From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Thanks for noticing, I've fixed the patch
(also some minor formatting issues I've noticed).
doc/muxers.texi | 20 +
libavformat/tee.c | 87 ++-
2 files changed, 106 i
From: Jan Sebechlebsky
This commit also adds new diff option for fate tests allowing do compare
multiple tuples of files.
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version:
- fixed out of tree build (previous version refered to SRC_PATH instead of
TARGET_PATH,
thanks to M
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- Refactoring based on Nicolas's comments
- Added TODO regarding boolean option parsing
doc/muxers.texi | 20 +
libavformat/tee.c | 87 ++
From: Jan Sebechlebsky
Hello,
I am re-sending the patch(es) adding support for fifo pseudo-muxer in tee muxer.
I will apply the first one in few days (since it already has been reviewed by
Nicolas,
and it differs only in issues Nicolas pointed out in review and these should be
fixed now).
The s
From: Jan Sebechlebsky
---
doc/muxers.texi | 9 +
libavformat/tee.c | 20
2 files changed, 29 insertions(+)
diff --git a/doc/muxers.texi b/doc/muxers.texi
index ced223e..139ced0 100644
--- a/doc/muxers.texi
+++ b/doc/muxers.texi
@@ -1603,6 +1603,11 @@ outputs and
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Chanages from the last version of patch:
-> dst->metadata dictionary is freed before entries are copied from
src->metadata
libavformat/internal.h | 9
libavformat/utils.c| 57
From: Jan Sebechlebsky
This ensures that AV_NOPTS_VALUE value is handled
correctly.
Signed-off-by: Jan Sebechlebsky
---
libavformat/tee.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/libavformat/tee.c b/libavformat/tee.c
index daddba5..b4158e1 100644
--- a/libavformat
From: Jan Sebechlebsky
mpeg4_unpack_bframes_bsf bitstream filters constructs
resulting packet using av_packet_from_data() function.
This function however modifies only buffer (data) and leaves
other fields untouched, so the content of other fields
of the output packet is undefined.
It is working
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
libavcodec/mpeg4_unpack_bframes_bsf.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/libavcodec/mpeg4_unpack_bframes_bsf.c
b/libavcodec/mpeg4_unpack_bframes_bsf.c
index aee8ccb..e227f58 100644
--- a/libavcodec
From: Jan Sebechlebsky
Set BSF EOF flag only if pkt == NULL or both data and
side data are not present in packet.
Signed-off-by: Jan Sebechlebsky
---
I believe that side data should be checked too, and
EOF flag set only when both data and side data are
not present.
I was testing new list
From: Jan Sebechlebsky
Specify av_bsf_packet() behaviour in case that the packet does
not contain any data more precisely in the comment.
Signed-off-by: Jan Sebechlebsky
---
libavcodec/avcodec.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libavcodec/avcodec.h b/li
From: Jan Sebechlebsky
Set BSF EOF flag only if pkt == NULL in av_bsf_send_packet().
Signed-off-by: Jan Sebechlebsky
---
I agree, it seems cleaner that way.
Thanks,
please apply this version of patch then and ignore
the patch changing the comment.
Regards,
Jan
libavcodec/bsf.c | 2 +-
From: Jan Sebechlebsky
Hello,
I am sending next version of fifo pseudo-muxer.
I will send patchset adding AVFMT_FLAG_NONBLOCK support requested
by Nicolas later - it turned out that there are some more things
to solve (av_write_frame does not support repeated calls in current state)
and test.
From: Jan Sebechlebsky
The fifo pseudo-muxer allows to separate encoder from the
actual output by using a first-in-first-out queue and
running actual muxer asynchronously in separate thread.
It can be configured to attempt transparent recovery
of output on failure.
Signed-off-by: Jan Sebechlebs
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 6d4c9f9..0e66170 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -164,6 +164,7 @@ Codecs:
exif.c, exif.hThilo Borgm
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
libavcodec/avcodec.h | 3 ++-
libavcodec/bsf.c | 3 +++
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index ca8dba8..36f7935 100644
--- a/libavcodec/avcodec.h
+++ b/libavco
From: Jan Sebechlebsky
Set BSF EOF flag only if pkt == NULL in av_bsf_send_packet().
Signed-off-by: Jan Sebechlebsky
---
libavcodec/bsf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/bsf.c b/libavcodec/bsf.c
index 88b7f29..9b9ada7 100644
--- a/libavcodec/bsf.c
From: Jan Sebechlebsky
---
libavcodec/avcodec.h | 74 ++
libavcodec/bsf.c | 284 +++
2 files changed, 358 insertions(+)
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index 36f7935..39106ee 100644
--- a/libavcodec/avcode
From: Jan Sebechlebsky
Use of this function can save unnecessary malloc operation
in bitstream filter.
Signed-off-by: Jan Sebechlebsky
---
libavcodec/bsf.c | 16
libavcodec/bsf.h | 11 +++
2 files changed, 27 insertions(+)
diff --git a/libavcodec/bsf.c b/libavcodec/bs
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
libavformat/tee.c | 123 +-
1 file changed, 58 insertions(+), 65 deletions(-)
diff --git a/libavformat/tee.c b/libavformat/tee.c
index b4158e1..1a055cd 100644
--- a/libavformat/tee.c
From: Jan Sebechlebsky
Add custom item name function for bsf list, which will
construct string description of filter chain. This is
done using lazy-initialization, so there is no overhead
if item name is never accessed.
Signed-off-by: Jan Sebechlebsky
Conflicts:
libavcodec/bsf.c
---
T
From: Jan Sebechlebsky
The actual implementation uses ':' divider, not '/' as
documented.
Signed-off-by: Jan Sebechlebsky
---
doc/bitstream_filters.texi | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/bitstream_filters.texi b/doc/bitstream_filters.texi
index 6c58d02..a8
From: Jan Sebechlebsky
compute_muxer_pkt_fields() stores the last seen timestamps in stream
and produces error if the same timestamp is presented again.
This is a problem if muxer works in non-blocking mode and calls
av_write_packet repeatedly with the same packet.
This patch saves stream fields
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
This adds two tests checking that the output of muxers is the same
as when fifo is used and stand-alone test program which covers
behaviour in failure scenarios.
libavformat/Makefile | 1 +
libavformat/tests/fifo_muxer.c
From: Jan Sebechlebsky
Add support for nonblocking calls.
Signed-off-by: Jan Sebechlebsky
---
libavformat/fifo.c | 70 +-
1 file changed, 59 insertions(+), 11 deletions(-)
diff --git a/libavformat/fifo.c b/libavformat/fifo.c
index bc8c973..e
From: Jan Sebechlebsky
Add comments regarding AVFMG_FLAG_NONBLOCK usage with muxers.
Signed-off-by: Jan Sebechlebsky
---
libavformat/avformat.h | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index 9173908..6898c8c 100
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
libavformat/avformat.h | 14 ++
libavformat/mux.c | 16
2 files changed, 30 insertions(+)
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index 9191c69..9173908 100644
--- a/libavformat/av
From: Jan Sebechlebsky
This makes av_write_trailer not to free the resources if write_trailer
call returns AVERROR(EAGAIN) allowing repeated calls of write_trailer of
non-blocking muxer.
Signed-off-by: Jan Sebechlebsky
---
libavformat/avformat.h | 6 +-
libavformat/mux.c | 3 +++
2 fi
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes from the last version of patch:
- I got rid of write header message, and pulled initial write_header
call out of the while loop as Nicolas originaly suggested.
Changelog| 1 +
configure
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 932e6fb..9fab34d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -164,6 +164,7 @@ Codecs:
exif.c, exif.hThilo Borgm
From: Jan Sebechlebsky
Move muxer deinitialization and private resources freeing
in a separate static function free_muxer(AVFormatContext*).
Signed-off-by: Jan Sebechlebsky
---
libavformat/mux.c | 31 ---
1 file changed, 20 insertions(+), 11 deletions(-)
diff --git
From: Jan Sebechlebsky
Restore original timestamps in write_packet() if the
actual write operation was not successfull. This allows
to pass the same packet to nonblocking muxer repeatedly
without corrupting the timestamps.
Signed-off-by: Jan Sebechlebsky
---
libavformat/mux.c | 9 +
1
From: Jan Sebechlebsky
Hello,
I am resending fifo muxer related patchset, now
also with patches adding support for nonblocking
calls (AVFMT_FLAG_NONBLOCK) and fate tests.
Regards,
Jan
Jan Sebechlebsky (11):
avformat: Add fifo pseudo-muxer
MAINTAINERS: Add myself as maintainer of fifo muxe
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
libavformat/tests/fifo_muxer.c | 139 +
tests/ref/fate/fifo-muxer-tst | 5 ++
2 files changed, 144 insertions(+)
diff --git a/libavformat/tests/fifo_muxer.c b/libavformat/tests/fifo_muxer.c
in
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes from the last version of patch:
- boolean AVOptions are now ints, this was the cause of fate test
segfault reported by Michael
Changelog| 1 +
configure| 1 +
doc/muxers.texi
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes from last version of patch:
- removed AVFMT_FLAG_NONBLOCK check and modified comment
so it states how function behaves with both blocking / non-blocking muxer
libavformat/avformat.h | 15 +++
libavformat/mux.c
From: Jan Sebechlebsky
Add comments regarding AVFMG_FLAG_NONBLOCK usage with muxers.
Signed-off-by: Jan Sebechlebsky
---
There are no changes in this patch, it is just rebased version
because of change in previous patch.
libavformat/avformat.h | 9 -
1 file changed, 8 insertions(+),
From: Jan Sebechlebsky
Add support for nonblocking calls.
Signed-off-by: Jan Sebechlebsky
---
Changes from the last version:
- boolean flags accessed from both threads are ints now
and are accessed with atomic operations.
- pthread_tryjoin_np is replaced by flag set before
fifo_consume
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes from last version:
- av_abort_output was renamed to avformat_write_abort
libavformat/tests/fifo_muxer.c | 139 +
tests/ref/fate/fifo-muxer-tst | 5 ++
2 files changed, 144 insertions
From: Jan Sebechlebsky
---
Changes from last version:
- fixed doxygen comments
- added av_bsf_list_append2() function
- changed names of array and length fields to follow naming convention
- idx and flushed_idx is now unsigned
- merged bsf_list_flush to bsf_list_filter to reduce code duplic
From: Jan Sebechlebsky
Add custom item name function for bsf list, which will
construct string description of filter chain. This is
done using lazy-initialization, so there is no overhead
if item name is never accessed.
Signed-off-by: Jan Sebechlebsky
---
No changes from last version, just reb
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
No changes from the last version, just rebased after Michael's commit.
libavformat/tee.c | 80 +++
1 file changed, 57 insertions(+), 23 deletions(-)
diff --git a/libavformat/tee.c b
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version done according Martons review:
- Documentation fixes as requested
- Rearanged message processing in while loop inside fifo_cosumer_thread,
added FIFO_WRITE_HEADER message type, message is initialized
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version:
- Removed empty lines at the end of fifo_muxer.c file
libavformat/Makefile | 1 +
libavformat/tests/fifo_muxer.c | 443 +
tests/Makefile
From: Jan Sebechlebsky
Add support for nonblocking calls.
Signed-off-by: Jan Sebechlebsky
---
No changes since the last version of the patch, just rebased
because of changes in the previous commit.
libavformat/fifo.c | 61 --
1 file change
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of patch:
- removed parse_bsfs() function I accidentaly left out
(it's replaced by av_bsf_list_parse_str())
libavformat/tee.c | 120 ++
1 file chang
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- Fixed thread include (old patch included pthread.h directly)
Changelog| 1 +
configure| 1 +
doc/muxers.texi | 93 +++
libavformat/Makef
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- Fixed whitespace and missing $(EXESUF) in fifo-muxer.mak
- Fixed "overflow with packet dropping" test which skipped write_trailer
call in case of failure.
libavformat/Makefile
From: Jan Sebechlebsky
This makes av_write_trailer not to free the resources if write_trailer
call returns AVERROR(EAGAIN) allowing repeated calls of write_trailer of
non-blocking muxer.
