kernel/u-boot.git
15 years agoMakefile: fix dependency problem with parallel builds
Wolfgang Denk [Tue, 16 Dec 2008 13:41:02 +0000 (14:41 +0100)]
Makefile: fix dependency problem with parallel builds

Parallel builds (using "make -jN") would occasionally fail with error
messages like
include/autoconf.mk:212: *** missing separator.  Stop.
Line numbers and affected boards were changing. Obviously some
Makefiles included autoconf.mk while it was still being written to.
As a fix, we now write to a temporary file first and then rename it,
so that it is really ready to use as soon as it appears.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoCoding style cleanup, update CHANGELOG.
Wolfgang Denk [Tue, 16 Dec 2008 00:02:17 +0000 (01:02 +0100)]
Coding style cleanup, update CHANGELOG.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agospi/stmicro: fix debug() display of cmd
Mike Frysinger [Thu, 11 Dec 2008 23:39:08 +0000 (18:39 -0500)]
spi/stmicro: fix debug() display of cmd

The stmicro_wait_ready() func tries to show the actual opcode that was sent
to the device, but instead it displays the array pointer.  Fix it to pull
out the opcode from the start of the array.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoenv_sf: support embedded environments
Mike Frysinger [Thu, 11 Dec 2008 11:23:37 +0000 (06:23 -0500)]
env_sf: support embedded environments

If both CONFIG_ENV_SECT_SIZE and CONFIG_ENV_SIZE are defined, and the sect
size is larger than the env size, then it means the env is embedded in a
block.  So we have to save/restore the part of the sector which is not the
environment.  Previously, saving the environment in SPI flash in this
setup would probably brick the board as the rest of the sector tends to
contain actual U-Boot data/code.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
15 years agoi2c: merge all i2c_reg_read() and i2c_reg_write() into inline functions
Timur Tabi [Wed, 3 Dec 2008 17:28:30 +0000 (11:28 -0600)]
i2c: merge all i2c_reg_read() and i2c_reg_write() into inline functions

All implementations of the functions i2c_reg_read() and
i2c_reg_write() are identical. We can save space and simplify the
code by converting these functions into inlines and putting them in
i2c.h.

Signed-off-by: Timur Tabi <timur@freescale.com>
Acked-By: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agolib_ppc: rework the flush_cache
Dave Liu [Fri, 5 Dec 2008 07:36:14 +0000 (15:36 +0800)]
lib_ppc: rework the flush_cache

- It is possible to miss flush/invalidate the last
  cache line, we fix it at here.
- add the volatile and memory clobber.

They are pointed by Scott Wood.

Signed-off-by: Dave Liu <daveliu@freescale.com>
15 years agoIntroduce addr_map library
Kumar Gala [Sat, 13 Dec 2008 23:20:28 +0000 (17:20 -0600)]
Introduce addr_map library

Add a library that helps in translating between virtual and physical
addresses.  This library can be useful as a simple means to implement
map_physmem() and virt_to_phys() for platforms that need functionality
beyond the simple 1:1 mapping.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoIntroduce virt_to_phys()
Kumar Gala [Sat, 13 Dec 2008 23:20:27 +0000 (17:20 -0600)]
Introduce virt_to_phys()

virt_to_phys() returns the physical address given a virtual. In most
cases this will be just the input value as the vast majority of
systems run in a 1:1 mode.

However in systems that are not running this way it should report the
physical address or ~0 if no mapping exists for the given virtual
address.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agoPOST Make: fix the sub-dir dependencies missing.
Yuri Tikhonov [Sun, 7 Dec 2008 21:12:50 +0000 (22:12 +0100)]
POST Make: fix the sub-dir dependencies missing.

Signed-off-by: Yuri Tikhonov <yur@emcraft.com>
15 years agoFix a typo in fw_env.config
Martin Michlmayr [Wed, 6 Aug 2008 11:44:05 +0000 (14:44 +0300)]
Fix a typo in fw_env.config

Reported-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoRemove unused CONFIG_ADDR_STREAMING defines
Peter Tyser [Mon, 1 Dec 2008 22:22:45 +0000 (16:22 -0600)]
Remove unused CONFIG_ADDR_STREAMING defines

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agocmd_mem: Remove unused variable
Peter Tyser [Mon, 24 Nov 2008 17:54:47 +0000 (11:54 -0600)]
cmd_mem: Remove unused variable

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agoFix new found CFG_
Jean-Christophe PLAGNIOL-VILLARD [Sun, 14 Dec 2008 09:29:39 +0000 (10:29 +0100)]
Fix new found CFG_

Also fix some minor typos.

