Cycles: explicitly skip self-intersection

Remember the last intersected primitive and skip any intersections with the
same primitive.

Ref D12954

28 files changed, 360 insertions, 78 deletions

diff --git a/intern/cycles/bvh/embree.cpp b/intern/cycles/bvh/embree.cpp
index 2e49e29d12b..616b6273e6a 100644
--- a/intern/cycles/bvh/embree.cpp
+++ b/intern/cycles/bvh/embree.cpp
@@ -66,6 +66,26 @@ static_assert(Object::MAX_MOTION_STEPS == Geometry::MAX_MOTION_STEPS,
  * as well as filtering for volume objects happen here.
  * Cycles' own BVH does that directly inside the traversal calls.
  */
+static void rtc_filter_intersection_func(const RTCFilterFunctionNArguments *args)
+{
+  /* Current implementation in Cycles assumes only single-ray intersection queries. */
+  assert(args->N == 1);
+
+  RTCHit *hit = (RTCHit *)args->hit;
+  CCLIntersectContext *ctx = ((IntersectContext *)args->context)->userRayExt;
+  const KernelGlobalsCPU *kg = ctx->kg;
+  const Ray *cray = ctx->ray;
+
+  if (kernel_embree_is_self_intersection(kg, hit, cray)) {
+    *args->valid = 0;
+  }
+}
+
+/* This gets called by Embree at every valid ray/object intersection.
+ * Things like recording subsurface or shadow hits for later evaluation
+ * as well as filtering for volume objects happen here.
+ * Cycles' own BVH does that directly inside the traversal calls.
+ */
 static void rtc_filter_occluded_func(const RTCFilterFunctionNArguments *args)
 {
   /* Current implementation in Cycles assumes only single-ray intersection queries. */
@@ -75,12 +95,16 @@ static void rtc_filter_occluded_func(const RTCFilterFunctionNArguments *args)
   RTCHit *hit = (RTCHit *)args->hit;
   CCLIntersectContext *ctx = ((IntersectContext *)args->context)->userRayExt;
   const KernelGlobalsCPU *kg = ctx->kg;
+  const Ray *cray = ctx->ray;
 
   switch (ctx->type) {
     case CCLIntersectContext::RAY_SHADOW_ALL: {
       Intersection current_isect;
       kernel_embree_convert_hit(kg, ray, hit, &current_isect);
-
+      if (intersection_skip_self_shadow(cray->self, current_isect.object, current_isect.prim)) {
+        *args->valid = 0;
+        return;
+      }
       /* If no transparent shadows or max number of hits exceeded, all light is blocked. */
       const int flags = intersection_get_shader_flags(kg, current_isect.prim, current_isect.type);
       if (!(flags & (SD_HAS_TRANSPARENT_SHADOW)) || ctx->num_hits >= ctx->max_hits) {
@@ -160,6 +184,10 @@ static void rtc_filter_occluded_func(const RTCFilterFunctionNArguments *args)
           break;
         }
       }
+      if (intersection_skip_self_local(cray->self, current_isect.prim)) {
+        *args->valid = 0;
+        return;
+      }
 
       /* No intersection information requested, just return a hit. */
       if (ctx->max_hits == 0) {
@@ -225,6 +253,11 @@ static void rtc_filter_occluded_func(const RTCFilterFunctionNArguments *args)
       if (ctx->num_hits < ctx->max_hits) {
         Intersection current_isect;
         kernel_embree_convert_hit(kg, ray, hit, &current_isect);
+        if (intersection_skip_self(cray->self, current_isect.object, current_isect.prim)) {
+          *args->valid = 0;
+          return;
+        }
+
         Intersection *isect = &ctx->isect_s[ctx->num_hits];
         ++ctx->num_hits;
         *isect = current_isect;
@@ -236,12 +269,15 @@ static void rtc_filter_occluded_func(const RTCFilterFunctionNArguments *args)
         }
         /* This tells Embree to continue tracing. */
         *args->valid = 0;
-        break;
       }
+      break;
     }
     case CCLIntersectContext::RAY_REGULAR:
     default:
-      /* Nothing to do here. */
+      if (kernel_embree_is_self_intersection(kg, hit, cray)) {
+        *args->valid = 0;
+        return;
+      }
       break;
   }
 }
@@ -257,6 +293,14 @@ static void rtc_filter_func_backface_cull(const RTCFilterFunctionNArguments *arg
     *args->valid = 0;
     return;
   }
+
+  CCLIntersectContext *ctx = ((IntersectContext *)args->context)->userRayExt;
+  const KernelGlobalsCPU *kg = ctx->kg;
+  const Ray *cray = ctx->ray;
+
+  if (kernel_embree_is_self_intersection(kg, hit, cray)) {
+    *args->valid = 0;
+  }
 }
 
 static void rtc_filter_occluded_func_backface_cull(const RTCFilterFunctionNArguments *args)
@@ -505,6 +549,7 @@ void BVHEmbree::add_triangles(const Object *ob, const Mesh *mesh, int i)
 
   rtcSetGeometryUserData(geom_id, (void *)prim_offset);
   rtcSetGeometryOccludedFilterFunction(geom_id, rtc_filter_occluded_func);
+  rtcSetGeometryIntersectFilterFunction(geom_id, rtc_filter_intersection_func);
   rtcSetGeometryMask(geom_id, ob->visibility_for_tracing());
 
   rtcCommitGeometry(geom_id);
@@ -767,6 +812,7 @@ void BVHEmbree::add_curves(const Object *ob, const Hair *hair, int i)
 
   rtcSetGeometryUserData(geom_id, (void *)prim_offset);
   if (hair->curve_shape == CURVE_RIBBON) {
+    rtcSetGeometryIntersectFilterFunction(geom_id, rtc_filter_intersection_func);
     rtcSetGeometryOccludedFilterFunction(geom_id, rtc_filter_occluded_func);
   }
   else {
diff --git a/intern/cycles/device/optix/device_impl.cpp b/intern/cycles/device/optix/device_impl.cpp
index 009661b2dec..cb6c36d5ea6 100644
--- a/intern/cycles/device/optix/device_impl.cpp
+++ b/intern/cycles/device/optix/device_impl.cpp
@@ -226,7 +226,7 @@ bool OptiXDevice::load_kernels(const uint kernel_features)
   pipeline_options.usesMotionBlur = false;
   pipeline_options.traversableGraphFlags =
       OPTIX_TRAVERSABLE_GRAPH_FLAG_ALLOW_SINGLE_LEVEL_INSTANCING;
-  pipeline_options.numPayloadValues = 6;
+  pipeline_options.numPayloadValues = 8;
   pipeline_options.numAttributeValues = 2; /* u, v */
   pipeline_options.exceptionFlags = OPTIX_EXCEPTION_FLAG_NONE;
   pipeline_options.pipelineLaunchParamsVariableName = "__params"; /* See globals.h */
diff --git a/intern/cycles/kernel/bvh/bvh.h b/intern/cycles/kernel/bvh/bvh.h
index 67804fb1d0d..1797bf60720 100644
--- a/intern/cycles/kernel/bvh/bvh.h
+++ b/intern/cycles/kernel/bvh/bvh.h
@@ -173,15 +173,16 @@ ccl_device_intersect bool scene_intersect(KernelGlobals kg,
   uint p3 = 0;
   uint p4 = visibility;
   uint p5 = PRIMITIVE_NONE;
+  uint p6 = ((uint64_t)ray) & 0xFFFFFFFF;
+  uint p7 = (((uint64_t)ray) >> 32) & 0xFFFFFFFF;
 
