Re: [LAU] yet another yoshimi bug fix, -> 0.022

Am Mittwoch 14 Oktober 2009 12:16:50 schrieb cal:
> Edgar Aichinger wrote:
> > Am Mittwoch 14 Oktober 2009 05:34:24 schrieb cal:
> >> cal wrote:
> >>> Jonathan E. Brickman wrote:
> >>> [ ... ]
> >>>
> >>>> Haven't found bugs in 0.022 yet, by the way, except the teeny one
> >>>> about patch renames not showing up in the main window.
> >>>
> >>> Just had a look at Zyn 2.4.0, and it behaves the same. The renamed
> >>> patch doesn't update in the main window until you click on the renamed
> >>> slot with READ checked. Instant update would be nice, but a bit tricky.
> >>
> >> Perhaps not so hard.
> >>
> >> 0.023 fix jack midi pitchbend, update master panel instrument label on
> >> bank slot rename
> >>
> >> <http://www.graggrag.com/?q=yoshimi>
> >>
> >> cheers.
> >> _______________________________________________
> >> Linux-audio-user mailing list
> >> Linux-audio-user@email-addr-hidden
> >> http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
> >
> > Since your first announcement of yoshimi here i've been downloading and
> > compiling several versions sucessfully, but haven't been able to run it.
>
> Umm, you're somewhat alone in this, which could suggest a local problem.
>
> > I get an immediate segmentation fault, now finally i found some time to
> > look into what's happening, via gdb. Here's the backtrace and output of
> > "l" (list) from gdb, i'll happily do whatever is needed to provide more
> > info as time permits:
> >
> > ed@email-addr-hidden:~/Documents/src/Synth/yoshimi-0.022> gdb yoshimi
>
> that doesn't seem to lead anywhere useful.

Too bad :(

> [ ... ]
>
> > Starting program: /home/ed/Documents/src/Synth/yoshimi-0.022/yoshimi
> > [Thread debugging using libthread_db enabled]
> >
> > Program received signal SIGSEGV, Segmentation fault.
> > 0x0813372b in PresetsStore::PresetsStore() ()
> > Current language: auto; currently asm
> > (gdb) bt
> > #0 0x0813372b in PresetsStore::PresetsStore() ()
> > #1 0x081337ef in global constructors keyed to presetsstore ()
> > #2 0x0815760d in __do_global_ctors_aux ()
> > #3 0x0807a53c in _init ()
> > #4 0x081574c9 in __libc_csu_init (argc=1, argv=0xafffed14,
> > envp=0xafffed1c) at elf-init.c:79
> > #5 0xa790c69c in __libc_start_main () from /lib/libc.so.6
> > #6 0x0807ba41 in _start () at ../sysdeps/i386/elf/start.S:119
> > (gdb) l
> > 1 /tmp/ccr5yeuc.s: No such file or directory.
> > in /tmp/ccr5yeuc.s
> > (gdb)
>
> That's a bit low on entertainment, and information.

Yeah that's why i asked here, I don't really know how to supply more useful
information.

>
> > I should add that i modified src/Params/PresetsStore.cpp a bit (without
> > understanding much of what i was doing, I don't know C++ well) - here's
> > the diff, but it didn't make a difference in gdb output anyway... :
> >
> > --- src/Params/PresetsStore.cpp.orig 2009-10-13 11:56:30.000000000
> > +0200 +++ src/Params/PresetsStore.cpp 2009-10-14 11:49:44.000000000 +0200
> > @@ -30,9 +30,9 @@
> > #include "Params/PresetsStore.h"
> >
> > PresetsStore presetsstore;
> > +string preset_extension = ".xpz";
> >
> > -PresetsStore::PresetsStore() :
> > - preset_extension(".xpz")
> > +PresetsStore::PresetsStore()
> > {
> > clipboard.data = NULL;
> > clipboard.type.clear();
>
> if you say so.
>
> What are your build options - anything obvious like building with X86_64
> flags on 32bit? I dunno.
>

Nah, while experimenting i tried to alter the 32bit flags with ccmake (using
-march=i686, -O2 or removing the -msse2 because i wasn't sure if my AthlonXP
CPU supports it) which led to even more problems. So i think they are in
effect, and I stayed with the flags ccmake determined by itself, "-O3 -
march=native -msse -msse2 -mfpmath=sse -ffast-math -fomit-frame-pointer"

Hm. Rather clueless now. Pretty weird...

Edgar
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
Received on Wed Oct 14 16:15:02 2009