On Sun, Dec 15, 2019 at 02:16:49AM +0100, Marton Balint wrote:
> Unfortunately the ld() and st() operations store the variables in the AVExpr
> context and not in the parser in order to keep the stored values between
> evaluations (which is a rarely(?) used undocumented(?) feature of AVExpr).
>
>
On 14/12/2019 23:15, Marton Balint wrote:
On Sat, 14 Dec 2019, Alexandre Heitor Schmidt wrote:
avfilter/drawtext: Added two special metadata tags which can, be used by
'drawtext' filter and others to have access to source file path, and its
basename.
The new field AVPacket->source_file will con
On Sun, 15 Dec 2019, Michael Niedermayer wrote:
On Sun, Dec 15, 2019 at 02:16:49AM +0100, Marton Balint wrote:
Unfortunately the ld() and st() operations store the variables in the AVExpr
context and not in the parser in order to keep the stored values between
evaluations (which is a rarely(?
On Fri, 6 Dec 2019, Michael Niedermayer wrote:
Signed-off-by: Michael Niedermayer
---
libavutil/eval.c | 28
libavutil/eval.h | 11 +++
2 files changed, 31 insertions(+), 8 deletions(-)
diff --git a/libavutil/eval.c b/libavutil/eval.c
index 62d2ae938b..d527
On Fri, 13 Dec 2019, Marton Balint wrote:
On Fri, 6 Dec 2019, Marton Balint wrote:
The documentation still mentions numerical constants in addition to textual
ones. It is also wrong to use distinct modes as flags and it disallows us
to
actually use the flags field for real flags in the
On 12/15/2019 3:40 AM, Jun Zhao wrote:
> From: Jun Zhao
>
> Before this change, the command:
>
> ffmpeg -y -i bear-av1.mp4 -c:v copy bear-av1.ivf -loglevel 99
>
> always dump the log message like:
>
> AVBSFContext @ 0x55ff99e78e00] Frame 4: size 320x240 upscaled 320 render
> 320x240 subsa
James Almer (12019-12-14):
> I'm not sure if this is a good idea. It takes any AVBufferRef as input,
> so it will crash on pretty much every one not created by an AVBufferPool
> (unless it checks that buf is not NULL before dereferencing it), or
> return something unrelated otherwise.
This is true
On 12/15/2019 11:24 AM, Nicolas George wrote:
> James Almer (12019-12-14):
>> I'm not sure if this is a good idea. It takes any AVBufferRef as input,
>> so it will crash on pretty much every one not created by an AVBufferPool
>> (unless it checks that buf is not NULL before dereferencing it), or
>>
On Sun, Dec 15, 2019 at 01:52:52PM +0100, Marton Balint wrote:
>
>
> On Sun, 15 Dec 2019, Michael Niedermayer wrote:
>
> >On Sun, Dec 15, 2019 at 02:16:49AM +0100, Marton Balint wrote:
> >>Unfortunately the ld() and st() operations store the variables in the AVExpr
> >>context and not in the par
On Sat, Dec 14, 2019 at 10:59:31PM +, Alexandre Heitor Schmidt wrote:
> avfilter/drawtext: Added two special metadata tags which can, be used by
> 'drawtext' filter and others to have access to source file path, and its
> basename.
>
> The new field AVPacket->source_file will contain a path to
On Sun, 15 Dec 2019, Michael Niedermayer wrote:
On Sun, Dec 15, 2019 at 01:52:52PM +0100, Marton Balint wrote:
On Sun, 15 Dec 2019, Michael Niedermayer wrote:
On Sun, Dec 15, 2019 at 02:16:49AM +0100, Marton Balint wrote:
Unfortunately the ld() and st() operations store the variables in
On 15/12/2019 15:04, Michael Niedermayer wrote:
> On Sat, Dec 14, 2019 at 10:59:31PM +, Alexandre Heitor Schmidt wrote:
>> avfilter/drawtext: Added two special metadata tags which can, be used by
>> 'drawtext' filter and others to have access to source file path, and its
>> basename.
>>
>> The
1st of 5 factorized patches; supersedes
https://patchwork.ffmpeg.org/patch/16272/
From 8ee0551584adc7020f485d7b9a237da7207fbdb6 Mon Sep 17 00:00:00 2001
From: Gyan Doshi
Date: Sun, 8 Dec 2019 15:43:42 +0530
Subject: [PATCH 1/5] avfilter/scale: store frame change eval
Better readability and allow
4th of 5 factorized patches; supersedes
https://patchwork.ffmpeg.org/patch/16272/
From 4428afaba039756f47c177f2533fe23c0bfb4bc1 Mon Sep 17 00:00:00 2001
From: Gyan Doshi
Date: Fri, 13 Dec 2019 00:12:53 +0530
Subject: [PATCH 4/5] avfilter/scale: add function to check expressions
Allows finer iden
3rd of 5 factorized patches; supersedes
https://patchwork.ffmpeg.org/patch/16272/
From 00b54948b88ae60aa3ab7c158b98e55cb8b967d3 Mon Sep 17 00:00:00 2001
From: Gyan Doshi
Date: Thu, 12 Dec 2019 22:54:31 +0530
Subject: [PATCH 3/5] avfilter/scale: separate exprs parse and eval
Will allow adding ani
2nd of 5 factorized patches; supersedes
https://patchwork.ffmpeg.org/patch/16272/
From f5055424fb9988644d413c2599fe7ed4526c8334 Mon Sep 17 00:00:00 2001
From: Gyan Doshi
Date: Sun, 8 Dec 2019 15:54:28 +0530
Subject: [PATCH 2/5] avfilter/scale2ref: update links and re-eval expr upon
ref frame cha
5th of 5 factorized patches; supersedes
https://patchwork.ffmpeg.org/patch/16272/
From a31f4fe482aaa1574ebb17ff5820a3a364b79fff Mon Sep 17 00:00:00 2001
From: Gyan Doshi
Date: Sun, 15 Dec 2019 18:56:06 +0530
Subject: [PATCH 5/5] avfilter/scale: add animation support
Width and height expressions
tis 2019-12-10 klockan 22:14 +0800 skrev zhilizhao:
> > On Dec 10, 2019, at 4:20 AM, Tomas Härdin wrote:
> >
> > lör 2019-12-07 klockan 01:06 +0800 skrev Zhao Zhili:
> > > This is a micro-optimization. Saving almost 200 reallocations makes
> > > it
> > > worth a try.
> > > ---
> > > fix commit me
Tomas Härdin (12019-12-15):
> Perhaps someone with more insight into lavfi wants to comment?
I intend to look at it, but it have been a hectic week here.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
___
ffmpeg-devel mailing
On 12/15/19, Kieran Kunhya wrote:
>> +static av_cold void braw_static_init(void)
>> +{
>> +INIT_VLC_STATIC(&dc_vlc, 13, 16, dc_bits, 1, 1, dc_codes, 2, 2,
>> 8192);
>> +INIT_VLC_STATIC(&ac_vlc, 18, 194, ac_bits, 1, 1, ac_codes, 4, 4,
>> 262144);
>> +}
>> +
>
>> +.caps_internal= FF_
On Sun, Dec 15, 2019 at 04:24:11PM +0100, Marton Balint wrote:
>
>
> On Sun, 15 Dec 2019, Michael Niedermayer wrote:
>
> >On Sun, Dec 15, 2019 at 01:52:52PM +0100, Marton Balint wrote:
> >>
> >>
> >>On Sun, 15 Dec 2019, Michael Niedermayer wrote:
> >>
> >>>On Sun, Dec 15, 2019 at 02:16:49AM +010
On Tue, 10 Dec 2019, Martin Storsjö wrote:
The stereo_interpolate functions add h_step to the values h
BUF_SIZE times. Within the stereo_interpolate C functions, the
values h (h0-h3, h00-h13) are declared as local float variables,
but the compiler is free to keep them in a register with extra
pr
On Tue, 10 Dec 2019, Martin Storsjö wrote:
In these cases, we must pass the full path of the file to ffprobe
(as the current working dir on the remote system, e.g. when invoked
with "ssh remote ffprobe ..." isn't the wanted one).
The input filename passed to ffprobe is also included in the outp
When testing on a memory limited system, these tests consume a
significant amount of memory and can often fail if testing by running
multiple processes in parallel.
---
Made it possible to override the setting when running make, by setting
LARGE_TESTS. The solution looks a bit different from what I
Signed-off-by: Michael Niedermayer
---
libavcodec/h264_ps.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/h264_ps.c b/libavcodec/h264_ps.c
index 74f12f8979..410e767cd9 100644
--- a/libavcodec/h264_ps.c
+++ b/libavcodec/h264_ps.c
@@ -186,7 +186,7 @@ static inline i
Fixes: Ticket7249 (No longer displaying a scary red message)
Signed-off-by: Michael Niedermayer
---
libavcodec/h264_ps.c | 4
1 file changed, 4 insertions(+)
diff --git a/libavcodec/h264_ps.c b/libavcodec/h264_ps.c
index e8738d8502..74f12f8979 100644
--- a/libavcodec/h264_ps.c
+++ b/libavc
On Tue, 10 Dec 2019, manuelyuan wrote:
I opened a ticket in https://trac.ffmpeg.org/ticket/8420
my patch may be not absolutely right, but this problem should get your
attention, thank you
I sent a patch that doesn't mess up all the details in the fragmentation,
but only changes the calculati
Keep all the existing data fields as they are (there's lots and
lots of nontrivial calculation and heuristics based on them in
their current form), but derive the duration as the difference
between the pts of the first packet to the maximum pts+duration
(not necessarily the last packet); use this d
On 12/15/2019 7:00 PM, Michael Niedermayer wrote:
> Fixes: Ticket7249 (No longer displaying a scary red message)
>
> Signed-off-by: Michael Niedermayer
> ---
> libavcodec/h264_ps.c | 4
> 1 file changed, 4 insertions(+)
>
> diff --git a/libavcodec/h264_ps.c b/libavcodec/h264_ps.c
> index e
> There is thread_once there. And similar is done for imm4.
> So do not see what you complain about.
I am clearly blind. sorry
Kieran
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe
On Sun, Dec 15, 2019 at 02:40:52PM +0800, Jun Zhao wrote:
> From: Jun Zhao
>
> Before this change, the command:
>
> ffmpeg -y -i bear-av1.mp4 -c:v copy bear-av1.ivf -loglevel 99
>
> always dump the log message like:
>
> AVBSFContext @ 0x55ff99e78e00] Frame 4: size 320x240 upscaled 320 rende
On Sat, Dec 14, 2019 at 11:03:10PM +0800, lance.lmw...@gmail.com wrote:
> From: Limin Wang
>
> Signed-off-by: Limin Wang
> ---
> libavcodec/bsf.c | 31 +--
> 1 file changed, 17 insertions(+), 14 deletions(-)
sounds like a good idea
thx
[...]
--
Michael GnuPG
On 12/14/2019 12:03 PM, lance.lmw...@gmail.com wrote:
> From: Limin Wang
>
> Signed-off-by: Limin Wang
> ---
> libavcodec/bsf.c | 31 +--
> 1 file changed, 17 insertions(+), 14 deletions(-)
>
> diff --git a/libavcodec/bsf.c b/libavcodec/bsf.c
> index c1653cd..a5a5a0
On Sat, Dec 14, 2019 at 11:19:14PM +0100, Andreas Rheinhardt wrote:
> bytestream2_get_bytes_left returns an unsigned int; as a result,
> it returns big positive numbers if an overread already happened,
> making it unsuitable for scenarios where one wants to allow this
> in a controlled way (because
On Sat, Dec 14, 2019 at 11:19:13PM +0100, Andreas Rheinhardt wrote:
> There can be at most 31 SPS and 255 PPS in the mp4/Matroska extradata.
> Given that each has a size of at most 2^16-1, the length of the output
> derived from these parameter sets can never overflow an ordinary 32 bit
> integer.
On Sun, Dec 15, 2019 at 11:53 PM Michael Niedermayer
wrote:
> On Sat, Dec 14, 2019 at 11:19:14PM +0100, Andreas Rheinhardt wrote:
> > bytestream2_get_bytes_left returns an unsigned int; as a result,
> > it returns big positive numbers if an overread already happened,
> > making it unsuitable for
On Mon, Dec 16, 2019 at 12:05:52AM +0100, Andreas Rheinhardt wrote:
> On Sun, Dec 15, 2019 at 11:53 PM Michael Niedermayer
> wrote:
>
> > On Sat, Dec 14, 2019 at 11:19:14PM +0100, Andreas Rheinhardt wrote:
> > > bytestream2_get_bytes_left returns an unsigned int; as a result,
> > > it returns big
Am So., 15. Dez. 2019 um 23:25 Uhr schrieb James Almer :
>
> On 12/15/2019 7:00 PM, Michael Niedermayer wrote:
> > Fixes: Ticket7249 (No longer displaying a scary red message)
> >
> > Signed-off-by: Michael Niedermayer
> > ---
> > libavcodec/h264_ps.c | 4
> > 1 file changed, 4 insertions(+)
On 12/15/2019 8:50 PM, Carl Eugen Hoyos wrote:
> Am So., 15. Dez. 2019 um 23:25 Uhr schrieb James Almer :
>>
>> On 12/15/2019 7:00 PM, Michael Niedermayer wrote:
>>> Fixes: Ticket7249 (No longer displaying a scary red message)
>>>
>>> Signed-off-by: Michael Niedermayer
>>> ---
>>> libavcodec/h264
Am Mo., 16. Dez. 2019 um 01:00 Uhr schrieb James Almer :
>
> On 12/15/2019 8:50 PM, Carl Eugen Hoyos wrote:
> > Am So., 15. Dez. 2019 um 23:25 Uhr schrieb James Almer :
> >>
> >> On 12/15/2019 7:00 PM, Michael Niedermayer wrote:
> >>> Fixes: Ticket7249 (No longer displaying a scary red message)
> >
This patchset is primarily concerned with fixing memleaks in the hls
muxer. I was drawn to this by a memleak created by ed897633; it is fixed
in the first patch.*
Looking through the code I found more improvable stuff; and some of this
is fixed in the patches that follow. Yet there is still stuff
hls_mux_init() currently leaks an AVDictionary if opening a dynamic
buffer fails or if avformat_init_output fails. This has been fixed by
moving the initialization resp. the freeing of the dictionary around:
In the former case to a place after opening the dynamic buffer, in the
latter to a place be
"If an error is detected, localtime_r() shall return a null pointer
and set errno to indicate the error." Yet in case this happened in
hls_init(), AVERROR(ENOMEM) has been returned.
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
hls_init() would allocate a buffer, although it is only needed in one of
two branches that follow. This commit moves the allocation to the branch
that actually needs the buffer.
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 11 +--
1 file changed, 5 insertions(+), 6 deleti
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
index 2b87070d62..965c05c560 100644
--- a/libavformat/hlsenc.c
+++ b/libavformat/hlsenc.c
@@ -2079,7 +2079,7 @@ static int par
When a parameter like e.g. language is contained more than once in the
part of var_stream_map pertaining to a single VariantStream, the later
one just overwrites the pointer to the earlier one, leading to a
memleak. This commit changes this by handling the situation gracefully:
The earlier string i
Before ed897633, the hls muxer would free its child AVFormatContexts
and reset the pointer to these contexts to NULL immediately afterwards;
ed897633 moved the former to later (into a separate function), but kept
the resetting, ensuring that the child context leaks.