   uint ray_mask = visibility & 0xFF;
-  uint ray_flags = OPTIX_RAY_FLAG_NONE;
+  uint ray_flags = OPTIX_RAY_FLAG_ENFORCE_ANYHIT;
   if (0 == ray_mask && (visibility & ~0xFF) != 0) {
     ray_mask = 0xFF;
-    ray_flags = OPTIX_RAY_FLAG_ENFORCE_ANYHIT;
   }
   else if (visibility & PATH_RAY_SHADOW_OPAQUE) {
-    ray_flags = OPTIX_RAY_FLAG_TERMINATE_ON_FIRST_HIT;
+    ray_flags |= OPTIX_RAY_FLAG_TERMINATE_ON_FIRST_HIT;
   }
 
   optixTrace(scene_intersect_valid(ray) ? kernel_data.bvh.scene : 0,
@@ -200,7 +201,9 @@ ccl_device_intersect bool scene_intersect(KernelGlobals kg,
              p2,
              p3,
              p4,
-             p5);
+             p5,
+             p6,
+             p7);
 
   isect->t = __uint_as_float(p0);
   isect->u = __uint_as_float(p1);
@@ -242,6 +245,7 @@ ccl_device_intersect bool scene_intersect(KernelGlobals kg,
   }
 
   MetalRTIntersectionPayload payload;
+  payload.self = ray->self;
   payload.u = 0.0f;
   payload.v = 0.0f;
   payload.visibility = visibility;
@@ -309,6 +313,7 @@ ccl_device_intersect bool scene_intersect(KernelGlobals kg,
     CCLIntersectContext ctx(kg, CCLIntersectContext::RAY_REGULAR);
     IntersectContext rtc_ctx(&ctx);
     RTCRayHit ray_hit;
+    ctx.ray = ray;
     kernel_embree_setup_rayhit(*ray, ray_hit, visibility);
     rtcIntersect1(kernel_data.bvh.scene, &rtc_ctx.context, &ray_hit);
     if (ray_hit.hit.geomID != RTC_INVALID_GEOMETRY_ID &&
@@ -356,6 +361,9 @@ ccl_device_intersect bool scene_intersect_local(KernelGlobals kg,
   uint p2 = pointer_pack_to_uint_0(local_isect);
   uint p3 = pointer_pack_to_uint_1(local_isect);
   uint p4 = local_object;
+  uint p6 = ((uint64_t)ray) & 0xFFFFFFFF;
+  uint p7 = (((uint64_t)ray) >> 32) & 0xFFFFFFFF;
+
   /* Is set to zero on miss or if ray is aborted, so can be used as return value. */
   uint p5 = max_hits;
 
@@ -379,7 +387,9 @@ ccl_device_intersect bool scene_intersect_local(KernelGlobals kg,
              p2,
              p3,
              p4,
-             p5);
+             p5,
+             p6,
+             p7);
 
   return p5;
 #  elif defined(__METALRT__)
@@ -417,6 +427,7 @@ ccl_device_intersect bool scene_intersect_local(KernelGlobals kg,
   }
 
   MetalRTIntersectionLocalPayload payload;
+  payload.self = ray->self;
   payload.local_object = local_object;
   payload.max_hits = max_hits;
   payload.local_isect.num_hits = 0;
@@ -460,6 +471,7 @@ ccl_device_intersect bool scene_intersect_local(KernelGlobals kg,
         kg, has_bvh ? CCLIntersectContext::RAY_SSS : CCLIntersectContext::RAY_LOCAL);
     ctx.lcg_state = lcg_state;
     ctx.max_hits = max_hits;
+    ctx.ray = ray;
     ctx.local_isect = local_isect;
     if (local_isect) {
       local_isect->num_hits = 0;
@@ -532,6 +544,8 @@ ccl_device_intersect bool scene_intersect_shadow_all(KernelGlobals kg,
   uint p3 = max_hits;
   uint p4 = visibility;
   uint p5 = false;
+  uint p6 = ((uint64_t)ray) & 0xFFFFFFFF;
+  uint p7 = (((uint64_t)ray) >> 32) & 0xFFFFFFFF;
 
   uint ray_mask = visibility & 0xFF;
   if (0 == ray_mask && (visibility & ~0xFF) != 0) {
@@ -555,7 +569,9 @@ ccl_device_intersect bool scene_intersect_shadow_all(KernelGlobals kg,
              p2,
              p3,
              p4,
-             p5);
+             p5,
+             p6,
+             p7);
 
   *num_recorded_hits = uint16_unpack_from_uint_0(p2);
   *throughput = __uint_as_float(p1);
@@ -588,6 +604,7 @@ ccl_device_intersect bool scene_intersect_shadow_all(KernelGlobals kg,
   }
 
   MetalRTIntersectionShadowPayload payload;
+  payload.self = ray->self;
   payload.visibility = visibility;
   payload.max_hits = max_hits;
   payload.num_hits = 0;
@@ -634,6 +651,7 @@ ccl_device_intersect bool scene_intersect_shadow_all(KernelGlobals kg,
     Intersection *isect_array = (Intersection *)state->shadow_isect;
     ctx.isect_s = isect_array;
     ctx.max_hits = max_hits;
+    ctx.ray = ray;
     IntersectContext rtc_ctx(&ctx);
     RTCRay rtc_ray;
     kernel_embree_setup_ray(*ray, rtc_ray, visibility);
@@ -685,6 +703,8 @@ ccl_device_intersect bool scene_intersect_volume(KernelGlobals kg,
   uint p3 = 0;
   uint p4 = visibility;
   uint p5 = PRIMITIVE_NONE;
+  uint p6 = ((uint64_t)ray) & 0xFFFFFFFF;
+  uint p7 = (((uint64_t)ray) >> 32) & 0xFFFFFFFF;
 
   uint ray_mask = visibility & 0xFF;
   if (0 == ray_mask && (visibility & ~0xFF) != 0) {
@@ -708,7 +728,9 @@ ccl_device_intersect bool scene_intersect_volume(KernelGlobals kg,
              p2,
              p3,
              p4,
-             p5);
+             p5,
+             p6,
+             p7);
 
   isect->t = __uint_as_float(p0);
   isect->u = __uint_as_float(p1);
@@ -744,6 +766,7 @@ ccl_device_intersect bool scene_intersect_volume(KernelGlobals kg,
   }
 
