project/bcm63xx/u-boot.git
13 years agoMIPS: Move Inca-IP targets to boards.cfg
Shinya Kuribayashi [Sat, 5 Feb 2011 09:33:36 +0000 (18:33 +0900)]
MIPS: Move Inca-IP targets to boards.cfg

At the same time, fix up CPU_CLOCK_RATE to have the CONFIG_ prefix to
work with boards.cfg.

Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
13 years agoMIPS: Move Qemu MIPS target to boards.cfg
Shinya Kuribayashi [Sat, 5 Feb 2011 09:25:44 +0000 (18:25 +0900)]
MIPS: Move Qemu MIPS target to boards.cfg

CONFIG_QEMU_MIPS is already provided by <configs/qemu-mips.h>, so we
don't generate it using the options fields in boards.cfg.

Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
13 years agoMIPS: Move Alchemy Au1x00 based boards to boards.cfg
Shinya Kuribayashi [Sat, 5 Feb 2011 08:43:22 +0000 (17:43 +0900)]
MIPS: Move Alchemy Au1x00 based boards to boards.cfg

CONFIG_GTH2 is already provided by <configs/gth2.h>, so we don't
generate it using the options fields in boards.cfg.

Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
13 years agocmd_ide: Fix an unused CONFIG_AU1X00 symbol to work as intended
Shinya Kuribayashi [Sat, 5 Feb 2011 10:07:00 +0000 (19:07 +0900)]
cmd_ide: Fix an unused CONFIG_AU1X00 symbol to work as intended

commit 8bde63eb3f79d68f693201528dafc8ae7aa087de ([MIPS] Rename Alchemy
processor configs into CONFIG_SOC_*) forgot to pick up this one.

Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
13 years agoMIPS: dbau1x00: Remove unused flash driver stub
Daniel Schwierzeck [Thu, 3 Feb 2011 13:17:10 +0000 (14:17 +0100)]
MIPS: dbau1x00: Remove unused flash driver stub

All dbau1x00 boards use the CFI driver so this stub driver is useless
and should not be compiled.

This patch fixes the error:

u-boot-git/board/dbau1x00/flash.c:34: multiple definition of `flash_init'
drivers/mtd/libmtd.o:u-boot-git/drivers/mtd/cfi_flash.c:2084: first defined here
board/dbau1x00/libdbau1x00.o: In function `write_buff':
u-boot-git/board/dbau1x00/flash.c:40: multiple definition of `write_buff'
drivers/mtd/libmtd.o:u-boot-git/drivers/mtd/cfi_flash.c:1265: first defined here

Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@googlemail.com>
Cc: Wolfgang Denk <wd@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
13 years agoMIPS: Purple: Fix multiple definition error on final linking of u-boot binary
Daniel Schwierzeck [Thu, 3 Feb 2011 13:17:09 +0000 (14:17 +0100)]
MIPS: Purple: Fix multiple definition error on final linking of u-boot binary

The linker of recent toolchains complains about multiple definitions
on final linking of u-boot binary. This patch removes all redundant
object files from u-boot.lds those are already added to .text section
by the linker.

That patch could not be tested but the resulting u-boot.map still looks
good. The start symbol is at 0xB0000000, the environment at 0xB0008000
so u-boot should boot.

Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@googlemail.com>
Cc: Wolfgang Denk <wd@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
13 years agoMIPS: VCT: Fix enabling of unwanted options if networking or USB support are disabled
Daniel Schwierzeck [Thu, 3 Feb 2011 13:17:08 +0000 (14:17 +0100)]
MIPS: VCT: Fix enabling of unwanted options if networking or USB support are disabled

Some VCT boards lacks the support of networking or USB.
Additionally that support is disabled in small image
configurations.

If CONFIG_CMD_NET should not used the CONFIG_CMD_NFS option
have to be disabled too. Otherwise the linker fails with
unresolved symbols.

If CONFIG_VCT_SMALL_IMAGE is set than CONFIG_CMD_NET and
CONFIG_CMD_USB are disabled at the end of vct.h.
This is not adequate because CONFIG_CMD_USB enables additional
options and the linker fails again with unresolved symbols.

This patch adds an early check against CONFIG_VCT_SMALL_IMAGE
so the additional options are only enabled if they are really
needed.

Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@googlemail.com>
Acked-by: Stefan Roese <sr@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@pobox.com>
13 years agoMerge branch 'master' of git://git.denx.de/u-boot-nand-flash
Wolfgang Denk [Fri, 4 Feb 2011 20:44:46 +0000 (21:44 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-nand-flash

13 years agoMerge branch 'master' of git://git.denx.de/u-boot-sh
Wolfgang Denk [Fri, 4 Feb 2011 19:38:27 +0000 (20:38 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-sh

13 years agosh: sh7785lcr: Fix out of tree building
Nobuhiro Iwamatsu [Thu, 3 Feb 2011 07:27:45 +0000 (16:27 +0900)]
sh: sh7785lcr: Fix out of tree building

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agoNAND: Fix saving of redundand environment
Alexander Holler [Thu, 20 Jan 2011 01:17:48 +0000 (02:17 +0100)]
NAND: Fix saving of redundand environment

When redundand environments are used the serial needs
to get increased, otherwise the old one will still be used.

