linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* 2.6 eicon/ and hardware/eicon/ drivers using the same symbols
@ 2003-09-25 10:15 Adrian Bunk
  2003-09-25 11:33 ` [PATCH] " Armin Schindler
  0 siblings, 1 reply; 5+ messages in thread
From: Adrian Bunk @ 2003-09-25 10:15 UTC (permalink / raw)
  To: mac; +Cc: isdn4linux, linux-kernel

I got the link error below in 2.6.0-test5-mm4 (but it doesn't seem to be 
specific to -mm).

It seems some drivers under eicon/ and hardware/eicon/ use the same 
symbols. Either some symbols should be renamed or Kconfig dependencies 
should ensure that you can't build two such drivers statically into the 
kernel at the same time.

cu
Adrian

<--  snip  -->

  LD      drivers/isdn/built-in.o
drivers/isdn/sc/built-in.o(.bss+0x0): multiple definition of `adapter'
drivers/isdn/hardware/built-in.o(.data+0x1134): first defined here
ld: Warning: size of symbol `adapter' changed from 4 in 
drivers/isdn/hardware/built-in.o to 16 in drivers/isdn/sc/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x97ac): In function `mem_inw':
: multiple definition of `mem_inw'
drivers/isdn/hardware/built-in.o(.text+0x10240): first defined here
ld: Warning: size of symbol `mem_inw' changed from 26 in 
drivers/isdn/hardware/built-in.o to 66 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9944): In function 
`mem_out_buffer':
: multiple definition of `mem_out_buffer'
drivers/isdn/hardware/built-in.o(.text+0x10378): first defined here
ld: Warning: size of symbol `mem_out_buffer' changed from 48 in 
drivers/isdn/hardware/built-in.o to 74 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9690): In function 
`DIVA_DIDD_Read':
: multiple definition of `DIVA_DIDD_Read'
drivers/isdn/hardware/built-in.o(.text+0xd454): first defined here
ld: Warning: size of symbol `DIVA_DIDD_Read' changed from 17 in 
drivers/isdn/hardware/built-in.o to 93 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9a78): In function `io_in_buffer':
: multiple definition of `io_in_buffer'
drivers/isdn/hardware/built-in.o(.text+0x1041c): first defined here
ld: Warning: size of symbol `io_in_buffer' changed from 105 in 
drivers/isdn/hardware/built-in.o to 73 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9a34): In function `io_inw':
: multiple definition of `io_inw'
drivers/isdn/hardware/built-in.o(.text+0x103f0): first defined here
ld: Warning: size of symbol `io_inw' changed from 44 in 
drivers/isdn/hardware/built-in.o to 65 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9768): In function `mem_in':
: multiple definition of `mem_in'
drivers/isdn/hardware/built-in.o(.text+0x10224): first defined here
ld: Warning: size of symbol `mem_in' changed from 25 in 
drivers/isdn/hardware/built-in.o to 66 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9c0c): In function `io_inc':
: multiple definition of `io_inc'
drivers/isdn/hardware/built-in.o(.text+0x105b8): first defined here
ld: Warning: size of symbol `io_inc' changed from 87 in 
drivers/isdn/hardware/built-in.o to 83 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9b7c): In function `io_outw':
: multiple definition of `io_outw'
drivers/isdn/hardware/built-in.o(.text+0x1051c): first defined here
ld: Warning: size of symbol `io_outw' changed from 47 in 
drivers/isdn/hardware/built-in.o to 68 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9990): In function `mem_inc':
: multiple definition of `mem_inc'
drivers/isdn/hardware/built-in.o(.text+0x103a8): first defined here
ld: Warning: size of symbol `mem_inc' changed from 25 in 
drivers/isdn/hardware/built-in.o to 80 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9bc0): In function 
`io_out_buffer':
: multiple definition of `io_out_buffer'
drivers/isdn/hardware/built-in.o(.text+0x1054c): first defined here
ld: Warning: size of symbol `io_out_buffer' changed from 105 in 
drivers/isdn/hardware/built-in.o to 73 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x983c): In function 
`mem_look_ahead':
: multiple definition of `mem_look_ahead'
drivers/isdn/hardware/built-in.o(.text+0x102c4): first defined here
ld: Warning: size of symbol `mem_look_ahead' changed from 74 in 
drivers/isdn/hardware/built-in.o to 118 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x99e0): In function `io_in':
: multiple definition of `io_in'
drivers/isdn/hardware/built-in.o(.text+0x103c4): first defined here
ld: Warning: size of symbol `io_in' changed from 44 in 
drivers/isdn/hardware/built-in.o to 81 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x98fc): In function `mem_outw':
: multiple definition of `mem_outw'
dridn/hardware/built-in.o(.text+0x10328): first defined here
ld: Warning: size of symbol `mem_outw' changed from 25 in 
drivers/isdn/hardware/built-in.o to 69 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x98b4): In function `mem_out':
: multiple definition of `mem_out'
drivers/isdn/hardware/built-in.o(.text+0x10310): first defined here
ld: Warning: size of symbol `mem_out' changed from 24 in 
drivers/isdn/hardware/built-in.o to 69 in drivers/isdn/eicon/built-in.o
drivers/isdn/eicon/built-in.o(.text+0x9b38): In function `io_out':
: multiple definition of `io_out'
drivers/isdn/hardware/built-in.o(.text+0x104ec): first defined here
ld: Warning: size of symbol `io_out' changed from 47 in 
drivers/isdn/hardware/built-in.o to 68 in drivers/isdn/eicon/built-in.o
...
make[2]: *** [drivers/isdn/built-in.o] Error 1

