* module registration changes
@ 2000-12-18 0:29 David Woodhouse
2000-12-18 13:04 ` Erwin Authried
0 siblings, 1 reply; 5+ messages in thread
From: David Woodhouse @ 2000-12-18 0:29 UTC (permalink / raw)
To: eauth; +Cc: mtd
Please could people test the new stuff I've just committed to bypass the
inter_module_get code by using weakly-linked symbols? Especially on 2.0.
This should also fix the dependency on link order which was introduced by
the removal of get_module_symbol().
--
dwmw2
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: module registration changes
2000-12-18 0:29 module registration changes David Woodhouse
@ 2000-12-18 13:04 ` Erwin Authried
2000-12-18 15:50 ` Nicolas Pitre
0 siblings, 1 reply; 5+ messages in thread
From: Erwin Authried @ 2000-12-18 13:04 UTC (permalink / raw)
To: David Woodhouse; +Cc: mtd
there were a couple of small changes necessary to compile on 2.0.
Here are some other changes that I would like to commit too, please let me
know if its ok.:
* Use #ifdef NO_MM in mtdchar.c to avoid vmalloc if no MMU is used
(uClinux).
* "list.h" is missing in the 2.0 kernel source. The best thing that I could
think of is to simply add it
to MTD in linux/mtd/list.h.
* Somehow initcalls.c must be used for 2.0 kernels conditionally. Either in
the Makefile, or just
#include from mtdcore.c? I don't know how to do that in the Makefile nicely.
Regards,
Erwin
----- Original Message -----
From: David Woodhouse <dwmw2@infradead.org>
To: <eauth@softsys.co.at>
Cc: <mtd@infradead.org>
Sent: Monday, December 18, 2000 1:29 AM
Subject: module registration changes
>
> Please could people test the new stuff I've just committed to bypass the
> inter_module_get code by using weakly-linked symbols? Especially on 2.0.
>
> This should also fix the dependency on link order which was introduced by
> the removal of get_module_symbol().
>
> --
> dwmw2
>
>
>
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: module registration changes
2000-12-18 13:04 ` Erwin Authried
@ 2000-12-18 15:50 ` Nicolas Pitre
2000-12-18 17:05 ` Erwin Authried
2000-12-19 14:52 ` David Woodhouse
0 siblings, 2 replies; 5+ messages in thread
From: Nicolas Pitre @ 2000-12-18 15:50 UTC (permalink / raw)
To: Erwin Authried; +Cc: David Woodhouse, mtd
On Mon, 18 Dec 2000, Erwin Authried wrote:
> there were a couple of small changes necessary to compile on 2.0.
> Here are some other changes that I would like to commit too, please let me
> know if its ok.:
>
> * Use #ifdef NO_MM in mtdchar.c to avoid vmalloc if no MMU is used
> (uClinux).
>
> * "list.h" is missing in the 2.0 kernel source. The best thing that I could
> think of is to simply add it
> to MTD in linux/mtd/list.h.
Well... I would suggest you add it where it belongs i.e.
include/linux/list.h.
Nicolas
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: module registration changes
2000-12-18 15:50 ` Nicolas Pitre
@ 2000-12-18 17:05 ` Erwin Authried
2000-12-19 14:52 ` David Woodhouse
1 sibling, 0 replies; 5+ messages in thread
From: Erwin Authried @ 2000-12-18 17:05 UTC (permalink / raw)
To: Nicolas Pitre; +Cc: mtd
Currently, I have two seperate directory trees for mtd and for the linux
kernel. With a link
to the mtd sources in the drivers directory, and a second link in
includes/linux to the includes/linux/mtd directory, both trees can
maintained with CVS completely seperately. This wouldn't be possible that
easy with files in the include/linux subdir.
-Erwin
----- Original Message -----
From: Nicolas Pitre <nico@cam.org>
To: Erwin Authried <eauth@softsys.co.at>
Cc: David Woodhouse <dwmw2@infradead.org>; <mtd@infradead.org>
Sent: Monday, December 18, 2000 4:50 PM
Subject: Re: module registration changes
>
>
> On Mon, 18 Dec 2000, Erwin Authried wrote:
>
> > there were a couple of small changes necessary to compile on 2.0.
> > Here are some other changes that I would like to commit too, please let
me
> > know if its ok.:
> >
> > * Use #ifdef NO_MM in mtdchar.c to avoid vmalloc if no MMU is used
> > (uClinux).
> >
> > * "list.h" is missing in the 2.0 kernel source. The best thing that I
could
> > think of is to simply add it
> > to MTD in linux/mtd/list.h.
>
> Well... I would suggest you add it where it belongs i.e.
> include/linux/list.h.
>
>
> Nicolas
>
>
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: module registration changes
2000-12-18 15:50 ` Nicolas Pitre
2000-12-18 17:05 ` Erwin Authried
@ 2000-12-19 14:52 ` David Woodhouse
1 sibling, 0 replies; 5+ messages in thread
From: David Woodhouse @ 2000-12-19 14:52 UTC (permalink / raw)
To: Nicolas Pitre; +Cc: Erwin Authried, mtd
nico@cam.org said:
> Well... I would suggest you add it where it belongs i.e. include/
> linux/list.h.
I agree. We need to patch 2.0 anyway to make the initcalls &c work. We
might as well add list.h in the right place in the same patch.
At this stage, I'd like to keep the main code as clean as possible for 2.4.
I want it to _work_ on 2.0, but I want it to be _pretty_ in 2.4, because I
have to justify it to Linus for 2.4, and I don't for 2.0 :)
Feel free to add an 'infrastructure' patch for 2.0 which does everything
you need and lets you just drop the contents of the drivers/mtd and
include/linux/mtd directories in place from CVS.
--
dwmw2
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2000-12-19 14:55 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2000-12-18 0:29 module registration changes David Woodhouse
2000-12-18 13:04 ` Erwin Authried
2000-12-18 15:50 ` Nicolas Pitre
2000-12-18 17:05 ` Erwin Authried
2000-12-19 14:52 ` David Woodhouse
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox