August 2020 - ffmpeg-cvslog - lists.ffmpeg.org

doc/decoders: Document default value of ac3's drc_scale option
by Aman Verma 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Aman Verma <amanraoverma(a)gmail.com> | Sun Aug 30 14:05:53 2020 -0400| [e4389ba0751a6738bac2b395662649efcecb60c1] | committer: Michael Niedermayer doc/decoders: Document default value of ac3's drc_scale option Signed-off-by: Aman Verma <amanraoverma(a)gmail.com> Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=e4389ba0751a6738b… --- doc/decoders.texi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/decoders.texi b/doc/decoders.texi index 9005714e3c..8649f83e28 100644 --- a/doc/decoders.texi +++ b/doc/decoders.texi @@ -107,7 +107,7 @@ the undocumented RealAudio 3 (a.k.a. dnet). @item -drc_scale @var{value} Dynamic Range Scale Factor. The factor to apply to dynamic range values -from the AC-3 stream. This factor is applied exponentially. +from the AC-3 stream. This factor is applied exponentially. The default value is 1. There are 3 notable scale factor ranges: @table @option @item drc_scale == 0

1 0

libavformat/dashenc: Reduce confusion in par error message
by Chris Miceli 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Chris Miceli <chris(a)miceli.net.au> | Mon Aug 31 14:16:39 2020 +1000| [41366522899f209d116d663ebedf00e3282e7bb2] | committer: Michael Niedermayer libavformat/dashenc: Reduce confusion in par error message In ticket #8754 there is discourse surrounding the error message which is printed upon a mismatched aspect ratio in derived encodings. This should make it clearer to the user as to the issues which they are experiencing. Reviewed-by: "Jeyapal, Karthick" <kjeyapal(a)akamai.com> Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=41366522899f209d1… --- libavformat/dashenc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavformat/dashenc.c b/libavformat/dashenc.c index dc3306a56a..2d757b3a87 100644 --- a/libavformat/dashenc.c +++ b/libavformat/dashenc.c @@ -1686,7 +1686,7 @@ static int dash_init(AVFormatContext *s) 1024 * 1024); if (as->par.num && av_cmp_q(par, as->par)) { - av_log(s, AV_LOG_ERROR, "Conflicting stream par values in Adaptation Set %d\n", os->as_idx); + av_log(s, AV_LOG_ERROR, "Conflicting stream aspect ratios values in Adaptation Set %d. Please ensure all adaptation sets have the same aspect ratio\n", os->as_idx); return AVERROR(EINVAL); } as->par = par;

1 0

hwcontext_vaapi: Don't require a render node when deriving from DRM
by Mark Thompson 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Mark Thompson <sw(a)jkqxz.net> | Fri Aug 28 23:15:41 2020 +0100| [303d252a4b4c104685dc46152c26abaf7ff2ce60] | committer: Mark Thompson hwcontext_vaapi: Don't require a render node when deriving from DRM The V4L2 driver does not actually have an associated DRM device at all, so users work around the requirement by giving libva an unrelated display-only device instead (which is fine, because it doesn't actually do anything with that device). This was broken by bc9b6358fb7315c0173de322472641766f6289da forcing a render node, because the display-only device did not have an associated render node to use. Fix that by just passing through the original non-render DRM fd if we can't find a render node. Reported-by: Paul Kocialkowski <paul.kocialkowski(a)bootlin.com> Tested-by: Paul Kocialkowski <paul.kocialkowski(a)bootlin.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=303d252a4b4c10468… --- libavutil/hwcontext_vaapi.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c index 38bdeb7820..2227d6ed69 100644 --- a/libavutil/hwcontext_vaapi.c +++ b/libavutil/hwcontext_vaapi.c @@ -1677,20 +1677,24 @@ static int vaapi_device_derive(AVHWDeviceContext *ctx, } else { render_node = drmGetRenderDeviceNameFromFd(src_hwctx->fd); if (!render_node) { - av_log(ctx, AV_LOG_ERROR, "Failed to find a render node " - "matching the DRM device.\n"); - return AVERROR(ENODEV); - } - fd = open(render_node, O_RDWR); - if (fd < 0) { - av_log(ctx, AV_LOG_ERROR, "Failed to open render node %s" - "matching the DRM device.\n", render_node); + av_log(ctx, AV_LOG_VERBOSE, "Using non-render node " + "because the device does not have an " + "associated render node.\n"); + fd = src_hwctx->fd; + } else { + fd = open(render_node, O_RDWR); + if (fd < 0) { + av_log(ctx, AV_LOG_VERBOSE, "Using non-render node " + "because the associated render node " + "could not be opened.\n"); + fd = src_hwctx->fd; + } else { + av_log(ctx, AV_LOG_VERBOSE, "Using render node %s " + "in place of non-render DRM device.\n", + render_node); + } free(render_node); - return AVERROR(errno); } - av_log(ctx, AV_LOG_VERBOSE, "Using render node %s in place " - "of non-render DRM device.\n", render_node); - free(render_node); } } #else

