platform/kernel/u-boot.git
15 years agoA driver for the S6E63D6 SPI display controller from Samsung
Guennadi Liakhovetski [Fri, 6 Feb 2009 09:37:41 +0000 (10:37 +0100)]
A driver for the S6E63D6 SPI display controller from Samsung

This is a driver for the S6E63D6 SPI OLED display controller from Samsung.
It only provides access to controller's registers so the client can freely
configure it.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
Acked-by: Anatolij Gustschin <agust@denx.de>
15 years agoi.MX31: support GPIO as a chip-select in the mxc_spi driver
Guennadi Liakhovetski [Fri, 13 Feb 2009 08:26:40 +0000 (09:26 +0100)]
i.MX31: support GPIO as a chip-select in the mxc_spi driver

Some SPI devices have special requirements on chip-select handling.
With this patch we can use a GPIO as a chip-select and strictly follow
the SPI_XFER_BEGIN and SPI_XFER_END flags.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoi.MX31: add a simple gpio driver
Guennadi Liakhovetski [Sat, 7 Feb 2009 00:18:07 +0000 (01:18 +0100)]
i.MX31: add a simple gpio driver

This is a minimal driver, so far only managing output. It will
be used by the mxc_spi.c driver.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoi.MX31: fix SPI driver for shorter than 32 bit
Guennadi Liakhovetski [Fri, 6 Feb 2009 23:09:12 +0000 (00:09 +0100)]
i.MX31: fix SPI driver for shorter than 32 bit

Fix setting the SPI Control register, 8 and 16-bit transfers
and a wrong pointer in the free routine in the mxc_spi driver.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoMPC86xx: set CONFIG_MAX_MEM_MAPPED to 2G by default
Becky Bruce [Mon, 23 Feb 2009 19:56:51 +0000 (13:56 -0600)]
MPC86xx: set CONFIG_MAX_MEM_MAPPED to 2G by default

Currently, we get 256MB as the default, but since all the 86xx
board configs define a 2G BAT mapping for RAM, raise default
to 2G.

Signed-off-by: Becky Bruce <beckyb@kernel.crashing.org>
Acked-by: Jon Loeliger <jdl@freescale.com>
15 years agompc8641hpcn: Indicate 36-bit addr map in boot messages
Becky Bruce [Thu, 12 Feb 2009 16:43:32 +0000 (10:43 -0600)]
mpc8641hpcn: Indicate 36-bit addr map in boot messages

If 36-bit addressing is enabled, print a message on the console
when we boot.

Signed-off-by: Becky Bruce <beckyb@kernel.crashing.org>
15 years agonetloop: speed up NetLoop
Heiko Schocher [Tue, 10 Feb 2009 08:38:52 +0000 (09:38 +0100)]
netloop: speed up NetLoop

NetLoop polls every cycle with getenv some environment variables.
This is horribly slow, especially when the environment is big.

This patch reads only the environment variables in NetLoop,
when they were changed.

Also moved the init part of the NetLoop function in a seperate
function.

Signed-off-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agosmc911x_eeprom: new example app for managing newer SMC parts
Mike Frysinger [Mon, 22 Dec 2008 07:56:07 +0000 (02:56 -0500)]
smc911x_eeprom: new example app for managing newer SMC parts

A forward port of the last version to work with the newer smc911x driver.
I only have a board with a LAN9218 part on it, so that is the only one
I've tested.  But there isn't anything in this that would make it terribly
chip specific afaik.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
CC: Sascha Hauer <s.hauer@pengutronix.de>
CC: Guennadi Liakhovetski <lg@denx.de>
CC: Magnus Lilja <lilja.magnus@gmail.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoAdded Vitesse VSC8211 definitions to TSEC driver
Pieter Henning [Mon, 23 Feb 2009 07:17:15 +0000 (23:17 -0800)]
Added Vitesse VSC8211 definitions to TSEC driver

Added the struct containing PHY settings for the Vitesse VSC8211 phy to
the phy_info list in tsec.c

Signed-off-by: Pieter Henning <phenning@vastech.co.za>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
15 years agoUpdate CHANGELOG; Prepare 2009.03-rc1 v2009.03-rc1
Wolfgang Denk [Sun, 22 Feb 2009 23:22:21 +0000 (00:22 +0100)]
Update CHANGELOG;  Prepare 2009.03-rc1

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoARM: synchronize mach-types.h with linux v2.6.29-rc5-315-g683fdc5
Wolfgang Denk [Sun, 22 Feb 2009 22:45:40 +0000 (23:45 +0100)]
ARM: synchronize mach-types.h with linux v2.6.29-rc5-315-g683fdc5

The file was generated from building versatile_defconfig.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Sun, 22 Feb 2009 21:51:32 +0000 (22:51 +0100)]
Merge branch 'master' of /home/wd/git/u-boot/custodians

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-arm
Wolfgang Denk [Sun, 22 Feb 2009 21:51:25 +0000 (22:51 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-arm

15 years agoMerge branch 'master' of /home/wd/git/u-boot/custodians
Wolfgang Denk [Sun, 22 Feb 2009 21:01:01 +0000 (22:01 +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 [Sun, 22 Feb 2009 21:00:58 +0000 (22:00 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-usb

