1 files changed, 121 insertions, 87 deletions

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 9461a238f5d5..40c6aff57256 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -27,29 +27,13 @@
  *
  */
 
+#include <linux/log2.h>
 #include <drm/drmP.h>
 #include "i915_drv.h"
 #include <drm/i915_drm.h>
 #include "i915_trace.h"
 #include "intel_drv.h"
 
-bool
-intel_ring_initialized(struct intel_engine_cs *ring)
-{
-	struct drm_device *dev = ring->dev;
-
-	if (!dev)
-		return false;
-
-	if (i915.enable_execlists) {
-		struct intel_context *dctx = ring->default_context;
-		struct intel_ringbuffer *ringbuf = dctx->engine[ring->id].ringbuf;
-
-		return ringbuf->obj;
-	} else
-		return ring->buffer && ring->buffer->obj;
-}
-
 int __intel_ring_space(int head, int tail, int size)
 {
 	int space = head - tail;
@@ -347,6 +331,7 @@ gen7_render_ring_flush(struct drm_i915_gem_request *req,
 	if (flush_domains) {
 		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
+		flags |= PIPE_CONTROL_DC_FLUSH_ENABLE;
 		flags |= PIPE_CONTROL_FLUSH_ENABLE;
 	}
 	if (invalidate_domains) {
@@ -419,6 +404,7 @@ gen8_render_ring_flush(struct drm_i915_gem_request *req,
 	if (flush_domains) {
 		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
+		flags |= PIPE_CONTROL_DC_FLUSH_ENABLE;
 		flags |= PIPE_CONTROL_FLUSH_ENABLE;
 	}
 	if (invalidate_domains) {
@@ -481,7 +467,7 @@ static void intel_ring_setup_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	u32 mmio = 0;
+	i915_reg_t mmio;
 
 	/* The ring status page addresses are no longer next to the rest of
 	 * the ring registers as of gen7.
@@ -524,7 +510,7 @@ static void intel_ring_setup_status_page(struct intel_engine_cs *ring)
 	 * invalidating the TLB?
 	 */
 	if (INTEL_INFO(dev)->gen >= 6 && INTEL_INFO(dev)->gen < 8) {
-		u32 reg = RING_INSTPM(ring->mmio_base);
+		i915_reg_t reg = RING_INSTPM(ring->mmio_base);
 
 		/* ring should be idle before issuing a sync flush*/
 		WARN_ON((I915_READ_MODE(ring) & MODE_IDLE) == 0);
@@ -733,7 +719,7 @@ static int intel_ring_workarounds_emit(struct drm_i915_gem_request *req)
 
 	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(w->count));
 	for (i = 0; i < w->count; i++) {
-		intel_ring_emit(ring, w->reg[i].addr);
+		intel_ring_emit_reg(ring, w->reg[i].addr);
 		intel_ring_emit(ring, w->reg[i].value);
 	}
 	intel_ring_emit(ring, MI_NOOP);
@@ -766,7 +752,8 @@ static int intel_rcs_ctx_init(struct drm_i915_gem_request *req)
 }
 
 static int wa_add(struct drm_i915_private *dev_priv,
-		  const u32 addr, const u32 mask, const u32 val)
+		  i915_reg_t addr,
+		  const u32 mask, const u32 val)
 {
 	const u32 idx = dev_priv->workarounds.count;
 
@@ -924,17 +911,15 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 	WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3,
 			  GEN9_DISABLE_OCL_OOB_SUPPRESS_LOGIC);
 
-	if ((IS_SKYLAKE(dev) && (INTEL_REVID(dev) == SKL_REVID_A0 ||
-	    INTEL_REVID(dev) == SKL_REVID_B0)) ||
-	    (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0)) {
-		/* WaDisableDgMirrorFixInHalfSliceChicken5:skl,bxt */
+	/* WaDisableDgMirrorFixInHalfSliceChicken5:skl,bxt */
+	if (IS_SKL_REVID(dev, 0, SKL_REVID_B0) ||
+	    IS_BXT_REVID(dev, 0, BXT_REVID_A1))
 		WA_CLR_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN5,
 				  GEN9_DG_MIRROR_FIX_ENABLE);
-	}
 
