FFmpeg
llvidencdsp.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2016 Alexandra Hájková
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or modify
7  * it under the terms of the GNU General Public License as published by
8  * the Free Software Foundation; either version 2 of the License, or
9  * (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14  * GNU General Public License for more details.
15  *
16  * You should have received a copy of the GNU General Public License along
17  * with FFmpeg; if not, write to the Free Software Foundation, Inc.,
18  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
19  */
20 
21 #include <string.h>
22 
23 #include "libavutil/common.h"
24 #include "libavutil/intreadwrite.h"
25 #include "libavutil/mem_internal.h"
26 
28 
29 #include "checkasm.h"
30 
31 #define randomize_buffers(buf, size) \
32  do { \
33  for (size_t j = 0; j < size & ~3; j += 4) \
34  AV_WN32(buf + j, rnd()); \
35  for (size_t j = 0; j < size; ++j) \
36  buf[j] = rnd(); \
37  } while (0)
38 
39 static const struct {uint8_t w, h, s;} planes[] = {
40  {16,16,16}, {21,23,25}, {32,17,48}, {15,128,16}, {128,127,128}
41 };
42 
43 #define MAX_STRIDE 128
44 #define MAX_HEIGHT 127
45 
47 {
48  int i;
49  LOCAL_ALIGNED_32(uint8_t, dst0, [MAX_STRIDE]);
50  LOCAL_ALIGNED_32(uint8_t, dst1, [MAX_STRIDE]);
51  LOCAL_ALIGNED_32(uint8_t, src0, [MAX_STRIDE]);
52  LOCAL_ALIGNED_32(uint8_t, src1, [MAX_STRIDE]);
53  LOCAL_ALIGNED_32(uint8_t, src2, [MAX_STRIDE]);
54  LOCAL_ALIGNED_32(uint8_t, src3, [MAX_STRIDE]);
55 
56  declare_func(void, uint8_t *dst, const uint8_t *src1,
57  const uint8_t *src2, intptr_t w);
58 
59  memset(dst0, 0, MAX_STRIDE);
60  memset(dst1, 0, MAX_STRIDE);
62  memcpy(src1, src0, MAX_STRIDE);
64  memcpy(src3, src2, MAX_STRIDE);
65 
66  if (check_func(c->diff_bytes, "diff_bytes")) {
67  for (i = 0; i < 5; i ++) {
68  call_ref(dst0, src0, src2, planes[i].w);
69  call_new(dst1, src1, src3, planes[i].w);
70  if (memcmp(dst0, dst1, planes[i].w))
71  fail();
72  }
73  bench_new(dst1, src0, src2, planes[4].w);
74  }
75 }
76 
78 {
79  enum {
80  BUF_SIZE = MAX_STRIDE + 15 /* to test misalignment */
81  };
82  uint8_t dst_ref[BUF_SIZE], dst_new[BUF_SIZE];
83  uint8_t src1[BUF_SIZE], src2[BUF_SIZE];
84 
85  declare_func(void, uint8_t *dst, const uint8_t *src1,
86  const uint8_t *src2, intptr_t w,
87  int *left, int *left_top);
88 
89  if (check_func(c->sub_median_pred, "sub_median_pred")) {
90  size_t width = 1 + rnd() % MAX_STRIDE;
91  size_t offset = rnd() & 0xF;
92  int left_ref = rnd() & 0xFF, top_ref = rnd() & 0xFF;
93  int left_new = left_ref, top_new = top_ref;
94 
95  memset(dst_ref, 0, sizeof(dst_ref));
96  memset(dst_new, 0, sizeof(dst_new));
97 
98  randomize_buffers(src1, sizeof(src1));
99  randomize_buffers(src2, sizeof(src2));
100 
101  call_ref(dst_ref + offset, src1 + offset, src2 + offset, width, &left_ref, &top_ref);
102  call_new(dst_new + offset, src1 + offset, src2 + offset, width, &left_new, &top_new);
103  if (left_new != left_ref || top_ref != top_new ||
104  memcmp(dst_ref, dst_new, width + offset))
105  fail();
106  bench_new(dst_new, src1, src2, MAX_STRIDE, &left_new, &top_new);
107  }
108 }
109 
111 {
112  int i;
113  LOCAL_ALIGNED_32(uint8_t, dst0, [MAX_STRIDE * MAX_HEIGHT]);
114  LOCAL_ALIGNED_32(uint8_t, dst1, [MAX_STRIDE * MAX_HEIGHT]);
117 
118  declare_func(void, uint8_t *dst, const uint8_t *src,
119  ptrdiff_t stride, ptrdiff_t width, int height);
120 
121  memset(dst0, 0, MAX_STRIDE * MAX_HEIGHT);
122  memset(dst1, 0, MAX_STRIDE * MAX_HEIGHT);
124  memcpy(src1, src0, MAX_STRIDE * MAX_HEIGHT);
125 
126  if (check_func(c->sub_left_predict, "sub_left_predict")) {
127  for (i = 0; i < 5; i ++) {
128  call_ref(dst0, src0, planes[i].s, planes[i].w, planes[i].h);
129  call_new(dst1, src1, planes[i].s, planes[i].w, planes[i].h);
130  if (memcmp(dst0, dst1, planes[i].w * planes[i].h))
131  fail();
132  break;
133  }
134  bench_new(dst1, src0, planes[4].s, planes[4].w, planes[4].h);
135  }
136 }
137 
139 {
142 
144  report("diff_bytes");
145 
147  report("sub_median_pred");
148 
150  report("sub_left_predict");
151 }
mem_internal.h
src1
const pixel * src1
Definition: h264pred_template.c:420
check_func
#define check_func(func,...)
Definition: checkasm.h:205
MAX_HEIGHT
#define MAX_HEIGHT
Definition: llvidencdsp.c:44
call_ref
#define call_ref(...)
Definition: checkasm.h:220
check_sub_left_pred
static void check_sub_left_pred(LLVidEncDSPContext *c)
Definition: llvidencdsp.c:110
fail
#define fail()
Definition: checkasm.h:214
checkasm.h
rnd
#define rnd()
Definition: checkasm.h:198
checkasm_check_llvidencdsp
void checkasm_check_llvidencdsp(void)
Definition: llvidencdsp.c:138
intreadwrite.h
check_sub_median_pred
static void check_sub_median_pred(LLVidEncDSPContext *c)
Definition: llvidencdsp.c:77
call_new
#define call_new(...)
Definition: checkasm.h:323
LOCAL_ALIGNED_32
#define LOCAL_ALIGNED_32(t, v,...)
Definition: mem_internal.h:132
planes
static const struct @548 planes[]
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
ff_llvidencdsp_init
av_cold void ff_llvidencdsp_init(LLVidEncDSPContext *c)
Definition: lossless_videoencdsp.c:100
height
#define height
Definition: dsp.h:89
dst
uint8_t ptrdiff_t const uint8_t ptrdiff_t int intptr_t intptr_t int int16_t * dst
Definition: dsp.h:87
h
uint8_t h
Definition: llvidencdsp.c:39
offset
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf offset
Definition: writing_filters.txt:86
lossless_videoencdsp.h
report
#define report
Definition: checkasm.h:217
bench_new
#define bench_new(...)
Definition: checkasm.h:408
i
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:256
src2
const pixel * src2
Definition: h264pred_template.c:421
common.h
s
uint8_t s
Definition: llvidencdsp.c:39
LLVidEncDSPContext
Definition: lossless_videoencdsp.h:25
check_diff_bytes
static void check_diff_bytes(LLVidEncDSPContext *c)
Definition: llvidencdsp.c:46
MAX_STRIDE
#define MAX_STRIDE
Definition: llvidencdsp.c:43
BUF_SIZE
@ BUF_SIZE
BUF_SIZE is bigger than necessary in order to test strides > block width.
