it will block other works of the list of hpwork when it waits to get buffer for too long. so use a separate workqueue for rpmsgdev.
Signed-off-by: liaoao <liaoao@xiaomi.com>
This commit expands RPC support for the OP-TEE driver using 2 files:
1) drivers/misc/optee_rpc.c
* Add support for RPCs that can be handled directly by the kernel.
* Can delegate RPC handling to optee_supplicant.c for RPCs that
need userspace interaction.
2) drivers/misc/optee_supplicant.c
* Enable communication between the userspace TEE supplicant and the
kernel driver.
Additional changes were needed to the following files:
1) drivers/misc/optee.c
* Add ioctls used SOLELY by the userspace TEE supplicant.
* Register /dev/teepriv0 if the supplicant is enabled in Kconfig
* Add OPTEE_ROLE_CA and OPTEE_ROLE_SUPPLICANT conditionals to
differentiate paths, between a normal Client Application (CA)
and the TEE supplicant.
* Change some functions from static to "public" to reuse them
in other C files.
* Adjust optee_to/from_msg_param() to work with RPCs.
2) drivers/misc/optee_smc.c
* Call the RPC handler from optee_rpc.c
3) drivers/misc/optee_msg.h
* Add definition needed for RPCs
4) drivers/misc/tee.h
* Add ioctl definitions
* Add TEE_SHM_SUPP flag, checked when unregistering supplicant
memory.
5) Documentation/guides/optee.rs
* Add documentation for RPCs and the supplicant.
6) drivers/misc/{CMakeLists.txt, Make.defs}
* Account for the new files.
7) drivers/misc/Kconfig
* Add DEV_OPTEE_SUPPLICANT option to enable/disable the supplicant
driver.
Signed-off-by: Theodore Karatapanis <tkaratapanis@census-labs.com>
The previous approach with memfd has 3 problems:
1) The close operation on the memfd isn't tied with optee_shm_close,
therefore close(fd) doesn't free the optee_shm struct allocated
by the kernel.
2) The kernel unnecessarily maps the file descriptor to its memory,
however only userspace should need to do that.
3) Since the kernel doesn't need to map the file descriptor we
don't need to unmap it.
To use anonymous mapping, the prototype of map_anonymous() was
moved from fs/mmap/fs_anonmap.h to include/nuttx/fs/fs.h. Since
fs_anonmap.h didn't contain any other information it is deleted.
A type from fs/mmap/fs_rammap.h was moved to the public :
include/nuttx/fs/fs.h as well.
Signed-off-by: Theodore Karatapanis <tkaratapanis@census-labs.com>
So far NuttX has supported OP-TEE interfacing over
local and RPMsg sockets. This commit introduces support
for direct invocation of OP-TEE through arm SMCs. The
SMC transport is enabled through CONFIG_DEV_OPTEE_SMC.
This SMC implementation has been tested only with arm64,
OP-TEE rev 4.4.
Note that it does not support reverse direction RPCs,
i.e. from the Secure World to the Normal World to
something like optee supplicant. A basic RPC handling
skeleton is provided with implementation for some
basic functions (alloc/free mem, and foreign interrupts)
but no supplicant command handling.
(+one minor change s/parm/param in arch/arm64/include/syscall
to satisfy the spellchecker during PR)
Signed-off-by: George Poulios <gpoulios@census-labs.com>
Prepares the ground for introducing new transports to
the OP-TEE driver. "transports" as in alternatives to
RPMsg and local network.
Signed-off-by: George Poulios <gpoulios@census-labs.com>
/dev/zero is a very commonly used config, for example use
/dev/zero to test the filesystem, mtd or block devices performance.
So let's default enable it when not enable DEFAULT_SMALL
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
The driver's main purpose is to support the porting of the open source
component optee_client (https://github.com/OP-TEE/optee_client) to nttux.
The basic function of the driver module is to convert the REE application layer data and send it to the TEE through rpmsg.
The main functions include
1 driver registration.
we need to register a device driver(/dev/tee0) through optee_register function.
2 open the driver
3 ioctl the driver
The ioctl command passes different parameters and commands, and interacts with the TEE through rpmsg.
4 close the driver
Signed-off-by: hujun5 <hujun5@xiaomi.com>
Like rpmsgdev and rpmsgmtd, rpmsgblk allow the local cpu to
access the block device in the remote cpu.
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
Before this commit, each rpmsgdev server only manages one file
no matter how many times the client opened.
It can't work fine with some complex drivers, such as input driver
(keyboard, touchscreen and button), because input driver will alloc
a new private open structure for every open operation to make sure
all the applications don't miss data when input driver is used by
multiple applications.
This commit solves this problem by making the files in server and
client be one-to-one.
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
