| 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
| ||
|
Message-ID: <20150929233253.GS21513@n2100.arm.linux.org.uk> Date: Wed, 30 Sep 2015 00:32:53 +0100 From: Russell King - ARM Linux <linux@....linux.org.uk> To: netdev@...r.kernel.org Subject: Poor IPv6 TCP performance in 4.3-rc3 Hi, I'm seeing really poor IPv6 performance compared to IPv4. I've checked using two different ARM platforms - an iMX6 platform using the FEC driver, and an Armada 38x using mvneta. The following was captured using iperf between the target system and my laptop. The problem only occurs one-way. The 4.3-rc3 platform is running iperf in server mode, the laptop is in client mode. Armada 38x: ipv6: [ 4] 0.0-23.9 sec 170 KBytes 58.3 Kbits/sec ipv4: [ 4] 0.0-10.0 sec 1.09 GBytes 936 Mbits/sec iMX6Q: ipv6: [ 4] 0.0-11.1 sec 640 KBytes 474 Kbits/sec ipv4: [ 4] 0.0-10.0 sec 655 MBytes 549 Mbits/sec iMX6D with 4.2: ipv6: [ 4] 0.0-10.0 sec 685 MBytes 574 Mbits/sec ipv4: [ 4] 0.0-10.0 sec 696 MBytes 583 Mbits/sec It looks like there's an IPv6 regression between 4.2 and 4.3-rc3. Turning GRO off on Armada 38x gives: ipv6: [ 4] 0.0-10.0 sec 1.08 GBytes 923 Mbits/sec ipv4: [ 5] 0.0-10.0 sec 1.09 GBytes 936 Mbits/sec I haven't started to debug yet, but I thought I'd post a heads-up in case it's a known problem. I'll try to get some packet logs on Thursday, and I'll try to bisect. -- FTTC broadband for 0.8mile line: currently at 9.6Mbps down 400kbps up according to speedtest.net. -- 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