Signed-off-by: Alexander Holler <holler@ahsoftware.de>
13 years agoPrepare v2011.03-rc1
Wolfgang Denk [Wed, 2 Feb 2011 21:37:32 +0000 (22:37 +0100)]
Prepare v2011.03-rc1

Signed-off-by: Wolfgang Denk <wd@denx.de>
13 years agoMinor Coding Style Cleanup.
Wolfgang Denk [Wed, 2 Feb 2011 21:36:10 +0000 (22:36 +0100)]
Minor Coding Style Cleanup.

Signed-off-by: Wolfgang Denk <wd@denx.de>
13 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Wed, 2 Feb 2011 20:57:19 +0000 (21:57 +0100)]
Merge branch 'master' of /home/wd/git/u-boot/custodians

13 years agoPrint compiler and linker version with the version command
Alexander Holler [Tue, 18 Jan 2011 23:24:40 +0000 (00:24 +0100)]
Print compiler and linker version with the version command

After years of unsuccessful research I've finally shamelessly stolen other
peoples intellectual properties to present the all-new and world-changing
updated version command:
-
U-Boot>> version

U-Boot 2010.12-00014-g7435056-dirty (Jan 18 2011 - 23:19:38)
MyBoard
gcc (GCC) 0.42 (Distro foobar)
GNU ld (GNU Binutils) 0.314159265
-
May the toolchain bugs rest in peace.

Signed-off-by: Alexander Holler <holler@ahsoftware.de>
13 years agochange email address in MAINTAINERS
Yoshihiro Shimoda [Wed, 2 Feb 2011 01:05:39 +0000 (10:05 +0900)]
change email address in MAINTAINERS

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
13 years agochange email address in MAINTAINERS
Yoshihiro Shimoda [Wed, 2 Feb 2011 01:05:39 +0000 (10:05 +0900)]
change email address in MAINTAINERS

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: add support for sh7757lcr board
Yoshihiro Shimoda [Wed, 2 Feb 2011 01:05:36 +0000 (10:05 +0900)]
sh: add support for sh7757lcr board

The R0P7757LC0030RL board has SH7757, 256MB DDR3-SDRAM, SPI ROM,
Ethernet, and more.

This patch supports the following functions:
 - 256MB DDR3-SDRAM
 - SPI ROM
 - Ethernet

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agospi: add support SuperH SPI module
Yoshihiro Shimoda [Mon, 31 Jan 2011 07:50:43 +0000 (16:50 +0900)]
spi: add support SuperH SPI module

SH7757 has SPI module. This patch supports it.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agonet: sh_eth: add cache handling
Yoshihiro Shimoda [Thu, 27 Jan 2011 01:06:08 +0000 (10:06 +0900)]
net: sh_eth: add cache handling

Some CPU needs cache handling. So this patch add the config of
CONFIG_SH_ETHER_CACHE_WRITEBACK, and it calls wback function.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agoREADME: add description of sh_eth driver
Yoshihiro Shimoda [Thu, 27 Jan 2011 01:06:03 +0000 (10:06 +0900)]
README: add description of sh_eth driver

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: add support the CONFIG_SYS_LDSCRIPT
Yoshihiro Shimoda [Thu, 27 Jan 2011 01:06:14 +0000 (10:06 +0900)]
sh: add support the CONFIG_SYS_LDSCRIPT

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for shmin board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 12:17:22 +0000 (21:17 +0900)]
sh: Remove config.mk for shmin board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for espt board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 12:15:44 +0000 (21:15 +0900)]
sh: Remove config.mk for espt board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for mpr2 board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 12:13:49 +0000 (21:13 +0900)]
sh: Remove config.mk for mpr2 board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for ms7750se board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 12:08:58 +0000 (21:08 +0900)]
sh: Remove config.mk for ms7750se board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for ms7722se board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 12:07:15 +0000 (21:07 +0900)]
sh: Remove config.mk for ms7722se board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for ms7720se board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 12:05:35 +0000 (21:05 +0900)]
sh: Remove config.mk for ms7720se board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for sh7785lcr board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 12:02:16 +0000 (21:02 +0900)]
sh: Remove config.mk for sh7785lcr board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for sh7763rdp board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 11:53:29 +0000 (20:53 +0900)]
sh: Remove config.mk for sh7763rdp board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for rsk7203 board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 11:51:55 +0000 (20:51 +0900)]
sh: Remove config.mk for rsk7203 board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for r7780mp board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 11:50:26 +0000 (20:50 +0900)]
sh: Remove config.mk for r7780mp board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for r2dplus board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 11:48:39 +0000 (20:48 +0900)]
sh: Remove config.mk for r2dplus board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for ap325rxa board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 11:46:35 +0000 (20:46 +0900)]
sh: Remove config.mk for ap325rxa board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agosh: Remove config.mk for MigoR board
Nobuhiro Iwamatsu [Mon, 17 Jan 2011 11:43:40 +0000 (20:43 +0900)]
sh: Remove config.mk for MigoR board

Move CONFIG_SYS_TEXT_BASE to the board's config file, and remove the
unnecessary config.mk file.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agoarmv7: add support for s5pc210 universal board
Minkyu Kang [Mon, 24 Jan 2011 06:33:50 +0000 (15:33 +0900)]
armv7: add support for s5pc210 universal board

This patch adds support for Samsung s5pc210 universal board

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
13 years agoarmv7: add support for S5PC210 SoC
Minkyu Kang [Mon, 24 Jan 2011 06:22:23 +0000 (15:22 +0900)]
armv7: add support for S5PC210 SoC

S5PC210 is a 32-bit RISC and Cortex-A9 Dual Core based micro-processor.

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
13 years agoS5P: serial: Use the inline function instead of static value
Minkyu Kang [Mon, 24 Jan 2011 05:43:25 +0000 (14:43 +0900)]
S5P: serial: Use the inline function instead of static value

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
13 years agoarm: a320evb: fixes for relocation support
Po-Yu Chuang [Sun, 19 Dec 2010 23:07:23 +0000 (23:07 +0000)]
arm: a320evb: fixes for relocation support

* add CONFIG_SYS_SDRAM_BASE and CONFIG_SYS_INIT_SP_ADDR
* do not update gd->bd in dram_init() because bd is unavailable then
* move CONFIG_SYS_TEXT_BASE from config.mk to a320evb.h
* remove config.mk

Signed-off-by: Po-Yu Chuang <ratbert@faraday-tech.com>
13 years agoarm926ejs: timer: Replace bss variable by gdr
Heiko Schocher [Thu, 20 Jan 2011 22:56:39 +0000 (22:56 +0000)]
arm926ejs: timer: Replace bss variable by gdr

Reuse the gd->tbl value for timestamp and add gd->lastinc for lastinc bss
values in the arm926ejs timers implementation.

The usage of bss values in drivers before initialisation of bss is forbidden.
In that special case some data in .rel.dyn gets corrupted.

This patch is similiar to the patch Dirk Behme posted
for the armv7/omap-common/timer.c and added suggestions
from Reinhard Meyer.

Tested on the arm926ejs mx27 based magnesium board
Tested on the arm926ejs kirkwood based suen3 board

Signed-off-by: Heiko Schocher <hs@denx.de>
cc: Albert ARIBAUD <albert.aribaud@free.fr>
cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Stefano Babic <sbabic@denx.de>
cc: Reinhard Meyer <u-boot@emk-elektronik.de>

13 years agoremove (double) LED initialization in arm920t start.s
Jens Scharsig [Sat, 18 Dec 2010 02:08:07 +0000 (02:08 +0000)]
remove (double) LED initialization in arm920t start.s

* remove LED initialization in front of relocation and bss init

Signed-off-by: Jens Scharsig <js_at_ng@scharsoft.de>
13 years agoarm1136: timer: Replace bss variable by gd
Heiko Schocher [Thu, 9 Dec 2010 22:01:15 +0000 (22:01 +0000)]
arm1136: timer: Replace bss variable by gd

Reuse the gd->tbl value for timestamp and add gd->lastinc for lastinc bss
values in the arm1136 timer driver for mx31 and omap24xx

The usage of bss values in drivers before initialisation of bss is forbidden.
In that special case some data in .rel.dyn gets corrupted.

This patch is similiar to the patch Dirk Behme posted
for the armv7/omap-common/timer.c

Tested on the mx31 based qong board

Signed-off-by: Heiko Schocher <hs@denx.de>
cc: Albert ARIBAUD <albert.aribaud@free.fr>
Acked-by: Albert ARIBAUD <albert.aribaud@free.fr>
13 years agoARM: */start.S: code cleanup
Liu Hui-R64343 [Wed, 22 Dec 2010 18:35:04 +0000 (18:35 +0000)]
ARM: */start.S: code cleanup

Remove the useless code from start.S

Signed-off-by: Jason Liu <r64343@freescale.com>
Tested-by: Andreas Bießmann <andreas.devel@googlemail.com>
13 years agoARM: Avoid compiler optimization for readb, writeb and friends.
Alexander Holler [Sun, 9 Jan 2011 12:19:44 +0000 (12:19 +0000)]
ARM: Avoid compiler optimization for readb, writeb and friends.

gcc 4.5.1 seems to ignore (at least some) volatile definitions,
avoid that as done in the kernel.

Reading C99 6.7.3 8 and the comment 114) there, I think it is a bug of that
gcc version to ignore the volatile type qualifier used e.g. in __arch_getl().
Anyway, using a definition as in the kernel headers avoids such optimizations when
gcc 4.5.1 is used.

Maybe the headers as used in the current linux-kernel should be used,
but to avoid large changes, I've just added a small change to the current headers.

