core: add support for SPMC at EL3

Adds support for SPMC at EL3 with CFG_CORE_EL3_SPMC. This is from OP-TEE
point of view almost identical to CFG_CORE_SEL2_SPMC with SPMC at S-EL2.

The previously S-

core: add support for SPMC at EL3

Adds support for SPMC at EL3 with CFG_CORE_EL3_SPMC. This is from OP-TEE
point of view almost identical to CFG_CORE_SEL2_SPMC with SPMC at S-EL2.

The previously S-EL2 specific functions mobj_ffa_sel2_spmc_new() and
mobj_ffa_sel2_spmc_delete() are renamed to mobj_ffa_spmc_new() and
mobj_ffa_spmc_delete() respectively since they are no longer reserved to
used only with SPMC at S-EL2.

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

show more ...

core: ffa: stop using FFA_MEMORY_REGION_FLAG_TIME_SLICE

FFA_MEMORY_REGION_FLAG_TIME_SLICE is an optional feature and might not
be available in the SPM. So drop this flag for better compatibility.

A

core: ffa: stop using FFA_MEMORY_REGION_FLAG_TIME_SLICE

FFA_MEMORY_REGION_FLAG_TIME_SLICE is an optional feature and might not
be available in the SPM. So drop this flag for better compatibility.

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

show more ...

core: ffa: correct response to FFA_INTERRUPT

When a secure interrupts is delivered by the SPM to OP-TEE the correct
response is FFA_MSG_WAIT, not FFA_SUCCESS_32 as was used previously. So
fix this.

core: ffa: correct response to FFA_INTERRUPT

When a secure interrupts is delivered by the SPM to OP-TEE the correct
response is FFA_MSG_WAIT, not FFA_SUCCESS_32 as was used previously. So
fix this.

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

show more ...

core: ffa: fix spmc_rxtx_map()

Adds the missing calls to virt_to_phys() when registering the rxtx
buffers with the SPMC.

Acked-by: Jerome Forissier <jerome@forissier.org>
Signed-off-by: Jens Wiklan

core: ffa: fix spmc_rxtx_map()

Adds the missing calls to virt_to_phys() when registering the rxtx
buffers with the SPMC.

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

show more ...

plat-vexpress: enable uart IT with TF-A and GICv3

With GICv3 TF-A doesn't need to know which interrupts are destined to
S-EL1 since this is programmed directly in the GICv3. On GICv2 that's
not poss

plat-vexpress: enable uart IT with TF-A and GICv3

With GICv3 TF-A doesn't need to know which interrupts are destined to
S-EL1 since this is programmed directly in the GICv3. On GICv2 that's
not possible so fix the ifdefs for this.

Always keep uart interrupts disabled with CFG_VIRTUALIZATION=y since
that requires more work to sort out.

Fixes: 41985789f3bd ("plat-vexpress: disable uart IT with TF-A and GICv3")
Acked-by: Jerome Forissier <jerome@forissier.org>
Signed-off-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

plat-sam: enable watchdog support in configuration

Enable watchdog support and watchdog SMC handler. Set the watchdog SMC
id to 0x2000500 (SiP SMC with func id 0x500).

Acked-by: Jens Wiklander <jen

plat-sam: enable watchdog support in configuration

Enable watchdog support and watchdog SMC handler. Set the watchdog SMC
id to 0x2000500 (SiP SMC with func id 0x500).

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

show more ...

dts: sama5d2: set the watchdog as secure

Enable the watchdog to be secure since dt_map_dev uses this attribute
to apply the correct MMU policy and we want the interrupt to be
redirected to the SAIC.

dts: sama5d2: set the watchdog as secure

Enable the watchdog to be secure since dt_map_dev uses this attribute
to apply the correct MMU policy and we want the interrupt to be
redirected to the SAIC.

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

show more ...

plat-sam: disable CFG_CORE_HAS_GENERIC_TIMER

sama5d2 platform does not have support for the ARM generic timer
extension. Disable CFG_CORE_HAS_GENERIC_TIMER and implement
plat_get_freq() using device

plat-sam: disable CFG_CORE_HAS_GENERIC_TIMER

sama5d2 platform does not have support for the ARM generic timer
extension. Disable CFG_CORE_HAS_GENERIC_TIMER and implement
plat_get_freq() using device-tree. Since clocks are probed early in
the boot process, the udelay function can be used at driver probe time
but not before clock probing of course.

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

show more ...

dts: sama5d2: add clocks for cpu

This clock will be used to retrieve the CPU clock frequency in
plat_get_freq().

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <c

dts: sama5d2: add clocks for cpu

This clock will be used to retrieve the CPU clock frequency in
plat_get_freq().

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

show more ...

delay: allow udelay() function to work without generic timers

sama5d2 processor is missing the optional ARM generic timers extension.
In order to allow using the udelay function, add a function base

delay: allow udelay() function to work without generic timers

sama5d2 processor is missing the optional ARM generic timers extension.
In order to allow using the udelay function, add a function based on
a wait loop. This support needs plat_get_freq() to be implemented by
the platform code and CFG_CORE_HAS_GENERIC_TIMER should be set to 'n'.

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

show more ...

core: sm: call wdt_sm_handler()

Call wdt_sm_handler() in order to handle watchdog SMCs.

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

core: split core/include/kernel/thread_private.h

Splits core/include/kernel/thread_private.h into one generic and one
architecture specific file.

Reviewed-by: Jerome Forissier <jerome@forissier.org

core: split core/include/kernel/thread_private.h

Splits core/include/kernel/thread_private.h into one generic and one
architecture specific file.

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

show more ...

core: move thread_defs.h into thread.h

Moves the defines in core/arch/arm/include/kernel/thread_defs.h into
core/include/kernel/thread.h.

Reviewed-by: Jerome Forissier <jerome@forissier.org>
Acked-

core: move thread_defs.h into thread.h

Moves the defines in core/arch/arm/include/kernel/thread_defs.h into
core/include/kernel/thread.h.

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

show more ...

core: split core/arch/arm/kernel/thread.c

Splits core/arch/arm/kernel/thread.c into one generic and one
architecture specific file.

Acked-by: Jerome Forissier <jerome@forissier.org>
Acked-by: Marou

core: split core/arch/arm/kernel/thread.c

Splits core/arch/arm/kernel/thread.c into one generic and one
architecture specific file.

Acked-by: Jerome Forissier <jerome@forissier.org>
Acked-by: Marouene Boubakri <marouene.boubakri@nxp.com>
Acked-by: Etienne Carriere <etienne.carriere@linaro.org>
Signed-off-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

core: split core/arch/arm/include/kernel/thread.h

Splits core/arch/arm/include/kernel/thread.h into one generic and one
architecture specific file.

Reviewed-by: Jerome Forissier <jerome@forissier.o

core: split core/arch/arm/include/kernel/thread.h

Splits core/arch/arm/include/kernel/thread.h into one generic and one
architecture specific file.

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

show more ...

core: move thread_private.h to an include directory

Moves core/arch/arm/kernel/thread_private.h to a include directory to be
included as <kernel/thread_private.h>.

Reviewed-by: Jerome Forissier <je

core: move thread_private.h to an include directory

Moves core/arch/arm/kernel/thread_private.h to a include directory to be
included as <kernel/thread_private.h>.

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

show more ...

plat-sam: pm: add support for setting suspend mode

PSCI allows entering platform suspend with SYSTEM_SUSPEND call which
is meant to enter the system in its deepest power state. sama5d2
platform supp

plat-sam: pm: add support for setting suspend mode

PSCI allows entering platform suspend with SYSTEM_SUSPEND call which
is meant to enter the system in its deepest power state. sama5d2
platform supports multiple suspend power states. Currently, Linux
supports the atmel.pm_modes command line option which allows to select
this suspend state. Since Linux uses PSCI SYSTEM_SUSPEND to enter
suspend mode, we are not able to pass information (such as done for
CPU_SUSPEND). In order to select the mode that will be entered by
SYSTEM_SUSPEND from normal world and thus select the desired suspend state,
SMCs are added to allow selecting and getting this power mode.

