These changes are the raw update to linux-4.4.6-rt14. Kernel sources

are taken from kernel.org, and rt patch from the rt wiki download page.

During the rebasing, the following patch collided:

Force tick interrupt and get rid of softirq magic(I70131fb85).

Collisions have been removed because its logic was found on the
source already.

Change-Id: I7f57a4081d9deaa0d9ccfc41a6c8daccdee3b769
Signed-off-by: José Pekkarinen <jose.pekkarinen@nokia.com>

63 files changed, 2096 insertions, 861 deletions

diff --git a/kernel/drivers/gpu/drm/radeon/atombios.h b/kernel/drivers/gpu/drm/radeon/atombios.h
index 92be50c39..ab89eed9d 100644
--- a/kernel/drivers/gpu/drm/radeon/atombios.h
+++ b/kernel/drivers/gpu/drm/radeon/atombios.h
@@ -7944,8 +7944,8 @@ typedef struct {
 typedef struct {
   AMD_ACPI_DESCRIPTION_HEADER SHeader;
   UCHAR TableUUID[16];    //0x24
-  ULONG VBIOSImageOffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the stucture.
-  ULONG Lib1ImageOffset;  //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the stucture.
+  ULONG VBIOSImageOffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the structure.
+  ULONG Lib1ImageOffset;  //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the structure.
   ULONG Reserved[4];      //0x3C
 }UEFI_ACPI_VFCT;
 
diff --git a/kernel/drivers/gpu/drm/radeon/atombios_dp.c b/kernel/drivers/gpu/drm/radeon/atombios_dp.c
index 447dbfa6c..bd73b4069 100644
--- a/kernel/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/kernel/drivers/gpu/drm/radeon/atombios_dp.c
@@ -179,6 +179,7 @@ radeon_dp_aux_transfer_atom(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 	switch (msg->request & ~DP_AUX_I2C_MOT) {
 	case DP_AUX_NATIVE_WRITE:
 	case DP_AUX_I2C_WRITE:
+	case DP_AUX_I2C_WRITE_STATUS_UPDATE:
 		/* The atom implementation only supports writes with a max payload of
 		 * 12 bytes since it uses 4 bits for the total count (header + payload)
 		 * in the parameter space.  The atom interface supports 16 byte
@@ -254,7 +255,7 @@ void radeon_dp_aux_init(struct radeon_connector *radeon_connector)
 #define DP_VOLTAGE_MAX         DP_TRAIN_VOLTAGE_SWING_LEVEL_3
 #define DP_PRE_EMPHASIS_MAX    DP_TRAIN_PRE_EMPH_LEVEL_3
 
-static void dp_get_adjust_train(u8 link_status[DP_LINK_STATUS_SIZE],
+static void dp_get_adjust_train(const u8 link_status[DP_LINK_STATUS_SIZE],
 				int lane_count,
 				u8 train_set[4])
 {
@@ -312,7 +313,7 @@ static int dp_get_max_dp_pix_clock(int link_rate,
 /***** radeon specific DP functions *****/
 
 int radeon_dp_get_max_link_rate(struct drm_connector *connector,
-				u8 dpcd[DP_DPCD_SIZE])
+				const u8 dpcd[DP_DPCD_SIZE])
 {
 	int max_link_rate;
 
@@ -329,7 +330,7 @@ int radeon_dp_get_max_link_rate(struct drm_connector *connector,
  * if the max lane# < low rate lane# then use max lane# instead.
  */
 static int radeon_dp_get_dp_lane_number(struct drm_connector *connector,
-					u8 dpcd[DP_DPCD_SIZE],
+					const u8 dpcd[DP_DPCD_SIZE],
 					int pix_clock)
 {
 	int bpp = convert_bpc_to_bpp(radeon_get_monitor_bpc(connector));
@@ -348,7 +349,7 @@ static int radeon_dp_get_dp_lane_number(struct drm_connector *connector,
 }
 
 static int radeon_dp_get_dp_link_clock(struct drm_connector *connector,
-				       u8 dpcd[DP_DPCD_SIZE],
+				       const u8 dpcd[DP_DPCD_SIZE],
 				       int pix_clock)
 {
 	int bpp = convert_bpc_to_bpp(radeon_get_monitor_bpc(connector));
diff --git a/kernel/drivers/gpu/drm/radeon/atombios_encoders.c b/kernel/drivers/gpu/drm/radeon/atombios_encoders.c
index dd39f434b..bb2921439 100644
--- a/kernel/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/kernel/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -237,6 +237,7 @@ void radeon_atom_backlight_init(struct radeon_encoder *radeon_encoder,
 	backlight_update_status(bd);
 
 	DRM_INFO("radeon atom DIG backlight initialized\n");
+	rdev->mode_info.bl_encoder = radeon_encoder;
 
 	return;
 
@@ -1624,8 +1625,14 @@ radeon_atom_encoder_dpms_avivo(struct drm_encoder *encoder, int mode)
 		} else
 			atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 		if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) {
-			args.ucAction = ATOM_LCD_BLON;
-			atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+			if (rdev->mode_info.bl_encoder) {
+				struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+
+				atombios_set_backlight_level(radeon_encoder, dig->backlight_level);
+			} else {
+				args.ucAction = ATOM_LCD_BLON;
+				atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+			}
 		}
 		break;
 	case DRM_MODE_DPMS_STANDBY:
@@ -1705,9 +1712,13 @@ radeon_atom_encoder_dpms_dig(struct drm_encoder *encoder, int mode)
 			if (ASIC_IS_DCE4(rdev))
 				atombios_dig_encoder_setup(encoder, ATOM_ENCODER_CMD_DP_VIDEO_ON, 0);
 		}
-		if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT))
-			atombios_dig_transmitter_setup(encoder,
-						       ATOM_TRANSMITTER_ACTION_LCD_BLON, 0, 0);
+		if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) {
+			if (rdev->mode_info.bl_encoder)
+				atombios_set_backlight_level(radeon_encoder, dig->backlight_level);
+			else
+				atombios_dig_transmitter_setup(encoder,
+							       ATOM_TRANSMITTER_ACTION_LCD_BLON, 0, 0);
+		}
 		if (ext_encoder)
 			atombios_external_encoder_setup(encoder, ext_encoder, ATOM_ENABLE);
 		break;
@@ -2299,8 +2310,7 @@ radeon_atom_encoder_mode_set(struct drm_encoder *encoder,
 	encoder_mode = atombios_get_encoder_mode(encoder);
 	if (connector && (radeon_audio != 0) &&
 	    ((encoder_mode == ATOM_ENCODER_MODE_HDMI) ||
-	     (ENCODER_MODE_IS_DP(encoder_mode) &&
-	      drm_detect_monitor_audio(radeon_connector_edid(connector)))))
+	     ENCODER_MODE_IS_DP(encoder_mode)))
 		radeon_audio_mode_set(encoder, adjusted_mode);
 }
 
diff --git a/kernel/drivers/gpu/drm/radeon/cayman_blit_shaders.c b/kernel/drivers/gpu/drm/radeon/cayman_blit_shaders.c
index 98d009e15..9fec4d09f 100644
--- a/kernel/drivers/gpu/drm/radeon/cayman_blit_shaders.c
+++ b/kernel/drivers/gpu/drm/radeon/cayman_blit_shaders.c
@@ -32,7 +32,7 @@
  * evergreen cards need to use the 3D engine to blit data which requires
  * quite a bit of hw state setup.  Rather than pull the whole 3D driver
  * (which normally generates the 3D state) into the DRM, we opt to use
- * statically generated state tables.  The regsiter state and shaders
+ * statically generated state tables.  The register state and shaders
  * were hand generated to support blitting functionality.  See the 3D
  * driver or documentation for descriptions of the registers and
  * shader instructions.
diff --git a/kernel/drivers/gpu/drm/radeon/cik.c b/kernel/drivers/gpu/drm/radeon/cik.c
index 845665362..f81fb2641 100644
--- a/kernel/drivers/gpu/drm/radeon/cik.c
+++ b/kernel/drivers/gpu/drm/radeon/cik.c
@@ -174,6 +174,31 @@ int cik_get_allowed_info_register(struct radeon_device *rdev,
 	}
 }
 
+/*
+ * Indirect registers accessor
+ */
+u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg)
+{
+	unsigned long flags;
+	u32 r;
+
+	spin_lock_irqsave(&rdev->didt_idx_lock, flags);
+	WREG32(CIK_DIDT_IND_INDEX, (reg));
+	r = RREG32(CIK_DIDT_IND_DATA);
+	spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
+	return r;
+}
+
+void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->didt_idx_lock, flags);
+	WREG32(CIK_DIDT_IND_INDEX, (reg));
+	WREG32(CIK_DIDT_IND_DATA, (v));
+	spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
+}
+
 /* get temperature in millidegrees */
 int ci_get_temp(struct radeon_device *rdev)
 {
@@ -4148,11 +4173,7 @@ void cik_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 	control |= ib->length_dw | (vm_id << 24);
 
 	radeon_ring_write(ring, header);
-	radeon_ring_write(ring,
-#ifdef __BIG_ENDIAN
-			  (2 << 0) |
-#endif
-			  (ib->gpu_addr & 0xFFFFFFFC));
+	radeon_ring_write(ring, (ib->gpu_addr & 0xFFFFFFFC));
 	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF);
 	radeon_ring_write(ring, control);
 }
@@ -8447,7 +8468,7 @@ restart_ih:
 	if (queue_dp)
 		schedule_work(&rdev->dp_work);
 	if (queue_hotplug)
-		schedule_work(&rdev->hotplug_work);
+		schedule_delayed_work(&rdev->hotplug_work, 0);
 	if (queue_reset) {
 		rdev->needs_reset = true;
 		wake_up_all(&rdev->fence_queue);
@@ -9605,6 +9626,9 @@ static void dce8_program_watermarks(struct radeon_device *rdev,
 		    (rdev->disp_priority == 2)) {
 			DRM_DEBUG_KMS("force priority to high\n");
 		}
+
+		/* Save number of lines the linebuffer leads before the scanout */
+		radeon_crtc->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay);
 	}
 
 	/* select wm A */
diff --git a/kernel/drivers/gpu/drm/radeon/cik_reg.h b/kernel/drivers/gpu/drm/radeon/cik_reg.h
index f667347d8..4e883fdc5 100644
--- a/kernel/drivers/gpu/drm/radeon/cik_reg.h
+++ b/kernel/drivers/gpu/drm/radeon/cik_reg.h
@@ -149,8 +149,30 @@
 
 #define KFD_CIK_SDMA_QUEUE_OFFSET		0x200
 
+#define SQ_IND_INDEX					0x8DE0
+#define SQ_CMD						0x8DEC
+#define SQ_IND_DATA					0x8DE4
+
+/*
+ * The TCP_WATCHx_xxxx addresses that are shown here are in dwords,
+ * and that's why they are multiplied by 4
+ */
+#define TCP_WATCH0_ADDR_H				(0x32A0*4)
+#define TCP_WATCH1_ADDR_H				(0x32A3*4)
+#define TCP_WATCH2_ADDR_H				(0x32A6*4)
+#define TCP_WATCH3_ADDR_H				(0x32A9*4)
+#define TCP_WATCH0_ADDR_L				(0x32A1*4)
+#define TCP_WATCH1_ADDR_L				(0x32A4*4)
+#define TCP_WATCH2_ADDR_L				(0x32A7*4)
+#define TCP_WATCH3_ADDR_L				(0x32AA*4)
+#define TCP_WATCH0_CNTL					(0x32A2*4)
+#define TCP_WATCH1_CNTL					(0x32A5*4)
+#define TCP_WATCH2_CNTL					(0x32A8*4)
+#define TCP_WATCH3_CNTL					(0x32AB*4)
+
+#define CPC_INT_CNTL					0xC2D0
+
 #define CP_HQD_IQ_RPTR					0xC970u
-#define AQL_ENABLE					(1U << 0)
 #define SDMA0_RLC0_RB_CNTL				0xD400u
 #define	SDMA_RB_VMID(x)					(x << 24)
 #define	SDMA0_RLC0_RB_BASE				0xD404u
@@ -184,4 +206,38 @@
 #define	SDMA0_CNTL					0xD010
 #define	SDMA1_CNTL					0xD810
 
+enum {
+	MAX_TRAPID = 8,		/* 3 bits in the bitfield.  */
+	MAX_WATCH_ADDRESSES = 4
+};
+
+enum {
+	ADDRESS_WATCH_REG_ADDR_HI = 0,
+	ADDRESS_WATCH_REG_ADDR_LO,
+	ADDRESS_WATCH_REG_CNTL,
+	ADDRESS_WATCH_REG_MAX
+};
+
+enum {				/*  not defined in the CI/KV reg file  */
+	ADDRESS_WATCH_REG_CNTL_ATC_BIT = 0x10000000UL,
+	ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK = 0x00FFFFFF,
+	ADDRESS_WATCH_REG_ADDLOW_MASK_EXTENSION = 0x03000000,
+	/* extend the mask to 26 bits in order to match the low address field */
+	ADDRESS_WATCH_REG_ADDLOW_SHIFT = 6,
+	ADDRESS_WATCH_REG_ADDHIGH_MASK = 0xFFFF
+};
+
+union TCP_WATCH_CNTL_BITS {
+	struct {
+		uint32_t mask:24;
+		uint32_t vmid:4;
+		uint32_t atc:1;
+		uint32_t mode:2;
+		uint32_t valid:1;
+	} bitfields, bits;
+	uint32_t u32All;
+	signed int i32All;
+	float f32All;
+};
+
 #endif
diff --git a/kernel/drivers/gpu/drm/radeon/cikd.h b/kernel/drivers/gpu/drm/radeon/cikd.h
index 0089d837a..391ff9d5d 100644
--- a/kernel/drivers/gpu/drm/radeon/cikd.h
+++ b/kernel/drivers/gpu/drm/radeon/cikd.h
@@ -1335,6 +1335,7 @@
 #       define CNTX_EMPTY_INT_ENABLE                    (1 << 20)
 #       define PRIV_INSTR_INT_ENABLE                    (1 << 22)
 #       define PRIV_REG_INT_ENABLE                      (1 << 23)
+#       define OPCODE_ERROR_INT_ENABLE                  (1 << 24)
 #       define TIME_STAMP_INT_ENABLE                    (1 << 26)
 #       define CP_RINGID2_INT_ENABLE                    (1 << 29)
 #       define CP_RINGID1_INT_ENABLE                    (1 << 30)
@@ -2147,9 +2148,12 @@
 #define VCE_CMD_IB_AUTO		0x00000005
 #define VCE_CMD_SEMAPHORE	0x00000006
 
-#define ATC_VMID0_PASID_MAPPING					0x339Cu
-#define	ATC_VMID_PASID_MAPPING_UPDATE_STATUS	0x3398u
-#define	ATC_VMID_PASID_MAPPING_VALID				(1U << 31)
+#define ATC_VMID_PASID_MAPPING_UPDATE_STATUS		0x3398u
+#define ATC_VMID0_PASID_MAPPING				0x339Cu
+#define ATC_VMID_PASID_MAPPING_PASID_MASK		(0xFFFF)
+#define ATC_VMID_PASID_MAPPING_PASID_SHIFT		0
+#define ATC_VMID_PASID_MAPPING_VALID_MASK		(0x1 << 31)
+#define ATC_VMID_PASID_MAPPING_VALID_SHIFT		31
 
 #define ATC_VM_APERTURE0_CNTL					0x3310u
 #define	ATS_ACCESS_MODE_NEVER						0
diff --git a/kernel/drivers/gpu/drm/radeon/dce6_afmt.c b/kernel/drivers/gpu/drm/radeon/dce6_afmt.c
index 44480c1b9..367a916f3 100644
--- a/kernel/drivers/gpu/drm/radeon/dce6_afmt.c
+++ b/kernel/drivers/gpu/drm/radeon/dce6_afmt.c
@@ -76,16 +76,35 @@ static void dce6_afmt_get_connected_pins(struct radeon_device *rdev)
 
 struct r600_audio_pin *dce6_audio_get_pin(struct radeon_device *rdev)
 {
-	int i;
+	struct drm_encoder *encoder;
+	struct radeon_encoder *radeon_encoder;
+	struct radeon_encoder_atom_dig *dig;
+	struct r600_audio_pin *pin = NULL;
+	int i, pin_count;
 
 	dce6_afmt_get_connected_pins(rdev);
 
 	for (i = 0; i < rdev->audio.num_pins; i++) {
-		if (rdev->audio.pin[i].connected)
-			return &rdev->audio.pin[i];
+		if (rdev->audio.pin[i].connected) {
+			pin = &rdev->audio.pin[i];
+			pin_count = 0;
+
+			list_for_each_entry(encoder, &rdev->ddev->mode_config.encoder_list, head) {
+				if (radeon_encoder_is_digital(encoder)) {
+					radeon_encoder = to_radeon_encoder(encoder);
+					dig = radeon_encoder->enc_priv;
+					if (dig->pin == pin)
+						pin_count++;
+				}
+			}
+
+			if (pin_count == 0)
+				return pin;
+		}
 	}
-	DRM_ERROR("No connected audio pins found!\n");
-	return NULL;
+	if (!pin)
+		DRM_ERROR("No connected audio pins found!\n");
+	return pin;
 }
 
 void dce6_afmt_select_pin(struct drm_encoder *encoder)
@@ -282,6 +301,14 @@ void dce6_dp_audio_set_dto(struct radeon_device *rdev,
 	 * is the numerator, DCCG_AUDIO_DTOx_MODULE is the denominator
 	 */
 	if (ASIC_IS_DCE8(rdev)) {
+		unsigned int div = (RREG32(DENTIST_DISPCLK_CNTL) &
+			DENTIST_DPREFCLK_WDIVIDER_MASK) >>
+			DENTIST_DPREFCLK_WDIVIDER_SHIFT;
+		div = radeon_audio_decode_dfs_div(div);
+
+		if (div)
+			clock = clock * 100 / div;
+
 		WREG32(DCE8_DCCG_AUDIO_DTO1_PHASE, 24000);
 		WREG32(DCE8_DCCG_AUDIO_DTO1_MODULE, clock);
 	} else {
diff --git a/kernel/drivers/gpu/drm/radeon/evergreen.c b/kernel/drivers/gpu/drm/radeon/evergreen.c
index feef136cd..2ad462896 100644
--- a/kernel/drivers/gpu/drm/radeon/evergreen.c
+++ b/kernel/drivers/gpu/drm/radeon/evergreen.c
@@ -35,6 +35,75 @@
 #include "evergreen_blit_shaders.h"
 #include "radeon_ucode.h"
 
+/*
+ * Indirect registers accessor
+ */
+u32 eg_cg_rreg(struct radeon_device *rdev, u32 reg)
+{
+	unsigned long flags;
+	u32 r;
+
+	spin_lock_irqsave(&rdev->cg_idx_lock, flags);
+	WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff));
+	r = RREG32(EVERGREEN_CG_IND_DATA);
+	spin_unlock_irqrestore(&rdev->cg_idx_lock, flags);
+	return r;
+}
+
+void eg_cg_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->cg_idx_lock, flags);
+	WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff));
+	WREG32(EVERGREEN_CG_IND_DATA, (v));
+	spin_unlock_irqrestore(&rdev->cg_idx_lock, flags);
+}
+
+u32 eg_pif_phy0_rreg(struct radeon_device *rdev, u32 reg)
+{
+	unsigned long flags;
+	u32 r;
+
+	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+	WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff));
+	r = RREG32(EVERGREEN_PIF_PHY0_DATA);
+	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+	return r;
+}
+
+void eg_pif_phy0_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+	WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff));
+	WREG32(EVERGREEN_PIF_PHY0_DATA, (v));
+	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+}
+
+u32 eg_pif_phy1_rreg(struct radeon_device *rdev, u32 reg)
+{
+	unsigned long flags;
+	u32 r;
+
+	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+	WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff));
+	r = RREG32(EVERGREEN_PIF_PHY1_DATA);
+	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+	return r;
+}
+
+void eg_pif_phy1_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+	WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff));
+	WREG32(EVERGREEN_PIF_PHY1_DATA, (v));
+	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+}
+
 static const u32 crtc_offsets[6] =
 {
 	EVERGREEN_CRTC0_REGISTER_OFFSET,
@@ -1335,44 +1404,20 @@ void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc)
  * @crtc_id: crtc to cleanup pageflip on
  * @crtc_base: new address of the crtc (GPU MC address)
  *
- * Does the actual pageflip (evergreen+).
- * During vblank we take the crtc lock and wait for the update_pending
- * bit to go high, when it does, we release the lock, and allow the
- * double buffered update to take place.
- * Returns the current update pending status.
+ * Triggers the actual pageflip by updating the primary
+ * surface base address (evergreen+).
  */
 void evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
-	u32 tmp = RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset);
-	int i;
-
-	/* Lock the graphics update lock */
-	tmp |= EVERGREEN_GRPH_UPDATE_LOCK;
-	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
 
 	/* update the scanout addresses */
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
-	       upper_32_bits(crtc_base));
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
-	       (u32)crtc_base);
-
 	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
 	       upper_32_bits(crtc_base));
 	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
 	       (u32)crtc_base);
-
-	/* Wait for update_pending to go high. */
-	for (i = 0; i < rdev->usec_timeout; i++) {
-		if (RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING)
-			break;
-		udelay(1);
-	}
-	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
-
-	/* Unlock the lock, so double-buffering can take place inside vblank */
-	tmp &= ~EVERGREEN_GRPH_UPDATE_LOCK;
-	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+	/* post the write */
+	RREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset);
 }
 
 /**
@@ -2327,6 +2372,9 @@ static void evergreen_program_watermarks(struct radeon_device *rdev,
 		c.full = dfixed_div(c, a);
 		priority_b_mark = dfixed_trunc(c);
 		priority_b_cnt |= priority_b_mark & PRIORITY_MARK_MASK;
+
+		/* Save number of lines the linebuffer leads before the scanout */
+		radeon_crtc->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay);
 	}
 
 	/* select wm A */
@@ -5299,7 +5347,7 @@ restart_ih:
 	if (queue_dp)
 		schedule_work(&rdev->dp_work);
 	if (queue_hotplug)
-		schedule_work(&rdev->hotplug_work);
+		schedule_delayed_work(&rdev->hotplug_work, 0);
 	if (queue_hdmi)
 		schedule_work(&rdev->audio_work);
 	if (queue_thermal && rdev->pm.dpm_enabled)
diff --git a/kernel/drivers/gpu/drm/radeon/evergreen_blit_shaders.c b/kernel/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
index d43383470..1a96ddb3e 100644
--- a/kernel/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
+++ b/kernel/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
@@ -32,7 +32,7 @@
  * evergreen cards need to use the 3D engine to blit data which requires
  * quite a bit of hw state setup.  Rather than pull the whole 3D driver
  * (which normally generates the 3D state) into the DRM, we opt to use
- * statically generated state tables.  The regsiter state and shaders
+ * statically generated state tables.  The register state and shaders
  * were hand generated to support blitting functionality.  See the 3D
  * driver or documentation for descriptions of the registers and
  * shader instructions.
diff --git a/kernel/drivers/gpu/drm/radeon/evergreen_cs.c b/kernel/drivers/gpu/drm/radeon/evergreen_cs.c
index c9e0fbbf7..46f87d4aa 100644
--- a/kernel/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/kernel/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -34,6 +34,8 @@
 #define MAX(a,b)                   (((a)>(b))?(a):(b))
 #define MIN(a,b)                   (((a)<(b))?(a):(b))
 
+#define REG_SAFE_BM_SIZE ARRAY_SIZE(evergreen_reg_safe_bm)
+
 int r600_dma_cs_next_reloc(struct radeon_cs_parser *p,
 			   struct radeon_bo_list **cs_reloc);
 struct evergreen_cs_track {
@@ -84,6 +86,7 @@ struct evergreen_cs_track {
 	u32			htile_surface;
 	struct radeon_bo	*htile_bo;
 	unsigned long		indirect_draw_buffer_size;
+	const unsigned		*reg_safe_bm;
 };
 
 static u32 evergreen_cs_get_aray_mode(u32 tiling_flags)
@@ -444,7 +447,7 @@ static int evergreen_cs_track_validate_cb(struct radeon_cs_parser *p, unsigned i
 		 * command stream.
 		 */
 		if (!surf.mode) {
-			volatile u32 *ib = p->ib.ptr;
+			uint32_t *ib = p->ib.ptr;
 			unsigned long tmp, nby, bsize, size, min = 0;
 
 			/* find the height the ddx wants */
@@ -1083,41 +1086,18 @@ static int evergreen_cs_parse_packet0(struct radeon_cs_parser *p,
 }
 
 /**
- * evergreen_cs_check_reg() - check if register is authorized or not
+ * evergreen_cs_handle_reg() - process registers that need special handling.
  * @parser: parser structure holding parsing context
  * @reg: register we are testing
  * @idx: index into the cs buffer
- *
- * This function will test against evergreen_reg_safe_bm and return 0
- * if register is safe. If register is not flag as safe this function
- * will test it against a list of register needind special handling.
  */
-static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
+static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 {
 	struct evergreen_cs_track *track = (struct evergreen_cs_track *)p->track;
 	struct radeon_bo_list *reloc;
-	u32 last_reg;
-	u32 m, i, tmp, *ib;
+	u32 tmp, *ib;
 	int r;
 
-	if (p->rdev->family >= CHIP_CAYMAN)
-		last_reg = ARRAY_SIZE(cayman_reg_safe_bm);
-	else
-		last_reg = ARRAY_SIZE(evergreen_reg_safe_bm);
-
-	i = (reg >> 7);
-	if (i >= last_reg) {
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
-		return -EINVAL;
-	}
-	m = 1 << ((reg >> 2) & 31);
-	if (p->rdev->family >= CHIP_CAYMAN) {
-		if (!(cayman_reg_safe_bm[i] & m))
-			return 0;
-	} else {
-		if (!(evergreen_reg_safe_bm[i] & m))
-			return 0;
-	}
 	ib = p->ib.ptr;
 	switch (reg) {
 	/* force following reg to 0 in an attempt to disable out buffer
@@ -1764,29 +1744,27 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	return 0;
 }
 
-static bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
+/**
+ * evergreen_is_safe_reg() - check if register is authorized or not
+ * @parser: parser structure holding parsing context
+ * @reg: register we are testing
+ *
+ * This function will test against reg_safe_bm and return true
+ * if register is safe or false otherwise.
+ */
+static inline bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg)
 {
-	u32 last_reg, m, i;
-
-	if (p->rdev->family >= CHIP_CAYMAN)
-		last_reg = ARRAY_SIZE(cayman_reg_safe_bm);
-	else
-		last_reg = ARRAY_SIZE(evergreen_reg_safe_bm);
+	struct evergreen_cs_track *track = p->track;
+	u32 m, i;
 
 	i = (reg >> 7);
-	if (i >= last_reg) {
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+	if (unlikely(i >= REG_SAFE_BM_SIZE)) {
 		return false;
 	}
 	m = 1 << ((reg >> 2) & 31);
-	if (p->rdev->family >= CHIP_CAYMAN) {
-		if (!(cayman_reg_safe_bm[i] & m))
-			return true;
-	} else {
-		if (!(evergreen_reg_safe_bm[i] & m))
-			return true;
-	}
-	dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+	if (!(track->reg_safe_bm[i] & m))
+		return true;
+
 	return false;
 }
 
