oweals/u-boot.git
15 years agoenv_onenand: change env_address type from unsigned long to loff_t
HeungJun Kim [Tue, 30 Jun 2009 05:42:22 +0000 (14:42 +0900)]
env_onenand: change env_address type from unsigned long to loff_t

If use the onenand boot, the env_relocate_spec() calls mtd->read(),
and the type of the argument #2 of mtd->read() was changed to loff_t.
But, the "env_addr" type is still unsigned long, thus this patch change
the type from unsigned long to loff_t.

Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: HeungJun, Kim <riverful.kim@samsung.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agofsl_elbc_nand: redirect the pointer of bbt pattern to RAM
Mingkai Hu [Thu, 18 Jun 2009 10:23:27 +0000 (18:23 +0800)]
fsl_elbc_nand: redirect the pointer of bbt pattern to RAM

The bbt descriptors contains the pointer to the bbt pattern which
are statically initialized memory struct. When relocated to RAM,
these pointers will continue point to NOR flash(or L2 SRAM, or
other boot device). If the contents of NOR flash changed or L2
SRAM disabled, it'll hang the system.

Signed-off-by: Mingkai Hu <Mingkai.hu@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agonand_spl: Fix cmd_ctrl usage in nand_boot.c.
Scott Wood [Wed, 24 Jun 2009 22:23:49 +0000 (17:23 -0500)]
nand_spl: Fix cmd_ctrl usage in nand_boot.c.

When adding large page NAND support to this file, I had a misunderstanding
about the exact semantics of NAND_CTRL_CHANGE (which isn't documented
anywhere I can find) -- it is apparently just a hint to drivers,
which aren't required to preserve the old value for subsequent
non-"change" invocations.

This change makes nand_boot.c no longer assume this.  Note that this
happened to work by chance with some NAND drivers, which don't preserve
the value, but treat 0 equivalently to NAND_CTRL_ALE.

I don't have hardware to test this, so any testing is appreciated.

Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoBug-fix in drivers mtd nand Makefile
kevin.morfitt@fearnside-systems.co.uk [Thu, 18 Jun 2009 17:41:03 +0000 (18:41 +0100)]
Bug-fix in drivers mtd nand Makefile

The S3C2410 NAND driver source file is included in the makefile instead of
the object file.

Signed-off-by: Kevin Morfitt <kevin.morfitt@fearnside-systems.co.uk>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agonand_spl: read environment early, when booting from NAND using nand_spl
Guennadi Liakhovetski [Mon, 18 May 2009 14:07:22 +0000 (16:07 +0200)]
nand_spl: read environment early, when booting from NAND using nand_spl

Currently, when booting from NAND using nand_spl, in the beginning the default
environment is used until later in boot process the dynamic environment is read
out. This way environment variables that must be interpreted early, like the
baudrate or "silent", cannot be modified dynamically and remain at their
default values. Fix this problem by reading out main and redundand (if used)
copies of the environment in the nand_spl code.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agomtd: nand: use loff_t for offset
Jean-Christophe PLAGNIOL-VILLARD [Sat, 16 May 2009 12:27:40 +0000 (14:27 +0200)]
mtd: nand: use loff_t for offset

nand_util currently uses size_t which is arch dependent and not always a
unsigned long.  Now use loff_t, as does the linux mtd layer.

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agonand/onenand: Fix missing argument checking for "markbad" command
Wolfgang Denk [Sun, 24 May 2009 15:34:33 +0000 (17:34 +0200)]
nand/onenand: Fix missing argument checking for "markbad" command

The "nand markbad" and "onenand markbad" commands did not check if an
argument was passed; if this was forgotten, no error was raised but
block 0 was marked as bad.

While fixing this bug, clean up the code a bit and allow to pass more
than one block address, thus allowing to mark several blocks as bad
in a single command invocation.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agomtd: nand: new base driver for memory mapped nand devices
Mike Frysinger [Tue, 26 May 2009 02:42:28 +0000 (22:42 -0400)]
mtd: nand: new base driver for memory mapped nand devices

The BF537-STAMP Blackfin board had a driver for working with NAND devices
that are simply memory mapped.  Since there is nothing Blackfin specific
about this, generalize the driver a bit so that everyone can leverage it.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoenv_nand: remove unused variable.
Guennadi Liakhovetski [Mon, 18 May 2009 14:06:45 +0000 (16:06 +0200)]
env_nand: remove unused variable.

Remove an unused "total" variable in multiple functions.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agodavinci_nand chipselect/init cleanup
David Brownell [Sun, 10 May 2009 22:43:01 +0000 (15:43 -0700)]
davinci_nand chipselect/init cleanup

Update chipselect handling in davinci_nand.c so that it can
handle 2 GByte chips the same way Linux does:  as one device,
even though it has two halves with independent chip selects.
For such chips the "nand info" command reports:

  Device 0: 2x nand0, sector size 128 KiB

Switch to use the default chipselect function unless the board
really needs its own.  The logic for the Sonata board moves out
of the driver into board-specific code.  (Which doesn't affect
current build breakage if its NAND support is enabled...)

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoNAND DaVinci: Update to ALE/CLE Mask values
Sandeep Paulraj [Sat, 9 May 2009 16:35:20 +0000 (12:35 -0400)]
NAND DaVinci: Update to ALE/CLE Mask values

All DaVinci SOC's use a CLE mask of 0x10 and an ALE mask of 0x8
except the DM646x. This was decided by the design team driving the design.
This patch updates the CLE and ALE values for DM646x.
Updated patches for DM646x will be sent shortly.
This applies to u-boot-nand-flash git

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoARM DaVinci: Changing ALE Mask Value
Sandeep Paulraj [Wed, 29 Apr 2009 13:47:09 +0000 (09:47 -0400)]
ARM DaVinci: Changing ALE Mask Value

The ALE mask used by DaVinci SOCs is wrong. The patch changes the mask value
from '0xa' to '0x8'. This is the mask we use for all TI releases.

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agodavinci_nand: cleanup II (CONFIG_SYS_DAVINCI_BROKEN_ECC)
David Brownell [Tue, 28 Apr 2009 20:19:53 +0000 (13:19 -0700)]
davinci_nand: cleanup II (CONFIG_SYS_DAVINCI_BROKEN_ECC)

Remove CONFIG_SYS_DAVINCI_BROKEN_ECC option.  It's not just nasty;
it's also unused by any current boards, and doesn't even match the
main U-Boot distributions from TI (which use soft ECC, or 4-bit ECC
on newer chips that support it).

DaVinci GIT kernels since 2.6.24, and mainline Linux since 2.6.30,
match non-BROKEN code paths for 1-bit HW ECC.  The BROKEN code paths
do seem to partially match what MontaVista/TI kernels (4.0/2.6.10,
and 5.0/2.6.18) do ... but only for small pages.  Large page support
is really broken (and it's unclear just what software it was trying
to match!), and the ECC layout was making three more bytes available
for use by filesystem (or whatever) code.

Since this option itself seems broken, remove it.  Add a comment
about the MV/TI compat issue, and the most straightforward way to
address it (should someone really need to solve it).

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agodavinci_nand: cleanup I (minor)
David Brownell [Tue, 28 Apr 2009 20:19:50 +0000 (13:19 -0700)]
davinci_nand: cleanup I (minor)

Minor cleanup for DaVinci NAND code:

 - Use I/O addresses from nand_chip; CONFIG_SYS_NAND_BASE won't
   be defined when there are multiple chipselect lines in use
   (as with common 2 GByte chips).

 - Cleanup handling of EMIF control registers
    * Only need one pointer pointing to them
    * Remove incorrect and unused struct supersetting them

 - Use the standard waitfunc; we don't need a custom version

 - Partial legacy cleanup:
    * Don't initialize every board like it's a DM6446 EVM
    * #ifdef a bit more code for BROKEN_ECC

Sanity checked with small page NAND on dm355 and dm6446 EVMs;
and large page on dm355 EVM (packaged as two devices, not one).

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-video
Wolfgang Denk [Tue, 7 Jul 2009 21:18:03 +0000 (23:18 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-video

15 years agoCoding style cleanup; update CHANGELOG
Wolfgang Denk [Tue, 7 Jul 2009 20:35:02 +0000 (22:35 +0200)]
Coding style cleanup; update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-arm
Wolfgang Denk [Tue, 7 Jul 2009 20:22:05 +0000 (22:22 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-arm

Conflicts:
drivers/spi/Makefile

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc5xxx
Wolfgang Denk [Tue, 7 Jul 2009 19:06:07 +0000 (21:06 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc5xxx

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-ubi
Wolfgang Denk [Tue, 7 Jul 2009 18:58:30 +0000 (20:58 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-ubi

15 years agoMerge branch 'sf' of git://git.denx.de/u-boot-blackfin
Wolfgang Denk [Tue, 7 Jul 2009 18:56:56 +0000 (20:56 +0200)]
Merge branch 'sf' of git://git.denx.de/u-boot-blackfin

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Tue, 7 Jul 2009 18:54:12 +0000 (20:54 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

15 years agoUBI: Fix build problem noticed on Apollon (arm/testing repo)
Stefan Roese [Mon, 29 Jun 2009 11:30:50 +0000 (13:30 +0200)]
UBI: Fix build problem noticed on Apollon (arm/testing repo)