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

show more ...

plat-sam: add sm_platform_handler() stub

Add sm_platform_handler() to handle SMC. For the moment, this is
stubbed and will allow to handle SiP specific SMC.

Acked-by: Jens Wiklander <jens.wiklander

plat-sam: add sm_platform_handler() stub

Add sm_platform_handler() to handle SMC. For the moment, this is
stubbed and will allow to handle SiP specific SMC.

Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Clément Léger <clement.leger@bootlin.com>

show more ...

core: mm: move tee_pager.h to core/include/mm

The tee_pager.h header file does not contain architecture-specific code,
move it from core/arch/arm/include/mm to core/include/mm

Signed-off-by: Maroue

core: mm: move tee_pager.h to core/include/mm

The tee_pager.h header file does not contain architecture-specific code,
move it from core/arch/arm/include/mm to core/include/mm

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

show more ...

core: mm: move pgt_cache.h to core/include/mm

The pgt_cache.h header file does not contain architecture-specific code,
move it from core/arch/arm/include/mm to core/include/mm

Signed-off-by: Maroue

core: mm: move pgt_cache.h to core/include/mm

The pgt_cache.h header file does not contain architecture-specific code,
move it from core/arch/arm/include/mm to core/include/mm

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

show more ...

core: kernel: move abort.h to core/include/kernel

The abort.h header file does not contain architecture-specific code, move
it from core/arch/arm/include/kernel/ to core/include/kernel

Signed-off-b

core: kernel: move abort.h to core/include/kernel

The abort.h header file does not contain architecture-specific code, move
it from core/arch/arm/include/kernel/ to core/include/kernel

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

show more ...

core: kernel: move spinlock.h to core/include/kernel

Header file spinlock.h does not contain architecture-specific code
and it is being included by lib code as well, therefore, move it
from core/arc

core: kernel: move spinlock.h to core/include/kernel

Header file spinlock.h does not contain architecture-specific code
and it is being included by lib code as well, therefore, move it
from core/arch/arm/include to core/include/kernel

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

show more ...

core: kernel: move spin_lock_debug.c to core/kernel

Source file spin_lock_debug.c does not contain architecture-specific code
therefore, move it from core/arch/arm/kernel to core/kernel and remove
u

core: kernel: move spin_lock_debug.c to core/kernel

Source file spin_lock_debug.c does not contain architecture-specific code
therefore, move it from core/arch/arm/kernel to core/kernel and remove
unused header thread_private.h

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

show more ...

core: ffa: add thread_spmc_register_secondary_ep()

Adds thread_spmc_register_secondary_ep() which replaces the previous
platforms specific functions ffa_secondary_cpu_ep_register() in
plat-vexpress

core: ffa: add thread_spmc_register_secondary_ep()

Adds thread_spmc_register_secondary_ep() which replaces the previous
platforms specific functions ffa_secondary_cpu_ep_register() in
plat-vexpress and plat-totalcompute.

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

show more ...

core: imx: Calculate CFG_TZDRAM_START and CFG_NSEC_DDR_0_SIZE

Move assignement in order to calculate the value instead of
hardcoded.

0x2000000 is the sum of 0x01e00000 and 0x00200000

Signed-off-by

core: imx: Calculate CFG_TZDRAM_START and CFG_NSEC_DDR_0_SIZE

Move assignement in order to calculate the value instead of
hardcoded.

0x2000000 is the sum of 0x01e00000 and 0x00200000

Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com>
Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
Acked-by: Clement Faure <clement.faure@nxp.com>

show more ...