2018-06-25 08:07:27

by Marcel Ziswiler

Subject: [PATCH] igb: allow optionally getting mac address from device tree

From: Marcel Ziswiler <[email protected]>

For Embedded use where e.g. i210/i211 chips may be used without
external EEPROMs but regardless of whether or not the MAC address is
actually programmed into the iNVM the boot loader may choose to pass
the MAC address to be used via device tree. Therefore, allow for
optionally getting the MAC address from device tree data e.g. as
follows (excerpt from a TK1 based board, local-mac-address to be
filled in by boot loader):

pcie@1003000 {
...

/* I210 Gigabit Ethernet Controller */
pci@2,0 {
...
status = "okay";
pcie@0 {
reg = <0 0 0 0 0>;
local-mac-address = [00 00 00 00 00 00];
};
};
};

Signed-off-by: Marcel Ziswiler <[email protected]>

---

drivers/net/ethernet/intel/igb/igb_main.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index f707709969ac..0abf3698b05c 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -37,6 +37,7 @@
#include <linux/dca.h>
#endif
#include <linux/i2c.h>
+#include <linux/of_net.h>
#include "igb.h"

#define MAJ 5
@@ -2931,6 +2932,7 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
const struct e1000_info *ei = igb_info_tbl[ent->driver_data];
int err, pci_using_dac;
u8 part_str[E1000_PBANUM_LENGTH];
+ const void *iap;

/* Catch broken hardware that put the wrong VF device ID in
* the PCIe SR-IOV capability.
@@ -3122,7 +3124,11 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
break;
}

- if (eth_platform_get_mac_address(&pdev->dev, hw->mac.addr)) {
+ /* try to get the MAC address from device tree data */
+ iap = of_get_mac_address(pdev->dev.of_node);
+ if (iap)
+ memcpy(hw->mac.addr, iap, ETH_ALEN);
+ else if (eth_platform_get_mac_address(&pdev->dev, hw->mac.addr)) {
/* copy the MAC address out of the NVM */
if (hw->mac.ops.read_mac_addr(hw))
dev_err(&pdev->dev, "NVM Read Error\n");
--
2.14.4

2018-06-25 08:08:17

by David Miller

Subject: Re: [PATCH] igb: allow optionally getting mac address from device tree

From: Marcel Ziswiler <[email protected]>
Date: Mon, 25 Jun 2018 10:00:42 +0200

> @@ -3122,7 +3124,11 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
> break;
> }
>
> - if (eth_platform_get_mac_address(&pdev->dev, hw->mac.addr)) {
> + /* try to get the MAC address from device tree data */
> + iap = of_get_mac_address(pdev->dev.of_node);
> + if (iap)
> + memcpy(hw->mac.addr, iap, ETH_ALEN);

This is exactly what eth_platform_get_mac_address() shoule be doing.

2018-06-25 11:27:26

by Marcel Ziswiler

Subject: Re: [PATCH] igb: allow optionally getting mac address from device tree

Hi David

On Mon, 2018-06-25 at 17:05 +0900, David Miller wrote:
> From: Marcel Ziswiler <[email protected]>
> Date: Mon, 25 Jun 2018 10:00:42 +0200
>
> > @@ -3122,7 +3124,11 @@ static int igb_probe(struct pci_dev *pdev,
> > const struct pci_device_id *ent)
> > break;
> > }
> >
> > - if (eth_platform_get_mac_address(&pdev->dev, hw-
> > >mac.addr)) {
> > + /* try to get the MAC address from device tree data */
> > + iap = of_get_mac_address(pdev->dev.of_node);
> > + if (iap)
> > + memcpy(hw->mac.addr, iap, ETH_ALEN);
>
> This is exactly what eth_platform_get_mac_address() shoule be doing.

Yes, you are absolutely right. That indeed already works just fine given a
proper device tree. Struggling with the device tree part I probably missed
noticing this.

Sorry for the noise and thanks for the tip!

Cheers

Marcel