On Wed, Dec 31, 2003 at 04:26:43PM +0100, Andreas Haumer wrote:
>...
> Hi!
Hi Andreas!
>...
> Here's a first report:
>...
> *) Unresolved symbol in kernel/drivers/net/wan/comx.o: proc_get_inode
>...
CONFIG_COMX=m always gave an unresolved symbol in kernel 2.4, and it
seems noone is interested in properly fixing it.
The patch below disallows a modular CONFIG_COMX.
cu
Adrian
--- linux-2.4.25-pre4-modular/drivers/net/wan/Config.in.old 2004-01-11 01:38:08.000000000 +0100
+++ linux-2.4.25-pre4-modular/drivers/net/wan/Config.in 2004-01-11 01:39:20.000000000 +0100
@@ -23,7 +23,7 @@
# COMX drivers
#
- tristate ' MultiGate (COMX) synchronous serial boards support' CONFIG_COMX
+ bool ' MultiGate (COMX) synchronous serial boards support' CONFIG_COMX
if [ "$CONFIG_COMX" != "n" ]; then
dep_tristate ' Support for COMX/CMX/HiCOMX boards' CONFIG_COMX_HW_COMX $CONFIG_COMX
dep_tristate ' Support for LoCOMX board' CONFIG_COMX_HW_LOCOMX $CONFIG_COMX
--- linux-2.4.25-pre4-modular/Documentation/Configure.help.old 2004-01-11 01:39:35.000000000 +0100
+++ linux-2.4.25-pre4-modular/Documentation/Configure.help 2004-01-11 01:40:14.000000000 +0100
@@ -11149,9 +11149,6 @@
You must say Y to "/proc file system support" (CONFIG_PROC_FS) to
use this driver.
- If you want to compile this as a module, say M and read
- <file:Documentation/modules.txt>. The module will be called comx.o.
-
Support for COMX/CMX/HiCOMX boards
CONFIG_COMX_HW_COMX
Hardware driver for the 'CMX', 'COMX' and 'HiCOMX' boards from the
Adrian Bunk wrote:
> On Wed, Dec 31, 2003 at 04:26:43PM +0100, Andreas Haumer wrote:
>
>>...
>>Hi!
>
>
> Hi Andreas!
>
>
>>...
>>Here's a first report:
>>...
>>*) Unresolved symbol in kernel/drivers/net/wan/comx.o: proc_get_inode
>>...
>
>
> CONFIG_COMX=m always gave an unresolved symbol in kernel 2.4, and it
> seems noone is interested in properly fixing it.
>
> The patch below disallows a modular CONFIG_COMX.
I agree with the intent...
At this point, I am tempted to simply comment it out of the Config.in,
and let interested parties backport bug fixes and crap from 2.6 if they
would like. The driver has had obvious bugs for a while...
Jeff
On Mon, Jan 12, 2004 at 05:24:49PM -0500, Jeff Garzik wrote:
> Adrian Bunk wrote:
> >
> >CONFIG_COMX=m always gave an unresolved symbol in kernel 2.4, and it
> >seems noone is interested in properly fixing it.
> >
> >The patch below disallows a modular CONFIG_COMX.
>
> I agree with the intent...
>
> At this point, I am tempted to simply comment it out of the Config.in,
> and let interested parties backport bug fixes and crap from 2.6 if they
> would like. The driver has had obvious bugs for a while...
In 2.6 the driver is marked BROKEN...
I only checked the compilation, if the driver compiled statically works
I don't think it's a good idea to completely disable it.
> Jeff
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Mon, Jan 12, 2004 at 05:24:49PM -0500, Jeff Garzik wrote:
> I agree with the intent...
>
> At this point, I am tempted to simply comment it out of the Config.in,
> and let interested parties backport bug fixes and crap from 2.6 if they
> would like. The driver has had obvious bugs for a while...
In 2.6 it's as buggy as in 2.4..
On Tue, Jan 13, 2004 at 09:57:11AM +0000, Christoph Hellwig wrote:
> On Mon, Jan 12, 2004 at 05:24:49PM -0500, Jeff Garzik wrote:
> > I agree with the intent...
> >
> > At this point, I am tempted to simply comment it out of the Config.in,
> > and let interested parties backport bug fixes and crap from 2.6 if they
> > would like. The driver has had obvious bugs for a while...
>
> In 2.6 it's as buggy as in 2.4..
The driver is full of obvious bugs, had been that way for years, has
userland API broken by design and maintainers who couldn't be arsed
to fix it.
By now I don't see how anything short of rm(1) will help that animal.