* [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes
@ 2024-07-09 9:34 Niklas Haas
2024-07-09 9:34 ` [FFmpeg-devel] [PATCH 2/2] avfilter/vf_scale: test return code of scale_frame() Niklas Haas
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: Niklas Haas @ 2024-07-09 9:34 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Niklas Haas
From: Niklas Haas <git@haasn.dev>
scale_frame() inconsistently handled the lifetime of `in`. Fixes a
possible double free and a possible memory leak.
The new code always has `scale_frame` take over ownership of the input
frame. I first tried writing this code in a way where the calling code
retains ownership, but this is nontrivial due to the presence of the
no-op short-circuit condition in which the input frame is directly
returned. (As an alternative, we could use av_frame_clone() instead, but
I wanted to avoid touching the original behavior in this commit)
---
libavfilter/vf_scale.c | 29 +++++++++++++++++------------
1 file changed, 17 insertions(+), 12 deletions(-)
diff --git a/libavfilter/vf_scale.c b/libavfilter/vf_scale.c
index 841075193e..a1a322ed9e 100644
--- a/libavfilter/vf_scale.c
+++ b/libavfilter/vf_scale.c
@@ -869,17 +869,20 @@ static int scale_field(ScaleContext *scale, AVFrame *dst, AVFrame *src,
return 0;
}
-static int scale_frame(AVFilterLink *link, AVFrame *in, AVFrame **frame_out)
+/* Takes over ownership of *frame_in, passes ownership of *frame_out to caller */
+static int scale_frame(AVFilterLink *link, AVFrame **frame_in,
+ AVFrame **frame_out)
{
AVFilterContext *ctx = link->dst;
ScaleContext *scale = ctx->priv;
AVFilterLink *outlink = ctx->outputs[0];
- AVFrame *out;
+ AVFrame *out, *in = *frame_in;
const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(link->format);
char buf[32];
int ret;
int frame_changed;
+ *frame_in = NULL;
*frame_out = NULL;
if (in->colorspace == AVCOL_SPC_YCGCO)
av_log(link->dst, AV_LOG_WARNING, "Detected unsupported YCgCo colorspace.\n");
@@ -922,11 +925,11 @@ static int scale_frame(AVFilterLink *link, AVFrame *in, AVFrame **frame_out)
ret = scale_parse_expr(ctx, NULL, &scale->w_pexpr, "width", scale->w_expr);
if (ret < 0)
- return ret;
+ goto err;
ret = scale_parse_expr(ctx, NULL, &scale->h_pexpr, "height", scale->h_expr);
if (ret < 0)
- return ret;
+ goto err;
}
if (ctx->filter == &ff_vf_scale2ref) {
@@ -957,7 +960,7 @@ FF_ENABLE_DEPRECATION_WARNINGS
link->dst->inputs[0]->sample_aspect_ratio.num = in->sample_aspect_ratio.num;
if ((ret = config_props(outlink)) < 0)
- return ret;
+ goto err;
}
scale:
@@ -971,10 +974,9 @@ scale:
out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
if (!out) {
- av_frame_free(&in);
- return AVERROR(ENOMEM);
+ ret = AVERROR(ENOMEM);
+ goto err;
}
- *frame_out = out;
av_frame_copy_props(out, in);
out->width = outlink->w;
@@ -999,9 +1001,12 @@ scale:
ret = sws_scale_frame(scale->sws, out, in);
}
- av_frame_free(&in);
if (ret < 0)
- av_frame_free(frame_out);
+ av_frame_free(&out);
+ *frame_out = out;
+
+err:
+ av_frame_free(&in);
return ret;
}
@@ -1058,7 +1063,7 @@ FF_ENABLE_DEPRECATION_WARNINGS
}
}
- ret = scale_frame(ctx->inputs[0], in, &out);
+ ret = scale_frame(ctx->inputs[0], &in, &out);
if (out) {
out->pts = av_rescale_q(fs->pts, fs->time_base, outlink->time_base);
return ff_filter_frame(outlink, out);
@@ -1077,7 +1082,7 @@ static int filter_frame(AVFilterLink *link, AVFrame *in)
AVFrame *out;
int ret;
- ret = scale_frame(link, in, &out);
+ ret = scale_frame(link, &in, &out);
if (out)
return ff_filter_frame(outlink, out);
--
2.44.0
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 4+ messages in thread
* [FFmpeg-devel] [PATCH 2/2] avfilter/vf_scale: test return code of scale_frame()
2024-07-09 9:34 [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes Niklas Haas
@ 2024-07-09 9:34 ` Niklas Haas
2024-07-10 9:03 ` [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes Anton Khirnov
2024-07-10 12:29 ` Michael Niedermayer
2 siblings, 0 replies; 4+ messages in thread
From: Niklas Haas @ 2024-07-09 9:34 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Niklas Haas
From: Niklas Haas <git@haasn.dev>
Instead of testing the returned frame against NULL, test the return code
itself, going more in line with the usual behavior of such functions.
