OpenGrok
Home
Sort by relevance
Sort by last modified time
Full Search
Definition
Symbol
File Path
History
|
|
Help
Searched
refs:L_x
(Results
1 - 15
of
15
) sorted by null
/frameworks/av/media/libstagefright/codecs/amrnb/common/src/
pow2.cpp
88
L_x
= Result of the Pow2() computation (Word32)
103
This function computes
L_x
= pow(2.0, exponent.fraction)
105
The function Pow2(
L_x
) is approximated by a table and linear interpolation.
109
3-
L_x
= table[i]<<16 - (table[i] - table[i+1]) * a * 2
110
4-
L_x
=
L_x
>> (30-exponent) (with rounding)
131
Word32
L_x
;
133
L_x
= L_mult (fraction, 32); //
L_x
= fraction<<6
134
i = extract_h (
L_x
); // Extract b10-b16 of fractio
[
all
...]
inv_sqrt.cpp
108
L_x
= input value (Word32)
115
L_y = inverse squareroot of
L_x
(Word32)
126
This function computes 1/sqrt(
L_x
), where
L_x
is positive.
127
If
L_x
is negative or zero, the result is 1 (3fff ffff).
129
The function 1/sqrt(
L_x
) is approximated by a table and linear
132
1- Normalization of
L_x
.
135
4- i = bit25-b31 of
L_x
; 16<=i<=63 because of normalization.
155
Word32
L_x
// (i) : input value
166
if (
L_x
<= (Word32) 0
[
all
...]
sqrt_l.cpp
90
L_x
= input value (Word32)
98
L_y = squareroot of
L_x
(Word32)
109
This function computes sqrt(
L_x
), where
L_x
is positive.
112
The function sqrt(
L_x
) is approximated by a table and linear
115
1- Normalization of
L_x
.
118
4- i = bit25-b31 of
L_x
; 16<=i<=63 because of normalization.
138
Word32
L_x
, // i : input value
156
if (
L_x
<= (Word32) 0)
167
e = norm_l (
L_x
) & 0xFFFE; // get next lower EVEN norm. ex
[
all
...]
log2.cpp
92
L_x
= input value of type Word32
117
This function computes logarithm (base2) of the input
L_x
, where
L_x
is
118
positive. If
L_x
is negative or zero, the result is 0.
120
This function first normalizes the input
L_x
and calls the function Log2_norm
164
Word32
L_x
, /* (i) : input value */
174
exp = norm_l(
L_x
);
175
result =
L_x
<< exp;
log2_norm.cpp
91
L_x
= normalized input value of type Word32
92
exp = number of shifts required to normalize
L_x
; it is of type Word16
114
The function Log2(
L_x
) calculates the logarithm of the normalized input
115
buffer
L_x
. The logarithm is approximated by a table and linear
116
interpolation. The following steps are used to compute Log2(
L_x
):
119
2. i = bit25-b31 of
L_x
; 32<=i<=63 (because of normalization).
138
Word32
L_x
, // (i) : input value (normalized)
139
Word16 exp, // (i) : norm_l (
L_x
)
147
if (
L_x
<= (Word32) 0)
156
L_x
= L_shr (L_x, 9)
[
all
...]
/frameworks/av/media/libstagefright/codecs/amrwbenc/src/
log2.c
20
* Purpose : Computes log2(
L_x
) *
40
* PURPOSE: Computes log2(
L_x
, exp), where
L_x
is positive and
42
* If
L_x
is negative or zero, the result is 0.
45
* The function Log2(
L_x
) is approximated by a table and linear
46
* interpolation. The following steps are used to compute Log2(
L_x
)
49
* 2- i = bit25-b31 of
L_x
; 32<=i<=63 (because of normalization).
57
Word32
L_x
, /* (i) : input value (normalized) */
58
Word16 exp, /* (i) : norm_l (
L_x
) */
65
if (
L_x
<= (Word32) 0
[
all
...]
math_op.c
47
| Compute 1/sqrt(
L_x
). |
48
| if
L_x
is negative or zero, result is 1 (7fffffff). |
52
| 1- Normalization of
L_x
. |
53
| 2- call Isqrt_n(
L_x
, exponant) |
54
| 3- L_y =
L_x
<< exponant |
58
Word32
L_x
/* (i) Q0 : input value (range: 0<=val<=7fffffff) */
63
exp = norm_l(
L_x
);
64
L_x
= (
L_x
<< exp); /*
L_x
is normalized *
[
all
...]
/frameworks/av/media/libstagefright/codecs/amrwbenc/inc/
log2.h
22
* Purpose : Computes log2(
L_x
)
48
Word32
L_x
, /* (i) : input value */
54
Word32
L_x
, /* (i) : input value (normalized) */
55
Word16 exp, /* (i) : norm_l (
L_x
) */
math_op.h
28
Word32
L_x
/* (i) Q0 : input value (range: 0<=val<=7fffffff) */
/frameworks/av/media/libstagefright/codecs/amrnb/common/include/
inv_sqrt.h
106
Word32
L_x
, /* (i) : input value */
log2.h
107
Word32
L_x
, /* (i) : input value */
sqrt_l.h
105
Word32
L_x
, /* i : input value, Q31 */
log2_norm.h
106
Word32
L_x
, /* (i) : input value (normalized) */
107
Word16 exp, /* (i) : norm_l (
L_x
) */
/frameworks/av/media/libstagefright/codecs/amrwb/src/
pvamrwb_math_op.cpp
36
one_ov_sqrt() : Compute 1/sqrt(
L_x
)
268
Compute 1/sqrt(
L_x
).
269
if
L_x
is negative or zero, result is 1 (7fffffff).
273
1- Normalization of
L_x
.
274
2- call Isqrt_n(
L_x
, exponant)
275
3- L_y =
L_x
<< exponant
278
int32
L_x
/* (i) Q0 : input value (range: 0<=val<=7fffffff) */
284
exp = normalize_amr_wb(
L_x
);
285
L_x
<<= exp; /*
L_x
is normalized *
[
all
...]
pvamrwb_math_op.h
77
int32
L_x
/* (i) Q0 : input value (range: 0<=val<=7fffffff) */
96
int32
L_x
, /* (i) : input value */
102
int32
L_x
, /* (i) : input value (normalized) */
103
int16 exp, /* (i) : norm_l (
L_x
) */
Completed in 2465 milliseconds