(EDP_PSR2_CTL, EDP_PSR2_ENABLE |
+				EDP_SU_TRACK_ENABLE | EDP_PSR2_TP2_TIME_100);
 }
 
 static bool intel_psr_match_conditions(struct intel_dp *intel_dp)
@@ -294,6 +319,12 @@ static bool intel_psr_match_conditions(struct intel_dp *intel_dp)
 		return false;
 	}
 
+	if (!IS_VALLEYVIEW(dev) && ((dev_priv->vbt.psr.full_link) ||
+				    (dig_port->port != PORT_A))) {
+		DRM_DEBUG_KMS("PSR condition failed: Link Standby requested/needed but not supported on this platform\n");
+		return false;
+	}
+
 	dev_priv->psr.source_ok = true;
 	return true;
 }
@@ -332,6 +363,7 @@ void intel_psr_enable(struct intel_dp *intel_dp)
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *crtc = to_intel_crtc(intel_dig_port->base.base.crtc);
 
 	if (!HAS_PSR(dev)) {
 		DRM_DEBUG_KMS("PSR not supported on this platform\n");
@@ -352,18 +384,20 @@ void intel_psr_enable(struct intel_dp *intel_dp)
 	if (!intel_psr_match_conditions(intel_dp))
 		goto unlock;
 
-	/* First we check VBT, but we must respect sink and source
-	 * known restrictions */
-	dev_priv->psr.link_standby = dev_priv->vbt.psr.full_link;
-	if ((intel_dp->psr_dpcd[1] & DP_PSR_NO_TRAIN_ON_EXIT) ||
-	    (IS_BROADWELL(dev) && intel_dig_port->port != PORT_A))
-		dev_priv->psr.link_standby = true;
-
 	dev_priv->psr.busy_frontbuffer_bits = 0;
 
 	if (HAS_DDI(dev)) {
 		hsw_psr_setup_vsc(intel_dp);
 
+		if (dev_priv->psr.psr2_support) {
+			/* PSR2 is restricted to work with panel resolutions upto 3200x2000 */
+			if (crtc->config->pipe_src_w > 3200 ||
+				crtc->config->pipe_src_h > 2000)
+				dev_priv->psr.psr2_support = false;
+			else
+				skl_psr_setup_su_vsc(intel_dp);
+		}
+
 		/* Avoid continuous PSR exit by masking memup and hpd */
 		I915_WRITE(EDP_PSR_DEBUG_CTL(dev), EDP_PSR_DEBUG_MASK_MEMUP |
 			   EDP_PSR_DEBUG_MASK_HPD | EDP_PSR_DEBUG_MASK_LPSP);
@@ -560,6 +594,48 @@ static void intel_psr_exit(struct drm_device *dev)
 }
 
 /**
+ * intel_psr_single_frame_update - Single Frame Update
+ * @dev: DRM device
+ *
+ * Some platforms support a single frame update feature that is used to
+ * send and update only one frame on Remote Frame Buffer.
+ * So far it is only implemented for Valleyview and Cherryview because
+ * hardware requires this to be done before a page flip.
+ */
+void intel_psr_single_frame_update(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_crtc *crtc;
+	enum pipe pipe;
+	u32 val;
+
+	/*
+	 * Single frame update is already supported on BDW+ but it requires
+	 * many W/A and it isn't really needed.
+	 */
+	if (!IS_VALLEYVIEW(dev))
+		return;
+
+	mutex_lock(&dev_priv->psr.lock);
+	if (!dev_priv->psr.enabled) {
+		mutex_unlock(&dev_priv->psr.lock);
+		return;
+	}
+
+	crtc = dp_to_dig_port(dev_priv->psr.enabled)->base.base.crtc;
+	pipe = to_intel_crtc(crtc)->pipe;
+	val = I915_READ(VLV_PSRCTL(pipe));
+
+	/*
+	 * We need to set this bit before writing registers for a flip.
+	 * This bit will be self-clear when it gets to the PSR active state.
+	 */
+	I915_WRITE(VLV_PSRCTL(pipe), val | VLV_EDP_PSR_SINGLE_FRAME_UPDATE);
+
+	mutex_unlock(&dev_priv->psr.lock);
+}
+
+/**
  * intel_psr_invalidate - Invalidade PSR
  * @dev: DRM device
  * @frontbuffer_bits: frontbuffer plane tracking bits
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 005b5e04de4d..3817a6f00d9e 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -908,57 +908,63 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t tmp;
 
-	/* WaDisablePartialInstShootdown:skl */
+	/* WaDisablePartialInstShootdown:skl,bxt */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
 			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE);
 
-	/* Syncing dependencies between camera and graphics */
+	/* Syncing dependencies between camera and graphics:skl,bxt */
 	WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3,
 			  GEN9_DISABLE_OCL_OOB_SUPPRESS_LOGIC);
 
-	if (INTEL_REVID(dev) == SKL_REVID_A0 ||
-	    INTEL_REVID(dev) == SKL_REVID_B0) {
-		/* WaDisableDgMirrorFixInHalfSliceChicken5:skl */
+	if ((IS_SKYLAKE(dev) && (INTEL_REVID(dev) == SKL_REVID_A0 ||
+	    INTEL_REVID(dev) == SKL_REVID_B0)) ||
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0)) {
+		/* WaDisableDgMirrorFixInHalfSliceChicken5:skl,bxt */
 		WA_CLR_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN5,
 				  GEN9_DG_MIRROR_FIX_ENABLE);
 	}
 
-	if (IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) {
-		/* WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken:skl */
+	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) ||
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0)) {
+		/* WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken:skl,bxt */
 		WA_SET_BIT_MASKED(GEN7_COMMON_SLICE_CHICKEN1,
 				  GEN9_RHWO_OPTIMIZATION_DISABLE);
 		WA_SET_BIT_MASKED(GEN9_SLICE_COMMON_ECO_CHICKEN0,
 				  DISABLE_PIXEL_MASK_CAMMING);
 	}
 
-	if (INTEL_REVID(dev) >= SKL_REVID_C0) {
-		/* WaEnableYV12BugFixInHalfSliceChicken7:skl */
+	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) >= SKL_REVID_C0) ||
+	    IS_BROXTON(dev)) {
+		/* WaEnableYV12BugFixInHalfSliceChicken7:skl,bxt */
 		WA_SET_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN7,
 				  GEN9_ENABLE_YV12_BUGFIX);
 	}
 
-	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
-		/*
-		 *Use Force Non-Coherent whenever executing a 3D context. This
-		 * is a workaround for a possible hang in the unlikely event
-		 * a TLB invalidation occurs during a PSD flush.
-		 */
-		/* WaForceEnableNonCoherent:skl */
-		WA_SET_BIT_MASKED(HDC_CHICKEN0,
-				  HDC_FORCE_NON_COHERENT);
-	}
-
-	/* Wa4x4STCOptimizationDisable:skl */
+	/* Wa4x4STCOptimizationDisable:skl,bxt */
 	WA_SET_BIT_MASKED(CACHE_MODE_1, GEN8_4x4_STC_OPTIMIZATION_DISABLE);
 
-	/* WaDisablePartialResolveInVc:skl */
+	/* WaDisablePartialResolveInVc:skl,bxt */
 	WA_SET_BIT_MASKED(CACHE_MODE_1, GEN9_PARTIAL_RESOLVE_IN_VC_DISABLE);
 
-	/* WaCcsTlbPrefetchDisable:skl */
+	/* WaCcsTlbPrefetchDisable:skl,bxt */
 	WA_CLR_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN5,
 			  GEN9_CCS_TLB_PREFETCH_ENABLE);
 
+	/* WaDisableMaskBasedCammingInRCC:skl,bxt */
+	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) == SKL_REVID_C0) ||
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0))
+		WA_SET_BIT_MASKED(SLICE_ECO_CHICKEN0,
+				  PIXEL_MASK_CAMMING_DISABLE);
+
+	/* WaForceContextSaveRestoreNonCoherent:skl,bxt */
+	tmp = HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT;
+	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) == SKL_REVID_F0) ||
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) >= BXT_REVID_B0))
+		tmp |= HDC_FORCE_CSR_NON_COHERENT_OVR_DISABLE;
+	WA_SET_BIT_MASKED(HDC_CHICKEN0, tmp);
+
 	return 0;
 }
 
@@ -1024,9 +1030,41 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 				  HDC_FENCE_DEST_SLM_DISABLE |
 				  HDC_BARRIER_PERFORMANCE_DISABLE);
 
+	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
+		/*
+		 *Use Force Non-Coherent whenever executing a 3D context. This
+		 * is a workaround for a possible hang in the unlikely event
+		 * a TLB invalidation occurs during a PSD flush.
+		 */
+		/* WaForceEnableNonCoherent:skl */
+		WA_SET_BIT_MASKED(HDC_CHICKEN0,
+				  HDC_FORCE_NON_COHERENT);
+	}
+
 	return skl_tune_iz_hashing(ring);
 }
 
+static int bxt_init_workarounds(struct intel_engine_cs *ring)
+{
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	gen9_init_workarounds(ring);
+
+	/* WaDisableThreadStallDopClockGating:bxt */
+	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
+			  STALL_DOP_GATING_DISABLE);
+
+	/* WaDisableSbeCacheDispatchPortSharing:bxt */
+	if (INTEL_REVID(dev) <= BXT_REVID_B0) {
+		WA_SET_BIT_MASKED(
+			GEN7_HALF_SLICE_CHICKEN1,
+			GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE);
+	}
+
+	return 0;
+}
+
 int init_workarounds_ring(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
@@ -1044,8 +1082,9 @@ int init_workarounds_ring(struct intel_engine_cs *ring)
 
 	if (IS_SKYLAKE(dev))
 		return skl_init_workarounds(ring);
-	else if (IS_GEN9(dev))
-		return gen9_init_workarounds(ring);
+
+	if (IS_BROXTON(dev))
+		return bxt_init_workarounds(ring);
 
 	return 0;
 }
@@ -1972,6 +2011,7 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
 	INIT_LIST_HEAD(&ring->execlist_queue);
+	i915_gem_batch_pool_init(dev, &ring->batch_pool);
 	ringbuf->size = 32 * PAGE_SIZE;
 	ringbuf->ring = ring;
 	memset(ring->semaphore.sync_seqno, 0, sizeof(ring->semaphore.sync_seqno));
@@ -2050,91 +2090,40 @@ void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 	cleanup_status_page(ring);
 
 	i915_cmd_parser_fini_ring(ring);
+	i915_gem_batch_pool_fini(&ring->batch_pool);
 
 	kfree(ringbuf);
 	ring->buffer = NULL;
 }
 
-static int intel_ring_wait_request(struct intel_engine_cs *ring, int n)
+static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
 {
 	struct intel_ringbuffer *ringbuf = ring->buffer;
 	struct drm_i915_gem_request *request;
+	unsigned space;
 	int ret;
 
 	if (intel_ring_space(ringbuf) >= n)
 		return 0;
 
 	list_for_each_entry(request, &ring->request_list, list) {
-		if (__intel_ring_space(request->postfix, ringbuf->tail,
-				       ringbuf->size) >= n) {
+		space = __intel_ring_space(request->postfix, ringbuf->tail,
+					   ringbuf->size);
+		if (space >= n)
 			break;
-		}
 	}
 
