Merge tag 'drm-intel-next-fixes-2017-04-27' of git://anongit.freedesktop.org/git/drm-intel into drm-next

drm/i915 and gvt fixes for drm-next/v4.12

* tag 'drm-intel-next-fixes-2017-04-27' of git://anongit.freedesktop.org/git/drm-intel:
  drm/i915: Confirm the request is still active before adding it to the await
  drm/i915: Avoid busy-spinning on VLV_GLTC_PW_STATUS mmio
  drm/i915/selftests: Allocate inode/file dynamically
  drm/i915: Fix system hang with EI UP masked on Haswell
  drm/i915: checking for NULL instead of IS_ERR() in mock selftests
  drm/i915: Perform link quality check unconditionally during long pulse
  drm/i915: Fix use after free in lpe_audio_platdev_destroy()
  drm/i915: Use the right mapping_gfp_mask for final shmem allocation
  drm/i915: Make legacy cursor updates more unsynced
  drm/i915: Apply a cond_resched() to the saturated signaler
  drm/i915: Park the signaler before sleeping
  drm/i915/gvt: fix a bounds check in ring_id_to_context_switch_event()
  drm/i915/gvt: Fix PTE write flush for taking runtime pm properly
  drm/i915/gvt: remove some debug messages in scheduler timer handler
  drm/i915/gvt: add mmio init for virtual display
  drm/i915/gvt: use directly assignment for structure copying
  drm/i915/gvt: remove redundant ring id check which cause significant CPU misprediction
  drm/i915/gvt: remove redundant platform check for mocs load/restore
  drm/i915/gvt: Align render mmio list to cacheline
  drm/i915/gvt: cleanup some too chatty scheduler message

7 files changed, 42 insertions, 40 deletions

diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index 94f2e70..41b2c3a 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -616,9 +616,6 @@ static inline u32 get_opcode(u32 cmd, int ring_id)
 {
 	struct decode_info *d_info;
 
-	if (ring_id >= I915_NUM_ENGINES)
-		return INVALID_OP;
-
 	d_info = ring_decode_info[ring_id][CMD_TYPE(cmd)];
 	if (d_info == NULL)
 		return INVALID_OP;
@@ -661,9 +658,6 @@ static inline void print_opcode(u32 cmd, int ring_id)
 	struct decode_info *d_info;
 	int i;
 
-	if (ring_id >= I915_NUM_ENGINES)
-		return;
-
 	d_info = ring_decode_info[ring_id][CMD_TYPE(cmd)];
 	if (d_info == NULL)
 		return;
@@ -2483,7 +2477,7 @@ static int cmd_parser_exec(struct parser_exec_state *s)
 
 	t1 = get_cycles();
 
-	memcpy(&s_before_advance_custom, s, sizeof(struct parser_exec_state));
+	s_before_advance_custom = *s;
 
 	if (info->handler) {
 		ret = info->handler(s);
diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c
index 4cf2b29..e0261fc 100644
--- a/drivers/gpu/drm/i915/gvt/display.c
+++ b/drivers/gpu/drm/i915/gvt/display.c
@@ -189,17 +189,44 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu)
 	}
 
 	if (intel_vgpu_has_monitor_on_port(vgpu, PORT_B)) {
-		vgpu_vreg(vgpu, SDEISR) |= SDE_PORTB_HOTPLUG_CPT;
 		vgpu_vreg(vgpu, SFUSE_STRAP) |= SFUSE_STRAP_DDIB_DETECTED;
+		vgpu_vreg(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) &=
+			~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
+			TRANS_DDI_PORT_MASK);
+		vgpu_vreg(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |=
+			(TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST |
+			(PORT_B << TRANS_DDI_PORT_SHIFT) |
+			TRANS_DDI_FUNC_ENABLE);
+		vgpu_vreg(vgpu, DDI_BUF_CTL(PORT_B)) |= DDI_BUF_CTL_ENABLE;
+		vgpu_vreg(vgpu, DDI_BUF_CTL(PORT_B)) &= ~DDI_BUF_IS_IDLE;
+		vgpu_vreg(vgpu, SDEISR) |= SDE_PORTB_HOTPLUG_CPT;
 	}
 
