Abdel,
Indeed mris_fix_topology had the same bug as mris_anatomical_stats (when running on Slackware Linux). I've posted a fixed version of mris_fix_topology here:
http://surfer.nmr.mgh.harvard.edu/pub/dist/
named:
mris_fix_topology.gz
mv this to your $FREESURFER_HOME/bin, then type:
gunzip mris_fix_topology.gz chmod a+x mris_fix_topology
Let me know if you encounter similar bugs. This fixed binary will appear in the next release.
Nick
On Fri, 2007-01-05 at 14:43 +0000, Abdel Douiri wrote:
Hi Nick We have done a new update to our Linux system and i have found the same error that i report a while ago (see bellow) But this time is for mris_fix_topology program. Can you please check it again for mris_fix_topology.c file, the error that you found was problem with sprintf! (sprintf(fname,"%s.mgz",fname) to strcat(fname,".mgz") ). Thanks Abdel
Abdel,
Thanks for testing this. The bug fix will appear in an upcoming release, but you can continue using the mris_anatomical_stats that you have now.
For those interested in the perils of testing a multi-platform app, the buggy line of code was:
sprintf(fname,"%s.mgz",fname)
Here the source and destination are the same, which, according to glibc docs for sprintf, is undefined behavior. The odd thing is that these Linux variants: RedHat 7.3, RedHat 9, Centos4, Debian and SuSE; and Mac OS X, all have a shared libc lib that handles the line of code properly. The libc included with Slackware, however, did not (and understandably so). libc.so is included not with Freesurfer, but with the OS, so this means that testing even on a half-dozen different platforms doesn't guarantee trouble-free performance on a seemingly identical platform.
The proper code in this case is:
strcat(fname,".mgz");
N.
On Fri, 2006-08-18 at 18:40 +0100, A. Douiri wrote:
that s works! Cheers!
At 01:51 18/08/2006, Nick Schmansky wrote:
Abdel,
I don't think there is a matlab alternative to mris_anatomical_stats.
However, there is a suspicious line of code in mris_anatomical_stats that I have fixed, and I'm wondering if you can give a new binary a try. It is found here:
http://surfer.nmr.mgh.harvard.edu/pub/dist/
and named:
mris_anatomical_stats.gz
To install, after downloading:
gunzip mris_anatomical_stats.gz chmod o+x mris_anatomical_stats mv mris_anatomical_stats $FREESURFER_HOME/bin
While I cannot recreate the error you see, I do not have a Slackware installation, and I am wondering if its core libs are revealing a bug in our code.
Let me know if it works.
Thanks,
Nick
Hi Nick, Thanks in advance for your help and i would like to ask is there an alternative in matlab for example to do mris_anatomical_stats? The detail of the outjkputs is:
the output of: $ mris_anatomical_stats --all-info ProgramName: mris_anatomical_stats ProgramArguments: --all-info ProgramVersion: $Name: stable3 $ TimeStamp: 06/08/17-10:28:55-GMT CVS: $Id: mris_anatomical_stats.c,v 1.35.2.1 2006/04/21 19:45:19 nicks Exp $ User: douiri Machine: img-179 Platform: Linux PlatformVersion: 2.6.16.22 CompilerName: GCC CompilerVersion: 30400
I tried with other subjects and also bert and i get the same problem:
$ mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats -b -a ../label/lh.aparc.annot -c ../stats/aparc.annot.ctab bert lh INFO: assuming MGZ format for volumes. computing statistics for each annotation in ../label/lh.aparc.annot. reading volume .mgz... zcat: /home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz.gz: No such file or directory mghRead(/home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz, -1): read error mris_anatomical_stats: could not read input volume .mgz
the freesurfer version $ cat $FREESURFER_HOME/build-stamp.txt freesurfer-Linux-centos4-stable-pub-v3.0.3
and the linux version is: $ cat /etc/slackware-version Slackware 10.2.0
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Nick Schmansky Sent: 17 August 2006 00:06 To: Abdel Douiri Cc: Freesurfer Mailing List Subject: RE: [Freesurfer] problem with mris_anatomical_stats
Abdel,
This problem is truly strange. The filename seems to just disappear.
