oweals/u-boot.git
14 years agox86: Place global data below stack before entering C
Graeme Russ [Thu, 7 Oct 2010 09:03:29 +0000 (20:03 +1100)]
x86: Place global data below stack before entering C

By reserving space for the Global Data immediately below the stack during
assembly level initialisation, the C declaration of the static global data
can be removed, along with the 'RAM Bootstrap' function. This results in
cleaner code, and the ability to pass boot-up flags from assembler into C

14 years agox86: Dont clobber %eax after getting memory size
Graeme Russ [Thu, 7 Oct 2010 09:03:28 +0000 (20:03 +1100)]
x86: Dont clobber %eax after getting memory size

By using another register, reduce code size by one instruction

14 years agox86: Don't clobber %ebx
Graeme Russ [Thu, 7 Oct 2010 09:03:28 +0000 (20:03 +1100)]
x86: Don't clobber %ebx

%ebx will hold low-level boot flags and must be preserved

14 years agox86: Remove usage of %ebp as a return pointer
Graeme Russ [Thu, 7 Oct 2010 09:03:27 +0000 (20:03 +1100)]
x86: Remove usage of %ebp as a return pointer

Using %ebp as a return pointer prevents creating 'load anywhere' images

14 years agox86: Move ECC initialisation outside RAM initialisation
Graeme Russ [Thu, 7 Oct 2010 09:03:26 +0000 (20:03 +1100)]
x86: Move ECC initialisation outside RAM initialisation

To allow for 'load anywhere' images, the %ebp return pointer 'hack' must
be removed, so we cannot have two 'calls' to get_mem_size

14 years agox86: Remove progress indication in low-level init
Graeme Russ [Thu, 7 Oct 2010 09:03:24 +0000 (20:03 +1100)]
x86: Remove progress indication in low-level init

Progress indication is not relocation friendly so remove it in
preperation for full relocatability support

14 years agox86: Fix %ss and %esp in register structure for interrupts
Graeme Russ [Thu, 7 Oct 2010 09:03:23 +0000 (20:03 +1100)]
x86: Fix %ss and %esp in register structure for interrupts

14 years agox86: Change compiler options
Graeme Russ [Thu, 7 Oct 2010 09:03:23 +0000 (20:03 +1100)]
x86: Change compiler options

Change to:
 - reparam=3
 - no-from-pointer
 - no-stack-protector
 - preferred-stack-boundary=2
 - no-top-level-reorder

These options make the code a little smaller and faster

14 years agox86: Coding Style Cleanup
Graeme Russ [Thu, 7 Oct 2010 09:03:21 +0000 (20:03 +1100)]
x86: Coding Style Cleanup

Perform some basic code cleanups of the x86 files

14 years agox86: Move loading of GTD to C code
Graeme Russ [Thu, 7 Oct 2010 09:03:21 +0000 (20:03 +1100)]
x86: Move loading of GTD to C code

Linux has C macros and code to load the GTD after switching to Protected
Mode. Using these greatly simplifies the assembler code

14 years agox86: use gc sections to reduce image size
Graeme Russ [Thu, 7 Oct 2010 09:03:20 +0000 (20:03 +1100)]
x86: use gc sections to reduce image size

Follow the discussion of Charles Manning and Mike Frysinger.
Using gc_sections helps reduce image size.

14 years agox86: zboot update
Graeme Russ [Thu, 7 Oct 2010 09:03:19 +0000 (20:03 +1100)]
x86: zboot update

The header of recent Linux Kernels includes the size of the image, and
therefore is not needed to be passed to zboot. Still process the third
parameter (size of image) in the event that an older kernel is being loaded

14 years agox86: Use TEXT_BASE in linker scripts
Graeme Russ [Thu, 7 Oct 2010 09:03:18 +0000 (20:03 +1100)]
x86: Use TEXT_BASE in linker scripts

Use TEXT_BASE rather than a hard-coded base address on x86 linker scripts.
This will allow any board to define its base link address without having
to modify the linker script

14 years agox86: Make CONFIG_RELOC_FIXUP_WORKS generic for all x86 boards
Graeme Russ [Thu, 7 Oct 2010 09:03:18 +0000 (20:03 +1100)]
x86: Make CONFIG_RELOC_FIXUP_WORKS generic for all x86 boards

Relocation is not board-specific for the x86 architectrure, so
CONFIG_RELOC_FIXUP_WORKS can be defined globally in the config.h

14 years agox86: Remove bi_env from do_bdinfo
Graeme Russ [Thu, 7 Oct 2010 09:03:17 +0000 (20:03 +1100)]
x86: Remove bi_env from do_bdinfo

Commit 55e97429d1e6cf0976711e4e0f29ea924b7e5917 removed the definition
from /arch/i386/include/asm/u-boot.h but not its usage in do_bdinfo()

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-blackfin
Wolfgang Denk [Tue, 5 Oct 2010 12:42:32 +0000 (14:42 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-blackfin

14 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Tue, 5 Oct 2010 12:37:25 +0000 (14:37 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

14 years agoMerge branch 'next' of git://git.denx.de/u-boot-video
Wolfgang Denk [Tue, 5 Oct 2010 12:31:48 +0000 (14:31 +0200)]
Merge branch 'next' of git://git.denx.de/u-boot-video

14 years agoppc44x: config GPIOs for USB on canyonlands board
Rupjyoti Sarmah [Fri, 1 Oct 2010 09:01:28 +0000 (14:31 +0530)]
ppc44x: config GPIOs for USB on canyonlands board

The GPIO 16 and 19 reconfiguration should be done once USB is initialized.
So moved the reconfiguration to the USB init  function.

Signed-off-by: Rupjyoti Sarmah <rsarmah@apm.com>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Remove some testing hacks from ppc4xx.h
Stefan Roese [Wed, 29 Sep 2010 14:59:57 +0000 (16:59 +0200)]
ppc4xx: Remove some testing hacks from ppc4xx.h

I accidentally left these hacks in the code while doing the big header
cleanup. Let's remove it now.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Use common ns16550 functions in 4xx UART POST driver
Stefan Roese [Wed, 29 Sep 2010 14:58:38 +0000 (16:58 +0200)]
ppc4xx: Use common ns16550 functions in 4xx UART POST driver

This patch changes the PPC4xx POST UART driver to use the common
NS16550 functions for receiving and sending. Additionally the
local function for SoC divisor setup are removed. Instead the
functions from arch/powerpc/cpu/ppc4xx/4xx_uart.c are used. This
removes code duplication.

Also the common CONFIG_SYS_NS16550_COMx defines are now used
to describe the POST UART's.

And a compile breakage is fixed, introduced by a git merge of
the ppc4xx/next branch into master. Now "ppc4xx.h" is moved to
"asm/ppc4xx.h". Fixed as well with this patch.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Small whitespace cleanup in canyonlands.c
Stefan Roese [Tue, 28 Sep 2010 06:06:06 +0000 (08:06 +0200)]
ppc4xx: Small whitespace cleanup in canyonlands.c

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx/fdt/flash: Fix bug in fdt_fixup_nor_flash_node()
Stefan Roese [Fri, 24 Sep 2010 11:51:50 +0000 (13:51 +0200)]
ppc4xx/fdt/flash: Fix bug in fdt_fixup_nor_flash_node()

This patch fixes a bug in fdt_fixup_nor_flash_node() when the reg
property has multiple reg tuples, like:

reg = <0 0x00000000 0x04000000
       0 0x04000000 0x04000000>;

In this case this function did not update the reg property correctly.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Add defines for COM3 & COM4 (UART2 & UART3) on 440EPx/GRx
Stefan Roese [Fri, 24 Sep 2010 09:20:41 +0000 (11:20 +0200)]
ppc4xx: Add defines for COM3 & COM4 (UART2 & UART3) on 440EPx/GRx

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Big lwmon5 board support rework/update
Sascha Laue [Thu, 19 Aug 2010 07:38:56 +0000 (09:38 +0200)]
ppc4xx: Big lwmon5 board support rework/update

This patch brings the lwmon5 board support up-to-date. Here a
summary of the changes:

lwmon5 board port related:
- GPIO's changed to control the LSB transmitter
- Reset USB PHY's upon power-up
- Enable CAN upon power-up
- USB init error workaround (errata CHIP_6)
- EBC: Enable burstmode and modify the timings for the GDC memory
- EBC: Speed up NOR flash timings

lwmon5 board POST related:
- Add FPGA memory test
- Add GDC memory test
- DSP POST reworked
- SYSMON POST: Fix handling of negative temperatures
- Add output for sysmon1 POST
- HW-watchdog min. time test reworked

Additionally some coding-style changes were done.

Signed-off-by: Sascha Laue <sascha.laue@liebherr.com>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoAPM821xx: Add bluestone board support
Tirumala Marri [Tue, 28 Sep 2010 21:15:21 +0000 (14:15 -0700)]
APM821xx: Add bluestone board support

Add support code for bluestone board wth APM821XX processor based.
This patch includes early board init, misc init, configure EBC,
initializes UIC, MAKEALL, board.cfg and MAINTAINERS file.

Signed-off-by: Tirumala R Marri <tmarri@apm.com
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoAPM821xx: Add CPU support
Tirumala Marri [Tue, 28 Sep 2010 21:15:14 +0000 (14:15 -0700)]
APM821xx: Add CPU support

APM821XX is a new line of SoCs which are derivatives of
PPC44X family of processors. This patch adds support of CPU, cache,
tlb, 32k ocm, bootstraps, PLB and AHB bus.