   MetalRTIntersectionPayload payload;
+  payload.self = ray->self;
   payload.visibility = visibility;
 
   typename metalrt_intersector_type::result_type intersection;
@@ -820,6 +843,7 @@ ccl_device_intersect uint scene_intersect_volume_all(KernelGlobals kg,
     ctx.isect_s = isect;
     ctx.max_hits = max_hits;
     ctx.num_hits = 0;
+    ctx.ray = ray;
     IntersectContext rtc_ctx(&ctx);
     RTCRay rtc_ray;
     kernel_embree_setup_ray(*ray, rtc_ray, visibility);
diff --git a/intern/cycles/kernel/bvh/embree.h b/intern/cycles/kernel/bvh/embree.h
index 9edd4f90a7e..19c4b9f6f3d 100644
--- a/intern/cycles/kernel/bvh/embree.h
+++ b/intern/cycles/kernel/bvh/embree.h
@@ -22,6 +22,8 @@
 #include "kernel/device/cpu/compat.h"
 #include "kernel/device/cpu/globals.h"
 
+#include "kernel/bvh/util.h"
+
 #include "util/vector.h"
 
 CCL_NAMESPACE_BEGIN
@@ -38,6 +40,9 @@ struct CCLIntersectContext {
   KernelGlobals kg;
   RayType type;
 
+  /* For avoiding self intersections */
+  const Ray *ray;
+
   /* for shadow rays */
   Intersection *isect_s;
   uint max_hits;
@@ -56,6 +61,7 @@ struct CCLIntersectContext {
   {
     kg = kg_;
     type = type_;
+    ray = NULL;
     max_hits = 1;
     num_hits = 0;
     num_recorded_hits = 0;
@@ -102,7 +108,34 @@ ccl_device_inline void kernel_embree_setup_rayhit(const Ray &ray,
 {
   kernel_embree_setup_ray(ray, rayhit.ray, visibility);
   rayhit.hit.geomID = RTC_INVALID_GEOMETRY_ID;
-  rayhit.hit.primID = RTC_INVALID_GEOMETRY_ID;
+  rayhit.hit.instID[0] = RTC_INVALID_GEOMETRY_ID;
+}
+
+ccl_device_inline bool kernel_embree_is_self_intersection(const KernelGlobals kg,
+                                                          const RTCHit *hit,
+                                                          const Ray *ray)
+{
+  bool status = false;
+  if (hit->instID[0] != RTC_INVALID_GEOMETRY_ID) {
+    const int oID = hit->instID[0] / 2;
+    if ((ray->self.object == oID) || (ray->self.light_object == oID)) {
+      RTCScene inst_scene = (RTCScene)rtcGetGeometryUserData(
+          rtcGetGeometry(kernel_data.bvh.scene, hit->instID[0]));
+      const int pID = hit->primID +
+                      (intptr_t)rtcGetGeometryUserData(rtcGetGeometry(inst_scene, hit->geomID));
+      status = intersection_skip_self_shadow(ray->self, oID, pID);
+    }
+  }
+  else {
+    const int oID = hit->geomID / 2;
+    if ((ray->self.object == oID) || (ray->self.light_object == oID)) {
+      const int pID = hit->primID + (intptr_t)rtcGetGeometryUserData(
+                                        rtcGetGeometry(kernel_data.bvh.scene, hit->geomID));
+      status = intersection_skip_self_shadow(ray->self, oID, pID);
+    }
+  }
+
+  return status;
 }
 
 ccl_device_inline void kernel_embree_convert_hit(KernelGlobals kg,
diff --git a/intern/cycles/kernel/bvh/local.h b/intern/cycles/kernel/bvh/local.h
index 4d0e6aac901..4ef6deef98d 100644
--- a/intern/cycles/kernel/bvh/local.h
+++ b/intern/cycles/kernel/bvh/local.h
@@ -157,7 +157,11 @@ ccl_device_inline
                 }
               }
 
+              /* Skip self intersection. */
               const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+              if (intersection_skip_self_local(ray->self, prim)) {
+                continue;
+              }
 
               if (triangle_intersect_local(kg,
                                            local_isect,
@@ -188,7 +192,11 @@ ccl_device_inline
                 }
               }
 
+              /* Skip self intersection. */
               const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+              if (intersection_skip_self_local(ray->self, prim)) {
+                continue;
+              }
 
               if (motion_triangle_intersect_local(kg,
                                                   local_isect,
diff --git a/intern/cycles/kernel/bvh/metal.h b/intern/cycles/kernel/bvh/metal.h
index 55456d15f50..5ab413d9314 100644
--- a/intern/cycles/kernel/bvh/metal.h
+++ b/intern/cycles/kernel/bvh/metal.h
@@ -15,6 +15,7 @@
  */
 
 struct MetalRTIntersectionPayload {
+  RaySelfPrimitives self;
   uint visibility;
   float u, v;
   int prim;
@@ -25,6 +26,7 @@ struct MetalRTIntersectionPayload {
 };
 
 struct MetalRTIntersectionLocalPayload {
+  RaySelfPrimitives self;
   uint local_object;
   uint lcg_state;
   short max_hits;
@@ -34,6 +36,7 @@ struct MetalRTIntersectionLocalPayload {
 };
 
 struct MetalRTIntersectionShadowPayload {
+  RaySelfPrimitives self;
   uint visibility;
 #if defined(__METALRT_MOTION__)
   float time;
diff --git a/intern/cycles/kernel/bvh/shadow_all.h b/intern/cycles/kernel/bvh/shadow_all.h
index 0fb86bfda77..59a7ba63045 100644
--- a/intern/cycles/kernel/bvh/shadow_all.h
+++ b/intern/cycles/kernel/bvh/shadow_all.h
@@ -160,6 +160,9 @@ ccl_device_inline
                                         kernel_tex_fetch(__prim_object, prim_addr) :
                                         object;
             const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+            if (intersection_skip_self_shadow(ray->self, prim_object, prim)) {
+              continue;
+            }
 
             switch (type & PRIMITIVE_ALL) {
               case PRIMITIVE_TRIANGLE: {
diff --git a/intern/cycles/kernel/bvh/traversal.h b/intern/cycles/kernel/bvh/traversal.h
index dc2d1422df6..17cd357a069 100644
--- a/intern/cycles/kernel/bvh/traversal.h
+++ b/intern/cycles/kernel/bvh/traversal.h
@@ -133,35 +133,29 @@ ccl_device_noinline bool BVH_FUNCTION_FULL_NAME(BVH)(KernelGlobals kg,
           --stack_ptr;
 
