WebSVN – Kolibri OS – Diff – /drivers/video/drm/i915/intel_ringbuffer.c

 #include "i915_drv.h"
 #include
 #include "i915_trace.h"
 #include "intel_drv.h"
+/* Early gen2 devices have a cacheline of just 32 bytes, using 64 is overkill,
+ * but keeps the logic simple. Indeed, the whole purpose of this macro is just
+ * to give some inclination as to some of the magic values used in the various
+ * workarounds!
+ */
+#define CACHELINE_BYTES 64
-static inline int ring_space(struct intel_ring_buffer *ring)
+static inline int __ring_space(int head, int tail, int size)
 {
-	int space = (ring->head & HEAD_ADDR) - (ring->tail + I915_RING_FREE_SPACE);
+	int space = head - (tail + I915_RING_FREE_SPACE);
 	if (space < 0)
-		space += ring->size;
+		space += size;
 	return space;
 }
+static inline int ring_space(struct intel_ringbuffer *ringbuf)
+{
+	return __ring_space(ringbuf->head & HEAD_ADDR, ringbuf->tail, ringbuf->size);
+}
-void __intel_ring_advance(struct intel_ring_buffer *ring)
+static bool intel_ring_stopped(struct intel_engine_cs *ring)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	return dev_priv->gpu_error.stop_rings & intel_ring_flag(ring);
+}
-{
+void __intel_ring_advance(struct intel_engine_cs *ring)
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+{
 	struct intel_ringbuffer *ringbuf = ring->buffer;
-	ring->tail &= ring->size - 1;
+	ringbuf->tail &= ringbuf->size - 1;
-	if (dev_priv->gpu_error.stop_rings & intel_ring_flag(ring))
+	if (intel_ring_stopped(ring))
 		return;
-	ring->write_tail(ring, ring->tail);
+	ring->write_tail(ring, ringbuf->tail);
+}
 static int
-gen2_render_ring_flush(struct intel_ring_buffer *ring,
+gen2_render_ring_flush(struct intel_engine_cs *ring,
 		       u32	invalidate_domains,
 	return 0;
+}
 static int
-gen4_render_ring_flush(struct intel_ring_buffer *ring,
+gen4_render_ring_flush(struct intel_engine_cs *ring,
 		  u32	invalidate_domains,
 		  u32	flush_domains)
+{
  * Post-sync nonzero is what triggered this second workaround, so we
  * can't use that one either.  Notify enable is IRQs, which aren't
  * really our business.  That leaves only stall at scoreboard.
  */
 static int
-intel_emit_post_sync_nonzero_flush(struct intel_ring_buffer *ring)
+intel_emit_post_sync_nonzero_flush(struct intel_engine_cs *ring)
+{
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 	return 0;
+}
 static int
-gen6_render_ring_flush(struct intel_ring_buffer *ring,
+gen6_render_ring_flush(struct intel_engine_cs *ring,
                          u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 	return 0;
+}
 static int
-gen7_render_ring_cs_stall_wa(struct intel_ring_buffer *ring)
+gen7_render_ring_cs_stall_wa(struct intel_engine_cs *ring)
+{
 	int ret;
 	intel_ring_advance(ring);
 	return 0;
 }
-static int gen7_ring_fbc_flush(struct intel_ring_buffer *ring, u32 value)
+static int gen7_ring_fbc_flush(struct intel_engine_cs *ring, u32 value)
+{
 	int ret;
 	ring->fbc_dirty = false;
 	return 0;
+}
 static int
-gen7_render_ring_flush(struct intel_ring_buffer *ring,
+gen7_render_ring_flush(struct intel_engine_cs *ring,
 		       u32 invalidate_domains, u32 flush_domains)
+{
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 	/*
 	return 0;
+}
+static int
+gen8_emit_pipe_control(struct intel_engine_cs *ring,
+		       u32 flags, u32 scratch_addr)
+{
+	int ret;
+	ret = intel_ring_begin(ring, 6);
+	if (ret)
+		return ret;
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+	intel_ring_emit(ring, flags);
+	intel_ring_emit(ring, scratch_addr);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_advance(ring);
+	return 0;
+}
 static int
-gen8_render_ring_flush(struct intel_ring_buffer *ring,
+gen8_render_ring_flush(struct intel_engine_cs *ring,
 		       u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 		flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_QW_WRITE;
 		flags |= PIPE_CONTROL_GLOBAL_GTT_IVB;
-	}
+		/* WaCsStallBeforeStateCacheInvalidate:bdw,chv */
+		ret = gen8_emit_pipe_control(ring,
+					     PIPE_CONTROL_CS_STALL |
+					     PIPE_CONTROL_STALL_AT_SCOREBOARD,
-	ret = intel_ring_begin(ring, 6);
+);
 	if (ret)
+		return ret;
-		return ret;
-	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+	}
-	intel_ring_emit(ring, flags);
-	intel_ring_emit(ring, scratch_addr);
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, 0);
-	intel_ring_advance(ring);
 	return 0;
+	return gen8_emit_pipe_control(ring, flags, scratch_addr);
+}
-static void ring_write_tail(struct intel_ring_buffer *ring,
+static void ring_write_tail(struct intel_engine_cs *ring,
 			    u32 value)
+{
-	drm_i915_private_t *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	I915_WRITE_TAIL(ring, value);
 }
-u32 intel_ring_get_active_head(struct intel_ring_buffer *ring)
+u64 intel_ring_get_active_head(struct intel_engine_cs *ring)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	u64 acthd;
+	if (INTEL_INFO(ring->dev)->gen >= 8)
-{
+		acthd = I915_READ64_2x32(RING_ACTHD(ring->mmio_base),
+					 RING_ACTHD_UDW(ring->mmio_base));
+	else if (INTEL_INFO(ring->dev)->gen >= 4)
-	drm_i915_private_t *dev_priv = ring->dev->dev_private;
+		acthd = I915_READ(RING_ACTHD(ring->mmio_base));
-	u32 acthd_reg = INTEL_INFO(ring->dev)->gen >= 4 ?
+	else
-			RING_ACTHD(ring->mmio_base) : ACTHD;
+		acthd = I915_READ(ACTHD);
-	return I915_READ(acthd_reg);
+	return acthd;
+}
-static void ring_setup_phys_status_page(struct intel_ring_buffer *ring)
+static void ring_setup_phys_status_page(struct intel_engine_cs *ring)
+{
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u32 addr;
-	addr = dev_priv->status_page_dmah->busaddr;
-	if (INTEL_INFO(ring->dev)->gen >= 4)
+	addr = dev_priv->status_page_dmah->busaddr;
-		addr |= (dev_priv->status_page_dmah->busaddr >> 28) & 0xf0;
-	I915_WRITE(HWS_PGA, addr);
-}
+	if (INTEL_INFO(ring->dev)->gen >= 4)
+		addr |= (dev_priv->status_page_dmah->busaddr >> 28) & 0xf0;
-static int init_ring_common(struct intel_ring_buffer *ring)
-{
+	I915_WRITE(HWS_PGA, addr);
-	struct drm_device *dev = ring->dev;
+}
-	drm_i915_private_t *dev_priv = dev->dev_private;
+static bool stop_ring(struct intel_engine_cs *ring)
-	struct drm_i915_gem_object *obj = ring->obj;
+{
-	int ret = 0;
-	u32 head;
+	struct drm_i915_private *dev_priv = to_i915(ring->dev);
-	gen6_gt_force_wake_get(dev_priv, FORCEWAKE_ALL);
+	if (!IS_GEN2(ring->dev)) {
+		I915_WRITE_MODE(ring, _MASKED_BIT_ENABLE(STOP_RING));
+		if (wait_for_atomic((I915_READ_MODE(ring) & MODE_IDLE) != 0, 1000)) {
+			DRM_ERROR("%s :timed out trying to stop ring\n", ring->name);
+			return false;
+		}
+	}
+	I915_WRITE_CTL(ring, 0);
+	I915_WRITE_HEAD(ring, 0);
+	ring->write_tail(ring, 0);
+	if (!IS_GEN2(ring->dev)) {
+		(void)I915_READ_CTL(ring);
+		I915_WRITE_MODE(ring, _MASKED_BIT_DISABLE(STOP_RING));
+	}
+	return (I915_READ_HEAD(ring) & HEAD_ADDR) == 0;
+}
-	if (I915_NEED_GFX_HWS(dev))
-		intel_ring_setup_status_page(ring);
-	else
+static int init_ring_common(struct intel_engine_cs *ring)
 		ring_setup_phys_status_page(ring);
 	struct drm_device *dev = ring->dev;
-	/* Stop the ring if it's running. */
+	struct drm_i915_private *dev_priv = dev->dev_private;
-	I915_WRITE_CTL(ring, 0);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
-	I915_WRITE_HEAD(ring, 0);
+	struct drm_i915_gem_object *obj = ringbuf->obj;
-	ring->write_tail(ring, 0);
+	int ret = 0;
-	head = I915_READ_HEAD(ring) & HEAD_ADDR;
-	/* G45 ring initialization fails to reset head to zero */
+	gen6_gt_force_wake_get(dev_priv, FORCEWAKE_ALL);
 	if (head != 0) {
-		DRM_DEBUG_KMS("%s head not reset to zero "
+	if (!stop_ring(ring)) {
-			      "ctl %08x head %08x tail %08x start %08x\n",
+		/* G45 ring initialization often fails to reset head to zero */
-			      ring->name,
+		DRM_DEBUG_KMS("%s head not reset to zero "
-			      I915_READ_CTL(ring),
+			      "ctl %08x head %08x tail %08x start %08x\n",
-			      I915_READ_HEAD(ring),
+			      ring->name,
+			      I915_READ_CTL(ring),
+			      I915_READ_HEAD(ring),
 			      I915_READ_TAIL(ring),
 			      I915_READ_START(ring));
+		if (!stop_ring(ring)) {
+			DRM_ERROR("failed to set %s head to zero "
+				  "ctl %08x head %08x tail %08x start %08x\n",
+				  ring->name,
+				  I915_READ_CTL(ring),
+				  I915_READ_HEAD(ring),
+				  I915_READ_TAIL(ring),
 				  I915_READ_START(ring));
-		I915_WRITE_HEAD(ring, 0);
+			ret = -EIO;
 			goto out;
 		if (I915_READ_HEAD(ring) & HEAD_ADDR) {
 			DRM_ERROR("failed to set %s head to zero "
 				  "ctl %08x head %08x tail %08x start %08x\n",
-				  ring->name,
+	if (I915_NEED_GFX_HWS(dev))
-				  I915_READ_CTL(ring),
+		intel_ring_setup_status_page(ring);
-				  I915_READ_HEAD(ring),
+	else
-				  I915_READ_TAIL(ring),
+		ring_setup_phys_status_page(ring);
 				  I915_READ_START(ring));
 	/* Enforce ordering by reading HEAD register back */
 	I915_READ_HEAD(ring);
 	/* Initialize the ring. This must happen _after_ we've cleared the ring
-	 * registers with the above sequence (the readback of the HEAD registers
-	 * also enforces ordering), otherwise the hw might lose the new ring
+	 * registers with the above sequence (the readback of the HEAD registers
-	 * register values. */
+	 * also enforces ordering), otherwise the hw might lose the new ring
-	I915_WRITE_START(ring, i915_gem_obj_ggtt_offset(obj));
+	 * register values. */
-	I915_WRITE_CTL(ring,
+	I915_WRITE_START(ring, i915_gem_obj_ggtt_offset(obj));
-			((ring->size - PAGE_SIZE) & RING_NR_PAGES)
+	I915_WRITE_CTL(ring,
-			| RING_VALID);
+			((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES)
+			| RING_VALID);
 	/* If the head is still not zero, the ring is dead */
 	if (wait_for((I915_READ_CTL(ring) & RING_VALID) != 0 &&
 		     I915_READ_START(ring) == i915_gem_obj_ggtt_offset(obj) &&
 		     (I915_READ_HEAD(ring) & HEAD_ADDR) == 0, 50)) {
 		DRM_ERROR("%s initialization failed "
-				"ctl %08x head %08x tail %08x start %08x\n",
+			  "ctl %08x (valid? %d) head %08x tail %08x start %08x [expected %08lx]\n",
 				ring->name,
-				I915_READ_CTL(ring),
+			  I915_READ_CTL(ring), I915_READ_CTL(ring) & RING_VALID,
-				I915_READ_HEAD(ring),
+			  I915_READ_HEAD(ring), I915_READ_TAIL(ring),
-				I915_READ_TAIL(ring),
+			  I915_READ_START(ring), (unsigned long)i915_gem_obj_ggtt_offset(obj));
-				I915_READ_START(ring));
+		ret = -EIO;
-		ret = -EIO;
+		goto out;
 		goto out;
 		DRM_ERROR("Failed to allocate seqno page\n");
 		ret = -ENOMEM;
 		goto err;
+	}
+	ret = i915_gem_object_set_cache_level(ring->scratch.obj, I915_CACHE_LLC);
+	if (ret)
-	i915_gem_object_set_cache_level(ring->scratch.obj, I915_CACHE_LLC);
+		goto err_unref;
-	ret = i915_gem_obj_ggtt_pin(ring->scratch.obj, 4096, true, false);
+	ret = i915_gem_obj_ggtt_pin(ring->scratch.obj, 4096, 0);
 	if (ret)
 		goto err_unref;
 	DRM_DEBUG_DRIVER("%s pipe control offset: 0x%08x\n",
 			 ring->name, ring->scratch.gtt_offset);
 	return 0;
 err_unpin:
-	i915_gem_object_unpin(ring->scratch.obj);
+	i915_gem_object_ggtt_unpin(ring->scratch.obj);
 err_unref:
 	drm_gem_object_unreference(&ring->scratch.obj->base);
 err:
 	return ret;
 }
-static int init_render_ring(struct intel_ring_buffer *ring)
+static int init_render_ring(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret = init_ring_common(ring);
+	if (ret)
-	struct drm_i915_private *dev_priv = dev->dev_private;
+		return ret;
 	int ret = init_ring_common(ring);
 	/* WaTimedSingleVertexDispatch:cl,bw,ctg,elk,ilk,snb */
-	if (INTEL_INFO(dev)->gen > 3)
+	if (INTEL_INFO(dev)->gen >= 4 && INTEL_INFO(dev)->gen < 7)
 		I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(VS_TIMER_DISPATCH));
 	/* We need to disable the AsyncFlip performance optimisations in order
 	 * to use MI_WAIT_FOR_EVENT within the CS. It should already be
 	 * programmed to '1' on all products.
+	 *
-	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv
+	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv,bdw,chv
+	 */
-	 */
+	if (INTEL_INFO(dev)->gen >= 6)
-	if (INTEL_INFO(dev)->gen >= 6)
+		I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(ASYNC_FLIP_PERF_DISABLE));
-		I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(ASYNC_FLIP_PERF_DISABLE));
+	/* Required for the hardware to program scanline values for waiting */
 	/* WaEnableFlushTlbInvalidationMode:snb */
-	/* Required for the hardware to program scanline values for waiting */
+	if (INTEL_INFO(dev)->gen == 6)
-	if (INTEL_INFO(dev)->gen == 6)
+		I915_WRITE(GFX_MODE,
-		I915_WRITE(GFX_MODE,
+			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_EXPLICIT));
 			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_ALWAYS));
 	/* WaBCSVCSTlbInvalidationMode:ivb,vlv,hsw */
 		if (IS_GEN7(dev))
 		 *  policy. [...] This bit must be reset.  LRA replacement
 		 *  policy is not supported."
 		 */
 		I915_WRITE(CACHE_MODE_0,
 			   _MASKED_BIT_DISABLE(CM0_STC_EVICT_DISABLE_LRA_SNB));
-		/* This is not explicitly set for GEN6, so read the register.
-		 * see intel_ring_mi_set_context() for why we care.