Signed-off-by: Alexander Holler <holler@ahsoftware.de>
Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Alessandro Rubini <rubini-list@gnudd.com>
Tested-by: Thomas Weber <weber@corscience.de>
Acked-by: Alexander Holler <holler@ahsoftware.de>
Tested-by: Alexander Holler <holler@ahsoftware.de>
13 years agoarmv7: s5pc1xx: don't use function pointer for clock functions
Minkyu Kang [Mon, 27 Dec 2010 06:55:48 +0000 (15:55 +0900)]
armv7: s5pc1xx: don't use function pointer for clock functions

Because of the bss area is cleared after relocation, we've lost pointers.
This patch fixed it.

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
13 years agoDaVinci: Remove incorrect CONFIG option
Sandeep Paulraj [Wed, 29 Dec 2010 21:09:27 +0000 (16:09 -0500)]
DaVinci: Remove incorrect CONFIG option

The option CONFIG_SOC_DM6447 seems to have ended up
in the code by mistake. It is not used anywhere and
there is no chip called DM6447.

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci Sonata: Fix Build Error
Sandeep Paulraj [Wed, 29 Dec 2010 19:42:56 +0000 (14:42 -0500)]
DaVinci Sonata: Fix Build Error

Fix a build error in the DaVinci Sonata Board

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci DM6467: Fix Build Error
Sandeep Paulraj [Wed, 29 Dec 2010 19:31:26 +0000 (14:31 -0500)]
DaVinci DM6467: Fix Build Error

This commit fixes build errors on the DM6467 port.

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci DM6467: Enhance board Support
Sandeep Paulraj [Tue, 28 Dec 2010 22:38:22 +0000 (17:38 -0500)]
DaVinci DM6467: Enhance board Support

Support for DM6467 was incomplete and the build failed
as well. Patches were sent to the list but have not been
added. This enhances the DM6467 support.
Some more patches will need to be sent to bring
it in line with what is available in internal TI
trees

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoARM: Update mach types
Sandeep Paulraj [Tue, 28 Dec 2010 21:42:57 +0000 (16:42 -0500)]
ARM: Update mach types

This commit updates the mach-types for ARM

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci DM6467: Added ET1011C (LSI) PHY support
Sandeep Paulraj [Tue, 28 Dec 2010 20:43:16 +0000 (15:43 -0500)]
DaVinci DM6467: Added ET1011C (LSI) PHY support

Added arch/arm/cpu/arm926ejs/davinci/et1011c.c for handling
ET1011C gigabit phy. which overrides get_link_speed function
from default implementation. This enables output of 125 MHz
reference clock on SYS_CLK pin.

Signed-off-by: Prakash PM <prakash.pm@ti.com>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci EMAC: Add name to Ethernet device
Sandeep Paulraj [Tue, 28 Dec 2010 19:42:27 +0000 (14:42 -0500)]
DaVinci EMAC: Add name to Ethernet device

Adds "DaVinci-EMAC" as the name of the device so that
it gets printed as "Using DaVinci-EMAC device"
during network access (dhcp, tftp) instead of empty name
in "Using" statement.This name also gets
reflected in 'ethact' env variable.

Signed-off-by: Hemant Pedanekar <hemantp@ti.com>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci EMAC: Fix davinci_eth_gigabit_enable
Sandeep Paulraj [Tue, 28 Dec 2010 19:37:33 +0000 (14:37 -0500)]
DaVinci EMAC: Fix davinci_eth_gigabit_enable

Enabling the gigabit was overwriting the
previous configuration by setting up only GIGAFORCE and
GIG bits of MAC control register.
Modified to retain previous configuration while
gigabit enabling.

Signed-off-by: Prakash PM <prakash.pm@ti.com>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDM365: Fix Build Error
Sandeep Paulraj [Tue, 28 Dec 2010 19:28:51 +0000 (14:28 -0500)]
DM365: Fix Build Error

After the merger of the next branch, the DM365 was
broken. A function used only by DA8xx based SOCs was
being incorrectly called. So fix it.

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci DM365: Adding MMC/SD support for DM365 EVM
Sandeep Paulraj [Sat, 18 Dec 2010 23:15:25 +0000 (18:15 -0500)]
DaVinci DM365: Adding MMC/SD support for DM365 EVM

The patch adds support for MMC/SD in the DM365 EVM

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDaVinci DM355: Adding MMC/SD support for DM355 EVM
Sandeep Paulraj [Sat, 18 Dec 2010 23:14:49 +0000 (18:14 -0500)]
DaVinci DM355: Adding MMC/SD support for DM355 EVM

The patch adds support for MMC/SD in the DM355 EVM

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoDavinci MMCSD Support
Sandeep Paulraj [Tue, 21 Dec 2010 01:01:21 +0000 (20:01 -0500)]
Davinci MMCSD Support

Added support for MMC/SD cards for Davinci.  This feature is enabled by
CONFIG_DAVINCI_MMC and is dependant on CONFIG_MMC and CONFIG_GENERIC_MMC
options. This is tested on DM355 and DM365 EVMs with both the available mmc
controllers.

Signed-off-by: Alagu Sankar <alagusankar@embwise.com>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoOMAP3: add CM-T35 board
Mike Rapoport [Sat, 18 Dec 2010 22:43:19 +0000 (17:43 -0500)]
OMAP3: add CM-T35 board

This patch adds support for CM-T35 board

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
13 years agoARM: fix broken build of ARM
Stefano Babic [Thu, 27 Jan 2011 06:03:49 +0000 (06:03 +0000)]
ARM: fix broken build of ARM

Commit 8aba9dceebb14144e07d19593111ee3a999c37fc breaks
ARM boards because for ARM the -pie option is used
for partial linking together with -r option.

The patch adds the -pie option to link u-boot.bin only.

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Jason Liu <liu.h.jason@gmail.com>
CC: lool@dooz.org
CC: Wolfgang Denk <wd@denx.de>
CC: Albert Aribaud <albert.aribaud@free.fr>
Tested-by: Alexander Holler <holler@ahsoftware.de>
13 years agoMXC: removed warnings from IMX51 ATA driver
Stefano Babic [Tue, 1 Feb 2011 07:13:45 +0000 (08:13 +0100)]
MXC: removed warnings from IMX51 ATA driver

Drop warnings due to unused variables.

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Marek Vasut <marek.vasut@gmail.com>
Acked-by: Marek Vasut <marek.vasut@gmail.com>
13 years agoMX5: Reuse the gd->tbl value for timestamp and add gd->lastinc for lastinc bss
Stefano Babic [Fri, 21 Jan 2011 20:16:15 +0000 (21:16 +0100)]
MX5: Reuse the gd->tbl value for timestamp and add gd->lastinc for lastinc bss

The usage of bss values in drivers before initialisation of bss is forbidden.
In that special case some data in .rel.dyn gets corrupted.

This patch is the same as recently applied for arm926js architecture.

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Heiko Schocher <hs@denx.de>
13 years agoAdd support for Freescale's mx35pdk board.
Stefano Babic [Thu, 20 Jan 2011 08:05:15 +0000 (08:05 +0000)]
Add support for Freescale's mx35pdk board.

The patch adds suupport for the Freescale's mx35pdk board
(known as well as mx35_3stack).

The board boots from the NOR flash. Following devices
are supported:
 - two ethernet devices (FEC and SMC911x on debug board)
 - I2C
 - PMIC (MC13892) via I2C interface
 - UART
 - NOR flash (64MB)
 - NAND flash (2GB)
 - basic access to mc9sdz60 registers via I2C interface

Signed-off-by: Stefano Babic <sbabic@denx.de>
13 years agoSPI: mxc_spi: replace fixed offsets with structures
Stefano Babic [Wed, 19 Jan 2011 22:46:33 +0000 (22:46 +0000)]
SPI: mxc_spi: replace fixed offsets with structures

This patch cleans driver code replacing all accesses
to registers with fixed offsets with a corresponding
structure.

Signed-off-by: Stefano Babic <sbabic@denx.de>
13 years agoSPI: mxc_spi: add SPI clock calculation and setup to the driver
Anatolij Gustschin [Wed, 19 Jan 2011 22:46:32 +0000 (22:46 +0000)]
SPI: mxc_spi: add SPI clock calculation and setup to the driver

The MXC SPI driver didn't calculate the SPI clock up to
now and just used highest possible divider 512 for DATA
RATE in the control register. This results in very low
transfer rates.

The patch adds code to calculate and setup the SPI clock
frequency for transfers.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Stefano Babic <sbabic@denx.de>
13 years agoSPI: mxc_spi: fix swapping bug and add missing swapping in unaligned rx case
Anatolij Gustschin [Thu, 20 Jan 2011 07:53:06 +0000 (07:53 +0000)]
SPI: mxc_spi: fix swapping bug and add missing swapping in unaligned rx case

We need to shift only one time in each cycle in the swapping loop
for unaligned tx case. Currently two byte shift operations are
performed in each loop cycle causing zero gaps in the transmited
data, so not all data scheduled for transmition is actually
transmited.

The proper swapping in unaligned rx case is missing, so add it
as we need to put the received data into the rx buffer in the
correct byte order.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Stefano Babic <sbabic@denx.de>
13 years agoSPI: mxc_spi: add support for i.MX35 processor
Stefano Babic [Wed, 19 Jan 2011 22:46:30 +0000 (22:46 +0000)]
SPI: mxc_spi: add support for i.MX35 processor

Signed-off-by: Stefano Babic <sbabic@denx.de>
13 years agoAdd basic support for Freescale's mc9sdz60
Stefano Babic [Wed, 19 Jan 2011 22:46:29 +0000 (22:46 +0000)]
Add basic support for Freescale's mc9sdz60

The patch adds helper funtions for basic access to the registers
of the MC9sdz60 chip (multifunctional device with RTC and CAN) via
I2C interface.

Signed-off-by: Stefano Babic <sbabic@denx.de>
13 years agoI2C: mxc_i2c: address failure with mx35 processor
Stefano Babic [Thu, 20 Jan 2011 07:51:31 +0000 (07:51 +0000)]
I2C: mxc_i2c: address failure with mx35 processor

There is sporadic failures when more as one I2C slave
is on the bus and the processor tries to communicate
with more as one slave.
The problem was seen on a mx35pdk (two I2C slaves,
PMIC controller and CAN/RTC chip).

