Re: [RFC PATCH V1 1/3] acpi: Eliminate all TPM related code if CONFIG_TPM is not set

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

On 6/14/21 5:53 AM, Philippe Mathieu-Daudé wrote:
> On 6/12/21 3:21 AM, Stefan Berger wrote:
>> Cc: M: Michael S. Tsirkin <mst@redhat.com>
>> Cc: Igor Mammedov <imammedo@redhat.com>
>> Signed-off-by: Stefan Berger <stefanb@linux.ibm.com>
>> ---
>>   hw/acpi/aml-build.c      |  2 ++
>>   hw/arm/virt-acpi-build.c |  2 ++
>>   hw/i386/acpi-build.c     | 20 ++++++++++++++++++++
>>   include/hw/acpi/tpm.h    |  4 ++++
>>   stubs/tpm.c              |  4 ----
>>   5 files changed, 28 insertions(+), 4 deletions(-)
>>
>> diff --git a/hw/acpi/aml-build.c b/hw/acpi/aml-build.c
>> index f0035d2b4a..d5103e6d7b 100644
>> --- a/hw/acpi/aml-build.c
>> +++ b/hw/acpi/aml-build.c
>> @@ -2044,6 +2044,7 @@ build_hdr:
>>                    "FACP", tbl->len - fadt_start, f->rev, oem_id, oem_table_id);
>>   }
>>   
>> +#ifdef CONFIG_TPM
>>   /*
>>    * build_tpm2 - Build the TPM2 table as specified in
>>    * table 7: TCG Hardware Interface Description Table Format for TPM 2.0
>> @@ -2101,6 +2102,7 @@ void build_tpm2(GArray *table_data, BIOSLinker *linker, GArray *tcpalog,
>>                    (void *)(table_data->data + tpm2_start),
>>                    "TPM2", table_data->len - tpm2_start, 4, oem_id, oem_table_id);
>>   }
>> +#endif
> This makes the ARM virt machine build to fail for missing
> the build_tpm2() symbol.
>
> You probably need to split the patch in 2 and rearrange the
> series:
>
> 1/ hw/i386    (current 1)
> 2/ hw/arm     (current 2)
> 3/ hw/acpi    (current 1)
> 4/ sysemu/tpm (current 3)


I took a bottom-up approach where I also eliminated all callers. So from 
what I can see build_tpm2() symbol has been eliminated by both i386 and 
aarch64, but ok, I will change it to a top-down approach.