This patch fixes a build problem noticed on Apollon by using
mtd_dev_by_eb() instead of "/" as done in the Linux UBI version.
So this brings the U-Boot UBI version more in sync with the Linux
version again.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agosf: Macronix additional chips supported
Prafulla Wadaskar [Mon, 6 Jul 2009 14:59:15 +0000 (20:29 +0530)]
sf: Macronix additional chips supported

new chips supported:-
MX25L1605D, MX25L3205D, MX25L6405D, MX25L12855E
out of which MX25L6405D and MX25L12855E tested on Kirkwood platforms

Modified the Macronix flash support to use 2 bytes of device id instead of 1
This was required to support MX25L12855E

Signed-off-by: Piyush Shah <spiyush@marvell.com>
Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agosf: sst: add sst25vf###b ids
Mike Frysinger [Fri, 19 Jun 2009 07:27:28 +0000 (03:27 -0400)]
sf: sst: add sst25vf###b ids

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agosf: sst: fix sector size
Mike Frysinger [Fri, 19 Jun 2009 07:20:06 +0000 (03:20 -0400)]
sf: sst: fix sector size

Looks like when I was encoding the sector sizes, I forgot to divide by 8
(due to the stupid marketing driven process that declares all sizes in
useless megabits and not megabytes).

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agonhk8815: fix MAKEALL
Jean-Christophe PLAGNIOL-VILLARD [Sat, 4 Jul 2009 23:06:06 +0000 (01:06 +0200)]
nhk8815: fix MAKEALL

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoMX31: Add NAND SPL boot support to i.MX31 PDK board.
Magnus Lilja [Sat, 4 Jul 2009 08:31:24 +0000 (10:31 +0200)]
MX31: Add NAND SPL boot support to i.MX31 PDK board.

Signed-off-by: Magnus Lilja <lilja.magnus@gmail.com>
15 years agoarm: Kirkwood: arch specific updated for ehci-Kirkwood driver support
Prafulla Wadaskar [Mon, 29 Jun 2009 15:25:54 +0000 (20:55 +0530)]
arm: Kirkwood: arch specific updated for ehci-Kirkwood driver support

This patch abstracts Kirkwood arch specific changes to support ehci-kirkwood driver

Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
15 years agoarm nomadik: use 1000 as HZ value and rewrite timer code
Alessandro Rubini [Mon, 29 Jun 2009 08:52:37 +0000 (10:52 +0200)]
arm nomadik: use 1000 as HZ value and rewrite timer code

This sets CONFIG_SYS_HZ to 1000 as required, and completely rewrites
timer code, which is now both correct and much smaller.  Unused
functions like udelay_masked() have been removed as no driver uses
them, even the ones that are not currently active for this board.
mtu.h is copied literally from the kernel sources.

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stericsson.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoarm nomadik: allow Nand and OneNand to coexists
Alessandro Rubini [Mon, 22 Jun 2009 07:18:57 +0000 (09:18 +0200)]
arm nomadik: allow Nand and OneNand to coexists

The evaluation kit has both Nand and OneNand, both drivers are there
and the two configurations only select a different default for the
jffs partition. This adds the OneNand driver and cleans up storage.

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stericsson.com>
15 years agoarm nomadik: cleanup reset
Alessandro Rubini [Mon, 22 Jun 2009 07:18:47 +0000 (09:18 +0200)]
arm nomadik: cleanup reset

There is only one public release of the Nomadik chip, so the ifdef
in reset code as well as a define in the config file are not needed

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stericsson.com>
15 years agoarm nomadik: rename board to nhk8815
Alessandro Rubini [Mon, 22 Jun 2009 07:18:37 +0000 (09:18 +0200)]
arm nomadik: rename board to nhk8815

This is an error in my side in the initial submission: nobody
calls it ""nmdk8815", it's "nomadik hardware kit", nhk8815, instead.

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stericsson.com>
15 years agoxscale: add support for the polaris board
Stefano Babic [Wed, 1 Jul 2009 18:40:41 +0000 (20:40 +0200)]
xscale: add support for the polaris board

The Polaris board is based on the TrizepsIV module of
Keith & Koep (http://www.keith-koep.com).

Signed-off-by: Stefano Babic <sbabic@denx.de>
15 years agoxscale: fix USB initialization for Trizepsiv module
Stefano Babic [Wed, 1 Jul 2009 02:33:56 +0000 (04:33 +0200)]
xscale: fix USB initialization for Trizepsiv module

Due to change in the usb_board_init() prototype, the USB for
the TrizepsIV was not correctly initialized.
Removed dummy print from usb_board_stop().

