Merge branch 'drm-next-3.13' of git://people.freedesktop.org/~agd5f/linux into drm-fixes

More fixes for radeon.  This adds new queries for tiling on CIK, and
fixes a crash in handling acpi atif backlight events on CIK.
Some fixes for radeon for 3.13.  Mostly CI stability fixes.  I think
I've tracked down the stability problems with dpm on Trinity/Richland,
so I'm going to enable that by default now.

* 'drm-next-3.13' of git://people.freedesktop.org/~agd5f/linux:
  drm/radeon: hook up backlight functions for CI and KV family.
  drm/radeon/cik: Add macrotile mode array query
  drm/radeon/cik: Return backend map information to userspace
  drm/radeon: enable DPM by default in TN asics
  drm/radeon: adjust TN dpm parameters for stability (v2)
  drm/radeon: use a single doorbell for cik kms compute
  drm/radeon/vm: don't attempt to update ptes if ib allocation fails
  drm/radeon: disable CIK CP semaphores for now
  drm/radeon: allow semaphore emission to fail
  drm/radeon: add semaphore trace point
  radeon: workaround pinning failure on low ram gpu
  radeon/i2c: do not count reg index in number of i2c byte we are writing.
  drm/radeon: cypress_dpm: Fix unused variable warning when CONFIG_ACPI=n
  drm: radeon: ni_dpm: Fix unused variable warning when CONFIG_ACPI=n

29 files changed, 346 insertions, 213 deletions

diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c
index deaf98cdca3a..0652ee0a2098 100644
--- a/drivers/gpu/drm/radeon/atombios_i2c.c
+++ b/drivers/gpu/drm/radeon/atombios_i2c.c
@@ -56,8 +56,10 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 			return -EINVAL;
 		}
 		args.ucRegIndex = buf[0];
-		if (num > 1)
-			memcpy(&out, &buf[1], num - 1);
+		if (num > 1) {
+			num--;
+			memcpy(&out, &buf[1], num);
+		}
 		args.lpI2CDataOut = cpu_to_le16(out);
 	} else {
 		if (num > ATOM_MAX_HW_I2C_READ) {
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index ae92aa041c6a..b43a3a3c9067 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -1560,17 +1560,17 @@ u32 cik_get_xclk(struct radeon_device *rdev)
  * cik_mm_rdoorbell - read a doorbell dword
  *
  * @rdev: radeon_device pointer
- * @offset: byte offset into the aperture
+ * @index: doorbell index
  *
  * Returns the value in the doorbell aperture at the
- * requested offset (CIK).
+ * requested doorbell index (CIK).
  */
-u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 offset)
+u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 index)
 {
-	if (offset < rdev->doorbell.size) {
-		return readl(((void __iomem *)rdev->doorbell.ptr) + offset);
+	if (index < rdev->doorbell.num_doorbells) {
+		return readl(rdev->doorbell.ptr + index);
 	} else {
-		DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", offset);
+		DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index);
 		return 0;
 	}
 }
@@ -1579,18 +1579,18 @@ u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 offset)
  * cik_mm_wdoorbell - write a doorbell dword
  *
  * @rdev: radeon_device pointer
- * @offset: byte offset into the aperture
+ * @index: doorbell index
  * @v: value to write
  *
  * Writes @v to the doorbell aperture at the
- * requested offset (CIK).
+ * requested doorbell index (CIK).
  */
-void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v)
+void cik_mm_wdoorbell(struct radeon_device *rdev, u32 index, u32 v)
 {
-	if (offset < rdev->doorbell.size) {
-		writel(v, ((void __iomem *)rdev->doorbell.ptr) + offset);
+	if (index < rdev->doorbell.num_doorbells) {
+		writel(v, rdev->doorbell.ptr + index);
 	} else {
-		DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", offset);
+		DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index);
 	}
 }
 
@@ -2427,6 +2427,7 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev)
 				gb_tile_moden = 0;
 				break;
 			}
+			rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden;
 			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
 		}
 	} else if (num_pipe_configs == 4) {
@@ -2773,6 +2774,7 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev)
 				gb_tile_moden = 0;
 				break;
 			}
+			rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden;
 			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
 		}
 	} else if (num_pipe_configs == 2) {
@@ -2990,6 +2992,7 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev)
 				gb_tile_moden = 0;
 				break;
 			}
+			rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden;
 			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
 		}
 	} else
@@ -3556,17 +3559,24 @@ void cik_fence_compute_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, 0);
 }
 
-void cik_semaphore_ring_emit(struct radeon_device *rdev,
+bool cik_semaphore_ring_emit(struct radeon_device *rdev,
 			     struct radeon_ring *ring,
 			     struct radeon_semaphore *semaphore,
 			     bool emit_wait)
 {
+/* TODO: figure out why semaphore cause lockups */
+#if 0
 	uint64_t addr = semaphore->gpu_addr;
 	unsigned sel = emit_wait ? PACKET3_SEM_SEL_WAIT : PACKET3_SEM_SEL_SIGNAL;
 
 	radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1));
 	radeon_ring_write(ring, addr & 0xffffffff);
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xffff) | sel);
+
+	return true;
+#else
+	return false;
+#endif
 }
 
 /**
@@ -3609,13 +3619,8 @@ int cik_copy_cpdma(struct radeon_device *rdev,
 		return r;
 	}
 
-	if (radeon_fence_need_sync(*fence, ring->idx)) {
-		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
-					    ring->idx);
-		radeon_fence_note_sync(*fence, ring->idx);
-	} else {
-		radeon_semaphore_free(rdev, &sem, NULL);
-	}
+	radeon_semaphore_sync_to(sem, *fence);
+	radeon_semaphore_sync_rings(rdev, sem, ring->idx);
 
 	for (i = 0; i < num_loops; i++) {
 		cur_size_in_bytes = size_in_bytes;
@@ -4052,7 +4057,7 @@ void cik_compute_ring_set_wptr(struct radeon_device *rdev,
 			       struct radeon_ring *ring)
 {
 	rdev->wb.wb[ring->wptr_offs/4] = cpu_to_le32(ring->wptr);
-	WDOORBELL32(ring->doorbell_offset, ring->wptr);
+	WDOORBELL32(ring->doorbell_index, ring->wptr);
 }
 
 /**
@@ -4393,10 +4398,6 @@ static int cik_cp_compute_resume(struct radeon_device *rdev)
 			return r;
 		}
 
-		/* doorbell offset */
-		rdev->ring[idx].doorbell_offset =
-			(rdev->ring[idx].doorbell_page_num * PAGE_SIZE) + 0;
-
 		/* init the mqd struct */
 		memset(buf, 0, sizeof(struct bonaire_mqd));
 