1 0

avcodec/sheervideo: use cached bitstream reader except for x32
by Paul B Mahol 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Aug 31 20:09:28 2020 +0200| [04d335e4365812b8db1d02ac59b5e3a51cc0f425] | committer: Paul B Mahol avcodec/sheervideo: use cached bitstream reader except for x32 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=04d335e4365812b8d… --- libavcodec/sheervideo.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/libavcodec/sheervideo.c b/libavcodec/sheervideo.c index 58776421d7..ba9aecbb11 100644 --- a/libavcodec/sheervideo.c +++ b/libavcodec/sheervideo.c @@ -23,6 +23,8 @@ #include <stdlib.h> #include <string.h> +#define CACHED_BITSTREAM_READER !ARCH_X86_32 + #include "libavutil/intreadwrite.h" #include "avcodec.h" #include "get_bits.h"

1 0

avfilter/f_graphmonitor: also show link EOF status in blue
by Paul B Mahol 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Aug 31 18:06:54 2020 +0200| [e417f84003d1dc62983024e902c9ebf306fdffe3] | committer: Paul B Mahol avfilter/f_graphmonitor: also show link EOF status in blue > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=e417f84003d1dc629… --- doc/filters.texi | 3 +++ libavfilter/f_graphmonitor.c | 9 +++++++++ 2 files changed, 12 insertions(+) diff --git a/doc/filters.texi b/doc/filters.texi index 5074538e9a..fb8757ff8f 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -11936,6 +11936,9 @@ Display video size or number of audio channels in case of audio used by filter l @item rate Display video frame rate or sample rate in case of audio used by filter link. + +@item eof +Display link output status. @end table @item rate, r diff --git a/libavfilter/f_graphmonitor.c b/libavfilter/f_graphmonitor.c index a9c4ba42f5..535e52b48b 100644 --- a/libavfilter/f_graphmonitor.c +++ b/libavfilter/f_graphmonitor.c @@ -47,6 +47,7 @@ typedef struct GraphMonitorContext { uint8_t yellow[4]; uint8_t red[4]; uint8_t green[4]; + uint8_t blue[4]; uint8_t bg[4]; } GraphMonitorContext; @@ -60,6 +61,7 @@ enum { MODE_FMT = 1 << 6, MODE_SIZE = 1 << 7, MODE_RATE = 1 << 8, + MODE_EOF = 1 << 9, }; #define OFFSET(x) offsetof(GraphMonitorContext, x) @@ -85,6 +87,7 @@ static const AVOption graphmonitor_options[] = { { "format", NULL, 0, AV_OPT_TYPE_CONST, {.i64=MODE_FMT}, 0, 0, VF, "flags" }, { "size", NULL, 0, AV_OPT_TYPE_CONST, {.i64=MODE_SIZE}, 0, 0, VF, "flags" }, { "rate", NULL, 0, AV_OPT_TYPE_CONST, {.i64=MODE_RATE}, 0, 0, VF, "flags" }, + { "eof", NULL, 0, AV_OPT_TYPE_CONST, {.i64=MODE_EOF}, 0, 0, VF, "flags" }, { "rate", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str = "25"}, 0, INT_MAX, VF }, { "r", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str = "25"}, 0, INT_MAX, VF }, { NULL } @@ -236,6 +239,11 @@ static void draw_items(AVFilterContext *ctx, AVFrame *out, drawtext(out, xpos, ypos, buffer, s->white); xpos += strlen(buffer) * 8; } + if (s->flags & MODE_EOF && ff_outlink_get_status(l)) { + snprintf(buffer, sizeof(buffer)-1, " | eof"); + drawtext(out, xpos, ypos, buffer, s->blue); + xpos += strlen(buffer) * 8; + } } static int create_frame(AVFilterContext *ctx, int64_t pts) @@ -352,6 +360,7 @@ static int config_output(AVFilterLink *outlink) s->yellow[0] = s->yellow[1] = 255; s->red[0] = 255; s->green[1] = 255; + s->blue[2] = 255; s->pts = AV_NOPTS_VALUE; s->next_pts = AV_NOPTS_VALUE; outlink->w = s->w;