-	if (&request->list == &ring->request_list)
+	if (WARN_ON(&request->list == &ring->request_list))
 		return -ENOSPC;
 
 	ret = i915_wait_request(request);
 	if (ret)
 		return ret;
 
-	i915_gem_retire_requests_ring(ring);
-
+	ringbuf->space = space;
 	return 0;
 }
 
-static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
-{
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ringbuffer *ringbuf = ring->buffer;
-	unsigned long end;
-	int ret;
-
-	ret = intel_ring_wait_request(ring, n);
-	if (ret != -ENOSPC)
-		return ret;
-
-	/* force the tail write in case we have been skipping them */
-	__intel_ring_advance(ring);
-
-	/* With GEM the hangcheck timer should kick us out of the loop,
-	 * leaving it early runs the risk of corrupting GEM state (due
-	 * to running on almost untested codepaths). But on resume
-	 * timers don't work yet, so prevent a complete hang in that
-	 * case by choosing an insanely large timeout. */
-	end = jiffies + 60 * HZ;
-
-	ret = 0;
-	trace_i915_ring_wait_begin(ring);
-	do {
-		if (intel_ring_space(ringbuf) >= n)
-			break;
-		ringbuf->head = I915_READ_HEAD(ring);
-		if (intel_ring_space(ringbuf) >= n)
-			break;
-
-		msleep(1);
-
-		if (dev_priv->mm.interruptible && signal_pending(current)) {
-			ret = -ERESTARTSYS;
-			break;
-		}
-
-		ret = i915_gem_check_wedge(&dev_priv->gpu_error,
-					   dev_priv->mm.interruptible);
-		if (ret)
-			break;
-
-		if (time_after(jiffies, end)) {
-			ret = -EBUSY;
-			break;
-		}
-	} while (1);
-	trace_i915_ring_wait_end(ring);
-	return ret;
-}
-
 static int intel_wrap_ring_buffer(struct intel_engine_cs *ring)
 {
 	uint32_t __iomem *virt;
@@ -2175,38 +2164,19 @@ int intel_ring_idle(struct intel_engine_cs *ring)
 		return 0;
 
 	req = list_entry(ring->request_list.prev,
-			   struct drm_i915_gem_request,
-			   list);
+			struct drm_i915_gem_request,
+			list);
 
-	return i915_wait_request(req);
+	/* Make sure we do not trigger any retires */
+	return __i915_wait_request(req,
+				   atomic_read(&to_i915(ring->dev)->gpu_error.reset_counter),
+				   to_i915(ring->dev)->mm.interruptible,
+				   NULL, NULL);
 }
 
-static int
-intel_ring_alloc_request(struct intel_engine_cs *ring)
+int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request)
 {
-	int ret;
-	struct drm_i915_gem_request *request;
-	struct drm_i915_private *dev_private = ring->dev->dev_private;
-
-	if (ring->outstanding_lazy_request)
-		return 0;
-
-	request = kzalloc(sizeof(*request), GFP_KERNEL);
-	if (request == NULL)
-		return -ENOMEM;
-
-	kref_init(&request->ref);
-	request->ring = ring;
-	request->ringbuf = ring->buffer;
-	request->uniq = dev_private->request_uniq++;
-
-	ret = i915_gem_get_seqno(ring->dev, &request->seqno);
-	if (ret) {
-		kfree(request);
-		return ret;
-	}
-
-	ring->outstanding_lazy_request = request;
+	request->ringbuf = request->ring->buffer;
 	return 0;
 }
 
@@ -2247,7 +2217,7 @@ int intel_ring_begin(struct intel_engine_cs *ring,
 		return ret;
 
 	/* Preallocate the olr before touching the ring */
-	ret = intel_ring_alloc_request(ring);
+	ret = i915_gem_request_alloc(ring, ring->default_context);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index c761fe05ad6f..e539314ae87e 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -2,6 +2,7 @@
 #define _INTEL_RINGBUFFER_H_
 
 #include <linux/hashtable.h>
+#include "i915_gem_batch_pool.h"
 
 #define I915_CMD_HASH_ORDER 9
 
@@ -117,6 +118,7 @@ struct intel_ringbuffer {
 };
 
 struct	intel_context;
+struct drm_i915_reg_descriptor;
 
 struct  intel_engine_cs {
 	const char	*name;
@@ -133,6 +135,13 @@ struct  intel_engine_cs {
 	struct		drm_device *dev;
 	struct intel_ringbuffer *buffer;
 
+	/*
+	 * A pool of objects to use as shadow copies of client batch buffers
+	 * when the command parser is enabled. Prevents the client from
+	 * modifying the batch contents after software parsing.
+	 */
+	struct i915_gem_batch_pool batch_pool;
+
 	struct intel_hw_status_page status_page;
 
 	unsigned irq_refcount; /* protected by dev_priv->irq_lock */
@@ -292,14 +301,14 @@ struct  intel_engine_cs {
 	/*
 	 * Table of registers allowed in commands that read/write registers.
 	 */
-	const u32 *reg_table;
+	const struct drm_i915_reg_descriptor *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;
+	const struct drm_i915_reg_descriptor *master_reg_table;
 	int master_reg_count;
 
 	/*
@@ -390,6 +399,8 @@ int intel_alloc_ringbuffer_obj(struct drm_device *dev,
 void intel_stop_ring_buffer(struct intel_engine_cs *ring);
 void intel_cleanup_ring_buffer(struct intel_engine_cs *ring);
 
+int intel_ring_alloc_request_extras(struct drm_i915_gem_request *request);
+
 int __must_check intel_ring_begin(struct intel_engine_cs *ring, int n);
 int __must_check intel_ring_cacheline_align(struct intel_engine_cs *ring);
 static inline void intel_ring_emit(struct intel_engine_cs *ring,
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index ce00e6994eeb..1a45385f4d66 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -49,6 +49,9 @@
  * present for a given platform.
  */
 
+#define GEN9_ENABLE_DC5(dev) 0
+#define SKL_ENABLE_DC6(dev) IS_SKYLAKE(dev)
+
 #define for_each_power_well(i, power_well, domain_mask, power_domains)	\
 	for (i = 0;							\
 	     i < (power_domains)->power_well_count &&			\
@@ -62,6 +65,9 @@
 	     i--)							 \
 		if ((power_well)->domains & (domain_mask))
 
+bool intel_display_power_well_is_enabled(struct drm_i915_private *dev_priv,
+				    int power_well_id);
+
 /*
  * We should only use the power well if we explicitly asked the hardware to
  * enable it, so check if it's enabled and also check if we've requested it to
@@ -308,7 +314,9 @@ static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 	BIT(POWER_DOMAIN_PORT_DDI_D_4_LANES) |		\
 	BIT(POWER_DOMAIN_INIT))
 #define SKL_DISPLAY_MISC_IO_POWER_DOMAINS (		\
-	SKL_DISPLAY_POWERWELL_1_POWER_DOMAINS)
+	SKL_DISPLAY_POWERWELL_1_POWER_DOMAINS |		\
+	BIT(POWER_DOMAIN_PLLS) |			\
+	BIT(POWER_DOMAIN_INIT))
 #define SKL_DISPLAY_ALWAYS_ON_POWER_DOMAINS (		\
 	(POWER_DOMAIN_MASK & ~(SKL_DISPLAY_POWERWELL_1_POWER_DOMAINS |	\
 	SKL_DISPLAY_POWERWELL_2_POWER_DOMAINS |		\
@@ -319,9 +327,246 @@ static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 	SKL_DISPLAY_MISC_IO_POWER_DOMAINS)) |		\
 	BIT(POWER_DOMAIN_INIT))
 
+#define BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS (		\
+	BIT(POWER_DOMAIN_TRANSCODER_A) |		\
+	BIT(POWER_DOMAIN_PIPE_B) |			\
+	BIT(POWER_DOMAIN_TRANSCODER_B) |		\
+	BIT(POWER_DOMAIN_PIPE_C) |			\
+	BIT(POWER_DOMAIN_TRANSCODER_C) |		\
+	BIT(POWER_DOMAIN_PIPE_B_PANEL_FITTER) |		\
+	BIT(POWER_DOMAIN_PIPE_C_PANEL_FITTER) |		\
+	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_AUX_B) |			\
+	BIT(POWER_DOMAIN_AUX_C) |			\
+	BIT(POWER_DOMAIN_AUDIO) |			\
+	BIT(POWER_DOMAIN_VGA) |				\
+	BIT(POWER_DOMAIN_INIT))
+#define BXT_DISPLAY_POWERWELL_1_POWER_DOMAINS (		\
+	BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS |		\
+	BIT(POWER_DOMAIN_PIPE_A) |			\
+	BIT(POWER_DOMAIN_TRANSCODER_EDP) |		\
+	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_A_2_LANES) |		\
+	BIT(POWER_DOMAIN_PORT_DDI_A_4_LANES) |		\
+	BIT(POWER_DOMAIN_AUX_A) |			\
+	BIT(POWER_DOMAIN_PLLS) |			\
+	BIT(POWER_DOMAIN_INIT))
+#define BXT_DISPLAY_ALWAYS_ON_POWER_DOMAINS (		\
+	(POWER_DOMAIN_MASK & ~(BXT_DISPLAY_POWERWELL_1_POWER_DOMAINS |	\
+	BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS)) |	\
+	BIT(POWER_DOMAIN_INIT))
+
+static void assert_can_enable_dc9(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	WARN(!IS_BROXTON(dev), "Platform doesn't support DC9.\n");
+	WARN((I915_READ(DC_STATE_EN) & DC_STATE_EN_DC9),
+		"DC9 already programmed to be enabled.\n");
+	WARN(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5,
+		"DC5 still not disabled to enable DC9.\n");
+	WARN(I915_READ(HSW_PWR_WELL_DRIVER), "Power well on.\n");
+	WARN(intel_irqs_enabled(dev_priv), "Interrupts not disabled yet.\n");
+
+	 /*
+	  * TODO: check for the following to verify the conditions to enter DC9
+	  * state are satisfied:
+	  * 1] Check relevant display engine registers to verify if mode set
+	  * disable sequence was followed.
+	  * 2] Check if display uninitialize sequence is initialized.
+	  */
+}
+
+static void assert_can_disable_dc9(struct drm_i915_private *dev_priv)
+{
+	WARN(intel_irqs_enabled(dev_priv), "Interrupts not disabled yet.\n");
+	WARN(!(I915_READ(DC_STATE_EN) & DC_STATE_EN_DC9),
+		"DC9 already programmed to be disabled.\n");
+	WARN(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5,
+		"DC5 still not disabled.\n");
+
+	 /*
+	  * TODO: check for the following to verify DC9 state was indeed
+	  * entered before programming to disable it:
+	  * 1] Check relevant display engine registers to verify if mode
+	  *  set disable sequence was followed.
+	  * 2] Check if display uninitialize sequence is initialized.
+	  */
+}
+
+void bxt_enable_dc9(struct drm_i915_private *dev_priv)
+{
+	uint32_t val;
+
+	assert_can_enable_dc9(dev_priv);
+
+	DRM_DEBUG_KMS("Enabling DC9\n");
+
+	val = I915_READ(DC_STATE_EN);
+	val |= DC_STATE_EN_DC9;
+	I915_WRITE(DC_STATE_EN, val);
+	POSTING_READ(DC_STATE_EN);
+}
+
+void bxt_disable_dc9(struct drm_i915_private *dev_priv)
+{
+	uint32_t val;
+
+	assert_can_disable_dc9(dev_priv);
+
+	DRM_DEBUG_KMS("Disabling DC9\n");
+
+	val = I915_READ(DC_STATE_EN);
+	val &= ~DC_STATE_EN_DC9;
+	I915_WRITE(DC_STATE_EN, val);
+	POSTING_READ(DC_STATE_EN);
+}
+
+static void gen9_set_dc_state_debugmask_memory_up(
+			struct drm_i915_private *dev_priv)
+{
+	uint32_t val;
+
+	/* The below bit doesn't need to be cleared ever afterwards */
+	val = I915_READ(DC_STATE_DEBUG);
+	if (!(val & DC_STATE_DEBUG_MASK_MEMORY_UP)) {
+		val |= DC_STATE_DEBUG_MASK_MEMORY_UP;
+		I915_WRITE(DC_STATE_DEBUG, val);
+		POSTING_READ(DC_STATE_DEBUG);
+	}
+}
+
+static void assert_can_enable_dc5(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	bool pg2_enabled = intel_display_power_well_is_enabled(dev_priv,
+					SKL_DISP_PW_2);
+
+	WARN(!IS_SKYLAKE(dev), "Platform doesn't support DC5.\n");
+	WARN(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
+	WARN(pg2_enabled, "PG2 not disabled to enable DC5.\n");
+
+	WARN((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5),
+				"DC5 already programmed to be enabled.\n");
+	WARN(dev_priv->pm.suspended,
+		"DC5 cannot be enabled, if platform is runtime-suspended.\n");
+
+	assert_csr_loaded(dev_priv);
+}
+
+static void assert_can_disable_dc5(struct drm_i915_private *dev_priv)
+{
+	bool pg2_enabled = intel_display_power_well_is_enabled(dev_priv,
+					SKL_DISP_PW_2);
+	/*
+	 * During initialization, the firmware may not be loaded yet.
+	 * We still want to make sure that the DC enabling flag is cleared.
+	 */
+	if (dev_priv->power_domains.initializing)
+		return;
+
+	WARN(!pg2_enabled, "PG2 not enabled to disable DC5.\n");
+	WARN(dev_priv->pm.suspended,
+		"Disabling of DC5 while platform is runtime-suspended should never happen.\n");
+}
+
+static void gen9_enable_dc5(struct drm_i915_private *dev_priv)
+{
+	uint32_t val;
+
+	assert_can_enable_dc5(dev_priv);
+
+	DRM_DEBUG_KMS("Enabling DC5\n");
+
+	gen9_set_dc_state_debugmask_memory_up(dev_priv);
+
+	val = I915_READ(DC_STATE_EN);
+	val &= ~DC_STATE_EN_UPTO_DC5_DC6_MASK;
+	val |= DC_STATE_EN_UPTO_DC5;
+	I915_WRITE(DC_STATE_EN, val);
+	POSTING_READ(DC_STATE_EN);
+}
+
+static void gen9_disable_dc5(struct drm_i915_private *dev_priv)
+{
+	uint32_t val;
+
+	assert_can_disable_dc5(dev_priv);
+
+	DRM_DEBUG_KMS("Disabling DC5\n");
+
+	val = I915_READ(DC_STATE_EN);
+	val &= ~DC_STATE_EN_UPTO_DC5;
+	I915_WRITE(DC_STATE_EN, val);
+	POSTING_READ(DC_STATE_EN);
+}
+
+static void assert_can_enable_dc6(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	WARN(!IS_SKYLAKE(dev), "Platform doesn't support DC6.\n");
+	WARN(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
+	WARN(I915_READ(UTIL_PIN_CTL) & UTIL_PIN_ENABLE,
+		"Backlight is not disabled.\n");
+	WARN((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
+		"DC6 already programmed to be enabled.\n");
+
+	assert_csr_loaded(dev_priv);
+}
+
+static void assert_can_disable_dc6(struct drm_i915_private *dev_priv)
+{
+	/*
+	 * During initialization, the firmware may not be loaded yet.
+	 * We still want to make sure that the DC enabling flag is cleared.
+	 */
+	if (dev_priv->power_domains.initializing)
+		return;
+
+	assert_csr_loaded(dev_priv);
+	WARN(!(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
+		"DC6 already programmed to be disabled.\n");
+}
+
+static void skl_enable_dc6(struct drm_i915_private *dev_priv)
+{
+	uint32_t val;
+
+	assert_can_enable_dc6(dev_priv);
+
+	DRM_DEBUG_KMS("Enabling DC6\n");
+
+	gen9_set_dc_state_debugmask_memory_up(dev_priv);
+
+	val = I915_READ(DC_STATE_EN);
+	val &= ~DC_STATE_EN_UPTO_DC5_DC6_MASK;
+	val |= DC_STATE_EN_UPTO_DC6;
+	I915_WRITE(DC_STATE_EN, val);
+	POSTING_READ(DC_STATE_EN);
+}
+
+static void skl_disable_dc6(struct drm_i915_private *dev_priv)
+{
+	uint32_t val;
+
+	assert_can_disable_dc6(dev_priv);
+
+	DRM_DEBUG_KMS("Disabling DC6\n");
+
+	val = I915_READ(DC_STATE_EN);
+	val &= ~DC_STATE_EN_UPTO_DC6;
+	I915_WRITE(DC_STATE_EN, val);
+	POSTING_READ(DC_STATE_EN);
+}
+
 static void skl_set_power_well(struct drm_i915_private *dev_priv,
 			struct i915_power_well *power_well, bool enable)
 {
+	struct drm_device *dev = dev_priv->dev;
 	uint32_t tmp, fuse_status;
 	uint32_t req_mask, state_mask;
 	bool is_enabled, enable_requested, check_fuse_status = false;
@@ -361,6 +606,25 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv,
 
 	if (enable) {
 		if (!enable_requested) {
+			WARN((tmp & state_mask) &&
+				!I915_READ(HSW_PWR_WELL_BIOS),
+				"Invalid for power well status to be enabled, unless done by the BIOS, \
+				when request is to disable!\n");
+			if ((GEN9_ENABLE_DC5(dev) || SKL_ENABLE_DC6(dev)) &&
+				power_well->data == SKL_DISP_PW_2) {
+				if (SKL_ENABLE_DC6(dev)) {
+					skl_disable_dc6(dev_priv);
+					/*
+					 * DDI buffer programming unnecessary during driver-load/resume
+					 * as it's already done during modeset initialization then.
+					 * It's also invalid here as encoder list is still uninitialized.
+					 */
+					if (!dev_priv->power_domains.initializing)
+						intel_prepare_ddi(dev);
+				} else {
+					gen9_disable_dc5(dev_priv);
+				}
+			}
 			I915_WRITE(HSW_PWR_WELL_DRIVER, tmp | req_mask);
 		}
 
@@ -377,6 +641,25 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv,
 			I915_WRITE(HSW_PWR_WELL_DRIVER,	tmp & ~req_mask);
 			POSTING_READ(HSW_PWR_WELL_DRIVER);
 			DRM_DEBUG_KMS("Disabling %s\n", power_well->name);
+
+			if ((GEN9_ENABLE_DC5(dev) || SKL_ENABLE_DC6(dev)) &&
+				power_well->data == SKL_DISP_PW_2) {
+				enum csr_state state;
+				/* TODO: wait for a completion event or
+				 * similar here instead of busy
+				 * waiting using wait_for function.
+				 */
+				wait_for((state = intel_csr_load_status_get(dev_priv)) !=
+						FW_UNINITIALIZED, 1000);
+				if (state != FW_LOADED)
+					DRM_ERROR("CSR firmware not ready (%d)\n",
+							state);
+				else
+					if (SKL_ENABLE_DC6(dev))
+						skl_enable_dc6(dev_priv);
+					else
+						gen9_enable_dc5(dev_priv);
+			}
 		}
 	}
 
