Re: [BUG] [2.6.19-rc4-mm2] can't compile drivers/acpi/processor_idle.c

Re: [BUG] [2.6.19-rc4-mm2] can't compile drivers/acpi/processor_idle.c

[Andrew Morton]

On Wed, 8 Nov 2006 15:01:41 +0900
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:

> 
> While compiling 2.6.19-rc4-mm2 on ia64, I met this compile error.
> ==
>   CC [M]  drivers/acpi/processor_idle.o
> drivers/acpi/processor_idle.c:43:22: asm/apic.h: No such file or directory
> drivers/acpi/processor_idle.c: In function `acpi_processor_power_seq_show':
> drivers/acpi/processor_idle.c:1202: warning: long long unsigned int format, u64 arg (arg 5)
> ==
> 
> This is because of acpi-include-apic-h.patch, maybe.
> ia64 doesn't have asm/acpi.h

That got fixed (by ugly means).

> my .config is attached.

But rc5-mm1 remains broken with that .config:

arch/ia64/pci/pci.c: In function `pci_acpi_scan_root':
arch/ia64/pci/pci.c:354: warning: implicit declaration of function `pxm_to_node'
...
arch/ia64/pci/built-in.o(.text+0xe92): In function `pci_acpi_scan_root':
: undefined reference to `pxm_to_node'

This bug exists in mainline.




Also,

drivers/built-in.o(.text+0xd9a72): In function `e1000_xmit_frame':
: undefined reference to `csum_ipv6_magic'

I don't know how this got broken.  ia64 seems to be the only architecture
which doesn't have an implementation of csum_ipv6_magic().  This bug
appears to be introduced by git-netdev-all.patch.

Re: [BUG] [2.6.19-rc4-mm2] can't compile drivers/acpi/processor_idle.c

[David Miller]

From: Andrew Morton <akpm@osdl.org>
Date: Tue, 7 Nov 2006 22:52:59 -0800

> Also,
> 
> drivers/built-in.o(.text+0xd9a72): In function `e1000_xmit_frame':
> : undefined reference to `csum_ipv6_magic'
> 
> I don't know how this got broken.  ia64 seems to be the only architecture
> which doesn't have an implementation of csum_ipv6_magic().  This bug
> appears to be introduced by git-netdev-all.patch.

There is a generic version, which e1000 would get if it included
the net/ip_checksum.h header file.

Re: [BUG] [2.6.19-rc4-mm2] can't compile drivers/acpi/processor_idle.c

[KAMEZAWA Hiroyuki]

On Tue, 7 Nov 2006 22:52:59 -0800
Andrew Morton <akpm@osdl.org> wrote:

> On Wed, 8 Nov 2006 15:01:41 +0900
> KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:
> 
> > 
> > While compiling 2.6.19-rc4-mm2 on ia64, I met this compile error.
> > ==
> >   CC [M]  drivers/acpi/processor_idle.o
> > drivers/acpi/processor_idle.c:43:22: asm/apic.h: No such file or directory
> > drivers/acpi/processor_idle.c: In function `acpi_processor_power_seq_show':
> > drivers/acpi/processor_idle.c:1202: warning: long long unsigned int format, u64 arg (arg 5)
> > ==
> > 
> > This is because of acpi-include-apic-h.patch, maybe.
> > ia64 doesn't have asm/acpi.h
> 
> That got fixed (by ugly means).

Ah, okay. I'll move to rc5-mm1. Thank you.

> 
> > my .config is attached.
> 
> But rc5-mm1 remains broken with that .config:
> 
> arch/ia64/pci/pci.c: In function `pci_acpi_scan_root':
> arch/ia64/pci/pci.c:354: warning: implicit declaration of function `pxm_to_node'
> ...
> arch/ia64/pci/built-in.o(.text+0xe92): In function `pci_acpi_scan_root':
> : undefined reference to `pxm_to_node'
> 
> This bug exists in mainline.
> 

How about this ? Maybe ia64 people's review is necessary.

-Kame
==
When ACPI && NUMA, pxm_to_node is used and it exists in drivers/acpi/numa.c

Signed-Off-By: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujtisu.com>

Index: linux-2.6.19-rc4-mm2/arch/ia64/Kconfig
===================================================================
--- linux-2.6.19-rc4-mm2.orig/arch/ia64/Kconfig	2006-11-08 14:15:21.000000000 +0900
+++ linux-2.6.19-rc4-mm2/arch/ia64/Kconfig	2006-11-08 16:16:40.000000000 +0900
@@ -353,6 +353,7 @@
 	bool "NUMA support"
 	depends on !IA64_HP_SIM && !FLATMEM
 	default y if IA64_SGI_SN2
+	select ACPI_NUMA if ACPI
 	help
 	  Say Y to compile the kernel to support NUMA (Non-Uniform Memory
 	  Access).  This option is for configuring high-end multiprocessor