path: root/src/mesa/pipe/softpipe/sp_surface.c

/**************************************************************************
 * 
 * Copyright 2003 Tungsten Graphics, Inc., Cedar Park, Texas.
 * 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 "pipe/p_defines.h"
#include "pipe/p_util.h"
#include "sp_context.h"
#include "sp_state.h"
#include "sp_surface.h"
#include "sp_tile_cache.h"

/**
 * Softpipe surface functions.
 * Basically, create surface of a particular type, then plug in default
 * read/write_quad and get/put_tile() functions.
 * Note that these quad funcs assume the buffer/region is in a linear
 * layout with Y=0=top.
 * If we had swizzled/AOS buffers the read/write quad functions could be
 * simplified a lot....
 */


/** Convert GLshort in [-32768,32767] to GLfloat in [-1.0,1.0] */
#define SHORT_TO_FLOAT(S)   ((2.0F * (S) + 1.0F) * (1.0F/65535.0F))

#define UNCLAMPED_FLOAT_TO_SHORT(us, f)  \
   us = ( (short) ( CLAMP((f), -1.0, 1.0) * 32767.0F) )



#if 0
#define CLIP_TILE \
   do { \
      assert(x + w <= ps->width);               \
      assert(y + h <= ps->height);              \
   } while(0)

#else
#define CLIP_TILE \
   do { \
      if (x + w > ps->width) \
         w = ps->width - x; \
      if (y + h > ps->height) \
         h = ps->height -y; \
   } while(0)
#endif


/*** PIPE_FORMAT_U_A8_R8_G8_B8 ***/

static void
a8r8g8b8_read_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                         float (*rrrr)[QUAD_SIZE])
{
   const unsigned *src
      = ((const unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_A8_R8_G8_B8);
#if 0
   assert(x < (int) sps->surface.width - 1);
   assert(y < (int) sps->surface.height - 1);
#endif
   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         const unsigned p = src[j];
         rrrr[0][i * 2 + j] = UBYTE_TO_FLOAT((p >> 16) & 0xff); /*R*/
         rrrr[1][i * 2 + j] = UBYTE_TO_FLOAT((p >>  8) & 0xff); /*G*/
         rrrr[2][i * 2 + j] = UBYTE_TO_FLOAT((p      ) & 0xff); /*B*/
         rrrr[3][i * 2 + j] = UBYTE_TO_FLOAT((p >> 24) & 0xff); /*A*/
      }
      src += sps->surface.region->pitch;
   }
}

static void
a8r8g8b8_write_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                          float (*rrrr)[QUAD_SIZE])
{
   unsigned *dst
      = ((unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_A8_R8_G8_B8);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         ubyte r, g, b, a;
         UNCLAMPED_FLOAT_TO_UBYTE(r, rrrr[0][i * 2 + j]); /*R*/
         UNCLAMPED_FLOAT_TO_UBYTE(g, rrrr[1][i * 2 + j]); /*G*/
         UNCLAMPED_FLOAT_TO_UBYTE(b, rrrr[2][i * 2 + j]); /*B*/
         UNCLAMPED_FLOAT_TO_UBYTE(a, rrrr[3][i * 2 + j]); /*A*/
         dst[j] = (a << 24) | (r << 16) | (g << 8) | b;
      }
      dst += sps->surface.region->pitch;
   }
}

static void
a8r8g8b8_get_tile(struct pipe_surface *ps,
                  unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const unsigned *src
      = ((const unsigned *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_A8_R8_G8_B8);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         const unsigned pixel = src[j];
         pRow[0] = UBYTE_TO_FLOAT((pixel >> 16) & 0xff);
         pRow[1] = UBYTE_TO_FLOAT((pixel >>  8) & 0xff);
         pRow[2] = UBYTE_TO_FLOAT((pixel >>  0) & 0xff);
         pRow[3] = UBYTE_TO_FLOAT((pixel >> 24) & 0xff);
         pRow += 4;
      }
      src += ps->region->pitch;
      p += w0 * 4;
   }
}


