On 22/02/16 18:20, Vittorio Giovara wrote:
> On Sun, Feb 21, 2016 at 5:51 AM, Luca Barbato wrote:
>> On 21/02/16 02:49, Vittorio Giovara wrote:
>>> On Sat, Feb 20, 2016 at 8:22 PM, Luca Barbato wrote:
On 20/02/16 23:40, Diego Biurrun wrote:
> Are
On Mon, Feb 22, 2016 at 08:17:25PM -0500, Vittorio Giovara wrote:
> --- a/tests/fate/microsoft.mak
> +++ b/tests/fate/microsoft.mak
> @@ -38,6 +38,9 @@ fate-wmv8-drm-nodec: CMD = framecrc -cryptokey
> 137381538c84c068111902a59c5cf6c34
> FATE_SAMPLES_AVCONV-$(call DEMDEC, ASF, WMV3) +=
On Mon, Feb 22, 2016 at 08:09:32PM -0500, Vittorio Giovara wrote:
> --- a/tests/fate/indeo.mak
> +++ b/tests/fate/indeo.mak
> @@ -1,6 +1,9 @@
> FATE_INDEO-$(call DEMDEC, AVI, INDEO2) += fate-indeo2
> fate-indeo2: CMD = framecrc -i $(TARGET_SAMPLES)/rt21/VPAR0026.AVI
>
> +FATE_INDEO-$(call
---
libavcodec/intrax8.c | 21 +++--
libavcodec/intrax8.h | 7 +--
libavcodec/vc1_block.c | 4 +++-
libavcodec/wmv2dec.c | 3 ++-
4 files changed, 21 insertions(+), 14 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index d9c156c..d0e7e93 100644
Use it instead of the embedded mpegvideo one. Update init function
signature to load it directly from the callers.
---
configure| 2 +-
libavcodec/intrax8.c | 14 --
libavcodec/intrax8.h | 6 +-
libavcodec/vc1dec.c | 2 +-
libavcodec/wmv2dec.c | 2 +-
5 files
---
This one bugs me since intrax8 decoding process requires no error resilience
whatsoever, and it's only there to make things work with mpegvideo.
I'd really really like to remove error resilience altogether, but that
would require adding two er function calls to the dependent decoders.
Would
---
libavcodec/intrax8.c | 17 ++---
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 3cd84dc..509a8d2 100644
--- a/libavcodec/intrax8.c
+++ b/libavcodec/intrax8.c
@@ -62,7 +62,7 @@ static av_cold void x8_vlc_init(void){
---
libavcodec/intrax8.c | 20
libavcodec/intrax8.h | 24
2 files changed, 24 insertions(+), 20 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 8b6de0d..9e9479b 100644
--- a/libavcodec/intrax8.c
+++ b/libavcodec/intrax8.c
---
libavcodec/intrax8.c | 7 +--
libavcodec/intrax8.h | 10 +-
libavcodec/vc1dec.c | 3 +--
libavcodec/wmv2dec.c | 2 +-
4 files changed, 4 insertions(+), 18 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 9c82ef9..c289dd4 100644
---
intrax8 needs tables from msmpeg4, which are normally supplied by either
VC1 or WMV2.
---
libavcodec/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 256dee3..ea47273 100644
--- a/libavcodec/Makefile
+++
This is assuming that intrax8 has no support for interlacing
(hence PICT_FRAME is set and last_frame is NULL).
---
libavcodec/intrax8.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index b3b1331..44cf141 100644
---
It will be needed in later commits.
