arm: Support new Xilinx Zynq platform

Add timer driver.

Signed-off-by: Michal Simek <monstr@monstr.eu>
CC: Joe Hershberger <joe.hershberger@gmail.com>
CC: Marek Vasut <marex@denx.de>
Acked-by: Marek Vasut <marex@denx.de>

serial: Add Zynq serial driver

The driver is used on Xilinx Zynq platform.

Signed-off-by: Michal Simek <monstr@monstr.eu>
CC: Joe Hershberger <joe.hershberger@gmail.com>
CC: Marek Vasut <marex@denx.de>
Acked-by: Marek Vasut <marex@denx.de>

arm: Fixed the offset for the no relocation.

When the u-boot address of destination equal to  __start,
no relocation. relocation offset(r9) = 0.

Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Tested-by: Stefano Babic <sbabic@denx.de>

arm: Add CONFIG_OF_BOARD_SETUP support to bootm

ARM boards need to change device tree settings as well

Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

arm: Remove additional config flags

These options are just duplicated from arch/arm/cpu/armv7/config.mk

Signed-off-by: Michal Simek <monstr@monstr.eu>

armv7 cpu_init_crit: Simplify code

We don't need to return to cpu_init_crit after calling lowlevel_init, so
lowlevel_init can directly return to the caller of cpu_init_crit.

Signed-off-by: Benoît Thébaudeau <benoit.thebaudeau@advansee.com>
Cc: Albert Aribaud <albert.u.boot@aribaud.net>

arm: work around assembler bug

Disable sibling call optimization based on binutils version.  This is
to work around a bug in the assember in binutils versions < 2.22.
Branches to weak symbols can be incorrectly optimized in thumb mode to
a short branch (b.n instruction) that won't reach when the symbol gets
preempted.

http://sourceware.org/bugzilla/show_bug.cgi?id=12532

Signed-off-by: Allen Martin <amartin@nvidia.com>
Acked-by: Tom Rini <trini@ti.com>

tools, config.mk: add binutils-version

Modeled after gcc-version, add function to get binutils version.

Signed-off-by: Allen Martin <amartin@nvidia.com>
Acked-by: Wolfgang Denk <wd@denx.de>

ARM: increase lmb stack space reservation to 4KB

The bootm initrd image copy to ram can collide with the stack in cases
where the print buffer size is large (i.e. 1K). The result is intermittent
initrd decompression errors depending on the initrd size MOD 4KB since
the initrd start address is 4KB aligned.

Cc: Albert ARIBAUD <albert.u.boot@aribaud.net>
Signed-off-by: Rob Herring <rob.herring@calxeda.com>

ARM: Add X600 board support (SPEAr600 based)

This patch adds support for the X600 SPEAr600 based board. Its also
the first SPEAr600 board that uses the newly introduced SPEAr600
SPL support. Xloader is not necessary any more. By using the new
"u-boot.spr" make target, one image will generated containing both,
U-Boot SPL (with mkimage header as needed by the SPEAr BootROM, and
the main U-Boot with mkimage header.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Amit Virdi <amit.virdi@st.com>
Cc: Vipin Kumar <vipin.kumar@st.com>

Merge remote-tracking branch 'u-boot-marvell/master'

lsxl: also turn off fan in power down mode

If while booting the power switch is in OFF position, turn off the fan,
too.

Signed-off-by: Michael Walle <michael@walle.cc>
Cc: Prafulla Wadaskar <prafulla@marvell.com>

km_kirkwood: enable MV88E6352_SWITCH support for kmnusa

This is required to configure the external 88e6352 switch on nusa.

Signed-off-by: Valentin Longchamp <valentin.longchamp@keymile.com>
cc: Holger Brunck <holger.brunck@keymile.com>
cc: Prafulla Wadaskar <prafulla@marvell.com>
Acked-by: Prafulla Wadaskar <Prafulla@marvell.com>

arm/km: add mv88e6352 configuration for kmnusa

The kmnusa board uses a mv88e6352 switch that is connected to the main
eth interface of the kirkwood. Therefore the switch must be configured
so that the kirkwood's egiga eth inferface can be used.

