On 19 August 2006 23:28, James R. Phillips wrote:
> I've built a test version of octave 2.1.73 on cygwin using the experimental
> gcc-3.4.4-2 compiler. Thanks to Brian Dessent for pointing out a packaging
> problem with the compiler and a workaround:
> http://www.cygwin.com/ml/cygwin/2006-07/msg00620.html .
Thanks for helping test the new compiler package!
> The build is successful, but several of the tests executed during "make
> check" issue error messages like the following:
>
> Running
> /home/dad/sources/octave/octave-2.1.73/test/octave.test/quad/quad.exp ...
> 102 [main] octave 1252 fork: child -1 - died waiting for longjmp before
> initialization, retry 0, exit code
> 0x80, errno 11
>
> Interestingly, none of these tests actually fail in the summary report.
>
> The only hits I get with Google on this error message ["died waiting for
> longjmp before initialization, retry 0"] relate to relatively recent issues
> with the cygwin dll. I'm not sure how to classify this issue, i.e. whether
> it is a cygwin issue, a compiler issue, or an octave issue. Also I'm not
> sure how serious it is, since the tests it occurred on appeared to complete
> successfully.
>
> Any help with analysis of these questions would be appreciated.
Change one parameter at a time. Keep all others constant. See what
changes.
First thing would be to experiment with a few different versions of the
cygwin dll, including the most recent snapshot. Re-run "make check" in the
exact same object dir without rebuilding the application. Then, keeping the
dll constant, try rolling your gcc install back to 3.4.4-1, building in a
fresh object dir, and running make check there. Please let us know what
results you get.
cheers,
DaveK
--
Can't think of a witty .sigline today....
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