<--  snip  -->

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH] Re: 2.6 eicon/ and hardware/eicon/ drivers using the same symbols
  2003-09-25 10:15 2.6 eicon/ and hardware/eicon/ drivers using the same symbols Adrian Bunk
@ 2003-09-25 11:33 ` Armin Schindler
  2003-09-25 11:38   ` Christoph Hellwig
  2003-09-26 22:50   ` [PATCH] let ISDN_DRV_SC depend on m Adrian Bunk
  0 siblings, 2 replies; 5+ messages in thread
From: Armin Schindler @ 2003-09-25 11:33 UTC (permalink / raw)
  To: Adrian Bunk; +Cc: isdn4linux, Linux Kernel Mailinglist

On Thu, 25 Sep 2003, Adrian Bunk wrote:
> I got the link error below in 2.6.0-test5-mm4 (but it doesn't seem to be
> specific to -mm).
>
> It seems some drivers under eicon/ and hardware/eicon/ use the same
> symbols. Either some symbols should be renamed or Kconfig dependencies
> should ensure that you can't build two such drivers statically into the
> kernel at the same time.

The legacy eicon driver in drivers/isdn/eicon is the old one and will be
removed as soon as all features went to the new driver.
Anyway this old driver was never meant to be non-module.

This patch should do it.

Armin



--- linux-2.5/drivers/isdn/eicon/Kconfig.orig	Thu Sep 25 13:28:07 2003
+++ linux-2.5/drivers/isdn/eicon/Kconfig	Thu Sep 25 13:27:01 2003
@@ -13,7 +13,7 @@
 choice
 	prompt "Eicon active card support"
 	optional
-	depends on ISDN_DRV_EICON && ISDN
+	depends on ISDN_DRV_EICON && ISDN && m

 config ISDN_DRV_EICON_DIVAS
 	tristate "Eicon driver"


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Re: 2.6 eicon/ and hardware/eicon/ drivers using the same symbols
  2003-09-25 11:33 ` [PATCH] " Armin Schindler
@ 2003-09-25 11:38   ` Christoph Hellwig
  2003-09-25 11:52     ` Armin Schindler
  2003-09-26 22:50   ` [PATCH] let ISDN_DRV_SC depend on m Adrian Bunk
  1 sibling, 1 reply; 5+ messages in thread
From: Christoph Hellwig @ 2003-09-25 11:38 UTC (permalink / raw)
  To: Armin Schindler; +Cc: Adrian Bunk, isdn4linux, Linux Kernel Mailinglist

On Thu, Sep 25, 2003 at 01:33:53PM +0200, Armin Schindler wrote:
> The legacy eicon driver in drivers/isdn/eicon is the old one and will be
> removed as soon as all features went to the new driver.
> Anyway this old driver was never meant to be non-module.

