[PATCH v2] net: fec: Fix build for MCF5272

[PATCH v2] net: fec: Fix build for MCF5272

[Guenter Roeck]

Commits 4c09eed9 (net: fec: Enable imx6 enet checksum acceleration) and
baa70a5c (net: fec: enable pause frame to improve rx prefomance for 1G
network) introduced functionality into the FEC driver which is not
supported on MCF5272. The registers used to implement this functionality
do not exist on MCF5272. Since register defines for MCF5272 are separate
from register defines for other chips, building images for MCF5272 fails,
complaining about several undefined symbols.

Adding the missing register defines is not an option, since the registers
do not exist on MCF5272. Disable the added functionality for MCF5272 builds.

Cc: Frank Li <Frank.Li@freescale.com>
Cc: Jim Baxter <jim_baxter@mentor.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v2: Improved explanation to clarify that the build fails due to missing register
    defines which can not be added since the registers don't exist on MCF5272. 

This problem exists in 3.9 as well, though only due to commit baa70a5c. The
patch does therefore not cleanly apply to 3.9. Options I can see are to ignore
the build error in 3.9, to back-port the patch manually, or to split it into
two separate patches. David, let me know which approach you prefer. 

 drivers/net/ethernet/freescale/fec_main.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index a667015..d48099f 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -516,6 +516,7 @@ fec_restart(struct net_device *ndev, int duplex)
 	/* Set MII speed */
 	writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED);
 
+#if !defined(CONFIG_M5272)
 	/* set RX checksum */
 	val = readl(fep->hwp + FEC_RACC);
 	if (fep->csum_flags & FLAG_RX_CSUM_ENABLED)
@@ -523,6 +524,7 @@ fec_restart(struct net_device *ndev, int duplex)
 	else
 		val &= ~FEC_RACC_OPTIONS;
 	writel(val, fep->hwp + FEC_RACC);
+#endif
 
 	/*
 	 * The phy interface and speed need to get configured
@@ -575,6 +577,7 @@ fec_restart(struct net_device *ndev, int duplex)
 #endif
 	}
 
+#if !defined(CONFIG_M5272)
 	/* enable pause frame*/
 	if ((fep->pause_flag & FEC_PAUSE_FLAG_ENABLE) ||
 	    ((fep->pause_flag & FEC_PAUSE_FLAG_AUTONEG) &&
@@ -592,6 +595,7 @@ fec_restart(struct net_device *ndev, int duplex)
 	} else {
 		rcntl &= ~FEC_ENET_FCE;
 	}
+#endif /* !defined(CONFIG_M5272) */
 
 	writel(rcntl, fep->hwp + FEC_R_CNTRL);
 
@@ -1205,7 +1209,9 @@ static int fec_enet_mii_probe(struct net_device *ndev)
 	/* mask with MAC supported features */
 	if (id_entry->driver_data & FEC_QUIRK_HAS_GBIT) {
 		phy_dev->supported &= PHY_GBIT_FEATURES;
+#if !defined(CONFIG_M5272)
 		phy_dev->supported |= SUPPORTED_Pause;
+#endif
 	}
 	else
 		phy_dev->supported &= PHY_BASIC_FEATURES;
@@ -1390,6 +1396,8 @@ static int fec_enet_get_ts_info(struct net_device *ndev,
 	}
 }
 
+#if !defined(CONFIG_M5272)
+
 static void fec_enet_get_pauseparam(struct net_device *ndev,
 				    struct ethtool_pauseparam *pause)
 {
@@ -1436,9 +1444,13 @@ static int fec_enet_set_pauseparam(struct net_device *ndev,
 	return 0;
 }
 
+#endif /* !defined(CONFIG_M5272) */
+
 static const struct ethtool_ops fec_enet_ethtool_ops = {
+#if !defined(CONFIG_M5272)
 	.get_pauseparam		= fec_enet_get_pauseparam,
 	.set_pauseparam		= fec_enet_set_pauseparam,
+#endif
 	.get_settings		= fec_enet_get_settings,
 	.set_settings		= fec_enet_set_settings,
 	.get_drvinfo		= fec_enet_get_drvinfo,
@@ -1874,10 +1886,12 @@ fec_probe(struct platform_device *pdev)
 	/* setup board info structure */
 	fep = netdev_priv(ndev);
 
