| 1 |
.. SPDX-License-Identifier: GPL-2.0 |
| 2 |
|
| 3 |
============= |
| 4 |
TPM Event Log |
| 5 |
============= |
| 6 |
|
| 7 |
This document briefly describes what TPM log is and how it is handed |
| 8 |
over from the preboot firmware to the operating system. |
| 9 |
|
| 10 |
Introduction |
| 11 |
============ |
| 12 |
|
| 13 |
The preboot firmware maintains an event log that gets new entries every |
| 14 |
time something gets hashed by it to any of the PCR registers. The events |
| 15 |
are segregated by their type and contain the value of the hashed PCR |
| 16 |
register. Typically, the preboot firmware will hash the components to |
| 17 |
who execution is to be handed over or actions relevant to the boot |
| 18 |
process. |
| 19 |
|
| 20 |
The main application for this is remote attestation and the reason why |
| 21 |
it is useful is nicely put in the very first section of [1]: |
| 22 |
|
| 23 |
"Attestation is used to provide information about the platform’s state |
| 24 |
to a challenger. However, PCR contents are difficult to interpret; |
| 25 |
therefore, attestation is typically more useful when the PCR contents |
| 26 |
are accompanied by a measurement log. While not trusted on their own, |
| 27 |
the measurement log contains a richer set of information than do the PCR |
| 28 |
contents. The PCR contents are used to provide the validation of the |
| 29 |
measurement log." |
| 30 |
|
| 31 |
UEFI event log |
| 32 |
============== |
| 33 |
|
| 34 |
UEFI provided event log has a few somewhat weird quirks. |
| 35 |
|
| 36 |
Before calling ExitBootServices() Linux EFI stub copies the event log to |
| 37 |
a custom configuration table defined by the stub itself. Unfortunately, |
| 38 |
the events generated by ExitBootServices() don't end up in the table. |
| 39 |
|
| 40 |
The firmware provides so called final events configuration table to sort |
| 41 |
out this issue. Events gets mirrored to this table after the first time |
| 42 |
EFI_TCG2_PROTOCOL.GetEventLog() gets called. |
| 43 |
|
| 44 |
This introduces another problem: nothing guarantees that it is not called |
| 45 |
before the Linux EFI stub gets to run. Thus, it needs to calculate and save the |
| 46 |
final events table size while the stub is still running to the custom |
| 47 |
configuration table so that the TPM driver can later on skip these events when |
| 48 |
concatenating two halves of the event log from the custom configuration table |
| 49 |
and the final events table. |
| 50 |
|
| 51 |
References |
| 52 |
========== |
| 53 |
|
| 54 |
- [1] https://trustedcomputinggroup.org/resource/pc-client-specific-platform-firmware-profile-specification/ |
| 55 |
- [2] The final concatenation is done in drivers/char/tpm/eventlog/efi.c |