Bug report (was: Mail delivery error)

Mon, 31 Mar 1997 03:11:32 +0200 (MET DST)

According to Olivier Lecarme:
> Following the indications in the GPC manual, I tried sending a bug
> report to bug-gpc@prep.ai.mit.edu, which does not exist. As a last
> resort, I'm trying the GPC mailing list, hoping it exists!

It does exist. :-)

>    ----- The following addresses had permanent fatal errors -----
> 
>     (expanded from: )

In fact we have a bug reporting system:   See the page "bugs" on
our WWW site,

    http://home.pages.de/~gnu-pascal/

but actually most bug reports go through this list, so you did it
right.  I expect that the mail reporting system will become more
useful when GPC is getting more stable and we will fine-tune the
compiler.

> I get the following message when compiling a somewhat complicated
> program (2352 lines):
> 
> gpc: Internal compiler error: program gpc1 got fatal signal 6
> 
> What can I do? Send you the whole program would probably be useless. Are
> there some options I could trigger in order to get more information?

Alpha versions of GPC have an option "-dY" which tells GPC to echo
each source line to stderr.  Then you can see where in the source GPC
crashes.

Another interesting option in this context is "-dy" which lets GPC
output parsing debugging information.  If you post the last "rule"
listed there, e.g.

    Reducing via rule 616 (line 3782),
    variable_or_function_access_maybe_assignment rest_of_statement
    -> assignment_or_call_statement

it might already tell me something about the error.  (The line number
refers to the GPC source file `gpc-parse.y'.)

> By the way, I encountered another problem with GPC: one program writes a
> file of some complicated record type; another program reads this file,
> declared in exactly the same way. However, it seems that every get(file)
> skips to the next *physical* record, instead of the next logical one.

This I don't understand ... could you please explain what you mean by
"physical" and "logical"?

Perhaps it's an alignment problem?  Within records, GPC aligns Integers
on 4-byte boundaries, etc.

Greetings,

    Peter

  Dipl.-Phys. Peter Gerwinski, Essen, Germany, free physicist and programmer
peter.gerwinski@uni-essen.de - http://home.pages.de/~peter.gerwinski/ [970201]
       maintainer GNU Pascal - http://home.pages.de/~gnu-pascal/ [970125]

Peter Gerwinski (peter@agnes.dida.physik.uni-essen.de)