@@ -1795,7 +1773,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 {
 	struct radeon_bo_list *reloc;
 	struct evergreen_cs_track *track;
-	volatile u32 *ib;
+	uint32_t *ib;
 	unsigned idx;
 	unsigned i;
 	unsigned start_reg, end_reg, reg;
@@ -2321,9 +2299,10 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad PACKET3_SET_CONFIG_REG\n");
 			return -EINVAL;
 		}
-		for (i = 0; i < pkt->count; i++) {
-			reg = start_reg + (4 * i);
-			r = evergreen_cs_check_reg(p, reg, idx+1+i);
+		for (reg = start_reg, idx++; reg <= end_reg; reg += 4, idx++) {
+			if (evergreen_is_safe_reg(p, reg))
+				continue;
+			r = evergreen_cs_handle_reg(p, reg, idx);
 			if (r)
 				return r;
 		}
@@ -2337,9 +2316,10 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad PACKET3_SET_CONTEXT_REG\n");
 			return -EINVAL;
 		}
-		for (i = 0; i < pkt->count; i++) {
-			reg = start_reg + (4 * i);
-			r = evergreen_cs_check_reg(p, reg, idx+1+i);
+		for (reg = start_reg, idx++; reg <= end_reg; reg += 4, idx++) {
+			if (evergreen_is_safe_reg(p, reg))
+				continue;
+			r = evergreen_cs_handle_reg(p, reg, idx);
 			if (r)
 				return r;
 		}
@@ -2594,8 +2574,11 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		} else {
 			/* SRC is a reg. */
 			reg = radeon_get_ib_value(p, idx+1) << 2;
-			if (!evergreen_is_safe_reg(p, reg, idx+1))
+			if (!evergreen_is_safe_reg(p, reg)) {
+				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
+					 reg, idx + 1);
 				return -EINVAL;
+			}
 		}
 		if (idx_value & 0x2) {
 			u64 offset;
@@ -2618,8 +2601,11 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		} else {
 			/* DST is a reg. */
 			reg = radeon_get_ib_value(p, idx+3) << 2;
-			if (!evergreen_is_safe_reg(p, reg, idx+3))
+			if (!evergreen_is_safe_reg(p, reg)) {
+				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
+					 reg, idx + 3);
 				return -EINVAL;
+			}
 		}
 		break;
 	case PACKET3_NOP:
@@ -2644,11 +2630,15 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 		if (track == NULL)
 			return -ENOMEM;
 		evergreen_cs_track_init(track);
-		if (p->rdev->family >= CHIP_CAYMAN)
+		if (p->rdev->family >= CHIP_CAYMAN) {
 			tmp = p->rdev->config.cayman.tile_config;
-		else
+			track->reg_safe_bm = cayman_reg_safe_bm;
+		} else {
 			tmp = p->rdev->config.evergreen.tile_config;
-
+			track->reg_safe_bm = evergreen_reg_safe_bm;
+		}
+		BUILD_BUG_ON(ARRAY_SIZE(cayman_reg_safe_bm) != REG_SAFE_BM_SIZE);
+		BUILD_BUG_ON(ARRAY_SIZE(evergreen_reg_safe_bm) != REG_SAFE_BM_SIZE);
 		switch (tmp & 0xf) {
 		case 0:
 			track->npipes = 1;
@@ -2757,7 +2747,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 	struct radeon_cs_chunk *ib_chunk = p->chunk_ib;
 	struct radeon_bo_list *src_reloc, *dst_reloc, *dst2_reloc;
 	u32 header, cmd, count, sub_cmd;
-	volatile u32 *ib = p->ib.ptr;
+	uint32_t *ib = p->ib.ptr;
 	u32 idx;
 	u64 src_offset, dst_offset, dst2_offset;
 	int r;
diff --git a/kernel/drivers/gpu/drm/radeon/evergreen_hdmi.c b/kernel/drivers/gpu/drm/radeon/evergreen_hdmi.c
index 9953356fe..3cf04a2f4 100644
--- a/kernel/drivers/gpu/drm/radeon/evergreen_hdmi.c
+++ b/kernel/drivers/gpu/drm/radeon/evergreen_hdmi.c
@@ -289,6 +289,16 @@ void dce4_dp_audio_set_dto(struct radeon_device *rdev,
 	 * number (coefficient of two integer numbers.  DCCG_AUDIO_DTOx_PHASE
 	 * is the numerator, DCCG_AUDIO_DTOx_MODULE is the denominator
 	 */
+	if (ASIC_IS_DCE41(rdev)) {
+		unsigned int div = (RREG32(DCE41_DENTIST_DISPCLK_CNTL) &
+			DENTIST_DPREFCLK_WDIVIDER_MASK) >>
+			DENTIST_DPREFCLK_WDIVIDER_SHIFT;
+		div = radeon_audio_decode_dfs_div(div);
+
+		if (div)
+			clock = 100 * clock / div;
+	}
+
 	WREG32(DCCG_AUDIO_DTO1_PHASE, 24000);
 	WREG32(DCCG_AUDIO_DTO1_MODULE, clock);
 }
diff --git a/kernel/drivers/gpu/drm/radeon/evergreend.h b/kernel/drivers/gpu/drm/radeon/evergreend.h
index 4aa5f7555..13b6029d6 100644
--- a/kernel/drivers/gpu/drm/radeon/evergreend.h
+++ b/kernel/drivers/gpu/drm/radeon/evergreend.h
@@ -511,6 +511,11 @@
 #define DCCG_AUDIO_DTO1_CNTL              0x05cc
 #       define DCCG_AUDIO_DTO1_USE_512FBR_DTO (1 << 3)
 
+#define DCE41_DENTIST_DISPCLK_CNTL			0x049c
+#       define DENTIST_DPREFCLK_WDIVIDER(x)		(((x) & 0x7f) << 24)
+#       define DENTIST_DPREFCLK_WDIVIDER_MASK		(0x7f << 24)
+#       define DENTIST_DPREFCLK_WDIVIDER_SHIFT		24
+
 /* DCE 4.0 AFMT */
 #define HDMI_CONTROL                         0x7030
 #       define HDMI_KEEPOUT_MODE             (1 << 0)
diff --git a/kernel/drivers/gpu/drm/radeon/ni.c b/kernel/drivers/gpu/drm/radeon/ni.c
index 64d3a7719..158872eb7 100644
--- a/kernel/drivers/gpu/drm/radeon/ni.c
+++ b/kernel/drivers/gpu/drm/radeon/ni.c
@@ -36,6 +36,31 @@
 #include "radeon_ucode.h"
 #include "clearstate_cayman.h"
 
+/*
+ * Indirect registers accessor
+ */
+u32 tn_smc_rreg(struct radeon_device *rdev, u32 reg)
+{
+	unsigned long flags;
+	u32 r;
+
+	spin_lock_irqsave(&rdev->smc_idx_lock, flags);
+	WREG32(TN_SMC_IND_INDEX_0, (reg));
+	r = RREG32(TN_SMC_IND_DATA_0);
+	spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
+	return r;
+}
+
+void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->smc_idx_lock, flags);
+	WREG32(TN_SMC_IND_INDEX_0, (reg));
+	WREG32(TN_SMC_IND_DATA_0, (v));
+	spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
+}
+
 static const u32 tn_rlc_save_restore_register_list[] =
 {
 	0x98fc,
@@ -2041,6 +2066,25 @@ static int cayman_startup(struct radeon_device *rdev)
 	if (r)
 		rdev->ring[R600_RING_TYPE_UVD_INDEX].ring_size = 0;
 
+	if (rdev->family == CHIP_ARUBA) {
+		r = radeon_vce_resume(rdev);
+		if (!r)
+			r = vce_v1_0_resume(rdev);
+
+		if (!r)
+			r = radeon_fence_driver_start_ring(rdev,
+							   TN_RING_TYPE_VCE1_INDEX);
+		if (!r)
+			r = radeon_fence_driver_start_ring(rdev,
+							   TN_RING_TYPE_VCE2_INDEX);
+
+		if (r) {
+			dev_err(rdev->dev, "VCE init error (%d).\n", r);
+			rdev->ring[TN_RING_TYPE_VCE1_INDEX].ring_size = 0;
+			rdev->ring[TN_RING_TYPE_VCE2_INDEX].ring_size = 0;
+		}
+	}
+
 	r = radeon_fence_driver_start_ring(rdev, CAYMAN_RING_TYPE_CP1_INDEX);
 	if (r) {
 		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
@@ -2118,6 +2162,21 @@ static int cayman_startup(struct radeon_device *rdev)
 			DRM_ERROR("radeon: failed initializing UVD (%d).\n", r);
 	}
 
+	if (rdev->family == CHIP_ARUBA) {
+		ring = &rdev->ring[TN_RING_TYPE_VCE1_INDEX];
+		if (ring->ring_size)
+			r = radeon_ring_init(rdev, ring, ring->ring_size, 0, 0x0);
+
+		ring = &rdev->ring[TN_RING_TYPE_VCE2_INDEX];
+		if (ring->ring_size)
+			r = radeon_ring_init(rdev, ring, ring->ring_size, 0, 0x0);
+
+		if (!r)
+			r = vce_v1_0_init(rdev);
+		if (r)
+			DRM_ERROR("radeon: failed initializing VCE (%d).\n", r);
+	}
+
 	r = radeon_ib_pool_init(rdev);
 	if (r) {
 		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
@@ -2273,6 +2332,19 @@ int cayman_init(struct radeon_device *rdev)
 		r600_ring_init(rdev, ring, 4096);
 	}
 
+	if (rdev->family == CHIP_ARUBA) {
+		r = radeon_vce_init(rdev);
+		if (!r) {
+			ring = &rdev->ring[TN_RING_TYPE_VCE1_INDEX];
+			ring->ring_obj = NULL;
+			r600_ring_init(rdev, ring, 4096);
+
+			ring = &rdev->ring[TN_RING_TYPE_VCE2_INDEX];
+			ring->ring_obj = NULL;
+			r600_ring_init(rdev, ring, 4096);
+		}
+	}
+
 	rdev->ih.ring_obj = NULL;
 	r600_ih_ring_init(rdev, 64 * 1024);
 
@@ -2326,6 +2398,8 @@ void cayman_fini(struct radeon_device *rdev)
 	radeon_irq_kms_fini(rdev);
 	uvd_v1_0_fini(rdev);
 	radeon_uvd_fini(rdev);
+	if (rdev->family == CHIP_ARUBA)
+		radeon_vce_fini(rdev);
 	cayman_pcie_gart_fini(rdev);
 	r600_vram_scratch_fini(rdev);
 	radeon_gem_fini(rdev);
@@ -2554,3 +2628,34 @@ void cayman_vm_flush(struct radeon_device *rdev, struct radeon_ring *ring,
 	radeon_ring_write(ring, PACKET3(PACKET3_PFP_SYNC_ME, 0));
 	radeon_ring_write(ring, 0x0);
 }
+
+int tn_set_vce_clocks(struct radeon_device *rdev, u32 evclk, u32 ecclk)
+{
+	struct atom_clock_dividers dividers;
+	int r, i;
+
+        r = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+					   ecclk, false, &dividers);
+	if (r)
+		return r;
+
+	for (i = 0; i < 100; i++) {
+		if (RREG32(CG_ECLK_STATUS) & ECLK_STATUS)
+			break;
+		mdelay(10);
+	}
+	if (i == 100)
+		return -ETIMEDOUT;
+
+	WREG32_P(CG_ECLK_CNTL, dividers.post_div, ~(ECLK_DIR_CNTL_EN|ECLK_DIVIDER_MASK));
+
+	for (i = 0; i < 100; i++) {
+		if (RREG32(CG_ECLK_STATUS) & ECLK_STATUS)
+			break;
+		mdelay(10);
+	}
+	if (i == 100)
+		return -ETIMEDOUT;
+
+	return 0;
+}
diff --git a/kernel/drivers/gpu/drm/radeon/nid.h b/kernel/drivers/gpu/drm/radeon/nid.h
index 3b2908389..47eb49b77 100644
--- a/kernel/drivers/gpu/drm/radeon/nid.h
+++ b/kernel/drivers/gpu/drm/radeon/nid.h
@@ -46,6 +46,13 @@
 
 #define DMIF_ADDR_CONFIG  				0xBD4
 
+/* fusion vce clocks */
+#define CG_ECLK_CNTL                                    0x620
+#       define ECLK_DIVIDER_MASK                        0x7f
+#       define ECLK_DIR_CNTL_EN                         (1 << 8)
+#define CG_ECLK_STATUS                                  0x624
+#       define ECLK_STATUS                              (1 << 0)
+
 /* DCE6 only */
 #define DMIF_ADDR_CALC  				0xC00
 
diff --git a/kernel/drivers/gpu/drm/radeon/r100.c b/kernel/drivers/gpu/drm/radeon/r100.c
index 04f2514f7..9e7e2bf03 100644
--- a/kernel/drivers/gpu/drm/radeon/r100.c
+++ b/kernel/drivers/gpu/drm/radeon/r100.c
@@ -806,7 +806,7 @@ int r100_irq_process(struct radeon_device *rdev)
 		status = r100_irq_ack(rdev);
 	}
 	if (queue_hotplug)
-		schedule_work(&rdev->hotplug_work);
+		schedule_delayed_work(&rdev->hotplug_work, 0);
 	if (rdev->msi_enabled) {
 		switch (rdev->family) {
 		case CHIP_RS400:
@@ -3217,6 +3217,9 @@ void r100_bandwidth_update(struct radeon_device *rdev)
 	uint32_t pixel_bytes1 = 0;
 	uint32_t pixel_bytes2 = 0;
 
+	/* Guess line buffer size to be 8192 pixels */
+	u32 lb_size = 8192;
+
 	if (!rdev->mode_info.mode_config_initialized)
 		return;
 
@@ -3631,6 +3634,13 @@ void r100_bandwidth_update(struct radeon_device *rdev)
 		DRM_DEBUG_KMS("GRPH2_BUFFER_CNTL from to %x\n",
 			  (unsigned int)RREG32(RADEON_GRPH2_BUFFER_CNTL));
 	}
+
+	/* Save number of lines the linebuffer leads before the scanout */
+	if (mode1)
+	    rdev->mode_info.crtcs[0]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode1->crtc_hdisplay);
+
+	if (mode2)
+	    rdev->mode_info.crtcs[1]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode2->crtc_hdisplay);
 }
 
 int r100_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
@@ -4090,6 +4100,28 @@ int r100_init(struct radeon_device *rdev)
 	return 0;
 }
 
+uint32_t r100_mm_rreg_slow(struct radeon_device *rdev, uint32_t reg)
+{
+	unsigned long flags;
+	uint32_t ret;
+
+	spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
+	writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
+	ret = readl(((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
+	spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
+	return ret;
+}
+
+void r100_mm_wreg_slow(struct radeon_device *rdev, uint32_t reg, uint32_t v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
+	writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
+	writel(v, ((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
+	spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
+}
+
 u32 r100_io_rreg(struct radeon_device *rdev, u32 reg)
 {
 	if (reg < rdev->rio_mem_size)
diff --git a/kernel/drivers/gpu/drm/radeon/r300.c b/kernel/drivers/gpu/drm/radeon/r300.c
index 08d68f3e1..718b12b03 100644
--- a/kernel/drivers/gpu/drm/radeon/r300.c
+++ b/kernel/drivers/gpu/drm/radeon/r300.c
@@ -50,6 +50,31 @@
  */
 
 /*
+ * Indirect registers accessor
+ */
+uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg)
+{
+	unsigned long flags;
+	uint32_t r;
+
+	spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
+	WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
+	r = RREG32(RADEON_PCIE_DATA);
+	spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
+	return r;
+}
+
+void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
+	WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
+	WREG32(RADEON_PCIE_DATA, (v));
+	spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
+}
+
+/*
  * rv370,rv380 PCIE GART
  */
 static int rv370_debugfs_pcie_gart_info_init(struct radeon_device *rdev);
diff --git a/kernel/drivers/gpu/drm/radeon/r600.c b/kernel/drivers/gpu/drm/radeon/r600.c
index 21e479fef..cc2fdf0be 100644
--- a/kernel/drivers/gpu/drm/radeon/r600.c
+++ b/kernel/drivers/gpu/drm/radeon/r600.c
@@ -108,6 +108,53 @@ static void r600_pcie_gen2_enable(struct radeon_device *rdev);
 extern int evergreen_rlc_resume(struct radeon_device *rdev);
 extern void rv770_set_clk_bypass_mode(struct radeon_device *rdev);
 
+/*
+ * Indirect registers accessor
+ */
+u32 r600_rcu_rreg(struct radeon_device *rdev, u32 reg)
+{
+	unsigned long flags;
+	u32 r;
+
+	spin_lock_irqsave(&rdev->rcu_idx_lock, flags);
+	WREG32(R600_RCU_INDEX, ((reg) & 0x1fff));
+	r = RREG32(R600_RCU_DATA);
+	spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags);
+	return r;
+}
+
+void r600_rcu_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->rcu_idx_lock, flags);
+	WREG32(R600_RCU_INDEX, ((reg) & 0x1fff));
+	WREG32(R600_RCU_DATA, (v));
+	spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags);
+}
+
+u32 r600_uvd_ctx_rreg(struct radeon_device *rdev, u32 reg)
+{
+	unsigned long flags;
+	u32 r;
+
+	spin_lock_irqsave(&rdev->uvd_idx_lock, flags);
+	WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff));
+	r = RREG32(R600_UVD_CTX_DATA);
+	spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags);
+	return r;
+}
+
+void r600_uvd_ctx_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rdev->uvd_idx_lock, flags);
+	WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff));
+	WREG32(R600_UVD_CTX_DATA, (v));
+	spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags);
+}
+
 /**
  * r600_get_allowed_info_register - fetch the register for the info ioctl
  *
@@ -4229,7 +4276,7 @@ restart_ih:
 		WREG32(IH_RB_RPTR, rptr);
 	}
 	if (queue_hotplug)
-		schedule_work(&rdev->hotplug_work);
+		schedule_delayed_work(&rdev->hotplug_work, 0);
 	if (queue_hdmi)
 		schedule_work(&rdev->audio_work);
 	if (queue_thermal && rdev->pm.dpm_enabled)
diff --git a/kernel/drivers/gpu/drm/radeon/r600_blit_shaders.c b/kernel/drivers/gpu/drm/radeon/r600_blit_shaders.c
index 34c8b2340..443cbe59b 100644
--- a/kernel/drivers/gpu/drm/radeon/r600_blit_shaders.c
+++ b/kernel/drivers/gpu/drm/radeon/r600_blit_shaders.c
@@ -32,7 +32,7 @@
  * R6xx+ cards need to use the 3D engine to blit data which requires
  * quite a bit of hw state setup.  Rather than pull the whole 3D driver
  * (which normally generates the 3D state) into the DRM, we opt to use
- * statically generated state tables.  The regsiter state and shaders
+ * statically generated state tables.  The register state and shaders
  * were hand generated to support blitting functionality.  See the 3D
  * driver or documentation for descriptions of the registers and
  * shader instructions.
diff --git a/kernel/drivers/gpu/drm/radeon/r600_cp.c b/kernel/drivers/gpu/drm/radeon/r600_cp.c
index 09e3f3992..e231eeafe 100644
--- a/kernel/drivers/gpu/drm/radeon/r600_cp.c
+++ b/kernel/drivers/gpu/drm/radeon/r600_cp.c
@@ -1837,7 +1837,7 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
 	SET_RING_HEAD(dev_priv, 0);
 	dev_priv->ring.tail = 0;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		rptr_addr = dev_priv->ring_rptr->offset
 			- dev->agp->base +
@@ -1863,7 +1863,7 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
 		     dev_priv->ring.size_l2qw);
 #endif
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* XXX */
 		radeon_write_agp_base(dev_priv, dev->agp->base);
@@ -1946,7 +1946,7 @@ int r600_do_cleanup_cp(struct drm_device *dev)
 	if (dev->irq_enabled)
 		drm_irq_uninstall(dev);
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		if (dev_priv->cp_ring != NULL) {
 			drm_legacy_ioremapfree(dev_priv->cp_ring, dev);
@@ -2089,7 +2089,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		}
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	/* XXX */
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		drm_legacy_ioremap_wc(dev_priv->cp_ring, dev);
@@ -2148,7 +2148,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		 * location in the card and on the bus, though we have to
 		 * align it down.
 		 */
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		/* XXX */
 		if (dev_priv->flags & RADEON_IS_AGP) {
 			base = dev->agp->base;
@@ -2175,7 +2175,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 				 base, dev_priv->gart_vm_start);
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	/* XXX */
 	if (dev_priv->flags & RADEON_IS_AGP)
 		dev_priv->gart_buffers_offset = (dev->agp_buffer_map->offset
@@ -2212,7 +2212,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 
 	dev_priv->ring.high_mark = RADEON_RING_HIGH_MARK;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* XXX turn off pcie gart */
 	} else
@@ -2483,7 +2483,7 @@ int r600_cp_dispatch_texture(struct drm_device *dev,
 	struct drm_buf *buf;
 	u32 *buffer;
 	const u8 __user *data;
-	int size, pass_size;
+	unsigned int size, pass_size;
 	u64 src_offset, dst_offset;
 
 	if (!radeon_check_offset(dev_priv, tex->offset)) {
diff --git a/kernel/drivers/gpu/drm/radeon/radeon.h b/kernel/drivers/gpu/drm/radeon/radeon.h
index 46eb0fa75..558056808 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon.h
+++ b/kernel/drivers/gpu/drm/radeon/radeon.h
@@ -268,6 +268,7 @@ struct radeon_clock {
 	uint32_t current_dispclk;
 	uint32_t dp_extclk;
 	uint32_t max_pixel_clock;
+	uint32_t vco_freq;
 };
 
 /*
@@ -467,7 +468,6 @@ struct radeon_bo_va {
 	/* protected by bo being reserved */
 	struct list_head		bo_list;
 	uint32_t			flags;
-	uint64_t			addr;
 	struct radeon_fence		*last_pt_update;
 	unsigned			ref_count;
 
@@ -719,7 +719,7 @@ struct radeon_doorbell {
 	resource_size_t		size;
 	u32 __iomem		*ptr;
 	u32			num_doorbells;	/* Number of doorbells actually reserved for radeon. */
-	unsigned long		used[DIV_ROUND_UP(RADEON_MAX_DOORBELLS, BITS_PER_LONG)];
+	DECLARE_BITMAP(used, RADEON_MAX_DOORBELLS);
 };
 
 int radeon_doorbell_get(struct radeon_device *rdev, u32 *page);
@@ -941,6 +941,9 @@ struct radeon_vm {
 	/* BOs freed, but not yet updated in the PT */
 	struct list_head	freed;
 
+	/* BOs cleared in the PT */
+	struct list_head	cleared;
+
 	/* contains the page directory */
 	struct radeon_bo	*page_directory;
 	unsigned		max_pde_used;
@@ -1656,6 +1659,7 @@ struct radeon_pm {
 	u8                      fan_max_rpm;
 	/* dpm */
 	bool                    dpm_enabled;
+	bool                    sysfs_initialized;
 	struct radeon_dpm       dpm;
 };
 
@@ -1709,8 +1713,6 @@ int radeon_uvd_send_upll_ctlreq(struct radeon_device *rdev,
  * VCE
  */
 #define RADEON_MAX_VCE_HANDLES	16
-#define RADEON_VCE_STACK_SIZE	(1024*1024)
-#define RADEON_VCE_HEAP_SIZE	(4*1024*1024)
 
 struct radeon_vce {
 	struct radeon_bo	*vcpu_bo;
@@ -1721,6 +1723,7 @@ struct radeon_vce {
 	struct drm_file		*filp[RADEON_MAX_VCE_HANDLES];
 	unsigned		img_size[RADEON_MAX_VCE_HANDLES];
 	struct delayed_work	idle_work;
+	uint32_t		keyselect;
 };
 
 int radeon_vce_init(struct radeon_device *rdev);
@@ -2412,7 +2415,7 @@ struct radeon_device {
 	struct r600_ih ih; /* r6/700 interrupt ring */
 	struct radeon_rlc rlc;
 	struct radeon_mec mec;
-	struct work_struct hotplug_work;
+	struct delayed_work hotplug_work;
 	struct work_struct dp_work;
 	struct work_struct audio_work;
 	int num_crtc; /* number of crtcs */
@@ -2435,6 +2438,7 @@ struct radeon_device {
 	atomic64_t			vram_usage;
 	atomic64_t			gtt_usage;
 	atomic64_t			num_bytes_moved;
+	atomic_t			gpu_reset_counter;
 	/* ACPI interface */
 	struct radeon_atif		atif;
 	struct radeon_atcs		atcs;
@@ -2456,7 +2460,6 @@ struct radeon_device {
 
 	/* amdkfd interface */
 	struct kfd_dev		*kfd;
-	struct radeon_sa_manager	kfd_bo;
 
 	struct mutex	mn_lock;
 	DECLARE_HASHTABLE(mn_hash, 7);
@@ -2472,38 +2475,24 @@ int radeon_gpu_wait_for_idle(struct radeon_device *rdev);
 
 #define RADEON_MIN_MMIO_SIZE 0x10000
 
+uint32_t r100_mm_rreg_slow(struct radeon_device *rdev, uint32_t reg);
+void r100_mm_wreg_slow(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 static inline uint32_t r100_mm_rreg(struct radeon_device *rdev, uint32_t reg,
 				    bool always_indirect)
 {
 	/* The mmio size is 64kb at minimum. Allows the if to be optimized out. */
 	if ((reg < rdev->rmmio_size || reg < RADEON_MIN_MMIO_SIZE) && !always_indirect)
 		return readl(((void __iomem *)rdev->rmmio) + reg);
-	else {
-		unsigned long flags;
-		uint32_t ret;
-
-		spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
-		writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
-		ret = readl(((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
-		spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
-
-		return ret;
-	}
+	else
+		return r100_mm_rreg_slow(rdev, reg);
 }
-
 static inline void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v,
 				bool always_indirect)
 {
 	if ((reg < rdev->rmmio_size || reg < RADEON_MIN_MMIO_SIZE) && !always_indirect)
 		writel(v, ((void __iomem *)rdev->rmmio) + reg);
-	else {
-		unsigned long flags;
-
-		spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
-		writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
-		writel(v, ((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
-		spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
-	}
+	else
+		r100_mm_wreg_slow(rdev, reg, v);
 }
 
 u32 r100_io_rreg(struct radeon_device *rdev, u32 reg);
@@ -2579,6 +2568,13 @@ static inline struct radeon_fence *to_radeon_fence(struct fence *f)
 		tmp_ |= ((val) & ~(mask));			\
 		WREG32_PLL(reg, tmp_);				\
 	} while (0)
+#define WREG32_SMC_P(reg, val, mask)				\
+	do {							\
+		uint32_t tmp_ = RREG32_SMC(reg);		\
+		tmp_ &= (mask);					\
+		tmp_ |= ((val) & ~(mask));			\
+		WREG32_SMC(reg, tmp_);				\
+	} while (0)
 #define DREG32_SYS(sqf, rdev, reg) seq_printf((sqf), #reg " : 0x%08X\n", r100_mm_rreg((rdev), (reg), false))
 #define RREG32_IO(reg) r100_io_rreg(rdev, (reg))
 #define WREG32_IO(reg, v) r100_io_wreg(rdev, (reg), (v))
@@ -2587,184 +2583,29 @@ static inline struct radeon_fence *to_radeon_fence(struct fence *f)
 #define WDOORBELL32(index, v) cik_mm_wdoorbell(rdev, (index), (v))
 
