At 03:57 PM 1/12/2005 +0300, Arkady V.Belousov wrote:
MD> Nothing on an application level should need to disable A20 and if it did, MD> it would crash things loaded in a UMB due to loss of memory access, not to --------------------------------------^^^ HMA ? MD> mention DOS image loaded high going away. Most interrupts would MD> immediately fail.
Why? They should fail only if they point into HMA and performed when A20 is off.
If you disable A20, then it's off, right? Important interrupts can go into DOS territory No HMA when an interrupt that goes there is triggered == bad situation. As I said, it was a decision made a long time ago before my time, you could research old discussions and talk with original developers if you want to know the exact reasons. I don't think it affects anything important.
As far as UMB's, I'm not sure if turning off the A20 line affects memory mapping from physical odd-address Mb. Seems like it should, but it would take time I'm unwilling to spend to create a definitive test. If A20 did affect it, that's another excellent reason not to support messing with it via application. Visualize a mouse driver going away.
MD> on my roadmap. May happen before March.
"May" == "can" or == "month May"? :) (Rhetoric questions, don't mind.)
The context points to use of "may" being a verbal auxiliary for the intransitive verb "happen", as part of the predicate following an implied pronoun simple subject "It". (Yes, I had to look a lot of that up.) Aren't you glad you asked?
Pool sharing coding will happen by the end of February, in the absence of unusual circumstance.
------------------------------------------------------- The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt _______________________________________________ Freedos-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freedos-devel
