]> git.sur5r.net Git - u-boot/log
u-boot
9 years agox86: Provide access to the IDT
Simon Glass [Wed, 29 Apr 2015 02:25:16 +0000 (20:25 -0600)]
x86: Provide access to the IDT

Add a function to return the address of the Interrupt Descriptor Table.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Store the GDT pointer in global_data
Simon Glass [Wed, 29 Apr 2015 02:25:15 +0000 (20:25 -0600)]
x86: Store the GDT pointer in global_data

When we start up additional CPUs we want them to use the same Global
Descriptor Table. Store the address of this in global_data so we can
reference it later.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Add an mfence macro
Simon Glass [Wed, 29 Apr 2015 02:25:14 +0000 (20:25 -0600)]
x86: Add an mfence macro

Provide access to this x86 instruction from C code.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Add defines for fixed MTRRs
Simon Glass [Wed, 29 Apr 2015 02:25:13 +0000 (20:25 -0600)]
x86: Add defines for fixed MTRRs

Add MSR numbers for the fixed MTRRs.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Add atomic operations
Simon Glass [Wed, 29 Apr 2015 02:25:12 +0000 (20:25 -0600)]
x86: Add atomic operations

Add a subset of this header file from Linux 4.0 to support atomic operations
in U-Boot.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agoAdd a 'cpu' command to print CPU information
Simon Glass [Wed, 29 Apr 2015 02:25:11 +0000 (20:25 -0600)]
Add a 'cpu' command to print CPU information

Add a simple command which provides access to a list of available CPUs along
with descriptions and basic information.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Add support for the Simple Firmware Interface (SFI)
Simon Glass [Wed, 29 Apr 2015 02:25:10 +0000 (20:25 -0600)]
x86: Add support for the Simple Firmware Interface (SFI)

This provides a way of passing information to Linux without requiring the
full ACPI horror. Provide a rudimentary implementation sufficient to be
recognised and parsed by Linux.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agodm: Implement a CPU uclass
Simon Glass [Wed, 29 Apr 2015 02:25:09 +0000 (20:25 -0600)]
dm: Implement a CPU uclass

It is useful to be able to keep track of the available CPUs in a multi-CPU
system. This uclass is mostly intended for use with SMP systems.

The uclass provides methods for getting basic information about each CPU.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agoAdd print_freq() to display frequencies nicely
Simon Glass [Wed, 29 Apr 2015 13:56:43 +0000 (07:56 -0600)]
Add print_freq() to display frequencies nicely

Add a function similar to print_size() that works for frequencies. It can
handle from Hz to GHz.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
9 years agoMove display_options functions to their own header
Simon Glass [Wed, 29 Apr 2015 02:25:07 +0000 (20:25 -0600)]
Move display_options functions to their own header

Before adding one more function, create a separate header to help reduce
the size of common.h. Add the missing function comments and tidy up.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Disable -Werror
Simon Glass [Wed, 29 Apr 2015 02:25:06 +0000 (20:25 -0600)]
x86: Disable -Werror

This is annoying during development and serves no useful purpose since
warnings are clearly displayed now that we are using Kbuild. Remove this
option.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Remove unwanted MMC debugging
Simon Glass [Wed, 29 Apr 2015 02:25:05 +0000 (20:25 -0600)]
x86: Remove unwanted MMC debugging

This printf() should not have made it into the code.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agodm: core: Add a function to bind a driver for a device tree node
Simon Glass [Wed, 29 Apr 2015 02:25:04 +0000 (20:25 -0600)]
dm: core: Add a function to bind a driver for a device tree node

Some device tree nodes do not have compatible strings but do require
drivers. This is pretty rare, and somewhat unfortunate. Add a function
to permit creation of a driver for any device tree node.

Signed-off-by: Simon Glass <sjg@chromium.org>
9 years agoFix comment nits in board_f.c
Simon Glass [Wed, 29 Apr 2015 02:25:03 +0000 (20:25 -0600)]
Fix comment nits in board_f.c

Try to make it a little clearer.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: fsp: Use reset_cpu()
Simon Glass [Wed, 29 Apr 2015 02:11:32 +0000 (20:11 -0600)]
x86: fsp: Use reset_cpu()

Now that reset_cpu() functions correctly, use it instead of directly
accessing the port on boards that use a Firmware Support Package (FSP).

