[Haskell-cafe] Debugging cause of indefinite thread blocking

2010-07-06 Thread Ben Lever 
Hi all,

I am making use of the Data.Array.Repa module to achieve data-parallelism. On 
running my program I get the error:

"thread blocked indefinitely on an MVar operation"

Two questions:
1. What could be some of the potential causes for the above error when the only 
use of parallelism is Repa arrays?
2. What are the best strategies for debugging he cause(s) of such an error?

Cheers,
Ben.

The information in this e-mail may be confidential and subject to legal 
professional privilege and/or copyright. National ICT Australia Limited accepts 
no liability for any damage caused by this email or its attachments.
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe

Re: [Haskell-cafe] Debugging cause of indefinite thread blocking

2010-07-07 Thread Claus Reinke 
I am making use of the Data.Array.Repa module to achieve data-parallelism. 
On running my program I get the error:


"thread blocked indefinitely on an MVar operation"


Haven't seen any responses yet, so here are some suggestions:


Two questions:
1. What could be some of the potential causes for the above error when the 
only use of parallelism is Repa arrays?


If you're sure the issue is in Repa, contact the Repa authors?

2. What are the best strategies for debugging he cause(s) of such an 
error?


If it is in your code, you could try replacing the Control.Concurrent
operations with variants that generate a helpful trace before calling
the originals. That approach was used in the Concurrent Haskell
Debugger: http://www.informatik.uni-kiel.de/~fhu/chd/

Google for "concurrent haskell debugger" to find related and
follow-on work on extensions of chd and alternative verification
approaches.

Newer GHCs have some internal runtime event logging features
(eg, link the code with -debug, then run with +RTS -Ds). See RTS
options in the GHC user manual (4.15.[67]).

Claus



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