 	if (intel_vgpu_has_monitor_on_port(vgpu, PORT_C)) {
 		vgpu_vreg(vgpu, SDEISR) |= SDE_PORTC_HOTPLUG_CPT;
+		vgpu_vreg(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) &=
+			~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
+			TRANS_DDI_PORT_MASK);
+		vgpu_vreg(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |=
+			(TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST |
+			(PORT_C << TRANS_DDI_PORT_SHIFT) |
+			TRANS_DDI_FUNC_ENABLE);
+		vgpu_vreg(vgpu, DDI_BUF_CTL(PORT_C)) |= DDI_BUF_CTL_ENABLE;
+		vgpu_vreg(vgpu, DDI_BUF_CTL(PORT_C)) &= ~DDI_BUF_IS_IDLE;
 		vgpu_vreg(vgpu, SFUSE_STRAP) |= SFUSE_STRAP_DDIC_DETECTED;
 	}
 
 	if (intel_vgpu_has_monitor_on_port(vgpu, PORT_D)) {
 		vgpu_vreg(vgpu, SDEISR) |= SDE_PORTD_HOTPLUG_CPT;
+		vgpu_vreg(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) &=
+			~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
+			TRANS_DDI_PORT_MASK);
+		vgpu_vreg(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |=
+			(TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST |
+			(PORT_D << TRANS_DDI_PORT_SHIFT) |
+			TRANS_DDI_FUNC_ENABLE);
+		vgpu_vreg(vgpu, DDI_BUF_CTL(PORT_D)) |= DDI_BUF_CTL_ENABLE;
+		vgpu_vreg(vgpu, DDI_BUF_CTL(PORT_D)) &= ~DDI_BUF_IS_IDLE;
 		vgpu_vreg(vgpu, SFUSE_STRAP) |= SFUSE_STRAP_DDID_DETECTED;
 	}
 
diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c
index 536bde8..dca989e 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.c
+++ b/drivers/gpu/drm/i915/gvt/execlist.c
@@ -56,8 +56,8 @@ static int context_switch_events[] = {
 
 static int ring_id_to_context_switch_event(int ring_id)
 {
-	if (WARN_ON(ring_id < RCS && ring_id >
-				ARRAY_SIZE(context_switch_events)))
+	if (WARN_ON(ring_id < RCS ||
+		    ring_id >= ARRAY_SIZE(context_switch_events)))
 		return -EINVAL;
 
 	return context_switch_events[ring_id];
@@ -687,9 +687,7 @@ static int submit_context(struct intel_vgpu *vgpu, int ring_id,
 	}
 
 	if (emulate_schedule_in)
-		memcpy(&workload->elsp_dwords,
-				&vgpu->execlist[ring_id].elsp_dwords,
-				sizeof(workload->elsp_dwords));
+		workload->elsp_dwords = vgpu->execlist[ring_id].elsp_dwords;
 
 	gvt_dbg_el("workload %p ring id %d head %x tail %x start %x ctl %x\n",
 			workload, ring_id, head, tail, start, ctl);
diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
index 6da4e44..c6f0077 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.c
+++ b/drivers/gpu/drm/i915/gvt/gtt.c
@@ -2294,12 +2294,15 @@ void intel_gvt_clean_gtt(struct intel_gvt *gvt)
 void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu)
 {
 	struct intel_gvt *gvt = vgpu->gvt;
+	struct drm_i915_private *dev_priv = gvt->dev_priv;
 	struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops;
 	u32 index;
 	u32 offset;
 	u32 num_entries;
 	struct intel_gvt_gtt_entry e;
 
+	intel_runtime_pm_get(dev_priv);
+
 	memset(&e, 0, sizeof(struct intel_gvt_gtt_entry));
 	e.type = GTT_TYPE_GGTT_PTE;
 	ops->set_pfn(&e, gvt->gtt.scratch_ggtt_mfn);
@@ -2314,6 +2317,8 @@ void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu)
 	num_entries = vgpu_hidden_sz(vgpu) >> PAGE_SHIFT;
 	for (offset = 0; offset < num_entries; offset++)
 		ops->set_entry(NULL, &e, index + offset, false, 0, vgpu);
+
+	intel_runtime_pm_put(dev_priv);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/gvt/render.c b/drivers/gpu/drm/i915/gvt/render.c
index a7b665e..c6e7972 100644
--- a/drivers/gpu/drm/i915/gvt/render.c
+++ b/drivers/gpu/drm/i915/gvt/render.c
@@ -44,7 +44,7 @@ struct render_mmio {
 	u32 value;
 };
 
