FFmpeg
cabac.h
Go to the documentation of this file.
1 /*
2  * Loongson SIMD optimized h264chroma
3  *
4  * Copyright (c) 2018 Loongson Technology Corporation Limited
5  * Contributed by Shiyou Yin <yinshiyou-hf@loongson.cn>
6  * Gu Xiwei(guxiwei-hf@loongson.cn)
7  *
8  * This file is part of FFmpeg.
9  *
10  * FFmpeg is free software; you can redistribute it and/or
11  * modify it under the terms of the GNU Lesser General Public
12  * License as published by the Free Software Foundation; either
13  * version 2.1 of the License, or (at your option) any later version.
14  *
15  * FFmpeg is distributed in the hope that it will be useful,
16  * but WITHOUT ANY WARRANTY; without even the implied warranty of
17  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18  * Lesser General Public License for more details.
19  *
20  * You should have received a copy of the GNU Lesser General Public
21  * License along with FFmpeg; if not, write to the Free Software
22  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
23  */
24 
25 #ifndef AVCODEC_MIPS_CABAC_H
26 #define AVCODEC_MIPS_CABAC_H
27 
28 #include "libavutil/attributes.h"
29 #include "libavcodec/cabac.h"
31 #include "config.h"
32 
33 #define get_cabac_inline get_cabac_inline_mips
35  uint8_t * const state){
36  mips_reg tmp0, tmp1, tmp2, bit;
37 
38  __asm__ volatile (
39  "lbu %[bit], 0(%[state]) \n\t"
40  "and %[tmp0], %[c_range], 0xC0 \n\t"
41  PTR_SLL "%[tmp0], %[tmp0], 0x01 \n\t"
42  PTR_ADDU "%[tmp0], %[tmp0], %[tables] \n\t"
43  PTR_ADDU "%[tmp0], %[tmp0], %[bit] \n\t"
44  /* tmp1: RangeLPS */
45  "lbu %[tmp1], %[lps_off](%[tmp0]) \n\t"
46 
47  PTR_SUBU "%[c_range], %[c_range], %[tmp1] \n\t"
48  PTR_SLL "%[tmp0], %[c_range], 0x11 \n\t"
49  "slt %[tmp2], %[tmp0], %[c_low] \n\t"
50  "beqz %[tmp2], 1f \n\t"
51  "move %[c_range], %[tmp1] \n\t"
52  "not %[bit], %[bit] \n\t"
53  PTR_SUBU "%[c_low], %[c_low], %[tmp0] \n\t"
54 
55  "1: \n\t"
56  /* tmp1: *state */
57  PTR_ADDU "%[tmp0], %[tables], %[bit] \n\t"
58  "lbu %[tmp1], %[mlps_off](%[tmp0]) \n\t"
59  /* tmp2: lps_mask */
60  PTR_ADDU "%[tmp0], %[tables], %[c_range] \n\t"
61  "lbu %[tmp2], %[norm_off](%[tmp0]) \n\t"
62 
63  "sb %[tmp1], 0(%[state]) \n\t"
64  "and %[bit], %[bit], 0x01 \n\t"
65  PTR_SLL "%[c_range], %[c_range], %[tmp2] \n\t"
66  PTR_SLL "%[c_low], %[c_low], %[tmp2] \n\t"
67 
68  "and %[tmp1], %[c_low], %[cabac_mask] \n\t"
69  "bnez %[tmp1], 1f \n\t"
70  PTR_ADDIU "%[tmp0], %[c_low], -0X01 \n\t"
71  "xor %[tmp0], %[c_low], %[tmp0] \n\t"
