[PATCH 3/6] mv643xx.c: Add basic device tree support.

Ian Molton [off-list ref] writes:
Hi,

[...]

I guess we need something for tx_csum_limit in the device tree too. It's
important for kirkwood and dove.

Arnaud

On 31/07/12 08:14, Arnaud Patard (Rtp) wrote:

Ian Molton [off-list ref] writes:
Hi,

[...]

quoted

@@ -2873,7 +2913,31 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
 	struct resource *res;
 	int err;
 
-	pd = pdev->dev.platform_data;
+	if (pdev->dev.of_node) {
+		struct device_node *np = NULL;
+
+		/* when all users of this driver use FDT, we can remove this */
+		pd = kzalloc(sizeof(*pd), GFP_ATOMIC);
+		if (!pd) {
+			dev_dbg(&pdev->dev, "Could not allocate platform data\n");
+			return -ENOMEM;
+		}
+
+		of_property_read_u32(pdev->dev.of_node,
+			"port_number", &pd->port_number);
+		of_property_read_u32(pdev->dev.of_node,
+			"phy_addr", &pd->phy_addr);

I guess we need something for tx_csum_limit in the device tree too. It's
important for kirkwood and dove.

It was my intention to keep the patch as non-invasive as possible - none
of the platforms using D-T at present require this.

Furthermore, git grep shows nothing in mach-kirkwood or mach-dove using
at all...

Are you sure this is required? It can always be added if anything
actually uses it :)

-Ian

On Tue, Jul 31, 2012 at 09:19:52AM +0100, Ian Molton wrote:

On 31/07/12 08:14, Arnaud Patard (Rtp) wrote:

quoted

Ian Molton [off-list ref] writes:
Hi,

[...]

quoted

@@ -2873,7 +2913,31 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
 	struct resource *res;
 	int err;
 
-	pd = pdev->dev.platform_data;
+	if (pdev->dev.of_node) {
+		struct device_node *np = NULL;
+
+		/* when all users of this driver use FDT, we can remove this */
+		pd = kzalloc(sizeof(*pd), GFP_ATOMIC);
+		if (!pd) {
+			dev_dbg(&pdev->dev, "Could not allocate platform data\n");
+			return -ENOMEM;
+		}
+
+		of_property_read_u32(pdev->dev.of_node,
+			"port_number", &pd->port_number);
+		of_property_read_u32(pdev->dev.of_node,
+			"phy_addr", &pd->phy_addr);

I guess we need something for tx_csum_limit in the device tree too. It's
important for kirkwood and dove.

It was my intention to keep the patch as non-invasive as possible - none
of the platforms using D-T at present require this.

Furthermore, git grep shows nothing in mach-kirkwood or mach-dove using
at all...

Are you sure this is required? It can always be added if anything
actually uses it :)

Hi Ian

Try using jumbo packets on kirkwood.

http://comments.gmane.org/gmane.linux.ports.arm.kernel/178785
http://www.spinics.net/lists/arm-kernel/msg186152.html

	Andrew

On 31/07/12 09:27, Andrew Lunn wrote:

On Tue, Jul 31, 2012 at 09:19:52AM +0100, Ian Molton wrote:

quoted

Are you sure this is required? It can always be added if anything
actually uses it :)

Hi Ian

Try using jumbo packets on kirkwood.

http://comments.gmane.org/gmane.linux.ports.arm.kernel/178785
http://www.spinics.net/lists/arm-kernel/msg186152.html

Ah, this postdates the D-T branches on which I am working being forked.

I'll fold this into my patchset.

-Ian

On 31/07/12 08:14, Arnaud Patard (Rtp) wrote:

Ian Molton[off-list ref]  writes:
Hi,

[...]

quoted

@@ -2873,7 +2913,31 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
  	struct resource *res;
  	int err;

-	pd = pdev->dev.platform_data;
+	if (pdev->dev.of_node) {
+		struct device_node *np = NULL;
+
+		/* when all users of this driver use FDT, we can remove this */
+		pd = kzalloc(sizeof(*pd), GFP_ATOMIC);
+		if (!pd) {
+			dev_dbg(&pdev->dev, "Could not allocate platform data\n");
+			return -ENOMEM;
+		}
+
+		of_property_read_u32(pdev->dev.of_node,
+			"port_number",&pd->port_number);
+		of_property_read_u32(pdev->dev.of_node,
+			"phy_addr",&pd->phy_addr);

I guess we need something for tx_csum_limit in the device tree too. It's
important for kirkwood and dove.

Is this something for all kirkwood devices? I think the best way of
doing this is to give the driver a new device-tree name and having
it set this field in the init sequence.

However we could also do with some of these as dt parameters just in
case people want to try and alter them for their own systems.

-- 
Ben Dooks				http://www.codethink.co.uk/
Senior Engineer				Codethink - Providing Genius