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- Added assert to the par
From: Jan Sebechlebsky
Move muxer deinitialization and private resources freeing
in a separate static function free_muxer(AVFormatContext*).
Signed-off-by: Jan Sebechlebsky
---
No changes since the last version, just rebased because
of changes in previous patch.
libavformat/mux.c | 31 +
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
No changes since the last version of the patch, just rebased
because of the changes in previous patches.
libavformat/avformat.h | 15 +++
libavformat/mux.c | 13 +
2 files changed, 28 insertions(+)
diff
From: Jan Sebechlebsky
Add comments regarding AVFMG_FLAG_NONBLOCK usage with muxers.
Add assert forbiding use of nonblocking muxer with
av_interleaved_write_frame.
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- added assert to the beginning of av_interleave
From: Jan Sebechlebsky
Add support for nonblocking calls.
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version:
- fixed wrong flag passed to av_thread_message_queue_recv()
- fixed memleak when queue is full in nonblocking mode
libavformat/fifo.c | 61 +
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
No changes since the last version of the patch, just rebased
because of changes in previous fate test patch.
libavformat/tests/fifo_muxer.c | 139 +
tests/ref/fate/fifo-muxer-tst | 5 ++
2
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of patch:
- Fixed make dependencies so the tests are not executed when
required components are disabled
libavformat/Makefile | 1 +
libavformat/tests/fifo_muxer.c | 443 +
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- fixed print_deinit_summary to use int instead of uint8_t
libavformat/Makefile | 1 +
libavformat/tests/fifo_muxer.c | 443 +
tests/Mak
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- Fixed documentation (apart from the Marton's suggestions I've
also changed example, since it used block_on_overflow option
from the earlier version of patch)
- Changed max_recovery_
From: Jan Sebechlebsky
Add support for nonblocking calls.
Signed-off-by: Jan Sebechlebsky
---
No changes since the last version of the patch, just rebased because
of changes in previous patch.
libavformat/fifo.c | 61 --
1 file changed, 55
From: Jan Sebechlebsky
Add custom item name function for bsf list, which will
construct string description of filter chain. This is
done using lazy-initialization, so there is no overhead
if item name is never accessed.
Signed-off-by: Jan Sebechlebsky
Conflicts:
libavcodec/bsf.c
---
C
From: Jan Sebechlebsky
---
Changes since the last version of the patch:
- removed av_freep(&bsf->item_name) from bsf_list_close since it belongs to
next commit
libavcodec/avcodec.h | 85
libavcodec/bsf.c | 278 +++
2 fil
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
doc/APIchanges | 7 +++
1 file changed, 7 insertions(+)
diff --git a/doc/APIchanges b/doc/APIchanges
index 209ab41..4868d58 100644
--- a/doc/APIchanges
+++ b/doc/APIchanges
@@ -15,6 +15,13 @@ libavutil: 2015-08-28
API change
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- fixed "s@item" in muxers.texi
- fixed second -> seconds in FIFO_DEFAULT_RECOVERY_WAIT_TIME_USEC comment
- removed AVFormat *oformat from FifoContext (it is local variable now)
- if reco
From: Jan Sebechlebsky
Add support for nonblocking calls.
Signed-off-by: Jan Sebechlebsky
---
No chanes since the last version of patch, rebased because of changes in the
patch adding fifo.
libavformat/fifo.c | 62 --
1 file changed, 56 in
From: Jan Sebechlebsky
Muxing might be running in a separate thread if actual muxer is
run inside of fifo pseudo-muxer. Callback should be therefore
thread-safe.
Signed-off-by: Jan Sebechlebsky
---
libavformat/avformat.h | 2 +-
libavformat/avio.h | 2 ++
2 files changed, 3 insertions(+),
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- Added note in fifo muxer documentation in muxers.texi
regarding thread-safety requirement for callbacks in AVFormatContext
Changelog| 1 +
configure|
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
I've noticed that conflicting patch was applied meanwhile, so I'm resending
this.
Please apply :)
doc/APIchanges | 7 +++
1 file changed, 7 insertions(+)
diff --git a/doc/APIchanges b/doc/APIchanges
index 74145b2..582563a 10064
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Conflicting patch was applied meanwhile, so I am resending this one.