@@ -4508,7 +4509,7 @@ static int cik_cp_compute_resume(struct radeon_device *rdev)
 				RREG32(CP_HQD_PQ_DOORBELL_CONTROL);
 			mqd->queue_state.cp_hqd_pq_doorbell_control &= ~DOORBELL_OFFSET_MASK;
 			mqd->queue_state.cp_hqd_pq_doorbell_control |=
-				DOORBELL_OFFSET(rdev->ring[idx].doorbell_offset / 4);
+				DOORBELL_OFFSET(rdev->ring[idx].doorbell_index);
 			mqd->queue_state.cp_hqd_pq_doorbell_control |= DOORBELL_EN;
 			mqd->queue_state.cp_hqd_pq_doorbell_control &=
 				~(DOORBELL_SOURCE | DOORBELL_HIT);
@@ -7839,14 +7840,14 @@ int cik_init(struct radeon_device *rdev)
 	ring = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX];
 	ring->ring_obj = NULL;
 	r600_ring_init(rdev, ring, 1024 * 1024);
-	r = radeon_doorbell_get(rdev, &ring->doorbell_page_num);
+	r = radeon_doorbell_get(rdev, &ring->doorbell_index);
 	if (r)
 		return r;
 
 	ring = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX];
 	ring->ring_obj = NULL;
 	r600_ring_init(rdev, ring, 1024 * 1024);
-	r = radeon_doorbell_get(rdev, &ring->doorbell_page_num);
+	r = radeon_doorbell_get(rdev, &ring->doorbell_index);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index 9c9529de20ee..0300727a4f70 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -130,7 +130,7 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev,
  * Add a DMA semaphore packet to the ring wait on or signal
  * other rings (CIK).
  */
-void cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
+bool cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
 				  struct radeon_ring *ring,
 				  struct radeon_semaphore *semaphore,
 				  bool emit_wait)
@@ -141,6 +141,8 @@ void cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SEMAPHORE, 0, extra_bits));
 	radeon_ring_write(ring, addr & 0xfffffff8);
 	radeon_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+
+	return true;
 }
 
 /**
@@ -443,13 +445,8 @@ int cik_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	if (radeon_fence_need_sync(*fence, ring->idx)) {
-		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
-					    ring->idx);
-		radeon_fence_note_sync(*fence, ring->idx);
-	} else {
-		radeon_semaphore_free(rdev, &sem, NULL);
-	}
+	radeon_semaphore_sync_to(sem, *fence);
+	radeon_semaphore_sync_rings(rdev, sem, ring->idx);
 
 	for (i = 0; i < num_loops; i++) {
 		cur_size_in_bytes = size_in_bytes;
diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c
index 91bb470de0a3..920e1e4a52c5 100644
--- a/drivers/gpu/drm/radeon/cypress_dpm.c
+++ b/drivers/gpu/drm/radeon/cypress_dpm.c
@@ -299,7 +299,9 @@ void cypress_program_response_times(struct radeon_device *rdev)
 static int cypress_pcie_performance_request(struct radeon_device *rdev,
 					    u8 perf_req, bool advertise)
 {
+#if defined(CONFIG_ACPI)
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
+#endif
 	u32 tmp;
 
 	udelay(10);
diff --git a/drivers/gpu/drm/radeon/evergreen_dma.c b/drivers/gpu/drm/radeon/evergreen_dma.c
index 6a0656d00ed0..a37b54436382 100644
--- a/drivers/gpu/drm/radeon/evergreen_dma.c
+++ b/drivers/gpu/drm/radeon/evergreen_dma.c
@@ -131,13 +131,8 @@ int evergreen_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	if (radeon_fence_need_sync(*fence, ring->idx)) {
-		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
-					    ring->idx);
-		radeon_fence_note_sync(*fence, ring->idx);
-	} else {
-		radeon_semaphore_free(rdev, &sem, NULL);
-	}
+	radeon_semaphore_sync_to(sem, *fence);
+	radeon_semaphore_sync_rings(rdev, sem, ring->idx);
 
 	for (i = 0; i < num_loops; i++) {
 		cur_size_in_dw = size_in_dw;
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index f26339028154..cdc003085a76 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -3445,9 +3445,9 @@ static int ni_enable_smc_cac(struct radeon_device *rdev,
 static int ni_pcie_performance_request(struct radeon_device *rdev,
 				       u8 perf_req, bool advertise)
 {
+#if defined(CONFIG_ACPI)
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
 
-#if defined(CONFIG_ACPI)
 	if ((perf_req == PCIE_PERF_REQ_PECI_GEN1) ||
             (perf_req == PCIE_PERF_REQ_PECI_GEN2)) {
 		if (eg_pi->pcie_performance_request_registered == false)
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 784983d78158..10abc4d5a6cc 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -869,13 +869,14 @@ void r100_fence_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, RADEON_SW_INT_FIRE);
 }
 
-void r100_semaphore_ring_emit(struct radeon_device *rdev,
+bool r100_semaphore_ring_emit(struct radeon_device *rdev,
 			      struct radeon_ring *ring,
 			      struct radeon_semaphore *semaphore,
 			      bool emit_wait)
 {
 	/* Unused on older asics, since we don't have semaphores or multiple rings */
 	BUG();
+	return false;
 }
 
 int r100_copy_blit(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 4e609e8a8d2b..9ad06732a78b 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2650,7 +2650,7 @@ void r600_fence_ring_emit(struct radeon_device *rdev,
 	}
 }
 
-void r600_semaphore_ring_emit(struct radeon_device *rdev,
+bool r600_semaphore_ring_emit(struct radeon_device *rdev,
 			      struct radeon_ring *ring,
 			      struct radeon_semaphore *semaphore,
 			      bool emit_wait)
@@ -2664,6 +2664,8 @@ void r600_semaphore_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1));
 	radeon_ring_write(ring, addr & 0xffffffff);
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | sel);
+
+	return true;
 }
 
 /**
@@ -2706,13 +2708,8 @@ int r600_copy_cpdma(struct radeon_device *rdev,
 		return r;
 	}
 
-	if (radeon_fence_need_sync(*fence, ring->idx)) {
-		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
-					    ring->idx);
-		radeon_fence_note_sync(*fence, ring->idx);
-	} else {
-		radeon_semaphore_free(rdev, &sem, NULL);
-	}
+	radeon_semaphore_sync_to(sem, *fence);
+	radeon_semaphore_sync_rings(rdev, sem, ring->idx);
 
