[FFmpeg-devel] [PATCH] avfilter/biquads: avoid clipping when using floating type
I think clipping isn't needed when using floating type.
Thanks
From 3ffce13b5b118d84b724f2aec342baa6536271da Mon Sep 17 00:00:00 2001
From: Muhammad Faiz
Date: Thu, 17 Jul 2014 10:49:04 +0700
Subject: [PATCH] avfilter/biquads: avoid clipping when using floating type
---
libavfilter/af_biquads.c | 22 +++---
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/libavfilter/af_biquads.c b/libavfilter/af_biquads.c
index 5bafad1..02bf9db 100644
--- a/libavfilter/af_biquads.c
+++ b/libavfilter/af_biquads.c
@@ -161,7 +161,7 @@ static int query_formats(AVFilterContext *ctx)
return 0;
}
-#define BIQUAD_FILTER(name, type, min, max) \
+#define BIQUAD_FILTER(name, type, min, max, need_clipping)\
static void biquad_## name (const void *input, void *output, int len, \
double *in1, double *in2, \
double *out1, double *out2, \
@@ -181,10 +181,10 @@ static void biquad_## name (const void *input, void
*output, int len, \
for (i = 0; i+1 < len; i++) { \
o2 = i2 * b2 + i1 * b1 + ibuf[i] * b0 + o2 * a2 + o1 * a1;\
i2 = ibuf[i]; \
-if (o2 < min) { \
+if (need_clipping && o2 < min) { \
av_log(NULL, AV_LOG_WARNING, "clipping\n"); \
obuf[i] = min;\
-} else if (o2 > max) {\
+} else if (need_clipping && o2 > max) { \
av_log(NULL, AV_LOG_WARNING, "clipping\n"); \
obuf[i] = max;\
} else { \
@@ -193,10 +193,10 @@ static void biquad_## name (const void *input, void
*output, int len, \
i++; \
o1 = i1 * b2 + i2 * b1 + ibuf[i] * b0 + o1 * a2 + o2 * a1;\
i1 = ibuf[i]; \
-if (o1 < min) { \
+if (need_clipping && o1 < min) { \
av_log(NULL, AV_LOG_WARNING, "clipping\n"); \
obuf[i] = min;\
-} else if (o1 > max) {\
+} else if (need_clipping && o1 > max) { \
av_log(NULL, AV_LOG_WARNING, "clipping\n"); \
obuf[i] = max;\
} else { \
@@ -209,10 +209,10 @@ static void biquad_## name (const void *input, void
*output, int len, \
i1 = ibuf[i]; \
o2 = o1; \
o1 = o0; \
-if (o0 < min) { \
+if (need_clipping && o0 < min) { \
av_log(NULL, AV_LOG_WARNING, "clipping\n"); \
obuf[i] = min;\
-} else if (o0 > max) {\
+} else if (need_clipping && o0 > max) { \
av_log(NULL, AV_LOG_WARNING, "clipping\n"); \
obuf[i] = max;\
} else { \
@@ -225,10 +225,10 @@ static void biquad_## name (const void *input, void
*output, int len, \
*out2 = o2; \
}
-BIQUAD_FILTER(s16, int16_t, INT16_MIN, INT16_MAX)
-BIQUAD_FILTER(s32, int32_t, INT32_MIN, INT32_MAX)
-BIQUAD_FILTER(flt, float, -1., 1.)
-BIQUAD_FILTER(dbl, double, -1., 1.)
+BIQUAD_FILTER(s16, int16_t, INT16_MIN, INT16_MAX, 1)
+BIQUAD_FILTER(s32, int32_t, INT32_MIN, INT32_MAX, 1)
+BIQUAD_FILTER(flt, float, -1., 1., 0)
+BIQUAD_FILTER(dbl, double, -1., 1., 0)
static int config_output(AVFilterLink *outlink)
{
--
1.8.3.1
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailma
[FFmpeg-devel] A few filter questions
Good day, I'm currently working on a video signature filter for ffmpeg. This allows you to fingerprint videos. This fingerprint is built up of 9mb/s of bits or 2-3 mb/s bits compressed. In this context a few questions come into my mind: - Should I print this whole bitstream to stdout/stderr at the end? Is it maybe a better choice to made an own stream out of this. But which type of stream this is? (btw, the video signature algorithm needs 90 following frames, so I can theoretically write every 90 frames something somewhere.) - If I print the whole bitstream to stdout/stderr (my current implementation), is there a possibility to use this later in an external program? The only other globally analyze filter I found is volumedetect. This filter at the end prints per print_stats the calculated results to the console. Is there a possibility within the API for an external program to use these values or do I have to grep the output? A similar example is AcousticID (a fingerprinting technique for audio). Currently chromaprint (the AcousticID library) provides an executable (fpcalc) to calculate AcousticID. It therefore uses FFmpeg to decode the audio and then its own library to calculate the fingerprint. The better way I think would be to have an ffmpeg filter for this. But is it possibly to use the calculated number in an external program without grepping the output? Another thing that came into my mind: Can filter force other filters to go into the filterchain? I see it, when I force GREY_8 only in my filter, it automatically enables the scale filter, too. The reason I asked is the lookup for my filter. Currently my filter analyzes a video and then produces a lot of numbers. To compare two videos and decide, wheather they match or not, these numbers has to be compared. I see three possibilities: 1. Write an VV->V filter. Reimplement (copy) the code from the V->V signature filter and give a boolean as output (match or match not). 2. Take the V->V filter and write a python (or whatever) script that fetch the output and calculates then the rest. 3. Write an VV->V filter, but enforce, that the normal signature filter is executed first to both streams, use the result and then calculate the matching type. Unfortunately I have no idea, how to do this and whether it is possible at all. Can you give me an advice? The last possibility also would allow something like twopass volume normalisation. Currently there is a volumedetect and volume filter. To normalize once could run volumedetect, then fetch the output, and put the values into the volume filter, but I currently don't see a way to do this automatically directly in ffmpeg. (Once the filter is in a good state, I will try to bring it upstream.) Best, Gerion ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] A few filter questions
On Thu, Jul 17, 2014 at 12:33:41PM +0200, Gerion Entrup wrote: > Good day, > > I'm currently working on a video signature filter for ffmpeg. This allows you > to > fingerprint videos. Oh, nice. > This fingerprint is built up of 9mb/s of bits or 2-3 mb/s bits compressed. > > In this context a few questions come into my mind: > - Should I print this whole bitstream to stdout/stderr at the end? Is it > maybe > a better choice to made an own stream out of this. But which type of stream > this is? How does the fingerprint looks like? Could it make sense as a gray video output fractal, or maybe some kind of audio signal? Also, you still have the string metadata possibility (git grep SET_META libavfilter). > (btw, the video signature algorithm needs 90 following frames, so I can > theoretically write every 90 frames something somewhere.) Do you cache all these frames or just update your caches/stats & drop them? > - If I print the whole bitstream to stdout/stderr (my current > implementation), > is there a possibility to use this later in an external program? The only > other globally analyze filter I found is volumedetect. This filter at the end > prints per print_stats the calculated results to the console. Is there a > possibility within the API for an external program to use these values or do > I > have to grep the output? stdout/stderr really isn't a good thing. Using metadata is way better because you can output them from ffprobe, and parse them according to various outputs (XML, CSV, JSON, ...). Another solution I can now think of is to simply pass an output file as option to the filter. That's typically how we do the 2-pass thing with vidstab filter. [...] > Another thing that came into my mind: Can filter force other filters to go > into > the filterchain? I see it, when I force GREY_8 only in my filter, it > automatically enables the scale filter, too. Some filter are inserted automatically for conversion & constraints, but that's not decided by the filters but the framework itself. > The reason I asked is the lookup > for my filter. Currently my filter analyzes a video and then produces a lot > of > numbers. To compare two videos and decide, wheather they match or not, these > numbers has to be compared. I see three possibilities: > 1. Write an VV->V filter. Reimplement (copy) the code from the V->V signature > filter and give a boolean as output (match or match not). > 2. Take the V->V filter and write a python (or whatever) script that fetch > the > output and calculates then the rest. > 3. Write an VV->V filter, but enforce, that the normal signature filter is > executed first to both streams, use the result and then calculate the > matching > type. Unfortunately I have no idea, how to do this and whether it is possible > at all. Can you give me an advice? > So if you output a file in the filter itself: ffmpeg -i video -vf fingerprint=video.sig -f null - ffmpeg -i another -vf fingerprint=video.sig:check=1 -f null - Or if you save the signature "stream" in a video (in gray8 for instance): ffmpeg -i video -vf fingerprint -c:v ffv1 sig.nut ffmpeg -i another -i sig.nut -vf '[0][1] fingerprint=mode=check' -f null - The 2nd method is "better" because it doesn't require file handling in the library, and it also allows stuff like using a diff filter (if you also apply fingerprint - not with mode=check - on `another`) Am I understanding right your wondering? > The last possibility also would allow something like twopass volume > normalisation. Currently there is a volumedetect and volume filter. To > normalize once could run volumedetect, then fetch the output, and put the > values into the volume filter, but I currently don't see a way to do this > automatically directly in ffmpeg. Check tools/normalize.py, it's using ebur128 and the metadata system. > > (Once the filter is in a good state, I will try to bring it upstream.) > Cool > Best, > Gerion > -- Clément B. pgpIUcBGsdmql.pgp Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavu/eval: add clip function
On date Monday 2014-07-07 20:58:50 +0200, Michael Niedermayer encoded:
> On Fri, Jul 04, 2014 at 04:40:50PM +0200, Stefano Sabatini wrote:
[...]
> > Subject: [PATCH] lavu/eval: add clip function
> >
> > TODO: bump micro
> > ---
> > doc/utils.texi | 3 +++
> > libavutil/eval.c| 14 +-
> > tests/ref/fate/eval | 9 +
> > 3 files changed, 25 insertions(+), 1 deletion(-)
> >
> > diff --git a/doc/utils.texi b/doc/utils.texi
> > index 5abfb0c..b92691f 100644
> > --- a/doc/utils.texi
> > +++ b/doc/utils.texi
> > @@ -782,6 +782,9 @@ large numbers (usually 2^53 and larger).
> > Round the value of expression @var{expr} upwards to the nearest
> > integer. For example, "ceil(1.5)" is "2.0".
> >
> > +@item clip(x, min, max)
> > +Return the value of @var{x} clipped between @var{min} and @var{max}.
> > +
> > @item cos(x)
> > Compute cosine of @var{x}.
> >
> > diff --git a/libavutil/eval.c b/libavutil/eval.c
> > index 4a313bf..1c53b79 100644
> > --- a/libavutil/eval.c
> > +++ b/libavutil/eval.c
> > @@ -147,7 +147,7 @@ struct AVExpr {
> > e_pow, e_mul, e_div, e_add,
> > e_last, e_st, e_while, e_taylor, e_root, e_floor, e_ceil, e_trunc,
> > e_sqrt, e_not, e_random, e_hypot, e_gcd,
> > -e_if, e_ifnot, e_print, e_bitand, e_bitor, e_between,
> > +e_if, e_ifnot, e_print, e_bitand, e_bitor, e_between, e_clip
> > } type;
> > double value; // is sign in other types
> > union {
> > @@ -187,6 +187,13 @@ static double eval_expr(Parser *p, AVExpr *e)
> >e->param[2] ? eval_expr(p,
> > e->param[2]) : 0);
> > case e_ifnot: return e->value * (!eval_expr(p, e->param[0]) ?
