Message-Id: <0362CDDC-AE9B-448C-BE7C-D563B12D5A61@126.com>
Date:   Mon, 6 Jun 2022 22:35:07 +0800
From:   梁礼学 <lianglixuehao@....com>
To:     Alexander H Duyck <alexander.duyck@...il.com>
Cc:     Paul Menzel <pmenzel@...gen.mpg.de>, anthony.l.nguyen@...el.com,
        intel-wired-lan@...ts.osuosl.org,
        Jesse Brandeburg <jesse.brandeburg@...el.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Netdev <netdev@...r.kernel.org>, lianglixue@...atwall.com.cn
Subject: Re: [PATCH v4] igb: Assign random MAC address instead of fail in case
 of invalid one

Hi,
thank you very much for your suggestion.

As you said, the way to cause ‘Invalid MAC address’ is not only through igb_set_eeprom,
but also some pre-production or uninitialized boards.

But if set by module parameters, especially in the case of CONFIG_IGB=y,
The situation may be more troublesome, because for most users, if the system does not properly load and generate 
the network card device, they can only ask the host supplier for help.But,

(1) If the invalid mac address is caused by igb_set_eeprom, it is relatively more convenient for most operations engineers 
to change the invalid mac address to the mac address they think should be valid by ethtool, which may still be Invalid.
At this time,assigned random MAC address which is valid by the driver enables the network card driver to continue to complete the loading.
As for what you mentioned, in this case if the user does not notice that the driver had used a random mac address,
it may lead to other problems.but the fact is that if the user deliberately sets a customized mac address, 
the user should pay attention to whether the mac address is successfully changed, and also pay attention to the 
expected result after changing the mac address.When users find that the custom mac address cannot 
be successfully changed to the customized one, they can continue debugging, which is easier than looking for 
the host supplier’s support from the very first time of “Invalid MAC address”.

(2) If the invalid mac address is caused during pre-production or initialization of the board, it is even more necessary 
to use a random mac address to complete the loading of the network card, because the user only cares about whether 
the network card is loaded, not what the valid MAC address is.

And I also noticed that ixgbvef_sw_init also uses a random valid mac address to continue loading the driver when 
the address is invalid. In addition, network card drivers such as marvell, broadcom, realtek, etc., when an invalid 
MAC address is detected, it also does not directly exit the driver loading, but uses a random valid MAC address.

> 2022年6月2日 23:57，Alexander H Duyck <alexander.duyck@...il.com> 写道：
> 
> On Wed, 2022-06-01 at 15:04 +0000, Lixue Liang wrote:
>> From: Lixue Liang <lianglixue@...atwall.com.cn>
>> 
>> In some cases, when the user uses igb_set_eeprom to modify the MAC
>> address to be invalid, the igb driver will fail to load. If there is no
>> network card device, the user must modify it to a valid MAC address by
>> other means.
>> 
>> Since the MAC address can be modified, then add a random valid MAC address
>> to replace the invalid MAC address in the driver can be workable, it can
>> continue to finish the loading, and output the relevant log reminder.
>> 
>> Signed-off-by: Lixue Liang <lianglixue@...atwall.com.cn>
>> ---
>> Changelog:
>> * v4:
>> - Change the igb_mian in the title to igb
>> - Fix dev_err message: replace "already assigned random MAC address" 
>>  with "Invalid MAC address. Assigned random MAC address" 
>> Suggested-by Tony <anthony.l.nguyen@...el.com>
>> 
>> * v3:
>> - Add space after comma in commit message 
>> - Correct spelling of MAC address
>> Suggested-by Paul <pmenzel@...gen.mpg.de>
>> 
>> * v2:
>> - Change memcpy to ether_addr_copy
>> - Change dev_info to dev_err
>> - Fix the description of the commit message
>> - Change eth_random_addr to eth_hw_addr_random
>> Reported-by: kernel test robot <lkp@...el.com>
>> 
>> drivers/net/ethernet/intel/igb/igb_main.c | 7 ++++---
>> 1 file changed, 4 insertions(+), 3 deletions(-)
>> 
>> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
>> index 34b33b21e0dc..5e3b162e50ac 100644
>> --- a/drivers/net/ethernet/intel/igb/igb_main.c
>> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
>> @@ -3359,9 +3359,10 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
>> 	eth_hw_addr_set(netdev, hw->mac.addr);
>> 
>> 	if (!is_valid_ether_addr(netdev->dev_addr)) {
>> -		dev_err(&pdev->dev, "Invalid MAC Address\n");
>> -		err = -EIO;
>> -		goto err_eeprom;
>> +		eth_hw_addr_random(netdev);
>> +		ether_addr_copy(hw->mac.addr, netdev->dev_addr);
>> +		dev_err(&pdev->dev,
>> +			"Invalid MAC address. Assigned random MAC address\n");
>> 	}
>> 
>> 	igb_set_default_mac_filter(adapter);
> 
> Losing the MAC address is one of the least destructive things you can
> do by poking the EEPROM manually. There are settings in there for other
> parts of the EEPROM for the NIC that can just as easily prevent the
> driver from loading, or worse yet even prevent it from appearing on the
> PCIe bus in some cases. So I don't see the user induced EEPROM
> corruption as a good justification for this patch as the user shouldn't
> be poking the EEPROM if they cannot do so without breaking things.
> 
> With that said I would be okay with adding this with the provision that
> there is a module parameter to turn on this funcitonality. The
> justification would be that the user is expecting to have a corrupted
> EEPROM because they are working with some pre-production board or
> uninitialized sample. This way if somebody is wanting to update the
> EEPROM on a bad board they can use the kernel to do it, but they have
> to explicitly enable this mode and not just have the fact that their
> EEPROM is corrupted hidden as error messages don't necessarily get
> peoples attention unless they are seeing some other issue.

Powered by blists - more mailing lists