From: Victor A. Wagner Jr. (vawjr_at_[hidden])
Date: 2004-09-20 03:52:49
At Sunday 2004-09-19 14:15, you wrote:
>Victor A. Wagner Jr. wrote:
> > At Sunday 2004-09-19 13:54, you wrote:
> >>I'm getting regular (but not consistent) exceptions when running the
> >>regression tests
> >>they always seem to happen when dealing with the mingw toolset
> >>"...error at 0x0040b7bc, location 0x00000040 cannot be written"
> > I tried deleting ALL of the .obj files and I get a "new" error this time
> > "...error at 0x7c911f52, location 0x00080173 cannot be read"
>Last time around we determined it was something wrong with the header
>scan caching. So if you absolutely need to have it work try deleting the
> >>How do I create a debug version of bjam so I can chase this down?
>Add a "--debug" option when building it.
ok, did that... and now I've gotten a breakpoint at an assertion failure
which is caused by the erroneous assumption in filent.c that some
is terminated by a '\0'.
it is NOT in this case....
the file (archive) in question is:
since I know next to nothing about this file format, the debugging has
gotten quite slow and it's now 0200 so I'm gonna call it quits for the night.
the file was created a couple days ago in a regression run (it appears from
the dates on the files).
so it's likely NOT bjam, but rather a bad library build. I'll keep you
> >>this is the line in my script that is executing when I get the errors
> >>python regression.py install --toolsets=mingw --bjam-toolset=msvc
>Although I don't know how to pass that to the regression.py, if that's
>how bjam is getting built.
>Yahoo! Groups Links
Victor A. Wagner Jr. http://rudbek.com
The five most dangerous words in the English language:
"There oughta be a law"
Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk