project/bcm63xx/u-boot.git
12 years agoPPC: Drop mv6436x_eth_initialize() from net/eth.c
Marek Vasut [Fri, 2 Mar 2012 22:39:31 +0000 (22:39 +0000)]
PPC: Drop mv6436x_eth_initialize() from net/eth.c

This function was defined as an extern in net/eth.c, drop that and use
standard means of calling it.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Reinhard Arlt <reinhard.arlt@esd-electronics.com>
12 years agoConvert cmd_usage() calls in common to use a return value
Simon Glass [Sat, 10 Dec 2011 08:44:01 +0000 (08:44 +0000)]
Convert cmd_usage() calls in common to use a return value

Change all files in common/ to use CMD_RET_USAGE instead of calling
cmd_usage() directly. I'm not completely sure about this patch since
the code since impact is small (100 byte or so on ARM) and it might
need splitting into smaller patches. But for now here it is.

Signed-off-by: Simon Glass <sjg@chromium.org>
12 years agoAdd cmd_process() to process commands in one place
Simon Glass [Tue, 14 Feb 2012 19:59:25 +0000 (19:59 +0000)]
Add cmd_process() to process commands in one place

We currently have the same code in hush.c and main.c. This brings the
code into one place.

As an added feature, if the command function returns CMD_RET_USAGE then
cmd_process() will print a usage message for the command before
returning the standard failure code of 1.

ARM code size increases about 32 bytes with this clean-up.

Signed-off-by: Simon Glass <sjg@chromium.org>
12 years agoRemove interleave of non-U-Boot code in hush
Simon Glass [Tue, 6 Dec 2011 19:47:53 +0000 (19:47 +0000)]
Remove interleave of non-U-Boot code in hush

There is a nasty interleave of #ifdefs in hush.c where the two code
paths have different indents. Remove this ickiness.

Signed-off-by: Simon Glass <sjg@chromium.org>
12 years agoCreate a single cmd_call() function to handle command execution
Simon Glass [Tue, 14 Feb 2012 19:59:23 +0000 (19:59 +0000)]
Create a single cmd_call() function to handle command execution

We should aim for a single point of entry to the commands, whichever
parser is used.

Signed-off-by: Simon Glass <sjg@chromium.org>
12 years agoDon't include standard parser if hush is used
Simon Glass [Tue, 14 Feb 2012 19:59:22 +0000 (19:59 +0000)]
Don't include standard parser if hush is used

This saves about 1KB of code space on ARM with CONFIG_SYS_HUSH_PARSER
defined.

Signed-off-by: Simon Glass <sjg@chromium.org>
12 years agoStop using builtin_run_command()
Simon Glass [Tue, 14 Feb 2012 19:59:21 +0000 (19:59 +0000)]
Stop using builtin_run_command()

Boards can select either the 'built-in' parser or the hush parser. We
should not call builtin_run_command() if we are using the hush parser.
We use run_command() instead, since it knows how to call the correct
parser.

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoRename run_command2() to run_command()
Simon Glass [Tue, 14 Feb 2012 19:59:20 +0000 (19:59 +0000)]
Rename run_command2() to run_command()

This is a more sensible name, so rename it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoRename run_command() to builtin_run_command()
Simon Glass [Tue, 14 Feb 2012 19:59:19 +0000 (19:59 +0000)]
Rename run_command() to builtin_run_command()

The current run_command() is only one of the parsing options - the other
is hush. We should not call run_command() when the hush parser is being
used. So we rename this function to better explain its purpose.

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoRemove CMD_PXE's static on run_command()
Simon Glass [Tue, 14 Feb 2012 19:59:18 +0000 (19:59 +0000)]
Remove CMD_PXE's static on run_command()

It really isn't clear why this is here and there is no comment, so
drop it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoRemove execute permissions from source files
Thomas Weber [Mon, 13 Feb 2012 18:21:54 +0000 (19:21 +0100)]
Remove execute permissions from source files

Signed-off-by: Thomas Weber <weber@corscience.de>
12 years agointegrator: remove fragile delay loop from PCI code
Linus Walleij [Sat, 3 Mar 2012 20:21:13 +0000 (21:21 +0100)]
integrator: remove fragile delay loop from PCI code

The reference implementation of the PCI initialization code almost
everywhere contain this fragile loop of "a few usecs", and its
use of volatile variables to delay a number of bus cycles is indeed
uncertain.

Reading the manual "Integrator/AP Users Guide", page 5-15 it is
clearly stated:

"Wait until 230ms after the end of the reset period before
accessing V360EPC internal registers. The V360EPC supports the
use of a serial configuration PROM and the software must wait for
the device to detect the absence of this PROM before accessing any
registers. The required delay is a function of the PCI Clock, but
at the lower frequency (25MHz) is 230ms".

So let's simply wait 230ms per the spec.

This solves the compilation error that looked like this:
pci.c: In function ‘pci_init_board’:
pci.c:286:18: warning: variable ‘j’ set but not used

Reported-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
12 years agopci: move pciauto_config_init() to pci.h
Linus Walleij [Sat, 3 Mar 2012 10:05:08 +0000 (10:05 +0000)]
pci: move pciauto_config_init() to pci.h

Fixing build regressions for the Integrator I get find that a few
boards try to work around the missing declaration of
pciauto_config_init() by declaring it in the local scope. This
does not make sense when the sibling functions are in <pci.h>
so move the function to the header, ridding the build error
in the Integrator and getting rid of the local declarations
here and there.

Reported-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
12 years agoppc: Change memsz variable to signed char
Simon Glass [Thu, 5 Jan 2012 17:54:57 +0000 (17:54 +0000)]
ppc: Change memsz variable to signed char

This seems to be unsigned char for no good reason. Tidy this up and
remove the casts.

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agom68k: Change memsz to a signed char to avoid warning
Simon Glass [Thu, 5 Jan 2012 17:54:56 +0000 (17:54 +0000)]
m68k: Change memsz to a signed char to avoid warning

There doesn't seem to be any reason for using uchar here, so change it
to char. This fixes a warning:

pointer targets in passing argument 1 of 'sprintf' differ in signedness

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoUSB: relax usbcore reset timings
Wolfgang Grandegger [Wed, 21 Dec 2011 00:01:09 +0000 (00:01 +0000)]
USB: relax usbcore reset timings

Following the corresponding Linux code, this patch relaxes reset timings
waiting at least 100ms after power to the ports. There are some reports
that it helps make enumeration work better on some high speed devices.
Furthermore, the wait is only done once after power has been enabled
on all ports.

