On Fri, Feb 16, 2018 at 11:31 AM Ard Biesheuvel <ard.biesheu...@linaro.org> wrote: > This is why I was leaning towards applying these patches: not breaking > userland is an important rule, but it does not imply every aspect of > behavior observable by userland is set in stone. In other words, I > agree with Peter that making this change does not *break* userland in > a way anyone is likely to care deeply about.
In some modes tpmtotp will run as non-root and expect to be able to read an EFI variable.
