Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.c++ > #4611 > unrolled thread

unkwon exception throwing

Started byPhilipp Kraus <philipp.kraus@flashpixx.de>
First post2011-05-03 12:40 +0200
Last post2011-05-03 15:11 +0200
Articles 10 — 7 participants

Back to article view | Back to comp.lang.c++


Contents

  unkwon exception throwing Philipp Kraus <philipp.kraus@flashpixx.de> - 2011-05-03 12:40 +0200
    Re: unkwon exception throwing gwowen <gwowen@gmail.com> - 2011-05-03 05:08 -0700
      Re: unkwon exception throwing Philipp Kraus <philipp.kraus@flashpixx.de> - 2011-05-03 14:27 +0200
        Re: unkwon exception throwing Öö Tiib <ootiib@hot.ee> - 2011-05-03 06:30 -0700
        Re: unkwon exception throwing Juha Nieminen <nospam@thanks.invalid> - 2011-05-03 14:29 +0000
          Re: unkwon exception throwing James Kanze <james.kanze@gmail.com> - 2011-05-03 14:57 -0700
            Re: unkwon exception throwing Juha Nieminen <nospam@thanks.invalid> - 2011-05-04 13:41 +0000
              Re: unkwon exception throwing Marc <marc.glisse@gmail.com> - 2011-05-04 16:38 +0000
    Re: unkwon exception throwing "Fred Zwarts" <F.Zwarts@KVI.nl> - 2011-05-03 14:41 +0200
      Re: unkwon exception throwing Philipp Kraus <philipp.kraus@flashpixx.de> - 2011-05-03 15:11 +0200

#4611 — unkwon exception throwing

FromPhilipp Kraus <philipp.kraus@flashpixx.de>
Date2011-05-03 12:40 +0200
Subjectunkwon exception throwing
Message-ID<ipom39$juf$1@online.de>
Hello,

I have written a numerical algorithm for optimizing and on large input 
data size an exception with "program fault" is thrown, but I don't have 
written this exception, so I would like to understand, how can I solve 
the error(s). What's the meaning of this exception (max stack / heap 
size is reached)? I know the method, which is throw the exception, but 
I don't know why the exception is created

Thanks a lot

Phil

[toc] | [next] | [standalone]


#4612

Fromgwowen <gwowen@gmail.com>
Date2011-05-03 05:08 -0700
Message-ID<4e469414-903c-467b-9526-74dee814bb0b@w24g2000yqb.googlegroups.com>
In reply to#4611
On May 3, 11:40 am, Philipp Kraus <philipp.kr...@flashpixx.de> wrote:
> Hello,
>
> I have written a numerical algorithm for optimizing and on large input
> data size an exception with "program fault" is thrown, but I don't have
> written this exception, so I would like to understand, how can I solve

Without any information on your platform, and compiler (at the very
least) no-one will be able to help, I'm afraid.

[toc] | [prev] | [next] | [standalone]


#4613

FromPhilipp Kraus <philipp.kraus@flashpixx.de>
Date2011-05-03 14:27 +0200
Message-ID<iposbk$ote$1@online.de>
In reply to#4612
On 2011-05-03 14:08:59 +0200, gwowen said:

> On May 3, 11:40 am, Philipp Kraus <philipp.kr...@flashpixx.de> wrote:
>> Hello,
>> 
>> I have written a numerical algorithm for optimizing and on large input
>> data size an exception with "program fault" is thrown, but I don't have
>> written this exception, so I would like to understand, how can I solve
> 
> Without any information on your platform, and compiler (at the very
> least) no-one will be able to help, I'm afraid.

Compiler is g++ under OSX 10.5 & 10.6 

[toc] | [prev] | [next] | [standalone]


#4618

FromÖö Tiib <ootiib@hot.ee>
Date2011-05-03 06:30 -0700
Message-ID<27fa3665-c8d0-45c7-815f-0e88f6b85437@a18g2000yqj.googlegroups.com>
In reply to#4613
On May 3, 3:27 pm, Philipp Kraus <philipp.kr...@flashpixx.de> wrote:
> On 2011-05-03 14:08:59 +0200, gwowen said:
>
> > On May 3, 11:40 am, Philipp Kraus <philipp.kr...@flashpixx.de> wrote:
> >> Hello,
>
> >> I have written a numerical algorithm for optimizing and on large input
> >> data size an exception with "program fault" is thrown, but I don't have
> >> written this exception, so I would like to understand, how can I solve

Try to describe your problem exactly and write down the messages
letter-by-letter since
1) you are software developer not aunt Annie and
2) most of us can not read minds.

> > Without any information on your platform, and compiler (at the very
> > least) no-one will be able to help, I'm afraid.
>
> Compiler is g++ under OSX 10.5 & 10.6

You tried it on two Macs? This is about 99.9% not OSX or compiler
problem but defect in your source code.

It feels that your running software is trying to tell you that it
throws unhandled exceptions and so it has to close. The message (that
you fail to reproduce here) likely tells that it is program's (as
opposed to its user's) fault that it closes.

You have to debug your program (run it in debugger or use post-mortem
debugger) to see what exception and from where has been actually
thrown. Otherwise we can't help.

