On 9 January 2016 at 17:21, Roman Yeryomin <[email protected]> wrote: > While trying to optimize ag71xx driver performance I've noticed a > weird thing: NAT performance may vary across reboots. > For example on Archer C7 the difference can be 80Mbps: 440 one boot, > 360 another. On AP152 the numbers differ but the difference across > reboots still present. > That is with iperf3, TCP download, 1m average, confirmed environment, > trunk version. > > Did someone notice this before? Is this driver specific? MIPS > specific? Something else? > > Unfortunately I don't have any other board with gigabit ports I can > test with currently. > Such behavior makes it very hard to test any optimizations. >
Really? nobody noticed this? Regards, Roman _______________________________________________ openwrt-devel mailing list [email protected] https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