CC: Remy Bohmer <linux@bohmer.net>
CC: Vincent Palatin <vpalatin@chromium.org>
Signed-off-by: Wolfgang Grandegger <wg@denx.de>
12 years agoUSB: move keyboard polling into kbd driver
amartin@nvidia.com [Tue, 20 Dec 2011 14:56:16 +0000 (14:56 +0000)]
USB: move keyboard polling into kbd driver

This moves keyboard polling logic from USB HCD drivers into USB
keyboard driver.  Remove usb_event_poll() as keyboard polling was
the only user of this API.  With this patch USB keyboard works with
EHCI controllers again.  Tested on a tegra2 seaboard.

Signed-off-by: Allen Martin <amartin@nvidia.com>
12 years agoUSB: reevaluate iomux stdin on USB kbd detect
amartin@nvidia.com [Fri, 23 Dec 2011 10:29:48 +0000 (10:29 +0000)]
USB: reevaluate iomux stdin on USB kbd detect

If CONSOLE_MUX is enabled, reevaluate console stdin when USB keyboard
device is detected.

Signed-off-by: Allen Martin <amartin@nvidia.com>
12 years agousb: add numeric keypad support to HID driver
Vincent Palatin [Mon, 9 Jan 2012 12:59:36 +0000 (12:59 +0000)]
usb: add numeric keypad support to HID driver

When keys are pressed on the numeric keypad, emit key codes for the numbers,
operators, dot and enter.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
12 years agoUSB: fix USB keyboard polling parameter
Vincent Palatin [Mon, 9 Jan 2012 08:35:09 +0000 (08:35 +0000)]
USB: fix USB keyboard polling parameter

When doing a "GET_REPORT" request on the keyboard control endpoint,
the report ID should 0 (ie report ID not used) rather than 1
as reports are not used in boot mode.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
12 years agoeth: remove usb-ethernet devices before re-enumerating them
Vincent Palatin [Mon, 9 Jan 2012 08:32:36 +0000 (08:32 +0000)]
eth: remove usb-ethernet devices before re-enumerating them

Fix the crash when running several times usb_init() with a USB ethernet
device plugged.

Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Wolfgang Grandegger <wg@denx.de>
12 years agoUSB: Make struct devrequest setup_packet local
Marek Vasut [Mon, 13 Feb 2012 18:58:18 +0000 (18:58 +0000)]
USB: Make struct devrequest setup_packet local

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Cc: Remy Bohmer <linux@bohmer.net>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoUSB: Separate out USB hub driver
Marek Vasut [Mon, 13 Feb 2012 18:58:17 +0000 (18:58 +0000)]
USB: Separate out USB hub driver

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Cc: Remy Bohmer <linux@bohmer.net>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoUSB: Staticize internal functions
Marek Vasut [Mon, 13 Feb 2012 18:58:16 +0000 (18:58 +0000)]
USB: Staticize internal functions

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Cc: Remy Bohmer <linux@bohmer.net>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agoMerge branch 'master' of git://git.denx.de/u-boot-nand-flash
Wolfgang Denk [Sat, 3 Mar 2012 14:15:07 +0000 (15:15 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-nand-flash

* 'master' of git://git.denx.de/u-boot-nand-flash:
  cmd_nand.c: Fix 'nand dump' after latest MTD resync
  mtd/nand:Fix wrong usage of is_blank() in fsl_ifc_run_command
  mtd/nand: Fix IFC driver to support 2K NAND page
  nand: reinstate lazy bad block scanning
  Revert "nand: make 1-bit software ECC configurable"

12 years agoMerge branch 'master' of git://git.denx.de/u-boot-nds32
Wolfgang Denk [Sat, 3 Mar 2012 14:13:41 +0000 (15:13 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-nds32

* 'master' of git://git.denx.de/u-boot-nds32:
  nds32/board.c: add PCI prompt at boot up
  nds32/ag101/watchdog.S: add linkage support
  nds32: add linkage support

12 years agoEthernut 5: fix build error
prabhakar.csengg@gmail.com [Thu, 23 Feb 2012 01:53:40 +0000 (01:53 +0000)]
Ethernut 5: fix build error

Fix build error for ethernut5 board due to prototype change
for function board_mmc_getcd().
ethernut5.c:238: error: conflicting types for 'board_mmc_getcd'
u-boot/include/mmc.h:318: note: previous declaration of 'board_mmc_getcd'
was here
make[2]: *** [ethernut5.o] Error 1

Signed-off-by: Prabhakar Lad <prabhakar.csengg@gmail.com>
Cc: Tim Schendekehl <tim.schendekehl@egnite.de>
Cc: Thierry Reding <thierry.reding@avionic-design.de>
Cc: Albert ARIBAUD <albert.u.boot@aribaud.net>
Cc: egnite GmbH <info@egnite.de>
12 years agocmd_nand.c: Fix 'nand dump' after latest MTD resync
Tom Rini [Thu, 23 Feb 2012 22:47:46 +0000 (15:47 -0700)]
cmd_nand.c: Fix 'nand dump' after latest MTD resync

With 2a8e0fc nand_do_read_ops changed in behavior slightly (keeping in sync
with the kernel which did this change in b64d39d8) such that the OOB data is
always copied into oobbuf and never appended to datbuf.  Within U-Boot only
the nand_dump function (for the dump nand subcommand) was expecting the OOB
data to only be appended to datbuf.  So we now change nand_dump to not
malloc extra space, correct the comment about datbuf and OOB data and switch
the pointer to oobbuf before printing.

Cc: Scott Wood <scottwood@freescale.com>
Signed-off-by: Tom Rini <trini@ti.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
12 years agomtd/nand:Fix wrong usage of is_blank() in fsl_ifc_run_command
Prabhakar Kushwaha [Fri, 20 Jan 2012 13:08:14 +0000 (18:38 +0530)]
mtd/nand:Fix wrong usage of is_blank() in fsl_ifc_run_command

Freescale IFC NAND Machine calculates ECC on 512byte sector and same is used in
fsl_ifc_run_command() during ECC status verification. Also this sector is passed
to is_blank() for blank checking. It is wrong at first place because
is_blank()'s implementation checks for Page size and OOB area size.
is_blank() should be called per page for main and OOB area verification.

Variables name are redefined to avoid confusion between buffer and ecc sector.

Signed-off-by: Poonam Aggrwal <poonam.aggrwal@freescale.com>
Signed-off-by: Prabhakar Kushwaha <prabhakar@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
12 years agomtd/nand: Fix IFC driver to support 2K NAND page
Prabhakar Kushwaha [Fri, 20 Jan 2012 13:09:05 +0000 (18:39 +0530)]
mtd/nand: Fix IFC driver to support 2K NAND page

1) OOB area should be updated irrespective of NAND page size. Earlier it was
updated only for 512byte NAND page.

2) During OOB update fbcr should be equal to OOB size.

Signed-off-by: Poonam Aggrwal <poonam.aggrwal@freescale.com>
Signed-off-by: Prabhakar Kushwaha <prabhakar@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
12 years agonand: reinstate lazy bad block scanning
Scott Wood [Mon, 20 Feb 2012 20:50:39 +0000 (14:50 -0600)]
nand: reinstate lazy bad block scanning

commit 2a8e0fc8b3dc31a3c571e439fbf04b882c8986be ("nand: Merge changes
from Linux nand driver") accidentally reverted commit
13f0fd94e3cae6f8a0d9fba5d367e311edc8ebde ("NAND: Scan bad blocks
lazily.").

Reinstate the change, as amended by commit
ff49ea8977b56916edd5b1766d9939010e30b181 ("NAND: Mark the BBT as scanned
prior to calling scan_bbt.").

Signed-off-by: Scott Wood <scottwood@freescale.com>
12 years agoRevert "nand: make 1-bit software ECC configurable"
Scott Wood [Mon, 13 Feb 2012 21:46:27 +0000 (15:46 -0600)]
Revert "nand: make 1-bit software ECC configurable"

This reverts commit 4fee6c2f295f932b8febdc7ce8731ba045695fa5.

It breaks boards that currently rely on soft-ecc, as pointed out here:
http://patchwork.ozlabs.org/patch/140872/

The reverted patch should be resubmitted with documentation, and with the
CONFIG_MTD_ECC_SOFT selected from every board that needs it.  We could
start by looking at what NAND driver the board selects, and whether
that driver ever asks for soft ECC.

Signed-off-by: Scott Wood <scottwood@freescale.com>
12 years agonds32/board.c: add PCI prompt at boot up
Macpaul Lin [Fri, 25 Nov 2011 09:14:51 +0000 (17:14 +0800)]
nds32/board.c: add PCI prompt at boot up

add PCI prompt at boot up for probing PCI device

Signed-off-by: Macpaul Lin <macpaul@andestech.com>
12 years agonds32/ag101/watchdog.S: add linkage support
Macpaul Lin [Wed, 30 Nov 2011 08:01:28 +0000 (16:01 +0800)]
nds32/ag101/watchdog.S: add linkage support

Add linkage support to watchdog.S.

Signed-off-by: Macpaul Lin <macpaul@andestech.com>
12 years agonds32: add linkage support
Macpaul Lin [Wed, 30 Nov 2011 07:33:55 +0000 (15:33 +0800)]
nds32: add linkage support

Add linkage support.

Signed-off-by: Macpaul Lin <macpaul@andestech.com>
12 years agoarmv7: omap3: leave outer cache enabled
Aneesh V [Thu, 16 Feb 2012 03:40:15 +0000 (03:40 +0000)]
armv7: omap3: leave outer cache enabled

Mainline kernel for OMAP3 doesn't enable L2 cache
It expects L2$ to be enabled by ROM-code/bootloader.

Leaving L2$ enabled can be troublesome in cases where
the L2 cache is not under CP15 control, such as in
Cortex-A9. This problem is explained in detail in
the commit dc7100f4080952798413fb63bb4134b22c57623a

However, this problem doesn't apply to Cortex-A8
because L2$ in Cortex-A8 is under CP15 control and
hence the generic armv7 maintenance opertions work
for it.

As such we can make an exception for OMAP3 and
leave the L2$ enabled when we jump to kernel. This
is done by removing the strongly-linked implementation
of v7_outer_cache_disable() and allowing it to fall
back to the weakly linked implementation that doesn't
do anything.

Signed-off-by: Aneesh V <aneesh@ti.com>
12 years agoat91: modified NAND flash timing on otc570 board
Daniel Gorsulowski [Wed, 25 Jan 2012 03:19:50 +0000 (03:19 +0000)]
at91: modified NAND flash timing on otc570 board

Signed-off-by: Daniel Gorsulowski <Daniel.Gorsulowski@esd.eu>
12 years agoat91: modified NAND flash timing on meesc board
Daniel Gorsulowski [Wed, 25 Jan 2012 03:19:49 +0000 (03:19 +0000)]
at91: modified NAND flash timing on meesc board

Signed-off-by: Daniel Gorsulowski <Daniel.Gorsulowski@esd.eu>
12 years agott01: fix hush parser config, add release config
Helmut Raiger [Mon, 30 Jan 2012 04:21:55 +0000 (04:21 +0000)]
tt01: fix hush parser config, add release config

Add PREBOOT, SILENT_CONSOLE and DEVICE_NULLDEV for release
build. Fixed bug in CONFIG_SYS_HUSH_PARSER define.

Signed-off-by: Helmut Raiger <helmut.raiger@hale.at>
12 years agott01: add video support
Helmut Raiger [Wed, 15 Feb 2012 22:40:12 +0000 (22:40 +0000)]
tt01: add video support

The video setup for the Epson display is provided. Addtionally
some extra info is displayed next to the Linux logo.
Make get_cpu_rev() publicly available (added to sys_proto.h).

Signed-off-by: Helmut Raiger <helmut.raiger@hale.at>
12 years agott01: fix environment size
Helmut Raiger [Mon, 30 Jan 2012 04:21:53 +0000 (04:21 +0000)]
tt01: fix environment size

size of environment must match erasable block size in the flash.

Signed-off-by: Helmut Raiger <helmut.raiger@hale.at>
12 years agoi.mx6q: mx6qsabrelite: Update the network configuration
Troy Kisky [Tue, 7 Feb 2012 14:08:50 +0000 (14:08 +0000)]
i.mx6q: mx6qsabrelite: Update the network configuration

Define CONFIG_PHY_MICREL, and
minimize the tx clock delay.

There is an issue with 1000 baseTx mode on early revs
of the SabreLite boards. The center tap pin 9 of the mag RJ45
USB combo was connected to the 3.3 filtered supply. Letting
this pin float solved the problem. Symptoms of the problem
were packets with many extra zeroes tacked on the end, and random
bit flips causing a high rate of CRC errors. 10/100 baseTx worked
fine on all revs. To disable 1000 baseTx for these boards, simply
define the environment variable disable_giga. ie.

setenv disable_giga 1

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
12 years agonet: phy: make board_phy_config responsible for calling drv->config
Troy Kisky [Tue, 7 Feb 2012 14:08:49 +0000 (14:08 +0000)]
net: phy: make board_phy_config responsible for calling drv->config

Boards may have things they want done before or after normal phy config.
Letting the boards call drv->config allows them more flexibilty.
Boards affected by this change are corenet_ds and mpc8544ds.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
12 years agonet: phy: add support for Micrel's KSZ9021
Troy Kisky [Tue, 7 Feb 2012 14:08:48 +0000 (14:08 +0000)]
net: phy: add support for Micrel's KSZ9021

Add the gigabit phy KSZ9021.
Also, add function ksz9021_phy_extended_write
/_read for access to the phys extended registers.
The environment variable "disable_giga"
can be used to disable 1000baseTx.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
12 years agonet: fec_mxc: add PHYLIB support
Troy Kisky [Tue, 7 Feb 2012 14:08:47 +0000 (14:08 +0000)]
net: fec_mxc: add PHYLIB support

Surround non PHYLIB routines miiphy_restart_aneg
and miiphy_wait_aneg with ifndef CONFIG_PHYLIB.
When later PHYLIB is required it is easy to delete
the non-PHYLIB code. This separation idea
came from Andy Fleming.

fec_miiphy_read, and fec_miiphy_write changed to
fec_phy_read, and fec_phy_write with argument changes.

Deprecated miiphy_register is no longer used. Both
old and new PHYLIB code now use mdio_register.

Cleanup some debug prints.

Inline function fec_miiphy_fec_to_eth is no longer necessary.
Moved to the single call location.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
12 years agonet: fec_mxc: add 1000 Mbps selection
Troy Kisky [Tue, 7 Feb 2012 14:08:46 +0000 (14:08 +0000)]
net: fec_mxc: add 1000 Mbps selection

Define FEC_QUIRK_ENET_MAC and add to
arch-mx6/imx-regs.h

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
12 years agointegrator: update the Integrator/CP config
Linus Walleij [Tue, 31 Jan 2012 12:20:17 +0000 (12:20 +0000)]
integrator: update the Integrator/CP config

The Integrator/CP config file was very outdated and gives a
non functional U-Boot. So:

- Set the CONFIG_ARCH_CINTEGRATOR arch flag correctly
- Add proper core module configuration options
- Skip cluttered command lista and just include
  <config_cmd_default.h>
- Kill off #if 0 network section and insert some example
  configuration that simply just boots from the network
  using TFTP if available.
- Delete flash control registers that have no business
  in this config file. These are already moved down to
  board/armltd/integrator/integrator-sc.h

Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
12 years agointegrator: rewrite the AP PCI driver
Linus Walleij [Mon, 30 Jan 2012 13:49:34 +0000 (13:49 +0000)]
integrator: rewrite the AP PCI driver

The PCI support for the Integrator AP has apparently never
been finished and I strongly suspect that it has never worked,
so let's fix it. This is a list of the more or less
un-splittable changes done in this driver rewrite:

- Replace the register definitions stashed into the config
  file (!) with a copy if the register file from the Linux
  kernels arch/arm/include/asm/hardware/pci_v3.h

- Delete the unreadable gigantic macros that perform the
  config accesses and replace them with copyedited code from
  Linux arch/arm/mach-integrator/pci_v3.c

- Rewrite the rest of the setup code to use the
  v3_[read|write][lwb]() accessors.

- Enable PCI by default in the AP board configuration.

- Fix checkpatch warnings and make code more conformant.

Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
12 years agointegrator: stop calling pci_init() twice
Linus Walleij [Sat, 21 Jan 2012 13:13:19 +0000 (13:13 +0000)]
integrator: stop calling pci_init() twice

The U-Boot startup infrastructure already makes sure pci_init() is
called at a proper time, calling it again from within the board
setup code will not make things better.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
12 years agomcx: Enable command line editing
Detlev Zundel [Wed, 8 Feb 2012 04:49:02 +0000 (04:49 +0000)]
mcx: Enable command line editing

Signed-off-by: Detlev Zundel <dzu@denx.de>
CC: Stefano Babic <sbabic@denx.de>
12 years agodm6467Tevm: Use a common configuration file for davinci_dm6467evm and davinci_dm6467Tevm
prabhakar.csengg@gmail.com [Sun, 12 Feb 2012 21:38:22 +0000 (21:38 +0000)]
dm6467Tevm: Use a common configuration file for davinci_dm6467evm and davinci_dm6467Tevm

In commit 2d575e46859dd9127a9cec731ace77523e6ea2ab a separate header
file was introduced for the DM6467T EVM, include/configs/davinci_dm6467Tevm.h.
The substantial difference between the davinci_dm6467evm and the
davinci_dm6467Tevm configuration is a single bit in the hardware revision
that is passed to the Linux kernel and davinci_dm6467evm has
REFCLK_FREQ = 27000000 where as davinci_dm6467Tevm.h has a
REFCLK_FREQ = 33000000.

This patch removes include/configs/davinci_dm6467Tevm.h. Instead the
include/configs/davinci_dm6467evm.h configuration is used for DM6467T EVMs and
renamed CFG_REFCLK_FREQ to CONFIG_REFCLK_FREQ and CONFIG_REFCLK_FREQ is defined
in boards.cfg.

Signed-off-by: Prabhakar Lad <prabhakar.csengg@gmail.com>
Cc: Tom Rini <trini@ti.com>
12 years agoOMAP3: Remove unused CONFIG_SYS_NAND_ECCSTEPS/TOTAL
Thomas Weber [Mon, 13 Feb 2012 07:59:07 +0000 (07:59 +0000)]
OMAP3: Remove unused CONFIG_SYS_NAND_ECCSTEPS/TOTAL

Signed-off-by: Thomas Weber <weber@corscience.de>
12 years agoTricorder: Switch env to generic mmc driver
Thomas Weber [Mon, 13 Feb 2012 03:16:53 +0000 (03:16 +0000)]
Tricorder: Switch env to generic mmc driver

Because the tricorder uses the CONFIG_OMAP_HSMMC the
initialization command is mmc rescan and not mmc init.

Signed-off-by: Thomas Weber <weber@corscience.de>
Signed-off-by: Tom Rini <trini@ti.com>
12 years agoomap3_beagle: Change MAINTAINERS entry
Tom Rini [Mon, 13 Feb 2012 06:19:36 +0000 (06:19 +0000)]
omap3_beagle: Change MAINTAINERS entry

Cc: Dirk Behme <dirk.behme@gmail.com>
Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Dirk Behme<dirk.behme@gmail.com>
12 years agoam335x_evm: Change MAINTAINERS entry
Tom Rini [Mon, 13 Feb 2012 06:02:01 +0000 (06:02 +0000)]
am335x_evm: Change MAINTAINERS entry

Signed-off-by: Tom Rini <trini@ti.com>
12 years agoam33xx: ddr_defs.h: Change DDR timings
Chase Maupin [Mon, 13 Feb 2012 05:57:27 +0000 (05:57 +0000)]
am33xx: ddr_defs.h: Change DDR timings

* For cold silicon the DDR timings need to be relaxed in order for
  the device to boot with DDR at 266MHz
* Fix proposed by James Doublesin

Signed-off-by: Chase Maupin <Chase.Maupin@ti.com>
12 years agoda850evm: read mac address from I2C EEPROM on AM18x EVM
Hadli, Manjunath [Thu, 9 Feb 2012 20:22:25 +0000 (20:22 +0000)]
da850evm: read mac address from I2C EEPROM on AM18x EVM

add support to read mac address for AM18x EVM manufactured from
Spectrum digital which have mac address stored in I2C EEPROM manufactured
by WINBOND. This patch reads mac address from I2C EEPROM and sets the
environment variable if not set, If mac address is already present in
environment compare it with the mac address in EEPROM and warn on
mismatch.Introduced a config option CONFIG_MAC_ADDR_IN_EEPROM to where
to look for the mac address.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Cc: Tom Rini <trini@ti.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Wolfgang Denk <wd@denx.de>
12 years agoda850evm: add support to read mac address from SPI flash
Hadli, Manjunath [Thu, 9 Feb 2012 20:22:24 +0000 (20:22 +0000)]
da850evm: add support to read mac address from SPI flash

add support to read mac address for da850/L138 evm manufactured by
Logic PD which store mac address in SPI flash manufactured by ST
Microelectronics. This patch adds support to read mac address from
SPI flash and set the mac address if it hasn't been set in environment,
If it has been set in the environment it compares the value one with
SPI flash and warns on mismatch. Introduced a config option
CONFIG_MAC_ADDR_IN_SPIFLASH indicating where to look mac address for.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Cc: Tom Rini <trini@ti.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Wolfgang Denk <wd@denx.de>
12 years agobeagle: enable DVI_PUP
Peter Meerwald [Wed, 8 Feb 2012 05:24:26 +0000 (05:24 +0000)]
beagle: enable DVI_PUP

this patch addresses the issue that there is no DVI output on beagleboard-xm in
u-boot; the TFP410 is not powered on, DVI_PUP needs to be set high

this is done differently on beagleboard-xm rev. B/C versus earlier boards:
beagleboard-xm B/C need to set GPIO2 of the TWL4030
earlier boards need to set GPIO170

note that the change occured somewhere inbetween beagleboard-xm rev. A2 and A3

note that beagleboard-xm rev. A and rev. B cannot be distinguished using
get_board_revision(), REVISION_XM_A/REVISION_XM_B is bogus

due to this glitch, the patch cannot work correctly on some beagleboard-xm
rev A/B boards

tested on beagleboard-xm rev. C (you should see orange color on a monitor)

note that the framebuffer is NOT yet set up, this could be done along the
lines below (for 800x480 resolution and RGB24) before calling omap3_dss_enable()

static void configure_frame_buffer() {
writel(0x80500000, (uint *) 0x48050480); // address
writel(0x80500000, (uint *) 0x48050484);
writel(0x01df031f, (uint *) 0x4805048c); // size
writel(0x00000091, (uint *) 0x480504a0); // RGB24
}

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
12 years agodavinci: cleanup davinci_sync_env_enetaddr() fucntion
Hadli, Manjunath [Thu, 9 Feb 2012 19:52:38 +0000 (19:52 +0000)]
davinci: cleanup davinci_sync_env_enetaddr() fucntion

check for the return status for eth_getenv_enetaddr_by_index()
and eth_setenv_enetaddr() functions and print appropriate message
on failure.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Cc: Tom Rini <trini@ti.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Wolfgang Denk <wd@denx.de>
Acked-by: Mike Frysinger <vapier@gentoo.org>
12 years agocommon/image.c: align usage of fdt_high with initrd_high
Shawn Guo [Mon, 9 Jan 2012 21:54:08 +0000 (21:54 +0000)]
common/image.c: align usage of fdt_high with initrd_high

The commit message of a28afca (Add uboot "fdt_high" enviroment variable)
states that fdt_high behaves similarly to the existing initrd_high.
But fdt_high actually has an outstanding difference from initrd_high.
The former specifies the start address, while the later specifies the
end address.

As fdt_high and initrd_high will likely be used together, it'd be nice
to have them behave same.  The patch changes the behavior of fdt_high
to have it aligned with initrd_high.

The document of fdt_high in README is updated with an example to
demonstrate the usage of this environment variable.

Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Acked-by: Simon Glass <sjg@chromium.org>
12 years agomx35pdk: Remove duplicate CPU revision and reset cause information
Fabio Estevam [Fri, 10 Feb 2012 06:29:15 +0000 (06:29 +0000)]
mx35pdk: Remove duplicate CPU revision and reset cause information

MX35PDK board does not need to print CPU revision and reset cause in board file
because this is printed by common code when CONFIG_DISPLAY_CPUINFO is selected

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx35: generic: Let get_reset_cause be defined only when CONFIG_DISPLAY_CPUINFO is...
Fabio Estevam [Fri, 10 Feb 2012 05:34:03 +0000 (05:34 +0000)]
mx35: generic: Let get_reset_cause be defined only when CONFIG_DISPLAY_CPUINFO is selected

get_reset_cause() function is only used inside print_cpuinfo(), so let it be
defined only when CONFIG_DISPLAY_CPUINFO is selected.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx51evk: Use gpio_direction_input prior to gpio_get_value
Fabio Estevam [Wed, 8 Feb 2012 02:34:45 +0000 (02:34 +0000)]
mx51evk: Use gpio_direction_input prior to gpio_get_value

Use gpio_direction_input prior to gpio_get_value.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53smd: Use gpio_direction_input prior to gpio_get_value
Fabio Estevam [Wed, 8 Feb 2012 02:34:44 +0000 (02:34 +0000)]
mx53smd: Use gpio_direction_input prior to gpio_get_value

Use gpio_direction_input prior to gpio_get_value.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53evk: Use gpio_direction_input prior to gpio_get_value
Fabio Estevam [Wed, 8 Feb 2012 02:34:43 +0000 (02:34 +0000)]
mx53evk: Use gpio_direction_input prior to gpio_get_value

Use gpio_direction_input prior to gpio_get_value.

Cc: Jason Liu <r64343@freescale.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Jason Liu <r64343@freescale.com>
12 years agomx53ard: Use gpio_direction_input prior to gpio_get_value
Fabio Estevam [Wed, 8 Feb 2012 02:34:42 +0000 (02:34 +0000)]
mx53ard: Use gpio_direction_input prior to gpio_get_value

Use gpio_direction_input prior to gpio_get_value.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx53loco: Use gpio_direction_input prior to gpio_get_value
Fabio Estevam [Wed, 8 Feb 2012 02:34:41 +0000 (02:34 +0000)]
mx53loco: Use gpio_direction_input prior to gpio_get_value

Use gpio_direction_input prior to gpio_get_value.

Cc: Jason Liu <r64343@freescale.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Stefano Babic <sbabic@denx.de>
Acked-by: Jason Liu <r64343@freescale.com>
12 years agomx6q: mx6qsabrelite: Conditionally define macros for environment in serial flash
Eric Nelson [Tue, 31 Jan 2012 07:52:10 +0000 (07:52 +0000)]
mx6q: mx6qsabrelite: Conditionally define macros for environment in serial flash

The default settings store the persistent environment on SD card
and not serial flash (SPI NOR).

To use SPI NOR to save the environment instead of SD card, edit
include/configs/mx6qsabrelite.h and

- undefine CONFIG_ENV_IS_IN_MMC
- define   CONFIG_ENV_IS_IN_SPI_FLASH

The SPI driver can take as chip select the controller's chip selects
as well as an external GPIO. The LSB byte has the value of the internal
chip select, the highest (thought as 16-bit value) contains the GPIO
number.

The GPIO used on Sabre Lite is GP3:19 == 83.

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
Acked-by: Jason Liu <jason.hui@linaro.org>
Tested-by: Jason Liu <jason.hui@linaro.org>
12 years agomx6q: mx6qsabrelite: Provide default serial flash bus and chip-select
Eric Nelson [Tue, 31 Jan 2012 07:52:09 +0000 (07:52 +0000)]
mx6q: mx6qsabrelite: Provide default serial flash bus and chip-select

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
Acked-by: Stefano Babic <sbabic@denx.de>
Acked-by: Jason Liu <jason.hui@linaro.org>
Tested-by: Jason Liu <jason.hui@linaro.org>
12 years agomx6q: mx6qsabrelite: Add ECSPI support to the Sabrelite platform
Eric Nelson [Tue, 31 Jan 2012 07:52:05 +0000 (07:52 +0000)]
mx6q: mx6qsabrelite: Add ECSPI support to the Sabrelite platform

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
Acked-by: Stefano Babic <sbabic@denx.de>
Acked-by: Jason Liu <jason.hui@linaro.org>
Tested-by: Jason Liu <jason.hui@linaro.org>
12 years agomx6q: Add support for ECSPI through mxc_spi driver
Eric Nelson [Tue, 31 Jan 2012 07:52:04 +0000 (07:52 +0000)]
mx6q: Add support for ECSPI through mxc_spi driver

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
Acked-by: Stefano Babic <sbabic@denx.de>
Acked-by: Jason Liu <jason.hui@linaro.org>
12 years agomxc_spi: move machine specifics into CPU headers
Eric Nelson [Tue, 31 Jan 2012 07:52:03 +0000 (07:52 +0000)]
mxc_spi: move machine specifics into CPU headers

Move (E)CSPI register declarations into the imx-regs.h files for each supported CPU

Introduce two new macros to control conditional setup
     MXC_CSPI - Used for processors with the Configurable Serial Peripheral Interface (MX3x)
     MXC_ECSPI - For processors with Enhanced Configurable... (MX5x, MX6x)

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
Acked-by: Dirk Behme <dirk.behme@de.bosch.com>
Acked-by: Stefano Babic <sbabic@denx.de>
Acked-by: Jason Liu <jason.hui@linaro.org>
Tested-by: Jason Liu <jason.hui@linaro.org>
12 years agomx6q: define GPIO macros for translating between ordinals and port:index
Eric Nelson [Tue, 31 Jan 2012 07:52:01 +0000 (07:52 +0000)]
mx6q: define GPIO macros for translating between ordinals and port:index

The interface to the mxc_gpio driver uses integer (ordinal) values to
refer to all GPIOs on the i.MX processors. The registers themselves
and much of the i.MX documentation are banked in groups of 32, and these
macros allow the use of the port:index numbering for clarity.

GPIO_NUMBER() converts to ordinal value from port:index
GPIO_PORT() returns the port of an ordinal value
GPIO_INDEX() returns the index or offset of the ordinal.

Discussion on the mailing list at
http://lists.denx.de/pipermail/u-boot/2012-January/116927.html

Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
Signed-off-by: Eric Nelson <eric.nelson@boundarydevices.com>
12 years agoMX27: add missing get_tbclk()
Stefano Babic [Tue, 7 Feb 2012 17:32:56 +0000 (18:32 +0100)]
MX27: add missing get_tbclk()

Signed-off-by: Stefano Babic <sbabic@denx.de>
12 years agoMX28: Fix get_timer() / get_tbclk() issue
Marek Vasut [Tue, 7 Feb 2012 06:47:31 +0000 (06:47 +0000)]
MX28: Fix get_timer() / get_tbclk() issue

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Detlev Zundel <dzu@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
12 years agomx28: fix SPL code to make USB booting work
Matthias Fuchs [Mon, 6 Feb 2012 23:32:42 +0000 (23:32 +0000)]
mx28: fix SPL code to make USB booting work

This patch fixes booting i.MX28 CPUs via USB download.
In this mode the CPU's bootrom implements a USB HID device that
accepts a bootstream.

When downloading the bootstream via USB, first the SPL code is
received and executed. Then the u-boot image is received and
called.

The USB bootmode is interrupt driven.

This patch fixes two things:

1) The ARM's fast interrupt mode is disabled when the SPL code
has been run. So save and restore the CPSR register.

2) Save and restore c1 control register: the exception vector
location needs to be set back to bootrom space to make the USB
interrupts work again. The SPL code needs to change this
option for the ram size probing.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd.eu>
Acked-by: Marek Vasut <marek.vasut@gmail.com>
12 years agoi.MX28: Fix VDDIO and VDDA setup
Marek Vasut [Mon, 30 Jan 2012 14:00:01 +0000 (14:00 +0000)]
i.MX28: Fix VDDIO and VDDA setup

The DC power STS shouldn't be checked if booting off 5V supply.

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Detlev Zundel <dzu@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Robert Deliën <robert@delien.nl>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Matthias Fuchs <matthias.fuchs@esd.eu>
12 years agoMX5/MX6: add missing get_ticks() and get_tbclk()
Stefano Babic [Mon, 6 Feb 2012 11:52:36 +0000 (12:52 +0100)]
MX5/MX6: add missing get_ticks() and get_tbclk()

commit f31a911fe (arm, post: add missing post_time_ms for arm)
enables get_ticks and get_tbclk for all arm based boards,
MX5/MX6 have not yet implemented.

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Dirk Behme <dirk.behme@de.bosch.com>
CC: Jason Liu <jason.hui@linaro.org>
CC: Marek Vasut <marek.vasut@gmail.com>
12 years agoMX31: add missing get_tbclk()
Stefano Babic [Sat, 4 Feb 2012 12:02:01 +0000 (13:02 +0100)]
MX31: add missing get_tbclk()

Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Helmut Raiger <helmut.raiger@hale.at>
12 years agoMX35: add missing get_ticks() and get_tbclk()
Stefano Babic [Sat, 4 Feb 2012 11:56:50 +0000 (12:56 +0100)]
MX35: add missing get_ticks() and get_tbclk()

commit f31a911fe (arm, post: add missing post_time_ms for arm)
enables get_ticks and get_tbclk for all arm based boards,
MX5/MX6 have not yet implemented.

Signed-off-by: Stefano Babic <sbabic@denx.de>
12 years agomx28evk: add SPI support
Matthias Fuchs [Wed, 18 Jan 2012 01:33:09 +0000 (01:33 +0000)]
mx28evk: add SPI support

This patch adds SPI support for the MX28EVK. Support for
an optionally installed SPI flash is also added. An example
configuration for redundant envrionment from SPI flash is also
added but disabled by default.

This patch has been tested on a MX28EVK Rev. D with an installed
SST25VF032B 32Mbit SPI flash.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd.eu>
Acked-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx28evk: add USB support
Matthias Fuchs [Wed, 18 Jan 2012 01:33:08 +0000 (01:33 +0000)]
mx28evk: add USB support

This patch enables USB host support on the MX28EVK board.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd.eu>
Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agomx28evk: add RTC support
Matthias Fuchs [Wed, 18 Jan 2012 01:33:07 +0000 (01:33 +0000)]
mx28evk: add RTC support

This patch adds support for the MX28 internal RTC
and enables u-boot's date command.

Signed-off-by: Matthias Fuchs <matthias.fuchs@esd.eu>
Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
12 years agoMerge branch 'master' of git://git.denx.de/u-boot-microblaze
Wolfgang Denk [Sun, 26 Feb 2012 21:17:40 +0000 (22:17 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-microblaze

* 'master' of git://git.denx.de/u-boot-microblaze:
  microblaze: Enable TFTP put command
  microblaze: avoid interrupt race conditions
  microblaze: fix build failure due to undefined reference to `get_ticks'

12 years agomicroblaze: Enable TFTP put command
Stephan Linz [Sat, 21 Jan 2012 21:02:57 +0000 (22:02 +0100)]
microblaze: Enable TFTP put command

Signed-off-by: Stephan Linz <linz@li-pro.net>
12 years agomicroblaze: avoid interrupt race conditions
Stephan Linz [Wed, 22 Feb 2012 18:12:43 +0000 (19:12 +0100)]
microblaze: avoid interrupt race conditions

The interrupt acknowledge action have to run after the
registered interrupt handler. So we have a chance to
bear out the corresponding interrupt request in the
corresponding controller hardware.

With this reordering, we optain a proper interrupt
handling for level triggered interrupt sources -- for
example the new axi_timer v1.02.a introduced in ISE 13.2.

Signed-off-by: Stephan Linz <linz@li-pro.net>
Acked-by: Michal Simek <monstr@monstr.eu>
12 years agomicroblaze: fix build failure due to undefined reference to `get_ticks'
Stephan Linz [Wed, 22 Feb 2012 21:39:57 +0000 (22:39 +0100)]
microblaze: fix build failure due to undefined reference to `get_ticks'

after commit "common: add possibility for readline_into_buffer
timeout" (sha1:9c34831) was applied. The Microblaze generic build
fails with error below:

common/libcommon.o: In function `cread_line':
/devel/u-boot/common/main.c:717: undefined reference to `get_ticks'
/devel/u-boot/common/main.c:717: undefined reference to `get_tbclk'
/devel/u-boot/common/main.c:720: undefined reference to `get_ticks'

Signed-off-by: Stephan Linz <linz@li-pro.net>
Acked-by: Michal Simek <monstr@monstr.eu>
12 years agoRevert "ARM: I2C: I2C Multi byte address support"
Tom Rini [Mon, 20 Feb 2012 18:49:16 +0000 (18:49 +0000)]
Revert "ARM: I2C: I2C Multi byte address support"

This reverts commits 2faa76196af4b3e93bcb9e38ed9090cbd3b06db3 as
this has introduced some large problems on all other platforms and have
more changes in them than the commit message implies.

Cc: Heiko Schocher <hs@denx.de>
Cc: Patil, Rachna <rachna@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
12 years agoRevert "ARM: AM33XX: Add AM33XX I2C driver support"
Tom Rini [Mon, 20 Feb 2012 18:49:15 +0000 (18:49 +0000)]
Revert "ARM: AM33XX: Add AM33XX I2C driver support"

This reverts commit 498cbdfe62a8330f6c89765bdd15e60328a26511 as we need
to revert the i2c changes that add the support for the platform.

