Re: BALR and BAL in AMODE=24

2019-01-22 Thread Tony Harminc
On Tue, 22 Jan 2019 at 07:44, John Gateley wrote: > The link information in the 24-bit addressing mode consists of the > instruction-length code (ILC), the condition code (CC), the program-mask > bits, and the rightmost 24 bits of the updated instruction address. > > I have never given much tho

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Paul Gilmartin
On Tue, 22 Jan 2019 17:13:04 +, Seymour J Metz wrote: >I've used the CC bits but never the ILC bits. > I've used them to tell whether I was on 370 hardware or 31-bit mode on XA hardware. I did not support 24-bit mode on XA hardware. -- gil --

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Seymour J Metz
I've used the CC bits but never the ILC bits. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 From: IBM Mainframe Discussion List on behalf of John Gateley Sent: Tuesday, January 22, 2019 7:43 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: BALR and

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Seymour J Metz
: Re: BALR and BAL in AMODE=24 I wonder what the purpose of this is. It branches to R14, usually the return address, unless you loaded the destination address in R14 first. I am used to BALR R14,15, where R15 is the (conventional) destination address. Kees. > -Original Message- > Fro

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Steve Thompson
Yes. I have worked on application programs that interrogated the high order byte to determine what logic path to take in the called routine. In certain other cases it was used after return to the caller to look at the cc from before the BAL/BALR and branch accordingly. Sent from my iPhone

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Ray Pearce
wab Sent: 22 January 2019 15:30 To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: BALR and BAL in AMODE=24 First thing is it loads the address of the next instruction into R14. It then jumps to that location. Doesn't that cause some processors to sync up the cache and instruction pipelines? A BA

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Mike Schwab
ntional) destination address. > > Kees. > > > > -Original Message- > > From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On > > Behalf Of esst...@juno.com > > Sent: 22 January, 2019 14:31 > > To: IBM-MAIN@LISTSERV.UA.EDU > > S

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Binyamin Dissen
On Tue, 22 Jan 2019 06:43:54 -0600 John Gateley wrote: :>The link information in the 24-bit addressing mode consists of the instruction-length code (ILC), the condition code (CC), the program-mask bits, and the rightmost 24 bits of the updated instruction address. :>I have never given much tho

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Vernooij, Kees (ITOP NM) - KLM
n List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On > Behalf Of esst...@juno.com > Sent: 22 January, 2019 14:31 > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: BALR and BAL in AMODE=24 > > I remember back in the CICS/VS days we would code a BALR 14,14 > It was a great Interview question

Re: BALR and BAL in AMODE=24

2019-01-22 Thread esst...@juno.com
I remember back in the CICS/VS days we would code a BALR 14,14 It was a great Interview question to explain the workings of a BALR instruction. -- Original Message -- From: John Gateley To: IBM-MAIN@LISTSERV.UA.EDU Subject: BALR and BAL in AMODE=24 Date: Tue, 22 Jan 2019 06:43:54

Re: BALR and BAL in AMODE=24

2019-01-22 Thread Joe Monk
In the 370 days, thats how we used to establish addressability! A BALR to a register with a zero in the link position merely loads the PSW instruction address into a register, like this: BALR 15,0 USING *,15 Joe On Tue, Jan 22, 2019 at 6:44 AM John Gateley wrote: > Hello. > > The link informat