 	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
 	radeon_ring_write(ring, (WAIT_UNTIL - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
diff --git a/drivers/gpu/drm/radeon/r600_dma.c b/drivers/gpu/drm/radeon/r600_dma.c
index 3b317456512a..7844d15c139f 100644
--- a/drivers/gpu/drm/radeon/r600_dma.c
+++ b/drivers/gpu/drm/radeon/r600_dma.c
@@ -311,7 +311,7 @@ void r600_dma_fence_ring_emit(struct radeon_device *rdev,
  * Add a DMA semaphore packet to the ring wait on or signal
  * other rings (r6xx-SI).
  */
-void r600_dma_semaphore_ring_emit(struct radeon_device *rdev,
+bool r600_dma_semaphore_ring_emit(struct radeon_device *rdev,
 				  struct radeon_ring *ring,
 				  struct radeon_semaphore *semaphore,
 				  bool emit_wait)
@@ -322,6 +322,8 @@ void r600_dma_semaphore_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_SEMAPHORE, 0, s, 0));
 	radeon_ring_write(ring, addr & 0xfffffffc);
 	radeon_ring_write(ring, upper_32_bits(addr) & 0xff);
+
+	return true;
 }
 
 /**
@@ -462,13 +464,8 @@ int r600_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	if (radeon_fence_need_sync(*fence, ring->idx)) {
-		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
-					    ring->idx);
-		radeon_fence_note_sync(*fence, ring->idx);
-	} else {
-		radeon_semaphore_free(rdev, &sem, NULL);
-	}
+	radeon_semaphore_sync_to(sem, *fence);
+	radeon_semaphore_sync_rings(rdev, sem, ring->idx);
 
 	for (i = 0; i < num_loops; i++) {
 		cur_size_in_dw = size_in_dw;
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index b9ee99258602..ecf2a3960c07 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -348,6 +348,7 @@ int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence **fence, i
 void radeon_fence_process(struct radeon_device *rdev, int ring);
 bool radeon_fence_signaled(struct radeon_fence *fence);
 int radeon_fence_wait(struct radeon_fence *fence, bool interruptible);
+int radeon_fence_wait_locked(struct radeon_fence *fence);
 int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring);
 int radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring);
 int radeon_fence_wait_any(struct radeon_device *rdev,
@@ -548,17 +549,20 @@ struct radeon_semaphore {
 	struct radeon_sa_bo		*sa_bo;
 	signed				waiters;
 	uint64_t			gpu_addr;
+	struct radeon_fence		*sync_to[RADEON_NUM_RINGS];
 };
 
 int radeon_semaphore_create(struct radeon_device *rdev,
 			    struct radeon_semaphore **semaphore);
-void radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring,
+bool radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring,
 				  struct radeon_semaphore *semaphore);
-void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
+bool radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
 				struct radeon_semaphore *semaphore);
+void radeon_semaphore_sync_to(struct radeon_semaphore *semaphore,
+			      struct radeon_fence *fence);
 int radeon_semaphore_sync_rings(struct radeon_device *rdev,
 				struct radeon_semaphore *semaphore,
-				int signaler, int waiter);
+				int waiting_ring);
 void radeon_semaphore_free(struct radeon_device *rdev,
 			   struct radeon_semaphore **semaphore,
 			   struct radeon_fence *fence);
@@ -645,13 +649,15 @@ void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg);
 /*
  * GPU doorbell structures, functions & helpers
  */
+#define RADEON_MAX_DOORBELLS 1024	/* Reserve at most 1024 doorbell slots for radeon-owned rings. */
+
 struct radeon_doorbell {
-	u32			num_pages;
-	bool			free[1024];
 	/* doorbell mmio */
-	resource_size_t			base;
-	resource_size_t			size;
-	void __iomem			*ptr;
+	resource_size_t		base;
+	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)];
 };
 
 int radeon_doorbell_get(struct radeon_device *rdev, u32 *page);
@@ -765,7 +771,6 @@ struct radeon_ib {
 	struct radeon_fence		*fence;
 	struct radeon_vm		*vm;
 	bool				is_const_ib;
-	struct radeon_fence		*sync_to[RADEON_NUM_RINGS];
 	struct radeon_semaphore		*semaphore;
 };
 
@@ -799,8 +804,7 @@ struct radeon_ring {
 	u32 pipe;
 	u32 queue;
 	struct radeon_bo	*mqd_obj;
-	u32 doorbell_page_num;
-	u32 doorbell_offset;
+	u32 doorbell_index;
 	unsigned		wptr_offs;
 };
 
@@ -921,7 +925,6 @@ int radeon_ib_get(struct radeon_device *rdev, int ring,
 		  struct radeon_ib *ib, struct radeon_vm *vm,
 		  unsigned size);
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib);
-void radeon_ib_sync_to(struct radeon_ib *ib, struct radeon_fence *fence);
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
 		       struct radeon_ib *const_ib);
 int radeon_ib_pool_init(struct radeon_device *rdev);
@@ -1638,7 +1641,7 @@ struct radeon_asic_ring {
 	/* command emmit functions */
 	void (*ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib);
 	void (*emit_fence)(struct radeon_device *rdev, struct radeon_fence *fence);
-	void (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp,
+	bool (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp,
 			       struct radeon_semaphore *semaphore, bool emit_wait);
 	void (*vm_flush)(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 
@@ -1979,6 +1982,7 @@ struct cik_asic {
 
 	unsigned tile_config;
 	uint32_t tile_mode_array[32];
+	uint32_t macrotile_mode_array[16];
 };
 
 union radeon_asic_config {
@@ -2239,8 +2243,8 @@ void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v,
 u32 r100_io_rreg(struct radeon_device *rdev, u32 reg);
 void r100_io_wreg(struct radeon_device *rdev, u32 reg, u32 v);
 
-u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 offset);
-void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v);
+u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 index);
+void cik_mm_wdoorbell(struct radeon_device *rdev, u32 index, u32 v);
 
 /*
  * Cast helper
@@ -2303,8 +2307,8 @@ void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v);
 #define RREG32_IO(reg) r100_io_rreg(rdev, (reg))
 #define WREG32_IO(reg, v) r100_io_wreg(rdev, (reg), (v))
 
-#define RDOORBELL32(offset) cik_mm_rdoorbell(rdev, (offset))
-#define WDOORBELL32(offset, v) cik_mm_wdoorbell(rdev, (offset), (v))
+#define RDOORBELL32(index) cik_mm_rdoorbell(rdev, (index))
+#define WDOORBELL32(index, v) cik_mm_wdoorbell(rdev, (index), (v))
 
