core: remove references to OPTEE_SMC_SHM_CACHED

Removes references to OPTEE_SMC_SHM_CACHED in architecture independent
code, the references are replaces by TEE_MATTR_CACHE_CACHED which is
more accur

core: remove references to OPTEE_SMC_SHM_CACHED

Removes references to OPTEE_SMC_SHM_CACHED in architecture independent
code, the references are replaces by TEE_MATTR_CACHE_CACHED which is
more accurate.

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

show more ...

core: vm.c: don't include sm/optee_smc.h

sm/optee_smc.h isn't needed in this file any longer so remove the
include statement.

Acked-by: Marouene Boubakri <marouene.boubakri@nxp.com>
Reviewed-by: Je

core: vm.c: don't include sm/optee_smc.h

sm/optee_smc.h isn't needed in this file any longer so remove the
include statement.

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

show more ...

core: mm: move tee_mm.c to core/mm

Move tee_mm.c from core/arch/arm/mm to core/mm to reuse it with new
architectures.

Signed-off-by: Marouene Boubakri <marouene.boubakri@nxp.com>
Reviewed-by: Etien

core: mm: move tee_mm.c to core/mm

Move tee_mm.c from core/arch/arm/mm to core/mm to reuse it with new
architectures.

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

show more ...

core: split core/arch/arm/mm/core_mmu.c

Splits core/arch/arm/mm/core_mmu.c into one generic and one architecture
specific file.

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

core: split core/arch/arm/mm/core_mmu.c

Splits core/arch/arm/mm/core_mmu.c into one generic and one architecture
specific file.

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

show more ...

core: mm: vm.c: remove arm.h and include config.h

Removes arm.h from include list as it is not used and explicitly add
config.h for IS_ENABLED macro.

Signed-off-by: Marouene Boubakri <marouene.boub

core: mm: vm.c: remove arm.h and include config.h

Removes arm.h from include list as it is not used and explicitly add
config.h for IS_ENABLED macro.

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: make mobj_get_va() more secure

Adds a length parameter to allow mobj_get_va() to check that the entire
va range requested is available.

Reviewed-by: Etienne Carriere <etienne.carriere@linaro.

core: make mobj_get_va() more secure

Adds a length parameter to allow mobj_get_va() to check that the entire
va range requested is available.

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

show more ...

core: Add vm_get_mobj

Return the mobj of a va.

Signed-off-by: Jelle Sels <jelle.sels@arm.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Acked-by: Etienne Carriere <etienne.carriere@li

core: Add vm_get_mobj

Return the mobj of a va.

Signed-off-by: Jelle Sels <jelle.sels@arm.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Acked-by: Etienne Carriere <etienne.carriere@linaro.org>

show more ...

core: mm : Enable GP bit for kernel mapping for userspace

Mark the kernel pages mapped in userspace as guarded.

Signed-off-by: Ruchika Gupta <ruchika.gupta@linaro.org>
Reviewed-by: Jerome Forissier

core: mm : Enable GP bit for kernel mapping for userspace

Mark the kernel pages mapped in userspace as guarded.

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

show more ...

core: mmu: fix overflow with high address in tee_mm_pool_t

In case of TA_RAM defined at the end of address range,
the high address will be defined outside the paddr_t limits
which ends in a 0 addres

core: mmu: fix overflow with high address in tee_mm_pool_t

In case of TA_RAM defined at the end of address range,
the high address will be defined outside the paddr_t limits
which ends in a 0 address usage.
The size must be used rather than the high address to
avoid this overflow issue. Update the corresponding files due
to API modification.

Signed-off-by: Lionel Debieve <lionel.debieve@foss.st.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Reviewed-by: Etienne Carriere <etienne.carriere@linaro.org>

show more ...

util: rename ALIGNMENT_IS_OK to IS_ALIGNED_WITH_TYPE

Implement the renamed macro using the IS_ALIGNED definition.

Signed-off-by: Jorge Ramirez-Ortiz <jorge@foundries.io>
Reviewed-by: Etienne Carrie

util: rename ALIGNMENT_IS_OK to IS_ALIGNED_WITH_TYPE

Implement the renamed macro using the IS_ALIGNED definition.

Signed-off-by: Jorge Ramirez-Ortiz <jorge@foundries.io>
Reviewed-by: Etienne Carriere <etienne.carriere@linaro.org>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

core_mmu: fix phys_to_virt() to check length

phys_to_virt() function without length parameter doesn`t
always have ability to find the correct mapping for
requested physical address. This is because

core_mmu: fix phys_to_virt() to check length

phys_to_virt() function without length parameter doesn`t
always have ability to find the correct mapping for
requested physical address. This is because physical
address can be mapped in the same time in different virtual
regions with different length. So the first found region
which contains the requested physical address possibly
doesn`t have enough mapped data. This is fixed by adding
the length parameter to phys_to_virt() function. Length
parameter can be set to 1 if caller knows that requested
(pa + len) doesn`t cross mapping granule boundary.

core_mmu_get_va() and io_pa_or_va() functions now are
take length parameter too as they based on phys_to_virt()
in case of MMU enabled.

