project/bcm63xx/u-boot.git
15 years ago85xx: Fix addrmap to include memory
Kumar Gala [Fri, 14 Aug 2009 21:43:22 +0000 (16:43 -0500)]
85xx: Fix addrmap to include memory

When we init the addrmap based on the TLB we will not end up getting
the TLB that covers memory if we are using SPD.  The reason is we
haven't relocated at the point that we setup the memory TLB and thus it
will not get setup in the addrmap.

Instead we can just walk over the TLB array after we've relocated and
see all the TLBs that have been set and use that information to populate
the initial addrmap.  By doing this we insure that we get the TLB
entries that cover memory.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoflash: Fix CFI buffer size bug
John Schmoller [Wed, 12 Aug 2009 15:55:47 +0000 (10:55 -0500)]
flash: Fix CFI buffer size bug

Fix bug introduced by 9c048b523413ae5f3ff34e00cf57569c3368ab51.

The cfi_flash.c driver cast the flash buffer size to a uchar in
flash_write_cfibuffer(). On some flash parts, (tested on Numonyx
part PC32F512M29EWH), the buffer size is 1KB. Remove the cast to
uchar to enable buffer sizes to be larger.

Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agotrab: rename spi_init()
Mike Frysinger [Thu, 13 Aug 2009 04:32:14 +0000 (00:32 -0400)]
trab: rename spi_init()

The local board-specific spi_init() function conflicts with the common SPI
layer, so rename it to something board-specific.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agozlib: add watchdog reset call
Giuseppe CONDORELLI [Wed, 29 Jul 2009 10:05:20 +0000 (06:05 -0400)]
zlib: add watchdog reset call

This patch adds watchdog reset call to allow its invokation during decompression
phase. This control was present on old zlib version and here it is
backported for those relevant routines. This patch is sent as a zlib separate
one beacuse it was not tested due to specific board lack.
zlib patches will be unified just in one when this will be validated through
tests.

Signed-off-by: Giuseppe Condorelli <giuseppe.condorelli@st.com>
15 years agozlib: updated to v.1.2.3
Giuseppe CONDORELLI [Wed, 29 Jul 2009 12:05:08 +0000 (08:05 -0400)]
zlib: updated to v.1.2.3

This patch updates zlib to the latest stable version.
Only relevant zlib parts were ported to u-boot tree, as already did for the
current zlib (0.95). New zlib guarantees a faster inflate performances
other then others improvements as explained at www.zlib.net.
It also includes Alessandro Rubini's patches to allow 0 as destination pointer
and to call watchdog reset if required by architecture.

Signed-off-by: Giuseppe Condorelli <giuseppe.condorelli@st.com>
Reviewed-by: Angelo Castello <angelo.castello@st.com>
Reviewed-by: Alessandro Rubini <rubini-list@gnudd.com>
15 years agodtt, lm81: move unneccessary printf into a debug printf
Heiko Schocher [Tue, 11 Aug 2009 08:37:58 +0000 (10:37 +0200)]
dtt, lm81: move unneccessary printf into a debug printf

Signed-off-by: Heiko Schocher <hs@denx.de>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc83xx
Wolfgang Denk [Tue, 11 Aug 2009 19:49:15 +0000 (21:49 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc83xx

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-video
Wolfgang Denk [Tue, 11 Aug 2009 19:47:42 +0000 (21:47 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-video

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-net
Wolfgang Denk [Tue, 11 Aug 2009 19:44:55 +0000 (21:44 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-net

15 years agoMIMC200: reduce LCD pixclock
Mark Jackson [Tue, 11 Aug 2009 10:33:47 +0000 (11:33 +0100)]
MIMC200: reduce LCD pixclock

The initial pixclock for the MIMC200 board is wrong (and causes
screen corruption due to DMA underruns).

This patch simply reduces the pixel clock to fix the problem.

Signed-off-by: Mark Jackson <mpfj@mimc.co.uk>
15 years agoUEC FIXED PHY: Determine fixed-phy port using UEC interface name.
Richard Retanubun [Wed, 1 Jul 2009 18:04:05 +0000 (14:04 -0400)]
UEC FIXED PHY: Determine fixed-phy port using UEC interface name.

Fixed a misunderstanding in the original implementation, 'devnum' that
was used in the cpu/ppc4xx/4xx_enet.c implementation was NOT the
PHY's SMI address, rather it was the number of the MAC interface on
the CPU. The equivalent of this for uec_phy will be the UEC number
stored in mii_info->dev->name. Usage example is updated for uec.

Signed-off-by: Richard Retanubun <RichardRetanubun@RuggedCom.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoAssigned a static SMI address to all UECs TBIPA address.
Richard Retanubun [Wed, 1 Jul 2009 18:03:15 +0000 (14:03 -0400)]
Assigned a static SMI address to all UECs TBIPA address.

It is set to 0x1F by default and can be overwritten on the board
header file by defining CONFIG_UTBIPAR_INIT_TBIPA. This allows
the CPU to simply "reserve" one SMI address instead of using
a different one for each UEC.

