Hi, On Fri, 1 Dec 2023 at 07:52, Ilias Apalodimas <ilias.apalodi...@linaro.org> wrote: > > Hi Sean, > > On Tue, Sep 12, 2023 at 02:47:25AM -0700, seanedm...@linux.microsoft.com > wrote: > > From: Stephen Carlson <stcar...@linux.microsoft.com> > > > > This implementation of the rollback uclass driver allows existing TPM2 > > devices declared in the device tree to be referenced for storing the OS > > anti-rollback counter, using the TPM2 non-volatile storage API. The > > rollback device must be a child of the TPM device. For example: > > > > tpm2 { > > compatible = "sandbox,tpm2"; > > > > rollback@1 { > > compatible = "tpm,rollback"; > > rollback-nv-index = <0x1001007>; > > }; > > }; > > > This node is part of the DT specification right? If we accept this, we > should figure out if we can add that to the specification.
For now I suggest adding a binding file to U-Boot and sending it upstream. We still seem to be having extreme difficulty actually getting things accepted upstream. The logic of this driver needs a few changes, I believe: - probe() should do nothing, to avoid probe failing due to bad counter, etc. - we should have a startup() method to actually init the counter - if that fails, we should have a setup() method to set up a new counter That way the caller is in charge of things. Of course you can put helper functions in the uclass. Regards, Simon