Signed-off-by: Stefano Babic <sbabic@denx.de>
15 years agonet: merge bugfix: Marvell Kirkwood gigabit ethernet driver
Prafulla Wadaskar [Wed, 1 Jul 2009 18:34:51 +0000 (20:34 +0200)]
net: merge bugfix: Marvell Kirkwood gigabit ethernet driver

This patch looks okay on u-boot-net.git/next branch
but when it was merged to u-boot.git/master the last line is missing

Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
Acked-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoat91: Add esd gmbh MEESC board support
Daniel Gorsulowski [Tue, 30 Jun 2009 19:03:37 +0000 (21:03 +0200)]
at91: Add esd gmbh MEESC board support

This patch adds support for esd gmbh MEESC board.
The MEESC is based on an Atmel AT91SAM9263 SoC.

Signed-off-by: Daniel Gorsulowski <Daniel.Gorsulowski@esd.eu>
15 years agoARM: Update mach-types
Jean-Christophe PLAGNIOL-VILLARD [Tue, 30 Jun 2009 19:03:35 +0000 (21:03 +0200)]
ARM: Update mach-types

update against linux v2.6.30

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91: Add CAN init function
Daniel Gorsulowski [Tue, 30 Jun 2009 21:03:33 +0000 (23:03 +0200)]
at91: Add CAN init function

To enable CAN init, CONFIG_CAN has to be defined in the board config file
and at91_can_hw_init() has to be called in the board specific code.

CAN is available on AT91SAM9263 and AT91CAP9 SoC.

Signed-off-by: Daniel Gorsulowski <Daniel.Gorsulowski@esd.eu>
15 years agoarm: Kirkwood: Correct header define
Simon Kagstrom [Tue, 30 Jun 2009 21:03:31 +0000 (23:03 +0200)]
arm: Kirkwood: Correct header define

Correct define typo (. -> ,)

Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
15 years agoMX31: Add basic support for Freescale i.MX31 PDK board.
Magnus Lilja [Tue, 30 Jun 2009 23:07:55 +0000 (01:07 +0200)]
MX31: Add basic support for Freescale i.MX31 PDK board.

Add support for Freescale's i.MX31 PDK board (a.k.a. 3 stack board).

This patch assumes that some other program performs the actual
NAND boot.

Signed-off-by: Magnus Lilja <lilja.magnus@gmail.com>
Acked-by: Fabio Estevam <fabioestevam@yahoo.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoS3C24x0: extract interrupts from timer
Jean-Christophe PLAGNIOL-VILLARD [Mon, 22 Jun 2009 22:12:01 +0000 (00:12 +0200)]
S3C24x0: extract interrupts from timer

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agodm355/pm9261: add missing CONFIG_NET_MULTI
Jean-Christophe PLAGNIOL-VILLARD [Sun, 28 Jun 2009 12:14:21 +0000 (14:14 +0200)]
dm355/pm9261: add missing CONFIG_NET_MULTI

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoarm920t/interrupts: Move conditional compilation to Makefile
Jean-Christophe PLAGNIOL-VILLARD [Mon, 22 Jun 2009 22:12:01 +0000 (00:12 +0200)]
arm920t/interrupts: Move conditional compilation to Makefile

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agomove L2 cache enable/disable function to cache.c in the omap3 SoC directory
Kim, Heung Jun [Sat, 20 Jun 2009 09:02:17 +0000 (11:02 +0200)]
move L2 cache enable/disable function to cache.c in the omap3 SoC directory

Signed-off-by: HeungJun, Kim <riverful.kim@samsung.com>
CC: Dirk Behme <dirk.behme@googlemail.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoARM DaVinci: EMIF settings
Thomas Lange [Sat, 20 Jun 2009 09:02:17 +0000 (11:02 +0200)]
ARM DaVinci: EMIF settings

NAND module should not modify EMIF registers unrelated to CS2
that is used for NAND, i.e. do not modify EWAIT config register
or registers for other Chip Selects.

Without this patch, EMIF configurations made in board_init()
will be invalidated.

Signed-off-by: Thomas Lange <thomas@corelatus.se>
15 years agoversatile: config coding style cleanup
Jean-Christophe PLAGNIOL-VILLARD [Sat, 20 Jun 2009 09:02:17 +0000 (11:02 +0200)]
versatile: config coding style cleanup

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Cc: Peter Pearse <peter.pearse@arm.com>
15 years agoarm: Kirkwood: Basic SOCs support
Prafulla Wadaskar [Sat, 20 Jun 2009 09:01:53 +0000 (11:01 +0200)]
arm: Kirkwood: Basic SOCs support

Kirkwood family controllers are highly integrated SOCs
based on Feroceon-88FR131/Sheeva-88SV131/arm926ejs cpu core.

