Up until now, the aderivative and aintegrate filters allocated
an audio buffer with exactly one sample per channel via
ff_get_audio_buffer(); if said buffer could not be allocated,
a frame would leak, as freeing it has been forgotten.
This commit instead uses a plain array for the audio buffer; it is
allocated when configuring the filter, removing the error condition
where the frame can leak. This fixes Coverity ticket #1197065.
Signed-off-by: Andreas Rheinhardt <andreas.rheinha...@outlook.com>
---
libavfilter/af_aderivative.c | 40 ++++++++++++++++++------------------
1 file changed, 20 insertions(+), 20 deletions(-)
diff --git a/libavfilter/af_aderivative.c b/libavfilter/af_aderivative.c
index 6b3e4dd0e4..559223d5ee 100644
--- a/libavfilter/af_aderivative.c
+++ b/libavfilter/af_aderivative.c
@@ -22,8 +22,8 @@
typedef struct ADerivativeContext {
const AVClass *class;
- AVFrame *prev;
- void (*filter)(void **dst, void **prv, const void **src,
+ void *prev;
+ void (*filter)(void **dst, void *prv, const void **src,
int nb_samples, int channels);
} ADerivativeContext;
@@ -63,22 +63,24 @@ static int query_formats(AVFilterContext *ctx)
}
#define DERIVATIVE(name, type) \
-static void aderivative_## name ##p(void **d, void **p, const void **s, \
+static void aderivative_## name ##p(void **d, void *p0, const void **s, \
int nb_samples, int channels) \
{ \
+ type *const p = p0; \
int n, c; \
\
for (c = 0; c < channels; c++) { \
const type *src = s[c]; \
type *dst = d[c]; \
- type *prv = p[c]; \
+ type prv = p[c]; \
\
for (n = 0; n < nb_samples; n++) { \
const type current = src[n]; \
\
- dst[n] = current - prv[0]; \
- prv[0] = current; \
+ dst[n] = current - prv; \
+ prv = current; \
} \
+ p[c] = prv; \
} \
}
@@ -88,22 +90,24 @@ DERIVATIVE(s16, int16_t)
DERIVATIVE(s32, int32_t)
#define INTEGRAL(name, type) \
-static void aintegral_## name ##p(void **d, void **p, const void **s, \
+static void aintegral_## name ##p(void **d, void *p0, const void **s, \
int nb_samples, int channels) \
{ \
+ type *const p = p0; \
int n, c; \
\
for (c = 0; c < channels; c++) { \
const type *src = s[c]; \
type *dst = d[c]; \
- type *prv = p[c]; \
+ type prv = p[c]; \
\
for (n = 0; n < nb_samples; n++) { \
const type current = src[n]; \
\
- dst[n] = current + prv[0]; \
- prv[0] = dst[n]; \
+ dst[n] = current + prv; \
+ prv = dst[n]; \
} \
+ p[c] = prv; \
} \
}
@@ -122,6 +126,10 @@ static int config_input(AVFilterLink *inlink)
case AV_SAMPLE_FMT_S16P: s->filter = aderivative_s16p; break;
}
+ s->prev = av_calloc(inlink->channels,
av_get_bytes_per_sample(inlink->format));
+ if (!s->prev)
+ return AVERROR(ENOMEM);
+
if (strcmp(ctx->filter->name, "aintegral"))
return 0;
@@ -146,15 +154,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in)
}
av_frame_copy_props(out, in);
- if (!s->prev) {
- s->prev = ff_get_audio_buffer(inlink, 1);
- if (!s->prev) {
- av_frame_free(&in);
- return AVERROR(ENOMEM);
- }
- }
-
- s->filter((void **)out->extended_data, (void **)s->prev->extended_data,
(const void **)in->extended_data,
+ s->filter((void **)out->extended_data, s->prev, (const void
**)in->extended_data,
in->nb_samples, in->channels);
av_frame_free(&in);
@@ -165,7 +165,7 @@ static av_cold void uninit(AVFilterContext *ctx)
{
ADerivativeContext *s = ctx->priv;
- av_frame_free(&s->prev);
+ av_freep(&s->prev);
}
static const AVFilterPad aderivative_inputs[] = {
--
2.30.2
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
