Rename rest of defconfig files of UniPhier SoC family to have the
prefix uniphier_.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

These boards are similar enough to be supported in a single defconfig
file.  Distinguish one from another by "DEVICE_TREE" from the command
line.  The how-to-build in doc/README.uniphier should be also updated.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

These two are similar enough to be merged into a single
defconfig file.  Distinguish one from another by "DEVICE_TREE"
from the command line.  The how-to-build in doc/README.uniphier
should be also updated.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Now this environment is run-time set to the DTB name U-Boot is really
running with.  Drop the static define.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

When we want to boot Linux with a DTB file downloaded from a TFTP
server or somewhere, we need to know the file name to be downloaded.

Assume the U-Boot configuration is shared among some similar boards.
If they are similar enough, the difference only appears in device
trees.  The build procedure would be like this:

 - Board A:  make foo_common_defconfig && make DEVICE_TREE=foo_board_a
 - Board B:  make foo_common_defconfig && make DEVICE_TREE=foo_board_b
 - Board C:  make foo_common_defconfig && make DEVICE_TREE=foo_board_c

In this case, the U-Boot image contains nothing about the DTB file name
it is running with.  (CONFIG_DEFAULT_DEVICE_TREE is not helpful for this
purpose because it is painful to change it from "make menuconfig" for
each board.)

This commit allows to lookup the DTB file name based on the compatible
string and set it to "fdt_file" environment.  Then "tftpboot $fdt_file"
will download the file we want.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

The UMC (Universal Memory Controller) and the DDR PHY block are
highly related to each other.  It is better to have both code in the
same directory.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Both "Model 1" and "Model 2" are supported for ProXstream2 and
PH1-LD6b boards.  It is useful to show the model number in the
boot banner.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

These are new SoCs from Socionext Inc.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Compile this file for U-Boot proper as well as SPL, so that the
U-Boot proper can call uniphier_get_board_param().
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

The DDR3 memory chips on ProXstream2 boards support up to 2133 MHz,
while only up to 1866MHz on PH1-LD6b boards.

Split the board data structure and change the DDR frequency of
ProXstream2 boards to 2133 MHz.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Move "gd->fdt_blob" from the caller to the callee so that this
function can be used more easily.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

We sometimes have to implement different code depending on the SoC
revision.  This commit adds functions to get the model/revision
number.

Note:
  Model number: incremented on major changes of the SoC
  Revision number: incremented on minor changes of the SoC

The "Model 2" exists for PH1-sLD3, ProXstream2/PH1-LD6b.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Sync device trees with Linux.

Linux commit: 06ff6b2d63210922a1b1d0f4997e29ce75b5e0c0
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

These nodes are not parsed by U-Boot for now, but syncing device trees
with Linux is helpful for easier diffing.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

UniPhier SoCs (except PH1-sLD3) have several nodes in common.
Factor out them into uniphier-common32.dtsi.  This improves the code
maintainability.

PH1-sLD3 is so old that it has more or less different register maps
than the others.  So, it cannot be included in this refactoring.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Return different error code depending on the reason so that the
caller can know the cause of the failure.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

The bit 27-24 of the DTCR register is described as RANKEN in the
DDR PHY databook.  Follow this abbreviation.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>

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

This commit adds support of USB 1.1 storage media on AXS103 board.
For some yet unknown reason USB 2.0 doesn't work on AXS103 board issuing
messages like this:
------------------------>8-------------------
AXS# usb start
starting USB...
USB0:   USB EHCI 1.00
scanning bus 0 for devices... EHCI timed out on TD - token=0x80008c80
unable to get device descriptor (error=-1)
1 USB Device(s) found
------------------------>8-------------------

As a work-around we're falling back to USB 1.1.
Indeed it is much slower but at least USB storage devices are usable on
AXS103.
Signed-off-by: NAlexey Brodkin <abrodkin@synopsys.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Simon Glass <sjg@chromium.org>

Marsboard A10 haven't any PMIC at all, so add CONFIG_SUNXI_NO_PMIC=y to
Marsboard_A10_defconfig
Signed-off-by: NAleksei Mamlin <mamlinav@gmail.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

We need to select DM_KEYBOARD now that the usb-kbd code has been converted
to this, otherwise usb keyboards do not work.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Some sunxi-based boards (such as the Olimex A20-SOM-EVB) have a
second MMC socket. This socket is not bootable hardware-wise,
i.e. u-boot itself cannot be loaded from it, but once u-boot has
started, the second socket can be used in the boot process
provided by config_distro_bootcmd.h.

If a second MMC socket is present, place it in the boot order
after the first MMC socket.
Signed-off-by: NKarsten Merker <merker@debian.org>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

The Olimex A20-SOM-EVB is an evaluation board for the Olimex
A20-SOM system-on-module. The baseboard provides a full-size SD
socket (connected to mmc3) in addition to the micro-SD socket on
the SOM itself (which is connected to mmc0).

Enable the mmc3 controller in the board defconfig.
Signed-off-by: NKarsten Merker <merker@debian.org>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

The Empire Electronix D709 tablet is a fairly standard 7" A13 tablet,
featuring usb-wifi, a micro-sd slot, micro-usb otg and headphone jack.

Empire Electronix is written on the back of the tablet, the D709 model
info can be found in the about tablet menu in android.

The PCB has no markings to speak of.

This dts file is identical to the one submitted to the upstream kernel.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Acked-by: NIan Campbell <ijc@hellion.org.uk>

As requested, remove Graeme's email address.
Signed-off-by: NSimon Glass <sjg@chromium.org>
Acked-by: NBin Meng <bmeng.cn@gmail.com>

Replace__CONFIG_SOCFPGA_CYCLONE5_COMMON_H__ with
__CONFIG_SOCFPGA_COMMON_H__ as the file is now called socfpga_common.h
Signed-off-by: NDinh Nguyen <dinguyen@opensource.altera.com>

Move the macro into the socfpga_dwmci_clksel().
Signed-off-by: NDinh Nguyen <dinguyen@opensource.altera.com>
Signed-off-by: NMarek Vasut <marex@denx.de>
[fix parenthesis in the sdmmc_mask]

The scan manager is not needed for the Arria10. Edit the makefile to
build the scan manager for arria5 and cyclone5 only.
Signed-off-by: NDinh Nguyen <dinguyen@opensource.altera.com>
Acked-by: NMarek Vasut <marex@denx.de>

In order to re-use as much Cyclone5 and Arria5 code as possible to support
the Arria10 platform, we need to wrap some of the code with #ifdef's. By
adding CONFIG_TARGET_SOCFPGA_GEN5, we can shorten the check by not having to check
for both AV || AV.
Signed-off-by: NDinh Nguyen <dinguyen@opensource.altera.com>

Now that the USB DWC2 probing is done from OF, remove this note to add
CONFIG_USB_DWC2_REG_ADDR.
Signed-off-by: NDinh Nguyen <dinguyen@opensource.altera.com>

Drop all the common board code, since it is not completely useless.
Signed-off-by: NMarek Vasut <marex@denx.de>
Cc: Chin Liang See <clsee@altera.com>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>

The SoCFPGA has reached a point where every single board code become
the same, since each and every single board is probed equally from OF.
Move the common board code into arch/arm/mach-socfpga/ .
Signed-off-by: NMarek Vasut <marex@denx.de>
Cc: Chin Liang See <clsee@altera.com>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>

We already have the CONFIG_SYS_BOARD variable, which defines the name
of the board. The value in CONFIG_HOSTNAME is exactly the same and is
thus just a duplicity, so switch it to reuse CONFIG_SYS_BOARD .
Signed-off-by: NMarek Vasut <marex@denx.de>
Cc: Chin Liang See <clsee@altera.com>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>

We already have the CONFIG_SYS_VENDOR variable, which defines the
manufacturer of the board. The value in CONFIG_G_DNL_MANUFACTURER
is just a duplicity, so switch it to reuse CONFIG_SYS_VENDOR .
Signed-off-by: NMarek Vasut <marex@denx.de>
Cc: Chin Liang See <clsee@altera.com>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>

There is no VT for this board, so remove this incorrect macro.
Signed-off-by: NMarek Vasut <marex@denx.de>
Cc: Chin Liang See <clsee@altera.com>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>