Signed-off-by: Andreas Rheinhar
Under certain circumstances hls_write_packet() would add options to an
AVDictionary. Said dictionary was never explicitly freed, instead it was
presumed that these options would be consumed when opening a new
IO-context. This left several possibilities for memleaks:
a) When no new IO-context would
The hls muxer allocates an array of VariantStreams, a structure that
contains pointers to objects that need to be freed on their own. This
means that the number of allocated VariantStreams needs to be correct
when they are freed; yet the number of VariantStreams is set in
update_variant_stream_info
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
index c612c3bbb8..16595f02cd 100644
--- a/libavformat/hlsenc.c
+++ b/libavformat/hlsenc.c
@@ -2684,9 +2684,7 @
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
index d738b08a3f..c612c3bbb8 100644
--- a/libavformat/hlsenc.c
+++ b/libavformat/hlsenc.c
@@ -2819,7 +2819,7 @@ static int hls
Several variables which are only used when the HLS_SINGLE_FILE bit is
not set have nevertheless been set regardless of whether this bit is set.
This has been changed; futhermore, these variables have been moved into
a smaller scope.
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 20
hls_init() would at first allocate the vtt_basename string, then
allocate the vtt_m3u8_name string followed by several operations that
may fail and then open the subtitles' output context. Yet upon freeing,
these strings were only freed when the subtitles' output context
existed, ensuring that they
This fixes memleaks in instances such as:
a) When an allocation fails at one of the two places in hls_init() where
the error is returned immediately without goto fail first.
b) When an error happens when writing the header.
c) When an allocation fails at one of the three places in
hls_write_trailer
Mainly includes reindentation, moving variables into smaller scopes (in
particular when this allows to remove unnecessary initializations) and
also returning directly (i.e. without a goto fail when possible).
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 175 +-
The check will be true at most once anyway.
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 17 +
1 file changed, 9 insertions(+), 8 deletions(-)
diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
index 16595f02cd..62a52b176f 100644
--- a/libavformat/hlsenc.c
Signed-off-by: Andreas Rheinhardt
---
libavformat/hlsenc.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
index 8f142fd475..b3f9582267 100644
--- a/libavformat/hlsenc.c
+++ b/libavformat/hlsenc.c
@@ -1614,6 +1614,8 @@
On 12/15/2019 9:03 PM, Carl Eugen Hoyos wrote:
> Am Mo., 16. Dez. 2019 um 01:00 Uhr schrieb James Almer :
>>
>> On 12/15/2019 8:50 PM, Carl Eugen Hoyos wrote:
>>> Am So., 15. Dez. 2019 um 23:25 Uhr schrieb James Almer :
On 12/15/2019 7:00 PM, Michael Niedermayer wrote:
> Fixes: Ticket
On Sun, Dec 15, 2019 at 07:49:41PM -0300, James Almer wrote:
> On 12/14/2019 12:03 PM, lance.lmw...@gmail.com wrote:
> > From: Limin Wang
> >
> > Signed-off-by: Limin Wang
> > ---
> > libavcodec/bsf.c | 31 +--
> > 1 file changed, 17 insertions(+), 14 deletions(-)
>
> 在 2019年12月16日，08:04，Andreas Rheinhardt 写道：
>
> Before ed897633, the hls muxer would free its child AVFormatContexts
> and reset the pointer to these contexts to NULL immediately afterwards;
> ed897633 moved the former to later (into a separate function), but kept
> the resetting, ensuring tha
> 在 2019年12月16日，08:04，Andreas Rheinhardt 写道：
>
> Signed-off-by: Andreas Rheinhardt
> ---
> libavformat/hlsenc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
> index 2b87070d62..965c05c560 100644
> --- a/libavformat/hlsenc.