Signed-off-by: Richard Retanubun <RichardRetanubun@RuggedCom.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agonet: kirkwood: updates: used eth_setenv_enetaddr api
Prafulla Wadaskar [Mon, 10 Aug 2009 14:13:06 +0000 (19:43 +0530)]
net: kirkwood: updates: used eth_setenv_enetaddr api

eth_setenv_enetaddr is avaible by upper layer
using this saves 204 bytes on total image size

used Local OUI instead of Marvell OUI for
random MAC address generation logic

Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoFix E1000 build warning on AP1000 board
Roy Zang [Mon, 10 Aug 2009 19:48:05 +0000 (03:48 +0800)]
Fix E1000 build warning on AP1000 board

 Fix E1000 build warning on AP1000 board
 Fix the build warning on AP1000 board:
 e1000.c:131: warning: 'e1000_read_eeprom' used but never defined
 e1000.c:2012: warning: 'e1000_set_phy_mode' defined but not used

Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoARM: Davinci DM355: Enabling DM9000 on DM355 EVM
Sandeep Paulraj [Mon, 10 Aug 2009 16:24:40 +0000 (12:24 -0400)]
ARM: Davinci DM355: Enabling DM9000 on DM355 EVM

Due to recent changes to the NET support on U-boot, DM9000
is no longer detected on the DM355 EVM.
This minor update enables DM9000 on the DM355 EVM.
Tested on the DM355 EVM

Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agonet: phy: bugfixes: mv88E61xx compiler warnings fixed
Prafulla Wadaskar [Mon, 10 Aug 2009 13:53:19 +0000 (19:23 +0530)]
net: phy: bugfixes: mv88E61xx compiler warnings fixed

1. mv88E61xx driver compiler warnings fixed
2. idstr if-else statements changed to switch() construct
   and added default case too.
   This fixed idstr may be uninitialized warning

Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoarm: A320: driver for FTMAC100 ethernet controller
Po-Yu Chuang [Mon, 10 Aug 2009 03:00:00 +0000 (11:00 +0800)]
arm: A320: driver for FTMAC100 ethernet controller

This patch adds an FTMAC100 ethernet driver for Faraday A320 evaluation board.

Signed-off-by: Po-Yu Chuang <ratbert@faraday-tech.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years ago85xx: Removed BEDBUG support from FSL 85xx boards
Kumar Gala [Mon, 10 Aug 2009 21:40:55 +0000 (16:40 -0500)]
85xx: Removed BEDBUG support from FSL 85xx boards

For some reason the MPC8544 enabled BEDBUG if PCI was enabled and that
got copied int the MPC8536, MPC8572 and P2020 DS boards.  The BEDBUG
support has never been made to work completely on e500/85xx so we
just disable it to save space and match the other FSL 85xx boards.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoPrepare 2009.08-rc2
Wolfgang Denk [Mon, 10 Aug 2009 08:39:12 +0000 (10:39 +0200)]
Prepare 2009.08-rc2

Update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoMinor coding style cleanup.
Wolfgang Denk [Mon, 10 Aug 2009 08:38:34 +0000 (10:38 +0200)]
Minor coding style cleanup.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agonet/tftp.c: fix warning: pointer targets differ in signedness
Wolfgang Denk [Mon, 10 Aug 2009 07:59:10 +0000 (09:59 +0200)]
net/tftp.c: fix warning: pointer targets differ in signedness

tftp.c:294: warning: pointer targets in passing argument 1 of 'strlen'
differ in signedness

This was only visible for the utx8245 board which seems to have DEBUG
enabled.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoARM Cortex A8: Remove bogus config.mk entries
Dirk Behme [Sat, 8 Aug 2009 14:06:47 +0000 (16:06 +0200)]
ARM Cortex A8: Remove bogus config.mk entries

Remove bogus config.mk entry, fix newline and remove redundant
omap3/config.mk

Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoOMAP3: Fix missing GPMC_CONFIG_CS0_BASE
Dirk Behme [Sat, 8 Aug 2009 10:46:09 +0000 (12:46 +0200)]
OMAP3: Fix missing GPMC_CONFIG_CS0_BASE

Applying two indepenent OMAP3 patches resulted in missing
GPMC_CONFIG_CS0_BASE. Patch "omap3: embedd gpmc_cs into gpmc
config struct" removes GPMC_CONFIG_CS0_BASE, independent patch
"omap3: bug fix for NOR boot support" introduces it's usage.
Re-introduce GPMC_CONFIG_CS0_BASE.

Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agomxc-mmc: sdhc host driver for MX2 and MX3 proccessor
Ilya Yanok [Mon, 8 Jun 2009 00:12:49 +0000 (04:12 +0400)]
mxc-mmc: sdhc host driver for MX2 and MX3 proccessor

This is a port of Linux driver for SDHC host controller hardware
found on Freescale's MX2 and MX3 processors. Uses new generic MMC
framework (CONFIG_GENERIC_MMC) and it looks like there are some
problems with a framework (at least on LE cpus). Some of these
problems are addressed in the following patches.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agokirkwood/gpio.h: remove duplicate definition
Alessandro Rubini [Fri, 7 Aug 2009 10:35:47 +0000 (12:35 +0200)]
kirkwood/gpio.h: remove duplicate definition

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
15 years agoAdd driver for the ST M95xxx SPI EEPROM
Albin Tonnerre [Fri, 7 Aug 2009 10:37:36 +0000 (12:37 +0200)]
Add driver for the ST M95xxx SPI EEPROM

This chip is used in a number of boards manufactured by Calao-Systems
which should be supported soon. This driver provides the necessary
spi_read and spi_write functions necessary to communicate with the chip.

Signed-off-by: Albin Tonnerre <albin.tonnerre@free-electrons.com>
15 years agoarm: Sheevaplug: Fixed NAND specific warning
Prafulla Wadaskar [Fri, 7 Aug 2009 16:57:32 +0000 (22:27 +0530)]
arm: Sheevaplug: Fixed NAND specific warning

It is recommended to define the macro CONFIG_SYS_64BIT_VSPRINTF
for NAND specific warning removal, same is done in this patch

Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
15 years agoDual-license IBM code contributions
Josh Boyer [Fri, 7 Aug 2009 17:53:20 +0000 (13:53 -0400)]
Dual-license IBM code contributions

It was brought to our attention that U-Boot contains code derived from the
IBM OpenBIOS source code originally provided with some of the older PowerPC
4xx development boards.  As a result, the original license of this code has
been carried in the various files for a number of years in the U-Boot project.

