Re: About cleaning up the thread stuff

[Kevin Ryde]

Marius Vollmer <address@hidden> writes:
>
> Maybe the debugger could still try to figure out whether
> there is a dead-lock,

That'd be nice, to see what each is apparently getting up to.

Speaking of threads, there's more low level things that need to be
reviewed for safety.

        - gethostbyname and friends definitely
        - string->symbol I think needs protection
        - symbol->keyword might be ok, but the SCM_DEFER_INTS
          could become some explicit mutex

I was doing some of the C bits with mutexes, but I'm thinking it may
be better just to demand the "_r" functions whenever threading is
enabled.  If the posix threading funcs exist then so should the _r's,
and those _r's will give maximum concurrency.