[MINC-users] nu_correct (volume_stats) assertion failed

Mon May 17 15:37:15 EDT 2010

Well whaddaya know, that was it!

Rebuilt EBTKS static only, rebuilt N3 against it and voilà!

Once again, thank you all for this quick and (almost) painless fix.

By the way, just so that I don't have to build everything again, what other
minc packages use ebtks?

Burt.

On Mon, May 17, 2010 at 14:42, Alexandre CARMEL-VEILLEUX <
acveilleux at mrs.mni.mcgill.ca> wrote:

> EBTKS is not built as a static library, I link directly to it. Maybe
> that's the problem? Because otherwise your environment looks like
> mine... I might have skipped a land mine by not dynamically linking
> EBTKS.
>
> Alex
>
> On Mon, May 17, 2010 at 02:36:05PM -0400, Burt Cr?peault wrote:
> > Date: Mon, 17 May 2010 14:36:05 -0400
> > From: Burt Cr?peault <burt.crepeault at crulrg.ulaval.ca>
> > Subject: Re: [MINC-users] nu_correct (volume_stats) assertion failed
> > To: MINC users mailing list <minc-users at bic.mni.mcgill.ca>
> >
> > Must be a case of bad fingers then because everything below appears very
> > similar.
> >
> > One difference I can see is that yours doesn't list EBTKS as a library of
> > volume_stats. Also, I built all the dependencies (libz, netcdf, hdf5)
> > directly in the build path with the same prefix directive. To allow
> > everything to work, I must set DYLD_LIBRARY_PATH and PERL5LIB as such:
> >
> > # echo $DYLD_LIBRARY_PATH
> > /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib
> >
> > # echo $PERL5LIB
> >
> /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib:/Network/MEDICS/EX/etc/minc-2.0.18/Darwin/Library/Perl/Updates/5.8.8:/Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/perl5/site_perl/5.8.8/darwin-thread-multi-2level
> >
> > # uname -a
> > Darwin beryllium.crulrg.local 9.8.0 Darwin Kernel Version 9.8.0: Wed Jul
> 15
> > 16:55:01 PDT 2009; root:xnu-1228.15.4~1/RELEASE_I386 i386
> >
> > # gcc --version
> > i686-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5465)
> >
> > # strings /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/libminc2.1.dylib
> |
> > grep '2\.0\.'
> > 2.0.18
> >
> > # otool -L volume_stats
> > volume_stats:
> >     /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/libvolume_io2.1.dylib
> > (compatibility version 3.0.0, current version 3.2.0)
> >     /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/libminc2.1.dylib
> > (compatibility version 3.0.0, current version 3.2.0)
> >     /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/libhdf5.0.dylib
> > (compatibility version 1.0.0, current version 1.0.0)
> >     /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/libz.1.dylib
> > (compatibility version 1.0.0, current version 1.2.5)
> >     /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/libnetcdf.4.dylib
> > (compatibility version 5.0.0, current version 5.0.0)
> >     /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current
> version
> > 111.0.0)
> >     /Network/MEDICS/EX/etc/minc-2.0.18/Darwin/lib/libEBTKS.0.dylib
> > (compatibility version 1.0.0, current version 1.0.0)
> >     /usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current
> version
> > 7.4.0)
> >     /usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current
> version
> > 1.0.0)
> >
> > # volume_stats -version
> > Assertion failed at line 32 in file templates/Pool.cc
> >
> > Burt.
> >
> >
> >
> >
> > On Mon, May 17, 2010 at 14:11, Alexandre CARMEL-VEILLEUX <
> > acveilleux at mrs.mni.mcgill.ca> wrote:
> >
> > > What's the exact platform you're building on?
> > >
> > > I'm using:
> > >
> > > uname -a:
> > > Darwin mrspc.bic.mni.mcgill.ca 9.8.0 Darwin Kernel Version 9.8.0: Wed
> Jul
> > > 15 16:55:01 PDT 2009; root:xnu-1228.15.4~1/RELEASE_I386 i386
> > >
> > > gcc:
> > > mrspc [/localscratch/compile/newtemp/bin] gcc --version
> > > i686-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5465)
> > >
> > > I've just built this tool line from scratch and it works:
> > >
> > > minc 2.0.18
> > > ebtks 1.6.2
> > > N3 1.11.0
> > >
> > >
> > > It's all very weird.
> > >
> > > Alex
> > >
> > > Extraneous output:
> > >
> > >
> > >
> > > mrspc [/localscratch/compile/newtemp/bin] strings
> > > /localscratch/compile/newtemp/lib/libminc2.1.dylib | grep '2\.0\.'
> > > 2.0.18
> > > mrspc [/localscratch/compile/newtemp/bin] otool -L volume_stats
> > > volume_stats:
> > >        /localscratch/compile/newtemp/lib/libvolume_io2.1.dylib
> > > (compatibility version 3.0.0, current version 3.2.0)
> > >        /localscratch/compile/newtemp/lib/libminc2.1.dylib
> (compatibility
> > > version 3.0.0, current version 3.2.0)
> > >        /usr/local/lib/libhdf5.0.dylib (compatibility version 1.0.0,
> current
> > > version 1.0.0)
> > >        /usr/lib/libz.1.dylib (compatibility version 1.0.0, current
> version
> > > 1.2.3)
> > >        /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current
> > > version 111.1.4)
> > >        /usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current
> > > version 7.4.0)
> > >        /usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current
> > > version 1.0.0)
