Dear Clara,
yes, this is a known problem. The core of the problem is that the non-linear transform constructed from the field has to be invertible to work in eddy. If the field is created by topup that is guaranteed to be the case, but it is not guaranteed with a measured fieldmap. I am working on a piece of software to process measured fieldmaps to be able to guarantee invertibility also for those.
Unfortunately progress has been painfully slow due to various other commitments, but it is worked on and it will get finished.
Jesper
On 21 Sep 2018, at 10:49, Clara Fonteneau <[log in to unmask]> wrote:
Dear Dan and FSL community,
I am running into the same problem with eddy_openmp, using the --field option instead of the --topup. Have you find a solution to this problem?
I get the same error message during the 4th iteration:
Segmentation violation, Address not mapped, Offending address = 0x4015414e0
eddy_openmp ) [0x558dd3] [
eddy_openmp ) [0x563397] [
eddy_openmp ) [0x56381d] [
eddy_openmp ) [0x47ce01] [
eddy_openmp ) [0x47e1ae] [
eddy_openmp ) [0x47205a] [
eddy_openmp ) [0x48dd1a] [
eddy_openmp ) [0x489824] [
eddy_openmp ) [0x491240] [
eddy_openmp ) [0x48a070] [
In the same line as your problem, the problem seem to be random (i.e., it can work fine with the same command once and then get this error message the another for the same subject and files).
It doesn't seem to be a RAM problem, top was running on the side in order to detect memory problems when the segmentation violation occurred.
Any answers or suggestions would be very much appreciated.
Thanks,
Clara
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1