NetBSD-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Lots of Interrupts when accessing a NFS share



> Joel CARNAT <joel%carnat.net@localhost> wrote:
>
>> When I run `find /mnt/nfs -type f -name "*whatever*"` from my C3, I see
>> Interrupts going from ~200 to ~8000/10000.
> [...]
>> The network driver ?
>> ==========
>> rtk0 at pci0 dev 8 function 0: Realtek 8139 10/100BaseTX (rev. 0x10)
>> rtk0: interrupting at irq 11
>> rtk0: Ethernet address 00:30:18:4a:2e:88
>> rlphy0 at rtk0 phy 7: Realtek internal PHY
>> ==========
>
> Oh, rtk.  You seems to have tough luck with this card (landisk has rtk
> too :).

yeah... my Plextor has this one too:
rtk0 at pci0 dev 0 function 0: Realtek 8139 10/100BaseTX (rev. 0x20)
rtk0: interrupting at irq 5
rtk0: Ethernet address 00:d0:a9:01:18:bf
rlphy0 at rtk0 phy 7: Realtek internal PHY
rlphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto

and network speed really <3 letters word> and make it unusable as a NAS
server.

> To quote a comment at the beginning of
> src/sys/dev/ic/rtl81x9.c
>
>  * The RealTek 8139 PCI NIC redefines the meaning of 'low end.' This is
>  * probably the worst PCI ethernet controller ever made, with the possible
>  * exception of the FEAST chip made by SMC. The 8139 supports bus-master
>  * DMA, but it has a terrible interface that nullifies any performance
>  * gains that bus-master DMA usually offers.
> [...more rant...]
>
> I really don't know if our rtk driver can be improved to squeeze the
> last drip of performance out this card if there's still any, but that
> comment certainly doesn't look encouraging.
>

Should a switch to tech-kern then or fill a PR ?

  Jo



Home | Main Index | Thread Index | Old Index