-	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) ||
-	    (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0)) {
-		/* WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken:skl,bxt */
+	/* WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken:skl,bxt */
+	if (IS_SKL_REVID(dev, 0, SKL_REVID_B0) ||
+	    IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
 		WA_SET_BIT_MASKED(GEN7_COMMON_SLICE_CHICKEN1,
 				  GEN9_RHWO_OPTIMIZATION_DISABLE);
 		/*
@@ -944,12 +929,10 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 		 */
 	}
 
-	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) >= SKL_REVID_C0) ||
-	    IS_BROXTON(dev)) {
-		/* WaEnableYV12BugFixInHalfSliceChicken7:skl,bxt */
+	/* WaEnableYV12BugFixInHalfSliceChicken7:skl,bxt */
+	if (IS_SKL_REVID(dev, SKL_REVID_C0, REVID_FOREVER) || IS_BROXTON(dev))
 		WA_SET_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN7,
 				  GEN9_ENABLE_YV12_BUGFIX);
-	}
 
 	/* Wa4x4STCOptimizationDisable:skl,bxt */
 	/* WaDisablePartialResolveInVc:skl,bxt */
@@ -961,24 +944,22 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 			  GEN9_CCS_TLB_PREFETCH_ENABLE);
 
 	/* WaDisableMaskBasedCammingInRCC:skl,bxt */
-	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) == SKL_REVID_C0) ||
-	    (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0))
+	if (IS_SKL_REVID(dev, SKL_REVID_C0, SKL_REVID_C0) ||
+	    IS_BXT_REVID(dev, 0, BXT_REVID_A1))
 		WA_SET_BIT_MASKED(SLICE_ECO_CHICKEN0,
 				  PIXEL_MASK_CAMMING_DISABLE);
 
 	/* WaForceContextSaveRestoreNonCoherent:skl,bxt */
 	tmp = HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT;
-	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) == SKL_REVID_F0) ||
-	    (IS_BROXTON(dev) && INTEL_REVID(dev) >= BXT_REVID_B0))
+	if (IS_SKL_REVID(dev, SKL_REVID_F0, SKL_REVID_F0) ||
+	    IS_BXT_REVID(dev, BXT_REVID_B0, REVID_FOREVER))
 		tmp |= HDC_FORCE_CSR_NON_COHERENT_OVR_DISABLE;
 	WA_SET_BIT_MASKED(HDC_CHICKEN0, tmp);
 
 	/* WaDisableSamplerPowerBypassForSOPingPong:skl,bxt */
-	if (IS_SKYLAKE(dev) ||
-	    (IS_BROXTON(dev) && INTEL_REVID(dev) <= BXT_REVID_B0)) {
+	if (IS_SKYLAKE(dev) || IS_BXT_REVID(dev, 0, BXT_REVID_B0))
 		WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3,
 				  GEN8_SAMPLER_POWER_BYPASS_DIS);
-	}
 
 	/* WaDisableSTUnitPowerOptimization:skl,bxt */
 	WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN2, GEN8_ST_PO_DISABLE);
@@ -1000,7 +981,7 @@ static int skl_tune_iz_hashing(struct intel_engine_cs *ring)
 		 * Only consider slices where one, and only one, subslice has 7
 		 * EUs
 		 */
-		if (hweight8(dev_priv->info.subslice_7eu[i]) != 1)
+		if (!is_power_of_2(dev_priv->info.subslice_7eu[i]))
 			continue;
 