> > > mrspc [/localscratch/compile/newtemp/bin] volume_stats -version
> > > The program <volume_stats> was built from:
> > > Package MNI N3, version 1.11.0, compiled by
> > > acveilleux at mrspc.bic.mni.mcgill.ca (i686-apple-darwin9.8.0) on
> 2010-05-17
> > > at 14:03:52
> > >
> > >
> > >
> > > On Mon, May 17, 2010 at 01:47:24PM -0400, Burt Cr?peault wrote:
> > > > Date: Mon, 17 May 2010 13:47:24 -0400
> > > > From: Burt Cr?peault <burt.crepeault at crulrg.ulaval.ca>
> > > > Subject: Re: [MINC-users] nu_correct (volume_stats) assertion failed
> > > > To: MINC users mailing list <minc-users at bic.mni.mcgill.ca>
> > > >
> > > > Ah, new builds, old libs... I used the exact same compiler flags, as
> per
> > > > your recommendations, so everything should work. Perhaps I should try
> > > > another build using 2.0.15 as a basis. Still, it'd be nice to know
> what
> > > the
> > > > problem is with 2.0.18.
> > > >
> > > > Burt.
> > > >
> > > >
> > > >
> > > >
> > > > On Mon, May 17, 2010 at 12:31, Alexandre CARMEL-VEILLEUX <
> > > > acveilleux at mrs.mni.mcgill.ca> wrote:
> > > >
> > > > > I've just tested a build of N3-1.11.0 with ebtks-1.6.2 on my build
> > > > > environment (Mac OS X 10.5.8, most of the libs as per the package I
> > > made
> > > > > last year, minc 2.0.15).
> > > > >
> > > > > 1.10.1 with 1.6.1 works for me
> > > > > 1.11.0 with 1.6.1 works for me
> > > > > 1.11.0 with 1.6.2 works for me
> > > > >
> > > > > Where works for me means: (a) -help works and (b) it prints out the
> > > > > stats of a random minc file I had on hand.
> > > > >
> > > > > I used this to build both tools:
> > > > >
> > > > > ./configure CFLAGS="-O2 -arch x86_64 -m64" CXXFLAGS="-O2 -arch
> x86_64
> > > -m64"
> > > > > \
> > > > > --prefix=//localscratch/compile/newtemp \
> > > > > --with-build-path=/localscratch/compile/newtemp:/usr/local/bic \
> > > > > --with-minc2
> > > > >
> > > > > /locascratch/compile/newtemp is where I built the new EBTKS.
> > > > >
> > > > >
> > > > > This is very annoying,
> > > > >
> > > > > Alex
> > > > >
> > > > > On Mon, May 17, 2010 at 12:12:02PM -0400, Burt Cr?peault wrote:
> > > > > > Date: Mon, 17 May 2010 12:12:02 -0400
> > > > > > From: Burt Cr?peault <burt.crepeault at crulrg.ulaval.ca>
> > > > > > Subject: Re: [MINC-users] nu_correct (volume_stats) assertion
> failed
> > > > > > To: MINC users mailing list <minc-users at bic.mni.mcgill.ca>
> > > > > >
> > > > > > I've put the original file here:
> > > > > >
> > > > >
> > >
> https://sodium.crulrg.ulaval.ca/WX/images.13671.bl.MR.T1.MP-RAGE.1034.1.10007.[mnc.rsh][mnc.dns].Br.80_p2.mnc.gz
> > > > > >
> > > > > > Andrew: yes, running a dry volume_stats command still fails.
> > > > > >
> > > > > > Burt.
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Mon, May 17, 2010 at 11:56, Andrew Janke <a.janke at gmail.com>
> > > wrote:
> > > > > >
> > > > > > > > Can't wait to get this OS X 64-bit build going... Man this is
> > > > > tough...
> > > > > > > > (sigh)
> > > > > > >
> > > > > > > Welcome to the EBTKS ornery beast! :)  Well actually that
> probably
> > > > > > > isn't fair on Alex, it's at least half C++'s fault for
> continually
> > > > > > > changing the spec!.  (but I digress...)
> > > > > > >
> > > > > > > > neon:Darwin:# volume_stats
> > > > > > > > Assertion failed at line 32 in file templates/Pool.cc
> > > > > > > >
> > > > > > > > I believe that Pool.cc file belongs to EBTKS, I'm using
> version
> > > > > 1.6.2. I
> > > > > > > > checked the output generated during its build, no apparent
> > > problem
> > > > > there.
> > > > > > > I
> > > > > > > > built it again, just to make sure, and did the same for N3.
> > > > > Everything
> > > > > > > went
> > > > > > > > well but the failed assertion error persists.
> > > > > > >
> > > > > > > > Any ideas?
> > > > > > >
> > > > > > > And this is just running volume_stats by itself?  ie: it was
> trying
> > > to
> > > > > > > produce the -help output?
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > Andrew Janke
> > > > > > > (a.janke at gmail.com || http://a.janke.googlepages.com/)
> > > > > > > Canberra->Australia    +61 (402) 700 883
> > > > > > > _______________________________________________
> > > > > > > MINC-users at bic.mni.mcgill.ca
> > > > > > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> > > > > > >
> > > > > > _______________________________________________
> > > > > > MINC-users at bic.mni.mcgill.ca
> > > > > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> > > > > _______________________________________________
> > > > > MINC-users at bic.mni.mcgill.ca
> > > > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> > > > >
> > > > _______________________________________________
> > > > MINC-users at bic.mni.mcgill.ca
> > > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> > > _______________________________________________
> > > MINC-users at bic.mni.mcgill.ca
> > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> > >
> > _______________________________________________
> > MINC-users at bic.mni.mcgill.ca
> > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>