aboutsummaryrefslogtreecommitdiff
path: root/mesalib/src/mesa/math/m_clip_tmp.h
blob: c6b8bd83d966cc012b3eb1478210decf33084217 (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
/*
 * Mesa 3-D graphics library
 * Version:  6.2
 *
 * Copyright (C) 1999-2004  Brian Paul   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, 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 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.
 */

/*
 * New (3.1) transformation code written by Keith Whitwell.
 */


/* KW: a clever asm implementation would nestle integer versions
 * of the outcode calculation underneath the division.  Gcc won't
 * do this, strangely enough, so I only do the divide in
 * the case where the cliptest passes.  This isn't essential,
 * and an asm implementation needn't replicate that behaviour.
 *
 * \param clip_vec vector of incoming clip-space coords
 * \param proj_vec vector of resultant NDC-space projected coords
 * \param clipMask resulting array of clip flags
 * \param orMask bitwise-OR of clipMask values
 * \param andMask bitwise-AND of clipMask values
 * \return proj_vec pointer
 */
static GLvector4f * _XFORMAPI TAG(cliptest_points4)( GLvector4f *clip_vec,
                                                     GLvector4f *proj_vec,
                                                     GLubyte clipMask[],
                                                     GLubyte *orMask,
                                                     GLubyte *andMask,
						     GLboolean viewport_z_clip )
{
   const GLuint stride = clip_vec->stride;
   const GLfloat *from = (GLfloat *)clip_vec->start;
   const GLuint count = clip_vec->count;
   GLuint c = 0;
   GLfloat (*vProj)[4] = (GLfloat (*)[4])proj_vec->start;
   GLubyte tmpAndMask = *andMask;
   GLubyte tmpOrMask = *orMask;
   GLuint i;
   STRIDE_LOOP {
      const GLfloat cx = from[0];
      const GLfloat cy = from[1];
      const GLfloat cz = from[2];
      const GLfloat cw = from[3];
#if defined(macintosh) || defined(__powerpc__)
      /* on powerpc cliptest is 17% faster in this way. */
      GLuint mask;
      mask = (((cw < cx) << CLIP_RIGHT_SHIFT));
      mask |= (((cw < -cx) << CLIP_LEFT_SHIFT));
      mask |= (((cw < cy) << CLIP_TOP_SHIFT));
      mask |= (((cw < -cy) << CLIP_BOTTOM_SHIFT));
      if (viewport_z_clip) {
	 mask |= (((cw < cz) << CLIP_FAR_SHIFT));
	 mask |= (((cw < -cz) << CLIP_NEAR_SHIFT));
      }
#else /* !defined(macintosh)) */
      GLubyte mask = 0;
      if (-cx + cw < 0) mask |= CLIP_RIGHT_BIT;
      if ( cx + cw < 0) mask |= CLIP_LEFT_BIT;
      if (-cy + cw < 0) mask |= CLIP_TOP_BIT;
      if ( cy + cw < 0) mask |= CLIP_BOTTOM_BIT;
      if (viewport_z_clip) {
	 if (-cz + cw < 0) mask |= CLIP_FAR_BIT;
	 if ( cz + cw < 0) mask |= CLIP_NEAR_BIT;
      }
#endif /* defined(macintosh) */

      clipMask[i] = mask;
      if (mask) {
	 c++;
	 tmpAndMask &= mask;
	 tmpOrMask |= mask;
	 vProj[i][0] = 0;
	 vProj[i][1] = 0;
	 vProj[i][2] = 0;
	 vProj[i][3] = 1;
      } else {
	 GLfloat oow = 1.0F / cw;
	 vProj[i][0] = cx * oow;
	 vProj[i][1] = cy * oow;
	 vProj[i][2] = cz * oow;
	 vProj[i][3] = oow;
      }
   }

   *orMask = tmpOrMask;
   *andMask = (GLubyte) (c < count ? 0 : tmpAndMask);

   proj_vec->flags |= VEC_SIZE_4;
   proj_vec->size = 4;
   proj_vec->count = clip_vec->count;
   return proj_vec;
}



