Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
tree: ab7291d368
Fetching contributors…

Cannot retrieve contributors at this time

file 439 lines (382 sloc) 12.738 kb
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439
/*
* Copyright 2007-2008 Maarten Maathuis
* Copyright 2008 Stephane Marchesin
*
* 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, sublicense,
* 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 NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS 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.
*/

#ifdef HAVE_CONFIG_H
#include "config.h"
#endif

#include "xf86xv.h"
#include <X11/extensions/Xv.h>
#include "exa.h"
#include "damage.h"
#include "dixstruct.h"
#include "fourcc.h"

#include "nv_include.h"
#include "nv_dma.h"

#include "nv30_shaders.h"

#include "hwdefs/nv30-40_3d.xml.h"

extern Atom xvSyncToVBlank, xvSetDefaults;

/*
* The filtering function used for video scaling. We use a cubic filter as defined in
* "Reconstruction Filters in Computer Graphics"
* Mitchell & Netravali in SIGGRAPH '88
*/
static float filter_func(float x)
{
const double B=0.75;
const double C=(1.0-B)/2.0;
double x1=fabs(x);
double x2=fabs(x)*x1;
double x3=fabs(x)*x2;

if (fabs(x)<1.0)
return ( (12.0-9.0*B-6.0*C)*x3+(-18.0+12.0*B+6.0*C)*x2+(6.0-2.0*B) )/6.0;
else
return ( (-B-6.0*C)*x3+(6.0*B+30.0*C)*x2+(-12.0*B-48.0*C)*x1+(8.0*B+24.0*C) )/6.0;
}

static int8_t f32tosb8(float v)
{
return (int8_t)(v*127.0);
}

/*
* Implements the filtering as described in
* "Fast Third-Order Texture Filtering"
* Sigg & Hardwiger in GPU Gems 2
*/
#define TABLE_SIZE 512
static void compute_filter_table(int8_t *t) {
int i;
float x;
for(i=0;i<TABLE_SIZE;i++) {
x=(i+0.5)/TABLE_SIZE;

float w0=filter_func(x+1.0);
float w1=filter_func(x);
float w2=filter_func(x-1.0);
float w3=filter_func(x-2.0);

t[4*i+2]=f32tosb8(1.0+x-w1/(w0+w1));
t[4*i+1]=f32tosb8(1.0-x+w3/(w2+w3));
t[4*i+0]=f32tosb8(w0+w1);
t[4*i+3]=f32tosb8(0.0);
}
}

static void
NV40_LoadFilterTable(ScrnInfoPtr pScrn)
{
NVPtr pNv = NVPTR(pScrn);

if (!pNv->xv_filtertable_mem) {
if (nouveau_bo_new(pNv->dev, NOUVEAU_BO_VRAM | NOUVEAU_BO_GART |
NOUVEAU_BO_MAP, 0,
TABLE_SIZE * sizeof(float) * 4,
&pNv->xv_filtertable_mem)) {
xf86DrvMsg(pScrn->scrnIndex, X_ERROR,
"Couldn't alloc filter table!\n");
return;
}

if (nouveau_bo_map(pNv->xv_filtertable_mem, NOUVEAU_BO_RDWR)) {
xf86DrvMsg(pScrn->scrnIndex, X_ERROR,
"Couldn't map filter table!\n");
return;
}

int8_t *t=pNv->xv_filtertable_mem->map;
compute_filter_table(t);
nouveau_bo_unmap(pNv->xv_filtertable_mem);
}
}

#define SWIZZLE(ts0x,ts0y,ts0z,ts0w,ts1x,ts1y,ts1z,ts1w) \
( \
NV30_3D_TEX_SWIZZLE_S0_X_##ts0x | NV30_3D_TEX_SWIZZLE_S0_Y_##ts0y | \
NV30_3D_TEX_SWIZZLE_S0_Z_##ts0z | NV30_3D_TEX_SWIZZLE_S0_W_##ts0w | \
NV30_3D_TEX_SWIZZLE_S1_X_##ts1x | NV30_3D_TEX_SWIZZLE_S1_Y_##ts1y | \
NV30_3D_TEX_SWIZZLE_S1_Z_##ts1z | NV30_3D_TEX_SWIZZLE_S1_W_##ts1w \
)