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoRemove compiler warning: target CPU does not support interworking
Sergei Poselenov [Fri, 19 Sep 2008 10:07:34 +0000 (12:07 +0200)]
Remove compiler warning: target CPU does not support interworking

This warning is issued by modern ARM-EABI GCC on non-thumb targets.

Signed-off-by: Vladimir Panfilov <pvr@emcraft.com>
Signed-off-by: Sergei Poselenov <sposelenov@emcraft.com>
15 years agoFix FIT and FDT support to have CONFIG_OF_LIBFDT and CONFIG_FIT independent
Jean-Christophe PLAGNIOL-VILLARD [Mon, 24 Nov 2008 12:33:51 +0000 (13:33 +0100)]
Fix FIT and FDT support to have CONFIG_OF_LIBFDT and CONFIG_FIT independent

FDT support is used for both FIT style images and for architectures
that can pass a fdt blob to an OS (ppc, m68k, sparc).

For other architectures and boards which do not pass a fdt blob to an
OS but want to use the new uImage format, we just need FIT support.

Now we can have the 4 following configurations :

1) FIT only             CONFIG_FIT
2) fdt blob only        CONFIG_OF_LIBFDT
3) both                 CONFIG_OF_LIBFDT & CONFIG_FIT
4) none                 none

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Fri, 12 Dec 2008 23:36:25 +0000 (00:36 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-mips
Wolfgang Denk [Fri, 12 Dec 2008 23:34:12 +0000 (00:34 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-mips

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-sh
Wolfgang Denk [Fri, 12 Dec 2008 23:26:19 +0000 (00:26 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-sh

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-microblaze
Wolfgang Denk [Fri, 12 Dec 2008 23:25:24 +0000 (00:25 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-microblaze

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-ubi
Wolfgang Denk [Fri, 12 Dec 2008 23:20:01 +0000 (00:20 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-ubi

15 years agoppc4xx: Disable EEPROM write access on PMC440 boards
Matthias Fuchs [Wed, 10 Dec 2008 14:13:32 +0000 (15:13 +0100)]
ppc4xx: Disable EEPROM write access on PMC440 boards

This patch disables EEPROM wrtie access by default on PMC440 board.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
15 years agoppc4xx: Fix Ethernet PHY LED configuration on PMC440 boards
Matthias Fuchs [Wed, 10 Dec 2008 14:12:56 +0000 (15:12 +0100)]
ppc4xx: Fix Ethernet PHY LED configuration on PMC440 boards

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
15 years agoMIPS: Flush data cache upon relocation
Stefan Roese [Tue, 18 Nov 2008 15:36:12 +0000 (16:36 +0100)]
MIPS: Flush data cache upon relocation

This patch now adds a flush to the data cache upon relocation. The
current implementation is missing this. Only a comment states that it
should be done. So let's really do it now.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
15 years agoMIPS: Add CONFIG_SKIP_LOWLEVEL_INIT
Stefan Roese [Tue, 18 Nov 2008 15:36:22 +0000 (16:36 +0100)]
MIPS: Add CONFIG_SKIP_LOWLEVEL_INIT

This patch adds the CONFIG_SKIP_LOWLEVEL_INIT option to start.S. This
enables support for boards where the lowlevel initialization is
already done when U-Boot runs (e.g. via OnChip ROM).

This will be used in the upcoming VCTH board support.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
15 years agoMIPS: Add board_early_init_f() to init_sequence
Stefan Roese [Wed, 12 Nov 2008 12:18:02 +0000 (13:18 +0100)]
MIPS: Add board_early_init_f() to init_sequence

This patch adds the board_early_init_f() call to the MIPS init
sequence. A weak dummy implementation is also added which can be
overridden by a board specific version.

This will be used by the upcoming VCTH board support.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
15 years agoMIPS: Add onenand_init() to board.c and move nand_init()
Stefan Roese [Wed, 12 Nov 2008 12:18:19 +0000 (13:18 +0100)]
MIPS: Add onenand_init() to board.c and move nand_init()

This patch adds a call to onenand_init() for OneNAND support and moves
the nand_init() call to an earlier place, so that the environment can
be used from NAND and OneNAND.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
15 years agosh: Update sh2/sh2a timer
Nobuhiro Iwamatsu [Tue, 9 Dec 2008 02:32:46 +0000 (11:32 +0900)]
sh: Update sh2/sh2a timer

Renesas SH2/SH2A timer broken.
This patch fix timer function.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: r2dplus fix register access
Jean-Christophe PLAGNIOL-VILLARD [Fri, 5 Dec 2008 06:27:37 +0000 (07:27 +0100)]
sh: r2dplus fix register access