           /* primitive intersection */
-          switch (type & PRIMITIVE_ALL) {
-            case PRIMITIVE_TRIANGLE: {
-              for (; prim_addr < prim_addr2; prim_addr++) {
-                kernel_assert(kernel_tex_fetch(__prim_type, prim_addr) == type);
-
-                const int prim_object = (object == OBJECT_NONE) ?
-                                            kernel_tex_fetch(__prim_object, prim_addr) :
-                                            object;
-                const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+          for (; prim_addr < prim_addr2; prim_addr++) {
+            kernel_assert(kernel_tex_fetch(__prim_type, prim_addr) == type);
+
+            const int prim_object = (object == OBJECT_NONE) ?
+                                        kernel_tex_fetch(__prim_object, prim_addr) :
+                                        object;
+            const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+            if (intersection_skip_self_shadow(ray->self, prim_object, prim)) {
+              continue;
+            }
 
+            switch (type & PRIMITIVE_ALL) {
+              case PRIMITIVE_TRIANGLE: {
                 if (triangle_intersect(
                         kg, isect, P, dir, isect->t, visibility, prim_object, prim, prim_addr)) {
                   /* shadow ray early termination */
                   if (visibility & PATH_RAY_SHADOW_OPAQUE)
                     return true;
                 }
+                break;
               }
-              break;
-            }
 #if BVH_FEATURE(BVH_MOTION)
-            case PRIMITIVE_MOTION_TRIANGLE: {
-              for (; prim_addr < prim_addr2; prim_addr++) {
-                kernel_assert(kernel_tex_fetch(__prim_type, prim_addr) == type);
-
-                const int prim_object = (object == OBJECT_NONE) ?
-                                            kernel_tex_fetch(__prim_object, prim_addr) :
-                                            object;
-                const int prim = kernel_tex_fetch(__prim_index, prim_addr);
-
+              case PRIMITIVE_MOTION_TRIANGLE: {
                 if (motion_triangle_intersect(kg,
                                               isect,
                                               P,
@@ -176,28 +170,21 @@ ccl_device_noinline bool BVH_FUNCTION_FULL_NAME(BVH)(KernelGlobals kg,
                   if (visibility & PATH_RAY_SHADOW_OPAQUE)
                     return true;
                 }
+                break;
               }
-              break;
-            }
 #endif /* BVH_FEATURE(BVH_MOTION) */
 #if BVH_FEATURE(BVH_HAIR)
-            case PRIMITIVE_CURVE_THICK:
-            case PRIMITIVE_MOTION_CURVE_THICK:
-            case PRIMITIVE_CURVE_RIBBON:
-            case PRIMITIVE_MOTION_CURVE_RIBBON: {
-              for (; prim_addr < prim_addr2; prim_addr++) {
+              case PRIMITIVE_CURVE_THICK:
+              case PRIMITIVE_MOTION_CURVE_THICK:
+              case PRIMITIVE_CURVE_RIBBON:
+              case PRIMITIVE_MOTION_CURVE_RIBBON: {
                 if ((type & PRIMITIVE_MOTION) && kernel_data.bvh.use_bvh_steps) {
                   const float2 prim_time = kernel_tex_fetch(__prim_time, prim_addr);
                   if (ray->time < prim_time.x || ray->time > prim_time.y) {
-                    continue;
+                    break;
                   }
                 }
 
-                const int prim_object = (object == OBJECT_NONE) ?
-                                            kernel_tex_fetch(__prim_object, prim_addr) :
-                                            object;
-                const int prim = kernel_tex_fetch(__prim_index, prim_addr);
-
                 const int curve_type = kernel_tex_fetch(__prim_type, prim_addr);
                 const bool hit = curve_intersect(
                     kg, isect, P, dir, isect->t, prim_object, prim, ray->time, curve_type);
@@ -206,26 +193,19 @@ ccl_device_noinline bool BVH_FUNCTION_FULL_NAME(BVH)(KernelGlobals kg,
                   if (visibility & PATH_RAY_SHADOW_OPAQUE)
                     return true;
                 }
+                break;
               }
-              break;
-            }
 #endif /* BVH_FEATURE(BVH_HAIR) */
 #if BVH_FEATURE(BVH_POINTCLOUD)
-            case PRIMITIVE_POINT:
-            case PRIMITIVE_MOTION_POINT: {
-              for (; prim_addr < prim_addr2; prim_addr++) {
+              case PRIMITIVE_POINT:
+              case PRIMITIVE_MOTION_POINT: {
                 if ((type & PRIMITIVE_MOTION) && kernel_data.bvh.use_bvh_steps) {
                   const float2 prim_time = kernel_tex_fetch(__prim_time, prim_addr);
                   if (ray->time < prim_time.x || ray->time > prim_time.y) {
-                    continue;
+                    break;
                   }
                 }
 
-                const int prim_object = (object == OBJECT_NONE) ?
-                                            kernel_tex_fetch(__prim_object, prim_addr) :
-                                            object;
-                const int prim = kernel_tex_fetch(__prim_index, prim_addr);
-
                 const int point_type = kernel_tex_fetch(__prim_type, prim_addr);
                 const bool hit = point_intersect(
                     kg, isect, P, dir, isect->t, prim_object, prim, ray->time, point_type);
@@ -234,10 +214,10 @@ ccl_device_noinline bool BVH_FUNCTION_FULL_NAME(BVH)(KernelGlobals kg,
                   if (visibility & PATH_RAY_SHADOW_OPAQUE)
                     return true;
                 }
+                break;
               }
-              break;
-            }
 #endif /* BVH_FEATURE(BVH_POINTCLOUD) */
+            }
           }
         }
         else {
diff --git a/intern/cycles/kernel/bvh/util.h b/intern/cycles/kernel/bvh/util.h
index bd79c6e19c6..ea86523c0f6 100644
--- a/intern/cycles/kernel/bvh/util.h
+++ b/intern/cycles/kernel/bvh/util.h
@@ -227,4 +227,25 @@ ccl_device_inline float intersection_curve_shadow_transparency(KernelGlobals kg,
   return (1.0f - u) * f0 + u * f1;
 }
 
