env: another attempt at fixing SPL build failures

I'm also seeing the build failure that commit

7d4776545b env: solve compilation error in SPL

tried to fix, namely that the reference to env_flags_validate from
env_htab cannot be satisfied when flags.o is not built in. However,
that commit got reverted by

d90fc9c3de Revert "env: solve compilation error in SPL"

Necessary, but not sufficient conditions to see this are

CONFIG_SPL=y (obviously)
CONFIG_SPL_ENV_SUPPORT=n (so flags.o does not get compiled)
CONFIG_SPL_LIBCOMMON_SUPPORT=y (so env/built-in.o is part of the SPL link)

Now, these are satisfied for e.g. imx6q_logic_defconfig. But that
builds just fine, and spl/u-boot-spl.map lists .data.env_htab among
the discarded (garbage collected) sections. Yet, on our
mpc8309-derived board, we do see the build failure, so perhaps the
linker works a bit differently on ppc than on ARM, or there's yet some
other configuration option needed to observe the break.

This is another attempt at solving it, which also cleans up
env/Makefile a bit: Introduce a def_bool y symbol CONFIG_ENV_SUPPORT
which complements CONFIG_(SPL/TPL)_SUPPORT. Then use
CONFIG_$(SPL_TPL_)ENV_SUPPORT to decide whether to include the five
basic env/*.o files. For attr.o, flags.o and callback.o, this
shouldn't change anything. Also, common.o and env.o still get
unconditionally built for U-boot proper. But for TPL/SPL, those two
are only included if CONFIG_(SPL/TPL)_SUPPORT is set.

Having that symbol should also allow simplifying conditionals such as

#if !defined(CONFIG_SPL_BUILD) || CONFIG_IS_ENABLED(ENV_SUPPORT)

found in drivers/reset/reset-socfpga.c to just
CONFIG_IS_ENABLED(ENV_SUPPORT).

Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>

configs: Resync with savedefconfig

Rsync all defconfig files using moveconfig.py

Signed-off-by: Tom Rini <trini@konsulko.com>

Merge https://gitlab.denx.de/u-boot/custodians/u-boot-marvell

- Clearfog: Add run-time board detection with TLV EEPROM support
  (Baruch)

ARM: mvebu: clearfog: add Clearfog Base serdes configuration

Clearfog Base carrier uses serdes lane #5 as USB host. Use EEPROM stored
device identification to configure the serdes accordingly when
available.

Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

ARM: mvebu: clearfog: add Clearfog GTR support

Select the serdes configuration table based on the platform identity
read from EEPROM TLV data. Clearfog GTR needs a slightly different
serdes configuration.

Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

ARM: mvebu: clearfog: run-time selection of DT file

Set the kernel device-tree file (fdtfile environment variable) based on
run-time detection of the platform.

Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

ARM: mvebu: clearfog: print TLV stored product name

Use the data from EEPROM TLV to display the board identity.

Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

ARM: mvebu: clearfog: read basic TLV data

Read RAM die capacity from the EEPROM TLV.

Follow the ONIE standard that defines the Vendor Extension entry type
for vendor specific data. We have no Private Enterprise Number at the
moment as the standard requires. Use the dummy all 0xff value for now.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Reviewed-by: Stefan Roese <sr@denx.de>

ARM: mvebu: clearfog: add support for EEPROM TLV info

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Reviewed-by: Stefan Roese <sr@denx.de>

cmd: add tlv_eeprom command

Add support for read/write of ONIE "Tlvinfo" EEPROM data format. TLV
stands for Type-Length-Value. The data format is described here:

  https://opencomputeproject.github.io/onie/design-spec/hw_requirements.html#board-eeprom-information-format

Based on U-Boot patch from the Open Compute project:

  https://github.com/opencomputeproject/onie/blob/ec87e872d46b9805565d2c6124b2f701ef1c07b1/patches/u-boot/common/feature-sys-eeprom-tlv-common.patch

Keep only I2C EEPROM support. Use the generic eeprom driver. Fix
checkpatch issues.

Add support for multiple EEPROM TLV stores on the same system. This is
useful in case of SOM and carrier that both provide ID and hardware
configuration information.

Add option to enable for SPL. This allows selection of RAM configuration
based on EEPROM stored board identification.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>

ARM: mvebu: clearfog: add EEPROM devices

Add device-tree description for the EEPROM devices on Armada 388 SOM rev
2.1.

