MVSFORUMS.com

dave · Beginner Joined: 29 Aug 2003 Posts: 5 Topics: 4

Can some telll me what are the advantages of 31-bit addressing over 24-bit addressing. I have gone through no. of internet sites but I have not got the satisfactory answer.

Can somebody please help me

Thax
Dave

kolusu · Posted: Fri Aug 29, 2003 5:07 am Post subject:

Dave,

In the early years when mainframe memory was limited, all of memory could be addressed using only the righmost 24 bits in an address register (hence, the term '24-bit addressing'). When memory went beyond that addressing scheme, a 'new' addressing scheme was implemented, in which the righmost 31 bits of a register are used (31-bit addressing). Both addressing schemes are valid below 16 megabytes, but only the 31-bit addressing scheme is valid above 16 megabytes. Options used when assembling/compiling and linkediting/binding programs that indicate your intentions: AMODE, which specifies the Addressing Mode, and RMODE which specifies the Residency Mode.

AMODE(24) indicates 24-bit (three-byte) addressing - memory below the line.
AMODE(31) indicates 31-bit addressing - memory above and below the line.
AMODE=ANY indicates the program may use either of the addressing technique.
Run Mode/Residency Mode
RMODE(24) indicates that the program must be loaded into memory below the line
RMODE(31) indicates that the program can be loaded either below or above the line.
RMODE=ANY indicates that the program can be run in either 24 bit (below)or 31 bit memory(above).

Hope this helps..

kolusu

dave · Beginner Joined: 29 Aug 2003 Posts: 5 Topics: 4

Thanks Kolusu
I got the point you explained. But what is the advantage of converting codes from 24 to 31. Is there performance issues. Or just to have more vitrual spaces.

Thax
Dave

to_agrawals · Beginner Joined: 12 Dec 2002 Posts: 26 Topics: 16

As far as I think, this problem surfaces in programs which are written in AMODE 24 and call programs written in AMODE 31. This creates problems. I too would like to know more about this and check if my thought is correct or needs correction.

Thanks
Saurabh

kolusu · Posted: Mon Sep 01, 2003 9:05 am Post subject:

dave,

In 24-bit addressing mode, the processor treats all virtual addresses as 24-bit values. This makes it impossible for a program in 24-bit addressing mode to address virtual storage with an address greater than 16,777,215 (16 megabytes) because that is the largest number that a 24-bit binary field can contain.

In 31-bit addressing mode, the processor treats all virtual addresses as 31-bit values.A virtual storage map of two gigabytes with MVS services to support programs executing or residing anywhere in virtual storage.Two program attributes that specify expected address length on entry and intended location in virtual storage. Bimodal operation, a capability of the processor that permits the execution of programs with 24-bit addresses as well as programs with 31-bit addresses.

Check this link for Understanding 31-Bit Addressing

Hope this helps...

cheers

kolusu

Mike Chantrey · Posted: Wed Sep 10, 2003 10:03 am Post subject:

Extra info:
On the Z-series machines running z/OS there is also 64 bit addressing. Storage above the limit for 31-bit addressing is known as 'above the bar' storage. This area is for data only - program code cannot execute here. Initially only assembler code can take advantage of 'above the bar' storage.

z/OS 1.1 to 1.4 can run with 31 bit hardware support or 64 bit hardware support, but IBM have recently announced that z/OS 1.5 onwards will only run on 64 bit capable hardware. So in 2007 when z/OS 1.4 goes out of support, there will be no supported MVS family operating system to run on 31 bit hardware. This is affecting second-hand values of such hardware already.
Note this does not affect software - 24 and 31 bit programs will still run unchanged on 64-bit hardware with z/OS in 64-bit mode.