[U-Boot] [PATCH v2 1/3] at91_emac.h: fix typo in register definition

[U-Boot] [PATCH v2 1/3] at91_emac.h: fix typo in register definition

[Andreas Bießmann]

Signed-off-by: Andreas Bie?mann <andreas.devel@googlemail.com>
Acked-by: Jens Scharsig <js_at_ng@scharsoft.de>
---
v2: 
 Acked-by: Jens Scharsig see http://article.gmane.org/gmane.comp.boot-loaders.u-boot/83624
 additionally add this patch to a series of other patches related to atmel_emac driver

 arch/arm/include/asm/arch-at91/at91_emac.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/arch/arm/include/asm/arch-at91/at91_emac.h b/arch/arm/include/asm/arch-at91/at91_emac.h
index 45ae333..0e2ff78 100644
--- a/arch/arm/include/asm/arch-at91/at91_emac.h
+++ b/arch/arm/include/asm/arch-at91/at91_emac.h
@@ -61,7 +61,7 @@ typedef struct at91_emac {
 	u32	 reserved2[3];
 	u32	 hsh;
 	u32	 hsl;
-	u32	 sh1l;
+	u32	 sa1l;
 	u32	 sa1h;
 	u32	 sa2l;
 	u32	 sa2h;
-- 
1.7.2.3

[U-Boot] [PATCH 2/3] at91_emac.c: fix 'Warning: eth device name has a space!'

[Andreas Bießmann]

This patch also removes conditional nameing of at91_emac driver whether it's
connection to PHY is RMII or MII.

Signed-off-by: Andreas Bie?mann <andreas.devel@googlemail.com>
---
 drivers/net/at91_emac.c |    6 +-----
 1 files changed, 1 insertions(+), 5 deletions(-)

diff --git a/drivers/net/at91_emac.c b/drivers/net/at91_emac.c
index d82459b..69e5409 100644
--- a/drivers/net/at91_emac.c
+++ b/drivers/net/at91_emac.c
@@ -500,11 +500,7 @@ int at91emac_register(bd_t *bis, unsigned long iobase)
 	memset(emacfix, 0, sizeof(emac_device));
 
 	memset(dev, 0, sizeof(*dev));
-#ifndef CONFIG_RMII
-	sprintf(dev->name, "AT91 EMAC");
-#else
-	sprintf(dev->name, "AT91 EMAC RMII");
-#endif
+	sprintf(dev->name, "at91_emac");
 	dev->iobase = iobase;
 	dev->priv = emacfix;
 	dev->init = at91emac_init;
-- 
1.7.2.3

[U-Boot] [PATCH 3/3] at91_emac.c: poll for IDLE when writing PHY

[Andreas Bießmann]

This patch replaces the unnecessary waiting in at91emac_read() and
at91emac_write() by checking the IDLE flag.

Signed-off-by: Andreas Bie?mann <andreas.devel@googlemail.com>
---
 drivers/net/at91_emac.c |   16 ++++++++++++++--
 1 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/net/at91_emac.c b/drivers/net/at91_emac.c
index 69e5409..219c8c7 100644
--- a/drivers/net/at91_emac.c
+++ b/drivers/net/at91_emac.c
@@ -127,13 +127,19 @@ void at91emac_DisableMDIO(at91_emac_t *at91mac)
 int  at91emac_read(at91_emac_t *at91mac, unsigned char addr,
 		unsigned char reg, unsigned short *value)
 {
+	unsigned long netstat;
 	at91emac_EnableMDIO(at91mac);
 
 	writel(AT91_EMAC_MAN_HIGH | AT91_EMAC_MAN_RW_R |
 		AT91_EMAC_MAN_REGA(reg) | AT91_EMAC_MAN_CODE_802_3 |
 		AT91_EMAC_MAN_PHYA(addr),
 		&at91mac->man);
-	udelay(10000);
+
+	do {
+		netstat = readl(&at91mac->sr);
+		DEBUG_AT91PHY("poll SR %08lx\n", netstat);
+	} while (!(netstat & AT91_EMAC_SR_IDLE));
+
 	*value = readl(&at91mac->man) & AT91_EMAC_MAN_DATA_MASK;
 
 	at91emac_DisableMDIO(at91mac);
@@ -146,6 +152,7 @@ int  at91emac_read(at91_emac_t *at91mac, unsigned char addr,
 int  at91emac_write(at91_emac_t *at91mac, unsigned char addr,
 		unsigned char reg, unsigned short value)
 {
+	unsigned long netstat;
 	DEBUG_AT91PHY("AT91PHY write %x REG(%d)=%x\n", at91mac, reg, &value)
 
 	at91emac_EnableMDIO(at91mac);
@@ -154,9 +161,14 @@ int  at91emac_write(at91_emac_t *at91mac, unsigned char addr,
 		AT91_EMAC_MAN_REGA(reg) | AT91_EMAC_MAN_CODE_802_3 |
 		AT91_EMAC_MAN_PHYA(addr) | (value & AT91_EMAC_MAN_DATA_MASK),
 		&at91mac->man);
-	udelay(10000);
+
+	do {
+		netstat = readl(&at91mac->sr);
+		DEBUG_AT91PHY("poll SR %08lx\n", netstat);
+	} while (!(netstat & AT91_EMAC_SR_IDLE));
 
 	at91emac_DisableMDIO(at91mac);
+
 	return 0;
 }
 
-- 
1.7.2.3

[U-Boot] [PATCH 2/3] at91_emac.c: fix 'Warning: eth device name has a space!'

[Reinhard Meyer]

Dear Andreas Bie?mann,
> +	sprintf(dev->name, "at91_emac");

I would think "emac" is sufficient. That they are on an AT91 system
should be obvious to the user.

And if, somehow, in the future, MACB and AT91_EMAC get merged, we
would not need to change this again :)

Reinhard

[U-Boot] [PATCH 2/3] at91_emac.c: fix 'Warning: eth device name has a space!'

[Reinhard Meyer]

Dear Andreas Bie?mann,
> This patch also removes conditional nameing of at91_emac driver whether it's
> connection to PHY is RMII or MII.
> 
> Signed-off-by: Andreas Bie?mann <andreas.devel@googlemail.com>

> +	sprintf(dev->name, "at91_emac");

Applied to u-boot-atmel/at91, with name shortened to "emac".

Thanks,
Reinhard

[U-Boot] [PATCH 3/3] at91_emac.c: poll for IDLE when writing PHY

[Reinhard Meyer]

Dear Andreas Bie?mann,

> This patch replaces the unnecessary waiting in at91emac_read() and
> at91emac_write() by checking the IDLE flag.
> 
> Signed-off-by: Andreas Bie?mann <andreas.devel@googlemail.com>
Applied to u-boot-atmel/at91.

Thanks,
Reinhard