oweals/u-boot.git
16 years agocmd_i2c: Fix help for CONFIG_I2C_CMD_TREE && !CONFIG_I2C_MULTI_BUS
Peter Tyser [Wed, 1 Oct 2008 17:25:04 +0000 (12:25 -0500)]
cmd_i2c: Fix help for CONFIG_I2C_CMD_TREE && !CONFIG_I2C_MULTI_BUS

Original code displayed:
 => help i2c
 i2c i2c speed [speed] - show or set I2C bus speed
 i2c md chip address[.0, .1, .2] [# of objects] - read from I2C device
 ...

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
16 years agoUpdate CHANGELOG
Wolfgang Denk [Tue, 14 Oct 2008 20:13:41 +0000 (22:13 +0200)]
Update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoDo not init SATA when disabled on 8536DS.
Jason Jin [Fri, 10 Oct 2008 03:41:01 +0000 (11:41 +0800)]
Do not init SATA when disabled on 8536DS.

SGMII and SATA share the serdes on MPC8536 CPU, When SATA disabled and the
driver still try to access the SATA registers, the cpu will hangup.
This patch try to fix this by reading the serdes status before the SATA
initialize.

Signed-off-by: Jason Jin <Jason.jin@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
16 years agofsl_diu: fix alignment error that caused malloc corruption
Nikita V. Youshchenko [Thu, 2 Oct 2008 20:03:55 +0000 (00:03 +0400)]
fsl_diu: fix alignment error that caused malloc corruption

When aligning malloc()ed screen_base, invalid offset was added.
This not only caused misaligned result (which did not cause hardware
misbehaviour), but - worse - caused screen_base + smem_len to
be out of malloc()ed space, which in turn caused breakage of
futher malloc()/free() operation.

This patch fixes screen_base alignment.

Also this patch makes memset() that cleans framebuffer to be executed
on first initialization of diu, not only on re-initialization. It looks
correct to clean the framebuffer instead of displaying random garbage;
I believe that was disabled only because that memset caused breakage
of malloc/free described above - which no longer happens with the fix
described above.

Signed-off-by: Nikita V. Youshchenko <yoush@debian.org>
16 years agoapi: Fix building with CONFIG_API
Matthias Fuchs [Wed, 24 Sep 2008 08:29:37 +0000 (10:29 +0200)]
api: Fix building with CONFIG_API

This patch fixes building with CONFIG_API and CONFIG_USB_STORAGE.

USB_MAX_STOR_DEV is defined in include/usb.h, but
needed in api/api_storage.c.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
16 years agoRemove unused CFG_EEPROM_PAGE_WRITE_ENABLE references
Peter Tyser [Tue, 23 Sep 2008 17:39:40 +0000 (12:39 -0500)]
Remove unused CFG_EEPROM_PAGE_WRITE_ENABLE references

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
16 years agoRemove CFG_EEPROM_PAGE* dependencies for temperature sensors
Peter Tyser [Tue, 23 Sep 2008 17:38:42 +0000 (12:38 -0500)]
Remove CFG_EEPROM_PAGE* dependencies for temperature sensors

The checks for CFG_EEPROM_PAGE_WRITE_ENABLE and
CFG_EEPROM_PAGE_WRITE_BITS in various temperature
sensor drivers are not necessary

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
16 years agocmd_spi: remove broken signed casting for display
Mike Frysinger [Tue, 14 Oct 2008 11:04:38 +0000 (07:04 -0400)]
cmd_spi: remove broken signed casting for display

Since we're working with unsigned data, you can't apply a signed pointer
cast and then attempt to print the result.  Otherwise you get wrong output
when the sign bit is set like "0xFF" incorrectly extended to "0xFFFFFFFF".

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
16 years agostrings cmd: drop old CONFIG_CFG_STRINGS define
Mike Frysinger [Tue, 14 Oct 2008 11:05:24 +0000 (07:05 -0400)]
strings cmd: drop old CONFIG_CFG_STRINGS define

We don't need CONFIG_CFG_STRINGS anymore now that we have the define
CONFIG_CMD_STRINGS and Makefile control.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
16 years agoFix the NAND size overflow issue.
Jason Jin [Fri, 19 Sep 2008 09:32:49 +0000 (17:32 +0800)]
Fix the NAND size overflow issue.

When the total size of all NAND devices exceeds 4 GiB, the size  will
overflow. This patch tries to fix this.

Note that we still have a problem when a single NAND device is bigger
than 4 GiB: then the overflow would actually happen earlier, i. e.
when storing the size in nand_info[].size, as nand_info[].size is an
"u_int32_t".