The current driver uses the IIF bit in the status register
to check if the bus is busy or not. According to the manual,
this is not correct, because the IIB bit should be checked.
Not only, to check if a transfer is finished must be checked
the ICF bit, and this is not tested at all.

This patch comes from analyse with a corresponding driver
provided by Freescale as part of the LTIB tool. Comparing
the two drivers, it appears that the current u-boot driver checks
the wrong bits, and depending on race condition, the transfer
can be successful or not.

The patch gets rid also of own debug function (DPRINTF),
replaced with the general debug().

Tested on Freescale mx35pdk.

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
13 years agoI2C: mxc_i2c: get rid of __REG access
Stefano Babic [Thu, 20 Jan 2011 07:50:44 +0000 (07:50 +0000)]
I2C: mxc_i2c: get rid of __REG access

This driver accesses to processor's register
via __REG macros, that are removed (or are planned
to be removed) and replaced by C structures.
This patches replaces all occurrencies of __REG macros.

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
13 years agomxc_i2c: Add support for the i.MX35 processor
Stefano Babic [Wed, 19 Jan 2011 22:46:26 +0000 (22:46 +0000)]
mxc_i2c: Add support for the i.MX35 processor

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
13 years agoserial_mxc: add support for Freescale's i.MX35 processor
Stefano Babic [Wed, 19 Jan 2011 22:46:25 +0000 (22:46 +0000)]
serial_mxc: add support for Freescale's i.MX35 processor

The patch adds UART support for the i.MX35 processor.

Signed-off-by: Stefano Babic <sbabic@denx.de>
13 years agoAdd support for MX35 processor
Stefano Babic [Thu, 20 Jan 2011 07:49:52 +0000 (07:49 +0000)]
Add support for MX35 processor

The patch adds basic support for the Freescale's i.MX35
(arm1136 based) processor.

The patch adds also a prototype for the initialization
of the FEC(ethernet controller) to netdev.h to avoid
warnings.

Signed-off-by: Stefano Babic <sbabic@denx.de>
13 years agoiMX5: EfikaMX: Preliminary board support
Marek Vasut [Wed, 19 Jan 2011 04:40:37 +0000 (04:40 +0000)]
iMX5: EfikaMX: Preliminary board support

Supported:
MMC
IDE
PMIC
SPI flash
LEDs

I can boot the kernel supplied by freescale/genesi with this from MMC card
and/or PATA disk.

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
13 years agoimximage: Add MX53 boot image support
Liu Hui-R64343 [Wed, 19 Jan 2011 09:40:26 +0000 (09:40 +0000)]
imximage: Add MX53 boot image support

This patch add the MX53 boot image support.

This patch has been tested on Freescale MX53EVK board
and MX51EVK board.

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agoMX51EVK: Use SWx macros in PMIC init
Marek Vasut [Wed, 19 Jan 2011 04:40:36 +0000 (04:40 +0000)]
MX51EVK: Use SWx macros in PMIC init

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
13 years agoMC13892: Add SWx buck switchers definitions
Marek Vasut [Wed, 19 Jan 2011 04:40:35 +0000 (04:40 +0000)]
MC13892: Add SWx buck switchers definitions

Define voltages configurable on SWx buck switchers.

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Acked-by: Stefano Babic <sbabic@denx.de>
13 years agomxc_nand: add support for i.MX35 processor
Stefano Babic [Fri, 14 Jan 2011 03:35:21 +0000 (03:35 +0000)]
mxc_nand: add support for i.MX35 processor

Signed-off-by: Stefano Babic <sbabic@denx.de>
Acked-by: Scott Wood <scottwood@freescale.com>
13 years agoBLOCK: Add freescale IMX51 PATA driver
Marek Vasut [Thu, 13 Jan 2011 07:27:55 +0000 (07:27 +0000)]
BLOCK: Add freescale IMX51 PATA driver

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Acked-by: Stefano Babic <sbabic@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
13 years agoMX5:MX53: add initial support for MX53EVK board
Liu Hui-R64343 [Mon, 3 Jan 2011 22:27:42 +0000 (22:27 +0000)]
MX5:MX53: add initial support for MX53EVK board

Add initial support for MX53EVK board support.
FEC, SD/MMC, UART, I2C, have been supported.

