The xTEA algorithm may not be right implemented. There are two different
behaved implementations, because the specification is not clear. I assumed
that the C code in the specification is right, so in mcrypt xtea behaves
exactly as the code shown in the specification. The different approach
is the code in ftp.funet.fi which seems to be more rational than the
code in the specification. But since there is no indication of the author of
that code I decided not to adopt this version, and I assumed that the
C specification is right.
Nikos
y~or5J ={Eeu磝Qk ᯘG{?+]ן?wM3X^歌>{7پK>on\jy R g/=fOroNVv~Y+ NGuÝHWyw[eQʨSb> >}Gmx[o[<{Ϯ_qF vM IENDB`
