#include "os/os_time.h"
#include "lp_scene_queue.h"
+#include "lp_context.h"
#include "lp_debug.h"
#include "lp_fence.h"
#include "lp_perf.h"
lp_rast_begin( struct lp_rasterizer *rast,
struct lp_scene *scene )
{
-
rast->curr_scene = scene;
LP_DBG(DEBUG_RAST, "%s\n", __FUNCTION__);
task->height = TILE_SIZE + y * TILE_SIZE > task->scene->fb.height ?
task->scene->fb.height - y * TILE_SIZE : TILE_SIZE;
+ task->thread_data.vis_counter = 0;
+ task->ps_invocations = 0;
+
/* reset pointers to color and depth tile(s) */
memset(task->color_tiles, 0, sizeof(task->color_tiles));
task->depth_tile = NULL;
}
+/**
+ * Examine a framebuffer object to determine if any of the colorbuffers
+ * use a pure integer format.
+ * XXX this could be a gallium utility function if useful elsewhere.
+ */
+static boolean
+is_fb_pure_integer(const struct pipe_framebuffer_state *fb)
+{
+ unsigned i;
+ for (i = 0; i < fb->nr_cbufs; i++) {
+ if (fb->cbufs[i] &&
+ util_format_is_pure_integer(fb->cbufs[i]->format)) {
+ return TRUE;
+ }
+ }
+ return FALSE;
+}
+
+
/**
* Clear the rasterizer's current color tile.
* This is a bin command called during bin processing.
unsigned i;
union util_color uc;
- if (util_format_is_pure_integer(scene->fb.cbufs[0]->format)) {
+ if (is_fb_pure_integer(&scene->fb)) {
/*
* We expect int/uint clear values here, though some APIs
* might disagree (but in any case util_pack_color()
for (i = 0; i < scene->fb.nr_cbufs; i++) {
enum pipe_format format = scene->fb.cbufs[i]->format;
- unsigned layer;
- uint8_t *map_layer = scene->cbufs[i].map;
if (util_format_is_pure_sint(format)) {
util_format_write_4i(format, arg.clear_color.i, 0, &uc, 0, 0, 0, 1, 1);
util_format_write_4ui(format, arg.clear_color.ui, 0, &uc, 0, 0, 0, 1, 1);
}
- for (layer = 0; layer <= scene->fb_max_layer; layer++) {
- util_fill_rect(map_layer,
- scene->fb.cbufs[i]->format,
- scene->cbufs[i].stride,
- task->x,
- task->y,
- task->width,
- task->height,
- &uc);
- map_layer += scene->cbufs[i].layer_stride;
- }
+ util_fill_box(scene->cbufs[i].map,
+ format,
+ scene->cbufs[i].stride,
+ scene->cbufs[i].layer_stride,
+ task->x,
+ task->y,
+ 0,
+ task->width,
+ task->height,
+ scene->fb_max_layer + 1,
+ &uc);
}
}
else {
clear_color[3]);
for (i = 0; i < scene->fb.nr_cbufs; i++) {
- unsigned layer;
- uint8_t *map_layer = scene->cbufs[i].map;
-
- for (layer = 0; layer <= scene->fb_max_layer; layer++) {
+ if (scene->fb.cbufs[i]) {
util_pack_color(arg.clear_color.f,
scene->fb.cbufs[i]->format, &uc);
- util_fill_rect(map_layer,
- scene->fb.cbufs[i]->format,
- scene->cbufs[i].stride,
- task->x,
- task->y,
- task->width,
- task->height,
- &uc);
- map_layer += scene->cbufs[i].layer_stride;
+
+ util_fill_box(scene->cbufs[i].map,
+ scene->fb.cbufs[i]->format,
+ scene->cbufs[i].stride,
+ scene->cbufs[i].layer_stride,
+ task->x,
+ task->y,
+ 0,
+ task->width,
+ task->height,
+ scene->fb_max_layer + 1,
+ &uc);
}
}
}
depth_stride = scene->zsbuf.stride;
}
+ /* Propagate non-interpolated raster state. */
+ task->thread_data.raster_state.viewport_index = inputs->viewport_index;
+
/* run shader on 4x4 block */
BEGIN_JIT_CALL(state, task);
variant->jit_function[RAST_WHOLE]( &state->jit_context,
/* color buffer */
for (i = 0; i < scene->fb.nr_cbufs; i++) {
- stride[i] = scene->cbufs[i].stride;
- color[i] = lp_rast_get_unswizzled_color_block_pointer(task, i, x, y, inputs->layer);
+ if (scene->fb.cbufs[i]) {
+ stride[i] = scene->cbufs[i].stride;
+ color[i] = lp_rast_get_unswizzled_color_block_pointer(task, i, x, y,
+ inputs->layer);
+ }
+ else {
+ stride[i] = 0;
+ color[i] = NULL;
+ }
}
/* depth buffer */
* allocated 4x4 blocks hence need to filter them out here.
