Enable i.MX thermal DM driver to mx6sabre_common.h file. Since the
thermal is used in init_sequence_f, so define the CONFIG_SYS_MALLOC_F_LEN
to support DM driver using in pre relocation phase.
Additional, thermal driver depends on ocotp, make sure to enable
CONFIG_MXC_OCOTP when CONFIG_IMX6_THERMAL is selected.
Simon Glass [Tue, 11 Nov 2014 00:16:51 +0000 (17:16 -0700)]
dm: tegra: Add platform data for the GPIO driver
Add platform data for the GPIO driver. It doesn't need to contain anything
since the GPIO driver will actually use information from the CONFIGs for
now. This merely serves to ensure that the GPIO driver is bound.
Simon Glass [Tue, 11 Nov 2014 00:16:48 +0000 (17:16 -0700)]
dm: Allow stdio registration to be dropped
Provide a CONFIG_DM_STDIO option to enable registering a serial device
with the stdio library. This is seldom useful in SPL, so disable it by
default when building for SPL.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Tom Rini <trini@ti.com>
Simon Glass [Tue, 11 Nov 2014 00:16:47 +0000 (17:16 -0700)]
dm: Allow device removal features to be dropped
For SPL we don't expect to need to remove a device. Save some code space
by dropping this feature. The board config can define
CONFIG_DM_DEVICE_REMOVE if this is in fact needed.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Tom Rini <trini@ti.com>
Simon Glass [Tue, 11 Nov 2014 00:16:44 +0000 (17:16 -0700)]
dm: arm: spl: Allow simple malloc() in SPL
For SPL it is sometimes useful to have a simple malloc() just to permit
driver model to work, in the cases where the full malloc() is not made
available by the board config.
Simon Glass [Tue, 11 Nov 2014 00:16:43 +0000 (17:16 -0700)]
dm: Split the simple malloc() implementation into its own file
The simple malloc() implementation is used when memory is tight. It provides
a simple buffer with an incrementing pointer.
At present the implementation is inside dlmalloc. Move it into its own file
so that it is easier to find.
Rather than using relocation as a signal that the full malloc() is
available, add a special GD_FLG_FULL_MALLOC_INIT flag. This signals that the
simple malloc() should no longer be used.
In some cases, such as SPL, even the code space used by the full malloc() is
wasteful. Add a CONFIG_SYS_MALLOC_SIMPLE option to provide only the simple
malloc. In this case the full malloc is not available at all. It saves about
1KB of code space and about 0.5KB of data on Thumb 2.
Acked-by: Tom Rini <trini@ti.com> Signed-off-by: Simon Glass <sjg@chromium.org>
Simon Glass [Wed, 29 Oct 2014 19:09:01 +0000 (13:09 -0600)]
dm: at91: Convert snapper9260 to use driver model
Convert this at91sam9260-based board to use driver model. This should serve
as an example for other similar boards. Serial and GPIO are supported so
far.
Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Andreas Bießmann <andreas.devel@googlemail.com>
Simon Glass [Wed, 29 Oct 2014 19:09:00 +0000 (13:09 -0600)]
dm: at91: Add driver model support for the serial driver
Add driver model support while retaining the existing legacy code. This
allows the driver to support boards that have converted to driver model
as well as those that have not.
Simon Glass [Wed, 29 Oct 2014 19:08:58 +0000 (13:08 -0600)]
dm: at91: Add platform data for GPIO on at91sam9260-based boards
These boards all have the same GPIO arrangement, so add some common platform
data that can be used by all boards. Remove the configs which are no longer
required.
Simon Glass [Thu, 13 Nov 2014 05:42:29 +0000 (22:42 -0700)]
x86: Rename chromebook-x86 to coreboot
Rename this vendor since it is intended to be used on any platform where
coreboot runs at reset and then loads U-Boot.
So far it is only tested on link. When other boards are supported it is
likely that we will need to move to multiple board names, all under the
'coreboot' vendor. So while it would be possible to remove the vendor for
now, that would be short-sighted.
Suggested-by: Bin Meng <bmeng.cn@gmail.com> Signed-off-by: Simon Glass <sjg@chromium.org>
Simon Glass [Thu, 13 Nov 2014 05:42:28 +0000 (22:42 -0700)]
x86: ivybridge: Implement SDRAM init
Implement SDRAM init using the Memory Reference Code (mrc.bin) provided in
the board directory and the SDRAM SPD information in the device tree. This
also needs the Intel Management Engine (me.bin) to work. Binary blobs
everywhere: so far we have MRC, ME and microcode.
SDRAM init works by setting up various parameters and calling the MRC. This
in turn does some sort of magic to work out how much memory there is and
the timing parameters to use. It also sets up the DRAM controllers. When
the MRC returns, we use the information it provides to map out the
available memory in U-Boot.