 /*
- * Indirect registers accessor
+ * Indirect registers accessors.
+ * They used to be inlined, but this increases code size by ~65 kbytes.
+ * Since each performs a pair of MMIO ops
+ * within a spin_lock_irqsave/spin_unlock_irqrestore region,
+ * the cost of call+ret is almost negligible. MMIO and locking
+ * costs several dozens of cycles each at best, call+ret is ~5 cycles.
  */
-static inline uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg)
-{
-	unsigned long flags;
-	uint32_t r;
-
-	spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
-	WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
-	r = RREG32(RADEON_PCIE_DATA);
-	spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
-	return r;
-}
-
-static inline void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
-	WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
-	WREG32(RADEON_PCIE_DATA, (v));
-	spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
-}
-
-static inline u32 tn_smc_rreg(struct radeon_device *rdev, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&rdev->smc_idx_lock, flags);
-	WREG32(TN_SMC_IND_INDEX_0, (reg));
-	r = RREG32(TN_SMC_IND_DATA_0);
-	spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
-	return r;
-}
-
-static inline void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->smc_idx_lock, flags);
-	WREG32(TN_SMC_IND_INDEX_0, (reg));
-	WREG32(TN_SMC_IND_DATA_0, (v));
-	spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
-}
-
-static inline u32 r600_rcu_rreg(struct radeon_device *rdev, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&rdev->rcu_idx_lock, flags);
-	WREG32(R600_RCU_INDEX, ((reg) & 0x1fff));
-	r = RREG32(R600_RCU_DATA);
-	spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags);
-	return r;
-}
-
-static inline void r600_rcu_wreg(struct radeon_device *rdev, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->rcu_idx_lock, flags);
-	WREG32(R600_RCU_INDEX, ((reg) & 0x1fff));
-	WREG32(R600_RCU_DATA, (v));
-	spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags);
-}
-
-static inline u32 eg_cg_rreg(struct radeon_device *rdev, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&rdev->cg_idx_lock, flags);
-	WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff));
-	r = RREG32(EVERGREEN_CG_IND_DATA);
-	spin_unlock_irqrestore(&rdev->cg_idx_lock, flags);
-	return r;
-}
-
-static inline void eg_cg_wreg(struct radeon_device *rdev, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->cg_idx_lock, flags);
-	WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff));
-	WREG32(EVERGREEN_CG_IND_DATA, (v));
-	spin_unlock_irqrestore(&rdev->cg_idx_lock, flags);
-}
-
-static inline u32 eg_pif_phy0_rreg(struct radeon_device *rdev, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
-	WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff));
-	r = RREG32(EVERGREEN_PIF_PHY0_DATA);
-	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
-	return r;
-}
-
-static inline void eg_pif_phy0_wreg(struct radeon_device *rdev, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
-	WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff));
-	WREG32(EVERGREEN_PIF_PHY0_DATA, (v));
-	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
-}
-
-static inline u32 eg_pif_phy1_rreg(struct radeon_device *rdev, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
-	WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff));
-	r = RREG32(EVERGREEN_PIF_PHY1_DATA);
-	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
-	return r;
-}
-
-static inline void eg_pif_phy1_wreg(struct radeon_device *rdev, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->pif_idx_lock, flags);
-	WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff));
-	WREG32(EVERGREEN_PIF_PHY1_DATA, (v));
-	spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
-}
-
-static inline u32 r600_uvd_ctx_rreg(struct radeon_device *rdev, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&rdev->uvd_idx_lock, flags);
-	WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff));
-	r = RREG32(R600_UVD_CTX_DATA);
-	spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags);
-	return r;
-}
-
-static inline void r600_uvd_ctx_wreg(struct radeon_device *rdev, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->uvd_idx_lock, flags);
-	WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff));
-	WREG32(R600_UVD_CTX_DATA, (v));
-	spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags);
-}
-
-
-static inline u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&rdev->didt_idx_lock, flags);
-	WREG32(CIK_DIDT_IND_INDEX, (reg));
-	r = RREG32(CIK_DIDT_IND_DATA);
-	spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
-	return r;
-}
-
-static inline void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rdev->didt_idx_lock, flags);
-	WREG32(CIK_DIDT_IND_INDEX, (reg));
-	WREG32(CIK_DIDT_IND_DATA, (v));
-	spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
-}
+uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg);
+void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
+u32 tn_smc_rreg(struct radeon_device *rdev, u32 reg);
+void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v);
+u32 r600_rcu_rreg(struct radeon_device *rdev, u32 reg);
+void r600_rcu_wreg(struct radeon_device *rdev, u32 reg, u32 v);
+u32 eg_cg_rreg(struct radeon_device *rdev, u32 reg);
+void eg_cg_wreg(struct radeon_device *rdev, u32 reg, u32 v);
+u32 eg_pif_phy0_rreg(struct radeon_device *rdev, u32 reg);
+void eg_pif_phy0_wreg(struct radeon_device *rdev, u32 reg, u32 v);
+u32 eg_pif_phy1_rreg(struct radeon_device *rdev, u32 reg);
+void eg_pif_phy1_wreg(struct radeon_device *rdev, u32 reg, u32 v);
+u32 r600_uvd_ctx_rreg(struct radeon_device *rdev, u32 reg);
+void r600_uvd_ctx_wreg(struct radeon_device *rdev, u32 reg, u32 v);
+u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg);
+void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v);
 
 void r100_pll_errata_after_index(struct radeon_device *rdev);
 
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_acpi.c b/kernel/drivers/gpu/drm/radeon/radeon_acpi.c
index 77e9d07c5..59acd0e5c 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_acpi.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_acpi.c
@@ -25,7 +25,6 @@
 #include <linux/acpi.h>
 #include <linux/slab.h>
 #include <linux/power_supply.h>
-#include <linux/vga_switcheroo.h>
 #include <acpi/video.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_agp.c b/kernel/drivers/gpu/drm/radeon/radeon_agp.c
index a9297b2c3..c77d349c5 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_agp.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_agp.c
@@ -28,7 +28,7 @@
 #include "radeon.h"
 #include <drm/radeon_drm.h>
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 
 struct radeon_agpmode_quirk {
 	u32 hostbridge_vendor;
@@ -54,6 +54,9 @@ static struct radeon_agpmode_quirk radeon_agpmode_quirk_list[] = {
 	/* Intel 82855PM host bridge / Mobility 9600 M10 RV350 Needs AGPMode 1 (lp #195051) */
 	{ PCI_VENDOR_ID_INTEL, 0x3340, PCI_VENDOR_ID_ATI, 0x4e50,
 		PCI_VENDOR_ID_IBM, 0x0550, 1},
+	/* Intel 82855PM host bridge / RV250/M9 GL [Mobility FireGL 9000/Radeon 9000] needs AGPMode 1 (Thinkpad T40p) */
+	{ PCI_VENDOR_ID_INTEL, 0x3340, PCI_VENDOR_ID_ATI, 0x4c66,
+		PCI_VENDOR_ID_IBM, 0x054d, 1},
 	/* Intel 82855PM host bridge / Mobility M7 needs AGPMode 1 */
 	{ PCI_VENDOR_ID_INTEL, 0x3340, PCI_VENDOR_ID_ATI, 0x4c57,
 		PCI_VENDOR_ID_IBM, 0x0530, 1},
@@ -123,7 +126,7 @@ static struct radeon_agpmode_quirk radeon_agpmode_quirk_list[] = {
 
 int radeon_agp_init(struct radeon_device *rdev)
 {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	struct radeon_agpmode_quirk *p = radeon_agpmode_quirk_list;
 	struct drm_agp_mode mode;
 	struct drm_agp_info info;
@@ -257,7 +260,7 @@ int radeon_agp_init(struct radeon_device *rdev)
 
 void radeon_agp_resume(struct radeon_device *rdev)
 {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	int r;
 	if (rdev->flags & RADEON_IS_AGP) {
 		r = radeon_agp_init(rdev);
@@ -269,7 +272,7 @@ void radeon_agp_resume(struct radeon_device *rdev)
 
 void radeon_agp_fini(struct radeon_device *rdev)
 {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->ddev->agp && rdev->ddev->agp->acquired) {
 		drm_agp_release(rdev->ddev);
 	}
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_asic.c b/kernel/drivers/gpu/drm/radeon/radeon_asic.c
index 8dbf5083c..1d4d4520a 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_asic.c
@@ -31,7 +31,6 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/radeon_drm.h>
 #include <linux/vgaarb.h>
-#include <linux/vga_switcheroo.h>
 #include "radeon_reg.h"
 #include "radeon.h"
 #include "radeon_asic.h"
@@ -1761,6 +1760,19 @@ static struct radeon_asic cayman_asic = {
 	},
 };
 
+static struct radeon_asic_ring trinity_vce_ring = {
+	.ib_execute = &radeon_vce_ib_execute,
+	.emit_fence = &radeon_vce_fence_emit,
+	.emit_semaphore = &radeon_vce_semaphore_emit,
+	.cs_parse = &radeon_vce_cs_parse,
+	.ring_test = &radeon_vce_ring_test,
+	.ib_test = &radeon_vce_ib_test,
+	.is_lockup = &radeon_ring_test_lockup,
+	.get_rptr = &vce_v1_0_get_rptr,
+	.get_wptr = &vce_v1_0_get_wptr,
+	.set_wptr = &vce_v1_0_set_wptr,
+};
+
 static struct radeon_asic trinity_asic = {
 	.init = &cayman_init,
 	.fini = &cayman_fini,
@@ -1794,6 +1806,8 @@ static struct radeon_asic trinity_asic = {
 		[R600_RING_TYPE_DMA_INDEX] = &cayman_dma_ring,
 		[CAYMAN_RING_TYPE_DMA1_INDEX] = &cayman_dma_ring,
 		[R600_RING_TYPE_UVD_INDEX] = &cayman_uvd_ring,
+		[TN_RING_TYPE_VCE1_INDEX] = &trinity_vce_ring,
+		[TN_RING_TYPE_VCE2_INDEX] = &trinity_vce_ring,
 	},
 	.irq = {
 		.set = &evergreen_irq_set,
@@ -1838,6 +1852,7 @@ static struct radeon_asic trinity_asic = {
 		.set_pcie_lanes = NULL,
 		.set_clock_gating = NULL,
 		.set_uvd_clocks = &sumo_set_uvd_clocks,
+		.set_vce_clocks = &tn_set_vce_clocks,
 		.get_temperature = &tn_get_temp,
 	},
 	.dpm = {
@@ -1929,6 +1944,8 @@ static struct radeon_asic si_asic = {
 		[R600_RING_TYPE_DMA_INDEX] = &si_dma_ring,
 		[CAYMAN_RING_TYPE_DMA1_INDEX] = &si_dma_ring,
 		[R600_RING_TYPE_UVD_INDEX] = &cayman_uvd_ring,
+		[TN_RING_TYPE_VCE1_INDEX] = &trinity_vce_ring,
+		[TN_RING_TYPE_VCE2_INDEX] = &trinity_vce_ring,
 	},
 	.irq = {
 		.set = &si_irq_set,
@@ -1973,6 +1990,7 @@ static struct radeon_asic si_asic = {
 		.set_pcie_lanes = &r600_set_pcie_lanes,
 		.set_clock_gating = NULL,
 		.set_uvd_clocks = &si_set_uvd_clocks,
+		.set_vce_clocks = &si_set_vce_clocks,
 		.get_temperature = &si_get_temp,
 	},
 	.dpm = {
@@ -2436,6 +2454,8 @@ int radeon_asic_init(struct radeon_device *rdev)
 		/* set num crtcs */
 		rdev->num_crtc = 4;
 		rdev->has_uvd = true;
+		rdev->cg_flags =
+			RADEON_CG_SUPPORT_VCE_MGCG;
 		break;
 	case CHIP_TAHITI:
 	case CHIP_PITCAIRN:
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_asic.h b/kernel/drivers/gpu/drm/radeon/radeon_asic.h
index a3ca8cd30..e0aa33262 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/kernel/drivers/gpu/drm/radeon/radeon_asic.h
@@ -694,6 +694,7 @@ int trinity_dpm_force_performance_level(struct radeon_device *rdev,
 void trinity_dpm_enable_bapm(struct radeon_device *rdev, bool enable);
 u32 trinity_dpm_get_current_sclk(struct radeon_device *rdev);
 u32 trinity_dpm_get_current_mclk(struct radeon_device *rdev);
+int tn_set_vce_clocks(struct radeon_device *rdev, u32 evclk, u32 ecclk);
 
 /* DCE6 - SI */
 void dce6_bandwidth_update(struct radeon_device *rdev);
@@ -745,6 +746,7 @@ void si_dma_vm_flush(struct radeon_device *rdev, struct radeon_ring *ring,
 u32 si_get_xclk(struct radeon_device *rdev);
 uint64_t si_get_gpu_clock_counter(struct radeon_device *rdev);
 int si_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
+int si_set_vce_clocks(struct radeon_device *rdev, u32 evclk, u32 ecclk);
 int si_get_temp(struct radeon_device *rdev);
 int si_get_allowed_info_register(struct radeon_device *rdev,
 				 u32 reg, u32 *val);
@@ -970,10 +972,14 @@ uint32_t vce_v1_0_get_wptr(struct radeon_device *rdev,
 			   struct radeon_ring *ring);
 void vce_v1_0_set_wptr(struct radeon_device *rdev,
 		       struct radeon_ring *ring);
+int vce_v1_0_load_fw(struct radeon_device *rdev, uint32_t *data);
+unsigned vce_v1_0_bo_size(struct radeon_device *rdev);
+int vce_v1_0_resume(struct radeon_device *rdev);
 int vce_v1_0_init(struct radeon_device *rdev);
 int vce_v1_0_start(struct radeon_device *rdev);
 
 /* vce v2.0 */
+unsigned vce_v2_0_bo_size(struct radeon_device *rdev);
 int vce_v2_0_resume(struct radeon_device *rdev);
 
 #endif
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_atombios.c b/kernel/drivers/gpu/drm/radeon/radeon_atombios.c
index 8f285244c..de9a2ffcf 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -437,7 +437,9 @@ static bool radeon_atom_apply_quirks(struct drm_device *dev,
 	}
 
 	/* Fujitsu D3003-S2 board lists DVI-I as DVI-D and VGA */
-	if (((dev->pdev->device == 0x9802) || (dev->pdev->device == 0x9806)) &&
+	if (((dev->pdev->device == 0x9802) ||
+	     (dev->pdev->device == 0x9805) ||
+	     (dev->pdev->device == 0x9806)) &&
 	    (dev->pdev->subsystem_vendor == 0x1734) &&
 	    (dev->pdev->subsystem_device == 0x11bd)) {
 		if (*connector_type == DRM_MODE_CONNECTOR_VGA) {
@@ -448,14 +450,6 @@ static bool radeon_atom_apply_quirks(struct drm_device *dev,
 		}
 	}
 
-	/* Fujitsu D3003-S2 board lists DVI-I as DVI-I and VGA */
-	if ((dev->pdev->device == 0x9805) &&
-	    (dev->pdev->subsystem_vendor == 0x1734) &&
-	    (dev->pdev->subsystem_device == 0x11bd)) {
-		if (*connector_type == DRM_MODE_CONNECTOR_VGA)
-			return false;
-	}
-
 	return true;
 }
 
@@ -1112,6 +1106,31 @@ union firmware_info {
 	ATOM_FIRMWARE_INFO_V2_2 info_22;
 };
 
+union igp_info {
+	struct _ATOM_INTEGRATED_SYSTEM_INFO info;
+	struct _ATOM_INTEGRATED_SYSTEM_INFO_V2 info_2;
+	struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 info_6;
+	struct _ATOM_INTEGRATED_SYSTEM_INFO_V1_7 info_7;
+	struct _ATOM_INTEGRATED_SYSTEM_INFO_V1_8 info_8;
+};
+
+static void radeon_atombios_get_dentist_vco_freq(struct radeon_device *rdev)
+{
+	struct radeon_mode_info *mode_info = &rdev->mode_info;
+	int index = GetIndexIntoMasterTable(DATA, IntegratedSystemInfo);
+	union igp_info *igp_info;
+	u8 frev, crev;
+	u16 data_offset;
+
+	if (atom_parse_data_header(mode_info->atom_context, index, NULL,
+			&frev, &crev, &data_offset)) {
+		igp_info = (union igp_info *)(mode_info->atom_context->bios +
+			data_offset);
+		rdev->clock.vco_freq =
+			le32_to_cpu(igp_info->info_6.ulDentistVCOFreq);
+	}
+}
+
 bool radeon_atom_get_clock_info(struct drm_device *dev)
 {
 	struct radeon_device *rdev = dev->dev_private;
@@ -1263,20 +1282,25 @@ bool radeon_atom_get_clock_info(struct drm_device *dev)
 		rdev->mode_info.firmware_flags =
 			le16_to_cpu(firmware_info->info.usFirmwareCapability.susAccess);
 
+		if (ASIC_IS_DCE8(rdev))
+			rdev->clock.vco_freq =
+				le32_to_cpu(firmware_info->info_22.ulGPUPLL_OutputFreq);
+		else if (ASIC_IS_DCE5(rdev))
+			rdev->clock.vco_freq = rdev->clock.current_dispclk;
+		else if (ASIC_IS_DCE41(rdev))
+			radeon_atombios_get_dentist_vco_freq(rdev);
+		else
+			rdev->clock.vco_freq = rdev->clock.current_dispclk;
+
+		if (rdev->clock.vco_freq == 0)
+			rdev->clock.vco_freq = 360000;	/* 3.6 GHz */
+
 		return true;
 	}
 
 	return false;
 }
 
-union igp_info {
-	struct _ATOM_INTEGRATED_SYSTEM_INFO info;
-	struct _ATOM_INTEGRATED_SYSTEM_INFO_V2 info_2;
-	struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 info_6;
-	struct _ATOM_INTEGRATED_SYSTEM_INFO_V1_7 info_7;
-	struct _ATOM_INTEGRATED_SYSTEM_INFO_V1_8 info_8;
-};
-
 bool radeon_atombios_sideport_present(struct radeon_device *rdev)
 {
 	struct radeon_mode_info *mode_info = &rdev->mode_info;
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/kernel/drivers/gpu/drm/radeon/radeon_atpx_handler.c
index 8bc7d0bbd..c4b4f298a 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_atpx_handler.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_atpx_handler.c
@@ -499,7 +499,7 @@ static int radeon_atpx_get_client_id(struct pci_dev *pdev)
 		return VGA_SWITCHEROO_DIS;
 }
 
-static struct vga_switcheroo_handler radeon_atpx_handler = {
+static const struct vga_switcheroo_handler radeon_atpx_handler = {
 	.switchto = radeon_atpx_switchto,
 	.power_state = radeon_atpx_power_state,
 	.init = radeon_atpx_init,
@@ -535,7 +535,7 @@ static bool radeon_atpx_detect(void)
 
 	if (has_atpx && vga_count == 2) {
 		acpi_get_name(radeon_atpx_priv.atpx.handle, ACPI_FULL_PATHNAME, &buffer);
-		printk(KERN_INFO "VGA switcheroo: detected switching method %s handle\n",
+		printk(KERN_INFO "vga_switcheroo: detected switching method %s handle\n",
 		       acpi_method_name);
 		radeon_atpx_priv.atpx_detected = true;
 		return true;
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_audio.c b/kernel/drivers/gpu/drm/radeon/radeon_audio.c
index d77dd1430..b214663b3 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_audio.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_audio.c
@@ -698,26 +698,37 @@ static void radeon_audio_hdmi_mode_set(struct drm_encoder *encoder,
 {
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 
 	if (!dig || !dig->afmt)
 		return;
 
-	radeon_audio_set_mute(encoder, true);
+	if (!connector)
+		return;
+
+	if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
+		radeon_audio_set_mute(encoder, true);
 
-	radeon_audio_write_speaker_allocation(encoder);
-	radeon_audio_write_sad_regs(encoder);
-	radeon_audio_write_latency_fields(encoder, mode);
-	radeon_audio_set_dto(encoder, mode->clock);
-	radeon_audio_set_vbi_packet(encoder);
-	radeon_hdmi_set_color_depth(encoder);
-	radeon_audio_update_acr(encoder, mode->clock);
-	radeon_audio_set_audio_packet(encoder);
-	radeon_audio_select_pin(encoder);
+		radeon_audio_write_speaker_allocation(encoder);
+		radeon_audio_write_sad_regs(encoder);
+		radeon_audio_write_latency_fields(encoder, mode);
+		radeon_audio_set_dto(encoder, mode->clock);
+		radeon_audio_set_vbi_packet(encoder);
+		radeon_hdmi_set_color_depth(encoder);
+		radeon_audio_update_acr(encoder, mode->clock);
+		radeon_audio_set_audio_packet(encoder);
+		radeon_audio_select_pin(encoder);
 
-	if (radeon_audio_set_avi_packet(encoder, mode) < 0)
-		return;
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
 
-	radeon_audio_set_mute(encoder, false);
+		radeon_audio_set_mute(encoder, false);
+	} else {
+		radeon_hdmi_set_color_depth(encoder);
+
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
+	}
 }
 
 static void radeon_audio_dp_mode_set(struct drm_encoder *encoder,
@@ -728,28 +739,24 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder,
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
 	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
-	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
-	struct radeon_connector_atom_dig *dig_connector =
-		radeon_connector->con_priv;
 
-	if (!connector)
+	if (!dig || !dig->afmt)
 		return;
 
-	if (!dig || !dig->afmt)
+	if (!connector)
 		return;
 
-	radeon_audio_write_speaker_allocation(encoder);
-	radeon_audio_write_sad_regs(encoder);
-	radeon_audio_write_latency_fields(encoder, mode);
-	if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev))
-		radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10);
-	else
-		radeon_audio_set_dto(encoder, dig_connector->dp_clock);
-	radeon_audio_set_audio_packet(encoder);
-	radeon_audio_select_pin(encoder);
+	if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
+		radeon_audio_write_speaker_allocation(encoder);
+		radeon_audio_write_sad_regs(encoder);
+		radeon_audio_write_latency_fields(encoder, mode);
+		radeon_audio_set_dto(encoder, rdev->clock.vco_freq * 10);
+		radeon_audio_set_audio_packet(encoder);
+		radeon_audio_select_pin(encoder);
 
-	if (radeon_audio_set_avi_packet(encoder, mode) < 0)
-		return;
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
+	}
 }
 
 void radeon_audio_mode_set(struct drm_encoder *encoder,
@@ -768,3 +775,15 @@ void radeon_audio_dpms(struct drm_encoder *encoder, int mode)
 	if (radeon_encoder->audio && radeon_encoder->audio->dpms)
 		radeon_encoder->audio->dpms(encoder, mode == DRM_MODE_DPMS_ON);
 }
+
+unsigned int radeon_audio_decode_dfs_div(unsigned int div)
+{
+	if (div >= 8 && div < 64)
+		return (div - 8) * 25 + 200;
+	else if (div >= 64 && div < 96)
+		return (div - 64) * 50 + 1600;
+	else if (div >= 96 && div < 128)
+		return (div - 96) * 100 + 3200;
+	else
+		return 0;
+}
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_audio.h b/kernel/drivers/gpu/drm/radeon/radeon_audio.h
index 059cc3012..5c70cceaa 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_audio.h
+++ b/kernel/drivers/gpu/drm/radeon/radeon_audio.h
@@ -79,5 +79,6 @@ void radeon_audio_fini(struct radeon_device *rdev);
 void radeon_audio_mode_set(struct drm_encoder *encoder,
 	struct drm_display_mode *mode);
 void radeon_audio_dpms(struct drm_encoder *encoder, int mode);
+unsigned int radeon_audio_decode_dfs_div(unsigned int div);
 
 #endif
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_bios.c b/kernel/drivers/gpu/drm/radeon/radeon_bios.c
index d27e4ccb8..21b673242 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_bios.c
@@ -30,7 +30,6 @@
 #include "radeon.h"
 #include "atom.h"
 
-#include <linux/vga_switcheroo.h>
 #include <linux/slab.h>
 #include <linux/acpi.h>
 /*
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_connectors.c b/kernel/drivers/gpu/drm/radeon/radeon_connectors.c
index 5a2cafb4f..340f3f549 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -1234,13 +1234,32 @@ radeon_dvi_detect(struct drm_connector *connector, bool force)
 	if (r < 0)
 		return connector_status_disconnected;
 
+	if (radeon_connector->detected_hpd_without_ddc) {
+		force = true;
+		radeon_connector->detected_hpd_without_ddc = false;
+	}
+
 	if (!force && radeon_check_hpd_status_unchanged(connector)) {
 		ret = connector->status;
 		goto exit;
 	}
 
-	if (radeon_connector->ddc_bus)
+	if (radeon_connector->ddc_bus) {
 		dret = radeon_ddc_probe(radeon_connector, false);
+
+		/* Sometimes the pins required for the DDC probe on DVI
+		 * connectors don't make contact at the same time that the ones
+		 * for HPD do. If the DDC probe fails even though we had an HPD
+		 * signal, try again later */
+		if (!dret && !force &&
+		    connector->status != connector_status_connected) {
+			DRM_DEBUG_KMS("hpd detected without ddc, retrying in 1 second\n");
+			radeon_connector->detected_hpd_without_ddc = true;
+			schedule_delayed_work(&rdev->hotplug_work,
+					      msecs_to_jiffies(1000));
+			goto exit;
+		}
+	}
 	if (dret) {
 		radeon_connector->detected_by_load = false;
 		radeon_connector_free_edid(connector);
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_cp.c b/kernel/drivers/gpu/drm/radeon/radeon_cp.c
index ea134a7d5..500287eff 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_cp.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_cp.c
@@ -762,7 +762,7 @@ static void radeon_cp_init_ring_buffer(struct drm_device * dev,
 			     ((dev_priv->gart_vm_start - 1) & 0xffff0000)
 			     | (dev_priv->fb_location >> 16));
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		radeon_write_agp_base(dev_priv, dev->agp->base);
 
@@ -791,7 +791,7 @@ static void radeon_cp_init_ring_buffer(struct drm_device * dev,
 	SET_RING_HEAD(dev_priv, cur_read_ptr);
 	dev_priv->ring.tail = cur_read_ptr;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		RADEON_WRITE(RADEON_CP_RB_RPTR_ADDR,
 			     dev_priv->ring_rptr->offset
@@ -1335,7 +1335,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		}
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		drm_legacy_ioremap_wc(dev_priv->cp_ring, dev);
 		drm_legacy_ioremap_wc(dev_priv->ring_rptr, dev);
@@ -1394,7 +1394,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		 * location in the card and on the bus, though we have to
 		 * align it down.
 		 */
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (dev_priv->flags & RADEON_IS_AGP) {
 			base = dev->agp->base;
 			/* Check if valid */
@@ -1424,7 +1424,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 			RADEON_READ(RADEON_CONFIG_APER_SIZE);
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP)
 		dev_priv->gart_buffers_offset = (dev->agp_buffer_map->offset
 						 - dev->agp->base
@@ -1455,7 +1455,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 
 	dev_priv->ring.high_mark = RADEON_RING_HIGH_MARK;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* Turn off PCI GART */
 		radeon_set_pcigart(dev_priv, 0);
@@ -1566,7 +1566,7 @@ static int radeon_do_cleanup_cp(struct drm_device * dev)
 	if (dev->irq_enabled)
 		drm_irq_uninstall(dev);
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		if (dev_priv->cp_ring != NULL) {
 			drm_legacy_ioremapfree(dev_priv->cp_ring, dev);
@@ -1625,7 +1625,7 @@ static int radeon_do_resume_cp(struct drm_device *dev, struct drm_file *file_pri
 