> 在 2019年12月16日，08:04，Andreas Rheinhardt 写道：
>
> "If an error is detected, localtime_r() shall return a null pointer
> and set errno to indicate the error." Yet in case this happened in
> hls_init(), AVERROR(ENOMEM) has been returned.
>
> Signed-off-by: Andreas Rheinhardt
> ---
> libavformat/
> 在 2019年12月16日，08:04，Andreas Rheinhardt 写道：
>
> When a parameter like e.g. language is contained more than once in the
> part of var_stream_map pertaining to a single VariantStream, the later
> one just overwrites the pointer to the earlier one, leading to a
> memleak. This commit changes this
> 在 2019年12月16日，08:04，Andreas Rheinhardt 写道：
>
> Several variables which are only used when the HLS_SINGLE_FILE bit is
> not set have nevertheless been set regardless of whether this bit is set.
> This has been changed; futhermore, these variables have been moved into
> a smaller scope.
>
> Si
> 在 2019年12月16日，08:04，Andreas Rheinhardt 写道：
>
> hls_init() would at first allocate the vtt_basename string, then
> allocate the vtt_m3u8_name string followed by several operations that
> may fail and then open the subtitles' output context. Yet upon freeing,
> these strings were only freed whe
> 在 2019年12月16日，08:04，Andreas Rheinhardt 写道：
>
> Signed-off-by: Andreas Rheinhardt
> ---
> libavformat/hlsenc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
> index d738b08a3f..c612c3bbb8 100644
> --- a/libavformat/hlsenc.
On Mon, Dec 16, 2019 at 4:07 AM Steven Liu wrote:
>
>
> > 在 2019年12月16日，08:04，Andreas Rheinhardt
> 写道：
> >
> > Several variables which are only used when the HLS_SINGLE_FILE bit is
> > not set have nevertheless been set regardless of whether this bit is set.
> > This has been changed; futhermore
On 12/15/2019 7:00 PM, Michael Niedermayer wrote:
> Fixes: Ticket7249 (No longer displaying a scary red message)
>
> Signed-off-by: Michael Niedermayer
> ---
> libavcodec/h264_ps.c | 4
> 1 file changed, 4 insertions(+)
>
> diff --git a/libavcodec/h264_ps.c b/libavcodec/h264_ps.c
> index e
On Fri, 06. Dec 14:22, Andriy Gelman wrote:
> From: Andriy Gelman
>
> payload_count is used to track the number of SEI payloads. It is also
> used to free the SEIs in cbs_h264_free_sei()/cbs_h265_free_sei().
>
> Currently, payload_count is set after for loop is completed. Hence if
> there is an
On Mon, Dec 16, 2019 at 6:37 AM Michael Niedermayer
wrote:
>
> On Sun, Dec 15, 2019 at 02:40:52PM +0800, Jun Zhao wrote:
> > From: Jun Zhao
> >
> > Before this change, the command:
> >
> > ffmpeg -y -i bear-av1.mp4 -c:v copy bear-av1.ivf -loglevel 99
> >
> > always dump the log message like:
> >
From: Limin Wang
Signed-off-by: Limin Wang
---
libavcodec/bsf.c | 49 +++--
1 file changed, 27 insertions(+), 22 deletions(-)
diff --git a/libavcodec/bsf.c b/libavcodec/bsf.c
index c1653cd..c6dce93 100644
--- a/libavcodec/bsf.c
+++ b/libavcodec/bsf.c
Is there still something wrong with this patch?
Cheers,
Jörg
> -Ursprüngliche Nachricht-
> Von: ffmpeg-devel Im Auftrag von Jörg
> Beckmann
> Gesendet: Montag, 9. Dezember 2019 16:05
> An: ffmpeg-devel@ffmpeg.org
> Betreff: [FFmpeg-devel] [PATCH v5] avformat/mov: Memory optimization with
72 matches
Mail list logo