Signed-off-by: Anton Rybakov <a.rybakov@omp.ru>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Acked-by: Etienne Carriere <etienne.carriere@linaro.org>
Tested-by: Etienne Carriere <etienne.carriere@linaro.org> (stm32mp1-157C_DK2)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx6dlsabreauto)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx6dlsabresd)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx6qpsabreauto)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx6sllevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx6ulevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx6ullevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx6ulzevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx7dsabresd)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx7ulpevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx8mmevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx8mnevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx8mqevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx8mpevk)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx8qmmek)
Tested-by: Clement Faure <clement.faure@nxp.com> (imx-mx8qxpmek)

show more ...

mm: fix mobj split by adding core_mmu_find_mapping_exclusive() helper

Fixes: ff01e2452169 ("mm: split mobj_tee_ram onto rw/rx parts")

This fixes mobj splitting onto RX/RW parts. Now split can be do

mm: fix mobj split by adding core_mmu_find_mapping_exclusive() helper

Fixes: ff01e2452169 ("mm: split mobj_tee_ram onto rw/rx parts")

This fixes mobj splitting onto RX/RW parts. Now split can be done
incorrectly if RX and RW regions doesn`t mapped contiguosly. Added helper
core_mmu_find_mapping_exclusive() allows to find unique mapping for
specified type and length independently of their order, so then RX/RW
regions for mobjects should be determined correctly.

Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Signed-off-by: Anton Rybakov <a.rybakov@omp.ru>

show more ...

core: mm: remove redundant mobj_put() in vm_map_pad()

When mobj_get_cattr() fails vm_map_pad() doesn't need to
call mobj_put() which is expected to balance mobj_get() called
only after mobj_get_catt

core: mm: remove redundant mobj_put() in vm_map_pad()

When mobj_get_cattr() fails vm_map_pad() doesn't need to
call mobj_put() which is expected to balance mobj_get() called
only after mobj_get_cattr() succeeds. The issue was introduced
in release 3.8.0 with struct mobj reference counting.

Signed-off-by: Gavin Liu <Gavin.Liu@mediatek.com>
Reviewed-by: Etienne Carriere <etienne.carriere@linaro.org>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

mm: split mobj_tee_ram onto rw/rx parts

Now mobj_tee_ram memory abstraction contains both TEE_RAM_RX and TEE_RAM_RW
regions joined together. This patch splits it to mobj_tee_ram_rx and
mobj_tee_ram_

mm: split mobj_tee_ram onto rw/rx parts

Now mobj_tee_ram memory abstraction contains both TEE_RAM_RX and TEE_RAM_RW
regions joined together. This patch splits it to mobj_tee_ram_rx and
mobj_tee_ram_rw to manage RX/RW memory objects separately.

Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Reviewed-by: Etienne Carriere <etienne.carriere@linaro.org>
Signed-off-by: Anton Rybakov <a.rybakov@omp.ru>

show more ...

core: make __rodata_unpaged() symbols __weak

Makes the __rodata_unpaged tagged symbols __weak and non-static in order
to be overridden in core/arch/arm/kernel/link_dummies_paged.c. This
makes sure t

core: make __rodata_unpaged() symbols __weak

Makes the __rodata_unpaged tagged symbols __weak and non-static in order
to be overridden in core/arch/arm/kernel/link_dummies_paged.c. This
makes sure that these symbols doesn't bring in further symbols in the
unpaged section.

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

show more ...

core: use separate sections for each __rodata_unpaged variable

Adds a mandatory argument to the macro __rodata_unpaged() to take the
name of the variable to put in the unpaged rodata section. This w

core: use separate sections for each __rodata_unpaged variable

Adds a mandatory argument to the macro __rodata_unpaged() to take the
name of the variable to put in the unpaged rodata section. This will
result in separate sections for each such variable and make it easier to
debug the pruning of the dependency tree for unpaged sections.

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

show more ...

core: pager: fix compiler warning with Clang

Function rwp_unpaged_iv_free() is reduced to a call to panic() when
CFG_WITH_PAGER=y and CFG_CORE_PAGE_TAG_AND_IV=y. In this case, Clang 12
suggests a no

core: pager: fix compiler warning with Clang

Function rwp_unpaged_iv_free() is reduced to a call to panic() when
CFG_WITH_PAGER=y and CFG_CORE_PAGE_TAG_AND_IV=y. In this case, Clang 12
suggests a noreturn attribute:

$ make -s CFG_WITH_PAGER=y COMPILER=clang
core/mm/fobj.c:322:1: warning: function 'rwp_unpaged_iv_free' could be
declared with attribute 'noreturn' [-Wmissing-noreturn]
{
^
1 warning generated.

However the attribute cannot be applied since it would be inappropriate
when CFG_CORE_PAGE_TAG_AND_IV != y. Therefore, disable the warning for
the file core/mm/fobj.c when the problematic configuration is enabled.

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

show more ...

core: rename struct tee_pager_area to vm_paged_region

Renames struct tee_pager_area to struct vm_paged_region and moves it
next to the declaration of struct vm_region. Since areas are now called
pag

core: rename struct tee_pager_area to vm_paged_region

Renames struct tee_pager_area to struct vm_paged_region and moves it
next to the declaration of struct vm_region. Since areas are now called
paged regions or regions also rename functions, variables and struct
members accordingly.

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

show more ...

core: mm: move mobj.c to core/mm

mobj is abstract and it is used by many sources which are not
architecture-specific such as core/kernel, core/pta and
core/tee. Therefore, move mobj.c to core/mm and

core: mm: move mobj.c to core/mm

mobj is abstract and it is used by many sources which are not
architecture-specific such as core/kernel, core/pta and
core/tee. Therefore, move mobj.c to core/mm and its
corresponding header file mobj.h to core/include/mm.

Signed-off-by: Marouene Boubakri <marouene.boubakri@nxp.com>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>

show more ...

core: mm: fix infinite loop in vm_pa2va()

Commit d6ad67f674e5 ("core: mm: change vm_pa2va() to return a virtual
address") moved the call to mobj_get_pa() up in a 'for' loop and added
a 'continue' st

core: mm: fix infinite loop in vm_pa2va()

Commit d6ad67f674e5 ("core: mm: change vm_pa2va() to return a virtual
address") moved the call to mobj_get_pa() up in a 'for' loop and added
a 'continue' statement. Moving it was wrong because at this point 'size'
is not yet updated which causes an infinite loop when the PA is not
found.

Move the call back to its original location but keep the 'continue'
which looks correct.

Fixes: d6ad67f674e5 ("core: mm: change vm_pa2va() to return a virtual address")
Signed-off-by: Jerome Forissier <jerome@forissier.org>
Reviewed-by: Jens Wiklander <jens.wiklander@linaro.org>
Reviewed-by: Etienne Carriere <etienne.carriere@linaro.org>

show more ...

core: mm: change vm_pa2va() to return a virtual address

Change vm_pa2va() to return target virtual address or NULL if the
physical address cannot be resolved which can happen when pager is
enabled a

core: mm: change vm_pa2va() to return a virtual address

Change vm_pa2va() to return target virtual address or NULL if the
physical address cannot be resolved which can happen when pager is
enabled and the target physical page belongs to the pager page pool.
This change makes vm_pa2va() helper function simpler and its only caller
doesn't differentiate error return codes.

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

show more ...

core: introduce CFG_CORE_PAGE_TAG_AND_IV

Introduces CFG_CORE_PAGE_TAG_AND_IV which defaults to enabled if TA
paging is enabled. Can be used to disable tag and IV paging for paged
read-write pages.

core: introduce CFG_CORE_PAGE_TAG_AND_IV

Introduces CFG_CORE_PAGE_TAG_AND_IV which defaults to enabled if TA
paging is enabled. Can be used to disable tag and IV paging for paged
read-write pages.

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

show more ...

core: fobj_rw_paged_alloc() store tags and IVs in paged area

fobj_rw_paged_alloc() is updated to store tags and IVs in a designated
paged area instead of storing them in the heap. This avoids large

core: fobj_rw_paged_alloc() store tags and IVs in paged area

fobj_rw_paged_alloc() is updated to store tags and IVs in a designated
paged area instead of storing them in the heap. This avoids large heap
allocations which also would suffer from a fragmented heap.

The previous ops_rw_paged and struct fobj_rwp are now replaced by
ops_rwp_unpaged_iv and struct fobj_rwp_unpaged_iv respectively. These
are now only used to support the area where other tags and IVs are
stored.

A new ops_rwp_paged_iv and struct fobj_rwp_paged_iv are added for
using the designated paged area.

A fobj based on the ops_rwp_unpaged_iv ops is allocated and registered
with the pager via a callback registered with driver_init_late().

This effectively enables paging of IV and tags for pages mapping a TA.

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

show more ...

core: clear user mappings from tables when removed

When a user mapping is removed clear it immediately from active or
cached translation tables.

Reviewed-by: Etienne Carriere <etienne.carriere@lina

core: clear user mappings from tables when removed

When a user mapping is removed clear it immediately from active or
cached translation tables.

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

show more ...

core: replace tee_mmu prefix with vm

Replaces the tee_mmu prefix with vm. tee_mmu.h is renamed to vm.h and
core/arch/arm/mm/tee_mmu.c is moved to core/mm/vm.c. Public functions
belonging to these fi

core: replace tee_mmu prefix with vm

Replaces the tee_mmu prefix with vm. tee_mmu.h is renamed to vm.h and
core/arch/arm/mm/tee_mmu.c is moved to core/mm/vm.c. Public functions
belonging to these files are renamed with a vm prefix.

Introduces: vm_map_param(), vm_clean_param(),
vm_buf_is_inside_private(), vm_buf_intersects_private(),
vm_buf_to_mboj_offs(), vm_buf_is_inside_um_private(),
vm_buf_intersects_um_private(), vm_add_rwmem(), vm_rem_rwmem(),
vm_va2pa(), vm_pa2va(), vm_check_access_rights(), vm_set_ctx() replacing
their tee_mmu_*() counterpart.

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

show more ...