public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* ima changes to shmem breaks !CONFIG_SHMEM
@ 2009-02-09 16:30 Mike Frysinger
  2009-02-09 21:23 ` James Morris
  0 siblings, 1 reply; 7+ messages in thread
From: Mike Frysinger @ 2009-02-09 16:30 UTC (permalink / raw)
  To: Mimi Zohar; +Cc: LKML, Serge Hallyn, James Morris

commit 1df9f0a73178718969ae47d813b8e7aab2cf073c broke default building
for Blackfin systems:
  CC      mm/shmem.o
mm/shmem.c: In function 'shmem_zero_setup':
mm/shmem.c:2670: error: implicit declaration of function 'ima_shm_check'
make[1]: *** [mm/shmem.o] Error 1

that's because the ima.h include was added behind the CONFIG_SHMEM
markings rather than everywhere, just like the function is used.
-mike

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: ima changes to shmem breaks !CONFIG_SHMEM
  2009-02-09 16:30 ima changes to shmem breaks !CONFIG_SHMEM Mike Frysinger
@ 2009-02-09 21:23 ` James Morris
  2009-02-09 21:27   ` Mike Frysinger
  0 siblings, 1 reply; 7+ messages in thread
From: James Morris @ 2009-02-09 21:23 UTC (permalink / raw)
  To: Mike Frysinger; +Cc: Mimi Zohar, LKML, Serge Hallyn

On Mon, 9 Feb 2009, Mike Frysinger wrote:

> commit 1df9f0a73178718969ae47d813b8e7aab2cf073c broke default building
> for Blackfin systems:
>   CC      mm/shmem.o
> mm/shmem.c: In function 'shmem_zero_setup':
> mm/shmem.c:2670: error: implicit declaration of function 'ima_shm_check'
> make[1]: *** [mm/shmem.o] Error 1
> 
> that's because the ima.h include was added behind the CONFIG_SHMEM
> markings rather than everywhere, just like the function is used.

Does this seem like the correct fix?

--- 

diff --git a/mm/shmem.c b/mm/shmem.c
index 7519988..8135fac 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -28,6 +28,7 @@
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/swap.h>
+#include <linux/ima.h>
 
 static struct vfsmount *shm_mnt;
 
@@ -59,7 +60,6 @@ static struct vfsmount *shm_mnt;
 #include <linux/highmem.h>
 #include <linux/seq_file.h>
 #include <linux/magic.h>
-#include <linux/ima.h>
 
 #include <asm/uaccess.h>
 #include <asm/div64.h>



^ permalink raw reply related	[flat|nested] 7+ messages in thread

* Re: ima changes to shmem breaks !CONFIG_SHMEM
  2009-02-09 21:23 ` James Morris
@ 2009-02-09 21:27   ` Mike Frysinger
  2009-02-09 21:49     ` Randy Dunlap
  0 siblings, 1 reply; 7+ messages in thread
From: Mike Frysinger @ 2009-02-09 21:27 UTC (permalink / raw)
  To: James Morris; +Cc: Mimi Zohar, LKML, Serge Hallyn

On Mon, Feb 9, 2009 at 16:23, James Morris wrote:
> On Mon, 9 Feb 2009, Mike Frysinger wrote:
>> commit 1df9f0a73178718969ae47d813b8e7aab2cf073c broke default building
>> for Blackfin systems:
>>   CC      mm/shmem.o
>> mm/shmem.c: In function 'shmem_zero_setup':
>> mm/shmem.c:2670: error: implicit declaration of function 'ima_shm_check'
>> make[1]: *** [mm/shmem.o] Error 1
>>
>> that's because the ima.h include was added behind the CONFIG_SHMEM
>> markings rather than everywhere, just like the function is used.
>
> Does this seem like the correct fix?

it converts the build error to a build warning:
  CC      mm/shmem.o
In file included from mm/shmem.c:32:
include/linux/ima.h:25: warning: 'struct linux_binprm' declared inside
parameter list
include/linux/ima.h:25: warning: its scope is only this definition or
declaration, which is probably not what you want
-mike

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: ima changes to shmem breaks !CONFIG_SHMEM
  2009-02-09 21:27   ` Mike Frysinger
@ 2009-02-09 21:49     ` Randy Dunlap
  2009-02-09 21:51       ` Mike Frysinger
  0 siblings, 1 reply; 7+ messages in thread
From: Randy Dunlap @ 2009-02-09 21:49 UTC (permalink / raw)
  To: Mike Frysinger; +Cc: James Morris, Mimi Zohar, LKML, Serge Hallyn

Mike Frysinger wrote:
> On Mon, Feb 9, 2009 at 16:23, James Morris wrote:
>> On Mon, 9 Feb 2009, Mike Frysinger wrote:
>>> commit 1df9f0a73178718969ae47d813b8e7aab2cf073c broke default building
>>> for Blackfin systems:
>>>   CC      mm/shmem.o
>>> mm/shmem.c: In function 'shmem_zero_setup':
>>> mm/shmem.c:2670: error: implicit declaration of function 'ima_shm_check'
>>> make[1]: *** [mm/shmem.o] Error 1
>>>
>>> that's because the ima.h include was added behind the CONFIG_SHMEM
>>> markings rather than everywhere, just like the function is used.
>> Does this seem like the correct fix?
> 
> it converts the build error to a build warning:
>   CC      mm/shmem.o
> In file included from mm/shmem.c:32:
> include/linux/ima.h:25: warning: 'struct linux_binprm' declared inside
> parameter list
> include/linux/ima.h:25: warning: its scope is only this definition or
> declaration, which is probably not what you want

Add to the other patch:

--- linux-next-20090209.orig/include/linux/ima.h
+++ linux-next-20090209/include/linux/ima.h
@@ -8,6 +8,7 @@
  */
 
 #include <linux/fs.h>
+struct linux_binprm;
 
 #ifndef _LINUX_IMA_H
 #define _LINUX_IMA_H



and it builds for me.

-- 
~Randy

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: ima changes to shmem breaks !CONFIG_SHMEM
  2009-02-09 21:49     ` Randy Dunlap
@ 2009-02-09 21:51       ` Mike Frysinger
  2009-02-09 21:56         ` Randy Dunlap
  0 siblings, 1 reply; 7+ messages in thread
From: Mike Frysinger @ 2009-02-09 21:51 UTC (permalink / raw)
  To: Randy Dunlap; +Cc: James Morris, Mimi Zohar, LKML, Serge Hallyn

On Mon, Feb 9, 2009 at 16:49, Randy Dunlap wrote:
> Mike Frysinger wrote:
>> On Mon, Feb 9, 2009 at 16:23, James Morris wrote:
>>> On Mon, 9 Feb 2009, Mike Frysinger wrote:
>>>> commit 1df9f0a73178718969ae47d813b8e7aab2cf073c broke default building
>>>> for Blackfin systems:
>>>>   CC      mm/shmem.o
>>>> mm/shmem.c: In function 'shmem_zero_setup':
>>>> mm/shmem.c:2670: error: implicit declaration of function 'ima_shm_check'
>>>> make[1]: *** [mm/shmem.o] Error 1
>>>>
>>>> that's because the ima.h include was added behind the CONFIG_SHMEM
>>>> markings rather than everywhere, just like the function is used.
>>> Does this seem like the correct fix?
>>
>> it converts the build error to a build warning:
>>   CC      mm/shmem.o
>> In file included from mm/shmem.c:32:
>> include/linux/ima.h:25: warning: 'struct linux_binprm' declared inside
>> parameter list
>> include/linux/ima.h:25: warning: its scope is only this definition or
>> declaration, which is probably not what you want
>
> Add to the other patch:
>
> --- linux-next-20090209.orig/include/linux/ima.h
> +++ linux-next-20090209/include/linux/ima.h
> @@ -8,6 +8,7 @@
>  */
>
>  #include <linux/fs.h>
> +struct linux_binprm;
>
>  #ifndef _LINUX_IMA_H
>  #define _LINUX_IMA_H
>
>
>
> and it builds for me.

yes, it builds w/out warnings now for me too.  any reason those are
outside of _LINUX_IMA_H ?  looks like they should both be inside of
that.
-mike

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: ima changes to shmem breaks !CONFIG_SHMEM
  2009-02-09 21:51       ` Mike Frysinger
@ 2009-02-09 21:56         ` Randy Dunlap
  2009-02-09 23:59           ` Mimi Zohar
  0 siblings, 1 reply; 7+ messages in thread
From: Randy Dunlap @ 2009-02-09 21:56 UTC (permalink / raw)
  To: Mike Frysinger; +Cc: Randy Dunlap, James Morris, Mimi Zohar, LKML, Serge Hallyn

Mike Frysinger wrote:
> On Mon, Feb 9, 2009 at 16:49, Randy Dunlap wrote:
>> Mike Frysinger wrote:
>>> On Mon, Feb 9, 2009 at 16:23, James Morris wrote:
>>>> On Mon, 9 Feb 2009, Mike Frysinger wrote:
>>>>> commit 1df9f0a73178718969ae47d813b8e7aab2cf073c broke default building
>>>>> for Blackfin systems:
>>>>>   CC      mm/shmem.o
>>>>> mm/shmem.c: In function 'shmem_zero_setup':
>>>>> mm/shmem.c:2670: error: implicit declaration of function 'ima_shm_check'
>>>>> make[1]: *** [mm/shmem.o] Error 1
>>>>>
>>>>> that's because the ima.h include was added behind the CONFIG_SHMEM
>>>>> markings rather than everywhere, just like the function is used.
>>>> Does this seem like the correct fix?
>>> it converts the build error to a build warning:
>>>   CC      mm/shmem.o
>>> In file included from mm/shmem.c:32:
>>> include/linux/ima.h:25: warning: 'struct linux_binprm' declared inside
>>> parameter list
>>> include/linux/ima.h:25: warning: its scope is only this definition or
>>> declaration, which is probably not what you want
>> Add to the other patch:
>>
>> --- linux-next-20090209.orig/include/linux/ima.h
>> +++ linux-next-20090209/include/linux/ima.h
>> @@ -8,6 +8,7 @@
>>  */
>>
>>  #include <linux/fs.h>
>> +struct linux_binprm;
>>
>>  #ifndef _LINUX_IMA_H
>>  #define _LINUX_IMA_H
>>
>>
>>
>> and it builds for me.
> 
> yes, it builds w/out warnings now for me too.  any reason those are
> outside of _LINUX_IMA_H ?  looks like they should both be inside of
> that.

Agreed, that change is also good.


Thanks.
-- 
~Randy

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: ima changes to shmem breaks !CONFIG_SHMEM
  2009-02-09 21:56         ` Randy Dunlap
@ 2009-02-09 23:59           ` Mimi Zohar
  0 siblings, 0 replies; 7+ messages in thread
From: Mimi Zohar @ 2009-02-09 23:59 UTC (permalink / raw)
  To: Randy Dunlap; +Cc: Mike Frysinger, James Morris, LKML, Serge Hallyn

On Mon, 2009-02-09 at 13:56 -0800, Randy Dunlap wrote:
> Mike Frysinger wrote:
> > On Mon, Feb 9, 2009 at 16:49, Randy Dunlap wrote:
> >> Mike Frysinger wrote:
> >>> On Mon, Feb 9, 2009 at 16:23, James Morris wrote:
> >>>> On Mon, 9 Feb 2009, Mike Frysinger wrote:
> >>>>> commit 1df9f0a73178718969ae47d813b8e7aab2cf073c broke default building
> >>>>> for Blackfin systems:
> >>>>>   CC      mm/shmem.o
> >>>>> mm/shmem.c: In function 'shmem_zero_setup':
> >>>>> mm/shmem.c:2670: error: implicit declaration of function 'ima_shm_check'
> >>>>> make[1]: *** [mm/shmem.o] Error 1
> >>>>>
> >>>>> that's because the ima.h include was added behind the CONFIG_SHMEM
> >>>>> markings rather than everywhere, just like the function is used.
> >>>> Does this seem like the correct fix?
> >>> it converts the build error to a build warning:
> >>>   CC      mm/shmem.o
> >>> In file included from mm/shmem.c:32:
> >>> include/linux/ima.h:25: warning: 'struct linux_binprm' declared inside
> >>> parameter list
> >>> include/linux/ima.h:25: warning: its scope is only this definition or
> >>> declaration, which is probably not what you want
> >> Add to the other patch:
> >>
> >> --- linux-next-20090209.orig/include/linux/ima.h
> >> +++ linux-next-20090209/include/linux/ima.h
> >> @@ -8,6 +8,7 @@
> >>  */
> >>
> >>  #include <linux/fs.h>
> >> +struct linux_binprm;
> >>
> >>  #ifndef _LINUX_IMA_H
> >>  #define _LINUX_IMA_H
> >>
> >>
> >>
> >> and it builds for me.
> > 
> > yes, it builds w/out warnings now for me too.  any reason those are
> > outside of _LINUX_IMA_H ?  looks like they should both be inside of
> > that.
> 
> Agreed, that change is also good.
> 
> 
> Thanks.

I posted the patch, forgetting to copy both Mike and you.  Could you
please verify the patch resolves the problem.

Thanks!

Mimi


^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2009-02-09 23:59 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-02-09 16:30 ima changes to shmem breaks !CONFIG_SHMEM Mike Frysinger
2009-02-09 21:23 ` James Morris
2009-02-09 21:27   ` Mike Frysinger
2009-02-09 21:49     ` Randy Dunlap
2009-02-09 21:51       ` Mike Frysinger
2009-02-09 21:56         ` Randy Dunlap
2009-02-09 23:59           ` Mimi Zohar

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox