I'm attempting to compare cortical surface area among three groups while controlling for estimated total intracranial volume (eTIV), as described on the Surface Based Group Analysis tutorial (MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysisV6.0). 

I ran the following command:

mri_glmfit --y three_group_comparison.area.lh.10.mgh --fsgd three_group_comparison.area.fsgd dods --C three_group_comparison.area.mtx --surf fsaverage lh --cortex --glmdir three_group_comparison.area.lh.glmdir --eres-save

Then I got the following error:

gdfRead(): reading three_group_comparison.area.fsgd
INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done.
Continuous Variable Means (all subjects)
0 eTIV 1.53351e+06 139455
Class Size and Means of each Continuous Variable
1 0 42 1566070.5417
2 1 34 1515147.1360
3 2 15 1483959.5000
INFO: gd2mtx_method is dods
Reading source surface /media/sjkim/hd2/subject_data/fsaverage/surf/lh.white
Number of vertices 163842
Number of faces    327680
Total area         65417.000000
AvgVtxArea       0.399269
AvgVtxDist       0.721953
StdVtxDist       0.195472

7.2.0
cwd /media/sjkim/hd2/subject_data/glm
cmdline mri_glmfit --y three_group_comparison.area.lh.10.mgh --fsgd three_group_comparison.area.fsgd dods --C three_group_comparison.area.mtx --surf fsaverage lh --cortex --glmdir three_group_comparison.area.lh.glmdir --eres-save
sysname  Linux
hostname sjkim-System-Product-Name
machine  x86_64
user     sjkim
FixVertexAreaFlag = 1
UseMaskWithSmoothing     1
OneSampleGroupMean 0
y    /media/sjkim/hd2/subject_data/glm/three_group_comparison.area.lh.10.mgh
logyflag 0
usedti  0
FSGD three_group_comparison.area.fsgd
labelmask  /media/sjkim/hd2/subject_data/fsaverage/label/lh.cortex.label
maskinv 0
glmdir three_group_comparison.area.lh.glmdir
IllCondOK 0
ReScaleX 1
DoFFx 0
Creating output directory three_group_comparison.area.lh.glmdir
Loading y from /media/sjkim/hd2/subject_data/glm/three_group_comparison.area.lh.10.mgh
   ... done reading.
INFO: gd2mtx_method is dods
Saving design matrix to three_group_comparison.area.lh.glmdir/Xg.dat
Computing normalized matrix
Normalized matrix condition is 833.575
Matrix condition is 1e+08
Found 149955 points in label.
Pruning voxels by thr: 1.175494e-38
Found 149953 voxels in mask
Saving mask to three_group_comparison.area.lh.glmdir/mask.mgh
Reshaping mriglm->mask...
search space = 74612.059149
DOF = 85
Starting fit and test
Fit completed in 0.0373833 minutes
Computing spatial AR1 on surface
Residual: ar1mn=0.997826, ar1std=0.000886, gstd=8.674183, fwhm=20.426140
Writing results
  three_group_comparison.area
    maxvox sig=3.0263  F=7.57212  at  index 107440 0 0    seed=1645215319
error: No such file or directory
error: MatrixMultiplyD(): m2 is null
 break /home/rd521/space_freesurfer/build/docker_ubuntu18/trunk/rd521-7.2.0/utils/matrix.cpp:596

I used FreeSurfer version 7.2.0 and Ubuntu 20.04.

I've searched the mail archives and similar errors have been reported in qdec, but I could not find any specific solution for this problem. Do you have any suggestions to solve this problem?

Thank you in advance,

Hyunwoo Jeong