platform/kernel/u-boot.git
11 years agoarm, am335x: add some missing registers and defines for lcd and epwm support
Heiko Schocher [Mon, 19 Aug 2013 14:38:57 +0000 (16:38 +0200)]
arm, am335x: add some missing registers and defines for lcd and epwm support

- add missing register defines in struct cm_perpl
  epwmss0clkctrl
  epwmss2clkctrl
  lcdcclkstctrl
- add missing register defines in struct cm_dpll
  clklcdcpixelclk
- add struct pwmss_regs
- add struct pwmss_ecap_regs
- add LCD Controller base LCD_CNTL_BASE
- add PWM0 controller base PWMSS0_BASE

Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Tom Rini <trini@ti.com>
11 years agoarm, am33xx: add defines for gmii_sel_register bits
Heiko Schocher [Mon, 19 Aug 2013 14:38:56 +0000 (16:38 +0200)]
arm, am33xx: add defines for gmii_sel_register bits

Signed-off-by: Heiko Schocher <hs@denx.de>
Acked-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agodra7xx_evm: Re-order and comment the networking related config options
Tom Rini [Tue, 20 Aug 2013 12:53:54 +0000 (08:53 -0400)]
dra7xx_evm: Re-order and comment the networking related config options

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoomap5_uevm: Better comment why we have TCA642X and the reset time
Tom Rini [Tue, 20 Aug 2013 12:53:53 +0000 (08:53 -0400)]
omap5_uevm: Better comment why we have TCA642X and the reset time

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:omap5: Clarify comments about SPL and DDR timings in common config
Tom Rini [Tue, 20 Aug 2013 12:53:52 +0000 (08:53 -0400)]
TI:omap5: Clarify comments about SPL and DDR timings in common config

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:am33xx: Move SPL YMODEM support to the per-board config
Tom Rini [Tue, 20 Aug 2013 12:53:51 +0000 (08:53 -0400)]
TI:am33xx: Move SPL YMODEM support to the per-board config

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Update README for customization
Tom Rini [Tue, 20 Aug 2013 12:53:50 +0000 (08:53 -0400)]
am335x_evm: Update README for customization

As this is a reference platform, update the README to note which IP
blocks are required for use due to design choices of the reference
rather than required by the SoC itself.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:armv7: Re-order slightly the generic CONFIG options, expand related comments
Tom Rini [Tue, 20 Aug 2013 12:53:49 +0000 (08:53 -0400)]
TI:armv7: Re-order slightly the generic CONFIG options, expand related comments

Re-group the general options to note things that can be removed safely
to reduce binary size when not required, and expand the comment about
what commands we do include to note it could be replaced with a specific
list of required one.  While at it, move the CMD parts of various
blocks IP to the end of the list for consistency.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Regroup USB options
Tom Rini [Tue, 20 Aug 2013 12:53:48 +0000 (08:53 -0400)]
am335x_evm: Regroup USB options

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Add comment by SPL SPI support
Tom Rini [Tue, 20 Aug 2013 12:53:47 +0000 (08:53 -0400)]
am335x_evm: Add comment by SPL SPI support

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:am335x: Better comment and organize the networking related options
Tom Rini [Tue, 20 Aug 2013 12:53:46 +0000 (08:53 -0400)]
TI:am335x: Better comment and organize the networking related options

While in here, drop CONFIG_BOOTP_DEFAULT as it is unused in the code.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoomap5: Expand CONFIG_SPL_MAX_SIZE and comment upon SRAM_SCRATCH_SPACE_ADDR
Tom Rini [Tue, 20 Aug 2013 12:53:45 +0000 (08:53 -0400)]
omap5: Expand CONFIG_SPL_MAX_SIZE and comment upon SRAM_SCRATCH_SPACE_ADDR

After examining both TRMs and doing some experimentation, we can rely on
using the start of the download area for CONFIG_SPL_TEXT_BASE and then
move SRAM_SCRATCH_SPACE_ADDR up, just like am335x.  This is required for
peripheral boot modes such as UART.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:armv7: Move CONFIG_SPL_LIBDISK_SUPPORT to MMC section
Tom Rini [Tue, 20 Aug 2013 12:53:44 +0000 (08:53 -0400)]
TI:armv7: Move CONFIG_SPL_LIBDISK_SUPPORT to MMC section

We only need this library when we're doing "disk" access to MMC/SD.
Update comment around the rest of CONFIG_SPL_LIB* to note that the
others are always required.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam33xx: Correct and expand comments on CONFIG_SPL_MAX_SIZE
Tom Rini [Tue, 20 Aug 2013 12:53:43 +0000 (08:53 -0400)]
am33xx: Correct and expand comments on CONFIG_SPL_MAX_SIZE

We had been allowing the max size to be larger than actually allowed by
the ROM.  Expand the commentary here to explain why we set these
locations.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoARM: igep00x0.h: Enable raw initrd support
Javier Martinez Canillas [Sun, 11 Aug 2013 16:20:00 +0000 (18:20 +0200)]
ARM: igep00x0.h: Enable raw initrd support

Now that IGEP base boards default environment use
the bootz command to boot a zImage instead of a
uImage, it makes sense to add support to supply a
raw initrd image to the kernel if needed.

Signed-off-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
11 years agoarm: omap3: fix SRAM copy and execution sequence
Albert ARIBAUD [Sat, 10 Aug 2013 17:03:59 +0000 (19:03 +0200)]
arm: omap3: fix SRAM copy and execution sequence

Fix size calculation in copy of go_to_speed into SRAM.
Use SRAM_CLK_CODE in call to SRAM-based go_to_speed.

Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
11 years agoARM: OMAP4460: sdp: Limit TPS mux config to 4460
Taras Kondratiuk [Tue, 6 Aug 2013 12:18:51 +0000 (15:18 +0300)]
ARM: OMAP4460: sdp: Limit TPS mux config to 4460

TPS mux config is 4460 specific, so it should be limited to 4460 only.

Signed-off-by: Taras Kondratiuk <taras@ti.com>
11 years agoARM: OMAP4470: Add Elpida EDB8164B3PF memory configuration
Lubomir Popov [Tue, 6 Aug 2013 12:18:50 +0000 (15:18 +0300)]
ARM: OMAP4470: Add Elpida EDB8164B3PF memory configuration

OMAP4470 SDP SoM has EDB8164B3PF PoP memory on board.
This memory has 4Gb x 2CS = 8Gb configuration.
Add configuration for runtime calculation and precalculated cases.

Patch is based on a draft Lubomir's patch [1].

[1] http://lists.denx.de/pipermail/u-boot/2013-April/150851.html

Signed-off-by: Lubomir Popov <lpopov@mm-sol.com>
[taras@ti.com: cleaned up patch and fixed precalculated values]
Signed-off-by: Taras Kondratiuk <taras@ti.com>
11 years agoARM: OMAP4470: Add voltage and dpll data
Taras Kondratiuk [Tue, 6 Aug 2013 12:18:49 +0000 (15:18 +0300)]
ARM: OMAP4470: Add voltage and dpll data

OMAP4470 reference design uses TWL6032 PMIC
with a following connection scheme:
  VDD_CORE = TWL6032 SMPS2
  VDD_MPU  = TWL6032 SMPS1
  VDD_IVA  = TWL6032 SMPS5

Set voltage and frequency values according to
OMAP4470 Data Manual Operating Condition Addendum v0.7

Signed-off-by: Taras Kondratiuk <taras@ti.com>
11 years agoARM: OMAP4470: Add OMAP4470 identification
Taras Kondratiuk [Tue, 6 Aug 2013 12:18:48 +0000 (15:18 +0300)]
ARM: OMAP4470: Add OMAP4470 identification

Signed-off-by: Taras Kondratiuk <taras@ti.com>
11 years agosdp4430: Initialize board id using CONFIG_MACH_TYPE
Oleksandr Tyshchenko [Tue, 6 Aug 2013 11:03:27 +0000 (14:03 +0300)]
sdp4430: Initialize board id using CONFIG_MACH_TYPE

Use CONFIG_MACH_TYPE generic macro to set the machine type
number in the common arm code instead of setting it in the
board code.

Signed-off-by: Oleksandr Tyshchenko <oleksandr.tyshchenko@ti.com>
11 years agoMerge branch 'u-boot-imx/master' into 'u-boot-arm/master'
Albert ARIBAUD [Sat, 17 Aug 2013 16:24:13 +0000 (18:24 +0200)]
Merge branch 'u-boot-imx/master' into 'u-boot-arm/master'

11 years agoAdd TI816X evm board support
TENART Antoine [Tue, 2 Jul 2013 10:06:00 +0000 (12:06 +0200)]
Add TI816X evm board support

Signed-off-by: Antoine Tenart <atenart@adeneo-embedded.com>
[trini: Change to SPDX, fix a few compiler warnings, adapt to
CONFIG_OMAP_COMMON]
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoAdd TI816X support
TENART Antoine [Tue, 2 Jul 2013 10:05:59 +0000 (12:05 +0200)]
Add TI816X support

Signed-off-by: Antoine Tenart <atenart@adeneo-embedded.com>
[trini: Fix warnings about vtp things in emif4.c, adapt AM43XX]
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoPrepare for TI816X : reuse existing code from TI814X
TENART Antoine [Tue, 2 Jul 2013 10:05:58 +0000 (12:05 +0200)]
Prepare for TI816X : reuse existing code from TI814X

Rename some CONFIG_TI814X to a more generic CONFIG_TI81XX

Signed-off-by: Antoine Tenart <atenart@adeneo-embedded.com>
[trini: Adapt for CONFIG_OMAP_COMMON changes, AM43XX]
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoarm, da850: add ipam390 board support
Heiko Schocher [Mon, 5 Aug 2013 14:00:38 +0000 (16:00 +0200)]
arm, da850: add ipam390 board support

add the am1808 based ipam390 board from Barix.

- 128MByte, DDR2, synchronous RAM 16bit databus to SDRAM
  interface
- 128MByte, NAND Flash, 8bit databus to the NANDFlash
  Interface
- Ethernet PHY Micrel KSZ8051R via RMII
- Console on UART 0
- booting fron nand flash
- spl falcon bootmode

Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Tom Rini <trini@ti.com>
11 years agoarm, da850: enable the correct uart in arch_cpu_init()
Heiko Schocher [Mon, 5 Aug 2013 14:00:37 +0000 (16:00 +0200)]
arm, da850: enable the correct uart in arch_cpu_init()

in arch_cpu_init() uart2 is fix enabled, without reference the
setting from CONFIG_SYS_NS16550_COM1. Use the setting from
CONFIG_SYS_NS16550_COM1 for enabling the console.

Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Albert ARIBAUD <albert.u.boot@aribaud.net>
Cc: Tom Rini <tom.rini@gmail.com>
Cc: Christian Riesch <christian.riesch@omicron.at>
11 years agobootstage: get more BOOTSTAGE_ID* in show_boot_progress()
Heiko Schocher [Mon, 5 Aug 2013 14:00:36 +0000 (16:00 +0200)]
bootstage: get more BOOTSTAGE_ID* in show_boot_progress()

In case CONFIG_BOOTSTAGE is not defined, call from bootstage_mark_name()
show_boot_progress(), so get more BOOTSTAGE_ID* ids in show_boot_progress()
if CONFIG_BOOTSTAGE is not defined.

Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Tom Rini <trini@ti.com>
Cc: Albert Aribaud <albert.u.boot@aribaud.net>
Acked-by: Simon Glass <sjg@chromium.org>
11 years agoarm/davinci/da850: add uart0_pins_rtscts and RMII_MHz_50_CLK in emac_pins_rmii pinmux
Heiko Schocher [Mon, 5 Aug 2013 14:00:35 +0000 (16:00 +0200)]
arm/davinci/da850: add uart0_pins_rtscts and RMII_MHz_50_CLK in emac_pins_rmii pinmux

Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Tom Rini <trini@ti.com>
11 years agoTI:armv7: Enable CONFIG_CMD_GPIO
Tom Rini [Fri, 9 Aug 2013 15:22:20 +0000 (11:22 -0400)]
TI:armv7: Enable CONFIG_CMD_GPIO

Add the generic "poke a GPIO" command, with the GPIO related defines.

Acked-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:armv7: Enable CONFIG_CMD_SPI
Tom Rini [Fri, 9 Aug 2013 15:22:19 +0000 (11:22 -0400)]
TI:armv7: Enable CONFIG_CMD_SPI

Add the generic "poke the SPI bus" command, with the SPI related
defines.

Acked-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:omap5/dra7xx: Convert to ti_armv7_common.h
Tom Rini [Fri, 9 Aug 2013 15:22:18 +0000 (11:22 -0400)]
TI:omap5/dra7xx: Convert to ti_armv7_common.h

Update omap5_common.h to use ti_armv7_common.h, and in turn update
dra7xx_evm.h and omap5_uevm.h slightly.  The biggest changes here are
that IP blocks which exist on the platform, and had clocks enabled,
now have the drivers being built as well.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoTI:am33xx: Create common config files for TI ARMv7 platforms, and AM33xx
Tom Rini [Fri, 9 Aug 2013 15:22:17 +0000 (11:22 -0400)]
TI:am33xx: Create common config files for TI ARMv7 platforms, and AM33xx

We create two new files, include/configs/ti_armv7_common.h for all of
the common IP blocks and related features / commands we share in
virtually all of our platforms.  We then create
include/configs/ti_am335x_common.h for everything common to the am335x
SoC leaving just the board specific parts to
include/configs/ti_am335x_common.h.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Bring in 'boot_fdt' logic from i.MX
Tom Rini [Fri, 9 Aug 2013 15:22:16 +0000 (11:22 -0400)]
am335x_evm: Bring in 'boot_fdt' logic from i.MX

Bring in the 'boot_fdt' environment variable that i.MX boards use to try
and load a device tree when booting.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoarm: spl: For Falcon Mode, set a default machid of ~0
Tom Rini [Fri, 9 Aug 2013 15:22:15 +0000 (11:22 -0400)]
arm: spl: For Falcon Mode, set a default machid of ~0

With device trees, boards do not always set CONFIG_MACH_TYPE now, so we
must not rely on this define being set.  The kernel uses ~0 to see if we
have a valid machine number or not, so set that as the default, invalid
machine, id and only fix if CONFIG_MACH_TYPE is set.

Acked-by: Dan Murphy <dmurphy@ti.com>
Tested-by: Heiko Schocher <hs@denx.de>
Cc: Albert ARIBAUD <albert.u.boot@aribaud.net>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Use default baud rate table
Tom Rini [Fri, 9 Aug 2013 15:22:14 +0000 (11:22 -0400)]
am335x_evm: Use default baud rate table

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam33xx: Stop using PHYS_DRAM_1 define
Tom Rini [Fri, 9 Aug 2013 15:22:13 +0000 (11:22 -0400)]
am33xx: Stop using PHYS_DRAM_1 define

We defined PHYS_DRAM_1 to 0x80000000 (start of DRAM) and then used this
for CONFIG_SYS_SDRAM_BASE.  But then we kept on referencing PHYS_DRAM_1
in other places.  Change to directly setting CONFIG_SYS_DRAM_BASE and
then using that name in code.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam33xx: CONFIG_DMA_COHERENT defines are unused, remove
Tom Rini [Fri, 9 Aug 2013 15:22:12 +0000 (11:22 -0400)]
am33xx: CONFIG_DMA_COHERENT defines are unused, remove

Acked-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoOMAP3: igep00x0: allow booting with a FDT from MMC
Javier Martinez Canillas [Wed, 7 Aug 2013 15:53:19 +0000 (17:53 +0200)]
OMAP3: igep00x0: allow booting with a FDT from MMC

IGEP boards now have Device Tree support in the mainline
kernel. To boot an IGEP board using a DT, a uEnv.txt plain
text file could be used to define a custom uenvcmd that will
be run by the default boot command.

It is more convenient to change the default boot command to
allow loading a FDT if it is stored in the boot dir of the
rootfs uSD/MMC partition.

If no FDT is found then the defaul command tries to boot a
zImage without a DT using legacy boot.

Signed-off-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
11 years agoARM: igep00x0.h: Enable the use of CMD_EXT4, CMD_FS_GENERIC and zImage.
Enric Balletbo i Serra [Wed, 7 Aug 2013 15:53:18 +0000 (17:53 +0200)]
ARM: igep00x0.h: Enable the use of CMD_EXT4, CMD_FS_GENERIC and zImage.

Able to load the kernel from some form of ext[234] or FAT. Also, with v3.9 and
later of the Linux Kernel, uImage isn't builtable anymore by default, so we
should switch to use the bootz command.

