similar to: State of Vorbis: Monty's current offline development

I'm running into memory leak and read-from/write-to unallocated errors in the cleanup phase of parallel oggenc. I see in vorbis_block_clear() that it references some fields on the vorbis_dsp_state that it cached during vorbis_block_init(), and conditionally does some cleanup based on the values of those fields. Does this fact effectively mean that you can't have multiple vorbis_block

A variation on questions that I've asked before... In working on the parallel version of oggenc (both threaded and MPI), a profiling run shows that the function vorbis_analysis() takes up the majority of the run time. This seems to be an obvious choice for parallelization -- send each vorbis_block to a different processor, and let them call vorbis_analsis() in parallel with each other.

Per Monty's suggestions from a while ago, I have [finally] gotten around to playing with different schemes for parallel oggenc. Monty's main suggestion was to have a single thread loop over reading samples and calling vorbis_analysis_blockout(), and then queueing up the resulting blocks to be processed through vorbis_analysis() in other threads (in parallel). To verify that this works, I

Monty -- I'm still running into problems with my threaded encoded, and I think I just figured out why. I have N threads running in parallel calling vorbis_analysis_blockout() and vorbis_analysis() to do the number crunching on the input samples. They all share a single vorbis_dsp_state -- my understanding was that this was ok; they only *read* from the vorbis_dsp_state, therefore not

I read some lines in libvorbis/include/vorbis/codec.h: /* vorbis_block is a single block of data to be processed as part of the analysis/synthesis stream; it belongs to a specific logical bitstream, but is independant from other vorbis_blocks belonging to that logical bitstream. *************************************************/ My question. Does this imply that vorbis_analysis

I apologize for bugging you all again, but I've been trying to wrap my head around the encoding process and I'm about to have an aneurysm :P I've gone through oggenc.c, encode.c, and audio.c and I think I have a basic idea of what happens. I got it down in pseudocode as follows: --- 1 get parameters 2 init vorbis_comment -- vorbis_comment_init() 3 init vorbis_info --

Hello all Can anyone point me to some form of description for the fields in the following structures : 1/ vorbis_dsp_state 2/ vorbis_block Thank you. regards ~previr _____________________________________________________________ Previr Rangroo STMicroelectronics Associate Systems Lab Engineer Plot No. 2 & 3, Audio Competence Center Sector 16A,

Hello all Can anyone point me to some form of description for the fields in the following structures : 1/ vorbis_dsp_state 2/ vorbis_block Thank you. regards ~previr _____________________________________________________________ Previr Rangroo STMicroelectronics Associate Systems Lab Engineer Plot No. 2 & 3, Audio Competence Center Sector 16A,

Using libvorbis (1.3.5) I wish to extract the raw vorbis packets.  I've built some simple code on the excellent libvorbis API overview on the xiph.org site, but the example relies on the ogg_packet struct for final output and input to decoder, and shows now examples on how to do without it. Taking a look at the vorbis_bitstream_flush() function, which in the overview is the last step before

Hi All, I am trying to build an Ogg/Vorbis plugin for RealOne Player through the HelixCommunity source. During runtime, however, I saw an access violation. I did some debugging and narrowed it down to the following function in the Vorbis source.. int vorbis_synthesis(vorbis_block *vb,ogg_packet *op){ vorbis_dsp_state *vd=vb->vd; backend_lookup_state *b=vd->backend_state;

I'm having trouble tracking down why it leaks, but below is an example program which shows--using valgrind--that vorbis_info_clear() leaks memory if called before vorbis_dsp_clear(), but not if called after vorbis_dsp_clear(). Just compile and run under valgrind, using the -l switch to the example program to trigger a leak. Tested under OS X 10.7 and Ubuntu 12.04. This may be by design, or

Hi folks, I'm doing a merge of my current branch onto the mainline (for testing) today. I believe it to be stable. Just a little more vorbisfile testing. After the merge, I have a few more patches to apply, then onto cascading/coupling. New stuff: Floor backend 1 and residue backend 1; both are present, but the mainline modes won't use either yet. Naturally, both are enabled for

> > Which part? > > Well, my biggest problem is dealing with files. As you have mentioned > that fill_buffer() is obsolete, what has replaced it? ogg_sync_buffer() > didn't seem to be what I was looking for, as far as I can tell... am I > headed in completely the wrong direction? Ah, OK, I understand where you're headed now. The libvorbis API is different than

I'm not sure that this is not my fault, but here it goes: I'm trying to encode a 22 kHz Mono stream. I started with the encoder example and made it take dynamic channels etc. This is how I init it: vorbis_info_init(&VorbisInfo); vorbis_encode_init(&VorbisInfo, 1, 22050, -1, 128000, -1); then I get a crash in: _vp_quantize_couple(vorbis_look_psy * 0x01e6a6ec,

Hi all, I tend to compile my code with all GCC warnings turned on. However, when I do this : vorbis_comment_add_tag(&vdata->vc,"ENCODER","libsndfile"); I get the following warning messages: warning: passing argument 2 of 'vorbis_comment_add_tag' discards qualifiers from pointer target type warning: passing argument 3 of

Hi, This mozilla bug report is a crash triggered by a Vorbis file with corrupt headers: https://bugzilla.mozilla.org/show_bug.cgi?id=481601 The patch to fix the crash adds a new vorbis_dsp_init() function to libvorbis, and calls that from fs_vorbis_init() in libfishsound: https://bug481601.bugzilla.mozilla.org/attachment.cgi?id=366150 The public function it adds is: void

you probably don't want the ~2% perf improvement below... but i figured i'd post it anyhow :) well... i do suggest the use of fabsf() rather than fabs() -- gcc doesn't seem smart enough to figure out fabs() when you ask it to compile "-mfpmath=sse". but the #else part uses properties of ieee-754 float layout which you may or may not be interested in (ab)using. basically

Okay, how do I drop a changeset/patchset/tag for you folks from SVN? At this point, I have written three examples of how to use the basics of the ogg streaming and decoding in Tremor. I heartily welcome any suggestions, improvements and corrections that you can point out in the code. The examples required me to make some small modifications to the main tremor library. However, the changes

I am trying to create a Graphical User Interface for a Vorbis Player under MS Visual C++ 6.0 I was initially trting to use the code from decoder_example.c but I keep getting linker errors. I have set up my project as told to by James Boer on this mailing list. This is as follows: This document was prepared from an email received from James Boer on the Ogg Vorbis mailing list. It gives a

Hi there. I'm getting this exception, Floating point division by zero, when encoding at, for example, 22050 hz. This doesn't happens in a specific song or audio, it mostly happens after several calls to vorbis_analysis. This also doensn't happens, at least so far, at 11025, 8820 or 44100. Also, it doesn't seem to happen at low qualities (0, -0.1). The exception is generated in