kernel/u-boot.git
15 years agoRemove sa1100.h
Shinya Kuribayashi [Thu, 16 Apr 2009 01:51:03 +0000 (10:51 +0900)]
Remove sa1100.h

sa1100.h is not used anywhere, then remove it.

$ find . -name '*.h' -empty -print
./include/sa1100.h
$ git grep 'sa1100.h' .
$

Signed-off-by: Shinya Kuribayashi <shinya.kuribayashi@necel.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoUBIFS: Remove tnc_commit.c which is not used in the read-only version
Stefan Roese [Tue, 14 Apr 2009 15:51:21 +0000 (17:51 +0200)]
UBIFS: Remove tnc_commit.c which is not used in the read-only version

I missed removing this file while implementing the UBIFS support. It's
not referenced at all, so let's remove it. Thanks to Artem Bityutskiy
for spotting.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoUBIFS: fix recovery bug
Adrian Hunter [Tue, 14 Apr 2009 15:50:38 +0000 (17:50 +0200)]
UBIFS: fix recovery bug

UBIFS did not recovery in a situation in which it could
have. The relevant function assumed there could not be
more nodes in an eraseblock after a corrupted node, but
in fact the last (NAND) page written might contain anything.
The correct approach is to check for empty space (0xFF bytes)
from then on.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agofs: Fix the wrong type of var
Gao Guanhua [Tue, 14 Apr 2009 06:37:35 +0000 (14:37 +0800)]
fs: Fix the wrong type of var

The filelen should be signed type, not unsigned type.
otherwise, The condition as below never take.
if (filelen < 0)

Signed-off-by: Gao Guanhua <B22826@freescale.com>
Signed-off-by: Dave Liu <daveliu@freescale.com>
15 years agoUpdate CHANGELOG
Wolfgang Denk [Mon, 27 Apr 2009 22:29:34 +0000 (00:29 +0200)]
Update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agodm9000 EEPROM reading bugfix
David Brownell [Fri, 17 Apr 2009 06:15:15 +0000 (23:15 -0700)]
dm9000 EEPROM reading bugfix

Make the U-Boot dm9000 driver read addresses from EEPROM just
like Linux does ... read six bytes, instead of reading twelve
bytes and then discarding every other one.

Using the right Ethernet address is a big win.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Ben Warren <biggerbadderben@gmail.com>
15 years agortl8169: fix cache coherency problem
Yoshihiro Shimoda [Wed, 25 Feb 2009 05:27:24 +0000 (14:27 +0900)]
rtl8169: fix cache coherency problem

Fix the problem that cannot access actual data when CPU data cache enabled.

Signed-off-by: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Tested-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Acked-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc83xx
Wolfgang Denk [Mon, 27 Apr 2009 21:41:01 +0000 (23:41 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc83xx

15 years agolib_arm/board.c: remove misleading "test-only" comment.
Wolfgang Denk [Sun, 26 Apr 2009 18:39:26 +0000 (20:39 +0200)]
lib_arm/board.c: remove misleading "test-only" comment.

For a long time, the print_cpuinfo() declaration in lib_arm/board.c
had been marked as "test-only", which is plain wrong considering
current usage.  Delete this misleading comment.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years ago83xx: searching "muram-data" by compatible property
Heiko Schocher [Fri, 24 Apr 2009 04:50:45 +0000 (06:50 +0200)]
83xx: searching "muram-data" by compatible property

if using CONFIG_BOOTCOUNT_LIMIT feature on a MPC8360 CPU
in the muram-data node, the reg entry needs to be updated.
This is done in fdt_fixup_muram(), but we should use
the compatible "fsl,qe-muram-data" for searching the
node instead of searching the muram-data node with
an absolute path.

Signed-off-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-net
Wolfgang Denk [Fri, 24 Apr 2009 11:36:08 +0000 (13:36 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-net

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-at91
Wolfgang Denk [Fri, 24 Apr 2009 11:31:02 +0000 (13:31 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-at91

15 years agompc83xx: MPC8360ERDK: fix environment offset configuration bug
Anatolij Gustschin [Thu, 23 Apr 2009 19:29:34 +0000 (21:29 +0200)]
mpc83xx: MPC8360ERDK: fix environment offset configuration bug

The size of U-Boot binary for MPC8360ERDK increased
(> 2 flash sectors now), so 'saveenv' will partially
overwrite U-Boot in flash and will brick the board.
This patch moves environment offset to fourth flash
sector and also fixes CONFIG_SYS_MONITOR_LEN.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agoNetLoop initialization bug
Michael Zaidman [Fri, 3 Apr 2009 22:43:00 +0000 (01:43 +0300)]
NetLoop initialization bug

The patch fixes the bug of partial initialization of global network
parameters.