Signed-off-by: Enric Balletbo i Serra <eballetbo@gmail.com>
11 years agoomap: emif: Set initial DDR PHY config first
Taras Kondratiuk [Tue, 6 Aug 2013 13:16:50 +0000 (16:16 +0300)]
omap: emif: Set initial DDR PHY config first

Commit "OMAP5: emif/ddr: Change emif settings as required for ES1.0 silicon"
(f40107345cbcd6e0d1747eda45e76c4e2a6df0db)
changed sequence to set final DDR PHY config register value at the beginning.
Looks like it was made by mistake and should be reverted.

Signed-off-by: Taras Kondratiuk <taras@ti.com>
11 years agoARM: omap24xx: remove remainders of dead board
Masahiro Yamada [Mon, 5 Aug 2013 01:49:42 +0000 (10:49 +0900)]
ARM: omap24xx: remove remainders of dead board

Since Commit 7f5eef9 removed OMAP2420H4 support,
arm1136/omap24xx has not been used at all.

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
11 years agoARM: omap3: Implement dpll5 (HSUSB clk) workaround for OMAP36xx/AM/DM37xx according...
Naumann Andreas [Tue, 9 Jul 2013 07:43:17 +0000 (09:43 +0200)]
ARM: omap3: Implement dpll5 (HSUSB clk) workaround for OMAP36xx/AM/DM37xx according to errata sprz318e.

In chapter 'Advisory 2.1 USB Host Clock Drift Causes USB Spec Non-compliance in Certain Configurations' of the TI Errata it is recommended to use certain div/mult values for the DPLL5 clock setup.
So far u-boot used the old 34xx values, so I added the errata recommended values specificly for 36xx init only.
Also, the FSEL registers exist no longer, so removed them from init.

Tested this on a AM3703 board with 19.2MHz oscillator, which previously couldnt lock the dpll5 (kernel complained). As a consequence the EHCI USB port wasnt usable in U-Boot and kernel. With this patch, kernel panics disappear and USB working fine in u-boot and kernel.

Signed-off-by: Andreas Naumann <anaumann@ultratronik.de>
[trini: Add extern to <asm/arch-omap3/clock.h>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoomap: overo: Use 200MHz SDRC timings for revision 1, 2 & 3 boards
Ash Charles [Wed, 24 Jul 2013 19:22:35 +0000 (12:22 -0700)]
omap: overo: Use 200MHz SDRC timings for revision 1, 2 & 3 boards

Gumstix uses 200Mhz RAM on revision 1, 2 & 3 COMs, so use 200MHz
timings rather than 165MHz.  Based on 6cf8bf44b1f8550e12f7f2a16e01890e5de8443d

Signed-off-by: Ash Charles <ashcharles@gmail.com>
11 years agoomap: overo: update support for Micron 1GB POP
Steve Sakoman [Wed, 24 Jul 2013 19:22:34 +0000 (12:22 -0700)]
omap: overo: update support for Micron 1GB POP

Signed-off-by: Ash Charles <ashcharles@gmail.com>
11 years agoARM: IGEP0033: Remove duplicate / unused #defines.
Enric Balletbo i Serra [Thu, 25 Jul 2013 07:27:40 +0000 (09:27 +0200)]
ARM: IGEP0033: Remove duplicate / unused #defines.

As config was originally based on am335x_evm.h we have also some
duplicate / unnused #defines.

Commit 15191c91 removed these #defines on various AM335x boards but not
for IGEP COM AQUILA. This patch simply removes them for this board.

Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Reviewed-by: Javier Martinez Canillas <javier@dowhile0.org>
11 years agoARM: IGEP0033: Add support to boot from NAND.
Enric Balletbo i Serra [Thu, 25 Jul 2013 07:27:39 +0000 (09:27 +0200)]
ARM: IGEP0033: Add support to boot from NAND.

Add to the default environment the possibily to boot from NAND using
a ubi rootfs. Also the partition scheme is set as follows:

                  Start      Size
    SPL         : 0x00000000 0x00080000 (512KiB)
    U-Boot      : 0x00080000 0x00100000 (1MiB)
    U-Boot Env  : 0x00180000 0x00020000 (128KiB)
    File System : 0x001C0000 -

The ubiboot script gets the kernel and the dtb file from the boot directory
of the File System.

Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Reviewed-by: Javier Martinez Canillas <javier@dowhile0.org>
11 years agoARM: IGEP0033: Remove CYGNUS name from header.
Enric Balletbo i Serra [Thu, 25 Jul 2013 07:27:38 +0000 (09:27 +0200)]
ARM: IGEP0033: Remove CYGNUS name from header.

We will not use CYGNUS names for any IGEP COM based on AM335x processor,
so, to avoid confusion, remove from headers.

Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Reviewed-by: Javier Martinez Canillas <javier@dowhile0.org>
11 years agoARM: IGEP0033: Add support for Flattened Device Tree.
Enric Balletbò i Serra [Thu, 25 Jul 2013 07:27:37 +0000 (09:27 +0200)]
ARM: IGEP0033: Add support for Flattened Device Tree.

Now, the default kernel to boot the IGEP COM AQUILA is device tree based. As
old kernel is deprecated we should adapt the boot commands to use DTB files.

Also, with v3.9 and later of the Linux Kernel, uImage isn't builtable anymore
by default, so we should switch to use the bootz command.

Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Reviewed-by: Javier Martinez Canillas <javier@dowhile0.org>
11 years agoARM: IGEP0033: Remove undef of CONFIG_CMD_MEMTEST
Enric Balletbo i Serra [Thu, 25 Jul 2013 07:27:36 +0000 (09:27 +0200)]
ARM: IGEP0033: Remove undef of CONFIG_CMD_MEMTEST

After commit:

  79cd2f814b1c75efd47161ac27f4cbebf768240f config_cmd_default.h: Remove CONFIG_CMD_MEMTEST

It's not necessary to undef the CONFIG_CMD_MEMTEST, so we can remove it from
configuration file.

Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Reviewed-by: Javier Martinez Canillas <javier@dowhile0.org>
11 years agoARM: AM43xx: Add config file
Lokesh Vutla [Tue, 30 Jul 2013 06:06:32 +0000 (11:36 +0530)]
ARM: AM43xx: Add config file

Add config file

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
11 years agoARM: AM43xx: Add build support
Lokesh Vutla [Tue, 30 Jul 2013 06:06:31 +0000 (11:36 +0530)]
ARM: AM43xx: Add build support

Add AM43xx support in the required places

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
11 years agoARM: OMAP: Add CONFIG_OMAP_COMMON
Lokesh Vutla [Tue, 30 Jul 2013 06:06:30 +0000 (11:36 +0530)]
ARM: OMAP: Add CONFIG_OMAP_COMMON

Adding a new CONFIG_OMAP_COMMON which is included by all boards
that needs to build cpu/armv7/omap-common folder.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
11 years agoARM: AM43xx: clocks: Add dpll and clock data
Lokesh Vutla [Tue, 30 Jul 2013 06:06:29 +0000 (11:36 +0530)]
ARM: AM43xx: clocks: Add dpll and clock data

Add dpll and clock data for AM43xx

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
11 years agoARM: AM43xx: Add header files
Lokesh Vutla [Tue, 30 Jul 2013 06:06:28 +0000 (11:36 +0530)]
ARM: AM43xx: Add header files

Adding the following data:
-> Prcm structure
-> Base addresses
-> Pin mux structure.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
11 years agoARM: AM43xx: Add Board files
Lokesh Vutla [Tue, 30 Jul 2013 06:06:27 +0000 (11:36 +0530)]
ARM: AM43xx: Add Board files

Add board specific information for AM43xx.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
11 years agomusb: Disable extra prints
Lokesh Vutla [Tue, 30 Jul 2013 05:18:55 +0000 (10:48 +0530)]
musb: Disable extra prints

There are many musb prints in SPL and U-Boot log.
These prints are required only during musb debug.
So replacing printk with pr_debug in musb_core.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Tested-by: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
11 years agoARM: AM33xx: Move s_init to a common place
Heiko Schocher [Tue, 30 Jul 2013 05:18:54 +0000 (10:48 +0530)]
ARM: AM33xx: Move s_init to a common place

s_init has the same outline for all the AM33xx based
board. So making it generic.
This also helps in addition of new Soc with minimal changes.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>
Tested-by: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
11 years agoARM: AM33xx: Cleanup clocks layer
Lokesh Vutla [Tue, 30 Jul 2013 05:18:53 +0000 (10:48 +0530)]
ARM: AM33xx: Cleanup clocks layer

Cleaning up the clocks layer.
This helps in addition of new Soc with minimal
changes.
This is derived from OMAP4 boards.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Tested-by: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
11 years agoARM: AM33xx: Cleanup dplls data
Lokesh Vutla [Tue, 30 Jul 2013 05:18:52 +0000 (10:48 +0530)]
ARM: AM33xx: Cleanup dplls data

Locking sequence for all the dplls is same.
In the current code same sequence is done repeatedly
for each dpll. Instead have a generic function
for locking dplls and pass dpll data to that function.

This is derived from OMAP4 boards.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Tested-by: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
11 years agoMerge git://git.denx.de/u-boot-arm
Stefano Babic [Wed, 31 Jul 2013 09:30:38 +0000 (11:30 +0200)]
Merge git://git.denx.de/u-boot-arm

Conflicts:
board/freescale/mx6qsabrelite/Makefile
board/freescale/mx6qsabrelite/mx6qsabrelite.c
include/configs/mx6qsabrelite.h

Signed-off-by: Stefano Babic <sbabic@denx.de>
11 years agogpio: omap5-uevm: Configure the tca6424 gpio expander
Dan Murphy [Thu, 11 Jul 2013 18:10:28 +0000 (13:10 -0500)]
gpio: omap5-uevm: Configure the tca6424 gpio expander

Configure the tca6424 gpio expander
This allows use of the debug and tri color LEDs.

Signed-off-by: Dan Murphy <dmurphy@ti.com>
11 years agogpio: tca642x: Add the tca642x gpio expander driver
Dan Murphy [Thu, 11 Jul 2013 18:10:27 +0000 (13:10 -0500)]
gpio: tca642x: Add the tca642x gpio expander driver

Add the tca642x gpio expander driver

Datasheet:
http://www.ti.com/product/tca6424a

Signed-off-by: Dan Murphy <dmurphy@ti.com>
11 years agoam335x_evm: Add am335x_boneblack variant
Justin Waters [Thu, 11 Jul 2013 13:55:03 +0000 (09:55 -0400)]
am335x_evm: Add am335x_boneblack variant

The BeagleBone Black differs from the other AM335x boards in a few
significant ways, so it makes sense to create a custom configuration
for it. In particular, it uses eMMC instead of NAND flash.

Signed-off-by: Justin Waters <justin.waters@timesys.com>
11 years agoam335x_evm: Add support for eMMC environment
Justin Waters [Thu, 11 Jul 2013 13:55:02 +0000 (09:55 -0400)]
am335x_evm: Add support for eMMC environment

Some boards, such as the BeagleBone Black, have an eMMC chip intstead
of NAND. We can use the eMMC boot partition to store the environment,
since it isn't used for anything else. This allows us to have a
configurable environment on those boards.

