Home > Archive > Fortran > March 2006 > Salford FTN95 Access violation on compile time
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Salford FTN95 Access violation on compile time
|
|
| Mathemagician 2006-03-27, 7:02 pm |
| Hi everyone,
I'm using Salford's Plato 3.15 and FTN 95 v.4.6.0 and I have the
following problem:
On compile time I get the message:
Access violation:
The instruction at address [..] attempted to read from location [..]
The [..] seem to be hexadecimal addresses.
The above lines are then followed by more addresses and C type function
names.
Unfortunately I cannot draw any information from that message in
regards to where the problem might be. Other error messages have always
given me at least some information on where to look for the problem,
but this one doesn't.
This error has popped up earlier occasionally and I remotely recall
that I always solved it somehow, but this time there were to many
changes in the source code to test them step by step and I don't know
where to look.
What is also strange is that if I make some modifications, save the
code, undo the modifications, save again and compile again, the error
will be gone in some scenarios, but not in others.
The total code is too large to post here and I have no idea which part
to post, so I cannot post any extracts.
Thanks for any help you can give.
Rainer
| |
| beliavsky@aol.com 2006-03-27, 7:02 pm |
| Mathemagician wrote:
> Hi everyone,
>
> I'm using Salford's Plato 3.15 and FTN 95 v.4.6.0 and I have the
> following problem:
>
> On compile time I get the message:
>
> Access violation:
> The instruction at address [..] attempted to read from location [..]
>
> The [..] seem to be hexadecimal addresses.
> The above lines are then followed by more addresses and C type function
> names.
According to a recent thread in this newsgroup
"Error when compiling with FTN95"
the latest version of the Salford compiler is 4.9.0
I suggest downloading the latest personal edition from
http://www.silverfrost.com/32/ftn95...nal_edition.asp and
seeing if that compiles your code. If it does, maybe you can upgrade.
I have had better experiences with g95 (free), Lahey, and Compaq Visual
Fortran than Salford.
| |
| Mathemagician 2006-03-27, 7:02 pm |
| Thank you for the fast response !
I will try that new compiler. :)
A hint on the apparently Salford-follow-up Silverfrost forum had me
delete all the MOD files before a new compilation. This removed the
access violation, at least as long as I repeat the deletion process
before each compilation. :)
I will possibly come back to you on the question of substitute
compilers if you don't mind. :)
| |
| Herman D. Knoble 2006-03-28, 7:02 pm |
| Latest version of Salford ftn95 is: 4.9.0
It's almost certain that this compiler bug was fixed between v4.6.0
and 4.9.0.
Skip Knoble
On 27 Mar 2006 12:27:20 -0800, "Mathemagician" <mathemagician@gmx.net> wrote:
-|Hi everyone,
-|
-|I'm using Salford's Plato 3.15 and FTN 95 v.4.6.0 and I have the
-|following problem:
-|
-|On compile time I get the message:
-|
-|Access violation:
-|The instruction at address [..] attempted to read from location [..]
-|
-|The [..] seem to be hexadecimal addresses.
-|The above lines are then followed by more addresses and C type function
-|names.
-|
-|Unfortunately I cannot draw any information from that message in
-|regards to where the problem might be. Other error messages have always
-|given me at least some information on where to look for the problem,
-|but this one doesn't.
-|This error has popped up earlier occasionally and I remotely recall
-|that I always solved it somehow, but this time there were to many
-|changes in the source code to test them step by step and I don't know
-|where to look.
-|
-|What is also strange is that if I make some modifications, save the
-|code, undo the modifications, save again and compile again, the error
-|will be gone in some scenarios, but not in others.
-|
-|The total code is too large to post here and I have no idea which part
-|to post, so I cannot post any extracts.
-|
-|Thanks for any help you can give.
-|Rainer
| |
| Mathemagician 2006-03-28, 7:02 pm |
| Thank you. Upgrading to 4.9.0 did indeed do the trick ! :)
Also thanks for calling it a "compiler bug". I know that I'm not very
experienced with these things and from the reactions in the Silverfrost
forum I was starting to think that I may be just too inexperienced. :)
| |
| Mathemagician 2006-03-28, 7:02 pm |
| Thank you. Upgrading to 4.9.0 did indeed do the trick ! :)
Also thanks for calling it a "compiler bug". I know that I'm not very
experienced with these things and from the reactions in the Silverfrost
forum I was starting to think that I may be just too inexperienced. :)
| |
| Richard E Maine 2006-03-28, 7:02 pm |
| Mathemagician <mathemagician@gmx.net> wrote:
> Thank you. Upgrading to 4.9.0 did indeed do the trick ! :)
>
> Also thanks for calling it a "compiler bug".
Crashing at compile time is *ALWAYS* a compiler bug. If it were
something wrong with your code, the compiler should XXXXX and say that
there is something wrong with your code - not just crash.
This is actually a requirement of the Fortran standard (but even if it
weren't required by the standard, it would be required by "normal user
requirements").
From section 1.5 of the standard
"A processor [that's basically standard-speak for compiler
- the distinction isn't relevant for this porpose] conforms
to this standard if
[...]
(8) It contains the capability to detect and report the reason for
rejecting a submitted program."
Note that run-time errors are completely different. There are
innumerable ways that your program might crash at run time, and 99% of
the time, it is the programmer's "fault". But a compile-time crash is a
compiler bug.
--
Richard Maine | Good judgment comes from experience;
email: my first.last at org.domain| experience comes from bad judgment.
org: nasa, domain: gov | -- Mark Twain
| |
| Mathemagician 2006-03-29, 8:01 am |
| Could someone give me links to where I can get these other compilers
g95, Lahey and Compaq Visual Fortran, as well as to the mentioned
Fortran standard ?
Thank you
Rainer
| |
|
|
| Mathemagician 2006-03-29, 8:01 am |
| Thank you :)
|
|
|
|
|