@@ -488,7 +771,7 @@ static void vlv_set_power_well(struct drm_i915_private *dev_priv,
 	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",
+		DRM_ERROR("timeout setting power well state %08x (%08x)\n",
 			  state,
 			  vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_CTRL));
 
@@ -666,8 +949,8 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 	if (wait_for(I915_READ(DISPLAY_PHY_STATUS) & PHY_POWERGOOD(phy), 1))
 		DRM_ERROR("Display PHY %d is not power up\n", phy);
 
-	I915_WRITE(DISPLAY_PHY_CONTROL, I915_READ(DISPLAY_PHY_CONTROL) |
-		   PHY_COM_LANE_RESET_DEASSERT(phy));
+	dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(phy);
+	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
 }
 
 static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
@@ -687,8 +970,8 @@ static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
 		assert_pll_disabled(dev_priv, PIPE_C);
 	}
 
-	I915_WRITE(DISPLAY_PHY_CONTROL, I915_READ(DISPLAY_PHY_CONTROL) &
-		   ~PHY_COM_LANE_RESET_DEASSERT(phy));
+	dev_priv->chv_phy_control &= ~PHY_COM_LANE_RESET_DEASSERT(phy);
+	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
 
 	vlv_set_power_well(dev_priv, power_well, false);
 }
@@ -746,7 +1029,7 @@ static void chv_set_pipe_power_well(struct drm_i915_private *dev_priv,
 	vlv_punit_write(dev_priv, PUNIT_REG_DSPFREQ, ctrl);
 
 	if (wait_for(COND, 100))
-		DRM_ERROR("timout setting power well state %08x (%08x)\n",
+		DRM_ERROR("timeout setting power well state %08x (%08x)\n",
 			  state,
 			  vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ));
 
@@ -950,18 +1233,6 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
 	BIT(POWER_DOMAIN_AUX_C) |		\
 	BIT(POWER_DOMAIN_INIT))
 
-#define CHV_PIPE_A_POWER_DOMAINS (	\
-	BIT(POWER_DOMAIN_PIPE_A) |	\
-	BIT(POWER_DOMAIN_INIT))
-
-#define CHV_PIPE_B_POWER_DOMAINS (	\
-	BIT(POWER_DOMAIN_PIPE_B) |	\
-	BIT(POWER_DOMAIN_INIT))
-
-#define CHV_PIPE_C_POWER_DOMAINS (	\
-	BIT(POWER_DOMAIN_PIPE_C) |	\
-	BIT(POWER_DOMAIN_INIT))
-
 #define CHV_DPIO_CMN_BC_POWER_DOMAINS (		\
 	BIT(POWER_DOMAIN_PORT_DDI_B_2_LANES) |	\
 	BIT(POWER_DOMAIN_PORT_DDI_B_4_LANES) |	\
@@ -977,17 +1248,6 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
 	BIT(POWER_DOMAIN_AUX_D) |		\
 	BIT(POWER_DOMAIN_INIT))
 
-#define CHV_DPIO_TX_D_LANES_01_POWER_DOMAINS (	\
-	BIT(POWER_DOMAIN_PORT_DDI_D_2_LANES) |	\
-	BIT(POWER_DOMAIN_PORT_DDI_D_4_LANES) |	\
-	BIT(POWER_DOMAIN_AUX_D) |		\
-	BIT(POWER_DOMAIN_INIT))
-
-#define CHV_DPIO_TX_D_LANES_23_POWER_DOMAINS (	\
-	BIT(POWER_DOMAIN_PORT_DDI_D_4_LANES) |	\
-	BIT(POWER_DOMAIN_AUX_D) |		\
-	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,
@@ -1145,110 +1405,33 @@ static struct i915_power_well chv_power_wells[] = {
 		.domains = VLV_ALWAYS_ON_POWER_DOMAINS,
 		.ops = &i9xx_always_on_power_well_ops,
 	},
-#if 0
 	{
 		.name = "display",
-		.domains = VLV_DISPLAY_POWER_DOMAINS,
-		.data = PUNIT_POWER_WELL_DISP2D,
-		.ops = &vlv_display_power_well_ops,
-	},
-#endif
-	{
-		.name = "pipe-a",
 		/*
-		 * FIXME: pipe A power well seems to be the new disp2d well.
-		 * At least all registers seem to be housed there. Figure
-		 * out if this a a temporary situation in pre-production
-		 * hardware or a permanent state of affairs.
+		 * Pipe A power well is the new disp2d well. Pipe B and C
+		 * power wells don't actually exist. Pipe A power well is
+		 * required for any pipe to work.
 		 */
-		.domains = CHV_PIPE_A_POWER_DOMAINS | VLV_DISPLAY_POWER_DOMAINS,
+		.domains = VLV_DISPLAY_POWER_DOMAINS,
 		.data = PIPE_A,
 		.ops = &chv_pipe_power_well_ops,
 	},
-#if 0
-	{
-		.name = "pipe-b",
-		.domains = CHV_PIPE_B_POWER_DOMAINS,
-		.data = PIPE_B,
-		.ops = &chv_pipe_power_well_ops,
-	},
-	{
-		.name = "pipe-c",
-		.domains = CHV_PIPE_C_POWER_DOMAINS,
-		.data = PIPE_C,
-		.ops = &chv_pipe_power_well_ops,
-	},
-#endif
 	{
 		.name = "dpio-common-bc",
-		/*
-		 * XXX: cmnreset for one PHY seems to disturb the other.
-		 * As a workaround keep both powered on at the same
-		 * time for now.
-		 */
-		.domains = CHV_DPIO_CMN_BC_POWER_DOMAINS | CHV_DPIO_CMN_D_POWER_DOMAINS,
+		.domains = CHV_DPIO_CMN_BC_POWER_DOMAINS,
 		.data = PUNIT_POWER_WELL_DPIO_CMN_BC,
 		.ops = &chv_dpio_cmn_power_well_ops,
 	},
 	{
 		.name = "dpio-common-d",
-		/*
-		 * XXX: cmnreset for one PHY seems to disturb the other.
-		 * As a workaround keep both powered on at the same
-		 * time for now.
-		 */
-		.domains = CHV_DPIO_CMN_BC_POWER_DOMAINS | CHV_DPIO_CMN_D_POWER_DOMAINS,
+		.domains = CHV_DPIO_CMN_D_POWER_DOMAINS,
 		.data = PUNIT_POWER_WELL_DPIO_CMN_D,
 		.ops = &chv_dpio_cmn_power_well_ops,
 	},
-#if 0
-	{
-		.name = "dpio-tx-b-01",
-		.domains = VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS |
-			   VLV_DPIO_TX_B_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,
-		.ops = &vlv_dpio_power_well_ops,
-		.data = PUNIT_POWER_WELL_DPIO_TX_B_LANES_23,
-	},
-	{
-		.name = "dpio-tx-c-01",
-		.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_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,
-	},
-	{
-		.name = "dpio-tx-d-01",
-		.domains = CHV_DPIO_TX_D_LANES_01_POWER_DOMAINS |
-			   CHV_DPIO_TX_D_LANES_23_POWER_DOMAINS,
-		.ops = &vlv_dpio_power_well_ops,
-		.data = PUNIT_POWER_WELL_DPIO_TX_D_LANES_01,
-	},
-	{
-		.name = "dpio-tx-d-23",
-		.domains = CHV_DPIO_TX_D_LANES_01_POWER_DOMAINS |
-			   CHV_DPIO_TX_D_LANES_23_POWER_DOMAINS,
-		.ops = &vlv_dpio_power_well_ops,
-		.data = PUNIT_POWER_WELL_DPIO_TX_D_LANES_23,
-	},
-#endif
 };
 
 static struct i915_power_well *lookup_power_well(struct drm_i915_private *dev_priv,
-						 enum punit_power_well power_well_id)
+						 int power_well_id)
 {
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
 	struct i915_power_well *power_well;
@@ -1262,6 +1445,18 @@ static struct i915_power_well *lookup_power_well(struct drm_i915_private *dev_pr
 	return NULL;
 }
 
+bool intel_display_power_well_is_enabled(struct drm_i915_private *dev_priv,
+				    int power_well_id)
+{
+	struct i915_power_well *power_well;
+	bool ret;
+
+	power_well = lookup_power_well(dev_priv, power_well_id);
+	ret = power_well->ops->is_enabled(dev_priv, power_well);
+
+	return ret;
+}
+
 static struct i915_power_well skl_power_wells[] = {
 	{
 		.name = "always-on",
@@ -1313,6 +1508,27 @@ static struct i915_power_well skl_power_wells[] = {
 	},
 };
 
+static struct i915_power_well bxt_power_wells[] = {
+	{
+		.name = "always-on",
+		.always_on = 1,
+		.domains = BXT_DISPLAY_ALWAYS_ON_POWER_DOMAINS,
+		.ops = &i9xx_always_on_power_well_ops,
+	},
+	{
+		.name = "power well 1",
+		.domains = BXT_DISPLAY_POWERWELL_1_POWER_DOMAINS,
+		.ops = &skl_power_well_ops,
+		.data = SKL_DISP_PW_1,
+	},
+	{
+		.name = "power well 2",
+		.domains = BXT_DISPLAY_POWERWELL_2_POWER_DOMAINS,
+		.ops = &skl_power_well_ops,
+		.data = SKL_DISP_PW_2,
+	}
+};
+
 #define set_power_wells(power_domains, __power_wells) ({		\
 	(power_domains)->power_wells = (__power_wells);			\
 	(power_domains)->power_well_count = ARRAY_SIZE(__power_wells);	\
@@ -1341,6 +1557,8 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv)
 		set_power_wells(power_domains, bdw_power_wells);
 	} else if (IS_SKYLAKE(dev_priv->dev)) {
 		set_power_wells(power_domains, skl_power_wells);
+	} else if (IS_BROXTON(dev_priv->dev)) {
+		set_power_wells(power_domains, bxt_power_wells);
 	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
 		set_power_wells(power_domains, chv_power_wells);
 	} else if (IS_VALLEYVIEW(dev_priv->dev)) {
@@ -1401,6 +1619,32 @@ static void intel_power_domains_resume(struct drm_i915_private *dev_priv)
 	mutex_unlock(&power_domains->lock);
 }
 