15 years agoMIPS: Fix GCC-4.2 'discards qualifiers from pointer target type' warnings
Shinya Kuribayashi [Sun, 22 Feb 2009 05:24:11 +0000 (14:24 +0900)]
MIPS: Fix GCC-4.2 'discards qualifiers from pointer target type' warnings

Compiling dbau1x00 and gth2 boards with GCC-4.2, you would see new warnings
like this:

skuribay@ubuntu:u-boot.git$ ./MAKEALL dbau1000
Configuring for dbau1x00 board...
au1x00_eth.c: In function 'au1x00_send':
au1x00_eth.c:158: warning: passing argument 1 of 'virt_to_phys' discards qualifiers from pointer target type
au1x00_eth.c: In function 'au1x00_recv':
au1x00_eth.c:211: warning: passing argument 1 of 'virt_to_phys' discards qualifiers from pointer target type
au1x00_eth.c: In function 'au1x00_init':
au1x00_eth.c:252: warning: passing argument 1 of 'virt_to_phys' discards qualifiers from pointer target type
au1x00_eth.c: In function 'au1x00_recv':
au1x00_eth.c:211: warning: passing argument 1 of 'virt_to_phys' discards qualifiers from pointer target type
au1x00_eth.c: In function 'au1x00_init':
au1x00_eth.c:252: warning: passing argument 1 of 'virt_to_phys' discards qualifiers from pointer target type
au1x00_eth.c: In function 'au1x00_send':
au1x00_eth.c:158: warning: passing argument 1 of 'virt_to_phys' discards qualifiers from pointer target type

We're passing a volatile pointer to a function which is expecting a non-
volatile pointer.  That's potentially dangerous, so gcc warns about it.
Confirmed with ELDK 4.2 (GCC 4.2.2) and Sourcey G++ 4.2 (GCC 4.2.3).

To fix this, we add a volatile attribute to the argument in question.
The virt_to_phys function in Linux kernel also does the same thing.

Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-blackfin
Wolfgang Denk [Sun, 22 Feb 2009 20:13:35 +0000 (21:13 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-blackfin

15 years agoOMAP3: Clean up MMC code
Dirk Behme [Fri, 20 Feb 2009 16:51:28 +0000 (17:51 +0100)]
OMAP3: Clean up MMC code

Clean up OMAP3 MMC code:

* Convert register access to struct & readx/writex style
* Replace hardcode values by macros
* Remove macro defined twice

Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoOMAP3: Pandora: Update pin mux
Dirk Behme [Thu, 12 Feb 2009 17:55:43 +0000 (18:55 +0100)]
OMAP3: Pandora: Update pin mux

Clock pin must have input enabled for MMC3 to work.
Also enable pull-ups for cmd/data lines to be consistent
with remaining MMC host pin setup.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
15 years agoOMAP3: Add OMAP3 auto detection
Dirk Behme [Thu, 12 Feb 2009 17:55:42 +0000 (18:55 +0100)]
OMAP3: Add OMAP3 auto detection

This patch adds OMAP3 cpu type auto detection based on OMAP3 register
and removes hardcoded values.

Signed-off-by: Steve Sakoman <sakoman@gmail.com>
Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoOMAP3: Beagle: Add board revision detection
Dirk Behme [Thu, 12 Feb 2009 17:55:41 +0000 (18:55 +0100)]
OMAP3: Beagle: Add board revision detection

With BeagleBoard revision C some HW changes are introduced (e.g. PinMUX)
which might need different software handling. For this, GPIO pin 171 (GPIO
module 6, offset 11) can be used to check for board revision. If this pin
is low, we have a rev C board. Else it must be a revision Ax or Bx board.

To handle board differences you can call function beagle_get_revision().
E.g.:

if (beagle_get_revision()) {

/* do special revision C stuff here */

}

Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoOMAP3: Overo: Clean up pin mux and GPIO configuration
Dirk Behme [Thu, 12 Feb 2009 17:55:40 +0000 (18:55 +0100)]
OMAP3: Overo: Clean up pin mux and GPIO configuration

* Make Overo GPIO114 an input for touchscreen PENDOWN
* Make Overo GPIO144-147 readable
* Make Overo EHCI pinmux match beagle rev c setup
* Adjust pinmux for SMSC911X network chip support
* Remove unnecessary GPIO setup
* Fix merge error in Makefile

Signed-off-by: Steve Sakoman <sakoman@gmail.com>
Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agonmdk8815: fix onenand support
Jean-Christophe PLAGNIOL-VILLARD [Sun, 22 Feb 2009 16:08:41 +0000 (17:08 +0100)]
nmdk8815: fix onenand support

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agonomadik/nand: fix 'ecc512' discards qualifiers from pointer target type
Jean-Christophe PLAGNIOL-VILLARD [Sun, 22 Feb 2009 16:56:50 +0000 (17:56 +0100)]
nomadik/nand: fix 'ecc512' discards qualifiers from pointer target type

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agodavinci: fix implicit declaration of function 'davinci_errata_workarounds'
Jean-Christophe PLAGNIOL-VILLARD [Sun, 22 Feb 2009 16:49:43 +0000 (17:49 +0100)]
davinci: fix implicit declaration of function 'davinci_errata_workarounds'

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoarm: add uart dcc support
Jean-Christophe PLAGNIOL-VILLARD [Sun, 22 Feb 2009 14:49:28 +0000 (15:49 +0100)]
arm: add uart dcc support

Serial driver via the EmbeddedICE macrocell's DCC channel using
co-processor 14.

