[PATCH] ARM: reenable DEBUG_SECTION_MISMATCH

[PATCH] ARM: reenable DEBUG_SECTION_MISMATCH

[Uwe Kleine-König]

For 2.6.38-rc4-git9 only 29 out of 133 defconfigs still produce section
mismatches.  These defconfigs produce 55 mismatches (weighted sum, so
maybe less uniq mismatches).

This is in my opinion enough to start scaring people about the remaining
problems.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
---
 lib/Kconfig.debug |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 3967c23..1130dd4 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -102,7 +102,7 @@ config HEADERS_CHECK
 
 config DEBUG_SECTION_MISMATCH
 	bool "Enable full Section mismatch analysis"
-	depends on UNDEFINED || (BLACKFIN)
+	depends on ARM || BLACKFIN
 	default y
 	# This option is on purpose disabled for now.
 	# It will be enabled when we are down to a reasonable number
-- 
1.7.2.3

Re: [PATCH] ARM: reenable DEBUG_SECTION_MISMATCH

[Randy Dunlap]

On Wed, 16 Feb 2011 18:23:13 +0100 Uwe Kleine-König wrote:

> For 2.6.38-rc4-git9 only 29 out of 133 defconfigs still produce section
> mismatches.  These defconfigs produce 55 mismatches (weighted sum, so
> maybe less uniq mismatches).
> 
> This is in my opinion enough to start scaring people about the remaining
> problems.

Yes, we have way too many of these and need more eyes on them.


> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

> ---
>  lib/Kconfig.debug |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
> index 3967c23..1130dd4 100644
> --- a/lib/Kconfig.debug
> +++ b/lib/Kconfig.debug
> @@ -102,7 +102,7 @@ config HEADERS_CHECK
>  
>  config DEBUG_SECTION_MISMATCH
>  	bool "Enable full Section mismatch analysis"
> -	depends on UNDEFINED || (BLACKFIN)
> +	depends on ARM || BLACKFIN

I would even like something like this:

	depends on ARM || BLACKFIN || KERNEL_DEVELOPER

or even
	select DEBUG_SECTION_MISMATCH if KERNEL_DEVELOPER

to make these warnings appear... but even with today's other warnings, lots
of them are ignored.


>  	default y
>  	# This option is on purpose disabled for now.
>  	# It will be enabled when we are down to a reasonable number
> -- 


---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

Re: [PATCH] ARM: reenable DEBUG_SECTION_MISMATCH

[Sam Ravnborg]

On Wed, Feb 16, 2011 at 06:23:13PM +0100, Uwe Kleine-König wrote:
> For 2.6.38-rc4-git9 only 29 out of 133 defconfigs still produce section
> mismatches.  These defconfigs produce 55 mismatches (weighted sum, so
> maybe less uniq mismatches).
> 
> This is in my opinion enough to start scaring people about the remaining
> problems.
> 
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> ---
>  lib/Kconfig.debug |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
> index 3967c23..1130dd4 100644
> --- a/lib/Kconfig.debug
> +++ b/lib/Kconfig.debug
> @@ -102,7 +102,7 @@ config HEADERS_CHECK
>  
>  config DEBUG_SECTION_MISMATCH
>  	bool "Enable full Section mismatch analysis"
> -	depends on UNDEFINED || (BLACKFIN)
> +	depends on ARM || BLACKFIN
>  	default y
>  	# This option is on purpose disabled for now.
>  	# It will be enabled when we are down to a reasonable number

If we are deciding to make this ARCH dependent then a blacklist is
much preferred.

I know that we have warnings lingering in the kernel for a long time.
But throwing these waring in the face of people for each build will
be eoung incentive to get most fixed.

Last I took a deeper look at this I had allyesconfig + allmodconfig
almost warning free on x86. But I recall that especially the HOTPLUG_CPU
stuff was tricky as they misuse the annotation.

We could try to enable it per default - and seem if people scream too much.

	Sam

[PATCH] kbuild: reenable section mismatch analysis

[Uwe Kleine-König]

This was disabled in commit

	e5f95c8 (kbuild: print only total number of section mismatces found)

because there were too many warnings.  Now we're down to a reasonable
number again, so we start scaring people with the details.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
---
 lib/Kconfig.debug |    4 ----
 1 files changed, 0 insertions(+), 4 deletions(-)

diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 2b97418..0f77ed1 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -102,11 +102,7 @@ config HEADERS_CHECK
 
 config DEBUG_SECTION_MISMATCH
 	bool "Enable full Section mismatch analysis"
-	depends on UNDEFINED || (BLACKFIN)
 	default y
-	# This option is on purpose disabled for now.
-	# It will be enabled when we are down to a reasonable number
-	# of section mismatch warnings (< 10 for an allyesconfig build)
 	help
 	  The section mismatch analysis checks if there are illegal
 	  references from one section to another section.
-- 
1.7.2.3

Re: [PATCH] kbuild: reenable section mismatch analysis

[Randy Dunlap]

On Fri, 18 Feb 2011 21:47:20 +0100 Uwe Kleine-König wrote:

> This was disabled in commit
> 
> 	e5f95c8 (kbuild: print only total number of section mismatces found)
> 
> because there were too many warnings.  Now we're down to a reasonable
> number again, so we start scaring people with the details.
> 
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

Acked-by: Randy Dunlap <rdunlap@xenotime.net>

Thanks.

