Merge airlied/drm-next into drm-misc-next

Backmerging to pick up a fix from drm-misc-next-fixes.

Signed-off-by: Sean Paul <seanpaul@chromium.org>

1 files changed, 100 insertions, 8 deletions

diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index 9b92b4e25a20..638abe84857c 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -52,6 +52,77 @@ static void set_context_pdp_root_pointer(
 		pdp_pair[i].val = pdp[7 - i];
 }
 
+static void update_shadow_pdps(struct intel_vgpu_workload *workload)
+{
+	struct intel_vgpu *vgpu = workload->vgpu;
+	int ring_id = workload->ring_id;
+	struct i915_gem_context *shadow_ctx = vgpu->submission.shadow_ctx;
+	struct drm_i915_gem_object *ctx_obj =
+		shadow_ctx->engine[ring_id].state->obj;
+	struct execlist_ring_context *shadow_ring_context;
+	struct page *page;
+
+	if (WARN_ON(!workload->shadow_mm))
+		return;
+
+	if (WARN_ON(!atomic_read(&workload->shadow_mm->pincount)))
+		return;
+
+	page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
+	shadow_ring_context = kmap(page);
+	set_context_pdp_root_pointer(shadow_ring_context,
+			(void *)workload->shadow_mm->ppgtt_mm.shadow_pdps);
+	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;
@@ -98,11 +169,18 @@ 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)
+#define COPY_REG_MASKED(name) {\
+		intel_gvt_hypervisor_read_gpa(vgpu, workload->ring_context_gpa \
+					      + RING_CTX_OFF(name.val),\
+					      &shadow_ring_context->name.val, 4);\
+		shadow_ring_context->name.val |= 0xffff << 16;\
+	}
 
-	COPY_REG(ctx_ctrl);
+	COPY_REG_MASKED(ctx_ctrl);
 	COPY_REG(ctx_timestamp);
 
 	if (ring_id == RCS) {
@@ -111,9 +189,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload)
 		COPY_REG(rcs_indirect_ctx_offset);
 	}
 #undef COPY_REG
-
-	set_context_pdp_root_pointer(shadow_ring_context,
-				     (void *)workload->shadow_mm->ppgtt_mm.shadow_pdps);
+#undef COPY_REG_MASKED
 
 	intel_gvt_hypervisor_read_gpa(vgpu,
 			workload->ring_context_gpa +
@@ -122,6 +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;
 }
@@ -381,6 +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)
@@ -509,6 +597,8 @@ static int prepare_workload(struct intel_vgpu_workload *workload)
 		return ret;
 	}
 
+	update_shadow_pdps(workload);
+
 	ret = intel_vgpu_sync_oos_pages(workload->vgpu);
 	if (ret) {
 		gvt_vgpu_err("fail to vgpu sync oos pages\n");
@@ -1049,10 +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;