Show all memory occupied by the program

[Default User]

Chris Hills wrote:

Ignore the netpolice.

Don't listen to this idiot. If you ignore the group consensus on
topicality, you'll quickly be marginalized. That means even when you
have a topical question, you'll not get answers because the majority
will be ignoring you.

You will get the best answers in the correct newsgroup.





Brian

 

[Mark McIntyre]

Yes This one is specific to the target in question However I could
have generated a similar file from any of the 30 compilers I have.

However as the majority of compilers are for free-standing targets it is
not off topic.

The majority of compilers emit textual error messages too. That
doesn't make specific text messages, or even the existence of text
mesages, topical. The standard is silent on how or indeed whether an
implementation must provide a textual error. It must merely emit one
diagnostic, which could equally be a beep, inverting video briefly, or
emitting a TCP packet.

You are COMPLETELY WRONG the map above give EXACT memory locations

Why is precisely why wittering on about offtopic stuff here is
irresponsible. I can't call you on that, whether its true or not,
because I'm no expert in your platform.

You have dug your own hole

Amusing.

The memory map given above give EXACT memory
locations.

On one very specific platform, using one very specific memory layout,
and with one very specific addressing strategy. It is completely
nonportable, probably totally unlike any other memory map and quite
off topic.

But as I said, you knew all this already...

Map files are used with Ice and Logic analysers to debug.

No sh.. sherlock?

I disagree by the way, map files are used to provide my GPS with
cutesy pictures to show me where the nearest kebab shop is.
--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

 

[Ian Collins]

So would you care to explain the mythical output below?

He won't. He gets his kicks form being rude, not admitting he is wrong.

 

[Chris Hills]

Why is precisely why wittering on about offtopic stuff here is
irresponsible. I can't call you on that, whether its true or not,
because I'm no expert in your platform.

I was talking about the majority of platforms. It's not my problem that
you only have experience of a very few platforms or targets.

On one very specific platform, using one very specific memory layout,
and with one very specific addressing strategy. It is completely
nonportable, probably totally unlike any other memory map and quite
off topic.

As I said the file I showed was specific but is an example of the map
file the majority of compilers will do. The majority of compilers will
give exect memory locations.

Wintel is one of the few exceptions. where it doesn't work.

I could give examples for another 30+ compilers including x86
targets.... all of which will give a similar file all with exact memory
locations that is the whole point of a map file.

But as I said, you knew all this already...

I know your experience is so limited as to be pointless in this
discussion

 

[Chris Hills]

Chris Hills wrote:




Don't listen to this idiot. If you ignore the group consensus on
topicality, you'll quickly be marginalized.

Which is why you see so many posts shouting OT.

Last time this discussion came up there were as many on one side as the
other and that has the case for the decade or so I have been no here.

 

[Walter Roberson]

I was talking about the majority of platforms. It's not my problem that
you only have experience of a very few platforms or targets.

No, you said "It has nothing to do with the OS." If you had said, "It
has little to do with the OS" you might have been right (perhaps,
allowed for the sake of argument), but you made an absolute statement
implying that *every* OS and compiler produces a memory map suitable
for the OP's muddy purposes. As your "nothing to do with the OS"
statement was in reply to a statement that the matter is OS specific,
your reply was an incorrect attempted refutation of the redirection.

If your reply had been to -add- the information that "On many
systems, the compiler produces a memory map file that may be
sufficient for the OP's needs", then there wouldn't be an issue.

 

[Keith Thompson]

No, you said "It has nothing to do with the OS." If you had said, "It
has little to do with the OS" you might have been right (perhaps,
allowed for the sake of argument), but you made an absolute statement
implying that *every* OS and compiler produces a memory map suitable
for the OP's muddy purposes. As your "nothing to do with the OS"
statement was in reply to a statement that the matter is OS specific,
your reply was an incorrect attempted refutation of the redirection.

[...]

And regardless of how much it has to do with the OS, it has very
little to do with the C programming language. On the one system I've
bothered to check, a memory map file is generated by the linker, not
by the compiler. My guess is that that's true in general. A linker
deals with object files, not with source files; it may not even be
aware of what language the original sources were written it.

So neither the question nor Chris Hills's answer to it had anything to
do with the C programming language. Perhaps comp.compilers or
comp.programming would have been a better place to redirect the
original query, rather than an OS-specific newsgroup, but it seems
clear that comp.lang.c isn't the right place for it, even if somebody
here happens to know the answer.

<OT>
The particular OS-specific solution I was thinking of was the
/proc/<PID>/maps pseudo-file available under Linux.
</OT>

 

[Default User]

Which is why you see so many posts shouting OT.

Last time this discussion came up there were as many on one side as
the other and that has the case for the decade or so I have been no
here.

I don't believe you. Every time this discussion comes up, there a few
people on the side of expansion, but very few. The overwhelming
majority, especially when weighted by typical contribution rate, favors
keeping the basic topicality that the group has employed for many years
now.




Brian

 

[Keith Thompson]

I don't believe you. Every time this discussion comes up, there a few
people on the side of expansion, but very few. The overwhelming
majority, especially when weighted by typical contribution rate, favors
keeping the basic topicality that the group has employed for many years
now.

And, to the best of my recollection, nobody has ever come up with a
coherent proposal for a new and more expansive set of topicality
guidelines, though I've asked for such proposals several times.

The set of guidelines encouraged by the so-called "netpolice" are the
*only* set of guidelines on which there's any kind of consensus
(unless you count the consensus of the trolls and spammers that
absolutely anything goes). If you don't like that situation, Chris,
then do something to change it. Tell us what guidelines you'd like to
see. It's not inconceivable that we might agree with you.

 

[Mark McIntyre]

I was talking about the majority of platforms. It's not my problem that
you only have experience of a very few platforms or targets.

Ah, I see - insults. Very good.

As I said the file I showed was specific

Good that we agree about something.

but is an example of the map
file the majority of compilers will do. The majority of compilers will
give exect memory locations.

Well, now, I don't claim to be the complete expert of this offtopic
field that you claim to be but putting it shortly: utter bollocks. You
can run the same app twice and it'll get loaded at a different address
each time, depending on other stuff in memory.....

Wintel is one of the few exceptions. where it doesn't work.

..... even on non-wintel platforms.

I could give examples for another 30+ compilers including x86
targets.... all of which will give a similar file all with exact memory
locations that is the whole point of a map file.

So what? Its still offtopic.

I know your experience is so limited as to be pointless in this
discussion

Ah, more insults, argument over then.
--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

 

[Mark McIntyre]

Last time this discussion came up there were as many on one side as the
other and that has the case for the decade or so I have been no here.

Not only does that last sentence not parse, but it also makes no sense
even if you fill in the gaps. However its not a problem, because
you're in the bitbucket.


--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

 

[Keith Thompson]

Linkage is part of the C std. From N869, 5.1.1.2:

8. All external object and function references are
resolved. Library components are linked to satisfy
external references to functions and objects not
defined in the current translation. All such
translator output is collected into a program image
which contains information needed for execution in its

Yes, but memory map files, and the operation of any particular linker,
are not.