Tim-Philipp Müller [Wed, 18 Nov 2015 11:46:45 +0000 (11:46 +0000)]
queue2: don't print criticals when receiving custom events in ring buffer mode
Downgrade from g_warning to GST_WARNING log message.
https://bugzilla.gnome.org/show_bug.cgi?id=758276
Sebastian Dröge [Wed, 18 Nov 2015 07:21:23 +0000 (09:21 +0200)]
ptp-helper: Disable multicast loopback
We're not really interested in our own packets and ignore them anyway.
Vineeth TM [Tue, 17 Nov 2015 00:17:58 +0000 (09:17 +0900)]
debugutils: Fix string memory leak
https://bugzilla.gnome.org/show_bug.cgi?id=758207
Thiago Santos [Mon, 16 Nov 2015 11:22:14 +0000 (08:22 -0300)]
baseparse: simplify code a bit
Avoid repeated checks for testing if a buffer is a header
Thiago Santos [Fri, 13 Nov 2015 23:44:57 +0000 (20:44 -0300)]
collectpads: handle buffer with dts-only when mapping to running time
Otherwise the buffer was left with the original values and later would
be compared with other buffers that were converted to runninn time,
leading to bad interleaving of multiple streams.
https://bugzilla.gnome.org/show_bug.cgi?id=757961
Thiago Santos [Fri, 13 Nov 2015 19:31:06 +0000 (16:31 -0300)]
baseparse: do not overwrite header buffer timestamps
baseparse tries to preserve timestamps from upstream if
it is running on a time segment and write that to
output buffers. It assumes the first DTS is going to be
segment.start and sets that to the first buffers. In case
the buffer is a header buffer, it had no timestamps and
will have only the DTS set due to this mechanism.
This patch prevents this by skipping this behavior for
header buffers.
https://bugzilla.gnome.org/show_bug.cgi?id=757961
Aleix Conchillo Flaqué [Thu, 18 Jun 2015 20:56:53 +0000 (13:56 -0700)]
identity: add drop-buffer-flags property
New property drop-buffer-flags that will discard buffers that have the
given flags set.
https://bugzilla.gnome.org/show_bug.cgi?id=751182
Ross Burton [Thu, 12 Nov 2015 19:46:44 +0000 (19:46 +0000)]
helpers: fix install race
The install hook needs to be a install-data-hook not an install-exec-hook as the
helpers are installed into helperdir which is considered data (only path
variables with "exec" in are considered executables).
The explicit dependency on install-helpersPROGRAMS was an attempt at solving
this, but this causes occasional races where install-helpersPROGRAMS can run
twice in parallel (once via install-all, once via the hook's dependency).
https://bugzilla.gnome.org/show_bug.cgi?id=758029
Heinrich Fink [Thu, 12 Nov 2015 15:14:18 +0000 (16:14 +0100)]
systemclock: Add test for gst_clock_get_resolution
In a series of time measurements, the diff between now and previous
timestamps is either 0 or at least as long as get_resolution returned.
https://bugzilla.gnome.org/show_bug.cgi?id=758012
Heinrich Fink [Thu, 12 Nov 2015 10:26:56 +0000 (11:26 +0100)]
systemclock: Use mach_time on Apple platforms
On iOS/OSX g_get_current_time was used by default. However, mach_time is
the preferred high-resolution monotonic clock to be used on Apple
platforms.
https://bugzilla.gnome.org/show_bug.cgi?id=758012
Thiago Santos [Wed, 11 Nov 2015 14:11:23 +0000 (11:11 -0300)]
tracers: log: add missing hooks
Log all possible hooks
Thiago Santos [Thu, 29 Oct 2015 11:40:32 +0000 (08:40 -0300)]
tracer: add element-change-state-pre/post hook
Helps catching when a state change is starting and ending.
It is also possible to track the end of state changes by checking the
async-done or state-change messages.
This is particularly important for elements that do async state changes.