 /*
  * Indirect registers accessor
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 50853c0cb49d..e354ce94cdd1 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2015,6 +2015,8 @@ static struct radeon_asic ci_asic = {
 		.bandwidth_update = &dce8_bandwidth_update,
 		.get_vblank_counter = &evergreen_get_vblank_counter,
 		.wait_for_vblank = &dce4_wait_for_vblank,
+		.set_backlight_level = &atombios_set_backlight_level,
+		.get_backlight_level = &atombios_get_backlight_level,
 		.hdmi_enable = &evergreen_hdmi_enable,
 		.hdmi_setmode = &evergreen_hdmi_setmode,
 	},
@@ -2114,6 +2116,8 @@ static struct radeon_asic kv_asic = {
 		.bandwidth_update = &dce8_bandwidth_update,
 		.get_vblank_counter = &evergreen_get_vblank_counter,
 		.wait_for_vblank = &dce4_wait_for_vblank,
+		.set_backlight_level = &atombios_set_backlight_level,
+		.get_backlight_level = &atombios_get_backlight_level,
 		.hdmi_enable = &evergreen_hdmi_enable,
 		.hdmi_setmode = &evergreen_hdmi_setmode,
 	},
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index f2833ee3a613..c9fd97b58076 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -80,7 +80,7 @@ int r100_irq_set(struct radeon_device *rdev);
 int r100_irq_process(struct radeon_device *rdev);
 void r100_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence);
-void r100_semaphore_ring_emit(struct radeon_device *rdev,
+bool r100_semaphore_ring_emit(struct radeon_device *rdev,
 			      struct radeon_ring *cp,
 			      struct radeon_semaphore *semaphore,
 			      bool emit_wait);
@@ -313,13 +313,13 @@ int r600_cs_parse(struct radeon_cs_parser *p);
 int r600_dma_cs_parse(struct radeon_cs_parser *p);
 void r600_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence);
-void r600_semaphore_ring_emit(struct radeon_device *rdev,
+bool r600_semaphore_ring_emit(struct radeon_device *rdev,
 			      struct radeon_ring *cp,
 			      struct radeon_semaphore *semaphore,
 			      bool emit_wait);
 void r600_dma_fence_ring_emit(struct radeon_device *rdev,
 			      struct radeon_fence *fence);
-void r600_dma_semaphore_ring_emit(struct radeon_device *rdev,
+bool r600_dma_semaphore_ring_emit(struct radeon_device *rdev,
 				  struct radeon_ring *ring,
 				  struct radeon_semaphore *semaphore,
 				  bool emit_wait);
@@ -566,10 +566,6 @@ int sumo_dpm_force_performance_level(struct radeon_device *rdev,
  */
 void cayman_fence_ring_emit(struct radeon_device *rdev,
 			    struct radeon_fence *fence);
-void cayman_uvd_semaphore_emit(struct radeon_device *rdev,
-			       struct radeon_ring *ring,
-			       struct radeon_semaphore *semaphore,
-			       bool emit_wait);
 void cayman_pcie_gart_tlb_flush(struct radeon_device *rdev);
 int cayman_init(struct radeon_device *rdev);
 void cayman_fini(struct radeon_device *rdev);
@@ -697,7 +693,7 @@ void cik_pciep_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 int cik_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
 void cik_sdma_fence_ring_emit(struct radeon_device *rdev,
 			      struct radeon_fence *fence);
-void cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
+bool cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
 				  struct radeon_ring *ring,
 				  struct radeon_semaphore *semaphore,
 				  bool emit_wait);
@@ -717,7 +713,7 @@ void cik_fence_gfx_ring_emit(struct radeon_device *rdev,
 			     struct radeon_fence *fence);
 void cik_fence_compute_ring_emit(struct radeon_device *rdev,
 				 struct radeon_fence *fence);
-void cik_semaphore_ring_emit(struct radeon_device *rdev,
+bool cik_semaphore_ring_emit(struct radeon_device *rdev,
 			     struct radeon_ring *cp,
 			     struct radeon_semaphore *semaphore,
 			     bool emit_wait);
@@ -807,7 +803,7 @@ void uvd_v1_0_stop(struct radeon_device *rdev);
 
 int uvd_v1_0_ring_test(struct radeon_device *rdev, struct radeon_ring *ring);
 int uvd_v1_0_ib_test(struct radeon_device *rdev, struct radeon_ring *ring);
-void uvd_v1_0_semaphore_emit(struct radeon_device *rdev,
+bool uvd_v1_0_semaphore_emit(struct radeon_device *rdev,
 			     struct radeon_ring *ring,
 			     struct radeon_semaphore *semaphore,
 			     bool emit_wait);
@@ -819,7 +815,7 @@ void uvd_v2_2_fence_emit(struct radeon_device *rdev,
 			 struct radeon_fence *fence);
 
 /* uvd v3.1 */
-void uvd_v3_1_semaphore_emit(struct radeon_device *rdev,
+bool uvd_v3_1_semaphore_emit(struct radeon_device *rdev,
 			     struct radeon_ring *ring,
 			     struct radeon_semaphore *semaphore,
 			     bool emit_wait);
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 26ca223d12d6..f41594b2eeac 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -159,7 +159,8 @@ static void radeon_cs_sync_rings(struct radeon_cs_parser *p)
 		if (!p->relocs[i].robj)
 			continue;
 
-		radeon_ib_sync_to(&p->ib, p->relocs[i].robj->tbo.sync_obj);
+		radeon_semaphore_sync_to(p->ib.semaphore,
+					 p->relocs[i].robj->tbo.sync_obj);
 	}
 }
 
@@ -411,9 +412,9 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
 		goto out;
 	}
 	radeon_cs_sync_rings(parser);
-	radeon_ib_sync_to(&parser->ib, vm->fence);
-	radeon_ib_sync_to(&parser->ib, radeon_vm_grab_id(
-		rdev, vm, parser->ring));
+	radeon_semaphore_sync_to(parser->ib.semaphore, vm->fence);
+	radeon_semaphore_sync_to(parser->ib.semaphore,
+				 radeon_vm_grab_id(rdev, vm, parser->ring));
 
 	if ((rdev->family >= CHIP_TAHITI) &&
 	    (parser->chunk_const_ib_idx != -1)) {
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index b9234c43f43d..39b033b441d2 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -251,28 +251,23 @@ void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg)
  */
 int radeon_doorbell_init(struct radeon_device *rdev)
 {
-	int i;
-
 	/* doorbell bar mapping */
 	rdev->doorbell.base = pci_resource_start(rdev->pdev, 2);
 	rdev->doorbell.size = pci_resource_len(rdev->pdev, 2);
 
