OpenHarmony/third_party_mesa3d
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Another vertex program checkpoint: clean-up of vertex attribute storage

Browse Source 
in vertex_buffer.   Improved vertex program pipeline stage such that
output registers can be processed in a loop.  Getting closer to where
we need to be in order to implement performance optimizations...
This commit is contained in:
Brian Paul
2002-01-06 03:54:12 +00:00
parent e2053d1690
commit 363049160d
10 changed files with 108 additions and 157 deletions
Download Patch File Download Diff File Expand all files Collapse all files

140
src/mesa/tnl/t_vb_program.c
View File

@@ -1,4 +1,4 @@
/* $Id: t_vb_program.c,v 1.6 2002/01/05 20:51:13 brianp Exp $ */
/* $Id: t_vb_program.c,v 1.7 2002/01/06 03:54:12 brianp Exp $ */
/*
* Mesa 3-D graphics library
@@ -98,16 +98,36 @@
#include "t_imm_exec.h"
/* WARNING: these values _MUST_ match the values in the OutputRegisters[]
* array in vpparse.c!!!
*/
#define VERT_RESULT_HPOS 0
#define VERT_RESULT_COL0 1
#define VERT_RESULT_COL1 2
#define VERT_RESULT_BFC0 3
#define VERT_RESULT_BFC1 4
#define VERT_RESULT_FOGC 5
#define VERT_RESULT_PSIZ 6
#define VERT_RESULT_TEX0 7
#define VERT_RESULT_TEX1 8
#define VERT_RESULT_TEX2 9
#define VERT_RESULT_TEX3 10
#define VERT_RESULT_TEX4 11
#define VERT_RESULT_TEX5 12
#define VERT_RESULT_TEX6 13
#define VERT_RESULT_TEX7 14
struct vp_stage_data {
GLvector4f clipCoords; /* post-modelview/projection coords */
GLvector4f ndcCoords; /* normalized device coords */
/* The results of running the vertex program go into these arrays. */
GLvector4f attribs[15];
/* These point to the attribs[VERT_RESULT_COL0, COL1, BFC0, BFC1] arrays */
struct gl_client_array color0[2]; /* front and back */
struct gl_client_array color1[2]; /* front and back */
GLvector4f texCoord[MAX_TEXTURE_UNITS];
GLvector4f fogCoord;
GLvector1f pointSize;
GLubyte *clipmask;
GLvector4f ndcCoords; /* normalized device coords */
GLubyte *clipmask; /* clip flags */
GLubyte ormask, andmask;
};
@@ -123,21 +143,8 @@ static GLboolean run_vp( GLcontext *ctx, struct gl_pipeline_stage *stage )
struct vp_machine *machine = &(ctx->VertexProgram.Machine);
GLint i;
/* convenience pointers */
GLfloat (*clip)[4] = (GLfloat (*)[4]) store->clipCoords.data;
GLfloat (*color0)[4] = (GLfloat (*)[4]) store->color0[0].Ptr;
GLfloat (*color1)[4] = (GLfloat (*)[4]) store->color1[0].Ptr;
GLfloat (*bfcolor0)[4] = (GLfloat (*)[4]) store->color0[1].Ptr;
GLfloat (*bfcolor1)[4] = (GLfloat (*)[4]) store->color1[1].Ptr;
GLfloat (*fog)[4] = (GLfloat (*)[4]) store->fogCoord.data;
GLfloat *pointSize = (GLfloat *) store->pointSize.data;
GLfloat (*texture0)[4] = (GLfloat (*)[4]) store->texCoord[0].data;
GLfloat (*texture1)[4] = (GLfloat (*)[4]) store->texCoord[1].data;
GLfloat (*texture2)[4] = (GLfloat (*)[4]) store->texCoord[2].data;
GLfloat (*texture3)[4] = (GLfloat (*)[4]) store->texCoord[3].data;
_mesa_init_tracked_matrices(ctx);
_mesa_init_vp_registers(ctx); /* sets temp regs to (0,0,0,1) */
_mesa_init_vp_registers(ctx); /* init temp and result regs */
for (i = 0; i < VB->Count; i++) {
GLuint attr;
@@ -185,33 +192,28 @@ static GLboolean run_vp( GLcontext *ctx, struct gl_pipeline_stage *stage )
machine->Registers[VP_OUT_COL0][3]);
#endif
/* store the attribute output registers into the VB arrays */
COPY_4V(clip[i], machine->Registers[VP_OUT_HPOS]);
COPY_4V(color0[i], machine->Registers[VP_OUT_COL0]);
COPY_4V(color1[i], machine->Registers[VP_OUT_COL1]);
COPY_4V(bfcolor0[i], machine->Registers[VP_OUT_BFC0]);
COPY_4V(bfcolor1[i], machine->Registers[VP_OUT_BFC1]);
fog[i][0] = machine->Registers[VP_OUT_FOGC][0];
pointSize[i] = machine->Registers[VP_OUT_PSIZ][0];
COPY_4V(texture0[i], machine->Registers[VP_OUT_TEX0]);
COPY_4V(texture1[i], machine->Registers[VP_OUT_TEX0]);
COPY_4V(texture2[i], machine->Registers[VP_OUT_TEX0]);
COPY_4V(texture3[i], machine->Registers[VP_OUT_TEX0]);
