project/bcm63xx/u-boot.git
14 years agoReplace CONFIG_SYS_GBL_DATA_SIZE by auto-generated value
Wolfgang Denk [Tue, 26 Oct 2010 12:34:52 +0000 (14:34 +0200)]
Replace CONFIG_SYS_GBL_DATA_SIZE by auto-generated value

CONFIG_SYS_GBL_DATA_SIZE has always been just a bad workarond for not
being able to use "sizeof(struct global_data)" in assembler files.
Recent experience has shown that manual synchronization is not
reliable enough.  This patch renames CONFIG_SYS_GBL_DATA_SIZE into
GENERATED_GBL_DATA_SIZE which gets automatically generated by the
asm-offsets tool.  In the result, all definitions of this value can be
deleted from the board config files.  We have to make sure that all
files that reference such data include the new <asm-offsets.h> file.

No other changes have been done yet, but it is obvious that similar
changes / simplifications can be done for other, related macro
definitions as well.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoinclude/asm-offsets.h: automatically generate assembler constants
Wolfgang Denk [Mon, 25 Oct 2010 22:08:35 +0000 (00:08 +0200)]
include/asm-offsets.h: automatically generate assembler constants

A recurrent issue is that certain C level constructs like sizeof() or
offsetof() cannot be used in assembler files, which is inconvenient
when such constructs are used in the definition of macro names etc.

To avoid duplication of such definitions (and thus another cause of
problems), we adapt the Linux way to automatically generate the
respective definitions from the respective C header files.

In Linux, this is implemented in include/linux/kbuild.h, Kbuild, and
arch/*/kernel/asm-offsets.c; we adapt the code from the Linux v2.6.36
kernel tree.

We also copy the concept of the include/generated/ directory which can
be used to hold other automatically generated files as well.

We start with an architecture-independent lib/asm-offsets.c which
generates include/generated/generic-asm-offsets.h (included by
include/asm-offsets.h, which is what will be referred to in the actual
source code).  Later this may be extended by architecture-specific
arch/*/lib/asm-offsets.c files that will generate a
include/generated/asm-offsets.h.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoRename CONFIG_SYS_INIT_RAM_END into CONFIG_SYS_INIT_RAM_SIZE
Wolfgang Denk [Tue, 26 Oct 2010 11:32:32 +0000 (13:32 +0200)]
Rename CONFIG_SYS_INIT_RAM_END into CONFIG_SYS_INIT_RAM_SIZE

CONFIG_SYS_INIT_RAM_END was a misnomer as it suggests this might be
some end address; to make the meaning more clear we rename it into
CONFIG_SYS_INIT_RAM_SIZE

No other code changes are performed in this patch, only minor editing
of white space (due to the changed length) and the comments was done,
where noticed.

Note that the code for the PATI and cmi_mpc5xx board configurations
looks seriously broken.  Last known maintainers on Cc:

Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Denis Peter <d.peter@mpl.ch>
Cc: Martin Winistoerfer <martinwinistoerfer@gmx.ch>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoMerge branch 'master' of git://git.denx.de/u-boot-blackfin
Wolfgang Denk [Tue, 26 Oct 2010 18:58:49 +0000 (20:58 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-blackfin

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Tue, 26 Oct 2010 18:57:41 +0000 (20:57 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-i2c
Wolfgang Denk [Tue, 26 Oct 2010 18:55:39 +0000 (20:55 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-i2c

14 years agoARM: Use consistent assembler syntax
Gray Remlin [Sun, 24 Oct 2010 15:18:31 +0000 (16:18 +0100)]
ARM: Use consistent assembler syntax

Signed-off-by: Gray Remlin <g_remlin@rocketmail.com>
Acked-by: Heiko Schocher <hs@denx.de>
14 years agoBlackfin: adi boards: set compiled size limits
Mike Frysinger [Tue, 19 Oct 2010 22:02:43 +0000 (18:02 -0400)]
Blackfin: adi boards: set compiled size limits

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf527-ezkit-v2: move to boards.cfg
Mike Frysinger [Tue, 19 Oct 2010 06:41:26 +0000 (02:41 -0400)]
Blackfin: bf527-ezkit-v2: move to boards.cfg

Now that the boards.cfg file supports options to mkconfig, we can move
the bf527-ezkit-v2 target out of the Makefile and into boards.cfg.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoppc4xx: Add Io and IoCon 405EP board support
Dirk Eibach [Thu, 21 Oct 2010 08:50:05 +0000 (10:50 +0200)]
ppc4xx: Add Io and IoCon 405EP board support

Board support for the Guntermann & Drunck CATCenter Io.
Board support for the Guntermann & Drunck IoCon.