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: r2dplus/lowlevel_init: coding style fix
Jean-Christophe PLAGNIOL-VILLARD [Tue, 2 Dec 2008 06:40:03 +0000 (07:40 +0100)]
sh: r2dplus/lowlevel_init: coding style fix

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: Changed value of CACHE_OC_NUM_ENTRIES and CACHE_OC_WAY_SHIFT
Nobuhiro Iwamatsu [Tue, 25 Nov 2008 02:05:19 +0000 (11:05 +0900)]
sh: Changed value of CACHE_OC_NUM_ENTRIES and CACHE_OC_WAY_SHIFT

SH4 is different a value of CACHE_OC_NUM_ENTRIES and
CACHE_OC_WAY_SHIFT every CPU.
This patch corrects these values.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: Update sh timer function
Nobuhiro Iwamatsu [Thu, 20 Nov 2008 07:44:42 +0000 (16:44 +0900)]
sh: Update sh timer function

Change to write/readX function and fix timer problem.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: Migo-R: Update BSC value
Nobuhiro Iwamatsu [Tue, 4 Nov 2008 02:58:58 +0000 (11:58 +0900)]
sh: Migo-R: Update BSC value

A value of BSC CS4 was wrong, Fixed it.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: Update ms7722se board config
Nobuhiro Iwamatsu [Mon, 17 Nov 2008 07:52:09 +0000 (16:52 +0900)]
sh: Update ms7722se board config

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: Update SuperH serial driver
Nobuhiro Iwamatsu [Mon, 17 Nov 2008 07:53:09 +0000 (16:53 +0900)]
sh: Update SuperH serial driver

The address of SCFSR register is wrong at SH7720/SH7721.
This patch fix this.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agosh: fix rsk7203 and MigoR out of tree build
Jean-Christophe PLAGNIOL-VILLARD [Tue, 11 Nov 2008 21:20:15 +0000 (22:20 +0100)]
sh: fix rsk7203 and MigoR out of tree build

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
15 years agoppc4xx: Update TEXT_BASE for CPCI405 boards
Matthias Fuchs [Wed, 10 Dec 2008 13:41:25 +0000 (14:41 +0100)]
ppc4xx: Update TEXT_BASE for CPCI405 boards

This patch fixes building U-Boot for CPCI405 boards.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Remove some features from ALPR to fit into 256k again
Stefan Roese [Tue, 9 Dec 2008 19:08:01 +0000 (20:08 +0100)]
ppc4xx: Remove some features from ALPR to fit into 256k again

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoUBI: Set ubi_dev.type back to DEV_TYPE_NONE upon failing initialization
Stefan Roese [Wed, 10 Dec 2008 09:32:59 +0000 (10:32 +0100)]
UBI: Set ubi_dev.type back to DEV_TYPE_NONE upon failing initialization

With this patch we set the type back to NONE upon failing UBI partition
initialization. Otherwise further calls to the UBI subsystem would try
to really access the non-existing UBI partition.

Thanks to Michael Lawnick for pointing this out.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoUBI: Return -ENOMEM upon failing malloc
Stefan Roese [Wed, 10 Dec 2008 09:28:33 +0000 (10:28 +0100)]
UBI: Return -ENOMEM upon failing malloc

Return with correct error code (-ENOMEM) from ubi_attach_mtd_dev() upon
failing malloc().

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoFix compile error in building MBX860T.
Ben Warren [Wed, 10 Dec 2008 07:34:15 +0000 (23:34 -0800)]
Fix compile error in building MBX860T.

Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agomicroblaze: Remove XUPV2P board
Michal Simek [Tue, 25 Nov 2008 10:42:20 +0000 (11:42 +0100)]
microblaze: Remove XUPV2P board

---

Microblaze platforms use generic settings and to have
many platforms is confusing that's why I decided to remove this
platform from U-BOOT. ml401 tree is sufficient for covering
all Microblaze platforms.

This change will go through microblaze custodian tree.

15 years agomicroblaze: Remove CONFIG_LIBFDT due to error in common files
Michal Simek [Mon, 24 Nov 2008 17:25:41 +0000 (18:25 +0100)]
microblaze: Remove CONFIG_LIBFDT due to error in common files

15 years agomicroblaze: Fix ml401 uart16550 setting
Michal Simek [Mon, 24 Nov 2008 10:43:00 +0000 (11:43 +0100)]
microblaze: Fix ml401 uart16550 setting

Signed-off-by: Michal Simek <monstr@monstr.eu>
15 years agomicroblaze: Set up relocation is done
Michal Simek [Mon, 24 Nov 2008 10:38:22 +0000 (11:38 +0100)]
microblaze: Set up relocation is done

15 years agotools/netconsole: new script for working with netconsole over UDP
Mike Frysinger [Wed, 10 Dec 2008 04:20:31 +0000 (23:20 -0500)]
tools/netconsole: new script for working with netconsole over UDP

While the doc/README.NetConsole does have a snippet for people to
create their own netcat script, it's a lot easier to make a simple
dedicated script and tell people to use it.