Signed-off-by: Justin Waters <justin.waters@timesys.com>
11 years agoAdd additional MLO images to .gitignore
Justin Waters [Thu, 11 Jul 2013 13:55:01 +0000 (09:55 -0400)]
Add additional MLO images to .gitignore

This rule catches images such as MLO.byteswap

Signed-off-by: Justin Waters <justin.waters@timesys.com>
11 years agoam335x_evm: Rework bootcmd to handle two MMC devs
Justin Waters [Thu, 11 Jul 2013 13:55:00 +0000 (09:55 -0400)]
am335x_evm: Rework bootcmd to handle two MMC devs

The BeagleBone Black can boot from either the MMC card
or eMMC chip on board. We should try both interfaces.

This modification also allows a graceful fallback if
a device exists but boot images are not present on it.

Changes for v2:

* Fix boot partition - it should always show up as mmcblk0p2
* Fix missing FDT load

Signed-off-by: Justin Waters <justin.waters@timesys.com>
11 years agoam335x_evm: Add command line editing
Justin Waters [Thu, 11 Jul 2013 13:54:59 +0000 (09:54 -0400)]
am335x_evm: Add command line editing

Many modern U-Boot ports enable command line editing and
a history buffer. The am335x_evm configuration is fairly
comprehensive as it is, so a few extra kb should not be
noticable, and it adds a very convenient feature.

Signed-off-by: Justin Waters <justin.waters@timesys.com>
11 years agoam335x_evm: Make NAND support modular
Justin Waters [Thu, 11 Jul 2013 13:54:58 +0000 (09:54 -0400)]
am335x_evm: Make NAND support modular

Give the user the ability to disable NAND support by defining
CONFIG_NO_NAND. This will allow custom hardware to easily support
this configuration.

Signed-off-by: Justin Waters <justin.waters@timesys.com>
[trini: Make apply on top of other series]
Signed-off-by: Tom Rini <trini@ti.com>
11 years agonet, phy, cpsw: fix gigabit register access
Heiko Schocher [Tue, 23 Jul 2013 13:32:36 +0000 (15:32 +0200)]
net, phy, cpsw: fix gigabit register access

accessing a lan9303 switch with the cpsw driver results in wrong
speed detection, as the switch sets the BMSR_ERCAP in BMSR
register, and follow read of the MII_STAT1000 register fails, as
the switch does not support it. Current code did not check,
if a phy_read() fails ... fix this.

Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Joe Hershberger <joe.hershberger@gmail.com>
Acked-by: Mugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoboard/ti/am335x/README: Document NOR programming
Tom Rini [Thu, 18 Jul 2013 19:13:05 +0000 (15:13 -0400)]
board/ti/am335x/README: Document NOR programming

The Beaglebone White may be populated with a memory cape that has a NOR
module.  Document how to program it.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Add support to boot from NOR.
Steve Kipisz [Thu, 18 Jul 2013 19:13:04 +0000 (15:13 -0400)]
am335x_evm: Add support to boot from NOR.

NOR requires that s_init be within the first 4KiB of the image so that
we can perform the rest of the required pinmuxing to talk with the rest
of NOR that we are found on.  When NOR_BOOT is set we save our
environment in NOR at 512KiB and a redundant copy at 768KiB.  We avoid
using SPL for this case and u-boot.bin is written directly to the start
of NOR.

We enclose the DMM-related parts of arch/arm/cpu/armv7/am33xx/emif4.c
with TI81xx checks as at this time U-Boot does not discard unused
sections in the main build and this code relies on functions specific to
(and only provided in) ti81xx-related code.

Cc: Albert ARIBAUD <albert.u.boot@aribaud.net>
Signed-off-by: Steve Kipisz <s-kipisz2@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Add support for the NOR module on the memory cape
Steve Kipisz [Thu, 18 Jul 2013 19:13:03 +0000 (15:13 -0400)]
am335x_evm: Add support for the NOR module on the memory cape

This patch adds support for the NOR module that attaches
to the memory cape for a Beaglebone board. This does not
add booting support; only support so that you can boot from
SD/MMC and see the NOR module so that it can be programmed.

Signed-off-by: Steve Kipisz <s-kipisz2@ti.com>
[trini: Clean up config changes slightly]
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam33xx: Correct gpmc_cfg->irqstatus/enable
Tom Rini [Thu, 18 Jul 2013 19:13:02 +0000 (15:13 -0400)]
am33xx: Correct gpmc_cfg->irqstatus/enable

Based on our usage of the GPMC, either with NOR or NAND we do not need
to be setting the irqstatus or irqenable bits and should clear them like
we have historically.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Rework board_is_foo() checks
Tom Rini [Thu, 18 Jul 2013 19:13:01 +0000 (15:13 -0400)]
am335x_evm: Rework board_is_foo() checks

We rework the various board_is_foo() checks to take a pointer to
struct am335x_baseboard_id rather than using a local copy in board.c.
This allows us to make use of the same checks in mux.c as well as fixing
problems when this code could be running from read-only memory.

Reviewed-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Update SPI_BOOT support, add MTDPARTS info
Tom Rini [Thu, 18 Jul 2013 19:12:59 +0000 (15:12 -0400)]
am335x_evm: Update SPI_BOOT support, add MTDPARTS info

- Style cleanup (# define -> #define)
- Due to ROM issues, redudant loading isn't feasible, so drop.
- Given extra space, increase max size of U-Boot to 512KiB
- Correct env size to match usage (we had not re-defined ENV_SIZE).
- Given extra space, keep env size as 128KiB, add redundant environment.