> ---
>  lib/Kconfig.debug |    4 ----
>  1 files changed, 0 insertions(+), 4 deletions(-)
> 
> diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
> index 2b97418..0f77ed1 100644
> --- a/lib/Kconfig.debug
> +++ b/lib/Kconfig.debug
> @@ -102,11 +102,7 @@ config HEADERS_CHECK
>  
>  config DEBUG_SECTION_MISMATCH
>  	bool "Enable full Section mismatch analysis"
> -	depends on UNDEFINED || (BLACKFIN)
>  	default y
> -	# This option is on purpose disabled for now.
> -	# It will be enabled when we are down to a reasonable number
> -	# of section mismatch warnings (< 10 for an allyesconfig build)
>  	help
>  	  The section mismatch analysis checks if there are illegal
>  	  references from one section to another section.
> -- 


---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

Re: [PATCH] kbuild: reenable section mismatch analysis

[Mike Frysinger]

2011/2/18 Uwe Kleine-König <u.kleine-koenig@pengutronix.de>:
> This was disabled in commit
>
>        e5f95c8 (kbuild: print only total number of section mismatces found)
>
> because there were too many warnings.  Now we're down to a reasonable
> number again, so we start scaring people with the details.

Acked-by: Mike Frysinger <vapier@gentoo.org>
-mike

Re: [PATCH] kbuild: reenable section mismatch analysis

[Michal Marek]

On 18.2.2011 21:47, Uwe Kleine-König wrote:
> This was disabled in commit
> 
> 	e5f95c8 (kbuild: print only total number of section mismatces found)
> 
> because there were too many warnings.  Now we're down to a reasonable
> number again, so we start scaring people with the details.
> 
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

OK, applied to kbuild-2.6.git#kbuild.

Thanks,
Michal

Re: [PATCH] kbuild: reenable section mismatch analysis

[Andrew Morton]

On Fri, 18 Feb 2011 21:47:20 +0100
Uwe Kleine-K__nig  <u.kleine-koenig@pengutronix.de> wrote:

> This was disabled in commit
> 
> 	e5f95c8 (kbuild: print only total number of section mismatces found)
> 
> because there were too many warnings.  Now we're down to a reasonable
> number again, so we start scaring people with the details.
> 
> Signed-off-by: Uwe Kleine-K__nig <u.kleine-koenig@pengutronix.de>
> ---
>  lib/Kconfig.debug |    4 ----
>  1 files changed, 0 insertions(+), 4 deletions(-)
> 
> diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
> index 2b97418..0f77ed1 100644
> --- a/lib/Kconfig.debug
> +++ b/lib/Kconfig.debug
> @@ -102,11 +102,7 @@ config HEADERS_CHECK
>  
>  config DEBUG_SECTION_MISMATCH
>  	bool "Enable full Section mismatch analysis"
> -	depends on UNDEFINED || (BLACKFIN)
>  	default y
> -	# This option is on purpose disabled for now.
> -	# It will be enabled when we are down to a reasonable number
> -	# of section mismatch warnings (< 10 for an allyesconfig build)
>  	help
>  	  The section mismatch analysis checks if there are illegal
>  	  references from one section to another section.

eaaaargh!  650 lines of crap added to my i386 allmodconfig build output.

I'll be locally opting out of this one, thanks.

Re: [PATCH] kbuild: reenable section mismatch analysis

[Randy Dunlap]

On Thu, 10 Mar 2011 16:41:00 -0800 Andrew Morton wrote:

> On Fri, 18 Feb 2011 21:47:20 +0100
> Uwe Kleine-K__nig  <u.kleine-koenig@pengutronix.de> wrote:
> 
> > This was disabled in commit
> > 
> > 	e5f95c8 (kbuild: print only total number of section mismatces found)
> > 
> > because there were too many warnings.  Now we're down to a reasonable
> > number again, so we start scaring people with the details.
> > 
> > Signed-off-by: Uwe Kleine-K__nig <u.kleine-koenig@pengutronix.de>
> > ---
> >  lib/Kconfig.debug |    4 ----
> >  1 files changed, 0 insertions(+), 4 deletions(-)
> > 
> > diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
> > index 2b97418..0f77ed1 100644
> > --- a/lib/Kconfig.debug
> > +++ b/lib/Kconfig.debug
> > @@ -102,11 +102,7 @@ config HEADERS_CHECK
> >  
> >  config DEBUG_SECTION_MISMATCH
> >  	bool "Enable full Section mismatch analysis"
> > -	depends on UNDEFINED || (BLACKFIN)
> >  	default y
> > -	# This option is on purpose disabled for now.
> > -	# It will be enabled when we are down to a reasonable number
> > -	# of section mismatch warnings (< 10 for an allyesconfig build)
> >  	help
> >  	  The section mismatch analysis checks if there are illegal
> >  	  references from one section to another section.
> 
> eaaaargh!  650 lines of crap added to my i386 allmodconfig build output.