72  PTR_SRA "%[tmp0], %[tmp0], 0x0f \n\t"
73  PTR_ADDU "%[tmp0], %[tmp0], %[tables] \n\t"
74  /* tmp2: ff_h264_norm_shift[x >> (CABAC_BITS - 1)] */
75  "lbu %[tmp2], %[norm_off](%[tmp0]) \n\t"
76 #if HAVE_BIGENDIAN
77  "lhu %[tmp0], 0(%[c_bytestream]) \n\t"
78 #else
79  "lhu %[tmp0], 0(%[c_bytestream]) \n\t"
80 #if HAVE_MIPS32R2 || HAVE_MIPS64R2
81  "wsbh %[tmp0], %[tmp0] \n\t"
82 #else
83  "and %[tmp1], %[tmp0], 0xff00ff00 \n\t"
84  "srl %[tmp1], %[tmp1], 8 \n\t"
85  "and %[tmp0], %[tmp0], 0x00ff00ff \n\t"
86  "sll %[tmp0], %[tmp0], 8 \n\t"
87  "or %[tmp0], %[tmp0], %[tmp1] \n\t"
88 #endif
89 #endif
90  PTR_SLL "%[tmp0], %[tmp0], 0x01 \n\t"
91  PTR_SUBU "%[tmp0], %[tmp0], %[cabac_mask] \n\t"
92 
93  "li %[tmp1], 0x07 \n\t"
94  PTR_SUBU "%[tmp1], %[tmp1], %[tmp2] \n\t"
95  PTR_SLL "%[tmp0], %[tmp0], %[tmp1] \n\t"
96  PTR_ADDU "%[c_low], %[c_low], %[tmp0] \n\t"
97 
98 #if UNCHECKED_BITSTREAM_READER
99  PTR_ADDIU "%[c_bytestream], %[c_bytestream], 0x02 \n\t"
100 #else
101  "slt %[tmp0], %[c_bytestream], %[c_bytestream_end] \n\t"
102  PTR_ADDIU "%[tmp2], %[c_bytestream], 0x02 \n\t"
103  "movn %[c_bytestream], %[tmp2], %[tmp0] \n\t"
104 #endif
105  "1: \n\t"
106  : [bit]"=&r"(bit), [tmp0]"=&r"(tmp0), [tmp1]"=&r"(tmp1), [tmp2]"=&r"(tmp2),
107  [c_range]"+&r"(c->range), [c_low]"+&r"(c->low),
108  [c_bytestream]"+&r"(c->bytestream)
109  : [state]"r"(state), [tables]"r"(ff_h264_cabac_tables),
111  [c_bytestream_end]"r"(c->bytestream_end),
112 #endif
113  [lps_off]"i"(H264_LPS_RANGE_OFFSET),
114  [mlps_off]"i"(H264_MLPS_STATE_OFFSET + 128),
115  [norm_off]"i"(H264_NORM_SHIFT_OFFSET),
116  [cabac_mask]"r"(CABAC_MASK)
117  : "memory"
118  );
119 
120  return bit;
121 }
122 
123 #define get_cabac_bypass get_cabac_bypass_mips
125 {
126  mips_reg tmp0, tmp1;
127  int res = 0;
128  __asm__ volatile(
129  PTR_SLL "%[c_low], %[c_low], 0x01 \n\t"
130  "and %[tmp0], %[c_low], %[cabac_mask] \n\t"
131  "bnez %[tmp0], 1f \n\t"
132 #if HAVE_BIGENDIAN
133  "lhu %[tmp1], 0(%[c_bytestream]) \n\t"
134 #else
135  "lhu %[tmp1], 0(%[c_bytestream]) \n\t"
136 #if HAVE_MIPS32R2 || HAVE_MIPS64R2
137  "wsbh %[tmp1], %[tmp1] \n\t"
138 #else
139  "and %[tmp0], %[tmp1], 0xff00ff00 \n\t"
140  "srl %[tmp0], %[tmp0], 8 \n\t"
141  "and %[tmp1], %[tmp1], 0x00ff00ff \n\t"
142  "sll %[tmp1], %[tmp1], 8 \n\t"
143  "or %[tmp1], %[tmp1], %[tmp0] \n\t"
144 #endif
145 #endif
146  PTR_SLL "%[tmp1], %[tmp1], 0x01 \n\t"
147  PTR_SUBU "%[tmp1], %[tmp1], %[cabac_mask] \n\t"
148  PTR_ADDU "%[c_low], %[c_low], %[tmp1] \n\t"
