[Haskell-cafe] Threadscope 0.2.2 goes in segmentation fault on Mac Os X 10.8.3

[Alfredo Di Napoli]

Hi Cafè,

I've tried installing threadscope, but when I run it from console with "threadscope" I get a laconic segmentation fault.

Some info to help the debugging:

* Installed gtk via brew

* gtk-demo runs correctly

* I'm using Mac Os X 10.8.3

* Running gtk and threadscope through a virtual environment (provided by hsenv)

* Using XQuartz 2.7.4

I know it's a bit difficult to debug this way, I can try debugging with gdb if it can help.

Cheers,

A.

[John Wiegley]

>>>>> Alfredo Di Napoli <alfredo....@gmail.com> writes:

> I know it's a bit difficult to debug this way, I can try debugging with gdb
> if it can help.

Yes, can you show us a backtrace from gdb, and also look in your CrashReports
log folder to see if it gives a bit more information on the state of the
process at the time it died?

Thanks,
--
John Wiegley
FP Complete Haskell tools, training and consulting
http://fpcomplete.com johnw on #haskell/irc.freenode.net

_______________________________________________
Haskell-Cafe mailing list
Haskel...@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe

[Alfredo Di Napoli]

Fair enough :)

Here is the gdb output:

(gdb) run

Starting program: /Users/adinapoli/Library/Haskell/ghc-7.6.2/bin/threadscope

Reading symbols for shared libraries ++++++++++++++++++....................................................................................................................................................................... done

Reading symbols for shared libraries . done

Reading symbols for shared libraries . done

Reading symbols for shared libraries . done

Reading symbols for shared libraries . done

Program received signal EXC_BAD_ACCESS, Could not access memory.

Reason: KERN_INVALID_ADDRESS at address: 0x0000000000000000

0x0000000000000000 in ?? ()

I have two hypothesis:

a) could be the RAM (tips about some RAM testing tool?)

b) could be some programs which is writing in that portion of memory, see:

https://discussions.apple.com/thread/3734077?start=0&tstart=0

[Tobias Müller]

Alfredo Di Napoli <alfredo....@gmail.com> wrote:
> Fair enough :)
> Here is the gdb output:
>
> (gdb) run Starting program:
> /Users/adinapoli/Library/Haskell/ghc-7.6.2/bin/threadscope Reading
> symbols for shared libraries
> ++++++++++++++++++.......................................................................................................................................................................
> done Reading symbols for shared libraries . done Reading symbols for
> shared libraries . done Reading symbols for shared libraries . done
> Reading symbols for shared libraries . done
>
> Program received signal EXC_BAD_ACCESS, Could not access memory. Reason:
> KERN_INVALID_ADDRESS at address: 0x0000000000000000 0x0000000000000000 in ?? ()

That's a NULL-pointer exception.

> I have two hypothesis:
>
> a) could be the RAM (tips about some RAM testing tool?)
> b) could be some programs which is writing in that portion of memory

Definitely an application error. NULL is never a valid memory address.
Missing NULL-pointer checks are a very common error source in low level
programming.

What do you mean by _some_ program? It's the program that you started
(threadscope).

Tobi

[Alfredo Di Napoli]

Hi Tobias,

>
>
> What do you mean by _some_ program? It's the program that you started
> (threadscope).

In a forum I've read that this error could be some third party app (for example one started at login or running as a daemon) which is conflicting and causing the error.
Unlikely, but i've reported the possibility for completeness.

Said that,has someone had any luck in running Threadscope on Mac OS X 10.8 at all?

Thanks,
A.