-static struct render_mmio gen8_render_mmio_list[] = {
+static struct render_mmio gen8_render_mmio_list[] __cacheline_aligned = {
 	{RCS, _MMIO(0x229c), 0xffff, false},
 	{RCS, _MMIO(0x2248), 0x0, false},
 	{RCS, _MMIO(0x2098), 0x0, false},
@@ -75,7 +75,7 @@ static struct render_mmio gen8_render_mmio_list[] = {
 	{BCS, _MMIO(0x22028), 0x0, false},
 };
 
-static struct render_mmio gen9_render_mmio_list[] = {
+static struct render_mmio gen9_render_mmio_list[] __cacheline_aligned = {
 	{RCS, _MMIO(0x229c), 0xffff, false},
 	{RCS, _MMIO(0x2248), 0x0, false},
 	{RCS, _MMIO(0x2098), 0x0, false},
@@ -204,9 +204,6 @@ static void load_mocs(struct intel_vgpu *vgpu, int ring_id)
 	if (WARN_ON(ring_id >= ARRAY_SIZE(regs)))
 		return;
 
-	if (!(IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)))
-		return;
-
 	offset.reg = regs[ring_id];
 	for (i = 0; i < 64; i++) {
 		gen9_render_mocs[ring_id][i] = I915_READ(offset);
@@ -242,9 +239,6 @@ static void restore_mocs(struct intel_vgpu *vgpu, int ring_id)
 	if (WARN_ON(ring_id >= ARRAY_SIZE(regs)))
 		return;
 
-	if (!(IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)))
-		return;
-
 	offset.reg = regs[ring_id];
 	for (i = 0; i < 64; i++) {
 		vgpu_vreg(vgpu, offset) = I915_READ(offset);
diff --git a/drivers/gpu/drm/i915/gvt/sched_policy.c b/drivers/gpu/drm/i915/gvt/sched_policy.c
index f849591..79ba4b34 100644
--- a/drivers/gpu/drm/i915/gvt/sched_policy.c
+++ b/drivers/gpu/drm/i915/gvt/sched_policy.c
@@ -133,9 +133,6 @@ static void try_to_schedule_next_vgpu(struct intel_gvt *gvt)
 	if (!scheduler->next_vgpu)
 		return;
 
-	gvt_dbg_sched("try to schedule next vgpu %d\n",
-			scheduler->next_vgpu->id);
-
 	/*
 	 * after the flag is set, workload dispatch thread will
 	 * stop dispatching workload for current vgpu
@@ -144,15 +141,10 @@ static void try_to_schedule_next_vgpu(struct intel_gvt *gvt)
 
 	/* still have uncompleted workload? */
 	for_each_engine(engine, gvt->dev_priv, i) {
-		if (scheduler->current_workload[i]) {
-			gvt_dbg_sched("still have running workload\n");
+		if (scheduler->current_workload[i])
 			return;
-		}
 	}
 
-	gvt_dbg_sched("switch to next vgpu %d\n",
-			scheduler->next_vgpu->id);
-
 	cur_time = ktime_get();
 	if (scheduler->current_vgpu) {
 		vgpu_data = scheduler->current_vgpu->sched_data;
@@ -224,17 +216,12 @@ static void tbs_sched_func(struct gvt_sched_data *sched_data)
 		list_del_init(&vgpu_data->lru_list);
 		list_add_tail(&vgpu_data->lru_list,
 				&sched_data->lru_runq_head);
-
-		gvt_dbg_sched("pick next vgpu %d\n", vgpu->id);
 	} else {
 		scheduler->next_vgpu = gvt->idle_vgpu;
 	}
 out:
-	if (scheduler->next_vgpu) {
-		gvt_dbg_sched("try to schedule next vgpu %d\n",
-				scheduler->next_vgpu->id);
+	if (scheduler->next_vgpu)
 		try_to_schedule_next_vgpu(gvt);
-	}
 }
 
 void intel_gvt_schedule(struct intel_gvt *gvt)
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index a77db23..bada32b 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -279,11 +279,8 @@ static struct intel_vgpu_workload *pick_next_workload(
 		goto out;
 	}
 
-	if (list_empty(workload_q_head(scheduler->current_vgpu, ring_id))) {
-		gvt_dbg_sched("ring id %d stop - no available workload\n",
-				ring_id);
+	if (list_empty(workload_q_head(scheduler->current_vgpu, ring_id)))
 		goto out;
-	}
 
 	/*
 	 * still have current workload, maybe the workload disptacher