sonny@charybdis.harris-atd.com (Bob Davis) (12/28/90)
All the information I have read about avoiding reentrancy to DOS seems to stress that the problem is in ruining the DOS internal stack areas. Seems to me that there must be more to it than just worrying about protection of the internal stacks. Aren't there certain RAM locations also used globally by DOS (aren't those areas starting at 40:00 and 50:00 examples of such regions?) that can be modified and therefore ruined by reentry even though measures have been taken to protect the stacks? In TSRs, all the "DOS Idle" (Int 28h), "InDOS" and "Critical Error" flag checking techniques do not protect me from TSR code that modifies global RAM storage areas and therefore ruins it for successful DOS reentrancy, do they? Is there really NO practical way to do context switching under DOS? No way to save ALL the bytes of the broad-sense environment (globally-used storage areas, stack regions, etc.) and then simply not worry about re-entering DOS code since we are going to restore the environment when we are done? What kills it for such an approach? Thanks. _____________________________________________________________________________ Bob Davis, UofALA alum \\ INTERNET: sonny@trantor.harris-atd.com | _ _ | Harris Corporation, ESS \\ UUCP: ...!uunet!x102a!trantor!sonny |_| |_| | | Advanced Technology Dept.\\ AETHER: K4VNO |==============|_/\/\/\|_| PO Box 37, MS 3A/1912 \\ VOICE: (407) 727-5886 | I SPEAK ONLY | |_| |_| | Melbourne, FL 32902 \\ FAX: (407) 729-2537 | FOR MYSELF. |_________|