+ccl_device_inline bool intersection_skip_self(ccl_private const RaySelfPrimitives &self,
+                                              const int object,
+                                              const int prim)
+{
+  return (self.prim == prim) && (self.object == object);
+}
+
+ccl_device_inline bool intersection_skip_self_shadow(ccl_private const RaySelfPrimitives &self,
+                                                     const int object,
+                                                     const int prim)
+{
+  return ((self.prim == prim) && (self.object == object)) ||
+         ((self.light_prim == prim) && (self.light_object == object));
+}
+
+ccl_device_inline bool intersection_skip_self_local(ccl_private const RaySelfPrimitives &self,
+                                                    const int prim)
+{
+  return (self.prim == prim);
+}
+
 CCL_NAMESPACE_END
diff --git a/intern/cycles/kernel/bvh/volume.h b/intern/cycles/kernel/bvh/volume.h
index c0746c8efc3..95bba4f071d 100644
--- a/intern/cycles/kernel/bvh/volume.h
+++ b/intern/cycles/kernel/bvh/volume.h
@@ -144,6 +144,9 @@ ccl_device_inline
                                             kernel_tex_fetch(__prim_object, prim_addr) :
                                             object;
                 const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+                if (intersection_skip_self(ray->self, prim_object, prim)) {
+                  continue;
+                }
 
                 int object_flag = kernel_tex_fetch(__object_flag, prim_object);
                 if ((object_flag & SD_OBJECT_HAS_VOLUME) == 0) {
@@ -164,6 +167,9 @@ ccl_device_inline
                                             kernel_tex_fetch(__prim_object, prim_addr) :
                                             object;
                 const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+                if (intersection_skip_self(ray->self, prim_object, prim)) {
+                  continue;
+                }
                 int object_flag = kernel_tex_fetch(__object_flag, prim_object);
                 if ((object_flag & SD_OBJECT_HAS_VOLUME) == 0) {
                   continue;
diff --git a/intern/cycles/kernel/bvh/volume_all.h b/intern/cycles/kernel/bvh/volume_all.h
index a88c9d95d46..9f53e987cf1 100644
--- a/intern/cycles/kernel/bvh/volume_all.h
+++ b/intern/cycles/kernel/bvh/volume_all.h
@@ -147,6 +147,9 @@ ccl_device_inline
                                             kernel_tex_fetch(__prim_object, prim_addr) :
                                             object;
                 const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+                if (intersection_skip_self(ray->self, prim_object, prim)) {
+                  continue;
+                }
                 int object_flag = kernel_tex_fetch(__object_flag, prim_object);
                 if ((object_flag & SD_OBJECT_HAS_VOLUME) == 0) {
                   continue;
@@ -188,6 +191,9 @@ ccl_device_inline
                                             kernel_tex_fetch(__prim_object, prim_addr) :
                                             object;
                 const int prim = kernel_tex_fetch(__prim_index, prim_addr);
+                if (intersection_skip_self(ray->self, prim_object, prim)) {
+                  continue;
+                }
                 int object_flag = kernel_tex_fetch(__object_flag, prim_object);
                 if ((object_flag & SD_OBJECT_HAS_VOLUME) == 0) {
                   continue;
diff --git a/intern/cycles/kernel/device/metal/kernel.metal b/intern/cycles/kernel/device/metal/kernel.metal
index 3303b541487..6b77940660f 100644
--- a/intern/cycles/kernel/device/metal/kernel.metal
+++ b/intern/cycles/kernel/device/metal/kernel.metal
@@ -40,6 +40,27 @@ struct TriangleIntersectionResult
 
 enum { METALRT_HIT_TRIANGLE, METALRT_HIT_BOUNDING_BOX };
 
+ccl_device_inline bool intersection_skip_self(ray_data const RaySelfPrimitives& self,
+                                              const int object,
+                                              const int prim)
+{
+  return (self.prim == prim) && (self.object == object);
+}
+
+ccl_device_inline bool intersection_skip_self_shadow(ray_data const RaySelfPrimitives& self,
+                                                     const int object,
+                                                     const int prim)
+{
+  return ((self.prim == prim) && (self.object == object)) ||
+         ((self.light_prim == prim) && (self.light_object == object));
+}
+
+ccl_device_inline bool intersection_skip_self_local(ray_data const RaySelfPrimitives& self,
+                                                    const int prim)
+{
+  return (self.prim == prim);
+}
+
 template<typename TReturn, uint intersection_type>
 TReturn metalrt_local_hit(constant KernelParamsMetal &launch_params_metal,
                           ray_data MetalKernelContext::MetalRTIntersectionLocalPayload &payload,
@@ -53,8 +74,8 @@ TReturn metalrt_local_hit(constant KernelParamsMetal &launch_params_metal,
 #ifdef __BVH_LOCAL__
   uint prim = primitive_id + kernel_tex_fetch(__object_prim_offset, object);
 
-  if (object != payload.local_object) {
-    /* Only intersect with matching object */
+  if ((object != payload.local_object) || intersection_skip_self_local(payload.self, prim)) {
+    /* Only intersect with matching object and skip self-intersecton. */
     result.accept = false;
     result.continue_search = true;
     return result;
@@ -166,6 +187,11 @@ bool metalrt_shadow_all_hit(constant KernelParamsMetal &launch_params_metal,
   }
 #  endif
 
+  if (intersection_skip_self_shadow(payload.self, object, prim)) {
+    /* continue search */
+    return true;
+  }
+
   float u = 0.0f, v = 0.0f;
   int type = 0;
   if (intersection_type == METALRT_HIT_TRIANGLE) {
@@ -322,21 +348,35 @@ inline TReturnType metalrt_visibility_test(constant KernelParamsMetal &launch_pa
   }
 #  endif
 
-#  ifdef __VISIBILITY_FLAG__
   uint visibility = payload.visibility;
+#  ifdef __VISIBILITY_FLAG__
   if ((kernel_tex_fetch(__objects, object).visibility & visibility) == 0) {
     result.accept = false;
     result.continue_search = true;
     return result;
   }
+#  endif
 
   /* Shadow ray early termination. */
   if (visibility & PATH_RAY_SHADOW_OPAQUE) {
-    result.accept = true;
-    result.continue_search = false;
-    return result;
+    if (intersection_skip_self_shadow(payload.self, object, prim)) {
+      result.accept = false;
+      result.continue_search = true;
+      return result;
+    }
+    else {
+      result.accept = true;
+      result.continue_search = false;
+      return result;
+    }
+  }
+  else {
+    if (intersection_skip_self(payload.self, object, prim)) {
+      result.accept = false;
+      result.continue_search = true;
+      return result;
+    }
   }
-#  endif
 
   result.accept = true;
   result.continue_search = true;
diff --git a/intern/cycles/kernel/device/optix/kernel.cu b/intern/cycles/kernel/device/optix/kernel.cu
index aa210b31a95..8e3d57bff8a 100644
--- a/intern/cycles/kernel/device/optix/kernel.cu
+++ b/intern/cycles/kernel/device/optix/kernel.cu
@@ -45,6 +45,11 @@ template<typename T> ccl_device_forceinline T *get_payload_ptr_2()
   return pointer_unpack_from_uint<T>(optixGetPayload_2(), optixGetPayload_3());
 }
 
+template<typename T> ccl_device_forceinline T *get_payload_ptr_6()
+{
+  return (T *)(((uint64_t)optixGetPayload_7() << 32) | optixGetPayload_6());
+}
+
 ccl_device_forceinline int get_object_id()
 {
 #ifdef __OBJECT_MOTION__
@@ -111,6 +116,12 @@ extern "C" __global__ void __anyhit__kernel_optix_local_hit()
     return optixIgnoreIntersection();
   }
 
