On Tue, 28 May 2002, Jason R Thorpe wrote:
   
   [fgen]
   > There are probably byte-order bugs in this -- should make sure it generates
   > the right output when used on a little-endian host.
   
   Erm, I haven't changed the code in any significant way, so if it's broken,
   the tgm builds are also broken.  Has anyone tested them on sparc{,64}?


i thought i tested fgen on i386 a year or two ago and it worked ok...