Signed-off-by: Dirk Eibach <eibach@gdsys.de>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoMerge branch 'master' of ssh://gemini/home/wd/git/u-boot/master
Wolfgang Denk [Mon, 25 Oct 2010 06:06:52 +0000 (08:06 +0200)]
Merge branch 'master' of ssh://gemini/home/wd/git/u-boot/master

14 years agoARMV7: OMAP: I2C driver: Fix bug found in 37XX testing
Steve Sakoman [Fri, 22 Oct 2010 20:48:00 +0000 (13:48 -0700)]
ARMV7: OMAP: I2C driver: Fix bug found in 37XX testing

On OMAP36/37XX the standard on chip pullups are not sufficient to
ensure proper i2c operation without external pullups or switching
to high speed mode and enabling special on chip pullups.

This is an issue for Beagle xM, which does not have external pullups
on the expansion board i2c lines.

The issue manifests itself as an AL (arbitration lost) error when
probing for a non-existent device (i.e. on a Beagle xM with no expansion
boards attached).  This issue does not occur on expansion boards that
include pullups or on Overo 37XX COM's since they include pull-ups.

This patch fixes the issue by checking for the AL bit in the i2c_probe
function.

Signed-off-by: Steve Sakoman <steve.sakoman@linaro.org>
14 years agohcu4, hcu5: fix out of tree building
Wolfgang Denk [Sun, 24 Oct 2010 14:49:12 +0000 (16:49 +0200)]
hcu4, hcu5: fix out of tree building

Out of tree building of the Netstal hcu4 and hcu5 boards failed like
that:

Assembler messages:
Fatal error: can't create /work/wd/tmp-ppc/board/netstal/hcu4/../common/fixed_sdram.o: No such file or directory
Assembler messages:
Fatal error: can't create /work/wd/tmp-ppc/board/netstal/hcu4/../common/nm_bsp.o: No such file or directory
make[1]: *** [/work/wd/tmp-ppc/board/netstal/hcu4/../common/fixed_sdram.o] Error 2

Adapt (and simplify) the respective Makefiles.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Niklaus Giger <niklaus.giger@netstal.com>
14 years agoMPC8315ERD: fix build error
Wolfgang Denk [Sun, 24 Oct 2010 14:07:23 +0000 (16:07 +0200)]
MPC8315ERD: fix build error

Commit 29c6fbe "MPC5121: Add USB EHCI support" renamed
CONFIG_SYS_MPC8xxx_USB_ADDR into CONFIG_SYS_FSL_USB_ADDR but missed
to update arch/powerpc/cpu/mpc83xx/cpu_init.c, resulting in:

cpu_init.c: In function 'cpu_init_f':
cpu_init.c:332: error: 'CONFIG_SYS_MPC8xxx_USB_ADDR' undeclared (first use in this function)
cpu_init.c:332: error: (Each undeclared identifier is reported only once
cpu_init.c:332: error: for each function it appears in.)
make[1]: *** [/work/wd/tmp-ppc/arch/powerpc/cpu/mpc83xx/cpu_init.o] Error 1

Fix this.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Kim Phillips <kim.phillips@freescale.com>
14 years agoVoVPN-GW_100MHz: drop unsupported board configuration
Wolfgang Denk [Sun, 24 Oct 2010 13:46:08 +0000 (15:46 +0200)]
VoVPN-GW_100MHz: drop unsupported board configuration

The 100MHz configuation of the VoVPN-GW has never been supported, so
drop it now.

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agolite5200b_PM: fix compile warning
Wolfgang Denk [Sun, 24 Oct 2010 13:37:12 +0000 (15:37 +0200)]
lite5200b_PM: fix compile warning

Fix warning:

icecube.c: In function 'lite5200b_wakeup':
icecube.c:83: warning: format '%08lx' expects type 'long unsigned int', but argument 2 has type 'void (*)(void)'

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoMAKEALL: drop non-existent i386 config
Mike Frysinger [Wed, 20 Oct 2010 07:34:19 +0000 (03:34 -0400)]
MAKEALL: drop non-existent i386 config

14 years agoARM: fix relocation support for onenand device.
Enric Balletbo i Serra [Tue, 19 Oct 2010 09:13:22 +0000 (11:13 +0200)]
ARM: fix relocation support for onenand device.

We also have to relocate the onenand command table manually, otherwise
onenand command don't work.

Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
14 years agoMVBLM7: make TSEC2 work again.
Andre Schwarz [Fri, 22 Oct 2010 09:21:46 +0000 (11:21 +0200)]
MVBLM7: make TSEC2 work again.

SICRH has been misconfigured, i.e. TSEC2 clock + D[0:3] are GPIOs.
Fix this to be RGMII signals again.

Signed-off-by: Andre Schwarz <andre.schwarz@matrix-vision.de>
14 years agohwconfig: Utilize getenv_f before relocation to allow for larger buffer
Kumar Gala [Fri, 22 Oct 2010 08:18:13 +0000 (03:18 -0500)]
hwconfig: Utilize getenv_f before relocation to allow for larger buffer

Since we use hwconfig in cases before relocation (like getting DDR
params on FSL PPC systems), we can have strings that exceed the early
small (32 byte) buffer size that getenv will handle.

So we explicitly allocate our own buffer on the stack and use if to
handle getting the hwconfig env string.  We currently utilize a string
length of 128 bytes.