SOC versions supported:-
1) 88F6281-A0       define CONFIG_KW88F6281_A0
2) 88F6192-A0       define CONFIG_KW88F6192_A0

Other supported features:-
1) get_random_hex() fucntion
2) PCI Express port initialization
3) NS16550 driver support

Contributors:
Yotam Admon <yotam@marvell.com>
Michael Blostein <michaelbl@marvell.com

Reviewed-by: Ronen Shitrit <rshitrit@marvell.com>
Acked-by: Stefan Rose <sr@denx.de>
Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
15 years agoarm: generic cache.h for ARM architectures
Prafulla Wadaskar [Sat, 20 Jun 2009 09:01:52 +0000 (11:01 +0200)]
arm: generic cache.h for ARM architectures

This patch is required for Kirkwood SoC support
may be used by other ARM architectures

Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
15 years agoOMAP3EVM: fix typo. replace CS6 by CS5, no functionality change
Matthias Ludwig [Sat, 20 Jun 2009 09:01:50 +0000 (11:01 +0200)]
OMAP3EVM: fix typo. replace CS6 by CS5, no functionality change

Signed-off-by: Matthias Ludwig <mludwig@ultratronik.de>
15 years agoat91sam9260/9263: add back up for the rst(reset controller).
Sedji Gaouaou [Wed, 24 Jun 2009 06:32:09 +0000 (08:32 +0200)]
at91sam9260/9263: add back up for the rst(reset controller).

On the boards at91sam9260ek, at91sam9263ek and afed9260, the rstc register was
set to 0 after being set to 500 ms for the PHY reset.
Do backup the old reset length and restore it after the MACB initialisation.

Signed-off-by: Sedji Gaouaou <sedji.gaouaou@atmel.com>
Signed-off-by: Stelian Pop <stelian@popies.net>
15 years agofsl: Fix compiler warnings from gcc-4.4 in sys_eeprom code
Kumar Gala [Fri, 3 Jul 2009 17:45:44 +0000 (12:45 -0500)]
fsl: Fix compiler warnings from gcc-4.4 in sys_eeprom code

sys_eeprom.c: In function 'do_mac':
sys_eeprom.c:323: warning: dereferencing type-punned pointer will break strict-aliasing rules
sys_eeprom.c: In function 'mac_read_from_eeprom':
sys_eeprom.c:395: warning: dereferencing type-punned pointer will break strict-aliasing rules

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago83xx: Add support for fsl_dma driver
Peter Tyser [Tue, 30 Jun 2009 22:15:51 +0000 (17:15 -0500)]
83xx: Add support for fsl_dma driver

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Reviewed-by: Ira W. Snyder <iws@ovro.caltech.edu>
Tested-by: Ira W. Snyder <iws@ovro.caltech.edu>
Acked-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago83xx: Replace CONFIG_ECC_INIT_VIA_DDRC references
Peter Tyser [Tue, 30 Jun 2009 22:15:50 +0000 (17:15 -0500)]
83xx: Replace CONFIG_ECC_INIT_VIA_DDRC references

Update 83xx architecture's CONFIG_ECC_INIT_VIA_DDRC references to
CONFIG_ECC_INIT_VIA_DDRCONTROLLER, which other Freescale architectures
use

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago8xxx: Second UART port added for MPC85xx, MPC83xx, MPC86xx processors
Poonam Aggrwal [Thu, 2 Jul 2009 10:45:01 +0000 (16:15 +0530)]
8xxx: Second UART port added for MPC85xx, MPC83xx, MPC86xx processors

Defining the next two configs allows to switch the serial port from the
console using the setenv stdin and stdout
  1. #define CONFIG_SERIAL_MULTI 1 /* Enable both serial ports */
  2. #define CONFIG_SYS_CONSOLE_IS_IN_ENV /* determine from environment */

Signed-off-by: Poonam Aggrwal <poonam.aggrwal@freescale.com>
Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago85xx: Adds GPIO registers to MPC85xx Memory Map.
Poonam Aggrwal [Thu, 2 Jul 2009 10:44:40 +0000 (16:14 +0530)]
85xx: Adds GPIO registers to MPC85xx Memory Map.

Signed-off-by: Poonam Aggrwal <poonam.aggrwal@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago86xx: XPedite5170 board support
Peter Tyser [Tue, 30 Jun 2009 22:26:01 +0000 (17:26 -0500)]
86xx: XPedite5170 board support

Initial support for Extreme Engineering Solutions XPedite5170 -
a MPC8640-based 3U VPX single board computer with a PMC/XMC
site.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_ddr: Fix DDR3 calculation of rank density with 8GB or more
Timur Tabi [Wed, 1 Jul 2009 21:51:59 +0000 (16:51 -0500)]
fsl_ddr: Fix DDR3 calculation of rank density with 8GB or more

The calculate for rank density in compute_ranksize() for DDR3 used all
integers for the expression, so the result was also a 32-bit integer, even
though the 'bsize' variable is a u64.  Fix the expression to calculate a
true 64-bit value.

Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_dma: Make DMA transactions snoopable
Peter Tyser [Tue, 30 Jun 2009 22:15:49 +0000 (17:15 -0500)]
fsl_dma: Make DMA transactions snoopable

Make DMA transactions snoopable so that CPUs can keep caches up-to-date.
This allows dma transactions to be used for operations such as memory
copies without any additional cache control operations.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_dma: Break out common memory initialization function
Peter Tyser [Tue, 30 Jun 2009 22:15:48 +0000 (17:15 -0500)]
fsl_dma: Break out common memory initialization function

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago8xxx: Move dma_init() call to common code
Peter Tyser [Tue, 30 Jun 2009 22:15:47 +0000 (17:15 -0500)]
8xxx: Move dma_init() call to common code

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_dma: Move dma function prototypes to common header file
Peter Tyser [Tue, 30 Jun 2009 22:15:46 +0000 (17:15 -0500)]
fsl_dma: Move dma function prototypes to common header file

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago8xxx: Rename dma_xfer() to dmacpy()
Peter Tyser [Tue, 30 Jun 2009 22:15:45 +0000 (17:15 -0500)]
8xxx: Rename dma_xfer() to dmacpy()

Also update dmacpy()'s argument order to match memcpy's and use
phys_addr_t/phy_size_t for address/size arguments

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_dma: Fix Channel Start bug in dma_check()
Peter Tyser [Tue, 30 Jun 2009 22:15:44 +0000 (17:15 -0500)]
fsl_dma: Fix Channel Start bug in dma_check()

The Channel Start (CS) bit in the Mode Register (MR) should actually be
cleared as the comment in the code suggests.  Previously, CS was being
set, not cleared.

Assuming normal operation of the DMA engine, this change shouldn't have
any real affect.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_dma: Add support for arbitrarily large transfers
Peter Tyser [Tue, 30 Jun 2009 22:15:43 +0000 (17:15 -0500)]
fsl_dma: Add support for arbitrarily large transfers

Support DMA transfers larger than the DMA controller's limit of
(2 ^ 26 - 1) bytes

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_dma: Use proper I/O access functions
Peter Tyser [Tue, 30 Jun 2009 22:15:42 +0000 (17:15 -0500)]
fsl_dma: Use proper I/O access functions

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_dma: Add bitfield definitions for common registers
Peter Tyser [Tue, 30 Jun 2009 22:15:41 +0000 (17:15 -0500)]
fsl_dma: Add bitfield definitions for common registers

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago8xxx: Break out DMA code to a common file
Peter Tyser [Tue, 30 Jun 2009 22:15:40 +0000 (17:15 -0500)]
8xxx: Break out DMA code to a common file

DMA support is now enabled via the CONFIG_FSL_DMA define instead of the
previous CONFIG_DDR_ECC

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoAtmel LCD driver GUARDTIME fix
Mark Jackson [Mon, 29 Jun 2009 14:59:10 +0000 (15:59 +0100)]
Atmel LCD driver GUARDTIME fix

This patch allows the guard time parameter to be set in
the Atmel LCDC driver.

By default, the previous value of 1 is used, unless the
setting is defined elsewhere.

Signed-off-by: Mark Jackson <mpfj@mimc.co.uk>
15 years ago85xx: Add pci e1000 Ethernet support for P2020 board
Roy Zang [Tue, 30 Jun 2009 05:56:23 +0000 (13:56 +0800)]
85xx: Add pci e1000 Ethernet support for P2020 board

Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago8xxx: Fix PCI bus address setup for 36-bit configs
Kumar Gala [Thu, 18 Jun 2009 13:39:42 +0000 (08:39 -0500)]
8xxx: Fix PCI bus address setup for 36-bit configs

We want the outbound PCI memory map to end at the 4G boundary so we
can maximize the amount of space available for inbound mappings if
we have large amounts of memory.

This matches the device tree setup in the kernel for the 36-bit physical
configs for the platforms that have one (MPC8641 HPCN & MPC8572 DS).

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years ago86xx: Add CPU_TYPE_ENTRY support
Kumar Gala [Thu, 18 Jun 2009 13:23:01 +0000 (08:23 -0500)]
86xx: Add CPU_TYPE_ENTRY support

Unify with 83xx and 85xx and use CPU_TYPE_ENTRY.  We are going to use
this to convey the # of cores and DDR width in the near future so its
good to keep in sync.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoBlackfin: TWI/I2C: fix pure writes
Peter Meerwald [Mon, 29 Jun 2009 19:48:33 +0000 (15:48 -0400)]
Blackfin: TWI/I2C: fix pure writes

If doing a pure write with register address and data (not a read/write
combo transfer), we don't set the initial transfer length properly which
ends up causing only the register address to be transferred.