What about just killing it off?  If users really want the old one
on 2.6 you can put up a tarball for them somewhere.

The driver is so horrinly ugly that it better goes away sooner than
later..


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Re: 2.6 eicon/ and hardware/eicon/ drivers using the same symbols
  2003-09-25 11:38   ` Christoph Hellwig
@ 2003-09-25 11:52     ` Armin Schindler
  0 siblings, 0 replies; 5+ messages in thread
From: Armin Schindler @ 2003-09-25 11:52 UTC (permalink / raw)
  To: Christoph Hellwig; +Cc: Linux Kernel Mailinglist

On Thu, 25 Sep 2003, Christoph Hellwig wrote:
> On Thu, Sep 25, 2003 at 01:33:53PM +0200, Armin Schindler wrote:
> > The legacy eicon driver in drivers/isdn/eicon is the old one and will be
> > removed as soon as all features went to the new driver.
> > Anyway this old driver was never meant to be non-module.
>
> What about just killing it off?  If users really want the old one
> on 2.6 you can put up a tarball for them somewhere.
>
> The driver is so horrinly ugly that it better goes away sooner than
> later..

Oh thank you! Okay, it's old, too old.

It's not just "the old one", this driver still supports more than 4 ISDN
cards, which are not yet supported by the other driver.

But maybe you are right...

Armin



^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH] let ISDN_DRV_SC depend on m
  2003-09-25 11:33 ` [PATCH] " Armin Schindler
  2003-09-25 11:38   ` Christoph Hellwig
@ 2003-09-26 22:50   ` Adrian Bunk
  1 sibling, 0 replies; 5+ messages in thread
From: Adrian Bunk @ 2003-09-26 22:50 UTC (permalink / raw)
  To: Armin Schindler; +Cc: isdn4linux, Linux Kernel Mailinglist, gpl-info

On Thu, Sep 25, 2003 at 01:33:53PM +0200, Armin Schindler wrote:
> On Thu, 25 Sep 2003, Adrian Bunk wrote:
> > I got the link error below in 2.6.0-test5-mm4 (but it doesn't seem to be
> > specific to -mm).
> >
> > It seems some drivers under eicon/ and hardware/eicon/ use the same
> > symbols. Either some symbols should be renamed or Kconfig dependencies
> > should ensure that you can't build two such drivers statically into the
> > kernel at the same time.
> 
> The legacy eicon driver in drivers/isdn/eicon is the old one and will be
> removed as soon as all features went to the new driver.
> Anyway this old driver was never meant to be non-module.
> 
> This patch should do it.

Yes, thanks, this works. The similar patch for ISDN_DRV_SC below is 
needed, too.

> Armin

cu
Adrian

--- linux-2.6.0-test5-mm4-no-smp-2.95/drivers/isdn/sc/Kconfig.old	2003-09-25 19:21:07.000000000 +0200
+++ linux-2.6.0-test5-mm4-no-smp-2.95/drivers/isdn/sc/Kconfig	2003-09-25 19:21:54.000000000 +0200
@@ -3,7 +3,7 @@
 #
 config ISDN_DRV_SC
 	tristate "Spellcaster support"
-	depends on ISDN && ISA
+	depends on ISDN && ISA && m
 	help
 	  This enables support for the Spellcaster BRI ISDN boards.  This
 	  driver currently builds only in a modularized version ( = code which

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2003-09-26 22:51 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-09-25 10:15 2.6 eicon/ and hardware/eicon/ drivers using the same symbols Adrian Bunk
2003-09-25 11:33 ` [PATCH] " Armin Schindler
2003-09-25 11:38   ` Christoph Hellwig
2003-09-25 11:52     ` Armin Schindler
2003-09-26 22:50   ` [PATCH] let ISDN_DRV_SC depend on m Adrian Bunk

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).