Miguel París Díaz [Wed, 11 Nov 2015 13:32:44 +0000 (14:32 +0100)]
pad: test for checking the order of the probe calls
https://bugzilla.gnome.org/show_bug.cgi?id=757197
Thiago Santos [Tue, 10 Nov 2015 17:41:35 +0000 (14:41 -0300)]
context: fix some copy and paste leftover in docs
Thiago Santos [Tue, 10 Nov 2015 17:37:27 +0000 (14:37 -0300)]
context: add some more documentation
Add a short paragraph on what means for a context to be persistent
Tim-Philipp Müller [Mon, 9 Nov 2015 18:02:09 +0000 (18:02 +0000)]
utils: use 'static inline' instead of 'inline static' for gtk-doc
gtk-doc doesn't seem to recognise the former variant.
Tim-Philipp Müller [Mon, 9 Nov 2015 17:59:16 +0000 (17:59 +0000)]
docs: remove dummy function declarations with G_INLINE_FUNCTION for gtk-doc
gtk-doc can handle static inline functions just fine these days,
there's no need for this stuff any more.
Duncan Palmer [Mon, 23 Feb 2015 03:16:19 +0000 (13:16 +1000)]
queue2: Add the avg-in-rate property.
https://bugzilla.gnome.org/show_bug.cgi?id=733959
Philippe Normand [Wed, 4 Nov 2015 11:02:51 +0000 (12:02 +0100)]
queue2: add overrun signal
Notifies that the queue2 is full, same as queue does
https://bugzilla.gnome.org/show_bug.cgi?id=733959
Thiago Santos [Wed, 30 Sep 2015 00:14:19 +0000 (21:14 -0300)]
pad: tests for accept-caps handling with proxy pads
Adds 3 new tests for testing accept-caps behavior with
proxy-caps pads.
1) A scenario where there is no proxy. The caps should be compared to the
template caps of the pad
2) A scenario where there is a compatible pad. The caps should be compared
to the proxied pad caps (and also with the template)
3) A scenario where there is an incompatible proxy pad. No caps should be
possible at all.
https://bugzilla.gnome.org/show_bug.cgi?id=754112
Thiago Santos [Thu, 5 Nov 2015 19:15:22 +0000 (16:15 -0300)]
pad: check template caps for proxy pads in accept-caps
Validate that the proxy pad indeed accepts the caps by also
comparing with the pad template caps, otherwise when the pad
had no internally linked pads it would always return true.
https://bugzilla.gnome.org/show_bug.cgi?id=754112
Thibault Saunier [Fri, 6 Nov 2015 11:58:42 +0000 (12:58 +0100)]
Fix build with -Werror=maybe-uninitialized
Thibault Saunier [Fri, 6 Nov 2015 11:05:18 +0000 (12:05 +0100)]
preset: Add a GST_PRESET_PATH env variable for presets to be usable uninstalled
And start setting the various uninstalled presets paths.
Anton Bondarenko [Thu, 5 Nov 2015 07:56:43 +0000 (08:56 +0100)]
filesink: continue element cleanup even if fclose fails
Sometimes filesink cleanup during stop may fail due to fclose error.
In this case object left partial cleanup with no file opened
but still holding old file descriptor.
It's not possible to change location property in a such state,
so next start will cause old file overwrite if 'append' does not set.
According to man page and POSIX standard about fclose behavior(extract):
------------------------------------------------------------------------
The fclose() function shall cause the stream pointed to by stream
to be flushed and the associated file to be closed.
...
Whether or not the call succeeds, the stream shall be disassociated
from the file and any buffer set by the setbuf() or setvbuf()
function shall be disassociated from the stream.
...
The fclose() function shall perform the equivalent of a close()
on the file descriptor that is associated with the stream
pointed to by stream.
After the call to fclose(), any use of stream results
in undefined behavior.
------------------------------------------------------------------------
So file is in 'closed' state no matter if fclose succeed or not.
And cleanup could be continued.
https://bugzilla.gnome.org/show_bug.cgi?id=757596
Jan Schmidt [Sat, 31 Oct 2015 13:04:27 +0000 (00:04 +1100)]
pad: Mark sticky events as sent on not-linked
Instead of re-sending sticky events over and over to a not-linked
pad, mark them as sent the first time. If the not-linked came from
downstream, it already received the events. If the pad is actually
not-linked, the sticky events will be rescheduled when the
pad is linked anyway.
