3dmaskdump fatal error

AFNI Message Board
1

Hello everyone,

I am running a search light analysis which has successfully completed with a radius of 14. I am now attempting to run at a larger radius = 28 and am running into a fatal error when some (not all) voxels are the center. I thought it might be due to the xball being drawn near a boundary, but that doesn’t seem to be the case.

This is the code I pass in which I loop through each voxel (e.g., ii=42).


3dmaskdump -overwrite -quiet -noijk -xball -11.25 55 -56 28 -o tmp.roi_42 /Volumes/NBL_Projects/NSF_DDK2/Processed_Data/DDK2001/Y1/proc/exemplar_level_stats.DDK2001_Y1_REML_mask+tlrc.BRIK"[2..35(3)]"

This is the error I receive:


Fatal Signal 11 (SIGSEGV) received
  3dmaskdump main
 Bottom of Debug Stack
** AFNI version = AFNI_21.3.10  Compile date = Dec  3 2021
** [[Precompiled binary macos_10.12_local: Dec  3 2021]]
** Program Death **
2

in debugging I can re run the command after it crashes and then it runs.

3

Hello,

Yes, there is a bug there where the program is not verifying that the ball is entirely within the dataset bounds. It will be fixed and rebuilt, hopefully by morning.

Thanks for letting us know!

  • rick
4

The problem has been fixed, and new binaries are available.

Thanks again for pointing this out.

  • rick