[RFC][PATCH] Do not set /proc inode->pid for non-pid-related inodes

[Serge E. Hallyn]

Quoting Dave Hansen (hansendc at us.ibm.com):
> On Thu, 2007-03-22 at 09:33 -0500, Serge E. Hallyn wrote:
> > 
> > 
> > I still prefer that we forego that kthread, and just work toward
> > allowing pid1 to exit.  Really I think the crufty /proc/<pid> handling
> > is the only reason we were going to punt on that for now. 
> 
> It's everything _but_ the /proc/<pid> stuff.  /proc/{mem,cpu}info and
> friends are the ones suspiciously tied to pid 1.

Oh, right.  I was looking at the proc_mnt->mnt_sb->s_root->d_inode->pid
being tied to find_get_pid(1) and misreading it.

So you're right it may be even easier than I thought to fix.

-serge