Peter Pietzuch
Peter Pietzuch
@Pengpeng-Microsoft can you drive this? Thanks.
Why is this P1? This should be P2 or P3.
It makes sense that you see those host-side threads, as the segfault happened during the host side termination sequence. This should be fairly easy to fix, as I suspect that,...
I've updated the original issue per our discussion today. Please add other thoughts/suggestions, thanks.
> I am not clear about this statement: > > > enforces attestation, so it becomes impossible to run an enclave application locally for testing; > > By `locally`, you...
> Symlinking `/opt/sgx-lkl` to `/opt/sgx-lkl-release` will be problematic for the case when the former prefix would be mounted into a container. I'm pretty sure you wouldn't be able to run...
> Well, if you have `/opt/sgx-lkl/debug/`, how can you have a symlink at `/opt/sgx-lkl`? Or did you mean to use dashes? If so, then it's fine of course. Right, the...
> Who would use the symlink? It’s a default for people who are confused by the build types? :-) I guess we don’t need the symlink.
@letmaik to me, a RELEASE build is not about what attestation enforces but rather about removing unnecessary code and functionality from the enclave, and thus getting a smaller TCB. By...
@wintersteiger @letmaik @jxyang can we also remove the invocation of oe_get_report here: https://github.com/lsds/sgx-lkl/blob/4419f27c4859bcec6abfa936963b1ff10c2aa975/src/enclave/enclave_init.c#L160 This not only breaks on SGX1 machines without FLC but also on SGX2 machines. When not running...