kolibrios-fun/drivers/video/Gallium/auxiliary/vl/vl_vertex_buffers.c
Sergey Semyonov (Serge) edb28b33f3 Gallium 3D: initial commit
git-svn-id: svn://kolibrios.org@3770 a494cfbc-eb01-0410-851d-a64ba20cac60
2013-07-06 10:06:47 +00:00

379 lines
9.0 KiB
C

/**************************************************************************
*
* Copyright 2010 Christian König
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sub license, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
* IN NO EVENT SHALL TUNGSTEN GRAPHICS AND/OR ITS SUPPLIERS BE LIABLE FOR
* ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
**************************************************************************/
#include <assert.h>
#include "util/u_format.h"
#include "vl_vertex_buffers.h"
#include "vl_types.h"
/* vertices for a quad covering a block */
static const struct vertex2f block_quad[4] = {
{0.0f, 0.0f}, {1.0f, 0.0f}, {1.0f, 1.0f}, {0.0f, 1.0f}
};
struct pipe_vertex_buffer
vl_vb_upload_quads(struct pipe_context *pipe)
{
struct pipe_vertex_buffer quad;
struct pipe_transfer *buf_transfer;
struct vertex2f *v;
unsigned i;
assert(pipe);
/* create buffer */
quad.stride = sizeof(struct vertex2f);
quad.buffer_offset = 0;
quad.buffer = pipe_buffer_create
(
pipe->screen,
PIPE_BIND_VERTEX_BUFFER,
PIPE_USAGE_STATIC,
sizeof(struct vertex2f) * 4
);
quad.user_buffer = NULL;
if(!quad.buffer)
return quad;
/* and fill it */
v = pipe_buffer_map
(
pipe,
quad.buffer,
PIPE_TRANSFER_WRITE | PIPE_TRANSFER_DISCARD_RANGE,
&buf_transfer
);
for (i = 0; i < 4; ++i, ++v) {
v->x = block_quad[i].x;
v->y = block_quad[i].y;
}
pipe_buffer_unmap(pipe, buf_transfer);
return quad;
}
struct pipe_vertex_buffer
vl_vb_upload_pos(struct pipe_context *pipe, unsigned width, unsigned height)
{
struct pipe_vertex_buffer pos;
struct pipe_transfer *buf_transfer;
struct vertex2s *v;
unsigned x, y;
assert(pipe);
/* create buffer */
pos.stride = sizeof(struct vertex2s);
pos.buffer_offset = 0;
pos.buffer = pipe_buffer_create
(
pipe->screen,
PIPE_BIND_VERTEX_BUFFER,
PIPE_USAGE_STATIC,
sizeof(struct vertex2s) * width * height
);
pos.user_buffer = NULL;
if(!pos.buffer)
return pos;
/* and fill it */
v = pipe_buffer_map
(
pipe,
pos.buffer,
PIPE_TRANSFER_WRITE | PIPE_TRANSFER_DISCARD_RANGE,
&buf_transfer
);
for ( y = 0; y < height; ++y) {
for ( x = 0; x < width; ++x, ++v) {
v->x = x;
v->y = y;
}
}
pipe_buffer_unmap(pipe, buf_transfer);
return pos;
}
static struct pipe_vertex_element
vl_vb_get_quad_vertex_element(void)
{
struct pipe_vertex_element element;
/* setup rectangle element */
element.src_offset = 0;
element.instance_divisor = 0;
element.vertex_buffer_index = 0;
element.src_format = PIPE_FORMAT_R32G32_FLOAT;
return element;
}
static void
vl_vb_element_helper(struct pipe_vertex_element* elements, unsigned num_elements,
unsigned vertex_buffer_index)
{
unsigned i, offset = 0;
assert(elements && num_elements);
for ( i = 0; i < num_elements; ++i ) {
elements[i].src_offset = offset;
elements[i].instance_divisor = 1;
elements[i].vertex_buffer_index = vertex_buffer_index;
offset += util_format_get_blocksize(elements[i].src_format);
}
}
void *
vl_vb_get_ves_ycbcr(struct pipe_context *pipe)
{
struct pipe_vertex_element vertex_elems[NUM_VS_INPUTS];
assert(pipe);
memset(&vertex_elems, 0, sizeof(vertex_elems));
vertex_elems[VS_I_RECT] = vl_vb_get_quad_vertex_element();
/* Position element */
vertex_elems[VS_I_VPOS].src_format = PIPE_FORMAT_R8G8B8A8_USCALED;
/* block num element */
vertex_elems[VS_I_BLOCK_NUM].src_format = PIPE_FORMAT_R32_FLOAT;
vl_vb_element_helper(&vertex_elems[VS_I_VPOS], 2, 1);
return pipe->create_vertex_elements_state(pipe, 3, vertex_elems);
}
void *
vl_vb_get_ves_mv(struct pipe_context *pipe)
{
struct pipe_vertex_element vertex_elems[NUM_VS_INPUTS];
assert(pipe);
memset(&vertex_elems, 0, sizeof(vertex_elems));
vertex_elems[VS_I_RECT] = vl_vb_get_quad_vertex_element();
/* Position element */
vertex_elems[VS_I_VPOS].src_format = PIPE_FORMAT_R16G16_SSCALED;
vl_vb_element_helper(&vertex_elems[VS_I_VPOS], 1, 1);
/* motion vector TOP element */
vertex_elems[VS_I_MV_TOP].src_format = PIPE_FORMAT_R16G16B16A16_SSCALED;
/* motion vector BOTTOM element */
vertex_elems[VS_I_MV_BOTTOM].src_format = PIPE_FORMAT_R16G16B16A16_SSCALED;
vl_vb_element_helper(&vertex_elems[VS_I_MV_TOP], 2, 2);
return pipe->create_vertex_elements_state(pipe, NUM_VS_INPUTS, vertex_elems);
}
bool
vl_vb_init(struct vl_vertex_buffer *buffer, struct pipe_context *pipe,
unsigned width, unsigned height)
{
unsigned i, size;
assert(buffer);
buffer->width = width;
buffer->height = height;
size = width * height;
for (i = 0; i < VL_NUM_COMPONENTS; ++i) {
buffer->ycbcr[i].resource = pipe_buffer_create
(
pipe->screen,
PIPE_BIND_VERTEX_BUFFER,
PIPE_USAGE_STREAM,
sizeof(struct vl_ycbcr_block) * size * 4
);
if (!buffer->ycbcr[i].resource)
goto error_ycbcr;
}
for (i = 0; i < VL_MAX_REF_FRAMES; ++i) {
buffer->mv[i].resource = pipe_buffer_create
(
pipe->screen,
PIPE_BIND_VERTEX_BUFFER,
PIPE_USAGE_STREAM,
sizeof(struct vl_motionvector) * size
);
if (!buffer->mv[i].resource)
goto error_mv;
}
vl_vb_map(buffer, pipe);
return true;
error_mv:
for (i = 0; i < VL_NUM_COMPONENTS; ++i)
pipe_resource_reference(&buffer->mv[i].resource, NULL);
error_ycbcr:
for (i = 0; i < VL_NUM_COMPONENTS; ++i)
pipe_resource_reference(&buffer->ycbcr[i].resource, NULL);
return false;
}
unsigned
vl_vb_attributes_per_plock(struct vl_vertex_buffer *buffer)
{
return 1;
}
struct pipe_vertex_buffer
vl_vb_get_ycbcr(struct vl_vertex_buffer *buffer, int component)
{
struct pipe_vertex_buffer buf;
assert(buffer);
buf.stride = sizeof(struct vl_ycbcr_block);
buf.buffer_offset = 0;
buf.buffer = buffer->ycbcr[component].resource;
buf.user_buffer = NULL;
return buf;
}
struct pipe_vertex_buffer
vl_vb_get_mv(struct vl_vertex_buffer *buffer, int motionvector)
{
struct pipe_vertex_buffer buf;
assert(buffer);
buf.stride = sizeof(struct vl_motionvector);
buf.buffer_offset = 0;
buf.buffer = buffer->mv[motionvector].resource;
buf.user_buffer = NULL;
return buf;
}
void
vl_vb_map(struct vl_vertex_buffer *buffer, struct pipe_context *pipe)
{
unsigned i;
assert(buffer && pipe);
for (i = 0; i < VL_NUM_COMPONENTS; ++i) {
buffer->ycbcr[i].vertex_stream = pipe_buffer_map
(
pipe,
buffer->ycbcr[i].resource,
PIPE_TRANSFER_WRITE | PIPE_TRANSFER_DISCARD_RANGE,
&buffer->ycbcr[i].transfer
);
}
for (i = 0; i < VL_MAX_REF_FRAMES; ++i) {
buffer->mv[i].vertex_stream = pipe_buffer_map
(
pipe,
buffer->mv[i].resource,
PIPE_TRANSFER_WRITE | PIPE_TRANSFER_DISCARD_RANGE,
&buffer->mv[i].transfer
);
}
}
struct vl_ycbcr_block *
vl_vb_get_ycbcr_stream(struct vl_vertex_buffer *buffer, int component)
{
assert(buffer);
assert(component < VL_NUM_COMPONENTS);
return buffer->ycbcr[component].vertex_stream;
}
unsigned
vl_vb_get_mv_stream_stride(struct vl_vertex_buffer *buffer)
{
assert(buffer);
return buffer->width;
}
struct vl_motionvector *
vl_vb_get_mv_stream(struct vl_vertex_buffer *buffer, int ref_frame)
{
assert(buffer);
assert(ref_frame < VL_MAX_REF_FRAMES);
return buffer->mv[ref_frame].vertex_stream;
}
void
vl_vb_unmap(struct vl_vertex_buffer *buffer, struct pipe_context *pipe)
{
unsigned i;
assert(buffer && pipe);
for (i = 0; i < VL_NUM_COMPONENTS; ++i) {
pipe_buffer_unmap(pipe, buffer->ycbcr[i].transfer);
}
for (i = 0; i < VL_MAX_REF_FRAMES; ++i) {
pipe_buffer_unmap(pipe, buffer->mv[i].transfer);
}
}
void
vl_vb_cleanup(struct vl_vertex_buffer *buffer)
{
unsigned i;
assert(buffer);
for (i = 0; i < VL_NUM_COMPONENTS; ++i) {
pipe_resource_reference(&buffer->ycbcr[i].resource, NULL);
}
for (i = 0; i < VL_MAX_REF_FRAMES; ++i) {
pipe_resource_reference(&buffer->mv[i].resource, NULL);
}
}