[qmtest] [PATCH] Use processes instead of threads

[Mark Mitchell]

Zack Weinberg wrote:

>It comes up in old (pre-job control) shells which would do
>backgrounding partially by blocking SIGINT in the child process.  I am
>not particularly concerned about this either, but it does seem more
>abstractly correct to me to record the signal mask at startup (before
>creating any threads) and restore that after fork.
>
Yes.

What I would do is have the QMTest core save the signal mask at startup 
in a global variable.  Then, the test classes would pass the value into 
executable.py.

As for NSIG, yes it's standard.  But, since we'll actually be doing is 
saving and restoring the same mask, I think all njs has to do is keep 
around a totally opaque sigset_t, for the moment.   It may be just as 
easy to use a vector of booleans, which would be the obvious 
representation for Python client code.

-- 
Mark Mitchell
CodeSourcery, LLC
(916) 791-8304
mark at codesourcery.com