-		 * TODO: consider explicitly setting the bit for GEN5
-		 */
-		ring->itlb_before_ctx_switch =
-			!!(I915_READ(GFX_MODE) & GFX_TLB_INVALIDATE_ALWAYS);
+	}
 	if (INTEL_INFO(dev)->gen >= 6)
 		I915_WRITE_IMR(ring, ~GT_PARITY_ERROR(dev));
 	return ret;
 }
-static void render_ring_cleanup(struct intel_ring_buffer *ring)
+static void render_ring_cleanup(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
 	if (ring->scratch.obj == NULL)
 		return;
 	if (INTEL_INFO(dev)->gen >= 5) {
-//       kunmap(sg_page(ring->scratch.obj->pages->sgl));
+//		kunmap(sg_page(ring->scratch.obj->pages->sgl));
-		i915_gem_object_unpin(ring->scratch.obj);
+		i915_gem_object_ggtt_unpin(ring->scratch.obj);
-	}
+	}
-	drm_gem_object_unreference(&ring->scratch.obj->base);
-	ring->scratch.obj = NULL;
+	drm_gem_object_unreference(&ring->scratch.obj->base);
+	ring->scratch.obj = NULL;
+}
+static int gen8_rcs_signal(struct intel_engine_cs *signaller,
+			   unsigned int num_dwords)
+{
+#define MBOX_UPDATE_DWORDS 8
+	struct drm_device *dev = signaller->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *waiter;
+	int i, ret, num_rings;
+	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
+#undef MBOX_UPDATE_DWORDS
-}
+	ret = intel_ring_begin(signaller, num_dwords);
+	if (ret)
+		return ret;
+	for_each_ring(waiter, dev_priv, i) {
+		u64 gtt_offset = signaller->semaphore.signal_ggtt[i];
+		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
+			continue;
+		intel_ring_emit(signaller, GFX_OP_PIPE_CONTROL(6));
+		intel_ring_emit(signaller, PIPE_CONTROL_GLOBAL_GTT_IVB |
+					   PIPE_CONTROL_QW_WRITE |
+					   PIPE_CONTROL_FLUSH_ENABLE);
+		intel_ring_emit(signaller, lower_32_bits(gtt_offset));
+		intel_ring_emit(signaller, upper_32_bits(gtt_offset));
+		intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
+		intel_ring_emit(signaller, 0);
+		intel_ring_emit(signaller, MI_SEMAPHORE_SIGNAL |
+					   MI_SEMAPHORE_TARGET(waiter->id));
-static void
+		intel_ring_emit(signaller, 0);
+	}
+	return 0;
+}
+static int gen8_xcs_signal(struct intel_engine_cs *signaller,
+			   unsigned int num_dwords)
+{
+#define MBOX_UPDATE_DWORDS 6
+	struct drm_device *dev = signaller->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *waiter;
+	int i, ret, num_rings;
+	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
+#undef MBOX_UPDATE_DWORDS
-update_mboxes(struct intel_ring_buffer *ring,
+	ret = intel_ring_begin(signaller, num_dwords);
+	if (ret)
+		return ret;
+	for_each_ring(waiter, dev_priv, i) {
+		u64 gtt_offset = signaller->semaphore.signal_ggtt[i];
+		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
+			continue;
+		intel_ring_emit(signaller, (MI_FLUSH_DW + 1) |
+					   MI_FLUSH_DW_OP_STOREDW);
+		intel_ring_emit(signaller, lower_32_bits(gtt_offset) |
-	    u32 mmio_offset)
+					   MI_FLUSH_DW_USE_GTT);
+		intel_ring_emit(signaller, upper_32_bits(gtt_offset));
+		intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
+		intel_ring_emit(signaller, MI_SEMAPHORE_SIGNAL |
+					   MI_SEMAPHORE_TARGET(waiter->id));
+		intel_ring_emit(signaller, 0);
+	}
+	return 0;
+}
+static int gen6_signal(struct intel_engine_cs *signaller,
+		       unsigned int num_dwords)
 {
+	struct drm_device *dev = signaller->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *useless;
+	int i, ret, num_rings;
+#define MBOX_UPDATE_DWORDS 3
+	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+	num_dwords += round_up((num_rings-1) * MBOX_UPDATE_DWORDS, 2);
+#undef MBOX_UPDATE_DWORDS
+	ret = intel_ring_begin(signaller, num_dwords);
-/* NB: In order to be able to do semaphore MBOX updates for varying number
+	if (ret)
- * of rings, it's easiest if we round up each individual update to a
+		return ret;
- * multiple of 2 (since ring updates must always be a multiple of 2)
+	for_each_ring(useless, dev_priv, i) {
+		u32 mbox_reg = signaller->semaphore.mbox.signal[i];
+		if (mbox_reg != GEN6_NOSYNC) {
+			intel_ring_emit(signaller, MI_LOAD_REGISTER_IMM(1));
+			intel_ring_emit(signaller, mbox_reg);
- * even though the actual update only requires 3 dwords.
+			intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
+		}
+	}
  */
-#define MBOX_UPDATE_DWORDS 4
+	/* If num_dwords was rounded, make sure the tail pointer is correct */
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+	if (num_rings % 2 == 0)
-	intel_ring_emit(ring, mmio_offset);
+		intel_ring_emit(signaller, MI_NOOP);
+ *
  * Update the mailbox registers in the *other* rings with the current seqno.
  * This acts like a signal in the canonical semaphore.
  */
 static int
-gen6_add_request(struct intel_ring_buffer *ring)
+gen6_add_request(struct intel_engine_cs *ring)
+{
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *useless;
-	int i, ret, num_dwords = 4;
+	int ret;
-	if (i915_semaphore_is_enabled(dev))
+	if (ring->semaphore.signal)
+	ret = ring->semaphore.signal(ring, 4);
-		num_dwords += ((I915_NUM_RINGS-1) * MBOX_UPDATE_DWORDS);
+	else
-#undef MBOX_UPDATE_DWORDS
 		ret = intel_ring_begin(ring, 4);
 	ret = intel_ring_begin(ring, num_dwords);
-	if (ret)
-		return ret;
-	if (i915_semaphore_is_enabled(dev)) {
-	for_each_ring(useless, dev_priv, i) {
-		u32 mbox_reg = ring->signal_mbox[i];
-		if (mbox_reg != GEN6_NOSYNC)
-			update_mboxes(ring, mbox_reg);
 	if (ret)
 		return ret;
 	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
 	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
+ *
  * @waiter - ring that is waiting
  * @signaller - ring which has, or will signal
  * @seqno - seqno which the waiter will block on
  */
 static int
-gen6_ring_sync(struct intel_ring_buffer *waiter,
+gen8_ring_sync(struct intel_engine_cs *waiter,
-		struct intel_ring_buffer *signaller,
+	       struct intel_engine_cs *signaller,
-		u32 seqno)
+	       u32 seqno)
+{
+	struct drm_i915_private *dev_priv = waiter->dev->dev_private;
 	int ret;
+	ret = intel_ring_begin(waiter, 4);
+	if (ret)
+		return ret;
+	intel_ring_emit(waiter, MI_SEMAPHORE_WAIT |
+				MI_SEMAPHORE_GLOBAL_GTT |
+				MI_SEMAPHORE_POLL |
+				MI_SEMAPHORE_SAD_GTE_SDD);
+	intel_ring_emit(waiter, seqno);
+	intel_ring_emit(waiter,
+			lower_32_bits(GEN8_WAIT_OFFSET(waiter, signaller->id)));
+	intel_ring_emit(waiter,
+			upper_32_bits(GEN8_WAIT_OFFSET(waiter, signaller->id)));
+	intel_ring_advance(waiter);
+	return 0;
+}
+static int
+gen6_ring_sync(struct intel_engine_cs *waiter,
+	       struct intel_engine_cs *signaller,
+		u32 seqno)
+{
 	u32 dw1 = MI_SEMAPHORE_MBOX |
 		  MI_SEMAPHORE_COMPARE |
 		  MI_SEMAPHORE_REGISTER;
+	u32 wait_mbox = signaller->semaphore.mbox.wait[waiter->id];
+	int ret;
 	/* Throughout all of the GEM code, seqno passed implies our current
 	 * seqno is >= the last seqno executed. However for hardware the
 	 * comparison is strictly greater than.
 	 */
-	seqno -= 1;
+	seqno -= 1;
 	WARN_ON(signaller->semaphore_register[waiter->id] ==
-		MI_SEMAPHORE_SYNC_INVALID);
+	WARN_ON(wait_mbox == MI_SEMAPHORE_SYNC_INVALID);
 	ret = intel_ring_begin(waiter, 4);
 	if (ret)
 		return ret;
-	/* If seqno wrap happened, omit the wait with no-ops */
 	if (likely(!i915_gem_has_seqno_wrapped(waiter->dev, seqno))) {
-	intel_ring_emit(waiter,
+	/* If seqno wrap happened, omit the wait with no-ops */
-				dw1 |
+	if (likely(!i915_gem_has_seqno_wrapped(waiter->dev, seqno))) {
-				signaller->semaphore_register[waiter->id]);
+		intel_ring_emit(waiter, dw1 | wait_mbox);
 	intel_ring_emit(waiter, seqno);
 	intel_ring_emit(ring__, 0);							\
 	intel_ring_emit(ring__, 0);							\
 } while (0)
 static int
-pc_render_add_request(struct intel_ring_buffer *ring)
+pc_render_add_request(struct intel_engine_cs *ring)
 {
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 	/* For Ironlake, MI_USER_INTERRUPT was deprecated and apparently
 			PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE);
 	intel_ring_emit(ring, ring->scratch.gtt_offset | PIPE_CONTROL_GLOBAL_GTT);
 	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, 0);
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128; /* write to separate cachelines */
+	scratch_addr += 2 * CACHELINE_BYTES; /* write to separate cachelines */
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
 	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE |
 			PIPE_CONTROL_WRITE_FLUSH |
 	return 0;
+}
 static u32
-gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+gen6_ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+{
 	/* Workaround to force correct ordering between irq and seqno writes on
 	 * ivb (and maybe also on snb) by reading from a CS register (like
+	 * ACTHD) before reading the status page. */
-	 * ACTHD) before reading the status page. */
+	if (!lazy_coherency) {
+		struct drm_i915_private *dev_priv = ring->dev->dev_private;
+		POSTING_READ(RING_ACTHD(ring->mmio_base));
 	if (!lazy_coherency)
 		intel_ring_get_active_head(ring);
 	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
 }
 static u32
-ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+{
 	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
+}
 static void
-ring_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+ring_set_seqno(struct intel_engine_cs *ring, u32 seqno)
 {
 	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
+}
 static u32
-pc_render_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+pc_render_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+{
 	return ring->scratch.cpu_page[0];
+}
 static void
-pc_render_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+pc_render_set_seqno(struct intel_engine_cs *ring, u32 seqno)
+{
 	ring->scratch.cpu_page[0] = seqno;
+}
 static bool
-gen5_ring_get_irq(struct intel_ring_buffer *ring)
+gen5_ring_get_irq(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 	if (!dev->irq_enabled)
 		return false;
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 	if (ring->irq_refcount++ == 0)
-		ilk_enable_gt_irq(dev_priv, ring->irq_enable_mask);
+		gen5_enable_gt_irq(dev_priv, ring->irq_enable_mask);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 	return true;
+}
 static void
-gen5_ring_put_irq(struct intel_ring_buffer *ring)
+gen5_ring_put_irq(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 	return true;
+}
 static void
-i9xx_ring_put_irq(struct intel_ring_buffer *ring)
+i9xx_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
+	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
 static bool
-i8xx_ring_get_irq(struct intel_ring_buffer *ring)
+i8xx_ring_get_irq(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 	return true;
+}
 static void
-i8xx_ring_put_irq(struct intel_ring_buffer *ring)
+i8xx_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 		POSTING_READ16(IMR);
+	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
-void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
+void intel_ring_setup_status_page(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u32 mmio = 0;
 	/* The ring status page addresses are no longer next to the rest of
 			mmio = RENDER_HWS_PGA_GEN7;
 			break;
 		case BCS:
 			mmio = BLT_HWS_PGA_GEN7;
 			break;
+		/*
+		 * VCS2 actually doesn't exist on Gen7. Only shut up
+		 * gcc switch check warning
+		 */
+		case VCS2:
 		case VCS:
 			mmio = BSD_HWS_PGA_GEN7;
 			break;
 		case VECS:
 			mmio = VEBOX_HWS_PGA_GEN7;
+	}
 	I915_WRITE(mmio, (u32)ring->status_page.gfx_addr);
+	POSTING_READ(mmio);
-	POSTING_READ(mmio);
+	/*
+	 * Flush the TLB for this page
+	 *
+	 * FIXME: These two bits have disappeared on gen8, so a question
+	 * arises: do we still need this and if so how should we go about
+	 * invalidating the TLB?
-	/* Flush the TLB for this page */
+	 */
+	if (INTEL_INFO(dev)->gen >= 6 && INTEL_INFO(dev)->gen < 8) {
+		u32 reg = RING_INSTPM(ring->mmio_base);
+		/* ring should be idle before issuing a sync flush*/
-	if (INTEL_INFO(dev)->gen >= 6) {
+		WARN_ON((I915_READ_MODE(ring) & MODE_IDLE) == 0);
 		u32 reg = RING_INSTPM(ring->mmio_base);
 		I915_WRITE(reg,
 			   _MASKED_BIT_ENABLE(INSTPM_TLB_INVALIDATE |
 					      INSTPM_SYNC_FLUSH));
 				  ring->name);
+	}
+}
 static int
-bsd_ring_flush(struct intel_ring_buffer *ring,
+bsd_ring_flush(struct intel_engine_cs *ring,
 	       u32     invalidate_domains,
 	       u32     flush_domains)
+{
 	intel_ring_advance(ring);
 	return 0;
+}
 static int
-i9xx_add_request(struct intel_ring_buffer *ring)
+i9xx_add_request(struct intel_engine_cs *ring)
+{
 	int ret;
 	return 0;
+}
 static bool
-gen6_ring_get_irq(struct intel_ring_buffer *ring)
+gen6_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 			I915_WRITE_IMR(ring,
 				       ~(ring->irq_enable_mask |
 					 GT_PARITY_ERROR(dev)));
 		else
 			I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
-		ilk_enable_gt_irq(dev_priv, ring->irq_enable_mask);
+		gen5_enable_gt_irq(dev_priv, ring->irq_enable_mask);
+	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
     return true;
+}
 static void
-gen6_ring_put_irq(struct intel_ring_buffer *ring)
+gen6_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long flags;
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 	if (--ring->irq_refcount == 0) {
 		if (HAS_L3_DPF(dev) && ring->id == RCS)
 			I915_WRITE_IMR(ring, ~GT_PARITY_ERROR(dev));
 		else
 			I915_WRITE_IMR(ring, ~0);
-		ilk_disable_gt_irq(dev_priv, ring->irq_enable_mask);
+		gen5_disable_gt_irq(dev_priv, ring->irq_enable_mask);
+	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
 static bool
-hsw_vebox_get_irq(struct intel_ring_buffer *ring)
+hsw_vebox_get_irq(struct intel_engine_cs *ring)
 		return false;
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 	if (ring->irq_refcount++ == 0) {
 		I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
-		snb_enable_pm_irq(dev_priv, ring->irq_enable_mask);
+		gen6_enable_pm_irq(dev_priv, ring->irq_enable_mask);
+	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 	return true;
 }
 static void
-hsw_vebox_put_irq(struct intel_ring_buffer *ring)
+hsw_vebox_put_irq(struct intel_engine_cs *ring)
+{
 		return;
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 	if (--ring->irq_refcount == 0) {
 		I915_WRITE_IMR(ring, ~0);
-		snb_disable_pm_irq(dev_priv, ring->irq_enable_mask);
+		gen6_disable_pm_irq(dev_priv, ring->irq_enable_mask);
+	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
 static bool
-gen8_ring_get_irq(struct intel_ring_buffer *ring)
+gen8_ring_get_irq(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
 	return true;
+}
 static void
-gen8_ring_put_irq(struct intel_ring_buffer *ring)
+gen8_ring_put_irq(struct intel_engine_cs *ring)
+{
 	struct drm_device *dev = ring->dev;
+	}
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
 static int
-i965_dispatch_execbuffer(struct intel_ring_buffer *ring,
+i965_dispatch_execbuffer(struct intel_engine_cs *ring,
-			 u32 offset, u32 length,
+			 u64 offset, u32 length,
 			 unsigned flags)
+{
 	int ret;
+}
 /* Just userspace ABI convention to limit the wa batch bo to a resonable size */
 #define I830_BATCH_LIMIT (256*1024)
 static int
-i830_dispatch_execbuffer(struct intel_ring_buffer *ring,
+i830_dispatch_execbuffer(struct intel_engine_cs *ring,
-				u32 offset, u32 len,
+				u64 offset, u32 len,
 				unsigned flags)
+{
 	int ret;
 	return 0;
+}
 static int
-i915_dispatch_execbuffer(struct intel_ring_buffer *ring,
+i915_dispatch_execbuffer(struct intel_engine_cs *ring,
-			 u32 offset, u32 len,
+			 u64 offset, u32 len,
 			 unsigned flags)
+{
 	intel_ring_advance(ring);
 	return 0;
 }
-static void cleanup_status_page(struct intel_ring_buffer *ring)
+static void cleanup_status_page(struct intel_engine_cs *ring)
+{
 	struct drm_i915_gem_object *obj;
 	obj = ring->status_page.obj;
 	if (obj == NULL)
 		return;
-//   kunmap(sg_page(obj->pages->sgl));
+//	kunmap(sg_page(obj->pages->sgl));
-    i915_gem_object_unpin(obj);
+	i915_gem_object_ggtt_unpin(obj);
 	drm_gem_object_unreference(&obj->base);
-	ring->status_page.obj = NULL;
 	ring->status_page.obj = NULL;
+}
+static int init_status_page(struct intel_engine_cs *ring)
+{
-static int init_status_page(struct intel_ring_buffer *ring)
+	struct drm_i915_gem_object *obj;
-	struct drm_device *dev = ring->dev;
+	if ((obj = ring->status_page.obj) == NULL) {
-	struct drm_i915_gem_object *obj;
+		unsigned flags;
-	int ret;
 	int ret;
-	obj = i915_gem_alloc_object(dev, 4096);
-	if (obj == NULL) {
-		DRM_ERROR("Failed to allocate status page\n");
-		ret = -ENOMEM;
+		obj = i915_gem_alloc_object(ring->dev, 4096);
-		goto err;
+	if (obj == NULL) {
+		DRM_ERROR("Failed to allocate status page\n");
+			return -ENOMEM;
+	}
+	ret = i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+	if (ret)
+		goto err_unref;
+		flags = 0;
+		if (!HAS_LLC(ring->dev))
+			/* On g33, we cannot place HWS above 256MiB, so
+			 * restrict its pinning to the low mappable arena.
