Paul Howard

Thank you, Ken. I fully appreciate that the simulator is not designed for a physical power outage - this was my initial assessment as well, but then I saw the comments about disorderly shutdown in the code, which made me less sure. But I think your response has cleared that up for me. If. what we are seeing is the expected behaviour for the circumstances described, then I'm happy to leave it at that. Thanks for the prompt response. All the best, Paul.

Quick follow up, the version of the simulator in this case is: https://sourceforge.net/projects/ibmswtpm2/files/ibmtpm1637.tar.gz Just in case that's significant! Paul.

Hi. I'm investigating a situation where the TPM simulator rejects a tpm2_startup command if it is issued after the previous simulator session was halted abruptly, such as due to a system power loss. The software stack is TSS and TPM tools from https://github.com/tpm2-software This has been observed on multiple Linux systems, including a Yocto image and also Ubuntu 18.04. Our observation is that the tpm2_shutdown command appears to write state back to the NVChip file that is essential for a successful...