Date: Wed, 23 Nov 2016 19:03:21 +0800
From: Jisheng Zhang <jszhang@marvell.com>
To: Arnd Bergmann <arnd@arndb.de>
CC: <linux-arm-kernel@lists.infradead.org>,
        Thomas Petazzoni <thomas.petazzoni@free-electrons.com>,
        Andrew Lunn <andrew@lunn.ch>, "Jason Cooper" <jason@lakedaemon.net>,
        <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        Gregory CLEMENT <gregory.clement@free-electrons.com>,
        Marcin Wojtas <mw@semihalf.com>,
        "David S. Miller" <davem@davemloft.net>,
        Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
Subject: Re: [PATCH net-next 1/4] net: mvneta: Convert to be 64 bits
 compatible
Message-ID: <20161123190321.2f027a3a@xhacker>
In-Reply-To: <9432400.S1OrxC027t@wuerfel>
References: <20161122164844.19566-1-gregory.clement@free-electrons.com>
        <2948812.F3se4ieqO6@wuerfel>    <20161123175341.4777595f@xhacker>
        <9432400.S1OrxC027t@wuerfel>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2659
Lines: 73

Hi Arnd,

On Wed, 23 Nov 2016 11:15:32 +0100 Arnd Bergmann wrote:

> On Wednesday, November 23, 2016 5:53:41 PM CET Jisheng Zhang wrote:
> > On Tue, 22 Nov 2016 22:04:12 +0100 Arnd Bergmann wrote:
> >   
> > > On Tuesday, November 22, 2016 5:48:41 PM CET Gregory CLEMENT wrote:  
> > > > +#ifdef CONFIG_64BIT
> > > > +       void *data_tmp;
> > > > +
> > > > +       /* In Neta HW only 32 bits data is supported, so in order to
> > > > +        * obtain whole 64 bits address from RX descriptor, we store
> > > > +        * the upper 32 bits when allocating buffer, and put it back
> > > > +        * when using buffer cookie for accessing packet in memory.
> > > > +        * Frags should be allocated from single 'memory' region,
> > > > +        * hence common upper address half should be sufficient.
> > > > +        */
> > > > +       data_tmp = mvneta_frag_alloc(pp->frag_size);
> > > > +       if (data_tmp) {
> > > > +               pp->data_high = (u64)upper_32_bits((u64)data_tmp) << 32;
> > > > +               mvneta_frag_free(pp->frag_size, data_tmp);
> > > > +       }
> > > >     
> > > 
> > > How does this work when the region spans a n*4GB address boundary?  
> > 
> > indeed. We also make use of this driver on 64bit platforms. We use
> > different solution to make the driver 64bit safe.
> > 
> > solA: make use of the reserved field in the mvneta_rx_desc, such
> > as reserved2 etc. Yes, the field is marked as "for future use, PnC", but
> > now it's not used at all. This is one possible solution however.  
> 
> Right, this sounds like the most straightforward choice.
> 
> > solB: allocate a shadow buf cookie during init, e.g
> > 
> > rxq->descs_bufcookie = kmalloc(rxq->size * sizeof(void*), GFP_KERNEL);
> > 
> > then modify mvneta_rx_desc_fill a bit to save the 64bit pointer in
> > the shadow buf cookie, e.g
> > static void mvneta_rx_desc_fill(struct mvneta_rx_desc *rx_desc,
> >                                 u32 phys_addr, u32 cookie,

sorry, this line should be:
u32 phys_addr, void *cookie

> > 				struct mvneta_rx_queue *rxq)
> > 
> > {
> > 	int i;
> > 
> > 	rx_desc->buf_cookie = cookie;
> > 	rx_desc->buf_phys_addr = phys_addr;
> > 	i = rx_desc - rxq->descs;
> > 	rxq->descs_bufcookie[i] = cookie;
> > }
> > 
> > then fetch the desc from the shadow buf cookie in all code path, such
> > as mvneta_rx() etc.
> > 
> > Both solutions should not have the problems pointed out by Arnd.  
> 
> Wait, since you compute an index 'i' here, can't you just store 'i'
> directly in the descriptor instead of the pointer?
> 

we need to store the pointer, it's to store the buffer allocated by
mvneta_frag_alloc()

Thanks,
Jisheng