 	DRM_DEBUG("Starting radeon_do_resume_cp()\n");
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* Turn off PCI GART */
 		radeon_set_pcigart(dev_priv, 0);
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_cursor.c b/kernel/drivers/gpu/drm/radeon/radeon_cursor.c
index fa661744a..afaf346bd 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_cursor.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_cursor.c
@@ -91,15 +91,34 @@ static void radeon_show_cursor(struct drm_crtc *crtc)
 	struct radeon_device *rdev = crtc->dev->dev_private;
 
 	if (ASIC_IS_DCE4(rdev)) {
+		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
+		       upper_32_bits(radeon_crtc->cursor_addr));
+		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+		       lower_32_bits(radeon_crtc->cursor_addr));
 		WREG32(RADEON_MM_INDEX, EVERGREEN_CUR_CONTROL + radeon_crtc->crtc_offset);
 		WREG32(RADEON_MM_DATA, EVERGREEN_CURSOR_EN |
 		       EVERGREEN_CURSOR_MODE(EVERGREEN_CURSOR_24_8_PRE_MULT) |
 		       EVERGREEN_CURSOR_URGENT_CONTROL(EVERGREEN_CURSOR_URGENT_1_2));
 	} else if (ASIC_IS_AVIVO(rdev)) {
+		if (rdev->family >= CHIP_RV770) {
+			if (radeon_crtc->crtc_id)
+				WREG32(R700_D2CUR_SURFACE_ADDRESS_HIGH,
+				       upper_32_bits(radeon_crtc->cursor_addr));
+			else
+				WREG32(R700_D1CUR_SURFACE_ADDRESS_HIGH,
+				       upper_32_bits(radeon_crtc->cursor_addr));
+		}
+
+		WREG32(AVIVO_D1CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+		       lower_32_bits(radeon_crtc->cursor_addr));
 		WREG32(RADEON_MM_INDEX, AVIVO_D1CUR_CONTROL + radeon_crtc->crtc_offset);
 		WREG32(RADEON_MM_DATA, AVIVO_D1CURSOR_EN |
 		       (AVIVO_D1CURSOR_MODE_24BPP << AVIVO_D1CURSOR_MODE_SHIFT));
 	} else {
+		/* offset is from DISP(2)_BASE_ADDRESS */
+		WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset,
+		       radeon_crtc->cursor_addr - radeon_crtc->legacy_display_base_addr);
+
 		switch (radeon_crtc->crtc_id) {
 		case 0:
 			WREG32(RADEON_MM_INDEX, RADEON_CRTC_GEN_CNTL);
@@ -228,34 +247,6 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 	return ret;
 }
 
-static void radeon_set_cursor(struct drm_crtc *crtc)
-{
-	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
-	struct radeon_device *rdev = crtc->dev->dev_private;
-
-	if (ASIC_IS_DCE4(rdev)) {
-		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
-		       upper_32_bits(radeon_crtc->cursor_addr));
-		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
-		       lower_32_bits(radeon_crtc->cursor_addr));
-	} else if (ASIC_IS_AVIVO(rdev)) {
-		if (rdev->family >= CHIP_RV770) {
-			if (radeon_crtc->crtc_id)
-				WREG32(R700_D2CUR_SURFACE_ADDRESS_HIGH,
-				       upper_32_bits(radeon_crtc->cursor_addr));
-			else
-				WREG32(R700_D1CUR_SURFACE_ADDRESS_HIGH,
-				       upper_32_bits(radeon_crtc->cursor_addr));
-		}
-		WREG32(AVIVO_D1CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
-		       lower_32_bits(radeon_crtc->cursor_addr));
-	} else {
-		/* offset is from DISP(2)_BASE_ADDRESS */
-		WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset,
-		       radeon_crtc->cursor_addr - radeon_crtc->legacy_display_base_addr);
-	}
-}
-
 int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 			    struct drm_file *file_priv,
 			    uint32_t handle,
@@ -324,7 +315,6 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 		radeon_crtc->cursor_hot_y = hot_y;
 	}
 
-	radeon_set_cursor(crtc);
 	radeon_show_cursor(crtc);
 
 	radeon_lock_cursor(crtc, false);
@@ -362,7 +352,6 @@ void radeon_cursor_reset(struct drm_crtc *crtc)
 		radeon_cursor_move_locked(crtc, radeon_crtc->cursor_x,
 					  radeon_crtc->cursor_y);
 
-		radeon_set_cursor(crtc);
 		radeon_show_cursor(crtc);
 
 		radeon_lock_cursor(crtc, false);
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_device.c b/kernel/drivers/gpu/drm/radeon/radeon_device.c
index 604c44d88..c566993a2 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_device.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_device.c
@@ -1080,6 +1080,22 @@ static bool radeon_check_pot_argument(int arg)
 }
 
 /**
+ * Determine a sensible default GART size according to ASIC family.
+ *
+ * @family ASIC family name
+ */
+static int radeon_gart_size_auto(enum radeon_family family)
+{
+	/* default to a larger gart size on newer asics */
+	if (family >= CHIP_TAHITI)
+		return 2048;
+	else if (family >= CHIP_RV770)
+		return 1024;
+	else
+		return 512;
+}
+
+/**
  * radeon_check_arguments - validate module params
  *
  * @rdev: radeon_device pointer
@@ -1097,27 +1113,17 @@ static void radeon_check_arguments(struct radeon_device *rdev)
 	}
 
 	if (radeon_gart_size == -1) {
-		/* default to a larger gart size on newer asics */
-		if (rdev->family >= CHIP_RV770)
-			radeon_gart_size = 1024;
-		else
-			radeon_gart_size = 512;
+		radeon_gart_size = radeon_gart_size_auto(rdev->family);
 	}
 	/* gtt size must be power of two and greater or equal to 32M */
 	if (radeon_gart_size < 32) {
 		dev_warn(rdev->dev, "gart size (%d) too small\n",
 				radeon_gart_size);
-		if (rdev->family >= CHIP_RV770)
-			radeon_gart_size = 1024;
-		else
-			radeon_gart_size = 512;
+		radeon_gart_size = radeon_gart_size_auto(rdev->family);
 	} else if (!radeon_check_pot_argument(radeon_gart_size)) {
 		dev_warn(rdev->dev, "gart size (%d) must be a power of 2\n",
 				radeon_gart_size);
-		if (rdev->family >= CHIP_RV770)
-			radeon_gart_size = 1024;
-		else
-			radeon_gart_size = 512;
+		radeon_gart_size = radeon_gart_size_auto(rdev->family);
 	}
 	rdev->mc.gtt_size = (uint64_t)radeon_gart_size << 20;
 
@@ -1191,7 +1197,7 @@ static void radeon_check_arguments(struct radeon_device *rdev)
  * radeon_switcheroo_set_state - set switcheroo state
  *
  * @pdev: pci dev pointer
- * @state: vga switcheroo state
+ * @state: vga_switcheroo state
  *
  * Callback for the switcheroo driver.  Suspends or resumes the
  * the asics before or after it is powered up using ACPI methods.
@@ -1567,10 +1573,12 @@ int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon)
 
 	drm_kms_helper_poll_disable(dev);
 
+	drm_modeset_lock_all(dev);
 	/* turn off display hw */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
 	}
+	drm_modeset_unlock_all(dev);
 
 	/* unpin the front buffers and cursors */
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
@@ -1728,9 +1736,11 @@ int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 	if (fbcon) {
 		drm_helper_resume_force_mode(dev);
 		/* turn on display hw */
+		drm_modeset_lock_all(dev);
 		list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
 		}
+		drm_modeset_unlock_all(dev);
 	}
 
 	drm_kms_helper_poll_enable(dev);
@@ -1772,6 +1782,8 @@ int radeon_gpu_reset(struct radeon_device *rdev)
 		return 0;
 	}
 
+	atomic_inc(&rdev->gpu_reset_counter);
+
 	radeon_save_bios_scratch_regs(rdev);
 	/* block TTM */
 	resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_display.c b/kernel/drivers/gpu/drm/radeon/radeon_display.c
index d2e9e9efc..642854b2e 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_display.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_display.c
@@ -322,8 +322,11 @@ void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id)
 	 * to complete in this vblank?
 	 */
 	if (update_pending &&
-	    (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, crtc_id, 0,
-							       &vpos, &hpos, NULL, NULL)) &&
+	    (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev,
+							       crtc_id,
+							       USE_REAL_VBLANKSTART,
+							       &vpos, &hpos, NULL, NULL,
+							       &rdev->mode_info.crtcs[crtc_id]->base.hwmode)) &&
 	    ((vpos >= (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100) ||
 	     (vpos < 0 && !ASIC_IS_AVIVO(rdev)))) {
 		/* crtc didn't flip in this target vblank interval,
@@ -400,6 +403,9 @@ static void radeon_flip_work_func(struct work_struct *__work)
 	struct drm_crtc *crtc = &radeon_crtc->base;
 	unsigned long flags;
 	int r;
+	int vpos, hpos, stat, min_udelay = 0;
+	unsigned repcnt = 4;
+	struct drm_vblank_crtc *vblank = &crtc->dev->vblank[work->crtc_id];
 
         down_read(&rdev->exclusive_lock);
 	if (work->fence) {
@@ -436,6 +442,53 @@ static void radeon_flip_work_func(struct work_struct *__work)
 	/* set the proper interrupt */
 	radeon_irq_kms_pflip_irq_get(rdev, radeon_crtc->crtc_id);
 
+	/* If this happens to execute within the "virtually extended" vblank
+	 * interval before the start of the real vblank interval then it needs
+	 * to delay programming the mmio flip until the real vblank is entered.
+	 * This prevents completing a flip too early due to the way we fudge
+	 * our vblank counter and vblank timestamps in order to work around the
+	 * problem that the hw fires vblank interrupts before actual start of
+	 * vblank (when line buffer refilling is done for a frame). It
+	 * complements the fudging logic in radeon_get_crtc_scanoutpos() for
+	 * timestamping and radeon_get_vblank_counter_kms() for vblank counts.
+	 *
+	 * In practice this won't execute very often unless on very fast
+	 * machines because the time window for this to happen is very small.
+	 */
+	while (radeon_crtc->enabled && --repcnt) {
+		/* GET_DISTANCE_TO_VBLANKSTART returns distance to real vblank
+		 * start in hpos, and to the "fudged earlier" vblank start in
+		 * vpos.
+		 */
+		stat = radeon_get_crtc_scanoutpos(rdev->ddev, work->crtc_id,
+						  GET_DISTANCE_TO_VBLANKSTART,
+						  &vpos, &hpos, NULL, NULL,
+						  &crtc->hwmode);
+
+		if ((stat & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE)) !=
+		    (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE) ||
+		    !(vpos >= 0 && hpos <= 0))
+			break;
+
+		/* Sleep at least until estimated real start of hw vblank */
+		min_udelay = (-hpos + 1) * max(vblank->linedur_ns / 1000, 5);
+		if (min_udelay > vblank->framedur_ns / 2000) {
+			/* Don't wait ridiculously long - something is wrong */
+			repcnt = 0;
+			break;
+		}
+		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+		usleep_range(min_udelay, 2 * min_udelay);
+		spin_lock_irqsave(&crtc->dev->event_lock, flags);
+	};
+
+	if (!repcnt)
+		DRM_DEBUG_DRIVER("Delay problem on crtc %d: min_udelay %d, "
+				 "framedur %d, linedur %d, stat %d, vpos %d, "
+				 "hpos %d\n", work->crtc_id, min_udelay,
+				 vblank->framedur_ns / 1000,
+				 vblank->linedur_ns / 1000, stat, vpos, hpos);
+
 	/* do the flip (mmio) */
 	radeon_page_flip(rdev, radeon_crtc->crtc_id, work->base);
 
@@ -1633,18 +1686,8 @@ int radeon_modeset_init(struct radeon_device *rdev)
 	radeon_fbdev_init(rdev);
 	drm_kms_helper_poll_init(rdev->ddev);
 
-	if (rdev->pm.dpm_enabled) {
-		/* do dpm late init */
-		ret = radeon_pm_late_init(rdev);
-		if (ret) {
-			rdev->pm.dpm_enabled = false;
-			DRM_ERROR("radeon_pm_late_init failed, disabling dpm\n");
-		}
-		/* set the dpm state for PX since there won't be
-		 * a modeset to call this.
-		 */
-		radeon_pm_compute_clocks(rdev);
-	}
+	/* do pm late init */
+	ret = radeon_pm_late_init(rdev);
 
 	return 0;
 }
@@ -1777,6 +1820,15 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  * \param dev Device to query.
  * \param crtc Crtc to query.
  * \param flags Flags from caller (DRM_CALLED_FROM_VBLIRQ or 0).
+ *              For driver internal use only also supports these flags:
+ *
+ *              USE_REAL_VBLANKSTART to use the real start of vblank instead
+ *              of a fudged earlier start of vblank.
+ *
+ *              GET_DISTANCE_TO_VBLANKSTART to return distance to the
+ *              fudged earlier start of vblank in *vpos and the distance
+ *              to true start of vblank in *hpos.
+ *
  * \param *vpos Location where vertical scanout position should be stored.
  * \param *hpos Location where horizontal scanout position should go.
  * \param *stime Target location for timestamp taken immediately before
@@ -1798,8 +1850,10 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  * unknown small number of scanlines wrt. real scanout position.
  *
  */
-int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int flags,
-			       int *vpos, int *hpos, ktime_t *stime, ktime_t *etime)
+int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
+			       unsigned int flags, int *vpos, int *hpos,
+			       ktime_t *stime, ktime_t *etime,
+			       const struct drm_display_mode *mode)
 {
 	u32 stat_crtc = 0, vbl = 0, position = 0;
 	int vbl_start, vbl_end, vtotal, ret = 0;
@@ -1808,48 +1862,49 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 	struct radeon_device *rdev = dev->dev_private;
 
 	/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
+	preempt_disable_rt();
 
 	/* Get optional system timestamp before query. */
 	if (stime)
 		*stime = ktime_get();
 
 	if (ASIC_IS_DCE4(rdev)) {
-		if (crtc == 0) {
+		if (pipe == 0) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC0_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC0_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 1) {
+		if (pipe == 1) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC1_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC1_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 2) {
+		if (pipe == 2) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC2_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC2_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 3) {
+		if (pipe == 3) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC3_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC3_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 4) {
+		if (pipe == 4) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC4_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC4_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 5) {
+		if (pipe == 5) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC5_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
@@ -1857,19 +1912,19 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
 	} else if (ASIC_IS_AVIVO(rdev)) {
-		if (crtc == 0) {
+		if (pipe == 0) {
 			vbl = RREG32(AVIVO_D1CRTC_V_BLANK_START_END);
 			position = RREG32(AVIVO_D1CRTC_STATUS_POSITION);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 1) {
+		if (pipe == 1) {
 			vbl = RREG32(AVIVO_D2CRTC_V_BLANK_START_END);
 			position = RREG32(AVIVO_D2CRTC_STATUS_POSITION);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
 	} else {
 		/* Pre-AVIVO: Different encoding of scanout pos and vblank interval. */
-		if (crtc == 0) {
+		if (pipe == 0) {
 			/* Assume vbl_end == 0, get vbl_start from
 			 * upper 16 bits.
 			 */
@@ -1883,7 +1938,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 1) {
+		if (pipe == 1) {
 			vbl = (RREG32(RADEON_CRTC2_V_TOTAL_DISP) &
 				RADEON_CRTC_V_DISP) >> RADEON_CRTC_V_DISP_SHIFT;
 			position = (RREG32(RADEON_CRTC2_VLINE_CRNT_VLINE) >> 16) & RADEON_CRTC_V_TOTAL;
@@ -1900,6 +1955,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 		*etime = ktime_get();
 
 	/* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */
+	preempt_enable_rt();
 
 	/* Decode into vertical and horizontal scanout position. */
 	*vpos = position & 0x1fff;
@@ -1914,14 +1970,44 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 	}
 	else {
 		/* No: Fake something reasonable which gives at least ok results. */
-		vbl_start = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vdisplay;
+		vbl_start = mode->crtc_vdisplay;
 		vbl_end = 0;
 	}
 
+	/* Called from driver internal vblank counter query code? */
+	if (flags & GET_DISTANCE_TO_VBLANKSTART) {
+	    /* Caller wants distance from real vbl_start in *hpos */
+	    *hpos = *vpos - vbl_start;
+	}
+
+	/* Fudge vblank to start a few scanlines earlier to handle the
+	 * problem that vblank irqs fire a few scanlines before start
+	 * of vblank. Some driver internal callers need the true vblank
+	 * start to be used and signal this via the USE_REAL_VBLANKSTART flag.
+	 *
+	 * The cause of the "early" vblank irq is that the irq is triggered
+	 * by the line buffer logic when the line buffer read position enters
+	 * the vblank, whereas our crtc scanout position naturally lags the
+	 * line buffer read position.
+	 */
+	if (!(flags & USE_REAL_VBLANKSTART))
+		vbl_start -= rdev->mode_info.crtcs[pipe]->lb_vblank_lead_lines;
+
 	/* Test scanout position against vblank region. */
 	if ((*vpos < vbl_start) && (*vpos >= vbl_end))
 		in_vbl = false;
 
+	/* In vblank? */
+	if (in_vbl)
+	    ret |= DRM_SCANOUTPOS_IN_VBLANK;
+
+	/* Called from driver internal vblank counter query code? */
+	if (flags & GET_DISTANCE_TO_VBLANKSTART) {
+		/* Caller wants distance from fudged earlier vbl_start */
+		*vpos -= vbl_start;
+		return ret;
+	}
+
 	/* Check if inside vblank area and apply corrective offsets:
 	 * vpos will then be >=0 in video scanout area, but negative
 	 * within vblank area, counting down the number of lines until
@@ -1930,38 +2016,12 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 
 	/* Inside "upper part" of vblank area? Apply corrective offset if so: */
 	if (in_vbl && (*vpos >= vbl_start)) {
-		vtotal = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vtotal;
+		vtotal = mode->crtc_vtotal;
 		*vpos = *vpos - vtotal;
 	}
 
 	/* Correct for shifted end of vbl at vbl_end. */
 	*vpos = *vpos - vbl_end;
 
-	/* In vblank? */
-	if (in_vbl)
-		ret |= DRM_SCANOUTPOS_IN_VBLANK;
-
-	/* Is vpos outside nominal vblank area, but less than
-	 * 1/100 of a frame height away from start of vblank?
-	 * If so, assume this isn't a massively delayed vblank
-	 * interrupt, but a vblank interrupt that fired a few
-	 * microseconds before true start of vblank. Compensate
-	 * by adding a full frame duration to the final timestamp.
-	 * Happens, e.g., on ATI R500, R600.
-	 *
-	 * We only do this if DRM_CALLED_FROM_VBLIRQ.
-	 */
-	if ((flags & DRM_CALLED_FROM_VBLIRQ) && !in_vbl) {
-		vbl_start = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vdisplay;
-		vtotal = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vtotal;
-
-		if (vbl_start - *vpos < vtotal / 100) {
-			*vpos -= vtotal;
-
-			/* Signal this correction as "applied". */
-			ret |= 0x8;
-		}
-	}
-
 	return ret;
 }
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_dp_mst.c b/kernel/drivers/gpu/drm/radeon/radeon_dp_mst.c
index 257b10be5..744f5c49c 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_dp_mst.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_dp_mst.c
@@ -246,9 +246,10 @@ radeon_dp_mst_connector_destroy(struct drm_connector *connector)
 	kfree(radeon_connector);
 }
 
-static void radeon_connector_dpms(struct drm_connector *connector, int mode)
+static int radeon_connector_dpms(struct drm_connector *connector, int mode)
 {
 	DRM_DEBUG_KMS("\n");
+	return 0;
 }
 
 static const struct drm_connector_funcs radeon_dp_mst_connector_funcs = {
@@ -264,7 +265,6 @@ static struct drm_connector *radeon_dp_add_mst_connector(struct drm_dp_mst_topol
 {
 	struct radeon_connector *master = container_of(mgr, struct radeon_connector, mst_mgr);
 	struct drm_device *dev = master->base.dev;
-	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_connector *radeon_connector;
 	struct drm_connector *connector;
 
@@ -283,15 +283,22 @@ static struct drm_connector *radeon_dp_add_mst_connector(struct drm_dp_mst_topol
 	radeon_connector->mst_encoder = radeon_dp_create_fake_mst_encoder(master);
 
 	drm_object_attach_property(&connector->base, dev->mode_config.path_property, 0);
+	drm_object_attach_property(&connector->base, dev->mode_config.tile_property, 0);
 	drm_mode_connector_set_path_property(connector, pathprop);
-	drm_reinit_primary_mode_group(dev);
 
-	mutex_lock(&dev->mode_config.mutex);
+	return connector;
+}
+
+static void radeon_dp_register_mst_connector(struct drm_connector *connector)
+{
+	struct drm_device *dev = connector->dev;
+	struct radeon_device *rdev = dev->dev_private;
+
+	drm_modeset_lock_all(dev);
 	radeon_fb_add_connector(rdev, connector);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	drm_connector_register(connector);
-	return connector;
 }
 
 static void radeon_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
@@ -303,14 +310,12 @@ static void radeon_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
 
 	drm_connector_unregister(connector);
 	/* need to nuke the connector */
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	/* dpms off */
 	radeon_fb_remove_connector(rdev, connector);
 
 	drm_connector_cleanup(connector);
-	mutex_unlock(&dev->mode_config.mutex);
-	drm_reinit_primary_mode_group(dev);
-
+	drm_modeset_unlock_all(dev);
 
 	kfree(connector);
 	DRM_DEBUG_KMS("\n");
@@ -326,6 +331,7 @@ static void radeon_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr)
 
 struct drm_dp_mst_topology_cbs mst_cbs = {
 	.add_connector = radeon_dp_add_mst_connector,
+	.register_connector = radeon_dp_register_mst_connector,
 	.destroy_connector = radeon_dp_destroy_mst_connector,
 	.hotplug = radeon_dp_mst_hotplug,
 };
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_drv.c b/kernel/drivers/gpu/drm/radeon/radeon_drv.c
index 7d620d4b3..5b6a6f5b3 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_drv.c
@@ -90,9 +90,10 @@
  *            CS to GPU on >= r600
  *   2.41.0 - evergreen/cayman: Add SET_BASE/DRAW_INDIRECT command parsing support
  *   2.42.0 - Add VCE/VUI (Video Usability Information) support
+ *   2.43.0 - RADEON_INFO_GPU_RESET_COUNTER
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	42
+#define KMS_DRIVER_MINOR	43
 #define KMS_DRIVER_PATCHLEVEL	0
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags);
 int radeon_driver_unload_kms(struct drm_device *dev);
@@ -104,10 +105,10 @@ void radeon_driver_preclose_kms(struct drm_device *dev,
 				struct drm_file *file_priv);
 int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon);
 int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon);
-u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc);
-int radeon_enable_vblank_kms(struct drm_device *dev, int crtc);
-void radeon_disable_vblank_kms(struct drm_device *dev, int crtc);
-int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
+u32 radeon_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe);
+int radeon_enable_vblank_kms(struct drm_device *dev, unsigned int pipe);
+void radeon_disable_vblank_kms(struct drm_device *dev, unsigned int pipe);
+int radeon_get_vblank_timestamp_kms(struct drm_device *dev, unsigned int pipe,
 				    int *max_error,
 				    struct timeval *vblank_time,
 				    unsigned flags);
@@ -123,10 +124,10 @@ void radeon_gem_object_close(struct drm_gem_object *obj,
 struct dma_buf *radeon_gem_prime_export(struct drm_device *dev,
 					struct drm_gem_object *gobj,
 					int flags);
-extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
-				      unsigned int flags,
-				      int *vpos, int *hpos, ktime_t *stime,
-				      ktime_t *etime);
+extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int crtc,
+				      unsigned int flags, int *vpos, int *hpos,
+				      ktime_t *stime, ktime_t *etime,
+				      const struct drm_display_mode *mode);
 extern bool radeon_is_px(struct drm_device *dev);
 extern const struct drm_ioctl_desc radeon_ioctls_kms[];
 extern int radeon_max_kms_ioctl;
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_drv.h b/kernel/drivers/gpu/drm/radeon/radeon_drv.h
index 46bd39382..0caafc7a6 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_drv.h
+++ b/kernel/drivers/gpu/drm/radeon/radeon_drv.h
@@ -404,9 +404,9 @@ extern int radeon_irq_emit(struct drm_device *dev, void *data, struct drm_file *
 extern int radeon_irq_wait(struct drm_device *dev, void *data, struct drm_file *file_priv);
 
 extern void radeon_do_release(struct drm_device * dev);
-extern u32 radeon_get_vblank_counter(struct drm_device *dev, int crtc);
-extern int radeon_enable_vblank(struct drm_device *dev, int crtc);
-extern void radeon_disable_vblank(struct drm_device *dev, int crtc);
+extern u32 radeon_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
+extern int radeon_enable_vblank(struct drm_device *dev, unsigned int pipe);
+extern void radeon_disable_vblank(struct drm_device *dev, unsigned int pipe);
 extern irqreturn_t radeon_driver_irq_handler(int irq, void *arg);
 extern void radeon_driver_irq_preinstall(struct drm_device * dev);
 extern int radeon_driver_irq_postinstall(struct drm_device *dev);
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_encoders.c b/kernel/drivers/gpu/drm/radeon/radeon_encoders.c
index ef99917f0..c6ee80216 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -194,7 +194,6 @@ static void radeon_encoder_add_backlight(struct radeon_encoder *radeon_encoder,
 			radeon_atom_backlight_init(radeon_encoder, connector);
 		else
 			radeon_legacy_backlight_init(radeon_encoder, connector);
-		rdev->mode_info.bl_encoder = radeon_encoder;
 	}
 }
 
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_fb.c b/kernel/drivers/gpu/drm/radeon/radeon_fb.c
index 634793ea8..26da2f4d7 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_fb.c
@@ -48,43 +48,13 @@ struct radeon_fbdev {
 	struct radeon_device *rdev;
 };
 
-/**
- * radeon_fb_helper_set_par - Hide cursor on CRTCs used by fbdev.
- *
- * @info: fbdev info
- *
- * This function hides the cursor on all CRTCs used by fbdev.
- */
-static int radeon_fb_helper_set_par(struct fb_info *info)
-{
-	int ret;
-
-	ret = drm_fb_helper_set_par(info);
-
-	/* XXX: with universal plane support fbdev will automatically disable
-	 * all non-primary planes (including the cursor)
-	 */
-	if (ret == 0) {
-		struct drm_fb_helper *fb_helper = info->par;
-		int i;
-
-		for (i = 0; i < fb_helper->crtc_count; i++) {
-			struct drm_crtc *crtc = fb_helper->crtc_info[i].mode_set.crtc;
-
-			radeon_crtc_cursor_set2(crtc, NULL, 0, 0, 0, 0, 0);
-		}
-	}
-
-	return ret;
-}
-
 static struct fb_ops radeonfb_ops = {
 	.owner = THIS_MODULE,
 	.fb_check_var = drm_fb_helper_check_var,
-	.fb_set_par = radeon_fb_helper_set_par,
-	.fb_fillrect = cfb_fillrect,
-	.fb_copyarea = cfb_copyarea,
-	.fb_imageblit = cfb_imageblit,
+	.fb_set_par = drm_fb_helper_set_par,
+	.fb_fillrect = drm_fb_helper_cfb_fillrect,
+	.fb_copyarea = drm_fb_helper_cfb_copyarea,
+	.fb_imageblit = drm_fb_helper_cfb_imageblit,
 	.fb_pan_display = drm_fb_helper_pan_display,
 	.fb_blank = drm_fb_helper_blank,
 	.fb_setcmap = drm_fb_helper_setcmap,
@@ -227,7 +197,6 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 	struct drm_mode_fb_cmd2 mode_cmd;
 	struct drm_gem_object *gobj = NULL;
 	struct radeon_bo *rbo = NULL;
-	struct device *device = &rdev->pdev->dev;
 	int ret;
 	unsigned long tmp;
 
@@ -250,25 +219,25 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 	rbo = gem_to_radeon_bo(gobj);
 
 	/* okay we have an object now allocate the framebuffer */
-	info = framebuffer_alloc(0, device);
-	if (info == NULL) {
-		ret = -ENOMEM;
+	info = drm_fb_helper_alloc_fbi(helper);
+	if (IS_ERR(info)) {
+		ret = PTR_ERR(info);
 		goto out_unref;
 	}
 
 	info->par = rfbdev;
+	info->skip_vt_switch = true;
 
 	ret = radeon_framebuffer_init(rdev->ddev, &rfbdev->rfb, &mode_cmd, gobj);
 	if (ret) {
 		DRM_ERROR("failed to initialize framebuffer %d\n", ret);
-		goto out_unref;
+		goto out_destroy_fbi;
 	}
 
 	fb = &rfbdev->rfb.base;
 
 	/* setup helper */
 	rfbdev->helper.fb = fb;
-	rfbdev->helper.fbdev = info;
 
 	memset_io(rbo->kptr, 0x0, radeon_bo_size(rbo));
 
@@ -288,11 +257,6 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 	drm_fb_helper_fill_var(info, &rfbdev->helper, sizes->fb_width, sizes->fb_height);
 
 	/* setup aperture base/size for vesafb takeover */
-	info->apertures = alloc_apertures(1);
-	if (!info->apertures) {
-		ret = -ENOMEM;
-		goto out_unref;
-	}
 	info->apertures->ranges[0].base = rdev->ddev->mode_config.fb_base;
 	info->apertures->ranges[0].size = rdev->mc.aper_size;
 
@@ -300,13 +264,7 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 
 	if (info->screen_base == NULL) {
 		ret = -ENOSPC;
-		goto out_unref;
-	}
-
-	ret = fb_alloc_cmap(&info->cmap, 256, 0);
-	if (ret) {
-		ret = -ENOMEM;
-		goto out_unref;
+		goto out_destroy_fbi;
 	}
 
 	DRM_INFO("fb mappable at 0x%lX\n",  info->fix.smem_start);
@@ -318,6 +276,8 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 	vga_switcheroo_client_fb_set(rdev->ddev->pdev, info);
 	return 0;
 
+out_destroy_fbi:
+	drm_fb_helper_release_fbi(helper);
 out_unref:
 	if (rbo) {
 
@@ -338,17 +298,10 @@ void radeon_fb_output_poll_changed(struct radeon_device *rdev)
 
 static int radeon_fbdev_destroy(struct drm_device *dev, struct radeon_fbdev *rfbdev)
 {
-	struct fb_info *info;
 	struct radeon_framebuffer *rfb = &rfbdev->rfb;
 
-	if (rfbdev->helper.fbdev) {
-		info = rfbdev->helper.fbdev;
-
-		unregister_framebuffer(info);
-		if (info->cmap.len)
-			fb_dealloc_cmap(&info->cmap);
-		framebuffer_release(info);
-	}
+	drm_fb_helper_unregister_fbi(&rfbdev->helper);
+	drm_fb_helper_release_fbi(&rfbdev->helper);
 
 	if (rfb->obj) {
 		radeonfb_destroy_pinned_object(rfb->obj);
@@ -444,3 +397,19 @@ void radeon_fb_remove_connector(struct radeon_device *rdev, struct drm_connector
 {
 	drm_fb_helper_remove_one_connector(&rdev->mode_info.rfbdev->helper, connector);
 }
+
+void radeon_fbdev_restore_mode(struct radeon_device *rdev)
+{
+	struct radeon_fbdev *rfbdev = rdev->mode_info.rfbdev;
+	struct drm_fb_helper *fb_helper;
+	int ret;
+
+	if (!rfbdev)
+		return;
+
+	fb_helper = &rfbdev->helper;
+
+	ret = drm_fb_helper_restore_fbdev_mode_unlocked(fb_helper);
+	if (ret)
+		DRM_DEBUG("failed to restore crtc mode\n");
+}
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_gem.c b/kernel/drivers/gpu/drm/radeon/radeon_gem.c
index 186d0b792..3dcc5733f 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_gem.c
@@ -429,7 +429,6 @@ int radeon_gem_mmap_ioctl(struct drm_device *dev, void *data,
 int radeon_gem_busy_ioctl(struct drm_device *dev, void *data,
 			  struct drm_file *filp)
 {
-	struct radeon_device *rdev = dev->dev_private;
 	struct drm_radeon_gem_busy *args = data;
 	struct drm_gem_object *gobj;
 	struct radeon_bo *robj;
@@ -441,10 +440,16 @@ int radeon_gem_busy_ioctl(struct drm_device *dev, void *data,
 		return -ENOENT;
 	}
 	robj = gem_to_radeon_bo(gobj);
-	r = radeon_bo_wait(robj, &cur_placement, true);
+
+	r = reservation_object_test_signaled_rcu(robj->tbo.resv, true);
+	if (r == 0)
+		r = -EBUSY;
+	else
+		r = 0;
+
+	cur_placement = ACCESS_ONCE(robj->tbo.mem.mem_type);
 	args->domain = radeon_mem_type_to_domain(cur_placement);
 	drm_gem_object_unreference_unlocked(gobj);
-	r = radeon_gem_handle_lockup(rdev, r);
 	return r;
 }
 
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_irq.c b/kernel/drivers/gpu/drm/radeon/radeon_irq.c
index 244b19bab..688afb62f 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_irq.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_irq.c
@@ -62,12 +62,12 @@ static void r500_vbl_irq_set_state(struct drm_device *dev, u32 mask, int state)
 		RADEON_WRITE(R500_DxMODE_INT_MASK, dev_priv->r500_disp_irq_reg);
 }
 
-int radeon_enable_vblank(struct drm_device *dev, int crtc)
+int radeon_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
 
 	if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_RS600) {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			r500_vbl_irq_set_state(dev, R500_D1MODE_INT_MASK, 1);
 			break;
@@ -75,12 +75,12 @@ int radeon_enable_vblank(struct drm_device *dev, int crtc)
 			r500_vbl_irq_set_state(dev, R500_D2MODE_INT_MASK, 1);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			return -EINVAL;
 		}
 	} else {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			radeon_irq_set_state(dev, RADEON_CRTC_VBLANK_MASK, 1);
 			break;
@@ -88,8 +88,8 @@ int radeon_enable_vblank(struct drm_device *dev, int crtc)
 			radeon_irq_set_state(dev, RADEON_CRTC2_VBLANK_MASK, 1);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			return -EINVAL;
 		}
 	}
@@ -97,12 +97,12 @@ int radeon_enable_vblank(struct drm_device *dev, int crtc)
 	return 0;
 }
 
-void radeon_disable_vblank(struct drm_device *dev, int crtc)
+void radeon_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
 
 	if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_RS600) {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			r500_vbl_irq_set_state(dev, R500_D1MODE_INT_MASK, 0);
 			break;
@@ -110,12 +110,12 @@ void radeon_disable_vblank(struct drm_device *dev, int crtc)
 			r500_vbl_irq_set_state(dev, R500_D2MODE_INT_MASK, 0);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			break;
 		}
 	} else {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			radeon_irq_set_state(dev, RADEON_CRTC_VBLANK_MASK, 0);
 			break;
@@ -123,8 +123,8 @@ void radeon_disable_vblank(struct drm_device *dev, int crtc)
 			radeon_irq_set_state(dev, RADEON_CRTC2_VBLANK_MASK, 0);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			break;
 		}
 	}
@@ -255,7 +255,7 @@ static int radeon_wait_irq(struct drm_device * dev, int swi_nr)
 	return ret;
 }
 
-u32 radeon_get_vblank_counter(struct drm_device *dev, int crtc)
+u32 radeon_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
 
@@ -264,18 +264,18 @@ u32 radeon_get_vblank_counter(struct drm_device *dev, int crtc)
 		return -EINVAL;
 	}
 
-	if (crtc < 0 || crtc > 1) {
-		DRM_ERROR("Invalid crtc %d\n", crtc);
+	if (pipe > 1) {
+		DRM_ERROR("Invalid crtc %u\n", pipe);
 		return -EINVAL;
 	}
 
 	if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_RS600) {
-		if (crtc == 0)
+		if (pipe == 0)
 			return RADEON_READ(R500_D1CRTC_FRAME_COUNT);
 		else
 			return RADEON_READ(R500_D2CRTC_FRAME_COUNT);
 	} else {
-		if (crtc == 0)
+		if (pipe == 0)
 			return RADEON_READ(RADEON_CRTC_CRNT_FRAME);
 		else
 			return RADEON_READ(RADEON_CRTC2_CRNT_FRAME);
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_irq_kms.c b/kernel/drivers/gpu/drm/radeon/radeon_irq_kms.c
index f682e5351..979f3bf65 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -74,11 +74,16 @@ irqreturn_t radeon_driver_irq_handler_kms(int irq, void *arg)
 static void radeon_hotplug_work_func(struct work_struct *work)
 {
 	struct radeon_device *rdev = container_of(work, struct radeon_device,
-						  hotplug_work);
+						  hotplug_work.work);
 	struct drm_device *dev = rdev->ddev;
 	struct drm_mode_config *mode_config = &dev->mode_config;
 	struct drm_connector *connector;
 
+	/* we can race here at startup, some boards seem to trigger
+	 * hotplug irqs when they shouldn't. */
+	if (!rdev->mode_info.mode_config_initialized)
+		return;
+
 	mutex_lock(&mode_config->mutex);
 	if (mode_config->num_connector) {
 		list_for_each_entry(connector, &mode_config->connector_list, head)
@@ -145,7 +150,13 @@ void radeon_driver_irq_preinstall_kms(struct drm_device *dev)
  */
 int radeon_driver_irq_postinstall_kms(struct drm_device *dev)
 {
-	dev->max_vblank_count = 0x001fffff;
+	struct radeon_device *rdev = dev->dev_private;
+
+	if (ASIC_IS_AVIVO(rdev))
+		dev->max_vblank_count = 0x00ffffff;
+	else
+		dev->max_vblank_count = 0x001fffff;
+
 	return 0;
 }
 
@@ -291,7 +302,7 @@ int radeon_irq_kms_init(struct radeon_device *rdev)
 		}
 	}
 
-	INIT_WORK(&rdev->hotplug_work, radeon_hotplug_work_func);
+	INIT_DELAYED_WORK(&rdev->hotplug_work, radeon_hotplug_work_func);
 	INIT_WORK(&rdev->dp_work, radeon_dp_work_func);
 	INIT_WORK(&rdev->audio_work, r600_audio_update_hdmi);
 
@@ -299,7 +310,7 @@ int radeon_irq_kms_init(struct radeon_device *rdev)
 	r = drm_irq_install(rdev->ddev, rdev->ddev->pdev->irq);
 	if (r) {
 		rdev->irq.installed = false;
-		flush_work(&rdev->hotplug_work);
+		flush_delayed_work(&rdev->hotplug_work);
 		return r;
 	}
 
@@ -322,7 +333,7 @@ void radeon_irq_kms_fini(struct radeon_device *rdev)
 		rdev->irq.installed = false;
 		if (rdev->msi_enabled)
 			pci_disable_msi(rdev->pdev);
-		flush_work(&rdev->hotplug_work);
+		flush_delayed_work(&rdev->hotplug_work);
 	}
 }
 
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_kfd.c b/kernel/drivers/gpu/drm/radeon/radeon_kfd.c
index 3db23007c..9a4d69e59 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_kfd.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_kfd.c
@@ -34,6 +34,13 @@
 
 #define CIK_PIPE_PER_MEC	(4)
 
+static const uint32_t watchRegs[MAX_WATCH_ADDRESSES * ADDRESS_WATCH_REG_MAX] = {
+	TCP_WATCH0_ADDR_H, TCP_WATCH0_ADDR_L, TCP_WATCH0_CNTL,
+	TCP_WATCH1_ADDR_H, TCP_WATCH1_ADDR_L, TCP_WATCH1_CNTL,
+	TCP_WATCH2_ADDR_H, TCP_WATCH2_ADDR_L, TCP_WATCH2_CNTL,
+	TCP_WATCH3_ADDR_H, TCP_WATCH3_ADDR_L, TCP_WATCH3_CNTL
+};
+
 struct kgd_mem {
 	struct radeon_bo *bo;
 	uint64_t gpu_addr;
@@ -66,7 +73,7 @@ static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
 
 static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
 				uint32_t hpd_size, uint64_t hpd_gpu_addr);
-
+static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id);
 static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id,
 			uint32_t queue_id, uint32_t __user *wptr);
 static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd);
@@ -79,6 +86,23 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type,
 static bool kgd_hqd_sdma_is_occupied(struct kgd_dev *kgd, void *mqd);
 static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
 				unsigned int timeout);
+static int kgd_address_watch_disable(struct kgd_dev *kgd);
+static int kgd_address_watch_execute(struct kgd_dev *kgd,
+					unsigned int watch_point_id,
+					uint32_t cntl_val,
+					uint32_t addr_hi,
+					uint32_t addr_lo);
+static int kgd_wave_control_execute(struct kgd_dev *kgd,
+					uint32_t gfx_index_val,
+					uint32_t sq_cmd);
+static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
+					unsigned int watch_point_id,
+					unsigned int reg_offset);
+
+static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid);
+static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
+							uint8_t vmid);
+static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid);
 
 static const struct kfd2kgd_calls kfd2kgd = {
 	.init_gtt_mem_allocation = alloc_gtt_mem,
@@ -89,12 +113,20 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.program_sh_mem_settings = kgd_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
 	.init_pipeline = kgd_init_pipeline,
+	.init_interrupts = kgd_init_interrupts,
 	.hqd_load = kgd_hqd_load,
 	.hqd_sdma_load = kgd_hqd_sdma_load,
 	.hqd_is_occupied = kgd_hqd_is_occupied,
 	.hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied,
 	.hqd_destroy = kgd_hqd_destroy,
 	.hqd_sdma_destroy = kgd_hqd_sdma_destroy,
+	.address_watch_disable = kgd_address_watch_disable,
+	.address_watch_execute = kgd_address_watch_execute,
+	.wave_control_execute = kgd_wave_control_execute,
+	.address_watch_get_offset = kgd_address_watch_get_offset,
+	.get_atc_vmid_pasid_mapping_pasid = get_atc_vmid_pasid_mapping_pasid,
+	.get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid,
+	.write_vmid_invalidate_request = write_vmid_invalidate_request,
 	.get_fw_version = get_fw_version
 };
 
@@ -371,8 +403,8 @@ static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
 	 * the SW cleared it.
 	 * So the protocol is to always wait & clear.
 	 */
-	uint32_t pasid_mapping = (pasid == 0) ? 0 :
-				(uint32_t)pasid | ATC_VMID_PASID_MAPPING_VALID;
+	uint32_t pasid_mapping = (pasid == 0) ? 0 : (uint32_t)pasid |
+					ATC_VMID_PASID_MAPPING_VALID_MASK;
 
 	write_register(kgd, ATC_VMID0_PASID_MAPPING + vmid*sizeof(uint32_t),
 			pasid_mapping);
@@ -407,6 +439,24 @@ static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
 	return 0;
 }
 
+static int kgd_init_interrupts(struct kgd_dev *kgd, uint32_t pipe_id)
+{
+	uint32_t mec;
+	uint32_t pipe;
+
+	mec = (pipe_id / CIK_PIPE_PER_MEC) + 1;
+	pipe = (pipe_id % CIK_PIPE_PER_MEC);
+
+	lock_srbm(kgd, mec, pipe, 0, 0);
+
+	write_register(kgd, CPC_INT_CNTL,
+			TIME_STAMP_INT_ENABLE | OPCODE_ERROR_INT_ENABLE);
+
+	unlock_srbm(kgd);
+
+	return 0;
+}
+
 static inline uint32_t get_sdma_base_addr(struct cik_sdma_rlc_registers *m)
 {
 	uint32_t retval;
@@ -646,6 +696,122 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd,
 	return 0;
 }
 
+static int kgd_address_watch_disable(struct kgd_dev *kgd)
+{
+	union TCP_WATCH_CNTL_BITS cntl;
+	unsigned int i;
+
+	cntl.u32All = 0;
+
+	cntl.bitfields.valid = 0;
+	cntl.bitfields.mask = ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK;
+	cntl.bitfields.atc = 1;
+
+	/* Turning off this address until we set all the registers */
+	for (i = 0; i < MAX_WATCH_ADDRESSES; i++)
+		write_register(kgd,
+				watchRegs[i * ADDRESS_WATCH_REG_MAX +
+					ADDRESS_WATCH_REG_CNTL],
+				cntl.u32All);
+
+	return 0;
+}
+
+static int kgd_address_watch_execute(struct kgd_dev *kgd,
+					unsigned int watch_point_id,
+					uint32_t cntl_val,
+					uint32_t addr_hi,
+					uint32_t addr_lo)
+{
+	union TCP_WATCH_CNTL_BITS cntl;
+
+	cntl.u32All = cntl_val;
+
+	/* Turning off this watch point until we set all the registers */
+	cntl.bitfields.valid = 0;
+	write_register(kgd,
+			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
+				ADDRESS_WATCH_REG_CNTL],
+			cntl.u32All);
+
+	write_register(kgd,
+			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
+				ADDRESS_WATCH_REG_ADDR_HI],
+			addr_hi);
+
+	write_register(kgd,
+			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
+				ADDRESS_WATCH_REG_ADDR_LO],
+			addr_lo);
+
+	/* Enable the watch point */
+	cntl.bitfields.valid = 1;
+
+	write_register(kgd,
+			watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX +
+				ADDRESS_WATCH_REG_CNTL],
+			cntl.u32All);
+
+	return 0;
+}
+
+static int kgd_wave_control_execute(struct kgd_dev *kgd,
+					uint32_t gfx_index_val,
+					uint32_t sq_cmd)
+{
+	struct radeon_device *rdev = get_radeon_device(kgd);
+	uint32_t data;
+
+	mutex_lock(&rdev->grbm_idx_mutex);
+
+	write_register(kgd, GRBM_GFX_INDEX, gfx_index_val);
+	write_register(kgd, SQ_CMD, sq_cmd);
+
+	/*  Restore the GRBM_GFX_INDEX register  */
+
+	data = INSTANCE_BROADCAST_WRITES | SH_BROADCAST_WRITES |
+		SE_BROADCAST_WRITES;
+
+	write_register(kgd, GRBM_GFX_INDEX, data);
+
+	mutex_unlock(&rdev->grbm_idx_mutex);
+
+	return 0;
+}
+
+static uint32_t kgd_address_watch_get_offset(struct kgd_dev *kgd,
+					unsigned int watch_point_id,
+					unsigned int reg_offset)
+{
+	return watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + reg_offset];
+}
+
+static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid)
+{
+	uint32_t reg;
+	struct radeon_device *rdev = (struct radeon_device *) kgd;
+
+	reg = RREG32(ATC_VMID0_PASID_MAPPING + vmid*4);
+	return reg & ATC_VMID_PASID_MAPPING_VALID_MASK;
+}
+
+static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
+							uint8_t vmid)
+{
+	uint32_t reg;
+	struct radeon_device *rdev = (struct radeon_device *) kgd;
+
+	reg = RREG32(ATC_VMID0_PASID_MAPPING + vmid*4);
+	return reg & ATC_VMID_PASID_MAPPING_PASID_MASK;
+}
+
+static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid)
+{
+	struct radeon_device *rdev = (struct radeon_device *) kgd;
+
+	return WREG32(VM_INVALIDATE_REQUEST, 1 << vmid);
+}
+
 static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 {
 	struct radeon_device *rdev = (struct radeon_device *) kgd;
@@ -679,7 +845,8 @@ static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 		hdr = (const union radeon_firmware_header *) rdev->rlc_fw->data;
 		break;
 
-	case KGD_ENGINE_SDMA:
+	case KGD_ENGINE_SDMA1:
+	case KGD_ENGINE_SDMA2:
 		hdr = (const union radeon_firmware_header *)
 							rdev->sdma_fw->data;
 		break;
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_kms.c b/kernel/drivers/gpu/drm/radeon/radeon_kms.c
index b0acf50d9..d290a8a09 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_kms.c
@@ -181,7 +181,9 @@ static void radeon_set_filp_rights(struct drm_device *dev,
 				   struct drm_file *applier,
 				   uint32_t *value)
 {
-	mutex_lock(&dev->struct_mutex);
+	struct radeon_device *rdev = dev->dev_private;
+
+	mutex_lock(&rdev->gem.mutex);
 	if (*value == 1) {
 		/* wants rights */
 		if (!*owner)
@@ -192,7 +194,7 @@ static void radeon_set_filp_rights(struct drm_device *dev,
 			*owner = NULL;
 	}
 	*value = *owner == applier ? 1 : 0;
-	mutex_unlock(&dev->struct_mutex);
+	mutex_unlock(&rdev->gem.mutex);
 }
 
 /*
@@ -579,6 +581,9 @@ static int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 	case RADEON_INFO_VA_UNMAP_WORKING:
 		*value = true;
 		break;
+	case RADEON_INFO_GPU_RESET_COUNTER:
+		*value = atomic_read(&rdev->gpu_reset_counter);
+		break;
 	default:
 		DRM_DEBUG_KMS("Invalid request %d\n", info->request);
 		return -EINVAL;
@@ -595,14 +600,17 @@ static int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file
  * Outdated mess for old drm with Xorg being in charge (void function now).
  */
 /**
- * radeon_driver_firstopen_kms - drm callback for last close
+ * radeon_driver_lastclose_kms - drm callback for last close
  *
  * @dev: drm dev pointer
  *
- * Switch vga switcheroo state after last close (all asics).
+ * Switch vga_switcheroo state after last close (all asics).
  */
 void radeon_driver_lastclose_kms(struct drm_device *dev)
 {
+	struct radeon_device *rdev = dev->dev_private;
+
+	radeon_fbdev_restore_mode(rdev);
 	vga_switcheroo_process_delayed_switch();
 }
 
@@ -721,10 +729,14 @@ void radeon_driver_preclose_kms(struct drm_device *dev,
 				struct drm_file *file_priv)
 {
 	struct radeon_device *rdev = dev->dev_private;
+
+	mutex_lock(&rdev->gem.mutex);
 	if (rdev->hyperz_filp == file_priv)
 		rdev->hyperz_filp = NULL;
 	if (rdev->cmask_filp == file_priv)
 		rdev->cmask_filp = NULL;
+	mutex_unlock(&rdev->gem.mutex);
+
 	radeon_uvd_free_handles(rdev, file_priv);
 	radeon_vce_free_handles(rdev, file_priv);
 }
@@ -743,6 +755,8 @@ void radeon_driver_preclose_kms(struct drm_device *dev,
  */
 u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc)
 {
+	int vpos, hpos, stat;
+	u32 count;
 	struct radeon_device *rdev = dev->dev_private;
 
 	if (crtc < 0 || crtc >= rdev->num_crtc) {
@@ -750,7 +764,53 @@ u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc)
 		return -EINVAL;
 	}
 
-	return radeon_get_vblank_counter(rdev, crtc);
+	/* The hw increments its frame counter at start of vsync, not at start
+	 * of vblank, as is required by DRM core vblank counter handling.
+	 * Cook the hw count here to make it appear to the caller as if it
+	 * incremented at start of vblank. We measure distance to start of
+	 * vblank in vpos. vpos therefore will be >= 0 between start of vblank
+	 * and start of vsync, so vpos >= 0 means to bump the hw frame counter
+	 * result by 1 to give the proper appearance to caller.
+	 */
+	if (rdev->mode_info.crtcs[crtc]) {
+		/* Repeat readout if needed to provide stable result if
+		 * we cross start of vsync during the queries.
+		 */
+		do {
+			count = radeon_get_vblank_counter(rdev, crtc);
+			/* Ask radeon_get_crtc_scanoutpos to return vpos as
+			 * distance to start of vblank, instead of regular
+			 * vertical scanout pos.
+			 */
+			stat = radeon_get_crtc_scanoutpos(
+				dev, crtc, GET_DISTANCE_TO_VBLANKSTART,
+				&vpos, &hpos, NULL, NULL,
+				&rdev->mode_info.crtcs[crtc]->base.hwmode);
+		} while (count != radeon_get_vblank_counter(rdev, crtc));
+
+		if (((stat & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE)) !=
+		    (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE))) {
+			DRM_DEBUG_VBL("Query failed! stat %d\n", stat);
+		}
+		else {
+			DRM_DEBUG_VBL("crtc %d: dist from vblank start %d\n",
+				      crtc, vpos);
+
+			/* Bump counter if we are at >= leading edge of vblank,
+			 * but before vsync where vpos would turn negative and
+			 * the hw counter really increments.
+			 */
+			if (vpos >= 0)
+				count++;
+		}
+	}
+	else {
+	    /* Fallback to use value as is. */
+	    count = radeon_get_vblank_counter(rdev, crtc);
+	    DRM_DEBUG_VBL("NULL mode info! Returned count may be wrong.\n");
+	}
+
+	return count;
 }
 
 /**
@@ -838,92 +898,52 @@ int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
 	/* Helper routine in DRM core does all the work: */
 	return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error,
 						     vblank_time, flags,
-						     drmcrtc, &drmcrtc->hwmode);
-}
-
-#define KMS_INVALID_IOCTL(name)						\
-static int name(struct drm_device *dev, void *data, struct drm_file	\
-		*file_priv)						\
-{									\
-	DRM_ERROR("invalid ioctl with kms %s\n", __func__);		\
-	return -EINVAL;							\
+						     &drmcrtc->hwmode);
 }
 
-/*
- * All these ioctls are invalid in kms world.
- */
-KMS_INVALID_IOCTL(radeon_cp_init_kms)
-KMS_INVALID_IOCTL(radeon_cp_start_kms)
-KMS_INVALID_IOCTL(radeon_cp_stop_kms)
-KMS_INVALID_IOCTL(radeon_cp_reset_kms)
-KMS_INVALID_IOCTL(radeon_cp_idle_kms)
-KMS_INVALID_IOCTL(radeon_cp_resume_kms)
-KMS_INVALID_IOCTL(radeon_engine_reset_kms)
-KMS_INVALID_IOCTL(radeon_fullscreen_kms)
-KMS_INVALID_IOCTL(radeon_cp_swap_kms)
-KMS_INVALID_IOCTL(radeon_cp_clear_kms)
-KMS_INVALID_IOCTL(radeon_cp_vertex_kms)
-KMS_INVALID_IOCTL(radeon_cp_indices_kms)
-KMS_INVALID_IOCTL(radeon_cp_texture_kms)
-KMS_INVALID_IOCTL(radeon_cp_stipple_kms)
-KMS_INVALID_IOCTL(radeon_cp_indirect_kms)
-KMS_INVALID_IOCTL(radeon_cp_vertex2_kms)
-KMS_INVALID_IOCTL(radeon_cp_cmdbuf_kms)
-KMS_INVALID_IOCTL(radeon_cp_getparam_kms)
-KMS_INVALID_IOCTL(radeon_cp_flip_kms)
-KMS_INVALID_IOCTL(radeon_mem_alloc_kms)
-KMS_INVALID_IOCTL(radeon_mem_free_kms)
-KMS_INVALID_IOCTL(radeon_mem_init_heap_kms)
-KMS_INVALID_IOCTL(radeon_irq_emit_kms)
-KMS_INVALID_IOCTL(radeon_irq_wait_kms)
-KMS_INVALID_IOCTL(radeon_cp_setparam_kms)
-KMS_INVALID_IOCTL(radeon_surface_alloc_kms)
-KMS_INVALID_IOCTL(radeon_surface_free_kms)
-
-
 const struct drm_ioctl_desc radeon_ioctls_kms[] = {
-	DRM_IOCTL_DEF_DRV(RADEON_CP_INIT, radeon_cp_init_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_START, radeon_cp_start_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_STOP, radeon_cp_stop_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_RESET, radeon_cp_reset_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_IDLE, radeon_cp_idle_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_RESUME, radeon_cp_resume_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_RESET, radeon_engine_reset_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_FULLSCREEN, radeon_fullscreen_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SWAP, radeon_cp_swap_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_CLEAR, radeon_cp_clear_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_VERTEX, radeon_cp_vertex_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_INDICES, radeon_cp_indices_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_TEXTURE, radeon_cp_texture_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_STIPPLE, radeon_cp_stipple_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_INDIRECT, radeon_cp_indirect_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_VERTEX2, radeon_cp_vertex2_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_CMDBUF, radeon_cp_cmdbuf_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_GETPARAM, radeon_cp_getparam_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_FLIP, radeon_cp_flip_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_ALLOC, radeon_mem_alloc_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_FREE, radeon_mem_free_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_INIT_HEAP, radeon_mem_init_heap_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_IRQ_EMIT, radeon_irq_emit_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_IRQ_WAIT, radeon_irq_wait_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SETPARAM, radeon_cp_setparam_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SURF_ALLOC, radeon_surface_alloc_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SURF_FREE, radeon_surface_free_kms, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_INIT, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_START, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_STOP, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_RESET, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_IDLE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_RESUME, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_RESET, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_FULLSCREEN, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SWAP, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CLEAR, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_VERTEX, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_INDICES, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_TEXTURE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_STIPPLE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_INDIRECT, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_VERTEX2, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CMDBUF, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_GETPARAM, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_FLIP, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_ALLOC, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_FREE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_INIT_HEAP, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_IRQ_EMIT, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_IRQ_WAIT, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SETPARAM, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SURF_ALLOC, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SURF_FREE, drm_invalid_op, DRM_AUTH),
 	/* KMS */
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_INFO, radeon_gem_info_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_CREATE, radeon_gem_create_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_MMAP, radeon_gem_mmap_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_DOMAIN, radeon_gem_set_domain_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_PREAD, radeon_gem_pread_ioctl, DRM_AUTH|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_PWRITE, radeon_gem_pwrite_ioctl, DRM_AUTH|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_WAIT_IDLE, radeon_gem_wait_idle_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_CS, radeon_cs_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_INFO, radeon_info_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_TILING, radeon_gem_set_tiling_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_GET_TILING, radeon_gem_get_tiling_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_BUSY, radeon_gem_busy_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_VA, radeon_gem_va_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_OP, radeon_gem_op_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_USERPTR, radeon_gem_userptr_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_INFO, radeon_gem_info_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_CREATE, radeon_gem_create_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_MMAP, radeon_gem_mmap_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_DOMAIN, radeon_gem_set_domain_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_PREAD, radeon_gem_pread_ioctl, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_PWRITE, radeon_gem_pwrite_ioctl, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_WAIT_IDLE, radeon_gem_wait_idle_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_CS, radeon_cs_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_INFO, radeon_info_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_TILING, radeon_gem_set_tiling_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_GET_TILING, radeon_gem_get_tiling_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_BUSY, radeon_gem_busy_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_VA, radeon_gem_va_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_OP, radeon_gem_op_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_USERPTR, radeon_gem_userptr_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 };
 int radeon_max_kms_ioctl = ARRAY_SIZE(radeon_ioctls_kms);
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/kernel/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
index 45715307d..30de43366 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
@@ -441,6 +441,7 @@ void radeon_legacy_backlight_init(struct radeon_encoder *radeon_encoder,
 	backlight_update_status(bd);
 
