Original commit message from CVS:
Another big set of changes. Connections are now also pullfunc based.
gstqueue has been updated, I don't know of any other connections offhand.
There are still a few things that need doing, specifically the concept
of a source or connection with connections to multiple thread contexts is
not dealt with. This may force us to move the threadstate from the
element to the pad, maybe keeping the element's copy for simple cases.
Then the Bin would create a structure to pass to the cothreaded _wrappers
of any such elements, which would detail the pads that are to be dealt with
by this particular cothread context.
That will speed things up to, since we don't have to look through the list
of all pads for every Src or Connection element for every iteration, we can
simply step through the list provided by the plan. Special case might even
have a single pad pointer sitting there to trump the list, if there's only
one (the common case anyway).
Task 23098 is tracking these changes. The main task 22588 depends on that
subtask, as well as 22240, which is a consistency check on PAD_DISABLED.
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
Original commit message from CVS:
Changes made to the DEBUG system. New header file gstdebug.h holds the
stuff to keep it out of gst.h's hair. DEBUG prints out the process id,
cothread id, source filename and line number. Two new macros DEBUG_ENTER
and DEBUG_LEAVE are used to show the entry and exit of a given function.
This eventually might be used to construct call trace graphs, even taking
cothreads into account. This would be quite useful in visualizing the
scheduling mechanism.
Minor changes to various debug messages.
Also sitting in gstdebug.h is a prototypical DEBUG_ENTER that's capable of
performing DEBUG_LEAVE automatically. It does this by utilizing a
little-known GCC extension that allows one to call a function with the
same parameters as the current function. The macro uses this to basically
call itself. A boolean is used to ensure that when it calls itself it
actually runs the body of the function. In the meantime it prints stuff
out before and after the real function, as well as constructing a
debugging string. This can be used eventually to provide call-wide data
on the DEBUG lines, instead of having to replicate data on each call to
DEBUG. More research is needed into how this would most cleanly be fit
into some other chunk of code, like GStreamer (I think of this DEBUG trick
as a separate project, sorta).
Unfortunately, the aforementioned DEBUG trick interacts quite poorly with
cothreads. Almost any time it's used in a function that has anything
remotely to do with a cothread context (as in, it runs in one), a segfault
results from the __builtin_apply call, which is the heart of the whole
thing. If someone who really knows assembly could analyze the resulting
code to see what's really going on, we might find a way to fix either the
macro or the cothreads (I'm thinking that there's something we missed in
constructing the cothreads themselves) so this works in all cases.
In the meantime, please insert both DEBUG_ENTER and DEBUG_LEAVE in your
functions. Be sure to put DEBUG_ENTER after your variable declarations
and before any functional code, not to put the function name in any DEBUG
strings (it's already there, trust me), and put a DEBUG_LEAVE if you care
enough.
Changes are going to happen in the way DEBUGs and other printouts occur,
so stay tuned.
Original commit message from CVS:
Added a check to see if the plugin actually exists before even bothering
to try to load it. Saves a fair amount of debugging spew with
_gst_plugin_spew enabled.
Original commit message from CVS:
Added multiple types to the pads.
Added first preview of the capabilities system.
Autoplugging is seriously broken with these (and other) changes.
Original commit message from CVS:
Enhanced debugging by making DEBUG() print out the cothread ID as well as
the process ID. cothread_getcurrent returns the current cothread ID, or
-1 if cothreads aren't set up in this process context.
Original commit message from CVS:
Rearranged cothread_switch a bit to move all the code for error conditions
to the end, using goto's to get there. This has the presumed advantage of
consolidating all the normally run code into one chunk, reducing jumps
(and the associated penalties in any modern processor) and limiting cache-
line usage. It may be instructive to look at the generated assembly for
this revision and the previous, to see if gcc is smart enough to do this
for us anyway.
If you want to turn off some of the checks (they are all checks for NULL
pointers, if you're curious) for a speed gain, disable the #define of
COTHREAD_PARANOID at the top.
Original commit message from CVS:
Added code to force the gsttypes plugin to load before gstelements, by
simply having gstelements.c request the plugin. Solves some dependency
problems. This is the correct method of doing this for now, though I had
a thought:
Have a static list of hard dependencies that the plugin system is responsible
for satisfying before even trying to load the plugin. Makes plugin design
easier.
Original commit message from CVS:
Fixed a dumb mistake: in gst_element_set_loop_function, it was forcing
the cothread's entry function to change. This is totally wrong, the
create_plan function in GstBin is responsible for setting that to its
own internal helper function. Code removed outright...
Original commit message from CVS:
Closed 22145 - gstelement.c: support hotswapping of loopfunc
Added a flag GST_ELEMENT_NEW_LOOPFUNC that's set when a new loopfunc is set,
which needs to be obeyed by the loopfuncs themselves. The assumption is
that only a loopfunc or an element that knows what it's doing will ever
change the loopfunc pointer, so they can be responsible for exitting
the loopfunc itself appropriately.
The loopfunc_wrapper function has been changed to call the loopfunc
pointer itself in a loop. If the loopfunc changes, the current loop is
supposed to exit, which causes the loopfunc_wrapper to start up the new
loopfunc.
Original commit message from CVS:
Closed task 22143 - gstelemnt.c: complete gst_element_connect
SImple matter of cleaning up the conflicting checks and actually
connecting the pads in question.
Original commit message from CVS:
gstbin.c: modified so create_plan occurs after the state change of all the
child elements.
gstelement.c: set_state now loops such that each element only deals with
one state change at a time, i.e. NULL->READY,READY->PLAYING,
instead of a single NULL->PLAYING.
Original commit message from CVS:
Reworked AC3 decoder. No seeking yet but at least we do not need ac3parse
anymore.
Reworked dvdsrc to read scrambled data (DeCSS not included). I have
modified DeCSS a bit to work in GStreamer. Can I release the code or is
there some lawyer that's going to sue me?
MPEG2 SSE motion compensation.
Tried to add PTS to the MPEG decoder but failed.
Original commit message from CVS:
Docs updates.
Added LICENSE info to headers/code where missing in gst directory
Added a bonobo wrapper for the media player (it shows up in gshell but
locks up when activating the component, anyone?)
Fixed some XML save/load problems with arguments.
Original commit message from CVS:
Added seeking to the avi decoder by implementing pull_region.
Fixes to the asyncdisksrc.
Added thread specific data to the cothreads.
Original commit message from CVS:
Make sure the Gdk thread lock is not held inside the gstreamer pipeline.
Fix the videosink to not hold the Gdk lock when emitting a signal.
All Gtk GUI apps using GStreamer should now handle the Gdk locks when
handling signals from the pipeline (as it should be)
Original commit message from CVS:
Code cleanup, make it adhere to the Gnome/Gtk+ code formatting, which is
quite clean and more readable.
Renamed parseavi to avidecoder
Implemented seeking/time display/pause/play/stop/clean exit to gstmediaplay
Added an element flag to indicate that it cannot deal with noncontigous
buffers. If such an element is found in the pipeline, seeking is disabled
for the complete stream (avidecoder cannot deal with seeking until we convert
it to a loop based element with pull_region to fetch the indeces etc...)
Original commit message from CVS:
Misc fixes and cleanups
A reworked gstplay. Now it is called gstmediaplay. gstplay is a custom
widget that can display media and is to become a bonobo component soon.
put the tables of the mpeg audio encoder in a header file.
maybe faster quantisation for the mpeg encoder.
Original commit message from CVS:
More Docs updates.
Added plugin documentation. I fear we need a gstdoc implementation
that loads plugins and does introspection on them. I think we should
automatically create the docs for the pads and mime types the plugins
provide. Does anyone have enough perl knowledge to add these features? I
allready changed the C code to output the pad definitions but my perl
knowledge is too limited, for now, to implement the rest of the needed
functionality...