Another EEPROM is now on Clearfog Pro carrier rev 2.2, and Clearfog Base
rev 1.3.

Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

Merge tag 'u-boot-imx-20200121' of https://gitlab.denx.de/u-boot/custodians/u-boot-imx

---------------------------------

imx8: cleanup, fix warnings
imx6ull: add VisionSOM SoM and EVK
mx7ulp: fix warning due network, cleanup
mx7dsabre: Fix dm probe pmic
imx6: fixed for vining2000

Travis: https://travis-ci.org/sbabic/u-boot-imx/builds/639512296

arm: mvebu: clearfog: enable both DDR clocks

Enabled both DDR clock signals to support Clearfog variants (currently,
Clearfog GTR) that need both clocks.

Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

ddr: marvell: a38x: allow board specific clock out setup

DDR clock out might be unrelated to the number of active chip-select.
For example, the board might have two DDR components, but only one
chip-select. The clk_enable mask allows the board to enable DDR clocks
regardless of active chip-selects.

Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

Merge tag '2020-01-20-ti-2020.04' of https://gitlab.denx.de/u-boot/custodians/u-boot-ti

K3 J721E:
* DMA support.
* MMC and ADMA support.
* EEPROM support.
* J721e High Security EVM support.
* USB DT nodes

K3 AM654:
* Fixed boot due to pmic probe error.
* USB support and DT nodes.
* ADMA support

DRA7xx/AM57xx:
* BBAI board support
* Clean up of net platform code under board/ti

AM33/AM43/Davinci:
* Reduce SPL size for omap3 boards.
* SPL DT support for da850-lcdk
* PLL divider fix for AM335x

Merge tag 'u-boot-stm32-20200117' of https://gitlab.denx.de/u-boot/custodians/u-boot-stm

- stm32mp1: split SOC and board and cleanup config

Merge branch '2020-01-17-reduce-size-of-common-h-even-more'

- Bring in Simon Glass's series that reduces what we have in <common.h>
  even more.

gitlab/azure: Update to latest container tag

- Based on bionic-20200112 tag from Ubuntu
- Add graphviz (Heinrich)

Signed-off-by: Tom Rini <trini@konsulko.com>

colibri-imx8x: Remove unused function detail_board_ddr_info()

detail_board_ddr_info() is not used anywhere, so simply remove it.

This function is only used by Layerscape, not by i.MX.

This was detected by the following sparse warning:

board/toradex/colibri-imx8x/colibri-imx8x.c:104:6: warning: no previous
prototype for ‘detail_board_ddr_info’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

apalis-imx8: Remove unused function detail_board_ddr_info()

detail_board_ddr_info() is not used anywhere, so simply remove it.

This function is only used by Layerscape, not by i.MX.

This was detected by the following sparse warning:

board/toradex/apalis-imx8/apalis-imx8.c:92:6: warning: no previous
prototype for ‘detail_board_ddr_info’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

imx8qxp_mek: Remove unused function detail_board_ddr_info()

detail_board_ddr_info() is not used anywhere, so simply remove it.

This function is only used by Layerscape, not by i.MX.

This was detected by the following sparse warning:

board/freescale/imx8qxp_mek/imx8qxp_mek.c:115:6: warning: no previous
prototype for ‘detail_board_ddr_info’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

imx8qm_mek: Remove unused function detail_board_ddr_info()

detail_board_ddr_info() is not used anywhere, so simply remove it.

This function is only used by Layerscape, not by i.MX.

This was detected by the following sparse warning:

board/freescale/imx8qm_mek/imx8qm_mek.c:102:6: warning: no previous
prototype for ‘detail_board_ddr_info’ [-Wmissing-prototypes]

imx8qm_rom7720_a1: Remove unused function detail_board_ddr_info()

detail_board_ddr_info() is not used anywhere, so simply remove it.

This function is only used by Layerscape, not by i.MX.