/*
* Texture 0 : filter table
* Texture 1 : Y data
* Texture 2 : UV data
*/
static Bool
NV40VideoTexture(ScrnInfoPtr pScrn, struct nouveau_bo *src, int offset,
uint16_t width, uint16_t height, uint16_t src_pitch, int unit)
{
NVPtr pNv = NVPTR(pScrn);
struct nouveau_channel *chan = pNv->chan;
uint32_t tex_reloc = NOUVEAU_BO_VRAM | NOUVEAU_BO_GART | NOUVEAU_BO_RD;
uint32_t card_fmt = 0;
uint32_t card_swz = 0;

switch(unit) {
case 0:
card_fmt = NV40_3D_TEX_FORMAT_FORMAT_A8R8G8B8;
card_swz = SWIZZLE(S1, S1, S1, S1, X, Y, Z, W);
break;
case 1:
card_fmt = NV40_3D_TEX_FORMAT_FORMAT_L8;
card_swz = SWIZZLE(S1, S1, S1, S1, X, X, X, X);
break;
case 2:
card_fmt = NV40_3D_TEX_FORMAT_FORMAT_A8L8;
#if X_BYTE_ORDER == X_BIG_ENDIAN
card_swz = SWIZZLE(S1, S1, S1, S1, Z, W, X, Y); /* x = V, y = U */
#else
card_swz = SWIZZLE(S1, S1, S1, S1, W, Z, Y, X); /* x = V, y = U */
#endif
break;
}

BEGIN_NV04(chan, NV30_3D(TEX_OFFSET(unit)), 8);
if (OUT_RELOCl(chan, src, offset, tex_reloc))
return FALSE;
if (unit==0) {
if (OUT_RELOCd(chan, src, card_fmt | 0x8000 |
NV30_3D_TEX_FORMAT_DIMS_1D |
NV30_3D_TEX_FORMAT_NO_BORDER |
(1 << NV40_3D_TEX_FORMAT_MIPMAP_COUNT__SHIFT),
tex_reloc | NOUVEAU_BO_OR,
NV30_3D_TEX_FORMAT_DMA0,
NV30_3D_TEX_FORMAT_DMA1))
return FALSE;
OUT_RING (chan, NV30_3D_TEX_WRAP_S_REPEAT |
NV30_3D_TEX_WRAP_T_CLAMP_TO_EDGE |
NV30_3D_TEX_WRAP_R_CLAMP_TO_EDGE);
} else {
if (OUT_RELOCd(chan, src, card_fmt | 0x8000 |
NV40_3D_TEX_FORMAT_LINEAR |
NV40_3D_TEX_FORMAT_RECT |
NV30_3D_TEX_FORMAT_DIMS_2D |
NV30_3D_TEX_FORMAT_NO_BORDER |
(1 << NV40_3D_TEX_FORMAT_MIPMAP_COUNT__SHIFT),
tex_reloc | NOUVEAU_BO_OR,
NV30_3D_TEX_FORMAT_DMA0,
NV30_3D_TEX_FORMAT_DMA1))
return FALSE;
OUT_RING (chan, NV30_3D_TEX_WRAP_S_CLAMP_TO_EDGE |
NV30_3D_TEX_WRAP_T_CLAMP_TO_EDGE |
NV30_3D_TEX_WRAP_R_CLAMP_TO_EDGE);
}

OUT_RING (chan, NV40_3D_TEX_ENABLE_ENABLE);
OUT_RING (chan, card_swz);
if (unit == 0)
OUT_RING (chan, NV30_3D_TEX_FILTER_SIGNED_ALPHA |
NV30_3D_TEX_FILTER_SIGNED_RED |
NV30_3D_TEX_FILTER_SIGNED_GREEN |
NV30_3D_TEX_FILTER_SIGNED_BLUE |
NV30_3D_TEX_FILTER_MIN_LINEAR |
NV30_3D_TEX_FILTER_MAG_LINEAR | 0x3fd6);
else
OUT_RING (chan, NV30_3D_TEX_FILTER_MIN_LINEAR |
NV30_3D_TEX_FILTER_MAG_LINEAR | 0x3fd6);
OUT_RING (chan, (width << 16) | height);
OUT_RING (chan, 0); /* border ARGB */