Each message is a bit longish, yes.

> I'll be locally opting out of this one, thanks.
> --


---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

Re: [PATCH] kbuild: reenable section mismatch analysis

[Mike Frysinger]

2011/3/10 Andrew Morton :
> On Fri, 18 Feb 2011 21:47:20 +0100 Uwe Kleine-K__nig wrote:
>> This was disabled in commit
>>
>>       e5f95c8 (kbuild: print only total number of section mismatces found)
>>
>> because there were too many warnings.  Now we're down to a reasonable
>> number again, so we start scaring people with the details.
>>
>> --- a/lib/Kconfig.debug
>> +++ b/lib/Kconfig.debug
>> @@ -102,11 +102,7 @@ config HEADERS_CHECK
>>
>>  config DEBUG_SECTION_MISMATCH
>>       bool "Enable full Section mismatch analysis"
>> -     depends on UNDEFINED || (BLACKFIN)
>>       default y
>> -     # This option is on purpose disabled for now.
>> -     # It will be enabled when we are down to a reasonable number
>> -     # of section mismatch warnings (< 10 for an allyesconfig build)
>>       help
>>         The section mismatch analysis checks if there are illegal
>>         references from one section to another section.
>
> eaaaargh!  650 lines of crap added to my i386 allmodconfig build output.

and it's going to keep staying all f-ed up until people start seeing
this stuff themselves when they write the bad code
-mike

Re: [PATCH] kbuild: reenable section mismatch analysis

[Andrew Morton]

On Thu, 10 Mar 2011 20:00:54 -0500
Mike Frysinger <vapier.adi@gmail.com> wrote:

> >
> > eaaaargh! __650 lines of crap added to my i386 allmodconfig build output.
> 
> and it's going to keep staying all f-ed up until people start seeing
> this stuff themselves when they write the bad code

That won't work - everyone will either disable the messages or will
train themselves to ignore them.

Re: [PATCH] kbuild: reenable section mismatch analysis

[Mike Frysinger]

On Thu, Mar 10, 2011 at 20:05, Andrew Morton wrote:
> On Thu, 10 Mar 2011 20:00:54 -0500 Mike Frysinger wrote:
>> > eaaaargh! __650 lines of crap added to my i386 allmodconfig build output.
>>
>> and it's going to keep staying all f-ed up until people start seeing
>> this stuff themselves when they write the bad code
>
> That won't work - everyone will either disable the messages or will
> train themselves to ignore them.

the status quo is obviously not working.  i spent quite a bit of time
in the past to get the warnings down to 0 for my builds and all ive
seen since then is the # creep back up.  if people arent being made
aware of the issue, they obviously arent going to do anything about
it.  some people might condition themselves to ignore it, but that
applies to any sort of warning message.
-mike

Re: [PATCH] kbuild: reenable section mismatch analysis

[Sam Ravnborg]

On Thu, Mar 10, 2011 at 04:41:00PM -0800, Andrew Morton wrote:
> On Fri, 18 Feb 2011 21:47:20 +0100
> Uwe Kleine-K__nig  <u.kleine-koenig@pengutronix.de> wrote:
> 
> > This was disabled in commit
> > 
> > 	e5f95c8 (kbuild: print only total number of section mismatces found)
> > 
> > because there were too many warnings.  Now we're down to a reasonable
> > number again, so we start scaring people with the details.
> > 
> > Signed-off-by: Uwe Kleine-K__nig <u.kleine-koenig@pengutronix.de>
> > ---
> >  lib/Kconfig.debug |    4 ----
> >  1 files changed, 0 insertions(+), 4 deletions(-)
> > 
> > diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
> > index 2b97418..0f77ed1 100644
> > --- a/lib/Kconfig.debug
> > +++ b/lib/Kconfig.debug
> > @@ -102,11 +102,7 @@ config HEADERS_CHECK
> >  
> >  config DEBUG_SECTION_MISMATCH
> >  	bool "Enable full Section mismatch analysis"
> > -	depends on UNDEFINED || (BLACKFIN)
> >  	default y
> > -	# This option is on purpose disabled for now.
> > -	# It will be enabled when we are down to a reasonable number
> > -	# of section mismatch warnings (< 10 for an allyesconfig build)
> >  	help
> >  	  The section mismatch analysis checks if there are illegal
> >  	  references from one section to another section.
> 
> eaaaargh!  650 lines of crap added to my i386 allmodconfig build output.
> 
> I'll be locally opting out of this one, thanks.

How about that we enable this in the merge window.
And if it continue to be unreasonable then drop it again
around rc5 time?

The likelyhood that warnings are fixed is bigger if people
see them and get annoyed.

	Sam