It does include a timeout to ensure that the system does not
totally freeze when there is nothing connected to read.

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoARM DaVinci: Add common peripherals and modules enable functions.
Hugo Villeneuve [Fri, 21 Nov 2008 19:35:56 +0000 (14:35 -0500)]
ARM DaVinci: Add common peripherals and modules enable functions.

Taken all the duplicated code for enabling common modules and apply
software workarounds from the board specific code into common
functions. Also added comments explaining the workarounds
(from TI errata documents) and replaced some numerical bit numbers
with more meaningful defines.

Signed-off-by: Hugo Villeneuve <hugo.villeneuve@lyrtech.com>
15 years agoEnable Ethernet for Nomadik 8815 Evaluation Kit
Alessandro Rubini [Mon, 9 Feb 2009 14:53:33 +0000 (15:53 +0100)]
Enable Ethernet for Nomadik 8815 Evaluation Kit

This trivially enables Ethernet support in the debug board
by setting up the proper chip select.

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stnwireless.com>
15 years agoNand driver for Nomadik SoC
Alessandro Rubini [Mon, 9 Feb 2009 14:53:31 +0000 (15:53 +0100)]
Nand driver for Nomadik SoC

This driver implements the ECC algorithm described in
the CPU data sheet and uses the OOB layout chosen in
already-released development systems (shipped with a custom-made
u-boot 1.3.1).

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stnwireless.com>
15 years agoAdded nomadik.h header
Alessandro Rubini [Mon, 9 Feb 2009 14:53:31 +0000 (15:53 +0100)]
Added nomadik.h header

Signed-off-by: Alessandro Rubini <rubini@unipv.it>
Acked-by: Andrea Gallo <andrea.gallo@stnwireless.com>
15 years agor8a66597-hcd: fix cannot use external hub
Yoshihiro Shimoda [Wed, 29 Oct 2008 11:05:18 +0000 (20:05 +0900)]
r8a66597-hcd: fix cannot use external hub

Fix the problem that cannot use external hub, because this driver
did not control correctly a DEVADDx register.

Signed-off-by: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Signed-off-by: Remy Bohmer <linux@bohmer.net>
15 years agoBlackfin: disable syscontrol code for now
Mike Frysinger [Thu, 19 Feb 2009 06:20:27 +0000 (01:20 -0500)]
Blackfin: disable syscontrol code for now

Looks like the initcode updates fell out of order during my merges.  The
patch that really fixes up this code is part of power-on overhaul and so
is too large for merging at this point.  Instead, we can disable the code
as no currently in-tree board depends on it.  The next merge window will
fix things up properly.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoBlackfin: bf537-stamp: fix I2C board defines
Mike Frysinger [Thu, 19 Feb 2009 06:19:49 +0000 (01:19 -0500)]
Blackfin: bf537-stamp: fix I2C board defines

The previous merge for cleaning up the I2C driver incorrectly reverted the
CFG_xxx rename for some of the I2C defines.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Heiko Schocher <hs@denx.de>
15 years agoCoding Style cleanup; update CHANGELOG
Wolfgang Denk [Sun, 22 Feb 2009 00:19:52 +0000 (01:19 +0100)]
Coding Style cleanup; update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoMakefile: fix cleanup
Wolfgang Denk [Sun, 22 Feb 2009 00:17:47 +0000 (01:17 +0100)]
Makefile: fix cleanup

Commit e4943ec5 moved the ARM boards to a vendor directory but forgot
to adapt the cleanup rules in the Makefile

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoCleanup the comment for m68k linux boot argument passing.
Richard Retanubun [Fri, 20 Feb 2009 18:01:56 +0000 (13:01 -0500)]
Cleanup the comment for m68k linux boot argument passing.

This patch clarifies the way m68k passes linux boot argument.
The one gotcha here is that the assembly instruction that
the compiler uses to jump to the kernel is 'jsr' which pushes the
program counter for the instruction after the jsr into the stack pointer.

Signed-off-by: Richard Retanubun <RichardRetanubun@RuggedCom.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agosh: Fix rsk7203 in tree build
Peter Griffin [Tue, 10 Feb 2009 16:44:45 +0000 (16:44 +0000)]
sh: Fix rsk7203 in tree build

Signed-off-by: Peter Griffin <pgriffin@mpc-data.co.uk>
15 years agobootm: Reduce the unnecessary memmove
Minkyu Kang [Wed, 18 Feb 2009 00:05:52 +0000 (09:05 +0900)]
bootm: Reduce the unnecessary memmove

Although load address and image start address are same address,
bootm command always does memmove.
That is unnecessary memmove and can be taken few milliseconds
(about 500 msec to 1000 msec).
If skip this memmove, we can reduce the boot time.