BEGIN_NV04(chan, NV40_3D(TEX_SIZE1(unit)), 1);
OUT_RING (chan, (1 << NV40_3D_TEX_SIZE1_DEPTH__SHIFT) |
(uint16_t) src_pitch);

return TRUE;
}

Bool
NV40GetSurfaceFormat(PixmapPtr ppix, int *fmt_ret)
{
switch (ppix->drawable.bitsPerPixel) {
case 32:
*fmt_ret = NV30_3D_RT_FORMAT_COLOR_A8R8G8B8;
break;
case 24:
*fmt_ret = NV30_3D_RT_FORMAT_COLOR_X8R8G8B8;
break;
case 16:
*fmt_ret = NV30_3D_RT_FORMAT_COLOR_R5G6B5;
break;
case 8:
*fmt_ret = NV30_3D_RT_FORMAT_COLOR_B8;
break;
default:
return FALSE;
}

return TRUE;
}

void
NV40StopTexturedVideo(ScrnInfoPtr pScrn, pointer data, Bool Exit)
{
}

#define VERTEX_OUT(sx,sy,dx,dy) do { \
BEGIN_NV04(chan, NV30_3D(VTX_ATTR_2F_X(8)), 4); \
OUT_RINGf (chan, (sx)); OUT_RINGf (chan, (sy)); \
OUT_RINGf (chan, (sx)/2.0); OUT_RINGf (chan, (sy)/2.0); \
BEGIN_NV04(chan, NV30_3D(VTX_ATTR_2I(0)), 1); \
OUT_RING (chan, ((dy)<<16)|(dx)); \
} while(0)

int
NV40PutTextureImage(ScrnInfoPtr pScrn,
struct nouveau_bo *src, int src_offset, int src_offset2,
int id, int src_pitch, BoxPtr dstBox,
int x1, int y1, int x2, int y2,
uint16_t width, uint16_t height,
uint16_t src_w, uint16_t src_h,
uint16_t drw_w, uint16_t drw_h,
RegionPtr clipBoxes, PixmapPtr ppix,
NVPortPrivPtr pPriv)
{
NVPtr pNv = NVPTR(pScrn);
struct nouveau_channel *chan = pNv->chan;
struct nouveau_bo *bo = nouveau_pixmap_bo(ppix);
Bool bicubic = pPriv->bicubic;
float X1, X2, Y1, Y2;
BoxPtr pbox;
int nbox;
int dst_format = 0;

if (drw_w > 4096 || drw_h > 4096) {
xf86DrvMsg(pScrn->scrnIndex, X_ERROR,
"XV: Draw size too large.\n");
return BadAlloc;
}

if (!NV40GetSurfaceFormat(ppix, &dst_format)) {
ErrorF("No surface format, bad.\n");
return BadImplementation;
}

pbox = REGION_RECTS(clipBoxes);
nbox = REGION_NUM_RECTS(clipBoxes);

if (MARK_RING(chan, 128, 1 + 1 + 3*2))
return BadImplementation;

/* Disable blending */
BEGIN_NV04(chan, NV30_3D(BLEND_FUNC_ENABLE), 1);
OUT_RING (chan, 0);

/* Setup surface */
BEGIN_NV04(chan, NV30_3D(RT_FORMAT), 3);
OUT_RING (chan, NV30_3D_RT_FORMAT_TYPE_LINEAR |
NV30_3D_RT_FORMAT_ZETA_Z24S8 | dst_format);
OUT_RING (chan, exaGetPixmapPitch(ppix));
if (OUT_RELOCl(chan, bo, 0, NOUVEAU_BO_VRAM | NOUVEAU_BO_WR)) {
MARK_UNDO(chan);
return BadImplementation;
}

NV40_LoadFilterTable(pScrn);

if (!NV40VideoTexture(pScrn, pNv->xv_filtertable_mem, 0, TABLE_SIZE,
1, 0 , 0) ||
!NV40VideoTexture(pScrn, src, src_offset, src_w, src_h,
src_pitch, 1)) {
MARK_UNDO(chan);
return BadImplementation;
}

/* We've got NV12 format, which means half width and half height texture of chroma channels. */
if (!NV40VideoTexture(pScrn, src, src_offset2, src_w/2,
src_h/2, src_pitch, 2)) {
MARK_UNDO(chan);
return BadImplementation;
}

NV40_LoadVtxProg(pScrn, &nv40_vp_video);

if (drw_w / 2 < src_w || drw_h / 2 < src_h)
bicubic = FALSE;

if (!NV40_LoadFragProg(pScrn, bicubic ?
&nv40_fp_yv12_bicubic :
&nv30_fp_yv12_bilinear)) {
MARK_UNDO(chan);
return BadImplementation;
}

