doc: rockchip: Add phyCORE-RK3288 RDK to board list

Signed-off-by: Wadim Egorov <w.egorov@phytec.de>
Acked-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: Add basic support for phyCORE-RK3288 SoM based carrier board

The phyCORE-RK3288 is a SoM (System on Module) containing a RK3288 SoC.
The module can be connected to different carrier boards.
It can be also equipped with different RAM, SPI flash and eMMC variants.
The Rapid Development Kit option is using the following setup:

  - 1 GB DDR3 RAM (2 Banks)
  - 1x 4 KB EEPROM
  - DP83867 Gigabit Ethernet PHY
  - 16 MB SPI Flash
  - 4 GB eMMC Flash

Add basic support for the PCM-947 carrier board, a RK3288 based development
board made by PHYTEC. This board works in a combination with
the phyCORE-RK3288 System on Module.

Signed-off-by: Wadim Egorov <w.egorov@phytec.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

power: regulator: rk8xx: Allow input current/charger shutdown configuration

The RK818 PMIC contains a charger. Add very basic charger functionality
to be able to regulate the USB input current and charger shutdown limits.

Signed-off-by: Wadim Egorov <w.egorov@phytec.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

power: regulator: rk8xx: Build get_ldo_reg only for SPL

Enabling CONFIG_SPL_POWER_SUPPORT will cause a compiler warning:
  ‘get_ldo_reg’ defined but not used [-Wunused-function]

Let's wrap get_ldo_reg(), rk808_ldo and rk818_ldo with ENABLE_DRIVER
which is only set for non SPL builds.

Signed-off-by: Wadim Egorov <w.egorov@phytec.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dts: rk3399: control vbus of typec by fixed regulator

Add fixed regulator for the port of typec0 and typec1 to control vbus
instead of gpio.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dts: rk3328: support and enable dwc2

Enable dwc2 controller and add fixed regulator for dwc2 controller to
control vbus.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3328: board: add support of dwc2 gadget

Probe dwc2 udc in the function of board_usb_start to enable
usb gadget function.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

usb: dwc2: use dev_read_bool() instead of fdt_getprop()

Use dev_read_bool() instead of fdt_getprop() to get the property
from DTS. And add a comment for "hnp-srp-disable" property to
fully describe its effect.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dts: rk3328: add fixed regulator node for xhci

The driver changes gpio to fixed regulator to control vbus, so add
fixed regulator node in DTS for xhci driver.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

usb: Kconfig: migrate USB_DWC2 to Kconfig

This change migrates the USB_DWC2 configuration item to Kconfig
and runs moveconfig to adjust header files and defconfig.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Split off into a separate patch:
Ran moveconfig to migrate other boards:
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dts: rk3328: add fixed regulator node for xhci

The driver changes gpio to fixed regulator to control vbus, so add
fixed regulator node in DTS for xhci driver.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

usb: host: xhci-rockchip: use fixed regulator to control vbus

Use fixed regulator to control the voltage of vbus. Enable vbus
supply when usb start and disable vbus supply when usb stop.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

usb: Kconfig: config USB_XHCI_ROCKCHIP depends on DM_REGULATOR and DM_USB

The xhci-rockchip driver depends on DM_REGULATOR and DM_USB.
So add dependent features for xhci-rockchip driver in Kconfig.

Signed-off-by: Meng Dongyang <daniel.meng@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3036: sync os_reg2 define with other soc

Rockchip using the same bit definition for dram info and write
to os_reg, the col and bw info is not correct and let's fix it.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3036 remove CONFIG_RAM from defconfig

rk3036 sdram driver does not use DM, remove CONFIG_RAM first.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: video: mipi: Modify format type for debug message

Modify format type for debug message.

Signed-off-by: Eric Gao <eric.gao@rock-chips.com>

rockchip: video: mipi: Modify variable type for arm32 compatibility

Some address relevant varibable is defined originally as u64. To
compatible with arm32, this patch change them to uintptr_t type.

Signed-off-by: Eric Gao <eric.gao@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: pwm: fix the register layout for the PWM controller

According to rk3288 spec, the pwm register order is:
    PWM_PWM0_CNT,
    PWM_PWM0_PERIOD_HPR,
    PWM_PWM0_DUTY_LPR,
    PWM_PWM0_CTRL
but the source code's order is:
  struct rk3288_pwm {
    u32 cnt;
    u32 duty_lpr;
    u32 period_hpr;
    u32 ctrl;
  };

So, correct it here. It is the same as RK3399.