+static void chv_phy_control_init(struct drm_i915_private *dev_priv)
+{
+	struct i915_power_well *cmn_bc =
+		lookup_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC);
+	struct i915_power_well *cmn_d =
+		lookup_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_D);
+
+	/*
+	 * DISPLAY_PHY_CONTROL can get corrupted if read. As a
+	 * workaround never ever read DISPLAY_PHY_CONTROL, and
+	 * instead maintain a shadow copy ourselves. Use the actual
+	 * power well state to reconstruct the expected initial
+	 * value.
+	 */
+	dev_priv->chv_phy_control =
+		PHY_LDO_SEQ_DELAY(PHY_LDO_DELAY_600NS, DPIO_PHY0) |
+		PHY_LDO_SEQ_DELAY(PHY_LDO_DELAY_600NS, DPIO_PHY1) |
+		PHY_CH_POWER_MODE(PHY_CH_SU_PSR, DPIO_PHY0, DPIO_CH0) |
+		PHY_CH_POWER_MODE(PHY_CH_SU_PSR, DPIO_PHY0, DPIO_CH1) |
+		PHY_CH_POWER_MODE(PHY_CH_SU_PSR, DPIO_PHY1, DPIO_CH0);
+	if (cmn_bc->ops->is_enabled(dev_priv, cmn_bc))
+		dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(DPIO_PHY0);
+	if (cmn_d->ops->is_enabled(dev_priv, cmn_d))
+		dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(DPIO_PHY1);
+}
+
 static void vlv_cmnlane_wa(struct drm_i915_private *dev_priv)
 {
 	struct i915_power_well *cmn =
@@ -1443,7 +1687,9 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv)
 
 	power_domains->initializing = true;
 
-	if (IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev)) {
+	if (IS_CHERRYVIEW(dev)) {
+		chv_phy_control_init(dev_priv);
+	} else if (IS_VALLEYVIEW(dev)) {
 		mutex_lock(&power_domains->lock);
 		vlv_cmnlane_wa(dev_priv);
 		mutex_unlock(&power_domains->lock);
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 987b81f31b0e..aa2fd751609c 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -242,7 +242,15 @@ static void intel_sdvo_write_sdvox(struct intel_sdvo *intel_sdvo, u32 val)
 
 	if (intel_sdvo->sdvo_reg == PCH_SDVOB) {
 		I915_WRITE(intel_sdvo->sdvo_reg, val);
-		I915_READ(intel_sdvo->sdvo_reg);
+		POSTING_READ(intel_sdvo->sdvo_reg);
+		/*
+		 * HW workaround, need to write this twice for issue
+		 * that may result in first write getting masked.
+		 */
+		if (HAS_PCH_IBX(dev)) {
+			I915_WRITE(intel_sdvo->sdvo_reg, val);
+			POSTING_READ(intel_sdvo->sdvo_reg);
+		}
 		return;
 	}
 
@@ -259,9 +267,9 @@ static void intel_sdvo_write_sdvox(struct intel_sdvo *intel_sdvo, u32 val)
 	for (i = 0; i < 2; i++)
 	{
 		I915_WRITE(GEN3_SDVOB, bval);
-		I915_READ(GEN3_SDVOB);
+		POSTING_READ(GEN3_SDVOB);
 		I915_WRITE(GEN3_SDVOC, cval);
-		I915_READ(GEN3_SDVOC);
+		POSTING_READ(GEN3_SDVOC);
 	}
 }
 
@@ -1429,6 +1437,7 @@ static void intel_disable_sdvo(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_sdvo *intel_sdvo = to_sdvo(encoder);
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	u32 temp;
 
 	intel_sdvo_set_active_outputs(intel_sdvo, 0);
@@ -1437,35 +1446,34 @@ static void intel_disable_sdvo(struct intel_encoder *encoder)
 						   DRM_MODE_DPMS_OFF);
 
 	temp = I915_READ(intel_sdvo->sdvo_reg);
-	if ((temp & SDVO_ENABLE) != 0) {
-		/* HW workaround for IBX, we need to move the port to
-		 * transcoder A before disabling it. */
-		if (HAS_PCH_IBX(encoder->base.dev)) {
-			struct drm_crtc *crtc = encoder->base.crtc;
-			int pipe = crtc ? to_intel_crtc(crtc)->pipe : -1;
-
-			if (temp & SDVO_PIPE_B_SELECT) {
-				temp &= ~SDVO_PIPE_B_SELECT;
-				I915_WRITE(intel_sdvo->sdvo_reg, temp);
-				POSTING_READ(intel_sdvo->sdvo_reg);
-
-				/* Again we need to write this twice. */
-				I915_WRITE(intel_sdvo->sdvo_reg, temp);
-				POSTING_READ(intel_sdvo->sdvo_reg);
-
-				/* Transcoder selection bits only update
-				 * effectively on vblank. */
-				if (crtc)
-					intel_wait_for_vblank(encoder->base.dev, pipe);
-				else
-					msleep(50);
-			}
-		}
 
-		intel_sdvo_write_sdvox(intel_sdvo, temp & ~SDVO_ENABLE);
+	temp &= ~SDVO_ENABLE;
+	intel_sdvo_write_sdvox(intel_sdvo, temp);
+
+	/*
+	 * HW workaround for IBX, we need to move the port
+	 * to transcoder A after disabling it to allow the
+	 * matching DP port to be enabled on transcoder A.
+	 */
+	if (HAS_PCH_IBX(dev_priv) && crtc->pipe == PIPE_B) {
+		temp &= ~SDVO_PIPE_B_SELECT;
+		temp |= SDVO_ENABLE;
+		intel_sdvo_write_sdvox(intel_sdvo, temp);
+
+		temp &= ~SDVO_ENABLE;
+		intel_sdvo_write_sdvox(intel_sdvo, temp);
 	}
 }
 
+static void pch_disable_sdvo(struct intel_encoder *encoder)
+{
+}
+
+static void pch_post_disable_sdvo(struct intel_encoder *encoder)
+{
+	intel_disable_sdvo(encoder);
+}
+
 static void intel_enable_sdvo(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
@@ -1478,14 +1486,9 @@ static void intel_enable_sdvo(struct intel_encoder *encoder)
 	bool success;
 
 	temp = I915_READ(intel_sdvo->sdvo_reg);
-	if ((temp & SDVO_ENABLE) == 0) {
-		/* HW workaround for IBX, we need to move the port
-		 * to transcoder A before disabling it, so restore it here. */
-		if (HAS_PCH_IBX(dev))
-			temp |= SDVO_PIPE_SEL(intel_crtc->pipe);
+	temp |= SDVO_ENABLE;
+	intel_sdvo_write_sdvox(intel_sdvo, temp);
 
-		intel_sdvo_write_sdvox(intel_sdvo, temp | SDVO_ENABLE);
-	}
 	for (i = 0; i < 2; i++)
 		intel_wait_for_vblank(dev, intel_crtc->pipe);
 
@@ -2291,10 +2294,11 @@ intel_sdvo_select_i2c_bus(struct drm_i915_private *dev_priv,
 	else
 		mapping = &dev_priv->sdvo_mappings[1];
 
-	if (mapping->initialized && intel_gmbus_is_port_valid(mapping->i2c_pin))
+	if (mapping->initialized &&
+	    intel_gmbus_is_valid_pin(dev_priv, mapping->i2c_pin))
 		pin = mapping->i2c_pin;
 	else
-		pin = GMBUS_PORT_DPB;
+		pin = GMBUS_PIN_DPB;
 
 	sdvo->i2c = intel_gmbus_get_adapter(dev_priv, pin);
 
@@ -2987,7 +2991,12 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
 	}
 
 	intel_encoder->compute_config = intel_sdvo_compute_config;
-	intel_encoder->disable = intel_disable_sdvo;
+	if (HAS_PCH_SPLIT(dev)) {
+		intel_encoder->disable = pch_disable_sdvo;
+		intel_encoder->post_disable = pch_post_disable_sdvo;
+	} else {
+		intel_encoder->disable = intel_disable_sdvo;
+	}
 	intel_encoder->pre_enable = intel_sdvo_pre_enable;
 	intel_encoder->enable = intel_enable_sdvo;
 	intel_encoder->get_hw_state = intel_sdvo_get_hw_state;
diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c
index 693ce8281970..8831fc579ade 100644
--- a/drivers/gpu/drm/i915/intel_sideband.c
+++ b/drivers/gpu/drm/i915/intel_sideband.c
@@ -49,7 +49,7 @@ static int vlv_sideband_rw(struct drm_i915_private *dev_priv, u32 devfn,
 		(port << IOSF_PORT_SHIFT) | (be << IOSF_BYTE_ENABLES_SHIFT) |
 		(bar << IOSF_BAR_SHIFT);
 
-	WARN_ON(!mutex_is_locked(&dev_priv->dpio_lock));
+	WARN_ON(!mutex_is_locked(&dev_priv->sb_lock));
 
 	if (wait_for((I915_READ(VLV_IOSF_DOORBELL_REQ) & IOSF_SB_BUSY) == 0, 5)) {
 		DRM_DEBUG_DRIVER("IOSF sideband idle wait (%s) timed out\n",
@@ -81,10 +81,10 @@ u32 vlv_punit_read(struct drm_i915_private *dev_priv, u32 addr)
 
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), IOSF_PORT_PUNIT,
 			SB_CRRDDA_NP, addr, &val);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	return val;
 }
@@ -93,10 +93,10 @@ void vlv_punit_write(struct drm_i915_private *dev_priv, u32 addr, u32 val)
 {
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), IOSF_PORT_PUNIT,
 			SB_CRWRDA_NP, addr, &val);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 }
 
 u32 vlv_bunit_read(struct drm_i915_private *dev_priv, u32 reg)
@@ -121,10 +121,10 @@ u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr)
 
