em 6.6.6 - watchdog timeout

[Jeremy Chadwick]

On Sat, Oct 20, 2007 at 11:21:10AM +1300, Philip Murray wrote:
> "me too" on a Supermicro 5015MT+, although I notice my em0 is also sharing 
> an interrupt with USB (uhci3)... not sure if that's the culprit.

I'm not aware of a 5015MT+ model.  Maybe you mean 5015M-MT+ or
5015M-T+?

We have two 5015M-T+ boxes, one running RELENG_7 and one running
RELENG_6, and neither exhibit this problem.  Here's relevant em(4)
information from both boxes; I do find the vmstat -i IRQ on the 2nd
box a little odd, but operationally it seems fine.

box1 (RELENG_6)
===============
em0: <Intel(R) PRO/1000 Network Connection Version - 6.2.9> port 0x4000-0x401f mem 0xe0200000-0xe021ffff irq 16 at device 0.0 on pci13
em1: <Intel(R) PRO/1000 Network Connection Version - 6.2.9> port 0x5000-0x501f mem 0xe0300000-0xe031ffff irq 17 at device 0.0 on pci14

irq16: em0 uhci3               117371806         11
irq17: em1                      49605005          4


box2 (RELENG_7)
===============
em0: <Intel(R) PRO/1000 Network Connection Version - 6.5.3> port 0x4000-0x401f mem 0xe8000000-0xe801ffff irq 16 at device 0.0 on pci13
em1: <Intel(R) PRO/1000 Network Connection Version - 6.5.3> port 0x5000-0x501f mem 0xe8200000-0xe821ffff irq 17 at device 0.0 on pci14

irq256: em0                       502854          4
irq257: em1                         5438          0


On box2, IRQ 16 is also shared with uhci3 and vgapci0, but vmstat
doesn't seem to show that.

uhci3: <UHCI (generic) USB controller> port 0x3060-0x307f irq 16 at device 29.3 on pci0
vgapci0: <VGA-compatible display> port 0x6000-0x60ff mem 0xe0000000-0xe7ffffff,0xe8300000-0xe830ffff irq 16 at device 0.0 on pci15

-- 
| Jeremy Chadwick                                    jdc at parodius.com |
| Parodius Networking                           http://www.parodius.com/ |
| UNIX Systems Administrator                      Mountain View, CA, USA |
| Making life hard for others since 1977.                  PGP: 4BD6C0CB |