Traneptora/FFmpeg
1
Fork 0
forked from FFmpeg/FFmpeg
Code Pull requests Activity

g723.1: remove useless uses of MUL64()

Browse source 
The operands in both cases are 16-bit so cannot overflow a 32-bit
destination.  In gain_scale() the inputs are reduced to 14-bit,
so even the shift cannot overflow.

Signed-off-by: Mans Rullgard <mans@mansr.com>
This commit is contained in:
Mans Rullgard 2012-08-10 16:42:54 +01:00
parent 5a43eba956
commit 1eb1f6f281
1 changed files with 4 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
libavcodec/g723_1.c
View file

@ -574,7 +574,7 @@ static int dot_product(const int16_t *a, const int16_t *b, int length)
int i, sum = 0;
for (i = 0; i < length; i++) {
int64_t prod = av_clipl_int32(MUL64(a[i], b[i]) << 1);
int64_t prod = av_clipl_int32((int64_t)(a[i] * b[i]) << 1);
sum = av_clipl_int32(sum + prod);
}
return sum;
@ -889,9 +889,9 @@ static void gain_scale(G723_1_Context *p, int16_t * buf, int energy)
num = energy;
denom = 0;
for (i = 0; i < SUBFRAME_LEN; i++) {
int64_t temp = buf[i] >> 2;
temp = av_clipl_int32(MUL64(temp, temp) << 1);
denom = av_clipl_int32(denom + temp);
int temp = buf[i] >> 2;
temp *= temp;
denom = av_clipl_int32((int64_t)denom + (temp << 1));
}
if (num && denom) {