> > eval_expr(p, e->param[1]) :
> >e->param[2] ? eval_expr(p,
> > e->param[2]) : 0);
> > +case e_clip: {
> > +double x = eval_expr(p, e->param[0]);
> > +double min = eval_expr(p, e->param[1]), max = eval_expr(p,
> > e->param[2]);
> > +if (isnan(min) || isnan(max) || isnan(x) || min > max)
> > +return NAN;
>
> > +return e->value * av_clip(eval_expr(p, e->param[0]), min, max);
>
> this should be av_clipd()
>
> should be ok otherwise
Applied, thanks.
--
FFmpeg = Frenzy & Fantastic Meaningful Plastic Ephemeral Game
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 1/2] drawtext: add expansion function eif()
On date Sunday 2014-07-06 13:46:31 +0300, Andrey Utkin encoded:
> 2014-07-05 16:19 GMT+03:00 Nicolas George :
> > Le sextidi 16 messidor, an CCXXII, Andrey Utkin a écrit :
> >> +@item eif
> >
> > I find the name obscure. Where does it come from?
>
> Yes, it is somewhat obscure. It comes from "expr_int_format".
>
> >> +if (!strchr("xXdu", argv[1][0])) {
> >
> > You should probably check that argv[1][1] is 0.
>
> Can be added to check for stray trailing symbols in passed argument,
> but this doesn't affect the algorithm, it uses only first character.
>
> >> +feclearexcept(FE_ALL_EXCEPT);
> >> +intval = res;
> >> +if ((ret = fetestexcept(FE_INVALID|FE_OVERFLOW|FE_UNDERFLOW))) {
> >> +av_log(ctx, AV_LOG_ERROR, "Conversion of floating-point result to
> >> int failed. Control register: 0x%08x. Conversion result: %d\n", ret,
> >> intval);
> >> +return AVERROR(EINVAL);
> >
> > Is this portable enough? Anyway, AFAIK, we never do that kind of test in the
> > code.
>
> This must be coming from C standard. I consulted with this CERT C
> Secure Coding Standards chapter:
> https://www.securecoding.cert.org/confluence/display/seccode/FLP03-C.+Detect+and+handle+floating-point+errors
Patch applied, further tweaking can be done later.
--
FFmpeg = Freak and Frenzy Mean Powerful Enchanting God
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2 v2] drawtext: add fontcolor_expr
On date Saturday 2014-07-05 11:59:35 +0200, Stefano Sabatini encoded: > On date Friday 2014-07-04 21:54:53 +0300, Andrey Utkin encoded: > > An option for fontcolor, dynamically evaluated > > --- > > doc/filters.texi | 16 > > libavfilter/vf_drawtext.c | 26 +- > > 2 files changed, 37 insertions(+), 5 deletions(-) > > No more comments from me, I'll apply both drawtext patches if I see no > comments from other developers in 1/2 days, thank you. Applied. -- FFmpeg = Faithless Fancy Majestic Political Elastic God ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/dv: implement fallback in dv_extract_pack()
On Jul 16, 2014, at 10:57 PM, Michael Niedermayer wrote: > Fixes Ticket2340 > Fixes Ticket2341 I tested with the samples from tickets 2340 and 2341. In both cases the patch fixes the originally reported issue. Thanks! Dave Rice ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] lavf/segment: only use reference frames for computing the segment end time
This avoids a systematic overestimate of the segments duration when there
are several streams.
Fix trac ticket #3724.
---
libavformat/segment.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/segment.c b/libavformat/segment.c
index 3de4623..7c3a3c5 100644
--- a/libavformat/segment.c
+++ b/libavformat/segment.c
@@ -728,7 +728,7 @@ static int seg_write_packet(AVFormatContext *s, AVPacket
*pkt)
seg->cur_entry.index = seg->segment_idx +
seg->segment_idx_wrap*seg->segment_idx_wrap_nb;
seg->cur_entry.start_time = (double)pkt->pts * av_q2d(st->time_base);
seg->cur_entry.start_pts = av_rescale_q(pkt->pts, st->time_base,
AV_TIME_BASE_Q);
-} else if (pkt->pts != AV_NOPTS_VALUE) {
+} else if (pkt->pts != AV_NOPTS_VALUE && pkt->stream_index ==
seg->reference_stream_index) {
seg->cur_entry.end_time =
FFMAX(seg->cur_entry.end_time, (double)(pkt->pts + pkt->duration)
* av_q2d(st->time_base));
}
--
1.8.3.2
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/dv: implement fallback in dv_extract_pack()
On Thu, Jul 17, 2014 at 07:37:58AM -0400, Dave Rice wrote: > On Jul 16, 2014, at 10:57 PM, Michael Niedermayer wrote: > > > Fixes Ticket2340 > > Fixes Ticket2341 > > I tested with the samples from tickets 2340 and 2341. In both cases the patch > fixes the originally reported issue. Thanks! applied thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB There will always be a question for which you do not know the correct answer. signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavf/segment: only use reference frames for computing the segment end time
On Thu, Jul 17, 2014 at 02:53:43PM +0200, Stefano Sabatini wrote: > This avoids a systematic overestimate of the segments duration when there > are several streams. > > Fix trac ticket #3724. > --- > libavformat/segment.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) probably ok [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB The worst form of inequality is to try to make unequal things equal. -- Aristotle signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] Issue 3052
Hi,
- Mail original -
> On Wed, Jul 16, 2014 at 05:40:35PM +0200, Benoit Fouet wrote:
> >
> >
> > - Mail original -
> > > Hi,
> > >
> > > I had a look at https://trac.ffmpeg.org/ticket/3052.
> > > The issue is that the last frame duration is lost. Is there any
> > > reason why the packet duration is not transmitted from the
> > > demuxer
> > > to the muxer?
> > > Attached is a dumb patch, just to understand a bit more why
> > > things
> > > are done the way they are...
> > >
> >
> > e.g., by dumb, I mean that the duration should be corrected with
> > the right timebase... But I just want to get feedback on the
> > forwarding part.
>
> pkt_duration and pkt_timebase where primarly intended for the decoder
> side, they could be used for encoder too but i guess filters dont
> update them currently
>
I'll try to address the filter case as a separate issue anyway, so this
shouldn't be an issue; at least for now.
> that is:
>
> @@ -2063,6 +2063,10 @@ FF_ENABLE_DEPRECATION_WARNINGS
> avpkt->size = 0;
> else if (!(avctx->codec->capabilities & CODEC_CAP_DELAY))
> avpkt->pts = avpkt->dts = frame->pts;
> +if (frame && av_codec_get_pkt_timebase(avctx).num)
> +avpkt->duration =
> av_rescale_q(av_frame_get_pkt_duration(frame),
> +
> av_codec_get_pkt_timebase(avctx),
> + avctx->time_base);
>
> if (needs_realloc && avpkt->data) {
> ret = av_buffer_realloc(&avpkt->buf, avpkt->size +
> FF_INPUT_BUFFER_PADDING_SIZE);
>
I'll try to add the time_base in the AVFrame then, so that the information is
there with the duration.
I should post a patch soon to do this, and test it on the GIF issue from issue
3052
--
Ben
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] Issue 3052
L'octidi 28 messidor, an CCXXII, Michael Niedermayer a écrit : > again, i suspect this is not updated by filters currently You are right, the filters do not update them currently. I believe changing that policy would not be a convenient way of solving that issue: the duration information raises consistency problems: what happens if next_frame->pts is not equal to cur_frame->pts + cur_frame->duration. Also, that would be abusing a field, since pkt_duration is the duration of a packet, not a frame. I believe a better solution would be to send EOF pseudo-frames forward in the filter chain instead of relying on the backward request_frame() calls to return EOF: it would solve this particular issue (a pseudo-frame has a timestamp) and other issues as well. This is one of the main things I intend to look at during this summer. Regards, -- Nicolas George signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] A few filter questions
Am Donnerstag 17 Juli 2014, 13:00:13 schrieb Clément Bœsch: > On Thu, Jul 17, 2014 at 12:33:41PM +0200, Gerion Entrup wrote: > > Good day, > > > > I'm currently working on a video signature filter for ffmpeg. This allows > > you to fingerprint videos. > > Oh, nice. > > > This fingerprint is built up of 9mb/s of bits or 2-3 mb/s bits compressed. Argh, fail, sorry. I meant: 9mb per hour of video (and 2-3 mb per hour). > > > > In this context a few questions come into my mind: > > - Should I print this whole bitstream to stdout/stderr at the end? Is it > > maybe a better choice to made an own stream out of this. But which type > > of stream this is? > > How does the fingerprint looks like? Could it make sense as a gray video > output fractal, or maybe some kind of audio signal? There a finesignatures per frame and coursesignatures per 90 finesignatures. coursesignature are binarized histograms (0 or 1 possible as count). finesignature is mainly a vector of 380 difference values between -128 and 127 which are ternarized into 0 1 or 2. (See the MPEG-7 Standard for more details). I doubt, this is a good video or audio stream. Definitely, interpreting this as video make sense in some way, but metadata looks more useful. > > Also, you still have the string metadata possibility (git grep SET_META > libavfilter). Hmm, thank you, I will take a look at it. If I see it right, it is used to fill a dictionary per frame with some kind of data? > > > (btw, the video signature algorithm needs 90 following frames, so I can > > > > theoretically write every 90 frames something somewhere.) > > Do you cache all these frames or just update your caches/stats & drop > them? ATM I don't cache the frames, but the whole signature. As said above, the coursesignatures (the part, which needs the 90 frames) is calculated only from the finesignatures (the finesignatures are cached, anyway). > > > - If I print the whole bitstream to stdout/stderr (my current > > implementation), is there a possibility to use this later in an external > > program? The only other globally analyze filter I found is volumedetect. > > This filter at the end prints per print_stats the calculated results to > > the console. Is there a possibility within the API for an external > > program to use these values or do I have to grep the output? > > stdout/stderr really isn't a good thing. Using metadata is way better > because you can output them from ffprobe, and parse them according to > various outputs (XML, CSV, JSON, ...). Sounds good… > > Another solution I can now think of is to simply pass an output file as > option to the filter. That's typically how we do the 2-pass thing with > vidstab filter. I don't like output files. If you want to write a program, that performs a lookup to signatures somewhere stored in a database and this program uses ffmpeg internally and then always has to write a file and read it again, it's not that elegant. (btw, an example for such a program is Musicbrainz Picard, but for AcousticID ;)) > > [...] > > > Another thing that came into my mind: Can filter force other filters to go > > into the filterchain? I see it, when I force GREY_8 only in my filter, it > > automatically enables the scale filter, too. > > Some filter are inserted automatically for conversion & constraints, but > that's not decided by the filters but the framework itself. > > > The reason I asked is the > > lookup > > > > for my filter. Currently my filter analyzes a video and then produces a > > lot of numbers. To compare two videos and decide, wheather they match or > > not, these numbers has to be compared. I see three possibilities: > > 1. Write an VV->V filter. Reimplement (copy) the code from the V->V > > signature filter and give a boolean as output (match or match not). > > 2. Take the V->V filter and write a python (or whatever) script that fetch > > the output and calculates then the rest. > > 3. Write an VV->V filter, but enforce, that the normal signature filter is > > executed first to both streams, use the result and then calculate the > > matching type. Unfortunately I have no idea, how to do this and whether > > it is possible at all. Can you give me an advice? > > So if you output a file in the filter itself: > ffmpeg -i video -vf fingerprint=video.sig -f null - > ffmpeg -i another -vf fingerprint=video.sig:check=1 -f null - > > Or if you save the signature "stream" in a video (in gray8 for instance): > ffmpeg -i video -vf fingerprint -c:v ffv1 sig.nut > ffmpeg -i another -i sig.nut -vf '[0][1] fingerprint=mode=check' -f null - > > The 2nd method is "better" because it doesn't require file handling in the > library, and it also allows stuff like using a diff filter (if you also > apply fingerprint - not with mode=check - on `another`) > > Am I understanding right your wondering? No ;), but anyway thanks for your
Re: [FFmpeg-devel] [PATCH] lavf/segment: only use reference frames for computing the segment end time
On date Thursday 2014-07-17 16:25:26 +0200, Michael Niedermayer encoded: > On Thu, Jul 17, 2014 at 02:53:43PM +0200, Stefano Sabatini wrote: > > This avoids a systematic overestimate of the segments duration when there > > are several streams. > > > > Fix trac ticket #3724. > > --- > > libavformat/segment.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > probably ok Thanks, applied. -- FFmpeg = Fascinating and Frightening Mere Purposeless Enhanced Glue ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] A few filter questions
On Thu, Jul 17, 2014 at 04:56:08PM +0200, Gerion Entrup wrote: [...] > > Also, you still have the string metadata possibility (git grep SET_META > > libavfilter). > Hmm, thank you, I will take a look at it. If I see it right, it is used to > fill > a dictionary per frame with some kind of data? > Strings only, so you'll have to find a serialization somehow. Maybe simply an ascii hex string or something. But yeah, it just allows you to map some key → value string couples to the frames passing by in the filter. How huge is the information to store per frame? [...] > > stdout/stderr really isn't a good thing. Using metadata is way better > > because you can output them from ffprobe, and parse them according to > > various outputs (XML, CSV, JSON, ...). > Sounds good… tools/normalize.py make use of such feature if you want examples (that's the -of option of ffprobe) [...] > > Am I understanding right your wondering? > No ;), but anyway thanks for your answer. In your 2nd method your filter is a > VV->V filter? Am I right, that this filter then also can take only one > stream? > Said in another way: Can a VV->V filter also behave as a V->V filter? Yes, fieldmatch is a (complex) example of this. But typically it's simply a filter with dynamic inputs, based on the user input. The simplest example would be the split filter. Look at it for an example of dynamic allocation of the number of inputs based on the user input (-vf split=4 is a V-> filter) [...] > > Check tools/normalize.py, it's using ebur128 and the metadata system. > Thats what I mean. Someone has to write an external script, which calls > ffmpeg/ffprobe two times, parse stdout of the first call and pass it to the > filteroptions of the second call. As I see, there is no direct way. Something > like: > ffmpeg -i foo -f:a volume=mode=autodetect normalized.opus We add a discussion several time for real time with that filter. If we do a 2-pass, that's simply because it's "more efficient". Typically, doing some live normalization can be done easily (we had patches for this): ebur128 already attaches some metadata to frames, so a following filter such as volume could reuse them, something like -filter_complex ebur128=metadata=1,volume=metadata. [...] -- Clément B. pgporzFGKSof0.pgp Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavfi: check refcount before merging.
Le septidi 27 messidor, an CCXXII, Michael Niedermayer a écrit : > LGTM Thanks. Then please merge the following commits from my tree: a65c0a3 lavd/x11grab: disable drawing mouse without XFixes. 36fbe3c lavd/x11grab: change error code for unsupported visuals. 16c6795 lavd/x11grab: check 32-bits color masks. 1d12df1 lavd/x11grab: add an option to disable MIT-SHM. 099aff5 lavfi: check refcount before merging. 8e29768 lavd/x11grab: reindent after last commit. Regards, -- Nicolas George signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] A few filter questions
Le nonidi 29 messidor, an CCXXII, Clément Bœsch a écrit : > We add a discussion several time for real time with that filter. If we do > a 2-pass, that's simply because it's "more efficient". Typically, doing > some live normalization can be done easily (we had patches for this): > ebur128 already attaches some metadata to frames, so a following filter > such as volume could reuse them, something like -filter_complex > ebur128=metadata=1,volume=metadata. I believe you are wrong in this paragraph: we do two passes for normalization because that is the only way of doing it without distortions: the level of volume adjustment depends on the whole stream. Normalization can be done in a single pass with distortions, but currently, no filter is capable of smoothing the measures computed by ebur128 to make the distortions inaudible. Patch welcome. Regards, -- Nicolas George signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] lavf/segment: sanitize segment end time in case last packet didn't have a defined duration
In particular, avoids to set segments with a 0 duration (e.g. segment
with a single reference frame for which duration is undefined).
---
libavformat/segment.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/libavformat/segment.c b/libavformat/segment.c
index db932f7..33a5558 100644
--- a/libavformat/segment.c
+++ b/libavformat/segment.c
@@ -48,6 +48,7 @@ typedef struct SegmentListEntry {
int64_t offset_pts;
char *filename;
struct SegmentListEntry *next;
+int64_t last_duration;
} SegmentListEntry;
typedef enum {
@@ -720,6 +721,10 @@ static int seg_write_packet(AVFormatContext *s, AVPacket
*pkt)
(pkt->pts != AV_NOPTS_VALUE &&
av_compare_ts(pkt->pts, st->time_base,
end_pts-seg->time_delta, AV_TIME_BASE_Q) >= 0))) {
+/* sanitize end time in case last packet didn't have a defined
duration */
+if (seg->cur_entry.last_duration == 0)
+seg->cur_entry.end_time = (double)pkt->pts * av_q2d(st->time_base);
+
if ((ret = segment_end(s, seg->individual_header_trailer, 0)) < 0)
goto fail;
@@ -735,6 +740,7 @@ static int seg_write_packet(AVFormatContext *s, AVPacket
*pkt)
} else if (pkt->pts != AV_NOPTS_VALUE && pkt->stream_index ==
seg->reference_stream_index) {
seg->cur_entry.end_time =
FFMAX(seg->cur_entry.end_time, (double)(pkt->pts + pkt->duration)
* av_q2d(st->time_base));
+seg->cur_entry.last_duration = pkt->duration;
}
if (seg->segment_frame_count == 0) {
--
1.8.3.2
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] lavf/segment: do not allow to create segments with no key-frames
Fix trac ticket #3749.
---
libavformat/segment.c | 16 +---
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/libavformat/segment.c b/libavformat/segment.c
index 72bf5e0..db932f7 100644
--- a/libavformat/segment.c
+++ b/libavformat/segment.c
@@ -94,7 +94,8 @@ typedef struct {
char *frames_str; ///< segment frame numbers specification string
int *frames; ///< list of frame number specification
int nb_frames; ///< number of elments in the frames array
-int frame_count;
+int frame_count; ///< total number of reference frames
+int segment_frame_count; ///< number of reference frames in the segment
int64_t time_delta;
int individual_header_trailer; /**< Set by a private option. */
@@ -109,7 +110,6 @@ typedef struct {
SegmentListEntry *segment_list_entries;
SegmentListEntry *segment_list_entries_end;
-int is_first_pkt; ///< tells if it is the first packet in the segment
} SegmentContext;
static void print_csv_escaped_str(AVIOContext *ctx, const char *str)
@@ -228,7 +228,7 @@ static int segment_start(AVFormatContext *s, int
write_header)
return err;
}
-seg->is_first_pkt = 1;
+seg->segment_frame_count = 0;
return 0;
}
@@ -647,7 +647,7 @@ static int seg_write_header(AVFormatContext *s)
avio_close(oc->pb);
goto fail;
}
-seg->is_first_pkt = 1;
+seg->segment_frame_count = 0;
if (oc->avoid_negative_ts > 0 && s->avoid_negative_ts < 0)
s->avoid_negative_ts = 1;
@@ -715,6 +715,7 @@ static int seg_write_packet(AVFormatContext *s, AVPacket
*pkt)
if (pkt->stream_index == seg->reference_stream_index &&
pkt->flags & AV_PKT_FLAG_KEY &&
+seg->segment_frame_count > 0 &&
(seg->cut_pending || seg->frame_count >= start_frame ||
(pkt->pts != AV_NOPTS_VALUE &&
av_compare_ts(pkt->pts, st->time_base,
@@ -736,11 +737,10 @@ static int seg_write_packet(AVFormatContext *s, AVPacket
*pkt)
FFMAX(seg->cur_entry.end_time, (double)(pkt->pts + pkt->duration)
* av_q2d(st->time_base));
}
-if (seg->is_first_pkt) {
+if (seg->segment_frame_count == 0) {
av_log(s, AV_LOG_VERBOSE, "segment:'%s' starts with packet stream:%d
pts:%s pts_time:%s frame:%d\n",
seg->avf->filename, pkt->stream_index,
av_ts2str(pkt->pts), av_ts2timestr(pkt->pts, &st->time_base),
seg->frame_count);
-seg->is_first_pkt = 0;
}
av_log(s, AV_LOG_DEBUG, "stream:%d start_pts_time:%s pts:%s pts_time:%s
dts:%s dts_time:%s",
@@ -764,8 +764,10 @@ static int seg_write_packet(AVFormatContext *s, AVPacket
*pkt)
ret = ff_write_chained(seg->avf, pkt->stream_index, pkt, s);
fail:
-if (pkt->stream_index == seg->reference_stream_index)
+if (pkt->stream_index == seg->reference_stream_index) {
seg->frame_count++;
+seg->segment_frame_count++;
+}
return ret;
}
--
1.8.3.2
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavfi: check refcount before merging.
On Thu, Jul 17, 2014 at 06:30:40PM +0200, Nicolas George wrote: > Le septidi 27 messidor, an CCXXII, Michael Niedermayer a écrit : > > LGTM > > Thanks. Then please merge the following commits from my tree: > > a65c0a3 lavd/x11grab: disable drawing mouse without XFixes. > 36fbe3c lavd/x11grab: change error code for unsupported visuals. > 16c6795 lavd/x11grab: check 32-bits color masks. > 1d12df1 lavd/x11grab: add an option to disable MIT-SHM. > 099aff5 lavfi: check refcount before merging. > 8e29768 lavd/x11grab: reindent after last commit. merged thanks [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB If a bugfix only changes things apparently unrelated to the bug with no further explanation, that is a good sign that the bugfix is wrong. signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 1/3] lavd/x11grab: disable drawing mouse without XFixes.
Le septidi 27 messidor, an CCXXII, Michael Niedermayer a écrit : > LGTM You just merged this series. Thanks. Regards, -- Nicolas George signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH]Ignore xing number of frames if filesize is wrong
Hi!
Attached patch fixes ticket #3777 for me, analyzed by Oliver Fromme.
Please comment, Carl Eugen
diff --git a/libavformat/mp3dec.c b/libavformat/mp3dec.c
index 8335388..cdef6d9 100644
--- a/libavformat/mp3dec.c
+++ b/libavformat/mp3dec.c
@@ -138,6 +138,7 @@ static void mp3_parse_info_tag(AVFormatContext *s, AVStream
*st,
MP3DecContext *mp3 = s->priv_data;
static const int64_t xing_offtbl[2][2] = {{32, 17}, {17,9}};
+int64_t fsize = avio_size(s->pb);
/* Check for Xing / Info tag */
avio_skip(s->pb, xing_offtbl[c->lsf == 1][c->nb_channels == 1]);
@@ -151,6 +152,9 @@ static void mp3_parse_info_tag(AVFormatContext *s, AVStream
*st,
mp3->frames = avio_rb32(s->pb);
if (v & XING_FLAG_SIZE)
mp3->header_filesize = avio_rb32(s->pb);
+if (fsize > 0 && mp3->header_filesize > 0 &&
+FFABS(fsize - (int64_t)mp3->header_filesize) / (float)FFMIN(fsize,
mp3->header_filesize) > 0.05)
+mp3->frames = 0;
if (v & XING_FLAG_TOC)
read_xing_toc(s, mp3->header_filesize, av_rescale_q(mp3->frames,
(AVRational){spf, c->sample_rate},
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]Ignore xing number of frames if filesize is wrong
Le nonidi 29 messidor, an CCXXII, Carl Eugen Hoyos a écrit :
> Hi!
>
> Attached patch fixes ticket #3777 for me, analyzed by Oliver Fromme.
>
> Please comment, Carl Eugen
> diff --git a/libavformat/mp3dec.c b/libavformat/mp3dec.c
> index 8335388..cdef6d9 100644
> --- a/libavformat/mp3dec.c
> +++ b/libavformat/mp3dec.c
> @@ -138,6 +138,7 @@ static void mp3_parse_info_tag(AVFormatContext *s,
> AVStream *st,
>
> MP3DecContext *mp3 = s->priv_data;
> static const int64_t xing_offtbl[2][2] = {{32, 17}, {17,9}};
> +int64_t fsize = avio_size(s->pb);
>
> /* Check for Xing / Info tag */
> avio_skip(s->pb, xing_offtbl[c->lsf == 1][c->nb_channels == 1]);
> @@ -151,6 +152,9 @@ static void mp3_parse_info_tag(AVFormatContext *s,
> AVStream *st,
> mp3->frames = avio_rb32(s->pb);
> if (v & XING_FLAG_SIZE)
> mp3->header_filesize = avio_rb32(s->pb);
> +if (fsize > 0 && mp3->header_filesize > 0 &&
> +FFABS(fsize - (int64_t)mp3->header_filesize) / (float)FFMIN(fsize,
> mp3->header_filesize) > 0.05)
I would suggest to avoid floating point arithmetic if possible. Possibly
something like that:
uint64_t min = FFMIN(fsize, mp3->header_filesize);
uint64_t delta = FFMAX(fsize, mp3->header_filesize) - min;
uint64_t tolerance = min / 20;
if (... && min - tolerance < 2 * tolerance)
I also find this version easier to understand.
And in any case, someone may correct me, but I believe nowadays double
should always preferred to float unless you need a lot of them and want to
reduce the memory use.
I can not judge on the correctness, though.
> +mp3->frames = 0;
> if (v & XING_FLAG_TOC)
> read_xing_toc(s, mp3->header_filesize, av_rescale_q(mp3->frames,
> (AVRational){spf, c->sample_rate},
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] Introduce avio_dump_contents() and use it in lavd/lavfi.c
This is a replacement for previously proposed API
av_bprint_fd_contents().
Side-effect: lavfi input device now accepts any URL as "graph_file"
option value.
---
libavdevice/lavfi.c | 30 +++---
libavformat/avio.h| 8
libavformat/aviobuf.c | 16
3 files changed, 39 insertions(+), 15 deletions(-)
diff --git a/libavdevice/lavfi.c b/libavdevice/lavfi.c
index d1904dd..0388a86 100644
--- a/libavdevice/lavfi.c
+++ b/libavdevice/lavfi.c
@@ -115,23 +115,23 @@ av_cold static int lavfi_read_header(AVFormatContext
*avctx)
}
if (lavfi->graph_filename) {
-uint8_t *file_buf, *graph_buf;
-size_t file_bufsize;
-ret = av_file_map(lavfi->graph_filename,
- &file_buf, &file_bufsize, 0, avctx);
+AVBPrint graph_file_pb;
+AVIOContext *avio = NULL;
+ret = avio_open(&avio, lavfi->graph_filename, AVIO_FLAG_READ);
if (ret < 0)
-goto end;
-
-/* create a 0-terminated string based on the read file */
-graph_buf = av_malloc(file_bufsize + 1);
-if (!graph_buf) {
-av_file_unmap(file_buf, file_bufsize);
-FAIL(AVERROR(ENOMEM));
+FAIL(ret);
+av_bprint_init(&graph_file_pb, 0, AV_BPRINT_SIZE_UNLIMITED);
+ret = avio_dump_contents(avio, &graph_file_pb);
+avio_close(avio);
+av_bprint_chars(&graph_file_pb, '\0', 1);
+if (!ret && !av_bprint_is_complete(&graph_file_pb))
+ret = AVERROR(ENOMEM);
+if (ret) {
+av_bprint_finalize(&graph_file_pb, NULL);
+FAIL(ret);
}
-memcpy(graph_buf, file_buf, file_bufsize);
-graph_buf[file_bufsize] = 0;
-av_file_unmap(file_buf, file_bufsize);
-lavfi->graph_str = graph_buf;
+if ((ret = av_bprint_finalize(&graph_file_pb, &lavfi->graph_str)))
+FAIL(ret);
}
if (!lavfi->graph_str)
diff --git a/libavformat/avio.h b/libavformat/avio.h
index 4004b6f..7f608fa 100644
--- a/libavformat/avio.h
+++ b/libavformat/avio.h
@@ -31,6 +31,7 @@
#include "libavutil/common.h"
#include "libavutil/dict.h"
#include "libavutil/log.h"
+#include "libavutil/bprint.h"
#include "libavformat/version.h"
@@ -500,4 +501,11 @@ int avio_pause(AVIOContext *h, int pause);
int64_t avio_seek_time(AVIOContext *h, int stream_index,
int64_t timestamp, int flags);
+/**
+ * Read contents of h into print buffer up to EOF.
+ *
+ * @return 0 for success, error code otherwise
+ */
+int avio_dump_contents(AVIOContext *h, AVBPrint *pb);
+
#endif /* AVFORMAT_AVIO_H */
diff --git a/libavformat/aviobuf.c b/libavformat/aviobuf.c
index 738459e..ba84873 100644
--- a/libavformat/aviobuf.c
+++ b/libavformat/aviobuf.c
@@ -953,6 +953,22 @@ int64_t avio_seek_time(AVIOContext *s, int stream_index,
return ret;
}
+int avio_dump_contents(AVIOContext *h, AVBPrint *pb)
+{
+int ret;
+char buf[1024];
+while (1) {
+ret = avio_read(h, buf, sizeof(buf));
+if (ret == AVERROR_EOF)
+return 0;
+if (ret <= 0)
+return ret;
+av_bprint_append_data(pb, buf, ret);
+if (!av_bprint_is_complete(pb))
+return AVERROR(ENOMEM);
+}
+}
+
/* output in a dynamic buffer */
typedef struct DynBuffer {
--
1.8.5.5
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 1/2] drawtext: add expansion function eif()
Damn, was just going to resubmit :( However, the only change was to add more comprehensive alias "expr_int_format". Do you think it's worth to submit such addition? -- Andrey Utkin ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 1/2] drawtext: add expansion function eif()
On date Thursday 2014-07-17 23:38:42 +0300, Andrey Utkin encoded: > Damn, was just going to resubmit :( However, the only change was to > add more comprehensive alias "expr_int_format". Do you think it's > worth to submit such addition? Sure. Also since the patch was committed just today, feel free to send changes which could affect the interface (the sooner the better). -- FFmpeg = Fostering and Fanciful Merciless Portable Erroneous Guru ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]Ignore xing number of frames if filesize is wrong
On Thursday 17 July 2014 09:41:40 pm Nicolas George wrote:
> Le nonidi 29 messidor, an CCXXII, Carl Eugen Hoyos a écrit :
> > Hi!
> >
> > Attached patch fixes ticket #3777 for me, analyzed by Oliver Fromme.
> >
> > Please comment, Carl Eugen
> >
> > diff --git a/libavformat/mp3dec.c b/libavformat/mp3dec.c
> > index 8335388..cdef6d9 100644
> > --- a/libavformat/mp3dec.c
> > +++ b/libavformat/mp3dec.c
> > @@ -138,6 +138,7 @@ static void mp3_parse_info_tag(AVFormatContext *s,
> > AVStream *st,
> >
> > MP3DecContext *mp3 = s->priv_data;
> > static const int64_t xing_offtbl[2][2] = {{32, 17}, {17,9}};
> > +int64_t fsize = avio_size(s->pb);
> >
> > /* Check for Xing / Info tag */
> > avio_skip(s->pb, xing_offtbl[c->lsf == 1][c->nb_channels == 1]);
> > @@ -151,6 +152,9 @@ static void mp3_parse_info_tag(AVFormatContext *s,
> > AVStream *st, mp3->frames = avio_rb32(s->pb);
> > if (v & XING_FLAG_SIZE)
> > mp3->header_filesize = avio_rb32(s->pb);
> >
> > +if (fsize > 0 && mp3->header_filesize > 0 &&
> > +FFABS(fsize - (int64_t)mp3->header_filesize) /
> > (float)FFMIN(fsize, mp3->header_filesize) > 0.05)
>
> I would suggest to avoid floating point arithmetic if possible. Possibly
> something like that:
>
> uint64_t min = FFMIN(fsize, mp3->header_filesize);
> uint64_t delta = FFMAX(fsize, mp3->header_filesize) - min;
> uint64_t tolerance = min / 20;
> if (... && min - tolerance < 2 * tolerance)
>
> I also find this version easier to understand.
>
> And in any case, someone may correct me, but I believe nowadays double
> should always preferred to float unless you need a lot of them and want to
> reduce the memory use.
>
> I can not judge on the correctness, though.
>
> > +mp3->frames = 0;
> > if (v & XING_FLAG_TOC)
> > read_xing_toc(s, mp3->header_filesize, av_rescale_q(mp3->frames,
> > (AVRational){spf,
> > c->sample_rate},
>
> Regards,
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]Ignore xing number of frames if filesize is wrong
On Thursday 17 July 2014 09:41:40 pm Nicolas George wrote:
> Le nonidi 29 messidor, an CCXXII, Carl Eugen Hoyos a écrit :
> >
> > Attached patch fixes ticket #3777 for me, analyzed by Oliver
> > Fromme.
> I would suggest to avoid floating point arithmetic if possible.
> Possibly something like that:
>
> uint64_t min = FFMIN(fsize, mp3->header_filesize);
> uint64_t delta = FFMAX(fsize, mp3->header_filesize) - min;
> uint64_t tolerance = min / 20;
> if (... && min - tolerance < 2 * tolerance)
>
> I also find this version easier to understand.
Thank you, new patch attached.
> I can not judge on the correctness, though.
The logic is copied from asfdec.c.
Please review, Carl Eugen
diff --git a/libavformat/mp3dec.c b/libavformat/mp3dec.c
index 8335388..cf5aa73 100644
--- a/libavformat/mp3dec.c
+++ b/libavformat/mp3dec.c
@@ -138,6 +138,7 @@ static void mp3_parse_info_tag(AVFormatContext *s, AVStream
*st,
MP3DecContext *mp3 = s->priv_data;
static const int64_t xing_offtbl[2][2] = {{32, 17}, {17,9}};
+uint64_t min, delta, fsize = avio_size(s->pb);
/* Check for Xing / Info tag */
avio_skip(s->pb, xing_offtbl[c->lsf == 1][c->nb_channels == 1]);
@@ -151,6 +152,10 @@ static void mp3_parse_info_tag(AVFormatContext *s,
AVStream *st,
mp3->frames = avio_rb32(s->pb);
if (v & XING_FLAG_SIZE)
mp3->header_filesize = avio_rb32(s->pb);
+min = FFMIN(fsize, mp3->header_filesize);
+delta = FFMAX(fsize, mp3->header_filesize) - min;
+if (fsize && mp3->header_filesize > 0 && delta > min >> 4)
+mp3->frames = 0;
if (v & XING_FLAG_TOC)
read_xing_toc(s, mp3->header_filesize, av_rescale_q(mp3->frames,
(AVRational){spf, c->sample_rate},
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]Ignore xing number of frames if filesize is wrong
Le nonidi 29 messidor, an CCXXII, Carl Eugen Hoyos a écrit :
> > if (... && min - tolerance < 2 * tolerance)
^^^ ^^^
This bit was a leftover from a slightly different version I tried at first,
it was wrong.
