[2.6 patch] fix arch/i386/pci/Makefile

[2.6 patch] fix arch/i386/pci/Makefile

[Adrian Bunk]

I got the following compile error in 2.6.7-mm2 (but it doesn't seem to 
be specific to -mm2):

<--  snip  -->

...
  LD      .tmp_vmlinux1
drivers/built-in.o(.text+0x6c24a): In function `acpi_pci_root_add':
: undefined reference to `pci_acpi_scan_root'
make: *** [.tmp_vmlinux1] Error 1

<--  snip  -->

This problem occurs with
  CONFIG_ACPI_PCI=y && (CONFIG_X86_VISWS=y || CONFIG_X86_NUMAQ=y)

The patch below fixes it.

Please apply
Adrian


--- linux-2.6.7-mm2-full/arch/i386/pci/Makefile.old	2004-06-25 02:08:29.000000000 +0200
+++ linux-2.6.7-mm2-full/arch/i386/pci/Makefile	2004-06-25 02:10:36.000000000 +0200
@@ -5,10 +5,11 @@
 obj-$(CONFIG_PCI_DIRECT)	+= direct.o
 
 pci-y				:= fixup.o
-pci-$(CONFIG_ACPI_PCI)		+= acpi.o
 pci-y				+= legacy.o irq.o
 
 pci-$(CONFIG_X86_VISWS)		:= visws.o fixup.o
 pci-$(CONFIG_X86_NUMAQ)		:= numa.o irq.o
 
+pci-$(CONFIG_ACPI_PCI)		+= acpi.o
+
 obj-y				+= $(pci-y) common.o

Re: [2.6 patch] fix arch/i386/pci/Makefile

[Andrew Morton]

Adrian Bunk <bunk@fs.tum.de> wrote:
>
> I got the following compile error in 2.6.7-mm2 (but it doesn't seem to 
>  be specific to -mm2):
> ..
>  drivers/built-in.o(.text+0x6c24a): In function `acpi_pci_root_add':
>  : undefined reference to `pci_acpi_scan_root'
>  make: *** [.tmp_vmlinux1] Error 1

> 
>  This problem occurs with
>    CONFIG_ACPI_PCI=y && (CONFIG_X86_VISWS=y || CONFIG_X86_NUMAQ=y)
> 
> ....
>  --- linux-2.6.7-mm2-full/arch/i386/pci/Makefile.old	2004-06-25 02:08:29.000000000 +0200
>  +++ linux-2.6.7-mm2-full/arch/i386/pci/Makefile	2004-06-25 02:10:36.000000000 +0200
>  @@ -5,10 +5,11 @@
>   obj-$(CONFIG_PCI_DIRECT)	+= direct.o
>   
>   pci-y				:= fixup.o
>  -pci-$(CONFIG_ACPI_PCI)		+= acpi.o
>   pci-y				+= legacy.o irq.o
>   
>   pci-$(CONFIG_X86_VISWS)		:= visws.o fixup.o
>   pci-$(CONFIG_X86_NUMAQ)		:= numa.o irq.o
>   
>  +pci-$(CONFIG_ACPI_PCI)		+= acpi.o
>  +

This causes my e100 NIC to not work.  Some initcall ordering dependency,
presumably.  A whole bunch of devices popped up on different IRQs.

Come to think about it, how can the above patch fix that linkage error
anyway?

Re: [2.6 patch] fix arch/i386/pci/Makefile

[Matthew Wilcox]

On Fri, Jun 25, 2004 at 02:15:14AM +0200, Adrian Bunk wrote:
> This problem occurs with
>   CONFIG_ACPI_PCI=y && (CONFIG_X86_VISWS=y || CONFIG_X86_NUMAQ=y)

That combination certainly doesn't make sense.  No VisWS or NUMAQ systems
have ACPI.  How did you manage to turn this on given:

menu "ACPI (Advanced Configuration and Power Interface) Support"
        depends on !X86_VISWS
        depends on !IA64_HP_SIM
        depends on IA64 || X86

If you apply the following patch, does it help?