 		/*
@@ -1038,11 +1019,7 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
-	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
-		/* WaDisableHDCInvalidation:skl */
-		I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
-			   BDW_DISABLE_HDC_INVALIDATION);
-
+	if (IS_SKL_REVID(dev, 0, SKL_REVID_D0)) {
 		/* WaDisableChickenBitTSGBarrierAckForFFSliceCS:skl */
 		I915_WRITE(FF_SLICE_CS_CHICKEN2,
 			   _MASKED_BIT_ENABLE(GEN9_TSG_BARRIER_ACK_DISABLE));
@@ -1051,23 +1028,23 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 	/* GEN8_L3SQCREG4 has a dependency with WA batch so any new changes
 	 * involving this register should also be added to WA batch as required.
 	 */
-	if (INTEL_REVID(dev) <= SKL_REVID_E0)
+	if (IS_SKL_REVID(dev, 0, SKL_REVID_E0))
 		/* WaDisableLSQCROPERFforOCL:skl */
 		I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) |
 			   GEN8_LQSC_RO_PERF_DIS);
 
 	/* WaEnableGapsTsvCreditFix:skl */
-	if (IS_SKYLAKE(dev) && (INTEL_REVID(dev) >= SKL_REVID_C0)) {
+	if (IS_SKL_REVID(dev, SKL_REVID_C0, REVID_FOREVER)) {
 		I915_WRITE(GEN8_GARBCNTL, (I915_READ(GEN8_GARBCNTL) |
 					   GEN9_GAPS_TSV_CREDIT_DISABLE));
 	}
 
 	/* WaDisablePowerCompilerClockGating:skl */
-	if (INTEL_REVID(dev) == SKL_REVID_B0)
+	if (IS_SKL_REVID(dev, SKL_REVID_B0, SKL_REVID_B0))
 		WA_SET_BIT_MASKED(HIZ_CHICKEN,
 				  BDW_HIZ_POWER_COMPILER_CLOCK_GATING_DISABLE);
 
-	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
+	if (IS_SKL_REVID(dev, 0, SKL_REVID_F0)) {
 		/*
 		 *Use Force Non-Coherent whenever executing a 3D context. This
 		 * is a workaround for a possible hang in the unlikely event
@@ -1076,21 +1053,23 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 		/* WaForceEnableNonCoherent:skl */
 		WA_SET_BIT_MASKED(HDC_CHICKEN0,
 				  HDC_FORCE_NON_COHERENT);
+
+		/* WaDisableHDCInvalidation:skl */
+		I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
+			   BDW_DISABLE_HDC_INVALIDATION);
 	}
 
-	if (INTEL_REVID(dev) == SKL_REVID_C0 ||
-	    INTEL_REVID(dev) == SKL_REVID_D0)
-		/* WaBarrierPerformanceFixDisable:skl */
+	/* WaBarrierPerformanceFixDisable:skl */
+	if (IS_SKL_REVID(dev, SKL_REVID_C0, SKL_REVID_D0))
 		WA_SET_BIT_MASKED(HDC_CHICKEN0,
 				  HDC_FENCE_DEST_SLM_DISABLE |
 				  HDC_BARRIER_PERFORMANCE_DISABLE);
 
 	/* WaDisableSbeCacheDispatchPortSharing:skl */
-	if (INTEL_REVID(dev) <= SKL_REVID_F0) {
+	if (IS_SKL_REVID(dev, 0, SKL_REVID_F0))
 		WA_SET_BIT_MASKED(
 			GEN7_HALF_SLICE_CHICKEN1,
 			GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE);
-	}
 
 	return skl_tune_iz_hashing(ring);
 }
@@ -1107,11 +1086,11 @@ static int bxt_init_workarounds(struct intel_engine_cs *ring)
 
 	/* WaStoreMultiplePTEenable:bxt */
 	/* This is a requirement according to Hardware specification */
-	if (INTEL_REVID(dev) == BXT_REVID_A0)
+	if (IS_BXT_REVID(dev, 0, BXT_REVID_A1))
 		I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
 
 	/* WaSetClckGatingDisableMedia:bxt */
-	if (INTEL_REVID(dev) == BXT_REVID_A0) {
+	if (IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
 		I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) &
 					    ~GEN8_DOP_CLOCK_GATE_MEDIA_ENABLE));
 	}
@@ -1121,7 +1100,7 @@ static int bxt_init_workarounds(struct intel_engine_cs *ring)
 			  STALL_DOP_GATING_DISABLE);
 
