On 09-10-24 08:50, Konstantin Belousov wrote:
Perhaps you can check for the presence of the symbol exit@FBSD_1.0 in the
backtrace to determine the situation.
We don't read .gnu.version and the version number hasn't changed as far
as I can see.
I can check osreldate to get an idea of the version.
However this looks like it is going to be a lot more difficult. I need
to be able to tell apart abnormal termination (where no locks are to be
expected) and normal termination (one lock expected).
If it's not possible/too difficult to work out how mnay locks should be
allowed on exit another option will be to try to fiddle around with what
happens during a call to exit(). That would mean something like ignoring
the lock count when called from exit().
And the last resort will be to just turn this check off.
A+
Paul