+#if !defined(CONFIG_M5272)
 	/* default enable pause frame auto negotiation */
 	if (pdev->id_entry &&
 	    (pdev->id_entry->driver_data & FEC_QUIRK_HAS_GBIT))
 		fep->pause_flag |= FEC_PAUSE_FLAG_AUTONEG;
+#endif
 
 	fep->hwp = devm_request_and_ioremap(&pdev->dev, r);
 	fep->pdev = pdev;
-- 
1.7.9.7

Re: [PATCH v2] net: fec: Fix build for MCF5272

[Fabio Estevam]

On Tue, Jun 18, 2013 at 12:38 PM, Guenter Roeck <linux@roeck-us.net> wrote:
> Commits 4c09eed9 (net: fec: Enable imx6 enet checksum acceleration) and
> baa70a5c (net: fec: enable pause frame to improve rx prefomance for 1G
> network) introduced functionality into the FEC driver which is not
> supported on MCF5272. The registers used to implement this functionality
> do not exist on MCF5272. Since register defines for MCF5272 are separate
> from register defines for other chips, building images for MCF5272 fails,
> complaining about several undefined symbols.
>
> Adding the missing register defines is not an option, since the registers
> do not exist on MCF5272. Disable the added functionality for MCF5272 builds.

Can you please show the build error message you got?

Also, it would be nice if FEC could be enabled by default on some m68k
defconfig, so that we could spot issues like this earlier.

Re: [PATCH v2] net: fec: Fix build for MCF5272

[Guenter Roeck]

On Tue, Jun 18, 2013 at 01:23:13PM -0300, Fabio Estevam wrote:
> On Tue, Jun 18, 2013 at 12:38 PM, Guenter Roeck <linux@roeck-us.net> wrote:
> > Commits 4c09eed9 (net: fec: Enable imx6 enet checksum acceleration) and
> > baa70a5c (net: fec: enable pause frame to improve rx prefomance for 1G
> > network) introduced functionality into the FEC driver which is not
> > supported on MCF5272. The registers used to implement this functionality
> > do not exist on MCF5272. Since register defines for MCF5272 are separate
> > from register defines for other chips, building images for MCF5272 fails,
> > complaining about several undefined symbols.
> >
> > Adding the missing register defines is not an option, since the registers
> > do not exist on MCF5272. Disable the added functionality for MCF5272 builds.
> 
> Can you please show the build error message you got?
> 
> Also, it would be nice if FEC could be enabled by default on some m68k
> defconfig, so that we could spot issues like this earlier.
> 
Try building m5272c3_defconfig; this is actually how I found the problem when
test-building v3.9.6-rc1.

make ARCH=m68k CROSS_COMPILE=m68k-linux-gnu- mrproper
make ARCH=m68k CROSS_COMPILE=m68k-linux-gnu- m5272c3_defconfig
make ARCH=m68k CROSS_COMPILE=m68k-linux-gnu-

results in:

drivers/net/ethernet/freescale/fec_main.c: In function ‘fec_restart’:
drivers/net/ethernet/freescale/fec_main.c:520:8: error: ‘FEC_RACC’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec_main.c:520:8: note: each undeclared identifier is reported only once for each function it appears in
drivers/net/ethernet/freescale/fec_main.c:585:3: error: ‘FEC_R_FIFO_RSEM’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec_main.c:586:3: error: ‘FEC_R_FIFO_RSFL’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec_main.c:587:3: error: ‘FEC_R_FIFO_RAEM’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec_main.c:588:3: error: ‘FEC_R_FIFO_RAFL’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec_main.c:591:3: error: ‘FEC_OPD’ undeclared (first use in this function)

with v3.10-rc6,

and

drivers/net/ethernet/freescale/fec.c: In function ‘fec_restart’:
drivers/net/ethernet/freescale/fec.c:530:3: error: ‘FEC_R_FIFO_RSEM’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec.c:530:3: note: each undeclared identifier is reported only once for each function it appears in
drivers/net/ethernet/freescale/fec.c:531:3: error: ‘FEC_R_FIFO_RSFL’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec.c:532:3: error: ‘FEC_R_FIFO_RAEM’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec.c:533:3: error: ‘FEC_R_FIFO_RAFL’ undeclared (first use in this function)
drivers/net/ethernet/freescale/fec.c:536:3: error: ‘FEC_OPD’ undeclared (first use in this function)

with 3.9.6.
 
Want me to add it to the commit log ?

Guenter

Re: [PATCH v2] net: fec: Fix build for MCF5272

[Fabio Estevam]

On Tue, Jun 18, 2013 at 1:32 PM, Guenter Roeck <linux@roeck-us.net> wrote:

> Want me to add it to the commit log ?

That would be nice, thanks.