Signed-off-by: Tirumala R Marri <tmarri@apm.com>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agohmi1001, mucmc52, uc100, uc101: move boards to vendor directory
Wolfgang Denk [Wed, 18 Aug 2010 12:14:17 +0000 (14:14 +0200)]
hmi1001, mucmc52, uc100, uc101: move boards to vendor directory

Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Heiko Schocher <hs@denx.de>
Cc: Stefan Roese <sr@denx.de>
Cc: Roderik Wildenburg <roderik.wildenburg@manroland.com>
14 years agoCONFIG_SYS_ARM_WITHOUT_RELOC: document feature removal
Wolfgang Denk [Wed, 29 Sep 2010 09:25:55 +0000 (11:25 +0200)]
CONFIG_SYS_ARM_WITHOUT_RELOC: document feature removal

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoBlackfin: otp: fix build after constification of args[]
Mike Frysinger [Sat, 2 Oct 2010 18:31:32 +0000 (14:31 -0400)]
Blackfin: otp: fix build after constification of args[]

The OTP code does a little shuffling of arguments that aren't really
necessary, so use a local variable instead to fix build errors now
that the args[] parameter is const.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf537-pnav: increase monitor len
Mike Frysinger [Fri, 1 Oct 2010 23:42:08 +0000 (19:42 -0400)]
Blackfin: bf537-pnav: increase monitor len

Building this board for parallel flash fills up the bss section and thus
fails to link, so bump up the monitor size a bit.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf527-ad7160-eval: fix GPIO CS define
Mike Frysinger [Fri, 1 Oct 2010 20:59:19 +0000 (16:59 -0400)]
Blackfin: bf527-ad7160-eval: fix GPIO CS define

Rather than use a hardcoded "7", use the new Blackfin global define.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf548-ezkit: bump SPI flash size up
Mike Frysinger [Wed, 29 Sep 2010 20:24:16 +0000 (20:24 +0000)]
Blackfin: bf548-ezkit: bump SPI flash size up

The current size used (256KiB) is smaller than the LDR created for
the bf548-ezkit, so 'run update' doesn't work correctly.  So bump
up the size a bit by making this flexible per-board config.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: fix MMC init output alignment
Mike Frysinger [Wed, 29 Sep 2010 18:46:42 +0000 (14:46 -0400)]
Blackfin: fix MMC init output alignment

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: blackvme: new board port
Wojtek Skulski [Mon, 27 Sep 2010 22:21:59 +0000 (18:21 -0400)]
Blackfin: blackvme: new board port

The board includes:
 * ADSP-BF561 rev. 0.5
 * 32-bit SDRAM (2 * MT48LC16M16A2TG or MT48LC32M16A2TG)
 * Gigabit Ether AX88180 (ASIX) + 88E1111 rev. B2 (Marvell)
 * SPI  boot flash on PF2 (M25P64 8MB, or M25P128 16 MB)
 * FPGA boot flash on PF3 (M25P64 8MB, or M25P128 16 MB)
 * Spartan6-LX150 (memory-mapped; both PPIs also connected)
 * See http://www.skutek.com/

Signed-off-by: Wojtek Skulski <skulski@pas.rochester.edu>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: propagate target cpu defines when building embedded env
Mike Frysinger [Sun, 26 Sep 2010 07:00:38 +0000 (07:00 +0000)]
Blackfin: propagate target cpu defines when building embedded env

Since we're no longer extracting the env from the target ELF file (since
upstream wouldn't take that change), we're back to the problem of cpu
defines not properly propagating to the env setup stage.  So the embedded
env built by the host compiler doesn't match the one that is linked into
the u-boot env.

Reported-by: Vivi Li <vivi.li@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: cmd_gpio: accept upper case pin names
Mike Frysinger [Fri, 24 Sep 2010 19:34:08 +0000 (15:34 -0400)]
Blackfin: cmd_gpio: accept upper case pin names

The intention all along was to accept pin names irrelevant of their case.
But I guess I forgot to test/implement support for that.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf526-ezbrd: enable BootROM-OOB layout when booting from NAND
Mike Frysinger [Tue, 21 Sep 2010 23:47:27 +0000 (19:47 -0400)]
Blackfin: bf526-ezbrd: enable BootROM-OOB layout when booting from NAND

We need to use the Blackfin BootROM-specific OOB layout when we boot out
of NAND as that is what the on-chip ROM expects.

Also need to increase the monitor size a little to accommodate the extra
NAND code overhead.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: adi config: enable nand lock/unlock support
Mike Frysinger [Tue, 21 Sep 2010 23:33:53 +0000 (19:33 -0400)]
Blackfin: adi config: enable nand lock/unlock support

We use the lock/unlock options in our default nand code, so enabl
support for the options.