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: quark: Use reset_cpu()
Simon Glass [Wed, 29 Apr 2015 02:11:31 +0000 (20:11 -0600)]
x86: quark: Use reset_cpu()

Now that reset_cpu() functions correctly, use it instead of directly
accessing the port.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: ivybridge: Use reset_cpu()
Simon Glass [Wed, 29 Apr 2015 02:11:30 +0000 (20:11 -0600)]
x86: ivybridge: Use reset_cpu()

Now that reset_cpu() functions correctly, use it instead of directly
accessing the port.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: Implement reset_cpu() correctly for modern CPUs
Simon Glass [Wed, 29 Apr 2015 02:11:29 +0000 (20:11 -0600)]
x86: Implement reset_cpu() correctly for modern CPUs

The existing code is pretty ancient and is unreliable on modern hardware.
Generally it will hang.

We can use port 0xcf9 to initiate reset on more modern hardware (say in the
last 10 years). Update the reset_cpu() function to do this, and add a new
'full reset' function to perform a full power cycle.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
9 years agox86: link: Add PCH driver to support SPI Flash
Simon Glass [Mon, 20 Apr 2015 13:07:03 +0000 (07:07 -0600)]
x86: link: Add PCH driver to support SPI Flash

U-Boot on coreboot does not have a driver for the PCH so cannot see the
SPI peripheral now that it has moved inside the PCH. Add a simple driver so
that SPI flash works again.

Signed-off-by: Simon Glass <sjg@chromium.org>
9 years agox86: Update chromebook_link instructions for binary blob
Simon Glass [Mon, 20 Apr 2015 04:05:37 +0000 (22:05 -0600)]
x86: Update chromebook_link instructions for binary blob

The MRC image is incorrect, or at least this one now does not seem to
work. Fix it.

Signed-off-by: Simon Glass <sjg@chromium.org>
9 years agox86: minnowmax: use the correct NOR in the configuration
Gabriel Huau [Sat, 25 Apr 2015 15:13:11 +0000 (08:13 -0700)]
x86: minnowmax: use the correct NOR in the configuration

The SPI NOR on the minnowboard max is a MICRON N25Q064A

Signed-off-by: Gabriel Huau <contact@huau-gabriel.fr>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Correct the typo in write_tables()
Bin Meng [Tue, 28 Apr 2015 10:37:03 +0000 (18:37 +0800)]
x86: Correct the typo in write_tables()

It should be #ifdef instead of #if.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Kconfig: Remove deprecated CONFIG_SYS_EXTRA_OPTIONS
Bin Meng [Mon, 27 Apr 2015 15:22:28 +0000 (23:22 +0800)]
x86: Kconfig: Remove deprecated CONFIG_SYS_EXTRA_OPTIONS

Currently all x86 boards still use CONFIG_SYS_EXTRA_OPTIONS to define
the text base address. Since it is deprecated, just remove it and use
CONFIG_SYS_TEXT_BASE directly.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Kconfig: Move DM_SPI & DM_SPI_FLASH to arch/Kconfig
Bin Meng [Mon, 27 Apr 2015 15:22:27 +0000 (23:22 +0800)]
x86: Kconfig: Move DM_SPI & DM_SPI_FLASH to arch/Kconfig

Since all x86 boards have been converted to use DM_SPI and
DM_SPI_FLASH, move them to arch/Kconfig x86 section.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Kconfig: MARK_GRAPHICS_MEM_WRCOMB cosmetics
Bin Meng [Mon, 27 Apr 2015 15:22:26 +0000 (23:22 +0800)]
x86: Kconfig: MARK_GRAPHICS_MEM_WRCOMB cosmetics

Remove the ending period of the MARK_GRAPHICS_MEM_WRCOMB option. Also
fix the indention of its help text.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Kconfig: Move platform options forward
Bin Meng [Mon, 27 Apr 2015 15:22:25 +0000 (23:22 +0800)]
x86: Kconfig: Move platform options forward

Move platform-specific options under in arch/x86/Kconfig forward right
after the board-specific options but before any architecture-specific
options. When it comes to the same Kconfig option, board-specific one
takes take the highest precedence, then platform-specific one, and
finally architecture-specific one.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Kconfig: Divide the target selection to vendor/model
Bin Meng [Mon, 27 Apr 2015 15:22:24 +0000 (23:22 +0800)]
x86: Kconfig: Divide the target selection to vendor/model

