Date: Thu, 3 Aug 2017 12:29:59 -0700
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: Michal Hocko <mhocko@kernel.org>
Cc: linux-kernel@vger.kernel.org, stable@vger.kernel.org,
        Michael Ellerman <mpe@ellerman.id.au>,
        Nicholas Piggin <npiggin@gmail.com>
Subject: Re: [PATCH 4.4 66/90] powerpc/numa: Fix percpu allocations to be
 NUMA aware
Message-ID: <20170803192959.GA2553@kroah.com>
References: <20170612152556.133240249@linuxfoundation.org>
 <20170612152600.887366512@linuxfoundation.org>
 <20170728135334.GC2278@dhcp22.suse.cz>
 <20170728224147.GB22460@kroah.com>
 <20170731064115.GA13036@dhcp22.suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20170731064115.GA13036@dhcp22.suse.cz>
User-Agent: Mutt/1.8.3 (2017-05-23)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1446
Lines: 31

On Mon, Jul 31, 2017 at 08:41:16AM +0200, Michal Hocko wrote:
> On Fri 28-07-17 15:41:47, Greg KH wrote:
> > On Fri, Jul 28, 2017 at 03:53:35PM +0200, Michal Hocko wrote:
> > > JFYI. We have encountered a regression after applying this patch on a
> > > large ppc machine. While the patch is the right thing to do it doesn't
> > > work well with the current vmalloc area size on ppc and large machines
> > > where NUMA nodes are very far from each other. Just for the reference
> > > the boot fails on such a machine with bunch of warning preceeding it.
> > > See http://lkml.kernel.org/r/20170724134240.GL25221@dhcp22.suse.cz
> > > 
> > > It seems the right thing to do is to enlarge the vmalloc space on ppc
> > > but this is not the case in the upstream kernel yet AFAIK. It is also
> > > questionable whether that is a stable material but I will decision on
> > > you here.
> > > 
> > > We have reverted this patch from our 4.4 based kernel.
> > 
> > But all is fine on newer kernels?  That is odd.
> 
> Newer kernels do not have enlarged vmalloc space yet AFAIK so they won't
> work properly eiter. This bug is quite rare though because you need a
> specific HW configuration to trigger the issue - namely NUMA nodes have
> to be far away from each other in the physical memory space.
> 
> > I'll be glad to drop it, but should it be dropped from all stable trees?
> 
> Yes from all stable backports.

Ok, now all reverted, thanks.

greg k-h