Re: [PATCH v2 5/5] optee: implement the RPMB subsystem interface for UFS

[Jorge Ramirez via U-Boot <u-boot@lists.u-boot-project.org>]

On 23/07/26 09:33:39, Neil Armstrong wrote:
> On 7/22/26 20:44, Jorge Ramirez wrote:
> > On 22/07/26 20:12:19, Jorge Ramirez wrote:
> > > On 22/07/26 17:43:16, Neil Armstrong wrote:
> > > > On 7/22/26 16:12, Jorge Ramirez wrote:
> > > > > On 22/07/26 10:31:04, neil.armstrong@linaro.org wrote:
> > > > > > Hi,
> > > > > > 
> > > > > > On 7/22/26 08:07, Jorge Ramirez-Ortiz wrote:
> > > > > > > OP-TEE's legacy RPMB supplicant interface assumes the RPMB device is eMMC
> > > > > > > and derives the key from an eMMC-shaped device ID, so it can never
> > > > > > > reproduce the key Linux derives for a UFS device; secure storage on
> > > > > > > UFS-only platforms instead needs the transport-agnostic RPMB subsystem
> > > > > > > interface, through which the normal world describes the real RPMB device
> > > > > > > to OP-TEE. Add a UFS backend for it so OP-TEE derives a key matching Linux
> > > > > > > and can use RPMB secure storage on UFS, leaving the legacy eMMC path
> > > > > > > untouched.
> > > > > > > 
> > > > > > > This subsystem backend currently supports UFS drives only; eMMC still uses
> > > > > > > the legacy interface. As the legacy interface is dropped in favour of the
> > > > > > > subsystem one, an eMMC backend should be added here so both transports are
> > > > > > > served through a single RPMB path.
> > > > > > > 
> > > > > > > Signed-off-by: Jorge Ramirez-Ortiz <jorge.ramirez@oss.qualcomm.com>
> > > > > > > ---
> > > > > > >     drivers/tee/optee/Makefile               |   1 +
> > > > > > >     drivers/tee/optee/optee_msg_supplicant.h |   8 ++
> > > > > > >     drivers/tee/optee/optee_private.h        |  41 ++++++++
> > > > > > >     drivers/tee/optee/rpmb.c                 | 124 +++++++++++++++++++++++
> > > > > > 
> > > > > > Why not rpmb_ufs ?
> > > > > 
> > > > > 
> > > > > because in the future (and IMO) rpmb.c should support UFS and EMMC using
> > > > > the new probe interface (eventually removing rpmb_legacy).
> > > > 
> > > > Let's see in the future, if somehow there's need for a common "new"
> > > > and emmc only legacy some rework will be needed anyway. The eMMC RPMB
> > > > hasn't been updated for a very long time...
> > > > 
> > > > Using the driver model to register an rpmb device attached to an eMMC
> > > > or UFS controller would the cleanest way to handle that, but we're not here now.
> > > > 
> > > > > 
> > > > > > 
> > > > > > It's fine to have 2 backend implementations, the emmc is implementing the
> > > > > > legacy optee API, just add a comment in the eMMC file about that.
> > > > > > 
> > > > > > Neil
> > > > > > 
> > > > > > >     drivers/tee/optee/supplicant.c           |   9 ++
> > > > > > >     5 files changed, 183 insertions(+)
> > > > > > >     create mode 100644 drivers/tee/optee/rpmb.c
> > > > > > > 
> > > > > > > diff --git a/drivers/tee/optee/Makefile b/drivers/tee/optee/Makefile
> > > > > > > index 8321cf53a19..7af19834c00 100644
> > > > > > > --- a/drivers/tee/optee/Makefile
> > > > > > > +++ b/drivers/tee/optee/Makefile
> > > > > > > @@ -4,3 +4,4 @@ obj-$(CONFIG_OPTEE) += core.o
> > > > > > >     obj-y += supplicant.o
> > > > > > >     obj-$(CONFIG_DM_I2C) += i2c.o
> > > > > > >     obj-$(CONFIG_SUPPORT_EMMC_RPMB) += rpmb_legacy.o
> > > > > > > +obj-$(CONFIG_SUPPORT_UFS_RPMB) += rpmb.o
> > > > > > > diff --git a/drivers/tee/optee/optee_msg_supplicant.h b/drivers/tee/optee/optee_msg_supplicant.h
> > > > > > > index 963cfd47824..b720d5d7b3f 100644
> > > > > > > --- a/drivers/tee/optee/optee_msg_supplicant.h
> > > > > > > +++ b/drivers/tee/optee/optee_msg_supplicant.h
> > > > > > > @@ -152,6 +152,14 @@
