CompuServe Thread

Forum unknown · Programming

#Misc.

3 messages in this thread
#79165From: Glenn NielsenAug 5, 1987 4:26 PM
Well sweet sucess!! Here's the info on the header checksum. It is the encoded longword EOR's of the header starting at the format byte to the end of the o/s recovery info, then AND with x55555555 to mask out the clock bits. Andy Finkel was a great deal of help answering my questions about these checksums. <Glenn
#79235From: Don Curtis/SYSOPAug 5, 1987 11:43 PM
Glenn, Ok, then if you presume that the "original" header checksum was 0x55555555 and then take the 5 longwords preceeding it and EOR them and then AND the result with that "original" checksum…you come up with the real checksum… correct? It also makes sense to ignore the framing and sync bytes…cause if they were wrong…you more than likely couldn't read the sector anyway. I appreciate the information…I've got it tucked away for future reference…thanks!
#79235From: Don Curtis/SYSOPAug 5, 1987 11:43 PM
Glenn, Ok, then if you presume that the "original" header checksum was 0x55555555 and then take the 5 longwords preceeding it and EOR them and then AND the result with that "original" checksum…you come up with the real checksum… correct? It also makes sense to ignore the framing and sync bytes…cause if they were wrong…you more than likely couldn't read the sector anyway. I appreciate the information…I've got it tucked away for future reference…thanks!