Messages in this thread | | | Date | Fri, 25 Aug 2000 02:03:14 -0700 | From | Mitchell Blank Jr <> | Subject | Re: SCO: "thread creation is about a thousand times faster than on native Linux" |
| |
Werner Almesberger wrote: > Where's the advantage over just making ordinary signal delivery do the > right thing ? The code's the same, and you don't give user space any > added flexibility.
Well from what I understand, people want to be able to signal threads individually and also signal the thread group with POSIX semantics. To me it seems the cleanest implementation would be to have an extra thread for accepting the signals for the task group. That way the logic can stay out of the hair of the non-pthread things (and could even be CONFIG_'ed out relatively painlessly for embedded uses)
It could certainly be done either way, I just suspect that having a kernel thread looping on "schedule()" and coping with signals would be easy.
Another advantage is that you wouldn't need sys_tgkill() - just have getpid() return the thread group master's PID, since it can be shot at with a normal sys_kill().
-Mitch - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
| |