static void
a8r8g8b8_put_tile(struct pipe_surface *ps,
                  unsigned x, unsigned y, unsigned w, unsigned h,
                  const float *p)
{
   unsigned *dst
      = ((unsigned *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_A8_R8_G8_B8);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      const float *pRow = p;
      for (j = 0; j < w; j++) {
         unsigned r, g, b, a;
         UNCLAMPED_FLOAT_TO_UBYTE(r, pRow[0]);
         UNCLAMPED_FLOAT_TO_UBYTE(g, pRow[1]);
         UNCLAMPED_FLOAT_TO_UBYTE(b, pRow[2]);
         UNCLAMPED_FLOAT_TO_UBYTE(a, pRow[3]);
         dst[j] = (a << 24) | (r << 16) | (g << 8) | b;
         pRow += 4;
      }
      dst += ps->region->pitch;
      p += w0 * 4;
   }
}


/*** PIPE_FORMAT_U_A1_R5_G5_B5 ***/

static void
a1r5g5b5_get_tile(struct pipe_surface *ps,
                  unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const ushort *src
      = ((const ushort *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   unsigned i, j;

   assert(ps->format == PIPE_FORMAT_U_A1_R5_G5_B5);

   for (i = 0; i < h; i++) {
      for (j = 0; j < w; j++) {
         const ushort pixel = src[j];
         p[0] = ((pixel >> 10) & 0x1f) * (1.0f / 31.0f);
         p[1] = ((pixel >>  5) & 0x1f) * (1.0f / 31.0f);
         p[2] = ((pixel      ) & 0x1f) * (1.0f / 31.0f);
         p[3] = ((pixel >> 15)       ) * 1.0f;
         p += 4;
      }
      src += ps->region->pitch;
   }
}



/*** PIPE_FORMAT_U_Z16 ***/

static void
z16_read_quad_z(struct softpipe_surface *sps,
                int x, int y, unsigned zzzz[QUAD_SIZE])
{
   const ushort *src
      = ((const ushort *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_U_Z16);

   /* converting ushort to unsigned: */
   zzzz[0] = src[0];
   zzzz[1] = src[1];
   src += sps->surface.region->pitch;
   zzzz[2] = src[0];
   zzzz[3] = src[1];
}

static void
z16_write_quad_z(struct softpipe_surface *sps,
                 int x, int y, const unsigned zzzz[QUAD_SIZE])
{
   ushort *dst
      = ((ushort *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_U_Z16);

   /* converting unsigned to ushort: */
   dst[0] = zzzz[0];
   dst[1] = zzzz[1];
   dst += sps->surface.region->pitch;
   dst[0] = zzzz[2];
   dst[1] = zzzz[3];
}

/**
 * Return as floats in [0,1].
 */
static void
z16_get_tile(struct pipe_surface *ps,
             unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const ushort *src
      = ((const ushort *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   const float scale = 1.0 / 65535.0;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_Z16);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         pRow[j] = src[j] * scale;
      }
      src += ps->region->pitch;
      p += w0;
   }
}




/*** PIPE_FORMAT_U_L8 ***/

static void
l8_read_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                   float (*rrrr)[QUAD_SIZE])
{
   const ubyte *src
      = ((const ubyte *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_L8);
   assert(x < (int) sps->surface.width - 1);
   assert(y < (int) sps->surface.height - 1);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         rrrr[0][i * 2 + j] =
         rrrr[1][i * 2 + j] =
         rrrr[2][i * 2 + j] = UBYTE_TO_FLOAT(src[j]);
         rrrr[3][i * 2 + j] = 1.0F;
      }
      src += sps->surface.region->pitch;
   }
}

static void
l8_write_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                    float (*rrrr)[QUAD_SIZE])
{
   ubyte *dst
      = ((ubyte *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_L8);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         ubyte r;
         UNCLAMPED_FLOAT_TO_UBYTE(r, rrrr[0][i * 2 + j]); /*R*/
         dst[j] = r;
      }
      dst += sps->surface.region->pitch;
   }
}

static void
l8_get_tile(struct pipe_surface *ps,
                  unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const ubyte *src
      = ((const ubyte *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_L8);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         pRow[0] =
         pRow[1] =
         pRow[2] = UBYTE_TO_FLOAT(src[j]);
         pRow[3] = 1.0;
         pRow += 4;
      }
      src += ps->region->pitch;
      p += w0 * 4;
   }
}