Reported-by: Vivi Li <vivi.li@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bct-brettl2: new board port
Peter Meerwald [Mon, 20 Sep 2010 18:08:57 +0000 (14:08 -0400)]
Blackfin: bct-brettl2: new board port

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: adi config: allow boards to tweak a little more
Mike Frysinger [Mon, 20 Sep 2010 19:14:58 +0000 (15:14 -0400)]
Blackfin: adi config: allow boards to tweak a little more

Let people easily override bootdelay and network settings.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: cmd_gpio: return gpio value to caller
Mike Frysinger [Mon, 20 Sep 2010 17:46:29 +0000 (13:46 -0400)]
Blackfin: cmd_gpio: return gpio value to caller

Make the GPIO command usable in a scripting environment by returning
the GPIO value rather than always 0.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf527-sdp: new board port
Mike Frysinger [Sun, 19 Sep 2010 20:28:40 +0000 (16:28 -0400)]
Blackfin: bf527-sdp: new board port

Support for the Blackfin System Development Platform (SDP) base module.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: adi config: add a hook for boards to append the env
Mike Frysinger [Sun, 19 Sep 2010 20:27:24 +0000 (16:27 -0400)]
Blackfin: adi config: add a hook for boards to append the env

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: adi config: add an "all spi flashes" option to unify board lists
Mike Frysinger [Sun, 19 Sep 2010 20:26:55 +0000 (16:26 -0400)]
Blackfin: adi config: add an "all spi flashes" option to unify board lists

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: support a 3rd gpio cfi pin
Peter Meerwald [Fri, 17 Sep 2010 17:02:42 +0000 (13:02 -0400)]
Blackfin: support a 3rd gpio cfi pin

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: cmd_gpio: document/extend input sub-option
Mike Frysinger [Fri, 13 Aug 2010 19:48:09 +0000 (15:48 -0400)]
Blackfin: cmd_gpio: document/extend input sub-option

The input sub command was missing from the help text, and it didn't show
the actual value currently read on the GPIO.  This allows people to read
the value of input pins.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: move CONFIG_BFIN_CPU to board config.mk
Mike Frysinger [Mon, 9 Aug 2010 21:57:47 +0000 (17:57 -0400)]
Blackfin: move CONFIG_BFIN_CPU to board config.mk

The CONFIG_BFIN_CPU option is largely used in the build system, so move
it out of the board config.h and into the board config.mk.  It'd be nice
to keep everything in the config.h, but the patch to extract that value
early was rejected.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf561: merge headers
Mike Frysinger [Mon, 26 Jul 2010 05:27:17 +0000 (01:27 -0400)]
Blackfin: bf561: merge headers

Only the BF561 port was using the common dual core headers, so merge them
into the BF561 specific headers.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf533: merge headers
Mike Frysinger [Mon, 26 Jul 2010 05:06:37 +0000 (01:06 -0400)]
Blackfin: bf533: merge headers

Only the BF533 port was using the common extended headers, so merge them
into the BF533 specific headers.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: unify core MMRs
Mike Frysinger [Mon, 26 Jul 2010 05:10:35 +0000 (01:10 -0400)]
Blackfin: unify core MMRs

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf52x: unify arch header duplication
Mike Frysinger [Mon, 26 Jul 2010 05:05:41 +0000 (01:05 -0400)]
Blackfin: bf52x: unify arch header duplication

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf537: unify arch header duplication
Mike Frysinger [Mon, 26 Jul 2010 05:04:38 +0000 (01:04 -0400)]
Blackfin: bf537: unify arch header duplication

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: punt old *p style volatile MMR defines
Mike Frysinger [Sun, 25 Jul 2010 20:39:45 +0000 (16:39 -0400)]
Blackfin: punt old *p style volatile MMR defines

These are unused now, so punt them to clean things up.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf527-ad7160-eval: convert from old style MMR macros
Mike Frysinger [Sun, 25 Jul 2010 21:19:36 +0000 (17:19 -0400)]
Blackfin: bf527-ad7160-eval: convert from old style MMR macros

The old MMR defines are being scrubbed, so convert the driver to use the
new standard helper macros.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: cm-bf548: video: convert from old style MMR macros
Mike Frysinger [Sun, 25 Jul 2010 21:19:16 +0000 (17:19 -0400)]
Blackfin: cm-bf548: video: convert from old style MMR macros

The old MMR defines are being scrubbed, so convert the driver to use the
new standard helper macros.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf548-ezkit: video: convert from old style MMR macros
Mike Frysinger [Sun, 25 Jul 2010 21:19:00 +0000 (17:19 -0400)]
Blackfin: bf548-ezkit: video: convert from old style MMR macros

The old MMR defines are being scrubbed, so convert the driver to use the
new standard helper macros.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf527-ezkit: video: convert from old style MMR macros
Mike Frysinger [Sun, 25 Jul 2010 21:18:42 +0000 (17:18 -0400)]
Blackfin: bf527-ezkit: video: convert from old style MMR macros

The old MMR defines are being scrubbed, so convert the driver to use the
new standard helper macros.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bf537-stamp: post: update old MMR style macros
Mike Frysinger [Sun, 25 Jul 2010 21:16:50 +0000 (17:16 -0400)]
Blackfin: bf537-stamp: post: update old MMR style macros

The old MMR defines are being scrubbed, so convert the driver to use the
new standard helper macros.