---
libavfilter/vf_scale.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/libavfilter/vf_scale.c b/libavfilter/vf_scale.c
index a1a322ed9e..ae7356fd7b 100644
--- a/libavfilter/vf_scale.c
+++ b/libavfilter/vf_scale.c
@@ -883,7 +883,6 @@ static int scale_frame(AVFilterLink *link, AVFrame **frame_in,
int frame_changed;
*frame_in = NULL;
- *frame_out = NULL;
if (in->colorspace == AVCOL_SPC_YCGCO)
av_log(link->dst, AV_LOG_WARNING, "Detected unsupported YCgCo colorspace.\n");
@@ -1064,14 +1063,15 @@ FF_ENABLE_DEPRECATION_WARNINGS
}
ret = scale_frame(ctx->inputs[0], &in, &out);
- if (out) {
- out->pts = av_rescale_q(fs->pts, fs->time_base, outlink->time_base);
- return ff_filter_frame(outlink, out);
- }
+ if (ret < 0)
+ goto err;
+
+ av_assert0(out);
+ out->pts = av_rescale_q(fs->pts, fs->time_base, outlink->time_base);
+ return ff_filter_frame(outlink, out);
err:
- if (ret < 0)
- av_frame_free(&in);
+ av_frame_free(&in);
return ret;
}
--
2.44.0
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes
2024-07-09 9:34 [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes Niklas Haas
2024-07-09 9:34 ` [FFmpeg-devel] [PATCH 2/2] avfilter/vf_scale: test return code of scale_frame() Niklas Haas
@ 2024-07-10 9:03 ` Anton Khirnov
2024-07-10 12:29 ` Michael Niedermayer
2 siblings, 0 replies; 4+ messages in thread
From: Anton Khirnov @ 2024-07-10 9:03 UTC (permalink / raw)
To: FFmpeg development discussions and patches; +Cc: Niklas Haas
Patchset looks ok.
--
Anton Khirnov
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes
2024-07-09 9:34 [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes Niklas Haas
2024-07-09 9:34 ` [FFmpeg-devel] [PATCH 2/2] avfilter/vf_scale: test return code of scale_frame() Niklas Haas
2024-07-10 9:03 ` [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes Anton Khirnov
@ 2024-07-10 12:29 ` Michael Niedermayer
2 siblings, 0 replies; 4+ messages in thread
From: Michael Niedermayer @ 2024-07-10 12:29 UTC (permalink / raw)
To: FFmpeg development discussions and patches
[-- Attachment #1.1: Type: text/plain, Size: 999 bytes --]
On Tue, Jul 09, 2024 at 11:34:36AM +0200, Niklas Haas wrote:
> From: Niklas Haas <git@haasn.dev>
>
> scale_frame() inconsistently handled the lifetime of `in`. Fixes a
> possible double free and a possible memory leak.
>
> The new code always has `scale_frame` take over ownership of the input
> frame. I first tried writing this code in a way where the calling code
> retains ownership, but this is nontrivial due to the presence of the
> no-op short-circuit condition in which the input frame is directly
> returned. (As an alternative, we could use av_frame_clone() instead, but
> I wanted to avoid touching the original behavior in this commit)
> ---
> libavfilter/vf_scale.c | 29 +++++++++++++++++------------
> 1 file changed, 17 insertions(+), 12 deletions(-)
patchset LGTM
thx
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
During times of universal deceit, telling the truth becomes a
revolutionary act. -- George Orwell
[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]
[-- Attachment #2: Type: text/plain, Size: 251 bytes --]
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2024-07-10 12:29 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-07-09 9:34 [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes Niklas Haas
2024-07-09 9:34 ` [FFmpeg-devel] [PATCH 2/2] avfilter/vf_scale: test return code of scale_frame() Niklas Haas
2024-07-10 9:03 ` [FFmpeg-devel] [PATCH 1/2] avfilter/vf_scale: fix frame lifetimes Anton Khirnov
2024-07-10 12:29 ` Michael Niedermayer
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
This inbox may be cloned and mirrored by anyone:
git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git
# If you have public-inbox 1.1+ installed, you may
# initialize and index your mirror using the following commands:
public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
ffmpegdev@gitmailbox.com
public-inbox-index ffmpegdev
Example config snippet for mirrors.
AGPL code for this site: git clone https://public-inbox.org/public-inbox.git