Re: [PATCH 3/6] ima-evm-utils: When using the IBM TSS, link in its library

[Ken Goldman <kgold@linux.ibm.com>]

On 10/6/2020 11:07 AM, Petr Vorel wrote:
> Hi Ken,
> 
>> This is a prerequisite for the code change that uses the TSS rather
>> than the command line tools.
> 
>> Signed-off-by: Ken Goldman <kgoldman@us.ibm.com>
>> ---
>>   src/Makefile.am | 1 +
>>   1 file changed, 1 insertion(+)
> 
>> diff --git a/src/Makefile.am b/src/Makefile.am
>> index d6c779f..bf18caf 100644
>> --- a/src/Makefile.am
>> +++ b/src/Makefile.am
>> @@ -26,6 +26,7 @@ if USE_PCRTSS
>>   evmctl_SOURCES += pcr_tss.c
>>   else
>>   evmctl_SOURCES += pcr_tsspcrread.c
>> +evmctl_LDADD += -libmtss
>>   endif
> But pcr_tsspcrread.c uses the binary (tsspcrread).

The idea is to eliminate the calls to command line utilities, which
are not as stable as the TSS library.  Patch 4/6 does that.

I'm a git newbie.  I thought the idea was to do the patches as
small pieces.  Thus, this one does the autotools piece.  Then
the next one does the C code.  Should they be combined?

> 
> pcr_tss.c uses TSS, but that's already covered:
> ldd src/.libs/evmctl |grep tss
> 	libtss2-rc.so.0 => /usr/lib64/libtss2-rc.so.0 (0x00007fb82514c000)
> 	libtss2-esys.so.0 => /usr/lib64/libtss2-esys.so.0 (0x00007fb824eba000)
> 	libtss2-sys.so.0 => /usr/lib64/libtss2-sys.so.0 (0x00007fb8244ba000)
> 	libtss2-mu.so.0 => /usr/lib64/libtss2-mu.so.0 (0x00007fb824276000)
> 
> Thus I thing this patch is invalid. Or do I miss something obvious?

There are two TSSes.  This one's for the IBM TSS.  Your traces are
for the ESAPI API (aka Intel TSS).

> 
> And other thing: could you please rebase your patches for next-testing branch?
> It's not applicable for master, next and next-testing.
> 
> Kind regards,
> Petr
>