Can you send me the output of:
mris_anatomical_stats --all-info
Also, does this error repeat? That is, if you try:
cd $SUBJECTS_DIR/01_00/scripts mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats \ -b -a ../label/lh.aparc.annot \ -c ../stats/aparc.annot.ctab 01_00 lh
does it fail again? And does this happen for other subjects?
Also, which freesurfer version is being used? To find out, type:
cat $FREESURFER_HOME/build-stamp.txt
And which Linux OS is it installed upon?
Nick
On Wed, 2006-08-16 at 17:16 +0100, Abdel Douiri wrote:
Thanks Bruce I have put the subject and that is "01_00" and the first and second autorecon was OK only the autorecon3 where it was the problem. Attached
the
log file. Thanks a lot for your help Abdel -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Bruce Fischl Sent: 16 August 2006 14:01 To: Abdel Douiri Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] problem with mris_anatomical_stats
Hi Abdel,
is that the full command line? If so, you need to specify a subject with -subjid. If not, could you send us the full command line and
recon-all.log?
cheers, Bruce
On Wed, 16 Aug 2006, Abdel Douiri wrote:
Hi all,
I will be grateful if someone can tell me what going wrong in the
following
problem:
When running "recon-all -autorecon3" the program stop with error
exactly in
the "mris_anatomical_stats" cmd stage and I have the following error:
mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats -b -a ../label/lh.aparc.annot -c ../stats/aparc.annot.ctab 01_00 lh
computing statistics for each annotation in ../label/lh.aparc.annot.
reading volume .mgz...
zcat:
/home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz.gz: No
such file or directory
mghRead(/home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz,
-1):
read error
mris_anatomical_stats: could not read input volume .mgz
INFO: assuming MGZ format for volumes.
Linux img-179 2.6.16.22 #2 SMP Mon Jun 26 15:49:31 BST 2006 i686
unknown
unknown GNU/Linux
recon-all exited with ERRORS at Tue Aug 15 20:52:49 BST 2006
Thanks
A
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Nick Thanks mris_fix_topology works now! But i have a similar bug a step later with: mris_make_surfaces Can you please check it. Thans for your help A
Nick Schmansky wrote:
Abdel,
Indeed mris_fix_topology had the same bug as mris_anatomical_stats (when running on Slackware Linux). I've posted a fixed version of mris_fix_topology here:
http://surfer.nmr.mgh.harvard.edu/pub/dist/
named:
mris_fix_topology.gz
mv this to your $FREESURFER_HOME/bin, then type:
gunzip mris_fix_topology.gz chmod a+x mris_fix_topology
Let me know if you encounter similar bugs. This fixed binary will appear in the next release.
Nick
On Fri, 2007-01-05 at 14:43 +0000, Abdel Douiri wrote:
Hi Nick We have done a new update to our Linux system and i have found the same error that i report a while ago (see bellow) But this time is for mris_fix_topology program. Can you please check it again for mris_fix_topology.c file, the error that you found was problem with sprintf! (sprintf(fname,"%s.mgz",fname) to strcat(fname,".mgz") ). Thanks Abdel
Abdel,
Thanks for testing this. The bug fix will appear in an upcoming release, but you can continue using the mris_anatomical_stats that you have now.
For those interested in the perils of testing a multi-platform app, the buggy line of code was:
sprintf(fname,"%s.mgz",fname)
Here the source and destination are the same, which, according to glibc docs for sprintf, is undefined behavior. The odd thing is that these Linux variants: RedHat 7.3, RedHat 9, Centos4, Debian and SuSE; and Mac OS X, all have a shared libc lib that handles the line of code properly. The libc included with Slackware, however, did not (and understandably so). libc.so is included not with Freesurfer, but with the OS, so this means that testing even on a half-dozen different platforms doesn't guarantee trouble-free performance on a seemingly identical platform.