1 0

avcodec/magicyuv: change bits used by 12bit tables
by Paul B Mahol 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Aug 31 19:53:30 2020 +0200| [57915df230b8241cb3636041c4113a2e1f8fa92f] | committer: Paul B Mahol avcodec/magicyuv: change bits used by 12bit tables Higher number slows decoder. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=57915df230b8241cb… --- libavcodec/magicyuv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavcodec/magicyuv.c b/libavcodec/magicyuv.c index af2f1687aa..a255f82d9c 100644 --- a/libavcodec/magicyuv.c +++ b/libavcodec/magicyuv.c @@ -152,7 +152,7 @@ static int huff_build12(VLC *vlc, uint8_t *len) } ff_free_vlc(vlc); - return ff_init_vlc_sparse(vlc, FFMIN(he[4095].len, 14), 4096, + return ff_init_vlc_sparse(vlc, FFMIN(he[4095].len, 12), 4096, bits, sizeof(*bits), sizeof(*bits), codes, sizeof(*codes), sizeof(*codes), syms, sizeof(*syms), sizeof(*syms), 0);

1 0

avcodec/magicyuv: use cached bitstream reader except for x32
by Paul B Mahol 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Aug 31 16:23:23 2020 +0200| [276d86a8b80378ac08f51dbd2698c4d234f0e5d6] | committer: Paul B Mahol avcodec/magicyuv: use cached bitstream reader except for x32 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=276d86a8b80378ac0… --- libavcodec/magicyuv.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/libavcodec/magicyuv.c b/libavcodec/magicyuv.c index dd947c3c2f..af2f1687aa 100644 --- a/libavcodec/magicyuv.c +++ b/libavcodec/magicyuv.c @@ -22,6 +22,8 @@ #include <stdlib.h> #include <string.h> +#define CACHED_BITSTREAM_READER !ARCH_X86_32 + #include "libavutil/pixdesc.h" #include "libavutil/qsort.h"

1 0

avcodec/magicyuv: invert symbols when building vlc
by Paul B Mahol 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Aug 31 16:19:55 2020 +0200| [445e5fb67ab1bd5b59e89efe7c14cda692a30e15] | committer: Paul B Mahol avcodec/magicyuv: invert symbols when building vlc Instead at every decoded symbol. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=445e5fb67ab1bd5b5… --- libavcodec/magicyuv.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/libavcodec/magicyuv.c b/libavcodec/magicyuv.c index 5d76274d54..dd947c3c2f 100644 --- a/libavcodec/magicyuv.c +++ b/libavcodec/magicyuv.c @@ -113,7 +113,7 @@ static int huff_build10(VLC *vlc, uint8_t *len) for (i = 1023; i >= 0; i--) { codes[i] = code >> (32 - he[i].len); bits[i] = he[i].len; - syms[i] = he[i].sym; + syms[i] = 1023 - he[i].sym; code += 0x80000000u >> (he[i].len - 1); } @@ -145,7 +145,7 @@ static int huff_build12(VLC *vlc, uint8_t *len) for (i = 4095; i >= 0; i--) { codes[i] = code >> (32 - he[i].len); bits[i] = he[i].len; - syms[i] = he[i].sym; + syms[i] = 4095 - he[i].sym; code += 0x80000000u >> (he[i].len - 1); } @@ -177,7 +177,7 @@ static int huff_build(VLC *vlc, uint8_t *len) for (i = 255; i >= 0; i--) { codes[i] = code >> (32 - he[i].len); bits[i] = he[i].len; - syms[i] = he[i].sym; + syms[i] = 255 - he[i].sym; code += 0x80000000u >> (he[i].len - 1); } @@ -259,7 +259,7 @@ static int magy_decode_slice10(AVCodecContext *avctx, void *tdata, if (pix < 0) return AVERROR_INVALIDDATA; - dst[x] = max - pix; + dst[x] = pix; } dst += stride; } @@ -389,7 +389,7 @@ static int magy_decode_slice(AVCodecContext *avctx, void *tdata, if (pix < 0) return AVERROR_INVALIDDATA; - dst[x] = 255 - pix; + dst[x] = pix; } dst += stride; }