 	DRM_INFO("radeon legacy LVDS backlight initialized\n");
+	rdev->mode_info.bl_encoder = radeon_encoder;
 
 	return;
 
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_mode.h b/kernel/drivers/gpu/drm/radeon/radeon_mode.h
index 9af2d8398..bba112628 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/kernel/drivers/gpu/drm/radeon/radeon_mode.h
@@ -367,6 +367,7 @@ struct radeon_crtc {
 	u32 line_time;
 	u32 wm_low;
 	u32 wm_high;
+	u32 lb_vblank_lead_lines;
 	struct drm_display_mode hw_mode;
 	enum radeon_output_csc output_csc;
 };
@@ -553,6 +554,7 @@ struct radeon_connector {
 	void *con_priv;
 	bool dac_load_detect;
 	bool detected_by_load; /* if the connection status was determined by load */
+	bool detected_hpd_without_ddc; /* if an HPD signal was detected on DVI, but ddc probing failed */
 	uint16_t connector_object_id;
 	struct radeon_hpd hpd;
 	struct radeon_router router;
@@ -686,6 +688,9 @@ struct atom_voltage_table
 	struct atom_voltage_table_entry entries[MAX_VOLTAGE_ENTRIES];
 };
 
+/* Driver internal use only flags of radeon_get_crtc_scanoutpos() */
+#define USE_REAL_VBLANKSTART 		(1 << 30)
+#define GET_DISTANCE_TO_VBLANKSTART	(1 << 31)
 
 extern void
 radeon_add_atom_connector(struct drm_device *dev,
@@ -753,7 +758,7 @@ extern bool radeon_dp_getdpcd(struct radeon_connector *radeon_connector);
 extern int radeon_dp_get_panel_mode(struct drm_encoder *encoder,
 				    struct drm_connector *connector);
 int radeon_dp_get_max_link_rate(struct drm_connector *connector,
-				u8 *dpcd);
+				const u8 *dpcd);
 extern void radeon_dp_set_rx_power_state(struct drm_connector *connector,
 					 u8 power_state);
 extern void radeon_dp_aux_init(struct radeon_connector *radeon_connector);
@@ -874,10 +879,10 @@ extern int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 				   int x, int y);
 extern void radeon_cursor_reset(struct drm_crtc *crtc);
 
-extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
-				      unsigned int flags,
-				      int *vpos, int *hpos, ktime_t *stime,
-				      ktime_t *etime);
+extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
+				      unsigned int flags, int *vpos, int *hpos,
+				      ktime_t *stime, ktime_t *etime,
+				      const struct drm_display_mode *mode);
 
 extern bool radeon_combios_check_hardcoded_edid(struct radeon_device *rdev);
 extern struct edid *
@@ -980,6 +985,7 @@ int radeon_fbdev_init(struct radeon_device *rdev);
 void radeon_fbdev_fini(struct radeon_device *rdev);
 void radeon_fbdev_set_suspend(struct radeon_device *rdev, int state);
 bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj);
+void radeon_fbdev_restore_mode(struct radeon_device *rdev);
 
 void radeon_fb_output_poll_changed(struct radeon_device *rdev);
 
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_object.c b/kernel/drivers/gpu/drm/radeon/radeon_object.c
index 676362769..fb6ad1438 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_object.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_object.c
@@ -33,6 +33,7 @@
 #include <linux/slab.h>
 #include <drm/drmP.h>
 #include <drm/radeon_drm.h>
+#include <drm/drm_cache.h>
 #include "radeon.h"
 #include "radeon_trace.h"
 
@@ -221,11 +222,17 @@ int radeon_bo_create(struct radeon_device *rdev,
 	if (!(rdev->flags & RADEON_IS_PCIE))
 		bo->flags &= ~(RADEON_GEM_GTT_WC | RADEON_GEM_GTT_UC);
 
+	/* Write-combined CPU mappings of GTT cause GPU hangs with RV6xx
+	 * See https://bugs.freedesktop.org/show_bug.cgi?id=91268
+	 */
+	if (rdev->family >= CHIP_RV610 && rdev->family <= CHIP_RV635)
+		bo->flags &= ~(RADEON_GEM_GTT_WC | RADEON_GEM_GTT_UC);
+
 #ifdef CONFIG_X86_32
 	/* XXX: Write-combined CPU mappings of GTT seem broken on 32-bit
 	 * See https://bugs.freedesktop.org/show_bug.cgi?id=84627
 	 */
-	bo->flags &= ~RADEON_GEM_GTT_WC;
+	bo->flags &= ~(RADEON_GEM_GTT_WC | RADEON_GEM_GTT_UC);
 #elif defined(CONFIG_X86) && !defined(CONFIG_X86_PAT)
 	/* Don't try to enable write-combining when it can't work, or things
 	 * may be slow
@@ -235,9 +242,16 @@ int radeon_bo_create(struct radeon_device *rdev,
 #warning Please enable CONFIG_MTRR and CONFIG_X86_PAT for better performance \
 	 thanks to write-combining
 
-	DRM_INFO_ONCE("Please enable CONFIG_MTRR and CONFIG_X86_PAT for "
-		      "better performance thanks to write-combining\n");
-	bo->flags &= ~RADEON_GEM_GTT_WC;
+	if (bo->flags & RADEON_GEM_GTT_WC)
+		DRM_INFO_ONCE("Please enable CONFIG_MTRR and CONFIG_X86_PAT for "
+			      "better performance thanks to write-combining\n");
+	bo->flags &= ~(RADEON_GEM_GTT_WC | RADEON_GEM_GTT_UC);
+#else
+	/* For architectures that don't support WC memory,
+	 * mask out the WC flag from the BO
+	 */
+	if (!drm_arch_can_wc_memory())
+		bo->flags &= ~RADEON_GEM_GTT_WC;
 #endif
 
 	radeon_ttm_placement_from_domain(bo, domain);
@@ -419,7 +433,6 @@ void radeon_bo_force_delete(struct radeon_device *rdev)
 	}
 	dev_err(rdev->dev, "Userspace still has active objects !\n");
 	list_for_each_entry_safe(bo, n, &rdev->gem.objects, list) {
-		mutex_lock(&rdev->ddev->struct_mutex);
 		dev_err(rdev->dev, "%p %p %lu %lu force free\n",
 			&bo->gem_base, bo, (unsigned long)bo->gem_base.size,
 			*((unsigned long *)&bo->gem_base.refcount));
@@ -427,8 +440,7 @@ void radeon_bo_force_delete(struct radeon_device *rdev)
 		list_del_init(&bo->list);
 		mutex_unlock(&bo->rdev->gem.mutex);
 		/* this should unref the ttm bo */
-		drm_gem_object_unreference(&bo->gem_base);
-		mutex_unlock(&rdev->ddev->struct_mutex);
+		drm_gem_object_unreference_unlocked(&bo->gem_base);
 	}
 }
 
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_pm.c b/kernel/drivers/gpu/drm/radeon/radeon_pm.c
index c1ba83a8d..60ab31517 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_pm.c
@@ -253,7 +253,6 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev)
 	    (rdev->pm.requested_power_state_index == rdev->pm.current_power_state_index))
 		return;
 
-	mutex_lock(&rdev->ddev->struct_mutex);
 	down_write(&rdev->pm.mclk_lock);
 	mutex_lock(&rdev->ring_lock);
 
@@ -268,7 +267,6 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev)
 			/* needs a GPU reset dont reset here */
 			mutex_unlock(&rdev->ring_lock);
 			up_write(&rdev->pm.mclk_lock);
-			mutex_unlock(&rdev->ddev->struct_mutex);
 			return;
 		}
 	}
@@ -304,7 +302,6 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev)
 
 	mutex_unlock(&rdev->ring_lock);
 	up_write(&rdev->pm.mclk_lock);
-	mutex_unlock(&rdev->ddev->struct_mutex);
 }
 
 static void radeon_pm_print_states(struct radeon_device *rdev)
@@ -720,10 +717,14 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	struct radeon_device *rdev = dev_get_drvdata(dev);
 	umode_t effective_mode = attr->mode;
 
-	/* Skip limit attributes if DPM is not enabled */
+	/* Skip attributes if DPM is not enabled */
 	if (rdev->pm.pm_method != PM_METHOD_DPM &&
 	    (attr == &sensor_dev_attr_temp1_crit.dev_attr.attr ||
-	     attr == &sensor_dev_attr_temp1_crit_hyst.dev_attr.attr))
+	     attr == &sensor_dev_attr_temp1_crit_hyst.dev_attr.attr ||
+	     attr == &sensor_dev_attr_pwm1.dev_attr.attr ||
+	     attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr ||
+	     attr == &sensor_dev_attr_pwm1_max.dev_attr.attr ||
+	     attr == &sensor_dev_attr_pwm1_min.dev_attr.attr))
 		return 0;
 
 	/* Skip fan attributes if fan is not present */
@@ -1062,7 +1063,6 @@ force:
 		radeon_dpm_print_power_state(rdev, rdev->pm.dpm.requested_ps);
 	}
 
-	mutex_lock(&rdev->ddev->struct_mutex);
 	down_write(&rdev->pm.mclk_lock);
 	mutex_lock(&rdev->ring_lock);
 
@@ -1078,10 +1078,6 @@ force:
 	/* update displays */
 	radeon_dpm_display_configuration_changed(rdev);
 
-	rdev->pm.dpm.current_active_crtcs = rdev->pm.dpm.new_active_crtcs;
-	rdev->pm.dpm.current_active_crtc_count = rdev->pm.dpm.new_active_crtc_count;
-	rdev->pm.dpm.single_display = single_display;
-
 	/* wait for the rings to drain */
 	for (i = 0; i < RADEON_NUM_RINGS; i++) {
 		struct radeon_ring *ring = &rdev->ring[i];
@@ -1097,6 +1093,10 @@ force:
 
 	radeon_dpm_post_set_power_state(rdev);
 
+	rdev->pm.dpm.current_active_crtcs = rdev->pm.dpm.new_active_crtcs;
+	rdev->pm.dpm.current_active_crtc_count = rdev->pm.dpm.new_active_crtc_count;
+	rdev->pm.dpm.single_display = single_display;
+
 	if (rdev->asic->dpm.force_performance_level) {
 		if (rdev->pm.dpm.thermal_active) {
 			enum radeon_dpm_forced_level level = rdev->pm.dpm.forced_level;
@@ -1113,7 +1113,6 @@ force:
 done:
 	mutex_unlock(&rdev->ring_lock);
 	up_write(&rdev->pm.mclk_lock);
-	mutex_unlock(&rdev->ddev->struct_mutex);
 }
 
 void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
@@ -1331,14 +1330,6 @@ static int radeon_pm_init_old(struct radeon_device *rdev)
 	INIT_DELAYED_WORK(&rdev->pm.dynpm_idle_work, radeon_dynpm_idle_work_handler);
 
 	if (rdev->pm.num_power_states > 1) {
-		/* where's the best place to put these? */
-		ret = device_create_file(rdev->dev, &dev_attr_power_profile);
-		if (ret)
-			DRM_ERROR("failed to create device file for power profile\n");
-		ret = device_create_file(rdev->dev, &dev_attr_power_method);
-		if (ret)
-			DRM_ERROR("failed to create device file for power method\n");
-
 		if (radeon_debugfs_pm_init(rdev)) {
 			DRM_ERROR("Failed to register debugfs file for PM!\n");
 		}
@@ -1396,20 +1387,6 @@ static int radeon_pm_init_dpm(struct radeon_device *rdev)
 		goto dpm_failed;
 	rdev->pm.dpm_enabled = true;
 
-	ret = device_create_file(rdev->dev, &dev_attr_power_dpm_state);
-	if (ret)
-		DRM_ERROR("failed to create device file for dpm state\n");
-	ret = device_create_file(rdev->dev, &dev_attr_power_dpm_force_performance_level);
-	if (ret)
-		DRM_ERROR("failed to create device file for dpm state\n");
-	/* XXX: these are noops for dpm but are here for backwards compat */
-	ret = device_create_file(rdev->dev, &dev_attr_power_profile);
-	if (ret)
-		DRM_ERROR("failed to create device file for power profile\n");
-	ret = device_create_file(rdev->dev, &dev_attr_power_method);
-	if (ret)
-		DRM_ERROR("failed to create device file for power method\n");
-
 	if (radeon_debugfs_pm_init(rdev)) {
 		DRM_ERROR("Failed to register debugfs file for dpm!\n");
 	}
@@ -1550,9 +1527,50 @@ int radeon_pm_late_init(struct radeon_device *rdev)
 	int ret = 0;
 
 	if (rdev->pm.pm_method == PM_METHOD_DPM) {
-		mutex_lock(&rdev->pm.mutex);
-		ret = radeon_dpm_late_enable(rdev);
-		mutex_unlock(&rdev->pm.mutex);
+		if (rdev->pm.dpm_enabled) {
+			if (!rdev->pm.sysfs_initialized) {
+				ret = device_create_file(rdev->dev, &dev_attr_power_dpm_state);
+				if (ret)
+					DRM_ERROR("failed to create device file for dpm state\n");
+				ret = device_create_file(rdev->dev, &dev_attr_power_dpm_force_performance_level);
+				if (ret)
+					DRM_ERROR("failed to create device file for dpm state\n");
+				/* XXX: these are noops for dpm but are here for backwards compat */
+				ret = device_create_file(rdev->dev, &dev_attr_power_profile);
+				if (ret)
+					DRM_ERROR("failed to create device file for power profile\n");
+				ret = device_create_file(rdev->dev, &dev_attr_power_method);
+				if (ret)
+					DRM_ERROR("failed to create device file for power method\n");
+				rdev->pm.sysfs_initialized = true;
+			}
+
+			mutex_lock(&rdev->pm.mutex);
+			ret = radeon_dpm_late_enable(rdev);
+			mutex_unlock(&rdev->pm.mutex);
+			if (ret) {
+				rdev->pm.dpm_enabled = false;
+				DRM_ERROR("radeon_pm_late_init failed, disabling dpm\n");
+			} else {
+				/* set the dpm state for PX since there won't be
+				 * a modeset to call this.
+				 */
+				radeon_pm_compute_clocks(rdev);
+			}
+		}
+	} else {
+		if ((rdev->pm.num_power_states > 1) &&
+		    (!rdev->pm.sysfs_initialized)) {
+			/* where's the best place to put these? */
+			ret = device_create_file(rdev->dev, &dev_attr_power_profile);
+			if (ret)
+				DRM_ERROR("failed to create device file for power profile\n");
+			ret = device_create_file(rdev->dev, &dev_attr_power_method);
+			if (ret)
+				DRM_ERROR("failed to create device file for power method\n");
+			if (!ret)
+				rdev->pm.sysfs_initialized = true;
+		}
 	}
 	return ret;
 }
@@ -1738,7 +1756,11 @@ static bool radeon_pm_in_vbl(struct radeon_device *rdev)
 	 */
 	for (crtc = 0; (crtc < rdev->num_crtc) && in_vbl; crtc++) {
 		if (rdev->pm.active_crtcs & (1 << crtc)) {
-			vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, crtc, 0, &vpos, &hpos, NULL, NULL);
+			vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev,
+								crtc,
+								USE_REAL_VBLANKSTART,
+								&vpos, &hpos, NULL, NULL,
+								&rdev->mode_info.crtcs[crtc]->base.hwmode);
 			if ((vbl_status & DRM_SCANOUTPOS_VALID) &&
 			    !(vbl_status & DRM_SCANOUTPOS_IN_VBLANK))
 				in_vbl = false;
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_sa.c b/kernel/drivers/gpu/drm/radeon/radeon_sa.c
index c507896ac..197b157b7 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_sa.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_sa.c
@@ -349,8 +349,13 @@ int radeon_sa_bo_new(struct radeon_device *rdev,
 			/* see if we can skip over some allocations */
 		} while (radeon_sa_bo_next_hole(sa_manager, fences, tries));
 
+		for (i = 0; i < RADEON_NUM_RINGS; ++i)
+			radeon_fence_ref(fences[i]);
+
 		spin_unlock(&sa_manager->wq.lock);
 		r = radeon_fence_wait_any(rdev, fences, false);
+		for (i = 0; i < RADEON_NUM_RINGS; ++i)
+			radeon_fence_unref(&fences[i]);
 		spin_lock(&sa_manager->wq.lock);
 		/* if we have nothing to wait for block */
 		if (r == -ENOENT) {
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_ttm.c b/kernel/drivers/gpu/drm/radeon/radeon_ttm.c
index edafd3c2b..e06ac546a 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -144,7 +144,7 @@ static int radeon_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
 		man->available_caching = TTM_PL_MASK_CACHING;
 		man->default_caching = TTM_PL_FLAG_CACHED;
 		man->flags = TTM_MEMTYPE_FLAG_MAPPABLE | TTM_MEMTYPE_FLAG_CMA;
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (rdev->flags & RADEON_IS_AGP) {
 			if (!rdev->ddev->agp) {
 				DRM_ERROR("AGP is not enabled for memory type %u\n",
@@ -461,7 +461,7 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 		/* system memory */
 		return 0;
 	case TTM_PL_TT:
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (rdev->flags & RADEON_IS_AGP) {
 			/* RADEON_IS_AGP is set only if AGP is active */
 			mem->bus.offset = mem->start << PAGE_SHIFT;
@@ -680,7 +680,7 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_bo_device *bdev,
 	struct radeon_ttm_tt *gtt;
 
 	rdev = radeon_get_rdev(bdev);
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
 		return ttm_agp_tt_create(bdev, rdev->ddev->agp->bridge,
 					 size, page_flags, dummy_read_page);
@@ -719,7 +719,7 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm)
 		return 0;
 
 	if (gtt && gtt->userptr) {
-		ttm->sg = kcalloc(1, sizeof(struct sg_table), GFP_KERNEL);
+		ttm->sg = kzalloc(sizeof(struct sg_table), GFP_KERNEL);
 		if (!ttm->sg)
 			return -ENOMEM;
 
@@ -736,7 +736,7 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm)
 	}
 
 	rdev = radeon_get_rdev(ttm->bdev);
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
 		return ttm_agp_tt_populate(ttm);
 	}
@@ -758,7 +758,7 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm)
 						       0, PAGE_SIZE,
 						       PCI_DMA_BIDIRECTIONAL);
 		if (pci_dma_mapping_error(rdev->pdev, gtt->ttm.dma_address[i])) {
-			while (--i) {
+			while (i--) {
 				pci_unmap_page(rdev->pdev, gtt->ttm.dma_address[i],
 					       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
 				gtt->ttm.dma_address[i] = 0;
@@ -787,7 +787,7 @@ static void radeon_ttm_tt_unpopulate(struct ttm_tt *ttm)
 		return;
 
 	rdev = radeon_get_rdev(ttm->bdev);
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
 		ttm_agp_tt_unpopulate(ttm);
 		return;
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_vce.c b/kernel/drivers/gpu/drm/radeon/radeon_vce.c
index 0de5711ac..7eb1ae758 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_vce.c
@@ -38,8 +38,10 @@
 #define VCE_IDLE_TIMEOUT_MS	1000
 
 /* Firmware Names */
+#define FIRMWARE_TAHITI	"radeon/TAHITI_vce.bin"
 #define FIRMWARE_BONAIRE	"radeon/BONAIRE_vce.bin"
 
+MODULE_FIRMWARE(FIRMWARE_TAHITI);
 MODULE_FIRMWARE(FIRMWARE_BONAIRE);
 
 static void radeon_vce_idle_work_handler(struct work_struct *work);
@@ -63,6 +65,14 @@ int radeon_vce_init(struct radeon_device *rdev)
 	INIT_DELAYED_WORK(&rdev->vce.idle_work, radeon_vce_idle_work_handler);
 
 	switch (rdev->family) {
+	case CHIP_TAHITI:
+	case CHIP_PITCAIRN:
+	case CHIP_VERDE:
+	case CHIP_OLAND:
+	case CHIP_ARUBA:
+		fw_name = FIRMWARE_TAHITI;
+		break;
+
 	case CHIP_BONAIRE:
 	case CHIP_KAVERI:
 	case CHIP_KABINI:
@@ -118,13 +128,17 @@ int radeon_vce_init(struct radeon_device *rdev)
 	rdev->vce.fw_version = (start << 24) | (mid << 16) | (end << 8);
 
 	/* we can only work with this fw version for now */
-	if (rdev->vce.fw_version != ((40 << 24) | (2 << 16) | (2 << 8)))
+	if ((rdev->vce.fw_version != ((40 << 24) | (2 << 16) | (2 << 8))) &&
+	    (rdev->vce.fw_version != ((50 << 24) | (0 << 16) | (1 << 8))) &&
+	    (rdev->vce.fw_version != ((50 << 24) | (1 << 16) | (2 << 8))))
 		return -EINVAL;
 
 	/* allocate firmware, stack and heap BO */
 
-	size = RADEON_GPU_PAGE_ALIGN(rdev->vce_fw->size) +
-	       RADEON_VCE_STACK_SIZE + RADEON_VCE_HEAP_SIZE;
+	if (rdev->family < CHIP_BONAIRE)
+		size = vce_v1_0_bo_size(rdev);
+	else
+		size = vce_v2_0_bo_size(rdev);
 	r = radeon_bo_create(rdev, size, PAGE_SIZE, true,
 			     RADEON_GEM_DOMAIN_VRAM, 0, NULL, NULL,
 			     &rdev->vce.vcpu_bo);
@@ -225,13 +239,17 @@ int radeon_vce_resume(struct radeon_device *rdev)
 		return r;
 	}
 
-	memcpy(cpu_addr, rdev->vce_fw->data, rdev->vce_fw->size);
+	memset(cpu_addr, 0, radeon_bo_size(rdev->vce.vcpu_bo));
+	if (rdev->family < CHIP_BONAIRE)
+		r = vce_v1_0_load_fw(rdev, cpu_addr);
+	else
+		memcpy(cpu_addr, rdev->vce_fw->data, rdev->vce_fw->size);
 
 	radeon_bo_kunmap(rdev->vce.vcpu_bo);
 
 	radeon_bo_unreserve(rdev->vce.vcpu_bo);
 
-	return 0;
+	return r;
 }
 
 /**
@@ -343,31 +361,31 @@ int radeon_vce_get_create_msg(struct radeon_device *rdev, int ring,
 
 	/* stitch together an VCE create msg */
 	ib.length_dw = 0;
-	ib.ptr[ib.length_dw++] = 0x0000000c; /* len */
-	ib.ptr[ib.length_dw++] = 0x00000001; /* session cmd */
-	ib.ptr[ib.length_dw++] = handle;
-
-	ib.ptr[ib.length_dw++] = 0x00000030; /* len */
-	ib.ptr[ib.length_dw++] = 0x01000001; /* create cmd */
-	ib.ptr[ib.length_dw++] = 0x00000000;
-	ib.ptr[ib.length_dw++] = 0x00000042;
-	ib.ptr[ib.length_dw++] = 0x0000000a;
-	ib.ptr[ib.length_dw++] = 0x00000001;
-	ib.ptr[ib.length_dw++] = 0x00000080;
-	ib.ptr[ib.length_dw++] = 0x00000060;
-	ib.ptr[ib.length_dw++] = 0x00000100;
-	ib.ptr[ib.length_dw++] = 0x00000100;
-	ib.ptr[ib.length_dw++] = 0x0000000c;
-	ib.ptr[ib.length_dw++] = 0x00000000;
-
-	ib.ptr[ib.length_dw++] = 0x00000014; /* len */
-	ib.ptr[ib.length_dw++] = 0x05000005; /* feedback buffer */
-	ib.ptr[ib.length_dw++] = upper_32_bits(dummy);
-	ib.ptr[ib.length_dw++] = dummy;
-	ib.ptr[ib.length_dw++] = 0x00000001;
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x0000000c); /* len */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000001); /* session cmd */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(handle);
+
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000030); /* len */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x01000001); /* create cmd */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000000);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000042);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x0000000a);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000001);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000080);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000060);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000100);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000100);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x0000000c);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000000);
+
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000014); /* len */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x05000005); /* feedback buffer */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(upper_32_bits(dummy));
+	ib.ptr[ib.length_dw++] = cpu_to_le32(dummy);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000001);
 
 	for (i = ib.length_dw; i < ib_size_dw; ++i)
-		ib.ptr[i] = 0x0;
+		ib.ptr[i] = cpu_to_le32(0x0);
 
 	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
@@ -410,21 +428,21 @@ int radeon_vce_get_destroy_msg(struct radeon_device *rdev, int ring,
 
 	/* stitch together an VCE destroy msg */
 	ib.length_dw = 0;
-	ib.ptr[ib.length_dw++] = 0x0000000c; /* len */
-	ib.ptr[ib.length_dw++] = 0x00000001; /* session cmd */
-	ib.ptr[ib.length_dw++] = handle;
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x0000000c); /* len */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000001); /* session cmd */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(handle);
 
-	ib.ptr[ib.length_dw++] = 0x00000014; /* len */
-	ib.ptr[ib.length_dw++] = 0x05000005; /* feedback buffer */
-	ib.ptr[ib.length_dw++] = upper_32_bits(dummy);
-	ib.ptr[ib.length_dw++] = dummy;
-	ib.ptr[ib.length_dw++] = 0x00000001;
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000014); /* len */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x05000005); /* feedback buffer */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(upper_32_bits(dummy));
+	ib.ptr[ib.length_dw++] = cpu_to_le32(dummy);
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000001);
 
-	ib.ptr[ib.length_dw++] = 0x00000008; /* len */
-	ib.ptr[ib.length_dw++] = 0x02000001; /* destroy cmd */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000008); /* len */
+	ib.ptr[ib.length_dw++] = cpu_to_le32(0x02000001); /* destroy cmd */
 
 	for (i = ib.length_dw; i < ib_size_dw; ++i)
-		ib.ptr[i] = 0x0;
+		ib.ptr[i] = cpu_to_le32(0x0);
 
