drm/i915/gvt: properly check per_ctx bb valid state

Need to check valid state for per_ctx bb and bypass batch buffer
combine for scan if necessary. Otherwise adding invalid MI batch
buffer start cmd for per_ctx bb will cause scan failure, which is
taken as -EFAULT now so vGPU would be put in failsafe. This trys
to fix that by checking per_ctx bb valid state. Also remove old
invalid WARNING that indirect ctx bb shouldn't depend on valid
per_ctx bb.

Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index 21c36e2..d4726a3 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -2723,6 +2723,9 @@ static int combine_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 	uint32_t per_ctx_start[CACHELINE_DWORDS] = {0};
 	unsigned char *bb_start_sva;
 
+	if (!wa_ctx->per_ctx.valid)
+		return 0;
+
 	per_ctx_start[0] = 0x18800001;
 	per_ctx_start[1] = wa_ctx->per_ctx.guest_gma;
 
diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c
index 91b4300..e5320b4 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.c
+++ b/drivers/gpu/drm/i915/gvt/execlist.c
@@ -701,8 +701,7 @@ static int submit_context(struct intel_vgpu *vgpu, int ring_id,
 			CACHELINE_BYTES;
 		workload->wa_ctx.per_ctx.guest_gma =
 			per_ctx & PER_CTX_ADDR_MASK;
-
-		WARN_ON(workload->wa_ctx.indirect_ctx.size && !(per_ctx & 0x1));
+		workload->wa_ctx.per_ctx.valid = per_ctx & 1;
 	}
 
 	if (emulate_schedule_in)
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h
index 0d431a9..93a49eb 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.h
+++ b/drivers/gpu/drm/i915/gvt/scheduler.h
@@ -68,6 +68,7 @@ struct shadow_indirect_ctx {
 struct shadow_per_ctx {
 	unsigned long guest_gma;
 	unsigned long shadow_gma;
+	unsigned valid;
 };
 
 struct intel_shadow_wa_ctx {