Backmerge tag 'v4.16-rc7' into drm-next

Linux 4.16-rc7

This was requested by Daniel, and things were getting
a bit hard to reconcile, most of the conflicts were
trivial though.

6 files changed, 133 insertions, 7 deletions

diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index c8454ac43fae..db6b94dda5df 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -471,6 +471,7 @@ struct parser_exec_state {
 	 * used when ret from 2nd level batch buffer
 	 */
 	int saved_buf_addr_type;
+	bool is_ctx_wa;
 
 	struct cmd_info *info;
 
@@ -1715,6 +1716,11 @@ static int perform_bb_shadow(struct parser_exec_state *s)
 	bb->accessing = true;
 	bb->bb_start_cmd_va = s->ip_va;
 
+	if ((s->buf_type == BATCH_BUFFER_INSTRUCTION) && (!s->is_ctx_wa))
+		bb->bb_offset = s->ip_va - s->rb_va;
+	else
+		bb->bb_offset = 0;
+
 	/*
 	 * ip_va saves the virtual address of the shadow batch buffer, while
 	 * ip_gma saves the graphics address of the original batch buffer.
@@ -2571,6 +2577,7 @@ static int scan_workload(struct intel_vgpu_workload *workload)
 	s.ring_tail = gma_tail;
 	s.rb_va = workload->shadow_ring_buffer_va;
 	s.workload = workload;
+	s.is_ctx_wa = false;
 
 	if ((bypass_scan_mask & (1 << workload->ring_id)) ||
 		gma_head == gma_tail)
@@ -2624,6 +2631,7 @@ static int scan_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 	s.ring_tail = gma_tail;
 	s.rb_va = wa_ctx->indirect_ctx.shadow_va;
 	s.workload = workload;
+	s.is_ctx_wa = true;
 
 	if (!intel_gvt_ggtt_validate_range(s.vgpu, s.ring_start, s.ring_size)) {
 		ret = -EINVAL;
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
index 520fe3d0a882..c16a492449d7 100644
--- a/drivers/gpu/drm/i915/gvt/kvmgt.c
+++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -750,6 +750,25 @@ static ssize_t intel_vgpu_rw(struct mdev_device *mdev, char *buf,
 	return ret == 0 ? count : ret;
 }
 
+static bool gtt_entry(struct mdev_device *mdev, loff_t *ppos)
+{
+	struct intel_vgpu *vgpu = mdev_get_drvdata(mdev);
+	unsigned int index = VFIO_PCI_OFFSET_TO_INDEX(*ppos);
+	struct intel_gvt *gvt = vgpu->gvt;
+	int offset;
+
+	/* Only allow MMIO GGTT entry access */
+	if (index != PCI_BASE_ADDRESS_0)
+		return false;
+
+	offset = (u64)(*ppos & VFIO_PCI_OFFSET_MASK) -
+		intel_vgpu_get_bar_gpa(vgpu, PCI_BASE_ADDRESS_0);
+
+	return (offset >= gvt->device_info.gtt_start_offset &&
+		offset < gvt->device_info.gtt_start_offset + gvt_ggtt_sz(gvt)) ?
+			true : false;
+}
+
 static ssize_t intel_vgpu_read(struct mdev_device *mdev, char __user *buf,
 			size_t count, loff_t *ppos)
 {
@@ -759,7 +778,21 @@ static ssize_t intel_vgpu_read(struct mdev_device *mdev, char __user *buf,
 	while (count) {
 		size_t filled;
 
-		if (count >= 4 && !(*ppos % 4)) {
+		/* Only support GGTT entry 8 bytes read */
+		if (count >= 8 && !(*ppos % 8) &&
+			gtt_entry(mdev, ppos)) {
+			u64 val;
+
+			ret = intel_vgpu_rw(mdev, (char *)&val, sizeof(val),
+					ppos, false);
+			if (ret <= 0)
+				goto read_err;
+
+			if (copy_to_user(buf, &val, sizeof(val)))
+				goto read_err;
+
+			filled = 8;
+		} else if (count >= 4 && !(*ppos % 4)) {
 			u32 val;
 
 			ret = intel_vgpu_rw(mdev, (char *)&val, sizeof(val),
@@ -819,7 +852,21 @@ static ssize_t intel_vgpu_write(struct mdev_device *mdev,
 	while (count) {
 		size_t filled;
 
