Commit graph

16845 commits

Author SHA1 Message Date
Sebastian Dröge
5664fd7635 buffer: Protect against failing to map input memory when merging memories
https://bugzilla.gnome.org/show_bug.cgi?id=762239
2016-02-18 11:09:36 +02:00
Matej Knopp
17d30e944b pad: Add test for blocking pull probe
https://bugzilla.gnome.org/show_bug.cgi?id=761211
2016-02-18 09:44:00 +02:00
Sebastian Dröge
b89fa4786b pad: PULL probes are called without a buffer so don't require any of the data flags to be set
https://bugzilla.gnome.org/show_bug.cgi?id=761211
2016-02-18 09:44:00 +02:00
Sebastian Dröge
a0b3a7f658 Revert "element: Don't hold state lock all the time while sending an event"
This reverts commit b427997119.

It breaks things that used to work before, even if the change by itself is
correct and the previous code is just working around deeper bugs in the async
state change code. Let's go back to what previously worked and then fix async
state changes in general.

https://bugzilla.gnome.org/show_bug.cgi?id=760532
2016-02-17 16:41:02 +02:00
Edward Hervey
899060d579 Revert "ghostpad: Do nothing in _internal_activate_push_default"
That commit would break scheduling reconfiguration with ghostpads

This reverts commit ab55ad7eaa.
2016-02-17 15:26:49 +01:00
Edward Hervey
8ae671cf1d check: Add test for checking scheduling reconfiguration with ghostpads
Showcases the regression introduced by this commit:
Commit: ab55ad7eaa

Author: Stian Selnes <stian@pexip.com>
Date:   Wed Jan 27 13:20:23 2016 +0100

ghostpad: Do nothing in _internal_activate_push_default
2016-02-17 15:26:41 +01:00
Havard Graff
87c2254b1d ghostpad: add some tests for activation
https://bugzilla.gnome.org/show_bug.cgi?id=761913
2016-02-17 12:26:12 +02:00
Stian Selnes
ab55ad7eaa ghostpad: Do nothing in _internal_activate_push_default
When calling gst_pad_activate_mode() on a ghostpad
gst_ghost_pad_activate_push_default() will be called. This will call
gst_pad_activate_mode() on the proxypad (which is internal of the
ghostpad), calling gst_ghost_pad_internal_activate_push_default(), which
again will call gst_pad_activate_mode() on the original ghostpad.

By simply returning TRUE in
gst_ghost_pad_internal_activate_push_default() the redundant call to
gst_pad_activate_mode() (for the same pad) is avoided.

https://bugzilla.gnome.org/show_bug.cgi?id=761913
2016-02-17 12:26:12 +02:00
Thiago Santos
6f4bb8850b registrychunks: remove unused macro
macro was added in 2011 and isn't used anymore
2016-02-16 17:53:10 -03:00
Sebastian Dröge
fd253ab1ab Revert "queue2: add overrun signal"
This reverts commit 8ae8b2723d.

It's not used anymore by anything and was considered a bad idea in general.
2016-02-16 19:11:59 +02:00
Evan Nemerson
d11e657412 docs: annotate C examples as such
https://bugzilla.gnome.org/show_bug.cgi?id=731292
2016-02-15 17:45:15 +00:00
Thiago Santos
ac12ad076f benchmark: tracerserialize: add missing return statement
tracerserialize.c:117:1: error: control reaches end of
non-void function [-Werror=return-type]
2016-02-15 12:34:17 -03:00
Thiago Santos
5b64123c54 protection/harness/systemclock: move declaration out of for loop initialization
C90 compilers complain about it
error: ‘for’ loop initial declarations are only allowed in C99 or C11 mode

Also run gst-indent on systemclock tests.
2016-02-15 12:33:46 -03:00
Havard Graff
69f5d28718 harness: always set our test-clock on the harnessed element
The integration is already so tight, there is no reason to
not further formalize it!

https://bugzilla.gnome.org/show_bug.cgi?id=761914
2016-02-15 10:09:42 +00:00
Tim-Philipp Müller
aafa79e281 harness: fix compilation
Was supposed to be squashed with 336c7bb6

https://bugzilla.gnome.org/show_bug.cgi?id=761910
2016-02-13 16:11:07 +00:00
Tim-Philipp Müller
39d3c06b18 harness: fix indentation 2016-02-13 16:10:11 +00:00
Stian Selnes
336c7bb69c harness: Fix MT issues when forwarding event/query to sink harness
https://bugzilla.gnome.org/show_bug.cgi?id=761910
2016-02-13 15:36:03 +00:00
Tim-Philipp Müller
7ae052e3eb gst-uninstalled: add new -bad libraries audio, player and wayland to paths
And remove egl which no longer exists.
2016-02-13 10:04:42 +00:00
Martin Kelly
86a46ee43b pushsrc: fix minor typos in header
https://bugzilla.gnome.org/show_bug.cgi?id=761970
2016-02-12 20:50:57 +00:00
Stian Selnes
778160be08 harness: Add event stress test functions with callback
Similar to the stress test functions for buffers that has a callback to
create the buffer to be pushed, it's useful to have functions that use a
callback to create the event to be pushed.