---
libavcodec/intrax8.c | 4 +++-
libavcodec/intrax8.h | 5 -
libavcodec/vc1dec.c | 2 +-
libavcodec/wmv2dec.c | 2 +-
4 files changed, 9 insertions(+), 4 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 7675d06..b03da68 100644
---
libavcodec/intrax8.c | 3 ++-
libavcodec/intrax8.h | 3 ++-
libavcodec/vc1_block.c | 2 +-
libavcodec/wmv2dec.c | 2 +-
4 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 8ac3417..b3b1331 100644
--- a/libavcodec/intrax8.c
+++
---
libavcodec/intrax8.c | 65 ++--
libavcodec/intrax8.h | 7 ++
libavcodec/vc1dec.c | 4 +++-
libavcodec/wmv2dec.c | 3 ++-
4 files changed, 44 insertions(+), 35 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index
---
libavcodec/vc1dec.c | 6 +++---
libavcodec/wmv2dec.c | 4 +---
2 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/libavcodec/vc1dec.c b/libavcodec/vc1dec.c
index d65c68a..e60a084 100644
--- a/libavcodec/vc1dec.c
+++ b/libavcodec/vc1dec.c
@@ -314,7 +314,7 @@ static void
These are not changed in the function, nor in the dependent decoders.
---
libavcodec/intrax8.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index ffb40b6..2613bf7 100644
--- a/libavcodec/intrax8.c
+++ b/libavcodec/intrax8.c
@@
---
libavcodec/intrax8.c | 35 ---
libavcodec/intrax8.h | 4 +++-
libavcodec/vc1_block.c | 2 +-
libavcodec/wmv2dec.c | 2 +-
4 files changed, 21 insertions(+), 22 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index
Sustain possible size changes via av_fast_malloc().
---
libavcodec/intrax8.c | 11 ---
libavcodec/intrax8.h | 2 ++
2 files changed, 10 insertions(+), 3 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index d0e7e93..4955a2b 100644
--- a/libavcodec/intrax8.c
+++
This allows to drop the assert since it would be redundant.
---
libavcodec/intrax8.c | 12 +---
libavcodec/intrax8.h | 3 ++-
2 files changed, 11 insertions(+), 4 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 3699f3b..b432f96 100644
---
---
libavcodec/intrax8.c | 23 +++
1 file changed, 11 insertions(+), 12 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 64c92fe..7af5f65 100644
--- a/libavcodec/intrax8.c
+++ b/libavcodec/intrax8.c
@@ -719,22 +719,21 @@ block_placed:
}
// FIXME
---
The test sample is already on FATE strangely enough.
Vittorio
tests/fate/microsoft.mak| 3 +
tests/ref/fate/wmv8-intrax8 | 475
2 files changed, 478 insertions(+)
create mode 100644 tests/ref/fate/wmv8-intrax8
diff --git
This is already performed in init_context_frame().
---
libavcodec/intrax8.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index b306aa3..9c82ef9 100644
--- a/libavcodec/intrax8.c
+++ b/libavcodec/intrax8.c
@@ -828,8 +828,6 @@ int
---
NOTE: both these elements are not actually used by the decoders:
vc1 modifies them only for intra (non-j) or interframes, wmv2 does
something similar. I'm not extremely confident in dropping the references,
at least not right now, however these two buffers could and should be
made local.
This buffers are just a way to store frame pointers and be able to
modify them without touching the original ones.
The two dependent decoders (WMV2 and VC1) do not need special care for
these fields: the former does not seem to use the dest buffers, while
the latter reinits them every time to the
---
libavcodec/intrax8.c | 5 +++--
libavcodec/intrax8.h | 7 +--
libavcodec/vc1_block.c | 2 +-
libavcodec/wmv2dec.c | 3 ++-
4 files changed, 11 insertions(+), 6 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 5aa9c36..d9c156c 100644
---
---
libavcodec/intrax8.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/libavcodec/intrax8.c b/libavcodec/intrax8.c
index 9e9479b..3699f3b 100644
--- a/libavcodec/intrax8.c
+++ b/libavcodec/intrax8.c
@@ -46,7 +46,7 @@ static VLC j_ac_vlc[2][2][8]; // [quant < 13],
---
configure| 2 +-
libavcodec/intrax8.c | 3 ++-
libavcodec/intrax8.h | 2 ++
3 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/configure b/configure
index d11b128..51141db 100755
--- a/configure
+++ b/configure
@@ -1860,7 +1860,7 @@ error_resilience_select="me_cmp"
---
tests/fate/indeo.mak| 3 ++
tests/ref/fate/indeo2-delta | 115
2 files changed, 118 insertions(+)
create mode 100644 tests/ref/fate/indeo2-delta
diff --git a/tests/fate/indeo.mak b/tests/fate/indeo.mak
index e725a6b..1a47d79 100644
---
From: Luca Barbato
Signed-off-by: Vittorio Giovara
---
libavcodec/indeo2data.h | 208
1 file changed, 104 insertions(+), 104 deletions(-)
diff --git a/libavcodec/indeo2data.h
From: Luca Barbato
Rename luma table to delta table and change how it is used.