/* Appears to be some kind of cache flush, needed here at least
* sometimes.. funky text rendering otherwise :)
*/
BEGIN_NV04(chan, NV40_3D(TEX_CACHE_CTL), 1);
OUT_RING (chan, 2);
BEGIN_NV04(chan, NV40_3D(TEX_CACHE_CTL), 1);
OUT_RING (chan, 1);

/* Just before rendering we wait for vblank in the non-composited case. */
if (pPriv->SyncToVBlank) {
FIRE_RING(chan);
NV11SyncToVBlank(ppix, dstBox);
}

/* These are fixed point values in the 16.16 format. */
X1 = (float)(x1>>16)+(float)(x1&0xFFFF)/(float)0x10000;
Y1 = (float)(y1>>16)+(float)(y1&0xFFFF)/(float)0x10000;
X2 = (float)(x2>>16)+(float)(x2&0xFFFF)/(float)0x10000;
Y2 = (float)(y2>>16)+(float)(y2&0xFFFF)/(float)0x10000;

BEGIN_NV04(chan, NV30_3D(VERTEX_BEGIN_END), 1);
OUT_RING (chan, NV30_3D_VERTEX_BEGIN_END_TRIANGLES);

while(nbox--) {
float tx1=X1+(float)(pbox->x1 - dstBox->x1)*(X2-X1)/(float)(drw_w);
float tx2=X1+(float)(pbox->x2 - dstBox->x1)*(src_w)/(float)(drw_w);
float ty1=Y1+(float)(pbox->y1 - dstBox->y1)*(Y2-Y1)/(float)(drw_h);
float ty2=Y1+(float)(pbox->y2 - dstBox->y1)*(src_h)/(float)(drw_h);
int sx1=pbox->x1;
int sx2=pbox->x2;
int sy1=pbox->y1;
int sy2=pbox->y2;

BEGIN_NV04(chan, NV30_3D(SCISSOR_HORIZ), 2);
OUT_RING (chan, (sx2 << 16) | 0);
OUT_RING (chan, (sy2 << 16) | 0);

VERTEX_OUT(tx1, ty1, sx1, sy1);
VERTEX_OUT(tx2+(tx2-tx1), ty1, sx2+(sx2-sx1), sy1);
VERTEX_OUT(tx1, ty2+(ty2-ty1), sx1, sy2+(sy2-sy1));

pbox++;
}

BEGIN_NV04(chan, NV30_3D(VERTEX_BEGIN_END), 1);
OUT_RING (chan, NV30_3D_VERTEX_BEGIN_END_STOP);

FIRE_RING (chan);

return Success;
}

/**
* NV40SetTexturePortAttribute
* sets the attribute "attribute" of port "data" to value "value"
* supported attributes:
* Sync to vblank.
*
* @param pScrenInfo
* @param attribute attribute to set
* @param value value to which attribute is to be set
* @param data port from which the attribute is to be set
*
* @return Success, if setting is successful
* BadValue/BadMatch, if value/attribute are invalid
*/
int
NV40SetTexturePortAttribute(ScrnInfoPtr pScrn, Atom attribute,
                       INT32 value, pointer data)
{
        NVPortPrivPtr pPriv = (NVPortPrivPtr)data;

        if (attribute == xvSyncToVBlank) {
                if ((value < 0) || (value > 1))
                        return BadValue;
                pPriv->SyncToVBlank = value;
        } else
        if (attribute == xvSetDefaults) {
                pPriv->SyncToVBlank = TRUE;
        } else
                return BadMatch;

        return Success;
}

/**
* NV40GetTexturePortAttribute
* reads the value of attribute "attribute" from port "data" into INT32 "*value"
* Sync to vblank.
*
* @param pScrn unused
* @param attribute attribute to be read
* @param value value of attribute will be stored here
* @param data port from which attribute will be read
* @return Success, if queried attribute exists
*/
int
NV40GetTexturePortAttribute(ScrnInfoPtr pScrn, Atom attribute,
                       INT32 *value, pointer data)
{
        NVPortPrivPtr pPriv = (NVPortPrivPtr)data;

        if(attribute == xvSyncToVBlank)
                *value = (pPriv->SyncToVBlank) ? 1 : 0;
        else
                return BadMatch;

        return Success;
}
Something went wrong with that request. Please try again.