This allows us to get rid of boot messages like:

env_buf too small [32]

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoenv_flash: Disable debug print statements
Peter Tyser [Fri, 22 Oct 2010 05:24:52 +0000 (00:24 -0500)]
env_flash: Disable debug print statements

With debug the follow is printed:
  => saveenv
  Saving Environment to Flash...
  Data to save 0x18000
  Data (start 0xfff48000, len 0x18000) saved at 0x7fe63f20
  Protect off FFF40000 ... FFF5FFFF
  Un-Protected 1 sectors
  Erasing Flash...
  . done
  Erased 1 sectors
  Writing to Flash... Restoring the rest of data to 0xfff48000 len 0x18000
  done
  Protected 1 sectors
  =>

Without debug:
  => saveenv
  Saving Environment to Flash...
  Un-Protected 1 sectors
  Erasing Flash...
  . done
  Erased 1 sectors
  Writing to Flash... done
  Protected 1 sectors
  =>

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
14 years agoMerge branch 'for-wd-master' of git://git.denx.de/u-boot-pxa
Wolfgang Denk [Sat, 23 Oct 2010 20:08:33 +0000 (22:08 +0200)]
Merge branch 'for-wd-master' of git://git.denx.de/u-boot-pxa

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Sat, 23 Oct 2010 19:59:42 +0000 (21:59 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-video
Wolfgang Denk [Sat, 23 Oct 2010 19:56:51 +0000 (21:56 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-video

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-usb
Wolfgang Denk [Sat, 23 Oct 2010 19:49:28 +0000 (21:49 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-usb

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-sh
Wolfgang Denk [Sat, 23 Oct 2010 19:45:57 +0000 (21:45 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-sh

14 years agoa4m072: add CONFIG_SYS_FLASH_BANKS_SIZES define
Ilya Yanok [Thu, 21 Oct 2010 15:20:13 +0000 (17:20 +0200)]
a4m072: add CONFIG_SYS_FLASH_BANKS_SIZES define

This patch adds CONFIG_SYS_FLASH_BANKS_SIZES define to make use of new
cfi_flash driver ability to detect flash chips that are bigger than a
corresponding address window (we have such situation on some revs of
a4m072).

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
14 years agocfi_flash: handle 'chip size exceeds address window' situation
Ilya Yanok [Thu, 21 Oct 2010 15:20:12 +0000 (17:20 +0200)]
cfi_flash: handle 'chip size exceeds address window' situation

On some boards we have flash mapped high in the address space with
considerably small window (say 0xFE000000 and 32MB). When we install
bigger chip (say 64MB) on such a board strange things happen
(flash_write() doesn't work at all, for ex). That's because cfi_flash
driver doesn't care about window size at all.
Of course, cleanest solution would probably be to just extend address
window to be able to map the whole flash but for legacy/compatibility
reasons some people prefer just truncate the flash size and never use
the upper part.
This patch adds an option for cfi_flash driver to handle this situation
properly. To achieve this we add the new function cfi_flash_bank_size()
which can be provided by the board code and weak-aliased to default
implementation that returns value from the CONFIG_SYS_FLASH_BANKS_SIZES
array if it's defined or 0 otherwise (the last case is added for
compatibility).
If non-zero flash bank size is provided and detected chip size is bigger
than provided address window size the warning will be displayed and
flash chip will be truncated.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Changed cfi_flash_bank_size() return type to unsigned long
to match caller function.
Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoehci-hcd.c: fix hanging under higher load
Wolfgang Denk [Fri, 22 Oct 2010 12:23:00 +0000 (14:23 +0200)]
ehci-hcd.c: fix hanging under higher load

This patch solves a problem with USB hanging under higher load on a
i.MX31 board.  It falls into class of typical USB problems and fixes:
if you don't understand the real cause, add a delay somewhere.

The problem appeared after introduction of ELF relocation, which
results in smaller code, which appears to run faster (probably because
it fits better in the cache); turning off the instruction cache,
adding debug printf()s and increasing the delay have all been found to
make the problem go away.

Moving the original "udelay(1)" up in the code to it's new place made
the problem appear much less frequently. Increasing the delay to 2
microseconds then made the code run reliably in all (hour-long) tests.
To be on the safe side, we set it to 5 microseconds here.

Signed-off-by: Heiko schocher <hs@denx.de>
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Remy Bohmer <linux@bohmer.net>
Cc: Stefano Babic <sbabic@denx.de>
14 years agousb_storage: constify us_direction lookup table
Mike Frysinger [Wed, 20 Oct 2010 11:16:04 +0000 (07:16 -0400)]
usb_storage: constify us_direction lookup table

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agousb: musb: constify a bit
Mike Frysinger [Wed, 20 Oct 2010 11:15:35 +0000 (07:15 -0400)]
usb: musb: constify a bit

These ep and root hub structures need not be writable.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoUSB: fix Queue Element Transfer Descriptor changes
Wolfgang Denk [Wed, 20 Oct 2010 19:08:17 +0000 (21:08 +0200)]
USB: fix Queue Element Transfer Descriptor changes

Commit 3ed1607 "USB: sync Queue Element Transfer Descriptor against
EHCI spec" added an "__attribute__ ((aligned (32)))" to the
declaration of struct qTD, as used for example in the Linux kernel as
well.