Reviewed-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Drop useless CONFIG_ENV_IS_NOWHERE
Tom Rini [Thu, 18 Jul 2013 19:12:58 +0000 (15:12 -0400)]
am335x_evm: Drop useless CONFIG_ENV_IS_NOWHERE

We always set a CONFIG_ENV_IS_...somewhere... so drop the initial define
of NOWHERE.

Reviewed-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Tom Rini <trini@ti.com>
11 years agoboard/ti/am335x/README: Document NAND programming
Tom Rini [Wed, 17 Jul 2013 16:24:30 +0000 (12:24 -0400)]
board/ti/am335x/README: Document NAND programming

The AM335x GP EVM ships with NAND.  Document programming of the chip
including the redundant locations that the ROM will check.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agomxc_ipuv3: fix memory alignment of framebuffer
Eric Nelson [Sat, 27 Jul 2013 00:53:45 +0000 (17:53 -0700)]
mxc_ipuv3: fix memory alignment of framebuffer

The frame-buffer on i.MX boards needs to be aligned for DMA.

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
11 years agoimx: nitrogen6x: mx6qsabrelite: Add support for DVI monitors
Robert Winkler [Tue, 23 Jul 2013 22:07:56 +0000 (15:07 -0700)]
imx: nitrogen6x: mx6qsabrelite: Add support for DVI monitors

A little background is probably appropriate for this patch.

Since "the beginning" of usage of the SABRE Lite and Nitrogen6x
boards, DVI detection has been somewhat broken.

Some (most) DVI monitors don't produce the "HPD" bit in
the PHY_STAT0 register, but do show proper toggling of the
RX_SENSE0..3 bits.

Creating a new the bit-mask to include all five bits and
modifying the 'hdmidet' command and internal detection
routines allows these monitors to function properly in U-Boot.

A related patch to our kernels allows things to work under
Linux:
        https://github.com/boundarydevices/linux-imx6/commit/7d8752905c118af9063738a533227de0b2f6ecd4

Signed-off-by: Robert Winkler <robert.winkler@boundarydevices.com>
Acked-by: Stefano Babic <sbabic@denx.de>
11 years agoimx: Add documentation for imx specific commands
Robert Winkler [Thu, 25 Jul 2013 00:57:06 +0000 (17:57 -0700)]
imx: Add documentation for imx specific commands

CONFIG_CMD_HDMIDETECT
CONFIG_CMD_BMODE

Signed-off-by: Robert Winkler <robert.winkler@boundarydevices.com>
Acked-by: Otavio Salvador <otavio@ossystems.com.br>
11 years agomx6qsabresd: Add splash screen support via HDMI
Pardeep Kumar Singla [Thu, 25 Jul 2013 17:12:14 +0000 (12:12 -0500)]
mx6qsabresd: Add splash screen support via HDMI

Signed-off-by: Pardeep Kumar Singla <b45784@freescale.com>
11 years agomx6: Factor out common HDMI setup code
Pardeep Kumar Singla [Thu, 25 Jul 2013 17:12:13 +0000 (12:12 -0500)]
mx6: Factor out common HDMI setup code

Instead of duplicating HDMI setup code for every mx6 board, factor out the common code

Signed-off-by: Pardeep Kumar Singla <b45784@freescale.com>
Acked-By: Eric Nelson <eric.nelson@boundarydevices.com>
11 years agoomap3_beagle: support booting from zImage and device tree as last option
Nishanth Menon [Mon, 15 Jul 2013 12:11:33 +0000 (07:11 -0500)]
omap3_beagle: support booting from zImage and device tree as last option

If no other bootoption works, try loading up device tree and zImage.

This is selected as the last option to allow backward compatibility as
well as support the recent trend in moving kernel boot to using zImage
and device tree.

NOTE: if uImage is present in bootpart, it will try this first and
will assume this is to be booted with bootm (so may be concatenated
image or plain vanilla ATAG MACHINE_ID based image)

Signed-off-by: Nishanth Menon <nm@ti.com>
11 years agoomap3_beagle: support findfdt and loadfdt for devicetree support
Nishanth Menon [Mon, 15 Jul 2013 12:11:32 +0000 (07:11 -0500)]
omap3_beagle: support findfdt and loadfdt for devicetree support

For folks not using concatenated device tree with uImage, having
an handy function to find and load device tree is very handy.

So introduce findfdt and loadfdt and run findfdt by default to make
it easier on user scripts.

Signed-off-by: Nishanth Menon <nm@ti.com>
11 years agoomap3_beagle: enable CMD_FS_GENERIC and simplify load of image/ramdisk
Nishanth Menon [Mon, 15 Jul 2013 12:11:31 +0000 (07:11 -0500)]
omap3_beagle: enable CMD_FS_GENERIC and simplify load of image/ramdisk

CMD_FS_GENERIC allows us to simplify where we load up our image from
either from ext2/fat etc. So, lets use that instead of cumbersome
options we currently use. Sticking with existing conventions,
defaults will be:
ramdisk=ramdisk.gz
bootpart=0:2 (second partition)
bootdir=/boot (/boot in second partition)

This matches with the default behavior, these can be overriden by
env files as needed.

Signed-off-by: Nishanth Menon <nm@ti.com>
11 years agobeagleboard: remove RevB support for BeagleBoard Xm
Nishanth Menon [Mon, 15 Jul 2013 12:11:30 +0000 (07:11 -0500)]
beagleboard: remove RevB support for BeagleBoard Xm

As reported in http://marc.info/?l=u-boot&m=137358037827735&w=2

There is no need for the "xMB" variant, as the gpio pins used for
identification where never changed from the xMA when the newer silcon
was used for the xMB, So rename XM A revision as AB revision
and report accordingly

Reported-by: Robert Nelson <robertcnelson@gmail.com>
Signed-off-by: Nishanth Menon <nm@ti.com>
11 years agoomap3_beagle: replace uImage.beagle with generic uImage
Nishanth Menon [Mon, 15 Jul 2013 12:11:29 +0000 (07:11 -0500)]
omap3_beagle: replace uImage.beagle with generic uImage

e682930867f7dfc4a01784fe452fad9e962d65a
(BeagleBoard: config: use uImage.beagle for tftp)

Introduced uImage.beagle which does not happen to be default output
file of Linux kernel build make uImage (output is uImage).