No changes since the last version.
Changelog| 1 +
configure| 1 +
doc/muxers.texi | 95 +++
libavformat/Makefile
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- fixed mistakes in docs (missing "is", 2xmisspelled "unsuccessful")
- removed braces around return statement in fifo_mux_init()
- fixed "return 0;" -> "return ret;" in fifo_write_header()
From: Jan Sebechlebsky
Add support for nonblocking calls.
Signed-off-by: Jan Sebechlebsky
---
No changes since the last version of the patch, just rebased because of
changes in previous fifo patch.
libavformat/fifo.c | 62 --
1 file change
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
I believe I have fixed handling input / output timebase and input parameters
to bitstream filters list.
libavformat/tee.c | 131 ++
1 file changed, 64 insertions(+), 67 deletions(-)
From: Jan Sebechlebsky
AV_OPT_SEARCH_CHILDREN flag must be passed to av_opt_set_dict()
to set options for private context.
Signed-off-by: Jan Sebechlebsky
---
libavcodec/bsf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/bsf.c b/libavcodec/bsf.c
index 2462e62.
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
Changes since the last version of the patch:
- added check for avcodec_parameters_copy return value
- fixed stray space
- rewritten cycle receiving packets from bsf so case when
av_interleaved_write_frame
returns EAGAIN is trea
From: Jan Sebechlebsky
This commit also adds new diff option for fate tests allowing do compare
multiple tuples of files.
Signed-off-by: Jan Sebechlebsky
---
tests/Makefile| 1 +
tests/fate-run.sh | 7
tests/fate/tee-muxer.mak | 2
From: Jan Sebechlebsky
Copy interrupt callback to slave format context to allow
user to interrupt IO. Copy format flags as well.
Signed-off-by: Jan Sebechlebsky
---
libavformat/tee.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libavformat/tee.c b/libavformat/tee.c
index 518af4a..d59a
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
This commit makes use of fifo muxer together with tee muxer
easier, fifo muxer does not have to be explicitly specified
for each slave. For the most simple use case it is sufficient
to turn fifo muxer on for all slaves by switching on us
From: Jan Sebechlebsky
Closing single slave operation is pulled out into separate
function close_slave(TeeSlave*).
Both close_slave and close_slaves function are moved before
open_slave function.
Signed-off-by: Jan Sebechlebsky
---
libavformat/tee.c | 58 ++-
From: Jan Sebechlebsky
Calling close_slave in case error is to be returned from open_slave
will free allocated resources.
Since failure can happen before bsfs array is initialized,
close_slave must check that bsfs is not NULL before accessing
tee_slave->bsfs[i] element.
Signed-off-by: Jan Sebec
From: Jan Sebechlebsky
Closing single slave operation is pulled out into separate
function close_slave(TeeSlave*).
Both close_slave and close_slaves function are moved before
open_slave function.
Signed-off-by: Jan Sebechlebsky
---
I've missed bad spacing in close_slave argument in previous ver
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
doc/muxers.texi | 14
libavformat/tee.c | 96 +--
2 files ch
From: Jan Sebechlebsky
Calling close_slave in case error is to be returned from open_slave
will free allocated resources.
Since failure can happen before bsfs array is initialized,
close_slave must check that bsfs is not NULL before accessing
tee_slave->bsfs[i] element.
Signed-off-by: Jan Sebec
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
I've just added topic to commit message title as Marton Balint suggested.
doc/muxers.texi | 14
libavformat/te
From: Jan Sebechlebsky
Closing single slave operation is pulled out into separate
function close_slave(TeeSlave*).
Both close_slave and close_slaves function are moved before
open_slave function.
Signed-off-by: Jan Sebechlebsky
---
I've fixed missing spaces and added topic to commit message ti
From: Jan Sebechlebsky
Closing single slave operation is pulled out into separate
function close_slave(TeeSlave*).
Both close_slave and close_slaves function are moved before
open_slave function.
Signed-off-by: Jan Sebechlebsky
---
libavformat/tee.c | 58 ++-
From: Jan Sebechlebsky
Calling close_slave in case error is to be returned from open_slave
will free allocated resources.