Let arch/x86/Kconfig prompt board vendor first, then select
the board model under that vendor. This way arch/x86/Kconfig
only needs concern board vendor and leave the supported target
list to board/<vendor>/Kconfig.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: quark: Turn on legacy segments decode
Bin Meng [Mon, 27 Apr 2015 06:16:02 +0000 (14:16 +0800)]
x86: quark: Turn on legacy segments decode

By default the legacy segments (A0000h-B0000h, E0000h-F0000h)
do not decode to system RAM. Turn on the decode 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>
9 years agox86: Check PIRQ routing table sanity in the F segment
Bin Meng [Mon, 27 Apr 2015 06:16:01 +0000 (14:16 +0800)]
x86: Check PIRQ routing table sanity in the F segment

Previously the PIRQ routing table sanity check was performed against
the original table provided by the platform codes. Now we switch to
check its sanity on the final table in the F segment as this one is
the one seen by the OS.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: minnowmax: add GPIO banks in the device tree
Gabriel Huau [Sat, 25 Apr 2015 20:16:57 +0000 (13:16 -0700)]
x86: minnowmax: add GPIO banks in the device tree

There are 6 banks:
    4 banks for CORE: available in S0 mode
    2 banks for SUS (Suspend): available in S0-S5 mode

Signed-off-by: Gabriel Huau <contact@huau-gabriel.fr>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: baytrail: fix the GPIOBASE address
Gabriel Huau [Sat, 25 Apr 2015 20:16:03 +0000 (13:16 -0700)]
x86: baytrail: fix the GPIOBASE address

The correct GPIOBASE address on the baytrail is 0x48

Signed-off-by: Gabriel Huau <contact@huau-gabriel.fr>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Correct Minnowboard instructions to use the right descriptor
Simon Glass [Sat, 25 Apr 2015 17:46:43 +0000 (11:46 -0600)]
x86: Correct Minnowboard instructions to use the right descriptor

The descriptor provided with the FSP does not seem to work. Update the
instructions to use the descriptor from the original Intel firmware.

Signed-off-by: Simon Glass <sjg@chromium.org>
9 years agobiosemu: Do not free vga_info->BIOSImage when it is 0xc0000
Bin Meng [Fri, 24 Apr 2015 07:48:05 +0000 (15:48 +0800)]
biosemu: Do not free vga_info->BIOSImage when it is 0xc0000

For x86, vga_info->BIOSImage points to 0xc0000 which cannot be freed.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agopci: Remove parameter 'class' of pci_rom_load()
Bin Meng [Fri, 24 Apr 2015 07:48:04 +0000 (15:48 +0800)]
pci: Remove parameter 'class' of pci_rom_load()

pci_rom_load() does not use its parameter 'class', so remove it.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agopci: Option rom class is a 24-bit number
Bin Meng [Fri, 24 Apr 2015 07:48:03 +0000 (15:48 +0800)]
pci: Option rom class is a 24-bit number

We should pass a u32 class number to pci_rom_probe() instead of a u16.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: queensbay: Implement PIRQ routing
Bin Meng [Fri, 24 Apr 2015 10:10:06 +0000 (18:10 +0800)]
x86: queensbay: Implement PIRQ routing

Implement Intel Queensbay platform-specific PIRQ routing support.
The chipset PIRQ routing setup is called in the arch_misc_init().

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Support platform PIRQ routing
Bin Meng [Fri, 24 Apr 2015 10:10:05 +0000 (18:10 +0800)]
x86: Support platform PIRQ routing

On x86 boards, platform chipset receives up to four different
interrupt signals from PCI devices (INTA/B/C/D), which in turn
will be routed to chipset internal PIRQ lines then routed to
8259 PIC finally if configuring the whole system to work under
the so-called PIC mode (in contrast to symmetric IO mode which
uses IOAPIC).

