On 10.9.2015. 9:57, Claudio Jeker wrote: > On Wed, Sep 09, 2015 at 11:57:57PM +0200, Hrvoje Popovski wrote: >> On 9.9.2015. 10:10, Claudio Jeker wrote: >>> This is a port of the em(4) IPL_MPSAFE changes made by kettenis@ to ix(4). >>> Seems to work for me but don't expect any miracles. >>> >>> Please test >>> >> >> Hi, >> >> i am testing your patch with bridged and routed setup and everything >> works nice. ix card is dual port 82599ES and i'm generating from 5Mpps >> to 14Mpps on box that is connected to ix1. in routed setup box connected >> to ix0 receives around 750Kpps and in bridge setup around 600Kpps. >> > > There are still some issues with this first version. Here a 2nd version > that pushes the kernel lock even a bit further and tries to solve some of > the problems between the interrupt handler and the ioctl code. > There is still something not fully right here since my torture loop of > ifconfig ix0 down; sleep 1; ifconfig up; sleep 5 > is looking up my box after a while. >
#!/bin/sh while true; do ifconfig ix0 down; ifconfig ix1 down; sleep 1; ifconfig ix0 up; ifconfig ix1 up; sleep 5; done this script is working for 3 hours without problems... will let script working over night and will generate traffic through openbsd box... i noticed that when ifconfig up is executed that interrupts on cpu0 went to 30% for that moment ... don't know is this is expected