This was detected by the following sparse warning:

board/advantech/imx8qm_rom7720_a1/imx8qm_rom7720_a1.c:110:6: warning: no previous
prototype for ‘detail_board_ddr_info’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>
Reviewed-by: Oliver Graute <oliver.graute@kococonnector.com>

imx8qm_mek: Include fdt_support.h header

Include fdt_support.h header file to fix the following sparse warning:

board/freescale/imx8qm_mek/imx8qm_mek.c:116:5: warning: no previous
prototype for ‘ft_board_setup’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

imx8qxp_mek: Include fdt_support.h header

Include fdt_support.h header file to fix the following sparse warning:

board/freescale/imx8qxp_mek/imx8qxp_mek.c:129:5: warning: no previous
prototype for ‘ft_board_setup’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

spl: Add prototype to function spl_board_boot_device()

Add prototype to function spl_board_boot_device to fix the following
sparse warning:

board/freescale/imx8mm_evk/spl.c:26:5: warning: no previous prototype
for ‘spl_board_boot_device’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

mx8mm_evk: spl: Staticize functions

Functions spl_dram_init() and power_init_board() are used only in
the scope of this file, so make them static to fix the following sparse
warnings:

board/freescale/imx8mm_evk/spl.c:40:6: warning: no previous prototype
for ‘spl_dram_init’ [-Wmissing-prototypes]

board/freescale/imx8mm_evk/spl.c:85:5: warning: no previous prototype
for ‘power_init_board’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

imx8m: clock_imx8mm: Staticize functions

Functions fracpll_configure(), decode_intpll(), decode_fracpll(),
get_root_src_clk() and get_root_clk() are used only in the scope of this
file, so make them static to fix the following sparse warnings:

arch/arm/mach-imx/imx8m/clock_imx8mm.c:50:5: warning: no previous
prototype for ‘fracpll_configure’ [-Wmissing-prototypes]
arch/arm/mach-imx/imx8m/clock_imx8mm.c:271:5: warning: no previous
prototype for ‘decode_intpll’ [-Wmissing-prototypes]
arch/arm/mach-imx/imx8m/clock_imx8mm.c:418:5: warning: no previous
prototype for ‘decode_fracpll’ [-Wmissing-prototypes]
arch/arm/mach-imx/imx8m/clock_imx8mm.c:483:5: warning: no previous
prototype for ‘get_root_src_clk’ [-Wmissing-prototypes]
arch/arm/mach-imx/imx8m/clock_imx8mm.c:527:5: warning: no previous
prototype for ‘get_root_clk’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

imx: fix typo in header file

Signed-off-by: Flavio Suligoi <f.suligoi@asem.it>

fsl: fix typo in header file

Signed-off-by: Flavio Suligoi <f.suligoi@asem.it>

imx8mm_evk: Include sys_proto.h header

Add sys_proto.h header file to fix the following sparse warning:

board/freescale/imx8mm_evk/imx8mm_evk.c:59:5: warning: no previous
prototype for ‘board_mmc_get_env_dev’ [-Wmissing-prototypes]

Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>

arm: dts: imx7ulp-evk: remove mux value from pad configuration

The mux mode is embedded in the PAD definition and therefore there is
no need to repeat it in the PAD configuration value (more over since
this information will be masked out when the configuration value is
applied).

Signed-off-by: Jorge Ramirez-Ortiz <jorge@foundries.io>
Reviewed-by: Fabio Estevam <festevam@gmail.com>

imx: imx8mm_evk: remove hdmi fw info from README

The imx8mm doesn't require the hdmi firmware.
Update also the fw version.

Signed-off-by: Flavio Suligoi <f.suligoi@asem.it>
Reviewed-by: Peng Fan <peng.fan@nxp.com>

mx7dsabre: Fix dm probe pmic

With commit 0d52bab4621 ("mx7dsabre: Enable DM_ETH")

Device Tree has been update and change pfuze3000 node

Signed-off-by: Joris Offouga <offougajoris@gmail.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>

mx7ulp: soc: s_init should only be executed once

On SPL enabled systems, the current s_init code (wdog, clock and ldo
init) is executed twice (by SPL and u-boot). This is not necessary and
might lead to boot issues (ie, starting PMC1 when it is already running).

Signed-off-by: Jorge Ramirez-Ortiz <jorge@foundries.io>
Reviewed-by: Fabio Estevam <festevam@gmail.com>

mx7ulp_evk_plugin: Disable CONFIG_NET

Currently the following build warning is seen:

===================== WARNING ======================
This board does not use CONFIG_DM_ETH (Driver Model
for Ethernet drivers). Please update the board to use
CONFIG_DM_ETH before the v2020.07 release. Failure to
update by the deadline may result in board removal.
See doc/driver-model/migration.rst for more info.
===================================================

Since the mx7ulp-evk board does not have networking support, explicitly
disable networking.

