netdev - Re: Re: [PATCH 18/23 v3] mlx4_core: Managing common port filters by

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-next>] [day] [month] [year] [list]

Date:	Wed, 10 Feb 2010 19:22:51 +0200
From:	Yevgeny Petrilin <yevgenyp@...lanox.co.il>
To:	or.gerlitz@...il.com
CC:	yevgenyp@...lanox.co.il, Roland Dreier <rdreier@...co.com>,
	netdev@...r.kernel.org, tziporet@...lanox.co.il,
	liranl@...lanox.co.il
Subject: Re: Re: [PATCH 18/23 v3] mlx4_core: Managing common port filters
 by

On Sunday -10,January,-28163 09:59 PM, Or Gerlitz [or.gerlitz@...il.com] wrote:
> Yevgeny Petrilin <yevgenyp@...lanox.co.il> wrote:
>> The Multicast filter configuration is done by the master, that manages 
>> the filter which is common for all the functions.
>> The VLAN filter is a bitwise OR of all the VLAN filters for all 
>> functions, the result is a false-positive filter.
> 
> Reading through patches "18/23 v3 mlx4_core: Managing common port filters..." &&
> "16/23 v3 mlx4_core: Multi-Function MCG support" I have difficulties to follow on the vlan and multicast filtering related changes and proposed architecture.
> 
> Can you explain them in a bit more detailed fashion? For example, in the multicast case I believe that currently a bloom filter is applied  at the port level and each packet which pass the filter and has no registered QP is consumed by the EN driver.
> At some point
> my understanding was that everyone (all PFs, all VFs, etc) move to exact match, is it correct?
> 
> As for the false-positive vlan filter, I understand that this means further filtering has to be applied by some driver. Where you thinking on the 802q driver? what about the case where a VF driver is assigned a vlan, does the mlx4 VF code does such filtering?
> 
> 
> Or.
The multicast bloom filter is a false-positive filter per port.
All multicast packets go through the Multicast tables mechanism where every PF/VF is registered for every multicast address it listens to.
We will add support for per-Function multicast promiscuous mode.

The deal with the VLAN filter is similar, it includes all VIDs that some PF/VF requested.
We plan to add a MAC+VLAN filter implementation, that would cancel the need in this filter.

Yevgeny

--
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