API: gst_harness_stress_push_event_with_cb_start()
API: gst_harness_stress_push_event_with_cb_start_full()
API: gst_harness_stress_send_upstream_event_with_cb_start()
API: gst_harness_stress_push_upstream_event_with_cb_start_full()

https://bugzilla.gnome.org/show_bug.cgi?id=761932
2016-02-12 19:32:29 +00:00
Havard Graff
668b3215b8 testclock: add crank method
And use it inside GstHarness

API: gst_test_clock_crank()

https://bugzilla.gnome.org/show_bug.cgi?id=761906
2016-02-12 17:08:32 +00:00
Havard Graff
cc4e4ae029 harness: enable empty harness creation and refactor around this
Also make the testclock a member of the harness, allowing some
more interactions with the clock prior to adding elements.

https://bugzilla.gnome.org/show_bug.cgi?id=761905
2016-02-12 16:42:43 +00:00
Stian Selnes
9d56c0e397 check: fix unused parameter compiler warning
https://bugzilla.gnome.org/show_bug.cgi?id=761919
2016-02-12 16:38:43 +00:00
Mikhail Fludkov
06ae41e0b2 harness: fix the race in blocking push mode
Depending on when gst_harness_pull was called - before the buffer reached
gst_harness_chain or after we can get different behaviors of the test
with enabled blocking push mode. The fix makes the behavior always the
same. In pull function we get the buffer first, thus making sure
gst_harness_chain waits for the signal, and emitting the signal after.

https://bugzilla.gnome.org/show_bug.cgi?id=761931
2016-02-12 12:32:22 +00:00
Stian Selnes
2450af1034 check: Add tcase_skip_broken_loop_test
https://bugzilla.gnome.org/show_bug.cgi?id=761917
2016-02-12 11:14:33 +00:00
Stian Selnes
ff7e6a14b0 harness: Fix docs for stress test functions
notify is not called per buffer, but when the thread is freed.

Comment about serialized events and OOB does not make sense for upstream
events.

https://bugzilla.gnome.org/show_bug.cgi?id=761909
2016-02-12 10:22:45 +00:00
Stian Selnes
dc4a3b71f0 harness: Unset sink_forward_pad before tearing down sink_harness
Set the sink_forward_pad to NULL before tearing down sink_harness to
avoid that the harness tries to forward events/queries to it while it's
tearing down.

https://bugzilla.gnome.org/show_bug.cgi?id=761904
2016-02-12 10:17:31 +00:00
Havard Graff
ed583832a1 harness: fix up docs to reference functions properly
https://bugzilla.gnome.org/show_bug.cgi?id=761901
2016-02-12 10:14:29 +00:00
Víctor Manuel Jáquez Leal
b92d639fbf bufferpool: pass acquire params to alloc_buffer
When allocating a new buffer in the pool, both the do_alloc_buffer() and the
vmethod, alloc_buffer(), receive the parameter GstBufferPoolAcquireParams.

Nonetheless, when default_acquire_buffer() calls the do_alloc_buffer() it does
not pass the received GstBufferPoolAcquireParams, so when the user pass those
parameters they are ignored by alloc_buffer() vmethod.

This one-liner patch pass the received acquire params to do_alloc_buffer().

https://bugzilla.gnome.org/show_bug.cgi?id=761824
2016-02-11 12:11:20 -05:00
Stian Selnes
2b0ff13d07 systemclock: Fix wait/unschedule race
Fixes a race where an entry is set to BUSY in
gst_system_clock_id_wait_jitter() and is UNSCHEDULED before
gst_system_clock_id_wait_jitter_unlocked() starts processing it. The
wakeup added by gst_system_clock_id_unschedule() must be cleaned up.

