Before the priming begins, we need to disable RMW (Read Modify Write)
and disable ECC verification for read accesses. By default, the EMIF
tool enables RMW and read accesses in the EMIF_ECC_CTRL_REG.
Signed-off-by: Krunal Bhargav <k-bhargav@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
If ECC is enabled, we need to ensure interleaving is disabled for higher
address space.
Signed-off-by: Krunal Bhargav <k-bhargav@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
ecc_address_range registers contains the start address and end address
of the DDR address space. But the ddr cmd driver is assuming the register
contains the start address and size of the DDR address space. Because
of this some valid ecc addresses are errored out as invalid address.
Fix this calculation.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
ecc_address_range registers contains the start address and end address
of the DDR address space. But the ddr driver is assuming the register
contains the start address and size of the DDR address space. Because
of this the ecc enabling is failing for the 2nd range of ecc addresses.
Fix this calculation.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Regulator should not be enabled at probe time if regulator-boot-on
property is not in the dt node.
"enable-active-high" property is only used to indicate the GPIO
polarity.
See kernel documentation :
- Documentation/devicetree/bindings/regulator/fixed-regulator.yaml
- Documentation/devicetree/bindings/regulator/gpio-regulator.yaml
Signed-off-by: Christophe Kerello <christophe.kerello@st.com>
Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
In 32bits mode upper bits need to be set to 0, otherwise controller will
try to DMA into not existing memory and stops with error.
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Oleksandr Rybalko <ray@ddteam.net>
This adds a document for tphy which supports physical layer
functionality for a number of controllers on MediaTek SoCs,
such as, USB2.0, USB3.0, PCIe, and SATA.
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
This adds a document for MT7623 PCIe controller.
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
This follows the linux header rules to avoid conflict bitfields.
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
This adds PCIe and its PHY nodes for MT7623.
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
The driver provides PHY for USB2, USB3.0, PCIe and SATA, and now
we just enable PCIe. As for the other functionalities will be
added gradually in upcoming days.
This is adapted from the Linux version.
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
This adds PCIe controller support for MT7623.
This is adapted from the Linux version.
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
This commit add a generic function board_init_f that
only initialize some device (for example serial). It
avoid to define a board function only to launch the
serial configuration.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
This commit enable the support of the spi-nor for the
broadcom reference board bcm968580xref.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Kursad Oney <kursad.oney@broadcom.com>
This commit add a spi-nor device in the bcm96850xref device tree.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Kursad Oney <kursad.oney@broadcom.com>
This commit add a hsspi controller in the bcm6858 device tree.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Kursad Oney <kursad.oney@broadcom.com>
This commit enable the support of the spi-nor for the
broadcom reference board bcm963158.
Signed-off-by: Kursad Oney <kursad.oney@broadcom.com>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
This change adds a spi nor flash device to the bcm963158 board.
Signed-off-by: Kursad Oney <kursad.oney@broadcom.com>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
This change adds the hsspi controller to the 63158 dtsi.
Signed-off-by: Kursad Oney <kursad.oney@broadcom.com>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
The Broadcom ARM implementations do not yet have a clock framework so
one can use a fixed clock as the root clock of the hsspi block. The
fixed clock does not have an "enable" routine, since it's always
enabled. So when we hit this issue, getting an ENOSYS return, do not
bail but continue initialization.
Similarly the block might already have been out of reset, say, when
we are booting from a SPI device. So if the reset signal is not configured
in the device tree, do not bail out and instead skip deasserting the reset.
Signed-off-by: Kursad Oney <kursad.oney@broadcom.com>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
Make the driver compatible with both big and little endian SOCs.
Replace big-endian calls with their raw equivalents, expect for
writing the command to FIFO. That still has to be in big-endian
format.
Signed-off-by: Kursad Oney <kursad.oney@broadcom.com>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
wait_for_bit_le32 and wait_for_bit_le16 use the raw I/O functions
which would default to big-endian on BE systems. Create the generic
equivalents to use the native endianness.
Signed-off-by: Kursad Oney <kursad.oney@broadcom.com>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
This IP exists in both MIPS and ARM cores, so we also
allow to use this driver on bcm6858 and bcm63158.