+  const int prim = optixGetPrimitiveIndex();
+  ccl_private Ray *const ray = get_payload_ptr_6<Ray>();
+  if (intersection_skip_self_local(ray->self, prim)) {
+    return optixIgnoreIntersection();
+  }
+
   const uint max_hits = optixGetPayload_5();
   if (max_hits == 0) {
     /* Special case for when no hit information is requested, just report that something was hit */
@@ -149,8 +160,6 @@ extern "C" __global__ void __anyhit__kernel_optix_local_hit()
     local_isect->num_hits = 1;
   }
 
-  const int prim = optixGetPrimitiveIndex();
-
   Intersection *isect = &local_isect->hits[hit];
   isect->t = optixGetRayTmax();
   isect->prim = prim;
@@ -185,6 +194,11 @@ extern "C" __global__ void __anyhit__kernel_optix_shadow_all_hit()
   }
 #  endif
 
+  ccl_private Ray *const ray = get_payload_ptr_6<Ray>();
+  if (intersection_skip_self_shadow(ray->self, object, prim)) {
+    return optixIgnoreIntersection();
+  }
+
   float u = 0.0f, v = 0.0f;
   int type = 0;
   if (optixIsTriangleHit()) {
@@ -314,6 +328,12 @@ extern "C" __global__ void __anyhit__kernel_optix_volume_test()
   if ((kernel_tex_fetch(__object_flag, object) & SD_OBJECT_HAS_VOLUME) == 0) {
     return optixIgnoreIntersection();
   }
+
+  const int prim = optixGetPrimitiveIndex();
+  ccl_private Ray *const ray = get_payload_ptr_6<Ray>();
+  if (intersection_skip_self(ray->self, object, prim)) {
+    return optixIgnoreIntersection();
+  }
 }
 
 extern "C" __global__ void __anyhit__kernel_optix_visibility_test()
@@ -330,18 +350,31 @@ extern "C" __global__ void __anyhit__kernel_optix_visibility_test()
 #  endif
 #endif
 
-#ifdef __VISIBILITY_FLAG__
   const uint object = get_object_id();
   const uint visibility = optixGetPayload_4();
+#ifdef __VISIBILITY_FLAG__
   if ((kernel_tex_fetch(__objects, object).visibility & visibility) == 0) {
     return optixIgnoreIntersection();
   }
+#endif
+
+  const int prim = optixGetPrimitiveIndex();
+  ccl_private Ray *const ray = get_payload_ptr_6<Ray>();
 
-  /* Shadow ray early termination. */
   if (visibility & PATH_RAY_SHADOW_OPAQUE) {
-    return optixTerminateRay();
+    if (intersection_skip_self_shadow(ray->self, object, prim)) {
+      return optixIgnoreIntersection();
+    }
+    else {
+      /* Shadow ray early termination. */
+      return optixTerminateRay();
+    }
+  }
+  else {
+    if (intersection_skip_self(ray->self, object, prim)) {
+      return optixIgnoreIntersection();
+    }
   }
-#endif
 }
 
 extern "C" __global__ void __closesthit__kernel_optix_hit()
diff --git a/intern/cycles/kernel/integrator/intersect_closest.h b/intern/cycles/kernel/integrator/intersect_closest.h
index df710dc1d82..4c5265189fa 100644
--- a/intern/cycles/kernel/integrator/intersect_closest.h
+++ b/intern/cycles/kernel/integrator/intersect_closest.h
@@ -328,6 +328,12 @@ ccl_device void integrator_intersect_closest(KernelGlobals kg,
 
   /* Scene Intersection. */
   Intersection isect ccl_optional_struct_init;
+  isect.object = OBJECT_NONE;
+  isect.prim = PRIM_NONE;
+  ray.self.object = last_isect_object;
+  ray.self.prim = last_isect_prim;
+  ray.self.light_object = OBJECT_NONE;
+  ray.self.light_prim = PRIM_NONE;
   bool hit = scene_intersect(kg, &ray, visibility, &isect);
 
   /* TODO: remove this and do it in the various intersection functions instead. */
diff --git a/intern/cycles/kernel/integrator/intersect_shadow.h b/intern/cycles/kernel/integrator/intersect_shadow.h
index 90422445fad..1ba8724826b 100644
--- a/intern/cycles/kernel/integrator/intersect_shadow.h
+++ b/intern/cycles/kernel/integrator/intersect_shadow.h
@@ -156,7 +156,10 @@ ccl_device void integrator_intersect_shadow(KernelGlobals kg, IntegratorShadowSt
   /* Read ray from integrator state into local memory. */
   Ray ray ccl_optional_struct_init;
   integrator_state_read_shadow_ray(kg, state, &ray);
-
+  ray.self.object = INTEGRATOR_STATE_ARRAY(state, shadow_isect, 0, object);
+  ray.self.prim = INTEGRATOR_STATE_ARRAY(state, shadow_isect, 0, prim);
+  ray.self.light_object = INTEGRATOR_STATE_ARRAY(state, shadow_isect, 1, object);
+  ray.self.light_prim = INTEGRATOR_STATE_ARRAY(state, shadow_isect, 1, prim);
   /* Compute visibility. */
   const uint visibility = integrate_intersect_shadow_visibility(kg, state);
 
diff --git a/intern/cycles/kernel/integrator/intersect_volume_stack.h b/intern/cycles/kernel/integrator/intersect_volume_stack.h
index 9fa5ff63ad2..aa7be879995 100644
--- a/intern/cycles/kernel/integrator/intersect_volume_stack.h
+++ b/intern/cycles/kernel/integrator/intersect_volume_stack.h
@@ -38,7 +38,10 @@ ccl_device void integrator_volume_stack_update_for_subsurface(KernelGlobals kg,
   Ray volume_ray ccl_optional_struct_init;
   volume_ray.P = from_P;
   volume_ray.D = normalize_len(to_P - from_P, &volume_ray.t);
-
+  volume_ray.self.object = INTEGRATOR_STATE(state, isect, object);
+  volume_ray.self.prim = INTEGRATOR_STATE(state, isect, prim);
+  volume_ray.self.light_object = OBJECT_NONE;
+  volume_ray.self.light_prim = PRIM_NONE;
   /* Store to avoid global fetches on every intersection step. */
   const uint volume_stack_size = kernel_data.volume_stack_size;
 