Signed-off-by: Fabio Estevam <festevam@gmail.com>

arm: dts: i.mx8x: add #cooling-cells properties

Fix dtb building warnings:
Warning (cooling_device_property): /thermal-zones/cpu-thermal0/cooling-maps/map0:
Missing property '#cooling-cells' in node /cpus/cpu@0 or bad phandle (referred from cooling-device[0])

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Stefano Babic <sbabic@denx.de>
Reviewed-by: Fabio Estevam <festevam@gmail.com>

ARM: imx: mx6ull: Add iMX6ULL VisionSOM SoM and EVK

Add iMX6ULL VisionSOM SoM and VisionCB-RT-STD evaluation board support.
The SoM has an iMX6ULL, 512 MiB of DRAM and microSD slot. The carrier
board has Ethernet, USB host port, USB OTG port.

Signed-off-by: Arkadiusz Karas <arkadiusz.karas@somlabs.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Stefano Babic <sbabic@denx.de>

ARM: imx: vining2000: Enable SPL SDP by default

Enable SPL SDP fallback boot option in default build.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@nxp.com>
Cc: Silvio Fricke <silvio.fricke@softing.com>
Cc: Stefano Babic <sbabic@denx.de>

ARM: imx: vining2000: Properly discern PFUZE100 and PFUZE200

The PFUZE100 and PFUZE200 PMICs can be discerned by bit 0 in DeviceID
register. Print the correct identification of the PMICs.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@nxp.com>
Cc: Silvio Fricke <silvio.fricke@softing.com>
Cc: Stefano Babic <sbabic@denx.de>

ARM: imx: vining2000: Clean up uSDHC4 setup

Simplify the uSDHC4 eMMC controller setup. This is the only eMMC
present on the system and only controller that is used, so drop
the extra logic.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@nxp.com>
Cc: Silvio Fricke <silvio.fricke@softing.com>
Cc: Stefano Babic <sbabic@denx.de>
Reviewed-by: Stefano Babic <sbabic@denx.de>

Merge branch 'master' of git://git.denx.de/u-boot

Signed-off-by: Stefano Babic <sbabic@denx.de>

arm: omap: fix MPU DPLL divisor for 800MHz clock

In locked condition CLKOUT = CLKINP * [M / (N + 1)].

Signed-off-by: Dario Binacchi <dariobin@libero.it>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

clk: sci-clk: add slack to clk-set-rate passed to firmware

Add slack to the clock frequency parameters passed to firmware within
clk_set_rate. min-freq is changed to 0 and max-rate is changed to
ULONG_MAX. This fixes certain issues with pll clock rounding when the
firmware is not able to set the frequency exactly to the target, the
current implementation fails if the available frequency is even 1Hz off
the target. With the change, the firmware still tries its best to set
the frequency as close as possible to the target.

Reported-by: Vishal Mahaveer <vishalm@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

MAINTAINERS: Update ARM TI entry

Take over TI maintainership from Tom.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: am65x_evm: Enable USB keyboard as second stdin

Enable USB keyboard to be used as input device at U-Boot prompt. Both
serial and USB keyboard will be active inputs simultaneously.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

am65x_evm_a53_defconfig: Enable configs to support USB and DFU

Enable configs related to USB Host mode, Peripheral mode and DFU.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: am65x_evm: Add DFU related env variables

Add env variables that set up dfu_alt_info for MMC/EMMC/OSPI. This
is required to allow update of firmware on these media.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: dts: k3-am654-base-board: enable USB1

Add pinmux for USB1 and enable it as a peripheral port in U-Boot
specific dtsi since U-Boot does not support OTG.
Disable USB0 as its not available on the baseboard.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: dts: k3-am65-main: add USB support

Add support for USB0 and USB1 instances on the AM6 SoC.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

phy: omap-usb2-phy: Add support for AM654 USB2 PHY

AM654 SoC has USB2 PHY which is similar to existing USB2 PHYs on OMAP
SoCs. Add support for the same.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

phy: omap-usb2-phy: Fix warnings when built for ARM64

Below warning is seen when this driver is built for devices with 64 bit
physical address space.

drivers/phy/omap-usb2-phy.c: In function ‘omap_usb2_phy_probe’:
drivers/phy/omap-usb2-phy.c:187:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
   priv->phy_base = (void *)base;
                    ^
Fix this by using dev_read_addr_ptr() instead of dev_read_addr().

