Dear Developers,
I have a setup of a pair of SuperMicro-branded XL710s connected with
passive crossover cable, one (eth4) is using firmware 5.0, the other (eth7)
is using fw 4.33. When I send couple of packets with no payload in GRE over
IP over ETH from eth7 to eth4, the receiving card stops processing further
packets. It seems that 4 empty-GRE packets in succession are enough to kill
the card's receivers (all ports stop working), more if there are other
packets inbetween. Transmit direction is not affected. After reboot, the
card comes back, reloading the driver (or unbind/bind) is not enough to fix
the problem.
14:31:24.299203 00:00:00:00:00:00 > ff:ff:ff:ff:ff:ff, ethertype IPv4
(0x0800), length 60: (tos 0x0, ttl 64, id 1, offset 0, flags [none], proto
GRE (47), length 24)
127.0.0.1 > 127.0.0.1: GREv0, Flags [none], proto unknown (0x0000),
length 4
gre-proto-0x0
0x0000: 4500 0018 0001 0000 402f 7cb4 7f00 0001 E.......@/|.....
0x0010: 7f00 0001 0000 0000 0000 0000 0000 0000 ................
0x0020: 0000 0000 0000 0000 0000 0000 0000 ..............
scapy 2.2.0 generates this packet from: Ether()/IP()/GRE()
Tested on stock Debian kernel, with in-kernel driver and one from DPDK-2.2.
Linux lab1 3.16.0-4-amd64 #1 SMP Debian 3.16.7-ckt25-2 (2016-04-08) x86_64
GNU/Linux
Best Regards,
Michał Mirosław
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit
http://communities.intel.com/community/wired