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agofsl_pmic: add I2C interface support
Liu Hui-R64343 [Mon, 3 Jan 2011 22:27:40 +0000 (22:27 +0000)]
fsl_pmic: add I2C interface support

This patch add I2C interface for fsl_pmic driver support

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agomxc_i2c: add support for MX53 processor
Liu Hui-R64343 [Mon, 3 Jan 2011 22:27:39 +0000 (22:27 +0000)]
mxc_i2c: add support for MX53 processor

This patch add I2C support for Freescale MX53 processor

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agomxc_gpio: add support for MX53 processor
Liu Hui-R64343 [Mon, 3 Jan 2011 22:27:38 +0000 (22:27 +0000)]
mxc_gpio: add support for MX53 processor

This patch add mxc_gpio support for Freescale MX53 processor

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agoserial_mxc: add support for MX53 processor
Liu Hui-R64343 [Mon, 3 Jan 2011 22:27:37 +0000 (22:27 +0000)]
serial_mxc: add support for MX53 processor

This patch add UART support for Freescale MX53 processor

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agofec_mxc: add support for MX53 processor
Liu Hui-R64343 [Mon, 3 Jan 2011 22:27:36 +0000 (22:27 +0000)]
fec_mxc: add support for MX53 processor

This patch add FEC support for Freescale MX53 processor

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agoMX5: Add initial support for MX53 processor
Liu Hui-R64343 [Mon, 3 Jan 2011 22:27:35 +0000 (22:27 +0000)]
MX5: Add initial support for MX53 processor

Add initial support for Freescale MX53 processor,

- Add the iomux support and the pin definition,
- Add the regs definition, clean up some unused def from mx51,
- Add the low level init support, make use the freq input of setup_pll macro

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agoMX51EVK: UART does not print out the early information
Liu Hui-R64343 [Thu, 23 Dec 2010 01:13:17 +0000 (01:13 +0000)]
MX51EVK: UART does not print out the early information

The early bootup information is not print out due to
the UART pin iomux not set up correctly before board_init

Add the board_early_init_f function and enable the
CONFIG_BOARD_EARLY_INIT_F. Move the UART pin setting
from board_init to board_early_init_f function.

This patch also move the FEC pin iomux setup to the
board_early_init_f.

Signed-off-by: Jason Liu <r64343@freescale.com>
13 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Mon, 31 Jan 2011 22:20:32 +0000 (23:20 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

13 years agoMerge branch 'master' of git://git.denx.de/u-boot-i2c
Wolfgang Denk [Mon, 31 Jan 2011 22:17:33 +0000 (23:17 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-i2c

13 years agoFix at91 includes in soft_i2c driver
Ryan Mallon [Wed, 26 Jan 2011 19:54:15 +0000 (08:54 +1300)]
Fix at91 includes in soft_i2c driver

Make at91 header includes in soft_i2c depend only on CONFIG_AT91FAMILY
rather than individual SoCs.

Signed-off-by: Ryan Mallon <ryan@bluewatersys.com>
Acked-by: Reinhard Meyer<u-boot@emk-elektronik.de>
13 years agompq101: initial support for Mercury Computer Systems MPQ101 board
Alex Dubov [Mon, 24 Jan 2011 05:59:10 +0000 (21:59 -0800)]
mpq101: initial support for Mercury Computer Systems MPQ101 board

Mpq101 is a RapidIO development board in AMC form factor, featuring MPC8548
processor, 512MB of hardwired DDR2 RAM, 128MB of hardwired NAND flash
memory, real time clock and additional serial EEPROM on i2c bus (enabled).
USB controller is available, but not presently enabled.

Additional board information is available at:
http://www.mc.com/products/boards/ensemble_mpq101_rapidio_powerquicc_iii.aspx

Environment is configured to precede the actual u-boot image so that it's
located at the beginning of flash erase block (made necessary by the recent
changes to the embedded environment handling). This is achieved by means of
custom ld script.

Signed-off-by: Alex Dubov <oakad@yahoo.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
13 years agop1022ds: fix pixis_reset altbank
York Sun [Wed, 26 Jan 2011 18:30:00 +0000 (10:30 -0800)]
p1022ds: fix pixis_reset altbank

Fix the bits for ngpixis to reset to alternative bank. Originally the mask
was 0xE0, which left it possible to reset to bank 3 if DIP switch is set to
boot from bank 1. Changing to 0xF0 gurantees to reset to bank 2.

Signed-off-by: York Sun <yorksun@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
13 years agolcd: align fb writing address for horizontal display offset
Liu Ying [Tue, 11 Jan 2011 07:29:58 +0000 (15:29 +0800)]
lcd: align fb writing address for horizontal display offset

CONFIG_SPLASH_SCREEN_ALIGN makes uboot support display
offset for splashimage. The framebuffer writing address
should be calculated according to different kinds of
framebuffer pixel format, i.e., bits per pixel value.

Signed-off-by: Liu Ying <Ying.Liu@freescale.com>
13 years agoDivides variable of linker flags to LDFLAGS-u-boot and LDFLAGS
Nobuhiro Iwamatsu [Thu, 6 Jan 2011 01:23:54 +0000 (10:23 +0900)]
Divides variable of linker flags to LDFLAGS-u-boot and LDFLAGS

Linker needs to use the proper endian/bfd flags even when doing partial linking.
LDFLAGS_u-boot sets linker option which is called it when U-boot is built
(u-boot final).
LDFLAGS sets necessary option by partial linking (use in cmd_link_o_target).

CC: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
13 years agoftpmu010: support faraday ftpmu010 driver
Macpaul Lin [Wed, 5 Jan 2011 09:12:23 +0000 (17:12 +0800)]
ftpmu010: support faraday ftpmu010 driver

Faraday's ftpmu010 is a power managemnet unit which support cpu
sleep and frequency scaling. It has been integrated into many SoC.

