svn commit: r189781 - stable/7/sys/kern

[Bruce Simpson]

Bruce M Simpson wrote:
>   Early MFC, as the test case in the PR is fairly complete and the
>   submitter also re-ran test case on -STABLE. It also bites Python
>   fairly hard, which will otherwise try to use POSIX sems for its
>   internal thread synchronization; this needs more in-depth testing.
>  

Note: FreeBSD's build of Python does NOT attempt to use POSIX sems by 
default, it
will only do this if explicitly patched to do so. It defaults to using 
condvars + mutexes
to build Python's internal thread locks.

I'd encourage folk to try the patch which does this,
I only posted the 'side-step' patch (i.e. use GNU Pth, needs to be 
explicitly
enabled in OPTIONS) but the changes needed are in here too:
    http://people.freebsd.org/~bms/dump/python26-fbsd-pth.patch

cheers
BMS