+			 * Though this restriction is not documented for
+			 * gen4, gen5, or byt, they also behave similarly
+			 * and hang if the HWS is placed at the top of the
+			 * GTT. To generalise, it appears that all !llc
+			 * platforms have issues with us placing the HWS
+			 * above the mappable region (even though we never
+			 * actualy map it).
+			 */
+			flags |= PIN_MAPPABLE;
+		ret = i915_gem_obj_ggtt_pin(obj, 4096, flags);
 		if (ret) {
 err_unref:
-	i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+			drm_gem_object_unreference(&obj->base);
-	ret = i915_gem_obj_ggtt_pin(obj, 4096, true, false);
-	if (ret != 0) {
-		goto err_unref;
-	}
 			return ret;
 	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(obj);
     ring->status_page.page_addr = (void*)MapIoMem((addr_t)sg_page(obj->pages->sgl),4096,PG_SW|0x100);
-	if (ring->status_page.page_addr == NULL) {
+		ring->status_page.obj = obj;
-		ret = -ENOMEM;
-		goto err_unpin;
-	}
-	ring->status_page.obj = obj;
-	memset(ring->status_page.page_addr, 0, PAGE_SIZE);
-	DRM_DEBUG_DRIVER("%s hws offset: 0x%08x\n",
 			ring->name, ring->status_page.gfx_addr);
-	return 0;
+	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(obj);
     ring->status_page.page_addr = (void*)MapIoMem((addr_t)sg_page(obj->pages->sgl),4096,PG_SW|0x100);
-err_unpin:
+	memset(ring->status_page.page_addr, 0, PAGE_SIZE);
 	i915_gem_object_unpin(obj);
     memset(ring->status_page.page_addr, 0, PAGE_SIZE);
     return 0;
 }
 static int intel_init_ring_buffer(struct drm_device *dev,
-			   struct intel_ring_buffer *ring)
+static void intel_destroy_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
-{
-	struct drm_i915_gem_object *obj;
+{
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	if (!ringbuf->obj)
-	int ret;
+		return;
-	ring->dev = dev;
+	iounmap(ringbuf->virtual_start);
-	INIT_LIST_HEAD(&ring->active_list);
+	i915_gem_object_ggtt_unpin(ringbuf->obj);
+	drm_gem_object_unreference(&ringbuf->obj->base);
-	INIT_LIST_HEAD(&ring->request_list);
+	ringbuf->obj = NULL;
+}
+static int intel_alloc_ringbuffer_obj(struct drm_device *dev,
+				      struct intel_ringbuffer *ringbuf)
-	ring->size = 32 * PAGE_SIZE;
-	memset(ring->sync_seqno, 0, sizeof(ring->sync_seqno));
-	init_waitqueue_head(&ring->irq_queue);
-	if (I915_NEED_GFX_HWS(dev)) {
-       ret = init_status_page(ring);
-       if (ret)
+{
-           return ret;
+	struct drm_i915_private *dev_priv = to_i915(dev);
-	} else {
-		BUG_ON(ring->id != RCS);
+	struct drm_i915_gem_object *obj;
-		ret = init_phys_status_page(ring);
+	int ret;
-		if (ret)
-			return ret;
+	if (ringbuf->obj)
-	}
+		return 0;
-	obj = NULL;
-	if (!HAS_LLC(dev))
+	obj = NULL;
-		obj = i915_gem_object_create_stolen(dev, ring->size);
-	if (obj == NULL)
+	if (!HAS_LLC(dev))
-        obj = i915_gem_alloc_object(dev, ring->size);
+		obj = i915_gem_object_create_stolen(dev, ringbuf->size);
-	if (obj == NULL) {
+	if (obj == NULL)
-		DRM_ERROR("Failed to allocate ringbuffer\n");
+		obj = i915_gem_alloc_object(dev, ringbuf->size);
-		ret = -ENOMEM;
+	if (obj == NULL)
-		goto err_hws;
+		return -ENOMEM;
 	/* mark ring buffers as read-only from GPU side by default */
-	ring->obj = obj;
+	obj->gt_ro = 1;
-	ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, true, false);
+	ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, PIN_MAPPABLE);
 	if (ret)
-		goto err_unref;
 		goto err_unref;
 	ret = i915_gem_object_set_to_gtt_domain(obj, true);
-	if (ret)
+	ret = i915_gem_object_set_to_gtt_domain(obj, true);
+	if (ret)
+		goto err_unpin;
+	ringbuf->virtual_start =
+		ioremap_wc(dev_priv->gtt.mappable_base + i915_gem_obj_ggtt_offset(obj),
+				ringbuf->size);
+	if (ringbuf->virtual_start == NULL) {
+		ret = -EINVAL;
+		goto err_unpin;
+	}
+	ringbuf->obj = obj;
+	return 0;
+err_unpin:
+	i915_gem_object_ggtt_unpin(obj);
+err_unref:
+	drm_gem_object_unreference(&obj->base);
+	return ret;
+}
+static int intel_init_ring_buffer(struct drm_device *dev,
+				  struct intel_engine_cs *ring)
+{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	int ret;
+	if (ringbuf == NULL) {
+		ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
+		if (!ringbuf)
+			return -ENOMEM;
+		ring->buffer = ringbuf;
-		goto err_unpin;
+	}
+	ring->dev = dev;
+	INIT_LIST_HEAD(&ring->active_list);
+	INIT_LIST_HEAD(&ring->request_list);
+	ringbuf->size = 32 * PAGE_SIZE;
 	memset(ring->semaphore.sync_seqno, 0, sizeof(ring->semaphore.sync_seqno));
+	init_waitqueue_head(&ring->irq_queue);
+	if (I915_NEED_GFX_HWS(dev)) {
+		ret = init_status_page(ring);
+		if (ret)
-	ring->virtual_start =
+			goto error;
+	} else {
-		ioremap_wc(dev_priv->gtt.mappable_base + i915_gem_obj_ggtt_offset(obj),
+		BUG_ON(ring->id != RCS);
-			   ring->size);
+		ret = init_phys_status_page(ring);
-	if (ring->virtual_start == NULL) {
+	if (ret)
-		DRM_ERROR("Failed to map ringbuffer.\n");
+			goto error;
-		ret = -EINVAL;
+	}
-		goto err_unpin;
-	}
+	ret = intel_alloc_ringbuffer_obj(dev, ringbuf);
+	if (ret) {
+		DRM_ERROR("Failed to allocate ringbuffer %s: %d\n", ring->name, ret);
+		goto error;
+	}
+	/* Workaround an erratum on the i830 which causes a hang if
+	 * the TAIL pointer points to within the last 2 cachelines
+	 * of the buffer.
 	 */
-	ret = ring->init(ring);
+	ringbuf->effective_size = ringbuf->size;
-	if (ret)
+	if (IS_I830(dev) || IS_845G(dev))
-		goto err_unmap;
-	/* Workaround an erratum on the i830 which causes a hang if
-	 * the TAIL pointer points to within the last 2 cachelines
-	 * of the buffer.