Also spruce it up a bit to make it user friendly.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agofs/fat: handle FAT on SATA
Sonic Zhang [Wed, 10 Dec 2008 04:20:18 +0000 (23:20 -0500)]
fs/fat: handle FAT on SATA

The FAT file system driver should also handle FAT on SATA devices.

Signed-off-by: Sonic Zhang <Sonic.Zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agolibfdt: Fix redefined uintptr_t warning for USE_HOSTCC
Jerry Van Baren [Mon, 24 Nov 2008 13:15:02 +0000 (08:15 -0500)]
libfdt: Fix redefined uintptr_t warning for USE_HOSTCC

Compiling U-Boot in an old OS environment (RedHat-7.3  :-)  gives the
following warnings from FDT:

include/libfdt_env.h:50: warning: redefinition of 'uintptr_t'
/usr/include/stdint.h:129: warning: 'uintptr_t' previously declared here

Fix: Protect the definition of uintptr_t when compiling on the host
system.

Signed-off-by: Gerald Van Baren <vanbaren@cideas.com>
15 years agoMoved sc520 PCI definitions to stand-alone file
Graeme Russ [Fri, 21 Nov 2008 21:43:29 +0000 (08:43 +1100)]
Moved sc520 PCI definitions to stand-alone file

Signed Off By: Graeme Russ <graeme.russ@gmail.com>

15 years agoFixed path to sc520 SSI include file
Graeme Russ [Fri, 21 Nov 2008 21:43:21 +0000 (08:43 +1100)]
Fixed path to sc520 SSI include file

Signed Off By: Graeme Russ <graeme.russ@gmail.com>

15 years agoFixed build error due to #define of _LINUX_STRING_H_ in 82559_eeprom.c
Graeme Russ [Thu, 20 Nov 2008 19:28:05 +0000 (06:28 +1100)]
Fixed build error due to #define of _LINUX_STRING_H_ in 82559_eeprom.c

Signed-off-by: Graeme Russ <graeme.russ@gmail.com>
15 years agoserial: Add vcth UART driver
Stefan Roese [Wed, 12 Nov 2008 12:30:10 +0000 (13:30 +0100)]
serial: Add vcth UART driver

This patch adds the UART driver for the upcoming VCTH board support.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agojffs2: cache data_crc results
Ilya Yanok [Thu, 13 Nov 2008 16:49:36 +0000 (19:49 +0300)]
jffs2: cache data_crc results

As we moved data_crc() invocation from jffs2_1pass_build_lists() to
jffs2_1pass_read_inode() data_crc is going to be calculated on each
inode access. This patch adds caching of data_crc() results. There
is no significant improvement in speed (because of flash access
caching added in previous patch I think, crc in RAM is really fast)
but this patch impacts memory usage -- every b_node structure uses
12 bytes instead of 8.

Signed-off-by: Alexey Neyman <avn@emcraft.com>
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agojffs2: summary support
Ilya Yanok [Thu, 13 Nov 2008 16:49:35 +0000 (19:49 +0300)]
jffs2: summary support

This patch adds support for reading fs information from summary
node instead of scanning full eraseblock.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agojffs2: add buffer to cache flash accesses
Ilya Yanok [Thu, 13 Nov 2008 16:49:34 +0000 (19:49 +0300)]
jffs2: add buffer to cache flash accesses

With this patch JFFS2 code allocates memory buffer of max_totlen size
(size of the largest node, calculated during scan time) and uses it to
store entire node. Speeds up loading. If malloc fails we use old ways
to do things.

Signed-off-by: Alexey Neyman <avn@emcraft.com>
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agojffs2: rewrite jffs2 scanning code based on Linux one
Ilya Yanok [Thu, 13 Nov 2008 16:49:33 +0000 (19:49 +0300)]
jffs2: rewrite jffs2 scanning code based on Linux one

Rewrites jffs2_1pass_build_lists() function in style of Linux's
jffs2_scan_medium() and jffs2_scan_eraseblock().
This includes:
 - Caching flash acceses
 - Smart dealing with free space

Signed-off-by: Alexey Neyman <avn@emcraft.com>
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agojffs2: add sector_size field to part_info structure
Ilya Yanok [Thu, 13 Nov 2008 16:49:32 +0000 (19:49 +0300)]
jffs2: add sector_size field to part_info structure

This patch adds sector_size field to part_info structure (used
by new JFFS2 code).

