On Sat, 26 Jun 2010 16:59:40 -0500 Chris Craddock <[email protected]> wrote:
:>> So your suggestion is to store and MVCK? :>Yes. That's what those instructions are for. So you demand that the SRB have a work area just for this? :>> What parts of code (or the operating system) would be "astonished" to find :>> that an SRB is running in a non-system key? :>Almost any service that accepts SRB mode callers is going to document that it also expects them to be in a system key. If you call one of those in a user key you have no idea whether it may have done/caused harm, even if it appeared to work. If you can find a system service that accepts calls from a user key SRB I'll eat my hat. If by "astonished" you mean failing because the call was incorrect, I guess one could also say that WTO is "astonished" if branch entered in AR mode. Or that OPEN is "astonished" if R1 does not point to valid storage. Or almost any service is "astonished" if called incorrectly. :>> Would you expect that a resolvable :>> page fault would cause MVS to abend the SRB because it isn't in key 0? :>No of course not. There's nothing to stop a supervisor state program from running any way it wants. Doesn't make it a good idea. You forgot (IM(H?)O) :>> If the callers areas are in Key8 it need not follow that the SRB work areas :>> are in key8. :>True, but if the SRB is running in key 8 it is pretty difficult to store anywhere else :-) And (again IMO) flipping keys is just as likely to end up with the program running in an unexpected condition after somebody makes a maintenance change. The SRB may not need a work area. And that is what code review and testing is for. -- Binyamin Dissen <[email protected]> http://www.dissensoftware.com Director, Dissen Software, Bar & Grill - Israel Should you use the mailblocks package and expect a response from me, you should preauthorize the dissensoftware.com domain. I very rarely bother responding to challenge/response systems, especially those from irresponsible companies.