Signed-off-by: Kursad Oney <kursad.oney@broadcom.com>
Reviewed-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
The A72 U-Boot code loads and boots a number of remote processors
including the C71x DSP, both the C66_0 and C66_1 DSPs, and the various
Main R5FSS Cores. In order to view the code loaded by the U-Boot by
remote cores, U-Boot should configure the memory region with right
memory attributes. Right now U-Boot carves out a memory region which
is not sufficient for all the images to be loaded. So, increase this
carve out region by 256MB.
Signed-off-by: Kedar Chitnis <kedarc@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
The A53 U-Boot code can load and boot the MCU domain R5F cores (either a
single core in LockStep mode or 2 cores in Split mode) to achieve various
early system functionalities. Change the memory attributes for the DDR
regions used by the remote processors so that the cores can see and
execute the proper code loaded by U-Boot.
These regions are currently limited to 0xa0000000 to 0xa2100000 as per
the DDR carveouts assigned for these R5F cores in the overall DDR memory
map.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
The A53 U-boot can support early booting of the MCU R5F remote processor(s)
from U-boot prompt to achieve various system usecases before booting the
Linux kernel. Update the default BOOTCOMMAND to provide an automatic and
easier way to start the MCU R5F cores through added environment variables.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Enable the R5F remoteproc driver for the AM65x GP EVM so that the
MCU domain R5F cores can be booted from A53 U-boot.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
The A72 U-boot can support early booting of any of the R5F or C66x
or C71x remote processors from U-boot prompt to achieve various system
usecases before booting the Linux kernel. Update the default BOOTCOMMAND
to provide an automatic and easier way to start various remote processors
through added environment variables.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
Add support to boot the MCU domain R5F Core0 remoteproc at U-boot prompt
on the AM65x EVM boards by using the 'boot_rprocs' and other env variables
defined in the common environment file k3_rproc.h, and updating the
'DEFAULT_RPROCS' macro.
The default configuration is to use the MCU R5F in Split mode, so both
the R5F Core0 and Core1 are started before loading and booting the Linux
kernel using the following firmware:
MCU R5FSS0 Core0 (Split) : 0 /lib/firmware/am65x-mcu-r5f0_0-fw
MCU R5FSS0 Core1 (Split) : 1 /lib/firmware/am65x-mcu-r5f0_1-fw
The MCU R5FSS was initially running the R5 SPL in LockStep mode with ATCM
disabled, and is actually shutdown to enable it to be reconfigured and
booted by either A53 U-Boot or Linux kernel in remoteproc mode and using
ATCM.
The MCU R5FSS would need to be reconfigured for Lockstep mode through
DT if a fault-tolerant/safety application were to be run on the cluster
with the DEFAULT_RPROCS macro updated to remove the Core1 firmware.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Add support to boot some remoteprocs at U-boot prompt on the J721E EVM
boards by using the 'boot_rprocs' and other env variables defined in the
common environment file k3_rproc.h, and updating the 'DEFAULT_RPROCS'
macro.
The list of R5F cores to be started before loading and booting the Linux
kernel are as follows, and in this order:
Main R5FSS0 (Split) Core1 : 3 /lib/firmware/j7-main-r5f0_1-fw
Main R5FSS1 (LockStep) : 4 /lib/firmware/j7-main-r5f1_0-fw
The MCU R5FSS0 and Main R5FSS1 are currently in LockStep mode, so the
equivalent Core1 rprocs (rproc #1 and #5) are not included. The Main
R5FSS0 Core0 (rproc #2) is already started by R5 SPL, so is not included
in the list either.
The DSP cores are started in the following order before loading and
booting the Linux kernel:
C66_0: 6 /lib/firmware/j7-c66_0-fw
C66_1: 7 /lib/firmware/j7-c66_1-fw
C71_0: 8 /lib/firmware/j7-c71_0-fw
The order of the rprocs to boot can be changed at runtime if desired by
overwriting the 'rproc_fw_binaries' environment variable at U-boot prompt.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Add a new file include/environment/ti/k3_rproc.h that defines
common environment variables useful for booting various remote
processors from U-Boot. This file is expected to be included in
the board config files with the EXTRA_ENV_RPROC_SETTINGS added
to CONFIG_EXTRA_ENV_SETTINGS and DEFAULT_RPROCS macro overwritten
to include the actual list of processors to be booted.
The 'boot_rprocs' variable just needs to be added to the board's
bootcmd to automatically boot the processors, and runtime control
can be achieved through the 'dorprocboot' variable.
The variables are currently defined to use MMC as the boot media,
and can be expanded in the future to include other boot media.
The immediate usage is intended for K3 J721E SoCs.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
The AM65x SoCs has a single dual-core Arm Cortex-R5F processor
subsystem/cluster (MCU_R5FSS0) within the MCU domain. This cluster
can be configured at boot time to be either run in a LockStep mode
or in an Asymmetric Multi Processing (AMP) fashion in Split-mode.
This subsystem has 64 KB each Tightly-Coupled Memory (TCM) internal
memories for each core split between two banks - ATCM and BTCM
(further interleaved into two banks). There are some IP integration
differences from standard Arm R5 clusters such as the absence of
an ACP port, presence of an additional TI-specific Region Address
Translater (RAT) module for translating 32-bit CPU addresses into
larger system bus addresses etc.
Add the DT node for the MCU domain R5F cluster/subsystem, the two
R5 cores are added as child nodes to the main cluster/subsystem node.
The cluster is configured to run in Split-mode by default, with the
ATCMs enabled to allow the R5 cores to execute code from DDR with
boot-strapping code from ATCM. The inter-processor communication
between the main A72 cores and these processors is achieved through
shared memory and Mailboxes.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
The J721E SoCs have a single TMS320C71x DSP Subsystem in the MAIN
voltage domain containing the next-generation C711 CPU core. The
subsystem has 32 KB of L1D configurable SRAM/Cache and 512 KB of
L2 configurable SRAM/Cache. This subsystem has a CMMU but is not
used currently. The inter-processor communication between the main
A72 cores and the C711 processor is achieved through shared memory
and a Mailbox. Add the DT node for this DSP processor sub-system
in the common k3-j721e-main.dtsi file.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
The J721E SoCs have two TMS320C66x DSP Core Subsystems (C66x CorePacs)
in the MAIN voltage domain, each with a C66x Fixed/Floating-Point DSP
Core, and 32 KB of L1P & L1D configurable SRAMs/Cache and an additional
288 KB of L2 configurable SRAM/Cache. These subsystems do not have
an MMU but contain a Region Address Translator (RAT) sub-module for
translating 32-bit processor addresses into larger bus addresses.
The inter-processor communication between the main A72 cores and
these processors is achieved through shared memory and Mailboxes.
Add the DT nodes for these DSP processor sub-systems in the common
k3-j721e-main.dtsi file.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
The J721E SoCs have 3 dual-core Arm Cortex-R5F processor (R5FSS)
subsystems/clusters. One R5F cluster (MCU_R5FSS0) is present within
the MCU domain, and the remaining two clusters are present in the
MAIN domain (MAIN_R5FSS0 & MAIN_R5FSS1). Each of these can be
configured at boot time to be either run in a LockStep mode or in
an Asymmetric Multi Processing (AMP) fashion in Split-mode. These
subsystems have 64 KB each Tightly-Coupled Memory (TCM) internal
memories for each core split between two banks - ATCM and BTCM
(further interleaved into two banks). There are some IP integration
differences from standard Arm R5 clusters such as the absence of
an ACP port, presence of an additional TI-specific Region Address
Translater (RAT) module for translating 32-bit CPU addresses into
larger system bus addresses etc.