However, it turns out that this attribute causes errors in "usb start"
(like "ERROR: NOT USB_CONFIG_DESC 7b" and similar). Drop the attribute
again.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Dan Lykowski <lykowdk@gmail.com>
Cc: Remy Bohmer <linux@bohmer.net>
Cc: Stefano Babic <sbabic@denx.de>
14 years agoehci-pci: print hccr, hcor and hc_lenght
Florian Fainelli [Fri, 15 Oct 2010 13:53:45 +0000 (15:53 +0200)]
ehci-pci: print hccr, hcor and hc_lenght

It is useful to know the EHCI-PCI hccr, hcor and hc_lenght to make sure it was
successfully registered, and at the correct location.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
14 years agoehci-hcd.c: fix hanging under higher load
Wolfgang Denk [Fri, 22 Oct 2010 12:08:13 +0000 (14:08 +0200)]
ehci-hcd.c: fix hanging under higher load

This patch solves a problem with USB hanging under higher load on a
i.MX31 board.  It falls into class of typical USB problems and fixes:
if you don't understand the real cause, add a delay somewhere.

The problem appeared after introduction of ELF relocation, which
results in smaller code, which appears to run faster (probably because
it fits better in the cache); turning off the instruction cache,
adding debug printf()s and increasing the delay have all been found to
make the problem go away.

Moving the original "udelay(1)" up in the code to it's new place made
the problem appear much less frequently. Increasing the delay to 2
microseconds then made the code run reliably in all (hour-long) tests.
To be on the safe side, we set it to 5 microseconds here.

Signed-off-by: Heiko schocher <hs@denx.de>
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Remy Bohmer <linux@bohmer.net>
Cc: Stefano Babic <sbabic@denx.de>
14 years agoXPedite5500 board support
John Schmoller [Fri, 22 Oct 2010 05:20:34 +0000 (00:20 -0500)]
XPedite5500 board support

Initial support for Extreme Engineering Solutions XPedite5500 -
a P2020-based PMC/XMC single board computer.

Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoxes: Add POST support for 8xxx boards
Peter Tyser [Fri, 22 Oct 2010 05:20:33 +0000 (00:20 -0500)]
xes: Add POST support for 8xxx boards

Add memory and I2C posts to the XPedite517x/520x/537x board families.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agompc86xx: Add generic POST word read/write support
Peter Tyser [Fri, 22 Oct 2010 05:20:32 +0000 (00:20 -0500)]
mpc86xx: Add generic POST word read/write support

The PIC's TFRR register doesn't affect hardware and is generally unused,
so use as storage for the POST word.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agopost/i2c: Add ability to ignore I2C devices
Peter Tyser [Fri, 22 Oct 2010 05:20:31 +0000 (00:20 -0500)]
post/i2c: Add ability to ignore I2C devices

Add the ability to not report an I2C POST error for a set of given I2C
addresses on bootup.  This is useful for cases when a device may or may
not be present, and neither case is considered an error.  For example:

- Some form factors such as XMC and Compact PCI Express have an I2C
  EEPROM whose address changes based on geographical address.  Eg
  installed in one slot its EEPROM address is, 0x50, in another its
  0x51, etc.  This allows multiple devices to have their EEPROMs present
  on the same I2C bus.  Thus the I2C devices present for an XMC or
  CPCIe card depend on if and where other cards are installed in the
  same system.

- Some cards have optional I2C devices.  Eg one hardware build
  configuration has different I2C devices than another and software
  can't determine if the optional device should be present or not.

- Some cards have optional daughtercards with I2C devices on them.

- I2C EEPROMs address range depends on their size.  Its possible to
  support differently size EEPROMs by only probing the EEPROM's base
  address and ignoring the other addresses that are impacted by its
  size.

A new CONFIG_SYS_POST_I2C_IGNORES define has been added which specifies
a list of I2C addresses for the I2C POST to ignore.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agopost/i2c: Rename I2C_ADDR_LIST to CONFIG_SYS_POST_I2C_ADDRS
Peter Tyser [Fri, 22 Oct 2010 05:20:30 +0000 (00:20 -0500)]
post/i2c: Rename I2C_ADDR_LIST to CONFIG_SYS_POST_I2C_ADDRS

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agopost/i2c: Don't probe address 0
Peter Tyser [Fri, 22 Oct 2010 05:20:29 +0000 (00:20 -0500)]
post/i2c: Don't probe address 0

According to the I2C specification device address 0 is the "general call
address", ie a broadcast address.  The I2C specification states that the
format of a general call uses at least 2 bytes, which U-Boot's probing
routine does not adhere to.

