diff mbox

[1/2] ARM: default PHYS_OFFSET if !MMU

Message ID 1384341351-4110-2-git-send-email-tom.leiming@gmail.com (mailing list archive)
State New, archived
Headers show

Commit Message

Ming Lei Nov. 13, 2013, 11:15 a.m. UTC
From: Jonathan Austin <jonathan.austin@arm.com>

Set CONFIG_PAGE_OFFSET as PHYS_OFFSET when !MMU so that
we can keep CONFIG_PAGE_OFFSET same with PAGE_OFFSET.
 
This patch and patch 2/2 fix regression on !MMU by
f6537f2f0eba4eba("scripts/kallsyms: filter symbols not
in kernel address space")

Cc: <stable@vger.kernel.org>
Signed-off-by: Jonathan Austin <jonathan.austin@arm.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
---
 arch/arm/Kconfig |    1 +
 1 file changed, 1 insertion(+)

Comments

Jonathan Austin Nov. 13, 2013, 11:46 a.m. UTC | #1
Hi Ming,

On 13/11/13 11:15, Ming Lei wrote:
> From: Jonathan Austin <jonathan.austin@arm.com>
> 
> Set CONFIG_PAGE_OFFSET as PHYS_OFFSET when !MMU so that
> we can keep CONFIG_PAGE_OFFSET same with PAGE_OFFSET.
>

I'm not so sure about this commit message - I don't think
it makes sense...Should it not be more like:

------->8-------
Subject: ARM: set PAGE_OFFSET = PHYS_OFFSET if !MMU

With !MMU we still set PAGE_OFFSET to 0xC0000000, which
makes little sense and causes a regression on !MMU after
f6537f2f0eba4eba ("scripts/kallsyms: filter symbols not
kernel address space")

This patch sets PAGE_OFFSET to PHYS_OFFSET in the !MMU
case, fixing the regression (Tested on Cortex-R7/Vexpress)
--------8<---------

Russell: Will you pick this patch up from the list or does
it need to go in through the patch system?

Jonny

> This patch and patch 2/2 fix regression on !MMU by
> f6537f2f0eba4eba("scripts/kallsyms: filter symbols not
> in kernel address space")
> 
> Cc: <stable@vger.kernel.org>
> Signed-off-by: Jonathan Austin <jonathan.austin@arm.com>
> Signed-off-by: Ming Lei <tom.leiming@gmail.com>
> ---
>   arch/arm/Kconfig |    1 +
>   1 file changed, 1 insertion(+)
> 
> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
> index 799ef94..db708c4 100644
> --- a/arch/arm/Kconfig
> +++ b/arch/arm/Kconfig
> @@ -1560,6 +1560,7 @@ endchoice
>   
>   config PAGE_OFFSET
>   	hex
> +	default PHYS_OFFSET if !MMU
>   	default 0x40000000 if VMSPLIT_1G
>   	default 0x80000000 if VMSPLIT_2G
>   	default 0xC0000000
>
Uwe Kleine-König Nov. 13, 2013, 1:39 p.m. UTC | #2
On Wed, Nov 13, 2013 at 11:46:49AM +0000, Jonathan Austin wrote:
> Hi Ming,
> 
> On 13/11/13 11:15, Ming Lei wrote:
> > From: Jonathan Austin <jonathan.austin@arm.com>
> > 
> > Set CONFIG_PAGE_OFFSET as PHYS_OFFSET when !MMU so that
> > we can keep CONFIG_PAGE_OFFSET same with PAGE_OFFSET.
> >
> 
> I'm not so sure about this commit message - I don't think
> it makes sense...Should it not be more like:
> 
> ------->8-------
> Subject: ARM: set PAGE_OFFSET = PHYS_OFFSET if !MMU
> 
> With !MMU we still set PAGE_OFFSET to 0xC0000000, which
> makes little sense and causes a regression on !MMU after
> f6537f2f0eba4eba ("scripts/kallsyms: filter symbols not
> kernel address space")
> 
> This patch sets PAGE_OFFSET to PHYS_OFFSET in the !MMU
> case, fixing the regression (Tested on Cortex-R7/Vexpress)
> --------8<---------
I like this commit log better, too.

As CONFIG_PAGE_OFFSET is only used in scripts/kallsyms.c and
scripts/link-vmlinux.sh for !MMU.

Booting on efm32 with this patch works just fine, so if you want add my
Tested-by. But I'm unsure how valuable it is as I don't see a breakage
without it.

Best regards
Uwe
Jonathan Austin Dec. 4, 2013, 4:39 p.m. UTC | #3
Hi all,

On 13/11/13 11:46, Jonathan Austin wrote:
> Hi Ming,
>
> On 13/11/13 11:15, Ming Lei wrote:
>> From: Jonathan Austin <jonathan.austin@arm.com>
>>
>> Set CONFIG_PAGE_OFFSET as PHYS_OFFSET when !MMU so that
>> we can keep CONFIG_PAGE_OFFSET same with PAGE_OFFSET.
>>
>
> I'm not so sure about this commit message - I don't think
> it makes sense...Should it not be more like:
>
> ------->8-------
> Subject: ARM: set PAGE_OFFSET = PHYS_OFFSET if !MMU
>
> With !MMU we still set PAGE_OFFSET to 0xC0000000, which
> makes little sense and causes a regression on !MMU after
> f6537f2f0eba4eba ("scripts/kallsyms: filter symbols not
> kernel address space")
>
> This patch sets PAGE_OFFSET to PHYS_OFFSET in the !MMU
> case, fixing the regression (Tested on Cortex-R7/Vexpress)
> --------8<---------
>
> Russell: Will you pick this patch up from the list or does
> it need to go in through the patch system?


What's the status of this patch? Shall I put it in the patch system with 
the commit message above? Or is it going via someone else?

Jonny

>
> Jonny
>
>> This patch and patch 2/2 fix regression on !MMU by
>> f6537f2f0eba4eba("scripts/kallsyms: filter symbols not
>> in kernel address space")
>>
>> Cc: <stable@vger.kernel.org>
>> Signed-off-by: Jonathan Austin <jonathan.austin@arm.com>
>> Signed-off-by: Ming Lei <tom.leiming@gmail.com>
>> ---
>>    arch/arm/Kconfig |    1 +
>>    1 file changed, 1 insertion(+)
>>
>> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
>> index 799ef94..db708c4 100644
>> --- a/arch/arm/Kconfig
>> +++ b/arch/arm/Kconfig
>> @@ -1560,6 +1560,7 @@ endchoice
>>
>>    config PAGE_OFFSET
>>    	hex
>> +	default PHYS_OFFSET if !MMU
>>    	default 0x40000000 if VMSPLIT_1G
>>    	default 0x80000000 if VMSPLIT_2G
>>    	default 0xC0000000
>>
>
>
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
>
diff mbox

Patch

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 799ef94..db708c4 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1560,6 +1560,7 @@  endchoice
 
 config PAGE_OFFSET
 	hex
+	default PHYS_OFFSET if !MMU
 	default 0x40000000 if VMSPLIT_1G
 	default 0x80000000 if VMSPLIT_2G
 	default 0xC0000000