> > > > > > >      */
> > > > > > >     #define OPTEE_MSG_RPC_CMD_I2C_TRANSFER 21
> > > > > > > +#define OPTEE_MSG_RPC_CMD_RPMB_PROBE_RESET	22
> > > > > > > +#define OPTEE_MSG_RPC_CMD_RPMB_PROBE_NEXT	23
> > > > > > > +#define OPTEE_MSG_RPC_CMD_RPMB_FRAMES		24
> > > > > > > +
> > > > > > > +#define OPTEE_RPC_SHM_TYPE_APPL		0
> > > > > > > +
> > > > > > > +#define OPTEE_RPC_RPMB_UFS		1
> > > > > > > +
> > > > > > >     /*
> > > > > > >      * Was OPTEE_MSG_RPC_CMD_SQL_FS, which isn't supported any longer
> > > > > > >      */
> > > > > > > diff --git a/drivers/tee/optee/optee_private.h b/drivers/tee/optee/optee_private.h
> > > > > > > index 1f07a27ee4b..2888257aefe 100644
> > > > > > > --- a/drivers/tee/optee/optee_private.h
> > > > > > > +++ b/drivers/tee/optee/optee_private.h
> > > > > > > @@ -9,6 +9,17 @@
> > > > > > >     #include <tee.h>
> > > > > > >     #include <log.h>
> > > > > > > +#ifdef CONFIG_SUPPORT_UFS_RPMB
> > > > > > > +/**
> > > > > > > + * struct optee_private - OP-TEE driver private data
> > > > > > > + * @rpmb_next_region:	next UFS RPMB region to report on PROBE_NEXT
> > > > > > > + * @rpmb_cur_region:	UFS RPMB region selected by the last PROBE_NEXT
> > > > > > > + */
> > > > > > > +struct optee_private {
> > > > > > > +	unsigned int rpmb_next_region;
> > > > > > > +	unsigned int rpmb_cur_region;
> > > > > > > +};
> > > > > > > +#else
> > > > > > >     /**
> > > > > > >      * struct optee_private - OP-TEE driver private data
> > > > > > >      * @rpmb_mmc:		mmc device for the RPMB partition
> > > > > > > @@ -22,6 +33,7 @@ struct optee_private {
> > > > > > >     	int rpmb_dev_id;
> > > > > > >     	int rpmb_original_part;
> > > > > > >     };
> > > > > > > +#endif
> > > > > > >     struct optee_msg_arg;
> > > > > > > @@ -60,6 +72,35 @@ static inline void optee_suppl_rpmb_release(struct udevice *dev)
> > > > > > >     }
> > > > > > >     #endif
> > > > > > > +#ifdef CONFIG_SUPPORT_UFS_RPMB
> > > > > > > +void optee_suppl_cmd_rpmb_probe_reset(struct udevice *dev,
> > > > > > > +				      struct optee_msg_arg *arg);
> > > > > > > +
> > > > > > > +void optee_suppl_cmd_rpmb_probe_next(struct udevice *dev,
> > > > > > > +				     struct optee_msg_arg *arg);
> > > > > > > +
> > > > > > > +void optee_suppl_cmd_rpmb_frames(struct udevice *dev,
> > > > > > > +				 struct optee_msg_arg *arg);
> > > > > > > +#else
> > > > > > > +static inline void optee_suppl_cmd_rpmb_probe_reset(struct udevice *dev,
> > > > > > > +						    struct optee_msg_arg *arg)
> > > > > > > +{
> > > > > > > +	arg->ret = TEE_ERROR_NOT_IMPLEMENTED;
> > > > > > > +}
> > > > > > > +
> > > > > > > +static inline void optee_suppl_cmd_rpmb_probe_next(struct udevice *dev,
> > > > > > > +						   struct optee_msg_arg *arg)
> > > > > > > +{
> > > > > > > +	arg->ret = TEE_ERROR_NOT_IMPLEMENTED;
> > > > > > > +}
> > > > > > > +
> > > > > > > +static inline void optee_suppl_cmd_rpmb_frames(struct udevice *dev,
> > > > > > > +					       struct optee_msg_arg *arg)
> > > > > > > +{
> > > > > > > +	arg->ret = TEE_ERROR_NOT_IMPLEMENTED;
> > > > > > > +}
> > > > > > > +#endif
> > > > > > > +
> > > > > > >     #ifdef CONFIG_DM_I2C
> > > > > > >     /**
> > > > > > >      * optee_suppl_cmd_i2c_transfer() - route I2C requests to an I2C chip
> > > > > > > diff --git a/drivers/tee/optee/rpmb.c b/drivers/tee/optee/rpmb.c
> > > > > > > new file mode 100644
> > > > > > > index 00000000000..9c3e0f031ad
> > > > > > > --- /dev/null
> > > > > > > +++ b/drivers/tee/optee/rpmb.c
> > > > > > > @@ -0,0 +1,124 @@
> > > > > > > +// SPDX-License-Identifier: GPL-2.0+
> > > > > > > +
> > > > > > > +#include <dm.h>
> > > > > > > +#include <tee.h>
> > > > > > > +#include <ufs.h>
> > > > > > > +
> > > > > > > +#include "optee_msg.h"