Not probing device address 0 will prevent possible issues with devices
that accept general calls.  Additionally, this change shouldn't reduce
POST coverage since each I2C device should still be accessed via its
own, unique address.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agopost/i2c: Clean up detection logic
Peter Tyser [Fri, 22 Oct 2010 05:20:28 +0000 (00:20 -0500)]
post/i2c: Clean up detection logic

The logic previously used in the I2C post was a bit convoluted.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agopost/i2c: General clean up
Peter Tyser [Fri, 22 Oct 2010 05:20:27 +0000 (00:20 -0500)]
post/i2c: General clean up

- Clean up ifdeffery
- Update coding style

No functional change should have occurred.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoxes: Make X-ES board names more generic
Peter Tyser [Fri, 22 Oct 2010 05:20:26 +0000 (00:20 -0500)]
xes: Make X-ES board names more generic

Some U-Boot images for X-ES boards support multiple products in the same
family.  For example, the XPedite5370, XPedite5371, and XPedite5372 are
similar enough that one U-Boot image can work on all 3 cards.  To make it
clear that a U-Boot image can work on boards of the same family, rename
the boards with the least significant digit of 'x'.

While we're at it, change the board config file and make targets to be
lowercase.

Also change the default uImage and fdt filenames to "board.uImage" and
"board.dtb" to be more generic.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoxes: Add board_flash_wp_on()
John Schmoller [Fri, 22 Oct 2010 05:20:25 +0000 (00:20 -0500)]
xes: Add board_flash_wp_on()

Add board_flash_wp_on() to check a pca9557 gpio pin to see
if non-volatile memory write protection is enabled.

Previously, write protected NOR flashes would fail initialization which
resulted in a bootup error such as:

 ...
 DTT:   53 C local / 64 C remote (adt7461@4c)
 DTT:   54 C local (ds1621@48)
 FLASH: Executed from FLASH1
 POST memory PASSED
 FLASH: ## Unknown FLASH on Bank 1 - Size = 0x00000000 = 0 MB
 ## Unknown FLASH on Bank 2 - Size = 0x00000000 = 0 MB
 *** failed ***
 ### ERROR ### Please RESET the board ###

With this patch, NOR flash initialization is skipped:

 ...
 DTT:   53 C local / 64 C remote (adt7461@4c)
 DTT:   54 C local (ds1621@48)
 FLASH: Executed from FLASH1
 POST memory PASSED
 FLASH: Uninitialized - Write Protect On
 L2:    1024 KB enabled
 NAND:  1024 MiB
 ...

Note that flash related commands such as flinfo and saveenv will error
out when flash write protection is enabled.

Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoxes: Consolidate checkboard()
John Schmoller [Fri, 22 Oct 2010 05:20:24 +0000 (00:20 -0500)]
xes: Consolidate checkboard()

Create a common checkboard() function to support all X-ES's Freescale
boards.

Also, add a get_board_derivative() function which reads hardware
strapping resistors to determine what model a board is.  This allows one
U-Boot image to support multiple boards.

Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agopci: Add ability to re-enumerate PCI buses
John Schmoller [Fri, 22 Oct 2010 05:20:23 +0000 (00:20 -0500)]
pci: Add ability to re-enumerate PCI buses

Add a new 'pci enum' command which re-enumerates the PCI buses.  This
command is enabled via the CONFIG_CMD_PCI_ENUM define and can be useful
in boards with FPGAs connected via PCI/PCIe, boards that support PCI
hot-plugging, or during PCI debug.

Also enable the 'pci enum' command for X-ES's Freescale-based boards.

Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Wolfgang Denk <wd@denx.de>
14 years agoxes: Use common PCI initialization code
Peter Tyser [Fri, 22 Oct 2010 05:20:22 +0000 (00:20 -0500)]
xes: Use common PCI initialization code

Common Freescale code for PCI initialization now exists, so migrate X-ES
boards to use it.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agocorenet_ds: Update CONFIG_SYS_GBL_DATA_SIZE to deal with growth in gd_t
Kumar Gala [Fri, 22 Oct 2010 06:32:56 +0000 (01:32 -0500)]
corenet_ds: Update CONFIG_SYS_GBL_DATA_SIZE to deal with growth in gd_t

The recent change the env code added an additional 32 bytes into gd_t
and that causes to grow pass the previous CONFIG_SYS_GBL_DATA_SIZE size.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years agoPXA: Remove "xsengine" board
Marek Vasut [Wed, 20 Oct 2010 20:01:12 +0000 (22:01 +0200)]
PXA: Remove "xsengine" board

This board is broken and it's not possible to repair it.

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: Remove "wepep250" board
Marek Vasut [Wed, 20 Oct 2010 19:37:14 +0000 (21:37 +0200)]
PXA: Remove "wepep250" board

This board is broken and impossible to repair without deep knowledge or
availability of the hardware.