Signed-off-by: Valentin Longchamp <valentin.longchamp@keymile.com>
Cc: Holger Brunck <holger.brunck@keymile.com>
Cc: Prafulla Wadaskar <prafulla@marvell.com>
Acked-By: Prafulla Wadaskar <prafulla@marvell.com>

edminiv2: increase malloc len to 256K

Malloc len of 128K caused a warning from
ehci_hcd asking for more.

Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Acked-by: Prafulla Wadaskar <prafulla@marvell.com>

ARM: kirkwood/orion5x: Use reset_cpu definition in include/common.h

include/common.h has the reset_cpu defined already. No need to
re-define here.

Signed-off-by: Dinh Nguyen <dinguyen@altera.com>
Signed-off-by: Pavel Machek <pavel@denx.de>
Acked-By: Prafulla Wadaskar <Prafulla@marvell.com>

kirkwood: ib62x0: Invert SATA activity LEDs

The hardware design of the IB-NAS62x0 causes the SATA activity
LEDs to be on when idle by default.  Reverse the polarity of the
activity LEDs in early init.

Signed-off-by: Simon Baatz <gmbnomis@gmail.com>
Cc: Luka Perkov <uboot@lukaperkov.net>
Cc: Prafulla Wadaskar <prafulla@marvell.com>
Acked-by: Luka Perkov <uboot@lukaperkov.net>

arm/km: use kw_sdram_size_adjust to adjust SDRAM size

Some boards may differ only in the SDRAM size. This function allows to
fix the size accordingly and we can use the same u-boot binary for both
boards.

Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
Signed-off-by: Gerlando Falauto <gerlando.falauto@keymile.com>
cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Valentin Longchamp <valentin.longchamp@keymile.com>

kirkwood: implement kw_sdram_size_adjust

Size of the SDRAM chips might differ between any two (otherwise
identical) instances of the same board.

So add a function kw_sdram_size_adjust() which reads out the current
ram size for a given bank, and adjusts the Kirkwood's SDRAM window size
register accordingly.

Signed-off-by: Gerlando Falauto <gerlando.falauto@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Valentin Longchamp <valentin.longchamp@keymile.com>

kirkwood: implement kw_sdram_bs_set()

Some boards might be equipped with different SDRAM configurations.
When that is the case, CPU CS Window Size Register (CS[0]n Size)
should be set to the biggest value through board.cfg file; then its
value can be fixed at runtime according to the detected SDRAM size.

Therefore, implement kw_sdram_bs_set().

Signed-off-by: Gerlando Falauto <gerlando.falauto@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Valentin Longchamp <valentin.longchamp@keymile.com>
cc: Marek Vasut <marex@denx.de>
Acked-by: Prafulla Wadaskar <Prafulla@marvell.com>

kirkwood: use c-struct for access to SDRAM addr decode registers

Remove the defines and do this with a C-struct.

Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Valentin Longchamp <valentin.longchamp@keymile.com>
cc: Gerlando Falauto <gerlando.falauto@keymile.com>
cc: Marek Vasut <marex@denx.de>
Acked-By: Prafulla Wadaskar <Prafulla@marvell.com>

kirkwood: add support for Iomega iConnect board

Add support for new board iConnect from Iomega.

More information about the device can be found here:

http://go.iomega.com/en/products/network-storage-desktop/wireless-data-station/network-hard-drive-iconnect/?partner=4735

Signed-off-by: Luka Perkov <uboot@lukaperkov.net>
Tested-by: Wojciech Dubowik <wojciech.dubowik@neratec.com>
Tested-by: Tim Fletcher <tim@night-shade.org.uk>

kirkwood: fix mpp.h coding style

Signed-off-by: Luka Perkov <uboot@lukaperkov.net>

ARM: add support for d2 Network v2

This patch adds support for the LaCie board d2 Network v2 which share
a lot of hardware caracteristics with the 2Big Network v2.

- CPU: Marvell 88F6281 1200Mhz
- SDRAM memory: 256MB DDR2 400Mhz
- 2 SATA ports: internal and eSATA
- Gigabit ethernet: PHY Marvell 88E1116R
- Flash memory: SPI NOR 512KB (Macronix MX25L4005A)
- i2c EEPROM: 512 bytes (24C04 type)
- 2 USB2 ports: host and host/device
- 1 push button
- 1 power switch
- 1 SATA LED (bi-color, blue and red)

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>

ARM: add support for Network Space v2 Lite and Mini

This patch adds support for the LaCie boards Network Space v2 (Lite and
Mini). This two boards are derived from the Network Space v2 and a lot
of hardware caracteristics are shared.