For the GPIO MMR usage, convert to the new GPIO framework.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bfin_mac: convert from old style MMR macros
Mike Frysinger [Sun, 25 Jul 2010 20:38:12 +0000 (16:38 -0400)]
Blackfin: bfin_mac: convert from old style MMR macros

The old MMR defines are being scrubbed, so convert the driver to use the
new standard helper macros.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: punt headers for non-existent BF541
Mike Frysinger [Sun, 25 Jul 2010 20:17:45 +0000 (16:17 -0400)]
Blackfin: punt headers for non-existent BF541

There is no BF541 processor variant, so punt headers for it.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: unify gpio cfi implementations
Mike Frysinger [Sat, 24 Jul 2010 07:54:19 +0000 (03:54 -0400)]
Blackfin: unify gpio cfi implementations

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bfin_sdh: clean up send_cmd
Mike Frysinger [Mon, 21 Jun 2010 20:56:54 +0000 (20:56 +0000)]
Blackfin: bfin_sdh: clean up send_cmd

Simplify the command setup and status checking steps, and add a proper
timeout to the status polling code to avoid possible infinite hangs.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: bfin_spi: add optional DMA support
Mike Frysinger [Wed, 5 May 2010 04:56:30 +0000 (00:56 -0400)]
Blackfin: bfin_spi: add optional DMA support

This moves the last piece from the old spi_flash driver to the new SPI
framework -- optional DMA RX support.  This typically cuts speeds by ~40%
at the cost of additional ~300 bytes.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: add support for BF51x parts
Mike Frysinger [Sun, 12 Oct 2008 01:55:21 +0000 (21:55 -0400)]
Blackfin: add support for BF51x parts

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoBlackfin: add support for BF538/BF539 processors
Mike Frysinger [Thu, 7 Aug 2008 17:10:41 +0000 (13:10 -0400)]
Blackfin: add support for BF538/BF539 processors

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
14 years agoMerge branch 'next' of /home/wd/git/u-boot/next
Wolfgang Denk [Tue, 28 Sep 2010 21:30:47 +0000 (23:30 +0200)]
Merge branch 'next' of /home/wd/git/u-boot/next

Conflicts:
include/ppc4xx.h

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoPrepare v2010.09 v2010.09
Wolfgang Denk [Tue, 28 Sep 2010 21:20:55 +0000 (23:20 +0200)]
Prepare v2010.09

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agompc512x: fix build issues
Wolfgang Denk [Tue, 28 Sep 2010 21:02:05 +0000 (23:02 +0200)]
mpc512x: fix build issues

Commit 800eb0964 "POST cleanup." removed file
arch/powerpc/cpu/mpc512x/common.c but failed to remove the reference
to it from arch/powerpc/cpu/mpc512x/Makefile which causes somewhat
obscure build errors:

make[1]: *** No rule to make target `/work/wd/tmp-ppc/arch/powerpc/cpu/mpc512x/.depend', needed by `_depend'.  Stop.

Fix these.

Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoARMV7: OMAP3: Update Beagle xM pinmux with USB hub and DVI gpio setup
Steve Sakoman [Mon, 20 Sep 2010 16:55:24 +0000 (09:55 -0700)]
ARMV7: OMAP3: Update Beagle xM pinmux with USB hub and DVI gpio setup

This patch adds missing pinmux setup for 4 GPIO signals used on the Beagle xM:
 - USB hub reset (gpio_56)
 - P8 USB hub reset (gpio_63)
 - DVI enable (gpio_129)
 - P8 DVI enable (gpio_170)

Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
14 years agoARMV7: OMAP4: Calculate SDRAM size
Aneesh V [Sun, 12 Sep 2010 05:02:55 +0000 (10:32 +0530)]
ARMV7: OMAP4: Calculate SDRAM size

Calculate the SDRAM size from DMM configuration registers instead of using
hard-coded values. This gives correct values for all different boards.

It's assumed that DMM sections do not overlap memory areas.

