On Wed, Oct 8, 2014 at 1:03 AM, Martin Galvan
<martin.gal...@tallertechnologies.com> wrote:
> On Tue, Oct 7, 2014 at 11:13 AM, Alistair Francis <alistai...@gmail.com>
> wrote:
>> The Netduino 2 machine won't run unless the reset_pc is based
>> on the ELF entry point.
>>
>> Signed-off-by: Alistair Francis <alistai...@gmail.com>
>> Signed-off-by: Peter Crosthwaite <crosthwaite.pe...@gmail.com>
>> ---
>> V2:
>> - Malloc straight away, thanks to Peter C
>>
>> hw/arm/armv7m.c | 19 ++++++++++++++++---
>> 1 file changed, 16 insertions(+), 3 deletions(-)
>>
>> diff --git a/hw/arm/armv7m.c b/hw/arm/armv7m.c
>> index 7169027..07b36e2 100644
>> --- a/hw/arm/armv7m.c
>> +++ b/hw/arm/armv7m.c
>> @@ -155,11 +155,19 @@ static void armv7m_bitband_init(void)
>>
>> /* Board init. */
>>
>> +typedef struct ARMV7MResetArgs {
>> + ARMCPU *cpu;
>> + uint32_t reset_pc;
>> +} ARMV7MResetArgs;
>> +
>> static void armv7m_reset(void *opaque)
>> {
>> - ARMCPU *cpu = opaque;
>> + ARMV7MResetArgs *args = opaque;
>> +
>> + cpu_reset(CPU(args->cpu));
>>
>> - cpu_reset(CPU(cpu));
>> + args->cpu->env.thumb = args->reset_pc & 1;
>> + args->cpu->env.regs[15] = args->reset_pc & ~1;
>> }
>>
>> /* Init CPU and memory for a v7-M based board.
>> @@ -180,6 +188,7 @@ qemu_irq *armv7m_init(MemoryRegion *system_memory, int
>> mem_size, int num_irq,
>> int i;
>> int big_endian;
>> MemoryRegion *hack = g_new(MemoryRegion, 1);
>> + ARMV7MResetArgs *reset_args = g_new0(ARMV7MResetArgs, 1);
>>
>> if (cpu_model == NULL) {
>> cpu_model = "cortex-m3";
>> @@ -234,7 +243,11 @@ qemu_irq *armv7m_init(MemoryRegion *system_memory, int
>> mem_size, int num_irq,
>> vmstate_register_ram_global(hack);
>> memory_region_add_subregion(system_memory, 0xfffff000, hack);
>>
>> - qemu_register_reset(armv7m_reset, cpu);
>> + *reset_args = (ARMV7MResetArgs) {
>> + .cpu = cpu,
>> + .reset_pc = entry,
>> + };
>> + qemu_register_reset(armv7m_reset, reset_args);
>> return pic;
>> }
>
> How does this differ from what's being done in arm_cpu_reset for
> ARMv7-M? What about the initial MSP?
So the problem I was having is that the standard linker meant that the
reset_handler wasn't being called. I have made some changes to the
linker file so now it calls the reset_handler. This patch is no longer
required. Sorry it took me so long to figure out what the problem was
Thanks,
Alistair
>
> --
>
> Martín Galván
>
> Software Engineer
>
> Taller Technologies Argentina
>
>
> San Lorenzo 47, 3rd Floor, Office 5
>
> Córdoba, Argentina
>
> Phone: 54 351 4217888 / +54 351 4218211
