vr(4) troubles for AMD Geode CS5536 chipset

[Adrian Chadd]

waaaaaaaaait.

Did you say that removing preemption from your kernel fixed your
performance issues?

If that's the case, we need to figure out why both you and i see that..



Adrian


On 3 September 2012 02:23, Eugene Grosbein <egrosbein at rdtc.ru> wrote:
> 04.09.2012 04:43, YongHyeon PYUN пишет:
>
>>> None. I've read its source and learned it prints its debug
>>> to the kernel dmesg buffer with sysctl dev.vr.1.stats=1
>>> and done that before, during and after noted failure -
>>> all counters are zero except of good frames conters (in/out).
>>
>> Ok, it seems you didn't encounter TX/RX MAC shutdown/restart
>> related issues.  To get more verbose debugging messages, you have
>> to define VR_SHOW_ERRORS in if_vr.c.
>
> I'll try this evening.
>
>> BTW, I see really poor bulk TCP performance on vr(4) with CURRENT
>> (< 12 Mbps). :-(
>> This is a quad-port VT6105 with Core2 Duo E6550.  Pre-r235334
>> restores the old good performance for me(> 86Mbps). However I can't
>> explain how taskq change shows this huge difference.
>
> Have you tried to remove options PREEMPTING from the kernel for newest driver?
> That helped me a lot - performance of new driver returned to the level of old one.
>
> Eugene Grosbein
>
> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"