summaryrefslogtreecommitdiff
path: root/src/gallium/auxiliary/util/u_blit.c
blob: 79e5c6cba91beee894335645a3905ec9f81d4e13 (plain)
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
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
/**************************************************************************
 *
 * Copyright 2008 VMware, Inc.
 * 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 VMWARE 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.
 *
 **************************************************************************/

/**
 * @file
 * Copy/blit pixel rect between surfaces
 *
 * @author Brian Paul
 */


#include "pipe/p_context.h"
#include "util/u_debug.h"
#include "pipe/p_defines.h"
#include "util/u_inlines.h"
#include "pipe/p_shader_tokens.h"
#include "pipe/p_state.h"

#include "util/u_blit.h"
#include "util/u_draw_quad.h"
#include "util/format/u_format.h"
#include "util/u_math.h"
#include "util/u_memory.h"
#include "util/u_sampler.h"
#include "util/u_texture.h"
#include "util/u_simple_shaders.h"

#include "cso_cache/cso_context.h"


struct blit_state
{
   struct pipe_context *pipe;
   struct cso_context *cso;

   struct pipe_blend_state blend_write_color;
   struct pipe_depth_stencil_alpha_state dsa_keep_depthstencil;
   struct pipe_rasterizer_state rasterizer;
   struct pipe_sampler_state sampler;
   struct pipe_viewport_state viewport;
   struct cso_velems_state velem;

   void *vs;
   void *fs[PIPE_MAX_TEXTURE_TYPES][4];

   struct pipe_resource *vbuf;  /**< quad vertices */
   unsigned vbuf_slot;

   float vertices[4][2][4];   /**< vertex/texcoords for quad */
};


/**
 * Create state object for blit.
 * Intended to be created once and re-used for many blit() calls.
 */
struct blit_state *
util_create_blit(struct pipe_context *pipe, struct cso_context *cso)
{
   struct blit_state *ctx;
   uint i;

   ctx = CALLOC_STRUCT(blit_state);
   if (!ctx)
      return NULL;

   ctx->pipe = pipe;
   ctx->cso = cso;

   /* disabled blending/masking */
   ctx->blend_write_color.rt[0].colormask = PIPE_MASK_RGBA;

   /* rasterizer */
   ctx->rasterizer.cull_face = PIPE_FACE_NONE;
   ctx->rasterizer.half_pixel_center = 1;
   ctx->rasterizer.bottom_edge_rule = 1;
   ctx->rasterizer.depth_clip_near = 1;
   ctx->rasterizer.depth_clip_far = 1;

   /* samplers */
   ctx->sampler.wrap_s = PIPE_TEX_WRAP_CLAMP_TO_EDGE;
   ctx->sampler.wrap_t = PIPE_TEX_WRAP_CLAMP_TO_EDGE;
   ctx->sampler.wrap_r = PIPE_TEX_WRAP_CLAMP_TO_EDGE;
   ctx->sampler.min_mip_filter = PIPE_TEX_MIPFILTER_NONE;
   ctx->sampler.min_img_filter = 0; /* set later */
   ctx->sampler.mag_img_filter = 0; /* set later */

   /* vertex elements state */
   ctx->velem.count = 2;
   for (i = 0; i < 2; i++) {
      ctx->velem.velems[i].src_offset = i * 4 * sizeof(float);
      ctx->velem.velems[i].instance_divisor = 0;
      ctx->velem.velems[i].vertex_buffer_index = 0;
      ctx->velem.velems[i].src_format = PIPE_FORMAT_R32G32B32A32_FLOAT;
   }

   ctx->vbuf = NULL;

   /* init vertex data that doesn't change */
   for (i = 0; i < 4; i++) {
      ctx->vertices[i][0][3] = 1.0f; /* w */
      ctx->vertices[i][1][3] = 1.0f; /* q */
   }

   return ctx;
}


/**
 * Destroy a blit context
 */
void
util_destroy_blit(struct blit_state *ctx)
{
   struct pipe_context *pipe = ctx->pipe;
   unsigned i, j;

   if (ctx->vs)
      pipe->delete_vs_state(pipe, ctx->vs);

   for (i = 0; i < ARRAY_SIZE(ctx->fs); i++) {
      for (j = 0; j < ARRAY_SIZE(ctx->fs[i]); j++) {
         if (ctx->fs[i][j])
            pipe->delete_fs_state(pipe, ctx->fs[i][j]);
      }
   }

   pipe_resource_reference(&ctx->vbuf, NULL);

   FREE(ctx);
}


/**
 * Helper function to set the fragment shaders.
 */
static inline void
set_fragment_shader(struct blit_state *ctx,
                    enum pipe_format format,
                    boolean src_xrbias,
                    enum pipe_texture_target pipe_tex)
{
   enum tgsi_return_type stype;
   unsigned idx;

   if (util_format_is_pure_uint(format)) {
      stype = TGSI_RETURN_TYPE_UINT;
      idx = 0;
   } else if (util_format_is_pure_sint(format)) {
      stype = TGSI_RETURN_TYPE_SINT;
      idx = 1;
   } else {
      stype = TGSI_RETURN_TYPE_FLOAT;
      idx = 2;
   }

   if (src_xrbias) {
      assert(stype == TGSI_RETURN_TYPE_FLOAT);
      idx = 3;
      if (!ctx->fs[pipe_tex][idx]) {
         enum tgsi_texture_type tgsi_tex =
            util_pipe_tex_to_tgsi_tex(pipe_tex, 0);
         ctx->fs[pipe_tex][idx] =
            util_make_fragment_tex_shader_xrbias(ctx->pipe, tgsi_tex);
      }
   }
   else if (!ctx->fs[pipe_tex][idx]) {
      enum tgsi_texture_type tgsi_tex = util_pipe_tex_to_tgsi_tex(pipe_tex, 0);

      /* OpenGL does not allow blits from signed to unsigned integer
       * or vice versa. */
      ctx->fs[pipe_tex][idx] =
         util_make_fragment_tex_shader_writemask(ctx->pipe, tgsi_tex,
                                                 TGSI_INTERPOLATE_LINEAR,
                                                 TGSI_WRITEMASK_XYZW,
                                                 stype, stype, false, false);
   }

   cso_set_fragment_shader_handle(ctx->cso, ctx->fs[pipe_tex][idx]);
}


/**
 * Helper function to set the vertex shader.
 */
static inline void
set_vertex_shader(struct blit_state *ctx)
{
   /* vertex shader - still required to provide the linkage between
    * fragment shader input semantics and vertex_element/buffers.
    */
   if (!ctx->vs) {
      const enum tgsi_semantic semantic_names[] = {
         TGSI_SEMANTIC_POSITION,
         TGSI_SEMANTIC_GENERIC
      };
      const uint semantic_indexes[] = { 0, 0 };
      ctx->vs = util_make_vertex_passthrough_shader(ctx->pipe, 2,
                                                    semantic_names,
                                                    semantic_indexes, FALSE);
   }

   cso_set_vertex_shader_handle(ctx->cso, ctx->vs);
}


/**
 * Get offset of next free slot in vertex buffer for quad vertices.
 */
static unsigned
get_next_slot(struct blit_state *ctx)
{
   const unsigned max_slots = 4096 / sizeof ctx->vertices;

   if (ctx->vbuf_slot >= max_slots) {
      pipe_resource_reference(&ctx->vbuf, NULL);
      ctx->vbuf_slot = 0;
   }

   if (!ctx->vbuf) {
      ctx->vbuf = pipe_buffer_create(ctx->pipe->screen,
                                     PIPE_BIND_VERTEX_BUFFER,
                                     PIPE_USAGE_STREAM,
                                     max_slots * sizeof ctx->vertices);
   }

   return ctx->vbuf_slot++ * sizeof ctx->vertices;
}




/**
 * Setup vertex data for the textured quad we'll draw.
 * Note: y=0=top
 *
 * FIXME: We should call util_map_texcoords2d_onto_cubemap
 * for cubemaps.
 */
static unsigned
setup_vertex_data_tex(struct blit_state *ctx,
                      enum pipe_texture_target src_target,
                      unsigned src_face,
                      float x0, float y0, float x1, float y1,
                      float s0, float t0, float s1, float t1,
                      float z)
{
   unsigned offset;

   ctx->vertices[0][0][0] = x0;
   ctx->vertices[0][0][1] = y0;
   ctx->vertices[0][0][2] = z;
   ctx->vertices[0][1][0] = s0; /*s*/
   ctx->vertices[0][1][1] = t0; /*t*/
   ctx->vertices[0][1][2] = 0;  /*r*/

   ctx->vertices[1][0][0] = x1;
   ctx->vertices[1][0][1] = y0;
   ctx->vertices[1][0][2] = z;
   ctx->vertices[1][1][0] = s1; /*s*/
   ctx->vertices[1][1][1] = t0; /*t*/
   ctx->vertices[1][1][2] = 0;  /*r*/