U-Boot normally moves itself to the top of RAM. On x86 the RAM is not
generally contiguous, and anyway some RAM may be above 4GB which doesn't
work in 32-bit mode. So we relocate to the top of the largest block of
RAM we can find below 4GB. Memory above 4GB is accessible with special
functions (see physmem).
It would be possible to build U-Boot in 64-bit mode but this wouldn't
necessarily provide any more memory, since the largest block is often below
4GB. Anyway U-Boot doesn't need huge amounts of memory - even a very large
ramdisk seldom exceeds 100-200MB. U-Boot has support for booting 64-bit
kernels directly so this does not pose a limitation in that area. Also there
are probably parts of U-Boot that will not work correctly in 64-bit mode.
The MRC is one.
There is some work remaining in this area. Since memory init is very slow
(over 500ms) it is possible to save the parameters in SPI flash to speed it
up next time. Suspend/resume support is not fully implemented, or at least
it is not efficient.
Simon Glass [Thu, 13 Nov 2014 05:42:27 +0000 (22:42 -0700)]
x86: ivybridge: Add LAPIC support
The local advanced programmable interrupt controller is not used much in
U-Boot but we do need to set it up. Add basic support for this, which will
be extended as needed.
Simon Glass [Thu, 13 Nov 2014 05:42:24 +0000 (22:42 -0700)]
x86: ivybridge: Add support for early GPIO init
When not relying on Coreboot for GPIO init the GPIOs must be set up
correctly. This is currently done statically through a rather ugly method.
As the GPIOs are figured out they can be moved to the device tree and set
up as needed rather than all at the start.
In this implementation, board files should call ich_gpio_set_gpio_map()
before the GPIO driver is used in order to provide the GPIO information.
We use the early PCI interface so that this driver can now be used before
relocation.
Simon Glass [Thu, 13 Nov 2014 05:42:18 +0000 (22:42 -0700)]
x86: Add msr read/write functions that use a structure
It is convenient to be able to adjust MSRs with a structure that splits the
two 32-bit halves into separate fields, as they are often dealt with
separately. Add a few functions to support this.
Simon Glass [Thu, 13 Nov 2014 05:42:16 +0000 (22:42 -0700)]
x86: Tidy up coreboot header usage
There is no need to explicitly write 'arch-coreboot' when including headers,
as when the arch directory points to coreboot the correct files will be
used.
Simon Glass [Thu, 13 Nov 2014 05:42:15 +0000 (22:42 -0700)]
x86: ivybridge: Add early LPC init so that serial works
The PCH (Platform Controller Hub) includes an LPC (Low Pin Count) device
which provides a serial port. This is accessible on Chromebooks, so enable
it early in the boot process.
Simon Glass [Thu, 13 Nov 2014 05:42:14 +0000 (22:42 -0700)]
x86: pci: Allow configuration before relocation
Add simple PCI access routines for x86 which permit use before relocation.
The normal PCI stack is still used, but for pre-relocation use there can
only ever be a single hose. After relocation, fall back to the normal
access, although even then on x86 machines there is normally only a single
PCI bus.
Simon Glass [Thu, 13 Nov 2014 05:42:11 +0000 (22:42 -0700)]
x86: Refactor PCI to permit alternate init
We want access PCI earlier in the init sequence, so refactor the code so
that it does not require use of a BSS variable to work. This will allow us
to use early malloc() to store information about a PCI hose.
Common PCI code moves to arch/x86/cpu/pci.c and a new
board_pci_setup_hose() function is provided by boards to set up the (single)
hose used by that board.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 13 Nov 2014 05:42:10 +0000 (22:42 -0700)]
x86: chromebook_link: Implement CAR support (cache as RAM)
Add support for CAR so that we have memory to use prior to DRAM init.
On link there is a total of 128KB of CAR available, although some is
used for the memory reference code.
Simon Glass [Thu, 13 Nov 2014 05:42:09 +0000 (22:42 -0700)]
x86: Emit post codes in startup code for Chromebooks
On x86 it is common to use 'post codes' which are 8-bit hex values emitted
from the code and visible to the user. Traditionally two 7-segment displays
were made available on the motherboard to show the last post code that was
emitted. This allows diagnosis of a boot problem since it is possible to
see where the code got to before it died.
On modern hardware these codes are not normally visible. On Chromebooks
they are displayed by the Embedded Controller (EC), so it is useful to emit
them. We must enable this feature for the EC to see the codes, so add an
option for this.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 13 Nov 2014 05:42:08 +0000 (22:42 -0700)]
x86: Build a .rom file which can be flashed to an x86 machine
On x86 machines U-Boot needs to be added to a large ROM image which is
then flashed onto the target board. The ROM has a particular format so it
makes sense for U-Boot to build this image automatically. Unfortunately
it relies on binary blobs so we cannot require this for the default
build as yet.