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
15 years agofpga: Fix Spartan III FPGA booting
Matthias Fuchs [Sun, 15 Feb 2009 21:29:15 +0000 (22:29 +0100)]
fpga: Fix Spartan III FPGA booting

This patch does some minor fixing of the Xilinx Spartan III
FPGA boot code:

- Fixed call order of post configuration callback and
  success message printing (result of copy-paste?)
- remove obsolete comment
- minor coding style cleanup

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
15 years agofpga: Fix Spartan II FPGA booting
Matthias Fuchs [Sun, 15 Feb 2009 21:28:36 +0000 (22:28 +0100)]
fpga: Fix Spartan II FPGA booting

This patch does some minor fixing of the Xilinx Spartan II
FPGA boot code:

- Fixed call order of post configuration callback and
  success message printing (result of copy-paste?)
- relocate post configuration callback only when it
  is implemented
- remove obsolete comment
- minor coding style cleanup

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
15 years agodrivers/serial/ns16550: move ifdef into Makefile COBJS-$(...)
Mike Frysinger [Thu, 12 Feb 2009 01:26:52 +0000 (20:26 -0500)]
drivers/serial/ns16550: move ifdef into Makefile COBJS-$(...)

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agolcd_putc bug fix for tab.
Derek Ou [Tue, 3 Feb 2009 23:00:07 +0000 (16:00 -0700)]
lcd_putc bug fix for tab.

Signed-off-by: Derek Ou <dou@siconix.com>
15 years agoMIPS: cpu/mips/Makefile: Add a missing START line
Shinya Kuribayashi [Mon, 2 Feb 2009 00:46:21 +0000 (09:46 +0900)]
MIPS: cpu/mips/Makefile: Add a missing START line

In the commit 79b51ff8205f0354d5300570614c1d2db499679c ([MIPS] cpu/mips/
Makefile: Split [CS]OBJS onto separate lines), I wrongly deleted a START
line.  This patch puts it back.

Signed-off-by: Shinya Kuribayashi <shinya.kuribayashi@necel.com>
15 years agoEnable ext2 support for TQM8xxL/M based boards
Wolfgang Denk [Sat, 21 Feb 2009 20:51:21 +0000 (21:51 +0100)]
Enable ext2 support for TQM8xxL/M based boards

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoARM:PXA Use new definitions in mmc.h
Tom Rix [Fri, 20 Feb 2009 02:47:50 +0000 (03:47 +0100)]
ARM:PXA Use new definitions in mmc.h

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoReduce the scope of PXA's mmc_read/mmc_write/mmc_bread functions
Andy Fleming [Fri, 20 Feb 2009 02:47:50 +0000 (03:47 +0100)]
Reduce the scope of PXA's mmc_read/mmc_write/mmc_bread functions

These names are being taken over by the new MMC framework.  Hopefuly
the PXA can be easily ported, and these functions will go away entirely.

Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agopxa: move mmc drivers to drivers/mmc
Jean-Christophe PLAGNIOL-VILLARD [Fri, 20 Feb 2009 02:47:50 +0000 (03:47 +0100)]
pxa: move mmc drivers to drivers/mmc

introduce new macro CONFIG_PXA_MMC to activate it

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agoARM:PXA Remove redefinition of mmc_cid and mmc_csd.
Tom Rix [Fri, 20 Feb 2009 01:27:22 +0000 (19:27 -0600)]
ARM:PXA Remove redefinition of mmc_cid and mmc_csd.

These structures are defined in the common mmc.h

This was compile checked on cerf250.

15 years agopxa: fixing get_timer to return time in miliseconds.
Micha Kalfon [Wed, 11 Feb 2009 17:50:11 +0000 (19:50 +0200)]
pxa: fixing get_timer to return time in miliseconds.

Fixing the get_timer function to return time in miliseconds instead of
ticks. Also fixed PXA boards to use the conventional value of 1000 for
CONFIG_SYS_HZ.

Signed-off-by: Micha Kalfon <smichak.uv@gmail.com>
15 years agoARM:OMAP3 Change mmc_init to mmc_legacy_init
Tom Rix [Thu, 19 Feb 2009 22:45:43 +0000 (16:45 -0600)]
ARM:OMAP3 Change mmc_init to mmc_legacy_init

omap3_mmc.c was changed to define mmc_legacy_init.
Remove unused functions.

Compile tested on all arm
Runtime tested on Zoom1.

Signed-off-by: Tom Rix <Tom.Rix@windriver.com>
15 years ago82xx, mgcoge: updates for 2009.03
Heiko Schocher [Thu, 19 Feb 2009 16:23:58 +0000 (17:23 +0100)]
82xx, mgcoge: updates for 2009.03

- activate CS4 for accessing the FPGA
- activate Rx buf len > 1 on SMC
- pram activated
- MTDPARTS_DEFAULT defined
- update the size of the flashes in the DTS
  before booting Linux
- MONITOR_LEN updated to 384k
- added CONFIG_HOSTNAME
- added CONFIG_ENV_BUFFER_PRINT
- Environment size reduced to 16k

Signed-off-by: Heiko Schocher <hs@denx.de>
15 years ago8xx, mgsuvd: updates for 2009.03
Heiko Schocher [Thu, 19 Feb 2009 16:24:01 +0000 (17:24 +0100)]
8xx, mgsuvd: updates for 2009.03

- activate Rx buf len > 1 on SMC
- pram activated
- MTDPARTS_DEFAULT defined
- update the size of the flash in the DTS
  before booting Linux
- MONITOR_LEN updated to 384k
- added CONFIG_HOSTNAME
- added CONFIG_ENV_BUFFER_PRINT
- Environment size reduced to 16k

Signed-off-by: Heiko Schocher <hs@denx.de>
15 years agoMMC: Don't use new framework code if not enabled
Dirk Behme [Wed, 18 Feb 2009 18:59:39 +0000 (19:59 +0100)]
MMC: Don't use new framework code if not enabled

Don't use code of new MMC framework in cmd_mmc if CONFIG_GENERIC_MMC
isn't enabled.