Signed-off-by: Eric Gao <eric.gao@rock-chips.com>
Edited the commit message:
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

Revert "mmc: dw_mmc: rockchip: select proper card clock"

The origin patch get rockchip dwmmc by name 'ciu', which lead
to the SPL not able to remove 'clock-names' node in dts.
I'm not saying this is not correct, but I would prefer to handle
this in dts or clock driver to save memory for SPL.
For example the rk3288 SPL size has out of memory if not enable
BACK_TO_BROM option, there are many other SoCs has less internal
memory than rk3288.

This reverts commit 480a9b834c661fc88217f50bdf780192c3018d66.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: firefly-rk3399: enable dwmmc driver for the board

Enable mmc_dw_rockchip driver, disable CONFIG_SPL_OF_PLATDATA
first for some dependence patches still not merged.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Tested-by: Vagrant Cascadian <vagrant@debian.org>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: firefly-rk3399: dts: enable sdmmc device

Enable sdmmc device and add the spl boot device sequence.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dwmmc: use max-frequency when OF_PLATDATA enabled

Since the 'clock-freq-min-max' is deprecated, we use max-frequency for
all rockchip SoC dwmmc controller.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3328: dtsi use max-frequency for mmc node

Since the 'clock-freq-min-max' is deprecated, we use max-frequency.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3288: dtsi use max-frequency for mmc node

Since the 'clock-freq-min-max' is deprecated, we use max-frequency.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3036: dtsi use max-frequency for mmc node

Since the 'clock-freq-min-max' is deprecated, we use max-frequency.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: add evb_rk3229 board

evb_rk3229 is a RK3229 based board, with:
- 8GB eMMC;
- 1GB DDR SDRAM;
- 2 USB2.0 HOST port;
- 1 MAC port;
- 1 HDMI port;
- IR;
- WiFi;

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk322x: add sysreset driver

Rockchip rk322x sysreset is much like rk3036 and other Rockchip SoCs,
only difference is that the target register address is different.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk322x: add basic soc support

Enable soc support for SPL and U-boot skeleton.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk322x: add dts file

The dts files are from kernel and with modify to adapt U-Boot.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk322x: add pinctrl driver

Add init pinctrl driver support for:
- i2c;
- spi;
- uart;
- pwm;
- emmc/sdmmc;

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk322x: add clock driver

Add clock driver init support for:
- cpu, bus clock init;
- emmc, sdmmc clock;
- ddr clock;

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Fixed format specified (%x -> %p) in clk_rk322x.c:
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: mkimage: add support for rk322x soc

Add support for rk322x package header in mkimage tool.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rv1108: disable CONFIG_RAM before we have driver

The rv1108 do not have DRAM driver now, so disable it first,
or else it will get conflict with the sdram common code.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: correct the bank0 ram size

The bank0 ram size should be the DRAM size minus reserved size,
the DRAM size may be 1GB, 2GB, 4GB, we can not hard code it.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Added DECLARE_GLOBAL_DATA_PTR for RK3328, RK3368 and RK3399:
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dts: rk3368: add dmc node

Add dmc node to enable sdram driver.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dts: rk3328: add dmc node

Add a dmc node for sdram driver.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3368: add sdram driver for U-Boot

Add sdram driver in U-Boot for get the correct sdram size from
sys_reg, so that U-Boot can co-work with Rockchip loader or SPL
to get different dram capability and then tell the kernel.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: rk3328: add sdram driver in U-Boot

Add sdram driver in U-Boot for get the correct sdram size from
sys_reg, so that U-Boot can co-work with Rockchip loader or SPL
to get different dram capability and then tell the kernel.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: use common sdram function

Replace the sdram_init() in board init and rockchip_sdram_size() in
sdram driver for all the Rockchip SoCs which enable CONFIG_RAM.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Make dram_init() in rk3036-board.c conditional on CONFIG_RAM:
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: add sdram_common for common functions

There are some functions like sdram_size_mb can be re-used for
different rockchip SoCs, just put them into common file.
Add board_get_usable_ram_top() for ram_top init base on
SDRAM_MAX_SIZE.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Added SDRAM_MAX_SIZE definition for RK3036:
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
fixup: 3036 fix for sdram_common

rockchip: rk3328: correct mem_region

According to rk3328 TRM:
0~0xff000000 is ddr space;
0xff000000~0xffffffff is device space.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>

rockchip: dm: convert fdt_get to dev_read

With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.