 	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
@@ -681,12 +699,12 @@ bool radeon_vce_semaphore_emit(struct radeon_device *rdev,
 {
 	uint64_t addr = semaphore->gpu_addr;
 
-	radeon_ring_write(ring, VCE_CMD_SEMAPHORE);
-	radeon_ring_write(ring, (addr >> 3) & 0x000FFFFF);
-	radeon_ring_write(ring, (addr >> 23) & 0x000FFFFF);
-	radeon_ring_write(ring, 0x01003000 | (emit_wait ? 1 : 0));
+	radeon_ring_write(ring, cpu_to_le32(VCE_CMD_SEMAPHORE));
+	radeon_ring_write(ring, cpu_to_le32((addr >> 3) & 0x000FFFFF));
+	radeon_ring_write(ring, cpu_to_le32((addr >> 23) & 0x000FFFFF));
+	radeon_ring_write(ring, cpu_to_le32(0x01003000 | (emit_wait ? 1 : 0)));
 	if (!emit_wait)
-		radeon_ring_write(ring, VCE_CMD_END);
+		radeon_ring_write(ring, cpu_to_le32(VCE_CMD_END));
 
 	return true;
 }
@@ -701,10 +719,10 @@ bool radeon_vce_semaphore_emit(struct radeon_device *rdev,
 void radeon_vce_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
 	struct radeon_ring *ring = &rdev->ring[ib->ring];
-	radeon_ring_write(ring, VCE_CMD_IB);
-	radeon_ring_write(ring, ib->gpu_addr);
-	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr));
-	radeon_ring_write(ring, ib->length_dw);
+	radeon_ring_write(ring, cpu_to_le32(VCE_CMD_IB));
+	radeon_ring_write(ring, cpu_to_le32(ib->gpu_addr));
+	radeon_ring_write(ring, cpu_to_le32(upper_32_bits(ib->gpu_addr)));
+	radeon_ring_write(ring, cpu_to_le32(ib->length_dw));
 }
 
 /**
@@ -720,12 +738,12 @@ void radeon_vce_fence_emit(struct radeon_device *rdev,
 	struct radeon_ring *ring = &rdev->ring[fence->ring];
 	uint64_t addr = rdev->fence_drv[fence->ring].gpu_addr;
 
-	radeon_ring_write(ring, VCE_CMD_FENCE);
-	radeon_ring_write(ring, addr);
-	radeon_ring_write(ring, upper_32_bits(addr));
-	radeon_ring_write(ring, fence->seq);
-	radeon_ring_write(ring, VCE_CMD_TRAP);
-	radeon_ring_write(ring, VCE_CMD_END);
+	radeon_ring_write(ring, cpu_to_le32(VCE_CMD_FENCE));
+	radeon_ring_write(ring, cpu_to_le32(addr));
+	radeon_ring_write(ring, cpu_to_le32(upper_32_bits(addr)));
+	radeon_ring_write(ring, cpu_to_le32(fence->seq));
+	radeon_ring_write(ring, cpu_to_le32(VCE_CMD_TRAP));
+	radeon_ring_write(ring, cpu_to_le32(VCE_CMD_END));
 }
 
 /**
@@ -747,7 +765,7 @@ int radeon_vce_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 			  ring->idx, r);
 		return r;
 	}
-	radeon_ring_write(ring, VCE_CMD_END);
+	radeon_ring_write(ring, cpu_to_le32(VCE_CMD_END));
 	radeon_ring_unlock_commit(rdev, ring, false);
 
 	for (i = 0; i < rdev->usec_timeout; i++) {
diff --git a/kernel/drivers/gpu/drm/radeon/radeon_vm.c b/kernel/drivers/gpu/drm/radeon/radeon_vm.c
index 9c3377ca1..3979632b9 100644
--- a/kernel/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/kernel/drivers/gpu/drm/radeon/radeon_vm.c
@@ -331,7 +331,6 @@ struct radeon_bo_va *radeon_vm_bo_add(struct radeon_device *rdev,
 	bo_va->it.start = 0;
 	bo_va->it.last = 0;
 	bo_va->flags = 0;
-	bo_va->addr = 0;
 	bo_va->ref_count = 1;
 	INIT_LIST_HEAD(&bo_va->bo_list);
 	INIT_LIST_HEAD(&bo_va->vm_status);
@@ -456,15 +455,15 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 
 	if (soffset) {
 		/* make sure object fit at this offset */
-		eoffset = soffset + size;
+		eoffset = soffset + size - 1;
 		if (soffset >= eoffset) {
 			r = -EINVAL;
 			goto error_unreserve;
 		}
 
 		last_pfn = eoffset / RADEON_GPU_PAGE_SIZE;
-		if (last_pfn > rdev->vm_manager.max_pfn) {
-			dev_err(rdev->dev, "va above limit (0x%08X > 0x%08X)\n",
+		if (last_pfn >= rdev->vm_manager.max_pfn) {
+			dev_err(rdev->dev, "va above limit (0x%08X >= 0x%08X)\n",
 				last_pfn, rdev->vm_manager.max_pfn);
 			r = -EINVAL;
 			goto error_unreserve;
@@ -479,7 +478,7 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 	eoffset /= RADEON_GPU_PAGE_SIZE;
 	if (soffset || eoffset) {
 		struct interval_tree_node *it;
-		it = interval_tree_iter_first(&vm->va, soffset, eoffset - 1);
+		it = interval_tree_iter_first(&vm->va, soffset, eoffset);
 		if (it && it != &bo_va->it) {
 			struct radeon_bo_va *tmp;
 			tmp = container_of(it, struct radeon_bo_va, it);
@@ -494,40 +493,38 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 	}
 
 	if (bo_va->it.start || bo_va->it.last) {
-		if (bo_va->addr) {
-			/* add a clone of the bo_va to clear the old address */
-			struct radeon_bo_va *tmp;
-			tmp = kzalloc(sizeof(struct radeon_bo_va), GFP_KERNEL);
-			if (!tmp) {
-				mutex_unlock(&vm->mutex);
-				r = -ENOMEM;
-				goto error_unreserve;
-			}
-			tmp->it.start = bo_va->it.start;
-			tmp->it.last = bo_va->it.last;
-			tmp->vm = vm;
-			tmp->addr = bo_va->addr;
-			tmp->bo = radeon_bo_ref(bo_va->bo);
-			spin_lock(&vm->status_lock);
-			list_add(&tmp->vm_status, &vm->freed);
-			spin_unlock(&vm->status_lock);
-
-			bo_va->addr = 0;
+		/* add a clone of the bo_va to clear the old address */
+		struct radeon_bo_va *tmp;
+		tmp = kzalloc(sizeof(struct radeon_bo_va), GFP_KERNEL);
+		if (!tmp) {
+			mutex_unlock(&vm->mutex);
+			r = -ENOMEM;
+			goto error_unreserve;
 		}
+		tmp->it.start = bo_va->it.start;
+		tmp->it.last = bo_va->it.last;
+		tmp->vm = vm;
+		tmp->bo = radeon_bo_ref(bo_va->bo);
 
 		interval_tree_remove(&bo_va->it, &vm->va);
+		spin_lock(&vm->status_lock);
 		bo_va->it.start = 0;
 		bo_va->it.last = 0;
+		list_del_init(&bo_va->vm_status);
+		list_add(&tmp->vm_status, &vm->freed);
+		spin_unlock(&vm->status_lock);
 	}
 
 	if (soffset || eoffset) {
+		spin_lock(&vm->status_lock);
 		bo_va->it.start = soffset;
-		bo_va->it.last = eoffset - 1;
+		bo_va->it.last = eoffset;
+		list_add(&bo_va->vm_status, &vm->cleared);
+		spin_unlock(&vm->status_lock);
 		interval_tree_insert(&bo_va->it, &vm->va);
 	}
 
 	bo_va->flags = flags;
-	bo_va->addr = 0;
 
 	soffset >>= radeon_vm_block_size;
 	eoffset >>= radeon_vm_block_size;
@@ -891,7 +888,7 @@ static void radeon_vm_fence_pts(struct radeon_vm *vm,
 	unsigned i;
 
 	start >>= radeon_vm_block_size;
-	end >>= radeon_vm_block_size;
+	end = (end - 1) >> radeon_vm_block_size;
 
 	for (i = start; i <= end; ++i)
 		radeon_bo_fence(vm->page_tables[i].bo, fence, true);
@@ -928,7 +925,16 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
 	}
 
 	spin_lock(&vm->status_lock);
-	list_del_init(&bo_va->vm_status);
+	if (mem) {
+		if (list_empty(&bo_va->vm_status)) {
+			spin_unlock(&vm->status_lock);
+			return 0;
+		}
+		list_del_init(&bo_va->vm_status);
+	} else {
+		list_del(&bo_va->vm_status);
+		list_add(&bo_va->vm_status, &vm->cleared);
+	}
 	spin_unlock(&vm->status_lock);
 
 	bo_va->flags &= ~RADEON_VM_PAGE_VALID;
@@ -954,10 +960,6 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
 		addr = 0;
 	}
 
-	if (addr == bo_va->addr)
-		return 0;
-	bo_va->addr = addr;
-
 	trace_radeon_vm_bo_update(bo_va);
 
 	nptes = bo_va->it.last - bo_va->it.start + 1;
@@ -1045,7 +1047,7 @@ int radeon_vm_clear_freed(struct radeon_device *rdev,
 			  struct radeon_vm *vm)
 {
 	struct radeon_bo_va *bo_va;
-	int r;
+	int r = 0;
 
 	spin_lock(&vm->status_lock);
 	while (!list_empty(&vm->freed)) {
@@ -1056,14 +1058,15 @@ int radeon_vm_clear_freed(struct radeon_device *rdev,
 		r = radeon_vm_bo_update(rdev, bo_va, NULL);
 		radeon_bo_unref(&bo_va->bo);
 		radeon_fence_unref(&bo_va->last_pt_update);
+		spin_lock(&vm->status_lock);
+		list_del(&bo_va->vm_status);
 		kfree(bo_va);
 		if (r)
-			return r;
+			break;
 
-		spin_lock(&vm->status_lock);
 	}
 	spin_unlock(&vm->status_lock);
-	return 0;
+	return r;
 
 }
 
@@ -1121,10 +1124,10 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev,
 	mutex_lock(&vm->mutex);
 	if (bo_va->it.start || bo_va->it.last)
 		interval_tree_remove(&bo_va->it, &vm->va);
+
 	spin_lock(&vm->status_lock);
 	list_del(&bo_va->vm_status);
-
-	if (bo_va->addr) {
+	if (bo_va->it.start || bo_va->it.last) {
 		bo_va->bo = radeon_bo_ref(bo_va->bo);
 		list_add(&bo_va->vm_status, &vm->freed);
 	} else {
@@ -1151,12 +1154,11 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev,
 	struct radeon_bo_va *bo_va;
 
 	list_for_each_entry(bo_va, &bo->va, bo_list) {
-		if (bo_va->addr) {
-			spin_lock(&bo_va->vm->status_lock);
-			list_del(&bo_va->vm_status);
+		spin_lock(&bo_va->vm->status_lock);
+		if (list_empty(&bo_va->vm_status) &&
+		    (bo_va->it.start || bo_va->it.last))
 			list_add(&bo_va->vm_status, &bo_va->vm->invalidated);
-			spin_unlock(&bo_va->vm->status_lock);
-		}
+		spin_unlock(&bo_va->vm->status_lock);
 	}
 }
 
@@ -1186,6 +1188,7 @@ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm)
 	spin_lock_init(&vm->status_lock);
 	INIT_LIST_HEAD(&vm->invalidated);
 	INIT_LIST_HEAD(&vm->freed);
+	INIT_LIST_HEAD(&vm->cleared);
 
 	pd_size = radeon_vm_directory_size(rdev);
 	pd_entries = radeon_vm_num_pdes(rdev);
diff --git a/kernel/drivers/gpu/drm/radeon/rs600.c b/kernel/drivers/gpu/drm/radeon/rs600.c
index 97a904835..6244f4e44 100644
--- a/kernel/drivers/gpu/drm/radeon/rs600.c
+++ b/kernel/drivers/gpu/drm/radeon/rs600.c
@@ -813,7 +813,7 @@ int rs600_irq_process(struct radeon_device *rdev)
 		status = rs600_irq_ack(rdev);
 	}
 	if (queue_hotplug)
-		schedule_work(&rdev->hotplug_work);
+		schedule_delayed_work(&rdev->hotplug_work, 0);
 	if (queue_hdmi)
 		schedule_work(&rdev->audio_work);
 	if (rdev->msi_enabled) {
diff --git a/kernel/drivers/gpu/drm/radeon/rs690.c b/kernel/drivers/gpu/drm/radeon/rs690.c
index 516ca27cf..6bc44c24e 100644
--- a/kernel/drivers/gpu/drm/radeon/rs690.c
+++ b/kernel/drivers/gpu/drm/radeon/rs690.c
@@ -207,6 +207,9 @@ void rs690_line_buffer_adjust(struct radeon_device *rdev,
 {
 	u32 tmp;
 
+	/* Guess line buffer size to be 8192 pixels */
+	u32 lb_size = 8192;
+
 	/*
 	 * Line Buffer Setup
 	 * There is a single line buffer shared by both display controllers.
@@ -243,6 +246,13 @@ void rs690_line_buffer_adjust(struct radeon_device *rdev,
 		tmp |= V_006520_DC_LB_MEMORY_SPLIT_D1_1Q_D2_3Q;
 	}
 	WREG32(R_006520_DC_LB_MEMORY_SPLIT, tmp);
+
+	/* Save number of lines the linebuffer leads before the scanout */
+	if (mode1)
+		rdev->mode_info.crtcs[0]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode1->crtc_hdisplay);
+
+	if (mode2)
+		rdev->mode_info.crtcs[1]->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode2->crtc_hdisplay);
 }
 
 struct rs690_watermark {
diff --git a/kernel/drivers/gpu/drm/radeon/rv730_dpm.c b/kernel/drivers/gpu/drm/radeon/rv730_dpm.c
index 3f5e1cf13..d37ba2cb8 100644
--- a/kernel/drivers/gpu/drm/radeon/rv730_dpm.c
+++ b/kernel/drivers/gpu/drm/radeon/rv730_dpm.c
@@ -464,7 +464,7 @@ void rv730_stop_dpm(struct radeon_device *rdev)
 	result = rv770_send_msg_to_smc(rdev, PPSMC_MSG_TwoLevelsDisabled);
 
 	if (result != PPSMC_Result_OK)
-		DRM_ERROR("Could not force DPM to low\n");
+		DRM_DEBUG("Could not force DPM to low\n");
 
 	WREG32_P(GENERAL_PWRMGT, 0, ~GLOBAL_PWRMGT_EN);
 
diff --git a/kernel/drivers/gpu/drm/radeon/rv770_dpm.c b/kernel/drivers/gpu/drm/radeon/rv770_dpm.c
index b9c770745..e830c8935 100644
--- a/kernel/drivers/gpu/drm/radeon/rv770_dpm.c
+++ b/kernel/drivers/gpu/drm/radeon/rv770_dpm.c
@@ -193,7 +193,7 @@ void rv770_stop_dpm(struct radeon_device *rdev)
 	result = rv770_send_msg_to_smc(rdev, PPSMC_MSG_TwoLevelsDisabled);
 
 	if (result != PPSMC_Result_OK)
-		DRM_ERROR("Could not force DPM to low.\n");
+		DRM_DEBUG("Could not force DPM to low.\n");
 
 	WREG32_P(GENERAL_PWRMGT, 0, ~GLOBAL_PWRMGT_EN);
 
@@ -1418,7 +1418,7 @@ int rv770_resume_smc(struct radeon_device *rdev)
 int rv770_set_sw_state(struct radeon_device *rdev)
 {
 	if (rv770_send_msg_to_smc(rdev, PPSMC_MSG_SwitchToSwState) != PPSMC_Result_OK)
-		return -EINVAL;
+		DRM_DEBUG("rv770_set_sw_state failed\n");
 	return 0;
 }
 
diff --git a/kernel/drivers/gpu/drm/radeon/si.c b/kernel/drivers/gpu/drm/radeon/si.c
index e15185b16..f878d6962 100644
--- a/kernel/drivers/gpu/drm/radeon/si.c
+++ b/kernel/drivers/gpu/drm/radeon/si.c
@@ -2376,6 +2376,9 @@ static void dce6_program_watermarks(struct radeon_device *rdev,
 		c.full = dfixed_div(c, a);
 		priority_b_mark = dfixed_trunc(c);
 		priority_b_cnt |= priority_b_mark & PRIORITY_MARK_MASK;
+
+		/* Save number of lines the linebuffer leads before the scanout */
+		radeon_crtc->lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay);
 	}
 
 	/* select wm A */
@@ -6848,7 +6851,7 @@ restart_ih:
 	if (queue_dp)
 		schedule_work(&rdev->dp_work);
 	if (queue_hotplug)
-		schedule_work(&rdev->hotplug_work);
+		schedule_delayed_work(&rdev->hotplug_work, 0);
 	if (queue_thermal && rdev->pm.dpm_enabled)
 		schedule_work(&rdev->pm.dpm.thermal.work);
 	rdev->ih.rptr = rptr;
@@ -6955,6 +6958,22 @@ static int si_startup(struct radeon_device *rdev)
 			rdev->ring[R600_RING_TYPE_UVD_INDEX].ring_size = 0;
 	}
 
+	r = radeon_vce_resume(rdev);
+	if (!r) {
+		r = vce_v1_0_resume(rdev);
+		if (!r)
+			r = radeon_fence_driver_start_ring(rdev,
+							   TN_RING_TYPE_VCE1_INDEX);
+		if (!r)
+			r = radeon_fence_driver_start_ring(rdev,
+							   TN_RING_TYPE_VCE2_INDEX);
+	}
+	if (r) {
+		dev_err(rdev->dev, "VCE init error (%d).\n", r);
+		rdev->ring[TN_RING_TYPE_VCE1_INDEX].ring_size = 0;
+		rdev->ring[TN_RING_TYPE_VCE2_INDEX].ring_size = 0;
+	}
+
 	/* Enable IRQ */
 	if (!rdev->irq.installed) {
 		r = radeon_irq_kms_init(rdev);
@@ -7023,6 +7042,23 @@ static int si_startup(struct radeon_device *rdev)
 		}
 	}
 
+	r = -ENOENT;
+
+	ring = &rdev->ring[TN_RING_TYPE_VCE1_INDEX];
+	if (ring->ring_size)
+		r = radeon_ring_init(rdev, ring, ring->ring_size, 0,
+				     VCE_CMD_NO_OP);
+
+	ring = &rdev->ring[TN_RING_TYPE_VCE2_INDEX];
+	if (ring->ring_size)
+		r = radeon_ring_init(rdev, ring, ring->ring_size, 0,
+				     VCE_CMD_NO_OP);
+
+	if (!r)
+		r = vce_v1_0_init(rdev);
+	else if (r != -ENOENT)
+		DRM_ERROR("radeon: failed initializing VCE (%d).\n", r);
+
 	r = radeon_ib_pool_init(rdev);
 	if (r) {
 		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
@@ -7081,6 +7117,7 @@ int si_suspend(struct radeon_device *rdev)
 	if (rdev->has_uvd) {
 		uvd_v1_0_fini(rdev);
 		radeon_uvd_suspend(rdev);
+		radeon_vce_suspend(rdev);
 	}
 	si_fini_pg(rdev);
 	si_fini_cg(rdev);
@@ -7188,6 +7225,17 @@ int si_init(struct radeon_device *rdev)
 		}
 	}
 
+	r = radeon_vce_init(rdev);
+	if (!r) {
+		ring = &rdev->ring[TN_RING_TYPE_VCE1_INDEX];
+		ring->ring_obj = NULL;
+		r600_ring_init(rdev, ring, 4096);
+
+		ring = &rdev->ring[TN_RING_TYPE_VCE2_INDEX];
+		ring->ring_obj = NULL;
+		r600_ring_init(rdev, ring, 4096);
+	}
+
 	rdev->ih.ring_obj = NULL;
 	r600_ih_ring_init(rdev, 64 * 1024);
 
@@ -7239,6 +7287,7 @@ void si_fini(struct radeon_device *rdev)
 	if (rdev->has_uvd) {
 		uvd_v1_0_fini(rdev);
 		radeon_uvd_fini(rdev);
+		radeon_vce_fini(rdev);
 	}
 	si_pcie_gart_fini(rdev);
 	r600_vram_scratch_fini(rdev);
@@ -7723,3 +7772,124 @@ static void si_program_aspm(struct radeon_device *rdev)
 		}
 	}
 }
+
+int si_vce_send_vcepll_ctlreq(struct radeon_device *rdev)
+{
+        unsigned i;
+
+        /* make sure VCEPLL_CTLREQ is deasserted */
+        WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~UPLL_CTLREQ_MASK);
+
+        mdelay(10);
+
+        /* assert UPLL_CTLREQ */
+        WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, UPLL_CTLREQ_MASK, ~UPLL_CTLREQ_MASK);
+
+        /* wait for CTLACK and CTLACK2 to get asserted */
+        for (i = 0; i < 100; ++i) {
+                uint32_t mask = UPLL_CTLACK_MASK | UPLL_CTLACK2_MASK;
+                if ((RREG32_SMC(CG_VCEPLL_FUNC_CNTL) & mask) == mask)
+                        break;
+                mdelay(10);
+        }
+
+        /* deassert UPLL_CTLREQ */
+        WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~UPLL_CTLREQ_MASK);
+
+        if (i == 100) {
+                DRM_ERROR("Timeout setting UVD clocks!\n");
+                return -ETIMEDOUT;
+        }
+
+        return 0;
+}
+
+int si_set_vce_clocks(struct radeon_device *rdev, u32 evclk, u32 ecclk)
+{
+	unsigned fb_div = 0, evclk_div = 0, ecclk_div = 0;
+	int r;
+
+	/* bypass evclk and ecclk with bclk */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL_2,
+		     EVCLK_SRC_SEL(1) | ECCLK_SRC_SEL(1),
+		     ~(EVCLK_SRC_SEL_MASK | ECCLK_SRC_SEL_MASK));
+
+	/* put PLL in bypass mode */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, VCEPLL_BYPASS_EN_MASK,
+		     ~VCEPLL_BYPASS_EN_MASK);
+
+	if (!evclk || !ecclk) {
+		/* keep the Bypass mode, put PLL to sleep */
+		WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, VCEPLL_SLEEP_MASK,
+			     ~VCEPLL_SLEEP_MASK);
+		return 0;
+	}
+
+	r = radeon_uvd_calc_upll_dividers(rdev, evclk, ecclk, 125000, 250000,
+					  16384, 0x03FFFFFF, 0, 128, 5,
+					  &fb_div, &evclk_div, &ecclk_div);
+	if (r)
+		return r;
+
+	/* set RESET_ANTI_MUX to 0 */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL_5, 0, ~RESET_ANTI_MUX_MASK);
+
+	/* set VCO_MODE to 1 */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, VCEPLL_VCO_MODE_MASK,
+		     ~VCEPLL_VCO_MODE_MASK);
+
+	/* toggle VCEPLL_SLEEP to 1 then back to 0 */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, VCEPLL_SLEEP_MASK,
+		     ~VCEPLL_SLEEP_MASK);
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~VCEPLL_SLEEP_MASK);
+
+	/* deassert VCEPLL_RESET */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~VCEPLL_RESET_MASK);
+
+	mdelay(1);
+
+	r = si_vce_send_vcepll_ctlreq(rdev);
+	if (r)
+		return r;
+
+	/* assert VCEPLL_RESET again */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, VCEPLL_RESET_MASK, ~VCEPLL_RESET_MASK);
+
+	/* disable spread spectrum. */
+	WREG32_SMC_P(CG_VCEPLL_SPREAD_SPECTRUM, 0, ~SSEN_MASK);
+
+	/* set feedback divider */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL_3, VCEPLL_FB_DIV(fb_div), ~VCEPLL_FB_DIV_MASK);
+
+	/* set ref divider to 0 */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~VCEPLL_REF_DIV_MASK);
+
+	/* set PDIV_A and PDIV_B */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL_2,
+		     VCEPLL_PDIV_A(evclk_div) | VCEPLL_PDIV_B(ecclk_div),
+		     ~(VCEPLL_PDIV_A_MASK | VCEPLL_PDIV_B_MASK));
+
+	/* give the PLL some time to settle */
+	mdelay(15);
+
+	/* deassert PLL_RESET */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~VCEPLL_RESET_MASK);
+
+	mdelay(15);
+
+	/* switch from bypass mode to normal mode */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~VCEPLL_BYPASS_EN_MASK);
+
+	r = si_vce_send_vcepll_ctlreq(rdev);
+	if (r)
+		return r;
+
+	/* switch VCLK and DCLK selection */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL_2,
+		     EVCLK_SRC_SEL(16) | ECCLK_SRC_SEL(16),
+		     ~(EVCLK_SRC_SEL_MASK | ECCLK_SRC_SEL_MASK));
+
+	mdelay(100);
+
+	return 0;
+}
diff --git a/kernel/drivers/gpu/drm/radeon/si_dpm.c b/kernel/drivers/gpu/drm/radeon/si_dpm.c
index 9dfcedec0..a82b891ae 100644
--- a/kernel/drivers/gpu/drm/radeon/si_dpm.c
+++ b/kernel/drivers/gpu/drm/radeon/si_dpm.c
@@ -1740,6 +1740,7 @@ struct ni_power_info *ni_get_pi(struct radeon_device *rdev);
 struct ni_ps *ni_get_ps(struct radeon_ps *rps);
 
 extern int si_mc_load_microcode(struct radeon_device *rdev);
+extern void vce_v1_0_enable_mgcg(struct radeon_device *rdev, bool enable);
 
 static int si_populate_voltage_value(struct radeon_device *rdev,
 				     const struct atom_voltage_table *table,
@@ -2926,9 +2927,61 @@ static struct si_dpm_quirk si_dpm_quirk_list[] = {
 	{ PCI_VENDOR_ID_ATI, 0x6810, 0x1462, 0x3036, 0, 120000 },
 	{ PCI_VENDOR_ID_ATI, 0x6811, 0x174b, 0xe271, 0, 120000 },
 	{ PCI_VENDOR_ID_ATI, 0x6810, 0x174b, 0xe271, 85000, 90000 },
+	{ PCI_VENDOR_ID_ATI, 0x6811, 0x1462, 0x2015, 0, 120000 },
+	{ PCI_VENDOR_ID_ATI, 0x6811, 0x1043, 0x2015, 0, 120000 },
 	{ 0, 0, 0, 0 },
 };
 
+static u16 si_get_lower_of_leakage_and_vce_voltage(struct radeon_device *rdev,
+						   u16 vce_voltage)
+{
+	u16 highest_leakage = 0;
+	struct si_power_info *si_pi = si_get_pi(rdev);
+	int i;
+
+	for (i = 0; i < si_pi->leakage_voltage.count; i++){
+		if (highest_leakage < si_pi->leakage_voltage.entries[i].voltage)
+			highest_leakage = si_pi->leakage_voltage.entries[i].voltage;
+	}
+
+	if (si_pi->leakage_voltage.count && (highest_leakage < vce_voltage))
+		return highest_leakage;
+
+	return vce_voltage;
+}
+
+static int si_get_vce_clock_voltage(struct radeon_device *rdev,
+				    u32 evclk, u32 ecclk, u16 *voltage)
+{
+	u32 i;
+	int ret = -EINVAL;
+	struct radeon_vce_clock_voltage_dependency_table *table =
+		&rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+
+	if (((evclk == 0) && (ecclk == 0)) ||
+	    (table && (table->count == 0))) {
+		*voltage = 0;
+		return 0;
+	}
+
+	for (i = 0; i < table->count; i++) {
+		if ((evclk <= table->entries[i].evclk) &&
+		    (ecclk <= table->entries[i].ecclk)) {
+			*voltage = table->entries[i].v;
+			ret = 0;
+			break;
+		}
+	}
+
+	/* if no match return the highest voltage */
+	if (ret)
+		*voltage = table->entries[table->count - 1].v;
+
+	*voltage = si_get_lower_of_leakage_and_vce_voltage(rdev, *voltage);
+
+	return ret;
+}
+
 static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 					struct radeon_ps *rps)
 {
@@ -2937,7 +2990,7 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 	bool disable_mclk_switching = false;
 	bool disable_sclk_switching = false;
 	u32 mclk, sclk;
-	u16 vddc, vddci;
+	u16 vddc, vddci, min_vce_voltage = 0;
 	u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
 	u32 max_sclk = 0, max_mclk = 0;
 	int i;
@@ -2956,6 +3009,16 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 		++p;
 	}
 
