This patch adds the basic CPU library code to support the Poseidon CPU
in TF-A. Poseidon is derived from HunterELP core, an implementation of
v9.2 architecture. Currently, Hunter CPU the predecessor to HunterELP,
is supported in TF-A. Accordingly the Hunter CPU library code has been
as the base and adapted here.
Signed-off-by: Jayanth Dodderi Chidanand <jayanthdodderi.chidanand@arm.com>
Change-Id: I406b4de156a67132e6a5523370115aaac933f18d
Remove stm32mp_incr_shrefcnt(), stm32mp_decr_shrefcnt(),
stm32mp_incr_refcnt() and stm32mp_decr_refcnt() that are unused.
The file is then just removed.
Change-Id: I09ee23c02317df5d8f71cbc355d3ed4a67ce2749
Signed-off-by: Etienne Carriere <etienne.carriere@st.com>
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Due to stm32mp_shres_helpers.h removal, the debug.h header is no more
included. It should then be added to stm32mp1_boot_device.c.
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Change-Id: I397911ac05fdff464c010cf3b2e04320a781b4aa
Add support for enabling the FWU multi bank boot feature on the
platform.
Currently, this feature is supported on the STM32MP157C-DK2 board,
which boots off a uSD card. Also, support has been enabled when
booting from a FIP image.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Change-Id: Ia69e858461e2daf599d41d66d7ff2ccae0c341c2
With the FWU Multi Bank update feature, the platform can boot from one
of multiple banks(partitions). Pass the value of bank from which the
platform has booted as boot index to the Update Agent. The Update
Agent will match this boot index value against the active_index field
in the metadata, and update the metadata if there is a mismatch.
Fow now, the mechanism to pass the boot index is platform specific. On
the STM32MP1 platform, the boot index value is passed through a
memorey mapped TAMP register on the SoC.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Change-Id: I0aa665ff9c1db95be8ae19ed8de6d866587d6850
Add support for reading the FWU metadata partition. The metadata
partition stores information on the current active bank along with
information on all the FWU updatable images on the platform. This
information is then used to identify the image to be booted.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Change-Id: I66bc5ac718c21a49c504e698b5b1f5c4daed2d08
With the FWU multi bank boot feature enabled, the platform can boot
from one of the multiple banks(partitions) containing the firmware
images. The bank whose firmware components are to be booted is read
from the FWU metadata structure -- the image to be booted is thus
derived by reading the metadata.
Read the metadata and set the image spec of the corresponding image
type to point to the partition from which the image is to be booted.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Change-Id: I3dfdc7e9202859e917ec4e1f7d1855aad42c6b70
Add GUID's for identifying the firmware image type. With the FWU
multi bank boot feature enabled, these GUID values are used to
identify the firmware image to be booted. This is done by matching
GUID values of images in the io policy table with the Image GUID value
that is read from the FWU metadata structure.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Change-Id: Id9751f02f95fc48ef68e4e3f9f0ddbf6d6319d3c
With the FWU multi bank feature enabled, the identification of
firmware image type is done using the image type GUID instead of
binary_type field.
Add GUID values for the FIP image which can be updated through
the FWU firmware update feature. The GUID values are used in
identifying the firmware images.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Change-Id: If7d9356aa8d2bb3fbcbc87100e6972f1a1862921
The metadata structure copy is passed to the platform routine to set
the image source to boot the platform from. This is done by reading
the metadata structure. Pass the metadata as a read-only copy to the
routine -- the routine only needs to consume the metadata values and
should not be able to update the metadata fields.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Change-Id: I399cad99ab89c71483e5a32a1de0e22df304f8b0
* changes:
feat(stm32mp1): add helper to enable high speed mode in low voltage
refactor(stm32mp1): add helpers for IO compensation cells
feat(stm32mp1): use clk_enable/disable functions
feat(stm32mp1): add timeout in IO compensation
Increase the stack size to avoid stack overflow
when the LOG_LEVEL compile option is set high.
