oweals/u-boot.git
17 years agoFix where the #ifdef CFG_BOOTMAPSZ is placed.
Jerry Van Baren [Wed, 15 Aug 2007 15:13:15 +0000 (11:13 -0400)]
Fix where the #ifdef CFG_BOOTMAPSZ is placed.

Commit 073e1b509980cefe6f53c2d7fbbcd135df1e3924 "Fix initrd/dtb
interaction" put the new code outside of the #if defined(CONFIG_OF_LIBFDT)
when it should have gone inside of the conditional.  As a result, it
broke non-LIBFDT board builds.

Also added a missing "not." to the comment.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoFix malloc size error in ahci_init_one.
Ed Swarthout [Tue, 14 Aug 2007 19:06:45 +0000 (14:06 -0500)]
Fix malloc size error in ahci_init_one.

Typically this causes scsi init to corrupt the
devlist and break the coninfo command.
Fix a compiler size warning.

Signed-off-by: Jason Jin <jason.jin@freescale.com>
Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years agoTI DaVinci - fix unsupported %hhx format
ksi@koi8.net [Tue, 14 Aug 2007 17:02:16 +0000 (10:02 -0700)]
TI DaVinci - fix unsupported %hhx format

Signed-off-by: Sergey Kubushyn <ksi@koi8.net>
17 years agoMerge with /home/wd/git/u-boot/custodian/u-boot-ppc4xx
Wolfgang Denk [Tue, 14 Aug 2007 16:43:14 +0000 (18:43 +0200)]
Merge ... /home/wd/git/u-boot/custodian/u-boot-ppc4xx

17 years agoCoding style cleanup. Update CHANGELOG.
Wolfgang Denk [Tue, 14 Aug 2007 16:42:36 +0000 (18:42 +0200)]
Coding style cleanup. Update CHANGELOG.

Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoMerge with /home/wd/git/u-boot/custodian/u-boot-arm
Wolfgang Denk [Tue, 14 Aug 2007 16:03:48 +0000 (18:03 +0200)]
Merge ... /home/wd/git/u-boot/custodian/u-boot-arm

17 years agoMerge with /home/wd/git/u-boot/work
Wolfgang Denk [Tue, 14 Aug 2007 15:38:34 +0000 (17:38 +0200)]
Merge with /home/wd/git/u-boot/work

17 years agoFix initrd/dtb interaction
Andy Fleming [Tue, 14 Aug 2007 15:32:59 +0000 (10:32 -0500)]
Fix initrd/dtb interaction

The original code would wrongly relocate the blob to be right before
the initrd if it existed.  The blob *must* be within CFG_BOOTMAPSZ,
if it is defined.  So we make two changes:

1) flag the blob for relocation whenever its address is above BOOTMAPSZ

2) If the blob is being relocated, relocate it before kbd, not initrd

Signed-off-by: Andy Fleming <afleming@freescale.com>
17 years agoSupply spi interface in at45.c
Peter Pearse [Tue, 14 Aug 2007 14:40:00 +0000 (15:40 +0100)]
Supply spi interface in at45.c

17 years agoMerge with git://www.denx.de/git/u-boot.git
Stefan Roese [Tue, 14 Aug 2007 14:36:29 +0000 (16:36 +0200)]
Merge with git://www.denx.de/git/u-boot.git

17 years agoPOST: Fix merge problem
Stefan Roese [Tue, 14 Aug 2007 13:12:01 +0000 (15:12 +0200)]
POST: Fix merge problem

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoCoding style cleanup
Stefan Roese [Tue, 14 Aug 2007 13:03:17 +0000 (15:03 +0200)]
Coding style cleanup

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoMerge with /home/stefan/git/u-boot/zeus
Stefan Roese [Tue, 14 Aug 2007 13:00:42 +0000 (15:00 +0200)]
Merge with /home/stefan/git/u-boot/zeus

17 years agoppc4xx: Add initial Zeus (PPC405EP) board support
Stefan Roese [Tue, 14 Aug 2007 12:44:41 +0000 (14:44 +0200)]
ppc4xx: Add initial Zeus (PPC405EP) board support

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoPOST: Add option for external ethernet loopback test
Stefan Roese [Tue, 14 Aug 2007 12:41:55 +0000 (14:41 +0200)]
POST: Add option for external ethernet loopback test

When CFG_POST_ETHER_EXT_LOOPBACK is defined, the ethernet POST
is not done using an internal loopback connection, but by assuming
that an external loopback connector is plugged into the board.

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoPOST: Add ppc405 support to cache and UART POST
Stefan Roese [Tue, 14 Aug 2007 12:39:44 +0000 (14:39 +0200)]
POST: Add ppc405 support to cache and UART POST

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoReplace lost end of at45.c.
Peter Pearse [Tue, 14 Aug 2007 09:46:32 +0000 (10:46 +0100)]
Replace lost end of at45.c.

17 years agoUpdate Makefiles for merged and split at45.c.
Peter Pearse [Tue, 14 Aug 2007 09:30:06 +0000 (10:30 +0100)]
Update Makefiles for merged and split at45.c.

17 years agoDelete the merged files.
Peter Pearse [Tue, 14 Aug 2007 09:21:06 +0000 (10:21 +0100)]
Delete the merged files.

17 years agoAdd the files.
Peter Pearse [Tue, 14 Aug 2007 09:14:05 +0000 (10:14 +0100)]
Add the files.

17 years agoAdd MACH_TYPE records for several AT91 boards.
Peter Pearse [Tue, 14 Aug 2007 09:10:52 +0000 (10:10 +0100)]
Add MACH_TYPE records for several AT91 boards.
Merge to two at45.c files into a common file, split to at45.c and spi.c
Fix spelling error in DM9161 PHY Support.
Initialize at91rm9200 board (and set LED).
Add PIO control for at91rm9200dk LEDs and Mux.
Change dataflash partition boundaries to be compatible with Linux 2.6.

Signed-off-by: Peter Pearse <peter.pearse@arm.com>
Signed-off-by: Ulf Samuelsson <ulf@atmel.com>

17 years agoCoding style cleanup, update CHANGELOG
Wolfgang Denk [Tue, 14 Aug 2007 07:54:46 +0000 (09:54 +0200)]
Coding style cleanup, update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoMerge with /home/wd/git/u-boot/custodian/u-boot-mpc85xx
Wolfgang Denk [Tue, 14 Aug 2007 07:53:38 +0000 (09:53 +0200)]
Merge ... /home/wd/git/u-boot/custodian/u-boot-mpc85xx

17 years agoCoding style cleanup; rebuild CHANGELOG
Wolfgang Denk [Tue, 14 Aug 2007 07:47:27 +0000 (09:47 +0200)]
Coding style cleanup; rebuild CHANGELOG

17 years ago85xxCDS: Add make targets for legacy systems.
Randy Vinson [Wed, 28 Feb 2007 02:42:22 +0000 (19:42 -0700)]
85xxCDS: Add make targets for legacy systems.

The PCI ID select values on the Arcadia main board differ depending
on the version of the hardware. The standard configuration supports
Rev 3.1. The legacy target supports Rev 2.x.

Signed-off-by Randy Vinson <rvinson@mvista.com>

17 years ago85xxCDS: Enable the VIA PCI-to-ISA bridge.
Andy Fleming [Tue, 14 Aug 2007 06:50:09 +0000 (01:50 -0500)]
85xxCDS: Enable the VIA PCI-to-ISA bridge.

Author: Randy Vinson <rvinson@linuxbox.(none)>

Enable the PCI-to-ISA bridge in the VIA Southbridge located on the
Arcadia main board.

Signed-off-by: Randy Vinson <rvinson@mvista.com>
Signed-off-by: York Sun <yorksun@freescale.com>
17 years agoAdd support for UEC to 8568
Andy Fleming [Tue, 14 Aug 2007 05:14:25 +0000 (00:14 -0500)]
Add support for UEC to 8568

Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
Signed-off-by: Andy Fleming <afleming@freescale.com>
17 years agoAdd PCI support for MPC8568MDS board
Haiying Wang [Tue, 19 Jun 2007 18:18:34 +0000 (14:18 -0400)]
Add PCI support for MPC8568MDS board

This patch is against u-boot-mpc85xx.git of www.denx.com

Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
Signed-off-by: Ebony Zhu <ebony.zhu@freescale.com>
17 years agoEmpirically set cpo and clk_adjust for mpc85xx DDR2 support
Haiying Wang [Tue, 19 Jun 2007 18:18:32 +0000 (14:18 -0400)]
Empirically set cpo and clk_adjust for mpc85xx DDR2 support

This patch is against u-boot-mpc85xx.git of www.denx.com

Setting cpo to 0x9 for frequencies higher than 333MHz is verified on
both MPC8548CDS board and MPC8568MDS board, especially for supporting
533MHz DDR2.

Setting clk_adjust to 0x6(3/4 late cycle) for MPC8568MDS board is for
DDR2 on all current board versions especially ver 1.92 or later to bring
up.

Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
17 years agoUse an absolute address when jumping out of 4k boot page
Kumar Gala [Tue, 7 Aug 2007 23:07:27 +0000 (18:07 -0500)]
Use an absolute address when jumping out of 4k boot page

On e500 when we leave the 4k boot page we should use an absolute address since
we don't know where the board code may want us to be really running at.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
17 years agoMPC85xx BA bits not set for 3-bit bank address DIMM
Andy Fleming [Mon, 13 Aug 2007 19:49:59 +0000 (14:49 -0500)]
MPC85xx BA bits not set for 3-bit bank address DIMM

The current implementation does not set the number of bank address bits
(BA) in the processor. The default assumes 2 logical bank bits. This
works fine for a DIMM that uses devices with 4 internal banks (SPD
byte17 = 0x4) but needs to be set appropriately for a DIMM that uses
devices with 8 internal banks (SPD byte17 = 0x8).

Signed-off-by: Greg Davis <DavisG@embeddedplanet.com>
17 years agoFix minor 85xx warnings
Andy Fleming [Mon, 13 Aug 2007 19:38:06 +0000 (14:38 -0500)]
Fix minor 85xx warnings

Some patches had inserted warnings into the build:
* mpc8560ads declared data without using it
* cpu_init declared ecm and immap without using it in all CONFIGs
* MPC8548CDS.h had its default filenames changed so that they contained
  "\m" in the paths.  Made the defaults not Windows-specific (or
  anything-specific)

Signed-off-by: Andy Fleming <afleming@freescale.com>
17 years ago8548cds PCIE support.
Ed Swarthout [Fri, 27 Jul 2007 06:50:52 +0000 (01:50 -0500)]
8548cds PCIE support.

Make the early L1 cache stack region guarded to prevent speculative
fetches outside the locked range.

Use _PHYS defines, not _MEM for cpu-side PCI memory mapped regions.
init.S whitespace cleanup.

Allow TEXT_BASE value to be specified on command line.  This allows it
to be set to 0xfffc0000 which cuts the uboot binary in half.

Clear and enable lbc and ecm errors.

Update last_busno in device-tree for pci and pcie.

Remove load of obsolete cpu/mpc85xx/pci.0

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years ago8544ds PCIE support
Ed Swarthout [Fri, 27 Jul 2007 06:50:51 +0000 (01:50 -0500)]
8544ds PCIE support

PCI1 LAW mapping should use CFG_PCI1_MEM_PHY and not _BASE address.

Enable LBC and ECM errors and clear error registers.

Add tftpflash env var to get uboot from tftp server and flash it.

Add pci/pcie convenience env vars to display register space:
  "run pcie3regs" to see all pcie3 ccsr registers
  "run pcie3cfg" to see all cfg registers
Whitespace cleanup and MPC8544DS.h

Enable CONFIG_INTERRUPTS.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years ago85xx start.S cleanup and exception support
Andy Fleming [Tue, 14 Aug 2007 06:34:21 +0000 (01:34 -0500)]
85xx start.S cleanup and exception support

From: Ed Swarthout <Ed.Swarthout@freescale.com>

Support external interrupts from platform to eliminate system hangs.
Define CONFIG_INTERRUPTS board configure option to enable.
Enable ecm, ddr, lbc, and pci/pcie error interrupts in PIC.

Remove extra cpu initialization redundant with hardware initialization.
Whitespace cleanup.

Define and use _START_OFFSET consistent with other processors using
ppc_asm.tmpl

Move additional code from .text to boot page to make room for
exception vectors at start of image.

Handle Machine Check, External and Critical exceptions.

Fix e500 machine check error determination in traps.c

TEXT_BASE can now be 0xfffc_0000 - which cuts binary image in half.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years agoAdd MPC8544DS README
Andy Fleming [Tue, 14 Aug 2007 06:33:18 +0000 (01:33 -0500)]
Add MPC8544DS README

Signed-off-by: Andy Fleming <afleming@freescale.com>
17 years ago85xx allow debugger to configure ddr.
Ed Swarthout [Fri, 27 Jul 2007 06:50:48 +0000 (01:50 -0500)]
85xx allow debugger to configure ddr.

Only check for mpc8548 rev 1 when compiled for 8548.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years agompc85xx L2 cache reporting and SRAM relocation option.
Ed Swarthout [Fri, 27 Jul 2007 06:50:47 +0000 (01:50 -0500)]
mpc85xx L2 cache reporting and SRAM relocation option.

Allow debugger to override flash cs0/cs1 settings to enable alternate
boot regions

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years agoe500 needs ppc_asm.tmp MCK_EXCEPTION
Ed Swarthout [Fri, 27 Jul 2007 06:50:46 +0000 (01:50 -0500)]
e500 needs ppc_asm.tmp MCK_EXCEPTION

Always define MCK_EXCEPTION macro - so e500 can use it too.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years agomulticast tftp: RFC2090
David Updegraff [Mon, 11 Jun 2007 15:41:07 +0000 (10:41 -0500)]
multicast tftp: RFC2090

Implemented IETF RFC2090, Multicast TFTP.  Initial implementation
on Realtek RTL8139 and Freescale TSEC.

Signed-off-by: David Updegraff <dave@cray.com>
Signed-off-by: Ben Warren <bwarren@qstreams.com>
17 years agoNew CONFIG_BOOTP_SERVERIP option
Wilson Callan [Sat, 28 Jul 2007 14:56:13 +0000 (10:56 -0400)]
New CONFIG_BOOTP_SERVERIP option

Added CONFIG_BOOTP_SERVERIP to allow the tftp server to be different
from the bootp server

Signed-off-by: Wilson Callan <wcallan@savantav.com>
Signed-off-by: Ben Warren <bwarren@qstreams.com>
17 years agoAdd ability to take MAC address from the environment to DM9000 driver
Mike Rapoport [Sun, 12 Aug 2007 05:48:27 +0000 (08:48 +0300)]
Add ability to take MAC address from the environment to DM9000 driver

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Ben Warren <bwarren@qstreams.com>
17 years agoSync'd u-boot-net with mainline
Ben Warren [Tue, 14 Aug 2007 01:26:03 +0000 (21:26 -0400)]
Sync'd u-boot-net with mainline

Merge git://www.denx.de/git/u-boot

Conflicts:

drivers/bcm570x.c
drivers/tigon3.c

17 years agoMerge with /home/wd/git/u-boot/custodian/u-boot-mpc86xx
Wolfgang Denk [Mon, 13 Aug 2007 20:00:25 +0000 (22:00 +0200)]
Merge ... /home/wd/git/u-boot/custodian/u-boot-mpc86xx

17 years agoMinor coding style cleanup. Update CHANGELOG.
Wolfgang Denk [Mon, 13 Aug 2007 19:57:53 +0000 (21:57 +0200)]
Minor coding style cleanup. Update CHANGELOG.

Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoMerge commit 'remotes/wd/master'
Jon Loeliger [Mon, 13 Aug 2007 16:01:52 +0000 (11:01 -0500)]
Merge commit 'remotes/wd/master'

Conflicts:

MAKEALL

With any luck, this is the last MAKEALL merge conflict!

17 years agoModify SBC8641D to use new Freescale PCI routines
Joe Hamman [Sat, 11 Aug 2007 11:54:58 +0000 (06:54 -0500)]
Modify SBC8641D to use new Freescale PCI routines

PCI-Express sockets 1 and 2 verified working with Intel Pro/1000 PT
adapter.

Signed-off-by: Joe Hamman <joe.hamman@embeddedspecialties.com>
Signde-off-by: Jon Loeliger <jdl@freescale.com>
17 years agoatmel_mci: Fix data timeout value
Haavard Skinnemoen [Fri, 29 Jun 2007 16:38:51 +0000 (18:38 +0200)]
atmel_mci: Fix data timeout value

Calculate the data timeout based on values from the CSD instead of
just using a hardcoded DTOR value. This is a backport of a similar fix
in BSP 2.0, with one additional fix: the DTOCYC value is rounded up
instead of down.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
17 years agoAVR32: Include <div64.h> instead of <asm/div64.h>
Haavard Skinnemoen [Mon, 13 Aug 2007 15:22:31 +0000 (17:22 +0200)]
AVR32: Include <div64.h> instead of <asm/div64.h>

include/asm-avr32/div64.h was recently moved to include/div64.h, but
cpu/at32ap/interrupts.c wasn't properly updated (an earlier version of
the patch was merged perhaps?)

This patch updates cpu/at32ap/interrupts.c so that the avr32 port
compiles again.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
17 years agoMerge commit 'upstream/master'
Haavard Skinnemoen [Mon, 13 Aug 2007 14:34:33 +0000 (16:34 +0200)]
Merge commit 'upstream/master'

17 years agoatmel_mci: Use 512 byte blocksize if possible
Haavard Skinnemoen [Wed, 27 Jun 2007 11:34:26 +0000 (13:34 +0200)]
atmel_mci: Use 512 byte blocksize if possible

Instead of always using the largest blocksize the card supports, check
if it can support smaller block sizes and use 512 bytes if possible.
Most cards do support this, and other parts of u-boot seem to have
trouble with block sizes different from 512 bytes.

Also enable underrun/overrun protection.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Acked-by: Hans-Christian Egtvedt <hcegtvedt@atmel.com>
17 years agoppc4xx: Fix problem in PLL clock calculation
Stefan Roese [Mon, 13 Aug 2007 07:05:33 +0000 (09:05 +0200)]
ppc4xx: Fix problem in PLL clock calculation

This patch was originall provided by David Mitchell <dmitchell@amcc.com>
and fixes a bug in the PLL clock calculation.

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoUpdate CHANGELOG
Wolfgang Denk [Sun, 12 Aug 2007 19:34:50 +0000 (21:34 +0200)]
Update CHANGELOG

17 years agoMinor alignment of output, 2nd try.
Wolfgang Denk [Sun, 12 Aug 2007 19:34:34 +0000 (21:34 +0200)]
Minor alignment of output, 2nd try.
Also update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoMinor alignment of output
Wolfgang Denk [Sun, 12 Aug 2007 18:35:49 +0000 (20:35 +0200)]
Minor alignment of output

Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoCleanup output on ADS5121 board
Wolfgang Denk [Sun, 12 Aug 2007 16:28:18 +0000 (18:28 +0200)]
Cleanup output on ADS5121 board

Signed-off-by: Wolfgang Denk
17 years agoMerge with git+ssh://gemini_vpn/home/wd/git/u-boot/master
Wolfgang Denk [Sun, 12 Aug 2007 16:12:38 +0000 (18:12 +0200)]
Merge ... git+ssh://gemini_vpn/home/wd/git/u-boot/master

17 years agoAdapt board configuration and fix kernel crash on MCC200 board.
Wolfgang Denk [Sun, 12 Aug 2007 13:11:38 +0000 (15:11 +0200)]
Adapt board configuration and fix kernel crash on MCC200 board.

The update procedure was modified to turn off the USB subsystem
before exit for MCC200 and TRAB. This is necessary as otherwise the
USB controller continues to write periodically to system memory!

MCC200-specific notes:
- the patch disables the magic key check for MCC200
- the patch contains the configuration changes made
  for the new revision of the board.

Signed-off-by: Sergei Poselenov <sposelenov@emcraft.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoAdjust default configuration of ADS5121 board.
Wolfgang Denk [Sun, 12 Aug 2007 12:47:54 +0000 (14:47 +0200)]
Adjust default configuration of ADS5121 board.

Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoClean up some remaining CFG_CMD_ -> CONFIG_CMD_ issues.
Wolfgang Denk [Sun, 12 Aug 2007 12:27:39 +0000 (14:27 +0200)]
Clean up some remaining CFG_CMD_ -> CONFIG_CMD_ issues.

Signed-off-by: Wolfgang Denk <wd@denx.de>
17 years agoMerge with /home/wd/git/u-boot/custodian/u-boot-fdt
Wolfgang Denk [Sat, 11 Aug 2007 00:14:05 +0000 (02:14 +0200)]
Merge ... /home/wd/git/u-boot/custodian/u-boot-fdt

17 years agoImprove error print messages.
Gerald Van Baren [Wed, 8 Aug 2007 01:14:22 +0000 (21:14 -0400)]
Improve error print messages.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agofdt: allow for builds that don't want env and bd_t nodes
Kim Phillips [Tue, 17 Jul 2007 18:57:04 +0000 (13:57 -0500)]
fdt: allow for builds that don't want env and bd_t nodes

protect fdt_env and fdt_bd_t invocations, fix codingstyle while in the
area.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
17 years agofdt: do board setup based on fdt address specified on bootm line
Kim Phillips [Tue, 17 Jul 2007 18:56:53 +0000 (13:56 -0500)]
fdt: do board setup based on fdt address specified on bootm line

The last fdt patch to bootm did board setup based on the address
specified by a prior fdt address command invocation.  The bootm
code, as its call to fdt_chosen does, should use the fdt specified
by the user on the bootm command.  Note this restores full
functionality for the 8360's existing default boot environment
values, e.g. 'run nfsboot' (i.e. no having to 'fdt addr $fdtaddr'
before booting a kernel).

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
17 years agoCall ft_board_setup() from the bootm command.
Gerald Van Baren [Wed, 11 Jul 2007 00:40:39 +0000 (20:40 -0400)]
Call ft_board_setup() from the bootm command.

In the patch titled "Create new fdt boardsetup command..." I removed the
call to ft_board_setup() from the routine fdt_chosen(), but I forgot
to add a direct call back into cmd_bootm.c

This fixes the oversight by adding the direct call to the bootm command.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoCreate new fdt boardsetup command, fix bug parsing [] form of set values.
Gerald Van Baren [Tue, 26 Jun 2007 03:25:28 +0000 (23:25 -0400)]
Create new fdt boardsetup command, fix bug parsing [] form of set values.

Previously ft_board_setup() was called by fdt_chosen() which was not
really correctly structured.  This splits ft_board_setup() out by creating
a new fdt boardsetup command.

Fix a bug when parsing fdt set command values which have the square
bracket form [00 11 22 33] - the length was updated incorrectly in when
parsing that form.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoTighten up the error messages.
Gerald Van Baren [Tue, 26 Jun 2007 00:55:58 +0000 (20:55 -0400)]
Tighten up the error messages.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoAsthetic improvements: error messages and line lengths.
Gerald Van Baren [Mon, 25 Jun 2007 23:52:23 +0000 (19:52 -0400)]
Asthetic improvements: error messages and line lengths.

Tighten up the error messages, split overlength lines.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoFix fdt_chosen() to call ft_board_setup(), clean up long lines.
Gerald Van Baren [Sat, 26 May 2007 02:08:57 +0000 (22:08 -0400)]
Fix fdt_chosen() to call ft_board_setup(), clean up long lines.

The fdt_chosen() function was adding/seting some properties ad-hoc
  improperly and duplicated (poorly) what was done in ft_board_setup()

Clean up long lines (setting properties, printing errors).

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoFor fdt_find_node_by_path(), handle the root path properly.
Gerald Van Baren [Tue, 22 May 2007 03:27:16 +0000 (23:27 -0400)]
For fdt_find_node_by_path(), handle the root path properly.

Also removes the special case root path detection in cmd_fdt.c since it
is no longer necessary.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoAdd fdt_find_node_by_type() and fdt_find_compatible_node() to LIBFDT
Gerald Van Baren [Fri, 18 May 2007 03:54:36 +0000 (23:54 -0400)]
Add fdt_find_node_by_type() and fdt_find_compatible_node() to LIBFDT

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
Acked-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoReplace fdt_node_offset() with fdt_find_node_by_path().
Gerald Van Baren [Thu, 7 Jun 2007 02:47:58 +0000 (22:47 -0400)]
Replace fdt_node_offset() with fdt_find_node_by_path().

The new name matches more closely the kernel's name, which is also
a much better description.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
Acked-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoFix cmd_fdt line lengths, refactor code.
Gerald Van Baren [Thu, 17 May 2007 02:39:59 +0000 (22:39 -0400)]
Fix cmd_fdt line lengths, refactor code.

Break lines that were greater than 80 characters in length.
Move the fdt print and property parsing code to separate static functions
  to reduce coding clutter in the fdt_cmd handling body.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoFDT command improvements.
Gerald Van Baren [Sat, 12 May 2007 13:47:25 +0000 (09:47 -0400)]
FDT command improvements.

Fix "fdt set" so that it will create a non-existing property.
Add "fdt mknode" to create nodes.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoReorganize and fix problems (returns) in the bootm command.
Gerald Van Baren [Sat, 12 May 2007 13:45:46 +0000 (09:45 -0400)]
Reorganize and fix problems (returns) in the bootm command.

Do *NOT* return after the "point of no return" has been passed.
  If something goes wrong, the board must be reset after that point.
Move the "Transferring control to Linux" debug message back to where it
  belongs: just before transferring control to linux.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoFix bugs in the CONFIG_OF_LIBFDT
Gerald Van Baren [Wed, 9 May 2007 01:27:35 +0000 (21:27 -0400)]
Fix bugs in the CONFIG_OF_LIBFDT

Stupid coding mistakes (identified by Timur Tabi, thanks).

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoImprove fdt move length handling.
Gerald Van Baren [Thu, 26 Apr 2007 02:47:15 +0000 (22:47 -0400)]
Improve fdt move length handling.

Make the length parameter optional: if not specified, do the move using
the current size unchanged.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoMinor code clean up.
Gerald Van Baren [Thu, 26 Apr 2007 02:23:36 +0000 (22:23 -0400)]
Minor code clean up.

Declare the variable fdt properly as extern.
Call the "set_fn" function pointer the "short way" without the full
  dereferencing syntax.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoImprove error messages, more informative.
Gerald Van Baren [Thu, 26 Apr 2007 01:24:27 +0000 (21:24 -0400)]
Improve error messages, more informative.

Print more than the raw libfdt error message strings.  This is especially
useful for cluing in the user when the bootm command aborts due to
blob problems.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agolibfdt: Conditionally compile based on CONFIG_OF_LIBFDT
Gerald Van Baren [Sat, 21 Apr 2007 02:46:53 +0000 (22:46 -0400)]
libfdt: Conditionally compile based on CONFIG_OF_LIBFDT

This is the way u-boot reduces configured-out code.  At Wolfgang
Grandegger and Wolfgang Denk's request, make libfdt conform.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
17 years agoMerge git://www.denx.de/git/u-boot
Gerald Van Baren [Fri, 10 Aug 2007 23:19:50 +0000 (19:19 -0400)]
Merge git://www.denx.de/git/u-boot

17 years agoadd image size and descriptors for Spartan 3E FPGA chips
Bruce Adler [Fri, 10 Aug 2007 21:54:47 +0000 (14:54 -0700)]
add image size and descriptors for Spartan 3E FPGA chips

Spartan 3E image sizes taken from Table 1-4 in Xilinx UG332 (v1.1)

Signed-off by: Bruce Adler <bruce.adler@ccpu.com>

17 years agomake MAKEALL more immune to merge conflicts
Kim Phillips [Fri, 10 Aug 2007 20:34:48 +0000 (15:34 -0500)]
make MAKEALL more immune to merge conflicts

..by placing board entries one per line, as suggested by jdl.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
17 years agoREADME: Remove outdated cpu type, board type, and NAME_config lists
Kim Phillips [Fri, 10 Aug 2007 18:28:25 +0000 (13:28 -0500)]
README: Remove outdated cpu type, board type, and NAME_config lists

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
17 years agompc83xx: Suppress the warning 'burstlen'
Dave Liu [Fri, 10 Aug 2007 07:48:59 +0000 (15:48 +0800)]
mpc83xx: Suppress the warning 'burstlen'

suppress the warning 'burstlen' of spd_sdram.

Signed-off-by: Dave Liu <daveliu@freescale.com>
17 years agoAdd support for SBC8641D. Config files.
Joe Hamman [Thu, 9 Aug 2007 20:11:03 +0000 (15:11 -0500)]
Add support for SBC8641D. Config files.

Add support for Wind River's SBC8641D reference board.

Signed-off by: Joe Hamman <joe.hamman@embeddedspecialties.com>
Acked-by: Wolfgang Denk <wd@denx.de>
Acked-by: Jon Loeliger <jdl@freescale.com>
17 years agoAdd support for SBC8641D. Board files.
Joe Hamman [Thu, 9 Aug 2007 20:10:53 +0000 (15:10 -0500)]
Add support for SBC8641D.  Board files.

Add support for Wind River's SBC8641D reference board.

Signed-off by: Joe Hamman <joe.hamman@embeddedspecialties.com>
Acked-by: Wolfgang Denk <wd@denx.de>
Acked-by: Jon Loeliger <jdl@freescale.com>
17 years agoMerge with git+ssh://sr@pollux.denx.org/home/wd/git/u-boot/from_stefan
Stefan Roese [Fri, 10 Aug 2007 18:38:35 +0000 (20:38 +0200)]
Merge ... git+ssh://sr@pollux.denx.org/home/wd/git/u-boot/from_stefan

17 years agoConding style cleanup
Stefan Roese [Fri, 10 Aug 2007 18:34:58 +0000 (20:34 +0200)]
Conding style cleanup

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoMerge with git://www.denx.de/git/u-boot.git
Stefan Roese [Fri, 10 Aug 2007 18:33:06 +0000 (20:33 +0200)]
Merge with git://www.denx.de/git/u-boot.git

17 years ago[ARM] TI DaVinci support, hopefully final
Sergey Kubushyn [Fri, 10 Aug 2007 18:26:18 +0000 (20:26 +0200)]
[ARM] TI DaVinci support, hopefully final

Add support for the following DaVinci boards:
- DV_EVM
- SCHMOOGIE
- SONATA

Changes:

- Split into separate board directories
- Removed changes to MTD_DEBUG (or whatever it's called)
- New CONFIG_CMD party line followed
- Some cosmetic fixes, cleanup etc.
- Patches against the latest U-Boot tree as of now.
- Fixed CONFIG_CMD_NET in net files.
- Fixed CONFIG_CMD_EEPROM for schmoogie.
- Made sure it compiles and works (forceenv() link problem) on SCHMOOGIE and
   DV_EVM. Can't check if it works on SONATA, don't have a board any more,
   but it at least compiles.

