netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] net: fsl: fec: handle 10Mbps speed in RMII mode
@ 2012-01-11 16:29 Eric Bénard
  2012-01-12 15:27 ` Shawn Guo
  0 siblings, 1 reply; 2+ messages in thread
From: Eric Bénard @ 2012-01-11 16:29 UTC (permalink / raw)
  To: netdev; +Cc: shawn.guo, davem, linux-arm-kernel, u.kleine-koenig

when the link is 10 Mbps and the mode is RMII, it's necessary
to set FRCONT to 1 in MIIGSK_CFGR to divide the RMII source
clock by 10 in order to support 10 Mbps operations.

Signed-off-by: Eric Bénard <eric@eukrea.com>
---
 drivers/net/ethernet/freescale/fec.c |   10 ++++++++--
 drivers/net/ethernet/freescale/fec.h |    4 ++++
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/freescale/fec.c b/drivers/net/ethernet/freescale/fec.c
index ddcbbb3..89532ab 100644
--- a/drivers/net/ethernet/freescale/fec.c
+++ b/drivers/net/ethernet/freescale/fec.c
@@ -476,6 +476,7 @@ fec_restart(struct net_device *ndev, int duplex)
 	} else {
 #ifdef FEC_MIIGSK_ENR
 		if (id_entry->driver_data & FEC_QUIRK_USE_GASKET) {
+			u32 miigsk;
 			/* disable the gasket and wait */
 			writel(0, fep->hwp + FEC_MIIGSK_ENR);
 			while (readl(fep->hwp + FEC_MIIGSK_ENR) & 4)
@@ -486,8 +487,13 @@ fec_restart(struct net_device *ndev, int duplex)
 			 *   RMII, 50 MHz, no loopback, no echo
 			 *   MII, 25 MHz, no loopback, no echo
 			 */
-			writel((fep->phy_interface == PHY_INTERFACE_MODE_RMII) ?
-					1 : 0, fep->hwp + FEC_MIIGSK_CFGR);
+			miigsk = (fep->phy_interface == PHY_INTERFACE_MODE_RMII)
+				? FEC_MIIGSK_RMII_MODE : FEC_MIIGSK_MII_MODE;
+			if (fep->phy_dev) {
+				if (fep->phy_dev->speed == SPEED_10)
+					miigsk |= FEC_MIIGSK_FRCONT_10M;
+			}
+			writel(miigsk, fep->hwp + FEC_MIIGSK_CFGR);
 
 
 			/* re-enable the gasket */
diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index 8b2c6d7..6870d9e 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -47,6 +47,10 @@
 #define FEC_MIIGSK_CFGR		0x300 /* MIIGSK Configuration reg */
 #define FEC_MIIGSK_ENR		0x308 /* MIIGSK Enable reg */
 
+#define FEC_MIIGSK_MII_MODE		0x00
+#define FEC_MIIGSK_RMII_MODE	0x01
+#define FEC_MIIGSK_FRCONT_10M	0x40
+
 #else
 
 #define FEC_ECNTRL		0x000 /* Ethernet control reg */
-- 
1.7.7.5


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] net: fsl: fec: handle 10Mbps speed in RMII mode
  2012-01-11 16:29 [PATCH] net: fsl: fec: handle 10Mbps speed in RMII mode Eric Bénard
@ 2012-01-12 15:27 ` Shawn Guo
  0 siblings, 0 replies; 2+ messages in thread
From: Shawn Guo @ 2012-01-12 15:27 UTC (permalink / raw)
  To: Eric Bénard; +Cc: netdev, davem, u.kleine-koenig, linux-arm-kernel

Hi Eric,

Though I do not have the 10 Mbps setup to test the patch, it looks
like the patch is fixing a real bug.

A few of nits below ...

On Wed, Jan 11, 2012 at 05:29:36PM +0100, Eric Bénard wrote:
> when the link is 10 Mbps and the mode is RMII, it's necessary
> to set FRCONT to 1 in MIIGSK_CFGR to divide the RMII source
> clock by 10 in order to support 10 Mbps operations.
> 
> Signed-off-by: Eric Bénard <eric@eukrea.com>
> ---
>  drivers/net/ethernet/freescale/fec.c |   10 ++++++++--
>  drivers/net/ethernet/freescale/fec.h |    4 ++++
>  2 files changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/ethernet/freescale/fec.c b/drivers/net/ethernet/freescale/fec.c
> index ddcbbb3..89532ab 100644
> --- a/drivers/net/ethernet/freescale/fec.c
> +++ b/drivers/net/ethernet/freescale/fec.c
> @@ -476,6 +476,7 @@ fec_restart(struct net_device *ndev, int duplex)
>  	} else {
>  #ifdef FEC_MIIGSK_ENR
>  		if (id_entry->driver_data & FEC_QUIRK_USE_GASKET) {
> +			u32 miigsk;

I would prefer to name it as 'cfgr' or 'miigsk_cfgr'.

>  			/* disable the gasket and wait */
>  			writel(0, fep->hwp + FEC_MIIGSK_ENR);
>  			while (readl(fep->hwp + FEC_MIIGSK_ENR) & 4)
> @@ -486,8 +487,13 @@ fec_restart(struct net_device *ndev, int duplex)
>  			 *   RMII, 50 MHz, no loopback, no echo
>  			 *   MII, 25 MHz, no loopback, no echo
>  			 */
> -			writel((fep->phy_interface == PHY_INTERFACE_MODE_RMII) ?
> -					1 : 0, fep->hwp + FEC_MIIGSK_CFGR);
> +			miigsk = (fep->phy_interface == PHY_INTERFACE_MODE_RMII)
> +				? FEC_MIIGSK_RMII_MODE : FEC_MIIGSK_MII_MODE;
> +			if (fep->phy_dev) {
> +				if (fep->phy_dev->speed == SPEED_10)
> +					miigsk |= FEC_MIIGSK_FRCONT_10M;
> +			}

This can be as simple as below?

                        if (fep->phy_dev && fep->phy_dev->speed == SPEED_10)
                                        miigsk |= FEC_MIIGSK_FRCONT_10M;

> +			writel(miigsk, fep->hwp + FEC_MIIGSK_CFGR);
>  
>  
>  			/* re-enable the gasket */
> diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
> index 8b2c6d7..6870d9e 100644
> --- a/drivers/net/ethernet/freescale/fec.h
> +++ b/drivers/net/ethernet/freescale/fec.h
> @@ -47,6 +47,10 @@
>  #define FEC_MIIGSK_CFGR		0x300 /* MIIGSK Configuration reg */
>  #define FEC_MIIGSK_ENR		0x308 /* MIIGSK Enable reg */
>  
> +#define FEC_MIIGSK_MII_MODE		0x00
> +#define FEC_MIIGSK_RMII_MODE	0x01
> +#define FEC_MIIGSK_FRCONT_10M	0x40
> +

I would prefer to define them as BM_MIIGSK_CFGR_*, so that we can
explicitly know what they are from the names, and also keep FEC_*
consistently being register offset.

Other than these minor comments:

Acked-by: Shawn Guo <shawn.guo@linaro.org>

Regards,
Shawn

>  #else
>  
>  #define FEC_ECNTRL		0x000 /* Ethernet control reg */
> -- 
> 1.7.7.5
> 

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2012-01-12 15:16 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-01-11 16:29 [PATCH] net: fsl: fec: handle 10Mbps speed in RMII mode Eric Bénard
2012-01-12 15:27 ` Shawn Guo

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).