Mmm.. that crash was our good old friend "ext2_new_block()", which others have also seen from time to time.

This probably won't happen with the newest Hijack kernels, as I've turned on "frame pointer" support in those again, and others have reported that their crashes disappeared after doing that.

Cheers