Signed-off-by: Jason Jin <Jason.jin@freescale.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoMerge branch 'master' of git://git.denx.de/u-boot-net
Wolfgang Denk [Tue, 14 Oct 2008 11:07:43 +0000 (13:07 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-net

16 years agoMerge branch 'master' of git://git.denx.de/u-boot-sh
Wolfgang Denk [Tue, 14 Oct 2008 10:59:35 +0000 (12:59 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-sh

16 years agoAX88180: new gigabit network driver
Louis Su [Wed, 9 Jul 2008 03:01:37 +0000 (11:01 +0800)]
AX88180: new gigabit network driver

Signed-off-by: Louis Su <louis@asix.com.tw>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
16 years agoenable 10/100M at VSC8601 at tsec driver
Andre Schwarz [Tue, 19 Aug 2008 14:07:03 +0000 (16:07 +0200)]
enable 10/100M at VSC8601 at tsec driver

Currently VSC8601 doesn't link with 10/100M partners if the
EEPROM/Strapping is not set up.
Setting the auto-neg register fixes this.

Signed-off-by: Andre Schwarz <andre.schwarz@matrix-vision.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
16 years agonet: ne2000: Divided a function of NE2000 driver
Nobuhiro Iwamatsu [Tue, 30 Sep 2008 06:02:53 +0000 (15:02 +0900)]
net: ne2000: Divided a function of NE2000 driver

get_prom function was used __attriute__ , but it is not enable.
ax88796.o does not do link besides ne2000.o. When ld is carried
out, get_prom function of ax88796.c is ignored.
This problem is a thing by specifications of ld.
I checked and test this patch on SuperH and MIPS.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
16 years agosh: rsk7203: Add smc911x driver support to board config file
Nobuhiro Iwamatsu [Tue, 14 Oct 2008 02:10:59 +0000 (11:10 +0900)]
sh: rsk7203: Add smc911x driver support to board config file

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
16 years agosh: Fix cannot execute a stand-alone application
Nobuhiro Iwamatsu [Thu, 9 Oct 2008 04:54:33 +0000 (13:54 +0900)]
sh: Fix cannot execute a stand-alone application

Address calculated in EXPORT_FUNC in SuperH was wrong, I revised it.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
16 years agofsl_pci_init do not scan bus when configured as an end-point
Ed Swarthout [Thu, 9 Oct 2008 04:38:00 +0000 (23:38 -0500)]
fsl_pci_init do not scan bus when configured as an end-point

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
16 years agoARM DaVinci: Remove redundant setting of GD_FLG_RELOC for sffsdr board.
Hugo Villeneuve [Tue, 16 Sep 2008 21:07:53 +0000 (17:07 -0400)]
ARM DaVinci: Remove redundant setting of GD_FLG_RELOC for sffsdr board.

This is no longer necessary now that the GD_FLG_RELOC flag is set for
all ARM boards.

Signed-off-by: Hugo Villeneuve <hugo.villeneuve@lyrtech.com>
16 years agoFix lzma uncompress call (image_start wrongly used instead image_len)
Luigi 'Comio' Mantellini [Sat, 13 Sep 2008 08:04:32 +0000 (10:04 +0200)]
Fix lzma uncompress call (image_start wrongly used instead image_len)

Signed-off-by: Luigi 'Comio' Mantellini <luigi.mantellini@idf-hit.com>
16 years agoMerge branch 'master' of git://git.denx.de/u-boot-cfi-flash
Wolfgang Denk [Mon, 13 Oct 2008 12:18:38 +0000 (14:18 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-cfi-flash

16 years agompc86xx: use r4 instead of r2 in lock_ram_in_cache and unlock_ram_in_cache
Nick Spence [Thu, 28 Aug 2008 21:09:15 +0000 (14:09 -0700)]
mpc86xx: use r4 instead of r2 in lock_ram_in_cache and unlock_ram_in_cache

This is needed in unlock_ram_in_cache() because it is called from C and
will corrupt the small data area anchor that is kept in R2.

lock_ram_in_cache() is modified similarly as good coding practice, but
is not called from C.

Signed-off-by: Nick Spence <nick.spence@freescale.com>
16 years ago86xx: remove redudant code with lib_ppc/interrupts.c
Kumar Gala [Tue, 19 Aug 2008 20:05:34 +0000 (15:05 -0500)]
86xx: remove redudant code with lib_ppc/interrupts.c

For some reason we duplicated the majority of code in lib_ppc/interrupts.c
Not know how that happened, but there is no good reason for it.

Use the interrupt_init_cpu() and timer_interrupt_cpu() since its why
they exist.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years agoCFI: cfi_flash write fix for AMD legacy
Ed Swarthout [Thu, 9 Oct 2008 06:26:36 +0000 (01:26 -0500)]
CFI: cfi_flash write fix for AMD legacy

The flash_unlock_seq requires a sector for AMD_LEGACY.
Fix a retcode check typeo.

Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoMerge branch 'master' of /home/stefan/git/u-boot/u-boot
Stefan Roese [Mon, 13 Oct 2008 09:17:31 +0000 (11:17 +0200)]
Merge branch 'master' of /home/stefan/git/u-boot/u-boot

16 years agoMerge branch 'master' of ssh://10.10.0.7/home/wd/git/u-boot/master
Wolfgang Denk [Sun, 12 Oct 2008 21:56:33 +0000 (23:56 +0200)]
Merge branch 'master' of ssh://10.10.0.7/home/wd/git/u-boot/master

16 years agoMerge branch 'master' of git://git.denx.de/u-boot-arm
Wolfgang Denk [Sun, 12 Oct 2008 21:55:12 +0000 (23:55 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-arm

16 years agoMerge branch 'master' of ssh://10.10.0.7/home/wd/git/u-boot/master
Wolfgang Denk [Sun, 12 Oct 2008 21:25:50 +0000 (23:25 +0200)]
Merge branch 'master' of ssh://10.10.0.7/home/wd/git/u-boot/master

16 years agoMerge branch 'master' of git://git.denx.de/u-boot-fdt
Wolfgang Denk [Sun, 12 Oct 2008 21:13:16 +0000 (23:13 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-fdt

16 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Sun, 12 Oct 2008 21:12:44 +0000 (23:12 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

16 years agoMerge branch 'master' of git://git.denx.de/u-boot-cfi-flash
Wolfgang Denk [Sun, 12 Oct 2008 21:12:12 +0000 (23:12 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-cfi-flash

16 years agoppc4xx: Fix USB 2.0 phy reset sequence
Matthias Fuchs [Tue, 7 Oct 2008 11:13:10 +0000 (13:13 +0200)]
ppc4xx: Fix USB 2.0 phy reset sequence

This patch fixes USB 2.0 communication issues on some DU440 boards.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoppc4xx: Add strapping mode for 667MHz CPU frequency on DU440 board
Matthias Fuchs [Tue, 7 Oct 2008 11:13:09 +0000 (13:13 +0200)]
ppc4xx: Add strapping mode for 667MHz CPU frequency on DU440 board

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoppc4xx: Fix DU440 GPIO configuration
Matthias Fuchs [Tue, 7 Oct 2008 11:13:08 +0000 (13:13 +0200)]
ppc4xx: Fix DU440 GPIO configuration

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoppc4xx: Update DU440 config
Matthias Fuchs [Tue, 7 Oct 2008 11:13:07 +0000 (13:13 +0200)]
ppc4xx: Update DU440 config

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoMPC8572DS: Fix compile warnings
Kumar Gala [Wed, 8 Oct 2008 20:36:39 +0000 (15:36 -0500)]
MPC8572DS: Fix compile warnings

Commit 445a7b38308eb05b41de74165b20855db58c7ee5 introduced the following
compile warnings:

cmd_i2c.c:112: warning: missing braces around initializer
cmd_i2c.c:112: warning: (near initialization for 'i2c_no_probes[0]')

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years ago85xx: Using proper I2C source clock divider for MPC8544
Wolfgang Grandegger [Tue, 30 Sep 2008 08:55:57 +0000 (10:55 +0200)]
85xx: Using proper I2C source clock divider for MPC8544

Measurements with our MPC8544 board showed that the I2C bus frequency
is wrong by a factor of 1.5. Obviously, the interpretation of the
MPC85xx_PORDEVSR2_SEC_CFG bit of the cfg_sec_freq register is not
correct. There seems to be an error in the 8544 RM.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
16 years agoFSL: Fix get_cpu_board_revision() return value.
Rafal Czubak [Wed, 8 Oct 2008 11:41:30 +0000 (13:41 +0200)]
FSL: Fix get_cpu_board_revision() return value.

get_cpu_board_revision() returned board revision based on information stored
in global static struct eeprom. It should instead use one from local struct
board_eeprom, to which the data is actually read from EEPROM. The bug led to
system hang after printing L1 cache information on U-Boot startup. The problem
was observed on MPC8555CDS system and possibly affects other Freescale MPC85xx
boards using CFG_I2C_EEPROM_CCID.

The change has been successfully tested on MPC8555CDS system.