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

dwc3-generic: Add support for AM654 USB controller

AM654 has DWC3 USB controller that is very similar to other TI SoCs. Add
a new compatible to enable the same.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: am65x_evm: Add Support for ADMA

Add Support for ADMA in a53 and r5 defconfigs.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

ARM: dts: omap3/omap35 Torpedo and SOM-LV: Unify and shrink SPL dtb

None of these boards boot, but the solution appears to be the same.
All the boards have SPL that is too large.  With a few defconfig
options removed, these corresponding options can be removed from
their respective SPL dtb files.

This patch unifies the DM37/OMAP35 boards' -u-boot.dtsi files
to remove gpio's, i2c, bandgap, thermal zones, unneeded uarts, and
unneeded MMC nodes.

Signed-off-by: Adam Ford <aford173@gmail.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: omap3/35_logic and omap3/35_logic_somlv: Reduce SPL size

Currently the DM37 and OMAP35 boards do not boot due to SPL
bring too large.

SPL doesn't need GPIO, I2C nor MMC sector access since it uses
a FAT file system.

This patch unifies all these boards to remove these unused features
from their defconfigs

Signed-off-by: Adam Ford <aford173@gmail.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: Add configs for J721e High Security EVM

Add new defconfig files for the J721e High Security EVM.

These defconfigs are the same as for the non-secure part, except for:
    CONFIG_TI_SECURE_DEVICE option set to 'y'
    CONFIG_FIT_IMAGE_POST_PROCESS option set to 'y'
    CONFIG_SPL_FIT_IMAGE_POST_PROCESS option set to 'y'
    CONFIG_BOOTCOMMAND uses FIT images for booting

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: K3: Increase default SYSFW image size allocation

The memory allocated to store the FIT image containing SYSFW and board
configuration data is statically defined to the largest size expected.
This was 276000 bytes but now needs to be grown to 277000 to make room
for the slightly larger SYSFW image used on J721e High-Security devices.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: K3: Disable ROM configured firewalls

ROM configures certain firewalls based on its usage, which includes
the one in front of boot peripherals. In specific case of boot
peripherals, ROM does not open up the full address space corresponding
to the peripherals. Like in OSPI, ROM only configures the firewall region
for 32 bit address space and mark 64bit address space flash regions
as in-accessible.

When security-cfg is initialized by sysfw, all the non-configured
firewalls are kept in bypass state using a global setting. Since ROM
configured firewalls for certain peripherals, these will not be touched.
So when bootloader touches any of the address space that ROM marked as
in-accessible, system raises a firewall exception causing boot hang.

It would have been ideal if sysfw cleans up the ROM configured boot
peripheral firewalls. Given the memory overhead to store this
information provided by ROM and the boot time increase in re configuring
the firewalls, it is concluded to clean this up in bootloaders.

So disable all the firewalls that ROM doesn't open up the full address
space.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: Venkateswara Rao Mandela <venkat.mandela@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: K3: Fix header comment match AM6 specific file function

This file used to be the common location for K3 init when AM6 was the
only device, but common code was moved to common.c and this file became
AM6 specific, correct this header text.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: ti: Factor out call to 'args_mmc' into MMC common environment

Both 'loadfit' and 'mmcloados' start with a call to 'args_mmc' so this
can be factored out to before eithers only call site. This also allows us
to remove that call from 'loadfit', which should not have been calling it
anyway as that command should not be MMC specific. Without the call to
'args_mmc' the command 'loadfit' becomes just a call to 'run_fit' so
remove the indirection and call 'run_fit' directly, this removes the need
for 'loadfit' command (which was misnamed anyway). Drop it.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: j721e_evm.h: Sync J721e environment configuration with AM65x

Some of the environment configuration in AM65x is not available in
J721e due to additions on one but not the other. These two platforms
are similar enough these common definitions should be factored out
to a common area, prepare for this by synchronizing them.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: am57xx: env: Fix DFU variables

Commit 8502fe84a4fc ("configs: am57xx_evm: define
CONFIG_SPL_LOAD_FIT_ADDRESS for SPL-DFU") implements incorrect ifdef
logic, which leads to DFU variables absence in non-SPL environment. Fix
that in order to bring back DFU variables, by reflecting the logic in
include/configs/dra7xx_evm.h.