/*** PIPE_FORMAT_U_A8 ***/

static void
a8_read_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                   float (*rrrr)[QUAD_SIZE])
{
   const ubyte *src
      = ((const ubyte *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_A8);
   assert(x < (int) sps->surface.width - 1);
   assert(y < (int) sps->surface.height - 1);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         rrrr[0][i * 2 + j] =
         rrrr[1][i * 2 + j] =
         rrrr[2][i * 2 + j] = 0.0F;
         rrrr[3][i * 2 + j] = UBYTE_TO_FLOAT(src[j]);
      }
      src += sps->surface.region->pitch;
   }
}

static void
a8_write_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                    float (*rrrr)[QUAD_SIZE])
{
   ubyte *dst
      = ((ubyte *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_A8);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         ubyte r;
         UNCLAMPED_FLOAT_TO_UBYTE(r, rrrr[3][i * 2 + j]); /*A*/
         dst[j] = r;
      }
      dst += sps->surface.region->pitch;
   }
}

static void
a8_get_tile(struct pipe_surface *ps,
                  unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const ubyte *src
      = ((const ubyte *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_A8);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         pRow[0] =
         pRow[1] =
         pRow[2] = 0.0;
         pRow[3] = UBYTE_TO_FLOAT(src[j]);
         pRow += 4;
      }
      src += ps->region->pitch;
      p += w0 * 4;
   }
}


/*** PIPE_FORMAT_S_R16_G16_B16_A16 ***/

static void
r16g16b16a16_get_tile(struct pipe_surface *ps,
                      unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const short *src
      = ((const short *) (ps->region->map + ps->offset))
      + (y * ps->region->pitch + x) * 4;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_S_R16_G16_B16_A16);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      const short *pixel = src;
      for (j = 0; j < w; j++) {
         pRow[0] = SHORT_TO_FLOAT(pixel[0]);
         pRow[1] = SHORT_TO_FLOAT(pixel[1]);
         pRow[2] = SHORT_TO_FLOAT(pixel[2]);
         pRow[3] = SHORT_TO_FLOAT(pixel[3]);
         pRow += 4;
         pixel += 4;
      }
      src += ps->region->pitch * 4;
      p += w0 * 4;
   }
}


static void
r16g16b16a16_put_tile(struct pipe_surface *ps,
                      unsigned x, unsigned y, unsigned w, unsigned h,
                      const float *p)
{
   short *dst
      = ((short *) (ps->region->map + ps->offset))
      + (y * ps->region->pitch + x) * 4;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_S_R16_G16_B16_A16);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      const float *pRow = p;
      for (j = 0; j < w; j++) {
         short r, g, b, a;
         UNCLAMPED_FLOAT_TO_SHORT(r, pRow[0]);
         UNCLAMPED_FLOAT_TO_SHORT(g, pRow[1]);
         UNCLAMPED_FLOAT_TO_SHORT(b, pRow[2]);
         UNCLAMPED_FLOAT_TO_SHORT(a, pRow[3]);
         dst[j*4+0] = r;
         dst[j*4+1] = g;
         dst[j*4+2] = b;
         dst[j*4+3] = a;
         pRow += 4;
      }
      dst += ps->region->pitch * 4;
      p += w0 * 4;
   }
}



/*** PIPE_FORMAT_U_I8 ***/

static void
i8_read_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                   float (*rrrr)[QUAD_SIZE])
{
   const ubyte *src
      = ((const ubyte *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_I8);
   assert(x < (int) sps->surface.width - 1);
   assert(y < (int) sps->surface.height - 1);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         rrrr[0][i * 2 + j] =
         rrrr[1][i * 2 + j] =
         rrrr[2][i * 2 + j] =
         rrrr[3][i * 2 + j] = UBYTE_TO_FLOAT(src[j]);
      }
      src += sps->surface.region->pitch;
   }
}

static void
i8_write_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                    float (*rrrr)[QUAD_SIZE])
{
   ubyte *dst
      = ((ubyte *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_I8);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         ubyte r;
         UNCLAMPED_FLOAT_TO_UBYTE(r, rrrr[0][i * 2 + j]); /*R*/
         dst[j] = r;
      }
      dst += sps->surface.region->pitch;
   }
}

static void
i8_get_tile(struct pipe_surface *ps,
            unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const ubyte *src
      = ((const ubyte *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_I8);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         pRow[0] =
         pRow[1] =
         pRow[2] =
         pRow[3] = UBYTE_TO_FLOAT(src[j]);
         pRow += 4;
      }
      src += ps->region->pitch;
      p += w0 * 4;
   }
}