+	if (rps->vce_active) {
+		rps->evclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].evclk;
+		rps->ecclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].ecclk;
+		si_get_vce_clock_voltage(rdev, rps->evclk, rps->ecclk,
+					 &min_vce_voltage);
+	} else {
+		rps->evclk = 0;
+		rps->ecclk = 0;
+	}
+
 	if ((rdev->pm.dpm.new_active_crtc_count > 1) ||
 	    ni_dpm_vblank_too_short(rdev))
 		disable_mclk_switching = true;
@@ -3036,6 +3099,13 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 		vddc = ps->performance_levels[0].vddc;
 	}
 
+	if (rps->vce_active) {
+		if (sclk < rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].sclk)
+			sclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].sclk;
+		if (mclk < rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].mclk)
+			mclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].mclk;
+	}
+
 	/* adjusted low state */
 	ps->performance_levels[0].sclk = sclk;
 	ps->performance_levels[0].mclk = mclk;
@@ -3085,6 +3155,8 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
                                               &ps->performance_levels[i]);
 
 	for (i = 0; i < ps->performance_level_count; i++) {
+		if (ps->performance_levels[i].vddc < min_vce_voltage)
+			ps->performance_levels[i].vddc = min_vce_voltage;
 		btc_apply_voltage_dependency_rules(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
 						   ps->performance_levels[i].sclk,
 						   max_limits->vddc,  &ps->performance_levels[i].vddc);
@@ -3111,7 +3183,6 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 		if (ps->performance_levels[i].vddc > rdev->pm.dpm.dyn_state.max_clock_voltage_on_dc.vddc)
 			ps->dc_compatible = false;
 	}
-
 }
 
 #if 0
@@ -5860,6 +5931,21 @@ static void si_set_pcie_lane_width_in_smc(struct radeon_device *rdev,
 	}
 }
 
+static void si_set_vce_clock(struct radeon_device *rdev,
+			     struct radeon_ps *new_rps,
+			     struct radeon_ps *old_rps)
+{
+	if ((old_rps->evclk != new_rps->evclk) ||
+	    (old_rps->ecclk != new_rps->ecclk)) {
+		/* turn the clocks on when encoding, off otherwise */
+		if (new_rps->evclk || new_rps->ecclk)
+			vce_v1_0_enable_mgcg(rdev, false);
+		else
+			vce_v1_0_enable_mgcg(rdev, true);
+		radeon_set_vce_clocks(rdev, new_rps->evclk, new_rps->ecclk);
+	}
+}
+
 void si_dpm_setup_asic(struct radeon_device *rdev)
 {
 	int r;
@@ -6548,6 +6634,7 @@ int si_dpm_set_power_state(struct radeon_device *rdev)
 		return ret;
 	}
 	ni_set_uvd_clock_after_set_eng_clock(rdev, new_ps, old_ps);
+	si_set_vce_clock(rdev, new_ps, old_ps);
 	if (eg_pi->pcie_performance_request)
 		si_notify_link_speed_change_after_state_change(rdev, new_ps, old_ps);
 	ret = si_set_power_state_conditionally_enable_ulv(rdev, new_ps);
@@ -6794,6 +6881,21 @@ static int si_parse_power_table(struct radeon_device *rdev)
 		power_state_offset += 2 + power_state->v2.ucNumDPMLevels;
 	}
 	rdev->pm.dpm.num_ps = state_array->ucNumEntries;
+
+	/* fill in the vce power states */
+	for (i = 0; i < RADEON_MAX_VCE_LEVELS; i++) {
+		u32 sclk, mclk;
+		clock_array_index = rdev->pm.dpm.vce_states[i].clk_idx;
+		clock_info = (union pplib_clock_info *)
+			&clock_info_array->clockInfo[clock_array_index * clock_info_array->ucEntrySize];
+		sclk = le16_to_cpu(clock_info->si.usEngineClockLow);
+		sclk |= clock_info->si.ucEngineClockHigh << 16;
+		mclk = le16_to_cpu(clock_info->si.usMemoryClockLow);
+		mclk |= clock_info->si.ucMemoryClockHigh << 16;
+		rdev->pm.dpm.vce_states[i].sclk = sclk;
+		rdev->pm.dpm.vce_states[i].mclk = mclk;
+	}
+
 	return 0;
 }
 
@@ -6838,10 +6940,11 @@ int si_dpm_init(struct radeon_device *rdev)
 	if (ret)
 		return ret;
 
-	ret = si_parse_power_table(rdev);
+	ret = r600_parse_extended_power_table(rdev);
 	if (ret)
 		return ret;
-	ret = r600_parse_extended_power_table(rdev);
+
+	ret = si_parse_power_table(rdev);
 	if (ret)
 		return ret;
 
diff --git a/kernel/drivers/gpu/drm/radeon/sid.h b/kernel/drivers/gpu/drm/radeon/sid.h
index 3afac3013..d1a7b58dd 100644
--- a/kernel/drivers/gpu/drm/radeon/sid.h
+++ b/kernel/drivers/gpu/drm/radeon/sid.h
@@ -915,6 +915,11 @@
 #define DCCG_AUDIO_DTO1_PHASE                           0x05c0
 #define DCCG_AUDIO_DTO1_MODULE                          0x05c4
 
+#define DENTIST_DISPCLK_CNTL				0x0490
+#	define DENTIST_DPREFCLK_WDIVIDER(x)		(((x) & 0x7f) << 24)
+#	define DENTIST_DPREFCLK_WDIVIDER_MASK		(0x7f << 24)
+#	define DENTIST_DPREFCLK_WDIVIDER_SHIFT		24
+
 #define AFMT_AUDIO_SRC_CONTROL                          0x713c
 #define		AFMT_AUDIO_SRC_SELECT(x)		(((x) & 7) << 0)
 /* AFMT_AUDIO_SRC_SELECT
@@ -1879,6 +1884,7 @@
 #define VCE_VCPU_CACHE_SIZE1				0x20030
 #define VCE_VCPU_CACHE_OFFSET2				0x20034
 #define VCE_VCPU_CACHE_SIZE2				0x20038
+#define VCE_VCPU_SCRATCH7				0x200dc
 #define VCE_SOFT_RESET					0x20120
 #define 	VCE_ECPU_SOFT_RESET			(1 << 0)
 #define 	VCE_FME_SOFT_RESET			(1 << 2)
@@ -1893,6 +1899,7 @@
 #define VCE_RB_RPTR					0x2018c
 #define VCE_RB_WPTR					0x20190
 #define VCE_CLOCK_GATING_A				0x202f8
+#	define CGC_DYN_CLOCK_MODE			(1 << 16)
 #define VCE_CLOCK_GATING_B				0x202fc
 #define VCE_UENC_CLOCK_GATING				0x205bc
 #define VCE_UENC_REG_CLOCK_GATING			0x205c0
@@ -1917,4 +1924,31 @@
 #define VCE_CMD_IB_AUTO					0x00000005
 #define VCE_CMD_SEMAPHORE				0x00000006
 
+/* discrete vce clocks */
+#define	CG_VCEPLL_FUNC_CNTL				0xc0030600
+#	define VCEPLL_RESET_MASK			0x00000001
+#	define VCEPLL_SLEEP_MASK			0x00000002
+#	define VCEPLL_BYPASS_EN_MASK			0x00000004
+#	define VCEPLL_CTLREQ_MASK			0x00000008
+#	define VCEPLL_VCO_MODE_MASK			0x00000600
+#	define VCEPLL_REF_DIV_MASK			0x003F0000
+#	define VCEPLL_CTLACK_MASK			0x40000000
+#	define VCEPLL_CTLACK2_MASK			0x80000000
+#define	CG_VCEPLL_FUNC_CNTL_2				0xc0030601
+#	define VCEPLL_PDIV_A(x)				((x) << 0)
+#	define VCEPLL_PDIV_A_MASK			0x0000007F
+#	define VCEPLL_PDIV_B(x)				((x) << 8)
+#	define VCEPLL_PDIV_B_MASK			0x00007F00
+#	define EVCLK_SRC_SEL(x)				((x) << 20)
+#	define EVCLK_SRC_SEL_MASK			0x01F00000
+#	define ECCLK_SRC_SEL(x)				((x) << 25)
+#	define ECCLK_SRC_SEL_MASK			0x3E000000
+#define	CG_VCEPLL_FUNC_CNTL_3				0xc0030602
+#	define VCEPLL_FB_DIV(x)				((x) << 0)
+#	define VCEPLL_FB_DIV_MASK			0x01FFFFFF
+#define	CG_VCEPLL_FUNC_CNTL_4				0xc0030603
+#define	CG_VCEPLL_FUNC_CNTL_5				0xc0030604
+#define	CG_VCEPLL_SPREAD_SPECTRUM			0xc0030606
+#	define VCEPLL_SSEN_MASK				0x00000001
+
 #endif
diff --git a/kernel/drivers/gpu/drm/radeon/trinity_dpm.c b/kernel/drivers/gpu/drm/radeon/trinity_dpm.c
index a5b02c575..d34bfcdab 100644
--- a/kernel/drivers/gpu/drm/radeon/trinity_dpm.c
+++ b/kernel/drivers/gpu/drm/radeon/trinity_dpm.c
@@ -336,6 +336,7 @@ static const u32 trinity_override_mgpg_sequences[] =
 	0x00000204, 0x00000000,
 };
 
+extern void vce_v1_0_enable_mgcg(struct radeon_device *rdev, bool enable);
 static void trinity_program_clk_gating_hw_sequence(struct radeon_device *rdev,
 						   const u32 *seq, u32 count);
 static void trinity_override_dynamic_mg_powergating(struct radeon_device *rdev);
@@ -985,6 +986,21 @@ static void trinity_set_uvd_clock_after_set_eng_clock(struct radeon_device *rdev
 	trinity_setup_uvd_clocks(rdev, new_rps, old_rps);
 }
 
+static void trinity_set_vce_clock(struct radeon_device *rdev,
+				  struct radeon_ps *new_rps,
+				  struct radeon_ps *old_rps)
+{
+	if ((old_rps->evclk != new_rps->evclk) ||
+	    (old_rps->ecclk != new_rps->ecclk)) {
+		/* turn the clocks on when encoding, off otherwise */
+		if (new_rps->evclk || new_rps->ecclk)
+			vce_v1_0_enable_mgcg(rdev, false);
+		else
+			vce_v1_0_enable_mgcg(rdev, true);
+		radeon_set_vce_clocks(rdev, new_rps->evclk, new_rps->ecclk);
+	}
+}
+
 static void trinity_program_ttt(struct radeon_device *rdev)
 {
 	struct trinity_power_info *pi = trinity_get_pi(rdev);
@@ -1246,6 +1262,7 @@ int trinity_dpm_set_power_state(struct radeon_device *rdev)
 		trinity_force_level_0(rdev);
 		trinity_unforce_levels(rdev);
 		trinity_set_uvd_clock_after_set_eng_clock(rdev, new_ps, old_ps);
+		trinity_set_vce_clock(rdev, new_ps, old_ps);
 	}
 	trinity_release_mutex(rdev);
 
@@ -1483,7 +1500,35 @@ static void trinity_adjust_uvd_state(struct radeon_device *rdev,
 	}
 }
 
+static int trinity_get_vce_clock_voltage(struct radeon_device *rdev,
+					 u32 evclk, u32 ecclk, u16 *voltage)
+{
+	u32 i;
+	int ret = -EINVAL;
+	struct radeon_vce_clock_voltage_dependency_table *table =
+		&rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+
+	if (((evclk == 0) && (ecclk == 0)) ||
+	    (table && (table->count == 0))) {
+		*voltage = 0;
+		return 0;
+	}
+
+	for (i = 0; i < table->count; i++) {
+		if ((evclk <= table->entries[i].evclk) &&
+		    (ecclk <= table->entries[i].ecclk)) {
+			*voltage = table->entries[i].v;
+			ret = 0;
+			break;
+		}
+	}
+
+	/* if no match return the highest voltage */
+	if (ret)
+		*voltage = table->entries[table->count - 1].v;
 
+	return ret;
+}
 
 static void trinity_apply_state_adjust_rules(struct radeon_device *rdev,
 					     struct radeon_ps *new_rps,
@@ -1496,6 +1541,7 @@ static void trinity_apply_state_adjust_rules(struct radeon_device *rdev,
 	u32 min_sclk = pi->sys_info.min_sclk; /* XXX check against disp reqs */
 	u32 sclk_in_sr = pi->sys_info.min_sclk; /* ??? */
 	u32 i;
+	u16 min_vce_voltage;
 	bool force_high;
 	u32 num_active_displays = rdev->pm.dpm.new_active_crtc_count;
 
@@ -1504,6 +1550,14 @@ static void trinity_apply_state_adjust_rules(struct radeon_device *rdev,
 
 	trinity_adjust_uvd_state(rdev, new_rps);
 
+	if (new_rps->vce_active) {
+		new_rps->evclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].evclk;
+		new_rps->ecclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].ecclk;
+	} else {
+		new_rps->evclk = 0;
+		new_rps->ecclk = 0;
+	}
+
 	for (i = 0; i < ps->num_levels; i++) {
 		if (ps->levels[i].vddc_index < min_voltage)
 			ps->levels[i].vddc_index = min_voltage;
@@ -1512,6 +1566,17 @@ static void trinity_apply_state_adjust_rules(struct radeon_device *rdev,
 			ps->levels[i].sclk =
 				trinity_get_valid_engine_clock(rdev, min_sclk);
 
+		/* patch in vce limits */
+		if (new_rps->vce_active) {
+			/* sclk */
+			if (ps->levels[i].sclk < rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].sclk)
+				ps->levels[i].sclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].sclk;
+			/* vddc */
+			trinity_get_vce_clock_voltage(rdev, new_rps->evclk, new_rps->ecclk, &min_vce_voltage);
+			if (ps->levels[i].vddc_index < min_vce_voltage)
+				ps->levels[i].vddc_index = min_vce_voltage;
+		}
+
 		ps->levels[i].ds_divider_index =
 			sumo_get_sleep_divider_id_from_clock(rdev, ps->levels[i].sclk, sclk_in_sr);
 
@@ -1733,6 +1798,19 @@ static int trinity_parse_power_table(struct radeon_device *rdev)
 		power_state_offset += 2 + power_state->v2.ucNumDPMLevels;
 	}
 	rdev->pm.dpm.num_ps = state_array->ucNumEntries;
+
+	/* fill in the vce power states */
+	for (i = 0; i < RADEON_MAX_VCE_LEVELS; i++) {
+		u32 sclk;
+		clock_array_index = rdev->pm.dpm.vce_states[i].clk_idx;
+		clock_info = (union pplib_clock_info *)
+			&clock_info_array->clockInfo[clock_array_index * clock_info_array->ucEntrySize];
+		sclk = le16_to_cpu(clock_info->sumo.usEngineClockLow);
+		sclk |= clock_info->sumo.ucEngineClockHigh << 16;
+		rdev->pm.dpm.vce_states[i].sclk = sclk;
+		rdev->pm.dpm.vce_states[i].mclk = 0;
+	}
+
 	return 0;
 }
 
@@ -1914,6 +1992,10 @@ int trinity_dpm_init(struct radeon_device *rdev)
 	if (ret)
 		return ret;
 
+	ret = r600_parse_extended_power_table(rdev);
+	if (ret)
+		return ret;
+
 	ret = trinity_parse_power_table(rdev);
 	if (ret)
 		return ret;
@@ -2000,6 +2082,7 @@ void trinity_dpm_fini(struct radeon_device *rdev)
 	}
 	kfree(rdev->pm.dpm.ps);
 	kfree(rdev->pm.dpm.priv);
+	r600_free_extended_power_table(rdev);
 }
 
 u32 trinity_dpm_get_sclk(struct radeon_device *rdev, bool low)
diff --git a/kernel/drivers/gpu/drm/radeon/vce_v1_0.c b/kernel/drivers/gpu/drm/radeon/vce_v1_0.c
index b44d9c842..a01efe39a 100644
--- a/kernel/drivers/gpu/drm/radeon/vce_v1_0.c
+++ b/kernel/drivers/gpu/drm/radeon/vce_v1_0.c
@@ -31,6 +31,23 @@
 #include "radeon_asic.h"
 #include "sid.h"
 
+#define VCE_V1_0_FW_SIZE	(256 * 1024)
+#define VCE_V1_0_STACK_SIZE	(64 * 1024)
+#define VCE_V1_0_DATA_SIZE	(7808 * (RADEON_MAX_VCE_HANDLES + 1))
+
+struct vce_v1_0_fw_signature
+{
+	int32_t off;
+	uint32_t len;
+	int32_t num;
+	struct {
+		uint32_t chip_id;
+		uint32_t keyselect;
+		uint32_t nonce[4];
+		uint32_t sigval[4];
+	} val[8];
+};
+
 /**
  * vce_v1_0_get_rptr - get read pointer
  *
@@ -82,6 +99,186 @@ void vce_v1_0_set_wptr(struct radeon_device *rdev,
 		WREG32(VCE_RB_WPTR2, ring->wptr);
 }
 
+void vce_v1_0_enable_mgcg(struct radeon_device *rdev, bool enable)
+{
+	u32 tmp;
+
+	if (enable && (rdev->cg_flags & RADEON_CG_SUPPORT_VCE_MGCG)) {
+		tmp = RREG32(VCE_CLOCK_GATING_A);
+		tmp |= CGC_DYN_CLOCK_MODE;
+		WREG32(VCE_CLOCK_GATING_A, tmp);
+
+		tmp = RREG32(VCE_UENC_CLOCK_GATING);
+		tmp &= ~0x1ff000;
+		tmp |= 0xff800000;
+		WREG32(VCE_UENC_CLOCK_GATING, tmp);
+
+		tmp = RREG32(VCE_UENC_REG_CLOCK_GATING);
+		tmp &= ~0x3ff;
+		WREG32(VCE_UENC_REG_CLOCK_GATING, tmp);
+	} else {
+		tmp = RREG32(VCE_CLOCK_GATING_A);
+		tmp &= ~CGC_DYN_CLOCK_MODE;
+		WREG32(VCE_CLOCK_GATING_A, tmp);
+
+		tmp = RREG32(VCE_UENC_CLOCK_GATING);
+		tmp |= 0x1ff000;
+		tmp &= ~0xff800000;
+		WREG32(VCE_UENC_CLOCK_GATING, tmp);
+
+		tmp = RREG32(VCE_UENC_REG_CLOCK_GATING);
+		tmp |= 0x3ff;
+		WREG32(VCE_UENC_REG_CLOCK_GATING, tmp);
+	}
+}
+
+static void vce_v1_0_init_cg(struct radeon_device *rdev)
+{
+	u32 tmp;
+
+	tmp = RREG32(VCE_CLOCK_GATING_A);
+	tmp |= CGC_DYN_CLOCK_MODE;
+	WREG32(VCE_CLOCK_GATING_A, tmp);
+
+	tmp = RREG32(VCE_CLOCK_GATING_B);
+	tmp |= 0x1e;
+	tmp &= ~0xe100e1;
+	WREG32(VCE_CLOCK_GATING_B, tmp);
+
+	tmp = RREG32(VCE_UENC_CLOCK_GATING);
+	tmp &= ~0xff9ff000;
+	WREG32(VCE_UENC_CLOCK_GATING, tmp);
+
+	tmp = RREG32(VCE_UENC_REG_CLOCK_GATING);
+	tmp &= ~0x3ff;
+	WREG32(VCE_UENC_REG_CLOCK_GATING, tmp);
+}
+
+int vce_v1_0_load_fw(struct radeon_device *rdev, uint32_t *data)
+{
+	struct vce_v1_0_fw_signature *sign = (void*)rdev->vce_fw->data;
+	uint32_t chip_id;
+	int i;
+
+	switch (rdev->family) {
+	case CHIP_TAHITI:
+		chip_id = 0x01000014;
+		break;
+	case CHIP_VERDE:
+		chip_id = 0x01000015;
+		break;
+	case CHIP_PITCAIRN:
+	case CHIP_OLAND:
+		chip_id = 0x01000016;
+		break;
+	case CHIP_ARUBA:
+		chip_id = 0x01000017;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	for (i = 0; i < le32_to_cpu(sign->num); ++i) {
+		if (le32_to_cpu(sign->val[i].chip_id) == chip_id)
+			break;
+	}
+
+	if (i == le32_to_cpu(sign->num))
+		return -EINVAL;
+
+	data += (256 - 64) / 4;
+	data[0] = sign->val[i].nonce[0];
+	data[1] = sign->val[i].nonce[1];
+	data[2] = sign->val[i].nonce[2];
+	data[3] = sign->val[i].nonce[3];
+	data[4] = cpu_to_le32(le32_to_cpu(sign->len) + 64);
+
+	memset(&data[5], 0, 44);
+	memcpy(&data[16], &sign[1], rdev->vce_fw->size - sizeof(*sign));
+
+	data += le32_to_cpu(data[4]) / 4;
+	data[0] = sign->val[i].sigval[0];
+	data[1] = sign->val[i].sigval[1];
+	data[2] = sign->val[i].sigval[2];
+	data[3] = sign->val[i].sigval[3];
+
+	rdev->vce.keyselect = le32_to_cpu(sign->val[i].keyselect);
+
+	return 0;
+}
+
+unsigned vce_v1_0_bo_size(struct radeon_device *rdev)
+{
+	WARN_ON(VCE_V1_0_FW_SIZE < rdev->vce_fw->size);
+	return VCE_V1_0_FW_SIZE + VCE_V1_0_STACK_SIZE + VCE_V1_0_DATA_SIZE;
+}
+
+int vce_v1_0_resume(struct radeon_device *rdev)
+{
+	uint64_t addr = rdev->vce.gpu_addr;
+	uint32_t size;
+	int i;
+
+	WREG32_P(VCE_CLOCK_GATING_A, 0, ~(1 << 16));
+	WREG32_P(VCE_UENC_CLOCK_GATING, 0x1FF000, ~0xFF9FF000);
+	WREG32_P(VCE_UENC_REG_CLOCK_GATING, 0x3F, ~0x3F);
+	WREG32(VCE_CLOCK_GATING_B, 0);
+
+	WREG32_P(VCE_LMI_FW_PERIODIC_CTRL, 0x4, ~0x4);
+
+	WREG32(VCE_LMI_CTRL, 0x00398000);
+	WREG32_P(VCE_LMI_CACHE_CTRL, 0x0, ~0x1);
+	WREG32(VCE_LMI_SWAP_CNTL, 0);
+	WREG32(VCE_LMI_SWAP_CNTL1, 0);
+	WREG32(VCE_LMI_VM_CTRL, 0);
+
+	WREG32(VCE_VCPU_SCRATCH7, RADEON_MAX_VCE_HANDLES);
+
+	addr += 256;
+	size = VCE_V1_0_FW_SIZE;
+	WREG32(VCE_VCPU_CACHE_OFFSET0, addr & 0x7fffffff);
+	WREG32(VCE_VCPU_CACHE_SIZE0, size);
+
+	addr += size;
+	size = VCE_V1_0_STACK_SIZE;
+	WREG32(VCE_VCPU_CACHE_OFFSET1, addr & 0x7fffffff);
+	WREG32(VCE_VCPU_CACHE_SIZE1, size);
+
+	addr += size;
+	size = VCE_V1_0_DATA_SIZE;
+	WREG32(VCE_VCPU_CACHE_OFFSET2, addr & 0x7fffffff);
+	WREG32(VCE_VCPU_CACHE_SIZE2, size);
+
+	WREG32_P(VCE_LMI_CTRL2, 0x0, ~0x100);
+
+	WREG32(VCE_LMI_FW_START_KEYSEL, rdev->vce.keyselect);
+
+	for (i = 0; i < 10; ++i) {
+		mdelay(10);
+		if (RREG32(VCE_FW_REG_STATUS) & VCE_FW_REG_STATUS_DONE)
+			break;
+	}
+
+	if (i == 10)
+		return -ETIMEDOUT;
+
+	if (!(RREG32(VCE_FW_REG_STATUS) & VCE_FW_REG_STATUS_PASS))
+		return -EINVAL;
+
+	for (i = 0; i < 10; ++i) {
+		mdelay(10);
+		if (!(RREG32(VCE_FW_REG_STATUS) & VCE_FW_REG_STATUS_BUSY))
+			break;
+	}
+
+	if (i == 10)
+		return -ETIMEDOUT;
+
+	vce_v1_0_init_cg(rdev);
+
+	return 0;
+}
+
 /**
  * vce_v1_0_start - start VCE block
  *
diff --git a/kernel/drivers/gpu/drm/radeon/vce_v2_0.c b/kernel/drivers/gpu/drm/radeon/vce_v2_0.c
index fbbe78fbd..cdeaab7c7 100644
--- a/kernel/drivers/gpu/drm/radeon/vce_v2_0.c
+++ b/kernel/drivers/gpu/drm/radeon/vce_v2_0.c
@@ -31,6 +31,10 @@
 #include "radeon_asic.h"
 #include "cikd.h"
 
+#define VCE_V2_0_FW_SIZE	(256 * 1024)
+#define VCE_V2_0_STACK_SIZE	(64 * 1024)
+#define VCE_V2_0_DATA_SIZE	(23552 * RADEON_MAX_VCE_HANDLES)
+
 static void vce_v2_0_set_sw_cg(struct radeon_device *rdev, bool gated)
 {
 	u32 tmp;
@@ -140,6 +144,12 @@ static void vce_v2_0_init_cg(struct radeon_device *rdev)
 	WREG32(VCE_CLOCK_GATING_B, tmp);
 }
 
+unsigned vce_v2_0_bo_size(struct radeon_device *rdev)
+{
+	WARN_ON(rdev->vce_fw->size > VCE_V2_0_FW_SIZE);
+	return VCE_V2_0_FW_SIZE + VCE_V2_0_STACK_SIZE + VCE_V2_0_DATA_SIZE;
+}
+
 int vce_v2_0_resume(struct radeon_device *rdev)
 {
 	uint64_t addr = rdev->vce.gpu_addr;
@@ -159,17 +169,17 @@ int vce_v2_0_resume(struct radeon_device *rdev)
 	WREG32(VCE_LMI_VCPU_CACHE_40BIT_BAR, addr >> 8);
 
 	addr &= 0xff;
-	size = RADEON_GPU_PAGE_ALIGN(rdev->vce_fw->size);
+	size = VCE_V2_0_FW_SIZE;
 	WREG32(VCE_VCPU_CACHE_OFFSET0, addr & 0x7fffffff);
 	WREG32(VCE_VCPU_CACHE_SIZE0, size);
 
 	addr += size;
-	size = RADEON_VCE_STACK_SIZE;
+	size = VCE_V2_0_STACK_SIZE;
 	WREG32(VCE_VCPU_CACHE_OFFSET1, addr & 0x7fffffff);
 	WREG32(VCE_VCPU_CACHE_SIZE1, size);
 
 	addr += size;
-	size = RADEON_VCE_HEAP_SIZE;
+	size = VCE_V2_0_DATA_SIZE;
 	WREG32(VCE_VCPU_CACHE_OFFSET2, addr & 0x7fffffff);
 	WREG32(VCE_VCPU_CACHE_SIZE2, size);