Signed-off-by: Rafal Czubak <rcz@semihalf.com>
16 years agoupdate uImage FIT multi documentation
Jean-Christophe PLAGNIOL-VILLARD [Tue, 30 Sep 2008 18:08:49 +0000 (20:08 +0200)]
update uImage FIT multi documentation

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
16 years agomove README.imx31 to doc/ and merge with README.mx31
Jean-Christophe PLAGNIOL-VILLARD [Tue, 30 Sep 2008 18:08:36 +0000 (20:08 +0200)]
move README.imx31 to doc/ and merge with README.mx31

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
16 years agoi.MX31: switch to CFG_HZ=1000
Guennadi Liakhovetski [Thu, 25 Sep 2008 18:54:37 +0000 (20:54 +0200)]
i.MX31: switch to CFG_HZ=1000

Switch to the standard CFG_HZ=1000 value, while at it, minor white-space
cleanup, remove CFG_CLKS_IN_HZ from config-headers. Tested on mx31ads,
provides 2% or 0.4% precision depending on the
CONFIG_MX31_TIMER_HIGH_PRECISION flag. Measured with stop-watch on 100s
boot-delay.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
16 years agoFix bug: in arch-arm, env_get_char dose not work fine
gnusercn [Wed, 8 Oct 2008 16:58:58 +0000 (18:58 +0200)]
Fix bug: in arch-arm, env_get_char dose not work fine

due to the arm implementation which supposed that U-Boot is in RAM
when we jump to start_armboot

Signed-off-by: gnusercn <gnusercn@gmail.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
16 years agoppc4xx: Reset and relock memory DLL after SDRAM_CLKTR change
Adam Graham [Mon, 6 Oct 2008 17:16:13 +0000 (10:16 -0700)]
ppc4xx: Reset and relock memory DLL after SDRAM_CLKTR change

After changing SDRAM_CLKTR phase value rerun the memory preload
initialization sequence (INITPLR) to reset and relock the memory
DLL. Changing the SDRAM_CLKTR memory clock phase coarse timing
adjustment effects the phase relationship of the internal, to the
PPC chip, and external, to the PPC chip, versions of MEMCLK_OUT.

Signed-off-by: Adam Graham <agraham@amcc.com>
Signed-off-by: Victor Gallardo <vgallardo@amcc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoRemove redundant #define for MPC8536DS
Haiying Wang [Fri, 3 Oct 2008 15:48:03 +0000 (11:48 -0400)]
Remove redundant #define for MPC8536DS

Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
16 years agoAdd ID EEPROM support for MPC8572DS
Haiying Wang [Fri, 3 Oct 2008 15:47:30 +0000 (11:47 -0400)]
Add ID EEPROM support for MPC8572DS

The ID EEPROM on MPC8572DS board locates on I2C bus 1. Its the storage for
system information like mac addresses etc. This patch enables it.

Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
16 years agoMinor fixes for I2C address on MPC8572DS
Haiying Wang [Fri, 3 Oct 2008 15:46:59 +0000 (11:46 -0400)]
Minor fixes for I2C address on MPC8572DS

MPC8572DS has two i2c buses. This patch moves the DDR SPD_EEPROM to i2c bus 1
according to the board spec, and adds the 2nd i2c bus offset.

Signed-off-by: Haiying Wang <Haiying.Wang@freescale.com>
16 years agoFix the incorrect DDR clk freq reporting on 8536DS
Jason Jin [Sat, 27 Sep 2008 06:40:57 +0000 (14:40 +0800)]
Fix the incorrect DDR clk freq reporting on 8536DS

On 8536DS board, When the DDR clk is set async mode(SW3[6:8] != 111),
The display is still sync mode DDR freq. This patch try to fix
this. The display DDR freq is now the actual freq in both
sync and async mode.

Signed-off-by: Jason Jin <Jason.jin@freescale.com>
16 years ago85xx: Remove setting of *cache-line-size in device trees
Kumar Gala [Tue, 7 Oct 2008 15:28:46 +0000 (10:28 -0500)]
85xx: Remove setting of *cache-line-size in device trees

ePAPR says if the *cache-block-size is the same as *cache-line-size
than we don't need the *cache-line-size property.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years agolibfdt: Fix error in documentation for fdt_get_alias_namelen()
Gerald Van Baren [Sat, 4 Oct 2008 11:56:06 +0000 (07:56 -0400)]
libfdt: Fix error in documentation for fdt_get_alias_namelen()

Oops, screwed up the function name in the documenting comment for this
function.  Trivial correction in this patch.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Gerald Van Baren <vanbaren@cideas.com>
16 years agolibfdt: Add function to explicitly expand aliases
David Gibson [Wed, 20 Aug 2008 06:55:14 +0000 (16:55 +1000)]
libfdt: Add function to explicitly expand aliases

Kumar has already added alias expansion to fdt_path_offset().
However, in some circumstances it may be convenient for the user of
libfdt to explicitly get the string expansion of an alias.  This patch
adds a function to do this, fdt_get_alias(), and uses it to implement
fdt_path_offset().

