plat-sunxi: Add Allwinner A64 support

This commit adds support for pine64[1] board based on
Allwinner's A64 SoC.

1. Build command

# make PLATFORM=sunxi-sun50i_a64

2. Pass optee_example_hello_

plat-sunxi: Add Allwinner A64 support

This commit adds support for pine64[1] board based on
Allwinner's A64 SoC.

1. Build command

# make PLATFORM=sunxi-sun50i_a64

2. Pass optee_example_hello_world and xtest.

[1]: https://www.pine64.org/

Acked-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by: Amit Singh Tomar <amittomer25@gmail.com>

show more ...

sm: pm_a32: typo fix

typo fix: do->pm

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

core: tee: crypt_utl: remove one crypto_init

Crypto_init is called twice in tee_cryp_init.

Signed-off-by: Silvano di Ninno <silvano.dininno@nxp.com>
Reviewed-by: Jerome Forissier <jerome.forissier@

core: tee: crypt_utl: remove one crypto_init

Crypto_init is called twice in tee_cryp_init.

Signed-off-by: Silvano di Ninno <silvano.dininno@nxp.com>
Reviewed-by: Jerome Forissier <jerome.forissier@linaro.org>

show more ...

abort.c: manipulate with VFP state only if thread is active

abort_handler() can be called both within and without thread
context. In the latter case it stops on
assert(thread_get_exceptions() & THRE

abort.c: manipulate with VFP state only if thread is active

abort_handler() can be called both within and without thread
context. In the latter case it stops on
assert(thread_get_exceptions() & THREAD_EXCP_FOREIGN_INTR)
in thread_kernel_save_vfp() and no information about abort is displayed.

This assert fires during some initialization stages and during
fast SMCs, because they are handled with foreign interrupts disabled.

To fix this, we should call thread_kernel_{save,restore}_vfp() only
when foreign interrupts are enabled.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Joakim Bech <joakim.bech@linaro.org>

show more ...

core: arm: ree_fs: fix free hash_ctx

Properly free hash_ctx by calling crypto_hash_free_ctx
instead of the generic free function.

Signed-off-by: Silvano di Ninno <silvano.dininno@nxp.com>
Reviewed-

core: arm: ree_fs: fix free hash_ctx

Properly free hash_ctx by calling crypto_hash_free_ctx
instead of the generic free function.

Signed-off-by: Silvano di Ninno <silvano.dininno@nxp.com>
Reviewed-by: Joakim Bech <joakim.bech@linaro.org>

show more ...

imx: cleanup configuration

Cleanup configuration to make it easy to add new platforms.
Make most configurations be common to CFG_MX6/7.
Normally only need to define CFG_DDR_SIZE and CFG_NS_ENTRY_ADD

imx: cleanup configuration

Cleanup configuration to make it easy to add new platforms.
Make most configurations be common to CFG_MX6/7.
Normally only need to define CFG_DDR_SIZE and CFG_NS_ENTRY_ADDR
to support new platforms.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Jerome Forissier <jerome.forissier@linaro.org>

show more ...

arm32: add assembly directive: .arch_extension sec

When compiling with -mcpu=cortex-a9, GCC 8.1 fails on the smc instruction:

$ make -s PLATFORM=stm CROSS_COMPILE32=<GCC8.1 path>/arm-linux-gnueabi

arm32: add assembly directive: .arch_extension sec

When compiling with -mcpu=cortex-a9, GCC 8.1 fails on the smc instruction:

$ make -s PLATFORM=stm CROSS_COMPILE32=<GCC8.1 path>/arm-linux-gnueabihf-
core/arch/arm/kernel/thread_a32.S: Assembler messages:
core/arch/arm/kernel/thread_a32.S:44: Error: selected processor does not support `smc #0' in ARM mode
[snip]
mk/compile.mk:146: recipe for target 'out/arm-plat-stm/core/arch/arm/kernel/thread_a32.o' failed
make: *** [out/arm-plat-stm/core/arch/arm/kernel/thread_a32.o] Error 1

Use the '.arch_extension sec' directive to allow the assembler to emit
the instruction.

Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Reviewed-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>

show more ...

core_self_tests.c: fix p1=realloc(p1) issue

This is invalid use of realloc, because it can cause memory leak.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jerome Forissier

core_self_tests.c: fix p1=realloc(p1) issue

This is invalid use of realloc, because it can cause memory leak.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jerome Forissier <jerome.forissier@linaro.org>

show more ...

imx: wdog: correct wdog_path

The prefix `0` is removed in Linux Kernel upstream code,
so let's drop it to let wdog work.

Linux Kernel commit 67b8d5c7081221efa252("Linux 4.17-rc5")

Signed-off-by: P

imx: wdog: correct wdog_path

The prefix `0` is removed in Linux Kernel upstream code,
so let's drop it to let wdog work.

Linux Kernel commit 67b8d5c7081221efa252("Linux 4.17-rc5")

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Jerome Forissier <jerome.forissier@linaro.org>

show more ...

plat-imx: add i.MX6 Hummingboard Edge platform flavors

Signed-off-by: Jordan Rhee <jordanrh@microsoft.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>

plat-imx: make platform flavor list diff friendly

Signed-off-by: Jordan Rhee <jordanrh@microsoft.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>

plat-hikey: embed ta/avb as early TA

In order to support AVB in U-boot embed the AVB ta as an early TA.

Reviewed-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by: Jens Wiklander <je

plat-hikey: embed ta/avb as early TA

In order to support AVB in U-boot embed the AVB ta as an early TA.

Reviewed-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

Add CFG_IN_TREE_EARLY_TAS

Adds CFG_IN_TREE_EARLY_TAS which is used to embed in-tree TAs as early
TAs in the OP-TEE binary.

Reviewed-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by:

Add CFG_IN_TREE_EARLY_TAS

Adds CFG_IN_TREE_EARLY_TAS which is used to embed in-tree TAs as early
TAs in the OP-TEE binary.

Reviewed-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

user_ta: should go for other TA stores on any load error

There seems to be an issue that if RPMB_FS is enabled in
OPTEE and TA is present in REE (normal file system), if
priority for secure storage

user_ta: should go for other TA stores on any load error

There seems to be an issue that if RPMB_FS is enabled in
OPTEE and TA is present in REE (normal file system), if
priority for secure storage TA is higher and RPMB
initialization fails, the error is returned and the
OPTEE doesn't goes to find the TA from REE TA store.

The issue is fixed by adding a 'continue' statement after
printing the respective error.

Reviewed-by: Jerome Forissier <jerome.forissier@linaro.org>
Signed-off-by: Sourabh Das <sourabhdas143@gmail.com>

show more ...

core: unwind: correct function args for print_stack_arm32/64

When CFG_TEE_CORE_LOG_LEVEL=0 to make, met build failure:
"
core/arch/arm/kernel/abort.c: In function '__print_stack_unwind_arm32':
core/

core: unwind: correct function args for print_stack_arm32/64

When CFG_TEE_CORE_LOG_LEVEL=0 to make, met build failure:
"
core/arch/arm/kernel/abort.c: In function '__print_stack_unwind_arm32':
core/arch/arm/kernel/abort.c:113:2: error: too many arguments to function 'print_stack_arm32'
print_stack_arm32(TRACE_ERROR, &state, exidx, exidx_sz, kernel_stack,
^~~~~~~~~~~~~~~~~
"

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

core: define syscall_t as void (*)(void)

syscall_t is currently typedef'ed as TEE_Result (*)(void). It is used to
represent a pointer to any system call, in the syscall table for instance.
As such,

core: define syscall_t as void (*)(void)

syscall_t is currently typedef'ed as TEE_Result (*)(void). It is used to
represent a pointer to any system call, in the syscall table for instance.
As such, the exact type behind syscall_t cannot reflect all the syscalls
since they have different prototypes. The current declaration with a
TEE_Result return type was probably chosen because it was a common
characteristic of all syscalls to return a TEE_Result.