- CPU: Marvell 88F6192 800Mhz
- SDRAM memory: 128MB DDR2 200Mhz
- 1 SATA port: internal
- Gigabit ethernet: PHY Marvell 88E1318
- Flash memory: SPI NOR 512KB (Macronix MX25L4005A)
- i2c EEPROM: 512 bytes (24C04 type)
- 2 USB2 ports (Lite only): host and host/device
- 1 push button
- 1 SATA LED (bi-color, blue and red)

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>

lacie_kw: add support for EFI partitions

Defines CONFIG_EFI_PARTITION for LaCie boards.

Additionally this patch defines CONFIG_DOS_PARTITION. Note that this
definition is implicit in mv_common.h when CONFIG_CMD_USB is enabled.

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>

ARM : Add support for MINI2440 (s3c2440).

Support of the MINI2440 board from FriendlyARM from
an old version of u-boot :
http://repo.or.cz/r/u-boot-openmoko/mini2440.git

Currently, supporting only boot from NOR.

Signed-off-by: Gabriel Huau <contact@huau-gabriel.fr>

ARM : Add GPIO Driver and IOMUX definition for S3C2440

It's now possible to use the gpio driver interface
for s3c2440. This patch add iomux definitions too.

Signed-off-by: Gabriel Huau <contact@huau-gabriel.fr>

rmobile: Fix build timer driver with BUILD_DIR

Rmobile common timer driver  diverts the same driver as SH architecture.
When it builds at the same place with source, it is no problem, but when
it builds out of source, it cannot build.
This patch revises this problem.

Reported-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>

i2c: sh: Remove irq_wait function

irq_wait function is not referred to from anywhere.

Reported-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>

doc/git-mailrc: Add 'rmobile' alias

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

rmobile: Add README

This add README of Renesas RMOBILE.
Based doc/README.omap3.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Acked-by: Tom Rini <trini@ti.com>

rmobile: armadillo-800eva: Remove CONFIG_SYS_NO_L2CACHE

armadillo-800eva needs this config.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

rmobile: armadillo-800eva: Add Support NFS and BOOTZ command

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

rmobile: armadillo-800eva: Add Support CONFIG_OF_LIBFDT

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

rmobile: armadillo-800eva: Change init function of SCIFA1

This initializes GPIO, without using PFC framework in
board_early_init_f function. It is because it cannot initialize
normally when PFC is used.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add cpu_eth_init function

This supports ethernet driver of RMOBILE R8A7740.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: armadillo-800eva Remove board_eth_init

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add support for ATMARK-TECHNO Armadillo-800EVA board

The Armadillo-800EVA board has Renesas R-Mobile R8A7740, 512MB DDR3-SDRAM,
Ethernet, and more.

This patch supports the following functions:
 - 512MB DDR3-SDRAM
 - Serial console (SCIF)
 - Ethernet MAC(MII) & PHY(SMSC)

Signed-off-by: Hideyuki Sano <hideyuki.sano.dn@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add support PFC of Renesas R8A7740

Renesas R8A7740 has GPIO based PFC. This privode framework of PFC.
The code included in this base from linux kernel.

Signed-off-by: Hideyuki Sano <hideyuki.sano.dn@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add support Renesas R8A7740

Renesas R8A7740 is CPU with Cortex-A9.
This supports the basic register definition and GPIO.

Signed-off-by: Hideyuki Sano <hideyuki.sano.dn@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: Add CONFIG_GLOBAL_TIMER to board config file

kzm9g board use global timer. But by commit 813ffda31, timer function of
rmobile was changed that global timer might be used, when CONFIG_GLOBAL_TIMER
was defined.
This add CONFIG_GLOBAL_TIMER to board config file.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add support TMU base timer function

Some rmobile SoC has TMU base timer function. This supports TMU.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Change initializing ICCICR register

There is rmobile without ICCICR.
ICCICR is initialized only when ICCICR is defined.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: remove unrelated config

Remove CONFIG_ARM_CORTEXA9.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: Add LIBFDT support

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: separate cpu_rev to integer and fraction

According to SoC document, revision info is separated to integer part and
fracton part.
So I separete rmobile_get_cpu_rev() to rmobile_get_cpu_rev_integer() and
rmobile_get_cpu_rev_fraction().

