Conflict between nfe(4) and x11/nvidia-driver
Pyun YongHyeon
pyunyh at gmail.com
Mon Jul 23 10:09:42 UTC 2007
On Mon, Jul 23, 2007 at 01:35:27PM +0400, Andrey V. Elsukov wrote:
> Hi, All.
>
> I have notebook with NVIDIA based network adapter and video.
> >From dmesg:
>
> nvidia0: <GeForce Go 6100> mem 0xc2000000-0xc2ffffff,0xd0000000-0xdfffffff,0xc1000000-0xc1ffffff irq 5 at device 5.0 on pci0
> nvidia0: [GIANT-LOCKED]
> nvidia0: [ITHREAD]
> ...
> nfe0: <NVIDIA nForce 430 MCP13 Networking Adapter> port 0x30b8-0x30bf mem 0xc0007000-0xc0007fff irq 5 at device 20.0 on pci0
> miibus0: <MII bus> on nfe0
> rlphy0: <RTL8201L 10/100 media interface> PHY 1 on miibus0
> rlphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
> nfe0: Ethernet address: 00:90:f5:4f:18:1b
> nfe0: [FILTER]
>
> Some time ago i've used nve(4) and nvidia-driver without problems.
> But nve(4) is buggy and sometimes have watchdog timeouts.
>
> After switching to nfe(4) i can't use x11/nvidia-driver.
> After some time of running, Xorg process takes 100% of resources and
> system not responds. But i can power off my system via ACPI.
> After reboot i have following messages in the /var/log/messages:
>
> Jul 20 16:26:51 btr-nb kernel: nfe0: watchdog timeout (missed Tx interrupts) -- recovering
> Jul 20 16:27:39 btr-nb last message repeated 3 times
> Jul 20 16:27:58 btr-nb kernel: nfe0: watchdog timeout (missed Tx interrupts) -- recovering
> Jul 20 16:28:08 btr-nb kernel: acpi_button0: power button pressed
>
> nfe(4) without x11/nvidia-driver and x11/nvidia-driver without nfe(4)
> works stable.
> But i want use both nfe(4) and x11/nvidia-driver...
> Can somebody suggest me some tricks?
>
Show me the output of "vmstat -i".
I guess nfe(4) shares interrupt with your video adapter.
If this is the case use polling(4) which may fix the issue.
--
Regards,
Pyun YongHyeon
More information about the freebsd-current
mailing list