 	/* WaDisableSbeCacheDispatchPortSharing:bxt */
-	if (INTEL_REVID(dev) <= BXT_REVID_B0) {
+	if (IS_BXT_REVID(dev, 0, BXT_REVID_B0)) {
 		WA_SET_BIT_MASKED(
 			GEN7_HALF_SLICE_CHICKEN1,
 			GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE);
@@ -1319,11 +1298,13 @@ static int gen6_signal(struct drm_i915_gem_request *signaller_req,
 		return ret;
 
 	for_each_ring(useless, dev_priv, i) {
-		u32 mbox_reg = signaller->semaphore.mbox.signal[i];
-		if (mbox_reg != GEN6_NOSYNC) {
+		i915_reg_t mbox_reg = signaller->semaphore.mbox.signal[i];
+
+		if (i915_mmio_reg_valid(mbox_reg)) {
 			u32 seqno = i915_gem_request_get_seqno(signaller_req);
+
 			intel_ring_emit(signaller, MI_LOAD_REGISTER_IMM(1));
-			intel_ring_emit(signaller, mbox_reg);
+			intel_ring_emit_reg(signaller, mbox_reg);
 			intel_ring_emit(signaller, seqno);
 		}
 	}
@@ -2004,11 +1985,35 @@ static int init_phys_status_page(struct intel_engine_cs *ring)
 
 void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
 {
-	iounmap(ringbuf->virtual_start);
+	if (HAS_LLC(ringbuf->obj->base.dev) && !ringbuf->obj->stolen)
+		vunmap(ringbuf->virtual_start);
+	else
+		iounmap(ringbuf->virtual_start);
 	ringbuf->virtual_start = NULL;
 	i915_gem_object_ggtt_unpin(ringbuf->obj);
 }
 
+static u32 *vmap_obj(struct drm_i915_gem_object *obj)
+{
+	struct sg_page_iter sg_iter;
+	struct page **pages;
+	void *addr;
+	int i;
+
+	pages = drm_malloc_ab(obj->base.size >> PAGE_SHIFT, sizeof(*pages));
+	if (pages == NULL)
+		return NULL;
+
+	i = 0;
+	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0)
+		pages[i++] = sg_page_iter_page(&sg_iter);
+
+	addr = vmap(pages, i, 0, PAGE_KERNEL);
+	drm_free_large(pages);
+
+	return addr;
+}
+
 int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 				     struct intel_ringbuffer *ringbuf)
 {
@@ -2016,21 +2021,39 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 	struct drm_i915_gem_object *obj = ringbuf->obj;
 	int ret;
 
-	ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, PIN_MAPPABLE);
-	if (ret)
-		return ret;
+	if (HAS_LLC(dev_priv) && !obj->stolen) {
+		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, 0);
+		if (ret)
+			return ret;
 
-	ret = i915_gem_object_set_to_gtt_domain(obj, true);
-	if (ret) {
-		i915_gem_object_ggtt_unpin(obj);
-		return ret;
-	}
+		ret = i915_gem_object_set_to_cpu_domain(obj, true);
+		if (ret) {
+			i915_gem_object_ggtt_unpin(obj);
+			return ret;
+		}
 
-	ringbuf->virtual_start = ioremap_wc(dev_priv->gtt.mappable_base +
-			i915_gem_obj_ggtt_offset(obj), ringbuf->size);
-	if (ringbuf->virtual_start == NULL) {
-		i915_gem_object_ggtt_unpin(obj);
-		return -EINVAL;
+		ringbuf->virtual_start = vmap_obj(obj);
+		if (ringbuf->virtual_start == NULL) {
+			i915_gem_object_ggtt_unpin(obj);
+			return -ENOMEM;
+		}
+	} else {
+		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, PIN_MAPPABLE);
+		if (ret)
+			return ret;
+
+		ret = i915_gem_object_set_to_gtt_domain(obj, true);
+		if (ret) {
+			i915_gem_object_ggtt_unpin(obj);
+			return ret;
+		}
+
+		ringbuf->virtual_start = ioremap_wc(dev_priv->gtt.mappable_base +
+						    i915_gem_obj_ggtt_offset(obj), ringbuf->size);
+		if (ringbuf->virtual_start == NULL) {
+			i915_gem_object_ggtt_unpin(obj);
+			return -EINVAL;
+		}
 	}
 