This covers the DRAM controller initialisation for the RK3188, RK3288
and RK3399... all of these read some of the tuning/setup/timing
parameters from the device-tree.

Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: ns16550: dm: convert fdt_get to dev_read

With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.

This covers the serial driver (ns16550 and compatible) used for the
Rockchip devices.

Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: net: dm: convert fdt_get to dev_read

With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.

This covers the Gigabit Ethernet MAC (i.e. common designware driver and
rockchip-specific wrapper).

Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: mmc: dm: convert fdt_get to dev_read

With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.

This covers the dw_mmc and sdhci wrapper drivers for Rockchip.

Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: xhci: dm: convert fdt_get to dev_read

With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.

This change covers the USB3 (xhci) driver for the Rockchip devices.

Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: spi: dm: convert fdt_get to dev_read

With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.

This change covers the rk_spi.c (SPI driver) used in Rockchip devices.

Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

rockchip: pinctrl: dm: convert fdt_get to dev_read

With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.

This change covers the pinctrl drivers for the Rockchip devices.

Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

Prepare v2017.07

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

km/common: remove unused function declarations

Cc: Wolfgang Denk <wd@denx.de>
Cc: Heiko Schocher <hs@denx.de>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>

cmd, nand: fix broken output of "nand info"

If the value of either "nand options" or "bbt options" has a zero in
the most significant nibble, the '0x' prefix will be isolated from the
value like shown below:
|Device 0: nand0, sector size 128 KiB
|  Page size       2048 b
|  OOB size          64 b
|  Erase size    131072 b
|  subpagesize     2048 b
|  options     0x40000200
|  bbt options 0x   60000

Change the format string to produce leading zeroes filling the gap.

Signed-off-by: Lothar Waßmann <LW@KARO-electronics.de>

libfdt: Drop -FDT_ERR_TOODEEP

This error code has not been upstreamed and is not really needed since it
is unlikely to be triggered. Drop it to maintain compatability with
upstream.

Reported-by: Peter Robinson <pbrobinson@gmail.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Peter Robinson <pbrobinson@gmail.com>

configs: Resync defconfigs

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

MAINTAINERS: drop bcm283x/rpi maintainership

It's been a while since I've touched U-Boot on the Raspberry Pi and
other things have been taking my time. Drop my maintainership for this
port.

Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>

powerpc, 8xx: Add support for MCR3000 board from CSSI

CS Systemes d'Information (CSSI) manufactures two boards, named MCR3000
and CMPC885 which are respectively based on MPC866 and MPC885 processors.

This patch adds support for the first board.

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>

powerpc, 8xx: move Serial driver to drivers/serial/

At the same time, move to Kconfig

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>

powerpc, 8xx: move SPI driver to drivers/spi/

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>

powerpc, 8xx: move FEC Ethernet driver in drivers/net

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>

powerpc, 8xx: Migrate to Kconfig

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Heiko Schocher <hs@denx.de>

powerpc, 8xx: Properly set CPM frequency in the device tree

For processors whose core runs at twice the bus frequency,
the fallback frequency calculation in Linux provides a wrong
result. Therefore, U-boot needs to pass the correct value.

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Heiko Schocher <hs@denx.de>

powerpc, 8xx: Handle checkpatch errors and some of the warnings/checks

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Heiko Schocher <hs@denx.de>

powerpc, 8xx: Implement GLL2 ERRATA

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Acked-by: Wolfgang Denk <wd@denx.de>
Reviewed-by: Heiko Schocher <hs@denx.de>

powerpc, 8xx: Use IO accessors to access IO memory

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Heiko Schocher <hs@denx.de>

powerpc, 8xx: move specific reginfo

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Heiko Schocher <hs@denx.de>

powerpc, 8xx: move immap.c in arch/powerpc/cpu/mpc8xx/

immap.c used to be common to several CPUs. It is now
only linked to the 8xx, so this patch moves it into
arch/powerpc/cpu/mpc8xx/

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Heiko Schocher <hs@denx.de>

powerpc: Partialy restore core of mpc8xx

CS Systemes d'Information (CSSI) manufactures 8xx boards for
critical communication systems. Those boards have been
running U-Boot since 2010 and will have to be maintained
until at least 2027.

commit 5b8e76c35ec312a3f73126bd1a2d2c0965b98a9f
("powerpc, 8xx: remove support for 8xx") orphaned those boards
by removing support for the mpc8xx CPU.

