Re: SCO: "thread creation is about a thousand times faster than on native Linux"

From: yodaiken@fsmlabs.com
Date: Thu Aug 24 2000 - 16:02:59 EST


On Thu, Aug 24, 2000 at 11:15:09PM +0200, Werner Almesberger wrote:
> yodaiken@fsmlabs.com wrote:
> > This call shifts delivered signal numbers up so that
> > "special" POSIX signals can be delivered to a process.
>
> Kind of a SIGOVERKILL ... yes, that could work.

Cool name. And it makes a kind of poetic sense that
kill(pid,SIGOVERKILL) should de-POSIX a process.

>
> > Make it a root pthread daemon.
>
> So you need a privileged launcher or VMS-style privileged libraries.
> Oh dear ...

Why not? Better than having it in the kernel. Think of Pthreads executables
as misc type.

> > Make SIG_KILL still work as usual, with the CLONE_PTHREAD flag
> > causing all children to be killed too, but with SUSPEND etc
> > coming in raw.
>
> Hmm, if you can do this, why do you need to special-case the other
> non-blockable signals then ?

SIG_KILL is the simplest case. No synchronization, no wierd I/O to
worry about ...

-- 
---------------------------------------------------------
Victor Yodaiken 
Finite State Machine Labs: The RTLinux Company.
 www.fsmlabs.com  www.rtlinux.com

- 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/



This archive was generated by hypermail 2b29 : Thu Aug 31 2000 - 21:00:14 EST