From: Hans-Bernhard Broeker <broeker AT physik DOT rwth-aachen DOT de>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: read a specific memory address
Date: 2 May 2001 15:22:27 GMT
Organization: Aachen University of Technology (RWTH)
Lines: 31
Message-ID: <9cp8nj$ap6$1@nets3.rz.RWTH-Aachen.DE>
References: <9cp464$ju8$1 AT news2 DOT dtag DOT de>
NNTP-Posting-Host: acp3bf.physik.rwth-aachen.de
X-Trace: nets3.rz.RWTH-Aachen.DE 988816947 11046 137.226.32.75 (2 May 2001 15:22:27 GMT)
X-Complaints-To: abuse AT rwth-aachen DOT de
NNTP-Posting-Date: 2 May 2001 15:22:27 GMT
Originator: broeker@
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Thomas Schachtner <thomas DOT schachtner AT gmx DOT de> wrote:

> I want to read a value that is stored on (Real-Mode-)Address: F000:0000

That's simple. Use the _farpeekb() function.

> And another value from e. g. 0x03847384 (a 32 bit memory address).

There is no such thing as an absolute memory address, from the point
of view of a DJGPP application. We use a segmented memory model
similar to the 'tiny' model of 16 bit PC programing times, but with
segments potentially 4 Gigabytes long.

You're probably referring to a linear adress space (i.e. the actual
physical address bits at the CPU's pins).  You need special function
of the __dpmi_*() family to map such addresses into the program's
memory space, which are not available in all DPMI environments.

> is the address the same as the value from a pointer?

No.

> Why not?

Because of the Memory Management Unit (MMU) built into all x86 CPUs
since the 386. It translates logical addresses into linear/physical
ones, based upon settings you make in the 'descriptor tables'.

-- 
Hans-Bernhard Broeker (broeker AT physik DOT rwth-aachen DOT de)
Even if all the snow were burnt, ashes would remain.