Fixes: 8502fe84a4fc ("configs: am57xx_evm: define CONFIG_SPL_LOAD_F...")
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: dts: k3-am654-r5-base-board: Fix power-domains for wkup_vtm0

wkup_vtm populates only 1 power-domain cell in it's node. But the
power-domain cell are defined as 2. Due to this the following warning
comes during build:

arch/arm/dts/k3-am654-r5-base-board.dtb: Warning (power_domains_property):
/interconnect@100000/interconnect@28380000/interconnect@42040000/
wkup_vtm@42050000:power-domains: property size (8) too small for cell size 2

Fix this by updating the power-domain cells.

Fixes: cfa6bd549c ("arm: dts: k3-am654-r5-base-board: Add VTM node")
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Keerthy <j-keerthy@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

power: regulator: tps62360_regulator: Convert ofdata_to_platdata to the missing probe

commit 29f7d05a347a ("dm: core: Move ofdata_to_platdata() call earlier")
introduces changes in the order of device_probe execution.
ofdata_to_platdata now comes before the probe function which resulted in
a deadlock and caused boot hang on AM6 devices.

Deadlock sequence: tps62360_regulator_ofdata_to_platdata --> i2c_get_chip
--> device_probe(tps62360) --> tps62360_regulator_ofdata_to_platdata

Hence convert ofdata_to_platdata to the missing probe function to fix the
hang.

Fixes: 22e8f18980d6 ("power: regulator: tps6236x: add support for tps6236x regulators")
Signed-off-by: Keerthy <j-keerthy@ti.com>
Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: mach-k3: security: Clean image out of cache before authentication

On K3 systems U-Boot runs on both an R5 and a large ARM cores (usually
A53 or A72). The large ARMs are coherent with the DMA controllers and
the SYSFW that perform authentication. And previously the R5 core did
not enable caches. Now that R5 does enable caching we need to be sure
to clean out any of the image that may still only be in cache before we
read it using external DMA for authentication.

Although not expected to happen, it may be possible that the data was
read back into cache after the flush but before the external operation,
in this case we must invalidate our stale local cached version.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: mach-k3: Warn when node to disable is not found

Not finding a node that we try to disable does not always need to be
fatal to boot but should at least print out a warning. Return error
from fdt_disable_node as it did fail to disable the node, but only
warn in the case of disabling the TRNG as this will not prevent boot.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

defconfigs: am43xx_hs_evm: Sync HS and non-HS defconfigs

Sync new additions to non-HS defconfig with HS defconfig.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

defconfigs: am335x_hs_evm: Sync HS and non-HS defconfigs

Sync new additions to non-HS defconfig with HS defconfig.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: j721e_evm_a72_defconfig: Enable I2C and EEPROM support

Enable I2C and EEPROM related configs for A72 SPL/U-Boot.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Andreas Dannenberg <dannenberg@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm64: dts: k3-j721e-common-proc-board: Fully enable wkup_i2c0 use

Make the wkup_i2c0 module usable across all stages of U-Boot by adding
the needed definitions including the associated pinmux definitions.

Signed-off-by: Andreas Dannenberg <dannenberg@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: j721e: Print board name and version during boot

Print the board name and ver along with the DT Model.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: j721e: Use EEPROM-based board detection

The TI J721E EVM system on module (SOM), the common processor board, and
the associated daughtercards have on-board I2C-based EEPROMs containing
board config data. Use the board detection infrastructure to do the
following:

1) Parse the J721E SOM EEPROM and populate items like board name, board
   HW and SW revision as well as board serial number into the TI common
   EEPROM data structure residing in SRAM scratch space
2) Check for presence of daughter card(s) by probing associated I2C
   addresses used for on-board EEPROMs containing daughter card-specific
   data. If such a card is found, parse the EEPROM data such as for
   additional Ethernet MAC addresses and populate those into U-Boot
   accordingly
3) Dynamically apply daughter card DTB overlays to the U-Boot (proper)
   DTB during SPL execution
4) Dynamically create an U-Boot ENV variable called name_overlays
   during U-Boot execution containing a list of daugherboard-specific
   DTB overlays based on daughercards found to be used during Kernel
   boot.