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: fix CPU info

CPU info register was read wrongly by mistake. And function rmobile_get_cpu_rev() was not called properly.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: remove unrelated config

Remove CONFIG_INTEGRATOR and CONFIG_ARCH_CINTEGRATOR. These are not for kzm9g.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: add NFS_TIMEOUT in config file

Set NFS_TIMEOUT to 10,000 msec.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: Fix CONFIG_BAUDRATE setting

The value of CONFIG_BAUDRATE is treated as string and put as initial value of
environment variable. If it begin with '(', it is wrongly parsed to 0 in number.
So I removed '(' and ')'.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: Add dummy member to struct sh73a0_rwdt

Add dummy member to struct sh73a0_rwdt in sh73a0.h.
Without this, initializing watch dog timer goes wrong.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Support build with gcc-4.6 or later

Latest rmobile code was tested by using old gcc (gcc-4.4).
When we use gcc-4.6 (or later), the build is made, but does not work.
This solves a problem not to work by add -march=armv5 to compiple option
when we built in gcc-4.6 (or later).
I tested by linaro's compiler version 2012.04-20120426.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: enable reset command

Do soft power on reset in U-Boot reset command.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: Modify bus controller setting for CS4

Problem:
Linux kernel hangs up when it write a file to NFS mounted directory.
Solution:
Modify bus controller setting for CS4, which connected smsc9221 ethernet
controller.

Detail:
Modify CS4BCR bit[29:28] (IWW[1:0]) from 00 to 01.
Modify CS4BCR bit[20:19] (IWRRD[1:0]) from 00 to 01.
Modify CS4BCR bit[17:16] (IWRRS[1:0]) from 00 to 01.
Modify CS4WCR bit[27:26] (WSW[1:0]) from 10 to 11
Modify CS4WCR bit[25:24] (WHW[1:0]) from 01 to 10
Modify CS4WCR bit[18:16] (WW[2:0]) from 101 to 111
Modify CS4WCR bit[13:11] (SW[2:0]) from 010 to 011
Modify CS4WCR bit[10:7] (WR[3:0]) from 1000 to 1011

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: change prompt to board specific

Change U-Boot prompt to board specific one.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: Adjust low level hardware setting

Adjust low level hardware setting in s_init.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: kzm9g: Modify sdram area

Reserve first 16MB for RT-CPU (as same as kernel config).

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add supoprt for KMC KZM-A9-GT board

The KZM-A9-GT board has Renesas R-Mobile SH73A0, 512MB DDR2-SDRAM,
USB, Ethernet, and more.

This patch supports the following functions:
- 512MB DDR2-SDRAM
- 16MB NOR Flash memory
- Serial console (SCIF)
- Ethernet (SMSC)
- I2C

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add support PFC of Renesas SH73A0

Renesas SH73A0 has GPIO based PFC. This privode framework of PFC.
The code included in this base from linux kernel.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add support Renesas SH73A0

Renesas SH73A0 is CPU with Cortex-A9.
This supports the basic register definition and GPIO.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

arm: rmobile: Add basic support for Renesas R-Mobile

This patch adds minimum support for R-Mobile. Only minimal support with timer.
This CPU can uses the peripheral of Renesas SuperH.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

ARMv7: Add register definition of global timer

ARMv7 has global timer. This provides the register definition of this timer.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>

omap4_panda: remove CONFIG_PANDA, not used

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>

am335x-evm config: decrease bootdelay to 1s and mount rootfs RO

A fast boot is important to the beaglebone, so save 2 seconds here by
decreasing bootdelay. This is still plenty time to break into the prompt,
I do that at least once a day.

Mount the rootfs RO by default, this is needed to make fsck succeed
without resorting to an initramfs.

Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>

am33xx: Fix fetching of mmc1 bootmode from bootrom for AM33XX

U-boot should not ignore getting the bootmode passed on from the bootrom.
With this, U-boot SPL knows it was loaded from MMC1 and use this info to
read full U-boot from MMC1 as well.