Jan Schmidt [Thu, 29 Oct 2015 07:53:29 +0000 (18:53 +1100)]
ghostpad: Allow deactivation with no peer.
Allow deactivation in pull-mode, since that implies we
had a peer, activated in pull mode, then the peer disa-peer-ed ;)
Thiago Santos [Wed, 28 Oct 2015 20:31:07 +0000 (17:31 -0300)]
element: emit tracer's element-new hook from 'constructed'
It allows to properly emitting it for all newly created elements
https://bugzilla.gnome.org/show_bug.cgi?id=757045
Thiago Santos [Wed, 28 Oct 2015 09:03:39 +0000 (06:03 -0300)]
stats: log the element-new hook properly
To be able to get the time the elements were created instead of
just logging them without a time
Sebastian Dröge [Wed, 28 Oct 2015 16:04:46 +0000 (18:04 +0200)]
pad: Unblock blocking pad probes when receiving FLUSH_START in send_event() too
Without this, flushing might not unblock the streaming thread and cause deadlocks.
https://bugzilla.gnome.org/show_bug.cgi?id=757257
Sebastian Dröge [Wed, 28 Oct 2015 11:04:25 +0000 (13:04 +0200)]
pad: Document the order in which pad probes are called
https://bugzilla.gnome.org/show_bug.cgi?id=757197
Sebastian Dröge [Tue, 27 Oct 2015 16:05:05 +0000 (18:05 +0200)]
pad: Append hooks instead of prepending to call them in the order they were added
https://bugzilla.gnome.org/show_bug.cgi?id=757197
Stian Selnes [Mon, 26 Oct 2015 11:50:12 +0000 (12:50 +0100)]
netaddressmeta: gst_buffer_get_net_address_meta() as function
Implement gst_buffer_get_net_address_meta() as a function instead
of a macro in order to get gobject-introspection to work.
https://bugzilla.gnome.org/show_bug.cgi?id=702921
Nirbheek Chauhan [Mon, 26 Oct 2015 12:37:11 +0000 (18:07 +0530)]
buffer: flesh out docs for gst_buffer_make_writable() a little
There is a similar explanation in gst_caps_make_writable, but the existing
documentation can be misleading since it does not define what 'is already
writable' means.
Also note when this function is meant to be used.
Nirbheek Chauhan [Mon, 26 Oct 2015 12:05:34 +0000 (17:35 +0530)]
caps: clarify docs for a few functions that they don't update things in-place
It is not necessarily clear from the existing introspection tags and
documentation alone.
Edward Hervey [Tue, 27 Oct 2015 07:48:07 +0000 (08:48 +0100)]
queue/queue2: Use GST_BUFFER_DTS_OR_PTS
The input of queue/queue2 might have DTS set, in which cas we want
to take that into account (instead of the PTS) to calculate position
and queue levels.
https://bugzilla.gnome.org/show_bug.cgi?id=756507
Edward Hervey [Tue, 13 Oct 2015 15:20:26 +0000 (17:20 +0200)]
multiqueue: Use buffer DTS if present, else PTS
In order to accurately determine the amount (in time) of data
travelling in queues, we should use an increasing value.
If buffers are encoded and potentially reordered, we should be
using their DTS (increasing) and not PTS (reordered)
https://bugzilla.gnome.org/show_bug.cgi?id=756507
Edward Hervey [Tue, 27 Oct 2015 07:33:41 +0000 (08:33 +0100)]
collectpads: Use GST_BUFFER_DTS_OR_PTS
Simplifies code a bit
Edward Hervey [Mon, 26 Oct 2015 06:59:54 +0000 (07:59 +0100)]
buffer: Add a GST_BUFFER_DTS_OR_PTS macro
API: GST_BUFFER_DTS_OR_PTS
Many scenarios/elements require dealing with streams of buffers that
might have DTS set (i.e. encoded data, potentially reordered)
To simplify getting the increasing "timestamp" of those buffers, create
a macro that will return the DTS if valid, and if not the PTS
Guillaume Desmottes [Tue, 6 Oct 2015 10:21:04 +0000 (12:21 +0200)]
doc: add GST_{PTR,SEGMENT}_FORMAT
Very useful formats in debug output so best to have them in the
generated doc.
