Simon Glass [Tue, 12 May 2015 20:55:05 +0000 (14:55 -0600)]
dm: serial: Don't support CONFIG_CONS_INDEX with device tree
This feature should be deprecated for new boards, and significantly adds
to SPL code size. Drop it. Instead, we can use stdout-path in the /chosen
node.
Simon Glass [Tue, 12 May 2015 20:55:02 +0000 (14:55 -0600)]
dm: ns16550: Support CONFIG_SYS_NS16550_MEM32 with driver model
This option is used by some boards, so support it with driver model. This
is really ugly - we should rewrite this driver once all users are moved to
driver model.
Simon Glass [Wed, 13 May 2015 13:02:27 +0000 (07:02 -0600)]
Remove typedefs from bmp_layout.h
We try to avoid typedefs and these ones are easy enough to remove. Before
changing this header in the next patch, remove the typedefs.
Signed-off-by: Simon Glass <sjg@chromium.org> Suggested-by: Joe Hershberger <joe.hershberger@gmail.com> Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Simon Glass [Wed, 13 May 2015 13:02:24 +0000 (07:02 -0600)]
arm: spl: Add an API to detect when U-Boot is started from SPL
For secure boot systems it is common to have a read-only U-Boot which starts
the machine and jumps to a read-write U-Boot for actual booting the OS. This
allows the read-write U-Boot to be upgraded without risk of permanently
bricking the machine. In the event that the read-write U-Boot is corrupted,
the read-only U-Boot can detect this with a checksum and boot into a
recovery flow.
To support this, add a way to detect when U-Boot is run from SPL as opposed
to some other method, such as booted directly (no SPL) or started from
another source (e.g. a primary U-Boot). This works by putting a special value
in r0.
For now we rely on board-specific code to actually check the register and
set a flag. At some point this could be generalised, perhaps by using a spare
register and passing a flag to _main and/or board_init_f().
This commit does not implement any feature, but merely provides the API for
boards to implement.
Simon Glass [Wed, 6 May 2015 20:00:16 +0000 (14:00 -0600)]
dm: tegra: usb: Move USB to driver model
Somehow this change was dropped in the various merges. I noticed when I
came to turn off the non-driver-model support for Tegra. We need to make
this change (and deal with any problems) before going further.
Change-Id: Ib9389a0d41008014eb0df0df98c27be65bc79ce6 Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Marek Vasut <marex@denx.de>
sunxi: Enable CONFIG_SYS_64BIT_LBA when AHCI is used
Due to absence of CONFIG_SYS_64BIT_LBA, u-boot-sunxi currently has
no support for the (GPT) partioning scheme of large disks > 2TB.
While the AHCI driver seems to handle this nicely, the problem is
that lbaint_t values get truncated to 32-bit.
This patch sets CONFIG_SYS_64BIT_LBA from sunxi_common.h for all
SoCs that support AHCI (CONFIG_SUNXI_AHCI).
Signed-off-by: Bernhard Nortmann <bernhard.nortmann@web.de> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Simon Glass [Fri, 5 Jun 2015 20:39:46 +0000 (14:39 -0600)]
tegra: config: nyan-big: Add options required by Chrome OS boot
We need to match the device tree in the FIT with the U-Boot model so we
can automatically select the right device tree. Also adjust the load address
so that the device tree is not in the way when a zImage kernel tries to
extract itself.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Warren <twarren@nvidia.com>
Simon Glass [Fri, 5 Jun 2015 20:39:35 +0000 (14:39 -0600)]
tegra: spi: Support slow SPI rates
Use the oscillator as the source clock when we cannot achieve a low-enough
speed with the peripheral clock. This happens when we request 3MHz on a SPI
clock, for example.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Warren <twarren@nvidia.com>
Simon Glass [Fri, 5 Jun 2015 20:39:33 +0000 (14:39 -0600)]
tegra: spi: Drop the claim_bus() method to correct delays
At present the driver does not properly honour the requested SPI CS
deactivation delay since the SPI bus is changed in the claim_bus() method.
Everything the claim_bus() method does can be done when the device is probed
(setting the speed and mode) and at the start of a new transfer (where the
fifo_status is already cleared). So drop this method.
Also, until the delay is complete, we should not touch the bus, so make sure
that spi_cs_activate() is called before other things are done in the xfer()
method.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Warren <twarren@nvidia.com>
Adam YH Lee [Mon, 1 Jun 2015 21:29:09 +0000 (14:29 -0700)]
pepper: Implement Board Detection mechanism
AM335x-based 'Gumstix Pepper' SBCs and variants use different types of
RAM (DDR2 vs DDR3 with DDR3 being the default). Detect the board type
by reading the factory-programmed EEPROM [1] and use this to select any
runtime boot options such as RAM type.
Stefan Roese [Mon, 18 May 2015 12:08:24 +0000 (14:08 +0200)]
autoboot.c: Add feature to stop autobooting via SHA256 encrypted password
This patch adds the feature to only stop the autobooting, and therefor
boot into the U-Boot prompt, when the input string / password matches
a values that is encypted via a SHA256 hash and saved in the environment.
This feature is enabled by defined these config options:
CONFIG_AUTOBOOT_KEYED
CONFIG_AUTOBOOT_STOP_STR_SHA256
Signed-off-by: Stefan Roese <sr@denx.de> Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Stefan Roese <sr@denx.de> Reviewed-by: Simon Glass <sjg@chromium.org>
[trini: Drop ip04 and pm9263 configs/ additions, those boards previously
set CONFIG_AUTOBOOT_PROMPT but never used it, re-run savedefconfig over
all boards that did change. Make digsy_mtc_* string include seconds to
match others and not warn. ] Signed-off-by: Tom Rini <trini@konsulko.com> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Stefan Roese [Mon, 18 May 2015 12:08:22 +0000 (14:08 +0200)]
autoboot.c: Remove CONFIG_AUTOBOOT_STOP_STR2 and CONFIG_AUTOBOOT_DELAY_STR2
These defines for a 2nd autoboot stop and delay string are nearly unused. Only
sc3 defines CONFIG_AUTOBOOT_DELAY_STR2. And a patch to remove this most likely
unmaintained board is also posted to the list.
By removing these defines the code will become cleaner and moving the remaining
compile options to Kconfig will get easier.
Signed-off-by: Stefan Roese <sr@denx.de> Reviewed-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
Stefan Roese [Fri, 29 May 2015 09:47:32 +0000 (11:47 +0200)]
Kconfig: Enable usage of escape char '\' in string values
I might have missed something, but I failed to use the escape char '\'
in strings. To pass a printf format string like "foo %d bar\n" via
Kconfig to the code.
Right now its not possible to use the escape character '\' in Kconfig
string values correctly to e.g. set this string value "test output\n".
The '\n' will be converted to 'n'.
The current implementation removes some of the '\' chars from the input
string in conf_set_sym_val(). Examples:
As you see in these examples, its impossible to generate a single '\'
charater in the output string as its needed for something like '\n'.
This patch now changes this behavior to not drop some backslashes in
conf_set_sym_val() and to not add new backslashes in the resulting
output string. Removing the function sym_escape_string_value()
completely as its not needed anymore.
Signed-off-by: Stefan Roese <sr@denx.de> Cc: Masahiro Yamada <yamada.masahiro@socionext.com> Reviewed-by: Simon Glass <sjg@chromium.org> Cc: Tom Rini <trini@konsulko.com>
Sanchayan Maity [Mon, 1 Jun 2015 13:07:25 +0000 (18:37 +0530)]
colibri_vf: Enable board specific USB initialisation for USB pen gpio
Add IOMUX for the pad used as USB pen. This needs to be driven low for
the Iris and Viola boards where it is pulled up high by default. This is
required for the USB host functionality to work on these boards. Use the
board specific weak initialisation function, to drive the pin low which
would be called on "usb start".
colibri_vf: Add separate defconfig for device tree support
Most of the drivers available for Vybrid are not yet converted
to OF model to use device tree model, only few drivers
like SPI and GPIO drivers use device trees.
Add separate defconfig for who needs to use device tree model.
Later this can be integrated to single defconfig.
Acked-by: Stefan Agner <stefan@agner.ch> Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
Add device tree files for Freescale Vybrid platform and
Toradex Colibri VF50, VF61 modules.
Device tree files are taken from upstream Kernel.
Removed the stuff which are not used/supported yet in U-Boot.
Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
dm: gpio: uclass: Add flag to control sequence numbering
Like SPI and I2C few GPIO controllers also have
multiple chip instances. This patch adds the
flag 'DM_UC_FLAG_SEQ_ALIAS' in gpio_uclass driver
to control device sequence numbering. By defalut
the dev->r_seq for gpio_uclass will alwalys
returns -1, which leads the gpio driver probe
failure when using the driver with device trees.
Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
Fabio Estevam [Thu, 28 May 2015 15:33:34 +0000 (12:33 -0300)]
mx6_common: Fix LOADADDR and SYS_TEXT_BASE for MX6SL and MX6SX
Commit 8183058188cd2d942 ("imx6: centralise common boot options in
mx6_common.h") broke boot on mx6sl and mx6sx by assuming that all mx6
SoCs use the same LOADADDR/SYS_TEXT_BASE range, which is not correct.
DDR on mx6sx/mx6sl starts at 0x80000000.
Adjust LOADADDR/SYS_TEXT_BASE to the proper values for mx6sx/mx6sl,
so that these SoCs can boot again.
Also, TQMA6 requires a custom CONFIG_SYS_TEXT_BASE value, so move
its setting prior to the inclusion of mx6_common.h.
Signed-off-by: Fabio Estevam <fabio.estevam at freescale.com>
Hans de Goede [Fri, 29 May 2015 13:09:48 +0000 (15:09 +0200)]
image-fit: Fix compiler warning in fit_conf_print()
This fixes the following compiler warning:
In file included from tools/common/image-fit.c:1:0:
./tools/../common/image-fit.c: In function ‘fit_conf_print’:
./tools/../common/image-fit.c:1470:27: warning: logical not is only applied
to the left hand side of comparison [-Wlogical-not-parentheses]
(const char **)&uname) > 0;
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Hans de Goede [Wed, 3 Jun 2015 18:08:37 +0000 (20:08 +0200)]
sunxi: Select CONFIG_CMD_NET and CONFIG_CMD_SETEXPR by default
Select CONFIG_CMD_NET and CONFIG_CMD_SETEXPR by default rather then
needing to have this in every sunxi defconfig file.
This also fixes the Merrii_A80_Optimus defconfig no longer building.
Cc: Maxin B. John <maxin.john@enea.com> Reported-by: Maxin B. John <maxin.john@enea.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
fdt: Pass the device serial number through devicetree
Before device-tree, the device serial number used to be passed to the kernel
using ATAGs (on ARM). This is now deprecated and all the handover to the kernel
should now be done using device-tree. Thus, this passes the serial-number
property to the kernel using the serial-number property of the root node, as
expected by the kernel.
The serial number is a string that somewhat represents the device's serial
number. It might come from some form of storage (e.g. an eeprom) and be
programmed at factory-time by the manufacturer or come from identification
bits available in e.g. the SoC.
Signed-off-by: Paul Kocialkowski <contact@paulk.fr> Reviewed-by: Simon Glass <sgj@chromium.org>
Hans de Goede [Tue, 2 Jun 2015 13:53:40 +0000 (15:53 +0200)]
sunxi: Sync dts files with the linux kernel
Copy over all the latest dts changes from mripard/sunxi/dt-for-4.2 ,
this gives us a proper dtsi file for the A33 rather then abusing
sun8i-a23.dtsi for this.
And this replaces our minimal (dummy) sun7i-a20-mk808c and
sun8i-a33-astar-mid756 dts files with proper ones.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 1 Jun 2015 14:37:24 +0000 (16:37 +0200)]
sunxi: Add new Mele_A1000G_quad defconfig
The Mele A1000G-quad and the Mele M9 have the same PCB, sofar we've been
using the same defconfig (and dts on the kernel side) for both models.
Unfortunately this does not work for the otg controller, on the M9 this
is routed to a micro-usb connector on the outside, while as on the
A1000G-quad it is connected to an usb to sata bridge.
This commit adds a new defconfig for the Mele-A1000G-quad to allow using
different otg controller settings on the 2 boards.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sun, 31 May 2015 17:26:54 +0000 (19:26 +0200)]
sunxi: usb_phy: Swap check for disconnect threshold
Before this commit the code for determining the disconnect threshold was
checking for sun4i or sun6i assuming that those where the exception and
that newer SoCs use a disconnect threshold of 2 like sun7i does.
But it turns out that newer SoCs actually use a disconnect threshold of 3
and sun5i and sun7i are the exceptions, so check for those instead.
Here are the settings from the various Allwinner SDK sources:
sun4i-a10: USBC_Phy_Write(usbc_no, 0x2a, 3, 2);
sun5i-a13: USBC_Phy_Write(usbc_no, 0x2a, 2, 2);
sun6i-a31: USBC_Phy_Write(usbc_no, 0x2a, 3, 2);
sun7i-a20: USBC_Phy_Write(usbc_no, 0x2a, 2, 2);
sun8i-a23: USBC_Phy_Write(usbc_no, 0x2a, 3, 2);
sun8i-h3: USBC_Phy_Write(usbc_no, 0x2a, 3, 2);
sun9i-a80: USBC_Phy_Write(usbc_no, 0x2a, 3, 2);
Note this commit makes no functional changes for sun4i - sun7i, and
changes the disconnect threshold for sun8i to match what Allwinner uses.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Simon Glass [Fri, 22 May 2015 21:42:17 +0000 (15:42 -0600)]
sandbox: dts: Add the real-time-clock test nodes back in
These were lost when the PMIC series was applied. Add them back so that the
tests pass again.
Reported-by: Joe Hershberger <joe.hershberger@gmail.com> Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Simon Glass [Fri, 22 May 2015 21:42:14 +0000 (15:42 -0600)]
dm: Sort the uclass IDs after the tegra/PMIC addition
Tidy up the sort order again.
Signed-off-by: Simon Glass <sjg@chromium.org> Reported-by: Joe Hershberger <joe.hershberger@gmail.com> Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Simon Glass [Mon, 11 May 2015 03:08:06 +0000 (21:08 -0600)]
dm: pci: Allow PCI bus numbering aliases
Commit 9cc36a2 'dm: core: Add a flag to control sequence numbering' changed
the default uclass behaviour to not support bus numbering. This is incorrect
for PCI and that commit should have enabled the flag for PCI.
Enable it so that PCI buses can be found and the 'pci' command works again.
Also add a test for this.
Simon Glass [Mon, 11 May 2015 03:07:27 +0000 (21:07 -0600)]
sandbox: Tidy up terminal restore
For some reason 'u-boot -D' does not restore the terminal correctly when
the 'reset' command is used. Call the terminal restore function explicitly
in this case.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
Andrew Bradford [Wed, 3 Jun 2015 16:37:39 +0000 (12:37 -0400)]
x86: baytrail: pci region 3 is not always mapped to end of ram
Baytrail physically maps the first 2 GB of SDRAM from 0x0 to 0x7FFFFFFF
and additional SDRAM is mapped from 0x100000000 and up. There is a
physical memory hole from 0x80000000 to 0xFFFFFFFF for other uses.
Because of this, PCI region 3 should only try to use up to the amount of
SDRAM or 0x80000000, which ever is less.
Signed-off-by: Andrew Bradford <andrew.bradford@kodakalaris.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 3 Jun 2015 01:20:06 +0000 (09:20 +0800)]
x86: qemu: Implement PIRQ routing
Support QEMU PIRQ routing via device tree on both i440fx and q35
platforms. With this commit, Linux booting on QEMU from U-Boot
has working ATA/SATA, USB and ethernet.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 3 Jun 2015 01:20:05 +0000 (09:20 +0800)]
x86: coreboot: Control I/O port 0xb2 writing via device tree
Writing 0xcb to I/O port 0xb2 (Advanced Power Management Control) causes
U-Boot to hang on QEMU q35 target. We introduce a config option in the
device tree "u-boot,no-apm-finalize" under /config node if we don't want
to do that.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 3 Jun 2015 01:20:04 +0000 (09:20 +0800)]
x86: qemu: Create separate i440fx and q35 device trees
Although the two qemu-x86 targets (i440fx and q35) share a lot in
common, they still have something that cannot easily handled in one
single device tree). Split to create two dedicated device tree files
and make the i440fx be the default build target.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Bin Meng [Mon, 25 May 2015 14:36:27 +0000 (22:36 +0800)]
x86: qemu: Adjust VGA initialization
As VGA option rom needs to run at C segment, although QEMU PAM emulation
seems to only guard E/F segments, for correctness, move VGA initialization
after PAM decode C/D/E/F segments.
Also since we already tested QEMU targets to differentiate I440FX and Q35
platforms, change to locate the VGA device via hardcoded b.d.f instead of
dynamic search for its vendor id & device id pair.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Bin Meng [Mon, 25 May 2015 14:36:26 +0000 (22:36 +0800)]
x86: qemu: Enable legacy IDE I/O ports decode
QEMU always decode legacy IDE I/O ports on PIIX chipset. However Linux ata_piix
driver does sanity check to see whether legacy ports decode is turned on.
To make Linux ata_piix driver happy, turn on the decode via IDE_TIMING register.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Bin Meng [Sat, 23 May 2015 16:12:33 +0000 (00:12 +0800)]
x86: qemu: Turn on legacy segments decode
By default the legacy segments C/D/E/F do not decode to system RAM.
Turn on the decode via Programmable Attribute Map (PAM) registers
so that we can write configuration tables in the F segment.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Bin Meng [Sat, 23 May 2015 16:12:32 +0000 (00:12 +0800)]
x86: qemu: Make host bridge (b.d.f=0.0.0) visible
The default weak version of pci_skip_dev() in drivers/pci/pci_common.c
skips the host bridge (b.d.f = 0.0.0) which is actually the i440fx/q35
chipset for QEMU targets. Define CONFIG_PCI_CONFIG_HOST_BRIDGE to make
it visible in the PCI configuration space.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
projects / u-boot / log