Import k2l specific DT files from Linux Kernel and enable
OF_CONTROL, DM, DM_SERIAL.

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

Import k2hk specific DT files from Linux Kernel and enable
OF_CONTROL, DM, DM_SERIAL.

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

Keystone dts files assumes that LPAE is enabled and top level root
node uses 64bit addresses. This breaks the keystone boot with
CONFIG_OF_CONTROL enabled. So do not use 64 bit addresse in U-Boot DT.

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

Import various generic dts files from Linux kernel so that
all keystone2 platforms can be DT in U-boot.

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

If CONFIG_SYS_MALLOC_F_LEN is enabled, the stack is moved down to the
specified size to make the malloc function available before relocation.
But on keystone platforms SYS_SPL_MALLOC is immediately preceding stack,
which is causing an overlap with this config enabled.
So leave a gap between malloc space and stack space.

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

With CONFIG_DM_SERIAL is enabled NS16550_init() cannot be
called directly. Driver probe should be taking care of this.
So call this function only when DM_SERIAL is not enabled.

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

Tested on J6Eco EVM.
Signed-off-by: Tom Rini <trini@konsulko.com>
Tested-by: Lokesh Vutla <lokeshvutla@ti.com>

Tested on J6Eco EVM
Signed-off-by: Tom Rini <trini@konsulko.com>

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

Use the sandbox environment for the basic tests.

Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Acked-by: Simon Glass <sjg@chromium.org>

Introduce dummy devices for sandbox remoteproc device and enable it by
default

Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Acked-by: Simon Glass <sjg@chromium.org>

Introduce a dummy driver for sandbox that allows us to verify basic
functionality. This is not meant to do anything functional - but is
more or less meant as a framework plumbing debug helper.

The sandbox remoteproc driver maintains absolutey no states and is a
simple driver which just is filled with empty hooks. Idea being to give
an approximate idea to implement own remoteproc driver using this as a
template.

Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Acked-by: Simon Glass <sjg@chromium.org>

Many System on Chip(SoC) solutions are complex with multiple processors
on the same die dedicated to either general purpose of specialized
functions. Many examples do exist in today's SoCs from various vendors.
Typical examples are micro controllers such as an ARM M3/M0 doing a
offload of specific function such as event integration or power
management or controlling camera etc.

Traditionally, the responsibility of loading up such a processor with a
firmware and communication has been with a High Level Operating
System(HLOS) such as Linux. However, there exists classes of products
where Linux would need to expect services from such a processor or the
delay of Linux and operating system being able to load up such a
firmware is unacceptable.

To address these needs, we need some minimal capability to load such a
system and ensure it is started prior to an Operating System(Linux or
any other) is started up.

NOTE: This is NOT meant to be a solve-all solution, instead, it tries to
address certain class of SoCs and products that need such a solution.

A very simple model is introduced here as part of the initial support
that supports microcontrollers with internal memory (no MMU, no
execution from external memory, or specific image format needs). This
basic framework can then (hopefully) be extensible to other complex SoC
processor support as need be.

Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Acked-by: Simon Glass <sjg@chromium.org>

enable ethernet driver model for am335x gp evm as cpsw supports
driver model

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>

enable ethernet driver model for am335x beagle bone black as cpsw
supports driver model

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>

adopt cpsw driver to device driver model

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

prepare driver for device model migration

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

Since usb ether gadget doesn't have support for driver model, so
not defining usb ether gadget when ethernet driver model is
defined.

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>

Prepare board file so that ethernet registration are
commented for DM conversion

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>

Add support for distro bootcmds and network booting while retaining
backwards compatibility with the current "legacy" setup. With these
changes the default boot sequence becomes:

* SD card (standard distro boot)
* SD card (legacy boot)
* EMMC (standard distro boot)
* EMMC (legacy boot)
* Nand (legacy boot)
* PXE (standard distro boot)
* DHCP (standard distro boot)

The older boot scripts have some overlap with what the distro
bootcommands to however i've left them unchanged to prevent introduction
of subtle bugs.

Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>

Add mandatory address variables for loading scripts and pxe configuration as
per README.distro

Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>

Hush has an oddity where using ${var} causes var to resolved in the the global
address space (iotw the environment) first and only afterwards will the local
variable space be searched.

This causes odd side-effects when iterating over the boot partitions
using ${bootpart} if the environment also has a bootpart variable (e.g. for
the various TI boards). Fix this by using the hopefully more unique
${distro_bootpart} instead of ${bootpart}.

Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>

Since there is a common function to grab the serial number from the die id bits,
it makes sense have one to parse that serial number and feed it to the serial
ATAG.

Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-by: Tom Rini <trini@konsulko.com>

This introduces omap_die_id_display to display the full die id.
There is no need to store it in an environment variable, that no boot script
is using anyway.

Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-by: Tom Rini <trini@konsulko.com>

Now that we have a common prototype to grab the omap die id, functions to figure
out a serial number and usb ethernet address can use it directly.
Those also get an omap_die_id prefix for better consistency.

Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-by: Tom Rini <trini@konsulko.com>

This introduces omap5 support for omap_die_id, which matches the common
omap_die_id definition. It replaces board-specific code to grab the die id bits.

Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-by: Tom Rini <trini@konsulko.com>

This introduces omap4 support for omap_die_id, which matches the common
omap_die_id definition. It replaces board-specific code to grab the die id bits.

Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-by: Tom Rini <trini@konsulko.com>

This replaces the previous get_dieid definition with omap_die_id, that matches
the common omap_die_id definition.

Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-by: Tom Rini <trini@konsulko.com>

This introduces a common definition for omap_die_id, that aims at providing a
common interface for accessing omap platform's die id bits.

Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-by: Tom Rini <trini@konsulko.com>

board_name environment variable was not getting set correctly for Pandaboard A4 and ES

Signed-off-by: David Batzle <dbatzle@dcbcyber.com>
CC: Albert Aribaud <albert.u.boot@aribaud.net>; Tom Rini <trini@ti.com>; Peter Robinson <pbrobinson@gmail.com>

Advantech SOM-6896 is a Broadwell U based COM Express Compact Module
Type 6. This patch adds support for it as a coreboot payload.

On board SATA and SPI are functional. On board Ethernet isn't functional
but since it's optional and ties up a PCIe x4 that is otherwise brought
out, this isn't a concern at the moment. USB doesn't work since the
xHCI driver appears to be broken.

Signed-off-by: George McCollister <george.mccollister@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>

This works correctly now, so enable it.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Dropped malloc() and adjusted commit message:
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

This code takes about 450ms without the MRC cache and about 27ms with the
cache. Add a debug timer so that this time can be displayed.

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

At present a missing $ causes this code to hang when using the MRC cache/
Fix it.

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

The RTC can fail, so check the return value for reads.

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

For consistency, use 'ret' to handle a return value.

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

The current check is incorrect and will fail when any non-zero byte is read.
Fix it.

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

At present this driver uses bind() to set up the device. The bind() method
should not touch the hardware, so move the init code to probe().

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

Add a comment to make it clear to which block the #endif relates.

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