https://bugzilla.gnome.org/show_bug.cgi?id=756115
Sebastian Dröge [Mon, 26 Oct 2015 08:53:35 +0000 (10:53 +0200)]
segment: Return -1 if gst_segment_to_stream_time_full() considers the position not inside the segment
Fixes GstPipeline unit test.
https://bugzilla.gnome.org/show_bug.cgi?id=756564
Florin Apostol [Sat, 24 Oct 2015 15:52:44 +0000 (16:52 +0100)]
uri: fix behaviour for merging uris ending in .. without following /
https://bugzilla.gnome.org/show_bug.cgi?id=757065
Florin Apostol [Sat, 24 Oct 2015 15:43:59 +0000 (16:43 +0100)]
uri: tests: added unit test for streams ending in .. without following /
https://bugzilla.gnome.org/show_bug.cgi?id=757065
Thibault Saunier [Thu, 27 Aug 2015 10:43:28 +0000 (12:43 +0200)]
debug: Dump pad properties values
Currently we only show element properties values, we should also show
pad properties values
https://bugzilla.gnome.org/show_bug.cgi?id=754166
Sebastian Dröge [Fri, 23 Oct 2015 17:04:42 +0000 (20:04 +0300)]
segment: Remove leftover debug g_print()
Vivia Nikolaidou [Thu, 15 Oct 2015 11:49:37 +0000 (14:49 +0300)]
segment: Add _full variants of all stream/running_time from/to segment position functions
See formula clarifications in design docs for calculation details.
https://bugzilla.gnome.org/show_bug.cgi?id=756564
Vivia Nikolaidou [Fri, 25 Sep 2015 22:29:07 +0000 (01:29 +0300)]
gst-uninstalled: Added env var for uninstalled PTP helper
https://bugzilla.gnome.org/show_bug.cgi?id=755651
Guillaume Desmottes [Thu, 22 Oct 2015 10:00:42 +0000 (12:00 +0200)]
basesink: rename argument of PREROLL_{COND,LOCK} macros
They take a GstBaseSink instance as argument at not a GstPad. Rename the
argument to 'obj' which is not miss leading and in line with
GST_BASE_SINK_PAD(obj).
https://bugzilla.gnome.org/show_bug.cgi?id=756954
Mathieu Duponchelle [Thu, 22 Oct 2015 08:05:14 +0000 (10:05 +0200)]
gstcontrolsource: Add missing (out) annotation
Tim-Philipp Müller [Wed, 21 Oct 2015 13:34:47 +0000 (14:34 +0100)]
Automatic update of common submodule
From b99800a to b319909
Sebastian Dröge [Wed, 21 Oct 2015 11:49:49 +0000 (14:49 +0300)]
pad: Fix docs/annotation of gst_pad_probe_info_get_buffer_list()
It's not get_bufferlist(). Because of that it was ignored by the docs and
G-I, leading to crashes because of broken ownership transfer.
https://bugzilla.gnome.org/show_bug.cgi?id=756898
Sebastian Dröge [Tue, 20 Oct 2015 14:29:42 +0000 (17:29 +0300)]
Use new GST_ENABLE_EXTRA_CHECKS #define
https://bugzilla.gnome.org/show_bug.cgi?id=756870
Sebastian Dröge [Wed, 21 Oct 2015 11:25:40 +0000 (14:25 +0300)]
Automatic update of common submodule
From 9aed1d7 to b99800a
Luis de Bethencourt [Tue, 20 Oct 2015 12:46:24 +0000 (13:46 +0100)]
stats: always free log
We always want to free the open file log if fopen() succeeded. Independently
of if fgets() succeeds or fails.