--- drivers/acpi/Kconfig        30 Mar 2004 12:42:25 -0000      1.11
+++ drivers/acpi/Kconfig        25 Jun 2004 18:00:29 -0000
@@ -3,7 +3,7 @@
 #
 
 menu "ACPI (Advanced Configuration and Power Interface) Support"
-       depends on !X86_VISWS
+       depends on !X86_VISWS && !X86_NUMAQ
        depends on !IA64_HP_SIM
        depends on IA64 || X86
 

-- 
"Next the statesmen will invent cheap lies, putting the blame upon 
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince 
himself that the war is just, and will thank God for the better sleep 
he enjoys after this process of grotesque self-deception." -- Mark Twain

Re: [2.6 patch] fix arch/i386/pci/Makefile

[Adrian Bunk]

On Fri, Jun 25, 2004 at 07:01:15PM +0100, Matthew Wilcox wrote:
> On Fri, Jun 25, 2004 at 02:15:14AM +0200, Adrian Bunk wrote:
> > This problem occurs with
> >   CONFIG_ACPI_PCI=y && (CONFIG_X86_VISWS=y || CONFIG_X86_NUMAQ=y)
> 
> That combination certainly doesn't make sense.  No VisWS or NUMAQ systems
> have ACPI.  How did you manage to turn this on given:
> 
> menu "ACPI (Advanced Configuration and Power Interface) Support"
>         depends on !X86_VISWS
>         depends on !IA64_HP_SIM
>         depends on IA64 || X86
> 
> If you apply the following patch, does it help?
> 
> --- drivers/acpi/Kconfig        30 Mar 2004 12:42:25 -0000      1.11
> +++ drivers/acpi/Kconfig        25 Jun 2004 18:00:29 -0000
> @@ -3,7 +3,7 @@
>  #
>  
>  menu "ACPI (Advanced Configuration and Power Interface) Support"
> -       depends on !X86_VISWS
> +       depends on !X86_VISWS && !X86_NUMAQ
>         depends on !IA64_HP_SIM
>         depends on IA64 || X86

Yes thanks, this fixes it.

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

Re: [2.6 patch] fix arch/i386/pci/Makefile

[Adrian Bunk]

On Thu, Jun 24, 2004 at 09:01:50PM -0700, Andrew Morton wrote:
> Adrian Bunk <bunk@fs.tum.de> wrote:
> >
> > I got the following compile error in 2.6.7-mm2 (but it doesn't seem to 
> >  be specific to -mm2):
> > ..
> >  drivers/built-in.o(.text+0x6c24a): In function `acpi_pci_root_add':
> >  : undefined reference to `pci_acpi_scan_root'
> >  make: *** [.tmp_vmlinux1] Error 1
> 
> > 
> >  This problem occurs with
> >    CONFIG_ACPI_PCI=y && (CONFIG_X86_VISWS=y || CONFIG_X86_NUMAQ=y)
> > 
> > ....
> >  --- linux-2.6.7-mm2-full/arch/i386/pci/Makefile.old	2004-06-25 02:08:29.000000000 +0200
> >  +++ linux-2.6.7-mm2-full/arch/i386/pci/Makefile	2004-06-25 02:10:36.000000000 +0200
> >  @@ -5,10 +5,11 @@
> >   obj-$(CONFIG_PCI_DIRECT)	+= direct.o
> >   
> >   pci-y				:= fixup.o
> >  -pci-$(CONFIG_ACPI_PCI)		+= acpi.o
> >   pci-y				+= legacy.o irq.o
> >   
> >   pci-$(CONFIG_X86_VISWS)		:= visws.o fixup.o
> >   pci-$(CONFIG_X86_NUMAQ)		:= numa.o irq.o
> >   
> >  +pci-$(CONFIG_ACPI_PCI)		+= acpi.o
> >  +
> 
> This causes my e100 NIC to not work.  Some initcall ordering dependency,
> presumably.  A whole bunch of devices popped up on different IRQs.
> 
> Come to think about it, how can the above patch fix that linkage error
> anyway?

A := overrides all previous := and += .


But Matthew's patch seems to be a better solution.


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