Signed-off-by: Aneesh V <aneesh@ti.com>
Tested-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
14 years agoARMV7: OMAP4: Fix Panda pinmux setting to enable Wifi/BT Module
Steve Sakoman [Wed, 22 Sep 2010 14:30:11 +0000 (07:30 -0700)]
ARMV7: OMAP4: Fix Panda pinmux setting to enable Wifi/BT Module

This patch corrects the pinmux settings to enable proper functioning
of the wifi/bluetooth module.

Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Sandeep Paulraj <s-paulraj@ti.com>
14 years agoixp/npe: Remove duplicated comment
Thomas Weber [Tue, 28 Sep 2010 12:03:32 +0000 (14:03 +0200)]
ixp/npe: Remove duplicated comment

Signed-off-by: Thomas Weber <weber@corscience.de>
14 years agobmw: Remove duplicated include of header file
Thomas Weber [Tue, 28 Sep 2010 12:03:33 +0000 (14:03 +0200)]
bmw: Remove duplicated include of header file

Signed-off-by: Thomas Weber <weber@corscience.de>
14 years agoAdd support for operating system OSE
Torkel Lundgren [Tue, 28 Sep 2010 09:05:36 +0000 (11:05 +0200)]
Add support for operating system OSE

Add OSE as operating system for mkimage and bootm.

Signed-off-by: Torkel Lundgren <torkel.lundgren@enea.com>
14 years agoREADME: Fix description of version numbering scheme
Thomas Weber [Tue, 28 Sep 2010 06:06:25 +0000 (08:06 +0200)]
README: Fix description of version numbering scheme

The version numbering scheme was changed in Oct, 2008.
This patch brings the documentation to the actual level.
The description is taken from:
http://www.denx.de/wiki/U-Boot/ReleaseCycle

Signed-off-by: Thomas Weber <weber@corscience.de>
Changed text slightly.
Signed-off-by: Wolfgang Denk <wd@denx.de>
14 years agoFix "ubi part" cmd re-entrancy
Karl Beldan [Thu, 23 Sep 2010 08:46:31 +0000 (10:46 +0200)]
Fix "ubi part" cmd re-entrancy

Commit 2ee951ba (UBI: Enable re-initializing of the "ubi part" command)
reset mtd_devs in ubi_exit() but missed ubi_init()'s failure path.

Signed-off-by: Karl Beldan <karl.beldan@gmail.com>
Cc: Stefan Roese <sr@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoatmel_lcd: Allow contrast polarity to be either positive or negative
Alexander Stein [Tue, 20 Jul 2010 06:55:40 +0000 (08:55 +0200)]
atmel_lcd: Allow contrast polarity to be either positive or negative

Signed-off-by: Alexander Stein <alexander.stein@systec-electronic.com>
14 years agop1022ds: use weak CFI flash accessors when DIU is enabled
Timur Tabi [Thu, 16 Sep 2010 21:35:44 +0000 (16:35 -0500)]
p1022ds: use weak CFI flash accessors when DIU is enabled

On the Freescale P1022, the DIU and the LBC share address pins, which means
that when the DIU is active (e.g. the console is on the DVI display), NOR flash
cannot be accessed.  So we use the weak accessor function feature of the CFI
flash code to temporarily switch the pin mux from DIU to LBC whenever we want
to read or write flash.  This has a significant performance penalty, but it's
the only way to make it work.

This change allows the 'saveenv' command to work when the video display is
enabled.  Erasing flash and writing to flash (with the 'cp' command) works,
but reading from flash (with the 'md' and 'cp' commands) does not.  Also, while
flash is being written, the video display will be blank.

Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
14 years agop1022ds: add video support
Timur Tabi [Thu, 23 Sep 2010 23:25:53 +0000 (01:25 +0200)]
p1022ds: add video support

Add support for the DIU controller.  If CONFIG_VIDEO is defined, then
the console will appear on a DVI monitor instead of the serial port.

Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
14 years agofsl_diu_fb: further refactoring of FSL DIU code
Anatolij Gustschin [Thu, 23 Sep 2010 23:06:37 +0000 (01:06 +0200)]
fsl_diu_fb: further refactoring of FSL DIU code

Move common code to the fsl_diu_fb.c file and remove obsolete
code from board files (aria, mpc8610hpcd and pdm360ng).
Move fsl_diu_fb.h file to the include directory.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
14 years agoMerge branch 'master' of git://git.denx.de/u-boot-ppc4xx
Wolfgang Denk [Thu, 23 Sep 2010 19:16:32 +0000 (21:16 +0200)]
Merge branch 'master' of git://git.denx.de/u-boot-ppc4xx

14 years ago83xx: Remove warmboot parameter from PCI init functions
Peter Tyser [Wed, 15 Sep 2010 00:13:50 +0000 (19:13 -0500)]
83xx: Remove warmboot parameter from PCI init functions

This change lays the groundwork for the BOOTFLAG_* flags being removed.