Here is an excerpt from session log on SCHMOOGIE...

U-Boot 1.2.0-g6c33c785-dirty (Aug  7 2007 - 13:07:17)

DRAM:  128 MB
NAND:  128 MiB
In:    serial
Out:   serial
Err:   serial
ARM Clock : 297MHz
DDR Clock : 162MHz
ETH PHY   : DP83848 @ 0x01
U-Boot > iprobe
Valid chip addresses: 1B 38 3A 3D 3F 50 5D 6F
U-Boot > ping 192.168.253.10
host 192.168.253.10 is alive
U-Boot >

Signed-off-by: Sergey Kubushyn <ksi@koi8.net>
Acked-by: Dirk Behme <dirk.behme@gmail.com>
Acked-by: Zach Sadecki <Zach.Sadecki@ripcode.com>
Acked-by: Stefan Roese <sr@denx.de>
17 years agofsl_pci_init cleanup.
Ed Swarthout [Fri, 27 Jul 2007 06:50:45 +0000 (01:50 -0500)]
fsl_pci_init cleanup.

Do not enable normal errors created during probe (master abort, perr,
and pcie Invalid Configuration access).

Add CONFIG_PCI_NOSCAN board option to prevent bus scan.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years agopciauto_setup_device bars_num fix
Ed Swarthout [Fri, 27 Jul 2007 06:50:44 +0000 (01:50 -0500)]
pciauto_setup_device bars_num fix

Passing bars_num=0 to pciauto_setup_device should assign no bars.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Shinya Kuribayashi <shinya.kuribayashi@necel.com>
Acked-by: Andy Fleming <afleming@freescale.com>
17 years ago8641hpcn: Do correct sized pointer math.
Jon Loeliger [Mon, 6 Aug 2007 22:39:44 +0000 (17:39 -0500)]
8641hpcn: Do correct sized pointer math.

When I rebased Ed's patch and cleaned up a few compilation
problems, I apparently rebased my brain on crack first.
Fix that by doing (char *) sized pointer math as needed.

Signed-off-by: Jon Loeliger <jdl@freescale.com>
17 years agocpu/86xx fixes.
Jon Loeliger [Thu, 2 Aug 2007 19:42:20 +0000 (14:42 -0500)]
cpu/86xx fixes.

Remove rev 1 fixes.
Always set PICGCR_MODE.
Enable machine check and provide board config option
to set and handle SoC error interrupts.

Include MSSSR0 in error message.

Isolate a RAMBOOT bit of code with #ifdef CFG_RAMBOOT.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Signed-off-by: Jon Loeliger <jdl@freescale.com>
17 years agoCoding style cleanup
Stefan Roese [Fri, 10 Aug 2007 08:42:25 +0000 (10:42 +0200)]
Coding style cleanup

Signed-off-by: Stefan Roese <sr@denx.de>
17 years agoMerge with git+ssh://atlas/home/wd/git/u-boot/master
Wolfgang Denk [Fri, 10 Aug 2007 08:35:10 +0000 (10:35 +0200)]
Merge ... git+ssh://atlas/home/wd/git/u-boot/master

17 years agoMerge with /home/wd/git/u-boot/work
Wolfgang Denk [Fri, 10 Aug 2007 08:34:10 +0000 (10:34 +0200)]
Merge with /home/wd/git/u-boot/work

17 years agoMake use of generic 64bit division in nand_util.c
Dirk Behme [Thu, 2 Aug 2007 15:42:08 +0000 (17:42 +0200)]
Make use of generic 64bit division in nand_util.c

Use generic 64bit division in nand_util.c. This makes nand_util.c
independent of any toolchain 64bit division.

Signed-off-by: Dirk Behme <dirk.behme@gmail.com>
17 years agoMove 64bit division from avr32 to generic lib
Dirk Behme [Thu, 2 Aug 2007 15:41:14 +0000 (17:41 +0200)]
Move 64bit division from avr32 to generic lib

Move the 64bit division from lib_avr32 to lib_generic. With this, all
boards can do_div/__div64_32 if needed, not only avr one. Code is put
to lib_generic, so no larger memory footprint if not used. No code
modifications. Thanks for proposal by HÃ¥vard Skinnemoen.

Signed-off-by: Dirk Behme <dirk.behme@gmail.com>