Simon Glass [Thu, 13 Nov 2014 05:42:07 +0000 (22:42 -0700)]
x86: Add chromebook_link board
This board is a 'bare' version of the existing 'link 'board. It does not
require coreboot to run, but is intended to start directly from the reset
vector.
This initial commit has place holders for a wide range of features. These
will be added in follow-on patches and series. So far it cannot be booted
as there is no ROM image produced, but it does build without errors.
Masahiro Yamada [Thu, 13 Nov 2014 03:28:41 +0000 (12:28 +0900)]
x86: use CONFIG_SYS_COREBOOT to descend into coreboot/ directory
The references of CONFIG_SYS_COREBOOT in arch/x86/cpu/coreboot/Makefile
are redundant because the build system descends into the directory
only when CONFIG_SYS_COREBOOT is defined.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Cc: Simon Glass <sjg@chromium.org> Acked-by: Simon Glass <sjg@chromium.org>
Simon Glass [Tue, 11 Nov 2014 01:00:22 +0000 (18:00 -0700)]
x86: Add ifdtool for working with Intel Flash Descriptor ROM images
Newer Intel chips require a Management Engine which requires a particular
format for the SPI flash that contains the boot loader. Add a tool that
supports creating and modifying these ROM images.
This tool is from Chrome OS but has been cleaned up to use U-Boot style
and to add comments. A few features have been added also.
Simon Glass [Tue, 11 Nov 2014 01:00:21 +0000 (18:00 -0700)]
dm: gpio: Add a function to read an ID from a list of GPIOs
For board IDs a common approach is to set aside several GPIOs for use in
determining the board ID. This can provide information about board features
and the revision.
Add a function that turns a list of GPIOs into an integer by assigning
each GPIO to a single bit.
Simon Glass [Tue, 11 Nov 2014 01:00:20 +0000 (18:00 -0700)]
dm: serial: Move current serial port pointer to global_data
In general we can't store things in the data section until we have inited
SDRAM. Some platforms allow this (e.g. those with SPL) but some don't. Move
the pointer to global_data so that it will work on all platforms.
Without this fix the serial port will not work prior to relocation with
driver model on some platforms.
Simon Glass [Tue, 11 Nov 2014 01:00:18 +0000 (18:00 -0700)]
Move early malloc() to before arch_cpu_init()
For some CPUs, having malloc() available very early is useful. There is no
reason to delay this since early malloc is allocated before board_init_f()
is called.
Move early malloc() init nearer to the start of the init sequence.
Bin Meng [Sun, 9 Nov 2014 14:19:13 +0000 (22:19 +0800)]
x86: Do TSC MSR calibration only for known/supported CPUs
Using MSR_PLATFORM_INFO (0xCE) to calibrate TSR will cause #GP on
processors which do not have this MSR. Instead only doing the MSR
calibration for known/supported CPUs.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Acked-by: Simon Glass <sjg@chromium.org> Tested-by: Simon Glass <sjg@chromium.org>
Simon Glass [Thu, 6 Nov 2014 20:20:10 +0000 (13:20 -0700)]
x86: Save the BIST value on reset
The built in self test value is available in register eax on start-up. Save
it so that it can be accessed later. Unfortunately we must wait until the
global_data is available before we can do this, so there is a little bit of
shuffling to keep it around.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 24 Oct 2014 00:58:47 +0000 (18:58 -0600)]
fdt: Allow ft_board_setup() to report failure
This function can fail if the device tree runs out of space. Rather than
silently booting with an incomplete device tree, allow the failure to be
detected.
Unfortunately this involves changing a lot of places in the code. I have
not changed behvaiour to return an error where one is not currently
returned, to avoid unexpected breakage.
Eventually it would be nice to allow boards to register functions to be
called to update the device tree. This would avoid all the many functions
to do this. However it's not clear yet if this should be done using driver
model or with a linker list. This work is left for later.
Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Anatolij Gustschin <agust@denx.de>
Masahiro Yamada [Thu, 6 Nov 2014 18:03:28 +0000 (03:03 +0900)]
include: move various macros to include/linux/kernel.h
U-Boot has imported various utility macros from Linux
scattering them to various places without consistency.
In include/common.h are min, max, min3, max3, ARRAY_SIZE, ALIGN,
container_of, DIV_ROUND_UP, etc.
In include/linux/compat.h are min_t, max_t, round_up, round_down,
etc.
We also have duplicated defines of min_t in some *.c files.
Moreover, we are suffering from too cluttered include/common.h.
This commit moves various macros that originate in
include/linux/kernel.h of Linux to their original position.
Note:
This commit simply moves the macros; the macros roundup,
min, max, min2, max3, ARRAY_SIZE are different
from those of Linux at this point.