This patch adds support for the J721E system on module boards containing
the actual SoC ("J721EX-PM2-SOM", accessed via CONFIG_EEPROM_CHIP_ADDRESS),
the common processor board ("J7X-BASE-CPB"), the Quad-Port Ethernet
Expansion Board ("J7X-VSC8514-ETH"), the infotainment board
("J7X-INFOTAN-EXP") as well as for the gateway/Ethernet switch/industrial
expansion board ("J7X-GESI-EXP").

Signed-off-by: Andreas Dannenberg <dannenberg@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

ti: common: board_detect: Handle EEPROM probe more gracefully

Use dm_i2c_probe() rather than i2c_get_chip() when trying to access
board-detection EEPROM devices. This has the advantage of more gracefully
handling the case when the EEPROM is not present by allowing to exit the
function early rather than failing and outputting an error message on the
I2C transactions that follow.

Signed-off-by: Andreas Dannenberg <dannenberg@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: beagleboneai: enable in am57xx_evm_defconfig

Adding the configurations to the evm_defconfig file

Signed-off-by: Jason Kridner <jdk@ti.com>
Signed-off-by: Caleb Robey <c-robey@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: beagleboneai: add dts file

BeagleBoard.org BeagleBone AI is an open source hardware single board
computer based on the Texas Instruments AM5729 SoC featuring dual-core
1.5GHz Arm Cortex-A15 processor, dual-core C66 digital signal
processor (DSP), quad-core embedded vision engine (EVE), Arm Cortex-M4
processors, dual programmable realtime unit industrial control
subsystems and more. The board features 1GB DDR3L, USB3.0 Type-C, USB
HS Type-A, microHDMI, 16GB eMMC flash, 1G Ethernet, 802.11ac 2/5GHz,
Bluetooth, and BeagleBone expansion headers.

For more information, refer to:
https://beaglebone.ai

The corresponding patch against the mainline linux kernel can be found
at: https://patchwork.kernel.org/patch/11254903/

This patch introduces the BeagleBone AI device tree.

Note that the device use the "ti,tpd12s016" component which is software
compatible with "ti,tpd12s015". Thus we only use the latter driver.

Signed-off-by: Jason Kridner <jdk@ti.com>
Signed-off-by: Caleb Robey <c-robey@ti.com>
Cc: Robert Nelson <robertcnelson@gmail.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: beagleboneai: add initial support

These are necessities for beaglebone ai boot. There is the addition of
CONFIG_SUPPORT_EMMC_CONFIG to the Kconfig file. This is present upstream
but not in 19.01 yet.

Signed-off-by: Jason Kridner <jdk@ti.com>
Signed-off-by: Caleb Robey <c-robey@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: beagleboneai: IODELAY and pinmux

This patch configures the pinmux settings for the BeagleBone AI after
the emmc read has completed.

Signed-off-by: Jason Kridner <jdk@ti.com>
Signed-off-by: Caleb Robey <c-robey@ti.com>
Cc: Robert Nelson <robertcnelson@gmail.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: beagleboneai: emmc read changes

BeagleBoard.org BeagleBone AI rev A1 does not include a board
identifier I2C EEPROM due to a design oversight. These boards have
been put into production and are generally available now.

The board identifier information, however, has been included in the
second eMMC linear boot partition (/dev/mmcblk1boot1).

This patch works by:
* First, looking for a board identifier I2C EEPROM and if not found,
* Then seeing if the boot mode matches BeagleBone AI with eMMC in the
  boot chain to make sure we don't enable eMMC pinmuxes on boards
  that don't support it, and
* Finally, initializes the eMMC pins and reading the header.

Signed-off-by: Jason Kridner <jdk@ti.com>
Signed-off-by: Caleb Robey <c-robey@ti.com>
Cc: Robert Nelson <robertcnelson@gmail.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: j721e_evm: Add configs for environment in eMMC

Add config to save and read back environment from eMMC.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: j721e_evm_a72: Add Support for GPT partitions

Introduce a default GPT partition table for eMMC.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: j721e_evm: Add configs for ADMA Support

Add configs for ADMA Support.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: dts: k3-j721e-common-proc-board: Add pinmux for SD card

Add pinmux for sdhci1 node connected to the SD card.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

arm: dts: k3-j721e-common-proc-board: Remove voltage-ranges from sdhci nodes

voltage-ranges properties are NOP. Remove them.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

mmc: am654_sdhci: Add Support for configuring PHY in J721e

Add Support for writing to PHY registers for J721e. There are number of
differences between the J721e 8 bit PHY, J721e 4 bit PHY and AM654 PHY.
Create a driver_data structure with an ops and flags field and use the
flags field to indicate these differences. The differences are as
follows:

1. The J721e 4 bit instance PHY does not have a DLL. Introduce a
DLL_PRESENT flag to make sure that DLL related registers are accessed
only where they are present. Also add a separate set_ios_post()
callback.

