Making of checksum using XOR

Subsonics · 02-20-2009, 09:37 AM

I´m trying to make a checksum of the content in a datapacket struct and want to make sure I get it right. Let´s say that the struct have three variables where on is an array like this:

Code:

typedef struct {
   char a;
   char b;
   char c[10];
} DataPacket;

Should I then make the checksum like this?

Code:

char checksum;

checksum = a ^ b;

for(i = 0; i < sizeof(c); i++){
   checksum ^= c[i];
}

matsp · 02-20-2009, 09:41 AM

Yes, you can. And it will be a slightly "better" checksum than using addition, because all bits have an equal chance of being detected as erroneous. But you can only detect an odd-number of failures for each bit.

An addition checksum fails to detect a double failure of the upper-most bit [assuming the checksum is equal size to the transmission unit, e.g. 1 byte checksum].

--
Mats

Subsonics · 02-20-2009, 09:54 AM

Perfect, Thank you.

iMalc · 02-20-2009, 12:54 PM

That's called an LRC.
In practice, both checksum and LRC are inferrior to a good CRC.

Subsonics · 02-20-2009, 01:19 PM

Aah, I see. I´m trying to comunicate with a device that uses checksum so I´m stuck with that.
The docs I have says "checksum (exclusive OR of above 124 bytes)" I just wanted to make sure that my approach would work or I would be stuck re-sending the first packet. :-)

02-20-2009, 09:37 AM	#1
Subsonics Registered User Join Date: Jan 2009 Posts: 151	Making of checksum using XOR I´m trying to make a checksum of the content in a datapacket struct and want to make sure I get it right. Let´s say that the struct have three variables where on is an array like this: Code: typedef struct { char a; char b; char c[10]; } DataPacket; Should I then make the checksum like this? Code: char checksum; checksum = a ^ b; for(i = 0; i < sizeof(c); i++){ checksum ^= c[i]; }
Subsonics is offline

02-20-2009, 09:41 AM	#2
matsp Kernel hacker Join Date: Jul 2007 Location: Farncombe, Surrey, England Posts: 15,686	Yes, you can. And it will be a slightly "better" checksum than using addition, because all bits have an equal chance of being detected as erroneous. But you can only detect an odd-number of failures for each bit. An addition checksum fails to detect a double failure of the upper-most bit [assuming the checksum is equal size to the transmission unit, e.g. 1 byte checksum]. -- Mats __________________ Compilers can produce warnings - make the compiler programmers happy: Use them! Please don't PM me for help - and no, I don't do help over instant messengers.
matsp is offline

02-20-2009, 09:54 AM	#3
Subsonics Registered User Join Date: Jan 2009 Posts: 151	Perfect, Thank you.
Subsonics is offline

02-20-2009, 12:54 PM	#4
iMalc Algorithm Dissector Join Date: Dec 2005 Location: New Zealand Posts: 2,476	That's called an LRC. In practice, both checksum and LRC are inferrior to a good CRC. __________________ My homepage Advice: Take only as directed - If symptoms persist, please see your debugger
iMalc is offline

02-20-2009, 01:19 PM	#5
Subsonics Registered User Join Date: Jan 2009 Posts: 151	Aah, I see. I´m trying to comunicate with a device that uses checksum so I´m stuck with that. The docs I have says "checksum (exclusive OR of above 124 bytes)" I just wanted to make sure that my approach would work or I would be stuck re-sending the first packet. :-)
Subsonics is offline

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
checksum sender vs reciever?	silhoutte75	C Programming	1	04-13-2008 01:27 PM
weird checksum function	sagitt13	C Programming	7	10-31-2006 01:25 AM
Help figuring out a checksum algorithm	barem23	A Brief History of Cprogramming.com	5	01-11-2003 04:24 AM
Reading Binary file to find Checksum	Abbila	C++ Programming	0	09-25-2002 09:52 PM