diff options
-rw-r--r-- | build_files/buildbot/config/blender_linux.cmake | 2 | ||||
-rw-r--r-- | intern/cycles/kernel/kernel_path.h | 2 | ||||
-rw-r--r-- | intern/cycles/kernel/kernel_path_branched.h | 2 | ||||
-rw-r--r-- | intern/cycles/kernel/kernel_path_surface.h | 8 | ||||
-rw-r--r-- | intern/cycles/kernel/kernel_path_volume.h | 8 | ||||
-rw-r--r-- | intern/cycles/kernel/kernel_shadow.h | 86 | ||||
-rw-r--r-- | intern/cycles/kernel/split/kernel_shadow_blocked_dl.h | 1 | ||||
-rw-r--r-- | source/blender/blenkernel/intern/pointcache.c | 8 | ||||
-rw-r--r-- | source/blender/blenkernel/intern/rigidbody.c | 13 | ||||
-rw-r--r-- | source/blender/editors/mesh/editmesh_knife.c | 12 | ||||
-rw-r--r-- | source/blender/gpu/intern/gpu_draw.c | 4 |
11 files changed, 102 insertions, 44 deletions
diff --git a/build_files/buildbot/config/blender_linux.cmake b/build_files/buildbot/config/blender_linux.cmake index 4fc6496728c..53e841ba536 100644 --- a/build_files/buildbot/config/blender_linux.cmake +++ b/build_files/buildbot/config/blender_linux.cmake @@ -70,7 +70,7 @@ set(FFMPEG_LIBRARIES ) # SndFile libraries -set(LIBSNDFILE_LIBRARY "/usr/lib/libsndfile.a;/usr/lib/libFLAC.a" CACHE STRING "" FORCE) +set(SNDFILE_LIBRARY "/usr/lib${MULTILIB}/libsndfile.a;/usr/lib${MULTILIB}/libFLAC.a" CACHE STRING "" FORCE) # OpenAL libraries set(OPENAL_ROOT_DIR "/opt/lib/openal" CACHE STRING "" FORCE) diff --git a/intern/cycles/kernel/kernel_path.h b/intern/cycles/kernel/kernel_path.h index ec8c297fbd5..3319e2c2435 100644 --- a/intern/cycles/kernel/kernel_path.h +++ b/intern/cycles/kernel/kernel_path.h @@ -84,7 +84,7 @@ ccl_device_noinline void kernel_path_ao(KernelGlobals *kg, light_ray.dP = sd->dP; light_ray.dD = differential3_zero(); - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &ao_shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &ao_shadow)) { path_radiance_accum_ao(L, state, throughput, ao_alpha, ao_bsdf, ao_shadow); } else { diff --git a/intern/cycles/kernel/kernel_path_branched.h b/intern/cycles/kernel/kernel_path_branched.h index c62c3a25405..dde40674ee6 100644 --- a/intern/cycles/kernel/kernel_path_branched.h +++ b/intern/cycles/kernel/kernel_path_branched.h @@ -54,7 +54,7 @@ ccl_device_inline void kernel_branched_path_ao(KernelGlobals *kg, light_ray.dP = sd->dP; light_ray.dD = differential3_zero(); - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &ao_shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &ao_shadow)) { path_radiance_accum_ao(L, state, throughput*num_samples_inv, ao_alpha, ao_bsdf, ao_shadow); } else { diff --git a/intern/cycles/kernel/kernel_path_surface.h b/intern/cycles/kernel/kernel_path_surface.h index 3d10736e90c..6c3a444e48a 100644 --- a/intern/cycles/kernel/kernel_path_surface.h +++ b/intern/cycles/kernel/kernel_path_surface.h @@ -67,7 +67,7 @@ ccl_device_noinline void kernel_branched_path_surface_connect_light( /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, throughput*num_samples_inv, &L_light, shadow, num_samples_inv, is_lamp); } @@ -104,7 +104,7 @@ ccl_device_noinline void