> > I also find this version easier to understand.
>
> Thank you, new patch attached.
>
> > I can not judge on the correctness, though.
>
> The logic is copied from asfdec.c.
>
> Please review, Carl Eugen
> diff --git a/libavformat/mp3dec.c b/libavformat/mp3dec.c
> index 8335388..cf5aa73 100644
> --- a/libavformat/mp3dec.c
> +++ b/libavformat/mp3dec.c
> @@ -138,6 +138,7 @@ static void mp3_parse_info_tag(AVFormatContext *s,
> AVStream *st,
>
> MP3DecContext *mp3 = s->priv_data;
> static const int64_t xing_offtbl[2][2] = {{32, 17}, {17,9}};
> +uint64_t min, delta, fsize = avio_size(s->pb);
>
> /* Check for Xing / Info tag */
> avio_skip(s->pb, xing_offtbl[c->lsf == 1][c->nb_channels == 1]);
> @@ -151,6 +152,10 @@ static void mp3_parse_info_tag(AVFormatContext *s,
> AVStream *st,
> mp3->frames = avio_rb32(s->pb);
> if (v & XING_FLAG_SIZE)
> mp3->header_filesize = avio_rb32(s->pb);
> +min = FFMIN(fsize, mp3->header_filesize);
> +delta = FFMAX(fsize, mp3->header_filesize) - min;
> +if (fsize && mp3->header_filesize > 0 && delta > min >> 4)
> +mp3->frames = 0;
> if (v & XING_FLAG_TOC)
> read_xing_toc(s, mp3->header_filesize, av_rescale_q(mp3->frames,
> (AVRational){spf, c->sample_rate},
No more remarks from me, but I do not maintain that part of the code.
Thanks.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] lavfi/drawtext: add alias "expr_int_format" to expansion function "eif"
---
doc/filters.texi | 2 +-
libavfilter/vf_drawtext.c | 3 ++-
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/doc/filters.texi b/doc/filters.texi
index 8cde277..a7919a3 100644
--- a/doc/filters.texi
+++ b/doc/filters.texi
@@ -3916,7 +3916,7 @@ example the text size is not known when evaluating the
expression, so
the constants @var{text_w} and @var{text_h} will have an undefined
value.
-@item eif
+@item expr_int_format, eif
Evaluate the expression's value and output as formatted integer.
First argument is expression to be evaluated, same as for @var{expr} function.
diff --git a/libavfilter/vf_drawtext.c b/libavfilter/vf_drawtext.c
index c744d93..b7a295f 100644
--- a/libavfilter/vf_drawtext.c
+++ b/libavfilter/vf_drawtext.c
@@ -949,7 +949,7 @@ static int func_eval_expr_int_format(AVFilterContext *ctx,
AVBPrint *bp,
if (argc == 3) {
ret = sscanf(argv[2], "%u", &positions);
if (ret != 1) {
-av_log(ctx, AV_LOG_ERROR, "eif(): Invalid number of positions"
+av_log(ctx, AV_LOG_ERROR, "expr_int_format(): Invalid number of
positions"
" to print: '%s'\n", argv[2]);
return AVERROR(EINVAL);
}
@@ -982,6 +982,7 @@ static const struct drawtext_function {
} functions[] = {
{ "expr", 1, 1, 0, func_eval_expr },
{ "e", 1, 1, 0, func_eval_expr },
+{ "expr_int_format", 2, 3, 0, func_eval_expr_int_format },
{ "eif", 2, 3, 0, func_eval_expr_int_format },
{ "pict_type", 0, 0, 0, func_pict_type },
{ "pts", 0, 2, 0, func_pts },
--
1.8.5.5
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavfi/drawtext: add alias "expr_int_format" to expansion function "eif"
Le decadi 30 messidor, an CCXXII, Andrey Utkin a écrit : > --- > doc/filters.texi | 2 +- > libavfilter/vf_drawtext.c | 3 ++- > 2 files changed, 3 insertions(+), 2 deletions(-) IMHO, since the function was added extremely recently, it is ok to just change the name. May I suggest: format_expr()? That way, if somebody extends it to accept more versatile format strings, including floats, the name still works. Regards, -- Nicolas George signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavfi/drawtext: add alias "expr_int_format" to expansion function "eif"
2014-07-18 1:05 GMT+03:00 Nicolas George : > Le decadi 30 messidor, an CCXXII, Andrey Utkin a écrit : >> --- >> doc/filters.texi | 2 +- >> libavfilter/vf_drawtext.c | 3 ++- >> 2 files changed, 3 insertions(+), 2 deletions(-) > > IMHO, since the function was added extremely recently, it is ok to just > change the name. Not principal on this, i just think it is natural, as long as we have "e" alias for "expr". I think "eif" is good to have, too. Aliases cost nothing there. > May I suggest: format_expr()? That way, if somebody extends it to accept > more versatile format strings, including floats, the name still works. Floats are output by "expr". And i'm afraid that making the formatting function more generic would add a lot of complexity (including safety checks). Actually now i cannot imagine how would it look like, and what are the use cases. -- Andrey Utkin ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] oss_audio: check all ioctl() return codes
Also uses a macro to simplify.
Signed-off-by: Timothy Gu
---
Found with clang's static analyzer.
---
libavdevice/oss_audio.c | 23 +++
1 file changed, 11 insertions(+), 12 deletions(-)
diff --git a/libavdevice/oss_audio.c b/libavdevice/oss_audio.c
index 734e565..3c5065c 100644
--- a/libavdevice/oss_audio.c
+++ b/libavdevice/oss_audio.c
@@ -87,8 +87,15 @@ static int audio_open(AVFormatContext *s1, int is_output,
const char *audio_devi
s->frame_size = AUDIO_BLOCK_SIZE;
+#define CHECK_IOCTL_ERROR(event) \
+if (err < 0) {\
+av_log(s1, AV_LOG_ERROR, #event ": %s\n", strerror(errno)); \
+goto fail;\
+}
+
/* select format : favour native format */
err = ioctl(audio_fd, SNDCTL_DSP_GETFMTS, &tmp);
+CHECK_IOCTL_ERROR(SNDCTL_DSP_GETFMTS)
#if HAVE_BIGENDIAN
if (tmp & AFMT_S16_BE) {
@@ -121,24 +128,15 @@ static int audio_open(AVFormatContext *s1, int is_output,
const char *audio_devi
return AVERROR(EIO);
}
err=ioctl(audio_fd, SNDCTL_DSP_SETFMT, &tmp);
-if (err < 0) {
-av_log(s1, AV_LOG_ERROR, "SNDCTL_DSP_SETFMT: %s\n", strerror(errno));
-goto fail;
-}
+CHECK_IOCTL_ERROR(SNDCTL_DSP_SETFMTS)
tmp = (s->channels == 2);
err = ioctl(audio_fd, SNDCTL_DSP_STEREO, &tmp);
-if (err < 0) {
-av_log(s1, AV_LOG_ERROR, "SNDCTL_DSP_STEREO: %s\n", strerror(errno));
-goto fail;
-}
+CHECK_IOCTL_ERROR(SNDCTL_DSP_STEREO)
tmp = s->sample_rate;
err = ioctl(audio_fd, SNDCTL_DSP_SPEED, &tmp);
-if (err < 0) {
-av_log(s1, AV_LOG_ERROR, "SNDCTL_DSP_SPEED: %s\n", strerror(errno));
-goto fail;
-}
+CHECK_IOCTL_ERROR(SNDCTL_DSP_SPEED)
s->sample_rate = tmp; /* store real sample rate */
s->fd = audio_fd;
@@ -146,6 +144,7 @@ static int audio_open(AVFormatContext *s1, int is_output,
const char *audio_devi
fail:
close(audio_fd);
return AVERROR(EIO);
+#undef CHECK_IOCTL_ERROR
}
static int audio_close(AudioData *s)
--
1.9.1
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] ffplay: remove dead code
Signed-off-by: Timothy Gu --- ffplay.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/ffplay.c b/ffplay.c index af0e199..8abbeae 100644 --- a/ffplay.c +++ b/ffplay.c @@ -775,8 +775,6 @@ static void blend_subrect(AVPicture *dst, const AVSubtitleRect *rect, int imgw, lum[0] = ALPHA_BLEND(a, lum[0], y, 0); YUVA_IN(y, u, v, a, p + BPP, pal); -u1 += u; -v1 += v; a1 += a; lum[1] = ALPHA_BLEND(a, lum[1], y, 0); cb[0] = ALPHA_BLEND(a1 >> 2, cb[0], u, 1); -- 1.9.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH 1/2] proresenc_anatoliy: use tables in proresdata.c instead of redefining them
Signed-off-by: Timothy Gu
---
libavcodec/Makefile | 4 ++--
libavcodec/proresenc_anatoliy.c | 43 ++---
2 files changed, 16 insertions(+), 31 deletions(-)
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 307ad22..e3f75f5 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -369,8 +369,8 @@ OBJS-$(CONFIG_PPM_DECODER) += pnmdec.o pnm.o
OBJS-$(CONFIG_PPM_ENCODER) += pnmenc.o
OBJS-$(CONFIG_PRORES_DECODER) += proresdec2.o proresdsp.o proresdata.o
OBJS-$(CONFIG_PRORES_LGPL_DECODER) += proresdec_lgpl.o proresdsp.o
proresdata.o
-OBJS-$(CONFIG_PRORES_ENCODER) += proresenc_anatoliy.o
-OBJS-$(CONFIG_PRORES_AW_ENCODER) += proresenc_anatoliy.o
+OBJS-$(CONFIG_PRORES_ENCODER) += proresenc_anatoliy.o proresdata.o
+OBJS-$(CONFIG_PRORES_AW_ENCODER) += proresenc_anatoliy.o proresdata.o
OBJS-$(CONFIG_PRORES_KS_ENCODER) += proresenc_kostya.o proresdata.o
OBJS-$(CONFIG_PTX_DECODER) += ptx.o
OBJS-$(CONFIG_QCELP_DECODER) += qcelpdec.o \
diff --git a/libavcodec/proresenc_anatoliy.c b/libavcodec/proresenc_anatoliy.c
index f471f49..b8531cd 100644
--- a/libavcodec/proresenc_anatoliy.c
+++ b/libavcodec/proresenc_anatoliy.c
@@ -32,6 +32,7 @@
#include "put_bits.h"
#include "bytestream.h"
#include "fdctdsp.h"
+#include "proresdata.h"
#define DEFAULT_SLICE_MB_WIDTH 8
@@ -52,17 +53,6 @@ static const int qp_start_table[4] = { 4, 1, 1, 1 };
static const int qp_end_table[4] = { 8, 9, 6, 6 };
static const int bitrate_table[5] = { 1000, 2100, 3500, 5400 };
-static const uint8_t progressive_scan[64] = {
- 0, 1, 8, 9, 2, 3, 10, 11,
-16, 17, 24, 25, 18, 19, 26, 27,
- 4, 5, 12, 20, 13, 6, 7, 14,
-21, 28, 29, 22, 15, 23, 30, 31,
-32, 33, 40, 48, 41, 34, 35, 42,
-49, 56, 57, 50, 43, 36, 37, 44,
-51, 58, 59, 52, 45, 38, 39, 46,
-53, 60, 61, 54, 47, 55, 62, 63
-};
-
static const uint8_t QMAT_LUMA[4][64] = {
{
4, 7, 9, 11, 13, 14, 15, 63,
@@ -196,10 +186,6 @@ static av_always_inline int get_level(int val)
return (val ^ sign) - sign;
}
-#define FIRST_DC_CB 0xB8
-
-static const uint8_t dc_codebook[7] = { 0x04, 0x28, 0x28, 0x4D, 0x4D, 0x70,
0x70};
-
static void encode_dc_coeffs(PutBitContext *pb, int16_t *in,
int blocks_per_slice, int *qmat)
{
@@ -211,48 +197,47 @@ static void encode_dc_coeffs(PutBitContext *pb, int16_t
*in,
code = TO_GOLOMB(prev_dc);
encode_codeword(pb, code, FIRST_DC_CB);
-code = 5; sign = 0; idx = 64;
+code = 3; sign = 0; idx = 64;
for (i = 1; i < blocks_per_slice; i++, idx += 64) {
new_dc= QSCALE(qmat, 0, in[idx] - 16384);
delta = new_dc - prev_dc;
diff_sign = DIFF_SIGN(delta, sign);
new_code = TO_GOLOMB2(get_level(delta), diff_sign);
-
-encode_codeword(pb, new_code, dc_codebook[FFMIN(code, 6)]);
-
-code = new_code;
+encode_codeword(pb, new_code, ff_prores_dc_codebook[code]);
+code = (new_code + (new_code & 1)) >> 1;
+code = FFMIN(code, 3);
sign = delta >> 31;
prev_dc = new_dc;
}
}
-static const uint8_t run_to_cb[16] = { 0x06, 0x06, 0x05, 0x05, 0x04, 0x29,
-0x29, 0x29, 0x29, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x4C };
-static const uint8_t lev_to_cb[10] = { 0x04, 0x0A, 0x05, 0x06, 0x04, 0x28,
-0x28, 0x28, 0x28, 0x4C };
-
static void encode_ac_coeffs(AVCodecContext *avctx, PutBitContext *pb,
int16_t *in, int blocks_per_slice, int *qmat)
{
int prev_run = 4;
int prev_level = 2;
-int run = 0, level, code, i, j;
+int run = 0, level, code, i, j, run_cb, lev_cb;
+run_cb = ff_prores_run_to_cb_index[4];
+lev_cb = ff_prores_lev_to_cb_index[2];
+
for (i = 1; i < 64; i++) {
-int indp = progressive_scan[i];
+int indp = ff_prores_progressive_scan[i];
for (j = 0; j < blocks_per_slice; j++) {
int val = QSCALE(qmat, indp, in[(j << 6) + indp]);
if (val) {
-encode_codeword(pb, run, run_to_cb[FFMIN(prev_run, 15)]);
+encode_codeword(pb, run, ff_prores_ac_codebook[run_cb]);
prev_run = run;
run= 0;
level = get_level(val);
code = level - 1;
-encode_codeword(pb, code, lev_to_cb[FFMIN(prev_level, 9)]);
+encode_codeword(pb, code, ff_prores_ac_codebook[lev_cb]);
prev_level = level;
+run_cb = ff_prores_run_to_cb_index[FFMIN(prev_run, 15)];
+lev_cb = ff_prores_lev_to_cb_index[FFMIN(prev_level, 9)];
put_bits(pb, 1, IS_NEGATIVE(val));
} else {
--
1.9.1
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman
[FFmpeg-devel] [PATCH 2/2] proresenc: move profile handling to global avcodec options
This allows users to use the two encoders with the same syntax.