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
16 years agocfi_flash: do not reset flash when probe fails
Mike Frysinger [Thu, 2 Oct 2008 05:55:38 +0000 (01:55 -0400)]
cfi_flash: do not reset flash when probe fails

The CFI flash driver starts at flash_init() which calls down into
flash_get_size().  This starts by calling flash_detect_cfi().  If said
function fails, flash_get_size() finishes by attempting to reset the
flash.  Unfortunately, it does this with an info->portwidth set to 0x10
which filters down into flash_make_cmd() and that happily smashes the
stack by sticking info->portwidth bytes into a cfiword_t variable that
lives on the stack.  On a 64bit system you probably won't notice, but
killing the last 8 bytes on a 32bit system usually leads to a corrupt
return address.  Which is what happens on a Blackfin system.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agofdt: Overwrite /chosen node in bootm if it already exists in the dtb
Stefan Roese [Fri, 26 Sep 2008 15:03:26 +0000 (17:03 +0200)]
fdt: Overwrite /chosen node in bootm if it already exists in the dtb

Set force parameter in fdt_chosen() call in do_bootm_linux() call.
Without this, the chosen node is not overwritten if it already
exists.

Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoFix an overflow case in fdt_offset_ptr() detected by GCC 4.3.
Jon Loeliger [Thu, 25 Sep 2008 16:02:17 +0000 (11:02 -0500)]
Fix an overflow case in fdt_offset_ptr() detected by GCC 4.3.

Using Gcc 4.3 detected this problem:

    ../dtc/libfdt/fdt.c: In function 'fdt_next_tag':
    ../dtc/libfdt/fdt.c:82: error: assuming signed overflow does not
    occur when assuming that (X + c) < X is always false

To fix the problem, treat the offset as an unsigned int.

The problem report and proposed fix were provided
by Steve Papacharalambous <stevep@freescale.com>.

Signed-off-by: Jon Loeliger <jdl@freescale.com>
16 years agolibfdt: Fix bugs in fdt_get_path()
David Gibson [Fri, 29 Aug 2008 04:19:13 +0000 (14:19 +1000)]
libfdt: Fix bugs in fdt_get_path()

The current implementation of fdt_get_path() has a couple of bugs,
fixed by this patch.

First, contrary to its documentation, on success it returns the length
of the node's path, rather than 0.  The testcase is correspondingly
wrong, and the patch fixes this as well.

Second, in some circumstances, it will return -FDT_ERR_BADOFFSET
instead of -FDT_ERR_NOSPACE when given insufficient buffer space.
Specifically this happens when there is insufficient space even to
hold the path's second last component.  This behaviour is corrected,
and the testcase updated to check it.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
16 years agoTQM5200: enable support for ATAPI devices
Wolfgang Denk [Wed, 1 Oct 2008 10:34:58 +0000 (12:34 +0200)]
TQM5200: enable support for ATAPI devices

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoRevert "ARM: set GD_FLG_RELOC for boards skipping relocation to RAM"
Jean-Christophe PLAGNIOL-VILLARD [Mon, 15 Sep 2008 03:48:25 +0000 (05:48 +0200)]
Revert "ARM: set GD_FLG_RELOC for boards skipping relocation to RAM"

we need this due to the arm implementation which supposed that U-Boot
is in RAM when we jump to start_armboot

This reverts commit f96b44cef897bd372beb86dde1b33637c119d84d.
in order to do it for all arm board

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
16 years agompc83xx: don't disable autoboot
Kim Phillips [Wed, 24 Sep 2008 13:46:25 +0000 (08:46 -0500)]
mpc83xx: don't disable autoboot

bootdelay set to -1 'permanently' disables autobooting, even if
bootcmd is specified.  Change to a positive value to allow
autobooting when a bootcmd is set.

Reported-by: Coray Tate <Coray.Tate@freescale.com>
Cc: Scott Wood <scottwood@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
16 years agompc83xx: Fix typo in include/mpc83xx.h
Nobuhiro Iwamatsu [Wed, 24 Sep 2008 01:42:15 +0000 (10:42 +0900)]
mpc83xx: Fix typo in include/mpc83xx.h

Fixed typo from CONIFG_MPC837X to CONFIG_MPC837X

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
16 years agompc83xx: add h/w flash protection to board configs
Kim Phillips [Tue, 23 Sep 2008 14:38:49 +0000 (09:38 -0500)]
mpc83xx: add h/w flash protection to board configs

the operating system may leave flash in a h/w locked state after writing.
This allows u-boot to continue to write flash by enabling h/w unlocking
by default.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
16 years agompc83xx: spd_sdram: fix ddr sdram base address assignment bug
Anton Vorontsov [Thu, 11 Sep 2008 17:35:36 +0000 (21:35 +0400)]
mpc83xx: spd_sdram: fix ddr sdram base address assignment bug

The spd_dram code shifts the base address, then masks 20 bits, but
forgets to shift the base address back. Fix this by just masking the
base address correctly.