-	/* limit to 4 MB for now */
-	if (rdev->doorbell.size > (4 * 1024 * 1024))
-		rdev->doorbell.size = 4 * 1024 * 1024;
+	rdev->doorbell.num_doorbells = min_t(u32, rdev->doorbell.size / sizeof(u32), RADEON_MAX_DOORBELLS);
+	if (rdev->doorbell.num_doorbells == 0)
+		return -EINVAL;
 
-	rdev->doorbell.ptr = ioremap(rdev->doorbell.base, rdev->doorbell.size);
+	rdev->doorbell.ptr = ioremap(rdev->doorbell.base, rdev->doorbell.num_doorbells * sizeof(u32));
 	if (rdev->doorbell.ptr == NULL) {
 		return -ENOMEM;
 	}
 	DRM_INFO("doorbell mmio base: 0x%08X\n", (uint32_t)rdev->doorbell.base);
 	DRM_INFO("doorbell mmio size: %u\n", (unsigned)rdev->doorbell.size);
 
-	rdev->doorbell.num_pages = rdev->doorbell.size / PAGE_SIZE;
+	memset(&rdev->doorbell.used, 0, sizeof(rdev->doorbell.used));
 
-	for (i = 0; i < rdev->doorbell.num_pages; i++) {
-		rdev->doorbell.free[i] = true;
-	}
 	return 0;
 }
 
@@ -290,40 +285,38 @@ void radeon_doorbell_fini(struct radeon_device *rdev)
 }
 
 /**
- * radeon_doorbell_get - Allocate a doorbell page
+ * radeon_doorbell_get - Allocate a doorbell entry
  *
  * @rdev: radeon_device pointer
- * @doorbell: doorbell page number
+ * @doorbell: doorbell index
  *
- * Allocate a doorbell page for use by the driver (all asics).
+ * Allocate a doorbell for use by the driver (all asics).
  * Returns 0 on success or -EINVAL on failure.
  */
 int radeon_doorbell_get(struct radeon_device *rdev, u32 *doorbell)
 {
-	int i;
-
-	for (i = 0; i < rdev->doorbell.num_pages; i++) {
-		if (rdev->doorbell.free[i]) {
-			rdev->doorbell.free[i] = false;
-			*doorbell = i;
-			return 0;
-		}
+	unsigned long offset = find_first_zero_bit(rdev->doorbell.used, rdev->doorbell.num_doorbells);
+	if (offset < rdev->doorbell.num_doorbells) {
+		__set_bit(offset, rdev->doorbell.used);
+		*doorbell = offset;
+		return 0;
+	} else {
+		return -EINVAL;
 	}
-	return -EINVAL;
 }
 
 /**
- * radeon_doorbell_free - Free a doorbell page
+ * radeon_doorbell_free - Free a doorbell entry
  *
  * @rdev: radeon_device pointer
- * @doorbell: doorbell page number
+ * @doorbell: doorbell index
  *
- * Free a doorbell page allocated for use by the driver (all asics)
+ * Free a doorbell allocated for use by the driver (all asics)
  */
 void radeon_doorbell_free(struct radeon_device *rdev, u32 doorbell)
 {
-	if (doorbell < rdev->doorbell.num_pages)
-		rdev->doorbell.free[doorbell] = true;
+	if (doorbell < rdev->doorbell.num_doorbells)
+		__clear_bit(doorbell, rdev->doorbell.used);
 }
 
 /*
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 1aee32213f66..9f5ff28864f6 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -76,9 +76,10 @@
  *   2.32.0 - new info request for rings working
  *   2.33.0 - Add SI tiling mode array query
  *   2.34.0 - Add CIK tiling mode array query
+ *   2.35.0 - Add CIK macrotile mode array query
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	34
+#define KMS_DRIVER_MINOR	35
 #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);
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index 281d14c22a47..d3a86e43c012 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -472,6 +472,36 @@ int radeon_fence_wait_any(struct radeon_device *rdev,
 }
 
 /**
+ * radeon_fence_wait_locked - wait for a fence to signal
+ *
+ * @fence: radeon fence object
+ *
+ * Wait for the requested fence to signal (all asics).
+ * Returns 0 if the fence has passed, error for all other cases.
+ */
+int radeon_fence_wait_locked(struct radeon_fence *fence)
+{
+	uint64_t seq[RADEON_NUM_RINGS] = {};
+	int r;
+
+	if (fence == NULL) {
+		WARN(1, "Querying an invalid fence : %p !\n", fence);
+		return -EINVAL;
+	}
+
+	seq[fence->ring] = fence->seq;
+	if (seq[fence->ring] == RADEON_FENCE_SIGNALED_SEQ)
+		return 0;
+
+	r = radeon_fence_wait_seq(fence->rdev, seq, false, false);
+	if (r)
+		return r;
+
+	fence->seq = RADEON_FENCE_SIGNALED_SEQ;
+	return 0;
+}
+
+/**
  * radeon_fence_wait_next_locked - wait for the next fence to signal
  *
  * @rdev: radeon device pointer
diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index 8a83b89d4709..3044e504f4ec 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -651,7 +651,7 @@ retry:
 	radeon_asic_vm_set_page(rdev, &ib, vm->pd_gpu_addr,
 				0, pd_entries, 0, 0);
 
-	radeon_ib_sync_to(&ib, vm->fence);
+	radeon_semaphore_sync_to(ib.semaphore, vm->fence);
 	r = radeon_ib_schedule(rdev, &ib, NULL);
 	if (r) {
 		radeon_ib_free(rdev, &ib);
@@ -1209,6 +1209,8 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 		return -ENOMEM;
 
 	r = radeon_ib_get(rdev, R600_RING_TYPE_DMA_INDEX, &ib, NULL, ndw * 4);
+	if (r)
+		return r;
 	ib.length_dw = 0;
 
 	r = radeon_vm_update_pdes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset);
@@ -1220,7 +1222,7 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 	radeon_vm_update_ptes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset,
 			      addr, radeon_vm_page_flags(bo_va->flags));
 