CID 1326055
https://bugzilla.gnome.org/show_bug.cgi?id=756864
Vivia Nikolaidou [Mon, 19 Oct 2015 13:50:51 +0000 (16:50 +0300)]
segment: Correct stream_time calc for negative applied rate
Updated gst_segment_position_from_stream_time and gst_segment_to_stream_time to reflect correct calculations for the case when the applied rate is negative.
Pasting from design docs:
===============================
Stream time is calculated using the buffer times and the preceding SEGMENT
event as follows:
stream_time = (B.timestamp - S.start) * ABS (S.applied_rate) + S.time
For negative rates, B.timestamp will go backwards from S.stop to S.start,
making the stream time go backwards.
===============================
Therefore, the calculation for applied_rate < 0 should be:
stream_time = (S.stop - B.timestamp) * ABS (S.applied_rate) + S.time
and the reverse:
B.timestamp = S.stop - (stream_time - S.time) / ABS (S.applied_rate)
https://bugzilla.gnome.org/show_bug.cgi?id=756810
Stefan Sauer [Mon, 19 Oct 2015 19:39:19 +0000 (21:39 +0200)]
tracer: rename the envvar to GST_TRACER_PLUGINS
The subsystem reused the GST_TRACE var that is allready in use by the alloc tracer.
Fixes #756760
Edward Hervey [Thu, 15 Oct 2015 14:32:42 +0000 (16:32 +0200)]
multiqueue: Improve incoming SEGMENT handling
Previously this code was just blindly setting the cached flow return
of downstream to GST_FLOW_OK when we get a SEGMENT.
The problem is that this can not be done blindly. If downstream was
not linked, the corresponding sinqlequeue source pad thread might be
waiting for the next ID to be woken up upon.
By blindly setting the cached return value to GST_FLOW_OK, and if that
stream was the only one that was NOT_LINKED, then the next time we
check (from any other thread) to see if we need to wake up a source pad
thread ... we won't even try, because none of the cached flow return
are equal to GST_FLOW_NOT_LINKED.
This would result in that thread never being woken up
https://bugzilla.gnome.org/show_bug.cgi?id=756645
Ting-Wei Lan [Sat, 26 Sep 2015 10:16:07 +0000 (18:16 +0800)]
gstutils: Fix build with clang -Werror=cast-align
https://bugzilla.gnome.org/show_bug.cgi?id=755657
Sebastian Dröge [Sat, 17 Oct 2015 19:13:08 +0000 (22:13 +0300)]
bin: Make sure to free all cached messages when going to NULL
An ASYNC READY->PAUSED might have failed without the bin code noticing during
the state change, in which case we will never get PAUSED->READY and would leak
messages.
https://bugzilla.gnome.org/show_bug.cgi?id=756611
Tim-Philipp Müller [Fri, 16 Oct 2015 14:59:49 +0000 (15:59 +0100)]
docs: mention xdot utility to view .dot files directly
Tim-Philipp Müller [Fri, 16 Oct 2015 11:00:50 +0000 (12:00 +0100)]
alloctrace: show details of events and messages leaked
So it's clearer what leaked.
Tim-Philipp Müller [Fri, 16 Oct 2015 10:54:18 +0000 (11:54 +0100)]
docs: manual: improve advanced metadata example a bit
Accept both filename and a URI as argument, and print
the error from the error message if there's an error.
https://bugzilla.gnome.org/show_bug.cgi?id=756630
Thiago Santos [Thu, 15 Oct 2015 22:05:21 +0000 (19:05 -0300)]
memory: fix typo in documentation
It should be 1.2 unless this is a flag from the future
Edward Hervey [Tue, 15 Sep 2015 16:08:18 +0000 (18:08 +0200)]
multiqueue: Accept STREAM_START after EOS
In the same way core now allows STREAM_START to remove the flushing
state from pads, we need to do the same thing in multiqueue
Sebastian Dröge [Wed, 14 Oct 2015 08:03:22 +0000 (11:03 +0300)]
segment: Convert function to macro in unit test to get proper line numbers on failures
https://bugzilla.gnome.org/show_bug.cgi?id=748316
Edward Hervey [Mon, 12 Oct 2015 15:29:26 +0000 (17:29 +0200)]
baseparse: Update internal position even if not linked
Our current position has nothing to do with being linked or not.
