[PATCH] An attempt to have an unlimitedly extendable sys_clone

Cedric Le Goater clg at fr.ibm.com
Mon Jan 21 02:36:38 PST 2008


Hello Pavel !

Pavel Emelyanov wrote:
>> Here's the unshare part if you want to fold that with your patch.
> 
> Thanks, Cedric. I'll pick that up.
> 
>> Thanks,
> 
> OK, thanks everyone, I'll try to cook the final patch today
> and send it to Andrew as an RFC.

As Al said it, it's probably not the direction to follow to extend the 
clone flags :) But the problem being real, I think we need to insist. 

First, we could simplify the API by using child_tidptr to pass a 64bits 
array of flags. This would still be a brutal syscall argument hijack but 
we can't do better without a new syscall.

Next, we could send the same code using a new syscall.

I have time to work on that. 

C.


More information about the Containers mailing list