Two stress tests are added. One test that triggers the specific issue
described above. The second stresses the code path where a wait is
rescheduled because the poll returned early.

https://bugzilla.gnome.org/show_bug.cgi?id=761586
2016-02-10 08:59:33 -03:00
Thiago Santos
77a98d98cb systemclock: handle unschedule of late entries
If the clockentry is too late and is unscheduled before it gets
a change to detect its lateness the wakeup count and the poll are
used but never properly cleaned up. This leaves it in a dirty state
that is going to mess with the next clock entry waiting requests.

https://bugzilla.gnome.org/show_bug.cgi?id=761586
2016-02-10 08:59:33 -03:00
Thiago Santos
682516402d tests: extend the AM_TESTS_ENVIRONMENT from check.mak
To get the CK_DEFAULT_TIMEOUT defined for all tests

https://bugzilla.gnome.org/show_bug.cgi?id=761472
2016-02-05 19:11:10 -03:00
Thiago Santos
e5a75e2689 Automatic update of common submodule
From 86e4663 to b64f03f
2016-02-05 18:01:52 -03:00
Tim-Philipp Müller
78a832ebd5 baseparse: fix stray discont flag set on outgoing buffers in push mode
We have no guarantees about what flags are set on buffers we take
out of the GstAdapter. If we push out multiple buffers from the
first input buffer (which will have discont set), only the first
buffer we push out should be flagged as discont, not all of the
buffers produced from that first initial input buffer.

Fixes issue where the first few mp3 frames/seconds of data in push
mode were skipped or garbled in some cases, and the discont flags
would also trip up decoders which were getting drained/flushed for
every buffer. This was a regression introduced in 1.6 apparently.
2016-02-04 19:04:41 +00:00
Thibault Saunier
438a83df58 controller: Do not unset uninitiallized GValue
In case the property was not interpollable we might never initialize
the GValue, we should thus never unset it.
2016-02-02 19:51:56 +01:00
Tim-Philipp Müller
81df1e1a34 docs: pwg: fix missing end of line semicolon in custom meta example 2016-02-02 16:34:51 +00:00
Tim-Philipp Müller
f90126a62e uri: add guard to make sure gstreamer is initialized
https://bugzilla.gnome.org/show_bug.cgi?id=761448
2016-02-02 10:56:35 +00:00
Tim-Philipp Müller
961ee1910e gst-uninstalled: add new rtsp server plugin location to plugins path 2016-02-01 18:41:55 +00:00
HoonHee Lee
f90fd86d5f baseparse: Try to generate caps on the srcpad before forwarding GAP event
To configure downstream elements and complete initial pre-rolling,
ensure we have default output caps before forwarding GAP event.

https://bugzilla.gnome.org/show_bug.cgi?id=753899
2016-01-29 10:49:24 +01:00
Alex Henrie
78ba7f9a8e typefindelement: Improve English grammar
https://bugzilla.gnome.org/show_bug.cgi?id=761273
2016-01-29 10:34:44 +01:00
Luis de Bethencourt
09397c442d docs: fix an other typo in clock chapter
Shockingly I missed this bigger typo in the previos fix
2016-01-27 12:45:23 +00:00
Luis de Bethencourt
130d6c8282 docs: fix typo in clock chapter 2016-01-27 12:27:44 +00:00
Vineeth TM
ca97a9882f tests:gstinfo: Fix string memory leak
info_fourcc test leaks string.

https://bugzilla.gnome.org/show_bug.cgi?id=761071
2016-01-25 09:41:41 -03:00
Tim-Philipp Müller
afaeec665a tests: info: make work without registry 2016-01-23 16:00:48 +00:00
Tim-Philipp Müller
63421675b1 tracer: rename GST_TRACER_PLUGINS env var to GST_TRACERS 2016-01-22 12:50:08 +00:00
Stefan Sauer
ec75b68984 tracerrecord: don't leak the spec structures
Change the gst_tracer_record_new() api to take the parameters the make the
spec structure directly. This allows us to own the top-level structure and
also collect the args so that we can take ownership of the sub-structures.

https://bugzilla.gnome.org/show_bug.cgi?id=760821
2016-01-22 10:07:26 +01:00
Tim-Philipp Müller
05a9655523 device: fix comparison in _has_classesv()
We're comparing a pointer type with '\0' here, which
probably isn't right, and the loop condition made sure
that classes[0] is != NULL already, so it's pointless.
Was probaby meant to check if the string pointed to is
not empty, so make it do that instead.
2016-01-21 15:45:30 +00:00
Stefan Sauer
a72368ebb3 test/tracerrecord: unref objects and free string 2016-01-21 08:25:24 +01:00
Florin Apostol
077db508a8 systemclock: tests: added stress test for async order
Keep inserting alarms at the beginning of the list. Due to
https://bugzilla.gnome.org/show_bug.cgi?id=760757
alarm thread will get confused and not serve them in order.
2016-01-20 13:47:20 +02:00
Florin Apostol
f0e94b4cdf systemclock: fixed race condition in handling alarms
When choosing the first entry from the list, gst_system_clock_async_thread
must set the entry state to busy before releasing the clock lock. Otherwise
a new entry could be added to the beginning of the list and
gst_system_clock_async_thread will be unaware and keep waiting on the entry
it has already chosen.

Also improved messages about expected state and bumped them to ERROR level
to detect unexpected state changes.

https://bugzilla.gnome.org/show_bug.cgi?id=760757
2016-01-20 13:47:20 +02:00