The proper code in this case is:
strcat(fname,".mgz");
N.
On Fri, 2006-08-18 at 18:40 +0100, A. Douiri wrote:
that s works! Cheers!
At 01:51 18/08/2006, Nick Schmansky wrote:
Abdel,
I don't think there is a matlab alternative to mris_anatomical_stats.
However, there is a suspicious line of code in mris_anatomical_stats that I have fixed, and I'm wondering if you can give a new binary a try. It is found here:
http://surfer.nmr.mgh.harvard.edu/pub/dist/
and named:
mris_anatomical_stats.gz
To install, after downloading:
gunzip mris_anatomical_stats.gz chmod o+x mris_anatomical_stats mv mris_anatomical_stats $FREESURFER_HOME/bin
While I cannot recreate the error you see, I do not have a Slackware installation, and I am wondering if its core libs are revealing a bug in our code.
Let me know if it works.
Thanks,
Nick
Hi Nick, Thanks in advance for your help and i would like to ask is there an alternative in matlab for example to do mris_anatomical_stats? The detail of the outjkputs is:
the output of: $ mris_anatomical_stats --all-info ProgramName: mris_anatomical_stats ProgramArguments: --all-info ProgramVersion: $Name: stable3 $ TimeStamp: 06/08/17-10:28:55-GMT CVS: $Id: mris_anatomical_stats.c,v 1.35.2.1 2006/04/21 19:45:19 nicks Exp $ User: douiri Machine: img-179 Platform: Linux PlatformVersion: 2.6.16.22 CompilerName: GCC CompilerVersion: 30400
I tried with other subjects and also bert and i get the same problem:
$ mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats -b -a ../label/lh.aparc.annot -c ../stats/aparc.annot.ctab bert lh INFO: assuming MGZ format for volumes. computing statistics for each annotation in ../label/lh.aparc.annot. reading volume .mgz... zcat: /home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz.gz: No such file or directory mghRead(/home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz, -1): read error mris_anatomical_stats: could not read input volume .mgz
the freesurfer version $ cat $FREESURFER_HOME/build-stamp.txt freesurfer-Linux-centos4-stable-pub-v3.0.3
and the linux version is: $ cat /etc/slackware-version Slackware 10.2.0
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Nick Schmansky Sent: 17 August 2006 00:06 To: Abdel Douiri Cc: Freesurfer Mailing List Subject: RE: [Freesurfer] problem with mris_anatomical_stats
Abdel,
This problem is truly strange. The filename seems to just disappear.
Can you send me the output of:
mris_anatomical_stats --all-info
Also, does this error repeat? That is, if you try:
cd $SUBJECTS_DIR/01_00/scripts mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats \ -b -a ../label/lh.aparc.annot \ -c ../stats/aparc.annot.ctab 01_00 lh
does it fail again? And does this happen for other subjects?
Also, which freesurfer version is being used? To find out, type:
cat $FREESURFER_HOME/build-stamp.txt
And which Linux OS is it installed upon?
Nick
On Wed, 2006-08-16 at 17:16 +0100, Abdel Douiri wrote:
Thanks Bruce I have put the subject and that is "01_00" and the first and second autorecon was OK only the autorecon3 where it was the problem. Attached
the
log file. Thanks a lot for your help Abdel -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Bruce Fischl Sent: 16 August 2006 14:01 To: Abdel Douiri Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] problem with mris_anatomical_stats
Hi Abdel,
is that the full command line? If so, you need to specify a subject with -subjid. If not, could you send us the full command line and
recon-all.log?
