* [PATCH] mm: fix setup_zone_pageset section mismatch
@ 2011-03-24 20:24 Randy Dunlap
2011-03-30 22:05 ` Andrew Morton
0 siblings, 1 reply; 5+ messages in thread
From: Randy Dunlap @ 2011-03-24 20:24 UTC (permalink / raw)
To: linux-mm; +Cc: akpm
From: Randy Dunlap <randy.dunlap@oracle.com>
Fix section mismatch warning:
setup_zone_pageset() is called from build_all_zonelists(),
which can be called at any time by NUMA sysctl handler
numa_zonelist_order_handler(),
so it should not be marked as __meminit.
WARNING: mm/built-in.o(.text+0xab17): Section mismatch in reference from the function build_all_zonelists() to the function .meminit.text:setup_zone_pageset()
The function build_all_zonelists() references
the function __meminit setup_zone_pageset().
This is often because build_all_zonelists lacks a __meminit
annotation or the annotation of setup_zone_pageset is wrong.
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
---
mm/page_alloc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- linux-2.6.38-git13.orig/mm/page_alloc.c
+++ linux-2.6.38-git13/mm/page_alloc.c
@@ -3511,7 +3511,7 @@ static void setup_pagelist_highmark(stru
pcp->batch = PAGE_SHIFT * 8;
}
-static __meminit void setup_zone_pageset(struct zone *zone)
+static void setup_zone_pageset(struct zone *zone)
{
int cpu;
---
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH] mm: fix setup_zone_pageset section mismatch
2011-03-24 20:24 [PATCH] mm: fix setup_zone_pageset section mismatch Randy Dunlap
@ 2011-03-30 22:05 ` Andrew Morton
2011-03-30 22:47 ` Randy Dunlap
0 siblings, 1 reply; 5+ messages in thread
From: Andrew Morton @ 2011-03-30 22:05 UTC (permalink / raw)
To: Randy Dunlap; +Cc: linux-mm, Paul Mundt
On Thu, 24 Mar 2011 13:24:35 -0700
Randy Dunlap <randy.dunlap@oracle.com> wrote:
> From: Randy Dunlap <randy.dunlap@oracle.com>
>
> Fix section mismatch warning:
> setup_zone_pageset() is called from build_all_zonelists(),
> which can be called at any time by NUMA sysctl handler
> numa_zonelist_order_handler(),
> so it should not be marked as __meminit.
>
> WARNING: mm/built-in.o(.text+0xab17): Section mismatch in reference from the function build_all_zonelists() to the function .meminit.text:setup_zone_pageset()
> The function build_all_zonelists() references
> the function __meminit setup_zone_pageset().
> This is often because build_all_zonelists lacks a __meminit
> annotation or the annotation of setup_zone_pageset is wrong.
>
> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
> ---
> mm/page_alloc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> --- linux-2.6.38-git13.orig/mm/page_alloc.c
> +++ linux-2.6.38-git13/mm/page_alloc.c
> @@ -3511,7 +3511,7 @@ static void setup_pagelist_highmark(stru
> pcp->batch = PAGE_SHIFT * 8;
> }
>
> -static __meminit void setup_zone_pageset(struct zone *zone)
> +static void setup_zone_pageset(struct zone *zone)
> {
> int cpu;
>
I already merged Paul Mundt's patch whcih marks build_all_zonelists()
as __ref. That seems a better solution?
I'm rather wondering if we did all this the right way anyway. The call
from build_all_zonelists() into setup_zone_pageset() is inside #ifdef
CONFIG_MEMORY_HOTPLUG, so there is clearly no bug here. But the build
system generated a warning anyway. Why'd it do that?
If we'd handled the section via
#ifdef CONFIG_MEMORY_HOTPLUG
#define __meminit
#else
#define __meminit __init
#endif
of similar then that would fix things. iirc we used to do it that
way...
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH] mm: fix setup_zone_pageset section mismatch
2011-03-30 22:05 ` Andrew Morton
@ 2011-03-30 22:47 ` Randy Dunlap
2011-03-30 22:53 ` Andrew Morton
0 siblings, 1 reply; 5+ messages in thread
From: Randy Dunlap @ 2011-03-30 22:47 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-mm, Paul Mundt
On 03/30/11 15:05, Andrew Morton wrote:
> On Thu, 24 Mar 2011 13:24:35 -0700
> Randy Dunlap <randy.dunlap@oracle.com> wrote:
>
>> From: Randy Dunlap <randy.dunlap@oracle.com>
>>
>> Fix section mismatch warning:
>> setup_zone_pageset() is called from build_all_zonelists(),
>> which can be called at any time by NUMA sysctl handler
>> numa_zonelist_order_handler(),
>> so it should not be marked as __meminit.