So, replace uImage.beagle with uImage

Signed-off-by: Nishanth Menon <nm@ti.com>
11 years agoomap3_beagle: remove JFFS2 support.
Nishanth Menon [Mon, 15 Jul 2013 12:11:28 +0000 (07:11 -0500)]
omap3_beagle: remove JFFS2 support.

We do not use JFFS2 by default and it conflicts with
CONFIG_CMD_FS_GENERIC (ls command is the same). Since most of our
BOOTCMD can be simplified by using the FS_GENERIC, dropping JFFS2

Signed-off-by: Nishanth Menon <nm@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
11 years agoARM: DRA7xx: Enable CPSW Ethernet support
Mugunthan V N [Mon, 8 Jul 2013 10:34:43 +0000 (16:04 +0530)]
ARM: DRA7xx: Enable CPSW Ethernet support

Enabling CPSW Ethernet support in DRA7xx EVM.

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agoARM: DRA7xx: Add CPSW and MDIO pinmux support
Mugunthan V N [Mon, 8 Jul 2013 10:34:42 +0000 (16:04 +0530)]
ARM: DRA7xx: Add CPSW and MDIO pinmux support

Adding CPSW Slave 0 and MDIO pinmux support for DRA7xx EVM

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agoARM: DRA7xx: Add CPSW support to DRA7xx EVM
Mugunthan V N [Mon, 8 Jul 2013 10:34:41 +0000 (16:04 +0530)]
ARM: DRA7xx: Add CPSW support to DRA7xx EVM

Adding support for CPSW Ethernet support found in DRA7xx EVM

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agoARM: DRA7xx: Enable GMAC clock control
Mugunthan V N [Mon, 8 Jul 2013 10:34:40 +0000 (16:04 +0530)]
ARM: DRA7xx: Enable GMAC clock control

Enabling CPSW module by enabling GMAC clock control

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agoARM: DRA7xx: Lock DPLL_GMAC
Lokesh Vutla [Mon, 8 Jul 2013 10:34:39 +0000 (16:04 +0530)]
ARM: DRA7xx: Lock DPLL_GMAC

Locking DPLL_GMAC

[mugunthanvnm@ti.com:Configure only if CPSW is selected]

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agodrivers: net: cpsw: Enable statistics for all port
Mugunthan V N [Mon, 8 Jul 2013 10:34:38 +0000 (16:04 +0530)]
drivers: net: cpsw: Enable statistics for all port

Enable hardware statistics for all ports, enabling only to host port is useless

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agodrivers: net: cpsw: remove hard coding bd ram for cpsw
Mugunthan V N [Mon, 8 Jul 2013 10:34:37 +0000 (16:04 +0530)]
drivers: net: cpsw: remove hard coding bd ram for cpsw

BD ram address may vary in various SOC, so removing the hardcoding and
passing the same information through platform data

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
11 years agoam335x_evm: Add basic README
Tom Rini [Mon, 8 Jul 2013 16:15:18 +0000 (12:15 -0400)]
am335x_evm: Add basic README

Add a README for the family of boards the am335x_evm covers, and include
instructions on preparing and using falcon mode, for various media.

Signed-off-by: Tom Rini <trini@ti.com>
Reviewed-by: Peter Korsgaard <jacmet@sunsite.dk>
Reviewed-by: Lukasz Majewski <l.majewski@samsung.com>
11 years agoam335x_evm: Correct CONFIG_CMD_SPL_WRITE_SIZE
Tom Rini [Mon, 8 Jul 2013 16:15:17 +0000 (12:15 -0400)]
am335x_evm: Correct CONFIG_CMD_SPL_WRITE_SIZE

We use CONFIG_CMD_SPL_WRITE_SIZE when reading/writing the args portion
of falcon mode to NAND.  Previously it was half the size of the
eraseblock which is too small, increase to eraseblock size.

Signed-off-by: Tom Rini <trini@ti.com>
11 years agoam335x_evm: Update eMMC falcon mode locations
Tom Rini [Mon, 8 Jul 2013 16:15:16 +0000 (12:15 -0400)]
am335x_evm: Update eMMC falcon mode locations

The previous location used for the "args" portion of falcon mode was too
small to allow for a device tree to be saved there, so move the location
slightly and increase the size.  In addition, our previous kernel
location was part of the area we set aside for U-Boot itself, so move it
up a bit higher.

Signed-off-by: Tom Rini <trini@ti.com>
Reviewed-by: Peter Korsgaard <jacmet@sunsite.dk>
11 years agoam335x_evm: Correct DFU ALT settings for falcon mode
Tom Rini [Mon, 8 Jul 2013 16:15:15 +0000 (12:15 -0400)]
am335x_evm: Correct DFU ALT settings for falcon mode

Now that we have falcon mode enabled, the partiton numbers for NAND have
changed, and we need to list entries for updating these parts of the
system.  While adding falcon mode entires for eMMC (raw), we round up
the limit on U-Boot for ease of math later.

Signed-off-by: Tom Rini <trini@ti.com>
Reviewed-by: Peter Korsgaard <jacmet@sunsite.dk>