-	radeon_ib_sync_to(&ib, vm->fence);
+	radeon_semaphore_sync_to(ib.semaphore, vm->fence);
 	r = radeon_ib_schedule(rdev, &ib, NULL);
 	if (r) {
 		radeon_ib_free(rdev, &ib);
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index bb8710531a1b..55d0b474bd37 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -340,7 +340,7 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		break;
 	case RADEON_INFO_BACKEND_MAP:
 		if (rdev->family >= CHIP_BONAIRE)
-			return -EINVAL;
+			*value = rdev->config.cik.backend_map;
 		else if (rdev->family >= CHIP_TAHITI)
 			*value = rdev->config.si.backend_map;
 		else if (rdev->family >= CHIP_CAYMAN)
@@ -449,6 +449,15 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 			return -EINVAL;
 		}
 		break;
+	case RADEON_INFO_CIK_MACROTILE_MODE_ARRAY:
+		if (rdev->family >= CHIP_BONAIRE) {
+			value = rdev->config.cik.macrotile_mode_array;
+			value_size = sizeof(uint32_t)*16;
+		} else {
+			DRM_DEBUG_KMS("macrotile mode array is cik+ only!\n");
+			return -EINVAL;
+		}
+		break;
 	case RADEON_INFO_SI_CP_DMA_COMPUTE:
 		*value = 1;
 		break;
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index 0c7b8c66301b..0b158f98d287 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -422,6 +422,7 @@ int radeon_crtc_do_set_base(struct drm_crtc *crtc,
 	/* Pin framebuffer & get tilling informations */
 	obj = radeon_fb->obj;
 	rbo = gem_to_radeon_bo(obj);
+retry:
 	r = radeon_bo_reserve(rbo, false);
 	if (unlikely(r != 0))
 		return r;
@@ -430,6 +431,33 @@ int radeon_crtc_do_set_base(struct drm_crtc *crtc,
 				     &base);
 	if (unlikely(r != 0)) {
 		radeon_bo_unreserve(rbo);
+
+		/* On old GPU like RN50 with little vram pining can fails because
+		 * current fb is taking all space needed. So instead of unpining
+		 * the old buffer after pining the new one, first unpin old one
+		 * and then retry pining new one.
+		 *
+		 * As only master can set mode only master can pin and it is
+		 * unlikely the master client will race with itself especialy
+		 * on those old gpu with single crtc.
+		 *
+		 * We don't shutdown the display controller because new buffer
+		 * will end up in same spot.
+		 */
+		if (!atomic && fb && fb != crtc->fb) {
+			struct radeon_bo *old_rbo;
+			unsigned long nsize, osize;
+
+			old_rbo = gem_to_radeon_bo(to_radeon_framebuffer(fb)->obj);
+			osize = radeon_bo_size(old_rbo);
+			nsize = radeon_bo_size(rbo);
+			if (nsize <= osize && !radeon_bo_reserve(old_rbo, false)) {
+				radeon_bo_unpin(old_rbo);
+				radeon_bo_unreserve(old_rbo);
+				fb = NULL;
+				goto retry;
+			}
+		}
 		return -EINVAL;
 	}
 	radeon_bo_get_tiling_flags(rbo, &tiling_flags, NULL);
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 866ace070b91..d1385ccc672c 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -1252,7 +1252,6 @@ int radeon_pm_init(struct radeon_device *rdev)
 	case CHIP_RS780:
 	case CHIP_RS880:
 	case CHIP_CAYMAN:
-	case CHIP_ARUBA:
 	case CHIP_BONAIRE:
 	case CHIP_KABINI:
 	case CHIP_KAVERI:
@@ -1284,6 +1283,7 @@ int radeon_pm_init(struct radeon_device *rdev)
 	case CHIP_BARTS:
 	case CHIP_TURKS:
 	case CHIP_CAICOS:
+	case CHIP_ARUBA:
 	case CHIP_TAHITI:
 	case CHIP_PITCAIRN:
 	case CHIP_VERDE:
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 18254e1c3e71..9214403ae173 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -61,7 +61,7 @@ int radeon_ib_get(struct radeon_device *rdev, int ring,
 		  struct radeon_ib *ib, struct radeon_vm *vm,
 		  unsigned size)
 {
-	int i, r;
+	int r;
 
 	r = radeon_sa_bo_new(rdev, &rdev->ring_tmp_bo, &ib->sa_bo, size, 256, true);
 	if (r) {
@@ -87,8 +87,6 @@ int radeon_ib_get(struct radeon_device *rdev, int ring,
 		ib->gpu_addr = radeon_sa_bo_gpu_addr(ib->sa_bo);
 	}
 	ib->is_const_ib = false;
-	for (i = 0; i < RADEON_NUM_RINGS; ++i)
-		ib->sync_to[i] = NULL;
 
 	return 0;
 }
@@ -109,25 +107,6 @@ void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib)
 }
 
 /**
- * radeon_ib_sync_to - sync to fence before executing the IB
- *
- * @ib: IB object to add fence to
- * @fence: fence to sync to
- *
- * Sync to the fence before executing the IB
- */
-void radeon_ib_sync_to(struct radeon_ib *ib, struct radeon_fence *fence)
-{
-	struct radeon_fence *other;
-
-	if (!fence)
-		return;
-
-	other = ib->sync_to[fence->ring];
-	ib->sync_to[fence->ring] = radeon_fence_later(fence, other);
-}
-
-/**
  * radeon_ib_schedule - schedule an IB (Indirect Buffer) on the ring
  *
  * @rdev: radeon_device pointer
@@ -151,8 +130,7 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
 		       struct radeon_ib *const_ib)
 {
 	struct radeon_ring *ring = &rdev->ring[ib->ring];
-	bool need_sync = false;
-	int i, r = 0;
+	int r = 0;
 
 	if (!ib->length_dw || !ring->ready) {
 		/* TODO: Nothings in the ib we should report. */
@@ -166,19 +144,15 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
 		dev_err(rdev->dev, "scheduling IB failed (%d).\n", r);
 		return r;
 	}
-	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-		struct radeon_fence *fence = ib->sync_to[i];
-		if (radeon_fence_need_sync(fence, ib->ring)) {
-			need_sync = true;
-			radeon_semaphore_sync_rings(rdev, ib->semaphore,
-						    fence->ring, ib->ring);
-			radeon_fence_note_sync(fence, ib->ring);
-		}
-	}
-	/* immediately free semaphore when we don't need to sync */
-	if (!need_sync) {
-		radeon_semaphore_free(rdev, &ib->semaphore, NULL);
+
+	/* sync with other rings */
+	r = radeon_semaphore_sync_rings(rdev, ib->semaphore, ib->ring);
+	if (r) {
+		dev_err(rdev->dev, "failed to sync rings (%d)\n", r);
+		radeon_ring_unlock_undo(rdev, ring);
+		return r;
 	}
+
 	/* if we can't remember our last VM flush then flush now! */
 	/* XXX figure out why we have to flush for every IB */
 	if (ib->vm /*&& !ib->vm->last_flush*/) {
diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c
index 8dcc20f53d73..2b42aa1914f2 100644
--- a/drivers/gpu/drm/radeon/radeon_semaphore.c
+++ b/drivers/gpu/drm/radeon/radeon_semaphore.c
@@ -29,12 +29,12 @@
  */
 #include <drm/drmP.h>
 #include "radeon.h"
-
+#include "radeon_trace.h"
 
 int radeon_semaphore_create(struct radeon_device *rdev,
 			    struct radeon_semaphore **semaphore)
 {
-	int r;
+	int i, r;
 
