mirror of https://git.ffmpeg.org/ffmpeg.git
mlpdec: only calculate output size once
This commit is contained in:
parent
caa845851d
commit
37b67f1bff
|
@ -906,6 +906,7 @@ static int output_data_internal(MLPDecodeContext *m, unsigned int substr,
|
|||
{
|
||||
SubStream *s = &m->substream[substr];
|
||||
unsigned int i, out_ch = 0;
|
||||
int out_size;
|
||||
int32_t *data_32 = (int32_t*) data;
|
||||
int16_t *data_16 = (int16_t*) data;
|
||||
|
||||
|
@ -914,8 +915,11 @@ static int output_data_internal(MLPDecodeContext *m, unsigned int substr,
|
|||
return AVERROR_INVALIDDATA;
|
||||
}
|
||||
|
||||
if (*data_size < m->avctx->channels * s->blockpos * (is32 ? 4 : 2))
|
||||
return -1;
|
||||
out_size = s->blockpos * m->avctx->channels *
|
||||
av_get_bytes_per_sample(m->avctx->sample_fmt);
|
||||
|
||||
if (*data_size < out_size)
|
||||
return AVERROR(EINVAL);
|
||||
|
||||
for (i = 0; i < s->blockpos; i++) {
|
||||
for (out_ch = 0; out_ch <= s->max_matrix_channel; out_ch++) {
|
||||
|
@ -928,7 +932,7 @@ static int output_data_internal(MLPDecodeContext *m, unsigned int substr,
|
|||
}
|
||||
}
|
||||
|
||||
*data_size = i * out_ch * (is32 ? 4 : 2);
|
||||
*data_size = out_size;
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
|
Loading…
Reference in New Issue