[U-Boot] [PATCH v5 4/7] arm, arm926ejs: Do not clear the V bit on DA850 SoCs

[Heiko Schocher]

Hello Sughosh,

Sughosh Ganu wrote:
> hi Christian,
> 
> On Tue, Jan 31, 2012 at 7:26 PM, Christian Riesch <
> christian.riesch at omicron.at> wrote:
> 
>> The V bit of the c1 register of CP15 should not be cleared
>> since the SoC has no valid memory at 0x00000000.
>>
>> Signed-off-by: Christian Riesch <christian.riesch at omicron.at>
>> Reported-by: Sughosh Ganu <urwithsughosh at gmail.com>
>> Cc: Albert Aribaud <albert.u.boot at aribaud.net>
>> Cc: Tom Rini <trini at ti.com>
>> ---
>>  arch/arm/cpu/arm926ejs/start.S |    5 ++++-
>>  1 files changed, 4 insertions(+), 1 deletions(-)
>>
>> diff --git a/arch/arm/cpu/arm926ejs/start.S
>> b/arch/arm/cpu/arm926ejs/start.S
>> index b39ed8a..b350480 100644
>> --- a/arch/arm/cpu/arm926ejs/start.S
>> +++ b/arch/arm/cpu/arm926ejs/start.S
>> @@ -372,7 +372,10 @@ flush_dcache:
>>         * disable MMU and D cache, and enable I cache
>>         */
>>        mrc     p15, 0, r0, c1, c0, 0
>> -       bic     r0, r0, #0x00002300     /* clear bits 13, 9:8 (--V- --RS)
>> */
>> +       bic     r0, r0, #0x00000300     /* clear bits 9:8 (---- --RS) */
>> +#ifndef CONFIG_SOC_DA850
>> +       bic     r0, r0, #0x00002000     /* clear bit 13 (--V- ----) */
>> +#endif
>>
> 
> Instead of checking for a particular SOC, can we introduce a generic
> config, something like CONFIG_EXCEPTION_VECTORS_LOW. This way, if other
> SOC's have a similar requirement, it won't be needed to keep adding checks
> here. It would also help in case this needs to be implemented for other arm
> cores, so that we can have a common config option for bypassing this V-bit
> clear. Just my suggestion. Maybe Tom and Albert can comment.

Yep, I vote for this too, also this config option should be documented
in the README.

> I will test your patch series on my hawkboard and let you know the results
> tomorrow. Thanks for clubbing it all together.

Yep, thanks!

Currently testing it on the enbw_cmc board, looks good. I test the
patches also on the cam_enc_4xx board ... and post the results.

bye,
Heiko
-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany