PCBUILD Archives

Personal Computer Hardware discussion List

PCBUILD@LISTSERV.ICORS.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Reply To:
PCBUILD - Personal Computer Hardware discussion List <[log in to unmask]>
Date:
Tue, 11 Jul 2000 10:48:28 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (39 lines)
On 10 Jul 2000, at 21:30, Frank R. Brown wrote:
> I agree with your comment about the vdm.  But what limitation
> in dos/windows would prevent multiple vdm's from running?
> nt has permitted multiple vdm's since its early versions.

For win95 to be sucessful, MS had to ensure maximum
compatibility with older apps and the ability to run on limited
hardware (4 MB of RAM etc).
A robust VDM mechanism breaks both these requirements.

> I've never understood the complete story.  But if dos/windows
> has complete pre-emptive multitasking, how can a single rogue
> app consistently hang the whole os?  (In nt, this *rarely* happens

There is a big difference in OS architecture.

For example, on win9x, the running application thread has access
to the memory space used by the system VM.
On NT, what appears to the application as the ring 3 system
DLLs are in fact just proxy stub forwarders (which pass the calls to
the separate, isolated system VM).

This is one of the reasons why on NT you don't have the OS
crashes, lockups etc that you get with win98/95.

There are other vulnerabilities, such as the 16 bit GDI code in
win9x that allow a single app to hang the system.



Gerard R Thomas
Port of Spain,  Trinidad and Tobago
mailto:[log in to unmask]  mailto:[log in to unmask]
PGP Key IDs: RSA:0x9DBCDE7D  DH/DSS:0xFF7155A2

                         PCBUILD's List Owner's:
                      Bob Wright<[log in to unmask]>
                       Drew Dunn<[log in to unmask]>

ATOM RSS1 RSS2