Signed-off-by: Hideyuki Nitta <hideyuki.nitta.jf@hitachi.com>
Signed-off-by: Toshiyuki Ogasahara <toshiyuki.ogasahara.bo@hitachi.com>
Signed-off-by: Yoshifumi Hosoya <yoshifumi.hosoya.wj@renesas.com>
Change-Id: I25047322763bff148dba13848a3a40f4c7cf90b7
Fixed an issue where the CPU and Cluster could not be turned OFF
when the SYSTEM_OFF has executed.
Signed-off-by: Hideyuki Nitta <hideyuki.nitta.jf@hitachi.com>
Signed-off-by: Toshiyuki Ogasahara <toshiyuki.ogasahara.bo@hitachi.com>
Signed-off-by: Yoshifumi Hosoya <yoshifumi.hosoya.wj@renesas.com>
Change-Id: Id476f815b58246ae0574c04ccb3eb201d09039b9
The LS1043A reference design board (RDB) is a computing, evaluation,
and development platform that supports the Layerscape LS1043A
architecture processor.
The old implementation in tf-a (plat/layerscape/board/ls1043/) is removed,
and this patch is adding it back, it is using the unified software
component and architecture with all the other Layerscape platforms.
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
Signed-off-by: Pankaj Gupta <pankaj.gupta@nxp.com>
Signed-off-by: Ruchika Gupta <ruchika.gupta@nxp.com>
Change-Id: I83eee2f9254267b148960b05e25b6c9ba86cf07e
The LS1043A processor was NXP's first quad-core, 64-bit Arm based
processor for embedded networking.
The old implementation in tf-a (plat/layerscape/board/ls1043/) is removed,
and this patch is adding it back, it is using the unified software
component and architecture with all the other Layerscape platforms.
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
Signed-off-by: Pankaj Gupta <pankaj.gupta@nxp.com>
Signed-off-by: Ruchika Gupta <ruchika.gupta@nxp.com>
Signed-off-by: rocket <rod.dorris@nxp.com>
Change-Id: Ia3877530fae6479bd4a33bbe46b0c0d28ab43160
Remove old implementation for Layerscape ls1043a platform, and
will added it back with unified software architecture of all
Layerscape platforms.
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
Change-Id: If038c19ab04d70050ec8e6ab2097b1c4f8324e87
Increate SoC name length as it is not enough for some
SoC personalities.
Signed-off-by: Biwen Li <biwen.li@nxp.com>
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
Change-Id: I2142b4b5162dd3c9ab3afefcdc859063836d8bcc
Use STM32MP_HEADER_RESERVED_SIZE macro instead of a fixed value 0x3000
in linker script.
Change-Id: I2702285c15aebaa1304a891c8aaabc949a912ba6
Signed-off-by: Yann Gautier <yann.gautier@st.com>
This new function is used to fill the register(s) responsible to enable
high speed mode for pad in low voltage (<2.7V).
Change-Id: Ib8abc6628bdf51bbe6a866bc6a9bcdeb4a84a8f4
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Add enable_io_comp_cell and disable_io_comp_cell local helpers
to enable or disable an IO compensation cell.
Change-Id: I65295298a7ece572ae939e2db93d10b188de0f9e
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Use the clock framework functions in SYSCFG driver instead of dedicated
functions.
Change-Id: Ifb50a5207e8cecef1c80d86e2de4d70ab6bf8b8b
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Use a timeout during IO compensation enable function, when
waiting for ready status. If timeout expires, print a warning
message, to indicate that the SoC recommendation is not followed.
Change-Id: I98c7dcb1364b832f4f4b5fc9a0b85a3741a8af4b
Signed-off-by: Nicolas Le Bayon <nicolas.le.bayon@st.com>
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Added some RCPM2 register offset definiton for register: IPSTPCR,
IPSTPACKR and POWMGTDCR, also added OVRD bit definiton of register
POWMGTDCR.
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
Change-Id: I301bc1401e053c2089b5eb3672c6e649c805a2ab
total_dram_size should be signed value because it is equal to return
value of init_ddr(), so if it is lower or equal zero, report
error as DDR is not initialized correctly.
