From fa3dd623e559e8e7004179f9594b090318df0d05 Mon Sep 17 00:00:00 2001
From: Min He <min.he@intel.com>
Date: Fri, 2 Mar 2018 10:00:25 +0800
Subject: [PATCH] drm/i915/gvt: keep oa config in shadow ctx
Git-commit: fa3dd623e559e8e7004179f9594b090318df0d05
Patch-mainline: v4.16-rc6
References: FATE#322643 bsc#1055900

When populating shadow ctx from guest, we should handle oa related
registers in hw ctx, so that they will not be overlapped by guest oa
configs. This patch made it possible to capture oa data from host for
both host and guests.

Signed-off-by: Min He <min.he@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Acked-by: Takashi Iwai <tiwai@suse.de>

---
 drivers/gpu/drm/i915/gvt/scheduler.c |   50 +++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/gvt/scheduler.h |    4 ++
 2 files changed, 54 insertions(+)

--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -52,6 +52,54 @@ static void set_context_pdp_root_pointer
 		pdp_pair[i].val = pdp[7 - i];
 }
 
+/*
+ * 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;
@@ -98,6 +146,7 @@ static int populate_shadow_context(struc
 	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)
@@ -122,6 +171,7 @@ static int populate_shadow_context(struc
 			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;
 }
--- 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 {