Dear Iwo,
the error you get is due to a change in MATLAB's setxor that took place
after the release of SPM8 you are using. A quick fix is to change
spm_check_installation.m line 380 so that it reads:
if isempty(ir) && isempty(il)
It should hopefully run successfully afterwards. Otherwise, what we
really want to see is the output of:
ls /home/login/spm8/toolbox/DARTEL/optimNn.*
Best regards,
Guillaume.
On 05/09/14 10:42, Iwo Bohr wrote:
> Hello John and Guillome,
>
>
>
> This is the what the script ‘spm_check_installation full’ returned in my
> case (incl. some errors):
>
>
>
> SPM is installed in: /home/login/spm8
>
> SPM version is SPM8 (5236, 04 Feb 13)
>
> SPM toolboxes: none
>
> MATLAB is installed in: /usr/local/MATLAB/R2013b
>
> MATLAB version is 8.2.0.701 (R2013b)
>
> MATLAB toolboxes: signal (v6.20) images (v8.3) stats (v8.3)
>
> Platform: GLNXA64 (maxsize=281474976710655)
>
> OS: Linux 3.9.10-100.fc17.x86_64 #1 SMP Sun Jul 14 01:31:27 UTC 2013 x86_64
>
> Java 1.7.0_11-b21 with Oracle Corporation Java HotSpot(TM) 64-Bit Server
> VM mixed mode
>
> Java support: jvm awt swing
>
> Monitor(s): [0 0 1434 810] (24bit)
>
> OpenGL version: 1.2 (1.5 Mesa 6.4.1)(Hardware)
>
> OpenGL renderer: Mesa project: www.mesa3d.org (Mesa GLX Indirect)
>
> MEX extension: mexa64
>
> C Compiler: GNU C ().
>
> C Compiler settings: gcc ('-O -DNDEBUG')
>
> C Source code permissions: dir rwxr-xr-x, file rw-r--r--
>
> ----------------------------------------------------------------------
>
> Directory : /home/login/spm8
>
> SVN Id does not match filename for file:
>
> /home/login/spm8/Zspm_defaults.m
>
> Directory : /home/login/spm8/external/fieldtrip/private
>
> SVN Id does not match filename for file:
>
> /home/login/spm8/external/fieldtrip/private/volumeedit.m
>
> Parsing local installation... ...done
>
> Downloading SPM information... ...done
>
> Parsing SPM information... ...done
>
> ----------------------------------------------------------------------
>
> Error using horzcat
>
> Dimensions of matrices being concatenated are not consistent.
>
>
>
> Error in spm_check_installation>compare_versions (line 380)
>
> if isempty([ir il])
>
>
>
> Error in spm_check_installation>check_full (line 364)
>
> compare_versions(l,r);
>
>
>
> Error in spm_check_installation (line 27)
>
> check_full;
>
>
>
> *From:*John Ashburner [mailto:[log in to unmask]]
> *Sent:* 04 September 2014 18:30
> *To:* Iwo Bohr
> *Cc:* [log in to unmask]
> *Subject:* Re: [SPM] DARTEL for GLNX64
>
>
>
> Is it possible that there is some form of installation problem.
> Guillaume suggests to me that you could try the following to make sure
> all the files have been extracted to the right place:
>
>
>
> spm_check_installation full
>
>
>
>
>
> On my SPM8 version (and in the official release) I see the following:
>
>
>
> john@ash:~/Documents/spm8$ ls toolbox/DARTEL/optimNn.*
>
> toolbox/DARTEL/optimNn.m toolbox/DARTEL/optimNn.mexmaci64
>
> toolbox/DARTEL/optimNn.mexa64 toolbox/DARTEL/optimNn.mexw32
>
> toolbox/DARTEL/optimNn.mexglx toolbox/DARTEL/optimNn.mexw64
>
> toolbox/DARTEL/optimNn.mexmaci
>
>
>
> Perhaps your version is missing one or more of these files.
>
> Best regards,
>
> -John
>
>
>
>
>
>
>
> On 4 September 2014 18:01, Iwo Bohr <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> It’s SPM8 .
>
> I can’t remember the exact error messages (should have noted it down, I
> know L) but it was something about uncompelled files…
>
> Well it’s easy to reproduce it for New Segment , because it happens at
> the very beginning of running it:
>
>
>
> “Running 'New Segment'
>
> Failed 'New Segment'
>
> Error using optimNn (line 68)
>
> Not compiled for GLNXA64 in MATLAB 8.2.0.701 (R2013b) (see make.m)
>
>
>
> In file "/home/login/spm8/toolbox/DARTEL/optimNn.m" (v1143), function
> "optimNn" at line 68.
>
> In file "/home/login/spm8/toolbox/Seg/spm_preproc8.m" (v4148), function
> "spm_preproc8" at line 818.
>
> In file "/home/login/spm8/toolbox/Seg/spm_preproc_run.m" (v4677),
> function "run_job" at line 106.
>
> In file "/home/login/spm8/toolbox/Seg/spm_preproc_run.m" (v4677),
> function "spm_preproc_run" at line 30.
>
> In file "/home/login/spm8/toolbox/Seg/tbx_cfg_preproc8.m" (v4337),
> function "spm_local_preproc_run" at line 363.”
>
>
>
> It was something similar while trying to run the batch with old Segment,
> but occurring at “run Dartel” component.
>
>
>
> Iwo
>
>
>
>
>
> *From:*John Ashburner [mailto:[log in to unmask]
> <mailto:[log in to unmask]>]
> *Sent:* 04 September 2014 17:48
> *To:* Iwo Bohr
> *Cc:* [log in to unmask] <mailto:[log in to unmask]>
> *Subject:* Re: [SPM] DARTEL for GLNX64
>
>
>
> Which version of SPM are you using? Older versions were developed
> before 64 bit systems were widely available. Also, what error messages
> were produced? These usually give some sort of clue about what's gone
> wrong.
>
>
>
> Best regards,
>
> -John
>
>
>
>
>
>
>
>
>
> On 4 September 2014 17:06, Iwo Bohr <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> Dear SPM developers,
>
> I've tried twice to run DARTEL batch (Seg+Dartel+MNI reg) on Linux64
> and twice I encountered an unpleasant surprise after a significant time
> of running it:
> using consecutively Old or New Segment as initial steps.
> In both cases it was due to the fact that some routines along the way
> weren't compiled for 64-bit system.
> For the Old Seg version it was more painful because the error occurred
> after two initial steps (seg+import) when everything was seemingly ok.
>
> Would it be possible to give an indication for the updates for these
> "troublesome" routines? Apparently according to the output of spm_update
> my download is ok.
> Also I think it would be helpful if some sanity check for the system
> would be introduced for out of date routines, it would save a lot of
> time spent on trying running out of date procedures. After all I
> downloaded my SPM not that long time ago (about 1 yr) when 32bit systems
> were already commonly in use...
>
> Many thanks,
> Iwo
>
> ---
> Iwo Bohr, PhD
> Research Associate,
> Newcastle Magnetic Resonance Centre
> Campus for Ageing and Vitality,
> Newcastle University,
> Newcastle upon Tyne, NE4 5PL, UK
> Tel. +44 191 208 1160 <tel:%2B44%20191%20208%201160>
> Webpage: http://www.ncl.ac.uk/magres/staff/profile/iwo.bohr
> http://scholar.google.co.uk/citations?user=-0AKmt4AAAAJ
>
>
> http://www.charliecroweappeal.com/
>
>
>
>
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|