Re: 2.6.1-mm4

Re: 2.6.1-mm4

[J.A. Magallon]

On 01.16, Andrew Morton wrote:
> 
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.1/2.6.1-mm4/
> 
> 

Net driver problem:

werewolf:/etc# modprobe --verbose 3c59x
insmod /lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko 
FATAL: Error inserting 3c59x (/lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko): Invalid argument

/var/messages:
Jan 18 01:03:00 werewolf kernel: 3c59x: falsely claims to have parameter rx_copybreak

Hardware:

00:12.0 Ethernet controller: 3Com Corporation 3c980-TX 10/100baseTX NIC [Python-T] (rev 78)
        Subsystem: 3Com Corporation: Unknown device 1000
        Flags: bus master, medium devsel, latency 32, IRQ 5
        I/O ports at ec00 [size=128]
        Memory at febfef80 (32-bit, non-prefetchable) [size=128]
        Expansion ROM at feba0000 [disabled] [size=128K]
        Capabilities: [dc] Power Management version 2

(if you answer from netdev, plz CC: me, I'm not subscribed. thanks)

TIA

-- 
J.A. Magallon <jamagallon()able!es>     \                 Software is like sex:
werewolf!able!es                         \           It's better when it's free
Mandrake Linux release 10.0 (Cooker) for i586
Linux 2.6.1-jam4 (gcc 3.3.2 (Mandrake Linux 10.0 3.3.2-4mdk))

Re: 2.6.1-mm4

[Andrew Morton]

"J.A. Magallon" <jamagallon@able.es> wrote:
>
> On 01.16, Andrew Morton wrote:
>  > 
>  > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.1/2.6.1-mm4/
>  > 
>  > 
> 
>  Net driver problem:
> 
>  werewolf:/etc# modprobe --verbose 3c59x
>  insmod /lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko 
>  FATAL: Error inserting 3c59x (/lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko): Invalid argument

hmm, cute.

--- 25/drivers/net/3c59x.c~3c59x-modprobe-fix	2004-01-17 21:49:14.000000000 -0800
+++ 25-akpm/drivers/net/3c59x.c	2004-01-17 21:49:18.000000000 -0800
@@ -211,11 +211,11 @@
 /* Set the copy breakpoint for the copy-only-tiny-frames scheme.
    Setting to > 1512 effectively disables this feature. */
 #ifndef __arm__
-static const int rx_copybreak = 200;
+static int rx_copybreak = 200;
 #else
 /* ARM systems perform better by disregarding the bus-master
    transfer capability of these cards. -- rmk */
-static const int rx_copybreak = 1513;
+static int rx_copybreak = 1513;
 #endif
 /* Allow setting MTU to a larger size, bypassing the normal ethernet setup. */
 static const int mtu = 1500;

_

Re: 2.6.1-mm4

[J.A. Magallon]

On 01.18, Andrew Morton wrote:
> "J.A. Magallon" <jamagallon@able.es> wrote:
> >
> > On 01.16, Andrew Morton wrote:
> >  > 
> >  > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.1/2.6.1-mm4/
> >  > 
> >  > 
> > 
> >  Net driver problem:
> > 
> >  werewolf:/etc# modprobe --verbose 3c59x
> >  insmod /lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko 
> >  FATAL: Error inserting 3c59x (/lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko): Invalid argument
> 
> hmm, cute.
> 

Yes.
It worked. 
I thought of this, but why this and not the other parameters ? Compiler bug ?

Witches...

-- 
J.A. Magallon <jamagallon()able!es>     \                 Software is like sex:
werewolf!able!es                         \           It's better when it's free
Mandrake Linux release 10.0 (Cooker) for i586
Linux 2.6.1-jam4 (gcc 3.3.2 (Mandrake Linux 10.0 3.3.2-4mdk))

Re: 2.6.1-mm4

[Andrew Morton]

"J.A. Magallon" <jamagallon@able.es> wrote:
>
> 
> On 01.18, Andrew Morton wrote:
> > "J.A. Magallon" <jamagallon@able.es> wrote:
> > >
> > > On 01.16, Andrew Morton wrote:
> > >  > 
> > >  > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.1/2.6.1-mm4/
> > >  > 
> > >  > 
> > > 
> > >  Net driver problem:
> > > 
> > >  werewolf:/etc# modprobe --verbose 3c59x
> > >  insmod /lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko 
> > >  FATAL: Error inserting 3c59x (/lib/modules/2.6.1-jam4/kernel/drivers/net/3c59x.ko): Invalid argument
> > 
> > hmm, cute.
> > 
> 
> Yes.
> It worked. 
> I thought of this, but why this and not the other parameters ? Compiler bug ?

Presumably, recent gcc's remove the variable altogether and just expand the
constant inline.  When the central module code checks for the parameter's
existence in the module's symbol table it errors out.

Re: 2.6.1-mm4

[Rusty Russell]

On Sun, 18 Jan 2004 00:17:08 -0800
Andrew Morton <akpm@osdl.org> wrote:

> Presumably, recent gcc's remove the variable altogether and just expand the
> constant inline.  When the central module code checks for the parameter's
> existence in the module's symbol table it errors out.

MODULE_PARM considered harmful.

Unfortunately, there's no easy way of fixing this, since MODULE_PARM()
is often used on variables which aren't declared yet 8(.  (I tried this
in an early patch).

Migrating to module_param() is the Right Thing here IMHO, which actually
takes the damn address,

Rusty.
-- 
   there are those who do and those who hang on and you don't see too
   many doers quoting their contemporaries.  -- Larry McVoy