/*
 * \param clip_vec vector of incoming clip-space coords
 * \param proj_vec vector of resultant NDC-space projected coords
 * \param clipMask resulting array of clip flags
 * \param orMask bitwise-OR of clipMask values
 * \param andMask bitwise-AND of clipMask values
 * \return clip_vec pointer
 */
static GLvector4f * _XFORMAPI TAG(cliptest_np_points4)( GLvector4f *clip_vec,
							GLvector4f *proj_vec,
							GLubyte clipMask[],
							GLubyte *orMask,
							GLubyte *andMask,
							GLboolean viewport_z_clip )
{
   const GLuint stride = clip_vec->stride;
   const GLuint count = clip_vec->count;
   const GLfloat *from = (GLfloat *)clip_vec->start;
   GLuint c = 0;
   GLubyte tmpAndMask = *andMask;
   GLubyte tmpOrMask = *orMask;
   GLuint i;
   (void) proj_vec;
   STRIDE_LOOP {
      const GLfloat cx = from[0];
      const GLfloat cy = from[1];
      const GLfloat cz = from[2];
      const GLfloat cw = from[3];
#if defined(macintosh) || defined(__powerpc__)
      /* on powerpc cliptest is 17% faster in this way. */
      GLuint mask;
      mask = (((cw < cx) << CLIP_RIGHT_SHIFT));
      mask |= (((cw < -cx) << CLIP_LEFT_SHIFT));
      mask |= (((cw < cy) << CLIP_TOP_SHIFT));
      mask |= (((cw < -cy) << CLIP_BOTTOM_SHIFT));
      if (viewport_z_clip) {
	 mask |= (((cw < cz) << CLIP_FAR_SHIFT));
	 mask |= (((cw < -cz) << CLIP_NEAR_SHIFT));
      }
#else /* !defined(macintosh)) */
      GLubyte mask = 0;
      if (-cx + cw < 0) mask |= CLIP_RIGHT_BIT;
      if ( cx + cw < 0) mask |= CLIP_LEFT_BIT;
      if (-cy + cw < 0) mask |= CLIP_TOP_BIT;
      if ( cy + cw < 0) mask |= CLIP_BOTTOM_BIT;
      if (viewport_z_clip) {
	 if (-cz + cw < 0) mask |= CLIP_FAR_BIT;
	 if ( cz + cw < 0) mask |= CLIP_NEAR_BIT;
      }
#endif /* defined(macintosh) */

      clipMask[i] = mask;
      if (mask) {
	 c++;
	 tmpAndMask &= mask;
	 tmpOrMask |= mask;
      }
   }

   *orMask = tmpOrMask;
   *andMask = (GLubyte) (c < count ? 0 : tmpAndMask);
   return clip_vec;
}


static GLvector4f * _XFORMAPI TAG(cliptest_points3)( GLvector4f *clip_vec,
                                                     GLvector4f *proj_vec,
                                                     GLubyte clipMask[],
                                                     GLubyte *orMask,
                                                     GLubyte *andMask,
						     GLboolean viewport_z_clip )
{
   const GLuint stride = clip_vec->stride;
   const GLuint count = clip_vec->count;
   const GLfloat *from = (GLfloat *)clip_vec->start;
   GLubyte tmpOrMask = *orMask;
   GLubyte tmpAndMask = *andMask;
   GLuint i;
   (void) proj_vec;
   STRIDE_LOOP {
      const GLfloat cx = from[0], cy = from[1], cz = from[2];
      GLubyte mask = 0;
      if (cx >  1.0)       mask |= CLIP_RIGHT_BIT;
      else if (cx < -1.0)  mask |= CLIP_LEFT_BIT;
      if (cy >  1.0)       mask |= CLIP_TOP_BIT;
      else if (cy < -1.0)  mask |= CLIP_BOTTOM_BIT;
      if (viewport_z_clip) {
	 if (cz >  1.0)       mask |= CLIP_FAR_BIT;
	 else if (cz < -1.0)  mask |= CLIP_NEAR_BIT;
      }
      clipMask[i] = mask;
      tmpOrMask |= mask;
      tmpAndMask &= mask;
   }

   *orMask = tmpOrMask;
   *andMask = tmpAndMask;
   return clip_vec;
}


static GLvector4f * _XFORMAPI TAG(cliptest_points2)( GLvector4f *clip_vec,
                                                     GLvector4f *proj_vec,
                                                     GLubyte clipMask[],
                                                     GLubyte *orMask,
                                                     GLubyte *andMask,
						     GLboolean viewport_z_clip )
{
   const GLuint stride = clip_vec->stride;
   const GLuint count = clip_vec->count;
   const GLfloat *from = (GLfloat *)clip_vec->start;
   GLubyte tmpOrMask = *orMask;
   GLubyte tmpAndMask = *andMask;
   GLuint i;
   (void) proj_vec;
   STRIDE_LOOP {
      const GLfloat cx = from[0], cy = from[1];
      GLubyte mask = 0;
      if (cx >  1.0)       mask |= CLIP_RIGHT_BIT;
      else if (cx < -1.0)  mask |= CLIP_LEFT_BIT;
      if (cy >  1.0)       mask |= CLIP_TOP_BIT;
      else if (cy < -1.0)  mask |= CLIP_BOTTOM_BIT;
      clipMask[i] = mask;
      tmpOrMask |= mask;
      tmpAndMask &= mask;
   }

   *orMask = tmpOrMask;
   *andMask = tmpAndMask;
   return clip_vec;
}


void TAG(init_c_cliptest)( void )
{
   _mesa_clip_tab[4] = TAG(cliptest_points4);
   _mesa_clip_tab[3] = TAG(cliptest_points3);
   _mesa_clip_tab[2] = TAG(cliptest_points2);

   _mesa_clip_np_tab[4] = TAG(cliptest_np_points4);
   _mesa_clip_np_tab[3] = TAG(cliptest_points3);
   _mesa_clip_np_tab[2] = TAG(cliptest_points2);
}