   ctx->vertices[2][0][0] = x1;
   ctx->vertices[2][0][1] = y1;
   ctx->vertices[2][0][2] = z;
   ctx->vertices[2][1][0] = s1;
   ctx->vertices[2][1][1] = t1;
   ctx->vertices[3][1][2] = 0;

   ctx->vertices[3][0][0] = x0;
   ctx->vertices[3][0][1] = y1;
   ctx->vertices[3][0][2] = z;
   ctx->vertices[3][1][0] = s0;
   ctx->vertices[3][1][1] = t1;
   ctx->vertices[3][1][2] = 0;

   if (src_target == PIPE_TEXTURE_CUBE ||
       src_target == PIPE_TEXTURE_CUBE_ARRAY) {
      /* Map cubemap texture coordinates inplace. */
      const unsigned stride =
         sizeof ctx->vertices[0] / sizeof ctx->vertices[0][0][0];
      util_map_texcoords2d_onto_cubemap(src_face,
                                        &ctx->vertices[0][1][0], stride,
                                        &ctx->vertices[0][1][0], stride,
                                        TRUE);
   }

   offset = get_next_slot(ctx);

   if (ctx->vbuf) {
      pipe_buffer_write_nooverlap(ctx->pipe, ctx->vbuf,
                                  offset, sizeof(ctx->vertices), ctx->vertices);
   }

   return offset;
}


/**
 * \return TRUE if two regions overlap, FALSE otherwise
 */
static boolean
regions_overlap(int srcX0, int srcY0,
                int srcX1, int srcY1,
                int dstX0, int dstY0,
                int dstX1, int dstY1)
{
   if (MAX2(srcX0, srcX1) <= MIN2(dstX0, dstX1))
      return FALSE; /* src completely left of dst */

   if (MAX2(dstX0, dstX1) <= MIN2(srcX0, srcX1))
      return FALSE; /* dst completely left of src */

   if (MAX2(srcY0, srcY1) <= MIN2(dstY0, dstY1))
      return FALSE; /* src completely above dst */

   if (MAX2(dstY0, dstY1) <= MIN2(srcY0, srcY1))
      return FALSE; /* dst completely above src */

   return TRUE; /* some overlap */
}


/**
 * Can we blit from src format to dest format with a simple copy?
 */
static boolean
formats_compatible(enum pipe_format src_format,
                   enum pipe_format dst_format)
{
   if (src_format == dst_format) {
      return TRUE;
   }
   else {
      const struct util_format_description *src_desc =
         util_format_description(src_format);
      const struct util_format_description *dst_desc =
         util_format_description(dst_format);
      return util_is_format_compatible(src_desc, dst_desc);
   }
}


/**
 * Copy pixel block from src surface to dst surface.
 * Overlapping regions are acceptable.
 * Flipping and stretching are supported.
 * \param filter  one of PIPE_TEX_FILTER_NEAREST/LINEAR
 * \param writemask  bitmask of PIPE_MASK_[RGBAZS].  Controls which channels
 *                   in the dest surface are sourced from the src surface.
 *                   Disabled color channels are sourced from (0,0,0,1).
 */
void
util_blit_pixels(struct blit_state *ctx,
                 struct pipe_resource *src_tex,
                 unsigned src_level,
                 int srcX0, int srcY0,
                 int srcX1, int srcY1,
                 int srcZ0,
                 struct pipe_surface *dst,
                 int dstX0, int dstY0,
                 int dstX1, int dstY1,
                 ASSERTED float z,
                 enum pipe_tex_filter filter,
                 uint writemask)
{
   struct pipe_context *pipe = ctx->pipe;
   enum pipe_format src_format, dst_format;
   const int srcW = abs(srcX1 - srcX0);
   const int srcH = abs(srcY1 - srcY0);
   boolean overlap;
   boolean is_stencil, is_depth, blit_depth, blit_stencil;
   const struct util_format_description *src_desc =
         util_format_description(src_tex->format);
   struct pipe_blit_info info;

   assert(filter == PIPE_TEX_FILTER_NEAREST ||
          filter == PIPE_TEX_FILTER_LINEAR);

   assert(src_level <= src_tex->last_level);

   /* do the regions overlap? */
   overlap = src_tex == dst->texture &&
             dst->u.tex.level == src_level &&
             dst->u.tex.first_layer == srcZ0 &&
      regions_overlap(srcX0, srcY0, srcX1, srcY1,
                      dstX0, dstY0, dstX1, dstY1);

   src_format = util_format_linear(src_tex->format);
   dst_format = util_format_linear(dst->texture->format);