>>
>> WARNING: mm/built-in.o(.text+0xab17): Section mismatch in reference from the function build_all_zonelists() to the function .meminit.text:setup_zone_pageset()
>> The function build_all_zonelists() references
>> the function __meminit setup_zone_pageset().
>> This is often because build_all_zonelists lacks a __meminit
>> annotation or the annotation of setup_zone_pageset is wrong.
>>
>> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
>> ---
>> mm/page_alloc.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> --- linux-2.6.38-git13.orig/mm/page_alloc.c
>> +++ linux-2.6.38-git13/mm/page_alloc.c
>> @@ -3511,7 +3511,7 @@ static void setup_pagelist_highmark(stru
>> pcp->batch = PAGE_SHIFT * 8;
>> }
>>
>> -static __meminit void setup_zone_pageset(struct zone *zone)
>> +static void setup_zone_pageset(struct zone *zone)
>> {
>> int cpu;
>>
>
> I already merged Paul Mundt's patch whcih marks build_all_zonelists()
> as __ref. That seems a better solution?
Merged where? mmotm?
2.6.39-rc1 still has this section mismatch warning.
If Paul's patch fixes the warning, I'm OK with it.
> I'm rather wondering if we did all this the right way anyway. The call
> from build_all_zonelists() into setup_zone_pageset() is inside #ifdef
> CONFIG_MEMORY_HOTPLUG, so there is clearly no bug here. But the build
> system generated a warning anyway. Why'd it do that?
It's a mystery.
> If we'd handled the section via
>
> #ifdef CONFIG_MEMORY_HOTPLUG
> #define __meminit
> #else
> #define __meminit __init
> #endif
>
> of similar then that would fix things. iirc we used to do it that
> way...
Yes, that older way made more sense to me.
--
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH] mm: fix setup_zone_pageset section mismatch
2011-03-30 22:47 ` Randy Dunlap
@ 2011-03-30 22:53 ` Andrew Morton
2011-03-31 0:28 ` Randy Dunlap
0 siblings, 1 reply; 5+ messages in thread
From: Andrew Morton @ 2011-03-30 22:53 UTC (permalink / raw)
To: Randy Dunlap; +Cc: linux-mm, Paul Mundt
On Wed, 30 Mar 2011 15:47:30 -0700
Randy Dunlap <randy.dunlap@oracle.com> wrote:
> ...
>
> >
> > I already merged Paul Mundt's patch whcih marks build_all_zonelists()
> > as __ref. That seems a better solution?
>
> Merged where? mmotm?
mm.
--- a/mm/page_alloc.c~mm-page_allocc-silence-build_all_zonelists-section-mismatch
+++ a/mm/page_alloc.c
@@ -3170,7 +3170,7 @@ static __init_refok int __build_all_zone
* Called with zonelists_mutex held always
* unless system_state == SYSTEM_BOOTING.
*/
-void build_all_zonelists(void *data)
+void __ref build_all_zonelists(void *data)
{
set_zonelist_order();
_
>
> ...
>
> > If we'd handled the section via
> >
> > #ifdef CONFIG_MEMORY_HOTPLUG
> > #define __meminit
> > #else
> > #define __meminit __init
> > #endif
> >
> > of similar then that would fix things. iirc we used to do it that
> > way...
>
> Yes, that older way made more sense to me.
I guess it was too simple.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH] mm: fix setup_zone_pageset section mismatch
2011-03-30 22:53 ` Andrew Morton
@ 2011-03-31 0:28 ` Randy Dunlap
0 siblings, 0 replies; 5+ messages in thread
From: Randy Dunlap @ 2011-03-31 0:28 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-mm, Paul Mundt
On 03/30/11 15:53, Andrew Morton wrote:
> On Wed, 30 Mar 2011 15:47:30 -0700
> Randy Dunlap <randy.dunlap@oracle.com> wrote:
>
>> ...
>>
>>>
>>> I already merged Paul Mundt's patch whcih marks build_all_zonelists()
>>> as __ref. That seems a better solution?
>>
>> Merged where? mmotm?
>
> mm.
>
> --- a/mm/page_alloc.c~mm-page_allocc-silence-build_all_zonelists-section-mismatch
> +++ a/mm/page_alloc.c
> @@ -3170,7 +3170,7 @@ static __init_refok int __build_all_zone
> * Called with zonelists_mutex held always
> * unless system_state == SYSTEM_BOOTING.
> */
> -void build_all_zonelists(void *data)
> +void __ref build_all_zonelists(void *data)
> {
> set_zonelist_order();
>
> _
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Thanks.
--
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2011-03-31 0:28 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-03-24 20:24 [PATCH] mm: fix setup_zone_pageset section mismatch Randy Dunlap
2011-03-30 22:05 ` Andrew Morton
2011-03-30 22:47 ` Randy Dunlap
2011-03-30 22:53 ` Andrew Morton
2011-03-31 0:28 ` Randy Dunlap
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).