Original commit message from CVS:
Added extra Gtk types for better editor properties (FILENAME, enums)
The disksrcs location property can now be set in the editor: we can
now make our first working mp3 player with gsteditor!!
Original commit message from CVS:
Fixed a severe timestamp bug in mpeg_play.
Create a new thread upon a multisrc element in autoplugging: the autoplug
example now correctly sets up 5 threads for an mpeg player.
Original commit message from CVS:
Fixed a bug in the typeloading.
Fixes to various elements so that correct types are returned.
Fixed flag collision with GtkObject.
Elements can now suggest a thread. not sure if this is the right way to
handle automatic thread creation.
Autoplugging now works with multiple sinks and thread setup. No threads
are created for intermediate elements yet, so MPEG may still be choppy.
Original commit message from CVS:
Loading and saving of XML pipeline descriptions.
GladeXML like operation implemented (you can retrieve parts of a pipeline)
Original commit message from CVS:
The cothreads were not initialized in its thread context resulting in
severe stack corruption. This was very hard to track down.
We should be able now to modify some plugins to a loop based setup so that
we can get rid of the mp3parse and mp1videoparse elements.
Modified the GList to a GSList in the queue.
Original commit message from CVS:
A first attempt to fix the queues in a cothreaded pipeline.
Some fixes to the thread handling.
Fix a bug in gstreamer-config : gthread was not included.
gst_bin_create_plan() is now done in the READY state.
a bin with only another bin in it will now work with gst_bin_iterate.
Added some examples for the queues.
Original commit message from CVS:
Patch from Chris Emerson for PPC. cothreads stuff was causing segfaults,
because the stack needs to be initialised with a particular structure on
PPC, unlike x86 where it simply has items pushed or popped.
Added Chris to AUTHORS, for his trouble.
Original commit message from CVS:
Added gst_pipeline_add_sink/src to allow multiple sink cases and
more complex autoplugging. Update docs too.
Simplified the pipeline autoplugging code.
Changed the cothread case: One iteration is now a push from the src
element. The disk source does not change its state anymore on eof.
Better type setting for the ac3 parser/decoder and mpeg2parse.
Original commit message from CVS:
Added check for vorbis.
Really compile with Xv extension if the library is found.
Changed the disksrc so that is goes to the PAUSE state if eof.
Original commit message from CVS:
Added a quick hack to allow loop based elements to finish intead of
looping till infinity.
Added compile time i386 or plain C getbits implementation selection.
The vorbis decoder now is our first loop based element!
Original commit message from CVS:
Fix on x86 architectures - was just checking for 386, D'oh.
Fix autogen to run autoheader _before_ autoconf, since it generates a
required file (config.h.in, which I just removed from CVS since its an
autogenerated one). Also fix messed up variables in gst/Makefile.am
Original commit message from CVS:
Begun to add support for non i386 architectures, with the help of Chris
Emerson. Added bits to configure to figure out architecture, and added
some alternatives for PPC. Should cause no problems on i386, but I'll
check after this commit, but doesn't yet work on PPC by a fair way...
Original commit message from CVS:
The vorbis encoder
A test program to convert an mp3 to a vorbis file.
A fix for the dynamic plugin loader factory_find was not working.
Don't forget to run gstreamer-register after installing the plugins!
Original commit message from CVS:
Due to popular demand :-), I added a vorbis decoder.
The encoder is not yet functional.
Small cosmetic changes to gstcpu.c
Beware:
You *need* to install libvorbis.a from the main vorbis CVS.
you *have* to change the line in libtool to
deplibs_check_method="pass_all"
because else the plugin shared library refuses to link against the
static libvorbis.a library. This is a hack. I have no intention in
including libvorbis into the gstreamer CVS tree and making it
libtool compatible.
Original commit message from CVS:
Added CPU detection.
Added SSE optimisation to mpeg_play
Modified the mpeg2 decoder and gstidct to use cpu detection.
Cleanups in the mpeg1 and mpeg2 decoders.