kernel_branched_path_surface_connect_light( /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, throughput*num_samples_inv, &L_light, shadow, num_samples_inv, is_lamp); } @@ -130,7 +130,7 @@ ccl_device_noinline void kernel_branched_path_surface_connect_light( /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, throughput*num_samples_adjust, &L_light, shadow, num_samples_adjust, is_lamp); } @@ -257,7 +257,7 @@ ccl_device_inline void kernel_path_surface_connect_light(KernelGlobals *kg, /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, throughput, &L_light, shadow, 1.0f, is_lamp); } diff --git a/intern/cycles/kernel/kernel_path_volume.h b/intern/cycles/kernel/kernel_path_volume.h index 3661432f0b7..c9c7f447c42 100644 --- a/intern/cycles/kernel/kernel_path_volume.h +++ b/intern/cycles/kernel/kernel_path_volume.h @@ -52,7 +52,7 @@ ccl_device_inline void kernel_path_volume_connect_light( /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, throughput, &L_light, shadow, 1.0f, is_lamp); } @@ -179,7 +179,7 @@ ccl_device void kernel_branched_path_volume_connect_light( /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, tp*num_samples_inv, &L_light, shadow, num_samples_inv, is_lamp); } @@ -228,7 +228,7 @@ ccl_device void kernel_branched_path_volume_connect_light( /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, tp*num_samples_inv, &L_light, shadow, num_samples_inv, is_lamp); } @@ -266,7 +266,7 @@ ccl_device void kernel_branched_path_volume_connect_light( /* trace shadow ray */ float3 shadow; - if(!shadow_blocked(kg, emission_sd, state, &light_ray, &shadow)) { + if(!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &shadow)) { /* accumulate */ path_radiance_accum_light(L, state, tp, &L_light, shadow, 1.0f, is_lamp); } diff --git a/intern/cycles/kernel/kernel_shadow.h b/intern/cycles/kernel/kernel_shadow.h index bb6bdc7fbd0..22e085e94da 100644 --- a/intern/cycles/kernel/kernel_shadow.h +++ b/intern/cycles/kernel/kernel_shadow.h @@ -119,12 +119,46 @@ ccl_device bool shadow_blocked_opaque(KernelGlobals *kg, # define SHADOW_STACK_MAX_HITS 64 +# ifdef __VOLUME__ +struct VolumeState { +# ifdef __SPLIT_KERNEL__ +# else + PathState ps; +# endif +}; + +/* Get PathState ready for use for volume stack evaluation. */ +ccl_device_inline PathState *shadow_blocked_volume_path_state( + KernelGlobals *kg, + VolumeState *volume_state, + ccl_addr_space PathState *state, + ShaderData *sd, + Ray *ray) +{ +# ifdef __SPLIT_KERNEL__ + ccl_addr_space PathState *ps = + &kernel_split_state.state_shadow[ccl_global_id(1) * ccl_global_size(0) + ccl_global_id(0)]; +# else + PathState *ps = &volume_state->ps; +# endif + *ps = *state; + /* We are checking for shadow on the "other" side of the surface, so need + * to discard volume we are currently at. + */ + if(dot(sd->Ng, ray->D) < 0.0f) { + kernel_volume_stack_enter_exit(kg, sd, ps->volume_stack); + } + return ps; +} +#endif // __VOLUME__ + /* Actual logic with traversal loop implementation which is free from device * specific tweaks. * * Note that hits array should be as big as max_hits+1. */ ccl_device bool shadow_blocked_transparent_all_loop(KernelGlobals *kg, + ShaderData *sd, ShaderData *shadow_sd, ccl_addr_space PathState *state, const uint visibility, @@ -143,6 +177,9 @@ ccl_device bool shadow_blocked_transparent_all_loop(KernelGlobals *kg, visibility, max_hits, &num_hits); +# ifdef __VOLUME__ + VolumeState volume_state; +# endif /* If no opaque surface found but we did find transparent hits, * shade them. */ @@ -153,13 +190,11 @@ ccl_device bool shadow_blocked_transparent_all_loop(KernelGlobals *kg, int bounce = state->transparent_bounce; Intersection *isect = hits; # ifdef __VOLUME__ -# ifdef __SPLIT_KERNEL__ - ccl_addr_space PathState *ps = &kernel_split_state.state_shadow[ccl_global_id(1) * ccl_global_size(0) + ccl_global_id(0)]; -# else - PathState ps_object; - PathState *ps = &ps_object; -# endif - *ps = *state; + PathState *ps = shadow_blocked_volume_path_state(kg, + &volume_state, + state, + sd, + ray); # endif sort_intersections(hits, num_hits); for(int hit = 0; hit < num_hits; hit++, isect++) { @@ -205,7 +240,12 @@ ccl_device bool shadow_blocked_transparent_all_loop(KernelGlobals *kg, # ifdef __VOLUME__ if(!blocked && state->volume_stack[0].shader != SHADER_NONE) { /* Apply attenuation from current volume shader. */ - kernel_volume_shadow(kg, shadow_sd, state, ray, shadow); + PathState *ps = shadow_blocked_volume_path_state(kg, + &volume_state, + state, + sd, + ray); + kernel_volume_shadow(kg, shadow_sd, ps, ray, shadow); } # endif return blocked; @@ -215,6 +255,7 @@ ccl_device bool shadow_blocked_transparent_all_loop(KernelGlobals *kg, * loop to help readability of the actual logic. */ ccl_device bool shadow_blocked_transparent_all(KernelGlobals *kg, + ShaderData *sd, ShaderData *shadow_sd, ccl_addr_space PathState *state, const uint visibility, @@ -250,6 +291,7 @@ ccl_device bool shadow_blocked_transparent_all(KernelGlobals *kg, # endif /* __KERNEL_GPU__ */ /* Invoke actual traversal. */ return shadow_blocked_transparent_all_loop(kg, + sd, shadow_sd, state, visibility, @@ -275,6 +317,7 @@ ccl_device bool shadow_blocked_transparent_all(KernelGlobals *kg, */ ccl_device bool shadow_blocked_transparent_stepped_loop( KernelGlobals *kg, + ShaderData *sd, ShaderData *shadow_sd, ccl_addr_space PathState *state, const uint visibility, @@ -284,18 +327,19 @@ ccl_device bool shadow_blocked_transparent_stepped_loop( const bool is_transparent_isect, float3 *shadow) { +# ifdef __VOLUME__ + VolumeState volume_state; +# endif if(blocked && is_transparent_isect) { float3 throughput = make_float3(1.0f, 1.0f, 1.0f); float3 Pend = ray->P + ray->D*ray->t; int bounce = state->transparent_bounce; # ifdef __VOLUME__ -# ifdef __SPLIT_KERNEL__ - ccl_addr_space PathState *ps = &kernel_split_state.state_shadow[ccl_global_id(1) * ccl_global_size(0) + ccl_global_id(0)]; -# else - PathState ps_object; - PathState *ps = &ps_object; -# endif - *ps = *state; + PathState *ps = shadow_blocked_volume_path_state(kg, + &volume_state, + state, + sd, + ray); # endif for(;;) { if(bounce >= kernel_data.integrator.transparent_max_bounce) { @@ -345,7 +389,12 @@ ccl_device bool shadow_blocked_transparent_stepped_loop( # ifdef __VOLUME__ if(!blocked && state->volume_stack[0].shader != SHADER_NONE) { /* Apply attenuation from current volume shader. */ - kernel_volume_shadow(kg, shadow_sd, state, ray, shadow); + PathState *ps = shadow_blocked_volume_path_state(kg, + &volume_state, + state, + sd, + ray); + kernel_volume_shadow(kg, shadow_sd, ps, ray, shadow); } # endif return blocked; @@ -353,6 +402,7 @@ ccl_device bool shadow_blocked_transparent_stepped_loop( ccl_device bool shadow_blocked_transparent_stepped( KernelGlobals *kg, + ShaderData *sd, ShaderData *shadow_sd, ccl_addr_space PathState *state, const uint visibility, @@ -370,6 +420,7 @@ ccl_device bool shadow_blocked_transparent_stepped( ? shader_transparent_shadow(kg, isect) : false; return shadow_blocked_transparent_stepped_loop(kg, + sd, shadow_sd, state, visibility, @@ -384,6 +435,7 @@ ccl_device bool shadow_blocked_transparent_stepped( #endif /* __TRANSPARENT_SHADOWS__ */ ccl_device_inline bool shadow_blocked(KernelGlobals *kg, + ShaderData *sd, ShaderData *shadow_sd, ccl_addr_space PathState *state, Ray *ray_input, @@ -452,6 +504,7 @@ ccl_device_inline bool shadow_blocked(KernelGlobals *kg, max_hits + 1 >= SHADOW_STACK_MAX_HITS) { return shadow_blocked_transparent_stepped_loop(kg, + sd, shadow_sd, state, visibility, @@ -463,6 +516,7 @@ ccl_device_inline bool shadow_blocked(KernelGlobals *kg, } # endif /* __KERNEL_GPU__ */ return shadow_blocked_transparent_all(kg, + sd, shadow_sd, state, visibility, diff --git a/intern/cycles/kernel/split/kernel_shadow_blocked_dl.h b/intern/cycles/kernel/split/kernel_shadow_blocked_dl.h index 19bfee6d039..b52f9a5eb81 100644 --- a/intern/cycles/kernel/split/kernel_shadow_blocked_dl.h +++ b/intern/cycles/kernel/split/kernel_shadow_blocked_dl.h @@ -89,6 +89,7 @@ ccl_device void kernel_shadow_blocked_dl(KernelGlobals *kg) float3 shadow; if(!shadow_blocked(kg, + sd, emission_sd, state, &ray, diff --git a/source/blender/blenkernel/intern/pointcache.c b/source/blender/blenkernel/intern/pointcache.c index 341a4d1fde5..5ec3156f4df 100644 --- a/source/blender/blenkernel/intern/pointcache.c +++ b/source/blender/blenkernel/intern/pointcache.c @@ -3619,7 +3619,13 @@ void BKE_ptcache_bake(PTCacheBaker *baker) psys_get_pointcache_start_end(scene, pid->calldata, &cache->startframe, &cache->endframe); } - if (((cache->flag & PTCACHE_BAKED) == 0) && (render || bake)) { + // XXX workaround for regression inroduced in ee3fadd, needs looking into + if (pid->type == PTCACHE_TYPE_RIGIDBODY) { + if ((cache->flag & PTCACHE_REDO_NEEDED || (cache->flag & PTCACHE_SIMULATION_VALID)==0) && (render || bake)) { + BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0); + } + } + else if (((cache->flag & PTCACHE_BAKED) == 0) && (render || bake)) { BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0); } diff --git a/source/blender/blenkernel/intern/rigidbody.c b/source/blender/blenkernel/intern/rigidbody.c index df048ce5036..f75b14579c7 100644 --- a/source/blender/blenkernel/intern/rigidbody.c +++ b/source/blender/blenkernel/intern/rigidbody.c @@ -1595,8 +1595,12 @@ void BKE_rigidbody_do_simulation(const struct EvaluationContext *eval_ctx, Scene BKE_ptcache_id_time(&pid, scene, ctime, &startframe, &endframe, NULL); cache = rbw->pointcache; + if (ctime <= startframe) { + rbw->ltime = startframe; + return; + } /* make sure we don't go out of cache frame range */ - if (ctime > endframe) { + else if (ctime > endframe) { ctime = endframe; } @@ -1610,12 +1614,9 @@ void BKE_rigidbody_do_simulation(const struct EvaluationContext *eval_ctx, Scene // RB_TODO deal with interpolated, old and baked results bool can_simulate = (ctime == rbw->ltime + 1) && !