/*** PIPE_FORMAT_U_A8_L8 ***/

static void
a8_l8_read_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                      float (*rrrr)[QUAD_SIZE])
{
   const ushort *src
      = ((const ushort *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_A8_L8);
   assert(x < (int) sps->surface.width - 1);
   assert(y < (int) sps->surface.height - 1);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         const ushort p = src[j];
         rrrr[0][i * 2 + j] =
         rrrr[1][i * 2 + j] =
         rrrr[2][i * 2 + j] = UBYTE_TO_FLOAT(p >> 8);
         rrrr[3][i * 2 + j] = UBYTE_TO_FLOAT(p & 0xff);
      }
      src += sps->surface.region->pitch;
   }
}

static void
a8_l8_write_quad_f_swz(struct softpipe_surface *sps, int x, int y,
                    float (*rrrr)[QUAD_SIZE])
{
   ushort *dst
      = ((ushort *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;
   unsigned i, j;

   assert(sps->surface.format == PIPE_FORMAT_U_A8_L8);

   for (i = 0; i < 2; i++) { /* loop over pixel row */
      for (j = 0; j < 2; j++) {  /* loop over pixel column */
         ubyte l, a;
         UNCLAMPED_FLOAT_TO_UBYTE(l, rrrr[0][i * 2 + j]); /*R*/
         UNCLAMPED_FLOAT_TO_UBYTE(a, rrrr[3][i * 2 + j]); /*A*/
         dst[j] = (l << 8) | a;
      }
      dst += sps->surface.region->pitch;
   }
}

static void
a8_l8_get_tile(struct pipe_surface *ps,
            unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const ushort *src
      = ((const ushort *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_A8_L8);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         const ushort p = src[j];
         pRow[0] =
         pRow[1] =
         pRow[2] = UBYTE_TO_FLOAT(p & 0xff);
         pRow[3] = UBYTE_TO_FLOAT(p >> 8);
         pRow += 4;
      }
      src += ps->region->pitch;
      p += w0 * 4;
   }
}




/*** PIPE_FORMAT_U_Z32 ***/

static void
z32_read_quad_z(struct softpipe_surface *sps,
                int x, int y, unsigned zzzz[QUAD_SIZE])
{
   const unsigned *src
      = ((unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_U_Z32);

   zzzz[0] = src[0];
   zzzz[1] = src[1];
   src += sps->surface.region->pitch;
   zzzz[2] = src[0];
   zzzz[3] = src[1];
}

static void
z32_write_quad_z(struct softpipe_surface *sps,
                 int x, int y, const unsigned zzzz[QUAD_SIZE])
{
   unsigned *dst
      = ((unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_U_Z32);

   dst[0] = zzzz[0];
   dst[1] = zzzz[1];
   dst += sps->surface.region->pitch;
   dst[0] = zzzz[2];
   dst[1] = zzzz[3];
}

/**
 * Return as floats in [0,1].
 */
static void
z32_get_tile(struct pipe_surface *ps,
             unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const uint *src
      = ((const uint *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   const double scale = 1.0 / (double) 0xffffffff;
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_U_Z16);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         pRow[j] = src[j] * scale;
      }
      src += ps->region->pitch;
      p += w0;
   }
}


/*** PIPE_FORMAT_S8_Z24 ***/

static void
s8z24_read_quad_z(struct softpipe_surface *sps,
                  int x, int y, unsigned zzzz[QUAD_SIZE])
{
   static const unsigned mask = 0x00ffffff;
   const unsigned *src
      = ((unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_S8_Z24);

   /* extract lower three bytes */
   zzzz[0] = src[0] & mask;
   zzzz[1] = src[1] & mask;
   src += sps->surface.region->pitch;
   zzzz[2] = src[0] & mask;
   zzzz[3] = src[1] & mask;
}

static void
s8z24_write_quad_z(struct softpipe_surface *sps,
                   int x, int y, const unsigned zzzz[QUAD_SIZE])
{
   static const unsigned mask = 0xff000000;
   unsigned *dst
      = ((unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_S8_Z24);
   assert(zzzz[0] <= 0xffffff);

   dst[0] = (dst[0] & mask) | zzzz[0];
   dst[1] = (dst[1] & mask) | zzzz[1];
   dst += sps->surface.region->pitch;
   dst[0] = (dst[0] & mask) | zzzz[2];
   dst[1] = (dst[1] & mask) | zzzz[3];
}

static void
s8z24_read_quad_stencil(struct softpipe_surface *sps,
                        int x, int y, ubyte ssss[QUAD_SIZE])
{
   const unsigned *src
      = ((unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_S8_Z24);

   ssss[0] = src[0] >> 24;
   ssss[1] = src[1] >> 24;
   src += sps->surface.region->pitch;
   ssss[2] = src[0] >> 24;
   ssss[3] = src[1] >> 24;
}

static void
s8z24_write_quad_stencil(struct softpipe_surface *sps,
                         int x, int y, const ubyte ssss[QUAD_SIZE])
{
   static const unsigned mask = 0x00ffffff;
   unsigned *dst
      = ((unsigned *) (sps->surface.region->map + sps->surface.offset))
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_S8_Z24);

   dst[0] = (dst[0] & mask) | (ssss[0] << 24);
   dst[1] = (dst[1] & mask) | (ssss[1] << 24);
   dst += sps->surface.region->pitch;
   dst[0] = (dst[0] & mask) | (ssss[2] << 24);
   dst[1] = (dst[1] & mask) | (ssss[3] << 24);
}


/**
 * Return Z component as float in [0,1].  Stencil part ignored.
 */
static void
s8z24_get_tile(struct pipe_surface *ps,
               unsigned x, unsigned y, unsigned w, unsigned h, float *p)
{
   const uint *src
      = ((const uint *) (ps->region->map + ps->offset))
      + y * ps->region->pitch + x;
   const double scale = 1.0 / ((1 << 24) - 1);
   unsigned i, j;
   unsigned w0 = w;

   assert(ps->format == PIPE_FORMAT_S8_Z24);

   CLIP_TILE;

   for (i = 0; i < h; i++) {
      float *pRow = p;
      for (j = 0; j < w; j++) {
         pRow[j] = (src[j] & 0xffffff) * scale;
      }
      src += ps->region->pitch;
      p += w0;
   }
}


/*** PIPE_FORMAT_U_S8 ***/

static void
s8_read_quad_stencil(struct softpipe_surface *sps,
                     int x, int y, ubyte ssss[QUAD_SIZE])
{
   const ubyte *src
      = sps->surface.region->map + sps->surface.offset
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_U_S8);

   ssss[0] = src[0];
   ssss[1] = src[1];
   src += sps->surface.region->pitch;
   ssss[2] = src[0];
   ssss[3] = src[1];
}

static void
s8_write_quad_stencil(struct softpipe_surface *sps,
                      int x, int y, const ubyte ssss[QUAD_SIZE])
{
   ubyte *dst
      = sps->surface.region->map + sps->surface.offset
      + y * sps->surface.region->pitch + x;

   assert(sps->surface.format == PIPE_FORMAT_U_S8);

   dst[0] = ssss[0];
   dst[1] = ssss[1];
   dst += sps->surface.region->pitch;
   dst[0] = ssss[2];
   dst[1] = ssss[3];
}


/**
 * Initialize the quad_read/write and get/put_tile() methods.
 */
void
softpipe_init_surface_funcs(struct softpipe_surface *sps)
{
   sps->tc = sp_create_tile_cache();

   assert(sps->surface.format);