Found this bug while trying to relocate a DDR memory at the base != 0.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
16 years agoMerge branch 'master' of /home/stefan/git/u-boot/u-boot
Stefan Roese [Tue, 23 Sep 2008 08:15:59 +0000 (10:15 +0200)]
Merge branch 'master' of /home/stefan/git/u-boot/u-boot

16 years agoppc4xx: Canyonlands: Remove unnecessary FDT warning upon DTB fixup
Stefan Roese [Mon, 22 Sep 2008 14:10:43 +0000 (16:10 +0200)]
ppc4xx: Canyonlands: Remove unnecessary FDT warning upon DTB fixup

Depending on the configuration jumper "SATA SELECT", U-Boot disabled
either one PCIe node or the SATA node in the device tree blob. This
patch removes the unnecessary and even confusing warning, when the node
is not found at all.

Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoAdd missing device types to dev_print() in part.c
Remy Bohmer [Fri, 19 Sep 2008 11:30:06 +0000 (13:30 +0200)]
Add missing device types to dev_print() in part.c

Signed-off-by: Remy Bohmer <linux@bohmer.net>
16 years agoFix DPRAM memory leak when CFG_ALLOC_DPRAM is defined, which
Wolfgang Denk [Mon, 22 Sep 2008 20:23:06 +0000 (22:23 +0200)]
Fix DPRAM memory leak when CFG_ALLOC_DPRAM is defined, which
eventually leads to a machine check. This change assures that DPRAM
is allocated only once in that case.

Signed-off-by: Gary Jennejohn <garyj@denx.de>
Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoFix Spartan-3 definitions.
Laurent Pinchart [Wed, 17 Sep 2008 15:57:34 +0000 (17:57 +0200)]
Fix Spartan-3 definitions.

A few Spartan-3 definitions erroneously use Spartan-3E size
constants. This patch fixes them.

Signed-off-by: Laurent Pinchart <laurentp@cse-semaphore.com>
16 years agoRemove duplicate Spartan-3E definition.
Laurent Pinchart [Wed, 17 Sep 2008 15:41:58 +0000 (17:41 +0200)]
Remove duplicate Spartan-3E definition.

Signed-off-by: Laurent Pinchart <laurentp@cse-semaphore.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agosocrates: change default mtest address range
Anatolij Gustschin [Wed, 17 Sep 2008 11:46:17 +0000 (13:46 +0200)]
socrates: change default mtest address range

Running mtest command on socrates without specifying
an address range crashes the board. This patch changes
default mtest address range to prevent this behavior.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
16 years agosocrates: fix crash after relocation
Anatolij Gustschin [Wed, 17 Sep 2008 10:34:45 +0000 (12:34 +0200)]
socrates: fix crash after relocation

Currently U-Boot crashes after relocation to RAM.
Changing the CPO value of the DDR SDRAM TIMING_CFG_2
register to READ_LAT + 1 (to the value it was before
conversion of socrates to new DDR code) fixes the
problem.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
16 years agosocrates: fix SPD EEPROM address
Anatolij Gustschin [Wed, 17 Sep 2008 09:45:51 +0000 (11:45 +0200)]
socrates: fix SPD EEPROM address

Commit be0bd8234b9777ecd63c4c686f72af070d886517
changed SPD EEPROM address to 0x51 and DDR SDRAM
detection stopped working. Change this address
back to 0x50.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
16 years agoMerge branch 'master' of /home/stefan/git/u-boot/u-boot
Stefan Roese [Mon, 22 Sep 2008 13:26:49 +0000 (15:26 +0200)]
Merge branch 'master' of /home/stefan/git/u-boot/u-boot

16 years agoRevert "ppc4xx: Fix DDR2 auto calibration on Kilauea 600MHz (200MHz PLB)"
Stefan Roese [Mon, 22 Sep 2008 09:06:50 +0000 (11:06 +0200)]
Revert "ppc4xx: Fix DDR2 auto calibration on Kilauea 600MHz (200MHz PLB)"

This reverts commit 3eec160a3a405b29ce9c06920f6427b9047dd8a8.

Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Sun, 21 Sep 2008 20:36:23 +0000 (22:36 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

16 years agosh: Fix compile warning
Nobuhiro Iwamatsu [Thu, 18 Sep 2008 10:34:36 +0000 (19:34 +0900)]
sh: Fix compile warning

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
16 years agosh: Add support watchdog for SH4A core
Nobuhiro Iwamatsu [Thu, 18 Sep 2008 10:04:26 +0000 (19:04 +0900)]
sh: Add support watchdog for SH4A core

Add support watchdog for SH4A core (SH7763, SH7780 and SH7785).
And fix some compile warning.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
16 years agosh: Fix typo in SH serial driver
Nobuhiro Iwamatsu [Wed, 17 Sep 2008 02:45:26 +0000 (11:45 +0900)]
sh: Fix typo in SH serial driver

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
16 years agosh: Add support any page size and empty_zero_page to SH Linux uImage
Nobuhiro Iwamatsu [Wed, 17 Sep 2008 02:08:36 +0000 (11:08 +0900)]
sh: Add support any page size and empty_zero_page to SH Linux uImage

Old U-Boot supported 4KB page size only. If this version, Linux
kernel can not get command line from U-Boot.
SH Linux kernel can change page size and empty_zero_page.
This patch support this function and fix promlem.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
16 years agosh: rsk7203: Add support pkt_data_pull and pkt_data_push function
Nobuhiro Iwamatsu [Thu, 28 Aug 2008 04:40:52 +0000 (13:40 +0900)]
sh: rsk7203: Add support pkt_data_pull and pkt_data_push function

Add function of smc911x, pkt_data_pull and pkt_data_push.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
16 years agoADS5121: fix typo in "rootpath" default setting
Wolfgang Denk [Thu, 18 Sep 2008 11:57:32 +0000 (13:57 +0200)]
ADS5121: fix typo in "rootpath" default setting

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoCoding style cleanup; update CHANEGLOG
Wolfgang Denk [Wed, 17 Sep 2008 08:17:55 +0000 (10:17 +0200)]
Coding style cleanup; update CHANEGLOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoppc4xx: Fix DDR2 auto calibration on Kilauea 600MHz (200MHz PLB)
Victor Gallardo [Tue, 16 Sep 2008 13:59:13 +0000 (06:59 -0700)]
ppc4xx: Fix DDR2 auto calibration on Kilauea 600MHz (200MHz PLB)

Signed-off-by: Victor Gallardo <vgallardo@amcc.com>
Signed-off-by: Adam Graham <agraham@amcc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
16 years agoSupport for multiple SGMII/TBI interfaces for TSEC ethernet
Peter Tyser [Tue, 16 Sep 2008 15:04:47 +0000 (10:04 -0500)]
Support for multiple SGMII/TBI interfaces for TSEC ethernet

Fix TBI PHY accesses to use the proper offset in CPU register space. The
previous code would incorrectly access the TBI PHY by reading/writing to CPU
register space at the same location as would be used to access external PHYs.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Andy Fleming <afleming@freescale.com>
16 years agoTQM8xx: Fix CFI flash driver support for all TQM8xx based boards
Wolfgang Denk [Tue, 16 Sep 2008 16:02:19 +0000 (18:02 +0200)]
TQM8xx: Fix CFI flash driver support for all TQM8xx based boards

After switching to using the CFI flash driver, the correct remapping
of the flash banks was forgotten.

Also, some boards were not adapted, and the old legacy flash driver
was not removed yet.

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoPrepare v2008.10-rc2 v2008.10-rc2
Wolfgang Denk [Sat, 13 Sep 2008 22:59:35 +0000 (00:59 +0200)]
Prepare v2008.10-rc2

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoCoding style cleanup, update CHANGELOG
Wolfgang Denk [Sat, 13 Sep 2008 00:23:05 +0000 (02:23 +0200)]
Coding style cleanup, update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoUpdate mailing list email and archive addresses
Peter Tyser [Wed, 10 Sep 2008 14:18:34 +0000 (09:18 -0500)]
Update mailing list email and archive addresses

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
16 years ago85xx: socrates: autoprobe Lime chip
u-boot@bugs.denx.de [Thu, 11 Sep 2008 13:40:01 +0000 (15:40 +0200)]
85xx: socrates: autoprobe Lime chip

This patch is an attempt to implement autoprobing for the Lime
presence on the bus.
Configure GPCM for Lime CS2 and try to access chip ID registers.
Second read atempt delivers register values if the chip is present.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
16 years ago85xx: socrates: Add support for new image format.
u-boot@bugs.denx.de [Thu, 11 Sep 2008 13:40:01 +0000 (15:40 +0200)]
85xx: socrates: Add support for new image format.

Signed-off-by: Detlev Zundel <dzu@denx.de>
16 years agosh: Fix compile error for r2dplus
Nobuhiro Iwamatsu [Thu, 11 Sep 2008 08:28:18 +0000 (17:28 +0900)]
sh: Fix compile error for r2dplus

netdev.h was not include by r2dplus.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
16 years agopowerpc: Fix bootm to boot up again with a Ramdisk
Heiko Schocher [Thu, 11 Sep 2008 06:11:23 +0000 (08:11 +0200)]
powerpc: Fix bootm to boot up again with a Ramdisk

Commit 2a1a2cb6 didnt remove the dummy mem reservation in fdt_chosen,
and this stopped Linux from booting with a Ramdisk. This patch fixes
this, by deleting the useless dummy mem reservation.

