Yes, and when it's an SVRB you have convenient access to the registers at the time the SVC was issued. That may be more helpful than the registers at the time of failure.
-- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 עַם יִשְׂרָאֵל חַי ________________________________________ From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> on behalf of Joseph Reichman <reichman...@gmail.com> Sent: Monday, November 27, 2023 10:55 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Abend producing SDWARBAD I know that the SDWARBAD can point to a SVRB, PRB ,IRB Typically though not always it’s a SVRB or else SDWANAME would be populated I don’t profess to have your kind of knowledge In this But I do have somewhat of an idea what to look for Thanks Get Outlook for iOS<https://aka.ms/o0ukef> ________________________________ From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> on behalf of Seymour J Metz <sme...@gmu.edu> Sent: Monday, November 27, 2023 10:50:32 AM To: IBM-MAIN@LISTSERV.UA.EDU <IBM-MAIN@LISTSERV.UA.EDU> Subject: Re: Abend producing SDWARBAD An SVRB is not a PRB. There is a mapping macro. In addition to the old PSW, an SVRB contains the general registers at the time of SVC invocation. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 עַם יִשְׂרָאֵל חַי ________________________________________ From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> on behalf of Joseph Reichman <reichman...@gmail.com> Sent: Monday, November 27, 2023 10:03 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Abend producing SDWARBAD I have been asked to update file 192 on cbttape File 192 is a general recovery program The easiest abend to deal with is that where the error PSW Is whitin your program Besides updating with 64 bit support I would like to update in a case where the error PSW does NOT point to anywhere whitin program Typically though not always in this case the SDWARBAD is populated Truth is throughout all the share sessions On recovery which I have seen online this type of scenario is not discussed Probably because dealing with it is very complex I would like to determine where whitin the user code the IBM service was called going through this will definitely help me command of the TCB/rb/cde chains and different type of rb Where information is stored for SVRB PRB’s ( such as the last few bytes of the prefix telling you what SVC was called by the PRB and abend code ) Thank you > On Nov 27, 2023, at 9:36 AM, Peter Relson <rel...@us.ibm.com> wrote: > > I have no comment on what you say is a "hang". > > Here's a simple example that does abend and does place an RB address in > SDWARBAD and does not retry (so that your recovery gets control). It assumes > AMODE 31 or AMODE 64 (since for AMODE 24 it would use the address x'FFF000' > which is valid). > LLILF 15,x'7FFFF000' > SVC 42 > > If assembled as > TEST CSECT > TEST AMODE 31 > LLILF 15,x'7FFFF000' > SVC 42 > END > And linkedited, then > TEST my.loadlib(TEST) > TEST > go > IKJ56641I TEST ENDED DUE TO ERROR > IKJ56640I SYSTEM ABEND CODE 0C4 REASON CODE 00000011 > TEST > > And using TEST on a more complete program with an ESTAEX in effect at the > time of the SVC 42 and setting a breakpoint at the ESTAEX routine does get > control at the ESTAEX routine, where +x'58' into the SDWA has the RB address. > And you get similar results with a type 2 SVC such as SVC 122-9 (setting reg > 1 to the bad value for that case) > > If you are truly hanging, then you have again omitted important details. > > What is it that you are intending to do with "SDWARBAD"? > > Peter Relson > z/OS Core Technology Design > > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