IBM is dual-licensing the IBM code contributions already present in U-Boot
under either the terms of the GNU General Public License version 2, or the
original code license already present.

Signed-off-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
15 years agofdt: Fix fdt_pci_dma_ranges handling of 64-bit ranges
Kumar Gala [Wed, 5 Aug 2009 14:03:54 +0000 (09:03 -0500)]
fdt: Fix fdt_pci_dma_ranges handling of 64-bit ranges

If the size of a region equal to 4G it can't be represnted in a 32-bit
BAR so we should have marked that case as MEM64.

Additionally bump the number of inbound windows up to 4 to handle the
fact that Freescale PPCs that have an implicit window for CCSRBAR.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoARM EABI: add new helper functions resp. function names
Wolfgang Denk [Thu, 6 Aug 2009 19:29:59 +0000 (21:29 +0200)]
ARM EABI: add new helper functions resp. function names

The ARM EABI defines new names for GCC helper functions,
and GCC seems to need some new functions as well.

This patch is a minimal-invasive approach to fix problems with EABI
conformant tool chains (to be used with "USE_PRIVATE_LIBGCC=yes").

Signed-off-by: Wolfgang Denk <wd@denx.de>
Tested-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agohush: Fix bogus free() call
Peter Tyser [Wed, 5 Aug 2009 21:18:44 +0000 (16:18 -0500)]
hush: Fix bogus free() call

An off-by-one error in hush.c resulted in an unintentional free() call
every time a command was executed

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agodigsy_mtc: Update default environment
Detlev Zundel [Wed, 5 Aug 2009 16:37:45 +0000 (18:37 +0200)]
digsy_mtc: Update default environment

Signed-off-by: Detlev Zundel <dzu@denx.de>
15 years agodigsy_mtc: Add delay in SPI transfers to the companion controller.
Detlev Zundel [Wed, 5 Aug 2009 16:37:44 +0000 (18:37 +0200)]
digsy_mtc: Add delay in SPI transfers to the companion controller.

While at it, remove initialization of variables which will be set
before usage in all cases.

Signed-off-by: Detlev Zundel <dzu@denx.de>
15 years agodigsy_mtc: minor fixes for mtc command help
Anatolij Gustschin [Wed, 5 Aug 2009 16:37:43 +0000 (18:37 +0200)]
digsy_mtc: minor fixes for mtc command help

Add mtc state subcommand description to the
help of mtc command.

Remove some newlines in description of commands
for proper help formating.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
15 years agodigsy_mtc: Add mtc state command.
Grzegorz Bernacki [Wed, 5 Aug 2009 16:37:42 +0000 (18:37 +0200)]
digsy_mtc: Add mtc state command.

Signed-off-by: Grzegorz Bernacki <gjb@semihalf.com>
15 years agoPartition support: remove newline from partition name
Wolfgang Denk [Tue, 28 Jul 2009 20:35:39 +0000 (22:35 +0200)]
Partition support: remove newline from partition name

Remove bogus newline character that got added to the .name field of
the disk_partition_t structure.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoFix LZMA string.h header inclusion issue and remove unused variables.
Luigi 'Comio' Mantellini [Tue, 28 Jul 2009 07:33:17 +0000 (09:33 +0200)]
Fix LZMA string.h header inclusion issue and remove unused variables.

Signed-off-by: Luigi 'Comio' Mantellini <luigi.mantellini@idf-hit.com>
15 years agobus_vcxk.c: fix warning: unused variable 'lineptr'
Jens Scharsig [Mon, 27 Jul 2009 13:28:42 +0000 (15:28 +0200)]
bus_vcxk.c: fix warning: unused variable 'lineptr'

Signed-off-by: Jens Scharsig <esw@bus-elektronik.de>
15 years agoenv: kill off default_environment_size
Mike Frysinger [Fri, 24 Jul 2009 21:51:27 +0000 (17:51 -0400)]
env: kill off default_environment_size

The only environment type that uses this variable is spi flash, and that is
only because it is reimplementing the common set_default_env() function.
So fix the spi flash code and kill off the default_environment_size in the
process.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoexport SPI functions to standalone apps
Mike Frysinger [Thu, 23 Jul 2009 20:37:48 +0000 (16:37 -0400)]
export SPI functions to standalone apps

While we're here, fix the broken #ifdef handling in _exports.h.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-net
Wolfgang Denk [Sat, 8 Aug 2009 10:08:09 +0000 (12:08 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-net

15 years agoomap3: bug fix for NOR boot support
Penda Naveen Kumar [Thu, 30 Jul 2009 18:36:36 +0000 (00:06 +0530)]
omap3: bug fix for NOR boot support

This patch provides bug fix, when omap3 uses nor boot.

Signed-off-by: Penda Naveen Kumar<pnaveen@ti.com>
Acked-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoFix examples for OMAP3 boards...
Michael Evans [Mon, 13 Jul 2009 19:13:45 +0000 (20:13 +0100)]
Fix examples for OMAP3 boards...

The attached patch corrects an error in the examples/Makefile which
causes the applications in the examples directory to hang on OMAP3
based boards. The current Makefile sets -Ttext during linking to
0x0c100000 which is outside of addressable SDRAM memory. The script
corrects the existing ifeq...else...endif logic to look at the VENDOR
tag rather than the CPU tag.

The patch affects the following configs: omap3_beagle_config,
omap3_overo_config, omap3_evm_config, omap3_pandora_config,
omap3_zoom1_config and omap3_zoom2_config.

Signed-off-by: Michael Evans <horse_dung@hotmail.com>
Edited commit message.
Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoomap3: use only fixed-size types inside ctrl_structs
Dirk Behme [Sat, 8 Aug 2009 07:30:23 +0000 (09:30 +0200)]
omap3: use only fixed-size types inside ctrl_structs

replace variable types in ctrl_structs for omap3 by those with
fixed size (u8, u16, u32).
Additional ifndef-protection is needed by examples which do not
compile when including asm/types.h

Signed-off-by: Matthias Ludwig <mludwig@ultratronik.de>
Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoomap3: replace all instances of gpmc config struct by one global
Dirk Behme [Sat, 8 Aug 2009 07:30:22 +0000 (09:30 +0200)]
omap3: replace all instances of gpmc config struct by one global

Signed-off-by: Matthias Ludwig <mludwig@ultratronik.de>
Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoomap3: remove typedefs for configuration structs
Dirk Behme [Sat, 8 Aug 2009 07:30:21 +0000 (09:30 +0200)]
omap3: remove typedefs for configuration structs

Signed-off-by: Matthias Ludwig <mludwig@ultratronik.de>
Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoAdd Intel E1000 PCIE card support
Roy Zang [Fri, 31 Jul 2009 05:34:02 +0000 (13:34 +0800)]
Add Intel E1000 PCIE card support

Based on Intel PRO/1000 Network Driver 7.3.20-k2
  Add Intel E1000 PCIE card support. The following cards are added:
  INTEL_82571EB_COPPER
  INTEL_82571EB_FIBER,
  INTEL_82571EB_SERDES
  INTEL_82571EB_QUAD_COPPER
  INTEL_82571PT_QUAD_COPPER
  INTEL_82571EB_QUAD_FIBER
  INTEL_82571EB_QUAD_COPPER_LOWPROFILE
  INTEL_82571EB_SERDES_DUAL
  INTEL_82571EB_SERDES_QUAD
  INTEL_82572EI_COPPER
  INTEL_82572EI_FIBER
  INTEL_82572EI_SERDES
  INTEL_82572EI
  INTEL_82573E
  INTEL_82573E_IAMT
  INTEL_82573L
  INTEL_82546GB_QUAD_COPPER_KSP3
  INTEL_80003ES2LAN_COPPER_DPT
  INTEL_80003ES2LAN_SERDES_DPT
  INTEL_80003ES2LAN_COPPER_SPT
  INTEL_80003ES2LAN_SERDES_SPT

 82571EB_COPPER dual ports,
 82572EI single port,
 82572EI_COPPER single port PCIE cards
 and
 82545EM_COPPER,
 82541GI_LF
 pci cards are tested on both  P2020 board
 and MPC8544DS board.

Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agonet: sync env ethaddr to device enetaddr in eth_init()
Mike Frysinger [Thu, 16 Jul 2009 01:31:28 +0000 (21:31 -0400)]
net: sync env ethaddr to device enetaddr in eth_init()

In the previous enetaddr refactoring, the assumption with commit 56b555a644
was that the eth layer would handle the env -> device enetaddr syncing.
This was not the case as eth_initialize() is called only once and the sync
occurs there.  So make sure the eth_init() function does the env -> device
sync with every network init.

Reported-by: Andrzej Wolski <awolski@poczta.fm>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agominor debug cleanups in ./net
Robin Getz [Thu, 23 Jul 2009 07:01:03 +0000 (03:01 -0400)]
minor debug cleanups in ./net

 Minor ./net cleanups - no functional changes
  - change #ifdef DEBUG printf(); #endif to just debug()
  - changed __FUNCTION__ to __func__
  - got rid of extra whitespace between function and opening brace
  - removed unnecessary braces on if statements

 gcc dead code elimination should make this functionally/size equivalent
 when DEBUG is not defined. (confirmed on Blackfin, with gcc 4.3.3).

Signed-off-by: Robin Getz <rgetz@blackfin.uclinux.org>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoomap3: embedd gpmc_cs into gpmc config struct
Matthias Ludwig [Tue, 19 May 2009 07:09:31 +0000 (09:09 +0200)]
omap3: embedd gpmc_cs into gpmc config struct

Embedd chip select configuration into struct for gpmc config
instead of having it completely separated as suggested by
Wolfgang Denk on
http://lists.denx.de/pipermail/u-boot/2009-May/052247.html

Signed-off-by: Matthias Ludwig <mludwig@ultratronik.de>
15 years agoMerge branch 'next' of git://git.denx.de/u-boot-coldfire
Wolfgang Denk [Tue, 4 Aug 2009 19:54:11 +0000 (21:54 +0200)]
Merge branch 'next' of git://git.denx.de/u-boot-coldfire

15 years agorm9200 lowevel_init: don't touch reserved/readonly registers
David Brownell [Fri, 17 Jul 2009 01:40:55 +0000 (18:40 -0700)]
rm9200 lowevel_init: don't touch reserved/readonly registers

For some reason the AT91rm9200 lowlevel init writes to a bunch of
reserved or read-only addresses.  All the boards seem to define the
value-to-be-written values as zero ... but they shouldn't actually
be writing *anything* there.

No documented erratum justifies these accesses.  It looks like maybe
some pre-release BDI-2000 setup code has been carried along by cargo
cult programming since at least late 2004 (per GIT history).

Here's a patch disabling what seems to be bogosity.  Tested on a
csb337; there were no behavioral changes.

Signed-off-by: David Brownell <david-b@pacbell.net>
on RM9200ek
Tested-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agopxa: Fix typo in GCDR(x)
David Hunter [Thu, 30 Jul 2009 21:32:49 +0000 (14:32 -0700)]
pxa: Fix typo in GCDR(x)

Fix a typo in the GCDR(x) macro. It's a good thing no one was using it.

Signed-off-by: David Hunter <hunterd42@gmail.com>
15 years agoAdd AT91SAM9260 to at91's lowlevel_init.S
Eric Benard [Sat, 18 Jul 2009 21:45:15 +0000 (23:45 +0200)]
Add AT91SAM9260 to at91's lowlevel_init.S

Needed for AT91SAM9260 NOR Boot on Eukrea's CPU9260.

Signed-off-by: Eric Benard <eric@eukrea.com>
15 years agoppc4xx: Remove check for PPC460EX from CompactCenter
Dirk Eibach [Thu, 30 Jul 2009 07:36:33 +0000 (09:36 +0200)]
ppc4xx: Remove check for PPC460EX from CompactCenter

Signed-off-by: Dirk Eibach <eibach@gdsys.de>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Add support for PPC460EX/460GT rev B chip to AMCC Canyonlands
Stefan Roese [Wed, 29 Jul 2009 06:46:10 +0000 (08:46 +0200)]
ppc4xx: Add support for PPC460EX/460GT rev B chip to AMCC Canyonlands

This patch is based on a diff created by Phong Vo from AMCC.

Signed-off-by: Phong Vo <pvo@amcc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Add basic support for AMCC PPC460EX/460GT rev B chips
Stefan Roese [Wed, 29 Jul 2009 06:45:27 +0000 (08:45 +0200)]
ppc4xx: Add basic support for AMCC PPC460EX/460GT rev B chips

This patch is based on a diff created by Phong Vo from AMCC.

Signed-off-by: Phong Vo <pvo@amcc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Canyonlands-NAND-boot: Support 2 Crucial 512MByte SODIMM's
Stefan Roese [Tue, 28 Jul 2009 13:12:04 +0000 (15:12 +0200)]
ppc4xx: Canyonlands-NAND-boot: Support 2 Crucial 512MByte SODIMM's

Some Canyonlands boards are equipped with different SODIMM's. This is no
problem with the "normal" NOR booting Canyonlands U-Boot, since it
automatically detects the SODIMM's via SPD data and correctly configures
them. But the NAND booting version is different. Here we only have 4k
of image size to completely setup the hardware, including DDR2 setup.
So we need to use a fixed DDR2 setup here. This doesn't work for different
SODIMM's right now.

Currently only this Crucial SODIMM is support:
CT6464AC667.8FB (dual ranked)

Now some boards are shipped with this SODIMM:
CT6464AC667.4FE (single ranked)

This patch now supports both SODIMM's by configuring first for the dual
ranked DIMM. A quick shows, if this module is really installed. If this test
fails, the DDR2 controller is re-configured for the single
ranked SODIMM.

Tested with those SODIMM's:

CT6464AC667.8FB (dual ranked)
CT6464AC667.4FE (single ranked)

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: amcc: Move "kernel_addr_r" etc to higher locations (> 16MB)
Stefan Roese [Tue, 28 Jul 2009 08:56:03 +0000 (10:56 +0200)]
ppc4xx: amcc: Move "kernel_addr_r" etc to higher locations (> 16MB)

This patch moves the load addresses for kernel, fdt and ramdisk to higher
addresses (>= 16MB). This enables booting of bigger kernel images (e.g.
lockdep enabled).

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: amcc: Set CONFIG_SYS_BOOTMAPSZ to 16MB for big kernels
Stefan Roese [Tue, 28 Jul 2009 08:50:32 +0000 (10:50 +0200)]
ppc4xx: amcc: Set CONFIG_SYS_BOOTMAPSZ to 16MB for big kernels

This patch changes CONFIG_SYS_BOOTMAPSZ from 8MB to 16MB which is the
initial TLB on 40x PPC's in the Linux kernel. With this change even bigger
Linux kernels (> 8MB) can be booted.

This patch also sets CONFIG_SYS_BOOTM_LEN to 16MB (default 8MB) to enable
decompression of bigger images.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years ago83xx, kmeter1, fix: update in the DTS the correct size for the first flash
Heiko Schocher [Tue, 28 Jul 2009 12:53:44 +0000 (14:53 +0200)]
83xx, kmeter1, fix: update in the DTS the correct size for the first flash

When updating the "reg" in the "/localbus/flash@f0000000,0" node
size was wrong updated for the first flash, because the total
size was filled in, instead of the right size for it.

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-i2c
Wolfgang Denk [Wed, 29 Jul 2009 22:36:25 +0000 (00:36 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-i2c

15 years agoUpdate Freescale copyrights to remove "All Rights Reserved"
Kumar Gala [Wed, 29 Jul 2009 02:49:52 +0000 (21:49 -0500)]
Update Freescale copyrights to remove "All Rights Reserved"

"All Rights Reserved" conflicts with the GPL.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
15 years agoarm nomadik: add i2c
Alessandro Rubini [Fri, 24 Jul 2009 09:27:14 +0000 (11:27 +0200)]
arm nomadik: add i2c

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stericsson.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoarm nomadik: add gpio support
Alessandro Rubini [Fri, 24 Jul 2009 09:27:03 +0000 (11:27 +0200)]
arm nomadik: add gpio support

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stericsson.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoOMAP3 Remove twl4030 defines
Tom Rix [Sun, 28 Jun 2009 17:52:32 +0000 (12:52 -0500)]
OMAP3 Remove twl4030 defines

These defines have been subplanted by the equivelent defines in
include/twl4030.h

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Heiko Schocher <hs@denx.de>
15 years agoOMAP3 Move twl4030 mmc function
Tom Rix [Sun, 28 Jun 2009 17:52:31 +0000 (12:52 -0500)]
OMAP3 Move twl4030 mmc function

Because twl4030 now has its own device files, move and rename
twl4030_mmc_config.

twl4030_mmc_config initializes the twl4030 power setting to
the mmc device.  Because it is in the twl4030 power domain, move
it out of drivers/mmc/omap3_mmc.c and into drivers/power/twl4030.c.

The function was renamed to twl4030_power_mmc_init because all
the functions in this file are to have the format

twl4030_power_<device>_<action>

In this case the suffix is mmc_init so
device : mmc
action : init

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Heiko Schocher <hs@denx.de>
15 years agoOMAP3 Move twl4030 power and led functions
Tom Rix [Sun, 28 Jun 2009 17:52:30 +0000 (12:52 -0500)]
OMAP3 Move twl4030 power and led functions

Because twl4030 now has its own device files, move exiting
omap3 power_init_r to a new location.

power_init_r is the only function in board/omap3/common.
It initializes the twl4030 power for the board and enables
the led.

The power part of the the function is moved to twl4030_power_init in
drivers/power/twl4030.c The power compilation is conditional on the
existing config variable CONFIG_TWL4030_POWER.

The led part is moved to twl4030_led_init in the new file
drivers/misc/twl4030_led.c  The led compilation is conditional on
the new config variable CONFIG_TWL4030_LED

The directory board/omap3/common was removed because power_init_r
was the only function in it.

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Heiko Schocher <hs@denx.de>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Wed, 29 Jul 2009 07:25:52 +0000 (09:25 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc83xx
Wolfgang Denk [Wed, 29 Jul 2009 07:15:36 +0000 (09:15 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc83xx

15 years agoTQM8xx* boards: set larger SMC Rx buffer len
Wolfgang Denk [Tue, 28 Jul 2009 20:13:52 +0000 (22:13 +0200)]
TQM8xx* boards: set larger SMC Rx buffer len

Commit 2b3f12c2 added support for configurable SMC Rx buffer length on
8xx systems. Enable this feature on TQM8xx* based boards.

This fixes the problem that pasting text in the middle of a line
(i. e. inserting in edit mode) did not work - only the first two
characters got inserted, the rest was lost.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoFix ext2load return code
Wolfgang Denk [Tue, 28 Jul 2009 20:07:37 +0000 (22:07 +0200)]
Fix ext2load return code

Make the ext2load command return 0 on success (instead of the file
length).

Also fix output format (get rid of random newlines) and some coding
style issues (long lines etc.).

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoext2: fix inode size and calculations
Weirich, Bernhard [Wed, 10 Jun 2009 12:00:37 +0000 (14:00 +0200)]
ext2: fix inode size and calculations

Signed-off-by: unsik Kim <donari75@gmail.com>
Signed-off-by: Bernhard Weirich <bernhard.weirich@riedel.net>
Signed-off-by: Wolfgang Denk <wd@denx.de>
Tested-by: Wolfgang Denk <wd@denx.de>
15 years agoTWL4030 Add power reset button
Tom Rix [Sun, 28 Jun 2009 17:52:29 +0000 (12:52 -0500)]
TWL4030 Add power reset button

The Zoom2 power reset button is on the top right side of the
main board.  Press and hold for about to 8 seconds to completely
reset the board.

Some of the beta boards have a hardware problem that prevents
using this feature.  If is difficult to further characterize the
boards that fail.  So disable resetting for all beta boards.

The Zoom1 reset button is the red circle on the top right,
front of the board.  Press and hold the button for 8 seconds to
completely reset the board.

After analyzing beagle, it was determined that other boards
that use the twl4030 for power managment can also make use
this function.

The resetting is done by the power management part of the twl4030.
Since there is no existing drivers/power, add one.

The compilation of power/twl4030.h is controlled by the config
variable CONFIG_TWL4030_POWER

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Heiko Schocher <hs@denx.de>
15 years agoTWL4030 Add initial support
Tom Rix [Sun, 28 Jun 2009 17:52:28 +0000 (12:52 -0500)]
TWL4030 Add initial support

The TWL4030 supplies many peripherals for OMAP3 boards. These include
power management, usb and, keyboard.

The product description is found here:

http://focus.ti.com/docs/prod/folders/print/tps65950.html

Product reference document, tps65950.pdf, is found here:

http://www.ti.com/lit/gpn/tps65950

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Heiko Schocher <hs@denx.de>
15 years agoOMAP I2C Fix the sampling clock.
Tom Rix [Sun, 28 Jun 2009 17:52:27 +0000 (12:52 -0500)]
OMAP I2C Fix the sampling clock.

This problem is seen on Zoom1 and Zoom2 in the startup and
when i2c probe is used

Before :

In:    serial
Out:   serial
Err:   serial
timed out in wait_for_bb: I2C_STAT=1000
timed out in wait_for_bb: I2C_STAT=1000
timed out in wait_for_bb: I2C_STAT=1000
timed out in wait_for_pin: I2C_STAT=1000
I2C read: I/O error
timed out in wait_for_bb: I2C_STAT=1000
timed out in wait_for_bb: I2C_STAT=1000
Die ID #327c00020000000004013ddd05026013
Hit any key to stop autoboot:  0
OMAP3 Zoom1# i2c probe
Valid chip addresses:timed out in wait_for_bb: I2C_STAT=1000
 02 03 04 05 06 07 08 09 0A 0B 0C 0D <snip>

After :

In:    serial
Out:   serial
Err:   serial
Die ID #327c00020000000004013ddd05026013
Hit any key to stop autoboot:  0
OMAP3 Zoom1# i2c probe
Valid chip addresses: 48 49 4A 4B

The addresses are for the twl4030.

The prescalar that converts the function clock to the sampling
clock is hardcoded to 0.  The reference manual recommends 7
if the function clock is 96MHz.

Instead of just changing the hardcoded values, the prescalar
is calculated from the value I2C_IP_CLK.

The i2c #defines are in kHz.  The speed passed into the
i2c init routine is in Hz.  To be consistent, change the
defines to be in Hz.

The timing calculations are based on what is done in the
linux 2.6.30 kernel in drivers/i2c/buses/i2c_omap.c as
apposed to what is done in TRM.

The major variables in the timing caculations are
specified as #defines that can be overriden as required.

