2009-11-13 12:29:32

by Stanislav O. Bezzubtsev

Subject: [PATCH] forcedeth: mac address fix

Set second bit of randomly generated mac. That marks MAC
as locally assigned. IEEE802.3, section one, 3.2.3.

Signed-off-by: Stanislav O. Bezzubtsev <[email protected]>
---
drivers/net/forcedeth.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c
index e1da466..b404c7a 100644
--- a/drivers/net/forcedeth.c
+++ b/drivers/net/forcedeth.c
@@ -5821,7 +5821,7 @@ static int __devinit nv_probe(struct pci_dev *pci_dev, const struct pci_device_i
dev->dev_addr);
dev_printk(KERN_ERR, &pci_dev->dev,
"Please complain to your hardware vendor. Switching to a random MAC.\n");
- dev->dev_addr[0] = 0x00;
+ dev->dev_addr[0] = 0x02; /* set local assignment bit */
dev->dev_addr[1] = 0x00;
dev->dev_addr[2] = 0x6c;
get_random_bytes(&dev->dev_addr[3], 3);
--
1.6.5

2009-11-14 03:52:16

by David Miller

Subject: Re: [PATCH] forcedeth: mac address fix

From: "Stanislav O\. Bezzubtsev" <[email protected]>
Date: Fri, 13 Nov 2009 15:22:55 +0300

> Set second bit of randomly generated mac. That marks MAC
> as locally assigned. IEEE802.3, section one, 3.2.3.
>
> Signed-off-by: Stanislav O. Bezzubtsev <[email protected]>

Do you know why this driver even has this bug?

It's because it does not use random_ether_addr() and tries to
do it by hand all by itself.

Please fix this bug for real by having the driver use the proper
interface to calculate a randomized ethernet address.

Thank you.