Hi Ilias, On Mon, Oct 19, 2026 at 10:47 AM Ilias Apalodimas <[email protected]> wrote: > > [...] > > > --- > > drivers/tee/optee/optee_private.h | 2 ++ > > drivers/tee/optee/rpmb.c | 11 +++++++++++ > > drivers/tee/optee/supplicant.c | 4 ++++ > > 3 files changed, 17 insertions(+) > > > > diff --git a/drivers/tee/optee/optee_private.h > > b/drivers/tee/optee/optee_private.h > > index daff3e04493..92e3b3bea10 100644 > > --- a/drivers/tee/optee/optee_private.h > > +++ b/drivers/tee/optee/optee_private.h > > @@ -42,6 +42,8 @@ struct optee_private { > > struct mmc *rpmb_mmc; > > int rpmb_dev_id; > > int rpmb_original_part; > > + /* Non-zero while a supplicant RPC from OP-TEE is being serviced */ > > + int in_supplicant; > > }; > > sandbox uses a different private (drivers/tee/sandbox.c). So you are > overwriting ta_avb_rollback_indexes in that code. oops, nice catch!
That is a v2 regression: v1 kept the counter in handle_rpc(), which only the OP-TEE driver reaches, and I moved it into optee_suppl_cmd() following Jan's suggestion without noticing that sandbox.c calls optee_suppl_cmd() too. Will fix in v3 > > [...] > > > Thanks > /Ilias