Avoids having stray segment updates fired every 2s
Florin Apostol [Wed, 7 Oct 2015 21:55:44 +0000 (22:55 +0100)]
pad: fix memory leak when sending events to an EOS pad
https://bugzilla.gnome.org/show_bug.cgi?id=756208
Sebastian Dröge [Wed, 7 Oct 2015 11:01:16 +0000 (12:01 +0100)]
tracers: Only link against libgstprintf.la if the debugging system is enabled
It does not exist otherwise and linking will fail.
Sebastian Dröge [Wed, 7 Oct 2015 10:25:52 +0000 (11:25 +0100)]
gst-ptp-helper: #include <sys/socket.h> to fix net/if.h include on OSX 10.6
In file included from gst-ptp-helper.c:40:0:
/usr/include/net/if.h:265:19: error: field 'ifru_addr' has incomplete type
struct sockaddr ifru_addr;
https://bugzilla.gnome.org/show_bug.cgi?id=756136
Stefan Sauer [Wed, 7 Oct 2015 10:22:34 +0000 (12:22 +0200)]
Revert "tracers: Only build getrusage() tracer if RUSAGE_THREAD is available"
This reverts commit
8ddbf76626a48420306869db1d171f854cc25310.
Stefan Sauer [Wed, 7 Oct 2015 10:21:56 +0000 (12:21 +0200)]
tracers/rusage: ifdef the RUSAGE_THREAD usage
Some versions of andoid don't seem to have it.
Sebastian Dröge [Wed, 7 Oct 2015 10:11:30 +0000 (11:11 +0100)]
tracers: Only build getrusage() tracer if RUSAGE_THREAD is available
Stefan Sauer [Tue, 6 Oct 2015 19:46:55 +0000 (21:46 +0200)]
win32: remove gst_tracer_quark_id_get_type
Revert addition from
777bbeea605051ae3d2fa7e02ad8589001e78ce0.
Stefan Sauer [Tue, 6 Oct 2015 16:52:38 +0000 (18:52 +0200)]
tracer: move prototype to the right header
Fixes the build when the tracing subsystem is disabled.
Stefan Sauer [Tue, 6 Oct 2015 16:49:46 +0000 (18:49 +0200)]
tracer: mark GstTracerQuarkId as non GEnum
This reverts
72ca02b1de4066eeae35c891e275386770117778 and marks the enum
accordingly.
Stefan Sauer [Tue, 6 Oct 2015 16:46:24 +0000 (18:46 +0200)]
tracers: disable the log tracer if debug logging is disabled
Stefan Sauer [Tue, 6 Oct 2015 16:45:41 +0000 (18:45 +0200)]
makefile.am: Remove obsolete Android build cruft
This is not needed any longer.
Stefan Sauer [Tue, 6 Oct 2015 12:01:03 +0000 (14:01 +0200)]
tracer: fix the build with debug (tracer) disabled
Sync the macro definitions. The dummy defines has argument mismatches.
Stefan Sauer [Tue, 6 Oct 2015 09:39:33 +0000 (11:39 +0200)]
tracer: fix the build with debug (tracer) disabled
Remove commas at the end of the macros.
Thiago Santos [Tue, 1 Sep 2015 19:39:30 +0000 (16:39 -0300)]
tracers: stats: add message structure to output
The name of the message is not enough. For example, state-change
is not enough to know the transition.
https://bugzilla.gnome.org/show_bug.cgi?id=754496
Thiago Santos [Mon, 5 Oct 2015 22:05:58 +0000 (19:05 -0300)]
gst: adding tracer quark id to gst init and deinit
Fixes issues at make check
Thiago Santos [Mon, 5 Oct 2015 21:50:48 +0000 (18:50 -0300)]
win32: libgstreamer: add tracer functions
Stefan Sauer [Mon, 5 Oct 2015 19:29:49 +0000 (21:29 +0200)]
tests: fix the tag test
The previous change (see bgo #756069) was causing us to free the same pointer
multiple times. If we actually get a sample back, the test fails, no need to
free anything in that case.
