Dear Opher,
could you try to install SPM8b as well and see if you observe the same
error message? The Linux 64bit MEX files provided with SPM8b have been
compiled on a different system than those in SPM5.
Also, when you compiled the MEX files yourself, did you also type "make
install" after the compilation?
All the best,
Guillaume.
John Ashburner wrote:
> If you also installed the updates and have a suitable version of MATLAB
> installed that is compatible with your version of Linux (see MathWorks web
> pages), then I'm all out of ideas. I still have an old computer, and have no
> experience of compiling for 64 bit.
>
> Best regards,
> -John
>
> On Thursday 08 January 2009 09:58, Opher Lubzens wrote:
>> I reinstalled SPM5 and remade it using the Makefile included, and I still
>> get the same error and output, including for the template T1.nii file.Had
>> anybody here had SPM show such problems with Linux64?
>>
>> Opher Lubzens
>>
>> --- On Tue, 1/6/09, John Ashburner <[log in to unmask]> wrote:
>>> From: John Ashburner <[log in to unmask]>
>>> Subject: Re: [SPM] mat2file DICOM conversion error
>>> To: [log in to unmask]
>>> Date: Tuesday, January 6, 2009, 6:28 PM
>>> I think Kathy could be on to something here. It is possible
>>> that re-installing
>>> SPM could help, because this segmentation fault should not
>>> normally occur.
>>> Try displaying one of the image files released with SPM.
>>> If this does not
>>> work, then an installation problem could be a likely cause.
>>> Another
>>> possibility is if your data are stored on some unusual sort
>>> of filesystem.
>>>
>>> Best regards,
>>> -John
>>>
>>> On Tuesday 06 January 2009 15:34, Kathy Pearson wrote:
>>>> Opher,
>>>>
>>>> I once had a similar problem when placing the SPM5
>>> distribution code into
>>>
>>>> CVS on Linux without first tagging the binaries as
>>> such. This caused
>>>
>>>> inappropriate conversion of the binary equivalent
>>> Carriage Return and Line
>>>
>>>> Feed text characters, leading to a crash. The same
>>> kind of thing might
>>>
>>>> happen if you've unzipped the distribution with
>>> WinZip using the "Smart
>>>
>>>> Tar" option to automatically translate what might
>>> be text files with UNIX
>>>
>>>> LF to Windows CR/LF.
>>>>
>>>> Kathy Pearson
>>>> UAB Psychology
>>>>
>>>> -----Original Message-----
>>>> From: SPM (Statistical Parametric Mapping)
>>> [mailto:[log in to unmask]] On
>>>
>>>> Behalf Of Opher Lubzens
>>>> Sent: Tuesday, January 06, 2009 7:48 AM
>>>> To: [log in to unmask]
>>>> Subject: Re: [SPM] mat2file DICOM conversion error
>>>>
>>>> Correction: I was using the wrong file-a foo.img from
>>> a previous attempt at
>>>
>>>> conversion to ANALYZE rather then a regular foo.nii
>>> file.When I use the
>>>
>>>> foo.nii with these commands I get this result:
>>>>
>>>> N =
>>>>
>>>> NIFTI object: 1-by-1
>>>> dat: [181x217x181 file_array]
>>>> mat: [4x4 double]
>>>> mat_intent: 'Aligned'
>>>> mat0: [4x4 double]
>>>> mat0_intent: 'Aligned'
>>>> timing: [1x1 struct]
>>>> descrip: 'AVERAGE:'
>>>> cal: [0 32767]
>>>>
>>>>
>>>>
>>>> ans =
>>>>
>>>> fname: 't1.nii'
>>>> dim: [181 217 181]
>>>> dtype: 'INT16-LE'
>>>> offset: 352
>>>> scl_slope: 1
>>>> scl_inter: 0
>>> ------------------------------------------------------------------------
>>>
>>>> Segmentation violation detected at Tue Jan 6
>>> 15:34:35 2009
>>>
>>> ------------------------------------------------------------------------
>>>
>>>> Configuration:
>>>> MATLAB Version: 7.5.0.338 (R2007b)
>>>> MATLAB License: 567156
>>>> Operating System: Linux 2.6.22.5-31-default #1 SMP
>>> 2007/09/21 22:29:00
>>>
>>>> UTC x86_64
>>>> GNU C Library: 2.6.1 stable
>>>> Window System: The X.Org Foundation (70200000),
>>> display :0.0
>>>
>>>> Current Visual: 0x23 (class 4, depth 16)
>>>> Processor ID: x86 Family 6 Model 7 Stepping 7,
>>> GenuineIntel
>>>
>>>> Virtual Machine: Java 1.6.0 with Sun Microsystems
>>> Inc. Java HotSpot(TM)
>>>
>>>> 64-Bit Server VM mixed mode
>>>> Default Charset: UTF-8
>>>>
>>>> Register State:
>>>> rax = 0000000000000000 rbx = 0000000000000001
>>>> rcx = 000000000634b4b0 rdx = 00002aaad032c160
>>>> rbp = 00000000407f76d0 rsi = 00000000407f83d4
>>>> rdi = 0000000000000000 rsp = 00000000407f7688
>>>> r8 = 0000000000000001 r9 = 0000000000000000
>>>> r10 = 00002aaad032c160 r11 = 000000000633e090
>>>> r12 = 0000000000000001 r13 = 00000000407f77d0
>>>> r14 = 00000000407f7bd0 r15 = 0000000000000004
>>>> rip = 00002aaacb37c55d flg = 0000000000010246
>>>>
>>>> Stack Trace:
>>>> [0] file2mat.mexa64:0x00002aaacb37c55d(0x407f7700,
>>> 0, 0x2aaad032c160,
>>>
>>>> 0x407f83d4)
>>>> [1] file2mat.mexa64:0x00002aaacb37c5e2(0x10e43879f,
>>> 0, 0x2aaad032c160,
>>>
>>>> 0x407f83d4)
>>>>
>>>>
>>>>
>>>>
>>>> Thank you for taking taking time to help me so
>>> closely.
>>>
>>>> Opher Lubzens
>>>>
>>>> --- On Tue, 1/6/09, John Ashburner
>>> <[log in to unmask]> wrote:
>>>>> From: John Ashburner
>>> <[log in to unmask]>
>>>
>>>>> Subject: Re: [SPM] mat2file DICOM conversion
>>> error
>>>
>>>>> To: [log in to unmask]
>>>>> Date: Tuesday, January 6, 2009, 2:22 PM
>>>>> If SPM won't convert your DICOM files, then
>>> maybe it is
>>>
>>>>> worth trying to figure
>>>>> out what is wrong with the foo.img file. Try
>>> pasting this
>>>
>>>>> into MATLAB and
>>>>> reporting back the results:
>>>>>
>>>>> P='foo.img';
>>>>> N=nifti(P)
>>>>> N.dat
>>>>> N.dat(1,1,1)
>>>>> t=dir(P)
>>>>>
>>>>> Best regards,
>>>>> -John
>>>>>
>>>>> On Tuesday 06 January 2009 12:12, Opher Lubzens
>>> wrote:
>>>>>> This error message is all the data I get.At
>>> first I
>>>
>>>>> thought it was because
>>>>>
>>>>>> the indices were int16 and the data was
>>> int32, but
>>>
>>>>> when I change them to
>>>>>
>>>>>> match I get an "Incompatible class
>>> types"
>>>
>>>>> error.And when I try to display
>>>>>
>>>>>> the dcm2nii nifti I get a "foo.img
>>> cannot be
>>>
>>>>> resampled" warning, followed
>>>>>
>>>>>> by a "wrong handle dimentions"
>>> error if I
>>>
>>>>> click anywhere in the view
>>>>>
>>>>>> windows.
>>>>>>
>>>>>> Opher Lubzens
>>>>>>
>>>>>> --- On Mon, 1/5/09, John Ashburner
>>>>> <[log in to unmask]> wrote:
>>>>>>> From: John Ashburner
>>>>> <[log in to unmask]>
>>>>>
>>>>>>> Subject: Re: [SPM] mat2file DICOM
>>> conversion
>>>
>>>>> error
>>>>>
>>>>>>> To: [log in to unmask]
>>>>>>> Date: Monday, January 5, 2009, 8:11 PM
>>>>>>> Was there anything more to the error
>>> message?
>>>
>>>>> Also, what
>>>>>
>>>>>>> sort of errors does
>>>>>>> SPM produce when you try to use NIfTI
>>> files
>>>
>>>>> generated by
>>>>>
>>>>>>> dcm2nii ?
>>>>>>>
>>>>>>> Best regards,
>>>>>>> -John
>>>>>>>
>>>>>>> On Sunday 04 January 2009 11:19, Opher
>>> Lubzens
>>>
>>>>> wrote:
>>>>>>>> Hello, I've been trying to
>>> import an
>>>
>>>>> inplane
>>>>>
>>>>>>> anatomy that I have in DICOM
>>>>>>>
>>>>>>>> format into nifti using the SPM5
>>> native
>>>
>>>>> DICOM import
>>>>>
>>>>>>> function.
>>>>>>>
>>>>>>>> Unfortunately I get a
>>> "Subscripted
>>>
>>>>> assignment
>>>>>
>>>>>>> dimension mismatch (2)" error
>>>>>>>
>>>>>>>> from mat2file, and I can't
>>> understand
>>>
>>>>> the
>>>>>
>>>>>>> mat2file.c code, since my copy
>>>>>>>
>>>>>>>> doesn't include
>>> documentation.Does
>>>
>>>>> anyone know of
>>>>>
>>>>>>> solution to this problem?
>>>>>>>
>>>>>>>> Opher Lubzens
>>>>>>>>
>>>>>>>> PS.I've tried using dcm2nii
>>> but SPM was
>>>
>>>>> unable to
>>>>>
>>>>>>> read the resulting nifti.
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|