We add two major APIs to aid this, one for routing PIRQ and the
other one for generating a PIRQ routing table.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Write configuration tables in last_stage_init()
Bin Meng [Fri, 24 Apr 2015 10:10:04 +0000 (18:10 +0800)]
x86: Write configuration tables in last_stage_init()

We can write the configuration table in last_stage_init() for all x86
boards, but not with coreboot since coreboot already has them.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Add a function to assign IRQ numbers to PCI device
Bin Meng [Fri, 24 Apr 2015 10:10:03 +0000 (18:10 +0800)]
x86: Add a function to assign IRQ numbers to PCI device

Add a function to assign an IRQ number to PCI device's interrupt
line register in its configuration space, so that the PCI device
can have its interrupt working under PIC mode after OS boots up.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Install a default e820 table in the __weak install_e820_map()
Bin Meng [Tue, 21 Apr 2015 04:21:36 +0000 (12:21 +0800)]
x86: Install a default e820 table in the __weak install_e820_map()

Create a default e820 table with 3 entries which is enough to boot
a Linux kernel.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Clean up arch/x86/include/asm/e820.h
Bin Meng [Tue, 21 Apr 2015 04:21:35 +0000 (12:21 +0800)]
x86: Clean up arch/x86/include/asm/e820.h

There are lots of unused codes defined in e820.h, clean it up.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Add alias for SPI node in the board dts
Bin Meng [Wed, 15 Apr 2015 04:00:11 +0000 (12:00 +0800)]
x86: Add alias for SPI node in the board dts

Since Intel ICH SPI driver has been converted to driver model, we need
add an alias for SPI node in the board dts files otherwise SPI flash
won't be detected due to 'invalid bus' error.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Set serial port IRQ for SMSC LPC47M
Bin Meng [Tue, 31 Mar 2015 03:25:03 +0000 (11:25 +0800)]
x86: Set serial port IRQ for SMSC LPC47M

For starting a Linux console on the superio serial port under
interrupt mode, the IRQ number must be configured.

Signed-off-by: Jian Luo <jian.luo4@boschrexroth.de>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Move CONFIG_ENV_IS_IN_SPI_FLASH to x86-common.h
Bin Meng [Tue, 31 Mar 2015 03:51:04 +0000 (11:51 +0800)]
x86: Move CONFIG_ENV_IS_IN_SPI_FLASH to x86-common.h

Since all x86 boards use spi flash as its bootloader storage media,
it makes sense to make CONFIG_ENV_IS_IN_SPI_FLASH a common option.
So far only minnowmax board does not support it so undefine it in
its board configuration file.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: queensbay: Avoid using PCH prefix
Bin Meng [Mon, 13 Apr 2015 11:03:42 +0000 (19:03 +0800)]
x86: queensbay: Avoid using PCH prefix

The prefix PCH was taken from ivybridge port. However Queensbay
platform official document does not mention PCH. It is composed
of TunnelCreek processor and Topcliff IOH chipset. For accuracy,
avoid using PCH prefix in the macro.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: Remove the old VGA driver
Bin Meng [Mon, 13 Apr 2015 11:05:02 +0000 (19:05 +0800)]
x86: Remove the old VGA driver

CONFIG_VIDEO_VGA is no longer needed thus remove it.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agox86: minnowmax: Remove CONFIG_VIDEO_X86 in the defconfig
Bin Meng [Mon, 13 Apr 2015 11:05:22 +0000 (19:05 +0800)]
x86: minnowmax: Remove CONFIG_VIDEO_X86 in the defconfig

CONFIG_VIDEO_X86 has been replaced by CONFIG_VIDEO_VESA.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agoMerge branch 'zynq' of git://www.denx.de/git/u-boot-microblaze
Tom Rini [Wed, 29 Apr 2015 10:46:33 +0000 (06:46 -0400)]
Merge branch 'zynq' of git://www.denx.de/git/u-boot-microblaze

9 years agoMerge branch 'master' of git://www.denx.de/git/u-boot-microblaze
Tom Rini [Wed, 29 Apr 2015 10:46:03 +0000 (06:46 -0400)]
Merge branch 'master' of git://www.denx.de/git/u-boot-microblaze

9 years agoARM: zynq: rename CONFIG_ZYNQ to CONFIG_ARCH_ZYNQ
Masahiro Yamada [Mon, 16 Mar 2015 07:43:24 +0000 (16:43 +0900)]
ARM: zynq: rename CONFIG_ZYNQ to CONFIG_ARCH_ZYNQ