Add the DT nodes for these two MAIN domain R5F cluster/subsystems,
the two R5 cores are each added as child nodes to the corresponding
main cluster node. Configure SS0 in split mode an SS1 in lockstep mode,
with the ATCMs enabled to allow the R5 cores to execute code from DDR
with boot-strapping code from ATCM.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
The J721E SoCs have 3 dual-core Arm Cortex-R5F processor (R5FSS)
subsystems/clusters. One R5F cluster (MCU_R5FSS0) is present within
the MCU domain, and the remaining two clusters are present in the
MAIN domain (MAIN_R5FSS0 & MAIN_R5FSS1). Each of these can be
configured at boot time to be either run in a LockStep mode or in
an Asymmetric Multi Processing (AMP) fashion in Split-mode. These
subsystems have 64 KB each Tightly-Coupled Memory (TCM) internal
memories for each core split between two banks - ATCM and BTCM
(further interleaved into two banks). There are some IP integration
differences from standard Arm R5 clusters such as the absence of
an ACP port, presence of an additional TI-specific Region Address
Translater (RAT) module for translating 32-bit CPU addresses into
larger system bus addresses etc.
Add the DT node for the MCU domain R5F cluster/subsystem, the two
R5 cores are added as child nodes to the main cluster/subsystem node.
The cluster is configured to run in LockStep mode by default, with the
ATCMs enabled to allow the R5 cores to execute code from DDR with
boot-strapping code from ATCM. The inter-processor communication
between the main A72 cores and these processors is achieved through
shared memory and Mailboxes.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Certain SoCs with K3 architecture have integrated a C66 Corepac DSP
subsystem and an advanced C71 DSPs. Introduce a remoteproc driver
that that does take care of loading an elf to any of the specified
DSPs and start it.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
Some Texas Instruments K3 family of SoCs have one of more Digital Signal
Processor (DSP) subsystems that are comprised of either a TMS320C66x
CorePac and/or a next-generation TMS320C71x CorePac processor subsystem.
Add the device tree bindings document for the C66x DSP devices on these
SoCs. The added example illustrates the DT nodes for the first C66x DSP
device present on the K3 J721E family of SoCs.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
SoCs with K3 architecture have an integrated Arm Cortex-R5F subsystem
that is comprised of dual-core Arm Cortex-R5F processor cores. This R5
subsytem can be configured at boot time to be either run in a LockStep
mode or in an Asymmetric Multi Processing (AMP) fashion in Split-mode.
This subsystem has each Tightly-Coupled Memory (TCM) internal memories
for each core split between two banks - TCMA and TCMB.
Add a remoteproc driver to support this subsystem to be able to load
and boot the R5 cores primarily in LockStep mode or split mode.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
The Texas Instruments K3 family of SoCs have one of more dual-core
Arm Cortex R5F processor subsystems/clusters (R5FSS). Add the device
tree bindings document for these R5F subsystem devices. These R5F
processors do not have an MMU, and so require fixed memory carveout
regions matching the firmware image addresses. The nodes require more
than one memory region, with the first memory region used for DMA
allocations at runtime. The remaining memory regions are reserved
and are used for the loading and running of the R5F remote processors.
The added example illustrates the DT nodes for the single R5FSS device
present on K3 AM65x family of SoCs.
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Power domain for the remote cores needs to be handled in a right
sequence as mandated by the spec. Introduce tisci helper apis
that can control power-domains of remote cores. TISCI clients
can use this api and control the remote cores power domain instead
of hooking it to power-domain layer.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Introduce rproc_elf_get_boot_addr() that returns the entry point of
the elf file. This api auto detects the 64/32 bit elf file and returns
the boot addr accordingly.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Introduce a common remoteproc elf loader and checker functions that
automatically detects the 64 bit elf file or 32 bit elf file and
loads/checks the sections accordingly.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Fabien Dessenne <fabien.dessenne@st.com>
The current rproc-elf-loader supports loading of only 32 bit elf files.
Introduce support for loading of 64 bit elf files in rproc-elf-loader.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Fabien Dessenne <fabien.dessenne@st.com>
rproc_elf32_load_image() rely on user to send a valid address for elf loading.
Instead do a sanity check on the address passed by user. This will help
all rproc elf users to not call sanity_check explicitly before calling
elf_loading.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Fabien Dessenne <fabien.dessenne@st.com>
Introduce a new parameter "size" that accepts size of the region to
remoteproc ops callback device_to_virt(). This can enforce more checks
on the region that device_to_virt() is dealing with.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Tested-by: Fabien Dessenne <fabien.dessenne@st.com>
Reviewed-by: Fabien Dessenne <fabien.dessenne@st.com>