-		if (count >= 4 && !(*ppos % 4)) {
+		/* Only support GGTT entry 8 bytes write */
+		if (count >= 8 && !(*ppos % 8) &&
+			gtt_entry(mdev, ppos)) {
+			u64 val;
+
+			if (copy_from_user(&val, buf, sizeof(val)))
+				goto write_err;
+
+			ret = intel_vgpu_rw(mdev, (char *)&val, sizeof(val),
+					ppos, true);
+			if (ret <= 0)
+				goto write_err;
+
+			filled = 8;
+		} else if (count >= 4 && !(*ppos % 4)) {
 			u32 val;
 
 			if (copy_from_user(&val, buf, sizeof(val)))
diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c
index 74a9c7b5516e..a5bac83d53a9 100644
--- a/drivers/gpu/drm/i915/gvt/mmio_context.c
+++ b/drivers/gpu/drm/i915/gvt/mmio_context.c
@@ -120,6 +120,7 @@ static struct engine_mmio gen9_engine_mmio_list[] __cacheline_aligned = {
 	{RCS, HALF_SLICE_CHICKEN3, 0xffff, true}, /* 0xe184 */
 	{RCS, GEN9_HALF_SLICE_CHICKEN5, 0xffff, true}, /* 0xe188 */
 	{RCS, GEN9_HALF_SLICE_CHICKEN7, 0xffff, true}, /* 0xe194 */
+	{RCS, GEN8_ROW_CHICKEN, 0xffff, true}, /* 0xe4f0 */
 	{RCS, TRVATTL3PTRDW(0), 0, false}, /* 0x4de0 */
 	{RCS, TRVATTL3PTRDW(1), 0, false}, /* 0x4de4 */
 	{RCS, TRNULLDETCT, 0, false}, /* 0x4de8 */
@@ -557,9 +558,11 @@ void intel_gvt_switch_mmio(struct intel_vgpu *pre,
 	 * performace for batch mmio read/write, so we need
 	 * handle forcewake mannually.
 	 */
+	intel_runtime_pm_get(dev_priv);
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 	switch_mmio(pre, next, ring_id);
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_runtime_pm_put(dev_priv);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index a55b4975c154..638abe84857c 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -75,6 +75,54 @@ static void update_shadow_pdps(struct intel_vgpu_workload *workload)
 	kunmap(page);
 }
 
+/*
+ * when populating shadow ctx from guest, we should not overrride oa related
+ * registers, so that they will not be overlapped by guest oa configs. Thus
+ * made it possible to capture oa data from host for both host and guests.
+ */
+static void sr_oa_regs(struct intel_vgpu_workload *workload,
+		u32 *reg_state, bool save)
+{
+	struct drm_i915_private *dev_priv = workload->vgpu->gvt->dev_priv;
+	u32 ctx_oactxctrl = dev_priv->perf.oa.ctx_oactxctrl_offset;
+	u32 ctx_flexeu0 = dev_priv->perf.oa.ctx_flexeu0_offset;
+	int i = 0;
+	u32 flex_mmio[] = {
+		i915_mmio_reg_offset(EU_PERF_CNTL0),
+		i915_mmio_reg_offset(EU_PERF_CNTL1),
+		i915_mmio_reg_offset(EU_PERF_CNTL2),
+		i915_mmio_reg_offset(EU_PERF_CNTL3),
+		i915_mmio_reg_offset(EU_PERF_CNTL4),
+		i915_mmio_reg_offset(EU_PERF_CNTL5),
+		i915_mmio_reg_offset(EU_PERF_CNTL6),
+	};
+
+	if (!workload || !reg_state || workload->ring_id != RCS)
+		return;
+
+	if (save) {
+		workload->oactxctrl = reg_state[ctx_oactxctrl + 1];
+
+		for (i = 0; i < ARRAY_SIZE(workload->flex_mmio); i++) {
+			u32 state_offset = ctx_flexeu0 + i * 2;
+
+			workload->flex_mmio[i] = reg_state[state_offset + 1];
+		}
+	} else {
+		reg_state[ctx_oactxctrl] =
+			i915_mmio_reg_offset(GEN8_OACTXCONTROL);
+		reg_state[ctx_oactxctrl + 1] = workload->oactxctrl;
+
+		for (i = 0; i < ARRAY_SIZE(workload->flex_mmio); i++) {
+			u32 state_offset = ctx_flexeu0 + i * 2;
+			u32 mmio = flex_mmio[i];
+
+			reg_state[state_offset] = mmio;
+			reg_state[state_offset + 1] = workload->flex_mmio[i];
+		}
+	}
+}
+
 static int populate_shadow_context(struct intel_vgpu_workload *workload)
 {
 	struct intel_vgpu *vgpu = workload->vgpu;
@@ -121,6 +169,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload)
 	page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
 	shadow_ring_context = kmap(page);
 
