Greetings,
I'm trying to compile Wine 0.9.44 as a 32-bit executable on an AMD
64-bit system (that is, with -m32, not in a chroot). I've done this
successfully in the past under 64Studio, but my recent crash & reinstall
has introduced something (or left out something) that leaves me unable
to build Wine. Here's the problem point :
gcc -c -I. -I. -I../../include -I../../include -D__WINESRC__
-D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing
-Wdeclaration-after-statement -Wwrite-strings -Wpointer-arith -g -O2
-D__i386__ -o interlocked.o interlocked.c
{standard input}: Assembler messages:
{standard input}:38: Error: suffix or operands invalid for `push'
{standard input}:39: Error: suffix or operands invalid for `push'
{standard input}:46: Error: suffix or operands invalid for `pop'
{standard input}:47: Error: suffix or operands invalid for `pop'
make[2]: *** [interlocked.o] Error 1
make[2]: Leaving directory `/home/dlphilp/src/wine-0.9.44/libs/port'
make[1]: *** [port] Error 2
make[1]: Leaving directory `/home/dlphilp/src/wine-0.9.44/libs'
make: *** [libs] Error 2
Here's the code in question :
__ASM_GLOBAL_FUNC(interlocked_cmpxchg64,
"push %ebx\n\t"
"push %esi\n\t"
"movl 12(%esp),%esi\n\t"
"movl 16(%esp),%ebx\n\t"
"movl 20(%esp),%ecx\n\t"
"movl 24(%esp),%eax\n\t"
"movl 28(%esp),%edx\n\t"
"lock; cmpxchg8b (%esi)\n\t"
"pop %esi\n\t"
"pop %ebx\n\t"
"ret");
And here's my GCC :
dlphilp@64studio:~/src/wine-0.9.44$ gcc --version
gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)
Maybe I need to use a different gcc version ? Whatever, there's nothing
on Google that clearly points to a solution, so I turn to the GLAD team
(Giants of Linux Audio Development) for guidance. Any suggestions or
advice that might move me towards resolving this difficulty ?
Best,
dp
Cedric GESTES wrote:
> try pushl/popl , it the 32bit version. but I dont know 64bit asm...
Thanks, Cedric, but I still got no joy. :(
In fact, I received exactly the same error message, i.e. referring to
push, not pushl as I'd expect. Any idea why that happened ?
Best,
dp
Hello,
my friends and I are developing a Computer Game and it should run on
Linux, Mac and Win. Now we're thinking about Audio Effects. Generell
reverb, compressor with sidechain (to compress the signal when speech is
active, so its clearly heard, no matters how the player has set his
audio balance. you know... like in Radio Stations).
We are using FMod4 and searching for a platform indipendent effect
system. Maybe there is LADSPA port to windows? Or have you other
suggestions?
greetings,
Nils
2007/9/6, Stefano D'Angelo <zanga.mail(a)gmail.com>:
>
> 2007/9/6, elthariel <elthariel(a)gmail.com>:
> > IMHO, i think a freshmeat-style site for linux audio plugins is just
> what we
> > need. it must say if the plugin work correctly from a technical point of
> > view. It may allows user to add tags on a plugins (like 'electro'
> > 'distortion' 'classical') and allows other user to rate the pertinence
> of
> > the tag.
>
> Hi all,
> what about setting up an Internet service to find, download and
> install plugins directly from audio applications?
> Maybe a library can be developed to do this, containing all the needed
> stuff to properly interface with the packaging system of each
> distro/OS (or in case there aren't any, download the source and tell
> the user to compile it) and maybe also downloading the packages
> directly from the repository of the given distro (or the official
> source tarballs).
> What do you think about it? Is it feasible?
> Just my 2 cents.
>
> Stefano
>
still IMHO, i think this is quite a good idea, the two service must refer to
the same data. It's a little bit in the way of ""libfreesound""
Hello everybody,
Finally, after what does seem like ages (mainly because there literally
has been ages) since the last release, here is finally what we have
decided to call MusE 0.9!
MusE is a multitrack midi/audio sequencer with support for internal and
external softsynths, in the same vein as popular commercial music
applications like Cubase, Logic, etc.
MusE has been in transition towards a redesign dubbed MusE 1.0 and there
wasn't supposed to be any more development in the old (0.7,0.8, now 0.9)
branch.
But, due to fresh developer interest and lots of angry users, we decided
to resurrect the old codebase :)
(I didn't mean the part about angry users, we love you all! :-) )
The MusE homepage is at:
http://www.muse-sequencer.org
The download is available at:
http://sourceforge.net/projects/lmuse/
Some major changes since 0.8.1:
- Working -old style- automation system, pretty much all parameters
should be automatable, audio and midi.
- Redesigned Solo/Mute system, works much better
- lots of "things" added
- lots of bugs fixed
(exchange quotations as needed)
Quite long winding excerpt from the ChangeLog:
* Fixed bug 1650953 - select and move segfault
* Fixed broken master list editor. Should cure bug 1687166
* Added auto part resizing to drum canvas
* Fixed intermittent crash when cutting a part.
* Fixed arranger part selection
* Fixed arranger part operations (move etc.) to use selected part.
* Fixed arranger keyboard left and right previous/next part selection.
* Fixed wave editor - Black selection bar intermittent drawing.
* Wave editor now allows you to open multiple selected wave parts
* Fixed intermittent arithmetic exception crash in wave viewer.
* Fixed bad wave drawing above certain zoom level in wave editor.
* fix for sustain pedal to work as suspected when looping.
* Feature request: 1650957 space bar should start/stop playback fixed
* Fixed bug 1645575 - Eraser tool causes segfault in drum editor
* Fix for bug #1074808 "NRPN - still 128 values?"
* Fixed some major issues with undo/redo.
* Prevented save and segfault upon load for routes with empty names.
* Add new soloing system.
* Enabled MIDI mixer strip solo buttons.
* Fixed loading of .med files, with plugins, created before muse-0.9pre1
* Fixed touch mode events storing, was not right.
* Sliders, knobs and labels respond to mousewheel and paging properly.
* Linked labels with sliders and knobs for paging and mouse wheeling
* Fixed integer controls (like Glame Bandpass 'stages').
* Fixed external plugin gui's to look like the built-in freeverb.
* solo is now united between midi and audio
* fixed issue with hanging notes for softsynths when stop is pressed
* fixed faulty off value for aux when set to max
* Fixed timing bug causing MusE to depend on jack buffer setting
* Added support for auto-scroll for all drag operations
* Adopted RT locking method from 1.0 branch.
* Fixed - DoubleLabel right click causes popup and runaway increment.
* Changing global tempo now updates things. (Like resizing wave tracks).
* Fixed ladspa plugin automation not working until plugin displayed.
* Select right item when stacked, for p-roll and arranger
* Apply effect rack before sending aux
* Auto assign internal softsynths upon creation
* New method for pasting parts, keep orig. spacing, fixes bug #1516294
* Muting of midi-tracks no longer mute input, bug #1092343
* Fixed issue with controllers when moving playhead, bug #1316542
* Cached peak files now updated when wave files are newer, bug #1469177
* Fixed bug with recording notes to existing part, bug #1552783
* asks when imported wave file has differing samplerate, bug #1428248
* Added rubberband scrolling in all applicable editors, bug #1057732
* Changed colours of audio-out and wave names to match icon colours
* Fixed aux sending
* Fix logarithmic feedback to plugin sliders
* Fixed multi-part undo for move,drag and copy in arranger, bug #1516298
* Fixed multiple note undo for move and drag in pianoroll
* Fixed cursor placement in List Editor upon delete events bug #1417718
* Fixed soloing. Also multiple soloing now allowed.
* Enabled solo buttons on the mixer strips.
* Fixed major issues with rack plugin moving, saving and loading.
* Fixed issue with multiple tempo changes and incorrect wave part length
* Changed plugin code to use double to stop certain errors
* Full LADSPA plugin automation.
* fluidsynth: fixed pitchbend events, fixes bug #1404212
* Editing volume and pan with keyboard on audio strip now works
* Fixed issue with audio record, fifo could overflow with small buffers
* fixed storing of files with relative path "under" the project path
Full ChangeLog:
http://lmuse.cvs.sourceforge.net/lmuse/muse/ChangeLog?revision=1.214.2.194
Please report bugs and/or feature requests to the tracker
(http://sourceforge.net/tracker/?group_id=93414).
Regards,
MusE Development team
Hi all,
we are developing Podcast support for Aqualung[1].
At present it is working well for RSS feeds, but we
would like to achieve compatibility with Atom feeds
as well. The problem is, we cannot seem to find
some good feeds we could test for compatibility;
everyone seems to provide their feeds in RSS.
So, can anyone provide some popular (or not)
feeds that use the Atom syndication format?
Thanks,
Tom
[1] http://aqualung.sf.net
Late one friday afternoon, a sound-engineer was sitting in his chair
admiring a brand new pair of flat frequency response monitors. He then
observed the following:
For two frequencies represented by sine-waves and an octave apart to
have the same relative loudness, the higher octave will need to have its
amplitude adjusted to half of that of the lower octave.[1]
Both frequencies will then force a membrane to travel the same distance
within a given timeframe - the higher will go half as far but twice as
often than the lower - and they will also both have the same speed or
steepness at the zero-crossing.
Surprisingly, the lower frequency consumes four times as much energy
than the higher[2], although it is apparently not doing any more actual
work.
Therefore, it is a better excersize to take a walk around the block,
rather than running around in small circles.
QED
cheers! // Jens M Andreasen
[1] Assuming that a saw wave has an even distribution
of harmonics and can be written as:
saw(t){ for(k = 1;k < inf;k++) rv += sin(k*t)/k; return rv * 2/M_PI}
[2] Ohms law ...
Ken Restivo <ken(a)restivo.org> writes:
> On Mon, Aug 27, 2007 at 02:05:55AM +0300, Nedko Arnaudov wrote:
>> a2jmidid is daemon for exposing legacy ALSA sequencer applications in
>> JACK MIDI system. It is based on jack-alsamidi-0.5 (jackd alsa seq midi
>> backend) by Dmitry Baikov. The main purpose is to ease usage of legacy,
>> not JACK-ified apps, in JACK MIDI enabled systems.
>>
>> It is tested, and works, here, with current SVN jack. ATM it may or may
>> not work with other variants.
>>
>> Planned features/improvements:
>> * One JACK client per ALSA sequencer client
>> * Improved build system (autotools) and support (better handling) for
>> other JACK variants (older MIDI API variants, jackdmp)
>> * More control on what ports to bridge (currently bridging is fixed to
>> non-hardware ports).
>> * Real daemonization with log file, init.d script, etc.
>>
>> If someone wants to contribute please, contact me, or send patches, or
>> request inclusion (Gna! a2jmidid project). As usual, packagers are more
>> than welcome too.
>>
>> Homepage with screenshots: http://home.gna.org/a2jmidid/
>>
>> Tarball download: http://download.gna.org/a2jmidid/
>>
>
> What is the difference between this application, and running jackd -X seq?
With current codebase, none (or minor, non-essential). Main purpose it
to handle hardware devices through raw midi in jackd and alsa seq apps
through a2jmidid simultaneously.
--
Nedko Arnaudov <GnuPG KeyID: DE1716B0>
