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 PHC | |
Open Source and information security mailing list archives
| ||
|
Date: Fri, 2 May 2014 10:58:55 +0200 From: Yegor Yefremov <yegorslists@...glemail.com> To: Mugunthan V N <mugunthanvnm@...com> Cc: David Miller <davem@...emloft.net>, netdev <netdev@...r.kernel.org> Subject: Re: [net PATCH 1/1] drivers: net: cpsw: Add default vlan for dual emac case also On Wed, Apr 16, 2014 at 3:58 PM, Yegor Yefremov <yegorslists@...glemail.com> wrote: > On Wed, Apr 16, 2014 at 7:57 AM, Mugunthan V N <mugunthanvnm@...com> wrote: >> On Tuesday 15 April 2014 07:16 PM, Yegor Yefremov wrote: >>> On Wed, Apr 9, 2014 at 8:56 PM, David Miller <davem@...emloft.net> wrote: >>>> From: Mugunthan V N <mugunthanvnm@...com> >>>> Date: Wed, 9 Apr 2014 11:34:40 +0530 >>>> >>>>> Dual EMAC works with VLAN segregation of the ports, so default vlan needs >>>>> to be added in dual EMAC case else default vlan will be tagged for all >>>>> egress packets and vlan unaware switches/servers will drop packets >>>>> from the EVM. >>> Hi Mugunthan, >>> >>> though this patch fixes drop packets issue, it introduces another >>> issue. Just connect both interfaces to one switch and configure >>> different subnets. As soon as you get a broadcast you have an endless >>> loop and network communication is not possible at all. Before this >>> patch you could operate as if you had really two independent NICs. >>> >>> >> >> This patch doesn't add any multicast or unicast entry in the ALE table, >> so there is no possible way that a broadcast packet is switched within >> the two external ports, and this patch removes the encapsulation of the >> default vlan id which is used for internal packet switching inside CPSW >> switch from host and any packet received from external port will be >> tagged with port vlan so internal switching will happen with that port >> vlan-id which will be only that physical port and host port. so the >> scenario which you mentioned cannot happen theoretically untill you >> enable bridge between the dual ethernet interface and connect both to >> the same interface. > > Have you tried the described setup? My system is Buildroot on > am335x-evmsk with almost empty /etc/network/interfaces > > # Configure Loopback > auto lo > iface lo inet loopback > > No bridges: > > # brctl show > bridge name bridge id STP enabled interfaces > > And following address setup made manually: > > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN > group default > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > inet 127.0.0.1/8 scope host lo > valid_lft forever preferred_lft forever > 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > state UP group default qlen 1000 > link/ether 00:18:31:93:f6:49 brd ff:ff:ff:ff:ff:ff > inet 192.168.1.234/24 brd 192.168.1.255 scope global eth0 > valid_lft forever preferred_lft forever > 3: eth1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast > state DOWN group default qlen 1000 > link/ether 56:90:76:5b:b5:f3 brd ff:ff:ff:ff:ff:ff > inet 192.168.4.234/24 brd 192.168.4.255 scope global eth1 > valid_lft forever preferred_lft forever > 4: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group > default qlen 1000 > link/ether de:ad:be:ef:00:00 brd ff:ff:ff:ff:ff:ff > > Routing table: > > # ip route show > 192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.234 > 192.168.4.0/24 dev eth1 proto kernel scope link src 192.168.4.234 > > Same behavior on 3.14.0-rc4-12740-gcc09039-dirty and 3.15.0-rc1-00002-ge0f8779 > > As soon as insert the cable into eth1, my switch shows very high > activity and the whole LAN in the company is dead till I remove the > cable. I have no problems with the version before the fix. I don't > really know, what happens inside the driver, but could make the same > setup (best of all in a LAN with many Windows PCs, that send lots of > broadcast) and let me know if you see the same behavior. This bug seems to be in 3.12 too. See this thread (http://e2e.ti.com/support/arm/sitara_arm/f/791/p/338292/1180954.aspx#1180954). Thanks. Yegor -- 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
Powered by blists - more mailing lists