Signed-off-by: Dirk Behme <dirk.behme@googlemail.com>
15 years agoTQM8xxL: make some room in low memory for future needs
Wolfgang Denk [Thu, 19 Feb 2009 12:53:29 +0000 (13:53 +0100)]
TQM8xxL: make some room in low memory for future needs

THe TQM8xxL use a ahnd-optimized linker script to efficiently use the
small boot sectors in the flash. This patch makes some room in the
first sector to prepare for a size increase of lib_generic/vsprintf.o
by a future patch.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoREADME: remove duplicate entry
Kim Phillips [Thu, 19 Feb 2009 00:06:18 +0000 (18:06 -0600)]
README: remove duplicate entry

it's been around since the original commit (2ad6b513) that added two
identical entries.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Wed, 18 Feb 2009 23:50:08 +0000 (00:50 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

15 years agoMerge branch 'master' of git://git.denx.de/u-boot-arm
Wolfgang Denk [Wed, 18 Feb 2009 23:46:32 +0000 (00:46 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-arm

15 years agoCoding style cleanup, update CHANGELOG
Wolfgang Denk [Wed, 18 Feb 2009 23:41:08 +0000 (00:41 +0100)]
Coding style cleanup, update CHANGELOG

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agosata_sil3114: fix compiler warning
Kim Phillips [Wed, 18 Feb 2009 23:43:59 +0000 (17:43 -0600)]
sata_sil3114: fix compiler warning

judging from other printfs in the same file, it seems ata should be
postpended with the interface number, not the address of the global
port variable.  Fixes this for current u-boot-mpc83xx tree:

Configuring for MPC8349ITX board...
sata_sil3114.c: In function 'sata_bus_softreset':
sata_sil3114.c:99: warning: format '%u' expects type 'unsigned int', but argument 2 has type 'struct sata_port *'
sata_sil3114.c:108: warning: format '%u' expects type 'unsigned int', but argument 2 has type 'struct sata_port *'

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agoCreate configuration option for restricted ns16550 functions
Ron Madrid [Wed, 18 Feb 2009 22:30:44 +0000 (14:30 -0800)]
Create configuration option for restricted ns16550 functions

This patch will create a configuration option for a minimum configuration for
the ns16550 serial driver at drivers/serial/ns16550.c and will apply this new
configuration option to the SIMPC8313.h config file in order to fix the NAND
bootstrap build error.  This option will exclude all functions with exception of
NS16550_putc and NS16550_init.  This will be used primarily to save space and
remove unused code from builds in which space is limited.

Signed-off-by: Ron Madrid <ron_madrid@sbcglobal.net>
15 years agomkconfig: include board config.h before asm/config.h
Kim Phillips [Wed, 18 Feb 2009 22:14:29 +0000 (16:14 -0600)]
mkconfig: include board config.h before asm/config.h

swapping the include order suppresses warnings for board configs
that define their own CONFIG_MAX_MEM_MAPPED:

In file included from /home/r1aaha/git/u-boot/include/config.h:5,
                from /home/r1aaha/git/u-boot/include/common.h:35,
                from simpc8313.c:26:
/home/r1aaha/git/u-boot/include/configs/SIMPC8313.h:81:1: warning:
"CONFIG_MAX_MEM_MAPPED" redefined
In file included from /home/r1aaha/git/u-boot/include/config.h:4,
                from /home/r1aaha/git/u-boot/include/common.h:35,
                from simpc8313.c:26:
/home/r1aaha/git/u-boot/include/asm/config.h:28:1: warning: this is
the location of the previous definition

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
15 years agoFix build errors after making flash_get_info() non-static
Wolfgang Denk [Wed, 18 Feb 2009 20:35:38 +0000 (21:35 +0100)]
Fix build errors after making flash_get_info() non-static

Fix for these build problems:
error: static declaration of 'flash_get_info' follows non-static declaration

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years agoppc4xx: PCIe: Change 16GB inbound memory to 4GB
Stefan Roese [Wed, 18 Feb 2009 12:18:00 +0000 (13:18 +0100)]
ppc4xx: PCIe: Change 16GB inbound memory to 4GB

This patch fixes a problem recently seen on some 4xx platforms. For
example on Kilauea PCIe slot #0.

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Some more PMC405 coding-style cleanup
Stefan Roese [Wed, 18 Feb 2009 13:05:37 +0000 (14:05 +0100)]
ppc4xx: Some more PMC405 coding-style cleanup

Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Update PMC405 board support
Matthias Fuchs [Sun, 15 Feb 2009 21:27:47 +0000 (22:27 +0100)]
ppc4xx: Update PMC405 board support

This patch prepares the good old PMC405 board support for
upcoming PMC405V2 patches.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoppc4xx: Cleanup PMC405 board support
Matthias Fuchs [Sun, 15 Feb 2009 21:26:54 +0000 (22:26 +0100)]
ppc4xx: Cleanup PMC405 board support

This patch fixes coding style for PMC405 board support.
Also some unneeded features/code is removed.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Signed-off-by: Stefan Roese <sr@denx.de>
15 years agoqong: changes to Dave/DENX Qong configuration
Ilya Yanok [Thu, 5 Feb 2009 03:08:20 +0000 (04:08 +0100)]
qong: changes to Dave/DENX Qong configuration