Original commit message from CVS:
Changes to gstreamer-config to include gtk+ libs
manual changes: queues, threads, programs
gsteditor does not crash anymore.
gstpipline new should return a GstElement *
fixed ac3dec for new getbits
fixes to gstreamer-launch
more efficient startup for gstplay.
Original commit message from CVS:
Faster and modular getbits implementation.
Fixed a bug in the audiosink that could lock up your box on bad MB.
Modified the plugins to use the new getbits functions.
Original commit message from CVS:
Externalized the plugin information in /etc/gstreamer/reg.xml
- no need to do a plugin_load_all()
- plugins are loaded when needed
- typedetect functions are loaded when needed (still buggy)
- no need to check for the existance of the plugin in the codecs
- added gstreamer-register to create the reg.xml file
- renamed launch to gstreamer-launch
- plugins need to register the typefactory they provide
modified the plugins to meet the new design
modified the plugins to correctly set their pad types
autoplugging can be done without loading the plugins now
Original commit message from CVS:
More docs (most of them just empty...)
Added automatic pad plugging.
Added automatic dynamic pad plugging.
Changed some codecs to correctly set their pad types.
Original commit message from CVS:
A first (rude) attempt at autoplug.
Autoplugging selects appropriate codecs to connect src to sink, adds
them to the pipeline and connect pads.
Autoplugging will run the typedetect plugin if the src pad has no MIME
type.
No autoplugging is done on the src and sink pads, it's hardcoded:
connect 'src to sink'.
No attempt at creating threads.
No attempt at dynamically autoplugging not yet existing pads.
Changes to (some) plugins to properly set their MIME types.
Original commit message from CVS:
Added the bufferpool handler.
This object is able to generate buffers and is passed between elements to
exchange buffers. Elements can also use this pool to efficiently generate
output buffers.
Original commit message from CVS:
Megapatch, changes which states are available, how they're used, and how
they're set. Also modifies the scheduling system, breaking pulled
buffers. Check mail archives for more details.
Original commit message from CVS:
Better error recovery in teh MPEG1 decoder.
Removed MPEG2 stuff from the MPEG1 decoderand make it ignore MPEG2
streams.
Some extra framerates in the MPEG2 decoder (from libmpeg2)
Round image size up to the nearest multiple of 16 (MPEG2 decoder)
Original commit message from CVS:
This is the audio/video sync release.
Changed the mpegvideoparser to parse complete pictures. Added the PTS
timestamps to the pictures.
Added PTS timestamps to the MPEG audio frames.
Made the clock a littlebit better.
Gstplay now uses two more threads one for video, one for audio playback.
Added the first QoS callbacks for the pads.
hopefully fix an mmx compilation problem.
Original commit message from CVS:
Added seeking to some of the plugins.
some MMX speedups in the MPEG decoders.
Better YUV to MMX conversion
implemented seeking to gstplay.
Original commit message from CVS:
Switched Cr and Cb in YUV2RGB
fixed the avi parser (gstriff had wrong size)
the system_encoder can now do video only system streams
sanitized the colorspace and scaler api.
added fast self modifying asm scaler.
worked on the typefind stuff.
added a cool media player (gstplay) it uses the typefinder to playback
avi, mpeg1 (system and video) streams. It is also a testbed and a real app.
Original commit message from CVS:
Reworked the audio resampler.
Added first frames skipping to the mp3 and ac3 parser (this allows you to
skip a number of frames before pushing them).
Removed wrong mmx code from the encoder.
Added Macroblocks per slice to the encoder.
added bilinear and bicubic interpolation to the videoscaler.
Added a MPEG1 multiplexer (no SCR or mux rate calculations yet)
Added some options (framerate to the encoder/decoder)
Original commit message from CVS:
Added a basic element: pipefilter. This element connects the stdin/stdout
of an external program into the pipeline. We now have virtually all of
the capabilities of sox, lame, ...
Original commit message from CVS:
Rewrote the mpeg1 system stream parser.
