Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 28 Apr 2020 13:25:14 -0400
From:   Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To:     =?iso-8859-1?Q?J=FCrgen_Gro=DF?= <jgross@suse.com>,
        joe.jin@oracle.com
Cc:     Peng Fan <peng.fan@nxp.com>,
        "boris.ostrovsky@oracle.com" <boris.ostrovsky@oracle.com>,
        "sstabellini@kernel.org" <sstabellini@kernel.org>,
        "xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>,
        "iommu@lists.linux-foundation.org" <iommu@lists.linux-foundation.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        dl-linux-imx <linux-imx@nxp.com>
Subject: Re: [PATCH] xen/swiotlb: correct the check for
 xen_destroy_contiguous_region
Message-ID: <20200428172514.GA24178@char.us.oracle.com>
References: <1588059225-11245-1-git-send-email-peng.fan@nxp.com>
 <1c01e97a-adcd-a703-55b5-8975b4ce4d2c@suse.com>
 <DB6PR0402MB2760A05135338B0CBB28123488AC0@DB6PR0402MB2760.eurprd04.prod.outlook.com>
 <dba804ea-4268-24ff-7447-ddef00e9e20c@suse.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <dba804ea-4268-24ff-7447-ddef00e9e20c@suse.com>
User-Agent: Mutt/1.9.1 (2017-09-22)
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Apr 28, 2020 at 12:19:41PM +0200, J=FCrgen Gro=DF wrote:
> On 28.04.20 10:25, Peng Fan wrote:

Adding Joe Jin.

Joe, didn't you have some ideas on how this could be implemented?

> > > Subject: Re: [PATCH] xen/swiotlb: correct the check for
> > > xen_destroy_contiguous_region
> > >=20
> > > On 28.04.20 09:33, peng.fan@nxp.com wrote:
> > > > From: Peng Fan <peng.fan@nxp.com>
> > > >=20
> > > > When booting xen on i.MX8QM, met:
> > > > "
> > > > [    3.602128] Unable to handle kernel paging request at virtual =
address
> > > 0000000000272d40
> > > > [    3.610804] Mem abort info:
> > > > [    3.613905]   ESR =3D 0x96000004
> > > > [    3.617332]   EC =3D 0x25: DABT (current EL), IL =3D 32 bits
> > > > [    3.623211]   SET =3D 0, FnV =3D 0
> > > > [    3.626628]   EA =3D 0, S1PTW =3D 0
> > > > [    3.630128] Data abort info:
> > > > [    3.633362]   ISV =3D 0, ISS =3D 0x00000004
> > > > [    3.637630]   CM =3D 0, WnR =3D 0
> > > > [    3.640955] [0000000000272d40] user address but active_mm is
> > > swapper
> > > > [    3.647983] Internal error: Oops: 96000004 [#1] PREEMPT SMP
> > > > [    3.654137] Modules linked in:
> > > > [    3.677285] Hardware name: Freescale i.MX8QM MEK (DT)
> > > > [    3.677302] Workqueue: events deferred_probe_work_func
> > > > [    3.684253] imx6q-pcie 5f000000.pcie: PCI host bridge to bus 0=
000:00
> > > > [    3.688297] pstate: 60000005 (nZCv daif -PAN -UAO)
> > > > [    3.688310] pc : xen_swiotlb_free_coherent+0x180/0x1c0
> > > > [    3.693993] pci_bus 0000:00: root bus resource [bus 00-ff]
> > > > [    3.701002] lr : xen_swiotlb_free_coherent+0x44/0x1c0
> > > > "
> > > >=20
> > > > In xen_swiotlb_alloc_coherent, if !(dev_addr + size - 1 <=3D dma_=
mask)
> > > > or range_straddles_page_boundary(phys, size) are true, it will cr=
eate
> > > > contiguous region. So when free, we need to free contiguous regio=
n use
> > > > upper check condition.
> > >=20
> > > No, this will break PV guests on x86.
> >=20
> > Could you share more details why alloc and free not matching for the =
check?
>=20
> xen_create_contiguous_region() is needed only in case:
>=20
> - the bus address is not within dma_mask, or
> - the memory region is not physically contiguous (can happen only for
>   PV guests)
>=20
> In any case it should arrange for the memory to be suitable for the
> DMA operation, so to be contiguous and within dma_mask afterwards. So
> xen_destroy_contiguous_region() should only ever called for areas
> which match above criteria, as otherwise we can be sure
> xen_create_contiguous_region() was not used for making the area DMA-abl=
e
> in the beginning.
>=20
> And this is very important in the PV case, as in those guests the page
> tables are containing the host-PFNs, not the guest-PFNS, and
> xen_create_contiguous_region() will fiddle with host- vs. guest-PFN
> arrangements, and xen_destroy_contiguous_region() is reverting this
> fiddling. Any call of xen_destroy_contiguous_region() for an area it
> was not intended to be called for might swap physical pages beneath
> random virtual addresses, which was the reason for this test to be
> added by me.
>=20
>=20
> Juergen
>=20
> >=20
> > Thanks,
> > Peng.
> >=20
> > >=20
> > > I think there is something wrong with your setup in combination wit=
h the ARM
> > > xen_create_contiguous_region() implementation.
> > >=20
> > > Stefano?
> > >=20
> > >=20
> > > Juergen
> > >=20
> > > >=20
> > > > Signed-off-by: Peng Fan <peng.fan@nxp.com>
> > > > ---
> > > >    drivers/xen/swiotlb-xen.c | 4 ++--
> > > >    1 file changed, 2 insertions(+), 2 deletions(-)
> > > >=20
> > > > diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.=
c
> > > > index b6d27762c6f8..ab96e468584f 100644
> > > > --- a/drivers/xen/swiotlb-xen.c
> > > > +++ b/drivers/xen/swiotlb-xen.c
> > > > @@ -346,8 +346,8 @@ xen_swiotlb_free_coherent(struct device *hwde=
v,
> > > size_t size, void *vaddr,
> > > >    	/* Convert the size to actually allocated. */
> > > >    	size =3D 1UL << (order + XEN_PAGE_SHIFT);
> > > >=20
> > > > -	if (!WARN_ON((dev_addr + size - 1 > dma_mask) ||
> > > > -		     range_straddles_page_boundary(phys, size)) &&
> > > > +	if (((dev_addr + size - 1 > dma_mask) ||
> > > > +	    range_straddles_page_boundary(phys, size)) &&
> > > >    	    TestClearPageXenRemapped(virt_to_page(vaddr)))
> > > >    		xen_destroy_contiguous_region(phys, order);
> > > >=20
> > > >=20
> >=20
>=20