Signed-off-by: Timothy Gu
---
libavcodec/avcodec.h| 5 +
libavcodec/options_table.h | 5 +
libavcodec/proresenc_anatoliy.c | 5 -
libavcodec/proresenc_kostya.c | 44 +
4 files changed, 33 insertions(+), 26 deletions(-)
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index 93ba4d0..cd29065 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -2862,6 +2862,11 @@ typedef struct AVCodecContext {
#define FF_PROFILE_JPEG2000_DCINEMA_2K 3
#define FF_PROFILE_JPEG2000_DCINEMA_4K 4
+#define FF_PROFILE_PRORES_PROXY 0
+#define FF_PROFILE_PRORES_LT1
+#define FF_PROFILE_PRORES_STANDARD 2
+#define FF_PROFILE_PRORES_HQ3
+#define FF_PROFILE_PRORES_ 4
#define FF_PROFILE_HEVC_MAIN1
#define FF_PROFILE_HEVC_MAIN_10 2
diff --git a/libavcodec/options_table.h b/libavcodec/options_table.h
index cbefa52..7d0a425 100644
--- a/libavcodec/options_table.h
+++ b/libavcodec/options_table.h
@@ -338,6 +338,11 @@ static const AVOption avcodec_options[] = {
{"dts_96_24", NULL, 0, AV_OPT_TYPE_CONST, {.i64 = FF_PROFILE_DTS_96_24 },
INT_MIN, INT_MAX, A|E, "profile"},
{"dts_hd_hra", NULL, 0, AV_OPT_TYPE_CONST, {.i64 = FF_PROFILE_DTS_HD_HRA },
INT_MIN, INT_MAX, A|E, "profile"},
{"dts_hd_ma", NULL, 0, AV_OPT_TYPE_CONST, {.i64 = FF_PROFILE_DTS_HD_MA },
INT_MIN, INT_MAX, A|E, "profile"},
+{"prores_proxy", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = FF_PROFILE_PRORES_PROXY
}, INT_MIN, INT_MAX, V|E, "profile" },
+{"prores_lt",NULL, 0, AV_OPT_TYPE_CONST, { .i64 = FF_PROFILE_PRORES_LT },
INT_MIN, INT_MAX, V|E, "profile" },
+{"prores_standard",NULL, 0, AV_OPT_TYPE_CONST, { .i64 =
FF_PROFILE_PRORES_STANDARD }, INT_MIN, INT_MAX, V|E, "profile" },
+{"prores_hq",NULL, 0, AV_OPT_TYPE_CONST, { .i64 = FF_PROFILE_PRORES_HQ },
INT_MIN, INT_MAX, V|E, "profile" },
+{"prores_", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = FF_PROFILE_PRORES_
}, INT_MIN, INT_MAX, V|E, "profile" },
{"level", NULL, OFFSET(level), AV_OPT_TYPE_INT, {.i64 = FF_LEVEL_UNKNOWN },
INT_MIN, INT_MAX, V|A|E, "level"},
{"unknown", NULL, 0, AV_OPT_TYPE_CONST, {.i64 = FF_LEVEL_UNKNOWN }, INT_MIN,
INT_MAX, V|A|E, "level"},
{"lowres", "decode at 1= 1/2, 2=1/4, 3=1/8 resolutions", OFFSET(lowres),
AV_OPT_TYPE_INT, {.i64 = 0 }, 0, INT_MAX, V|A|D},
diff --git a/libavcodec/proresenc_anatoliy.c b/libavcodec/proresenc_anatoliy.c
index b8531cd..b66f405 100644
--- a/libavcodec/proresenc_anatoliy.c
+++ b/libavcodec/proresenc_anatoliy.c
@@ -36,11 +36,6 @@
#define DEFAULT_SLICE_MB_WIDTH 8
-#define FF_PROFILE_PRORES_PROXY 0
-#define FF_PROFILE_PRORES_LT1
-#define FF_PROFILE_PRORES_STANDARD 2
-#define FF_PROFILE_PRORES_HQ3
-
static const AVProfile profiles[] = {
{ FF_PROFILE_PRORES_PROXY,"apco"},
{ FF_PROFILE_PRORES_LT, "apcs"},
diff --git a/libavcodec/proresenc_kostya.c b/libavcodec/proresenc_kostya.c
index 93bcde7..9505df0 100644
--- a/libavcodec/proresenc_kostya.c
+++ b/libavcodec/proresenc_kostya.c
@@ -40,14 +40,6 @@
#define MAX_PLANES 4
enum {
-PRORES_PROFILE_PROXY = 0,
-PRORES_PROFILE_LT,
-PRORES_PROFILE_STANDARD,
-PRORES_PROFILE_HQ,
-PRORES_PROFILE_,
-};
-
-enum {
QUANT_MAT_PROXY = 0,
QUANT_MAT_LT,
QUANT_MAT_STANDARD,
@@ -1214,6 +1206,19 @@ static av_cold int encode_init(AVCodecContext *avctx)
avctx->codec_tag = ctx->profile_info->tag;
+if (avctx->profile == FF_PROFILE_UNKNOWN) {
+avctx->profile = FF_PROFILE_PRORES_STANDARD;
+av_log(avctx, AV_LOG_INFO,
+ "no profile specified. encoding with ProRes standard (apcn)
profile\n");
+} else if (avctx->profile < FF_PROFILE_PRORES_PROXY
+|| avctx->profile > FF_PROFILE_PRORES_) {
+av_log(avctx, AV_LOG_ERROR,
+ "unknown profile %d. Supported profiles: prores_proxy,
prores_lt,\n"
+ "prores_standard, prores_hq, prores_.\n",
+ avctx->profile);
+return AVERROR(EINVAL);
+}
+
av_log(avctx, AV_LOG_DEBUG,
"profile %d, %d slices, interlacing: %s, %d bits per MB\n",
ctx->profile, ctx->slices_per_picture * ctx->pictures_per_frame,
@@ -1230,19 +1235,6 @@ static av_cold int encode_init(AVCodecContext *avctx)
static const AVOption options[] = {
{ "mbs_per_slice", "macroblocks per slice", OFFSET(mbs_per_slice),
AV_OPT_TYPE_INT, { .i64 = 8 }, 1, MAX_MBS_PER_SLICE, VE },
-{ "profile", NULL, OFFSET(profile), AV_OPT_TYPE_INT,
-{ .i64 = PRORES_PROFILE_STANDARD },
-PRORES_PROFILE_PROXY, PRORES_PROFILE_, VE, "profile" },
-{ "proxy", NULL, 0, AV_OPT_TYPE_CONST, { .i64 =
PRORES_PROFILE_PROXY },
-0, 0, VE, "profile" },
-{ "lt",NULL, 0, AV_OPT_TYPE_CONST, { .i64 = PRORES_PROFI
Re: [FFmpeg-devel] [Updated PATCH 2/4] armv6: Accelerate ff_fft_calc for general case (nbits != 4)
On Fri, Jul 11, 2014 at 11:32:08AM +0100, Ben Avison wrote: > The previous implementation targeted DTS Coherent Acoustics, which only > requires nbits == 4 (fft16()). This case was (and still is) linked directly > rather than being indirected through ff_fft_calc_vfp(), but now the full > range from radix-4 up to radix-65536 is available. This benefits other codecs > such as AAC and AC3. > > The implementaion is based upon the C version, with each routine larger than > radix-16 calling a hierarchy of smaller FFT functions, then performing a > post-processing pass. This pass benefits a lot from loop unrolling to > counter the long pipelines in the VFP. A relaxed calling standard also > reduces the overhead of the call hierarchy, and avoiding the excessive > inlining performed by GCC probably helps with I-cache utilisation too. > > I benchmarked the result by measuring the number of gperftools samples that > hit anywhere in the AAC decoder (starting from aac_decode_frame()) or > specifically in the FFT routines (fft4() to fft512() and pass()) for the > same sample AAC stream: > > Before After > Mean StdDev Mean StdDev Confidence Change > Audio decode 2245.5 53.1 1599.6 43.8100.0% +40.4% > FFT routines 940.6 22.0 348.1 20.8100.0% +170.2% > --- > libavcodec/arm/fft_init_arm.c |8 +- > libavcodec/arm/fft_vfp.S | 284 > +++-- > 2 files changed, 275 insertions(+), 17 deletions(-) merged a variant of this patch [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Complexity theory is the science of finding the exact solution to an approximation. Benchmarking OTOH is finding an approximation of the exact signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] hevc: propagate error code from set_sps()
Signed-off-by: Timothy Gu --- libavcodec/hevc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libavcodec/hevc.c b/libavcodec/hevc.c index afb2baa..d3108d7 100644 --- a/libavcodec/hevc.c +++ b/libavcodec/hevc.c @@ -3299,7 +3299,8 @@ static int hevc_update_thread_context(AVCodecContext *dst, } if (s->sps != s0->sps) -ret = set_sps(s, s0->sps); +if ((ret = set_sps(s, s0->sps)) < 0) +return ret; s->seq_decode = s0->seq_decode; s->seq_output = s0->seq_output; -- 1.9.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 1/2] proresenc_anatoliy: use tables in proresdata.c instead of redefining them
On Thu, Jul 17, 2014 at 06:04:06PM -0700, Timothy Gu wrote:
> Signed-off-by: Timothy Gu
> ---
> libavcodec/Makefile | 4 ++--
> libavcodec/proresenc_anatoliy.c | 43
> ++---
> 2 files changed, 16 insertions(+), 31 deletions(-)
it appears this makes the code slightly slower
before patch:
233505 decicycles in coeffs, 1048576 runs, 0 skips
233491 decicycles in coeffs, 1048576 runs, 0 skips
233492 decicycles in coeffs, 1048575 runs, 1 skips
with patch:
234461 decicycles in coeffs, 1048576 runs, 0 skips
234602 decicycles in coeffs, 1048576 runs, 0 skips
234546 decicycles in coeffs, 1048576 runs, 0 skips
tested with:
diff --git a/libavcodec/proresenc_anatoliy.c b/libavcodec/proresenc_anatoliy.c
index b8531cd..4f00b5c 100644
--- a/libavcodec/proresenc_anatoliy.c
+++ b/libavcodec/proresenc_anatoliy.c
@@ -290,10 +290,10 @@ static int encode_slice_plane(AVCodecContext *avctx, int
mb_count,
blocks_per_slice = mb_count << (2 - chroma);
init_put_bits(&pb, buf, buf_size << 3);
-
+START_TIMER
encode_dc_coeffs(&pb, blocks, blocks_per_slice, qmat);
encode_ac_coeffs(avctx, &pb, blocks, blocks_per_slice, qmat);
-
+STOP_TIMER("coeffs")
flush_put_bits(&pb);
return put_bits_ptr(&pb) - pb.buf;
}
and
./ffmpeg -y -i matrixbench_mpeg2.mpg -an -threads 1 -vcodec prores -f null -
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
Old school: Use the lowest level language in which you can solve the problem
conveniently.
