[PATCH] early_pfn_to_nid needs to be __meminit

[PATCH] early_pfn_to_nid needs to be __meminit

[Stephen Rothwell]

since it is referenced by memmap_init_zone (which is __meminit) via the
early_pfn_in_nid macro when CONFIG_NODES_SPAN_OTHER_NODES is set (which
basically means PowerPC 64).

This removes a section mismatch warning in those circumstances.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 mm/page_alloc.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 6fd0b74..3297fa9 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2298,7 +2298,7 @@ static int __meminit next_active_region_index_in_nid(int index, int nid)
  * was used and there are no special requirements, this is a convenient
  * alternative
  */
-int __init early_pfn_to_nid(unsigned long pfn)
+int __meminit early_pfn_to_nid(unsigned long pfn)
 {
 	int i;
 
-- 
1.5.1.3

Section mismatch warnings (was Re: [PATCH] early_pfn_to_nid needs to be __meminit)

[Gabriel Paubert]

On Thu, May 10, 2007 at 02:25:52AM +1000, Stephen Rothwell wrote:
> since it is referenced by memmap_init_zone (which is __meminit) via the
> early_pfn_in_nid macro when CONFIG_NODES_SPAN_OTHER_NODES is set (which
> basically means PowerPC 64).
> 
> This removes a section mismatch warning in those circumstances.

Speaking of this, I just tried to compile an official (Linus' git tree) 
kernel for my old PMac G4 and I get a lot of section mismatch warnings at 
the end of the compilation:

  LD      vmlinux
  SYSMAP  System.map
  SYSMAP  .tmp_System.map
  MODPOST vmlinux
WARNING: "fee_restarts" [arch/powerpc/kernel/built-in] is COMMON symbol
WARNING: "ee_restarts" [arch/powerpc/kernel/built-in] is COMMON symbol
WARNING: arch/powerpc/kernel/built-in.o - Section mismatch: reference to .init.data:.got2 from dt_string_start (offset 0x8)
WARNING: arch/powerpc/kernel/built-in.o - Section mismatch: reference to .init.data:.got2 from dt_string_end (offset 0xc)
...
WARNING: arch/powerpc/kernel/built-in.o - Section mismatch: reference to .init.data:.got2 from dt_struct_end (offset 0x20c)
WARNING: arch/powerpc/kernel/built-in.o - Section mismatch: reference to .init.data:.got2 from disp_BAT (offset 0x234)
WARNING: "primary_pteg_full" [arch/powerpc/mm/built-in] is COMMON symbol
WARNING: "next_slot" [arch/powerpc/mm/built-in] is COMMON symbol
WARNING: "htab_hash_searches" [arch/powerpc/mm/built-in] is COMMON symbol
WARNING: arch/powerpc/mm/built-in.o - Section mismatch: reference to .init.text:early_get_page from .text between 'pte_alloc_one_kernel' (at offset 0xf50) and 'v_mapped_by_bats'
WARNING: arch/powerpc/platforms/built-in.o - Section mismatch: reference to .init.data:.got2 from  (offset 0x0)
...
WARNING: arch/powerpc/platforms/built-in.o - Section mismatch: reference to .init.text:pmac_pcibios_after_init from .machine.desc after 'mach_powermac' (at offset 0xc4)
WARNING: mm/built-in.o - Section mismatch: reference to .init.text:set_up_list3s from .text between 'kmem_cache_create' (at
offset 0x1de88) and 'kmem_cache_shrink'
WARNING: mm/built-in.o - Section mismatch: reference to .init.text:set_up_list3s from .text between 'kmem_cache_create' (at
offset 0x1dec0) and 'kmem_cache_shrink'

I find these 50 or so warnings so scary that I've not yet tried 
to boot the kernel. Note that this is a non-modular kernel.

	Gabriel

Re: Section mismatch warnings (was Re: [PATCH] early_pfn_to_nid needs to be __meminit)

[Linas Vepstas]

On Wed, May 09, 2007 at 06:51:15PM +0200, Gabriel Paubert wrote:
> On Thu, May 10, 2007 at 02:25:52AM +1000, Stephen Rothwell wrote:
> > This removes a section mismatch warning in those circumstances.
> 
> Speaking of this, I just tried to compile an official (Linus' git tree) 
> kernel for my old PMac G4 and I get a lot of section mismatch warnings at 
> the end of the compilation:
> 
> WARNING: arch/powerpc/mm/built-in.o - Section mismatch: reference to .init.text:early_get_page from .text between 'pte_alloc_one_kernel' (at offset 0xf50) and 'v_mapped_by_bats'
> 
> I find these 50 or so warnings so scary that I've not yet tried 
> to boot the kernel. Note that this is a non-modular kernel.

I'm getting oodles of these on an older -mm2 tree. The kernels seem to
work fine. Yes, they should be fixed but I'm up to my proverbial eyballs
in alligators.

--linas

Re: Section mismatch warnings (was Re: [PATCH] early_pfn_to_nid needs to be __meminit)

[Josh Boyer]

On Wed, 2007-05-09 at 14:45 -0500, Linas Vepstas wrote:
> On Wed, May 09, 2007 at 06:51:15PM +0200, Gabriel Paubert wrote:
> > On Thu, May 10, 2007 at 02:25:52AM +1000, Stephen Rothwell wrote:
> > > This removes a section mismatch warning in those circumstances.
> > 
> > Speaking of this, I just tried to compile an official (Linus' git tree) 
> > kernel for my old PMac G4 and I get a lot of section mismatch warnings at 
> > the end of the compilation:
> > 
> > WARNING: arch/powerpc/mm/built-in.o - Section mismatch: reference to .init.text:early_get_page from .text between 'pte_alloc_one_kernel' (at offset 0xf50) and 'v_mapped_by_bats'
> > 
> > I find these 50 or so warnings so scary that I've not yet tried 
> > to boot the kernel. Note that this is a non-modular kernel.
> 
> I'm getting oodles of these on an older -mm2 tree. The kernels seem to
> work fine. Yes, they should be fixed but I'm up to my proverbial eyballs
> in alligators.

Yeah, I see all kinds of this too.  Does anyone have an idea why they're
being spit out?  Or if it's related to a particular binutils version,
etc?

josh

Re: Section mismatch warnings (was Re: [PATCH] early_pfn_to_nid needs to be __meminit)

[Yasunori Goto]

Add to CC: Sam-san. :-)

> On Wed, 2007-05-09 at 14:45 -0500, Linas Vepstas wrote:
> > On Wed, May 09, 2007 at 06:51:15PM +0200, Gabriel Paubert wrote:
> > > On Thu, May 10, 2007 at 02:25:52AM +1000, Stephen Rothwell wrote:
> > > > This removes a section mismatch warning in those circumstances.
> > > 
> > > Speaking of this, I just tried to compile an official (Linus' git tree) 
> > > kernel for my old PMac G4 and I get a lot of section mismatch warnings at 
> > > the end of the compilation:
> > > 
> > > WARNING: arch/powerpc/mm/built-in.o - Section mismatch: reference to .init.text:early_get_page from .text between 'pte_alloc_one_kernel' (at offset 0xf50) and 'v_mapped_by_bats'
> > > 
> > > I find these 50 or so warnings so scary that I've not yet tried 
> > > to boot the kernel. Note that this is a non-modular kernel.
> > 
> > I'm getting oodles of these on an older -mm2 tree. The kernels seem to
> > work fine. Yes, they should be fixed but I'm up to my proverbial eyballs
> > in alligators.
> 
> Yeah, I see all kinds of this too.  Does anyone have an idea why they're
> being spit out?  Or if it's related to a particular binutils version,
> etc?

These warnings are showed by scripts/mod/modposts.c. 
It checks section mismatch stricter than before.

Some of them are not defined as __init even if it's called at just
boottime. This is simple to fix.

Second case is that the functions judge which functions
should call (boottime or normal) by themselves.
pte_alloc_one_kernel() looks its case.
In this case, they must be registered in whitelist of modpost.c

Other possibility is they are just rare called functions.
It should be investigated....

Bye.

-- 
Yasunori Goto