While we're here, fix the i2c_write() parameter description of the buffer.

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agospi: Add Marvell Kirkwood SPI driver
Prafulla Wadaskar [Fri, 29 May 2009 19:43:33 +0000 (01:13 +0530)]
spi: Add Marvell Kirkwood SPI driver

This patch adds a SPI driver for the Marvell Kirkwood SoC's.

Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
15 years agos3c64xx: move the reset_cpu function
Minkyu Kang [Thu, 25 Jun 2009 10:21:33 +0000 (19:21 +0900)]
s3c64xx: move the reset_cpu function

Because of the reset_cpu is soc specific, should be move to soc
And read reset value from SYS_ID register instead of hard code
this patch also supports s3c6410

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
15 years agointegrator: merge integratorap and integratorcp
Jean-Christophe PLAGNIOL-VILLARD [Sat, 16 May 2009 22:58:37 +0000 (00:58 +0200)]
integrator: merge integratorap and integratorcp

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Peter Pearse <peter.pearse@arm.com>
15 years agointegratorap/cp: use cfi driver
Jean-Christophe PLAGNIOL-VILLARD [Sat, 16 May 2009 22:58:36 +0000 (00:58 +0200)]
integratorap/cp: use cfi driver

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Peter Pearse <peter.pearse@arm.com>
15 years agointegratorap/cp/versatile: remove non used functions
Jean-Christophe PLAGNIOL-VILLARD [Sat, 16 May 2009 22:58:36 +0000 (00:58 +0200)]
integratorap/cp/versatile: remove non used functions

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Peter Pearse <peter.pearse@arm.com>
15 years agointegratorcp: split timer support
Jean-Christophe PLAGNIOL-VILLARD [Sat, 16 May 2009 22:58:36 +0000 (00:58 +0200)]
integratorcp: split timer support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Peter Pearse <peter.pearse@arm.com>
15 years agointegratorap: split timer support
Jean-Christophe PLAGNIOL-VILLARD [Sat, 16 May 2009 22:58:36 +0000 (00:58 +0200)]
integratorap: split timer support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Peter Pearse <peter.pearse@arm.com>
15 years agointegratorap: split pci support
Jean-Christophe PLAGNIOL-VILLARD [Sat, 16 May 2009 22:58:36 +0000 (00:58 +0200)]
integratorap: split pci support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Peter Pearse <peter.pearse@arm.com>
15 years agoarm: add support for CONFIG_GENERIC_MMC
Ilya Yanok [Mon, 8 Jun 2009 00:12:50 +0000 (04:12 +0400)]
arm: add support for CONFIG_GENERIC_MMC

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agoserial_mx31: allow it to work with mx27 too and rename to serial_mxc
Ilya Yanok [Mon, 8 Jun 2009 00:12:46 +0000 (04:12 +0400)]
serial_mx31: allow it to work with mx27 too and rename to serial_mxc

UART hardware on i.MX27 is the same as on the i.MX31 so we just
need to provide the driver with correct address of the registers.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agomx27: basic cpu support
Ilya Yanok [Mon, 8 Jun 2009 00:12:45 +0000 (04:12 +0400)]
mx27: basic cpu support

This patch adds generic code to support Freescale's i.MX27 SoCs.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agoi.MX31: Create a common device file.
Magnus Lilja [Sat, 13 Jun 2009 18:50:03 +0000 (20:50 +0200)]
i.MX31: Create a common device file.

Signed-off-by: Magnus Lilja <lilja.magnus@gmail.com>
15 years agoARM: Add macros.h to be used in assembler file.
Jean-Christophe PLAGNIOL-VILLARD [Sat, 13 Jun 2009 18:50:02 +0000 (20:50 +0200)]
ARM: Add macros.h to be used in assembler file.

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoMX31: Add NAND SPL for i.MX31.
Magnus Lilja [Sat, 13 Jun 2009 18:50:01 +0000 (20:50 +0200)]
MX31: Add NAND SPL for i.MX31.

This patch adds the NAND SPL framework needed to boot i.MX31 boards
from NAND.

It has been tested on a i.MX31 PDK board with large page NAND. Small
page NANDs should work as well, but this has not been tested.

Note: The i.MX31 NFC uses a non-standard layout for large page NANDs,
whether this is compatible with a particular setup depends on how
the NAND device is programmed by the flash programmer (e.g. JTAG
debugger).

The patch is based on the work by Maxim Artamonov.

Signed-off-by: Maxim Artamonov <scn1874@yandex.ru>
Signed-off-by: Magnus Lilja <lilja.magnus@gmail.com>
15 years agoARM1136: Introduce CONFIG_PRELOADER macro.
Magnus Lilja [Sat, 13 Jun 2009 18:50:00 +0000 (20:50 +0200)]
ARM1136: Introduce CONFIG_PRELOADER macro.