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agoARM: zynq: move SoC headers to mach-zynq/include/mach
Masahiro Yamada [Mon, 16 Mar 2015 07:43:23 +0000 (16:43 +0900)]
ARM: zynq: move SoC headers to mach-zynq/include/mach

Move arch/arm/include/asm/arch-zynq/*
  -> arch/arm/mach-zynq/include/mach/*

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agoARM: zynq: move SoC sources to mach-zynq
Masahiro Yamada [Mon, 16 Mar 2015 07:43:22 +0000 (16:43 +0900)]
ARM: zynq: move SoC sources to mach-zynq

Move arch/arm/cpu/armv7/zynq/* -> arch/arm/mach-zynq/*

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agoARM: zynq: pass "-mfpu=neon" only to lowlevel_init.S
Masahiro Yamada [Mon, 16 Mar 2015 07:43:21 +0000 (16:43 +0900)]
ARM: zynq: pass "-mfpu=neon" only to lowlevel_init.S

The comment line in arch/arm/cpu/armv7/zynq/config.mk says that
the option "-mfpu=neon" is necessary for compiling lowlevel_init.S.
We do not have to give it to all the source files.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynqmp: Enable SDHCI0 options
Michal Simek [Wed, 15 Apr 2015 13:21:37 +0000 (15:21 +0200)]
zynqmp: Enable SDHCI0 options

Enable SDHCI0 for zynqmp.
Add empty gpio.h because of sdhci requirement.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynqmp: Enable FS_GENERIC option
Siva Durga Prasad Paladugu [Fri, 13 Mar 2015 12:13:49 +0000 (17:43 +0530)]
zynqmp: Enable FS_GENERIC option

Provide an option to write filesystem independend commands.

Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynqmp: Add SPI driver support for ZynqMP
Siva Durga Prasad Paladugu [Wed, 15 Apr 2015 06:18:48 +0000 (11:48 +0530)]
zynqmp: Add SPI driver support for ZynqMP

Added the SPI driver support for ZynqMP
The controller is same as zynq SPI controller

Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
9 years agozynqmp: i2c: Enable i2c driver for zynqMP
Siva Durga Prasad Paladugu [Tue, 3 Mar 2015 09:31:44 +0000 (15:01 +0530)]
zynqmp: i2c: Enable i2c driver for zynqMP

Enable the i2c driver for ZynqMP
Also enable the eeprom for read and writes
to eeprom on ZynqMP
ZynqMP uses the same i2c controller as in Zynq

Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynqmp: Add support for EMMC bootmode
Michal Simek [Wed, 15 Apr 2015 13:02:28 +0000 (15:02 +0200)]
zynqmp: Add support for EMMC bootmode

Add support for EMMC bootmode.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynqmp: Add support for emulation platform - Veloce
Michal Simek [Wed, 15 Apr 2015 12:59:19 +0000 (14:59 +0200)]
zynqmp: Add support for emulation platform - Veloce

Add support for Veloce - zynqmp emulation platform.
Platform doesn't support SDHCI.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynq: timer: Fix wrong timer calculation
Siva Durga Prasad Paladugu [Mon, 13 Apr 2015 05:27:04 +0000 (10:57 +0530)]
zynq: timer: Fix wrong timer calculation

Fix wrong timer calculation in get_timer_masked incase of
overflow.
This fixes the issue of getting wrong time from get_timer()
calls.

Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynqmp: sdhci: Remove the quirk SDHCI_QUIRK_NO_CD
Siva Durga Prasad Paladugu [Tue, 8 Jul 2014 10:01:04 +0000 (15:31 +0530)]
zynqmp: sdhci: Remove the quirk SDHCI_QUIRK_NO_CD

Remove the quirk SDHCI_QUIRK_NO_CD as it is not
required.

Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynqmp: Add support for R5 sw loading
Michal Simek [Wed, 15 Apr 2015 11:36:40 +0000 (13:36 +0200)]
zynqmp: Add support for R5 sw loading

Add support for loading sw for R5 with enabling for zynqmp.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
9 years agozynqmp: caches: Enable dcache for zynqmp
Siva Durga Prasad Paladugu [Sat, 6 Dec 2014 07:27:51 +0000 (12:57 +0530)]
zynqmp: caches: Enable dcache for zynqmp

Define the mmu table till 2MB granularity
enable dcaches for zynqmp.

Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynq: slcr: Disable all level shifters
Siva Durga Prasad Paladugu [Mon, 2 Mar 2015 10:33:46 +0000 (16:03 +0530)]
zynq: slcr: Disable all level shifters

Disable all level shifters before enabling
the PS-to-PL level shifters as it would
be good to disable all level shifters before
enabling the PS-to-PL in order to ensure that
it is in proper state

Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agoARM: zynq: drop legacy ps7_init.c/h support
Masahiro Yamada [Tue, 14 Apr 2015 07:50:50 +0000 (16:50 +0900)]
ARM: zynq: drop legacy ps7_init.c/h support

We are about to change the location for ps7_init files, breaking the
current work-flows.  It is good time to drop the legacy ps7_init.c/h
support.

Going forward, please use ps7_init_gpl.c/h all the time.
If you are still using old Xilinx tools that are only able to
generate ps7_init.c/h, rename them into ps7_init_gpl.c/h.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Suggested-by: Sören Brinkmann <soren.brinkmann@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynq: Add Zynq PicoZed board support
Nathan Rossi [Tue, 14 Apr 2015 04:58:10 +0000 (14:58 +1000)]
zynq: Add Zynq PicoZed board support

The PicoZed is a System-on-Module board which is marketed as part of
the ZedBoard/MicroZed/etc. collection. It includes a Zynq-7000
processor.

This patch adds support that covers all the variants of the PicoZed
including the SKUs with Z7010/Z7020 and Z7015/Z7030 Zynq chips. This
patch set however only covers support for the System-on-Module and does
not cover any extra components that are available on carrier boards
(except those that are fanned out of the module itself).

More information on this board, its variants and available carrier
boards is available at: http://zedboard.org/product/picozed

Signed-off-by: Nathan Rossi <nathan.rossi@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agoserial: zynq: Add support for slow emulation platform
Michal Simek [Wed, 15 Apr 2015 11:05:06 +0000 (13:05 +0200)]
serial: zynq: Add support for slow emulation platform

On slow platforms not all baudrate setting is valid.
Check it directly in the driver and setup maximum possible
frequency.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agozynq: Enable GPIO driver and GPIO commands
Michal Simek [Wed, 25 Mar 2015 12:35:04 +0000 (13:35 +0100)]
zynq: Enable GPIO driver and GPIO commands

Enable GPIO driver and GPIO commands.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agogpio: add Xilinx Zynq PS GPIO driver
Andrea Scian [Fri, 20 Mar 2015 15:00:25 +0000 (16:00 +0100)]
gpio: add Xilinx Zynq PS GPIO driver

Most of the code is taken (and adapted) from Linux kernel driver.

Just add CONFIG_ZYNQ_GPIO to you config to enable it

Signed-off-by: Andrea Scian <andrea.scian@dave.eu>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agomicroblaze: Fix EMAC Lite initialization
Nathan Rossi [Tue, 14 Apr 2015 06:16:39 +0000 (16:16 +1000)]
microblaze: Fix EMAC Lite initialization

It is possible for CONFIG_XILINX_EMACLITE to be defined without
XILINX_EMACLITE_BASEADDR being defined as the EMAC Lite driver support
OF init. Check that the driver is enabled and the base address is
available before initializing with a static base address.

Signed-off-by: Nathan Rossi <nathan.rossi@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
9 years agoMerge branch 'master' of git://www.denx.de/git/u-boot-socfpga
Tom Rini [Wed, 29 Apr 2015 00:48:43 +0000 (20:48 -0400)]
Merge branch 'master' of git://www.denx.de/git/u-boot-socfpga

9 years agoMerge branch 'master' of git://www.denx.de/git/u-boot-dm
Tom Rini [Tue, 28 Apr 2015 23:37:20 +0000 (19:37 -0400)]
Merge branch 'master' of git://www.denx.de/git/u-boot-dm

9 years agodm: core: Correct bug introduced in uclass_first/next_device()
Simon Glass [Sat, 25 Apr 2015 04:33:07 +0000 (22:33 -0600)]
dm: core: Correct bug introduced in uclass_first/next_device()

These functions now rely on uclass_find_first/next_device() and assume that
they will either return failure (-ve error code) or a device. In fact,
coming to the end of a list is not considered failure and they return 0
in that case.

