On Thu, 23 Dec 2004, Christian Weisgerber wrote:
> After updating my 5.3-STABLE/alpha from Dec 4 to Dec 22 on RELENG_5,
> network _receiving_ throughput on its re(4) interface has collapsed
> to a maximum of 0.5-1.0 MB/s. (Figures from scp. Yes, I realize
> this is not the most suitable test, bu
Scott Long:
> I suspect that the buffers are being bounced all over the place in the
> if_re driver. Can you send me the output of 'sysctl hw.busdma' after
> the system has been under load?
sysctl: unknown oid 'hw.busdma'
Strangely enough, I'm running the affected Dec 22 kernel at the
moment, b
Christian Weisgerber wrote:
Robert Watson:
Could you use a tool like netperf to see whether the slowdown is specific
to TCP, or affects UDP also? There have been some TCP tweaks and
bugfixes, and this would help isolate that. Seeing the results of a
netperf run with the UDP_RR and UDP_STREAM tes
Robert Watson:
> Could you use a tool like netperf to see whether the slowdown is specific
> to TCP, or affects UDP also? There have been some TCP tweaks and
> bugfixes, and this would help isolate that. Seeing the results of a
> netperf run with the UDP_RR and UDP_STREAM tests in the "before" a
On Thu, 23 Dec 2004, Christian Weisgerber wrote:
> After updating my 5.3-STABLE/alpha from Dec 4 to Dec 22 on RELENG_5,
> network _receiving_ throughput on its re(4) interface has collapsed to a
> maximum of 0.5-1.0 MB/s. (Figures from scp. Yes, I realize this is not
> the most suitable test, b
After updating my 5.3-STABLE/alpha from Dec 4 to Dec 22 on RELENG_5,
network _receiving_ throughput on its re(4) interface has collapsed
to a maximum of 0.5-1.0 MB/s. (Figures from scp. Yes, I realize
this is not the most suitable test, but the box is not CPU starved
and used to receive data at s
