[U-Boot] [PATCH] arm: bugfix: Move vector table before jumping relocated code

[Tetsuyuki Kobayashi <koba@kmckk.co.jp>]

Hi Tom and all,

(2012/06/28 3:23), Tom Rini wrote:
> On Mon, Jun 25, 2012 at 09:40:57PM +0900, Tetsuyuki Kobayashi wrote:
>
>> Interrupts and exceptions doesn't work in relocated code.
>> It badly use IRQ_STACK_START_IN in rom area as interrupt stack.
>> It is because the vecotr table is not moved to ram area.
>> This patch moves vector table before jumping relocated code.
>>
>> Signed-off-by: Tetsuyuki Kobayashi<koba@kmckk.co.jp>
>> ---
>>   arch/arm/cpu/armv7/start.S |   12 ++++++++++++
>>   1 file changed, 12 insertions(+)
>>
>> diff --git a/arch/arm/cpu/armv7/start.S b/arch/arm/cpu/armv7/start.S
>> index 52f7f6e..5098f7b 100644
>> --- a/arch/arm/cpu/armv7/start.S
>> +++ b/arch/arm/cpu/armv7/start.S
>> @@ -277,6 +277,18 @@ jump_2_ram:
>>   	mcr     p15, 0, r0, c7, c10, 4	@ DSB
>>   	mcr     p15, 0, r0, c7, c5, 4	@ ISB
>>   #endif
>> +/*
>> + * Move vector table
>> + */
>> +#if !defined(CONFIG_TEGRA2)
>> +#if !(defined(CONFIG_OMAP44XX)&&  defined(CONFIG_SPL_BUILD))
>> +	/* Set vector address in CP15 VBAR register */
>> +	ldr     r0, =_start
>> +	add     r0, r0, r9
>> +	mcr     p15, 0, r0, c12, c0, 0  @Set VBAR
>> +#endif
>> +#endif /* !Tegra2 */
>> +	
>>   	ldr	r0, _board_init_r_ofs
>>   	adr	r1, _start
>>   	add	lr, r0, r1
>
> I think this code should get boot tested on a few platforms to make sure
> it's OK.  As such, I've tried on am335x and omap3 and they're still
> fine.
>
> Tested-by: Tom Rini<trini@ti.com>
>
Tom, thank you for testing.

Easy way to test this patch.
Jump any bad address using go command on U-boot prompt.

Good restult:

 > go 0x0badc0de
## Starting application at 0x0BADC0DE ...
undefined instruction
pc : [<0badc0e0>]          lr : [<5ff91354>]
sp : 5feefb98  ip : 0000001c     fp : 5feefbdd
r10: 00000002  r9 : 5feefbdd     r8 : 5feeff68
r7 : 5feeff14  r6 : 0badc0de     r5 : 5feefed4  r4 : 00000002
r3 : 0badc0de  r2 : 5feefed4     r1 : 5feefed4  r0 : 00000001
Flags: nZCv  IRQs off  FIQs off  Mode SVC_32
Resetting CPU ...

resetting ...


Bad result:
 > go 0x0badc0de

(.. hang up without any crash dump.)