Re: Annoying problem with udev

[Richard Mortimer <richm@oldelvet.org.uk>]

Apologies. It seems I forgot to reply all earlier! I had use cas instead 
of cp in a variable name. Other comments inline below.

On 02/12/2010 20:13, Frans van Berckel wrote:
> > > Got this build error facing up.
> > >
> > > /usr/bin/make -j5   ARCH=sparc \
> > > 			     image
> > > make[1]: Entering directory `/usr/src/linux-2.6.34.7'
> > >     CHK     include/linux/version.h
> > >     CHK     include/generated/utsrelease.h
> > >     CALL    scripts/checksyscalls.sh
> > >     CHK     include/generated/compile.h
> > >     CC      drivers/net/cassini.o
> > > drivers/net/cassini.c: In function ‘cas_get_vpd_info’:
> > > drivers/net/cassini.c:3382: warning: ISO C90 forbids mixed declarations and code
> > > drivers/net/cassini.c:3384: error: ‘cas’ undeclared (first use in this function)
> > > drivers/net/cassini.c:3384: error: (Each undeclared identifier is reported only once
> > > drivers/net/cassini.c:3384: error: for each function it appears in.)
> > > make[3]: *** [drivers/net/cassini.o] Error 1
> > > make[2]: *** [drivers/net] Error 2
> > > make[1]: *** [drivers] Error 2
> >
> > Hi,
> >
> > Ah a stupid naming error! Just edit  cas->of_node to be cp->of_node at
> > line 5329.
> >
> > I've attached a revised cassini.c.diff if you prefer to use patch.
> > First off remove the old patch
> >
> > patch -R -p0<  cassini.c.diff
> > then save/apply the revised one (sorry I used the same filenames)
> > patch -p0<  cassini.c
>
> Hi Richard, I love Nano. But his warning is printed, beside that it's
> building nicely now.
>
> /usr/bin/make -j5   ARCH=sparc \
> 			     image
> make[1]: Entering directory `/usr/src/linux-2.6.34.7'
>    CHK     include/linux/version.h
>    CHK     include/generated/utsrelease.h
>    CALL    scripts/checksyscalls.sh
>    CHK     include/generated/compile.h
>    CC      drivers/net/cassini.o
> drivers/net/cassini.c: In function ‘cas_get_vpd_info’:
> drivers/net/cassini.c:3382: warning: ISO C90 forbids mixed declarations
> and code

Yes that is because the C90 standard does not like the const unsigned 
char *addr half way down the function. Easy to fix but doesn't spoil the 
experiment.

>    LD      drivers/net/built-in.o
>    LD      drivers/built-in.o
>    LD      vmlinux.o
>    MODPOST vmlinux.o
>
> But how is this looking? The mac addresses are the right ones. So your
> idea, does what it got to do.

That is good news. Time to push the issue upstream. I'll fire off an
with the relevant info later.

Regards

Richard

> deblnxsrv225:~# dmesg | grep cassini
> [    0.000000] Linux version 2.6.34.7-cassini-2 (root@deblnxsrv225) (gcc
> version 4.4.5 (Debian 4.4.5-8) ) #3 SMP Thu Dec 2 20:26:09 CET 2010
> [   88.378444] cassini: cassini.c:v1.6 (21 May 2008)
> [   88.382765] cassini 0000:00:02.0: eth0: Sun Cassini+ (64bit/66MHz
> PCI/Cu) Ethernet[6] 00:03:ba:62:8d:61
> [   88.387758] cassini 0003:00:01.0: eth1: Sun Cassini+ (64bit/66MHz
> PCI/Cu) Ethernet[30] 00:03:ba:62:8d:62
> [   92.465010] usb usb1: Manufacturer: Linux 2.6.34.7-cassini-2 ohci_hcd
> [   92.612613] usb usb2: Manufacturer: Linux 2.6.34.7-cassini-2 ohci_hcd
> [  102.959308] cassini 0000:00:02.0: eth0: Link up at 100 Mbps,
> full-duplex
> [  102.969107] cassini 0000:00:02.0: eth0: Pause is enabled (rxfifo:
> 16384 off: 13312 on: 11776)
>
> Frans van Berckel