*/
if ((x % TILE_SIZE) < task->width && (y % TILE_SIZE) < task->height) {
+ /* not very accurate would need a popcount on the mask */
+ /* always count this not worth bothering? */
+ task->ps_invocations += 1 * variant->ps_inv_multiplier;
+
+ /* Propagate non-interpolated raster state. */
+ task->thread_data.raster_state.viewport_index = inputs->viewport_index;
+
/* run shader on 4x4 block */
BEGIN_JIT_CALL(state, task);
variant->jit_function[RAST_EDGE_TEST](&state->jit_context,
{
struct llvmpipe_query *pq = arg.query_obj;
- assert(task->query[pq->type] == NULL);
-
switch (pq->type) {
case PIPE_QUERY_OCCLUSION_COUNTER:
- task->thread_data.vis_counter = 0;
+ case PIPE_QUERY_OCCLUSION_PREDICATE:
+ pq->start[task->thread_index] = task->thread_data.vis_counter;
break;
- case PIPE_QUERY_PRIMITIVES_GENERATED:
- case PIPE_QUERY_PRIMITIVES_EMITTED:
- case PIPE_QUERY_SO_STATISTICS:
case PIPE_QUERY_PIPELINE_STATISTICS:
- case PIPE_QUERY_SO_OVERFLOW_PREDICATE:
+ pq->start[task->thread_index] = task->ps_invocations;
break;
default:
assert(0);
break;
}
-
- task->query[pq->type] = pq;
}
const union lp_rast_cmd_arg arg)
{
struct llvmpipe_query *pq = arg.query_obj;
- assert(task->query[pq->type] == pq || pq->type == PIPE_QUERY_TIMESTAMP);
switch (pq->type) {
case PIPE_QUERY_OCCLUSION_COUNTER:
- pq->count[task->thread_index] += task->thread_data.vis_counter;
+ case PIPE_QUERY_OCCLUSION_PREDICATE:
+ pq->end[task->thread_index] +=
+ task->thread_data.vis_counter - pq->start[task->thread_index];
+ pq->start[task->thread_index] = 0;
break;
case PIPE_QUERY_TIMESTAMP:
- pq->count[task->thread_index] = os_time_get_nano();
+ pq->end[task->thread_index] = os_time_get_nano();
break;
- case PIPE_QUERY_PRIMITIVES_GENERATED:
- case PIPE_QUERY_PRIMITIVES_EMITTED:
- case PIPE_QUERY_SO_STATISTICS:
case PIPE_QUERY_PIPELINE_STATISTICS:
- case PIPE_QUERY_SO_OVERFLOW_PREDICATE:
+ pq->end[task->thread_index] +=
+ task->ps_invocations - pq->start[task->thread_index];
+ pq->start[task->thread_index] = 0;
break;
default:
assert(0);
break;
}
-
- if (task->query[pq->type] == pq) {
- task->query[pq->type] = NULL;
- }
}
{
unsigned i;
- for (i = 0; i < PIPE_QUERY_TYPES; ++i) {
- if (task->query[i]) {
- lp_rast_end_query(task, lp_rast_arg_query(task->query[i]));
- }
+ for (i = 0; i < task->scene->num_active_queries; ++i) {
+ lp_rast_end_query(task, lp_rast_arg_query(task->scene->active_queries[i]));
}
/* debug */
lp_rast_begin_query,
lp_rast_end_query,
lp_rast_set_state,
+ lp_rast_triangle_32_1,
+ lp_rast_triangle_32_2,
+ lp_rast_triangle_32_3,
+ lp_rast_triangle_32_4,
+ lp_rast_triangle_32_5,
+ lp_rast_triangle_32_6,
+ lp_rast_triangle_32_7,
+ lp_rast_triangle_32_8,
+ lp_rast_triangle_32_3_4,
+ lp_rast_triangle_32_3_16,
+ lp_rast_triangle_32_4_16
};
if (rast->num_threads == 0) {
/* no threading */
+ unsigned fpstate = util_fpstate_get();
+
+ /* Make sure that denorms are treated like zeros. This is
+ * the behavior required by D3D10. OpenGL doesn't care.
+ */
+ util_fpstate_set_denorms_to_zero(fpstate);
lp_rast_begin( rast, scene );
lp_rast_end( rast );
+ util_fpstate_set(fpstate);
+
rast->curr_scene = NULL;
}
else {
struct lp_rasterizer_task *task = (struct lp_rasterizer_task *) init_data;
struct lp_rasterizer *rast = task->rast;
boolean debug = false;
+ unsigned fpstate = util_fpstate_get();
+
+ /* Make sure that denorms are treated like zeros. This is
+ * the behavior required by D3D10. OpenGL doesn't care.
+ */
+ util_fpstate_set_denorms_to_zero(fpstate);
while (1) {
/* wait for work */
pipe_semaphore_signal(&task->work_done);
}
- return NULL;
+ return 0;
}
}
-/** Return number of rasterization threads */
-unsigned
-lp_rast_get_num_threads( struct lp_rasterizer *rast )
-{
- return rast->num_threads;
-}
-
-
projects / mesa.git / blobdiff