@@ -91,6 +94,10 @@ ccl_device void integrator_volume_stack_init(KernelGlobals kg, IntegratorState s
    * fewest hits. */
   volume_ray.D = make_float3(0.0f, 0.0f, 1.0f);
   volume_ray.t = FLT_MAX;
+  volume_ray.self.object = OBJECT_NONE;
+  volume_ray.self.prim = PRIM_NONE;
+  volume_ray.self.light_object = OBJECT_NONE;
+  volume_ray.self.light_prim = PRIM_NONE;
 
   int stack_index = 0, enclosed_index = 0;
 
diff --git a/intern/cycles/kernel/integrator/shade_shadow.h b/intern/cycles/kernel/integrator/shade_shadow.h
index a68fcaa7a64..10ec48c8637 100644
--- a/intern/cycles/kernel/integrator/shade_shadow.h
+++ b/intern/cycles/kernel/integrator/shade_shadow.h
@@ -83,7 +83,10 @@ ccl_device_inline void integrate_transparent_volume_shadow(KernelGlobals kg,
   /* Setup shader data. */
   Ray ray ccl_optional_struct_init;
   integrator_state_read_shadow_ray(kg, state, &ray);
-
+  ray.self.object = OBJECT_NONE;
+  ray.self.prim = PRIM_NONE;
+  ray.self.light_object = OBJECT_NONE;
+  ray.self.light_prim = PRIM_NONE;
   /* Modify ray position and length to match current segment. */
   const float start_t = (hit == 0) ? 0.0f :
                                      INTEGRATOR_STATE_ARRAY(state, shadow_isect, hit - 1, t);
diff --git a/intern/cycles/kernel/integrator/shade_surface.h b/intern/cycles/kernel/integrator/shade_surface.h
index 9f6077e5d66..3ca9e773591 100644
--- a/intern/cycles/kernel/integrator/shade_surface.h
+++ b/intern/cycles/kernel/integrator/shade_surface.h
@@ -182,6 +182,11 @@ ccl_device_forceinline void integrate_surface_direct_light(KernelGlobals kg,
 
   /* Write shadow ray and associated state to global memory. */
   integrator_state_write_shadow_ray(kg, shadow_state, &ray);
+  // Save memory by storing the light and object indices in the shadow_isect
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 0, object) = ray.self.object;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 0, prim) = ray.self.prim;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 1, object) = ray.self.light_object;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 1, prim) = ray.self.light_prim;
 
   /* Copy state from main path to shadow path. */
   const uint16_t bounce = INTEGRATOR_STATE(state, path, bounce);
@@ -364,6 +369,10 @@ ccl_device_forceinline void integrate_surface_ao(KernelGlobals kg,
   ray.D = ao_D;
   ray.t = kernel_data.integrator.ao_bounces_distance;
   ray.time = sd->time;
+  ray.self.object = sd->object;
+  ray.self.prim = sd->prim;
+  ray.self.light_object = OBJECT_NONE;
+  ray.self.light_prim = PRIM_NONE;
   ray.dP = differential_zero_compact();
   ray.dD = differential_zero_compact();
 
@@ -375,6 +384,10 @@ ccl_device_forceinline void integrate_surface_ao(KernelGlobals kg,
 
   /* Write shadow ray and associated state to global memory. */
   integrator_state_write_shadow_ray(kg, shadow_state, &ray);
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 0, object) = ray.self.object;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 0, prim) = ray.self.prim;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 1, object) = ray.self.light_object;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 1, prim) = ray.self.light_prim;
 
   /* Copy state from main path to shadow path. */
   const uint16_t bounce = INTEGRATOR_STATE(state, path, bounce);
diff --git a/intern/cycles/kernel/integrator/shade_volume.h b/intern/cycles/kernel/integrator/shade_volume.h
index 00fa256d894..107d5ec1795 100644
--- a/intern/cycles/kernel/integrator/shade_volume.h
+++ b/intern/cycles/kernel/integrator/shade_volume.h
@@ -791,6 +791,10 @@ ccl_device_forceinline void integrate_volume_direct_light(
 
   /* Write shadow ray and associated state to global memory. */
   integrator_state_write_shadow_ray(kg, shadow_state, &ray);
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 0, object) = ray.self.object;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 0, prim) = ray.self.prim;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 1, object) = ray.self.light_object;
+  INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_isect, 1, prim) = ray.self.light_prim;
 
   /* Copy state from main path to shadow path. */
   const uint16_t bounce = INTEGRATOR_STATE(state, path, bounce);
@@ -878,6 +882,9 @@ ccl_device_forceinline bool integrate_volume_phase_scatter(
   INTEGRATOR_STATE_WRITE(state, ray, dP) = differential_make_compact(sd->dP);
   INTEGRATOR_STATE_WRITE(state, ray, dD) = differential_make_compact(phase_domega_in);
 #  endif
+  // Save memory by storing last hit prim and object in isect
+  INTEGRATOR_STATE_WRITE(state, isect, prim) = sd->prim;
+  INTEGRATOR_STATE_WRITE(state, isect, object) = sd->object;
 
   /* Update throughput. */
   const float3 throughput = INTEGRATOR_STATE(state, path, throughput);
diff --git a/intern/cycles/kernel/integrator/shadow_state_template.h b/intern/cycles/kernel/integrator/shadow_state_template.h
index 625a429d3db..86fcabdcd82 100644
--- a/intern/cycles/kernel/integrator/shadow_state_template.h
+++ b/intern/cycles/kernel/integrator/shadow_state_template.h
@@ -61,6 +61,7 @@ KERNEL_STRUCT_MEMBER(shadow_ray, packed_float3, D, KERNEL_FEATURE_PATH_TRACING)
 KERNEL_STRUCT_MEMBER(shadow_ray, float, t, KERNEL_FEATURE_PATH_TRACING)
 KERNEL_STRUCT_MEMBER(shadow_ray, float, time, KERNEL_FEATURE_PATH_TRACING)
 KERNEL_STRUCT_MEMBER(shadow_ray, float, dP, KERNEL_FEATURE_PATH_TRACING)
+KERNEL_STRUCT_MEMBER(shadow_ray, int, object, KERNEL_FEATURE_PATH_TRACING)
 KERNEL_STRUCT_END(shadow_ray)
 
 /*********************** Shadow Intersection result **************************/
diff --git a/intern/cycles/kernel/integrator/subsurface.h b/intern/cycles/kernel/integrator/subsurface.h
index 59b0cd2596c..5a09fc51d75 100644
--- a/intern/cycles/kernel/integrator/subsurface.h
+++ b/intern/cycles/kernel/integrator/subsurface.h
@@ -57,7 +57,6 @@ ccl_device int subsurface_bounce(KernelGlobals kg,
 
   /* Pass along object info, reusing isect to save memory. */
   INTEGRATOR_STATE_WRITE(state, subsurface, Ng) = sd->Ng;
-  INTEGRATOR_STATE_WRITE(state, isect, object) = sd->object;
 
   uint32_t path_flag = (INTEGRATOR_STATE(state, path, flag) & ~PATH_RAY_CAMERA) |
                        ((sc->type == CLOSURE_BSSRDF_BURLEY_ID) ? PATH_RAY_SUBSURFACE_DISK :
diff --git a/intern/cycles/kernel/integrator/subsurface_disk.h b/intern/cycles/kernel/integrator/subsurface_disk.h
index cc6f5048cda..f5641d1fa5e 100644
--- a/intern/cycles/kernel/integrator/subsurface_disk.h
+++ b/intern/cycles/kernel/integrator/subsurface_disk.h
@@ -99,6 +99,10 @@ ccl_device_inline bool subsurface_disk(KernelGlobals kg,
   ray.dP = ray_dP;
   ray.dD = differential_zero_compact();
   ray.time = time;
+  ray.self.object = OBJECT_NONE;
+  ray.self.prim = PRIM_NONE;
+  ray.self.light_object = OBJECT_NONE;
+  ray.self.light_prim = OBJECT_NONE;
 