 	*semaphore = kmalloc(sizeof(struct radeon_semaphore), GFP_KERNEL);
 	if (*semaphore == NULL) {
@@ -50,54 +50,121 @@ int radeon_semaphore_create(struct radeon_device *rdev,
 	(*semaphore)->waiters = 0;
 	(*semaphore)->gpu_addr = radeon_sa_bo_gpu_addr((*semaphore)->sa_bo);
 	*((uint64_t*)radeon_sa_bo_cpu_addr((*semaphore)->sa_bo)) = 0;
+
+	for (i = 0; i < RADEON_NUM_RINGS; ++i)
+		(*semaphore)->sync_to[i] = NULL;
+
 	return 0;
 }
 
-void radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring,
+bool radeon_semaphore_emit_signal(struct radeon_device *rdev, int ridx,
 			          struct radeon_semaphore *semaphore)
 {
-	--semaphore->waiters;
-	radeon_semaphore_ring_emit(rdev, ring, &rdev->ring[ring], semaphore, false);
+	struct radeon_ring *ring = &rdev->ring[ridx];
+
+	trace_radeon_semaphore_signale(ridx, semaphore);
+
+	if (radeon_semaphore_ring_emit(rdev, ridx, ring, semaphore, false)) {
+		--semaphore->waiters;
+
+		/* for debugging lockup only, used by sysfs debug files */
+		ring->last_semaphore_signal_addr = semaphore->gpu_addr;
+		return true;
+	}
+	return false;
 }
 
-void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
+bool radeon_semaphore_emit_wait(struct radeon_device *rdev, int ridx,
 			        struct radeon_semaphore *semaphore)
 {
-	++semaphore->waiters;
-	radeon_semaphore_ring_emit(rdev, ring, &rdev->ring[ring], semaphore, true);
+	struct radeon_ring *ring = &rdev->ring[ridx];
+
+	trace_radeon_semaphore_wait(ridx, semaphore);
+
+	if (radeon_semaphore_ring_emit(rdev, ridx, ring, semaphore, true)) {
+		++semaphore->waiters;
+
+		/* for debugging lockup only, used by sysfs debug files */
+		ring->last_semaphore_wait_addr = semaphore->gpu_addr;
+		return true;
+	}
+	return false;
+}
+
+/**
+ * radeon_semaphore_sync_to - use the semaphore to sync to a fence
+ *
+ * @semaphore: semaphore object to add fence to
+ * @fence: fence to sync to
+ *
+ * Sync to the fence using this semaphore object
+ */
+void radeon_semaphore_sync_to(struct radeon_semaphore *semaphore,
+			      struct radeon_fence *fence)
+{
+        struct radeon_fence *other;
+
+        if (!fence)
+                return;
+
+        other = semaphore->sync_to[fence->ring];
+        semaphore->sync_to[fence->ring] = radeon_fence_later(fence, other);
 }
 
-/* caller must hold ring lock */
+/**
+ * radeon_semaphore_sync_rings - sync ring to all registered fences
+ *
+ * @rdev: radeon_device pointer
+ * @semaphore: semaphore object to use for sync
+ * @ring: ring that needs sync
+ *
+ * Ensure that all registered fences are signaled before letting
+ * the ring continue. The caller must hold the ring lock.
+ */
 int radeon_semaphore_sync_rings(struct radeon_device *rdev,
 				struct radeon_semaphore *semaphore,
-				int signaler, int waiter)
+				int ring)
 {
-	int r;
+	int i, r;
 
-	/* no need to signal and wait on the same ring */
-	if (signaler == waiter) {
-		return 0;
-	}
+        for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+		struct radeon_fence *fence = semaphore->sync_to[i];
 
-	/* prevent GPU deadlocks */
-	if (!rdev->ring[signaler].ready) {
-		dev_err(rdev->dev, "Trying to sync to a disabled ring!");
-		return -EINVAL;
-	}
+		/* check if we really need to sync */
+                if (!radeon_fence_need_sync(fence, ring))
+			continue;
 
-	r = radeon_ring_alloc(rdev, &rdev->ring[signaler], 8);
-	if (r) {
-		return r;
-	}
-	radeon_semaphore_emit_signal(rdev, signaler, semaphore);
-	radeon_ring_commit(rdev, &rdev->ring[signaler]);
+		/* prevent GPU deadlocks */
+		if (!rdev->ring[i].ready) {
+			dev_err(rdev->dev, "Syncing to a disabled ring!");
+			return -EINVAL;
+		}
 
-	/* we assume caller has already allocated space on waiters ring */
-	radeon_semaphore_emit_wait(rdev, waiter, semaphore);
+		/* allocate enough space for sync command */
+		r = radeon_ring_alloc(rdev, &rdev->ring[i], 16);
+		if (r) {
+			return r;
+		}
 
-	/* for debugging lockup only, used by sysfs debug files */
-	rdev->ring[signaler].last_semaphore_signal_addr = semaphore->gpu_addr;
-	rdev->ring[waiter].last_semaphore_wait_addr = semaphore->gpu_addr;
+		/* emit the signal semaphore */
+		if (!radeon_semaphore_emit_signal(rdev, i, semaphore)) {
+			/* signaling wasn't successful wait manually */
+			radeon_ring_undo(&rdev->ring[i]);
+			radeon_fence_wait_locked(fence);
+			continue;
+		}
+
+		/* we assume caller has already allocated space on waiters ring */
+		if (!radeon_semaphore_emit_wait(rdev, ring, semaphore)) {
+			/* waiting wasn't successful wait manually */
+			radeon_ring_undo(&rdev->ring[i]);
+			radeon_fence_wait_locked(fence);
+			continue;
+		}
+
+		radeon_ring_commit(rdev, &rdev->ring[i]);
+		radeon_fence_note_sync(fence, ring);
+	}
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_trace.h b/drivers/gpu/drm/radeon/radeon_trace.h
index 811bca691b36..9f0e18172b6e 100644
--- a/drivers/gpu/drm/radeon/radeon_trace.h
+++ b/drivers/gpu/drm/radeon/radeon_trace.h
@@ -111,6 +111,42 @@ DEFINE_EVENT(radeon_fence_request, radeon_fence_wait_end,
 	    TP_ARGS(dev, seqno)
 );
 