Cc: pprakash@ti.com
Cc: trini@ti.com
Signed-off-by: Joel A Fernandes <joelagnel@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>

OMAP: networking support for SPL

This patch adds support for networking in SPL. Some devices are
capable of loading SPL via network so it makes sense to load the
main U-Boot binary via network too. This patch tries to use
existing network code as much as possible. Unfortunately, it depends
on environment which in turn depends on other code so SPL size
is increased significantly. No effort was done to decouple network
code and environment so far.

Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Signed-off-by: Tom Rini <trini@ti.com>

am335x_evm: enable networking in SPL

This patch adds support for networking in SPL on TI AM335x based
boards. Vendor Class Identifier used by SPL during BOOTP is
"AM335x U-Boot SPL".

Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com>

OMAP: spl: call timer_init() from SPL

We need to initialize timer properly, otherwise all delays
inside SPL will be wrong.

Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com>

net/bootp: add VCI support for BOOTP also

Vendor Class Identifier option is common to BOOTP and DHCP and
can be useful without PXE. So send VCI in both BOOTP and DHCP
requests if CONFIG_BOOTP_VCI_STRING is defined.

Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com>
Signed-off-by: Tom Rini <trini@ti.com>

davinci: ea20: add some configs and default environmet variables

Signed-off-by: Bastian Ruppert <Bastian.Ruppert@Sewerin.de>
CC: Tom Rini <trini@ti.com>
Acked-by: Stefano Babic <sbabic@denx.de>

da850/omap-l138: davinci_emac: Suppress auto negotiation if needed

negotiation is enabled in RMII mode. Some boards based on da850 need
to suppress this procedure.

CC: Rajashekhara, Sudhakar <sudhakar.raj@ti.com>
CC: Lad, Prabhakar <prabhakar.lad@ti.com>
CC: Hadli, Manjunath <manjunath.hadli@ti.com>
CC: sbabic@denx.de
Acked-by: Stefano Babic <sbabic@denx.de>
CC: Tom Rini <trini@ti.com>
Signed-off-by: Bastian Ruppert <Bastian.Ruppert@Sewerin.de>
Acked-by: Prabhakar Lad <prabhakar.lad@ti.com>

video: cfb_console: add function to plot the logo area black

Signed-off-by: Bastian Ruppert <Bastian.Ruppert@Sewerin.de>
CC: Anatolij Gustschin <agust@denx.de>
CC: Tom Rini <trini@ti.com>
CC: Stefano Babic <sbabic@denx.de>

video: cfb_console: logo can be positioned via the splashpos variable

Extend the driver for placing the video/bmp logo as specified
by "splashpos" environment variable.

Signed-off-by: Bastian Ruppert <Bastian.Ruppert@Sewerin.de>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
CC: Stefano Babic <sbabic@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>

davinci: ea20: the console is always set to the serial line

Do not allow to overwrite it when video is enabled.

Signed-off-by: Bastian Ruppert <Bastian.Ruppert@Sewerin.de>
CC: Tom Rini <trini@ti.com>
Acked-by: Stefano Babic <sbabic@denx.de>

davinci: ea20: reorganisation LCD startup

Signed-off-by: Bastian Ruppert <Bastian.Ruppert@Sewerin.de>
CC: Tom Rini <trini@ti.com>
Acked-by: Stefano Babic <sbabic@denx.de>

devkit8000: add rootwait to mmcboot option

Signed-off-by: Andreas Bießmann <biessmann@corscience.de>
cc: Thomas Weber <weber@corscience.de>

Merge remote-tracking branch 'u-boot/master'

sandbox: Add asm/errno.h

This file is required for all archs. Fixes a sandbox build break on ext4.

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

cmd_fat.c: fix build warning

Commit cfda5aeab89d73779e26f0d34cf10f64caa67431
(cmd_fat: use common get_device_and_partition function)
introduced a warning:
cmd_fat.c: In function 'do_fat_fswrite':
cmd_fat.c:178:8: warning: unused variable 'ep' [-Wunused-variable]

Fix it.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Reported-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Tested-by: Albert ARIBAUD <albert.u.boot@aribaud.net>

common: add zip command support

Signed-off-by: Lei Wen <leiwen@marvell.com>

lib: add gzip lib function callback

Signed-off-by: Lei Wen <leiwen@marvell.com>

lib: zlib: remove the limitation for cannot using 0 as start

We often need the requirement that compressing those memory range start
from 0, but the default deflate code in zlib prevent us to do this.
Considering the special case of uboot, that it could access all memory
range, it is reasonable to be able to also take the address space from 0
into compression.