   switch (sps->surface.format) {
   case PIPE_FORMAT_U_A8_R8_G8_B8:
      sps->read_quad_f_swz = a8r8g8b8_read_quad_f_swz;
      sps->write_quad_f_swz = a8r8g8b8_write_quad_f_swz;
      sps->surface.get_tile = a8r8g8b8_get_tile;
      sps->surface.put_tile = a8r8g8b8_put_tile;
      break;
   case PIPE_FORMAT_U_A1_R5_G5_B5:
      sps->surface.get_tile = a1r5g5b5_get_tile;
      break;
   case PIPE_FORMAT_U_L8:
      sps->read_quad_f_swz = l8_read_quad_f_swz;
      sps->write_quad_f_swz = l8_write_quad_f_swz;
      sps->surface.get_tile = l8_get_tile;
      break;
   case PIPE_FORMAT_U_A8:
      sps->read_quad_f_swz = a8_read_quad_f_swz;
      sps->write_quad_f_swz = a8_write_quad_f_swz;
      sps->surface.get_tile = a8_get_tile;
      break;
   case PIPE_FORMAT_U_I8:
      sps->read_quad_f_swz = i8_read_quad_f_swz;
      sps->write_quad_f_swz = i8_write_quad_f_swz;
      sps->surface.get_tile = i8_get_tile;
      break;
   case PIPE_FORMAT_U_A8_L8:
      sps->read_quad_f_swz = a8_l8_read_quad_f_swz;
      sps->write_quad_f_swz = a8_l8_write_quad_f_swz;
      sps->surface.get_tile = a8_l8_get_tile;
      break;

   case PIPE_FORMAT_S_R16_G16_B16_A16:
      sps->surface.get_tile = r16g16b16a16_get_tile;
      sps->surface.put_tile = r16g16b16a16_put_tile;
      break;

   case PIPE_FORMAT_U_Z16:
      sps->read_quad_z = z16_read_quad_z;
      sps->write_quad_z = z16_write_quad_z;
      sps->surface.get_tile = z16_get_tile;
      break;
   case PIPE_FORMAT_U_Z32:
      sps->read_quad_z = z32_read_quad_z;
      sps->write_quad_z = z32_write_quad_z;
      sps->surface.get_tile = z32_get_tile;
      break;
   case PIPE_FORMAT_S8_Z24:
      sps->read_quad_z = s8z24_read_quad_z;
      sps->write_quad_z = s8z24_write_quad_z;
      sps->read_quad_stencil = s8z24_read_quad_stencil;
      sps->write_quad_stencil = s8z24_write_quad_stencil;
      sps->surface.get_tile = s8z24_get_tile;
      break;

   case PIPE_FORMAT_U_S8:
      sps->read_quad_stencil = s8_read_quad_stencil;
      sps->write_quad_stencil = s8_write_quad_stencil;
      break;
   default:
      assert(0);
   }
}


static struct pipe_surface *
softpipe_surface_alloc(struct pipe_context *pipe, unsigned pipeFormat)
{
   struct softpipe_surface *sps = CALLOC_STRUCT(softpipe_surface);
   if (!sps)
      return NULL;

   assert(pipeFormat < PIPE_FORMAT_COUNT);

   sps->surface.format = pipeFormat;
   sps->surface.refcount = 1;
   softpipe_init_surface_funcs(sps);

   return &sps->surface;
}





/**
 * Called via pipe->get_tex_surface()
 * XXX is this in the right place?
 */
struct pipe_surface *
softpipe_get_tex_surface(struct pipe_context *pipe,
                         struct pipe_mipmap_tree *mt,
                         unsigned face, unsigned level, unsigned zslice)
{
   struct pipe_surface *ps;
   unsigned offset;  /* in bytes */

   offset = mt->level[level].level_offset;

   if (mt->target == PIPE_TEXTURE_CUBE) {
      offset += mt->level[level].image_offset[face] * mt->cpp;
   }
   else if (mt->target == PIPE_TEXTURE_3D) {
      offset += mt->level[level].image_offset[zslice] * mt->cpp;
   }
   else {
      assert(face == 0);
      assert(zslice == 0);
   }

   ps = pipe->surface_alloc(pipe, mt->format);
   if (ps) {
      assert(ps->format);
      assert(ps->refcount);
      pipe_region_reference(&ps->region, mt->region);
      ps->width = mt->level[level].width;
      ps->height = mt->level[level].height;
      ps->offset = offset;
   }
   return ps;
}


void
sp_init_surface_functions(struct softpipe_context *sp)
{
   sp->pipe.surface_alloc = softpipe_surface_alloc;
}