+DECLARE_EVENT_CLASS(radeon_semaphore_request,
+
+	    TP_PROTO(int ring, struct radeon_semaphore *sem),
+
+	    TP_ARGS(ring, sem),
+
+	    TP_STRUCT__entry(
+			     __field(int, ring)
+			     __field(signed, waiters)
+			     __field(uint64_t, gpu_addr)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->ring = ring;
+			   __entry->waiters = sem->waiters;
+			   __entry->gpu_addr = sem->gpu_addr;
+			   ),
+
+	    TP_printk("ring=%u, waiters=%d, addr=%010Lx", __entry->ring,
+		      __entry->waiters, __entry->gpu_addr)
+);
+
+DEFINE_EVENT(radeon_semaphore_request, radeon_semaphore_signale,
+
+	    TP_PROTO(int ring, struct radeon_semaphore *sem),
+
+	    TP_ARGS(ring, sem)
+);
+
+DEFINE_EVENT(radeon_semaphore_request, radeon_semaphore_wait,
+
+	    TP_PROTO(int ring, struct radeon_semaphore *sem),
+
+	    TP_ARGS(ring, sem)
+);
+
 #endif
 
 /* This part must be outside protection */
diff --git a/drivers/gpu/drm/radeon/rv770_dma.c b/drivers/gpu/drm/radeon/rv770_dma.c
index f9b02e3d6830..aca8cbe8a335 100644
--- a/drivers/gpu/drm/radeon/rv770_dma.c
+++ b/drivers/gpu/drm/radeon/rv770_dma.c
@@ -66,13 +66,8 @@ int rv770_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	if (radeon_fence_need_sync(*fence, ring->idx)) {
-		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
-					    ring->idx);
-		radeon_fence_note_sync(*fence, ring->idx);
-	} else {
-		radeon_semaphore_free(rdev, &sem, NULL);
-	}
+	radeon_semaphore_sync_to(sem, *fence);
+	radeon_semaphore_sync_rings(rdev, sem, ring->idx);
 
 	for (i = 0; i < num_loops; i++) {
 		cur_size_in_dw = size_in_dw;
diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c
index 8e8f46133532..59be2cfcbb47 100644
--- a/drivers/gpu/drm/radeon/si_dma.c
+++ b/drivers/gpu/drm/radeon/si_dma.c
@@ -195,13 +195,8 @@ int si_copy_dma(struct radeon_device *rdev,
 		return r;
 	}
 
-	if (radeon_fence_need_sync(*fence, ring->idx)) {
-		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
-					    ring->idx);
-		radeon_fence_note_sync(*fence, ring->idx);
-	} else {
-		radeon_semaphore_free(rdev, &sem, NULL);
-	}
+	radeon_semaphore_sync_to(sem, *fence);
+	radeon_semaphore_sync_rings(rdev, sem, ring->idx);
 
 	for (i = 0; i < num_loops; i++) {
 		cur_size_in_bytes = size_in_bytes;
diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c
index 9364129ba292..d700698a1f22 100644
--- a/drivers/gpu/drm/radeon/trinity_dpm.c
+++ b/drivers/gpu/drm/radeon/trinity_dpm.c
@@ -1873,9 +1873,9 @@ int trinity_dpm_init(struct radeon_device *rdev)
 	pi->enable_sclk_ds = true;
 	pi->enable_gfx_power_gating = true;
 	pi->enable_gfx_clock_gating = true;
-	pi->enable_mg_clock_gating = true;
-	pi->enable_gfx_dynamic_mgpg = true; /* ??? */
-	pi->override_dynamic_mgpg = true;
+	pi->enable_mg_clock_gating = false;
+	pi->enable_gfx_dynamic_mgpg = false;
+	pi->override_dynamic_mgpg = false;
 	pi->enable_auto_thermal_throttling = true;
 	pi->voltage_drop_in_dce = false; /* need to restructure dpm/modeset interaction */
 	pi->uvd_dpm = true; /* ??? */
diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c
index 7266805d9786..d4a68af1a279 100644
--- a/drivers/gpu/drm/radeon/uvd_v1_0.c
+++ b/drivers/gpu/drm/radeon/uvd_v1_0.c
@@ -357,7 +357,7 @@ int uvd_v1_0_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
  *
  * Emit a semaphore command (either wait or signal) to the UVD ring.
  */
-void uvd_v1_0_semaphore_emit(struct radeon_device *rdev,
+bool uvd_v1_0_semaphore_emit(struct radeon_device *rdev,
 			     struct radeon_ring *ring,
 			     struct radeon_semaphore *semaphore,
 			     bool emit_wait)
@@ -372,6 +372,8 @@ void uvd_v1_0_semaphore_emit(struct radeon_device *rdev,
 
 	radeon_ring_write(ring, PACKET0(UVD_SEMA_CMD, 0));
 	radeon_ring_write(ring, emit_wait ? 1 : 0);
+
+	return true;
 }
 
 /**
diff --git a/drivers/gpu/drm/radeon/uvd_v3_1.c b/drivers/gpu/drm/radeon/uvd_v3_1.c
index 5b6fa1f62d4e..d722db2cf340 100644
--- a/drivers/gpu/drm/radeon/uvd_v3_1.c
+++ b/drivers/gpu/drm/radeon/uvd_v3_1.c
@@ -37,7 +37,7 @@
  *
  * Emit a semaphore command (either wait or signal) to the UVD ring.
  */
-void uvd_v3_1_semaphore_emit(struct radeon_device *rdev,
+bool uvd_v3_1_semaphore_emit(struct radeon_device *rdev,
 			     struct radeon_ring *ring,
 			     struct radeon_semaphore *semaphore,
 			     bool emit_wait)
@@ -52,4 +52,6 @@ void uvd_v3_1_semaphore_emit(struct radeon_device *rdev,
 
 	radeon_ring_write(ring, PACKET0(UVD_SEMA_CMD, 0));
 	radeon_ring_write(ring, 0x80 | (emit_wait ? 1 : 0));
+
+	return true;
 }
diff --git a/include/uapi/drm/radeon_drm.h b/include/uapi/drm/radeon_drm.h
index 46d41e8b0dcc..2f3f7ea8c77b 100644
--- a/include/uapi/drm/radeon_drm.h
+++ b/include/uapi/drm/radeon_drm.h
@@ -981,6 +981,8 @@ struct drm_radeon_cs {
 #define RADEON_INFO_SI_TILE_MODE_ARRAY	0x16
 /* query if CP DMA is supported on the compute ring */
 #define RADEON_INFO_SI_CP_DMA_COMPUTE	0x17
+/* CIK macrotile mode array */
+#define RADEON_INFO_CIK_MACROTILE_MODE_ARRAY	0x18
 
 
 struct drm_radeon_info {