netdev - Re: [PATCH net-next 1/3] dcbnl: adding DCBX engine capability

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <4D1B6DA0.50506@intel.com>
Date:	Wed, 29 Dec 2010 09:19:28 -0800
From:	John Fastabend <john.r.fastabend@...el.com>
To:	Shmulik Ravid <shmulikr@...adcom.com>
CC:	"davem@...emloft.net" <davem@...emloft.net>,
	Eilon Greenstein <eilong@...adcom.com>,
	"Liu, Lucy" <lucy.liu@...el.com>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next 1/3] dcbnl: adding DCBX engine capability

On 12/29/2010 5:59 AM, Shmulik Ravid wrote:
> 
> On Tue, 2010-12-28 at 16:05 -0800, John Fastabend wrote:
> 
>> I would like to use these bits for guests using a VF as well. The
>>  problem is multiple lldp agents advertising dcbx tlvs on the same link
>>  is not spec compliant. In the VF case there may or may not be a
>>  hardware lldp engine all the VF driver (ie ixgbevf) should need to
>>  know is that some other entity is managing the DCB attributes.
>>
>> To reflect this I would propose changing DCB_CAP_DCBX_HW and the comments by removing "HW". The two ideas I had were DCB_CAP_DCBX_READONLY or DCB_CAP_DCBX_LLD_MANAGED. Admittedly a bit of a nitpick but its a bit confusing to set the DCBX_HW bit when there really is no HW engine in the 82599 adapter case.
>>
>> Otherwise this all looks good to me. I was hoping someone would get around to this. Thanks a lot!
>>
>> -- John.
>>
> 
> I see your point, I like DCB_CAP_DCBX_LLD_MANAGED better. I will change
> it an resubmit the patches. DCB_CAP_DCBX_HW implies 3 things:
> 1. DCBX negotiation is managed by some other entity.
> 2. The user can use the 'get' routines to get the negotiation results.
> 3. The user can use the 'set' routines to set the initial configuration
> for the negotiation. 
> I think No 3. is irrelevant to VFs, that is you don't want multiple VF
> drivers trying to change the initial configuration settings. I can think
> of 2 ways to make this distinction, the first is for the VF driver not
> to support the 'set' routines (or always return an error value). The
> second is to add another attribute flag: DCB_CAP_DCBX_LLD_CFG which will
> indicate exactly No 3. while DCB_CAP_DCBX_LLD_MANAGED will indicate No
> 1. and 2. The VF driver will declare DCB_CAP_DCBX_LLD_MANAGED only and a
> driver for an embedded DCBX engine will declare both flags. What do you
> think?
> 
> Thanks
> Shmulik.
> 
>  
> 
> 

I think having the VF driver not support the 'set' routines is good enough. This is inline with how we handle other operations not supported by the lower layer device. Adding another bit would work as well but it seems simpler to only add flags that are needed.

Then dcbnl should return EOPNOTSUPP for this case.

-- John
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html