A while ago Mario and I (really!) wrote a small library for communicating
with the TPM, called trunks. When it was almost finished I realized we
could get our task done differently, and never used it. It started out as
a Chromium repo, but I moved it to github for convenience (with the
blessings of our open source gurus):
https://github.com/semenzato/trunks
Its main advantage is size. It's much smaller than trousers, so it's
easier to hack. Also, IIRC, the size of the binary is about 60k and could
be smaller if one needs only a subset of the commands. Compare (again
IIRC) to about 1MB for trousers.
Most of the functionality goes into serializing (marshalling/unmarshalling)
TPM commands and structures. The library encodes the structure layouts
into bytecodes, which the serializing code interprets (compare to Trousers
which uses a C function for each structure). Given the typical speed of a
TPM, the interpretation overhead is negligible.
I am happy to maintain it if anybody is going to send patches.
On Thu, Jun 12, 2014 at 5:37 AM, Johnson, Douglas <[email protected]>
wrote:
>
>
>
>
> I'm working on a project for a small embedded system that will have a TPM
> but will not have an operating system. I'm trying to determine if Trousers
> will work in that environment. The TPM is connected to an FPGA containing
> a 32 bit processor using SPI. I'll write a driver conforming to the TCG PC
> Client-Specific TPM Interface Specification. I don't need TCP/IP access.
> I have flash, so I might have a filesystem. Single threaded operation is
> fine.
>
>
>
> So my questions are:
>
>
>
> Is trousers a reasonable starting point for me?
>
> Can trousers be made to work without an OS?
>
> What OS specific functions does trousers require?
>
> What areas of the code contain OS specific functions?
>
> Are there other software stacks that I should consider? (I have
> downloaded tpmdd as an example driver.)
>
>
>
>
>
> Regards,
>
> Doug Johnson
>
>
> ------------------------------------------------------------------------------
> HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
> Find What Matters Most in Your Big Data with HPCC Systems
> Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
> Leverages Graph Analysis for Fast Processing & Easy Data Exploration
> http://p.sf.net/sfu/hpccsystems
> _______________________________________________
> TrouSerS-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/trousers-users
>
>
------------------------------------------------------------------------------
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration
http://p.sf.net/sfu/hpccsystems
_______________________________________________
TrouSerS-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/trousers-users