Upon u-boot's start up the first ping command causes a failure of the
consequent TFTP command. It happens in the recently added mechanism of
the NetLoop initialization where initialization of global network
parameters is separated in the NetInitLoop routine which is called per
env_id change. Thus, ping request will initialize the network parameters
necessary for ping operation only, afterwards the env_changed_id will be
set to the env_id that will prevent all following initialization requests
from other protocols.
The problem is that the initialized by ping subset of network parameters
is not sufficient for other protocols and particularly for TFTP which
requires the NetServerIp also.

Signed-off-by: Michael Zaidman <michael.zaidman@gmail.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agonet: fix ULI 526x macro usage in netdev.h
Timur Tabi [Thu, 9 Apr 2009 15:27:05 +0000 (10:27 -0500)]
net: fix ULI 526x macro usage in netdev.h

Change netdev.h to use CONFIG_ULI526X instead of CONFIG_ULI526.  CONFIG_ULI526X
is used everywhere else, so that's the correct macro name.  Without this fix,
Ethernet will not work on the Freescale MPC8610 HPCD.

Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agofix DaVinci NS16550_REG_SIZE regression
David Brownell [Sun, 12 Apr 2009 22:38:06 +0000 (15:38 -0700)]
fix DaVinci NS16550_REG_SIZE regression

Update the DaVinci DM6446 boards to use the new convention
for CONFIG_SYS_NS16550_REG_SIZE ... the size hasn't changed
from the original 4 bytes, but these chips are little-endian.

(Resolves a regression added recently by the include/ns16550.h
patch to "Unify structure declaration for registers".  The code
previously worked just fine because the registers were accessed
as host-endian words, not as bytes.)

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agocmd_nand: drop duplicate NULL ptr check
Mike Frysinger [Mon, 13 Apr 2009 02:29:20 +0000 (22:29 -0400)]
cmd_nand: drop duplicate NULL ptr check

The first if statement checks for NULL ptrs, so there is no need to check
it again in later else cases (such as .oob).

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
CC: Scott Wood <scottwood@freescale.com>
15 years agoCMD_UBI != MTD_PARTITIONS
David Brownell [Mon, 13 Apr 2009 15:03:38 +0000 (08:03 -0700)]
CMD_UBI != MTD_PARTITIONS

Fix dependency goofage:  it should certainly be possible to have the
partition support without bringing in UBI commands.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Stefan Roese <sr@denx.de>
15 years agoat91sam9/at91cap: fix CONFIG_SYS_HZ to 1000
Jean-Christophe PLAGNIOL-VILLARD [Thu, 16 Apr 2009 19:30:48 +0000 (21:30 +0200)]
at91sam9/at91cap: fix CONFIG_SYS_HZ to 1000

The timer has been rewrote with a precision at ~0,18%

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Tested-by: Sergey Lapin <slapin@ossfans.org>
Tested-by: Eric BENARD <ebenard@free.fr>
15 years agoat91: add support for the PM9263 board of Ronetix GmbH
Ilko Iliev [Thu, 16 Apr 2009 19:30:48 +0000 (21:30 +0200)]
at91: add support for the PM9263 board of Ronetix GmbH

The PM9263 board is based on the AT91SAM9263-EK board.

Here is the page on Ronetix website:
http://www.ronetix.at/starter_kit_9263.html

Signed-off-by: Ilko Iliev <iliev@ronetix.at>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91sam9/at91cap: improve clock framework
Jean-Christophe PLAGNIOL-VILLARD [Thu, 16 Apr 2009 19:30:44 +0000 (21:30 +0200)]
at91sam9/at91cap: improve clock framework

calculate dynamically the clock rate and pllb setting for usb

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoppc4xx: Disable POST memory test on NAND-booting Kilauea
Stefan Roese [Wed, 15 Apr 2009 12:08:48 +0000 (14:08 +0200)]
ppc4xx: Disable POST memory test on NAND-booting Kilauea

Don't run the memory POST on the NAND-booting version. It will
overwrite part of the U-Boot image which is already loaded from NAND
to SDRAM. We were just lucky that it booted at all with this SDRAM
test enabled.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Disable POST memory test on NAND-booting Sequoia
Stefan Roese [Wed, 15 Apr 2009 12:06:26 +0000 (14:06 +0200)]
ppc4xx: Disable POST memory test on NAND-booting Sequoia

