On Mon, 3 Aug 2009 10:43:29 -0500, Rick Fochtman <rfocht...@ync.net> wrote:
> >>... One way to greatly relieve >>it would be to allow (part of) LPA and JPA to reside above-the-bar. >> >> >That may happen yet, but I wouldn't hold my breath waiting for it. There >would have to be major modifications the most of the Operating System, Yes. >as well as the hardware, before RM=64 would be viable. I think the hardware is ready, and has been since the introduction of z/Architecture. There are limitations. For example, CCWs need to be below the bar, but these don't affect the ability to have code running above the bar. Indeed, Linux is able to run code above the bar today. >Think of the size >of a simple SYSUDUMP, and how it would be generated. Of course, with larger address spaces, dumps get bigger >How would you do I/O above the bar? Another CCW change? Not necessarily. Format 2 IDAW and MIDAW both support 64-bit data addresses. IAC, it is not a RM=64 problem. >Not to mention all the interrupt >handlers and the interrupt mechanism in the hardware. Again, this has been ready since the introduction of z/Architecture. The Interrupt old and new PSW fixed storage definitions are changed. They are at new locations and they are all 16 bytes. Bigger problems for IBM are the places where the PSW is stored as an 8-byte value and where program addresses are stored in fullwords. RB, SDWA and PIE, for example. -- Tom Marchant ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