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agojffs2: fix searching for latest version in jffs2_1pass_list_inodes()
Ilya Yanok [Thu, 13 Nov 2008 16:49:31 +0000 (19:49 +0300)]
jffs2: fix searching for latest version in jffs2_1pass_list_inodes()

We need to update i_version inside cycle to find really latest version
inside jffs2_1pass_list_inodes(). With that fixed we can use isize inside
dump_inode() instead of calling expensive jffs2_1pass_read_inode().

Signed-off-by: Alexey Neyman <avn@emcraft.com>
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agoevb64260: fix "cast to pointer from integer of different size" warnings
Wolfgang Denk [Tue, 9 Dec 2008 22:13:51 +0000 (23:13 +0100)]
evb64260: fix "cast to pointer from integer of different size" warnings

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Tue, 9 Dec 2008 21:30:22 +0000 (22:30 +0100)]
Merge branch 'master' of /home/wd/git/u-boot/custodians

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-usb
Wolfgang Denk [Tue, 9 Dec 2008 21:30:11 +0000 (22:30 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-usb

15 years agoUSB: descriptor handling
Stefan Althoefer [Sun, 7 Dec 2008 18:39:11 +0000 (19:39 +0100)]
USB: descriptor handling

Hi,

I found a bug when working with the u-boot USB subsystem on IXP425 processor
(big endian Xscale aka ARMv5).
I recognized that the second usb_endpoint_descriptor of the attached memory
stick was corrupted.

The reason for this are the packed structures below (either u-boot and
u-boot-usb):

--------------
/* Endpoint descriptor */
struct usb_endpoint_descriptor {
unsigned char  bLength;
unsigned char  bDescriptorType;
unsigned char  bEndpointAddress;
unsigned char  bmAttributes;
unsigned short wMaxPacketSize;
unsigned char  bInterval;
unsigned char  bRefresh;
unsigned char  bSynchAddress;

} __attribute__ ((packed));
/* Interface descriptor */
struct usb_interface_descriptor {
unsigned char  bLength;
unsigned char  bDescriptorType;
unsigned char  bInterfaceNumber;
unsigned char  bAlternateSetting;
unsigned char  bNumEndpoints;
unsigned char  bInterfaceClass;
unsigned char  bInterfaceSubClass;
unsigned char  bInterfaceProtocol;
unsigned char  iInterface;

unsigned char  no_of_ep;
unsigned char  num_altsetting;
unsigned char  act_altsetting;
struct usb_endpoint_descriptor ep_desc[USB_MAXENDPOINTS];
} __attribute__ ((packed));
------------

As usb_endpoint_descriptor is only 7byte in length, the start of all
odd ep_desc[] structures is not word aligned. This makes wMaxPacketSize
of these structures also not word aligned.

ARMv5 Architecture however does not support non-aligned multibyte
data type (see A2.8 of ARM Architecture Reference Manual).

Signed-off-by: Stefan Althoefer <stefan.althoefer@web.de>
Signed-off-by: Remy Böhmer <linux@bohmer.net>
15 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Tue, 9 Dec 2008 20:16:38 +0000 (21:16 +0100)]
Merge branch 'master' of /home/wd/git/u-boot/custodians

15 years agodrivers/fsl_pci_init: Fix compile warning
Kumar Gala [Tue, 9 Dec 2008 16:27:33 +0000 (10:27 -0600)]
drivers/fsl_pci_init: Fix compile warning

fsl_pci_init.c: In function 'fsl_pci_setup_inbound_windows':
fsl_pci_init.c:122: warning: comparison is always true due to limited range of data type

The check only makes sense if we are CONFIG_PHYS_64BIT

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
15 years agousbtty/omap: update to current API
Jean-Christophe PLAGNIOL-VILLARD [Sun, 7 Dec 2008 08:45:35 +0000 (09:45 +0100)]
usbtty/omap: update to current API

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Remy Böhmer <linux@bohmer.net>
15 years agovideo: fix FADS823 and RRvision compiling issues
Anatolij Gustschin [Tue, 9 Dec 2008 16:52:05 +0000 (17:52 +0100)]
video: fix FADS823 and RRvision compiling issues

Since commit 561858ee building for FADS823 and RRvision
doesn't work. Let's include version.h and timestamp.h
unconditionally to fix the problem.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
15 years agoUBI: Fix size parsing in "ubi create"
Stefan Roese [Tue, 2 Dec 2008 09:53:47 +0000 (10:53 +0100)]
UBI: Fix size parsing in "ubi create"

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoUBI: Enable re-initializing of the "ubi part" command
Stefan Roese [Thu, 27 Nov 2008 13:07:09 +0000 (14:07 +0100)]
UBI: Enable re-initializing of the "ubi part" command

With this patch now, the user can call "ubi part" multiple times to
re-connect the UBI device to another MTD partition.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoMTD: Fix problem based on non-working relocation (list head mtd_partitions)
Stefan Roese [Thu, 27 Nov 2008 13:05:15 +0000 (14:05 +0100)]
MTD: Fix problem based on non-working relocation (list head mtd_partitions)

Don't use LIST_HEAD() but initialize the struct via INIT_LIST_HEAD() upon
first call of add_mtd_partitions(). Otherwise this won't work on platforms
where the relocation is broken (like MIPS or PPC).

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-at91
Wolfgang Denk [Tue, 9 Dec 2008 00:06:14 +0000 (01:06 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-at91

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-net
Wolfgang Denk [Tue, 9 Dec 2008 00:00:24 +0000 (01:00 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-net

15 years agoSection name should be ".data", not "data"
Trent Piepho [Thu, 13 Nov 2008 01:29:48 +0000 (17:29 -0800)]
Section name should be ".data", not "data"

Signed-off-by: Trent Piepho <tpiepho@freescale.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoMAKEALL: Automatically use parallel builds
Wolfgang Denk [Mon, 8 Dec 2008 23:39:08 +0000 (00:39 +0100)]
MAKEALL: Automatically use parallel builds

Add logic to the MAKEALL script to determine the number of CPU cores
on the system, and run a parallel build if there is more than one.
Usually this significantrly accelerates builds.

Allow to manually adjust the number of parallel make jobs by using
the "BUILD_NCPUS" environment variable.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agovxworks.h: Fix build problem introduced by commits 29a4c24d/e9084b23
Wolfgang Denk [Mon, 8 Dec 2008 23:24:30 +0000 (00:24 +0100)]
vxworks.h: Fix build problem introduced by commits 29a4c24d/e9084b23

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoavr32/bootm: remove unused variable 'ret'
Jean-Christophe PLAGNIOL-VILLARD [Tue, 11 Nov 2008 05:08:59 +0000 (06:08 +0100)]
avr32/bootm: remove unused variable 'ret'

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Acked-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
15 years agoRemove unneeded CONFIG_SHELL references
Peter Tyser [Wed, 12 Nov 2008 19:06:48 +0000 (13:06 -0600)]
Remove unneeded CONFIG_SHELL references

Make should be using the bash shell by default which makes
CONFIG_SHELL unnecessary

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agoUse bash for default GNU Make shell application
Peter Tyser [Wed, 12 Nov 2008 18:33:20 +0000 (12:33 -0600)]
Use bash for default GNU Make shell application

Some Make script commands rely on bash-specific features like brace
expansion, so default to bash for the SHELL variable with a fallback
to the standard sh shell

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agojffs2: rename devices_init () in common/jffs2.c
Heiko Schocher [Wed, 12 Nov 2008 08:50:45 +0000 (09:50 +0100)]
jffs2: rename devices_init () in common/jffs2.c

rename devices_init () in common/jffs2.c to
jffs2_devices_init (), because there is also a
devices_init () in common/devices.c.

Signed-off-by: Heiko Schocher <hs@denx.de>
15 years agoSPARC: Fixed compiler error introduced by commit c160a9544743
Daniel Hellstrom [Mon, 10 Nov 2008 12:46:20 +0000 (12:46 +0000)]
SPARC: Fixed compiler error introduced by commit c160a9544743

This patch fixes a build error for the SPARC platform. It was
introduced by commit c160a9544743e80e8889edb2275538e7764ce334.

Signed-off-by: Daniel Hellstrom <daniel@gaisler.com>
15 years agomgsuvd add the board-specific part of the HDLC driver
Gary Jennejohn [Sun, 9 Nov 2008 11:50:59 +0000 (12:50 +0100)]
mgsuvd add the board-specific part of the HDLC driver

Signed-off-by: Gary Jennejohn <garyj@denx.de>
15 years agomgcoge add the board-specific part of the HDLC driver
Gary Jennejohn [Sun, 9 Nov 2008 11:45:03 +0000 (12:45 +0100)]
mgcoge add the board-specific part of the HDLC driver

Signed-off-by: Gary Jennejohn <garyj@denx.de>
15 years agokeymile add the common parts of the HDLC driver
Gary Jennejohn [Sun, 9 Nov 2008 11:36:15 +0000 (12:36 +0100)]
keymile add the common parts of the HDLC driver

This implements the ICN protocol used across the backplane and is
needed by all the keymile boards.

Signed-off-by: Gary Jennejohn <garyj@denx.de>
15 years agodrivers/bios_emulator: Move conditional compilation to Makefile
Jean-Christophe PLAGNIOL-VILLARD [Fri, 7 Nov 2008 21:46:22 +0000 (22:46 +0100)]
drivers/bios_emulator: Move conditional compilation to Makefile

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agocommon/cmd_ide.c: Corrected endian order printing for compact flash serial number.
Richard Retanubun [Thu, 6 Nov 2008 19:01:51 +0000 (14:01 -0500)]
common/cmd_ide.c: Corrected endian order printing for compact flash serial number.

Corrected endian order printing for compact flash serial number.

Signed-off-by: Richard Retanubun <RichardRetanubun@RuggedCom.com>
15 years agoIOMUX: Add console multiplexing support.
Gary Jennejohn [Thu, 6 Nov 2008 14:04:23 +0000 (15:04 +0100)]
IOMUX: Add console multiplexing support.

Modifications to support console multiplexing.  This is controlled using
CONFIG_SYS_CONSOLE_MUX in the board configuration file.

This allows a user to specify multiple console devices in the environment
with a command like this: setenv stdin serial,nc.  As a result, the user can
enter text on both the serial and netconsole interfaces.

All devices - stdin, stdout and stderr - can be set in this manner.

1) common/iomux.c and include/iomux.h contain the environment setting
implementation.
2) doc/README.iomux contains a somewhat more detailed description.
3) The implementation in (1) is called from common/cmd_nvedit.c to
handle setenv and from common/console.c to handle initialization of
input/output devices at boot time.
4) common/console.c also contains the code needed to poll multiple console
devices for input and send output to all devices registered for output.
5) include/common.h includes iomux.h and common/Makefile generates iomux.o
when CONFIG_SYS_CONSOLE_MUX is set.

Signed-off-by: Gary Jennejohn <garyj@denx.de>
15 years agostrings: use puts() rather than printf()
Mike Frysinger [Tue, 4 Nov 2008 21:03:46 +0000 (16:03 -0500)]
strings: use puts() rather than printf()

When running `strings` on really long strings, the stack tends to get
smashed due to printf().  Switch to puts() instead since we're only passing
the data through.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoUse new CONFIG_SYS_VXWORKS parameters for Netstal boards
Niklaus Giger [Mon, 3 Nov 2008 21:16:18 +0000 (22:16 +0100)]
Use new CONFIG_SYS_VXWORKS parameters for Netstal boards

Signed-off-by: Niklaus Giger <niklaus.giger@member.fsf.org>
15 years agocmd_elf.c: Cleanup bootvx and handle new CONFIG_SYS_VXWORKS parameters
Niklaus Giger [Mon, 3 Nov 2008 21:15:34 +0000 (22:15 +0100)]
cmd_elf.c: Cleanup bootvx and handle new CONFIG_SYS_VXWORKS parameters

- fix size too small by one in sprintf
- changed old (pre 2004) device name ibmEmac to emac
- boot device may be overriden in board config
- servername may be defined in board config
- additional parameters may be defined in board config
- fixed some line wrappings
- replaced  redundant MAX define by max

Signed-off-by: Niklaus Giger <niklaus.giger@member.fsf.org>
15 years agoAdd vxworks.h to handle CONFIG_SYS_VXWORKS parameters
Niklaus Giger [Mon, 3 Nov 2008 21:14:36 +0000 (22:14 +0100)]
Add vxworks.h to handle CONFIG_SYS_VXWORKS parameters

Signed-off-by: Niklaus Giger <niklaus.giger@member.fsf.org>
15 years agoREADME: Document CONFIG_SYS parameters for vxworks
Niklaus Giger [Mon, 3 Nov 2008 21:13:47 +0000 (22:13 +0100)]
README: Document CONFIG_SYS parameters for vxworks

Signed-off-by: Niklaus Giger <niklaus.giger@member.fsf.org>
15 years agolcd: Let the board code show board-specific info cleanup
Peter Tyser [Fri, 31 Oct 2008 16:12:38 +0000 (11:12 -0500)]
lcd: Let the board code show board-specific info cleanup

remove unneeded version.h from lcd.c

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoUpdate U-Boot's build timestamp on every compile
Peter Tyser [Mon, 3 Nov 2008 15:30:59 +0000 (09:30 -0600)]
Update U-Boot's build timestamp on every compile

Use the GNU 'date' command to auto-generate a new U-Boot
timestamp on every compile.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agoRemove redundant armv4 flag from arm926ejs compile flags
Remy Bohmer [Thu, 4 Dec 2008 21:25:57 +0000 (22:25 +0100)]
Remove redundant armv4 flag from arm926ejs compile flags

Currently the arm926ejs tree has the armv4 option set during compilation.
This flag does not belong here because a arm926 CPU is always a armv5 CPU.