[toc] | [prev] | [next] | [standalone]


#4621

FromJuha Nieminen <nospam@thanks.invalid>
Date2011-05-03 14:29 +0000
Message-ID<4dc01141$0$2840$7b1e8fa0@news.nbl.fi>
In reply to#4613
Philipp Kraus <philipp.kraus@flashpixx.de> wrote:
> Compiler is g++ under OSX 10.5 & 10.6 

  Compile the program with the -g compiler option and then run the
program with gdb (iow. "gdb ./yourprogram" and then "run"), and if/when
it crashes it should show the line where it does. (If the line is
nondescriptive eg. because it's in some system library, write "bt"
to get a backtrace. The first line in the backtrace which is in your
own code is usually a good indicative.)

[toc] | [prev] | [next] | [standalone]


#4648

FromJames Kanze <james.kanze@gmail.com>
Date2011-05-03 14:57 -0700
Message-ID<1871ed9e-573d-4e74-914b-178a795831af@c41g2000yqm.googlegroups.com>
In reply to#4621
On May 3, 3:29 pm, Juha Nieminen <nos...@thanks.invalid> wrote:
> Philipp Kraus <philipp.kr...@flashpixx.de> wrote:
> > Compiler is g++ under OSX 10.5 & 10.6

>   Compile the program with the -g compiler option and then run the
> program with gdb (iow. "gdb ./yourprogram" and then "run"), and if/when
> it crashes it should show the line where it does. (If the line is
> nondescriptive eg. because it's in some system library, write "bt"
> to get a backtrace. The first line in the backtrace which is in your
> own code is usually a good indicative.)

Except when it's not:-).  Corrupting the free space arena is a
frequent cause of such problems, and of course, the "crash"
occurs in an allocation totally unrelated to where the error
was.

(Such problems are really fun to debug:-).  To top it off, the
last time I had to attack one, 1) I'd just entered a new
company, and had never seen the half a million lines of code
before I was asked to debug it, and 2) the actual problem was
due to the compiler generating wrong code in a tight loop.  Both
very exceptional cases, which multiply the difficulty by several
orders of magnitude.  On the other hand, without such difficult
problems, I'd be hard pressed to justify how much I'm paid, so I
won't complain too much.)

--
James Kanze

[toc] | [prev] | [next] | [standalone]


#4672

FromJuha Nieminen <nospam@thanks.invalid>
Date2011-05-04 13:41 +0000
Message-ID<4dc1578a$0$2871$7b1e8fa0@news.nbl.fi>
In reply to#4648
James Kanze <james.kanze@gmail.com> wrote:
> Except when it's not:-).  Corrupting the free space arena is a
> frequent cause of such problems, and of course, the "crash"
> occurs in an allocation totally unrelated to where the error
> was.

  valgrind is a good tool to find such problems. I don't know if it's
available for Mac, though.

[toc] | [prev] | [next] | [standalone]


#4682

FromMarc <marc.glisse@gmail.com>
Date2011-05-04 16:38 +0000
Message-ID<iprve8$7pm$1@news-rocq.inria.fr>
In reply to#4672
Juha Nieminen  wrote:

>   valgrind is a good tool to find such problems. I don't know if it's
> available for Mac, though.

It is, but the port isn't as mature as the linux one.

[toc] | [prev] | [next] | [standalone]


#4615

From"Fred Zwarts" <F.Zwarts@KVI.nl>
Date2011-05-03 14:41 +0200
Message-ID<ipot5f$83j$1@news.albasani.net>
In reply to#4611
"Philipp Kraus" <philipp.kraus@flashpixx.de> wrote in message 
news:ipom39$juf$1@online.de...
> Hello,
>
> I have written a numerical algorithm for optimizing and on large input 
> data size an exception with "program fault" is thrown, but I don't have 
> written this exception, so I would like to understand, how can I solve the 
> error(s). What's the meaning of this exception (max stack / heap size is 
> reached)? I know the method, which is throw the exception, but I don't 
> know why the exception is created

What exactly do you mean with an 'exception with "program fault"'?
Is it an exception of type string with value "program fault" that is thrown?
Probably, no one, except the author of this exception, knows what it means.

If you know the method in which the exception is thrown
and you can show us the code, maybe someone does understand
under which conditions the exception is thrown. 

[toc] | [prev] | [next] | [standalone]


#4616

FromPhilipp Kraus <philipp.kraus@flashpixx.de>
Date2011-05-03 15:11 +0200
Message-ID<ipouts$r2i$1@online.de>
In reply to#4615
On 2011-05-03 14:41:19 +0200, Fred Zwarts said:

> 
> "Philipp Kraus" <philipp.kraus@flashpixx.de> wrote in message 
> news:ipom39$juf$1@online.de...
>> Hello,
>> 
>> I have written a numerical algorithm for optimizing and on large input 
>> data size an exception with "program fault" is thrown, but I don't have 
>> > written this exception, so I would like to understand, how can I 
>> solve the > error(s). What's the meaning of this exception (max stack / 
>> heap size is > reached)? I know the method, which is throw the 
>> exception, but I don't know why the exception is created
> 
> What exactly do you mean with an 'exception with "program fault"'?
> Is it an exception of type string with value "program fault" that is thrown?
> Probably, no one, except the author of this exception, knows what it means.
> 
> If you know the method in which the exception is thrown
> and you can show us the code, maybe someone does understand
> under which conditions the exception is thrown.

I've found the exception, it comes from an external component (but it 
was not documentated)

[toc] | [prev] | [standalone]


Back to top | Article view | comp.lang.c++


csiph-web