Don't run the memory POST on the NAND-booting version. It will
overwrite part of the U-Boot image which is already loaded from NAND
to SDRAM. We were just lucky that it booted at all with this SDRAM
test enabled.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Remove unused code for Sequoia NAND booting version
Stefan Roese [Wed, 15 Apr 2009 09:32:53 +0000 (11:32 +0200)]
ppc4xx: Remove unused code for Sequoia NAND booting version

The current define of get_bus_freq() in the CONFIG_NAND_SPL #ifdef is not
used at all. This patch changes it's define to the currently used value of
133333333 and removes the unnecessary code.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Add "booting from NAND" to 4xx NAND-booting targets
Stefan Roese [Wed, 15 Apr 2009 08:50:48 +0000 (10:50 +0200)]
ppc4xx: Add "booting from NAND" to 4xx NAND-booting targets

This additional text in the bootup log helps to see if the board is
configured for NAND-booting. Especially helpful for boards that can
boot from NOR and NAND (e.g. most of the AMCC eval boards).

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Fixup chip-selects in dtb for NAND-booting Sequoia
Stefan Roese [Wed, 8 Apr 2009 08:36:22 +0000 (10:36 +0200)]
ppc4xx: Fixup chip-selects in dtb for NAND-booting Sequoia

Currently the NOR & NAND support in Linux only works for the "standard"
Sequoia, the version booting for NOR flash. The NAND-booting version
has the chip-selects swapped. Here the chip-select mappings:

"Standard" NOR-booting version:
CS0 NOR
CS3 NAND

NAND-booting version:
CS0 NAND
CS3 NOR

With this path the dtb gets fixed-up, so that the correct chip-select
numbers are patched in the dtb enabling correct NOR & NAND support
in Linux on the NAND-booting Sequoia version.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-blackfin
Wolfgang Denk [Tue, 7 Apr 2009 19:53:07 +0000 (21:53 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-blackfin

15 years agoBlackfin: audit UART for all known anomalies
Mike Frysinger [Sat, 4 Apr 2009 13:10:27 +0000 (09:10 -0400)]
Blackfin: audit UART for all known anomalies

There is no code change here, just new comments, but this keeps me from
having to do another audit from scratch in the future.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: add check for anomaly 05000362
Mike Frysinger [Sat, 4 Apr 2009 12:40:13 +0000 (08:40 -0400)]
Blackfin: add check for anomaly 05000362

DESCRIPTION:
The column address width settings for banks 2 and 3 are misconnected in
the SDRAM controller.  Accesses to bank 2 will result in an error if the
Column Address Width for bank 3 (EB3CAW ) is not set to be the same as
that of bank 2.

WORKAROUND:
If using bank 2, make sure that banks 2 and 3 have the same column address
width settings in the EBIU_SDBCTL register.  This must be the case
regardless of whether or not bank 3 is enabled.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: add comment about anomaly 05000430 avoidance
Mike Frysinger [Sat, 4 Apr 2009 12:29:55 +0000 (08:29 -0400)]
Blackfin: add comment about anomaly 05000430 avoidance

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: add workaround for anomaly 05000242
Mike Frysinger [Sat, 4 Apr 2009 12:10:22 +0000 (08:10 -0400)]
Blackfin: add workaround for anomaly 05000242

DESCRIPTION:
If the DF bit is set prior to a hardware reset, the PLL will continue to
divide CLKIN by 2 after the hardware reset, but the DF bit itself will be
cleared in the PLL_CTL register.

WORKAROUND:
Reprogram the PLL with DF cleared if the desire is to not divide CLKIN by
2 after reset.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: add workaround for anomaly 05000171
Mike Frysinger [Sat, 4 Apr 2009 12:09:24 +0000 (08:09 -0400)]
Blackfin: add workaround for anomaly 05000171

DESCRIPTION:
The Boot ROM is executed at power up/reset and changes the value of the
SICA_IWR registers from their default reset value of 0xFFFF, but does not
restore them.

WORKAROUND:
User code should not rely on the default value of these registers.  Set
the desired values explicitly.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: update anomaly sheets
Mike Frysinger [Sat, 4 Apr 2009 12:22:36 +0000 (08:22 -0400)]
Blackfin: update anomaly sheets

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: nand: flush peripheral before polling it
Graf Yang [Sat, 4 Apr 2009 11:45:57 +0000 (07:45 -0400)]
Blackfin: nand: flush peripheral before polling it

We need to make sure the data written to the nand flash controller makes
it there before we start polling its status register.  Otherwise, we may
get stale data and return before the controller is actually ready.

Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Scott Wood <scottwood@freescale.com>
15 years agoSupport for PXA27X UDC.
Remy Bohmer [Sun, 5 Apr 2009 09:43:28 +0000 (11:43 +0200)]
Support for PXA27X UDC.

This Patch adds Support for PXA27X UDC.
(Rebased to drivers/usb reorganisation)

Signed-off-by: Vivek Kutal <vivek.kutal@azingo.com>
Signed-off-by: Remy Bohmer <linux@bohmer.net>
15 years agodrivers/usb: regorganisation
Jean-Christophe PLAGNIOL-VILLARD [Fri, 3 Apr 2009 10:46:58 +0000 (12:46 +0200)]
drivers/usb: regorganisation

move to linux usb driver organisation

as following

drivers/usb/gadget
drivers/usb/host
drivers/usb/musb

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Remy Bohmer <linux@bohmer.net>
15 years agoBlackfin: spi: make cs deassert function deterministic
Todor I Mollov [Sat, 4 Apr 2009 10:53:06 +0000 (06:53 -0400)]
Blackfin: spi: make cs deassert function deterministic

Blackfin SPI driver was not driving the SPI chip-select high before
putting the chip-select signals into tri-state mode.  This is probably
something that slipped by unnoticed in most designs.  If the signals are
put directly into a tri-state mode, then the board is relying on the
pull-up resistors to pull up the chip-select before the next transaction.
Most of the time this is fine, except when you have two transactions that
follow each other very closely, such as the flash erase and read status
register commands.  In this case I was seeing a 500ns separation between
the transactions.  In my setup, with a 10kOhm pull-up, it would meet
timing spec about half the time and resulted in intermittent errors.  (A
stronger pull up would fix this, but our design is targeted for low power
consumption and a 3.3kOhm @ 3.3v is 3.3mW of needless power consumption.)
I modified the spi_cs_deactivate() function in bfin_spi.c to drive the
chip-selects high before putting them into tri-state.  For me, this
resulted in a rise time of 5ns instead of the previous rise time of about
1us, and fully satisfied the timing spec of the chip.

Signed-off-by: Todor I Mollov <tmollov@ucsd.edu>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-arm
Wolfgang Denk [Sun, 5 Apr 2009 21:04:30 +0000 (23:04 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-arm

15 years agointegratorap: fix PCI support
Jean-Christophe PLAGNIOL-VILLARD [Sat, 13 Dec 2008 20:08:05 +0000 (21:08 +0100)]
integratorap: fix PCI support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoOMAP3: Update Overo pin mux for new expansion board
Dirk Behme [Mon, 30 Mar 2009 19:15:23 +0000 (21:15 +0200)]
OMAP3: Update Overo pin mux for new expansion board

A new Overo expansion board uses GPIO 14, 21, 22 and 23 for LED's and
switches. This patch changes the pinmux configuration for those pins.
They were previously set up for unused MMC3_DAT4-7.

Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoarm: unify reset command
Jean-Christophe PLAGNIOL-VILLARD [Sun, 5 Apr 2009 11:08:03 +0000 (13:08 +0200)]
arm: unify reset command

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoarm: clean cache management
Jean-Christophe PLAGNIOL-VILLARD [Sun, 5 Apr 2009 11:06:31 +0000 (13:06 +0200)]
arm: clean cache management

unify arm cache management except for non standard cache as ARM7TDMI

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoarm: update co-processor 15 access
Jean-Christophe PLAGNIOL-VILLARD [Sun, 5 Apr 2009 11:02:43 +0000 (13:02 +0200)]
arm: update co-processor 15 access

import system.h from linux

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoARM: add the imx31_phycore_eet target to MAINTAINERS
Guennadi Liakhovetski [Sat, 4 Apr 2009 22:42:02 +0000 (00:42 +0200)]
ARM: add the imx31_phycore_eet target to MAINTAINERS

imx31_phycore_eet is a variant of the imx31_phycore board with a few
extensions, which justifies a separate entry in the MAINTAINERS list,
whereas normally all entries sharing a single configuration file and a
board/ directory have only one entry in MAINTAINERS.

Reported-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
15 years agoARM: fix out-of-tree build of imx31_phycore_eet
Guennadi Liakhovetski [Sat, 4 Apr 2009 22:37:07 +0000 (00:37 +0200)]
ARM: fix out-of-tree build of imx31_phycore_eet

Fix out-of-tree build of the imx31_phycore_eet target.

Reported-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
15 years agompc5200: reduce delays in i2c
Jon Smirl [Sat, 4 Apr 2009 21:44:51 +0000 (17:44 -0400)]
mpc5200: reduce delays in i2c

The previous code waited 1000us before checking i2c
status. Measurement shows i2c is usually ready in
under 50us. Change the polling interval to 15us,
loop 6,667 times to keep the polling timeout constant
at 100ms.

15 years agoOMAP: Fix compile issue
Sanjeev Premi [Fri, 3 Apr 2009 08:30:07 +0000 (14:00 +0530)]
OMAP: Fix compile issue

Fixes this compile error:
board.c: In function 'do_switch_ecc':
board.c:339: error: 'cmd_tbl_t' has no member named 'help'
make[1]: *** [board.o] Error 1
make[1]: Leaving directory `/db/psp_git/users/a0756819/u-boot/cpu/arm_cortexa8/omap3'
make: *** [cpu/arm_cortexa8/omap3/libomap3.a] Error 2

This is due to the fact that current command uses long
help for the usage print even if the CONFIG_SYS_LONGHELP
is not enabled. (Thanks Jean-Christophe for explanation).

Signed-off-by: Sanjeev Premi <premi@ti.com>
15 years agos3c64xx: fix the wrong gpio offset
Minkyu Kang [Fri, 3 Apr 2009 00:56:16 +0000 (09:56 +0900)]
s3c64xx: fix the wrong gpio offset

This patch fix the wrong gpio offset

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
15 years agoMove machine specific code to board at s3c64xx (v2)
Kyungmin Park [Wed, 26 Nov 2008 01:18:13 +0000 (10:18 +0900)]
Move machine specific code to board at s3c64xx (v2)

Move machine specific code to smdk6400.
Some board use OneNAND instead of NAND.

Some register MP0_CS_CFG[5:0] are controled by both h/w and s/w.
So it's better to use macro instead of hard-coded value.

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
15 years agoZOOM1 Remove legacy NAND defines
Tom Rix [Thu, 2 Apr 2009 03:02:19 +0000 (22:02 -0500)]
ZOOM1 Remove legacy NAND defines

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
15 years agoOMAP3 Fix multiline formatting in board init files.
Tom Rix [Thu, 2 Apr 2009 03:02:20 +0000 (22:02 -0500)]
OMAP3 Fix multiline formatting in board init files.

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
15 years agoUpdate CHANGELOG, coding style cleanup.
Wolfgang Denk [Sat, 4 Apr 2009 22:27:57 +0000 (00:27 +0200)]
Update CHANGELOG, coding style cleanup.

15 years agoOne more fix for building env_embedded.o
Wolfgang Denk [Sat, 4 Apr 2009 22:18:44 +0000 (00:18 +0200)]
One more fix for building env_embedded.o

In addition to the changes for CONFIG_ENV_IS_IN_FLASH as done in
commit afcbce07, we also need to do the same for
CONFIG_ENV_IS_IN_EEPROM and CONFIG_ENV_IS_IN_NVRAM.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agocommon/image.c: Relocate strings in tables.
Scott Wood [Thu, 2 Apr 2009 21:15:10 +0000 (16:15 -0500)]
common/image.c: Relocate strings in tables.

Without this, u-boot can crash or print garbage if the original link
address no longer points to a valid string.

Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agompc8260: Fill in brg's clock-frequency in device tree.
Scott Wood [Thu, 2 Apr 2009 21:10:36 +0000 (16:10 -0500)]
mpc8260: Fill in brg's clock-frequency in device tree.

Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoMPC8260ADS: Add nfsboot/ramboot to default environment.
Scott Wood [Thu, 2 Apr 2009 23:20:43 +0000 (18:20 -0500)]
MPC8260ADS: Add nfsboot/ramboot to default environment.

This brings it in line with other Freescale boards.

Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoUBI/cfi-mtd: Fix mtd name for multiple chips
Andreas Huber [Thu, 2 Apr 2009 15:15:34 +0000 (17:15 +0200)]
UBI/cfi-mtd: Fix mtd name for multiple chips

On platforms with multiple NOR chips, currently only the first one
can be selected using the "ubi part" command. This patch fixes this
problem by using different names for the NOR "mtd devices".

It also changes the name of the NOR MTD device from "cfi-mtd" to
"norX" (X indexing the device numer) to better match the mtdparts
defaults.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Andreas Huber <andreas.huber@keymile.com>
15 years agotools/setlocalversion: use git svn instead of git-svn
Peter Korsgaard [Tue, 31 Mar 2009 15:52:16 +0000 (17:52 +0200)]
tools/setlocalversion: use git svn instead of git-svn

Use the new "git <subcmd>" syntax instead of the deprecated
"git-<subcmd>".

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
15 years agoeSPI: add the eSPI register support
Mingkai Hu [Tue, 31 Mar 2009 06:09:40 +0000 (14:09 +0800)]
eSPI: add the eSPI register support

Signed-off-by: Mingkai Hu <Mingkai.hu@freescale.com>
15 years agocmd_mmc: add support for device command for selecting mmc device
Minkyu Kang [Mon, 30 Mar 2009 05:55:51 +0000 (14:55 +0900)]
cmd_mmc: add support for device command for selecting mmc device

This patch improves device command for selecting mmc device

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
15 years agoSmall fix to m5282evb
Alan Carvalho de Assis [Sat, 28 Mar 2009 22:50:16 +0000 (19:50 -0300)]
Small fix to m5282evb

This is just a small fix to get u-boot on m5282evb.

Signed-off-by: Alan Carvalho de Assis <acassis@gmail.com>
15 years agoUBIFS: add R/O compatibility
Artem Bityutskiy [Fri, 27 Mar 2009 09:21:14 +0000 (10:21 +0100)]
UBIFS: add R/O compatibility

Now UBIFS is supported by u-boot. If we ever decide to change the
media format, then people will have to upgrade their u-boots to
mount new format images. However, very often it is possible to
preserve R/O forward-compatibility, even though the write
forward-compatibility is not preserved.

This patch introduces a new super-block field which stores the
R/O compatibility version.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Acked-by: Adrian Hunter <Adrian.Hunter@nokia.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agomore command usage cleanup
Mike Frysinger [Tue, 24 Mar 2009 02:27:34 +0000 (22:27 -0400)]
more command usage cleanup

Fix up a few dangling commands like in "Command usage cleanup" commit.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoSeparate mtdparts command from jffs2
Ladislav Michl [Mon, 23 Mar 2009 11:06:07 +0000 (12:06 +0100)]
Separate mtdparts command from jffs2

On Thu, Mar 19, 2009 at 01:30:36PM +0100, Stefan Roese wrote:
> Currently the mtdparts commands are included in the jffs2 command support.
> This doesn't make sense anymore since other commands (e.g. UBI) use this
> infrastructure as well now. This patch separates the mtdparts commands from
> the jffs2 commands making it possible to only select mtdparts when no JFFS2
> support is needed.

... and to make it useful for NAND chips as well, we should also remove now
unrelated CONFIG_JFFS2_NAND. Note that struct part_info etc is in
jffs2/load_kernel.h which is a bit misleading filename for that purpose,
but that can be fixed later (tm).

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
15 years agosmc91111_eeprom: move board-specific init into SMC91111_EEPROM_INIT()
Mike Frysinger [Mon, 23 Mar 2009 02:18:01 +0000 (22:18 -0400)]
smc91111_eeprom: move board-specific init into SMC91111_EEPROM_INIT()

Rather than sticking Blackfin-specific stuff into the eeprom example, use
an indirect macro so that any board can override it with their own magic
sauce in their board config file.

Also fix some spurious semi-colons in defines while I'm at it ...

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
CC: Ben Warren <biggerbadderben@gmail.com>
15 years agompc52xx phy: initialize only when needed
Sascha Hauer [Sat, 21 Mar 2009 13:38:46 +0000 (09:38 -0400)]
mpc52xx phy: initialize only when needed

Do not initialize phy on startup, instead initialize it
when we actually need it.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-at91
Wolfgang Denk [Sat, 4 Apr 2009 20:15:46 +0000 (22:15 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-at91

15 years agoat91sam9263ek: enable hush and auto complete support
Jean-Christophe PLAGNIOL-VILLARD [Mon, 30 Mar 2009 14:51:40 +0000 (16:51 +0200)]
at91sam9263ek: enable hush and auto complete support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Stelian Pop <stelian@popies.net>
15 years agoat91: move dataflash spi driver to drivers/spi
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:44 +0000 (23:26 +0100)]
at91: move dataflash spi driver to drivers/spi

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91: move usb driver to drivers/usb
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:44 +0000 (23:26 +0100)]
at91: move usb driver to drivers/usb

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91rm9200: Reset update
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:43 +0000 (23:26 +0100)]
at91rm9200: Reset update

Update the rm9200 reset sequence to try executing a board-specific reset
function and move specific board reset to board.

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91rm9200: move serial shutdown code to serial drivers
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:43 +0000 (23:26 +0100)]
at91rm9200: move serial shutdown code to serial drivers

introduce serial_exit for this purpose. Use it only when the rm9200
serial driver is active

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91rm9200: move serial driver to drivers/serial
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:43 +0000 (23:26 +0100)]
at91rm9200: move serial driver to drivers/serial

add CONFIG_AT91RM9200_USART to activate the driver

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoAdd support for the AT91RM9200EK Board.
Ulf Samuelsson [Fri, 27 Mar 2009 22:26:43 +0000 (23:26 +0100)]
Add support for the AT91RM9200EK Board.

The AT91RM9200-EK Evaluation Board supports the AT91RM9200
ARM9-based 32-bit RISC microcontroller and enables real-time code development
and evaluation.

Here is the chip page on Atmel website:
http://www.atmel.com/dyn/products/tools_card.asp?tool_id=3507

with
- NOR (cfi driver)
- DataFlash
- USB OHCI
- Net
- I2C (hard)

Signed-off-by: Ulf Samuelsson <ulf@atmel.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoadd dataflash mmc mux missing support
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:42 +0000 (23:26 +0100)]
add dataflash mmc mux missing support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91rm9200dk: Move conditional compilation to Makefile
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:42 +0000 (23:26 +0100)]
at91rm9200dk: Move conditional compilation to Makefile

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91: rename DATAFLASH_MMC_SELECT to CONFIG_DATAFLASH_MMC_SELECT
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:42 +0000 (23:26 +0100)]
at91: rename DATAFLASH_MMC_SELECT to CONFIG_DATAFLASH_MMC_SELECT

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91sam9: add watchdog support
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 22:26:42 +0000 (23:26 +0100)]
at91sam9: add watchdog support

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoat91sam9/at91cap: spi init add hardware chip select support
Jean-Christophe PLAGNIOL-VILLARD [Fri, 27 Mar 2009 12:14:52 +0000 (13:14 +0100)]
at91sam9/at91cap: spi init add hardware chip select support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agonetstar: fix crc32.c dependancy location
Jean-Christophe PLAGNIOL-VILLARD [Sat, 4 Apr 2009 17:15:40 +0000 (19:15 +0200)]
netstar: fix crc32.c dependancy location

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Sat, 4 Apr 2009 18:05:18 +0000 (20:05 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

15 years agoMPC8260ADS: Define CONFIG_HAS_ETH0.
Scott Wood [Fri, 3 Apr 2009 20:24:40 +0000 (15:24 -0500)]
MPC8260ADS: Define CONFIG_HAS_ETH0.

This is required so that the MAC address will be updated in the device tree.

Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoPQ2FADS: Enable PCI.
Scott Wood [Fri, 3 Apr 2009 20:26:45 +0000 (15:26 -0500)]
PQ2FADS: Enable PCI.

PCI on PQ2FADS is very similar to PCI on MPC8272ADS.

Signed-off-by: Scott Wood <scottwood@freescale.com>
15 years agoFix quoting bug introduced by commit 74de7aef
Wolfgang Denk [Sat, 4 Apr 2009 14:57:50 +0000 (16:57 +0200)]
Fix quoting bug introduced by commit 74de7aef

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoFix implicit declaration of function 'htons'
Wolfgang Denk [Sat, 4 Apr 2009 14:14:51 +0000 (16:14 +0200)]
Fix implicit declaration of function 'htons'

Include <netinet/in.h> instead of <linux/in.h> to get htons()
prototpye.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agofsl_law: Fix bug in calculation of LAW sizing
Kumar Gala [Sat, 4 Apr 2009 15:21:02 +0000 (10:21 -0500)]
fsl_law: Fix bug in calculation of LAW sizing

In set_ddr_laws() when we determined how much of the size requested
to be mapped was covered by the the first LAW we needed to recalculate
the size based on what was actually mapped.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_pci: Move prototypes into fsl_pci.h and remove explicit externs
Kumar Gala [Thu, 2 Apr 2009 18:57:05 +0000 (13:57 -0500)]
fsl_pci: Move prototypes into fsl_pci.h and remove explicit externs