2. The J721e 8 bit instance is not muxed with anything else inside the
SoC and hence the IOMUX_ENABLE filed does not exist. Add a flag which is
used to indicate the presence of this field.

3. The register field used to select DLL frequency is 3 bit wide in
J721e as compared to 2 bits in AM65x. Add another flag that
distinguishes these fields.

4. The strobe select field is 8 bit wide as compared to 4 bit wide for
AM65x. Add yet another flag to indicate this difference. Strobe select
is used only for HS400 speed mode, support for which has not been added
in AM65x.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

mmc: am654_sdhci: Get Xin clock by name

Get clk_xin by name instead of by index to avoid having to put clocks in
the same order in all devices.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

dma: ti: k3-udma: Fix build warnings when building for 32 bit platforms

Cast pointers properly so as to avoid warnings when driver is built for
32 bit platforms

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

dma: ti: k3-udma: Fix ring push operation for 32 bit cores

UDMA always expects 64 bit address pointer of the transfer descriptor in
the Ring. But on 32 bit cores like R5, pointer is always 32 bit in size.
Therefore copy over 32 bit pointer value to 64 bit variable before
pushing it over to the ring, so that upper 32 bits are 0s.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

dma: ti: k3-udma: Switch to exposed ring mode

Exposed ring mode works well with 32 bit and 64 bit cores without need
for Proxies for 32 bit cores. Therefore switch to exposed ring mode.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

dma: ti: k3-udma: Fix debug prints during enabling MEM_TO_DEV transfers

Fix up the debug prints that were dumping state of TCHAN RT registers to
use tchan for MEM_TO_DEV transfers.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

dma: ti: k3-udma: Remove coherency check for cache ops

Remove redundant coherency checks before calling cache ops in UDMA
driver. This is now handled in arch specific cache operation
implementation based on Kconfig option

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

soc: ti: k3-navss-ringacc: Get SYSFW reference from DT phandle

Instead of looking getting reference to SYSFW device using name which
is not guaranteed to be constant, use phandle supplied in the DT node to
get reference to SYSFW

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

soc: ti: k3-navss-ringacc: Flush/invalidate caches on ring push/pop

Flush caches when pushing an element to ring and invalidate caches when
popping an element from ring in Exposed Ring mode. Otherwise DMA
transfers don't work properly in R5 SPL (with caches enabled) where the
core is not in coherency domain.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

dma: ti: k3-udma: Query DMA channels allocated from Resource Manager

On K3 SoCs, DMA channels are shared across multiple entities, therefore
U-Boot DMA driver needs to query resource range from centralised
resource management controller i.e SystemFirmware and use DMA channels
allocated for A72 host. Add support for the same.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

lib: Import few bitmap functions from Linux

Import few basic bitmap functions (bitmap_{weight,fill,set,clear,or}())
and their dependencies from Linux. These are required for upcoming DMA
resource allocation support for TI's K3 SoCs.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: am43xx: remove net platform code

The TI AM43xx platform has DM_ETH and OF_CONTROL enabled,
so remove networking platform code.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

board: ti: dra7-evm: remove net platform code

The DRA7 has DM_ETH and OF_CONTROL enabled, so remove networking platform
code.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

ARM: dts: da850-lcdk: Update DTS files for SPL device tree support

Currently, the da850-lcdk uses SPL_OF_PLATDATA and manually loads
the necessary source code instead of using the auto-generated,
because the drivers don't properly autogenerate the code.

This patch simply enables the various device tree options to
mimic the da850-evm which doesn't need or use OF_PLATDATA for
device tree support.  It does not disable OF_PLATDATA.

Signed-off-by: Adam Ford <aford173@gmail.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

thermal: ti-bandgap: Fix adc value datatype

The CORE_TEMP_SENSOR_MPU register gives a raw adc value which needs to
be indexed into a lookup table to get the actual temperature. Fix the
naming and datatype of the adc value variable.

Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>

configs: j721e_evm_a72_defconfig: Enable USB related configs

Enable USB host and device related configs.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>