Date: Fri, 11 Aug 2017 16:24:57 +0200
From: Michal Hocko <mhocko@kernel.org>
To: Florian Weimer <fweimer@redhat.com>
Cc: Colm =?iso-8859-1?Q?MacC=E1rthaigh?= <colm@allcosts.net>,
        Kees Cook <keescook@chromium.org>,
        Mike Kravetz <mike.kravetz@oracle.com>, Rik van Riel <riel@redhat.com>,
        Will Drewry <wad@chromium.org>, akpm@linux-foundation.org,
        dave.hansen@intel.com, kirill@shutemov.name, linux-api@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org, luto@amacapital.net,
        mingo@kernel.org
Subject: Re: [PATCH v2 0/2] mm,fork,security: introduce MADV_WIPEONFORK
Message-ID: <20170811142457.GP30811@dhcp22.suse.cz>
References: <20170807134648.GI32434@dhcp22.suse.cz>
 <1502117991.6577.13.camel@redhat.com>
 <20170810130531.GS23863@dhcp22.suse.cz>
 <CAAF6GDc2hsj-XJj=Rx2ZF6Sh3Ke6nKewABXfqQxQjfDd5QN7Ug@mail.gmail.com>
 <20170810153639.GB23863@dhcp22.suse.cz>
 <CAAF6GDeno6RpHf1KORVSxUL7M-CQfbWFFdyKK8LAWd_6PcJ55Q@mail.gmail.com>
 <20170810170144.GA987@dhcp22.suse.cz>
 <CAAF6GDdFjS612mx1TXzaVk1J-Afz9wsAywTEijO2TG4idxabiw@mail.gmail.com>
 <20170811140653.GO30811@dhcp22.suse.cz>
 <c8cda773-b28d-f35f-7f18-6735584cb173@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <c8cda773-b28d-f35f-7f18-6735584cb173@redhat.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1134
Lines: 24

On Fri 11-08-17 16:11:44, Florian Weimer wrote:
> On 08/11/2017 04:06 PM, Michal Hocko wrote:
> 
> > I am sorry to look too insisting here (I have still hard time to reconcile
> > myself with the madvise (ab)use) but if we in fact want minherit like
> > interface why don't we simply add minherit and make the code which wants
> > to use that interface easier to port? Is the only reason that hooking
> > into madvise is less code? If yes is that a sufficient reason to justify
> > the (ab)use of madvise? If there is a general consensus on that part I
> > will shut up and won't object anymore. Arguably MADV_DONTFORK would fit
> > into minherit API better as well.
> 
> It does, OpenBSD calls it MAP_INHERIT_NONE.
> 
> Could you implement MAP_INHERIT_COPY and MAP_INHERIT_SHARE as well?  Or
> is changing from MAP_SHARED to MAP_PRIVATE and back impossible?

I haven't explored those two very much. Their semantic seems rather
awkward, especially map_inherit_share one. I guess MAP_INHERIT_COPY
would be doable. Do we have to support all modes or a missing support
would disqualify the syscall completely?
-- 
Michal Hocko
SUSE Labs