Re: [PATCH v3 4/8] tpm: Split TPM_TIS_IRQ into TPM_TIS_ISA_IRQ and TPM_TIS_SYSBUS_IRQ

[Stefan Berger <stefanb@linux.ibm.com>]

On 6/17/20 4:12 AM, Auger Eric wrote:
> Hi Stefan,
>
> On 6/16/20 10:57 PM, Stefan Berger wrote:
>> Before the enablement of interrupts on PCs, split the TPM_TIS_IRQ
>> into TPM_TIS_ISA_IRQ for PCs and TPM_TIS_SYSBUS_IRQ for ARM.
>>
>> Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
>> ---
>>   hw/tpm/tpm_tis_isa.c    | 2 +-
>>   hw/tpm/tpm_tis_sysbus.c | 3 ++-
>>   include/hw/acpi/tpm.h   | 3 ++-
>>   3 files changed, 5 insertions(+), 3 deletions(-)
>>
>> diff --git a/hw/tpm/tpm_tis_isa.c b/hw/tpm/tpm_tis_isa.c
>> index ed6d422f05..27222a9a49 100644
>> --- a/hw/tpm/tpm_tis_isa.c
>> +++ b/hw/tpm/tpm_tis_isa.c
>> @@ -97,7 +97,7 @@ static void tpm_tis_isa_reset(DeviceState *dev)
>>   }
>>   
>>   static Property tpm_tis_isa_properties[] = {
>> -    DEFINE_PROP_UINT32("irq", TPMStateISA, state.irq_num, TPM_TIS_IRQ),
>> +    DEFINE_PROP_UINT32("irq", TPMStateISA, state.irq_num, TPM_TIS_ISA_IRQ),
> irq_num is a uint32 prop while the default value we are going to assign
> is -1. I guess we will end up with irq_num = 255?

We can make it ~0, if that's better. Otherwise it does seem to become 
0xff ff ff ff with the current -1.


    Stefan