sukadev@us.ibm.com writes:

> Ok. I thought they were low priority but required eventually.

If the target is not the current task change is required.  We gain
very little by changing our accessors of the current task.

The important part is that when someone else looks at us we
go through a struct pid and pid_nr so they get the relative view
of what is going on.

The tricky thing is that we find all references to pids.  The well
defined functions are easy.  We can grep for them and see if they
are using anything besides current.

Any time we store a pid we need to store it as a struct pid.

Right now by messing with pid_nr(task_xxx(current)) you are generating
a lot of unnecessary changes, and to some extent hiding where real
change needs to happen.

These things all follow from the fundamentals of what we are doing.

> | I have largely given up on review this patch set until that is fixed.
>
> I am sending out the patches with the noise cancelled :-) Would like
> to send them out to akpm in a few days.

Unfortunately I won't have a chance to do anything until sometime next
week.  I'm off on a short trip without a computer.

Eric
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers