From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vlad Yasevich Subject: Re: [PATCH net-next v2] bridge: vlan: allow to suppress local mac install for all vlans Date: Fri, 28 Aug 2015 08:31:02 -0400 Message-ID: <55E05486.5090500@redhat.com> References: <1440549295-3979-1-git-send-email-razor@blackwall.org> <20150825.194222.390859854071446877.davem@davemloft.net> <20150825.230641.773630246486190390.davem@davemloft.net> <55DE98AF.8000503@cumulusnetworks.com> <2125A434-6529-4D5A-BA6B-9F64C6B7A8C0@cumulusnetworks.com> <55DFA1A3.30601@redhat.com> <691CF770-DDF3-4AC9-B99C-9640992037C5@cumulusnetworks.com> Reply-To: vyasevic@redhat.com Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Cc: bridge@lists.linux-foundation.org, netdev@vger.kernel.org, roopa , wkok@cumulusnetworks.com, David Miller To: Nikolay Aleksandrov Return-path: In-Reply-To: <691CF770-DDF3-4AC9-B99C-9640992037C5@cumulusnetworks.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: bridge-bounces@lists.linux-foundation.org Errors-To: bridge-bounces@lists.linux-foundation.org List-Id: netdev.vger.kernel.org On 08/27/2015 10:17 PM, Nikolay Aleksandrov wrote: >=20 >> On Aug 27, 2015, at 4:47 PM, Vlad Yasevich wrote= : >> >> On 08/27/2015 05:02 PM, Nikolay Aleksandrov wrote: >>> >>>> On Aug 26, 2015, at 9:57 PM, roopa wrote= : >>>> >>>> On 8/26/15, 4:33 AM, Nikolay Aleksandrov wrote: >>>>>> On Aug 25, 2015, at 11:06 PM, David Miller w= rote: >>>>>> >>>>>> From: Nikolay Aleksandrov >>>>>> Date: Tue, 25 Aug 2015 22:28:16 -0700 >>>>>> >>>>>>> Certainly, that should be done and I will look into it, but the >>>>>>> essence of this patch is a bit different. The problem here is not >>>>>>> the size of the fdb entries, it=E2=80=99s more the number of them= - having >>>>>>> 96000 entries (even if they were 1 byte ones) is just way too muc= h >>>>>>> especially when the fdb hash size is small and static. We could w= ork >>>>>>> on making it dynamic though, but still these type of local entrie= s >>>>>>> per vlan per port can easily be avoided with this option. >>>>>> 96000 bits can be stored in 12k. Get where I'm going with this? >>>>>> >>>>>> Look at the problem sideways. >>>>> Oh okay, I misunderstood your previous comment. I=E2=80=99ll look i= nto that. >>>>> >>>> I just wanted to add the other problems we have had with keeping the= se macs (mostly from userspace POV): >>>> - add/del netlink notification storms >>>> - and large netlink dumps >>>> >>>> In addition to in-kernel optimizations, will be nice to have a solut= ion that reduces the burden on userspace. That will need a newer netlink = dump format for fdbs. Considering all the changes needed, Nikolays patch = seems less intrusive. >>> >>> Right, we need to take these into account as well. I=E2=80=99ll conti= nue the discussion on this (or restart it) because >>> I looked into using a bitmap for the local entries only and while it = fixes the scalability issue, it presents >>> a few new ones which are mostly related to the fact that these entrie= s now exist only without a vlan >>> and if a new mac comes along which matches one of these but is in a v= lan, the entry will get created >>> in br_fdb_update() unless we add a second lookup, but that will slow = down the learning path. >>> Also this change requires an update of every fdb function that uses t= he vid as a key (every fdb function?!) >>> because now we can have the mac in two places instead of one which is= a pretty big churn with lots >>> of conditionals all over the place and I don=E2=80=99t like it. Addin= g this complexity for the local addresses only >>> seems like an overkill, so I think to drop this issue for now. >> >> I seem to recall Roopa and I and maybe a few others have discussing th= is a few >> years ago at plumbers, I can't remember the details any more. All the= se local >> addresses add a ton of confusion. Does anyone (Stephen?) remember wha= t the >> original reason was for all these local addresses? I wonder if we can = have >> a nob to disable all of them (not just per vlan)? That might be clean= er and >> easier to swallow. >> >=20 > Right, this would be the easiest way and if the others agree - I=E2=80=99= ll post a patch for it so we can > have some way to resolve it today and even if we fix the scalability is= sue, this is still a valid case > that some people don=E2=80=99t want local fdbs installed automatically. > Any objections to this ? >=20 >>> This patch (that works around the initial problem) also has these iss= ues. >>> Note that one way to take care of this in a more straight-forward way= would be to have each entry >>> with some sort of a bitmap (like Vlad has tried earlier) and then we = can combine the paths so most >>> of these issues disappear, but that will not be easy as was already c= ommented earlier. I=E2=80=99ve looked >>> briefly into doing this with rhashtable so we can keep the memory foo= tprint for each entry relatively >>> small but it still affects the performance and we can have thousands = of resizes happening.=20 >>> >> >> So, one of the earlier approaches that I've tried (before rhashtable w= as >> in the kernel) was to have a hash of vlan ids each with a data structu= re >> pointing to a list of ports for a given vlan as well as a list of fdbs= for >> a given vlan. As far as scalability goes, that's really the best appr= oach. >> It would also allow us to do packet accounting per vlan. The only con= cern >> at the time was performance of ingress lookup. I think rhashtables m= ight >> help with this as well as ability to grow the footprint of the vlan ha= sh >> table dynamically. >> >> -vlad >> > I=E2=80=99ll look into it but I=E2=80=99m guessing the learning will be= come a more complicated process with additional=20 > allocations and some hash handling. I don't remember learning being all that complicated. The hash only chan= ged under rtnl when vlans were added/removed. The nice this is that we wouldn't ne= ed to rebalance, because if the vlan is removed all fdb links get removed to= o. They don't move to another bucket (But that was with static hash. Need to loo= k at rhash in more detail). If you want, I might still have patches hanging around on my machine that= had a hash table implementation. I can send them to you. -vlad >=20 >>> On the notification side if we can fix that, we can actually delete t= he 96000 entries without creating a >>> huge notification storm and do a user-land workaround of the original= issue, so I=E2=80=99ll look into that next. >>> >>> Any comments or ideas are very welcome. >>> >>> Thank you, >>> Nik >=20