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
-	mutex_lock(&dev_priv->dpio_lock);
+	mutex_lock(&dev_priv->sb_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), IOSF_PORT_NC,
 			SB_CRRDDA_NP, addr, &val);
-	mutex_unlock(&dev_priv->dpio_lock);
+	mutex_unlock(&dev_priv->sb_lock);
 
 	return val;
 }
@@ -213,7 +213,7 @@ u32 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg,
 		   enum intel_sbi_destination destination)
 {
 	u32 value = 0;
-	WARN_ON(!mutex_is_locked(&dev_priv->dpio_lock));
+	WARN_ON(!mutex_is_locked(&dev_priv->sb_lock));
 
 	if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0,
 				100)) {
@@ -243,7 +243,7 @@ void intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 {
 	u32 tmp;
 
-	WARN_ON(!mutex_is_locked(&dev_priv->dpio_lock));
+	WARN_ON(!mutex_is_locked(&dev_priv->sb_lock));
 
 	if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0,
 				100)) {
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index a4c0a04b5044..8193a35388d7 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -33,6 +33,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_rect.h>
+#include <drm/drm_atomic.h>
 #include <drm/drm_plane_helper.h>
 #include "intel_drv.h"
 #include <drm/i915_drm.h>
@@ -165,17 +166,6 @@ void intel_pipe_update_end(struct intel_crtc *crtc, u32 start_vbl_count)
 			  pipe_name(pipe), start_vbl_count, end_vbl_count);
 }
 
-static void intel_update_primary_plane(struct intel_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	int reg = DSPCNTR(crtc->plane);
-
-	if (crtc->primary_enabled)
-		I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE);
-	else
-		I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
-}
-
 static void
 skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 		 struct drm_framebuffer *fb,
@@ -190,72 +180,24 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	const int pipe = intel_plane->pipe;
 	const int plane = intel_plane->plane + 1;
-	u32 plane_ctl, stride_div;
+	u32 plane_ctl, stride_div, stride;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key = &intel_plane->ckey;
 	unsigned long surf_addr;
+	u32 tile_height, plane_offset, plane_size;
+	unsigned int rotation;
+	int x_offset, y_offset;
+	struct intel_crtc_state *crtc_state = to_intel_crtc(crtc)->config;
+	int scaler_id;
 
 	plane_ctl = PLANE_CTL_ENABLE |
 		PLANE_CTL_PIPE_CSC_ENABLE;
 
-	switch (fb->pixel_format) {
-	case DRM_FORMAT_RGB565:
-		plane_ctl |= PLANE_CTL_FORMAT_RGB_565;
-		break;
-	case DRM_FORMAT_XBGR8888:
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888 | PLANE_CTL_ORDER_RGBX;
-		break;
-	case DRM_FORMAT_XRGB8888:
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888;
-		break;
-	/*
-	 * XXX: For ARBG/ABGR formats we default to expecting scanout buffers
-	 * to be already pre-multiplied. We need to add a knob (or a different
-	 * DRM_FORMAT) for user-space to configure that.
-	 */
-	case DRM_FORMAT_ABGR8888:
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888 |
-			     PLANE_CTL_ORDER_RGBX |
-			     PLANE_CTL_ALPHA_SW_PREMULTIPLY;
-		break;
-	case DRM_FORMAT_ARGB8888:
-		plane_ctl |= PLANE_CTL_FORMAT_XRGB_8888 |
-			     PLANE_CTL_ALPHA_SW_PREMULTIPLY;
-		break;
-	case DRM_FORMAT_YUYV:
-		plane_ctl |= PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_YUYV;
-		break;
-	case DRM_FORMAT_YVYU:
-		plane_ctl |= PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_YVYU;
-		break;
-	case DRM_FORMAT_UYVY:
-		plane_ctl |= PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_UYVY;
-		break;
-	case DRM_FORMAT_VYUY:
-		plane_ctl |= PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_VYUY;
-		break;
-	default:
-		BUG();
-	}
-
-	switch (fb->modifier[0]) {
-	case DRM_FORMAT_MOD_NONE:
-		break;
-	case I915_FORMAT_MOD_X_TILED:
-		plane_ctl |= PLANE_CTL_TILED_X;
-		break;
-	case I915_FORMAT_MOD_Y_TILED:
-		plane_ctl |= PLANE_CTL_TILED_Y;
-		break;
-	case I915_FORMAT_MOD_Yf_TILED:
-		plane_ctl |= PLANE_CTL_TILED_YF;
-		break;
-	default:
-		MISSING_CASE(fb->modifier[0]);
-	}
+	plane_ctl |= skl_plane_ctl_format(fb->pixel_format);
+	plane_ctl |= skl_plane_ctl_tiling(fb->modifier[0]);
 
-	if (drm_plane->state->rotation == BIT(DRM_ROTATE_180))
-		plane_ctl |= PLANE_CTL_ROTATE_180;
+	rotation = drm_plane->state->rotation;
+	plane_ctl |= skl_plane_ctl_rotation(rotation);
 
 	intel_update_sprite_watermarks(drm_plane, crtc, src_w, src_h,
 				       pixel_size, true,
@@ -264,6 +206,8 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	stride_div = intel_fb_stride_alignment(dev, fb->modifier[0],
 					       fb->pixel_format);
 
+	scaler_id = to_intel_plane_state(drm_plane->state)->scaler_id;
+
 	/* Sizes are 0 based */
 	src_w--;
 	src_h--;
@@ -283,31 +227,65 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 
 	surf_addr = intel_plane_obj_offset(intel_plane, obj);
 
-	I915_WRITE(PLANE_OFFSET(pipe, plane), (y << 16) | x);
-	I915_WRITE(PLANE_STRIDE(pipe, plane), fb->pitches[0] / stride_div);
-	I915_WRITE(PLANE_POS(pipe, plane), (crtc_y << 16) | crtc_x);
-	I915_WRITE(PLANE_SIZE(pipe, plane), (crtc_h << 16) | crtc_w);
+	if (intel_rotation_90_or_270(rotation)) {
+		/* stride: Surface height in tiles */
+		tile_height = intel_tile_height(dev, fb->pixel_format,
+						fb->modifier[0]);
+		stride = DIV_ROUND_UP(fb->height, tile_height);
+		plane_size = (src_w << 16) | src_h;
+		x_offset = stride * tile_height - y - (src_h + 1);
+		y_offset = x;
+	} else {
+		stride = fb->pitches[0] / stride_div;
+		plane_size = (src_h << 16) | src_w;
+		x_offset = x;
+		y_offset = y;
+	}
+	plane_offset = y_offset << 16 | x_offset;
+
+	I915_WRITE(PLANE_OFFSET(pipe, plane), plane_offset);
+	I915_WRITE(PLANE_STRIDE(pipe, plane), stride);
+	I915_WRITE(PLANE_SIZE(pipe, plane), plane_size);
+
+	/* program plane scaler */
+	if (scaler_id >= 0) {
+		uint32_t ps_ctrl = 0;
+
+		DRM_DEBUG_KMS("plane = %d PS_PLANE_SEL(plane) = 0x%x\n", plane,
+			PS_PLANE_SEL(plane));
+		ps_ctrl = PS_SCALER_EN | PS_PLANE_SEL(plane) |
+			crtc_state->scaler_state.scalers[scaler_id].mode;
+		I915_WRITE(SKL_PS_CTRL(pipe, scaler_id), ps_ctrl);
+		I915_WRITE(SKL_PS_PWR_GATE(pipe, scaler_id), 0);
+		I915_WRITE(SKL_PS_WIN_POS(pipe, scaler_id), (crtc_x << 16) | crtc_y);
+		I915_WRITE(SKL_PS_WIN_SZ(pipe, scaler_id),
+			((crtc_w + 1) << 16)|(crtc_h + 1));
+
+		I915_WRITE(PLANE_POS(pipe, plane), 0);
+	} else {
+		I915_WRITE(PLANE_POS(pipe, plane), (crtc_y << 16) | crtc_x);
+	}
+
 	I915_WRITE(PLANE_CTL(pipe, plane), plane_ctl);
 	I915_WRITE(PLANE_SURF(pipe, plane), surf_addr);
 	POSTING_READ(PLANE_SURF(pipe, plane));
 }
 
 static void
-skl_disable_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc)
+skl_disable_plane(struct drm_plane *dplane, struct drm_crtc *crtc, bool force)
 {
-	struct drm_device *dev = drm_plane->dev;
+	struct drm_device *dev = dplane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_plane *intel_plane = to_intel_plane(drm_plane);
+	struct intel_plane *intel_plane = to_intel_plane(dplane);
 	const int pipe = intel_plane->pipe;
 	const int plane = intel_plane->plane + 1;
 
 	I915_WRITE(PLANE_CTL(pipe, plane), 0);
 
-	/* Activate double buffered register update */
 	I915_WRITE(PLANE_SURF(pipe, plane), 0);
 	POSTING_READ(PLANE_SURF(pipe, plane));
 
-	intel_update_sprite_watermarks(drm_plane, crtc, 0, 0, 0, false, false);
+	intel_update_sprite_watermarks(dplane, crtc, 0, 0, 0, false, false);
 }
 
 static void
@@ -360,7 +338,6 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 	struct drm_device *dev = dplane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(dplane);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	int pipe = intel_plane->pipe;
 	int plane = intel_plane->plane;
@@ -448,8 +425,6 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 		linear_offset += src_h * fb->pitches[0] + src_w * pixel_size;
 	}
 
-	intel_update_primary_plane(intel_crtc);
-
 	if (key->flags) {
 		I915_WRITE(SPKEYMINVAL(pipe, plane), key->min_value);
 		I915_WRITE(SPKEYMAXVAL(pipe, plane), key->max_value);
@@ -476,33 +451,26 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 	I915_WRITE(SPCNTR(pipe, plane), sprctl);
 	I915_WRITE(SPSURF(pipe, plane), i915_gem_obj_ggtt_offset(obj) +
 		   sprsurf_offset);
-
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+	POSTING_READ(SPSURF(pipe, plane));
 }
 
 static void
-vlv_disable_plane(struct drm_plane *dplane, struct drm_crtc *crtc)
+vlv_disable_plane(struct drm_plane *dplane, struct drm_crtc *crtc, bool force)
 {
 	struct drm_device *dev = dplane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(dplane);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	int plane = intel_plane->plane;
 
-	intel_update_primary_plane(intel_crtc);
-
 	I915_WRITE(SPCNTR(pipe, plane), 0);
 
