AX.25 kernel module development

AX.25 kernel module development

[Øyvind Hanssen]

Please excuse me if this has been discussed before.. 

There are some talk here about that the AX.25 Linux kernel code is broken. 
There has been attempts to create patches for the kernel; some are good but
arent (yet) included in the mainstream kernel, and there has been a serious 
attempt to rewrite the kernel stack as well (?) And it seems to be some 
uncertainty wrt. what is happening in Linux 2.5 (?)

It seem complicated to maintain patches for the Linux kernel the 
way it has been done, and I notice that someone has even suggested to 
implement the whole thing in userspace instead, at least to make it easier
to develop new protocols and drivers... 

What about keeping the core AX.25 stuff and drivers in kernel but split it
out from the mainstream Linux codebase and do the development in a separate
project instead; like e.g. the ALSA project has done until now? I guess 
that it would be easier to compile, maintain and add features to such 
a package than getting patches accepted for the mainstream Linux kernel. 

It is possible that a small part must remain in the kernel (socket API?), but
we shouldnt need to change this part too often. 

73 de LA7ECA, Øyvind
-
To unsubscribe from this list: send the line "unsubscribe linux-hams" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: AX.25 kernel module development

[M Taylor]

On Tue, Apr 08, 2003 at 02:36:15PM +0200, Øyvind Hanssen wrote:
> Please excuse me if this has been discussed before.. 
> 
> attempt to rewrite the kernel stack as well (?) And it seems to be some 
> uncertainty wrt. what is happening in Linux 2.5 (?)

Some (all?) of Ralf's patches are in the 2.5.x (~66), and the rest of his
work is available via bitkeeper.

> What about keeping the core AX.25 stuff and drivers in kernel but split it
> out from the mainstream Linux codebase and do the development in a separate
> project instead; like e.g. the ALSA project has done until now? I guess 

Gak, no please. ALSA has stablized and intergrated into 2.5 kernel for a
reason. There is no need for a massive disruption of the AX.25 protocol(s)
and drivers, they are simple and small compared to the rest of the kernel.

Considering that a growing number of users (percentage wise) seem to
find building a custom stock kernel a challenge that they would rather
avoid, making the build process more complex would be a major stumbling
block for these users who are finding it hard enough. Plus, given how
little the amateur radio section is actually changed, I think it would
simply be creating work.

If there were more than the 2-3 people working on the kernel(ax25) code
and more than ~5 drivers being developed I might consider it worthwhile.

-
To unsubscribe from this list: send the line "unsubscribe linux-hams" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html