Marcin Kolny [Wed, 3 Jun 2015 23:50:34 +0000 (01:50 +0200)]
tracer: add missing hooks
Add following hooks: element-new, element-add-pad, element-remove-pad,
bin-add-pre, bin-add-post, bin-remove-pre, bin-remove-post, pad-link-pre,
pad-link-post, pad-unlink-pre, pad-unlink-post.
https://bugzilla.gnome.org/show_bug.cgi?id=733187
Thiago Santos [Fri, 13 Mar 2015 18:31:40 +0000 (18:31 +0000)]
tracer: gststats: add thread-id to log line
Thiago Santos [Fri, 13 Mar 2015 13:10:42 +0000 (13:10 +0000)]
tracer: add pad query hooks
Mathieu Duponchelle [Thu, 15 Jan 2015 05:32:48 +0000 (06:32 +0100)]
tracer: strdup the passed parameters.
Mathieu Duponchelle [Tue, 13 Jan 2015 21:11:34 +0000 (22:11 +0100)]
tracer: Use GST_TIME_ARGS when printing with GST_TIME_FORMAT.
Stefan Sauer [Thu, 16 Oct 2014 08:42:05 +0000 (10:42 +0200)]
tracing: update docs
Stefan Sauer [Thu, 2 Oct 2014 17:52:03 +0000 (19:52 +0200)]
stats: TIMESTAMP -> PTS
Stefan Sauer [Mon, 22 Sep 2014 07:55:56 +0000 (09:55 +0200)]
tracing: rename the global api to gst_tracing
This makes it more obvious what is the api for tracer elements and what is api
for the global state.
Stefan Sauer [Thu, 18 Sep 2014 06:28:48 +0000 (08:28 +0200)]
stats: fix cpu stats printing
Only print them if we have them. Also scale them by 10.0 as the are in
per-mille now.
Stefan Sauer [Thu, 18 Sep 2014 06:26:19 +0000 (08:26 +0200)]
tracers: eliminate var_args
Register tracer hooks as GCallback. Use macros for hook dispatch and cast the
hook functions back to the appropriate type.
Stefan Sauer [Wed, 17 Sep 2014 07:41:46 +0000 (09:41 +0200)]
stats: don't warn on ".class" log lines
Stefan Sauer [Wed, 17 Sep 2014 06:38:02 +0000 (08:38 +0200)]
tracers: code cleanups
Move static variables to instance variables. Add finalize methods. Remove code
that is commented out. Cleanup locking code.
Stefan Sauer [Mon, 15 Sep 2014 20:27:11 +0000 (22:27 +0200)]
tracer: use GQuark or strings for the hook id
This way one can define new tracing probes without changing the core. We are
using our own quark table, as 1) we only want to initialize them if we're
tracing, 2) we want to share them with the tracers.
Stefan Sauer [Mon, 15 Sep 2014 11:15:17 +0000 (13:15 +0200)]
tracer: simplify hook api
Instead of a single invoke() function and a 'mask', register to individual
hooks. This avoids one level of indirection and allows us to remove the
hook enums. The message enms are now renamed to hook enums.
Stefan Sauer [Fri, 12 Sep 2014 16:43:52 +0000 (18:43 +0200)]
tracer: drop the HookId hid from the invoke method
The MessageId is more detailed and anyway needed to interpret the varargs.
Stefan Sauer [Fri, 12 Sep 2014 09:17:41 +0000 (11:17 +0200)]
stats: fixup doc name and remove commented code
Stefan Sauer [Fri, 12 Sep 2014 06:40:01 +0000 (08:40 +0200)]
tracers: add metadata for the logged values
Stefan Sauer [Thu, 11 Sep 2014 11:02:51 +0000 (13:02 +0200)]
rusage: improve cpu load meassurements
Get the number of cpus and scale process cpu-load accordingly. Switch the
cpuload to be per-mille to get smoother graphs. Add a bit more logging and use
the _OBJECT variant.