+		ringbuf->effective_size -= 2 * CACHELINE_BYTES;
-	 */
-	ring->effective_size = ring->size;
 	if (IS_I830(ring->dev) || IS_845G(ring->dev))
-		ring->effective_size -= 128;
+	ret = i915_cmd_parser_init_ring(ring);
+	if (ret)
-	return 0;
+		goto error;
-err_unmap:
+	ret = ring->init(ring);
-	iounmap(ring->virtual_start);
+	if (ret)
-err_unpin:
+		goto error;
-	i915_gem_object_unpin(obj);
-err_unref:
-	drm_gem_object_unreference(&obj->base);
-	ring->obj = NULL;
-err_hws:
+	return 0;
-//   cleanup_status_page(ring);
-	return ret;
-}
-void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring)
-{
-	struct drm_i915_private *dev_priv;
-	int ret;
 error:
-	if (ring->obj == NULL)
+	kfree(ringbuf);
-		return;
+	ring->buffer = NULL;
 	return ret;
 	/* Disable the ring buffer. The ring must be idle at this point */
-	dev_priv = ring->dev->dev_private;
-	ret = intel_ring_idle(ring);
-	if (ret && !i915_reset_in_progress(&dev_priv->gpu_error))
-		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
-			  ring->name, ret);
+void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
-	I915_WRITE_CTL(ring, 0);
-	iounmap(ring->virtual_start);
+{
 	struct drm_i915_private *dev_priv = to_i915(ring->dev);
-    i915_gem_object_unpin(ring->obj);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	drm_gem_object_unreference(&ring->obj->base);
+	if (!intel_ring_initialized(ring))
-	ring->obj = NULL;
+		return;
 	ring->preallocated_lazy_request = NULL;
-	ring->outstanding_lazy_seqno = 0;
+	intel_stop_ring_buffer(ring);
+	WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
+	intel_destroy_ringbuffer_obj(ringbuf);
-	if (ring->cleanup)
-		ring->cleanup(ring);
-//   cleanup_status_page(ring);
+	ring->preallocated_lazy_request = NULL;
-}
+	ring->outstanding_lazy_seqno = 0;
-static int intel_ring_wait_seqno(struct intel_ring_buffer *ring, u32 seqno)
+	if (ring->cleanup)
 		ring->cleanup(ring);
-	int ret;
-	ret = i915_wait_seqno(ring, seqno);
-	if (!ret)
-		i915_gem_retire_requests_ring(ring);
-	return ret;
-}
-static int intel_ring_wait_request(struct intel_ring_buffer *ring, int n)
+//	cleanup_status_page(ring);
-	struct drm_i915_gem_request *request;
+	i915_cmd_parser_fini_ring(ring);
-	u32 seqno = 0;
-	int ret;
-	i915_gem_retire_requests_ring(ring);
-	if (ring->last_retired_head != -1) {
-		ring->head = ring->last_retired_head;
 		ring->last_retired_head = -1;
-		ring->space = ring_space(ring);
+	kfree(ringbuf);
-		if (ring->space >= n)
+	ring->buffer = NULL;
-			return 0;
+}
 static int intel_ring_wait_request(struct intel_engine_cs *ring, int n)
-	list_for_each_entry(request, &ring->request_list, list) {
+{
-		int space;
-		if (request->tail == -1)
+	struct intel_ringbuffer *ringbuf = ring->buffer;
-			continue;
+	struct drm_i915_gem_request *request;
-		space = request->tail - (ring->tail + I915_RING_FREE_SPACE);
-		if (space < 0)
+	u32 seqno = 0;
-			space += ring->size;
+	int ret;
 		if (space >= n) {
-			seqno = request->seqno;
+	if (ringbuf->last_retired_head != -1) {
-			break;
+		ringbuf->head = ringbuf->last_retired_head;
 		ringbuf->last_retired_head = -1;
+		ringbuf->space = ring_space(ringbuf);
-		/* Consume this request in case we need more space than
+		if (ringbuf->space >= n)
-		 * is available and so need to prevent a race between
+			return 0;
 		 * updating last_retired_head and direct reads of
 		 * I915_RING_HEAD. It also provides a nice sanity check.
-		 */
+	list_for_each_entry(request, &ring->request_list, list) {
-		request->tail = -1;
+		if (__ring_space(request->tail, ringbuf->tail, ringbuf->size) >= n) {
 			seqno = request->seqno;
-	if (seqno == 0)
+			break;
 		return -ENOSPC;
+	}
 	ret = intel_ring_wait_seqno(ring, seqno);
-	if (ret)
+	if (seqno == 0)
-		return ret;
+		return -ENOSPC;
 	ret = i915_wait_seqno(ring, seqno);
-	if (WARN_ON(ring->last_retired_head == -1))
+	if (ret)
-		return -ENOSPC;
+		return ret;
-	ring->head = ring->last_retired_head;
+	i915_gem_retire_requests_ring(ring);
-	ring->last_retired_head = -1;
+	ringbuf->head = ringbuf->last_retired_head;
-	ring->space = ring_space(ring);
+	ringbuf->last_retired_head = -1;
-	if (WARN_ON(ring->space < n))
+	ringbuf->space = ring_space(ringbuf);
-		return -ENOSPC;
+	return 0;
+}
 	return 0;
-}
+static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
+{
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	unsigned long end;
+	int ret;
 static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
-{
+	ret = intel_ring_wait_request(ring, n);
-	struct drm_device *dev = ring->dev;
+	if (ret != -ENOSPC)
-	struct drm_i915_private *dev_priv = dev->dev_private;
+		return ret;
 	unsigned long end;
-	int ret;
+	/* force the tail write in case we have been skipping them */
+	__intel_ring_advance(ring);
-	ret = intel_ring_wait_request(ring, n);
+	/* With GEM the hangcheck timer should kick us out of the loop,
-	if (ret != -ENOSPC)
+	 * leaving it early runs the risk of corrupting GEM state (due
-		return ret;
+	 * to running on almost untested codepaths). But on resume
 	 * timers don't work yet, so prevent a complete hang in that
-	/* force the tail write in case we have been skipping them */
+	 * case by choosing an insanely large timeout. */
-	__intel_ring_advance(ring);
+	end = jiffies + 60 * HZ;
 	trace_i915_ring_wait_begin(ring);
-	/* With GEM the hangcheck timer should kick us out of the loop,
+	do {
-	 * leaving it early runs the risk of corrupting GEM state (due
+		ringbuf->head = I915_READ_HEAD(ring);
-	 * to running on almost untested codepaths). But on resume
+		ringbuf->space = ring_space(ringbuf);
-	 * timers don't work yet, so prevent a complete hang in that
+		if (ringbuf->space >= n) {
-	 * case by choosing an insanely large timeout. */
+			ret = 0;
-	end = GetTimerTicks() + 60 * HZ;
+			break;
+		}
 	do {
 		ring->head = I915_READ_HEAD(ring);
-		ring->space = ring_space(ring);
+		msleep(1);
 	return i915_wait_seqno(ring, seqno);
+}
 static int
-intel_ring_alloc_seqno(struct intel_ring_buffer *ring)
+intel_ring_alloc_seqno(struct intel_engine_cs *ring)
+{
 	if (ring->outstanding_lazy_seqno)
+	}
 	return i915_gem_get_seqno(ring->dev, &ring->outstanding_lazy_seqno);
 }
-static int __intel_ring_prepare(struct intel_ring_buffer *ring,
+static int __intel_ring_prepare(struct intel_engine_cs *ring,
+			      int bytes)
 			      int bytes)
-{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	int ret;
-	if (unlikely(ring->tail + bytes > ring->effective_size)) {
+	if (unlikely(ringbuf->tail + bytes > ringbuf->effective_size)) {
 		ret = intel_wrap_ring_buffer(ring);
 		if (unlikely(ret))
 			return ret;
+	}
-	if (unlikely(ring->space < bytes)) {
+	if (unlikely(ringbuf->space < bytes)) {
 		ret = ring_wait_for_space(ring, bytes);
 		if (unlikely(ret))
 			return ret;
+	}
 	return 0;
+}
-int intel_ring_begin(struct intel_ring_buffer *ring,
+int intel_ring_begin(struct intel_engine_cs *ring,
 		     int num_dwords)
 	/* Preallocate the olr before touching the ring */
 	ret = intel_ring_alloc_seqno(ring);
 	if (ret)
 		return ret;
-	ring->space -= num_dwords * sizeof(uint32_t);
+	ring->buffer->space -= num_dwords * sizeof(uint32_t);
 	return 0;
+}
-}
 /* Align the ring tail to a cacheline boundary */
+int intel_ring_cacheline_align(struct intel_engine_cs *ring)
+{
+	int num_dwords = (ring->buffer->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
+	int ret;
+	if (num_dwords == 0)
+		return 0;
+	num_dwords = CACHELINE_BYTES / sizeof(uint32_t) - num_dwords;
+	ret = intel_ring_begin(ring, num_dwords);
+	if (ret)
+		return ret;
+	while (num_dwords--)
+		intel_ring_emit(ring, MI_NOOP);
+	intel_ring_advance(ring);
+	return 0;
+}
+void intel_ring_init_seqno(struct intel_engine_cs *ring, u32 seqno)
-void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
+{
 	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	BUG_ON(ring->outstanding_lazy_seqno);
-	if (INTEL_INFO(ring->dev)->gen >= 6) {
+	if (INTEL_INFO(dev)->gen == 6 || INTEL_INFO(dev)->gen == 7) {
 		I915_WRITE(RING_SYNC_0(ring->mmio_base), 0);
 		I915_WRITE(RING_SYNC_1(ring->mmio_base), 0);
-		if (HAS_VEBOX(ring->dev))
+		if (HAS_VEBOX(dev))
 			I915_WRITE(RING_SYNC_2(ring->mmio_base), 0);
 	}
 	ring->set_seqno(ring, seqno);
 	ring->hangcheck.seqno = seqno;
+}
-static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring,
+static void gen6_bsd_ring_write_tail(struct intel_engine_cs *ring,
 	 */
        I915_WRITE(GEN6_BSD_SLEEP_PSMI_CONTROL,
 		   _MASKED_BIT_DISABLE(GEN6_BSD_SLEEP_MSG_DISABLE));
+}
-static int gen6_bsd_ring_flush(struct intel_ring_buffer *ring,
+static int gen6_bsd_ring_flush(struct intel_engine_cs *ring,
 			   u32 invalidate, u32 flush)
+{
 	uint32_t cmd;
 	intel_ring_advance(ring);
 	return 0;
+}
 static int
-gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
+gen8_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
-			      u32 offset, u32 len,
+			      u64 offset, u32 len,
 			      unsigned flags)
+{
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	bool ppgtt = dev_priv->mm.aliasing_ppgtt != NULL &&
 	if (ret)
 		return ret;
 	/* FIXME(BDW): Address space and security selectors. */
 	intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8));
-	intel_ring_emit(ring, offset);
+	intel_ring_emit(ring, lower_32_bits(offset));
-	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, upper_32_bits(offset));
 	intel_ring_emit(ring, MI_NOOP);
 	intel_ring_advance(ring);
 	return 0;
 }
 static int
-hsw_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
+hsw_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
-			      u32 offset, u32 len,
+			      u64 offset, u32 len,
 			      unsigned flags)
 	return 0;
+}
 static int
-gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
+gen6_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
-			      u32 offset, u32 len,
+			      u64 offset, u32 len,
 			      unsigned flags)
+{
        return 0;
+}
 /* Blitter support (SandyBridge+) */
-static int gen6_ring_flush(struct intel_ring_buffer *ring,
+static int gen6_ring_flush(struct intel_engine_cs *ring,
 			  u32 invalidate, u32 flush)
+{
 	struct drm_device *dev = ring->dev;
 	return 0;
+}
 int intel_init_render_ring_buffer(struct drm_device *dev)
 {
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct drm_i915_gem_object *obj;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	int ret;
 	ring->name = "render ring";
-	ring->id = RCS;
-	ring->mmio_base = RENDER_RING_BASE;
-	if (INTEL_INFO(dev)->gen >= 6) {
-       ring->add_request = gen6_add_request;
-		ring->flush = gen7_render_ring_flush;
+	ring->id = RCS;
+	ring->mmio_base = RENDER_RING_BASE;
+	if (INTEL_INFO(dev)->gen >= 8) {
+		if (i915_semaphore_is_enabled(dev)) {
+			obj = i915_gem_alloc_object(dev, 4096);
+			if (obj == NULL) {
+				DRM_ERROR("Failed to allocate semaphore bo. Disabling semaphores\n");
+				i915.semaphores = 0;
+			} else {
+				i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+				ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_NONBLOCK);
+				if (ret != 0) {
+					drm_gem_object_unreference(&obj->base);
+					DRM_ERROR("Failed to pin semaphore bo. Disabling semaphores\n");
+					i915.semaphores = 0;
+				} else
+					dev_priv->semaphore_obj = obj;
 		if (INTEL_INFO(dev)->gen == 6)
 		ring->flush = gen6_render_ring_flush;
-		if (INTEL_INFO(dev)->gen >= 8) {
+		ring->add_request = gen6_add_request;
+		ring->flush = gen8_render_ring_flush;
+		ring->irq_get = gen8_ring_get_irq;
+		ring->irq_put = gen8_ring_put_irq;
+		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
+		ring->get_seqno = gen6_ring_get_seqno;
+		ring->set_seqno = ring_set_seqno;
+		if (i915_semaphore_is_enabled(dev)) {
+			WARN_ON(!dev_priv->semaphore_obj);
-			ring->flush = gen8_render_ring_flush;
+			ring->semaphore.sync_to = gen8_ring_sync;
+			ring->semaphore.signal = gen8_rcs_signal;
+			GEN8_RING_SEMAPHORE_INIT;
+		}
+	} else if (INTEL_INFO(dev)->gen >= 6) {
+       ring->add_request = gen6_add_request;
-			ring->irq_get = gen8_ring_get_irq;
+		ring->flush = gen7_render_ring_flush;
-			ring->irq_put = gen8_ring_put_irq;
+		if (INTEL_INFO(dev)->gen == 6)
-		} else {
-		ring->irq_get = gen6_ring_get_irq;
+		ring->flush = gen6_render_ring_flush;
-		ring->irq_put = gen6_ring_put_irq;
+		ring->irq_get = gen6_ring_get_irq;
 		ring->irq_put = gen6_ring_put_irq;
+		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
-		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
+		ring->get_seqno = gen6_ring_get_seqno;
+		ring->set_seqno = ring_set_seqno;
+		if (i915_semaphore_is_enabled(dev)) {
+		ring->semaphore.sync_to = gen6_ring_sync;
+		ring->semaphore.signal = gen6_signal;
+		/*
+			 * The current semaphore is only applied on pre-gen8
+			 * platform.  And there is no VCS2 ring on the pre-gen8
+			 * platform. So the semaphore between RCS and VCS2 is
-		ring->get_seqno = gen6_ring_get_seqno;
+			 * initialized as INVALID.  Gen8 will initialize the
-		ring->set_seqno = ring_set_seqno;
+			 * sema between VCS2 and RCS later.
-		ring->sync_to = gen6_ring_sync;
+		 */
-		ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_RV;
-		ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_RV;
+		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_RB;
-		ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_RB;
+		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_RVE;
-		ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_RVE;
+		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->signal_mbox[RCS] = GEN6_NOSYNC;
+		ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
+		ring->semaphore.mbox.signal[VCS] = GEN6_VRSYNC;
+		ring->semaphore.mbox.signal[BCS] = GEN6_BRSYNC;
-		ring->signal_mbox[VCS] = GEN6_VRSYNC;
+		ring->semaphore.mbox.signal[VECS] = GEN6_VERSYNC;
-		ring->signal_mbox[BCS] = GEN6_BRSYNC;
+		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 		ring->signal_mbox[VECS] = GEN6_VERSYNC;
 	} else if (IS_GEN5(dev)) {
        ring->add_request = pc_render_add_request;
 			ring->irq_put = i9xx_ring_put_irq;
+		}
 		ring->irq_enable_mask = I915_USER_INTERRUPT;
+	}
 	ring->write_tail = ring_write_tail;
 	if (IS_HASWELL(dev))
 		ring->dispatch_execbuffer = hsw_ring_dispatch_execbuffer;
 	else if (IS_GEN8(dev))
 		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
 	else if (INTEL_INFO(dev)->gen >= 6)
 	ring->init = init_render_ring;
 	ring->cleanup = render_ring_cleanup;
 	/* Workaround batchbuffer to combat CS tlb bug. */
-	if (HAS_BROKEN_CS_TLB(dev)) {
-		struct drm_i915_gem_object *obj;
-		int ret;
 	if (HAS_BROKEN_CS_TLB(dev)) {
 		obj = i915_gem_alloc_object(dev, I830_BATCH_LIMIT);
 		if (obj == NULL) {
 			DRM_ERROR("Failed to allocate batch bo\n");
 			return -ENOMEM;
 		}
-		ret = i915_gem_obj_ggtt_pin(obj, 0, true, false);
+		ret = i915_gem_obj_ggtt_pin(obj, 0, 0);
 		if (ret != 0) {
 			drm_gem_object_unreference(&obj->base);
 			DRM_ERROR("Failed to ping batch bo\n");
+}
 #if 0
 int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size)
 {
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	int ret;
+	if (ringbuf == NULL) {
+		ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
+		if (!ringbuf)
+			return -ENOMEM;
+		ring->buffer = ringbuf;
 	int ret;
 	ring->name = "render ring";
 	ring->id = RCS;
 	ring->mmio_base = RENDER_RING_BASE;
+	if (INTEL_INFO(dev)->gen >= 6) {
-	if (INTEL_INFO(dev)->gen >= 6) {
+		/* non-kms not supported on gen6+ */
-		/* non-kms not supported on gen6+ */
+		ret = -ENODEV;
-		return -ENODEV;
+		goto err_ringbuf;
+	}
 	ring->dev = dev;
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
-	ring->size = size;
+	ringbuf->size = size;
-	ring->effective_size = ring->size;
+	ringbuf->effective_size = ringbuf->size;
 	if (IS_I830(ring->dev) || IS_845G(ring->dev))
-		ring->effective_size -= 128;
+		ringbuf->effective_size -= 2 * CACHELINE_BYTES;
-	ring->virtual_start = ioremap_wc(start, size);
+	ringbuf->virtual_start = ioremap_wc(start, size);
-	if (ring->virtual_start == NULL) {