+	sr_oa_regs(workload, (u32 *)shadow_ring_context, true);
 #define COPY_REG(name) \
 	intel_gvt_hypervisor_read_gpa(vgpu, workload->ring_context_gpa \
 		+ RING_CTX_OFF(name.val), &shadow_ring_context->name.val, 4)
@@ -149,6 +198,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload)
 			sizeof(*shadow_ring_context),
 			I915_GTT_PAGE_SIZE - sizeof(*shadow_ring_context));
 
+	sr_oa_regs(workload, (u32 *)shadow_ring_context, false);
 	kunmap(page);
 	return 0;
 }
@@ -408,6 +458,17 @@ static int prepare_shadow_batch_buffer(struct intel_vgpu_workload *workload)
 			goto err;
 		}
 
+		/* For privilge batch buffer and not wa_ctx, the bb_start_cmd_va
+		 * is only updated into ring_scan_buffer, not real ring address
+		 * allocated in later copy_workload_to_ring_buffer. pls be noted
+		 * shadow_ring_buffer_va is now pointed to real ring buffer va
+		 * in copy_workload_to_ring_buffer.
+		 */
+
+		if (bb->bb_offset)
+			bb->bb_start_cmd_va = workload->shadow_ring_buffer_va
+				+ bb->bb_offset;
+
 		/* relocate shadow batch buffer */
 		bb->bb_start_cmd_va[1] = i915_ggtt_offset(bb->vma);
 		if (gmadr_bytes == 8)
@@ -1078,10 +1139,12 @@ int intel_vgpu_setup_submission(struct intel_vgpu *vgpu)
 
 	bitmap_zero(s->shadow_ctx_desc_updated, I915_NUM_ENGINES);
 
-	s->workloads = kmem_cache_create("gvt-g_vgpu_workload",
-			sizeof(struct intel_vgpu_workload), 0,
-			SLAB_HWCACHE_ALIGN,
-			NULL);
+	s->workloads = kmem_cache_create_usercopy("gvt-g_vgpu_workload",
+						  sizeof(struct intel_vgpu_workload), 0,
+						  SLAB_HWCACHE_ALIGN,
+						  offsetof(struct intel_vgpu_workload, rb_tail),
+						  sizeof_field(struct intel_vgpu_workload, rb_tail),
+						  NULL);
 
 	if (!s->workloads) {
 		ret = -ENOMEM;
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h
index bab4097aa6d7..486ed57a4ad1 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.h
+++ b/drivers/gpu/drm/i915/gvt/scheduler.h
@@ -110,6 +110,10 @@ struct intel_vgpu_workload {
 	/* shadow batch buffer */
 	struct list_head shadow_bb;
 	struct intel_shadow_wa_ctx wa_ctx;
+
+	/* oa registers */
+	u32 oactxctrl;
+	u32 flex_mmio[7];
 };
 
 struct intel_vgpu_shadow_bb {
@@ -120,6 +124,7 @@ struct intel_vgpu_shadow_bb {
 	u32 *bb_start_cmd_va;
 	unsigned int clflush;
 	bool accessing;
+	unsigned long bb_offset;
 };
 
 #define workload_q_head(vgpu, ring_id) \
diff --git a/drivers/gpu/drm/i915/gvt/trace.h b/drivers/gpu/drm/i915/gvt/trace.h
index fc7831a62121..82093f1e8612 100644
--- a/drivers/gpu/drm/i915/gvt/trace.h
+++ b/drivers/gpu/drm/i915/gvt/trace.h
@@ -333,7 +333,7 @@ TRACE_EVENT(render_mmio,
 	TP_PROTO(int old_id, int new_id, char *action, unsigned int reg,
 		 unsigned int old_val, unsigned int new_val),
 
-	TP_ARGS(old_id, new_id, action, reg, new_val, old_val),
+	TP_ARGS(old_id, new_id, action, reg, old_val, new_val),
 
 	TP_STRUCT__entry(
 		__field(int, old_id)