Re: Porting vfork()

Perry Harrington (pedward@sun4.apsoft.com)
Thu, 7 Jan 1999 22:30:48 -0800 (PST)

>
> Date: Fri, 8 Jan 1999 01:06:58 -0500 (EST)
> From: Kenneth Albanowski <kjahds@kjahds.com>
>
> On Thu, 7 Jan 1999, Perry Harrington wrote:
>
> > I discussed this issue with Linus as well, and concluded that
> > vforking within a cloned process can be achieved via putting the
> > sleep_on in the task structure. I'm going to be working on this
> > tonight and I'll see what I can do.
>
> I do think a sempahore would be appropriate to avoid the
> possibility of recursion.
>
> Last time I checked BSD doesn't even fully handle the recursive
> vfork() case, and thus we should not need to either. Just make sure
> the kernel retains it's own internal integrity in such a case, nothing
> more.

For the time being I added another wait queue to struct task, and I do
a sleep_on in the parent. When the child exits, in do_exit, just after
notify_parent, it does a wake_up on current->p_pptr->vfork_sleep.

I think that most of the handling could be eliminated if we simply
reused the chldexit queue that wait4 uses. Essentially forking then
waiting, but wait it signal interruptable.

The reason I did it the way I did it, is so you could vfork from a
cloned thread, and theoretically a recursive vfork should work without
error.

I haven't tested it yet, just got all the code in and built a kernel, next
I'll write a test proggy.

Just to be clear, I'd need to do a SYS_190() to call it, right? If not,
how do you make direct syscalls?

>
> Later,
> David S. Miller
> davem@dm.cobaltmicro.com
>

--Perry

-- 
Perry Harrington       Linux rules all OSes.    APSoft      ()
email: perry@apsoft.com 			Think Blue. /\


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/