Cc: Heiko Schocher <hs@denx.de>
Cc: Patil, Rachna <rachna@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
12 years agoMerge branch 'master' of git://git.denx.de/u-boot-mmc
Wolfgang Denk [Fri, 17 Feb 2012 22:54:46 +0000 (23:54 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-mmc

* 'master' of git://git.denx.de/u-boot-mmc:
  mmc: make mmc_send_status() more reliable
  mmc: fix card busy polling
  Tegra: mmc: Fixed handling of interrupts in timeouts.
  omap_hsmmc: Wait for CMDI to be clear

12 years agoMerge branch 'master' of git://git.denx.de/u-boot-mpc85xx
Wolfgang Denk [Fri, 17 Feb 2012 22:54:17 +0000 (23:54 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-mpc85xx

* 'master' of git://git.denx.de/u-boot-mpc85xx:
  powerpc/8xxx:Add MPH controller support in USB device-tree fixup
  powerpc/8xxx: Cleanup USB device-tree fixup

12 years agoMerge branch 'trini@ti.com' of git://git.denx.de/u-boot-staging
Wolfgang Denk [Fri, 17 Feb 2012 22:52:52 +0000 (23:52 +0100)]
Merge branch 'trini@ti.com' of git://git.denx.de/u-boot-staging

* 'trini@ti.com' of git://git.denx.de/u-boot-staging:
  usb_ether: fix typo
  cmd_eeprom: typo
  doc: complete, typos

12 years agoMerge branch 'master' of git://git.denx.de/u-boot-x86
Wolfgang Denk [Fri, 17 Feb 2012 22:47:29 +0000 (23:47 +0100)]
Merge branch 'master' of git://git.denx.de/u-boot-x86

* 'master' of git://git.denx.de/u-boot-x86:
  x86: Convert board_init_f_r to a processing loop
  x86: Split init functions out of board.c
  x86: Move relocation code out of board.c
  x86: Move setup_pcat_compatibility() out of board.c
  x86: Move do_go_exec() out of board.c
  CHECKPATCH: arch/x86/lib/*
  x86: Tweak IDT and GDT for alignment and readability
  x86: Allow cache before copy to RAM
  x86: Create weak init_cache() and default enable_caches() functions
  x86: Set GD_FLG_RELOC after entering in-RAM copy of U-Boot
  x86: Use fs for global data
  x86: Rework relocation calculations
  x86: Simplify Flash-to-RAM code execution transition
  x86: Rework Global Descriptor Table loading
  x86: Remove GDR related magic numbers
  x86: Speed up copy-to-RAM and clear BSS operations
  x86: Import glibc memcpy implementation

12 years agommc: make mmc_send_status() more reliable
Jan Kloetzke [Sun, 5 Feb 2012 22:29:12 +0000 (22:29 +0000)]
mmc: make mmc_send_status() more reliable

Align the card status polling with the Linux kernel and retry the
command at least five times. Also some cards apparently mishandle the
status bits, so make sure to check the card state too.

Signed-off-by: Jan Kloetzke <jan.kloetzke@dspg.com>
Cc: Andy Fleming <afleming@gmail.com>
12 years agommc: fix card busy polling
Jan Kloetzke [Sun, 5 Feb 2012 22:29:11 +0000 (22:29 +0000)]
mmc: fix card busy polling

A MMC/SD card may always go into the programming state (and hence be
busy) after a block write. Therefore always check the card state, even
after single block writes. On the other hand there is no need to check
the card status after a read.

Also make sure that errors during busy polling are propagated upwards.

Signed-off-by: Jan Kloetzke <jan.kloetzke@dspg.com>
Cc: Andy Fleming <afleming@gmail.com>
12 years agoTegra: mmc: Fixed handling of interrupts in timeouts.
Tom Warren [Tue, 7 Feb 2012 06:17:16 +0000 (06:17 +0000)]
Tegra: mmc: Fixed handling of interrupts in timeouts.

We are seeing occasional timeouts in the Tegra mmc code when
we are reading from external MMC cards. These don't seem to be
detrimental if they are handled properly.  This CL properly
clears the "normal interrupt status register" (norintsts) in
error conditions.  If we don't do this, when we come back into
mmc_send_cmd() the register will still contain status from the
last transaction.

Signed-off-by: Doug Anderson <dianders@chromium.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
12 years agoomap_hsmmc: Wait for CMDI to be clear
Tom Rini [Mon, 30 Jan 2012 11:22:25 +0000 (11:22 +0000)]
omap_hsmmc: Wait for CMDI to be clear

Before we can send a command we need both the DATI (command inhibit on
mmc_dat line) bit and CMDI (command inhibit on mmc_cmd line) are clear.
The previous behavior of only checking on DATI was insufficient on some
cards and incorrect behavior in any case.  This makes the code check
for both bits being clear and makes the error print more clear as
to what happened.  DATI_CMDDIS is removed as it was unused elsewhere
in the code and stood for 'DATI is set, cmds are disabled still'.

Fix originally spotted by Peter Bigot.

Tested-by: Peter A. Bigot <bigotp@acm.org>
Tested-by: Robert Nelson <robertcnelson@gmail.com>
Signed-off-by: Tom Rini <trini@ti.com>
Tested-by: Andreas Müller <schnitzeltony@googlemail.com>
12 years agopowerpc/8xxx:Add MPH controller support in USB device-tree fixup
ramneek mehresh [Fri, 10 Feb 2012 00:36:43 +0000 (00:36 +0000)]
powerpc/8xxx:Add MPH controller support in USB device-tree fixup

Add support for fixing usb mode and phy type for
MPH(Multi Port Host) USB controllers in device-tree nodes.
Required for socs like P3060, P5020, etc having MPH USB controller

Signed-off-by: Ramneek Mehresh <ramneek.mehresh@freescale.com>
12 years agopowerpc/8xxx: Cleanup USB device-tree fixup
ramneek mehresh [Mon, 6 Feb 2012 19:17:29 +0000 (19:17 +0000)]
powerpc/8xxx: Cleanup USB device-tree fixup

Some code cleanup done for USB device-tree fixup:
- handling error value returned from fdt_fixup_usb_mode_phy_type()
- using ARRAY_SIZE macro
- using snprintf instead of sprintf

Signed-off-by: Ramneek Mehresh <ramneek.mehresh@freescale.com>