> > > > > > > +#include "optee_msg_supplicant.h"
> > > > > > > +#include "optee_private.h"
> > > > > > > +
> > > > > > > +#define UFS_RPMB_CONTROLLER	0
> > > > > > 
> > > > > > Can this be a config option instead which defaults to 0 ?
> > > > > 
> > > > > um, I dont think so - this assumes there is only one controller. If there
> > > > > are more than one, we would need a mechanism to select which one from
> > > > > the enumerated values (maybe a DT alias (ie ufs-rpmb ?) I think the binding order
> > > > > changes as we enable/disable devices)
> > > > 
> > > > It could collide with other SCSI devices like SATA devices over PCIe.
> > > > 
> > > > Usually we try not to hardcode the devices ids.
> > > 
> > > 
> > > right...shall I go ahead with a DT alias approach?
> > 
> > 
> > something like this?
> > 
> > 
> >    8 ⋮  8 │ #include "optee_msg_supplicant.h"
> >    9 ⋮  9 │ #include "optee_private.h"
> >   10 ⋮ 10 │
> >   11 ⋮    │-#define UFS_RPMB_CONTROLLER        0
> >      ⋮ 11 │+static int optee_rpmb_get_scsi_dev(struct udevice **scsi_devp)
> >      ⋮ 12 │+{
> >      ⋮ 13 │+        struct udevice *ufs_dev, *scsi_dev;
> >      ⋮ 14 │+        ofnode node;
> >      ⋮ 15 │+        int ret;
> >      ⋮ 16 │+
> >      ⋮ 17 │+        node = ofnode_get_aliases_node("ufs-rpmb");
> >      ⋮ 18 │+        if (ofnode_valid(node))
> >      ⋮ 19 │+                ret = uclass_get_device_by_ofnode(UCLASS_UFS, node, &ufs_dev);
> >      ⋮ 20 │+        else
> >      ⋮ 21 │+                ret = uclass_get_device(UCLASS_UFS, 0, &ufs_dev);
> >      ⋮ 22 │+        if (ret)
> >      ⋮ 23 │+                return ret;
> >      ⋮ 24 │+
> >      ⋮ 25 │+        ret = device_get_child(ufs_dev, 0, &scsi_dev);
> >      ⋮ 26 │+        if (ret)
> >      ⋮ 27 │+                return ret;
> >      ⋮ 28 │+
> >      ⋮ 29 │+        *scsi_devp = scsi_dev;
> >      ⋮ 30 │+
> >      ⋮ 31 │+        return 0;
> >      ⋮ 32 │+}
> >   12 ⋮ 33 │
> >   13 ⋮ 34 │ void optee_suppl_cmd_rpmb_probe_reset(struct udevice *dev,
> >   14 ⋮ 35 │                                       struct optee_msg_arg *arg)
> > 
> > ────────────────────────────────────────────────────────────────┐
> > • 76: void optee_suppl_cmd_rpmb_probe_next(struct udevice *dev, │
> > ────────────────────────────────────────────────────────────────┘
> >   55 ⋮ 76 │                 return;
> >   56 ⋮ 77 │         }
> >   57 ⋮ 78 │
> >   58 ⋮    │-        if (uclass_get_device(UCLASS_SCSI, UFS_RPMB_CONTROLLER, &scsi_dev)) {
> >      ⋮ 79 │+        if (optee_rpmb_get_scsi_dev(&scsi_dev)) {
> >   59 ⋮ 80 │                 arg->ret = TEE_ERROR_ITEM_NOT_FOUND;
> >   60 ⋮ 81 │                 return;
> >   61 ⋮ 82 │         }
> > 
> > ─────────────────────────────────────────────────────────────┐
> > • 124: void optee_suppl_cmd_rpmb_frames(struct udevice *dev, │
> > ─────────────────────────────────────────────────────────────┘
> >   103⋮ 124│                 return;
> >   104⋮ 125│         }
> >   105⋮ 126│
> >   106⋮    │-        if (uclass_get_device(UCLASS_SCSI, UFS_RPMB_CONTROLLER, &scsi_dev)) {
> >      ⋮ 127│+        if (optee_rpmb_get_scsi_dev(&scsi_dev)) {
> >   107⋮ 128│                 arg->ret = TEE_ERROR_ITEM_NOT_FOUND;
> >   108⋮ 129│                 return;
> >   109⋮ 130│         }
> > 
> 
> No I was just thinking as a Kconfig option along CONFIG_SUPPORT_UFS_RPMB but
> I just didn't find where you added the CONFIG_SUPPORT_UFS_RPMB option...
> 
> Neil

well ok, sure I can do the Kconfig change.

I havent enabled RPMB just yet in configs/qcom_lemans_defconfig until
the OP-TEE support is merged.

This is what I have on my development branch:

# Route OP-TEE RPMB secure storage over the UFS RPMB (no eMMC on this SoC).                                        
CONFIG_SUPPORT_UFS_RPMB=y