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: Remove "delta" board
Marek Vasut [Wed, 20 Oct 2010 18:48:43 +0000 (20:48 +0200)]
PXA: Remove "delta" board

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: zipitz2: Add missing CONFIG_SYS_TEXT_BASE
Marek Vasut [Wed, 20 Oct 2010 20:02:16 +0000 (22:02 +0200)]
PXA: zipitz2: Add missing CONFIG_SYS_TEXT_BASE

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: vpac270: Add missing CONFIG_SYS_TEXT_BASE
Marek Vasut [Wed, 20 Oct 2010 19:30:03 +0000 (21:30 +0200)]
PXA: vpac270: Add missing CONFIG_SYS_TEXT_BASE

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: xm250: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 19:57:34 +0000 (21:57 +0200)]
PXA: xm250: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: xaeniax: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 19:54:19 +0000 (21:54 +0200)]
PXA: xaeniax: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: trizepsiv: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 19:28:14 +0000 (21:28 +0200)]
PXA: trizepsiv: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: pxa255_idp: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 19:20:07 +0000 (21:20 +0200)]
PXA: pxa255_idp: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: pleb2: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 19:10:25 +0000 (21:10 +0200)]
PXA: pleb2: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: palmtc: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 19:06:23 +0000 (21:06 +0200)]
PXA: palmtc: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: palmld: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 19:04:13 +0000 (21:04 +0200)]
PXA: palmld: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: lubbock: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 18:55:44 +0000 (20:55 +0200)]
PXA: lubbock: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: innokom: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 18:52:21 +0000 (20:52 +0200)]
PXA: innokom: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: csb226: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 18:43:41 +0000 (20:43 +0200)]
PXA: csb226: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: cradle: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 18:17:51 +0000 (20:17 +0200)]
PXA: cradle: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: colibri_pxa270: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 18:15:11 +0000 (20:15 +0200)]
PXA: colibri_pxa270: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: cerf250: Fix for reloc
Marek Vasut [Wed, 20 Oct 2010 16:56:41 +0000 (18:56 +0200)]
PXA: cerf250: Fix for reloc

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: Use pxa-regs.h register definitions in start.S
Marek Vasut [Wed, 20 Oct 2010 18:09:09 +0000 (20:09 +0200)]
PXA: Use pxa-regs.h register definitions in start.S

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoPXA: Implement ELF relocation
Marek Vasut [Wed, 20 Oct 2010 17:36:39 +0000 (19:36 +0200)]
PXA: Implement ELF relocation

Based on patch by: Albert Aribaud <albert.aribaud@free.fr>

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
14 years agoMX51: Add video support to vision2 board
Stefano Babic [Thu, 21 Oct 2010 08:34:39 +0000 (10:34 +0200)]
MX51: Add video support to vision2 board

The patch adds support for LCD to the vision2 board.

Signed-off-by: Stefano Babic <sbabic@denx.de>
14 years agoMX51: Add video support
Stefano Babic [Wed, 13 Oct 2010 10:17:14 +0000 (12:17 +0200)]
MX51: Add video support

Add framebuffer driver for the MX51 processor
working on the IPUv3 internal graphic processor.
The port is based on the driver found in the kernel
delivered by Freescale as part of i.MX BSP:

[kernel 2.6.31 commit cc4fe714041805997b601fe8e5dd585d8a99297f]

[agust@denx.de: some style fixes and dead code removal]
Signed-off-by: Stefano Babic <sbabic@denx.de>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
14 years agoMX51: Add IPU driver for video support
Stefano Babic [Wed, 13 Oct 2010 10:16:35 +0000 (12:16 +0200)]
MX51: Add IPU driver for video support

The patch is a porting of the IPU Linux driver
developed by Freescale to have framebuffer
functionalities in u-boot. The port is based on
kernel 2.6.31 commit cc4fe714041805997b601fe8e5dd585d8a99297f,
as delivered by Freescale [i.MX BSP].
Most features are dropped from the original driver and
only LCD support is the goal of this porting.

Signed-off-by: Stefano Babic <sbabic@denx.de>
14 years agoAdd linux framebuffer header include/linux/fb.h
Stefano Babic [Wed, 13 Oct 2010 10:14:29 +0000 (12:14 +0200)]
Add linux framebuffer header include/linux/fb.h

[port of linux 2.6.34 commit 6a9ee8af344e3bd7dbd61e67037096cdf7f83289]

Signed-off-by: Stefano Babic <sbabic@denx.de>
14 years agoled_display: remove unused DISPLAY_MARK define
Ilya Yanok [Thu, 21 Oct 2010 15:20:11 +0000 (17:20 +0200)]
led_display: remove unused DISPLAY_MARK define

DISPLAY_MARK subcommand of display_set() is not used anywhere so
we can remove it safely.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
14 years agoa4m072: change 'display' command implementation
Ilya Yanok [Thu, 21 Oct 2010 15:20:10 +0000 (17:20 +0200)]
a4m072: change 'display' command implementation

This patch changes implementation of 'display' command for a4m072
that uses 7-segment LED display as customer requested:

a) The "display" command shall _not_ turn on the decimal point.

b) Exception: "display ." shall turn on (only) the decimal point.

c) Hex digits (0-9, A-F, a-f) shall be displayed as usual.

d) Letters U, P, Y, L, S, T, H shall be displayed as usual (lower
   case letters identical to upper case letters)

e) 'I' (and 'i') shall be displayed like '1'

f) 'O' (and 'o') shall be displayed like '0'

g) all other Characters shall be displayed like ' ' (all segments off).

