From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wei Liu Subject: Re: [PATCH XEN v8 24/29] tools/libs/call: linux: touch newly allocated pages after madvise lockdown Date: Tue, 19 Jan 2016 14:26:25 +0000 Message-ID: <20160119142625.GF1691@citrix.com> References: <1452864168.32341.97.camel@citrix.com> <1452864188-2417-1-git-send-email-ian.campbell@citrix.com> <1452864188-2417-25-git-send-email-ian.campbell@citrix.com> <20160119132420.GZ1691@citrix.com> <1453210855.29930.36.camel@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Return-path: Content-Disposition: inline In-Reply-To: <1453210855.29930.36.camel@citrix.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Campbell Cc: ian.jackson@eu.citrix.com, Roger Pau =?iso-8859-1?Q?Monn=E9?= , Wei Liu , xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org On Tue, Jan 19, 2016 at 01:40:55PM +0000, Ian Campbell wrote: > On Tue, 2016-01-19 at 13:24 +0000, Wei Liu wrote: > > On Fri, Jan 15, 2016 at 01:23:03PM +0000, Ian Campbell wrote: > > > This avoids a potential issue with a fork after allocation but before > > > madvise. > > > = > > > Signed-off-by: Ian Campbell > > > --- > > > v7: New, replacing "tools/libs/call: linux: avoid forking between mmap > > > =A0=A0=A0=A0and madvise". > > > --- > > > =A0tools/libs/call/linux.c | 14 +++++++++++++- > > > =A01 file changed, 13 insertions(+), 1 deletion(-) > > > = > > > diff --git a/tools/libs/call/linux.c b/tools/libs/call/linux.c > > > index 3641e41..651f380 100644 > > > --- a/tools/libs/call/linux.c > > > +++ b/tools/libs/call/linux.c > > = > > I didn't notice you only handled this for Linux until now. > > = > > I think FreeBSD and NetBSD need similar treatment, too? But then current > > BSD* code doesn't even support DONTFORK in madvise. > = > I think any updates to the *BSD side should come as separate improvements. > = > If they are necessary at all, the Linux stuff comes from Linux making use > of particular behaviours on mlock()'s memory (specifically, IIRC, doing > things such as THP and NUMA balancing which can cause page faults on the > mlock'd pages despite them being locked) which are (possibly) allowed by > POSIX, but *BSD may not take the same interpretation. > = > > Adding Roger for more input. > > = > > The changes in this patch look fine to me. > = > May I take that as an Ack? > = Yes.