/* copy the output registers into the VB->attribs arrays */
/* XXX (optimize) could use a conditional and smaller loop limit here */
for (attr = 0; attr < 15; attr++) {
COPY_4V( store->attribs[attr].data[i],
machine->Registers[VP_OUTPUT_REG_START + attr] );
}
}
VB->ClipPtr = &store->clipCoords;
/* Setup the VB pointers so that the next pipeline stages get
* their data from the right place (the program output arrays).
*/
VB->ClipPtr = &store->attribs[VERT_RESULT_HPOS];
VB->ClipPtr->size = 4;
VB->ClipPtr->count = VB->Count;
VB->ColorPtr[0] = &store->color0[0];
VB->ColorPtr[1] = &store->color0[1];
VB->SecondaryColorPtr[0] = &store->color1[0];
VB->SecondaryColorPtr[1] = &store->color1[1];
VB->AttribPtr[VERT_ATTRIB_FOG] = &store->fogCoord;
VB->PointSizePtr = &store->pointSize;
VB->TexCoordPtr[0] = &store->texCoord[0];
VB->TexCoordPtr[1] = &store->texCoord[1];
VB->TexCoordPtr[2] = &store->texCoord[2];
VB->TexCoordPtr[3] = &store->texCoord[3];
VB->FogCoordPtr = &store->attribs[VERT_RESULT_FOGC];
VB->PointSizePtr = &store->attribs[VERT_RESULT_PSIZ];
for (i = 0; i < ctx->Const.MaxTextureUnits; i++)
VB->TexCoordPtr[i] = &store->attribs[VERT_RESULT_TEX0 + i];
/* Cliptest and perspective divide. Clip functions must clear
* the clipmask.
@@ -304,23 +306,9 @@ static GLboolean run_validate_program( GLcontext *ctx,
}
#if 0
static void alloc_4chan( struct gl_client_array *a, GLuint sz )
static void init_client_array( struct gl_client_array *a, GLvector4f *vec )
{
a->Ptr = ALIGN_MALLOC( sz * sizeof(GLchan) * 4, 32 );
a->Size = 4;
a->Type = CHAN_TYPE;
a->Stride = 0;
a->StrideB = sizeof(GLchan) * 4;
a->Enabled = 0;
a->Flags = 0;
}
#endif
static void alloc_4float( struct gl_client_array *a, GLuint sz )
{
a->Ptr = ALIGN_MALLOC( sz * sizeof(GLfloat) * 4, 32 );
a->Ptr = vec->data;
a->Size = 4;
a->Type = GL_FLOAT;
a->Stride = 0;
@@ -347,19 +335,19 @@ static GLboolean run_init_vp( GLcontext *ctx,
if (!store)
return GL_FALSE;
/* The output of a vertex program is: */
_mesa_vector4f_alloc( &store->clipCoords, 0, size, 32 );
_mesa_vector4f_alloc( &store->ndcCoords, 0, size, 32 );
alloc_4float( &store->color0[0], size );
alloc_4float( &store->color0[1], size );
alloc_4float( &store->color1[0], size );
alloc_4float( &store->color1[1], size );
for (i = 0 ; i < ctx->Const.MaxTextureUnits ; i++)
_mesa_vector4f_alloc( &store->texCoord[i], 0, VB->Size, 32 );
_mesa_vector4f_alloc( &store->fogCoord, 0, size, 32 );
_mesa_vector1f_alloc( &store->pointSize, 0, size, 32 );
store->clipmask = (GLubyte *) ALIGN_MALLOC(sizeof(GLubyte)*size, 32 );
/* Allocate arrays of vertex output values */
for (i = 0; i < 15; i++)
_mesa_vector4f_alloc( &store->attribs[i], 0, size, 32 );
/* Make the color0[] and color1[] arrays point into the attribs[] arrays */
init_client_array( &store->color0[0], &store->attribs[VERT_RESULT_COL0] );
init_client_array( &store->color0[1], &store->attribs[VERT_RESULT_COL1] );
init_client_array( &store->color1[0], &store->attribs[VERT_RESULT_BFC0] );
init_client_array( &store->color1[1], &store->attribs[VERT_RESULT_BFC1] );
/* a few other misc allocations */
_mesa_vector4f_alloc( &store->ndcCoords, 0, size, 32 );
store->clipmask = (GLubyte *) ALIGN_MALLOC(sizeof(GLubyte)*size, 32 );
/* Now validate the stage derived data...
*/
@@ -401,17 +389,13 @@ static void dtr( struct gl_pipeline_stage *stage )
if (store) {
GLuint i;
_mesa_vector4f_free( &store->clipCoords );
/* free the vertex program result arrays */
for (i = 0; i < 15; i++)
_mesa_vector4f_free( &store->attribs[i] );
/* free misc arrays */
_mesa_vector4f_free( &store->ndcCoords );
ALIGN_FREE( store->color0[0].Ptr );
ALIGN_FREE( store->color0[1].Ptr );
ALIGN_FREE( store->color1[0].Ptr );
ALIGN_FREE( store->color1[1].Ptr );
for (i = 0 ; i < MAX_TEXTURE_UNITS ; i++)
if (store->texCoord[i].data)
_mesa_vector4f_free( &store->texCoord[i] );
_mesa_vector4f_free( &store->fogCoord );
_mesa_vector1f_free( &store->pointSize );
ALIGN_FREE( store->clipmask );
FREE( store );