Hi all, it seems like the support for fftw got lost somewhere in the various revisions of configure.in. There's just the check for the library left but no real setting of a preprocessor definition or the likes. Has it ever been present in devel_0_38 at all? Tim, please fix!
best greetings, Thomas
sorry for the noise... i found the old code and checked it in...
Thomas Grill wrote:
Hi all, it seems like the support for fftw got lost somewhere in the various revisions of configure.in. There's just the check for the library left but no real setting of a preprocessor definition or the likes. Has it ever been present in devel_0_38 at all? Tim, please fix!
best greetings, Thomas
PD-dev mailing list PD-dev@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-dev
and one more (it's the last one, i promise): Tim please have a look at the fftw implementation. The pvoc example crashes with a segmentation fault when fftw is enabled.
good luck, Thomas
Thomas Grill wrote:
Hi all, it seems like the support for fftw got lost somewhere in the various revisions of configure.in. There's just the check for the library left but no real setting of a preprocessor definition or the likes. Has it ever been present in devel_0_38 at all? Tim, please fix!
best greetings, Thomas
PD-dev mailing list PD-dev@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-dev
Tim please have a look at the fftw implementation. The pvoc example crashes with a segmentation fault when fftw is enabled.
works fine for me ... the only problem, i'm aware of at the moment is a segfault related to the garray locks... (iirc, when closing a patch, possibly related to the gui changes from 37 to 38)
would it be possible, that you send be a gdb backtrace?
cheers ... tim
Tim please have a look at the fftw implementation. The pvoc example crashes with a segmentation fault when fftw is enabled.
hm ... that's really strange ... the segfault is related to the threaded soundfiler ... but only occurs if fftw is enabled ...
i'll have to think about that ... tim
hi thomas ...
Tim please have a look at the fftw implementation. The pvoc example crashes with a segmentation fault when fftw is enabled.
i looked into it ...
i'm not sure, what's the reason for the behaviour, but it seems to be the same problem that we had when trying to change the asio implementation ... we both tried different approaches to let the pd scheduler wait for the asio callback ... the segfault occured in the same location somewhere in the dsp chain ...
the question is ... why is the dsp chain corrupted ... that all is getting really strange ...
cheers ... tim
it seems like the support for fftw got lost somewhere in the various revisions of configure.in. There's just the check for the library left
for linux, i'm trying to maintain the autobuild system (configure.ac)
i'd suggest to use that ... tim
-
Thomas Grill -
Tim Blechmann