1. Changes to the default environment:
  - "bootcmd" defined as "run flash_self"
  - "saveenv" command removed from "update"
  - "uboot" changed to "u-boot" (also in "load")
  - "addmtd" variable defined (and added to all boot commands)
2. CONFIG_CMD_JFFS2 defined to enable "mtdparts" command
3. MTDIDS_DEFAULT and MTDPARTS_DEFAULT defined
4. CONFIG_SYS_CBSIZE changed from 256 to 512. That solves the problem
with truncated "bootargs" environment variable.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
15 years agocommon/console: avoid ifdef CONFIG_CONSOLE_MUX when it's possible
Jean-Christophe PLAGNIOL-VILLARD [Sun, 1 Feb 2009 16:07:52 +0000 (17:07 +0100)]
common/console: avoid ifdef CONFIG_CONSOLE_MUX when it's possible

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agocommon/console: coding style cleanup
Jean-Christophe PLAGNIOL-VILLARD [Sun, 1 Feb 2009 16:07:51 +0000 (17:07 +0100)]
common/console: coding style cleanup

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
15 years agompc8xx_pcmcia: move CONFIG_8xx out of .c file and into Makefile
Mike Frysinger [Fri, 30 Jan 2009 01:02:23 +0000 (20:02 -0500)]
mpc8xx_pcmcia: move CONFIG_8xx out of .c file and into Makefile

Move the CONFIG_8xx mpc8xx_pcmcia.c protection out of the C file and
into the Makefile so we avoid pointless compiling of the file.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agodisk: convert part_* files to COBJ-$(CONFIG_XXX) style
Mike Frysinger [Fri, 30 Jan 2009 01:02:07 +0000 (20:02 -0500)]
disk: convert part_* files to COBJ-$(CONFIG_XXX) style

Move the CONFIG_XXX out of the part_XXX.c file and into Makefile to
avoid pointless compiles.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agoinclude/image.h: Ease grepping of image_* functions
Petri Lehtinen [Thu, 29 Jan 2009 08:35:40 +0000 (10:35 +0200)]
include/image.h: Ease grepping of image_* functions

Because the functions have been defined using macros, grepping for
their definitions is not possible. This patch adds the real function
names in comments.

Signed-off-by: Petri Lehtinen <petri.lehtinen@inoi.fi>
Acked-by: Mike Frysinger <vapier@gentoo.org>
15 years agorename CONFIG_CMD_ENV to CONFIG_CMD_SAVEENV
Mike Frysinger [Thu, 29 Jan 2009 00:08:14 +0000 (19:08 -0500)]
rename CONFIG_CMD_ENV to CONFIG_CMD_SAVEENV

The CONFIG_CMD_ENV option controls enablement of the `saveenv` command
rather than a generic "env" command, or anything else related to the
environment.  So, let's make sure the define is named accordingly.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
15 years agodisable imls command if no flash is defined
Valeriy Glushkov [Fri, 23 Jan 2009 18:02:17 +0000 (20:02 +0200)]
disable imls command if no flash is defined

Default CONFIG_CMD_IMLS must be disabled when CONFIG_SYS_NO_FLASH is defined

Signed-off-by: Valeriy Glushkov <gvv@lstec.com>
15 years agoAPI: Improve glue mid-layer of the API demo application.
Rafal Jaworowski [Fri, 23 Jan 2009 12:27:18 +0000 (13:27 +0100)]
API: Improve glue mid-layer of the API demo application.

- Extend ub_dev_read() and ub_dev_recv() so they return the length actually
read, which allows for better control and error handling (this introduces
additional error code API_ESYSC returned by the glue mid-layer).

- Clean up definitions naming and usage.

- Other minor cosmetics.

Note these changes do not touch the API proper, so the interface between
U-Boot and standalone applications remains unchanged.

Signed-off-by: Rafal Jaworowski <raj@semihalf.com>
15 years agoAPI: Only output test data when reading was successful.
Rafal Jaworowski [Fri, 23 Jan 2009 12:27:17 +0000 (13:27 +0100)]
API: Only output test data when reading was successful.

Signed-off-by: Rafal Czubak <rcz@semihalf.com>
15 years agoAPI: Provide syscall entry point for the ARM architecture.
Rafal Jaworowski [Fri, 23 Jan 2009 12:27:16 +0000 (13:27 +0100)]
API: Provide syscall entry point for the ARM architecture.

Signed-off-by: Rafal Czubak <rcz@semihalf.com>
Acked-by: Rafal Jaworowski <raj@semihalf.com>
15 years agoAPI: Use stack pointer as API signature search hint in the glue layer.
Rafal Jaworowski [Fri, 23 Jan 2009 12:27:15 +0000 (13:27 +0100)]
API: Use stack pointer as API signature search hint in the glue layer.

De-hardcode range in RAM we search for the API signature. Instead use the stack
pointer as a hint to narrow down the range in which the signature could reside
(it is malloc'ed on the U-Boot heap, and is hoped to remain in some proximity
from stack area). Adjust PowerPC code in API demo to the new scheme.