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
Change-Id: Idbc40da103f60f10cb18c5306e97b764c1a9d372
Add helper function to disable the load-store prefetch.
Signed-off-by: Jiafei Pan <Jiafei.Pan@nxp.com>
Change-Id: I36d7be37e0b800ab1e5842a56cfd04d779338868
* changes:
feat(stm32mp1): enable BL2_IN_XIP_MEM to remove relocation sections
refactor(stm32mp1): reduce MMU memory regions and split XLAT by context
feat(st): map 2MB for ROM code
fix(stm32mp1): restrict DEVICE2 mapping in BL2
Total Compute has ETE and TRBE tracing components and they have
to be enabled to capture the execution trace of the processor.
Signed-off-by: Davidson K <davidson.kumaresan@arm.com>
Change-Id: I3c86c11be2c655a61ecefa3eb2e4e3951577a113
Because the BL2 is not relocated, the usage of BL2_IN_XIP_MEM
can be used. It reduces the binary size by removing all relocation
sections. XIP will not be used when STM32MP_USE_STM32IMAGE is
defined. Introduce new definitions for SEPARATE_CODE_AND_RODATA.
Signed-off-by: Lionel Debieve <lionel.debieve@st.com>
Change-Id: Ifd76f14e5bc98990bf84e0bfd4ee0b4e49a9a293
Simplify the BL2 MMU mapping and reduce the memory regions
number. Split the XLAT define between BL2 and BL32 as binaries
do not share the same tables anymore.
Signed-off-by: Lionel Debieve <lionel.debieve@st.com>
Change-Id: Iaf09e72b4cc29acbe376f6f1cd2a8116c793ba26
This allows reducing MMU tables, and as there is nothing after ROM code
in memory mapping, this has no impact.
Change-Id: If51facb96a523770465cb06eb1ab400f75d26db3
Signed-off-by: Yann Gautier <yann.gautier@foss.st.com>
Only NAND memory map area can be of interest for BL2 in the
DEVICE2 area. Map DEVICE2 under STM32MP_RAW_NAND flag.
Change-Id: I7e3b39579e4a2525b25cb1987d6ec38038d0de2b
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Change BL31 load address to 0x970000. This was done by Change-Id
I96d572fc. But then changed back to 0x960000 by Change-Id I8308c629.
However, 0x970000 is the correct value thus we change it back again.
Signed-off-by: Ying-Chun Liu (PaulLiu) <paul.liu@linaro.org>
Change-Id: Ia0db4877123b89072f723d18e2bcce25ef38f47d
Disabled CRYPTO_SUPPORT option for Renesas platform as it does not
follow the TF-A authentication mechanism where Trusted-Boot mandates
Crypto module support.
Change-Id: I3aa771e983e3dde083dd8a861f25c0714ffd707f
Signed-off-by: Manish V Badarkhe <Manish.Badarkhe@arm.com>
As Measured-Boot and Trusted-Boot are orthogonal, removed
Trusted-Boot's dependency on Measured-Boot by allowing them
to apply the Crypto module changes independently using the
CRYPTO_SUPPORT build flag.
Change-Id: I5a420e5d84f3fefe0c0092d822dab981e6390bbf
Signed-off-by: Manish V Badarkhe <Manish.Badarkhe@arm.com>
* changes:
refactor(st-ddr): move basic tests in a dedicated file
refactor(st-ddr): reorganize generic and specific elements
feat(stm32mp1): allow configuration of DDR AXI ports number
refactor(st-ddr): update parameter array initialization
feat(st-ddr): add read valid training support
refactor(stm32mp1): remove the support of calibration result
fix(st-ddr): correct DDR warnings
As the UART is already initialized, no need to check for UART clock
or reset in next BL. An issue can appear if the next BL device tree
(e.g HW_CONFIG) doesn't use the same clocks or resets (like SCMI ones).
Signed-off-by: Yann Gautier <yann.gautier@st.com>
Change-Id: I044ef2386abe2d3dba5a53c3685440d64ca50a4f