2006-09-29 03:41:36

by Randy Dunlap

[permalink] [raw]
Subject: [PATCH] fix EMBEDDED + SYSCTL menu

From: Randy Dunlap <[email protected]>

SYSCTL should still depend on EMBEDDED. This unbreaks the
EMBEDDED menu (from the recent SYSCTL_SYCALL menu option patch).

Fix typos in new SYSCTL_SYSCALL menu.

Signed-off-by: Randy Dunlap <[email protected]>
---
init/Kconfig | 14 +++++++-------
1 files changed, 7 insertions(+), 7 deletions(-)

--- linux-2618-g10.orig/init/Kconfig
+++ linux-2618-g10/init/Kconfig
@@ -257,6 +257,9 @@ config CC_OPTIMIZE_FOR_SIZE

If unsure, say N.

+config SYSCTL
+ bool
+
menuconfig EMBEDDED
bool "Configure standard kernel features (for small systems)"
help
@@ -272,11 +275,8 @@ config UID16
help
This enables the legacy 16-bit UID syscall wrappers.

-config SYSCTL
- bool
-
config SYSCTL_SYSCALL
- bool "Sysctl syscall support"
+ bool "Sysctl syscall support" if EMBEDDED
default n
select SYSCTL
---help---
@@ -285,11 +285,11 @@ config SYSCTL_SYSCALL
and use. The interface in /proc/sys is now the primary and what
everyone uses.

- Nothing has been using the binary sysctl interface for some time
+ Nothing has been using the binary sysctl interface for some
time now so nothing should break if you disable sysctl syscall
- support, and you kernel will get marginally smaller.
+ support, and your kernel will get marginally smaller.

- Unless you have an application that uses the sys_syscall interface
+ Unless you have an application that uses the sys_sysctl interface
you should probably say N here.

config KALLSYMS


---


2006-09-29 16:05:44

by Adrian Bunk

[permalink] [raw]
Subject: Re: [PATCH] fix EMBEDDED + SYSCTL menu

On Thu, Sep 28, 2006 at 08:42:51PM -0700, Randy Dunlap wrote:
> From: Randy Dunlap <[email protected]>
>
> SYSCTL should still depend on EMBEDDED. This unbreaks the
> EMBEDDED menu (from the recent SYSCTL_SYCALL menu option patch).
>
> Fix typos in new SYSCTL_SYSCALL menu.
>
> Signed-off-by: Randy Dunlap <[email protected]>
> ---
> init/Kconfig | 14 +++++++-------
> 1 files changed, 7 insertions(+), 7 deletions(-)
>
> --- linux-2618-g10.orig/init/Kconfig
> +++ linux-2618-g10/init/Kconfig
> @@ -257,6 +257,9 @@ config CC_OPTIMIZE_FOR_SIZE
>
> If unsure, say N.
>
> +config SYSCTL
> + bool
> +
> menuconfig EMBEDDED
> bool "Configure standard kernel features (for small systems)"
> help
> @@ -272,11 +275,8 @@ config UID16
> help
> This enables the legacy 16-bit UID syscall wrappers.
>
> -config SYSCTL
> - bool
> -

ACK

> config SYSCTL_SYSCALL
> - bool "Sysctl syscall support"
> + bool "Sysctl syscall support" if EMBEDDED
> default n
> select SYSCTL
> ---help---
>...

You could achieve the same by removing the option...

Simply move SYSCTL_SYSCALL to the same place you are moving SYSCTL to
without fiddling with the dependencies.