The variables and their defaults are

I2C_IP_CLK SYSTEM_CLOCK_96
I2C_INTERNAL_SAMPLING_CLK 19200000
I2C_FASTSPEED_SCLL_TRIM 6
I2C_FASTSPEED_SCLH_TRIM 6
I2C_HIGHSPEED_PHASE_ONE_SCLL_TRIM I2C_FASTSPEED_SCLL_TRIM
I2C_HIGHSPEED_PHASE_ONE_SCLH_TRIM I2C_FASTSPEED_SCLH_TRIM
I2C_HIGHSPEED_PHASE_TWO_SCLL_TRIM I2C_FASTSPEED_SCLL_TRIM
I2C_HIGHSPEED_PHASE_TWO_SCLH I2C_FASTSPEED_SCLH_TRIM

This was runtime verified on Zoom1, Zoom2, Beagle and Overo.
The 400kHz and 3.4M cases were verifed on test Zoom1,
Zoom2, Beagle and Overo configurations.

Testing for omap2 will be done in a second step as Nishanth
and Jean-Christophe commented.

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Heiko Schocher <hs@denx.de>
15 years agoarm, i2c: added support for the TWSI I2C Interface
Heiko Schocher [Mon, 20 Jul 2009 07:59:37 +0000 (09:59 +0200)]
arm, i2c: added support for the TWSI I2C Interface

Signed-off-by: Heiko Schocher <hs@denx.de>
15 years agoppc4xx: Fix problem with NOR range assignment in Canyonlands ft_board_setup
Stefan Roese [Mon, 27 Jul 2009 07:13:38 +0000 (09:13 +0200)]
ppc4xx: Fix problem with NOR range assignment in Canyonlands ft_board_setup

This patch fixes the problem, that the current fdt board fixup code only
set's one range, the one for NOR. By this it's overwriting the already
correctly configured values done in __ft_board_setup(). Just remove this
now unneeded NOR fixup and all the ranges are correctly defined.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Dirk Eibach <eibach@gdsys.de>
Cc: Felix Radensky <felix@embedded-sol.com>
15 years agoppc4xx: Add some NAND-booting bootstrap entries to Kilauea chip_config cmd
Stefan Roese [Mon, 27 Jul 2009 05:42:48 +0000 (07:42 +0200)]
ppc4xx: Add some NAND-booting bootstrap entries to Kilauea chip_config cmd

This patch adds some I2C bootstrap setting for NAND booting to the Kilauea
chip_config command ("533-nand" and "600-nand").

Additionally some incorrectly indented lines are fixed.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Kilauea: Fix SDRAM init in NAND booting version
Stefan Roese [Mon, 27 Jul 2009 05:42:37 +0000 (07:42 +0200)]
ppc4xx: Kilauea: Fix SDRAM init in NAND booting version

DDR2 Auto-calibration needs to be disabled on the NAND booting PPC4xx
targets. Otherwise the configured fixed init values for some DDR2
controller registers (e.g. RQDC) are not initialized at all resulting
in a non working SDRAM.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Fix Arches DDR2 initialization
Stefan Roese [Mon, 27 Jul 2009 08:53:43 +0000 (10:53 +0200)]
ppc4xx: Fix Arches DDR2 initialization

Testing on AMCC Arches with the latest U-Boot version yielded that DDR2
initialization is currently broken. U-Boot hangs upon relocation to SDRAM
or crashes with random traps. This patch fixes this problem. Arches now
uses a different WRDTR and CLKTR default setting than Canyonlands/Glacier.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Add GDsys CompactCenter board support.
Dirk Eibach [Mon, 27 Jul 2009 06:49:48 +0000 (08:49 +0200)]
ppc4xx: Add GDsys CompactCenter board support.

Board support for the Guntermann & Drunck CompactCenter and
DevCon-Center.
Based on the AMCC Canyonlands board support by Stefan Roese.

Signed-off-by: Dirk Eibach <eibach@gdsys.de>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agompc83xx: Add esd VME8349 board support
Reinhard Arlt [Sat, 25 Jul 2009 04:19:12 +0000 (06:19 +0200)]
mpc83xx: Add esd VME8349 board support

This patch adds support for the esd VME8349 board equipped with the
MPC8349. It's a VME PMC carrier board equipped with the Tundra
TSI148 VME-bridge.

Signed-off-by: Reinhard Arlt <reinhard.arlt@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agosbc8349: combine HRCW flash and u-boot image flash
Paul Gortmaker [Thu, 23 Jul 2009 21:10:55 +0000 (17:10 -0400)]
sbc8349: combine HRCW flash and u-boot image flash

Up to this point in time, the sbc8349 board was storing the u-boot
image in flash 2x.  One for the HRCW value at the beginning of
flash (0xff80_0000), and once close to the end of flash (0xfff8_0000)
for the actual image that got executed.

This moves the TEXT_BASE to be the beginning of flash, which makes
the second copy of the image redundant, and frees up the flash
from the end of the environment storage to the end of the flash
device itself.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agompc83xx: CONFIG_83XX_GENERIC_PCI is now synonymous with CONFIG_PCI; remove the former
Kim Phillips [Thu, 23 Jul 2009 19:09:38 +0000 (14:09 -0500)]
mpc83xx: CONFIG_83XX_GENERIC_PCI is now synonymous with CONFIG_PCI; remove the former

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agoPrepare 2009.08-rc1 (again, after fixing last minute issues).
Wolfgang Denk [Mon, 27 Jul 2009 08:37:37 +0000 (10:37 +0200)]
Prepare 2009.08-rc1 (again, after fixing last minute issues).

Update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agocpu/arm920t/start.S: include <common.h> to have ROUND() defined
Wolfgang Denk [Mon, 27 Jul 2009 08:06:39 +0000 (10:06 +0200)]
cpu/arm920t/start.S: include <common.h> to have ROUND() defined

Commit fcd3c87e made include/common.h usable by assembler code but
failed to update cpu/arm920t/start.S

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoat91cap9adk: fix #ifdef/#endif pairing (2nd try)
Wolfgang Denk [Mon, 27 Jul 2009 08:01:11 +0000 (10:01 +0200)]
at91cap9adk: fix #ifdef/#endif pairing (2nd try)

Commit 7024aa14 was supposed to fix the #ifdef/#endif pairing in
include/configs/at91cap9adk.h, but did not cate all problems.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoFix include/common.h for boards with CONFIG_STATUS_LED
Wolfgang Denk [Mon, 27 Jul 2009 07:58:14 +0000 (09:58 +0200)]
Fix include/common.h for boards with CONFIG_STATUS_LED

The reordering of include/common.h by commit fcd3c87e495f3c48 broke
boards with status LED support, resulting in
error: #error Status LED configuration missing
errors. Undo this reordering to avoid this issue.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoABI: fix build problems due to now needed div64 routine.
Wolfgang Denk [Mon, 27 Jul 2009 07:19:15 +0000 (09:19 +0200)]
ABI: fix build problems due to now needed div64 routine.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoPMC405DE: fix out of tree building
Wolfgang Denk [Mon, 27 Jul 2009 06:50:59 +0000 (08:50 +0200)]
PMC405DE: fix out of tree building

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoPrepare 2009.08-rc1
Wolfgang Denk [Sun, 26 Jul 2009 22:24:55 +0000 (00:24 +0200)]
Prepare 2009.08-rc1

Update CHANGELOG, minor coding style fix.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoadd WATCHDOG_RESET to allow LZMA kernel decompression on slow machines
rhabarber1848@web.de [Fri, 24 Jul 2009 06:16:30 +0000 (08:16 +0200)]
add WATCHDOG_RESET to allow LZMA kernel decompression on slow machines

Signed-off-by: rhabarber1848@web.de
15 years agoLess verbose output when loading vxworks 6.x images
Niklaus Giger [Thu, 23 Jul 2009 21:31:58 +0000 (23:31 +0200)]
Less verbose output when loading vxworks 6.x images

Loading vxWorks 5.x images resulted just into 3 or 4 lines of output.
With vxWorks 6.x and the new GCC it emits about 30 lines, which is
far too noisy in my opinion.

Signed-off-by: Niklaus Giger <niklaus.giger@member.fsf.org>
15 years agoMake include/common.h usable by assembler code
Wolfgang Denk [Thu, 23 Jul 2009 22:17:48 +0000 (00:17 +0200)]
Make include/common.h usable by assembler code

Commit 70ebf316 factored out the ROUND() macro into include/common.h,
not realizing that the primary use of this macro on AT91 systems was
in start.S where common.h was not included, and could not be included
because it contains a lot of C code which the assembler doesn't
understand.

This patch wraps such code in common.h in a "#ifndef __ASSEMBLY__"
construct, and then adds an include to cpu/arm926ejs/start.S thus
solving the problem.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoarm: add _lshrdi3.S
Heiko Schocher [Thu, 23 Jul 2009 11:27:04 +0000 (13:27 +0200)]
arm: add _lshrdi3.S

Signed-off-by: Heiko Schocher <hs@denx.de>
15 years agoMake linking against libgcc configurable
Wolfgang Denk [Thu, 23 Jul 2009 11:15:59 +0000 (13:15 +0200)]
Make linking against libgcc configurable

Many (especially ARM) tool chains seem to come with broken or
otherwise unusable (for the purposes of builing U-Boot) run-time
support libraries `libgcc.a'. By using the "USE_PRIVATE_LIBGCC"
setting we allow to use alternative libraries instead.

"USE_PRIVATE_LIBGCC" can either be set as an environment variable in
the shell, or as a command line argument when running "make", i. e.
$ make USE_PRIVATE_LIBGCC=yes
or
$ USE_PRIVATE_LIBGCC=yes
$ export USE_PRIVATE_LIBGCC
$ make

The value of "USE_PRIVATE_LIBGCC" is the name of the directory which
contains the alternative run-time support library `libgcc.a'. The
special value "yes" selects the directory $(OBJTREE)/lib_$(ARCH) .

Note that not all architectures provide an alternative `libgcc.a' in
their lib_$(ARCH) directories - so far, only ARM does.

Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Cc: Prafulla Wadaskar <prafulla@marvell.com>
cc: Stefan Roese <sr@denx.de>

15 years agoUse do_div from div64.h for vsprintf
Dirk Behme [Wed, 22 Jul 2009 15:51:56 +0000 (17:51 +0200)]
Use do_div from div64.h for vsprintf

Use do_div from div64.h for vsprintf in case of 64bit division.
For 32bit division, do_div from div64.h can't be used as it
needs a 64bit parameter.

Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
Acked-by: Stefan Roese <sr@denx.de>
CC: Simon Kagstrom <simon.kagstrom@netinsight.net>
15 years agoFix compiler warnings after loff_t change
Kyungmin Park [Mon, 20 Jul 2009 00:47:47 +0000 (09:47 +0900)]
Fix compiler warnings after loff_t change

Now 'env_addr' type is loff_t so use correct field type.

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
15 years agoRevert "ppc: Unlock cache-as-ram in a consistent manner"
Wolfgang Denk [Sun, 26 Jul 2009 21:28:02 +0000 (23:28 +0200)]
Revert "ppc: Unlock cache-as-ram in a consistent manner"

This reverts commit 982adfc610669482a32127282fe489857a92cfe3.

This patch causes problems on MPC83xx boards - flash recognition stops
working.

Signed-off-by: Wolfgang Denk <wd@denx.de>