1 0

avfilter/vsrc_gradients: add speed option
by Paul B Mahol 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Aug 31 11:04:38 2020 +0200| [eb9fa00bcc7138516009c6e060e07b510806fe61] | committer: Paul B Mahol avfilter/vsrc_gradients: add speed option > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=eb9fa00bcc7138516… --- doc/filters.texi | 3 +++ libavfilter/vsrc_gradients.c | 4 +++- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/doc/filters.texi b/doc/filters.texi index 40d493fcae..5074538e9a 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -22577,6 +22577,9 @@ for the accepted syntax. If not specified, or the expressed duration is negative, the video is supposed to be generated forever. + +@item speed +Set speed of gradients rotation. @end table diff --git a/libavfilter/vsrc_gradients.c b/libavfilter/vsrc_gradients.c index f62cedad8f..9de0544f5b 100644 --- a/libavfilter/vsrc_gradients.c +++ b/libavfilter/vsrc_gradients.c @@ -38,6 +38,7 @@ typedef struct GradientsContext { AVRational frame_rate; int64_t pts; int64_t duration; ///< duration expressed in microseconds + float speed; uint8_t color_rgba[8][4]; int nb_colors; @@ -75,6 +76,7 @@ static const AVOption gradients_options[] = { {"seed", "set the seed", OFFSET(seed), AV_OPT_TYPE_INT64, {.i64=-1}, -1, UINT32_MAX, FLAGS }, {"duration", "set video duration", OFFSET(duration), AV_OPT_TYPE_DURATION, {.i64=-1}, -1, INT64_MAX, FLAGS },\ {"d", "set video duration", OFFSET(duration), AV_OPT_TYPE_DURATION, {.i64=-1}, -1, INT64_MAX, FLAGS },\ + {"speed", "set gradients rotation speed", OFFSET(speed), AV_OPT_TYPE_FLOAT,{.dbl=0.01}, 0.00001, 1, FLAGS },\ {NULL}, }; @@ -250,7 +252,7 @@ static int gradients_request_frame(AVFilterLink *outlink) AVFilterContext *ctx = outlink->src; GradientsContext *s = ctx->priv; AVFrame *frame = ff_get_video_buffer(outlink, s->w, s->h); - float angle = fmodf(s->pts / 100.f, 2.f * M_PI); + float angle = fmodf(s->pts * s->speed, 2.f * M_PI); const float w2 = s->w / 2.f; const float h2 = s->h / 2.f;

1 0

avfilter/vsrc_gradients: do not use (l)lrint variants for double
by Paul B Mahol 31 Aug '20

31 Aug '20

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Aug 31 10:55:59 2020 +0200| [823eb009cb93b9717e3532e4325233bf85946a73] | committer: Paul B Mahol avfilter/vsrc_gradients: do not use (l)lrint variants for double > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=823eb009cb93b9717… --- libavfilter/vsrc_gradients.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/libavfilter/vsrc_gradients.c b/libavfilter/vsrc_gradients.c index 410c3cb69b..f62cedad8f 100644 --- a/libavfilter/vsrc_gradients.c +++ b/libavfilter/vsrc_gradients.c @@ -98,20 +98,20 @@ static uint32_t lerp_color(uint8_t c0[4], uint8_t c1[4], float x) { const float y = 1.f - x; - return (lrint(c0[0] * y + c1[0] * x)) << 0 | - (lrint(c0[1] * y + c1[1] * x)) << 8 | - (lrint(c0[2] * y + c1[2] * x)) << 16 | - (lrint(c0[3] * y + c1[3] * x)) << 24; + return (lrintf(c0[0] * y + c1[0] * x)) << 0 | + (lrintf(c0[1] * y + c1[1] * x)) << 8 | + (lrintf(c0[2] * y + c1[2] * x)) << 16 | + (lrintf(c0[3] * y + c1[3] * x)) << 24; } static uint64_t lerp_color16(uint8_t c0[4], uint8_t c1[4], float x) { const float y = 1.f - x; - return (llrint((c0[0] * y + c1[0] * x) * 256)) << 0 | - (llrint((c0[1] * y + c1[1] * x) * 256)) << 16 | - (llrint((c0[2] * y + c1[2] * x) * 256)) << 32 | - (llrint((c0[3] * y + c1[3] * x) * 256)) << 48; + return (llrintf((c0[0] * y + c1[0] * x) * 256)) << 0 | + (llrintf((c0[1] * y + c1[1] * x) * 256)) << 16 | + (llrintf((c0[2] * y + c1[2] * x) * 256)) << 32 | + (llrintf((c0[3] * y + c1[3] * x) * 256)) << 48; } static uint32_t lerp_colors(uint8_t arr[3][4], int nb_colors, float step)

1 0