However, this type causes compilation warnings with GCC 8.1:

core/arch/arm/tee/arch_svc.c:43:36: warning: cast between incompatible function types from ‘void (*)(long unsigned int)’ to ‘TEE_Result (*)(void)’ {aka ‘unsigned int (*)(void)’} [-Wcast-function-type]
#define SYSCALL_ENTRY(_fn) { .fn = (syscall_t)_fn }
^
core/arch/arm/tee/arch_svc.c:50:2: note: in expansion of macro ‘SYSCALL_ENTRY’
SYSCALL_ENTRY(syscall_sys_return),
^~~~~~~~~~~~~

The solution is to use 'void (*)(void)' instead, as explained in the GCC
documentation:

-Wcast-function-type

Warn when a function pointer is cast to an incompatible function
pointer. [...] The function type void (*) (void) is special and matches
everything, which can be used to suppress this warning. [...]

Link: [1] https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html
Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Acked-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

drivers: hi16xx_rng: replace mutex with spinlock

The mutex in hw_get_random_byte() protects a very short section of code.
A spinlock is more lightweight and therefore better suited to the task.

Sig

drivers: hi16xx_rng: replace mutex with spinlock

The mutex in hw_get_random_byte() protects a very short section of code.
A spinlock is more lightweight and therefore better suited to the task.

Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

thread: move stacks to separate sections

With this change it is possible to move tmp and abt stacks to kernel
memory area, while leaving thread stacks in tee memory.

Signed-off-by: Volodymyr Babchu

thread: move stacks to separate sections

With this change it is possible to move tmp and abt stacks to kernel
memory area, while leaving thread stacks in tee memory.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

link_dummy.ld: provide __data_start symbol

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

linker.h: declare __data_start as non-const

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

asid: move asid allocator from tee_mmu.c to core_mmu.c

ASIDs will be allocated for individual virtrual guests, so
allocator should reside in more generic place.

Also, comment for MMU_NUM_ASIDS was

asid: move asid allocator from tee_mmu.c to core_mmu.c

ASIDs will be allocated for individual virtrual guests, so
allocator should reside in more generic place.

Also, comment for MMU_NUM_ASIDS was updated.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

mmu: align va of memory regions to pa modulo PGDIR_SIZE

If pa % PGDIR_SIZE == va % PGDIR_SIZE, then we can effectively map
large smallpage-aligned regions. Most of the region can be mapped
with supe

mmu: align va of memory regions to pa modulo PGDIR_SIZE

If pa % PGDIR_SIZE == va % PGDIR_SIZE, then we can effectively map
large smallpage-aligned regions. Most of the region can be mapped
with super blocks and only ends will be mapped using small pages.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

generic_ram_layout: align TA_RAM to SMALL_PAGE_SIZE

This enables more optimal memory usage, as there will be no unused
holes in memory mappings.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail

generic_ram_layout: align TA_RAM to SMALL_PAGE_SIZE

This enables more optimal memory usage, as there will be no unused
holes in memory mappings.

Signed-off-by: Volodymyr Babchuk <vlad.babchuk@gmail.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

core: arm: sm: initialize PMCR.DP to 1 and save/restore PMCR

Introduce CFG_SM_NO_CYCLE_COUNTING to intitialize PMCR.DP to 1 and
save/restore PMCR on world switch. Similar to what is done in ARM TF
c

core: arm: sm: initialize PMCR.DP to 1 and save/restore PMCR

Introduce CFG_SM_NO_CYCLE_COUNTING to intitialize PMCR.DP to 1 and
save/restore PMCR on world switch. Similar to what is done in ARM TF
commit 3e61b2b54336 ("Init and save / restore of PMCR_EL0 / PMCR") [1].

The purpose of this is to (hopefully) make attacks such as CLKSCREW [2]
harder to mount, although it is likely that timing information could be
obtained via other means.

Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Link: [1] https://github.com/ARM-software/arm-trusted-firmware/commit/3e61b2b54336
Link: [2] https://www.usenix.org/system/files/conference/usenixsecurity17/sec17-tang.pdf
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Acked-by: Etienne Carriere <etienne.carriere@linaro.org>

show more ...

core: arm: sm: rename struct sm_mode_regs to sm_unbanked_regs

struct sm_mode_regs will soon be used to store one non-banked register
other then the mode registers (PMCR). Rename it to sm_unbanked_re

core: arm: sm: rename struct sm_mode_regs to sm_unbanked_regs

struct sm_mode_regs will soon be used to store one non-banked register
other then the mode registers (PMCR). Rename it to sm_unbanked_regs.

Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Acked-by: Etienne Carriere <etienne.carriere@linaro.org>

show more ...