New school: Use the highest level language in which the latest supercomputer
can solve the problem without the user falling asleep waiting.
signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH 3/4] smacker: remove dead code
Signed-off-by: Timothy Gu
---
libavcodec/smacker.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/libavcodec/smacker.c b/libavcodec/smacker.c
index 644beb3..518bdad 100644
--- a/libavcodec/smacker.c
+++ b/libavcodec/smacker.c
@@ -438,7 +438,6 @@ static int decode_frame(AVCodecContext *avctx, void *data,
int *got_frame,
bw = avctx->width >> 2;
bh = avctx->height >> 2;
blocks = bw * bh;
-out = smk->pic->data[0];
stride = smk->pic->linesize[0];
while(blk < blocks) {
int type, run, mode;
@@ -499,7 +498,6 @@ static int decode_frame(AVCodecContext *avctx, void *data,
int *got_frame,
out += stride;
out[0] = out[1] = pix & 0xFF;
out[2] = out[3] = pix >> 8;
-out += stride;
break;
case 2:
for(i = 0; i < 2; i++) {
--
1.9.1
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH 2/4] kerndeint: remove dead assignments
Signed-off-by: Timothy Gu --- libavfilter/vf_kerndeint.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavfilter/vf_kerndeint.c b/libavfilter/vf_kerndeint.c index 1f8e091..5130208 100644 --- a/libavfilter/vf_kerndeint.c +++ b/libavfilter/vf_kerndeint.c @@ -154,10 +154,10 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *inpic) h = plane == 0 ? inlink->h : FF_CEIL_RSHIFT(inlink->h, kerndeint->vsub); bwidth = kerndeint->tmp_bwidth[plane]; -srcp = srcp_saved = inpic->data[plane]; +srcp_saved= inpic->data[plane]; src_linesize = inpic->linesize[plane]; psrc_linesize = kerndeint->tmp_linesize[plane]; -dstp = dstp_saved = outpic->data[plane]; +dstp_saved= outpic->data[plane]; dst_linesize = outpic->linesize[plane]; srcp = srcp_saved + (1 - order) * src_linesize; dstp = dstp_saved + (1 - order) * dst_linesize; -- 1.9.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH 1/4] mpeg: remove unused assignment
Signed-off-by: Timothy Gu --- libavformat/mpeg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavformat/mpeg.c b/libavformat/mpeg.c index c53bab3..5891cea 100644 --- a/libavformat/mpeg.c +++ b/libavformat/mpeg.c @@ -204,8 +204,8 @@ static long mpegps_psm_parse(MpegDemuxContext *m, AVIOContext *pb) /* skip program_stream_info */ avio_skip(pb, ps_info_length); -es_map_length = avio_rb16(pb); /* Ignore es_map_length, trust psm_length */ +avio_rb16(pb); es_map_length = psm_length - ps_info_length - 10; /* at least one es available? */ -- 1.9.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH 4/4] a64multienc: remove dead assignment
Signed-off-by: Timothy Gu --- libavcodec/a64multienc.c | 1 - 1 file changed, 1 deletion(-) diff --git a/libavcodec/a64multienc.c b/libavcodec/a64multienc.c index 3067774..9760298 100644 --- a/libavcodec/a64multienc.c +++ b/libavcodec/a64multienc.c @@ -351,7 +351,6 @@ static int a64multi_encode_frame(AVCodecContext *avctx, AVPacket *pkt, /* advance pointers */ buf += charset_size; -charset += charset_size; } /* write x frames to buf */ -- 1.9.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] proresenc: move profile handling to global avcodec options
On Thu, Jul 17, 2014 at 06:04:07PM -0700, Timothy Gu wrote: > This allows users to use the two encoders with the same syntax. > > Signed-off-by: Timothy Gu > --- > libavcodec/avcodec.h| 5 + > libavcodec/options_table.h | 5 + > libavcodec/proresenc_anatoliy.c | 5 - > libavcodec/proresenc_kostya.c | 44 > + seems to break fate --- ./tests/ref/vsynth/vsynth1-prores_ks2014-07-17 16:00:41.213032673 +0200 +++ tests/data/fate/vsynth1-prores_ks 2014-07-18 04:40:34.717993203 +0200 @@ -1,4 +0,0 @@ -fe41a284da97ea5ec8866ca9a55b84da *tests/data/fate/vsynth1-prores_ks.mov -3858911 tests/data/fate/vsynth1-prores_ks.mov -100eb002413fe7a632d440dfbdf7e3ff *tests/data/fate/vsynth1-prores_ks.out.rawvideo -stddev:3.17 PSNR: 38.09 MAXDIFF: 39 bytes: 7603200/ 7603200 [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB The real ebay dictionary, page 1 "Used only once"- "Some unspecified defect prevented a second use" "In good condition" - "Can be repaird by experienced expert" "As is" - "You wouldnt want it even if you were payed for it, if you knew ..." signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH 1/2] alpha/idctdsp: move disabled cruft out of dsputil
Signed-off-by: James Almer
---
libavcodec/alpha/dsputil_alpha.c | 72
libavcodec/alpha/idctdsp_alpha.c | 72
2 files changed, 72 insertions(+), 72 deletions(-)
diff --git a/libavcodec/alpha/dsputil_alpha.c b/libavcodec/alpha/dsputil_alpha.c
index d99a74b..e00497d 100644
--- a/libavcodec/alpha/dsputil_alpha.c
+++ b/libavcodec/alpha/dsputil_alpha.c
@@ -24,78 +24,6 @@
#include "dsputil_alpha.h"
#include "asm.h"
-#if 0
-/* These functions were the base for the optimized assembler routines,
- and remain here for documentation purposes. */
-static void put_pixels_clamped_mvi(const int16_t *block, uint8_t *pixels,
- ptrdiff_t line_size)
-{
-int i = 8;
-uint64_t clampmask = zap(-1, 0xaa); /* 0x00ff00ff00ff00ff */
-
-do {
-uint64_t shorts0, shorts1;
-
-shorts0 = ldq(block);
-shorts0 = maxsw4(shorts0, 0);
-shorts0 = minsw4(shorts0, clampmask);
-stl(pkwb(shorts0), pixels);
-
-shorts1 = ldq(block + 4);
-shorts1 = maxsw4(shorts1, 0);
-shorts1 = minsw4(shorts1, clampmask);
-stl(pkwb(shorts1), pixels + 4);
-
-pixels += line_size;
-block += 8;
-} while (--i);
-}
-
-void add_pixels_clamped_mvi(const int16_t *block, uint8_t *pixels,
-ptrdiff_t line_size)
-{
-int h = 8;
-/* Keep this function a leaf function by generating the constants
- manually (mainly for the hack value ;-). */
-uint64_t clampmask = zap(-1, 0xaa); /* 0x00ff00ff00ff00ff */
-uint64_t signmask = zap(-1, 0x33);
-signmask ^= signmask >> 1; /* 0x8000800080008000 */
-
-do {
-uint64_t shorts0, pix0, signs0;
-uint64_t shorts1, pix1, signs1;
-
-shorts0 = ldq(block);
-shorts1 = ldq(block + 4);
-
-pix0= unpkbw(ldl(pixels));
-/* Signed subword add (MMX paddw). */
-signs0 = shorts0 & signmask;
-shorts0 &= ~signmask;
-shorts0 += pix0;
-shorts0 ^= signs0;
-/* Clamp. */
-shorts0 = maxsw4(shorts0, 0);
-shorts0 = minsw4(shorts0, clampmask);
-
-/* Next 4. */
-pix1= unpkbw(ldl(pixels + 4));
-signs1 = shorts1 & signmask;
-shorts1 &= ~signmask;
-shorts1 += pix1;
-shorts1 ^= signs1;
-shorts1 = maxsw4(shorts1, 0);
-shorts1 = minsw4(shorts1, clampmask);
-
-stl(pkwb(shorts0), pixels);
-stl(pkwb(shorts1), pixels + 4);
-
-pixels += line_size;
-block += 8;
-} while (--h);
-}
-#endif
-
av_cold void ff_dsputil_init_alpha(DSPContext *c, AVCodecContext *avctx)
{
/* amask clears all bits that correspond to present features. */
diff --git a/libavcodec/alpha/idctdsp_alpha.c b/libavcodec/alpha/idctdsp_alpha.c
index 73f4821..1050697 100644
--- a/libavcodec/alpha/idctdsp_alpha.c
+++ b/libavcodec/alpha/idctdsp_alpha.c
@@ -33,6 +33,78 @@ void (*put_pixels_clamped_axp_p)(const int16_t *block,
uint8_t *pixels,
void (*add_pixels_clamped_axp_p)(const int16_t *block, uint8_t *pixels,
int line_size);
+#if 0
+/* These functions were the base for the optimized assembler routines,
+ and remain here for documentation purposes. */
+static void put_pixels_clamped_mvi(const int16_t *block, uint8_t *pixels,
+ ptrdiff_t line_size)
+{
+int i = 8;
+uint64_t clampmask = zap(-1, 0xaa); /* 0x00ff00ff00ff00ff */
+
+do {
+uint64_t shorts0, shorts1;
+
+shorts0 = ldq(block);
+shorts0 = maxsw4(shorts0, 0);
+shorts0 = minsw4(shorts0, clampmask);
+stl(pkwb(shorts0), pixels);
+
+shorts1 = ldq(block + 4);
+shorts1 = maxsw4(shorts1, 0);
+shorts1 = minsw4(shorts1, clampmask);
+stl(pkwb(shorts1), pixels + 4);
+
+pixels += line_size;
+block += 8;
+} while (--i);
+}
+
+void add_pixels_clamped_mvi(const int16_t *block, uint8_t *pixels,
+ptrdiff_t line_size)
+{
+int h = 8;
+/* Keep this function a leaf function by generating the constants
+ manually (mainly for the hack value ;-). */
+uint64_t clampmask = zap(-1, 0xaa); /* 0x00ff00ff00ff00ff */
+uint64_t signmask = zap(-1, 0x33);
+signmask ^= signmask >> 1; /* 0x8000800080008000 */
+
+do {
+uint64_t shorts0, pix0, signs0;
+uint64_t shorts1, pix1, signs1;
+
+shorts0 = ldq(block);
+shorts1 = ldq(block + 4);
+
+pix0= unpkbw(ldl(pixels));
+/* Signed subword add (MMX paddw). */
+signs0 = shorts0 & signmask;
+shorts0 &= ~signmask;
+shorts0 += pix0;
+shorts0 ^= signs0;
+/* Clamp. */
+shorts0 = maxsw4(shorts0, 0);
+shorts0 = minsw4(shorts0, clampmask);
+
+/* Next 4. */
+pix1= unpkbw(ldl(pixels + 4));
+signs1 = s
[FFmpeg-devel] [PATCH 2/2] alpha/me_cmp: move code out of dsputil
Signed-off-by: James Almer
---
Untested.
