On 9 September 2014 11:14, Peter Maydell <peter.mayd...@linaro.org> wrote:
> On 5 September 2014 16:15, Ard Biesheuvel <ard.biesheu...@linaro.org> wrote:
>> When booting with -bios or -pflash rather than -kernel, we need to make sure
>> reset handlers are registered.
>>
>> Signed-off-by: Ard Biesheuvel <ard.biesheu...@linaro.org>
>> ---
>> hw/arm/boot.c | 4 ++++
>> 1 file changed, 4 insertions(+)
>>
>> diff --git a/hw/arm/boot.c b/hw/arm/boot.c
>> index 8f5649a250fd..0cfc11d42962 100644
>> --- a/hw/arm/boot.c
>> +++ b/hw/arm/boot.c
>> @@ -473,6 +473,10 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info
>> *info)
>> }
>> }
>>
>> + for (; cs; cs = CPU_NEXT(cs)) {
>> + qemu_register_reset(do_cpu_reset, ARM_CPU(cs));
>> + }
>> +
>> /* If no kernel specified, do nothing; we will start from address 0
>> * (typically a boot ROM image) in the same way as hardware.
>> */
>
> Andreas, with QOM CPUs who is supposed to be responsible
> for causing them to be reset when qemu_system_reset()
> happens? At the moment for ARM it doesn't happen at all
> unless you happened to pass -kernel. This patch makes
> boot.c register the reset handlers, but that seems
> a bit odd to me (among other things, it won't work
> for v7M).
>
Has there been any discussion on this topic in the mean time?
Cheers,
Ard.
