1. Re: why you can't find ramdrive in your system
----- Original Message -----
From: Norm Goundry <bonk1000 at HOTMAIL.COM>
To: <EUPHORIA at LISTSERV.MUOHIO.EDU>
Sent: Monday, August 16, 1999 1:53 PM
Subject: Re: why you can't find ramdrive in your system
> The reason you can't 'find' windows:ramdrive.sys active in your operating
> system is simply because IT ISN'T. If you read my email carefully you
will
> notice that I stated that it was used during the INSTALL of the operating
> system. It (ramdrive) has to strap itself in as E:, that is why you have
> to leave it alone in regards to it picking its drive letter. It erases
> itself after it has done its prepatory work (one such necessary thing is
> for it to establish HIMEM so that Windows can do its install within it).
> It is NOT active once the install is completed; I never said that it was.
Ok, i looked again in Unauthorized Windows95 (by Andrew Schulman), page 50,
in chapter2: Watching Chicago Boot, at a SoftICE dump of Chicago booting up,
and while drivespace and himem.sys is loaded, ramdrive was not loaded. Himem
was loaded even tho it wasn't in the config.sys. After loading some dos7
version 16bit code, Win95 goes on to load it's VXDs in XMS, and begins
taking some operations from 16bit dos and moving them into 32bit dos7.
Specifically mentioned is V86MMGR taking over from himem.sys (page 57-59).
Ramdrive has nothing to do with ems/xms drivers, or the processor mode,, it
*can* use himem.sys to use xms, or the emm api to use the ems page, but
ramdrive also has switches to use only the lower 640k in real or v86 mode.
> By the way, when you said that you built your own ramdrive, do you mean
> that you implemented the ramdrive.sys application, or did you really build
> a program (as opposed to scripting commands as boot macros)? Pardon me
for
> being so abrupt, but Building and Using are two different things, and both
> are necessary in their place, but this is a programmer's forum, not an
> application batching forum. Sorry about the smoke, but at least that is
> better than flames.
I'm glad you don't flame without asking if it's appropriate first. I wrote
the ramdrive in Borland's TurboPascal V5 (i think,, it was quite a while
ago, could have been v7...), and i wrote some inline assy code in 16bit 386
code. I've also written assy code to do dos calls not supported in TP7, like
seek in a text file. And assy code using pascal's inline statement to do
bios calls. When i was giving up on pascal doing what i wanted it to, i was
exploring code to alter the PSPs for apps, to see if i could get dynamic
linking on demand. And since i have some of the compiler's source code, i
was considering function overloading. After some consideration, my wishlist
was too big, and that's when i stopped any significant coding. Sorry to
disappoint you.
> I do wish to thank you from getting some of us away from the constant
> filing of Linux stuff. It is great that porting is taking place, but Kat
> at least brings some relief in the form of a different subject. 
At least.
sigh.
Kat,
who simply leaves when poked at enough.