149 #if UNCHECKED_BITSTREAM_READER
150  PTR_ADDIU "%[c_bytestream], %[c_bytestream], 0x02 \n\t"
151 #else
152  "slt %[tmp0], %[c_bytestream], %[c_bytestream_end] \n\t"
153  PTR_ADDIU "%[tmp1], %[c_bytestream], 0x02 \n\t"
154  "movn %[c_bytestream], %[tmp1], %[tmp0] \n\t"
155 #endif
156  "1: \n\t"
157  PTR_SLL "%[tmp1], %[c_range], 0x11 \n\t"
158  "slt %[tmp0], %[c_low], %[tmp1] \n\t"
159  PTR_SUBU "%[tmp1], %[c_low], %[tmp1] \n\t"
160  "movz %[res], %[one], %[tmp0] \n\t"
161  "movz %[c_low], %[tmp1], %[tmp0] \n\t"
162  : [tmp0]"=&r"(tmp0), [tmp1]"=&r"(tmp1), [res]"+&r"(res),
163  [c_range]"+&r"(c->range), [c_low]"+&r"(c->low),
164  [c_bytestream]"+&r"(c->bytestream)
165  : [cabac_mask]"r"(CABAC_MASK),
167  [c_bytestream_end]"r"(c->bytestream_end),
168 #endif
169  [one]"r"(0x01)
170  : "memory"
171  );
172  return res;
173 }
174 
175 #define get_cabac_bypass_sign get_cabac_bypass_sign_mips
177 {
178  mips_reg tmp0, tmp1;
179  int res = val;
180  __asm__ volatile(
181  PTR_SLL "%[c_low], %[c_low], 0x01 \n\t"
182  "and %[tmp0], %[c_low], %[cabac_mask] \n\t"
183  "bnez %[tmp0], 1f \n\t"
184 #if HAVE_BIGENDIAN
185  "lhu %[tmp1], 0(%[c_bytestream]) \n\t"
186 #else
187  "lhu %[tmp1], 0(%[c_bytestream]) \n\t"
188 #if HAVE_MIPS32R2 || HAVE_MIPS64R2
189  "wsbh %[tmp1], %[tmp1] \n\t"
190 #else
191  "and %[tmp0], %[tmp1], 0xff00ff00 \n\t"
192  "srl %[tmp0], %[tmp0], 8 \n\t"
193  "and %[tmp1], %[tmp1], 0x00ff00ff \n\t"
194  "sll %[tmp1], %[tmp1], 8 \n\t"
195  "or %[tmp1], %[tmp1], %[tmp0] \n\t"
196 #endif
197 #endif
198  PTR_SLL "%[tmp1], %[tmp1], 0x01 \n\t"
199  PTR_SUBU "%[tmp1], %[tmp1], %[cabac_mask] \n\t"
200  PTR_ADDU "%[c_low], %[c_low], %[tmp1] \n\t"
201 #if UNCHECKED_BITSTREAM_READER
202  PTR_ADDIU "%[c_bytestream], %[c_bytestream], 0x02 \n\t"
203 #else
204  "slt %[tmp0], %[c_bytestream], %[c_bytestream_end] \n\t"
205  PTR_ADDIU "%[tmp1], %[c_bytestream], 0x02 \n\t"
206  "movn %[c_bytestream], %[tmp1], %[tmp0] \n\t"
207 #endif
208  "1: \n\t"
209  PTR_SLL "%[tmp1], %[c_range], 0x11 \n\t"
210  "slt %[tmp0], %[c_low], %[tmp1] \n\t"
211  PTR_SUBU "%[tmp1], %[c_low], %[tmp1] \n\t"
212  "movz %[c_low], %[tmp1], %[tmp0] \n\t"
213  PTR_SUBU "%[tmp1], %[zero], %[res] \n\t"
214  "movn %[res], %[tmp1], %[tmp0] \n\t"
215  : [tmp0]"=&r"(tmp0), [tmp1]"=&r"(tmp1), [res]"+&r"(res),
216  [c_range]"+&r"(c->range), [c_low]"+&r"(c->low),
217  [c_bytestream]"+&r"(c->bytestream)
218  : [cabac_mask]"r"(CABAC_MASK),
220  [c_bytestream_end]"r"(c->bytestream_end),
221 #endif
222  [zero]"r"(0x0)