This change has the small affect of delaying 100ms on PCI initialization
after a warm boot as opposed to the optimal 1ms on some boards.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
included the mpc8308_p1m board.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agompc83xx: extend CONFIG_SYS_BOOTMAPSZ increase to mpc8308_p1m
Kim Phillips [Wed, 22 Sep 2010 20:36:27 +0000 (15:36 -0500)]
mpc83xx: extend CONFIG_SYS_BOOTMAPSZ increase to mpc8308_p1m

continuation of commit 39da1ba923d55f316f9f1bb3a960e4ed91dc17ac:
"e300: increase CONFIG_SYS_BOOTMAPSZ to allow booting large kernels"

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agompc83xx: fix pcie build warning
Kim Phillips [Wed, 22 Sep 2010 20:31:01 +0000 (15:31 -0500)]
mpc83xx: fix pcie build warning

Configuring for MPC8308RDB board...
pcie.c: In function 'mpc83xx_pcie_register_hose':
pcie.c:143: warning: assignment makes pointer from integer without a cast

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agompc8308_p1m: support for MPC8308 P1M board
Ilya Yanok [Fri, 17 Sep 2010 21:41:50 +0000 (23:41 +0200)]
mpc8308_p1m: support for MPC8308 P1M board

This patch provides support for MPC8308 P1M board with the following
set of features:
 Dual UART is supported
 NOR flash is supported
 Both TSEC Ethernet controllers are supported
 PCI Express initialization is supported
 Both I2C controllers are supported

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agoMPC8308RDB: various clean ups
Ilya Yanok [Fri, 17 Sep 2010 21:41:49 +0000 (23:41 +0200)]
MPC8308RDB: various clean ups

This patch cleans up the Freescale MPC8308RDB Development board support.
Things fixed:
 - Removed unused PCIE2 definitions from configuration
 - SICR{L,H} defines used for System I/O Configuration Registers values
   instead of hardcoding
 - CONFIG_SYS_SCCR_PCIEXP1CM used to enable PCIE clock instead of
   writing to SCCR from the board code
 - sleep mode stuff removed as MPC8308 has no support for deep sleep and
   PMCCR1 register. board_early_init_f() removed.
 - MPC8308 has no ERRATA for DDR controller so workaround removed
 - 'assignment in if statement' issues solved
 - use LBLAWAR_* defines instead of hardcoding

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agompc8308: add SICR{L,H} fields definitions
Ilya Yanok [Fri, 17 Sep 2010 21:41:48 +0000 (23:41 +0200)]
mpc8308: add SICR{L,H} fields definitions

This patch adds defines to set supported fields in System I/O
Configuration Registers High and Low on Freescale MPC8308 CPU.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agompc83xx: add support for setting PCIE clocks
Ilya Yanok [Fri, 17 Sep 2010 21:41:47 +0000 (23:41 +0200)]
mpc83xx: add support for setting PCIE clocks

This patch adds support for setting PCIE clocks in cpu_init.c by
providing CONFIG_SYS_SCCR_PCIEXP{1,2} in configuration.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agompc83xx/pcie: make it compile with PCIE2 unconfigured
Ilya Yanok [Fri, 17 Sep 2010 21:41:46 +0000 (23:41 +0200)]
mpc83xx/pcie: make it compile with PCIE2 unconfigured

MPC8308 has only one PCIE host controller so we want it to compile
without CONFIG_SYS_PCIE2_CFG_{BASE,SIZE} defined.

Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agoe300: increase CONFIG_SYS_BOOTMAPSZ to allow booting large kernels
Ira W. Snyder [Fri, 10 Sep 2010 22:42:32 +0000 (15:42 -0700)]
e300: increase CONFIG_SYS_BOOTMAPSZ to allow booting large kernels

Newer Linux kernels can overrun the initial memory window used for
booting with their BSS area. When this happens, they overwrite the FDT
and silently fail to boot.

On e300 CPUs, the Linux kernel uses an initial BAT covering the first
256MB of RAM. See arch/powerpc/kernel/head_32.S for details. Increase
the value of CONFIG_SYS_BOOTMAPSZ to accommodate the maximum value
allowed by Linux. This will allow very large kernels to boot.

Signed-off-by: Ira W. Snyder <iws@ovro.caltech.edu>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agompc831xerdb: enable mtdparts for NAND
Scott Wood [Mon, 30 Aug 2010 23:04:52 +0000 (18:04 -0500)]
mpc831xerdb: enable mtdparts for NAND

The default partition table matches the .dts files for these boards in
Linux.  This allows these partitions to be used by name with U-Boot's
"nand" command.

Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
14 years agoppc4xx: Disable trace broadcast for 44x non debug mode
Victor Gallardo [Thu, 16 Sep 2010 18:32:04 +0000 (11:32 -0700)]
ppc4xx: Disable trace broadcast for 44x non debug mode

By default the trace broadcast is enabled on 44x systems.

To reduce power consumption when instruction tracing is
not needed, disable trace broadcast.

Check External Debug Mode (EDM) bit to detect if it should be
disabled or not.

Resetting system via a debugger will set the DBCR0[EDM] bit.
Resetting via u-boot or OS will not.

Signed-off-by: Victor Gallardo <vgallardo@apm.com>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Cleanup of PVR detection code in cpu.c
Stefan Roese [Fri, 3 Sep 2010 11:27:02 +0000 (13:27 +0200)]
ppc4xx: Cleanup of PVR detection code in cpu.c

This patch cleans the PVR detection code in check_cpu() up a bit.
Basically the strings are better seperated, resulting in an easier
to understand and maintain code version.

The #ifdef's couldn't be removed easily because of two reasons:

- Some SoC revisions have the same PVR, so need a way to differentiate
  between those two SoC's.
- In some case statements registers only available in this SoC variant
  are referenced.

Instead I moved the CONFIG_440 #ifdef a bit, so that 405 platforms don't
add this 440 detection code and vice versa. Resulting in this U-Boot
image size change:

405EX (Kilauea):     408 bytes less
440EPx (Sequoia):    604 bytes less
460EX (Canyonlands): 564 bytes less

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
14 years agoppx4xx: remove unused functionality for DU405 boards
Matthias Fuchs [Tue, 21 Sep 2010 07:38:04 +0000 (09:38 +0200)]
ppx4xx: remove unused functionality for DU405 boards

Remove some unused functionality to make U-Boot build again.
Especially PCI is not used on the board.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd.eu>
Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoRemove unused CONFIG_SERIAL_SOFTWARE_FIFO feature
Stefan Roese [Wed, 15 Sep 2010 07:33:25 +0000 (09:33 +0200)]
Remove unused CONFIG_SERIAL_SOFTWARE_FIFO feature

This patch removes the completely unused CONFIG_SERIAL_SOFTWARE_FIFO
feature from U-Boot. It has only been implemented for PPC4xx and was not
used at all. So let's remove it and make the code smaller and cleaner.

Signed-off-by: Stefan Roese <sr@denx.de>
Acked-by: Detlev Zundel <dzu@denx.de>
14 years agoppc4xx: Remove now unused CONFIG_UART1_CONSOLE
Stefan Roese [Tue, 21 Sep 2010 08:24:36 +0000 (10:24 +0200)]
ppc4xx: Remove now unused CONFIG_UART1_CONSOLE

CONFIG_UART1_CONSOLE was a PPC4xx specific implementation and is now
removed since the move from the 4xx UART driver to the common NS16550
UART driver. Let's remove all references to this define now.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Use common NS16550 driver for PPC4xx UART
Stefan Roese [Mon, 20 Sep 2010 14:05:31 +0000 (16:05 +0200)]
ppc4xx: Use common NS16550 driver for PPC4xx UART

This patch removes the PPC4xx UART driver. Instead the common NS16550
driver is used, since all PPC4xx SoC's use this peripheral device.

The file 4xx_uart.c now only implements the UART clock calculation
function which also sets the SoC internal UART divisors.

All PPC4xx board config headers are changed to use this common NS16550
driver now.

Tested on these boards:
acadia, canyonlands, katmai, kilauea, sequoia, zeus

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Big header cleanup part 2, mostly PPC405 related
Stefan Roese [Sun, 12 Sep 2010 04:21:37 +0000 (06:21 +0200)]
ppc4xx: Big header cleanup part 2, mostly PPC405 related

This cleanup is done by creating header files for all SoC versions and
moving the SoC specific defines into these special headers. This way the
common header ppc405.h and ppc440.h can be cleaned up finally.

As a part from this cleanup, the GPIO definitions for PPC405EP are
corrected. The high and low parts of the registers (for example
CONFIG_SYS_GPIO0_OSRL vs. CONFIG_SYS_GPIO0_OSRH) have been defined in
the wrong order. This patch now fixes this issue by switching these
xxxH and xxxL values. This brings the GPIO 405EP port in sync with all
other PPC4xx ports.

Signed-off-by: Stefan Roese <sr@denx.de>
14 years agoppc4xx: Big header cleanup, mostly PPC440 related
Stefan Roese [Sat, 11 Sep 2010 07:31:43 +0000 (09:31 +0200)]
ppc4xx: Big header cleanup, mostly PPC440 related

This patch starts a bit PPC4xx header cleanup. First patch mostly
touches PPC440 files. A later patch will touch the PPC405 files as well.

This cleanup is done by creating header files for all SoC versions and
moving the SoC specific defines into these special headers. This way the
common header ppc405.h and ppc440.h can be cleaned up finally.

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