This commit partially restores support for the 8xx, with the
following limitations:
- Restores support for MPC866 and MPC885 only
- Does not restore IDE, PCMCIA, I2C, USB
- Does not restore examples
- Does not restore POST
- Does not restore Ethernet on SCC
- Does not restore console on SCC
- Does not restore bedbug and kgdb support

As the 866 and 885 do not support the following features,
they are not restored either:
- VIDEO / LCD
- RTC clock

The CPM uCODE patch is not restored either, because:
- 866 and 885 already have support for I2C and SPI relocation
without a uCODE patch
- relocation of SMC, I2C or SPI is only needed for using SCCs
for Ethernet or QMC

The dynamic setup/calculation of clocks is removed, we
expect the target being use with the clock and PLPRCR register
defined in the configuration.
All the clock settings for 8xx prior to 866 is removed as
well as we now only support 866 and 885.

This code is mature and addresses mature boards. Therefore
all code enclosed in '#if 0/#endif' and '#if XX_DEBUG/#endif'
is unneeded.

The following files are not restored by this patch:

- arch/powerpc/cpu/mpc8xx/bedbug_860.c
- arch/powerpc/cpu/mpc8xx/fec.h
- arch/powerpc/cpu/mpc8xx/kgdb.S
- arch/powerpc/cpu/mpc8xx/plprcr_write.S
- arch/powerpc/cpu/mpc8xx/scc.c
- arch/powerpc/cpu/mpc8xx/upatch.c
- arch/powerpc/cpu/mpc8xx/video.c
- arch/powerpc/include/asm/status_led.h
- arch/powerpc/lib/ide.c
- arch/powerpc/lib/ide.h
- doc/README.MPC866
- drivers/pcmcia/mpc8xx_pcmcia.c
- drivers/rtc/mpc8xx.c
- drivers/usb/gadget/mpc8xx_udc.c
- drivers/video/mpc8xx_lcd.c
- examples/standalone/test_burst.c
- examples/standalone/test_burst.h
- examples/standalone/test_burst_lib.S
- examples/standalone/timer.c
- include/mpc823_lcd.h
- include/usb/mpc8xx_udc.h
- post/cpu/mpc8xx/Makefile
- post/cpu/mpc8xx/cache.c
- post/cpu/mpc8xx/cache_8xx.S
- post/cpu/mpc8xx/ether.c
- post/cpu/mpc8xx/spr.c
- post/cpu/mpc8xx/uart.c
- post/cpu/mpc8xx/usb.c
- post/cpu/mpc8xx/watchdog.c

Some of the restored files are not located in a proper location.
In order to keep traceability of the changes, they will be
moved to their correct location and moved to Kconfig in a
followup patch.

This patch also declares CSSI as point of contact for the update
of the 8xx platform, as those boards are the only ones still
being maintained on the 8xx area. A later patch will add
those boards to the tree.

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>

am335x: sl50: Enable CONFIG_AUTOBOOT_KEYED

On startup the SL50 board halt at U-Boot prompt. Use CONFIG_AUTOBOOT_KEYED
to enable autoboot for this board and define the <SPACE> key to get the
U-Boot prompt.

Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>

board/BuR/brppt1: fix MMC boot

since commit
'd5abcf94c7123167725fc22ace342f0d455093c1' -
ti: boot: Register the MMC controllers in SPL in the same way as in u-boot

MMC boot on brppt1 board is broken, with this commit we make our board
    working again.

Signed-off-by: Hannes Schmelzer <hannes.schmelzer@br-automation.com>

wdt: Unify option of timeout value

There is no need to duplicate same option with different name.

Kill HW_WATCHDOG_TIMEOUT_MS in favor of WATCHDOG_TIMEOUT_MSECS.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Heiko Schocher <hs@denx.de>

wdt: Fix spelling Resettting -> Resetting

Fix spelling Resettting -> Resetting.
No functional change intended.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Heiko Schocher <hs@denx.de>

avr32: Retire AVR32 for good

AVR32 is gone. It's already more than two years for no support in Buildroot,
even longer there is no support in GCC (last version is heavily patched 4.2.4).

Linux kernel v4.12 got rid of it (and v4.11 didn't build successfully).

There is no good point to keep this support in U-Boot either.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

stm32: Correct positioning of declaration

The current code gives a warning:

arch/arm/mach-stm32/stm32f7/soc.c: In function 'arch_cpu_init':
arch/arm/mach-stm32/stm32f7/soc.c:38:2: error: 'for' loop initial
    declarations are only allowed in C99 or C11 mode
  for (int i = 0; i < ARRAY_SIZE(stm32_region_config); i++)
  ^
arch/arm/mach-stm32/stm32f7/soc.c:38:2: note: use option -std=c99,
    -std=gnu99, -std=c11 or -std=gnu11 to compile your code

Fix it by moving the declaration to the top of the function.

Signed-off-by: Simon Glass <sjg@chromium.org>
Series-cc trini

km/common: remove unused code

The 82xx board mgcoge3ne was removed from the codebase, so this is dead
code.

Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
Cc: Heiko Schocher <hs@denx.de>
Reviewed-by: Heiko Schocher <hs@denx.de>

configs: Odroid C2: enable device tree overlays

The Odroid C2 comes with several expansion options such as an RTC module.
As these expansions are not considered in the Linux device tree they have
to be enabled via device tree overlays.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>

mkimage: fix display of image types list

Since commit 5b9d44df2307f (mkimage: Display a better list of available image
types) mkimage usage text suggest to "use -T to see a list of available image
types". Unfortunately, commit 02221f29deb8 (mkimage: Convert to use getopt())
broke that feature, because getopt() fails when -T has no option argument.

Add a pseudo image type name 'list' that lists all image types. Update the
usage text accordingly.

Cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>

CHIP: add device tree overlay support

CHIP and CHIP Pro devices have options of DIP addon boards some of which need
overlays from the beginning so it makes sense to enable device tree overlays.

Signed-off-by: Peter Robinson <pbrobinson@gmail.com>

96boards: dragonboard/hikey: add device tree overlay support

96boards CE devices such as the DragonBoard and Hikey devices have numerous
mezzanine options some of which need overlays from the beginning so it makes
sense to enable device tree overlays.

Signed-off-by: Peter Robinson <pbrobinson@gmail.com>

am335x: beagle/evm: add device tree overlay support

TI am33xx devices such as the BeagleBone devices have numerous cape options such
as screens some of which need overlays from the beginning so it makes sense to
enable evice tree overlays.

Signed-off-by: Peter Robinson <pbrobinson@gmail.com>

Raspberry Pi: add device tree overlay support

Raspberry Pi has numerous HAT and other HW expansion options such as screens and
cameras some of which need overlays from the beginning so it makes sense to
enable evice tree overlays.

Signed-off-by: Peter Robinson <pbrobinson@gmail.com>

arm64: use psci reset on snapdragon

This actually works on snapdragon.. not sure why we weren't using it.
Fixes reboot/poweroff when using UEFI.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Alexander Graf <agraf@suse.de>

igep003x: Falcon mode

Implement spl_start_uboot to let Falcon mode work.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Acked-by: Heiko Schocher <hs@denx.de>

igep00x0: Enable UBI fastmap

Fastmap significantly reduces flash scan time, enable it by default.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Acked-by: Heiko Schocher <hs@denx.de>

spl: Make UBI fastmap support Kconfig selectable

Fastmap was always enabled in ubispl, make it selectable by
CONFIG_MTD_UBI_FASTMAP.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Acked-by: Heiko Schocher <hs@denx.de>

odroid-c2: Populate serial# environment variable from efuse

gpio: add static to get_function()

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

serial: make serial_stub_* to static functions

Add missing static to serial_stub_puts().

Unexport serial_stub_{getc,tstc} because they are used locally.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

Merge tag 'signed-efi-next' of git://github.com/agraf/u-boot

Patch queue for efi - 2017-07-04

Highlights this time: bugfixes. With these changes, OpenBSD should
be more happy.

efi_loader: abort on unsupported relocation type

If a relocation type is not supported loading the EFI binary
should be aborted.

Writing a message only is insufficient.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
[agraf: use a() != b coding style]
Signed-off-by: Alexander Graf <agraf@suse.de>

efi_loader: Add efi-next git tree to MAINTAINERS file

The efi-next tree lives on github, not the usual denx git.
Reflect this in the MAINTAINERS file so that people can
find it.

Signed-off-by: Alexander Graf <agraf@suse.de>

efi_loader: efi_handle_protocol set attributes

UEFI spec 2.7 indicates that HandleProtocol can be implemented
by calling OpenProtocol with
attributes = EFI_OPEN_PROTOCOL_BY_HANDLE_PROTOCOL.

Currently we pass attributes = 0 to efi_open_protocol. 0 is not a
valid value when calling OpenProtocol. This does not cause any errors
yet because our implementation of OpenProtocol is incomplete.

We should pass the correct value to enable a fully compliant
implementation of OpenProtocol in the future.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>

Prepare v2017.07-rc3

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

powerpc: remove 4xx support

There was for long time no activity in the 4xx area.
We need to go further and convert to Kconfig, but it
turned out, nobody is interested anymore in 4xx,
so remove it.

Signed-off-by: Heiko Schocher <hs@denx.de>

drivers, block: remove sil680 driver

driver is not used anymore, so remove it.

Signed-off-by: Heiko Schocher <hs@denx.de>

efi_loader: add static to local functions

These are locally used in lib/efi_loader/efi_boottime.c

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>

efi_loader: check CreateEvent() parameters

Add some of the invalid parameter checks described in the UEFI
specification for CreateEvent().  This does not include checking
the validity of the type and tpl parameters.

Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Acked-By: Heinrich Schuchardt <xypron.glpk@gmx.de>
[agraf: fix checkpatch.pl indent warning]
Signed-off-by: Alexander Graf <agraf@suse.de>

efi_loader: run CreateEvent() notify function based on flags

The UEFI specification states that the tpl, function and context
arguments are to be ignored if neither EVT_NOTIFY_WAIT or
EVT_NOTIFY_SIGNAL are specified.  This matches observed behaviour with
an AMI EDK2 based UEFI implementation.

Skip calling the notify function if neither flag is present.

Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Acked-By: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>

efi_loader: Add check for fallback fdt memory reservation

When running bootefi, we allocate new space but never check whether
the allocation succeeded. This patch adds a check so that in case
things go wrong, we at least know they did.

Signed-off-by: Alexander Graf <agraf@suse.de>

pico-imx7d: Remove bouncing email

Wig Cheng's email bounces, so remove it from the maintainers list.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>

doc: restore doc/README.fsl-clk

doc/README.fsl-clk was removed in commit
5b8e76c35ec31 ("powerpc, 8xx: remove support for 8xx")
allthought CONFIG_SYS_FSL_CLK is defined in
arch/arm/cpu/armv8/fsl-layerscape/Kconfig and still in use
in the following configs:

./include/configs/mx53loco.h:21:#define CONFIG_SYS_FSL_CLK
./include/configs/m53evk.h:16:#define CONFIG_SYS_FSL_CLK
./include/configs/mx25pdk.h:17:#define CONFIG_SYS_FSL_CLK
./include/configs/usbarmory.h:14:#define CONFIG_SYS_FSL_CLK
./include/configs/ls1021aqds.h:14:#define CONFIG_SYS_FSL_CLK
./include/configs/mx53cx9020.h:22:#define CONFIG_SYS_FSL_CLK
./include/configs/colibri_vf.h:17:#define CONFIG_SYS_FSL_CLK
./include/configs/mx35pdk.h:21:#define CONFIG_SYS_FSL_CLK
./include/configs/woodburn_common.h:19:#define CONFIG_SYS_FSL_CLK
./include/configs/mx7_common.h:25:#define CONFIG_SYS_FSL_CLK
./include/configs/ls1021aiot.h:12:#define CONFIG_SYS_FSL_CLK
./include/configs/ls1021atwr.h:14:#define CONFIG_SYS_FSL_CLK
./include/configs/mx53ard.h:21:#define CONFIG_SYS_FSL_CLK
./include/configs/mx53smd.h:21:#define CONFIG_SYS_FSL_CLK
./include/configs/mx51evk.h:16:#define CONFIG_SYS_FSL_CLK
./include/configs/mx6_common.h:31:#define CONFIG_SYS_FSL_CLK
./include/configs/vf610twr.h:14:#define CONFIG_SYS_FSL_CLK
./include/configs/mx53evk.h:21:#define CONFIG_SYS_FSL_CLK

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Heiko Schocher <hs@denx.de>

.travis.yml: All DENX boards are now under Aries

DENX hardware is now under Aries Embedded, update the job.

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