summaryrefslogtreecommitdiffstats
blob: 236a2c96ef906b440e60c8db62c387b131743a68 (plain) (blame)
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
/* libs/pixelflinger/t32cb16blend.S
**
** Copyright 2010, The Android Open Source Project
**
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
**
**     http://www.apache.org/licenses/LICENSE-2.0
**
** Unless required by applicable law or agreed to in writing, software
** distributed under the License is distributed on an "AS IS" BASIS,
** WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
** See the License for the specific language governing permissions and
** limitations under the License.
*/

#ifdef DEBUG
#define DBG
#else
#define DBG #
#endif

/*
 * blend one of 2 16bpp RGB pixels held in dreg selected by shift
 * with the 32bpp ABGR pixel held in src and store the result in fb
 *
 * Assumes that the dreg data is little endian and that
 * the the second pixel (shift==16) will be merged into
 * the fb result
 *
 * Uses $t0,$t6,$t7,$t8
 */

#if __mips==32 && __mips_isa_rev>=2
    .macro pixel dreg src fb shift
    /*
     * sA = s >> 24
     * f = 0x100 - (sA + (sA>>7))
     */
DBG .set    noat
DBG rdhwr   $at,$2
DBG .set    at

    srl  $t7,\src,24
    srl  $t6,$t7,7
    addu $t7,$t6
    li   $t6,0x100
    subu $t7,$t6,$t7

    /* red */
    ext  $t8,\dreg,\shift+6+5,5         # dst[\shift:15..11]
    mul  $t6,$t8,$t7
    ext  $t0,\dreg,\shift+5,6           # start green extraction dst[\shift:10..5]
    ext  $t8,\src,3,5               # src[7..3]
    srl  $t6,8
    addu $t8,$t6
.if \shift!=0
    sll  $t8,\shift+11
    or   \fb,$t8
.else
    sll  \fb,$t8,11
.endif

    /* green */
    mul  $t8,$t0,$t7
    ext  $t0,\dreg,\shift,5         # start blue extraction dst[\shift:4..0]
    ext  $t6,\src,2+8,6             # src[15..10]
    srl  $t8,8
    addu $t8,$t6

    /* blue */
    mul  $t0,$t0,$t7
    sll  $t8, $t8, \shift+5
    or   \fb, \fb, $t8
    ext  $t6,\src,(3+8+8),5
    srl  $t8,$t0,8
    addu $t8,$t6
    sll  $t8, $t8, \shift
    or   \fb, \fb, $t8

DBG .set    noat
DBG rdhwr $t8,$2
DBG subu  $t8,$at
DBG sltu  $at,$t8,$v0
DBG movn  $v0,$t8,$at
DBG sgtu  $at,$t8,$v1
DBG movn  $v1,$t8,$at
DBG .set    at
    .endm

#else

    .macro pixel dreg src fb shift
    /*
     * sA = s >> 24
     * f = 0x100 - (sA + (sA>>7))
     */
DBG .set    push
DBG .set    noat
DBG .set    mips32r2
DBG rdhwr   $at,$2
DBG .set    pop

    srl  $t7,\src,24
    srl  $t6,$t7,7
    addu $t7,$t6
    li   $t6,0x100
    subu $t7,$t6,$t7

    /*
     * red
     * dR = (d >> (6 + 5)) & 0x1f;
     * dR = (f*dR)>>8
     * sR = (s >> (   3)) & 0x1f;
     * sR += dR
     * fb |= sR << 11
     */
    srl  $t8,\dreg,\shift+6+5
.if \shift==0
    and  $t8,0x1f
.endif
    mul  $t8,$t8,$t7
    srl  $t6,\src,3
    and  $t6,0x1f
    srl  $t8,8
    addu $t8,$t6
.if \shift!=0
    sll  $t8,\shift+11
    or   \fb,$t8
.else
    sll  \fb,$t8,11
.endif

        /*
     * green
     * dG = (d >> 5) & 0x3f
     * dG = (f*dG) >> 8
     * sG = (s >> ( 8+2))&0x3F;
     */
    srl  $t8,\dreg,\shift+5
    and  $t8,0x3f
    mul  $t8,$t8,$t7
    srl  $t6,\src,8+2
    and  $t6,0x3f
    srl  $t8,8
    addu $t8,$t6
    sll  $t8,\shift + 5
    or   \fb,$t8

    /* blue */
.if \shift!=0
    srl  $t8,\dreg,\shift
    and  $t8,0x1f
.else
    and  $t8,\dreg,0x1f
.endif
    mul  $t8,$t8,$t7
    srl  $t6,\src,(8+8+3)
    and  $t6,0x1f
    srl  $t8,8
    addu $t8,$t6
.if \shift!=0
    sll  $t8,\shift
.endif
    or   \fb,$t8
DBG .set    push
DBG .set    noat
DBG .set    mips32r2
DBG rdhwr   $t8,$2
DBG subu    $t8,$at
DBG sltu    $at,$t8,$v0
DBG movn    $v0,$t8,$at
DBG sgtu    $at,$t8,$v1
DBG movn    $v1,$t8,$at
DBG .set    pop
    .endm
#endif

    .text
    .align

    .global scanline_t32cb16blend_mips
    .ent    scanline_t32cb16blend_mips
scanline_t32cb16blend_mips:
DBG li    $v0,0xffffffff
DBG li    $v1,0
    /* Align the destination if necessary */
    and   $t0,$a0,3
    beqz  $t0,aligned

    /* as long as there is at least one pixel */
    beqz  $a2,done

    lw    $t4,($a1)
    addu  $a0,2
    addu  $a1,4
    beqz  $t4,1f
    lhu   $t3,-2($a0)
    pixel $t3,$t4,$t1,0
    sh    $t1,-2($a0)
1:  subu  $a2,1

aligned:
    /* Check to see if its worth unrolling the loop */
    subu  $a2,4
    bltz  $a2,tail

    /* Process 4 pixels at a time */
fourpixels:
    /* 1st pair of pixels */
    lw    $t4,0($a1)
    lw    $t5,4($a1)
    addu  $a0,8
    addu  $a1,16

    /* both are zero, skip this pair */
    or    $t3,$t4,$t5
    beqz  $t3,1f

    /* load the destination */
    lw    $t3,-8($a0)

    pixel $t3,$t4,$t1,0
    andi  $t1, 0xFFFF
    pixel $t3,$t5,$t1,16
    sw    $t1,-8($a0)

1:
    /* 2nd pair of pixels */
    lw    $t4,-8($a1)
    lw    $t5,-4($a1)

    /* both are zero, skip this pair */
    or    $t3,$t4,$t5
    beqz  $t3,1f

    /* load the destination */
    lw    $t3,-4($a0)

    pixel $t3,$t4,$t1,0
    andi  $t1, 0xFFFF
    pixel $t3,$t5,$t1,16
    sw    $t1,-4($a0)

1:  subu  $a2,4
    bgtz  $a2,fourpixels

tail:
    /* the pixel count underran, restore it now */
    addu  $a2,4

    /* handle the last 0..3 pixels */
    beqz  $a2,done
onepixel:
    lw    $t4,($a1)
    addu  $a0,2
    addu  $a1,4
    beqz  $t4,1f
    lhu   $t3,-2($a0)
    pixel $t3,$t4,$t1,0
    sh    $t1,-2($a0)
1:  subu  $a2,1
    bnez  $a2,onepixel
done:
DBG .set    push
DBG .set    mips32r2
DBG rdhwr   $a0,$3
DBG mul     $v0,$a0
DBG mul     $v1,$a0
DBG .set    pop
    j     $ra
    .end    scanline_t32cb16blend_mips