[PATCH] MIPS: EARLY_PRINTK: Fixup of dependency

[PATCH] MIPS: EARLY_PRINTK: Fixup of dependency

[Wu Zhangjin]

From: Wu Zhangjin <wuzhangjin@gmail.com>

In the old version, if the platform select SYS_HAS_EARLY_PRINTK and
users not select DEBUG_KERNEL, there is no interface for user to enable
or disable the EARLY_PRINTK option, it will be enabled all the time.
this will waste a little bit of memory and slow down the booting of
kernel but the users never know it.

This patch will only enable that option when the DEBUG_KERNEL is
enabled.

Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
---
 arch/mips/Kconfig |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index b342197..d2446d5 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -824,8 +824,8 @@ config DMA_NEED_PCI_MAP_STATE
 	bool
 
 config EARLY_PRINTK
-	bool "Early printk" if EMBEDDED && DEBUG_KERNEL
-	depends on SYS_HAS_EARLY_PRINTK
+	bool "Early printk" if EMBEDDED
+	depends on SYS_HAS_EARLY_PRINTK && DEBUG_KERNEL
 	default y
 	help
 	  This option enables special console drivers which allow the kernel
-- 
1.6.2.1

Re: [PATCH] MIPS: EARLY_PRINTK: Fixup of dependency

[Manuel Lauss]

Hi Wu!

On Tue, Nov 24, 2009 at 10:33 AM, Wu Zhangjin <wuzhangjin@gmail.com> wrote:
[...]
> This patch will only enable that option when the DEBUG_KERNEL is
> enabled.

How about making it independent from DEBUG_KERNEL altogether?  If find
it useful even without full debug info.

Manuel

Re: [PATCH] MIPS: EARLY_PRINTK: Fixup of dependency

[Ralf Baechle]

On Tue, Nov 24, 2009 at 05:24:57PM +0100, Manuel Lauss wrote:

> On Tue, Nov 24, 2009 at 10:33 AM, Wu Zhangjin <wuzhangjin@gmail.com> wrote:
> [...]
> > This patch will only enable that option when the DEBUG_KERNEL is
> > enabled.
> 
> How about making it independent from DEBUG_KERNEL altogether?  If find
> it useful even without full debug info.

DEBUG_INFO controlls the generation of ELF debug information.  DEBUG_KERNEL
only hides most of the debugging options.

  Ralf

Re: [PATCH] MIPS: EARLY_PRINTK: Fixup of dependency

[Kevin Hickey]

On Tue, 2009-11-24 at 16:30 +0000, Ralf Baechle wrote:
> On Tue, Nov 24, 2009 at 05:24:57PM +0100, Manuel Lauss wrote:
> 
> > On Tue, Nov 24, 2009 at 10:33 AM, Wu Zhangjin <wuzhangjin@gmail.com> wrote:
> > [...]
> > > This patch will only enable that option when the DEBUG_KERNEL is
> > > enabled.
> > 
> > How about making it independent from DEBUG_KERNEL altogether?  If find
> > it useful even without full debug info.

I agree with Manuel here.  I often build release kernels that benefit
from EARLY_PRINTK.  Why not make EARLY_PRINTK a selectable option in the
config?  Coupling it to DEBUG_KERNEL seems confusing. 

=Kevin
> 
> DEBUG_INFO controlls the generation of ELF debug information.  DEBUG_KERNEL
> only hides most of the debugging options.
> 
>   Ralf
> 

Re: [PATCH] MIPS: EARLY_PRINTK: Fixup of dependency

[Wu Zhangjin]

On Wed, 2009-11-25 at 08:42 -0600, Kevin Hickey wrote:
> On Tue, 2009-11-24 at 16:30 +0000, Ralf Baechle wrote:
> > On Tue, Nov 24, 2009 at 05:24:57PM +0100, Manuel Lauss wrote:
> > 
> > > On Tue, Nov 24, 2009 at 10:33 AM, Wu Zhangjin <wuzhangjin@gmail.com> wrote:
> > > [...]
> > > > This patch will only enable that option when the DEBUG_KERNEL is
> > > > enabled.
> > > 
> > > How about making it independent from DEBUG_KERNEL altogether?  If find
> > > it useful even without full debug info.
> 
> I agree with Manuel here.  I often build release kernels that benefit
> from EARLY_PRINTK.  Why not make EARLY_PRINTK a selectable option in the
> config?  Coupling it to DEBUG_KERNEL seems confusing. 

Hello,

Ralf have moved the EARLY_PRINTK to Kconfig.debug and removed the
dependency on DEBUG_KERNEL in his -queue repository ;) Just as the X86
and some other ARCHs does:

http://www.linux-mips.org/git?p=linux-queue.git

[...]
+config EARLY_PRINTK
+       bool "Early printk" if EMBEDDED
+       depends on SYS_HAS_EARLY_PRINTK
+       default y
[...]

So, it is okay now, please ignore this patch ;)

Regards,
	Wu Zhangjin

> 
> =Kevin
> > 
> > DEBUG_INFO controlls the generation of ELF debug information.  DEBUG_KERNEL
> > only hides most of the debugging options.
> > 
> >   Ralf
> > 
> 

Re: [PATCH] MIPS: EARLY_PRINTK: Fixup of dependency

[Ralf Baechle]

On Thu, Nov 26, 2009 at 12:15:36AM +0800, Wu Zhangjin wrote:
> From: Wu Zhangjin <wuzhangjin@gmail.com>
> Date:   Thu, 26 Nov 2009 00:15:36 +0800
> To: Kevin Hickey <khickey@netlogicmicro.com>
> Cc: Ralf Baechle <ralf@linux-mips.org>,
> 	Manuel Lauss <manuel.lauss@googlemail.com>,
> 	Linux-MIPS <linux-mips@linux-mips.org>
> Subject: Re: [PATCH] MIPS: EARLY_PRINTK: Fixup of dependency
> Content-Type: text/plain; charset="UTF-8"
> 
> On Wed, 2009-11-25 at 08:42 -0600, Kevin Hickey wrote:
> > On Tue, 2009-11-24 at 16:30 +0000, Ralf Baechle wrote:
> > > On Tue, Nov 24, 2009 at 05:24:57PM +0100, Manuel Lauss wrote:
> > > 
> > > > On Tue, Nov 24, 2009 at 10:33 AM, Wu Zhangjin <wuzhangjin@gmail.com> wrote:
> > > > [...]
> > > > > This patch will only enable that option when the DEBUG_KERNEL is
> > > > > enabled.
> > > > 
> > > > How about making it independent from DEBUG_KERNEL altogether?  If find
> > > > it useful even without full debug info.
> > 
> > I agree with Manuel here.  I often build release kernels that benefit
> > from EARLY_PRINTK.  Why not make EARLY_PRINTK a selectable option in the
> > config?  Coupling it to DEBUG_KERNEL seems confusing. 
> 
> Hello,
> 
> Ralf have moved the EARLY_PRINTK to Kconfig.debug and removed the
> dependency on DEBUG_KERNEL in his -queue repository ;) Just as the X86
> and some other ARCHs does:
> 
> http://www.linux-mips.org/git?p=linux-queue.git
> 
> [...]
> +config EARLY_PRINTK
> +       bool "Early printk" if EMBEDDED
> +       depends on SYS_HAS_EARLY_PRINTK
> +       default y
> [...]
> 
> So, it is okay now, please ignore this patch ;)

Okay, dropped then.

Thanks!

  Ralf