 	return 0;
@@ -2070,10 +2093,14 @@ intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size)
 	int ret;
 
 	ring = kzalloc(sizeof(*ring), GFP_KERNEL);
-	if (ring == NULL)
+	if (ring == NULL) {
+		DRM_DEBUG_DRIVER("Failed to allocate ringbuffer %s\n",
+				 engine->name);
 		return ERR_PTR(-ENOMEM);
+	}
 
 	ring->ring = engine;
+	list_add(&ring->link, &engine->buffers);
 
 	ring->size = size;
 	/* Workaround an erratum on the i830 which causes a hang if
@@ -2089,8 +2116,9 @@ intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size)
 
 	ret = intel_alloc_ringbuffer_obj(engine->dev, ring);
 	if (ret) {
-		DRM_ERROR("Failed to allocate ringbuffer %s: %d\n",
-			  engine->name, ret);
+		DRM_DEBUG_DRIVER("Failed to allocate ringbuffer %s: %d\n",
+				 engine->name, ret);
+		list_del(&ring->link);
 		kfree(ring);
 		return ERR_PTR(ret);
 	}
@@ -2102,6 +2130,7 @@ void
 intel_ringbuffer_free(struct intel_ringbuffer *ring)
 {
 	intel_destroy_ringbuffer_obj(ring);
+	list_del(&ring->link);
 	kfree(ring);
 }
 
@@ -2117,14 +2146,17 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
 	INIT_LIST_HEAD(&ring->execlist_queue);
+	INIT_LIST_HEAD(&ring->buffers);
 	i915_gem_batch_pool_init(dev, &ring->batch_pool);
 	memset(ring->semaphore.sync_seqno, 0, sizeof(ring->semaphore.sync_seqno));
 
 	init_waitqueue_head(&ring->irq_queue);
 
 	ringbuf = intel_engine_create_ringbuffer(ring, 32 * PAGE_SIZE);
-	if (IS_ERR(ringbuf))
-		return PTR_ERR(ringbuf);
+	if (IS_ERR(ringbuf)) {
+		ret = PTR_ERR(ringbuf);
+		goto error;
+	}
 	ring->buffer = ringbuf;
 
 	if (I915_NEED_GFX_HWS(dev)) {
@@ -2153,8 +2185,7 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	return 0;
 
 error:
-	intel_ringbuffer_free(ringbuf);
-	ring->buffer = NULL;
+	intel_cleanup_ring_buffer(ring);
 	return ret;
 }
 
@@ -2167,12 +2198,14 @@ void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 
 	dev_priv = to_i915(ring->dev);
 
-	intel_stop_ring_buffer(ring);
-	WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
+	if (ring->buffer) {
+		intel_stop_ring_buffer(ring);
+		WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
 
-	intel_unpin_ringbuffer_obj(ring->buffer);
-	intel_ringbuffer_free(ring->buffer);
-	ring->buffer = NULL;
+		intel_unpin_ringbuffer_obj(ring->buffer);
+		intel_ringbuffer_free(ring->buffer);
+		ring->buffer = NULL;
+	}
 
 	if (ring->cleanup)
 		ring->cleanup(ring);
@@ -2181,6 +2214,7 @@ void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 
 	i915_cmd_parser_fini_ring(ring);
 	i915_gem_batch_pool_fini(&ring->batch_pool);
+	ring->dev = NULL;
 }
 
 static int ring_wait_for_space(struct intel_engine_cs *ring, int n)