libavcodec/alpha/Makefile | 5 ++-
libavcodec/alpha/dsputil_alpha.c | 40 --
libavcodec/alpha/dsputil_alpha.h | 32 -
.../alpha/{motion_est_alpha.c => me_cmp_alpha.c} | 27 ---
.../{motion_est_mvi_asm.S => me_cmp_mvi_asm.S} | 0
libavcodec/alpha/pixblockdsp_alpha.c | 1 -
6 files changed, 24 insertions(+), 81 deletions(-)
delete mode 100644 libavcodec/alpha/dsputil_alpha.c
delete mode 100644 libavcodec/alpha/dsputil_alpha.h
rename libavcodec/alpha/{motion_est_alpha.c => me_cmp_alpha.c} (88%)
rename libavcodec/alpha/{motion_est_mvi_asm.S => me_cmp_mvi_asm.S} (100%)
diff --git a/libavcodec/alpha/Makefile b/libavcodec/alpha/Makefile
index 42cabfe..796d976 100644
--- a/libavcodec/alpha/Makefile
+++ b/libavcodec/alpha/Makefile
@@ -1,7 +1,6 @@
OBJS-$(CONFIG_BLOCKDSP) += alpha/blockdsp_alpha.o
-OBJS-$(CONFIG_DSPUTIL) += alpha/dsputil_alpha.o\
- alpha/motion_est_alpha.o \
- alpha/motion_est_mvi_asm.o
+OBJS-$(CONFIG_ME_CMP) += alpha/me_cmp_alpha.o \
+ alpha/me_cmp_mvi_asm.o
OBJS-$(CONFIG_HPELDSP) += alpha/hpeldsp_alpha.o\
alpha/hpeldsp_alpha_asm.o
OBJS-$(CONFIG_IDCTDSP) += alpha/idctdsp_alpha.o\
diff --git a/libavcodec/alpha/dsputil_alpha.c b/libavcodec/alpha/dsputil_alpha.c
deleted file mode 100644
index e00497d..000
--- a/libavcodec/alpha/dsputil_alpha.c
+++ /dev/null
@@ -1,40 +0,0 @@
-/*
- * Alpha optimized DSP utils
- * Copyright (c) 2002 Falk Hueffner
- *
- * This file is part of FFmpeg.
- *
- * FFmpeg is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation; either
- * version 2.1 of the License, or (at your option) any later version.
- *
- * FFmpeg is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with FFmpeg; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
- */
-
-#include "libavutil/attributes.h"
-#include "libavcodec/dsputil.h"
-#include "dsputil_alpha.h"
-#include "asm.h"
-
-av_cold void ff_dsputil_init_alpha(DSPContext *c, AVCodecContext *avctx)
-{
-/* amask clears all bits that correspond to present features. */
-if (amask(AMASK_MVI) == 0) {
-c->sad[0] = pix_abs16x16_mvi_asm;
-c->sad[1] = pix_abs8x8_mvi;
-c->pix_abs[0][0]= pix_abs16x16_mvi_asm;
-c->pix_abs[1][0]= pix_abs8x8_mvi;
-c->pix_abs[0][1]= pix_abs16x16_x2_mvi;
-c->pix_abs[0][2]= pix_abs16x16_y2_mvi;
-c->pix_abs[0][3]= pix_abs16x16_xy2_mvi;
-}
-
-}
diff --git a/libavcodec/alpha/dsputil_alpha.h b/libavcodec/alpha/dsputil_alpha.h
deleted file mode 100644
index a48765f..000
--- a/libavcodec/alpha/dsputil_alpha.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * This file is part of FFmpeg.
- *
- * FFmpeg is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation; either
- * version 2.1 of the License, or (at your option) any later version.
- *
- * FFmpeg is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with FFmpeg; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
- */
-
-#ifndef AVCODEC_ALPHA_DSPUTIL_ALPHA_H
-#define AVCODEC_ALPHA_DSPUTIL_ALPHA_H
-
-#include
-#include
-
-int pix_abs8x8_mvi(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int
h);
-int pix_abs16x16_mvi_asm(void *v, uint8_t *pix1, uint8_t *pix2, int line_size,
int h);
-int pix_abs16x16_x2_mvi(void *v, uint8_t *pix1, uint8_t *pix2, int line_size,
int h);
-int pix_abs16x16_y2_mvi(void *v, uint8_t *pix1, uint8_t *pix2, int line_size,
int h);
-int pix_abs16x16_xy2_mvi(void *v, uint8_t *pix1, uint8_t *pix2, int line_size,
int h);
-
-
-#endif /* AVCODEC_ALPHA_DSPUTIL_ALPHA_H */
diff --git a/libavcodec/alpha/motion_est_alpha.c
b/libavcodec/alpha/me_cmp_alpha.c
similarity index 88%
rename from libavcodec/al
Re: [FFmpeg-devel] [PATCH 2/4] kerndeint: remove dead assignments
On Thu, Jul 17, 2014 at 07:25:41PM -0700, Timothy Gu wrote: > Signed-off-by: Timothy Gu > --- > libavfilter/vf_kerndeint.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) applied thanks [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB I am the wisest man alive, for I know one thing, and that is that I know nothing. -- Socrates signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 4/4] a64multienc: remove dead assignment
On Thu, Jul 17, 2014 at 07:25:43PM -0700, Timothy Gu wrote: > Signed-off-by: Timothy Gu > --- > libavcodec/a64multienc.c | 1 - > 1 file changed, 1 deletion(-) applied thanks [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Freedom in capitalist society always remains about the same as it was in ancient Greek republics: Freedom for slave owners. -- Vladimir Lenin signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 1/2] alpha/idctdsp: move disabled cruft out of dsputil
On Thu, Jul 17, 2014 at 11:45:34PM -0300, James Almer wrote: > Signed-off-by: James Almer > --- > libavcodec/alpha/dsputil_alpha.c | 72 > > libavcodec/alpha/idctdsp_alpha.c | 72 > > 2 files changed, 72 insertions(+), 72 deletions(-) tested applied thanks [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB I do not agree with what you have to say, but I'll defend to the death your right to say it. -- Voltaire signature.asc Description: Digital signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] alpha/me_cmp: move code out of dsputil
On Thu, Jul 17, 2014 at 11:45:35PM -0300, James Almer wrote:
> Signed-off-by: James Almer
> ---
> Untested.
>
> libavcodec/alpha/Makefile | 5 ++-
> libavcodec/alpha/dsputil_alpha.c | 40
> --
> libavcodec/alpha/dsputil_alpha.h | 32 -
> .../alpha/{motion_est_alpha.c => me_cmp_alpha.c} | 27 ---
> .../{motion_est_mvi_asm.S => me_cmp_mvi_asm.S} | 0
> libavcodec/alpha/pixblockdsp_alpha.c | 1 -
> 6 files changed, 24 insertions(+), 81 deletions(-)
> delete mode 100644 libavcodec/alpha/dsputil_alpha.c
> delete mode 100644 libavcodec/alpha/dsputil_alpha.h
> rename libavcodec/alpha/{motion_est_alpha.c => me_cmp_alpha.c} (88%)
> rename libavcodec/alpha/{motion_est_mvi_asm.S => me_cmp_mvi_asm.S} (100%)
tested
applied
thanks
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
Those who are too smart to engage in politics are punished by being
governed by those who are dumber. -- Plato
signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavfi/drawtext: add alias "expr_int_format" to expansion function "eif"
On date Friday 2014-07-18 01:00:40 +0300, Andrey Utkin encoded:
> ---
> doc/filters.texi | 2 +-
> libavfilter/vf_drawtext.c | 3 ++-
> 2 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/doc/filters.texi b/doc/filters.texi
> index 8cde277..a7919a3 100644
> --- a/doc/filters.texi
> +++ b/doc/filters.texi
> @@ -3916,7 +3916,7 @@ example the text size is not known when evaluating the
> expression, so
> the constants @var{text_w} and @var{text_h} will have an undefined
> value.
>
> -@item eif
> +@item expr_int_format, eif
> Evaluate the expression's value and output as formatted integer.
>
> First argument is expression to be evaluated, same as for @var{expr}
> function.
> diff --git a/libavfilter/vf_drawtext.c b/libavfilter/vf_drawtext.c
> index c744d93..b7a295f 100644
> --- a/libavfilter/vf_drawtext.c
> +++ b/libavfilter/vf_drawtext.c
> @@ -949,7 +949,7 @@ static int func_eval_expr_int_format(AVFilterContext
> *ctx, AVBPrint *bp,
> if (argc == 3) {
> ret = sscanf(argv[2], "%u", &positions);
> if (ret != 1) {
> -av_log(ctx, AV_LOG_ERROR, "eif(): Invalid number of positions"
> +av_log(ctx, AV_LOG_ERROR, "expr_int_format(): Invalid number of
> positions"
> " to print: '%s'\n", argv[2]);
> return AVERROR(EINVAL);
> }
> @@ -982,6 +982,7 @@ static const struct drawtext_function {
> } functions[] = {
> { "expr", 1, 1, 0, func_eval_expr },
> { "e", 1, 1, 0, func_eval_expr },
> +{ "expr_int_format", 2, 3, 0, func_eval_expr_int_format },
> { "eif", 2, 3, 0, func_eval_expr_int_format },
> { "pict_type", 0, 0, 0, func_pict_type },
> { "pts", 0, 2, 0, func_pts },
LGTM.
--
FFmpeg = Frightening & Furious Mega Perfectionist Ermetic Gospel
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavfi/drawtext: add alias "expr_int_format" to expansion function "eif"
On Fri, Jul 18, 2014 at 12:05:42AM +0200, Nicolas George wrote: > Le decadi 30 messidor, an CCXXII, Andrey Utkin a écrit : > > --- > > doc/filters.texi | 2 +- > > libavfilter/vf_drawtext.c | 3 ++- > > 2 files changed, 3 insertions(+), 2 deletions(-) > > IMHO, since the function was added extremely recently, it is ok to just > change the name. > It's not present in 2.3 release? [...] -- Clément B. pgpDHEq12DjM8.pgp Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