Currently CONFIG_ONENAND_IPL is used in a number of #ifdef's
in start.S. In preparation for adding support for NAND SPL
the macro CONFIG_PRELOADER is introducted and replaces the
CONFIG_ONENAND_IPL in start.S.

Signed-off-by: Magnus Lilja <lilja.magnus@gmail.com>
15 years agoat91: unify nor boot support
Jean-Christophe PLAGNIOL-VILLARD [Sat, 13 Jun 2009 10:50:04 +0000 (12:50 +0200)]
at91: unify nor boot support

the lowlevel init sequence is the same so unify it

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91sam9263ek: add nor flash support
Jean-Christophe PLAGNIOL-VILLARD [Sat, 13 Jun 2009 10:48:36 +0000 (12:48 +0200)]
at91sam9263ek: add nor flash support

this will allow you to store use it for the env and to boot directly U-Boot from

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91: add support for the PM9261 board of Ronetix GmbH
Ilko Iliev [Fri, 12 Jun 2009 19:20:39 +0000 (21:20 +0200)]
at91: add support for the PM9261 board of Ronetix GmbH

The PM9261 board is based on the AT91SAM9261-EK board.

Here is the page on Ronetix website:
http://www.ronetix.at/starter_kit_9261.html

Signed-off-by: Ilko Iliev <iliev@ronetix.at>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agopm9263: use macro instead of hardcode value for the lowlevel_init
Jean-Christophe PLAGNIOL-VILLARD [Fri, 12 Jun 2009 19:20:38 +0000 (21:20 +0200)]
pm9263: use macro instead of hardcode value for the lowlevel_init

optimize a few the RAM init

Signed-off-by: Ilko Iliev <iliev@ronetix.at>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agopm9263: lowlevel init update
Jean-Christophe PLAGNIOL-VILLARD [Fri, 12 Jun 2009 19:20:37 +0000 (21:20 +0200)]
pm9263: lowlevel init update

move PSRAM init to pm9263.c
this will allow us after to make the nor lowlevel_init generic

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoBlackfin: move ALL += u-boot.ldr to blackfin_config.mk
Mike Frysinger [Mon, 15 Jun 2009 04:25:19 +0000 (00:25 -0400)]
Blackfin: move ALL += u-boot.ldr to blackfin_config.mk

The way the ALL variable is used allows for config.mk's to add more
targets themselves without having to clutter up the top level Makefile.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: fix SPI flash speed define name
Mike Frysinger [Mon, 15 Jun 2009 02:29:35 +0000 (22:29 -0400)]
Blackfin: fix SPI flash speed define name

The SPI flash define is named CONFIG_SF_DEFAULT_SPEED, not
CONFIG_SF_DEFAULT_HZ, so fix the typos in the Blackfin boards.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: enable -O2 in lib_generic/ for ADI/Bluetechnix boards
Mike Frysinger [Mon, 15 Jun 2009 02:26:31 +0000 (22:26 -0400)]
Blackfin: enable -O2 in lib_generic/ for ADI/Bluetechnix boards

Building the compression code in lib_generic/ with -O2 rather than -Os
gives a nice speed boost without too much code size increase.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: bf548-ezkit: bump up monitor size
Mike Frysinger [Mon, 15 Jun 2009 01:23:27 +0000 (21:23 -0400)]
Blackfin: bf548-ezkit: bump up monitor size

The latest version of U-Boot got a bit fatter in the BSS section which
caused overflows in the RAM region, so increase the monitor size.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: bf533-stamp/bf537-stamp: fix env settings for SPI flash
Vivi Li [Fri, 12 Jun 2009 10:53:22 +0000 (10:53 +0000)]
Blackfin: bf533-stamp/bf537-stamp: fix env settings for SPI flash

The SPI flash layer is much stricter about sector usage than the eeprom
layer we used to use, so update the env settings to better match the
sector alignment of the flashes we use.

Signed-off-by: Vivi Li <vivi.li@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: bump up default JTAG console timeout
Vivi Li [Fri, 12 Jun 2009 10:33:23 +0000 (10:33 +0000)]
Blackfin: bump up default JTAG console timeout

The debug tools that interface with the other side of the JTAG console
got much slower when generalizing things, so bump up the default timeout
value on the U-Boot side to cope.  Hopefully at some point we can improve
the debug tools to speed things back up.

Signed-off-by: Vivi Li <vivi.li@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Mon, 15 Jun 2009 09:16:04 +0000 (11:16 +0200)]
Merge branch 'master' of /home/wd/git/u-boot/custodians

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-net
Wolfgang Denk [Mon, 15 Jun 2009 09:15:54 +0000 (11:15 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-net