Fixed a motion compensation bug (was a bug on Intels site)
Faster/Better/more accurate motion detection in the mpeg encoder
Added an audio resampler (convert 48000 kHz to 44100 for example) does not work
Fixed rate control in the mpeg2 encoder.
Original commit message from CVS:
Fixed the mpeg2 encoder. Added MMX optimisations for motion estimation.
Added another mpeg1 encoder.
Added Two filters: smooth and median (a median cut filter)
Made the queue pull buffers.
Original commit message from CVS:
More changes to the mpeg parser and encoder.
It seems like the mpeg2 decoder does not like some MPEG1 videos. Use
mpeg_play for now (eg. AlienSong)
Original commit message from CVS:
Added MMX optimized yuv2rgb (AlienSong now plays back at only 6% CPU)
Added mpeg1 picture skipping and fixed a buffer overflow.
Added a system clock. The audiosink can now adjust the clock.
Fixed incorrect behaviour on 8, 15, 16, 24 and 32 bits displays.
Cleanup of the videosink, it now uses the color conversion library when
needed.
Original commit message from CVS:
Added a colorspace library and removed yuv2rgb conversion from the
videosink.
The aviencoder should now more or less work.
Original commit message from CVS:
Documentation updates. All standard library objects and standard
elements are documented. Modified some of the elements to more
accuratly report about their arguments so the documentation builds
more reasonable output.
Added aviencoder and jpegencoder elements (not working yet)
Original commit message from CVS:
Added jpeg decoder. the avi decoder now configures the jpeg codec if
an MJPG encoded avi is found. Fixed riff parsing. readded rgb_draw
functions to the videosink. jpeglib is used to decode the jpeg image.
Original commit message from CVS:
Added the riff code into a seperate library in /libs dir. The plugin
loader also checks the /libs path.
The avi parser now uses the riff library.
WAV is broken.
Original commit message from CVS:
Various cleanups. Moved the overlay code to a specialized widget.
One error I cannot seem to fix: switching desktops does not disable the
overlay.
Original commit message from CVS:
This is a rather large patch. Switched on -Wall compiler flag and fixed
the warnings.
Made the Video for Linux more like it should be.
Original commit message from CVS:
Added a pull function in the gstpad. Modified the videosink to implement
the pull. This function allows a source element to request a buffer
from the destination. This is much more efficient because the
videosink can then pass a buffer with SHM to the element, which does
not require an aditional memcpy.
removed scaling from the videosink. I need something better.
Original commit message from CVS:
Speedup in mpg123 parsing. speedup in mp1videoparse. rearanged the
MPEG player got rid of some memcpy. bit handling changes.
MMX code for the IDCT and motion compensation in mpeg_play.
Almost as fast as the commercial mpeg player mtv, but with a much
better video quality :-)
Original commit message from CVS:
The first functional video sink... Removed all of the video stuff
from the MPEG video decoder. Fixed a bug in smoothwave.
The MPEG video decoder still does the YUV->RGB conversion.
Original commit message from CVS:
Made queue default bigger.
fixed parsing errors in mp3parse mpeg1parse mp1videoparse: more than 2
zeros and a 1 is also a sync.
fixed MPEG1 video SKIP_PICTURE which caused a segfault. AlienSong now
plays as it should do. Skips are currently ignored and give some error
on the console, need to clean this up.
Original commit message from CVS:
Fixed the queue length (fixed length 5 for now).
fixed mpeg1 video rate control.
AlienSong segfaults sometimes. My other movies don't....
Original commit message from CVS:
Changed the way state is dealt with when a child is added to the bin. The
note states that the COMPLETE state should probably reflect nothing more
than whether or not there is a child in the bin, not whether or not all
children are COMPLETE. I need to write out a few scenarios for complex
pipeline manipulations to figure out how all the states should interact.
The idea is to maintain the ability to dynamically recofigure the
pipeline.
Original commit message from CVS:
Compile a test program to ensure that we have working atomic resource
counting.
A few small changes (include headers, fix a cast) to stop compiler
warnings.