cheers, Bruce
On Wed, 16 Aug 2006, Abdel Douiri wrote:
> Hi all, > > > > I will be grateful if someone can tell me what going wrong in the > following
> problem: > > When running "recon-all -autorecon3" the program stop with error > exactly in
> the "mris_anatomical_stats" cmd stage and I have the following error: > > > > mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats -b -a > ../label/lh.aparc.annot -c ../stats/aparc.annot.ctab 01_00 lh > > > > computing statistics for each annotation in ../label/lh.aparc.annot. > > reading volume .mgz... > > zcat: > /home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz.gz: No
> such file or directory > > mghRead(/home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz, > -1):
> read error > > mris_anatomical_stats: could not read input volume .mgz > > INFO: assuming MGZ format for volumes. > > Linux img-179 2.6.16.22 #2 SMP Mon Jun 26 15:49:31 BST 2006 i686 > unknown
> unknown GNU/Linux > > > > recon-all exited with ERRORS at Tue Aug 15 20:52:49 BST 2006 > > > > > > Thanks > > A > > > _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Abdel,
I've put a fixed version of mris_make_surfaces for centos4 here:
http://surfer.nmr.mgh.harvard.edu/pub/dist/
Also, I've checked the remaining downstream binaries (run by recon-all) and confirmed that the bug does not exist in those (hopefully!).
Nick
On Tue, 2007-01-09 at 16:11 +0000, A. Douiri wrote:
Hi Nick Thanks mris_fix_topology works now! But i have a similar bug a step later with: mris_make_surfaces Can you please check it. Thans for your help A
Nick Schmansky wrote:
Abdel,
Indeed mris_fix_topology had the same bug as mris_anatomical_stats (when running on Slackware Linux). I've posted a fixed version of mris_fix_topology here:
http://surfer.nmr.mgh.harvard.edu/pub/dist/
named:
mris_fix_topology.gz
mv this to your $FREESURFER_HOME/bin, then type:
gunzip mris_fix_topology.gz chmod a+x mris_fix_topology
Let me know if you encounter similar bugs. This fixed binary will appear in the next release.
Nick
On Fri, 2007-01-05 at 14:43 +0000, Abdel Douiri wrote:
Hi Nick We have done a new update to our Linux system and i have found the same error that i report a while ago (see bellow) But this time is for mris_fix_topology program. Can you please check it again for mris_fix_topology.c file, the error that you found was problem with sprintf! (sprintf(fname,"%s.mgz",fname) to strcat(fname,".mgz") ). Thanks Abdel
Abdel,
Thanks for testing this. The bug fix will appear in an upcoming release, but you can continue using the mris_anatomical_stats that you have now.
For those interested in the perils of testing a multi-platform app, the buggy line of code was:
sprintf(fname,"%s.mgz",fname)
Here the source and destination are the same, which, according to glibc docs for sprintf, is undefined behavior. The odd thing is that these Linux variants: RedHat 7.3, RedHat 9, Centos4, Debian and SuSE; and Mac OS X, all have a shared libc lib that handles the line of code properly. The libc included with Slackware, however, did not (and understandably so). libc.so is included not with Freesurfer, but with the OS, so this means that testing even on a half-dozen different platforms doesn't guarantee trouble-free performance on a seemingly identical platform.
The proper code in this case is:
strcat(fname,".mgz");
N.
On Fri, 2006-08-18 at 18:40 +0100, A. Douiri wrote:
that s works! Cheers!
At 01:51 18/08/2006, Nick Schmansky wrote:
Abdel,
I don't think there is a matlab alternative to mris_anatomical_stats.
However, there is a suspicious line of code in mris_anatomical_stats that I have fixed, and I'm wondering if you can give a new binary a try. It is found here:
http://surfer.nmr.mgh.harvard.edu/pub/dist/
and named:
mris_anatomical_stats.gz
To install, after downloading:
gunzip mris_anatomical_stats.gz chmod o+x mris_anatomical_stats mv mris_anatomical_stats $FREESURFER_HOME/bin
While I cannot recreate the error you see, I do not have a Slackware installation, and I am wondering if its core libs are revealing a bug in our code.
Let me know if it works.