-	/* Activate double buffered register update */
 	I915_WRITE(SPSURF(pipe, plane), 0);
-
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+	POSTING_READ(SPSURF(pipe, plane));
 
 	intel_update_sprite_watermarks(dplane, crtc, 0, 0, 0, false, false);
 }
 
-
 static void
 ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		 struct drm_framebuffer *fb,
@@ -514,7 +482,6 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	enum pipe pipe = intel_plane->pipe;
 	u32 sprctl, sprscale = 0;
@@ -595,8 +562,6 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		}
 	}
 
-	intel_update_primary_plane(intel_crtc);
-
 	if (key->flags) {
 		I915_WRITE(SPRKEYVAL(pipe), key->min_value);
 		I915_WRITE(SPRKEYMAX(pipe), key->max_value);
@@ -626,29 +591,24 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	I915_WRITE(SPRCTL(pipe), sprctl);
 	I915_WRITE(SPRSURF(pipe),
 		   i915_gem_obj_ggtt_offset(obj) + sprsurf_offset);
-
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+	POSTING_READ(SPRSURF(pipe));
 }
 
 static void
-ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
+ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc, bool force)
 {
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 
-	intel_update_primary_plane(intel_crtc);
-
 	I915_WRITE(SPRCTL(pipe), I915_READ(SPRCTL(pipe)) & ~SPRITE_ENABLE);
 	/* Can't leave the scaler enabled... */
 	if (intel_plane->can_scale)
 		I915_WRITE(SPRSCALE(pipe), 0);
-	/* Activate double buffered register update */
-	I915_WRITE(SPRSURF(pipe), 0);
 
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+	I915_WRITE(SPRSURF(pipe), 0);
+	POSTING_READ(SPRSURF(pipe));
 }
 
 static void
@@ -662,7 +622,6 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	int pipe = intel_plane->pipe;
 	unsigned long dvssurf_offset, linear_offset;
@@ -735,8 +694,6 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		linear_offset += src_h * fb->pitches[0] + src_w * pixel_size;
 	}
 
-	intel_update_primary_plane(intel_crtc);
-
 	if (key->flags) {
 		I915_WRITE(DVSKEYVAL(pipe), key->min_value);
 		I915_WRITE(DVSKEYMAX(pipe), key->max_value);
@@ -761,109 +718,32 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	I915_WRITE(DVSCNTR(pipe), dvscntr);
 	I915_WRITE(DVSSURF(pipe),
 		   i915_gem_obj_ggtt_offset(obj) + dvssurf_offset);
-
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+	POSTING_READ(DVSSURF(pipe));
 }
 
 static void
-ilk_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
+ilk_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc, bool force)
 {
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 
-	intel_update_primary_plane(intel_crtc);
-
 	I915_WRITE(DVSCNTR(pipe), 0);
 	/* Disable the scaler */
 	I915_WRITE(DVSSCALE(pipe), 0);
 
-	/* Flush double buffered register updates */
 	I915_WRITE(DVSSURF(pipe), 0);
-
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
-}
-
-/**
- * intel_post_enable_primary - Perform operations after enabling primary plane
- * @crtc: the CRTC whose primary plane was just enabled
- *
- * Performs potentially sleeping operations that must be done after the primary
- * plane is enabled, such as updating FBC and IPS.  Note that this may be
- * called due to an explicit primary plane update, or due to an implicit
- * re-enable that is caused when a sprite plane is updated to no longer
- * completely hide the primary plane.
- */
-void
-intel_post_enable_primary(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-
-	/*
-	 * BDW signals flip done immediately if the plane
-	 * is disabled, even if the plane enable is already
-	 * armed to occur at the next vblank :(
-	 */
-	if (IS_BROADWELL(dev))
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-
-	/*
-	 * FIXME IPS should be fine as long as one plane is
-	 * enabled, but in practice it seems to have problems
-	 * when going from primary only to sprite only and vice
-	 * versa.
-	 */
-	hsw_enable_ips(intel_crtc);
-
-	mutex_lock(&dev->struct_mutex);
-	intel_fbc_update(dev);
-	mutex_unlock(&dev->struct_mutex);
-}
-
-/**
- * intel_pre_disable_primary - Perform operations before disabling primary plane
- * @crtc: the CRTC whose primary plane is to be disabled
- *
- * Performs potentially sleeping operations that must be done before the
- * primary plane is enabled, such as updating FBC and IPS.  Note that this may
- * be called due to an explicit primary plane update, or due to an implicit
- * disable that is caused when a sprite plane completely hides the primary
- * plane.
- */
-void
-intel_pre_disable_primary(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);
-
-	mutex_lock(&dev->struct_mutex);
-	if (dev_priv->fbc.crtc == intel_crtc)
-		intel_fbc_disable(dev);
-	mutex_unlock(&dev->struct_mutex);
-
-	/*
-	 * FIXME IPS should be fine as long as one plane is
-	 * enabled, but in practice it seems to have problems
-	 * when going from primary only to sprite only and vice
-	 * versa.
-	 */
-	hsw_disable_ips(intel_crtc);
-}
-
-static bool colorkey_enabled(struct intel_plane *intel_plane)
-{
-	return intel_plane->ckey.flags != I915_SET_COLORKEY_NONE;
+	POSTING_READ(DVSSURF(pipe));
 }
 
 static int
 intel_check_sprite_plane(struct drm_plane *plane,
 			 struct intel_plane_state *state)
 {
+	struct drm_device *dev = plane->dev;
 	struct intel_crtc *intel_crtc = to_intel_crtc(state->base.crtc);
+	struct intel_crtc_state *crtc_state;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
 	struct drm_framebuffer *fb = state->base.fb;
 	int crtc_x, crtc_y;
@@ -874,9 +754,13 @@ intel_check_sprite_plane(struct drm_plane *plane,
 	const struct drm_rect *clip = &state->clip;
 	int hscale, vscale;
 	int max_scale, min_scale;
+	bool can_scale;
 	int pixel_size;
+	int ret;
 
 	intel_crtc = intel_crtc ? intel_crtc : to_intel_crtc(plane->crtc);
+	crtc_state = state->base.state ?
+		intel_atomic_get_crtc_state(state->base.state, intel_crtc) : NULL;
 
 	if (!fb) {
 		state->visible = false;
@@ -895,13 +779,29 @@ intel_check_sprite_plane(struct drm_plane *plane,
 		return -EINVAL;
 	}
 
+	/* setup can_scale, min_scale, max_scale */
+	if (INTEL_INFO(dev)->gen >= 9) {
+		/* use scaler when colorkey is not required */
+		if (intel_plane->ckey.flags == I915_SET_COLORKEY_NONE) {
+			can_scale = 1;
+			min_scale = 1;
+			max_scale = skl_max_scale(intel_crtc, crtc_state);
+		} else {
+			can_scale = 0;
+			min_scale = DRM_PLANE_HELPER_NO_SCALING;
+			max_scale = DRM_PLANE_HELPER_NO_SCALING;
+		}
+	} else {
+		can_scale = intel_plane->can_scale;
+		max_scale = intel_plane->max_downscale << 16;
+		min_scale = intel_plane->can_scale ? 1 : (1 << 16);
+	}
+
 	/*
 	 * FIXME the following code does a bunch of fuzzy adjustments to the
 	 * coordinates and sizes. We probably need some way to decide whether
 	 * more strict checking should be done instead.
 	 */
-	max_scale = intel_plane->max_downscale << 16;
-	min_scale = intel_plane->can_scale ? 1 : (1 << 16);
 
 	drm_rect_rotate(src, fb->width << 16, fb->height << 16,
 			state->base.rotation);
@@ -972,7 +872,7 @@ intel_check_sprite_plane(struct drm_plane *plane,
 			 * Must keep src and dst the
 			 * same if we can't scale.
 			 */
-			if (!intel_plane->can_scale)
+			if (!can_scale)
 				crtc_w &= ~1;
 
 			if (crtc_w == 0)
@@ -984,7 +884,7 @@ intel_check_sprite_plane(struct drm_plane *plane,
 	if (state->visible && (src_w != crtc_w || src_h != crtc_h)) {
 		unsigned int width_bytes;
 
-		WARN_ON(!intel_plane->can_scale);
+		WARN_ON(!can_scale);
 
 		/* FIXME interlacing min height is 6 */
 
@@ -998,18 +898,18 @@ intel_check_sprite_plane(struct drm_plane *plane,
 		width_bytes = ((src_x * pixel_size) & 63) +
 					src_w * pixel_size;
 
-		if (src_w > 2048 || src_h > 2048 ||
-		    width_bytes > 4096 || fb->pitches[0] > 4096) {
+		if (INTEL_INFO(dev)->gen < 9 && (src_w > 2048 || src_h > 2048 ||
+		    width_bytes > 4096 || fb->pitches[0] > 4096)) {
 			DRM_DEBUG_KMS("Source dimensions exceed hardware limits\n");
 			return -EINVAL;
 		}
 	}
 
 	if (state->visible) {
-		src->x1 = src_x;
-		src->x2 = src_x + src_w;
-		src->y1 = src_y;
-		src->y2 = src_y + src_h;
+		src->x1 = src_x << 16;
+		src->x2 = (src_x + src_w) << 16;
+		src->y1 = src_y << 16;
+		src->y2 = (src_y + src_h) << 16;
 	}
 
 	dst->x1 = crtc_x;
@@ -1022,23 +922,10 @@ finish:
 	 * If the sprite is completely covering the primary plane,
 	 * we can disable the primary and save power.
 	 */
-	state->hides_primary = fb != NULL && drm_rect_equals(dst, clip) &&
-		!colorkey_enabled(intel_plane);
-	WARN_ON(state->hides_primary && !state->visible && intel_crtc->active);
-
 	if (intel_crtc->active) {
-		if (intel_crtc->primary_enabled == state->hides_primary)
-			intel_crtc->atomic.wait_for_flips = true;
-
-		if (intel_crtc->primary_enabled && state->hides_primary)
-			intel_crtc->atomic.pre_disable_primary = true;
-
 		intel_crtc->atomic.fb_bits |=
 			INTEL_FRONTBUFFER_SPRITE(intel_crtc->pipe);
 
-		if (!intel_crtc->primary_enabled && !state->hides_primary)
-			intel_crtc->atomic.post_enable_primary = true;
-
 		if (intel_wm_need_update(plane, &state->base))
 			intel_crtc->atomic.update_wm = true;
 
