How to determine contents of data segment?

#11 03-03-2008, 08:16 PM

On Mar 3, 3:15 pm, Robert Riebisch <Robert.Riebi...@arcor.de> wrote:
> Jim Leonard wrote:
> > So, does anyone know of a nice .map analyzer utility out there, to
> > prevent me from re-inventing the wheel? :-)
>
> Erm... What do you need this tool for? All info is already in the .map
> file.

Yes, but I am not going to visually inspect a ~200KB .MAP file looking
for the largest chunks of the DATA segment.

I will probably write a small utility to analyze the .MAP file,
determine the sizes of every element, and then print out the 10
largest ones or something.

#12 03-04-2008, 02:05 AM

Jim Leonard wrote :
>
> I will probably write a small utility to analyze the .MAP file,
> determine the sizes of every element, and then print out the 10
> largest ones or something.

Import the map file into an Excel sheet with the hex offsets in column
A and the symbol names in column B.
Then use the formula "=HEX2BIN(A2)-HEX2BIN(A1)" in column C - that is,
depending on which row is on top.
Copy column C to column D with values only - then sort on column D.

--

/klaus

#13 03-04-2008, 10:30 AM

On Mar 4, 1:05 am, Klaus Jorgensen <k...@no.spam> wrote:
> Import the map file into an Excel sheet with the hex offsets in column
> A and the symbol names in column B.
> Then use the formula "=HEX2BIN(A2)-HEX2BIN(A1)" in column C - that is,
> depending on which row is on top.
> Copy column C to column D with values only - then sort on column D.

I'll try that tonight, thanks very much!

Always learning, even after 20 years of working with TP... This
exercise has clued me into a few things:

- foo:string is the same as foo:string[255] (knew this already but
it's a good reminder never to use empty strings again)

- PChar arrays (aka defined as const foo:PChar='testing') take up
space in the data segment :-/

- The compiler takes advantage of the fact that PChar arrays are
pointers to null-terminated strings and is smart enough to optimize
100% identical contents by only having one copy and the others
reference it via pointers. I was not expecting that at all; that is
very cool.

#14 03-05-2008, 10:12 AM

On Mar 4, 1:05 am, Klaus Jorgensen <k...@no.spam> wrote:
> Then use the formula "=HEX2BIN(A2)-HEX2BIN(A1)" in column C - that is,

Hm... Excel 2003's HEX2BIN claims in the helpfile that it doesn't
support numbers over 1FF.

I'll write my utility anyway :-)

#15 03-05-2008, 12:05 PM

Jim Leonard wrote :
> On Mar 4, 1:05 am, Klaus Jorgensen <k...@no.spam> wrote:
>> Then use the formula "=HEX2BIN(A2)-HEX2BIN(A1)" in column C - that is,
>
> Hm... Excel 2003's HEX2BIN claims in the helpfile that it doesn't
> support numbers over 1FF.
>
> I'll write my utility anyway :-)

Sorry - my fingers or keyboard must have had an off day....
Should have been HEX2DEC and not HEX2BIN.

--

/klaus

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode