A long time ago, the majority of system ABEND codes of the format X'abc' were 
organized so that 'bc' was the hexadecimal equivalent of the SVC number and the 
'a' part was a modifier.  E.g., there were many x37 ABEND codes, all 
"belonging" to SVC 37 (hex) = 55 (decimal) = EOV function within DADSM.  But 
there have always been exceptions.  In more recent years, the number of 
exceptions has been increasing rapidly. 



After lookiing up many ABEND codes, the pattern became obvious, but I do not 
remember ever seeing a statement from IBM that the pattern was guaranteed 
without exceptions for all time. 

Bill Fairchild 
Franklin, TN 
----- Original Message -----
From: "John McKown" <[email protected]> 
To: [email protected] 
Sent: Thursday, April 4, 2013 7:05:01 AM 
Subject: Re: IBM Hijacking User SVC ABEND code S0F8 - Was (Re: DataSpaces) 

I used to think that the Sxhh abends "belonged" to SVC x'hh' also. But it 
appear more like SVC x'hh' will try to use an Sxhh abend code if it needs 
to abend. But I don't think that is even guaranteed by IBM. I guess to be 
sure, one must look in the appropriate documentation. Which can become 
difficult if somebody gets, say, an SAED-50 (chosen at random) abend. Quick 
- what blew up and why? That's why I love QuickRef. It is fairly 
comprehensive in its look up capability. LookAt is good too,  for IBM stuff.

Reply via email to