Path: utzoo!attcan!uunet!ogicse!ucsd!ucselx!petunia!news
From: cambler@polyslo.CalPoly.EDU (Fubar)
Newsgroups: comp.mail.uucp
Subject: Checksumming
Message-ID: <26fb29cf.5149@petunia.CalPoly.EDU>
Date: 22 Sep 90 09:07:27 GMT
Reply-To: cambler@polyslo.CalPoly.EDU (Fubar)
Organization: Fantasy, Incorporated: Reality None of Our Business.
Lines: 23



I'm noticing interesting bugs in checksumming. I hope someone can
help out.

Of the 4 digit checksum, in the form 0xABCD (eg. 0x3F51 or 0x1234),
the A digit is always OK or off by 8, the B digit is always OK or off by 1,
the C digit is always OK or off by 8, and the D digit is always OK. Is there
some massaging that I am missing?

final_checksum = 0xAAAA - (checksum(data, SIZE) ^ (header[4] ^ 0xff));

right? the control byte is header[4]? [DLE][K][C0][C1][C][X] [data...]
                                        0   1  2   3   4  5


Argph!

-- 
++Christopher(); --- cambler@polyslo.calpoly.edu --- chris@fubarsys.slo.ca.us
---
Ever get the feeling that there is someone out there who understands you
completely?