+	if (ringbuf->virtual_start == NULL) {
+		DRM_ERROR("can not ioremap virtual address for"
-		DRM_ERROR("can not ioremap virtual address for"
+			  " ring buffer\n");
-			  " ring buffer\n");
+		ret = -ENOMEM;
-		return -ENOMEM;
+		goto err_ringbuf;
+	}
 	if (!I915_NEED_GFX_HWS(dev)) {
 		ret = init_phys_status_page(ring);
+		if (ret)
+			goto err_vstart;
+	}
+	return 0;
+err_vstart:
-		if (ret)
+	iounmap(ringbuf->virtual_start);
-			return ret;
+err_ringbuf:
 	kfree(ringbuf);
 	ring->buffer = NULL;
-	return 0;
+	return ret;
 }
 #endif
 int intel_init_bsd_ring_buffer(struct drm_device *dev)
 				GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT;
 			ring->irq_get = gen8_ring_get_irq;
 			ring->irq_put = gen8_ring_put_irq;
 			ring->dispatch_execbuffer =
 				gen8_ring_dispatch_execbuffer;
+			if (i915_semaphore_is_enabled(dev)) {
+				ring->semaphore.sync_to = gen8_ring_sync;
+				ring->semaphore.signal = gen8_xcs_signal;
+				GEN8_RING_SEMAPHORE_INIT;
+			}
 		} else {
 		ring->irq_enable_mask = GT_BSD_USER_INTERRUPT;
 		ring->irq_get = gen6_ring_get_irq;
 		ring->irq_put = gen6_ring_put_irq;
 			ring->dispatch_execbuffer =
 				gen6_ring_dispatch_execbuffer;
+			if (i915_semaphore_is_enabled(dev)) {
+		ring->semaphore.sync_to = gen6_ring_sync;
+		ring->semaphore.signal = gen6_signal;
+		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VR;
+		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VB;
+		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_VVE;
+		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.signal[RCS] = GEN6_RVSYNC;
+		ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
+		ring->semaphore.mbox.signal[BCS] = GEN6_BVSYNC;
+		ring->semaphore.mbox.signal[VECS] = GEN6_VEVSYNC;
+		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+			}
+		}
-		ring->sync_to = gen6_ring_sync;
-		ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VR;
-		ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_VB;
-		ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_VVE;
-		ring->signal_mbox[RCS] = GEN6_RVSYNC;
-		ring->signal_mbox[VCS] = GEN6_NOSYNC;
-		ring->signal_mbox[BCS] = GEN6_BVSYNC;
-		ring->signal_mbox[VECS] = GEN6_VEVSYNC;
 	} else {
 		ring->mmio_base = BSD_RING_BASE;
 		ring->flush = bsd_ring_flush;
 		ring->add_request = i9xx_add_request;
 		ring->get_seqno = ring_get_seqno;
 	ring->init = init_ring_common;
 	return intel_init_ring_buffer(dev, ring);
+}
+/**
+ * Initialize the second BSD ring for Broadwell GT3.
+ * It is noted that this only exists on Broadwell GT3.
+ */
+int intel_init_bsd2_ring_buffer(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring = &dev_priv->ring[VCS2];
+	if ((INTEL_INFO(dev)->gen != 8)) {
+		DRM_ERROR("No dual-BSD ring on non-BDW machine\n");
+		return -EINVAL;
+	}
+	ring->name = "bsd2 ring";
+	ring->id = VCS2;
+	ring->write_tail = ring_write_tail;
+	ring->mmio_base = GEN8_BSD2_RING_BASE;
+	ring->flush = gen6_bsd_ring_flush;
+	ring->add_request = gen6_add_request;
+	ring->get_seqno = gen6_ring_get_seqno;
+	ring->set_seqno = ring_set_seqno;
+	ring->irq_enable_mask =
+			GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT;
+	ring->irq_get = gen8_ring_get_irq;
+	ring->irq_put = gen8_ring_put_irq;
+	ring->dispatch_execbuffer =
+			gen8_ring_dispatch_execbuffer;
+	if (i915_semaphore_is_enabled(dev)) {
+		ring->semaphore.sync_to = gen8_ring_sync;
+		ring->semaphore.signal = gen8_xcs_signal;
+		GEN8_RING_SEMAPHORE_INIT;
+	}
+	ring->init = init_ring_common;
+	return intel_init_ring_buffer(dev, ring);
+}
 int intel_init_blt_ring_buffer(struct drm_device *dev)
 {
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[BCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[BCS];
 		ring->irq_enable_mask =
 			GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT;
 		ring->irq_get = gen8_ring_get_irq;
 		ring->irq_put = gen8_ring_put_irq;
 		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.sync_to = gen8_ring_sync;
+			ring->semaphore.signal = gen8_xcs_signal;
+			GEN8_RING_SEMAPHORE_INIT;
+		}
 	} else {
 	ring->irq_enable_mask = GT_BLT_USER_INTERRUPT;
 	ring->irq_get = gen6_ring_get_irq;
 	ring->irq_put = gen6_ring_put_irq;
 	ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.signal = gen6_signal;
+	ring->semaphore.sync_to = gen6_ring_sync;
+	/*
+			 * The current semaphore is only applied on pre-gen8
+			 * platform.  And there is no VCS2 ring on the pre-gen8
+			 * platform. So the semaphore between BCS and VCS2 is
+			 * initialized as INVALID.  Gen8 will initialize the
+			 * sema between BCS and VCS2 later.
+	 */
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_BR;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_BV;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_BVE;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_RBSYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_VBSYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_VEBSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+		}
+	}
-	ring->sync_to = gen6_ring_sync;
-	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_BR;
-	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_BV;
-	ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_BVE;
-	ring->signal_mbox[RCS] = GEN6_RBSYNC;
-	ring->signal_mbox[VCS] = GEN6_VBSYNC;
-	ring->signal_mbox[BCS] = GEN6_NOSYNC;
-	ring->signal_mbox[VECS] = GEN6_VEBSYNC;
 	ring->init = init_ring_common;
 	return intel_init_ring_buffer(dev, ring);
+}
 int intel_init_vebox_ring_buffer(struct drm_device *dev)
 {
-	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[VECS];
+	struct intel_engine_cs *ring = &dev_priv->ring[VECS];
 		ring->irq_enable_mask =
 			GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT;
 		ring->irq_get = gen8_ring_get_irq;
 		ring->irq_put = gen8_ring_put_irq;
 		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.sync_to = gen8_ring_sync;
+			ring->semaphore.signal = gen8_xcs_signal;
+			GEN8_RING_SEMAPHORE_INIT;
+		}
 	} else {
 	ring->irq_enable_mask = PM_VEBOX_USER_INTERRUPT;
 	ring->irq_get = hsw_vebox_get_irq;
 	ring->irq_put = hsw_vebox_put_irq;
 	ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+	ring->semaphore.sync_to = gen6_ring_sync;
+	ring->semaphore.signal = gen6_signal;
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VER;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_VEV;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VEB;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_RVESYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_VVESYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+		}
+	}
-	ring->sync_to = gen6_ring_sync;
-	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VER;
-	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_VEV;
-	ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_VEB;
-	ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->signal_mbox[RCS] = GEN6_RVESYNC;
-	ring->signal_mbox[VCS] = GEN6_VVESYNC;
-	ring->signal_mbox[BCS] = GEN6_BVESYNC;
-	ring->signal_mbox[VECS] = GEN6_NOSYNC;
 	ring->init = init_ring_common;
 	return intel_init_ring_buffer(dev, ring);
+}
 int
-intel_ring_flush_all_caches(struct intel_ring_buffer *ring)
+intel_ring_flush_all_caches(struct intel_engine_cs *ring)
+{
 	int ret;
 	ring->gpu_caches_dirty = false;
 	return 0;
+}
 int
-intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring)
+intel_ring_invalidate_all_caches(struct intel_engine_cs *ring)
+{
 	uint32_t flush_domains;
 	int ret;
 	trace_i915_gem_ring_flush(ring, I915_GEM_GPU_DOMAINS, flush_domains);
 	ring->gpu_caches_dirty = false;
 	return 0;
+}
+void
+intel_stop_ring_buffer(struct intel_engine_cs *ring)
+{
+	int ret;
+	if (!intel_ring_initialized(ring))
+		return;
+	ret = intel_ring_idle(ring);
+	if (ret && !i915_reset_in_progress(&to_i915(ring->dev)->gpu_error))
+		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
+			  ring->name, ret);
+	stop_ring(ring);

Subversion Repositories Kolibri OS

(root)/drivers/video/drm/i915/intel_ringbuffer.c – Rev 4560 → 5060