Signed-off-by: Remy Bohmer <linux@bohmer.net>
15 years agoat91: Choose environment variables location within make config target
Nicolas Ferre [Sat, 6 Dec 2008 12:11:14 +0000 (13:11 +0100)]
at91: Choose environment variables location within make config target

This patch adds the possiblity to choose the media where the environment will
be located. This allow to choose this fundamental configuration without editing
config files.

Documentation file added.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Stelian Pop <stelian@popies.net>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agolwmon, tqm8xx: Fix build errors
Anatolij Gustschin [Mon, 3 Nov 2008 14:30:34 +0000 (15:30 +0100)]
lwmon, tqm8xx: Fix build errors

Commit 6b59e03e0237a40a2305ea385defdfd92000978b
lcd: Let the board code show board-specific info

introduced some bugs which prevent U-Boot building
for lwmon board if CONFIG_LCD_INFO_BELOW_LOGO will
be defined in the board configuration.

Also "LCD enabled" building for TQM823L doesn't work
since this commit.

This patch fixes above-mentioned issues.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
15 years agoignore .gdb_history files
Mike Frysinger [Sun, 2 Nov 2008 05:18:18 +0000 (01:18 -0400)]
ignore .gdb_history files

When using gdb, history files will often get generated.  So ignore them.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoFPGA: move fpga drivers to drivers/fpga
Jean-Christophe PLAGNIOL-VILLARD [Fri, 31 Oct 2008 11:26:55 +0000 (12:26 +0100)]
FPGA: move fpga drivers to drivers/fpga

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agonet: Fix TftpStart() ip:filename bug
Peter Tyser [Mon, 1 Dec 2008 22:29:38 +0000 (16:29 -0600)]
net: Fix TftpStart() ip:filename bug

The TftpStart() function modifies the 'BootFile'
string when 'BootFile' contains both an IP address
and filename (eg 1.2.3.4:/path/file). This causes
subsequent calls to TftpStart to incorrectly parse
the TFTP filename and server IP address to use.
For example:

=> tftp 0x100000 10.52.0.62:/home/ptyser/non_existant
Speed: 100, half duplex
Using eTSEC1 device
TFTP from server 10.52.0.62; our IP address is 10.52.253.79
                 ^^^^^^^^^^ CORRECT
Filename '/home/ptyser/non_existant'.
          ^^^^^^^^^^^^^^^^^^^^^^^^^ CORRECT
Load address: 0x100000
Loading: *
TFTP error: 'File not found' (1)
Starting again

eTSEC2: No link.
Speed: 100, half duplex
Using eTSEC1 device
TFTP from server 10.52.0.33; our IP address is 10.52.253.79
                 ^^^^^^^^^^ WRONG
Filename '10.52.0.62'.
          ^^^^^^^^^^ WRONG
Load address: 0x100000
Loading: *
TFTP error: 'File not found' (1)
Starting again

TftpStart() was modified to not modify the 'BootFile' string.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agonet: Add additional IP fragmentation check
Peter Tyser [Mon, 1 Dec 2008 22:26:21 +0000 (16:26 -0600)]
net: Add additional IP fragmentation check

Ignore IP packets which have the "more fragments" flag bit
set.  This flag indicates the IP packet is fragmented and
must be ignored by U-Boot.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agonet: Define IP flag field values
Peter Tyser [Mon, 1 Dec 2008 22:26:20 +0000 (16:26 -0600)]
net: Define IP flag field values

These defines were pulled from the "Add simple
IP/UDP fragmentation support" patch from Frank
Haverkamp <haver@vnet.ibm.com>.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agonet: tsec: Fix Marvell 88E1121R phy init
Anatolij Gustschin [Tue, 2 Dec 2008 09:31:04 +0000 (10:31 +0100)]
net: tsec: Fix Marvell 88E1121R phy init

This patch tries to ensure that phy interrupt pin
won't be asserted after booting. We experienced
following issues with current 88E1121R phy init:

Marvell 88E1121R phy can be hardware-configured
to share MDC/MDIO and interrupt pins for both ports
P0 and P1 (e.g. as configured on socrates board).
Port 0 interrupt pin will be shared by both ports
in such configuration. After booting Linux and
configuring eth0 interface, port 0 phy interrupts
are enabled. After rebooting without proper eth0
interface shutdown port 0 phy interrupts remain
enabled so any change on port 0 (link status, etc.)
cause assertion of the interrupt. Now booting Linux
and configuring eth1 interface will cause permanent
phy interrupt storm as the registered phy 1 interrupt
handler doesn't acknowledge phy 0 interrupts. This
of course should be fixed in Linux driver too.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Acked-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>