From: Michael Glassford (glassfordm_at_[hidden])
Date: 2004-01-12 14:17:13
> An other possible solution would be to reconsider my version that can
> do without an explicit DLL. (Michael Glassford, are you listening?
> the thread monitor still necessary.)
As I mentioned before, I'm always listening.
I'm also very interested in the static/dynamic linking issue and in looking
at your pseudo-dll approach, but I have other Boost.Thread priorities to get
to first and, unfortunately, limited time for dealing with Boost.Thread.
Did you ever look into the issues you mentioned with your pseudo-dll (quoted
below from another thread)?
> I know that there are some things that will need more consideration.
> Robustness of temporary file path, possible race conditions,
> (or speed?) to name a few that come to mind. Also issues of putting
> the stub file compilation into bjam build is not yet obvious to me.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk