Re: [PATCH] if_dc cleanups..

Le Wednesday 31 May 2006 23:53, John Baldwin a écrit :
> >
> > one side effect seems to be more watchdog timeouts :
> >
> > dc0: watchdog timeout
> > dc0: failed to force tx and rx to idle state
> >
> > on a macroscopic level, the NIC is still usable (enough to get a cvs
> > update
>
> of
>
> > the full world from an NFS repository).
>
> It shouldn't have affect the operation of the device at all.  Do the probe
> messages in dmesg match the dmesg lines w/o the patch exactly?  Do you have
> before and after dmesg's that I can compare?

sorry, I did not come back : the watchdog must be due to bad PCI routing (a 
switch of slot resulted in the dc and USB boards working again)

This is how the NIC appeared on an older kernel : (in another PCI slot, 
perhaps ?)
+dc0: <82c169 PNIC 10/100BaseTX> port 0x9400-0x94ff mem 0xd8000000-0xd80000ff 
irq 19 at device 9.0 on pci0
+miibus0: <MII bus> on dc0
+bmtphy0: <BCM5201 10/100baseTX PHY> on miibus0
+bmtphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
+dc0: Ethernet address: 00:02:e3:08:63:af

this is exactly how the NIC appears today :
dc0: <82c169 PNIC 10/100BaseTX> port 0xc400-0xc4ff mem 0xd8002000-0xd80020ff 
irq 19 at device 9.0 on pci0
miibus0: <MII bus> on dc0
bmtphy0: <BCM5201 10/100baseTX PHY> on miibus0
bmtphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
dc0: Ethernet address: 00:02:e3:08:63:af

using the sources with your patches :
# ls -la /usr/src/sys/dev/dc
total 624
drwxr-xr-x    3 root  wheel    512 May 29 23:18 .
drwxr-xr-x  174 root  wheel   2560 May 31 20:34 ..
drwxr-xr-x    2 root  wheel    512 May 29 19:53 CVS
-rw-r--r--    1 root  wheel  12440 May 25 09:25 dcphy.c
-rw-r--r--    1 root  wheel  12392 Oct 18  2005 dcphy.c.orig
-rw-r--r--    1 root  wheel  99043 May 25 09:25 if_dc.c
-rw-r--r--    1 root  wheel  98871 May 12 22:51 if_dc.c.orig
-rw-r--r--    1 root  wheel  34406 May 25 09:25 if_dcreg.h
-rw-r--r--    1 root  wheel  37329 Mar 17 06:57 if_dcreg.h.orig

	TfH