linux-kernel - Re: [PATCH 03/14] net: axienet: Fix DMA descriptor cleanup path

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20200110154328.6676215f@donnerap.cambridge.arm.com>
Date:   Fri, 10 Jan 2020 15:43:28 +0000
From:   Andre Przywara <andre.przywara@....com>
To:     Radhey Shyam Pandey <radheys@...inx.com>
Cc:     "David S . Miller" <davem@...emloft.net>,
        Michal Simek <michals@...inx.com>,
        Robert Hancock <hancock@...systems.ca>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 03/14] net: axienet: Fix DMA descriptor cleanup path

On Fri, 10 Jan 2020 15:14:46 +0000
Radhey Shyam Pandey <radheys@...inx.com> wrote:

Hi Radhey,

thanks for having a look!

> > -----Original Message-----
> > From: Andre Przywara <andre.przywara@....com>
> > Sent: Friday, January 10, 2020 5:24 PM
> > To: David S . Miller <davem@...emloft.net>; Radhey Shyam Pandey
> > <radheys@...inx.com>
> > Cc: Michal Simek <michals@...inx.com>; Robert Hancock
> > <hancock@...systems.ca>; netdev@...r.kernel.org; linux-arm-
> > kernel@...ts.infradead.org; linux-kernel@...r.kernel.org
> > Subject: [PATCH 03/14] net: axienet: Fix DMA descriptor cleanup path
> > 
> > When axienet_dma_bd_init() bails out during the initialisation process,
> > it might do so with parts of the structure already allocated and
> > initialised, while other parts have not been touched yet. Before
> > returning in this case, we call axienet_dma_bd_release(), which does not
> > take care of this corner case.
> > This is most obvious by the first loop happily dereferencing
> > lp->rx_bd_v, which we actually check to be non NULL *afterwards*.
> > 
> > Make sure we only unmap or free already allocated structures, by:
> > - directly returning with -ENOMEM if nothing has been allocated at all
> > - checking for lp->rx_bd_v to be non-NULL *before* using it
> > - only unmapping allocated DMA RX regions
> > 
> > This avoids NULL pointer dereferences when initialisation fails.
> > 
> > Signed-off-by: Andre Przywara <andre.przywara@....com>
> > ---
> >  .../net/ethernet/xilinx/xilinx_axienet_main.c | 43 ++++++++++++-------
> >  1 file changed, 28 insertions(+), 15 deletions(-)
> > 
> > diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
> > b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
> > index 97482cf093ce..7e90044cf2d9 100644
> > --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
> > +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
> > @@ -160,24 +160,37 @@ static void axienet_dma_bd_release(struct
> > net_device *ndev)
> >  	int i;
> >  	struct axienet_local *lp = netdev_priv(ndev);
> > 
> > +	/* If we end up here, tx_bd_v must have been DMA allocated. */
> > +	dma_free_coherent(ndev->dev.parent,
> > +			  sizeof(*lp->tx_bd_v) * lp->tx_bd_num,
> > +			  lp->tx_bd_v,
> > +			  lp->tx_bd_p);
> > +
> > +	if (!lp->rx_bd_v)
> > +		return;
> > +
> >  	for (i = 0; i < lp->rx_bd_num; i++) {
> > -		dma_unmap_single(ndev->dev.parent, lp->rx_bd_v[i].phys,
> > -				 lp->max_frm_size, DMA_FROM_DEVICE);
> > +		/* A NULL skb means this descriptor has not been initialised
> > +		 * at all.
> > +		 */
> > +		if (!lp->rx_bd_v[i].skb)
> > +			break;
> > +
> >  		dev_kfree_skb(lp->rx_bd_v[i].skb);
> > -	}
> > 
> > -	if (lp->rx_bd_v) {
> > -		dma_free_coherent(ndev->dev.parent,
> > -				  sizeof(*lp->rx_bd_v) * lp->rx_bd_num,
> > -				  lp->rx_bd_v,
> > -				  lp->rx_bd_p);
> > -	}
> > -	if (lp->tx_bd_v) {
> > -		dma_free_coherent(ndev->dev.parent,
> > -				  sizeof(*lp->tx_bd_v) * lp->tx_bd_num,
> > -				  lp->tx_bd_v,
> > -				  lp->tx_bd_p);
> > +		/* For each descriptor, we programmed cntrl with the (non-
> > zero)
> > +		 * descriptor size, after it had been successfully allocated.
> > +		 * So a non-zero value in there means we need to unmap it.
> > +		 */  
> 
> > +		if (lp->rx_bd_v[i].cntrl)  
> 
> I think it should ok to unmap w/o any check?

Do you mean because .phys would be 0 if not initialised? AFAIK 0 can be a valid DMA address, so there is no special check for that, and unmapping DMA address 0 will probably go wrong at some point. So it's unlike kfree(NULL).

Cheers,
Andre.


> > +			dma_unmap_single(ndev->dev.parent, lp-  
> > >rx_bd_v[i].phys,  
> > +					 lp->max_frm_size,
> > DMA_FROM_DEVICE);
> >  	}
> > +
> > +	dma_free_coherent(ndev->dev.parent,
> > +			  sizeof(*lp->rx_bd_v) * lp->rx_bd_num,
> > +			  lp->rx_bd_v,
> > +			  lp->rx_bd_p);
> >  }
> > 
> >  /**
> > @@ -207,7 +220,7 @@ static int axienet_dma_bd_init(struct net_device
> > *ndev)
> >  					 sizeof(*lp->tx_bd_v) * lp-  
> > >tx_bd_num,  
> >  					 &lp->tx_bd_p, GFP_KERNEL);
> >  	if (!lp->tx_bd_v)
> > -		goto out;
> > +		return -ENOMEM;
> > 
> >  	lp->rx_bd_v = dma_alloc_coherent(ndev->dev.parent,
> >  					 sizeof(*lp->rx_bd_v) * lp-  
> > >rx_bd_num,  
> > --
> > 2.17.1  
>