223  : "memory"
224  );
225 
226  return res;
227 }
228 #endif /* AVCODEC_MIPS_CABAC_H */
cabac.h
PTR_SLL
#define PTR_SLL
Definition: asmdefs.h:57
get_cabac_bypass_sign_mips
static av_always_inline int get_cabac_bypass_sign_mips(CABACContext *c, int val)
Definition: cabac.h:176
mips_reg
#define mips_reg
Definition: asmdefs.h:46
bit
#define bit(string, value)
Definition: cbs_mpeg2.c:58
tables
Writing a table generator This documentation is preliminary Parts of the API are not good and should be changed Basic concepts A table generator consists of two *_tablegen c and *_tablegen h The h file will provide the variable declarations and initialization code for the tables
Definition: tablegen.txt:10
val
static double val(void *priv, double ch)
Definition: aeval.c:77
ff_h264_cabac_tables
const uint8_t ff_h264_cabac_tables[512+4 *2 *64+4 *64+63]
mmiutils.h
CABAC_MASK
#define CABAC_MASK
Definition: cabac.h:39
get_cabac_bypass_mips
static av_always_inline int get_cabac_bypass_mips(CABACContext *c)
Definition: cabac.h:124
c
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
Definition: undefined.txt:32
H264_LPS_RANGE_OFFSET
#define H264_LPS_RANGE_OFFSET
Definition: cabac.h:34
PTR_SUBU
#define PTR_SUBU
Definition: asmdefs.h:52
attributes.h
av_always_inline
#define av_always_inline
Definition: attributes.h:49
UNCHECKED_BITSTREAM_READER
#define UNCHECKED_BITSTREAM_READER
Definition: bitstream.h:57
H264_NORM_SHIFT_OFFSET
#define H264_NORM_SHIFT_OFFSET
Definition: cabac.h:33
__asm__
__asm__(".macro parse_r var r\n\t" "\\var = -1\n\t" _IFC_REG(0) _IFC_REG(1) _IFC_REG(2) _IFC_REG(3) _IFC_REG(4) _IFC_REG(5) _IFC_REG(6) _IFC_REG(7) _IFC_REG(8) _IFC_REG(9) _IFC_REG(10) _IFC_REG(11) _IFC_REG(12) _IFC_REG(13) _IFC_REG(14) _IFC_REG(15) _IFC_REG(16) _IFC_REG(17) _IFC_REG(18) _IFC_REG(19) _IFC_REG(20) _IFC_REG(21) _IFC_REG(22) _IFC_REG(23) _IFC_REG(24) _IFC_REG(25) _IFC_REG(26) _IFC_REG(27) _IFC_REG(28) _IFC_REG(29) _IFC_REG(30) _IFC_REG(31) ".iflt \\var\n\t" ".error \"Unable to parse register name \\r\"\n\t" ".endif\n\t" ".endm")
PTR_ADDU
#define PTR_ADDU
Definition: asmdefs.h:49
get_cabac_inline_mips
static av_always_inline int get_cabac_inline_mips(CABACContext *c, uint8_t *const state)
Definition: cabac.h:34
PTR_ADDIU
#define PTR_ADDIU
Definition: asmdefs.h:50
zero
#define zero
Definition: regdef.h:64
state
static struct @345 state
H264_MLPS_STATE_OFFSET
#define H264_MLPS_STATE_OFFSET
Definition: cabac.h:35
CABACContext
Definition: cabac.h:41
PTR_SRA
#define PTR_SRA
Definition: asmdefs.h:55