Signed-off-by: Rafal Czubak <rcz@semihalf.com>
Signed-off-by: Rafal Jaworowski <raj@semihalf.com>
15 years agoTQM8260: fix locations of kernel and ramdisk images in flash
Wolfgang Denk [Tue, 17 Feb 2009 09:26:38 +0000 (10:26 +0100)]
TQM8260: fix locations of kernel and ramdisk images in flash

After introducing redundant environment the kernel images was
overlapping with environment.

Signed-off-by: Wolfgang Denk <wd@denx.de>
15 years ago83xx: Add eSDHC support on 8379 EMDS board
Andy Fleming [Thu, 30 Oct 2008 21:50:14 +0000 (16:50 -0500)]
83xx: Add eSDHC support on 8379 EMDS board

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years ago85xx: Add eSDHC support for 8536 DS
Andy Fleming [Thu, 30 Oct 2008 21:51:33 +0000 (16:51 -0500)]
85xx: Add eSDHC support for 8536 DS

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years agoAdd support for the Freescale eSDHC found on 8379 and 8536 SoCs
Andy Fleming [Thu, 30 Oct 2008 21:47:16 +0000 (16:47 -0500)]
Add support for the Freescale eSDHC found on 8379 and 8536 SoCs

This uses the new MMC framework

Some contributions by Dave Liu <daveliu@freescale.com>

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years agoAdd MMC Framework
Andy Fleming [Thu, 30 Oct 2008 21:41:01 +0000 (16:41 -0500)]
Add MMC Framework

Here's a new framework (based roughly off the linux one) for managing
MMC controllers.  It handles all of the standard SD/MMC transactions,
leaving the host drivers to implement only what is necessary to
deal with their specific hardware.

This also hooks the infrastructure into the PowerPC board code
(similar to how the ethernet infrastructure now hooks in)

Some of this code was contributed by Dave Liu <daveliu@freescale.com>

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years agoEliminated arch-specific mmc header requirement
Andy Fleming [Thu, 30 Oct 2008 21:31:39 +0000 (16:31 -0500)]
Eliminated arch-specific mmc header requirement

The current MMC infrastructure relies on the existence of an
arch-specific header file.  This isn't necessary, and a couple
drivers were forced to implement dummy files to meet this requirement.
Instead, we move the stuff in those header files into a more appropriate
place, and eliminate the stubs and the #include of asm/arch/mmc.h

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years agoConvert mmc_init to mmc_legacy_init
Andy Fleming [Thu, 30 Oct 2008 21:21:00 +0000 (16:21 -0500)]
Convert mmc_init to mmc_legacy_init

This is to get it out of the way of incoming MMC framework

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years agoEliminate support for using MMC as memory
Andy Fleming [Thu, 30 Oct 2008 21:19:25 +0000 (16:19 -0500)]
Eliminate support for using MMC as memory

MMC cards are not memory, so we stop treating them that way.

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years ago32bit BUg fix for DDR2 on 8572
Poonam_Aggrwal-b10812 [Sun, 4 Jan 2009 03:16:38 +0000 (08:46 +0530)]
32bit BUg fix for DDR2 on 8572

This errata fix is required for 32 bit DDR2 controller on 8572.
May  also be required for P10XX20XX platforms

Signed-off-by: Poonam_Agarwal-b10812 <b10812@lc1106.zin33.ap.freescale.net>
15 years agoTQM85xx: Fix a couple warnings in TQM8548 build
Andy Fleming [Mon, 16 Feb 2009 15:40:20 +0000 (09:40 -0600)]
TQM85xx: Fix a couple warnings in TQM8548 build

The ecm variable in sdram.c was being declared for all 8548, but only
used by specific 8548 boards, so we make that variable require those
specific boards, too

The nand code was using an index "i" into a table, and then re-using "i"
to set addresses for each upm.  However, then it relied on the old value
of i still being there to enable things.  Changed the second "i" to "j"

Signed-off-by: Andy Fleming <afleming@freescale.com>
15 years agoMPC85xx: TQM8548: workaround for erratum DDR 19 and 20
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:26 +0000 (18:38 +0100)]
MPC85xx: TQM8548: workaround for erratum DDR 19 and 20

This patch adds the workaround for erratum DDR20 according to MPC8548
Device Errata document, Rev. 1: "CKE signal may not function correctly
after assertion of HRESET". Furthermore, the bug DDR19 is fixed in
processor version 2.1 and the work-around must be removed.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years agoMPC85xx: TQM8548: use cache for AG and BE variants
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:25 +0000 (18:38 +0100)]
MPC85xx: TQM8548: use cache for AG and BE variants

This patch makes accesses to the system memory cachable by removing the
caching-inhibited and guarded flags from the relevant TLB entries for
the TQM8548_BE and TQM8548_AG modules. FYI, the Freescale MPC85* boards
are configured similarly.

This results in a big averall performace improvement. TFTP downloads,
NAND Flash accesses, kernel boots, etc. are much faster.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years agoMPC85xx: TQM8548_AG: add 1 GiB DDR2-SDRAM configuration
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:24 +0000 (18:38 +0100)]
MPC85xx: TQM8548_AG: add 1 GiB DDR2-SDRAM configuration

This patch add support for the 1 GiB DDR2-SDRAM on the TQM8548_AG
module.

Signed-off-by: Jens Gehrlein <sew_s@tqs.de>
Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years agoMPC85xx: TQM8548: fix SDRAM timing for 533 MHz
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:23 +0000 (18:38 +0100)]
MPC85xx: TQM8548: fix SDRAM timing for 533 MHz

According to new TQM8548 timing specification:
Refresh Recovery: 34 -> 53 clocks
CKE pulse width:  1 -> 3 cycles
Window for four activities: 13 -> 14 cycles

Signed-off-by: Jens Gehrlein <sew_s@tqs.de>
Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years agoMPC85xx: TQM8548: add support for the TQM8548_AG module
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:22 +0000 (18:38 +0100)]
MPC85xx: TQM8548: add support for the TQM8548_AG module

The TQM8548_AG is a variant of the TQM8548 module with 1 GiB memory,
CAN and without PCI/PCI-X and RTC. U-Boot can be built for this module
with "$ make TQM8548_AG_config".

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years agoMPC85xx: TQM8548: add support for the TQM8548_BE module
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:21 +0000 (18:38 +0100)]
MPC85xx: TQM8548: add support for the TQM8548_BE module

The TQM8548_BE is a variant of the TQM8548 module with NAND and CAN
interface. With NAND support, the image is significantly larger and
TEXT_BASE is adjusted accordingly. U-Boot can be built for this
module with "$ make TQM8548_BE_config".

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years agoMPC85xx: TQM85xx: make standard PCI/PCI-X configurable
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:20 +0000 (18:38 +0100)]
MPC85xx: TQM85xx: make standard PCI/PCI-X configurable

The TQM8548_AG module does not have the standard PCI/PCI-X interface
connected but just the PCI Express interface . So far it was not
possible to disable it without disabling the complete PCI interface
(CONFIG_PCI) including PCI Express.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years agoMPC85xx: TQM85xx: fix flash protection for boot loader
Wolfgang Grandegger [Wed, 11 Feb 2009 17:38:19 +0000 (18:38 +0100)]
MPC85xx: TQM85xx: fix flash protection for boot loader

As the reset vector is located at 0xfffffffc, all flash sectors from the
beginning of the U-Boot binary to 0xffffffff must be protected. On the
TQM8548-AG having small sectors at the end of the flash it happened that
the last two sector were not protected and an "erase all" left an
un-bootable system behind:

Bank # 2: CFI conformant FLASH (32 x 16)  Size: 32 MB in 270 Sectors
  AMD Standard command set, Manufacturer ID: 0xEC, Device ID: 0x257E
  Erase timeout: 8192 ms, write timeout: 1 ms

  FFFA0000 E RO   FFFC0000   RO   FFFE0000   RO   FFFE4000   RO   FFFE8000   RO
  FFFEC000   RO   FFFF0000   RO   FFFF4000   RO   FFFF8000 E      FFFFC000

The same bug seems to be in drivers/mtd/cfi_flash.c:flash_init() and many
board BSPs as well.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
15 years ago86xx: Update CPU info output on bootup
Peter Tyser [Fri, 6 Feb 2009 20:30:40 +0000 (14:30 -0600)]
86xx: Update CPU info output on bootup

- Update style of 86xx CPU information on boot to more closely
  match 85xx boards
- Fix detection of 8641/8641D
- Use strmhz() to display frequencies
- Display L1 information
- Display L2 cache size
- Fixed CPU/SVR version output

== Before ==
Freescale PowerPC
CPU:
    Core: E600 Core 0, Version: 0.2, (0x80040202)
    System: Unknown, Version: 2.1, (0x80900121)
    Clocks: CPU:1066 MHz, MPX: 533 MHz, DDR: 266 MHz, LBC: 133 MHz
    L2: Enabled
Board: X-ES XPedite5170 3U VPX SBC

== After ==
CPU:   8641D, Version: 2.1, (0x80900121)
Core:  E600 Core 0, Version: 2.2, (0x80040202)
Clock Configuration:
       CPU:1066.667 MHz, MPX:533.333 MHz
       DDR:266.667 MHz (533.333 MT/s data rate), LBC:133.333 MHz
L1:    D-cache 32 KB enabled
       I-cache 32 KB enabled
L2:    512 KB enabled
Board: X-ES XPedite5170 3U VPX SBC

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years ago86xx: Update Global Utilities structure
Peter Tyser [Thu, 5 Feb 2009 17:25:24 +0000 (11:25 -0600)]
86xx: Update Global Utilities structure

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years ago86xx: Reset update
Peter Tyser [Thu, 5 Feb 2009 17:25:25 +0000 (11:25 -0600)]
86xx: Reset update

Update the 86xx reset sequence to try executing a board-specific reset
function.  If the board-specific reset is not implemented or does not
succeed, then assert #HRESET_REQ.  Using #HRESET_REQ is a more standard
reset procedure than the previous method and allows all board
peripherals to be reset if needed.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
15 years agofsl-ddr: Allow system to boot if we have more than 4G of memory
Kumar Gala [Wed, 11 Feb 2009 05:53:40 +0000 (23:53 -0600)]
fsl-ddr: Allow system to boot if we have more than 4G of memory

Previously if we >=4G of memory and !CONFIG_PHYS_64BIT we'd report
an error and hang.  Instead of doing that since DDR is mapped in the
lowest priority LAWs we setup the DDR controller and the max amount
of memory we report back is what we can map (CONFIG_MAX_MEM_MAPPED)

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Becky Bruce <beckyb@kernel.crashing.org>