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agofsl_pci: Renamed immap_fsl_pci.h to fsl_pci.h
Kumar Gala [Thu, 2 Apr 2009 18:22:48 +0000 (13:22 -0500)]
fsl_pci: Renamed immap_fsl_pci.h to fsl_pci.h

Rename the pci header for FSL HW so we can move some prototypes
in there and stop doing explicit externs

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoFix building of env_embedded.o
Wolfgang Denk [Sat, 4 Apr 2009 14:10:40 +0000 (16:10 +0200)]
Fix building of env_embedded.o

Since commit a706bfc7 common/env_embedded.o and tools/envcrc were
only built when CONFIG_ENV_IS_EMBEDDED was set, but this breaks
building for many boards.

We always have to build these files when CONFIG_ENV_IS_IN_FLASH is
set.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agorename include/zlib.h to include/u-boot/zlib.h
Jean-Christophe PLAGNIOL-VILLARD [Sat, 4 Apr 2009 10:49:11 +0000 (12:49 +0200)]
rename include/zlib.h to include/u-boot/zlib.h

Some systems have zlib.h installed in /usr/include/. This isn't the
desired file for u-boot code - we want the one in include/zlib.h.
This rename will avoid the conflict.

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agotools/Makefile: fix image.c dependancy location
Jean-Christophe PLAGNIOL-VILLARD [Sat, 4 Apr 2009 10:46:31 +0000 (12:46 +0200)]
tools/Makefile: fix image.c dependancy location

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agotools: add ncb to gitignore
Mike Frysinger [Sat, 4 Apr 2009 11:42:25 +0000 (07:42 -0400)]
tools: add ncb to gitignore

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoinclude/ns16550.h: Unify structure declaration for registers
Detlev Zundel [Fri, 3 Apr 2009 14:45:46 +0000 (16:45 +0200)]
include/ns16550.h: Unify structure declaration for registers

Instead of special casing the different access patterns, use common
code with light macros sprinkled in to accomodate for the different
layouts of the register structure.

Note that this also changes the types of the registers for the
"positively packed (>1)" cases.  As the registers truly are unsigned
chars, this is surely the Right Thing, but it is a semantic change.
Note that for this case depending on the endianness on the bus, we may
see a change of behaviour.

Signed-off-by: Detlev Zundel <dzu@denx.de>
15 years agoDelete now unused tools/Makefile.win32
Peter Tyser [Fri, 13 Mar 2009 23:54:52 +0000 (18:54 -0500)]
Delete now unused tools/Makefile.win32

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agoAdd support for building native win32 tools
Peter Tyser [Fri, 13 Mar 2009 23:54:51 +0000 (18:54 -0500)]
Add support for building native win32 tools

Add support for compiling the host tools in the tools directory using
the MinGW toolchain.  This produces executables which can be used on
standard Windows computers without requiring cygwin.

One must specify the MinGW compiler and strip utilities as if they
were the host toolchain in order to build win32 executables, eg:

make HOSTCC=i586-mingw32msvc-gcc HOSTSTRIP=i586-mingw32msvc-strip tools

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agotools/Makefile: Simplify HOST_CFLAGS/HOST_LDFLAGS generation
Peter Tyser [Fri, 13 Mar 2009 23:54:50 +0000 (18:54 -0500)]
tools/Makefile: Simplify HOST_CFLAGS/HOST_LDFLAGS generation

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agotools/Makefile: Create generic build rules
Peter Tyser [Fri, 13 Mar 2009 23:54:49 +0000 (18:54 -0500)]
tools/Makefile: Create generic build rules

Create a few generic build rules to replace the current method which has
1 build target for each file

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agotools/Makefile: Use auto-generated object file dependencies
Peter Tyser [Fri, 13 Mar 2009 23:54:48 +0000 (18:54 -0500)]
tools/Makefile: Use auto-generated object file dependencies

Files in the SRCS variable have their dependencies automatically
generated so remove duplicate explicit dependencies

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agotools/Makefile: Remove symlinks for remaining source files
Peter Tyser [Fri, 13 Mar 2009 23:54:47 +0000 (18:54 -0500)]
tools/Makefile: Remove symlinks for remaining source files

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agotools/Makefile: Dynamically generate libfdt object dependencies
Peter Tyser [Fri, 13 Mar 2009 23:54:46 +0000 (18:54 -0500)]
tools/Makefile: Dynamically generate libfdt object dependencies

Add the libfdt files to the SRCS variable so that they have their
dependencies automatically generated

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>