#2192: occasional segmentation faults when using Control.Parallel
-----------------------+----------------------------------------------------
Reporter: ivant | Owner:
Type: bug | Status: new
Priority: normal | Component: Compiler
Version: 6.8.2 | Severity: normal
Keywords: | Testcase:
Architecture: x86 | Os: Linux
-----------------------+----------------------------------------------------
The attached program fails quite often with segmentation faults or
internal errors (unpredictably, in consecutive runs, without changing the
source):
{{{
% time ./parallel +RTS -N2
zsh: segmentation fault (core dumped) ./parallel +RTS -N2
./parallel +RTS -N2 2.79s user 0.42s system 107% cpu 2.988 total
% time ./parallel +RTS -N2
parallel: internal error: evacuate: strange closure type 65284
(GHC version 6.8.2 for i386_unknown_linux)
Please report this as a GHC bug:
http://www.haskell.org/ghc/reportabug
zsh: abort (core dumped) ./parallel +RTS -N2
./parallel +RTS -N2 3.60s user 0.51s system 108% cpu 3.781 total
% time ./parallel +RTS -N2
zsh: segmentation fault (core dumped) ./parallel +RTS -N2
./parallel +RTS -N2 5.09s user 0.46s system 111% cpu 4.982 total
% time ./parallel +RTS -N2
500000
./parallel +RTS -N2 8.34s user 0.14s system 108% cpu 7.796 total
}}}
The program can fail both when it is compiled with and without -O2.
Changing the value of maxSparks makes a difference (with value 2 it is
finishing successfully always (or probably more often)).
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/2192>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