Thanks,
Nick
Hi Nick, Thanks in advance for your help and i would like to ask is there an alternative in matlab for example to do mris_anatomical_stats? The detail of the outjkputs is:
the output of: $ mris_anatomical_stats --all-info ProgramName: mris_anatomical_stats ProgramArguments: --all-info ProgramVersion: $Name: stable3 $ TimeStamp: 06/08/17-10:28:55-GMT CVS: $Id: mris_anatomical_stats.c,v 1.35.2.1 2006/04/21 19:45:19 nicks Exp $ User: douiri Machine: img-179 Platform: Linux PlatformVersion: 2.6.16.22 CompilerName: GCC CompilerVersion: 30400
I tried with other subjects and also bert and i get the same problem:
$ mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats -b -a ../label/lh.aparc.annot -c ../stats/aparc.annot.ctab bert lh INFO: assuming MGZ format for volumes. computing statistics for each annotation in ../label/lh.aparc.annot. reading volume .mgz... zcat: /home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz.gz: No such file or directory mghRead(/home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz, -1): read error mris_anatomical_stats: could not read input volume .mgz
the freesurfer version $ cat $FREESURFER_HOME/build-stamp.txt freesurfer-Linux-centos4-stable-pub-v3.0.3
and the linux version is: $ cat /etc/slackware-version Slackware 10.2.0
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Nick Schmansky Sent: 17 August 2006 00:06 To: Abdel Douiri Cc: Freesurfer Mailing List Subject: RE: [Freesurfer] problem with mris_anatomical_stats
Abdel,
This problem is truly strange. The filename seems to just disappear.
Can you send me the output of:
mris_anatomical_stats --all-info
Also, does this error repeat? That is, if you try:
cd $SUBJECTS_DIR/01_00/scripts mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats \ -b -a ../label/lh.aparc.annot \ -c ../stats/aparc.annot.ctab 01_00 lh
does it fail again? And does this happen for other subjects?
Also, which freesurfer version is being used? To find out, type:
cat $FREESURFER_HOME/build-stamp.txt
And which Linux OS is it installed upon?
Nick
On Wed, 2006-08-16 at 17:16 +0100, Abdel Douiri wrote:
> Thanks Bruce > I have put the subject and that is "01_00" and the first and second > autorecon was OK only the autorecon3 where it was the problem. Attached > the
> log file. > Thanks a lot for your help > Abdel > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of Bruce > Fischl > Sent: 16 August 2006 14:01 > To: Abdel Douiri > Cc: freesurfer@nmr.mgh.harvard.edu > Subject: Re: [Freesurfer] problem with mris_anatomical_stats > > Hi Abdel, > > is that the full command line? If so, you need to specify a subject with > -subjid. If not, could you send us the full command line and > recon-all.log?
> cheers, > Bruce > > > On Wed, 16 Aug 2006, Abdel Douiri wrote: > > > > Hi all, > > > > > > > > I will be grateful if someone can tell me what going wrong in the > > > following > > > problem: > > > > When running "recon-all -autorecon3" the program stop with error > > > exactly > in > > > the "mris_anatomical_stats" cmd stage and I have the following error: > > > > > > > > mris_anatomical_stats -mgz -f ../stats/lh.aparc.stats -b -a > > ../label/lh.aparc.annot -c ../stats/aparc.annot.ctab 01_00 lh > > > > > > > > computing statistics for each annotation in ../label/lh.aparc.annot. > > > > reading volume .mgz... > > > > zcat: > > > /home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz.gz: > No > > > such file or directory > > > > mghRead(/home/samba/user/douiri/freesurfer/subjects/01_00/scripts/.mgz, > > > -1): > > > read error > > > > mris_anatomical_stats: could not read input volume .mgz > > > > INFO: assuming MGZ format for volumes. > > > > Linux img-179 2.6.16.22 #2 SMP Mon Jun 26 15:49:31 BST 2006 i686 > > > unknown > > > unknown GNU/Linux > > > > > > > > recon-all exited with ERRORS at Tue Aug 15 20:52:49 BST 2006 > > > > > > > > > > > > Thanks > > > > A > > > > > > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
A. Douiri -
Nick Schmansky