(cache->flag & PTCACHE_BAKED); - if (cache->flag & PTCACHE_OUTDATED || cache->last_exact == 0) { - rbw->ltime = cache->startframe; - } - - if (BKE_ptcache_read(&pid, ctime, can_simulate)) { + if (BKE_ptcache_read(&pid, ctime, can_simulate) == PTCACHE_READ_EXACT) { BKE_ptcache_validate(cache, (int)ctime); + rbw->ltime = ctime; return; } diff --git a/source/blender/editors/mesh/editmesh_knife.c b/source/blender/editors/mesh/editmesh_knife.c index 5aa10620936..433163b2fab 100644 --- a/source/blender/editors/mesh/editmesh_knife.c +++ b/source/blender/editors/mesh/editmesh_knife.c @@ -1229,6 +1229,7 @@ static bool knife_ray_intersect_face( for (; tri_i < tottri; tri_i++) { const float *lv1, *lv2, *lv3; + float ray_tri_uv[2]; tri = kcd->em->looptris[tri_i]; if (tri[0]->f != f) @@ -1240,7 +1241,7 @@ static bool knife_ray_intersect_face( * tesselation edge and might not hit either tesselation tri with * an exact test; * we will exclude hits near real edges by a later test */ - if (isect_ray_tri_epsilon_v3(v1, raydir, lv1, lv2, lv3, &lambda, NULL, KNIFE_FLT_EPS)) { + if (isect_ray_tri_epsilon_v3(v1, raydir, lv1, lv2, lv3, &lambda, ray_tri_uv, KNIFE_FLT_EPS)) { /* check if line coplanar with tri */ normal_tri_v3(tri_norm, lv1, lv2, lv3); plane_from_point_normal_v3(tri_plane, lv1, tri_norm); @@ -1249,8 +1250,7 @@ static bool knife_ray_intersect_face( { return false; } - copy_v3_v3(hit_cageco, v1); - madd_v3_v3fl(hit_cageco, raydir, lambda); + interp_v3_v3v3v3_uv(hit_cageco, lv1, lv2, lv3, ray_tri_uv); /* Now check that far enough away from verts and edges */ lst = knife_get_face_kedges(kcd, f); for (ref = lst->first; ref; ref = ref->next) { @@ -1262,11 +1262,7 @@ static bool knife_ray_intersect_face( return false; } } - - transform_point_by_tri_v3( - hit_co, hit_cageco, - tri[0]->v->co, tri[1]->v->co, tri[2]->v->co, - lv1, lv2, lv3); + interp_v3_v3v3v3_uv(hit_co, tri[0]->v->co, tri[1]->v->co, tri[2]->v->co, ray_tri_uv); return true; } } diff --git a/source/blender/gpu/intern/gpu_draw.c b/source/blender/gpu/intern/gpu_draw.c index 601de2249f5..1ab1cba2504 100644 --- a/source/blender/gpu/intern/gpu_draw.c +++ b/source/blender/gpu/intern/gpu_draw.c @@ -579,10 +579,10 @@ int GPU_verify_image( if (do_color_management) { srgb_frect = MEM_mallocN(ibuf->x * ibuf->y * sizeof(float) * 4, "floar_buf_col_cor"); gpu_verify_high_bit_srgb_buffer(srgb_frect, ibuf); - frect = srgb_frect + texwinsy * ibuf->x + texwinsx; + frect = srgb_frect + (4 * (texwinsy * ibuf->x + texwinsx)); } else { - frect = ibuf->rect_float + texwinsy * ibuf->x + texwinsx; + frect = ibuf->rect_float + (ibuf->channels * (texwinsy * ibuf->x + texwinsx)); } } else { |