CC: libav-sta...@libav.org
Signed-off-by: Vittorio Giovara
---
libavcodec/indeo2.c | 15 +++--
libavcodec/indeo2data.h | 170
On 22/02/16 13:17, Diego Biurrun wrote:
> On Mon, Feb 22, 2016 at 11:13:47AM +0100, Luca Barbato wrote:
>> And use the delta table as it should be used.
>> Fix some banding artefacts.
>
> FixES
>
>> --- a/libavcodec/indeo2data.h
>> +++ b/libavcodec/indeo2data.h
>> @@ -27,56 +27,56 @@
>> static
Quoting Mark Thompson (2016-02-20 19:32:30)
> ---
> configure| 3 +
> libavfilter/Makefile | 1 +
> libavfilter/allfilters.c | 1 +
> libavfilter/vf_vaapi_scale.c | 575
> +++
> 4 files changed, 580 insertions(+)
>
On Mon, Feb 22, 2016 at 5:13 AM, Luca Barbato wrote:
> And use the delta table as it should be used.
> Fix some banding artefacts.
Imho the most important thing is fixing the banding in the video not
updating the tables
May I suggest
"indeo2: Fix banding artefacts
Update
On Sun, Feb 21, 2016 at 5:51 AM, Luca Barbato wrote:
> On 21/02/16 02:49, Vittorio Giovara wrote:
>> On Sat, Feb 20, 2016 at 8:22 PM, Luca Barbato wrote:
>>> On 20/02/16 23:40, Diego Biurrun wrote:
Are you on Git HEAD and did you distclean? The wmv2
On Sun, Feb 21, 2016 at 6:53 AM, Anton Khirnov wrote:
> ---
> doc/examples/transcode_aac.c | 66
>
> 1 file changed, 49 insertions(+), 17 deletions(-)
probably ok
--
Vittorio
___
On Sun, Feb 21, 2016 at 6:53 AM, Anton Khirnov wrote:
> AVStream.codec is now deprecated. Allocate a separate codec context
> instead.
> ---
> avplay.c | 53 +++--
> 1 file changed, 35 insertions(+), 18 deletions(-)
probably ok
On Sun, Feb 21, 2016 at 6:53 AM, Anton Khirnov wrote:
> ---
> doc/examples/output.c | 60
> ---
> 1 file changed, 43 insertions(+), 17 deletions(-)
ok i think
--
Vittorio
___
On Mon, Feb 22, 2016 at 8:03 AM, Diego Biurrun wrote:
> On Sat, Aug 23, 2014 at 10:59:53AM -0400, Reinhard Tartler wrote:
>> This option facilitates testing shared libarary builds: for instance
>> fate builders do no longer need to set LD_LIBRARY_PATH as the binaries will
>> get
On Mon, Feb 22, 2016 at 5:23 AM, Anton Khirnov wrote:
> Quoting Diego Biurrun (2016-02-21 15:07:24)
>> On Sun, Feb 21, 2016 at 12:53:25PM +0100, Anton Khirnov wrote:
>> > --- a/avprobe.c
>> > +++ b/avprobe.c
>> > @@ -619,31 +619,25 @@ static void show_stream(AVFormatContext
On Sat, Aug 23, 2014 at 10:59:53AM -0400, Reinhard Tartler wrote:
> This option facilitates testing shared libarary builds: for instance
> fate builders do no longer need to set LD_LIBRARY_PATH as the binaries will
> get the right search paths hardcoded into their executable file.