   /* See whether we will blit depth or stencil. */
   is_depth = util_format_has_depth(src_desc);
   is_stencil = util_format_has_stencil(src_desc);

   blit_depth = is_depth && (writemask & PIPE_MASK_Z);
   blit_stencil = is_stencil && (writemask & PIPE_MASK_S);

   if (is_depth || is_stencil) {
      assert((writemask & PIPE_MASK_RGBA) == 0);
      assert(blit_depth || blit_stencil);
   }
   else {
      assert((writemask & PIPE_MASK_ZS) == 0);
      assert(!blit_depth);
      assert(!blit_stencil);
   }

   /*
    * XXX: z parameter is deprecated. dst->u.tex.first_layer
    * specificies the destination layer.
    */
   assert(z == 0.0f);

   /*
    * Check for simple case:  no format conversion, no flipping, no stretching,
    * no overlapping, same number of samples.
    * Filter mode should not matter since there's no stretching.
    */
   if (formats_compatible(src_format, dst_format) &&
       src_tex->nr_samples == dst->texture->nr_samples &&
       is_stencil == blit_stencil &&
       is_depth == blit_depth &&
       srcX0 < srcX1 &&
       dstX0 < dstX1 &&
       srcY0 < srcY1 &&
       dstY0 < dstY1 &&
       (dstX1 - dstX0) == (srcX1 - srcX0) &&
       (dstY1 - dstY0) == (srcY1 - srcY0) &&
       !overlap) {
      struct pipe_box src_box;
      src_box.x = srcX0;
      src_box.y = srcY0;
      src_box.z = srcZ0;
      src_box.width = srcW;
      src_box.height = srcH;
      src_box.depth = 1;
      pipe->resource_copy_region(pipe,
                                 dst->texture, dst->u.tex.level,
                                 dstX0, dstY0, dst->u.tex.first_layer,/* dest */
                                 src_tex, src_level,
                                 &src_box);
      return;
   }

   memset(&info, 0, sizeof info);
   info.dst.resource = dst->texture;
   info.dst.level = dst->u.tex.level;
   info.dst.box.x = dstX0;
   info.dst.box.y = dstY0;
   info.dst.box.z = dst->u.tex.first_layer;
   info.dst.box.width = dstX1 - dstX0;
   info.dst.box.height = dstY1 - dstY0;
   assert(info.dst.box.width >= 0);
   assert(info.dst.box.height >= 0);
   info.dst.box.depth = 1;
   info.dst.format = dst_format;
   info.src.resource = src_tex;
   info.src.level = src_level;
   info.src.box.x = srcX0;
   info.src.box.y = srcY0;
   info.src.box.z = srcZ0;
   info.src.box.width = srcX1 - srcX0;
   info.src.box.height = srcY1 - srcY0;
   info.src.box.depth = 1;
   info.src.format = src_format;
   info.mask = writemask;
   info.filter = filter;
   info.scissor_enable = 0;

   pipe->blit(pipe, &info);
}


/**
 * Copy pixel block from src sampler view to dst surface.
 *
 * The sampler view's first_level field indicates the source
 * mipmap level to use.
 *
 * The sampler view's first_layer indicate the layer to use, but for
 * cube maps it must point to the first face.  Face is passed in src_face.
 *
 * The main advantage over util_blit_pixels is that it allows to specify
 * swizzles in pipe_sampler_view::swizzle_?.
 *
 * But there is no control over blitting Z and/or stencil.
 */
void
util_blit_pixels_tex(struct blit_state *ctx,
                     struct pipe_sampler_view *src_sampler_view,
                     int srcX0, int srcY0,
                     int srcX1, int srcY1,
                     unsigned src_face,
                     struct pipe_surface *dst,
                     int dstX0, int dstY0,
                     int dstX1, int dstY1,
                     float z, enum pipe_tex_filter filter,
                     boolean src_xrbias)
{
   boolean normalized = src_sampler_view->texture->target != PIPE_TEXTURE_RECT;
   struct pipe_framebuffer_state fb;
   float s0, t0, s1, t1;
   unsigned offset;
   struct pipe_resource *tex = src_sampler_view->texture;

   assert(filter == PIPE_TEX_FILTER_NEAREST ||
          filter == PIPE_TEX_FILTER_LINEAR);

   assert(tex);
   assert(tex->width0 != 0);
   assert(tex->height0 != 0);

   s0 = (float) srcX0;
   s1 = (float) srcX1;
   t0 = (float) srcY0;
   t1 = (float) srcY1;

   if (normalized) {
      /* normalize according to the mipmap level's size */
      int level = src_sampler_view->u.tex.first_level;
      float w = (float) u_minify(tex->width0, level);
      float h = (float) u_minify(tex->height0, level);
      s0 /= w;
      s1 /= w;
      t0 /= h;
      t1 /= h;
   }

   assert(ctx->pipe->screen->is_format_supported(ctx->pipe->screen, dst->format,
                                                 PIPE_TEXTURE_2D,
                                                 dst->texture->nr_samples,
                                                 dst->texture->nr_storage_samples,
                                                 PIPE_BIND_RENDER_TARGET));

