View previous topic :: View next topic
|
Author |
Message |
monica1
New User
Joined: 05 Jan 2017 Posts: 5 Location: Bangalore
|
|
|
|
Hi,
I am trying to find the line of code which caused the abend(S0CF) in an assembler module. The abend occurred when transaction 900I was invoked. MODULE IN ERROR RDAARRYB is just a link card that is used to link-edit a number of modules.
The dump is as shown below.
Code: |
0ABENDING TASK ST3 EPA 11C2ECA8 MODULE IN ERROR RDAARRYB CSECT *
0PSW AT ENTRY TO ABEND 078D0E00 92AEFA7A TEA
DATA AT PSW 12AEFA74 - 50007D20 D1247020 D1004180
Task code 900I ,initial program DS2900 ,current prog DS2900 , 4 l
0REGS 0-7 1213C680 80000000 1213BC98 12100984 12101114 1213C08C 9213C0A8
REGS 8-F 92AEFA6C 121006D8 121048DA 12AEF910 12AE7030 1213C5D4 92AEFA3E
0FLTR 0-3 C054395800000001 0000000000000000 0000000000000000 00000000
0FLTR 4-7 0000000000000000 0000000000000000 FFFFFFFF00000000 00000000
0FLTR 8-B 0000000000000000 0000000000000000 0000000000000000 00000000
0FLTR C-F 0000000000000000 0000000000000000 0000000000000000 00000000
1DCF JOB DLIVE2D2 PROG DS900I SEQ 0011 TIME 11:20:34 DAT
-REQUEST BLOCK -------------------------- at location 00BA7100
0PRB PSW 078D0000 80DA7EF6 WC-L-IC 00020001 SZ-STAB 00110002 RB-LINK
REGS 0-7 000001F0 12DA3794 12DA3EF0 000102DC 00020188 00BA7F50 11A371C0
REGS 8-F 00040E40 00BA7380 00022000 92DA2228 11C2FB94 12DA3720 12DA2E7A
|
On further going down the dump, I can see that the module that caused the error was AGETFID.
Code: |
12AEF910 12AE7030 1213C5D4 92AEFA3E 00000000
---- at location 00BCE070
0004000F SZ-STAB 00110082 RB-LINK 00BCE0F8
11C2ECA8 00000006 000556B4 000102E4 11C2C000
11C2BB98 00040E40 00350F68 00044360 11C2B250
INT 91C2ECA8 -------------------------------
---- at location 1213C5D4
fset 015CE8, ASM CSECT AGETFID ASSEMBLED_ON_11/27/12_AT_04.45
LSA 1213C86C RET 92AEFA3F EPA 12AF1C78 R0 1213C680
R3 12100984 R4 12101114 R5 1213C08C R6 9213C0A8
R9 121006D8 R10 92AE67EA R11 12AEF910 R12 12AE7030
---- at location 1213BFBC
fset 013980, ASM CSECT LFRNC ASSEMBLED_ON_06/06/11_AT_11.11
LSA 1213C5D4 RET 92AE680E EPA 12AEF910 R0 1213C148
R3 1218F800 R4 00044360 R5 00041C6C R6 00000002
R9 12169000 R10 92AE67EA R11 12AE6030 R12 12AE7030
SEQ 0011 TIME 11:20:34 DATE 17.128 TERM 03761 ABEND S0CF
|
I tried subtracting the EPA from PSW (2AF1C78 - 2AEFA7A) as suggested in many places in google but it is giving me a negative number which is incorrect.
Could you help me in analyzing the assembler dump?
Regards,
Monica.
[/code] |
|
Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006 Posts: 6966 Location: porcelain throne
|
|
|
|
Quote: |
I tried subtracting the EPA from PSW (2AF1C78 - 2AEFA7A) as suggested in many places in google but it is giving me a negative number which is incorrect. |
2AF1C78
2AEFA7A
if i were you,
i would be tempted to overview my subtraction. |
|
Back to top |
|
|
Willy Jensen
Active Member
Joined: 01 Sep 2015 Posts: 712 Location: Denmark
|
|
|
|
2AF1C78 - 2AEFA7A = 21FE according to my calculator. What were you using for the calculation? |
|
Back to top |
|
|
steve-myers
Active Member
Joined: 30 Nov 2013 Posts: 917 Location: The Universe
|
|
|
|
S0CF - I had to go to Principles of Operation to look that one up! But that's the first thing you should do when faced with an unusual (like S0CF) ABEND code.
I suspect you have a wild branch here, so "normal" debug procedures won't help you very much. I'd look at the registers to see if any are reasonably close to the failing address. Especially if the address is in register 15, I'd next look at register 14 and see where that leads you. Good luck. |
|
Back to top |
|
|
steve-myers
Active Member
Joined: 30 Nov 2013 Posts: 917 Location: The Universe
|
|
|
|
Next, I'd try to construct the instructions from the DATA AT PSW area. I get
----5000
7D20D124
7020D100
4180----
The 7D20D124 appears to be the failing instruction, which tracks to the S0CF ABEND code. The 7020D100 instruction also seems reasonable to follow the 7D instruction. I had to go to Principles of Operation to lookup both instructions. 4180 (which I didn't have to lookup) also seems legitimate
Regardless, I'm probably just spinning my wheels here, since you likely have a wild branch to what may be legitimate code, but I could be wrong! That's all I'm going do here; you have plenty of leads to research. |
|
Back to top |
|
|
|