Re: [PATCH] mm: mempolicy: turn vma_set_policy() into vma_dup_policy()

From: Andrew Morton <akpm@linux-foundation.org>
To: David Rientjes <rientjes@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>,
	KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
	Mel Gorman <mgorman@suse.de>, Rik van Riel <riel@redhat.com>,
	Andi Kleen <andi@firstfloor.org>,
	linux-kernel@vger.kernel.org,
	Rusty Russell <rusty@rustcorp.com.au>
Subject: Re: [PATCH] mm: mempolicy: turn vma_set_policy() into vma_dup_policy()
Date: Thu, 11 Jul 2013 15:33:13 -0700	[thread overview]
Message-ID: <20130711153313.7e37a9095eacb1153ade516c@linux-foundation.org> (raw)
In-Reply-To: <alpine.DEB.2.02.1307111524400.2458@chino.kir.corp.google.com>

On Thu, 11 Jul 2013 15:27:46 -0700 (PDT) David Rientjes <rientjes@google.com> wrote:

> On Thu, 11 Jul 2013, Andrew Morton wrote:
> 
> > On Thu, 11 Jul 2013 15:13:03 -0700 (PDT) David Rientjes <rientjes@google.com> wrote:
> > 
> > > > +int vma_dup_policy(struct vm_area_struct *src, struct vm_area_struct *dst)
> > > > +{
> > > > +	struct mempolicy *pol = mpol_dup(vma_policy(src));
> > > > +
> > > > +	if (IS_ERR(pol))
> > > > +		return PTR_ERR(pol);
> > > 
> > > PTR_ERR() returns long, so vma_dup_policy() needs to return long.
> > 
> > hm, really?  vma_dup_policy() returns an errno and errnos have type "int".
> > 
> > Arguably it is PTR_ERR() which returns the wrong type...
> > 
> 
> PTR_ERR() may not imply IS_ERR(),

Well why not.  Are we saying that code can legitimately convert the
PTR_ERR() return value back into a pointer?  If so that sounds nuts.

> which I believe Rusty is trying to fix 
> with his series that fixes up abuses of PTR_ERR().

Well I hope all of this will be completely documented in err.h. 
Otherwise the abuses will simply continue.