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
14 years agoa4m072: changes to the default environment
Ilya Yanok [Thu, 21 Oct 2010 15:20:09 +0000 (17:20 +0200)]
a4m072: changes to the default environment

The default values for 'addip' and 'norargs' changed per customer
request. Everything else cleaned up to fit into 80 symbol line.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
14 years agoFix building for 83xx boards with USB support
Anatolij Gustschin [Thu, 21 Oct 2010 16:30:35 +0000 (18:30 +0200)]
Fix building for 83xx boards with USB support

Commit 29c6fbe0471afd7ffa41fcb2103eec5b53294897 broke
building for 83xx boards with USB support:

ehci-fsl.c: In function 'ehci_hcd_init':
ehci-fsl.c:43: error: 'CONFIG_SYS_FSL_USB_ADDR' undeclared (first use in this function)
ehci-fsl.c:43: error: (Each undeclared identifier is reported only once
ehci-fsl.c:43: error: for each function it appears in.)
make[1]: *** [ehci-fsl.o] Error 1

Signed-off-by: Anatolij Gustschin <agust@denx.de>
14 years agosh: sh4: Move to boards.cfg
Nobuhiro Iwamatsu [Tue, 19 Oct 2010 16:35:28 +0000 (01:35 +0900)]
sh: sh4: Move to boards.cfg

Move the sh4 target boards out of the Makefile and into boards.cfg.
And fix sh4 of MAKEALL.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agosh: sh3: Move to boards.cfg
Nobuhiro Iwamatsu [Tue, 19 Oct 2010 16:28:29 +0000 (01:28 +0900)]
sh: sh3: Move to boards.cfg

Move the sh3 target boards out of the Makefile and into boards.cfg.
And fix sh3 of MAKEALL.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agosh: rsk7203: Move to boards.cfg
Nobuhiro Iwamatsu [Tue, 19 Oct 2010 16:22:25 +0000 (01:22 +0900)]
sh: rsk7203: Move to boards.cfg

Move the rsk7203 target out of the Makefile and into boards.cfg.
And fix sh2 of MAKEALL.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agosh: sh7785lcr: Add CONFIG_SYS_TEXT_BASE for 32bit mode
Nobuhiro Iwamatsu [Tue, 19 Oct 2010 16:17:08 +0000 (01:17 +0900)]
sh: sh7785lcr: Add CONFIG_SYS_TEXT_BASE for 32bit mode

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agosh: Fix warning about uninitialized value of ramdisk_flags
Nobuhiro Iwamatsu [Tue, 19 Oct 2010 08:14:15 +0000 (17:14 +0900)]
sh: Fix warning about uninitialized value of ramdisk_flags

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agosh: Rename TEXT_BASE to CONFIG_SYS_TEXT_BASE
Nobuhiro Iwamatsu [Tue, 19 Oct 2010 08:07:07 +0000 (17:07 +0900)]
sh: Rename TEXT_BASE to CONFIG_SYS_TEXT_BASE

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agosh: sh7785lcr: Add CONFIG_SH_SDRAM_OFFSET
Nobuhiro Iwamatsu [Tue, 5 Oct 2010 07:58:05 +0000 (16:58 +0900)]
sh: sh7785lcr: Add CONFIG_SH_SDRAM_OFFSET

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agosh: Add support load and boot of Initrd.
Nobuhiro Iwamatsu [Tue, 28 Sep 2010 03:14:57 +0000 (12:14 +0900)]
sh: Add support load and boot of Initrd.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
14 years agoppc: Don't initialize write protected NOR flashes
John Schmoller [Wed, 29 Sep 2010 19:05:44 +0000 (14:05 -0500)]
ppc: Don't initialize write protected NOR flashes

If a NOR flash is write protected it can not be initialized/detected so
add the ability for boards to skip NOR initialization on bootup.  A
board can skip NOR initialization by implementing the
board_flash_wp_on() function.

Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
CC: sr@denx.de
Acked-by: Stefan Roese <sr@denx.de>
14 years agoMPC52xx, motionpro: update default configuration
Wolfgang Denk [Wed, 20 Oct 2010 06:31:45 +0000 (08:31 +0200)]
MPC52xx, motionpro: update default configuration

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoftrtc010.c : enhance code according to original datasheet
Macpaul Lin [Tue, 19 Oct 2010 12:14:36 +0000 (20:14 +0800)]
ftrtc010.c : enhance code according to original datasheet

Add missing codes according to original datasheet.
This patch also makes ftrtc010 could be adapted to PCLK and EXT_CLK.

Signed-off-by: Macpaul Lin <macpaul@andestech.com>
14 years agoMerge branch 'master' of git://git.denx.de/u-boot-i2c
Wolfgang Denk [Wed, 20 Oct 2010 19:23:22 +0000 (21:23 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-i2c

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Wed, 20 Oct 2010 19:20:24 +0000 (21:20 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

14 years agoARMV7: OMAP: I2C driver: Restructure i2c_probe function
Steve Sakoman [Wed, 20 Oct 2010 13:07:47 +0000 (06:07 -0700)]
ARMV7: OMAP: I2C driver: Restructure i2c_probe function

This patch removes the "magic number" delays and instead
monitors state changes in the status register bits.