This patch also move ftpmu010 to a proper place for later enhancement.

Signed-off-by: Macpaul Lin <macpaul@andestech.com>
13 years agopowerpc: Fix FPU post related link warnings
Kumar Gala [Tue, 25 Jan 2011 09:00:08 +0000 (03:00 -0600)]
powerpc: Fix FPU post related link warnings

If we built POST on PPC's that didn't enable CONFIG_SYS_POST_FPU we'd
get the following warning with newer toolchains:

powerpc-linux-gnu-ld: Warning: lib_powerpc/fpu/libpostpowerpcfpu.o
      uses hard float, libpost.o uses soft float

We actually worked around this sometime ago with the following commit:

commit ce82ff05388b5ddafdf6082ef0776cce72c40b1c
Author: Yuri Tikhonov <yur@emcraft.com>
Date:   Sat Dec 20 14:54:21 2008 +0300

   FPU POST: fix warnings when building with 2.18 binutils

However, this only took into effect if CONFIG_SYS_POST_FPU was enabled.
We can simply move the GNU_FPOST_ATTR out of the CONFIG_SYS_POST_FPU
ifdef block to address the issue.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
13 years agoUEC: Fix compiler warnings introduced by linux/mii.h change
Kumar Gala [Wed, 19 Jan 2011 09:36:40 +0000 (03:36 -0600)]
UEC: Fix compiler warnings introduced by linux/mii.h change

Patch 8ef583a0 [miiphy: convert to linux/mii.h] introduced the following
compiler warnings in the uec ethernet driver:

In file included from /local/home/galak/git/u-boot-85xx/include/miiphy.h:37:0,
                 from uec.c:32:
/local/home/galak/git/u-boot-85xx/include/linux/mii.h:133:0: warning: "LPA_1000FULL" redefined
uec_phy.h:34:0: note: this is the location of the previous definition
/local/home/galak/git/u-boot-85xx/include/linux/mii.h:134:0: warning: "LPA_1000HALF" redefined
uec_phy.h:35:0: note: this is the location of the previous definition
In file included from /local/home/galak/git/u-boot-85xx/include/miiphy.h:37:0,
                 from uec_phy.c:27:
/local/home/galak/git/u-boot-85xx/include/linux/mii.h:133:0: warning: "LPA_1000FULL" redefined
uec_phy.h:34:0: note: this is the location of the previous definition
/local/home/galak/git/u-boot-85xx/include/linux/mii.h:134:0: warning: "LPA_1000HALF" redefined
uec_phy.h:35:0: note: this is the location of the previous definition

Fix them be removing the duplication in the uec code and utlizing the
linux/mii.h version instead.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
13 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Tue, 25 Jan 2011 20:13:04 +0000 (21:13 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

13 years agopowerpc/8xxx: Fix compile warning when build for a DDR1 or DDR2 board
Kumar Gala [Thu, 20 Jan 2011 07:53:15 +0000 (01:53 -0600)]
powerpc/8xxx: Fix compile warning when build for a DDR1 or DDR2 board

ctrl_regs.c: In function 'set_ddr_sdram_mode_2':
ctrl_regs.c:690:6: warning: unused variable 'i'

'i' is only used by DDR3 code.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
13 years agoCleanup .boards.depend when using an objtree
Loïc Minier [Wed, 19 Jan 2011 12:16:30 +0000 (13:16 +0100)]
Cleanup .boards.depend when using an objtree

.boards.depend was created in the source tree even when calling make
with O=objtree, and distclean O=objtree wouldn't clean it.  Create
.boards.depend in objtree instead as to clean it up properly.

Reported-by: Loïc Minier <loic.minier@linaro.org>
Signed-off-by: Wolfgang Denk <wd@denx.de>
13 years agoDon't add symlink in srctree when using an objtree
Loïc Minier [Wed, 19 Jan 2011 12:16:29 +0000 (13:16 +0100)]
Don't add symlink in srctree when using an objtree

When building with srctree != objtree, the build creates arch/soc/cpu
specific symlinks in the source tree.  This means that the same source
tree can't be used for multiple builds at the same time.  Also, these
symlinks in the source tree are only cleaned up if one passes the same
O= to distclean.

When srctree != objtree, mkconfig creates an $objtree/include2 directory
in the objtree to host the asm -> arch/$arch/include/asm symlink so that
"#include <asm>" can be used.  But it also creates another identical
symlink in $objtree/include.

Then, mkconfig creates two symlinks:
$objtree/include/asm/arch -> arch/$arch/include/asm/arch-$cpu (or $soc)
$objtree/include/asm/proc -> arch/$arch/include/asm/proc-armv (on arm)
but because $objtree/include/asm points at $srctree already, the two
symlinks are created under $srctree.

To fix this, create a real $objtree/include/asm directory, instead of a
symlink.  Update cleanup code accordingly.

Signed-off-by: Loïc Minier <loic.minier@linaro.org>