Re: [PATCH net-next 08/40] net: fec: move cbd_bufaddr assignment closer to the mapping function

[Troy Kisky]

On 1/28/2016 3:02 PM, Arnd Bergmann wrote:
> On Thursday 28 January 2016 14:25:32 Troy Kisky wrote:
>> Signed-off-by: Troy Kisky 
>> ---
>>  drivers/net/ethernet/freescale/fec_main.c | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> [note: missing changelog?]
> 
>> diff --git a/drivers/net/ethernet/freescale/fec_main.c 
>> b/drivers/net/ethernet/freescale/fec_main.c
>> index b87f80d..15a93f90 100644
>> --- a/drivers/net/ethernet/freescale/fec_main.c
>> +++ b/drivers/net/ethernet/freescale/fec_main.c
>> @@ -476,6 +476,8 @@ static int fec_enet_txq_submit_skb(struct 
>> fec_enet_priv_tx_q *txq,
>>  estatus |= BD_ENET_TX_TS;
>>  }
>>  }
>> +bdp->cbd_bufaddr = addr;
>> +bdp->cbd_datlen = buflen;
>>  
>>  if (fep->bufdesc_ex) {
>>  
>> @@ -499,8 +501,6 @@ static int fec_enet_txq_submit_skb(struct 
>> fec_enet_priv_tx_q *txq,
>>  /* Save skb pointer */
>>  txq->tx_skbuff[index] = skb;
>>  
>> -bdp->cbd_datlen = buflen;
>> -bdp->cbd_bufaddr = addr;
>>  /* Make sure the updates to rest of the descriptor are performed before
>>   * transferring ownership.
>>   */
>>
> 
> This patch and others in the series conflicts with the bugfix "net: fec: make
> driver endian-safe" that Johannes sent this week. Can you include his fix
> in your series and ensure that all descriptor accesses are done in
> an endian-safe way?
> 
>   Arnd

Sure, I'll rebase and send the first few when net-next is updated.

Thanks
Troy

Re: [PATCH net-next 08/40] net: fec: move cbd_bufaddr assignment closer to the mapping function

[Arnd Bergmann]

On Thursday 28 January 2016 14:25:32 Troy Kisky wrote:
> Signed-off-by: Troy Kisky 
> ---
>  drivers/net/ethernet/freescale/fec_main.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

[note: missing changelog?]

> diff --git a/drivers/net/ethernet/freescale/fec_main.c 
> b/drivers/net/ethernet/freescale/fec_main.c
> index b87f80d..15a93f90 100644
> --- a/drivers/net/ethernet/freescale/fec_main.c
> +++ b/drivers/net/ethernet/freescale/fec_main.c
> @@ -476,6 +476,8 @@ static int fec_enet_txq_submit_skb(struct 
> fec_enet_priv_tx_q *txq,
>   estatus |= BD_ENET_TX_TS;
>   }
>   }
> + bdp->cbd_bufaddr = addr;
> + bdp->cbd_datlen = buflen;
>  
>   if (fep->bufdesc_ex) {
>  
> @@ -499,8 +501,6 @@ static int fec_enet_txq_submit_skb(struct 
> fec_enet_priv_tx_q *txq,
>   /* Save skb pointer */
>   txq->tx_skbuff[index] = skb;
>  
> - bdp->cbd_datlen = buflen;
> - bdp->cbd_bufaddr = addr;
>   /* Make sure the updates to rest of the descriptor are performed before
>* transferring ownership.
>*/
> 

This patch and others in the series conflicts with the bugfix "net: fec: make
driver endian-safe" that Johannes sent this week. Can you include his fix
in your series and ensure that all descriptor accesses are done in
an endian-safe way?

Arnd

Re: [PATCH net-next 08/40] net: fec: move cbd_bufaddr assignment closer to the mapping function

[Johannes Berg]

On Thu, 2016-01-28 at 23:02 +0100, Arnd Bergmann wrote:
> On Thursday 28 January 2016 14:25:32 Troy Kisky wrote:
> > Signed-off-by: Troy Kisky 
> > ---
> >  drivers/net/ethernet/freescale/fec_main.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> [note: missing changelog?]
> 
> > diff --git a/drivers/net/ethernet/freescale/fec_main.c
> > b/drivers/net/ethernet/freescale/fec_main.c
> > index b87f80d..15a93f90 100644
> > --- a/drivers/net/ethernet/freescale/fec_main.c
> > +++ b/drivers/net/ethernet/freescale/fec_main.c
> > @@ -476,6 +476,8 @@ static int fec_enet_txq_submit_skb(struct
> > fec_enet_priv_tx_q *txq,
> >     estatus |= BD_ENET_TX_TS;
> >     }
> >     }
> > +   bdp->cbd_bufaddr = addr;
> > +   bdp->cbd_datlen = buflen;
> >  
> >     if (fep->bufdesc_ex) {
> >  
> > @@ -499,8 +501,6 @@ static int fec_enet_txq_submit_skb(struct
> > fec_enet_priv_tx_q *txq,
> >     /* Save skb pointer */
> >     txq->tx_skbuff[index] = skb;
> >  
> > -   bdp->cbd_datlen = buflen;
> > -   bdp->cbd_bufaddr = addr;
> >     /* Make sure the updates to rest of the descriptor are
> > performed before
> >      * transferring ownership.
> >      */
> > 
> 
> This patch and others in the series conflicts with the bugfix "net:
> fec: make driver endian-safe" that Johannes sent this week. Can you
> include his fix in your series and ensure that all descriptor
> accesses are done in an endian-safe way?
> 

My fixes are already in net.git, so you should base on a current
net.git or on net-next.git after net.git is merged there (which davem
usually does after Linus merges from him, IIRC)

johannes

[PATCH net-next 08/40] net: fec: move cbd_bufaddr assignment closer to the mapping function

[Troy Kisky]

Signed-off-by: Troy Kisky 
---
 drivers/net/ethernet/freescale/fec_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/freescale/fec_main.c 
b/drivers/net/ethernet/freescale/fec_main.c
index b87f80d..15a93f90 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -476,6 +476,8 @@ static int fec_enet_txq_submit_skb(struct 
fec_enet_priv_tx_q *txq,
estatus |= BD_ENET_TX_TS;
}
}
+   bdp->cbd_bufaddr = addr;
+   bdp->cbd_datlen = buflen;
 
if (fep->bufdesc_ex) {
 
@@ -499,8 +501,6 @@ static int fec_enet_txq_submit_skb(struct 
fec_enet_priv_tx_q *txq,
/* Save skb pointer */
txq->tx_skbuff[index] = skb;
 
-   bdp->cbd_datlen = buflen;
-   bdp->cbd_bufaddr = addr;
/* Make sure the updates to rest of the descriptor are performed before
 * transferring ownership.
 */
-- 
2.5.0