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

2006-09-29 16:23:30

by Randy Dunlap

[permalink] [raw]
Subject: Re: [PATCH] fix EMBEDDED + SYSCTL menu

On Fri, 29 Sep 2006 18:05:21 +0200 Adrian Bunk wrote:

> On Thu, Sep 28, 2006 at 08:42:51PM -0700, Randy Dunlap wrote:
> > From: Randy Dunlap <[email protected]>
> >
> > SYSCTL should still depend on EMBEDDED. This unbreaks the
> > EMBEDDED menu (from the recent SYSCTL_SYCALL menu option patch).
> >
> > Fix typos in new SYSCTL_SYSCALL menu.
> >
> > Signed-off-by: Randy Dunlap <[email protected]>
> > ---
> > init/Kconfig | 14 +++++++-------
> > 1 files changed, 7 insertions(+), 7 deletions(-)
> >
> > --- linux-2618-g10.orig/init/Kconfig
> > +++ linux-2618-g10/init/Kconfig
> > @@ -257,6 +257,9 @@ config CC_OPTIMIZE_FOR_SIZE
> >
> > If unsure, say N.
> >
> > +config SYSCTL
> > + bool
> > +
> > menuconfig EMBEDDED
> > bool "Configure standard kernel features (for small systems)"
> > help
> > @@ -272,11 +275,8 @@ config UID16
> > help
> > This enables the legacy 16-bit UID syscall wrappers.
> >
> > -config SYSCTL
> > - bool
> > -
>
> ACK
>
> > config SYSCTL_SYSCALL
> > - bool "Sysctl syscall support"
> > + bool "Sysctl syscall support" if EMBEDDED
> > default n
> > select SYSCTL
> > ---help---
> >...
>
> You could achieve the same by removing the option...
>
> Simply move SYSCTL_SYSCALL to the same place you are moving SYSCTL to
> without fiddling with the dependencies.

Yes, I realize that (I even had that patch earlier).
I don't care which way it's done. Eric, any preference here?

---
~Randy

2006-09-30 00:57:18

by Eric W. Biederman

[permalink] [raw]
Subject: Re: [PATCH] fix EMBEDDED + SYSCTL menu

Randy Dunlap <[email protected]> writes:

> On Fri, 29 Sep 2006 18:05:21 +0200 Adrian Bunk wrote:
>
>> On Thu, Sep 28, 2006 at 08:42:51PM -0700, Randy Dunlap wrote:
>> > From: Randy Dunlap <[email protected]>
>> >
>> > SYSCTL should still depend on EMBEDDED. This unbreaks the
>> > EMBEDDED menu (from the recent SYSCTL_SYCALL menu option patch).
>> >
>> > Fix typos in new SYSCTL_SYSCALL menu.
>> >
>> > Signed-off-by: Randy Dunlap <[email protected]>
>> > ---
>> > init/Kconfig | 14 +++++++-------
>> > 1 files changed, 7 insertions(+), 7 deletions(-)
>> >
>> > --- linux-2618-g10.orig/init/Kconfig
>> > +++ linux-2618-g10/init/Kconfig
>> > @@ -257,6 +257,9 @@ config CC_OPTIMIZE_FOR_SIZE
>> >
>> > If unsure, say N.
>> >
>> > +config SYSCTL
>> > + bool
>> > +
>> > menuconfig EMBEDDED
>> > bool "Configure standard kernel features (for small systems)"
>> > help
>> > @@ -272,11 +275,8 @@ config UID16
>> > help
>> > This enables the legacy 16-bit UID syscall wrappers.
>> >
>> > -config SYSCTL
>> > - bool
>> > -
>>
>> ACK
>>
>> > config SYSCTL_SYSCALL
>> > - bool "Sysctl syscall support"
>> > + bool "Sysctl syscall support" if EMBEDDED
>> > default n
>> > select SYSCTL
>> > ---help---
>> >...
>>
>> You could achieve the same by removing the option...
>>
>> Simply move SYSCTL_SYSCALL to the same place you are moving SYSCTL to
>> without fiddling with the dependencies.
>
> Yes, I realize that (I even had that patch earlier).
> I don't care which way it's done. Eric, any preference here?

Not really. The point of this patch is to bring the sys_sysctl discussion
to a conclusion. Either we don't support it and no one uses it. Or
someone actually uses it so we need to support it. The stupid case in
glibc where it tests for an SMP kernel when it should be using uname
and has a fallback into /proc/sys doesn't count.

Eric

p.s. I'm confused I only saw this message through lkml. I'm wondering
why haven't I gotten it normally yet.