Definition: cavsdsp.c:34
left
Tag MUST be and< 10hcoeff half pel interpolation filter coefficients, hcoeff[0] are the 2 middle coefficients[1] are the next outer ones and so on, resulting in a filter like:...eff[2], hcoeff[1], hcoeff[0], hcoeff[0], hcoeff[1], hcoeff[2] ... the sign of the coefficients is not explicitly stored but alternates after each coeff and coeff[0] is positive, so ...,+,-,+,-,+,+,-,+,-,+,... hcoeff[0] is not explicitly stored but found by subtracting the sum of all stored coefficients with signs from 32 hcoeff[0]=32 - hcoeff[1] - hcoeff[2] - ... a good choice for hcoeff and htaps is htaps=6 hcoeff={40,-10, 2} an alternative which requires more computations at both encoder and decoder side and may or may not be better is htaps=8 hcoeff={42,-14, 6,-2}ref_frames minimum of the number of available reference frames and max_ref_frames for example the first frame after a key frame always has ref_frames=1spatial_decomposition_type wavelet type 0 is a 9/7 symmetric compact integer wavelet 1 is a 5/3 symmetric compact integer wavelet others are reserved stored as delta from last, last is reset to 0 if always_reset||keyframeqlog quality(logarithmic quantizer scale) stored as delta from last, last is reset to 0 if always_reset||keyframemv_scale stored as delta from last, last is reset to 0 if always_reset||keyframe FIXME check that everything works fine if this changes between framesqbias dequantization bias stored as delta from last, last is reset to 0 if always_reset||keyframeblock_max_depth maximum depth of the block tree stored as delta from last, last is reset to 0 if always_reset||keyframequant_table quantization tableHighlevel bitstream structure:==============================--------------------------------------------|Header|--------------------------------------------|------------------------------------|||Block0||||split?||||yes no||||......... intra?||||:Block01 :yes no||||:Block02 :....... ..........||||:Block03 ::y DC ::ref index:||||:Block04 ::cb DC ::motion x :||||......... :cr DC ::motion y :||||....... ..........|||------------------------------------||------------------------------------|||Block1|||...|--------------------------------------------|------------ ------------ ------------|||Y subbands||Cb subbands||Cr subbands||||--- ---||--- ---||--- ---|||||LL0||HL0||||LL0||HL0||||LL0||HL0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||LH0||HH0||||LH0||HH0||||LH0||HH0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HL1||LH1||||HL1||LH1||||HL1||LH1|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HH1||HL2||||HH1||HL2||||HH1||HL2|||||...||...||...|||------------ ------------ ------------|--------------------------------------------Decoding process:=================------------|||Subbands|------------||||------------|Intra DC||||LL0 subband prediction ------------|\ Dequantization ------------------- \||Reference frames|\ IDWT|------- -------|Motion \|||Frame 0||Frame 1||Compensation . OBMC v -------|------- -------|--------------. \------> Frame n output Frame Frame<----------------------------------/|...|------------------- Range Coder:============Binary Range Coder:------------------- The implemented range coder is an adapted version based upon "Range encoding: an algorithm for removing redundancy from a digitised message." by G. N. N. Martin. The symbols encoded by the Snow range coder are bits(0|1). The associated probabilities are not fix but change depending on the symbol mix seen so far. bit seen|new state ---------+----------------------------------------------- 0|256 - state_transition_table[256 - old_state];1|state_transition_table[old_state];state_transition_table={ 0, 0, 0, 0, 0, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 194, 194, 195, 196, 197, 198, 199, 200, 201, 202, 202, 204, 205, 206, 207, 208, 209, 209, 210, 211, 212, 213, 215, 215, 216, 217, 218, 219, 220, 220, 222, 223, 224, 225, 226, 227, 227, 229, 229, 230, 231, 232, 234, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 248, 0, 0, 0, 0, 0, 0, 0};FIXME Range Coding of integers:------------------------- FIXME Neighboring Blocks:===================left and top are set to the respective blocks unless they are outside of the image in which case they are set to the Null block top-left is set to the top left block unless it is outside of the image in which case it is set to the left block if this block has no larger parent block or it is at the left side of its parent block and the top right block is not outside of the image then the top right block is used for top-right else the top-left block is used Null block y, cb, cr are 128 level, ref, mx and my are 0 Motion Vector Prediction:=========================1. the motion vectors of all the neighboring blocks are scaled to compensate for the difference of reference frames scaled_mv=(mv *(256 *(current_reference+1)/(mv.reference+1))+128)> the median of the scaled left
Definition: snow.txt:386
randomize_buffers
#define randomize_buffers(buf, size)
Definition: llvidencdsp.c:31
src0
const pixel *const src0
Definition: h264pred_template.c:419
w
uint8_t w
Definition: llvidencdsp.c:39
declare_func
#define declare_func(ret,...)
Definition: checkasm.h:209
stride
#define stride
Definition: h264pred_template.c:536
width
#define width
Definition: dsp.h:89
src
#define src
Definition: vp8dsp.c:248