Since failure can happen before bsfs array is initialized,
close_slave must check that bsfs is not NULL before accessing
tee_slave->bsfs[i] element.
Slave muxer expects writ
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
doc/muxers.texi | 14
libavformat/tee.c | 102 +-
2 files c
Hello,
I'm sorry for delay. I made a few modifications to the patchset from the
last version apart from issues pointed by Marton:
- I discovered another source of leaks - slave muxers often allocate memory
in write_header call and free them in write_trailer. write_trailer is called
if write_heade
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
Thanks for pointing on the problem with testing enum for negative value.
I've fixed that, so parse_slave_failure_policy_o
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
Sorry I've missed those. It is fixed in this patch.
Regards,
Jan S.
doc/muxers.texi | 14 +
libavformat/tee.
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
This is embarassing, sorry for that! You're right in both, I've tried
to merge my old file with the changes in previous p
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
doc/muxers.texi | 14
libavformat/tee.c | 95 ---
2 files ch
Understood :)
I've modified the patch so the error is printed in that case.
I'm sending both yet unapplied patches (So you don't have to
search in mailing list for the most recent version of second one).
Can you apply them, if it's OK now?
I guess I can then create patch to replace calls to depr
From: Jan Sebechlebsky
In open_slave failure can happen before bsfs array is initialized,
close_slave must check that bsfs is not NULL before accessing
tee_slave->bsfs[i] element.
Slave muxer expects write_trailer to be called if it's
write_header suceeded (so resources allocated in write_header
From: Jan Sebechlebsky
Adds per slave option 'onfail' to the tee muxer allowing an output to
fail,so other slave outputs can continue.
Signed-off-by: Jan Sebechlebsky
---
Changes from last version:
-> Use av_err2str in tee_process_slave_failure instead of combination
of av_strerror
From: Jan Sebechlebsky
TeeSlave.bsfs is array of pointers to AVBitStreamFilterContext,
so element size should be really size of a pointer, not size
of TeeSlave structure.
Signed-off-by: Jan Sebechlebsky
---
libavformat/tee.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/l
From: Jan Sebechlebsky
Replace av_copy_packet and deprecated av_dup_packet by
creating reference using av_packet_ref.
Signed-off-by: Jan Sebechlebsky
---
This should be effectively the same as calling av_packet_clone,
but without dynamic memory allocation (reuses local AVPacket pkt).
libav
From: Jan Sebechlebsky
This fixes ticket #5487 - mjpeg2jpeg bitstream filter causes
segmentation fault with header-less mjpeg.
Signed-off-by: Jan Sebechlebsky
---
libavcodec/mjpeg2jpeg_bsf.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/libavcodec/mjpeg2jpeg_bsf.c b/li
From: Jan Sebechlebsky
This fixes ticket #5487 - mjpeg2jpeg bitstream filter causes
segmentation fault with header-less mjpeg.
Signed-off-by: Jan Sebechlebsky
---
libavcodec/mjpeg2jpeg_bsf.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libavcodec/mjpeg2jpeg_bsf.c b/libavcodec/mjpeg2jp
From: Jan Sebechlebsky
TeeSlave.bsfs is array of pointers to AVBitStreamFilterContext,
so element size should be really size of a pointer, not size
of TeeSlave structure.
Signed-off-by: Jan Sebechlebsky
---
I've rewritten sizeof as suggested :)
libavformat/tee.c | 2 +-
1 file changed, 1 ins
From: Jan Sebechlebsky
Signed-off-by: Jan Sebechlebsky
---
libavformat/tee.c | 171 --
1 file changed, 139 insertions(+), 32 deletions(-)
diff --git a/libavformat/tee.c b/libavformat/tee.c
index 806beaa..ff0918b 100644
--- a/libavformat/tee.c
From: Jan Sebechlebsky
Fifo was reallocating always to twice of the requested size.
This fixes it to reallocate to requested size, or twice of the
original size - whichever is greater.
Signed-off-by: Jan Sebechlebsky
---
I believe the intended behaviour was as described in commit message
and
1 - 100 of 126 matches
Mail list logo