   /* save state (restored below) */
   cso_save_state(ctx->cso, (CSO_BIT_BLEND |
                             CSO_BIT_DEPTH_STENCIL_ALPHA |
                             CSO_BIT_RASTERIZER |
                             CSO_BIT_SAMPLE_MASK |
                             CSO_BIT_MIN_SAMPLES |
                             CSO_BIT_FRAGMENT_SAMPLERS |
                             CSO_BIT_FRAGMENT_SAMPLER_VIEWS |
                             CSO_BIT_STREAM_OUTPUTS |
                             CSO_BIT_VIEWPORT |
                             CSO_BIT_FRAMEBUFFER |
                             CSO_BIT_PAUSE_QUERIES |
                             CSO_BIT_FRAGMENT_SHADER |
                             CSO_BIT_VERTEX_SHADER |
                             CSO_BIT_TESSCTRL_SHADER |
                             CSO_BIT_TESSEVAL_SHADER |
                             CSO_BIT_GEOMETRY_SHADER |
                             CSO_BIT_VERTEX_ELEMENTS |
                             CSO_BIT_AUX_VERTEX_BUFFER_SLOT));

   /* set misc state we care about */
   cso_set_blend(ctx->cso, &ctx->blend_write_color);
   cso_set_depth_stencil_alpha(ctx->cso, &ctx->dsa_keep_depthstencil);
   cso_set_sample_mask(ctx->cso, ~0);
   cso_set_min_samples(ctx->cso, 1);
   cso_set_rasterizer(ctx->cso, &ctx->rasterizer);
   cso_set_vertex_elements(ctx->cso, &ctx->velem);
   cso_set_stream_outputs(ctx->cso, 0, NULL, NULL);

   /* sampler */
   ctx->sampler.normalized_coords = normalized;
   ctx->sampler.min_img_filter = filter;
   ctx->sampler.mag_img_filter = filter;
   {
      const struct pipe_sampler_state *samplers[] = {&ctx->sampler};
      cso_set_samplers(ctx->cso, PIPE_SHADER_FRAGMENT, 1, samplers);
   }

   /* viewport */
   ctx->viewport.scale[0] = 0.5f * dst->width;
   ctx->viewport.scale[1] = 0.5f * dst->height;
   ctx->viewport.scale[2] = 0.5f;
   ctx->viewport.translate[0] = 0.5f * dst->width;
   ctx->viewport.translate[1] = 0.5f * dst->height;
   ctx->viewport.translate[2] = 0.5f;
   cso_set_viewport(ctx->cso, &ctx->viewport);

   /* texture */
   cso_set_sampler_views(ctx->cso, PIPE_SHADER_FRAGMENT, 1, &src_sampler_view);

   /* shaders */
   set_fragment_shader(ctx, src_sampler_view->format,
                       src_xrbias,
                       src_sampler_view->texture->target);
   set_vertex_shader(ctx);
   cso_set_tessctrl_shader_handle(ctx->cso, NULL);
   cso_set_tesseval_shader_handle(ctx->cso, NULL);
   cso_set_geometry_shader_handle(ctx->cso, NULL);

   /* drawing dest */
   memset(&fb, 0, sizeof(fb));
   fb.width = dst->width;
   fb.height = dst->height;
   fb.nr_cbufs = 1;
   fb.cbufs[0] = dst;
   cso_set_framebuffer(ctx->cso, &fb);

   /* draw quad */
   offset = setup_vertex_data_tex(ctx,
                                  src_sampler_view->texture->target,
                                  src_face,
                                  (float) dstX0 / dst->width * 2.0f - 1.0f,
                                  (float) dstY0 / dst->height * 2.0f - 1.0f,
                                  (float) dstX1 / dst->width * 2.0f - 1.0f,
                                  (float) dstY1 / dst->height * 2.0f - 1.0f,
                                  s0, t0, s1, t1,
                                  z);

   util_draw_vertex_buffer(ctx->pipe, ctx->cso, ctx->vbuf, 0,
                           offset,
                           PIPE_PRIM_TRIANGLE_FAN,
                           4,  /* verts */
                           2); /* attribs/vert */

   /* restore state we changed */
   cso_restore_state(ctx->cso);
}