When booting with a Ramdisk, a fix offset FDT_RAMDISK_OVERHEAD is now
added to of_size, so we dont need anymore a dummy mem reservation.

I measured the value of FDT_RAMDISK_OVERHEAD on a MPC8270 based
system (=0x44 bytes) and rounded it up to 0x80).

Signed-off-by: Heiko Schocher <hs@denx.de>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
16 years agoAdd support for LZMA uncompression algorithm.
Luigi 'Comio' Mantellini [Mon, 8 Sep 2008 00:46:13 +0000 (02:46 +0200)]
Add support for LZMA uncompression algorithm.

Signed-off-by: Luigi 'Comio' Mantellini <luigi.mantellini@idf-hit.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
16 years agoMerge branch 'master' of git://git.denx.de/u-boot-nand-flash
Wolfgang Denk [Fri, 12 Sep 2008 23:45:56 +0000 (01:45 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-nand-flash

16 years agofsl_elbc_nand: ecclayout cleanups
Anton Vorontsov [Fri, 27 Jun 2008 19:04:20 +0000 (23:04 +0400)]
fsl_elbc_nand: ecclayout cleanups

This patch deletes oobavail assignments, they're calculated by the nand
core code in nand_scan_tail, plus current oobavail values are wrong for
the LP NANDs.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
16 years agofsl_elbc_nand: implement support for flash-based BBT
Anton Vorontsov [Fri, 27 Jun 2008 19:04:13 +0000 (23:04 +0400)]
fsl_elbc_nand: implement support for flash-based BBT

This patch implements support for flash-based BBT for chips working
through ELBC NAND controller, so that NAND core will not have to re-scan
for bad blocks on every boot.

Because ELBC controller may provide HW-generated ECCs we should adjust
bbt pattern and bbt version positions in the OOB free area.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
16 years agofsl_elbc_nand: fix OOB workability for large page NAND chips
Anton Vorontsov [Fri, 27 Jun 2008 19:04:04 +0000 (23:04 +0400)]
fsl_elbc_nand: fix OOB workability for large page NAND chips

For large page chips, nand_bbt is looking into OOB area, and checking
for "0xff 0xff" pattern at OOB offset 0. That is, two bytes should be
reserved for bbt means.

But ELBC driver is specifying ecclayout so that oobfree area starts at
offset 1, so only one byte left for the bbt purposes.

This causes problems with any OOB users, namely JFFS2: after first mount
JFFS2 will fill all OOBs with "erased marker", so OOBs will contain:

  OOB Data: ff 19 85 20 03 00 ff ff ff 00 00 08 ff ff ff ff
  OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
  OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
  OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff

And on the next boot, NAND core will rescan for bad blocks, then will
see "0xff 0x19" pattern, and will mark all blocks as bad ones.

To fix the issue we should implement our own bad block pattern: just one
byte at OOB start. Though, this will work only for x8 chips. For x16
chips two bytes must be checked. Since ELBC driver does not support x16
NANDs (yet), we're safe for now.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
16 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Fri, 12 Sep 2008 14:14:28 +0000 (16:14 +0200)]
Merge branch 'master' of /home/wd/git/u-boot/custodians

16 years agoMerge branch 'Makefile-next' of git://git.denx.de/u-boot-arm
Wolfgang Denk [Fri, 12 Sep 2008 14:13:12 +0000 (16:13 +0200)]
Merge branch 'Makefile-next' of git://git.denx.de/u-boot-arm

16 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Fri, 12 Sep 2008 13:24:54 +0000 (15:24 +0200)]
Merge branch 'master' of /home/wd/git/u-boot/custodians

16 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Fri, 12 Sep 2008 13:23:20 +0000 (15:23 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

16 years agoMPC512x: reduce timeout waiting for Ethernet autonegotiation to 2.5s
Wolfgang Denk [Fri, 12 Sep 2008 11:52:21 +0000 (13:52 +0200)]
MPC512x: reduce timeout waiting for Ethernet autonegotiation to 2.5s

Signed-off-by: Wolfgang Denk <wd@denx.de>
16 years agoppc4xx: Enable device tree (FDT) support in zeus board port
Stefan Roese [Thu, 11 Sep 2008 11:05:56 +0000 (13:05 +0200)]
ppc4xx: Enable device tree (FDT) support in zeus board port

Signed-off-by: Stefan Roese <sr@denx.de>