Signed-off-by: Steve Sakoman <steve.sakoman@linaro.org>
Tested-by: Heiko Schocher <hs@denx.de>
14 years agoARMV7: OMAP: I2C driver: Restructure i2c_write_byte function
Steve Sakoman [Wed, 20 Oct 2010 13:07:46 +0000 (06:07 -0700)]
ARMV7: OMAP: I2C driver: Restructure i2c_write_byte function

This patch removes the "magic number" delays and instead
monitors state changes in the status register bits.

Signed-off-by: Steve Sakoman <steve.sakoman@linaro.org>
Tested-by: Heiko Schocher <hs@denx.de>
14 years agoARMV7: OMAP: I2C driver: Restructure i2c_read_byte function
Steve Sakoman [Wed, 20 Oct 2010 13:07:45 +0000 (06:07 -0700)]
ARMV7: OMAP: I2C driver: Restructure i2c_read_byte function

This patch removes the "magic number" delays and instead
monitors state changes in the status register bits.

Signed-off-by: Steve Sakoman <steve.sakomanlinaro.org>
Tested-by: Heiko Schocher <hs@denx.de>
14 years agoARMV7: OMAP: I2C driver: Use same timeout value as linux kernel driver
Steve Sakoman [Wed, 20 Oct 2010 13:07:44 +0000 (06:07 -0700)]
ARMV7: OMAP: I2C driver: Use same timeout value as linux kernel driver

This patch matches the poll interval (1 millisecond) and timeout (1 second)
used in the linux driver. It also adds a return value of 0 in the event of
a timeout error and cleans up some formatting errors in that section of the
code.

Signed-off-by: Steve Sakoman <steve.sakoman@linaro.org>
Tested-by: Heiko Schocher <hs@denx.de>
14 years agoppc4xx: Fix default environment for AMCC boards
Stefan Roese [Fri, 15 Oct 2010 06:41:01 +0000 (08:41 +0200)]
ppc4xx: Fix default environment for AMCC boards

Commit 0ad7f0950a9bc0a69b3cd5f34ccf7da25fcf1c6d [ppc4xx: cleanup
default environment for AMCC boards] broke the default env for
many PPC4xx boards. The '\0' character got removed at the end
of some environment commands like "update". This patch adds the
missing '\0' characters again.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Enable NOR flash support in sequoia_ramboot target
Stefan Roese [Mon, 11 Oct 2010 13:50:10 +0000 (15:50 +0200)]
ppc4xx: Enable NOR flash support in sequoia_ramboot target

Until now, the Sequoia RAM-booting image disabled NOR flash support
as this image was mainly created for NAND-only boards. This patch
now enables NOR flash support for this RAM-booting version as well.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx/POST: Add board specific UART POST test to lwmon5
Stefan Roese [Thu, 7 Oct 2010 12:16:25 +0000 (14:16 +0200)]
ppc4xx/POST: Add board specific UART POST test to lwmon5

Add CONFIG_POST_UART to implement a board specific UART POST test.
This is done since lwmon5 needs to set POST_ALWAYS to run this
test on each reboot. And we don't want to change the default
behavious of this this.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Change tsr/tcr macros to upper case
Stefan Roese [Mon, 4 Oct 2010 09:09:40 +0000 (11:09 +0200)]
ppc4xx: Change tsr/tcr macros to upper case

Remove uneccessary functions to access the TCR/TSR registers as well.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoFix compile warning in uli526x driver
Kumar Gala [Wed, 20 Oct 2010 06:55:39 +0000 (01:55 -0500)]
Fix compile warning in uli526x driver

uli526x.c: In function 'uli526x_init_one':
uli526x.c:314:2: warning: dereferencing type-punned pointer will break strict-aliasing rules
uli526x.c:314:2: warning: dereferencing type-punned pointer will break strict-aliasing rules

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years ago85xx: Use gc-sections to reduce image size
Peter Tyser [Wed, 29 Sep 2010 19:05:56 +0000 (14:05 -0500)]
85xx: Use gc-sections to reduce image size

On an XPedite5370 over 11KBytes were saved:
  Before:
       text    data     bss     dec     hex filename
   332456   33364   33476  399296   617c0 ./u-boot

  After:
      text    data     bss     dec     hex filename
   321075   33836   33476  388387   5ed23 ./u-boot

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
14 years ago86xx: Use gc-sections to reduce image size
Peter Tyser [Wed, 29 Sep 2010 19:05:55 +0000 (14:05 -0500)]
86xx: Use gc-sections to reduce image size

On an XPedite5170 over 11KBytes were saved:
  Before:
     text    data     bss     dec     hex filename
   319488   28700   33204  381392   5d1d0 ./u-boot

  After:
     text    data     bss     dec     hex filename
   307663   29144   33204  370011   5a55b ./u-boot

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>