Signed-off-by: Lei Wen <leiwen@marvell.com>

lib: zlib: include deflate into zlib build

Add a new config CONFIG_GZIP_ENABLED, if enabled, the uboot bin would
include zlib's deflate method which could be used for compressing.

Signed-off-by: Lei Wen <leiwen@marvell.com>

lib: zlib: import trees file from 1.2.5

Signed-off-by: Lei Wen <leiwen@marvell.com>

lib: zlib: import deflate source file from 1.2.5

Signed-off-by: Lei Wen <leiwen@marvell.com>

Merge remote-tracking branch 'u-boot-imx/master'

Merge remote-tracking branch 'u-boot-atmel/master'

remove unnecessary code in ata_piix

We set sata_curr_device to 0 right after returning from init_sata(), so there's
no point in setting it to the last scanned driver at this point.
Note: there are more duplicities with cmd_sata, but those might be required,
as the code seems to reset the entire controller on every scan, ignoring the
requested port number.

Signed-off-by: Pavel Herrmann <morpheus.ibis@gmail.com>

disk: allow - or empty string to fall back to $bootdevice

Commit 10a37fd "disk: get_device_and_partition() "auto" partition"
prevented the use of "-" on the command-line to request fallback to the
$bootdevice environment variable instead. This patch allows that, or an
empty string "" to be used.

Tested:
setenv bootfile /boot/zImage
setenv bootdevice 0:1
ext2load mmc 0:1
ext2load mmc -
ext2load mmc ""

Signed-off-by: Stephen Warren <swarren@nvidia.com>

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

SPL: Add support for loading image from ram in SPL.

Signed-off-by: Pavel Machek <pavel@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>

SPL: Rework how we inform about un-headered images

First, remove the puts from the case where we don't have an mkimage
header as this is somewhat common and intentional for no-arg target
images.  Second, rework the final switch statement in board_init_r to,
in the case of !CONFIG_SPL_OS_BOOT be only about doing debug prints
about if we know what the magic is or not (the CONFIG_SPL_OS_BOOT case
is unchanged).  Then we call jump_to_image_no_args().  This gives us the
same behavior as before but with slightly smaller code.

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

SPL: Make un-supported boot device puts a debug instead

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

SPL: Enable use of custom defined U-Boot entry point

By setting CONFIG_SYS_UBOOT_START boards can now use a different entry
point for their U-Boot image. So the U-Boot entry point is not fixed
to CONFIG_SYS_TEXT_BASE any more.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>

SPL: SPI: Enhance spi_spl_load to match the other load functions

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

SPL: Add option to skip copying of the mkimage header

On some system (e.g. powerpc), the load-address and entry-point is
located at address 0. So the current approach to load the image
(payload) including the header to the address "load-address - 64"
can't work here.

This patch adds an flag to skip this copying including header to
the SPL framework. By setting SPL_COPY_PAYLOAD_ONLY, only the
playload will be copied. This will be used by the SPL NOR flash
driver on powerpc.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>

SPL: Add NOR flash booting support

SPL NOR flash booting support is quite simple. Only copying of the
images is needed.

On MPC5xxx we need to make sure to only use the standard memcpy()
implementation and not the MPC5xxx specific one. As the MPC5xxx
version has some complexity which is not needed for this SPL
booting.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>

da850: Add README.da850

This file documents when to build for da850evm and when to build for
da850_am18xxevm.  It also documents how to write the u-boot.ais file to
persistent storage (such as SPI), in some cases as well as how to write
a recovery image.

Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Christian Riesch <christian.riesch@omicron.at>
Acked-by: Prabhakar Lad <prabhakar.lad@ti.com>

hawkboard: Update config file to work with common spl framework

The common spl framework expects the u-boot payload size through
CONFIG_SYS_MONITOR_LEN. Define the macro with the u-boot's size. With
this change, CONFIG_SYS_NAND_U_BOOT_SIZE is no longer required. Delete
the same.

Signed-off-by: Tom Rini <trini@ti.com>
Signed-off-by: Sughosh Ganu <urwithsughosh@gmail.com>