   /* Intersect with the same object. if multiple intersections are found it
    * will use at most BSSRDF_MAX_HITS hits, a random subset of all hits. */
diff --git a/intern/cycles/kernel/integrator/subsurface_random_walk.h b/intern/cycles/kernel/integrator/subsurface_random_walk.h
index 7a8b467e199..43676fccfe5 100644
--- a/intern/cycles/kernel/integrator/subsurface_random_walk.h
+++ b/intern/cycles/kernel/integrator/subsurface_random_walk.h
@@ -195,6 +195,7 @@ ccl_device_inline bool subsurface_random_walk(KernelGlobals kg,
   const float time = INTEGRATOR_STATE(state, ray, time);
   const float3 Ng = INTEGRATOR_STATE(state, subsurface, Ng);
   const int object = INTEGRATOR_STATE(state, isect, object);
+  const int prim = INTEGRATOR_STATE(state, isect, prim);
 
   /* Sample diffuse surface scatter into the object. */
   float3 D;
@@ -211,6 +212,10 @@ ccl_device_inline bool subsurface_random_walk(KernelGlobals kg,
   ray.time = time;
   ray.dP = ray_dP;
   ray.dD = differential_zero_compact();
+  ray.self.object = object;
+  ray.self.prim = prim;
+  ray.self.light_object = OBJECT_NONE;
+  ray.self.light_prim = PRIM_NONE;
 
 #ifndef __KERNEL_GPU_RAYTRACING__
   /* Compute or fetch object transforms. */
@@ -377,7 +382,15 @@ ccl_device_inline bool subsurface_random_walk(KernelGlobals kg,
      * If yes, we will later use backwards guided sampling in order to have a decent
      * chance of connecting to it.
      * TODO: Maybe use less than 10 times the mean free path? */
-    ray.t = (bounce == 0) ? max(t, 10.0f / (min3(sigma_t))) : t;
+    if (bounce == 0) {
+      ray.t = max(t, 10.0f / (min3(sigma_t)));
+    }
+    else {
+      ray.t = t;
+      /* After the first bounce the object can intersect the same surface again */
+      ray.self.object = OBJECT_NONE;
+      ray.self.prim = PRIM_NONE;
+    }
     scene_intersect_local(kg, &ray, &ss_isect, object, NULL, 1);
     hit = (ss_isect.num_hits > 0);
 
diff --git a/intern/cycles/kernel/light/light.h b/intern/cycles/kernel/light/light.h
index 6e445f862db..b9c0b533518 100644
--- a/intern/cycles/kernel/light/light.h
+++ b/intern/cycles/kernel/light/light.h
@@ -418,8 +418,8 @@ ccl_device bool light_sample_from_intersection(KernelGlobals kg,
   LightType type = (LightType)klight->type;
   ls->type = type;
   ls->shader = klight->shader_id;
-  ls->object = PRIM_NONE;
-  ls->prim = PRIM_NONE;
+  ls->object = isect->object;
+  ls->prim = isect->prim;
   ls->lamp = lamp;
   /* todo: missing texture coordinates */
   ls->t = isect->t;
diff --git a/intern/cycles/kernel/light/sample.h b/intern/cycles/kernel/light/sample.h
index 7dbc783b1bb..65e87e77c36 100644
--- a/intern/cycles/kernel/light/sample.h
+++ b/intern/cycles/kernel/light/sample.h
@@ -257,6 +257,12 @@ ccl_device_inline void shadow_ray_setup(ccl_private const ShaderData *ccl_restri
   ray->dP = differential_make_compact(sd->dP);
   ray->dD = differential_zero_compact();
   ray->time = sd->time;
+
+  /* Fill in intersection surface and light details. */
+  ray->self.prim = sd->prim;
+  ray->self.object = sd->object;
+  ray->self.light_prim = ls->prim;
+  ray->self.light_object = ls->object;
 }
 
 /* Create shadow ray towards light sample. */
diff --git a/intern/cycles/kernel/svm/ao.h b/intern/cycles/kernel/svm/ao.h
index 678f49c8ccd..e3abc9d69ff 100644
--- a/intern/cycles/kernel/svm/ao.h
+++ b/intern/cycles/kernel/svm/ao.h
@@ -74,6 +74,10 @@ ccl_device float svm_ao(
     ray.D = D.x * T + D.y * B + D.z * N;
     ray.t = max_dist;
     ray.time = sd->time;
+    ray.self.object = sd->object;
+    ray.self.prim = sd->prim;
+    ray.self.light_object = OBJECT_NONE;
+    ray.self.light_prim = PRIM_NONE;
     ray.dP = differential_zero_compact();
     ray.dD = differential_zero_compact();
 
diff --git a/intern/cycles/kernel/svm/bevel.h b/intern/cycles/kernel/svm/bevel.h
index 57c0288a96f..98b663299da 100644
--- a/intern/cycles/kernel/svm/bevel.h
+++ b/intern/cycles/kernel/svm/bevel.h
@@ -196,6 +196,10 @@ ccl_device float3 svm_bevel(
     ray.dP = differential_zero_compact();
     ray.dD = differential_zero_compact();
     ray.time = sd->time;
+    ray.self.object = OBJECT_NONE;
+    ray.self.prim = PRIM_NONE;
+    ray.self.light_object = OBJECT_NONE;
+    ray.self.light_prim = PRIM_NONE;
 
     /* Intersect with the same object. if multiple intersections are found it
      * will use at most LOCAL_MAX_HITS hits, a random subset of all hits. */
diff --git a/intern/cycles/kernel/types.h b/intern/cycles/kernel/types.h
index 3d9a8b403ac..d4cb22d4af4 100644
--- a/intern/cycles/kernel/types.h
+++ b/intern/cycles/kernel/types.h
@@ -512,12 +512,21 @@ typedef struct differential {
 
 /* Ray */
 
+typedef struct RaySelfPrimitives {
+  int prim;         /* Primitive the ray is starting from */
+  int object;       /* Instance prim is a part of */
+  int light_prim;   /* Light primitive */
+  int light_object; /* Light object */
+} RaySelfPrimitives;
+
 typedef struct Ray {
   float3 P;   /* origin */
   float3 D;   /* direction */
   float t;    /* length of the ray */
   float time; /* time (for motion blur) */
 
+  RaySelfPrimitives self;
+
 #ifdef __RAY_DIFFERENTIALS__
   float dP;
   float dD;