>
> This option
On Mon, Feb 22, 2016 at 11:13:47AM +0100, Luca Barbato wrote:
> --- a/libavcodec/indeo2data.h
> +++ b/libavcodec/indeo2data.h
> @@ -95,47 +95,151 @@ static const uint16_t ir2_codes[IR2_CODES][2] = {
> +static const uint8_t ir2_delta_table[4][256] = {
> +{
> +0x80, 0x80, 0x84, 0x84,
On Mon, Feb 22, 2016 at 11:13:47AM +0100, Luca Barbato wrote:
> And use the delta table as it should be used.
> Fix some banding artefacts.
FixES
> --- a/libavcodec/indeo2data.h
> +++ b/libavcodec/indeo2data.h
> @@ -27,56 +27,56 @@
> static const uint16_t ir2_codes[IR2_CODES][2] = {
> #ifdef
On Mon, Feb 22, 2016 at 11:13:46AM +0100, Luca Barbato wrote:
> ---
> libavcodec/indeo2data.h | 208
>
> 1 file changed, 104 insertions(+), 104 deletions(-)
LGTM
Diego
___
libav-devel mailing list
On 22/02/16 11:23, Anton Khirnov wrote:
> The demuxers do not export those fields, they are codec properties. If
> someone really cares about avprobe exporting them, it should open a
> codec and do some decoding.
>
Improving avprobe can happen later IMHO.
lu
On 22/02/16 11:12, Anton Khirnov wrote:
> From: Maxym Dmytrychenko
>
> The QSV runtime expects the sync point address passed to
> MFXVideoENCODE_EncodeFrameAsync() to be valid until
> MFXVideoCORE_SyncOperation().
>
> Signed-off-by: Anton Khirnov
Quoting Diego Biurrun (2016-02-21 14:59:23)
> On Sun, Feb 21, 2016 at 12:53:18PM +0100, Anton Khirnov wrote:
> > --- a/libavdevice/xcbgrab.c
> > +++ b/libavdevice/xcbgrab.c
> > @@ -544,13 +544,14 @@ static int create_stream(AVFormatContext *s)
> >
> > -st->codec->codec_type =
Quoting Diego Biurrun (2016-02-21 15:07:24)
> On Sun, Feb 21, 2016 at 12:53:25PM +0100, Anton Khirnov wrote:
> > --- a/avprobe.c
> > +++ b/avprobe.c
> > @@ -619,31 +619,25 @@ static void show_stream(AVFormatContext *fmt_ctx, int
> > stream_idx)
> >
> > -probe_str("codec_type",
Quoting Luca Barbato (2016-02-19 14:02:21)
> Reported-By: Maxim Koitsov
> ---
>
> Thanks to Hendrik for the suggestion.
>
> libavformat/matroska.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/libavformat/matroska.c b/libavformat/matroska.c
>
And use the delta table as it should be used.
Fix some banding artefacts.
CC: libav-sta...@libav.org
---
libavcodec/indeo2.c | 15 +--
libavcodec/indeo2data.h | 278 +---
2 files changed, 200 insertions(+), 93 deletions(-)
diff --git
---
libavcodec/indeo2data.h | 208
1 file changed, 104 insertions(+), 104 deletions(-)
diff --git a/libavcodec/indeo2data.h b/libavcodec/indeo2data.h
index ed8d83c..077e1df 100644
--- a/libavcodec/indeo2data.h
+++ b/libavcodec/indeo2data.h
@@
From: Maxym Dmytrychenko
The QSV runtime expects the sync point address passed to
MFXVideoENCODE_EncodeFrameAsync() to be valid until
MFXVideoCORE_SyncOperation().
Signed-off-by: Anton Khirnov
---
libavcodec/qsvenc.c | 25
On 19/02/16 14:02, Luca Barbato wrote:
> Reported-By: Maxim Koitsov
> ---
>
> Thanks to Hendrik for the suggestion.
>
Ok on irc by j45
___
libav-devel mailing list
libav-devel@libav.org
On 21/02/16 12:53, Anton Khirnov wrote:
> The changes are:
> - different duration in flv, due to lavf no longer randomly guessing the
> duration of the last frame
> - fixed timestamps for some obscure demuxers
> - in the movenc test, zero bitrate is now written in esds, instead of
> the lavc
53 matches
Mail list logo