@@ -1053,6 +940,13 @@ finish:
 		}
 	}
 
+	if (INTEL_INFO(dev)->gen >= 9) {
+		ret = skl_update_scaler_users(intel_crtc, crtc_state, intel_plane,
+			state, 0);
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 
@@ -1074,22 +968,20 @@ intel_commit_sprite_plane(struct drm_plane *plane,
 	plane->fb = fb;
 
 	if (intel_crtc->active) {
-		intel_crtc->primary_enabled = !state->hides_primary;
-
 		if (state->visible) {
 			crtc_x = state->dst.x1;
 			crtc_y = state->dst.y1;
 			crtc_w = drm_rect_width(&state->dst);
 			crtc_h = drm_rect_height(&state->dst);
-			src_x = state->src.x1;
-			src_y = state->src.y1;
-			src_w = drm_rect_width(&state->src);
-			src_h = drm_rect_height(&state->src);
+			src_x = state->src.x1 >> 16;
+			src_y = state->src.y1 >> 16;
+			src_w = drm_rect_width(&state->src) >> 16;
+			src_h = drm_rect_height(&state->src) >> 16;
 			intel_plane->update_plane(plane, crtc, fb,
 						  crtc_x, crtc_y, crtc_w, crtc_h,
 						  src_x, src_y, src_w, src_h);
 		} else {
-			intel_plane->disable_plane(plane, crtc);
+			intel_plane->disable_plane(plane, crtc, false);
 		}
 	}
 }
@@ -1119,6 +1011,16 @@ int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
 	}
 
 	intel_plane = to_intel_plane(plane);
+
+	if (INTEL_INFO(dev)->gen >= 9) {
+		/* plane scaling and colorkey are mutually exclusive */
+		if (to_intel_plane_state(plane->state)->scaler_id >= 0) {
+			DRM_ERROR("colorkey not allowed with scaler\n");
+			ret = -EINVAL;
+			goto out_unlock;
+		}
+	}
+
 	intel_plane->ckey = *set;
 
 	/*
@@ -1139,14 +1041,14 @@ int intel_plane_restore(struct drm_plane *plane)
 	if (!plane->crtc || !plane->state->fb)
 		return 0;
 
-	return plane->funcs->update_plane(plane, plane->crtc, plane->state->fb,
-				  plane->state->crtc_x, plane->state->crtc_y,
-				  plane->state->crtc_w, plane->state->crtc_h,
-				  plane->state->src_x, plane->state->src_y,
-				  plane->state->src_w, plane->state->src_h);
+	return drm_plane_helper_update(plane, plane->crtc, plane->state->fb,
+				       plane->state->crtc_x, plane->state->crtc_y,
+				       plane->state->crtc_w, plane->state->crtc_h,
+				       plane->state->src_x, plane->state->src_y,
+				       plane->state->src_w, plane->state->src_h);
 }
 
-static uint32_t ilk_plane_formats[] = {
+static const uint32_t ilk_plane_formats[] = {
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_YVYU,
@@ -1154,7 +1056,7 @@ static uint32_t ilk_plane_formats[] = {
 	DRM_FORMAT_VYUY,
 };
 
-static uint32_t snb_plane_formats[] = {
+static const uint32_t snb_plane_formats[] = {
 	DRM_FORMAT_XBGR8888,
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_YUYV,
@@ -1163,7 +1065,7 @@ static uint32_t snb_plane_formats[] = {
 	DRM_FORMAT_VYUY,
 };
 
-static uint32_t vlv_plane_formats[] = {
+static const uint32_t vlv_plane_formats[] = {
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_ABGR8888,
 	DRM_FORMAT_ARGB8888,
@@ -1255,14 +1157,10 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 		}
 		break;
 	case 9:
-		/*
-		 * FIXME: Skylake planes can be scaled (with some restrictions),
-		 * but this is for another time.
-		 */
-		intel_plane->can_scale = false;
-		intel_plane->max_downscale = 1;
+		intel_plane->can_scale = true;
 		intel_plane->update_plane = skl_update_plane;
 		intel_plane->disable_plane = skl_disable_plane;
+		state->scaler_id = -1;
 
 		plane_formats = skl_plane_formats;
 		num_plane_formats = ARRAY_SIZE(skl_plane_formats);
@@ -1276,6 +1174,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 	intel_plane->plane = plane;
 	intel_plane->check_plane = intel_check_sprite_plane;
 	intel_plane->commit_plane = intel_commit_sprite_plane;
+	intel_plane->ckey.flags = I915_SET_COLORKEY_NONE;
 	possible_crtcs = (1 << pipe);
 	ret = drm_universal_plane_init(dev, &intel_plane->base, possible_crtcs,
 				       &intel_plane_funcs,
@@ -1286,16 +1185,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 		goto out;
 	}
 
-	if (!dev->mode_config.rotation_property)
-		dev->mode_config.rotation_property =
-			drm_mode_create_rotation_property(dev,
-							  BIT(DRM_ROTATE_0) |
-							  BIT(DRM_ROTATE_180));
-
-	if (dev->mode_config.rotation_property)
-		drm_object_attach_property(&intel_plane->base.base,
-					   dev->mode_config.rotation_property,
-					   state->base.rotation);
+	intel_create_rotation_property(dev, intel_plane);
 
 	drm_plane_helper_add(&intel_plane->base, &intel_plane_helper_funcs);
 
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index ff2a74651dd4..a6d8a3ee7750 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -383,6 +383,26 @@ void intel_uncore_sanitize(struct drm_device *dev)
 	intel_disable_gt_powersave(dev);
 }
 
+static void __intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
+					 enum forcewake_domains fw_domains)
+{
+	struct intel_uncore_forcewake_domain *domain;
+	enum forcewake_domain_id id;
+
+	if (!dev_priv->uncore.funcs.force_wake_get)
+		return;
+
+	fw_domains &= dev_priv->uncore.fw_domains;
+
+	for_each_fw_domain_mask(domain, fw_domains, dev_priv, id) {
+		if (domain->wake_count++)
+			fw_domains &= ~(1 << id);
+	}
+
+	if (fw_domains)
+		dev_priv->uncore.funcs.force_wake_get(dev_priv, fw_domains);
+}
+
 /**
  * intel_uncore_forcewake_get - grab forcewake domain references
  * @dev_priv: i915 device instance
@@ -400,41 +420,39 @@ void intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
 				enum forcewake_domains fw_domains)
 {
 	unsigned long irqflags;
-	struct intel_uncore_forcewake_domain *domain;
-	enum forcewake_domain_id id;
 
 	if (!dev_priv->uncore.funcs.force_wake_get)
 		return;
 
 	WARN_ON(dev_priv->pm.suspended);
 
-	fw_domains &= dev_priv->uncore.fw_domains;
-
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-
-	for_each_fw_domain_mask(domain, fw_domains, dev_priv, id) {
-		if (domain->wake_count++)
-			fw_domains &= ~(1 << id);
-	}
-
-	if (fw_domains)
-		dev_priv->uncore.funcs.force_wake_get(dev_priv, fw_domains);
-
+	__intel_uncore_forcewake_get(dev_priv, fw_domains);
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
 /**
- * intel_uncore_forcewake_put - release a forcewake domain reference
+ * intel_uncore_forcewake_get__locked - grab forcewake domain references
  * @dev_priv: i915 device instance
- * @fw_domains: forcewake domains to put references
+ * @fw_domains: forcewake domains to get reference on
  *
- * This function drops the device-level forcewakes for specified
- * domains obtained by intel_uncore_forcewake_get().
+ * See intel_uncore_forcewake_get(). This variant places the onus
+ * on the caller to explicitly handle the dev_priv->uncore.lock spinlock.
  */
-void intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
-				enum forcewake_domains fw_domains)
+void intel_uncore_forcewake_get__locked(struct drm_i915_private *dev_priv,
+					enum forcewake_domains fw_domains)
+{
+	assert_spin_locked(&dev_priv->uncore.lock);
+
+	if (!dev_priv->uncore.funcs.force_wake_get)
+		return;
+
+	__intel_uncore_forcewake_get(dev_priv, fw_domains);
+}
+
+static void __intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
+					 enum forcewake_domains fw_domains)
 {
-	unsigned long irqflags;
 	struct intel_uncore_forcewake_domain *domain;
 	enum forcewake_domain_id id;
 
@@ -443,8 +461,6 @@ void intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
 
 	fw_domains &= dev_priv->uncore.fw_domains;
 
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-
 	for_each_fw_domain_mask(domain, fw_domains, dev_priv, id) {
 		if (WARN_ON(domain->wake_count == 0))
 			continue;
@@ -455,10 +471,48 @@ void intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
 		domain->wake_count++;
 		fw_domain_arm_timer(domain);
 	}
+}
 
+/**
+ * intel_uncore_forcewake_put - release a forcewake domain reference
+ * @dev_priv: i915 device instance
+ * @fw_domains: forcewake domains to put references
+ *
+ * This function drops the device-level forcewakes for specified
+ * domains obtained by intel_uncore_forcewake_get().
+ */
+void intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
+				enum forcewake_domains fw_domains)
+{
+	unsigned long irqflags;
+
+	if (!dev_priv->uncore.funcs.force_wake_put)
+		return;
+
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+	__intel_uncore_forcewake_put(dev_priv, fw_domains);
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
+/**
+ * intel_uncore_forcewake_put__locked - grab forcewake domain references
+ * @dev_priv: i915 device instance
+ * @fw_domains: forcewake domains to get reference on
+ *
+ * See intel_uncore_forcewake_put(). This variant places the onus
+ * on the caller to explicitly handle the dev_priv->uncore.lock spinlock.
+ */
+void intel_uncore_forcewake_put__locked(struct drm_i915_private *dev_priv,
+					enum forcewake_domains fw_domains)
+{
+	assert_spin_locked(&dev_priv->uncore.lock);
+
+	if (!dev_priv->uncore.funcs.force_wake_put)
+		return;
+
+	__intel_uncore_forcewake_put(dev_priv, fw_domains);
+}
+
 void assert_forcewakes_inactive(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore_forcewake_domain *domain;