All of lore.kernel.org
 help / color / mirror / Atom feed
* 6PACK - Another question
@ 2005-11-02  5:39 Ray Wells
  2005-11-02  9:13 ` Ralf Baechle DL5RB
  0 siblings, 1 reply; 3+ messages in thread
From: Ray Wells @ 2005-11-02  5:39 UTC (permalink / raw)
  To: linux-hams

Hi All,

In the documentation /.../6pack.txt, the author says the driver has only 
been tested as a module and not built-in to the kernel.

Has anyone on this list used the 6pack driver in the built-in version 
with kernel 2.2.20?

I always opt for built-in drivers for stuff I know I'll use and maybe 
this has been my downfall in this case :-(

No drama, of course, to recompile the kernel with the driver as a module 
but I may as well test the waters first :-)

Regards

Ray Wells VK2TV

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: 6PACK - Another question
  2005-11-02  5:39 6PACK - Another question Ray Wells
@ 2005-11-02  9:13 ` Ralf Baechle DL5RB
  2005-11-03  2:54   ` Ray Wells
  0 siblings, 1 reply; 3+ messages in thread
From: Ralf Baechle DL5RB @ 2005-11-02  9:13 UTC (permalink / raw)
  To: Ray Wells; +Cc: linux-hams

On Wed, Nov 02, 2005 at 04:39:26PM +1100, Ray Wells wrote:

> In the documentation /.../6pack.txt, the author says the driver has only 
> been tested as a module and not built-in to the kernel.
> 
> Has anyone on this list used the 6pack driver in the built-in version 
> with kernel 2.2.20?

Linux 2.2, that's extremly old - ages from before I took over it's
maintainership.  But the code actually looks okay.  Sure you don't want
to upgrade to a piece of software from this decade?

> I always opt for built-in drivers for stuff I know I'll use and maybe 
> this has been my downfall in this case :-(

For custom kernels that will save a tiny amount of memory and give a small,
almost unmeassurable performance gain.  You lose the advantage of for
example being able to re-initialize a subsystem by unloading and reloading
it's module.

Given all that what learned about problems with the 2.2 module loader we
can probably declare it a buggy horror trip from today's perspective.

> No drama, of course, to recompile the kernel with the driver as a module 
> but I may as well test the waters first :-)

The 2.2 6pack driver looks ok, certainly up to the standards of those days
and I wouldn't expect any problems with it from linking it.

73 de DL5RB op Ralf

--
Loc. JN47BS / CQ 14 / ITU 28 / DOK A21

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: 6PACK - Another question
  2005-11-02  9:13 ` Ralf Baechle DL5RB
@ 2005-11-03  2:54   ` Ray Wells
  0 siblings, 0 replies; 3+ messages in thread
From: Ray Wells @ 2005-11-03  2:54 UTC (permalink / raw)
  To: Ralf Baechle DL5RB; +Cc: linux-hams



Ralf Baechle DL5RB wrote:

>On Wed, Nov 02, 2005 at 04:39:26PM +1100, Ray Wells wrote:
>
>  
>
>>In the documentation /.../6pack.txt, the author says the driver has only 
>>been tested as a module and not built-in to the kernel.
>>
>>Has anyone on this list used the 6pack driver in the built-in version 
>>with kernel 2.2.20?
>>    
>>
>
>Linux 2.2, that's extremly old - ages from before I took over it's
>maintainership.  But the code actually looks okay.  Sure you don't want
>to upgrade to a piece of software from this decade?
>  
>

I've stuck with 2.2.x kernels because they have been reliable. When I 
started looking at 2.4.x kernels there appeared to be unresolved issues 
with the USCC driver, and since I run a USCC>4 card I decided to stick 
with what was already working well. I still can't see any compelling 
reason to move away from the 2.2.20 kernel.

>  
>
>>I always opt for built-in drivers for stuff I know I'll use and maybe 
>>this has been my downfall in this case :-(
>>    
>>
>
>For custom kernels that will save a tiny amount of memory and give a small,
>almost unmeassurable performance gain.  You lose the advantage of for
>example being able to re-initialize a subsystem by unloading and reloading
>it's module.
>
>Given all that what learned about problems with the 2.2 module loader we
>can probably declare it a buggy horror trip from today's perspective.
>  
>
>>No drama, of course, to recompile the kernel with the driver as a module 
>>but I may as well test the waters first :-)
>>    
>>
>
>The 2.2 6pack driver looks ok, certainly up to the standards of those days
>and I wouldn't expect any problems with it from linking it.
>  
>
I'll recompile with 6pack as a module and see what happens. I really 
don't have the time to spend sorting out the problems that will arise 
from changing to a leter kernel. KISS has served me well for 6 years so, 
if I can't get 6pack working under 2.2.20 I'll just abandon it and stay 
with KISS.

>73 de DL5RB op Ralf
>
>--
>Loc. JN47BS / CQ 14 / ITU 28 / DOK A21
>
>
>  
>
Cheers ... Ray VK2TV

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2005-11-03  2:54 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-11-02  5:39 6PACK - Another question Ray Wells
2005-11-02  9:13 ` Ralf Baechle DL5RB
2005-11-03  2:54   ` Ray Wells

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.