The logic to deal with this was replaced in commit acb9ca2a with just using
uclass_get_device_tail(). Add back the missing logic. This bug was
caught by unit tests but since they were broken for other reasons at the
time, this was not noticed.

Signed-off-by: Simon Glass <sjg@chromium.org>
9 years agoserial: ns16550: Remove hard-coded baud_divisor setting
Axel Lin [Sat, 25 Apr 2015 02:53:14 +0000 (10:53 +0800)]
serial: ns16550: Remove hard-coded baud_divisor setting

This was accidentally added by commit dd0b0122bacc
"serial: ns16550: Add an option to specify the debug UART register shift".
Remove it.

Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agodm: core: drop device removal error path correctly
Masahiro Yamada [Fri, 24 Apr 2015 08:28:40 +0000 (17:28 +0900)]
dm: core: drop device removal error path correctly

Trivial bug fix for commit 5a87c4174d18 (dm: core: Drop device
removal error path when not supported).

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agoMerge branch 'master' of git://www.denx.de/git/u-boot-imx
Tom Rini [Tue, 28 Apr 2015 16:15:13 +0000 (12:15 -0400)]
Merge branch 'master' of git://www.denx.de/git/u-boot-imx

9 years agoMerge branch 'master' of git://git.denx.de/u-boot-spi
Tom Rini [Tue, 28 Apr 2015 11:28:43 +0000 (07:28 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-spi

9 years agosf: Fix to compute proper sector_size
Jagannadha Sutradharudu Teki [Mon, 27 Apr 2015 15:34:15 +0000 (21:04 +0530)]
sf: Fix to compute proper sector_size

Upto now flash sector_size is assigned from params which isn't
necessarily a sector size from vendor, so based on the SECT_*
flags from flash_params the erase_size will compute and it will
become the sector_size finally.

Bug report (from Bin Meng):
=> sf probe
SF: Detected SST25VF016B with page size 256 Bytes, erase size 4 KiB,
total 2 MiB, mapped at ffe00000

=> sf erase 0 +100
SF: 65536 bytes @ 0x0 Erased: OK

Signed-off-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
Reported-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
9 years agodm: sf: Make SST flash write op work again
Bin Meng [Fri, 24 Apr 2015 11:51:10 +0000 (19:51 +0800)]
dm: sf: Make SST flash write op work again

With SPI flash moving to driver model, commit fbb0991 "dm: Convert
spi_flash_probe() and 'sf probe' to use driver model" ignored the
SST flash-specific write op (byte program & word program), which
actually broke the SST flash from wroking.

This commit makes SST flash work again under driver model, by adding
SST flash-specific handling in the spi_flash_std_write().

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
9 years agodm: sf: Save flash flags to struct spi_flash
Bin Meng [Tue, 28 Apr 2015 07:59:54 +0000 (13:29 +0530)]
dm: sf: Save flash flags to struct spi_flash

Add a new member 'flags' in struct spi_flash to store the flash flags
during spi_flash_validate_params().

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
9 years agospi: omap3: Fix timeout handling
David Dueck [Wed, 1 Apr 2015 12:20:24 +0000 (14:20 +0200)]
spi: omap3: Fix timeout handling

The timeout value is never reset during the transfer. This means that when
transferring more data we eventually trigger the timeout.

This was reported on the mailing list:
"Spansion SPI flash read timeout with AM335x"

Signed-off-by: David Dueck <davidcdueck@googlemail.com>
CC: Tom Rini <trini@konsulko.com>
CC: Stefan Roese <sr@denx.de>
CC: Andy Pont <andy.pont@sdcsystems.com>
Tested-by: David Dueck <davidcdueck@googlemail.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
9 years agomx6cuboxi: Load the correct 'fdtfile' variable
Fabio Estevam [Sat, 25 Apr 2015 21:47:21 +0000 (18:47 -0300)]
mx6cuboxi: Load the correct 'fdtfile' variable

Instead of hardcoding the 'fdtfile' variable, let's detect the SoC and
board variant on the fly and change the dtb name.

Based on the scheme done on am335x board.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Tested-By: Vagrant Cascadian <vagrant@debian.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
9 years agomx6cuboxi: Use more standard namings for fdt variables
Fabio Estevam [Sat, 25 Apr 2015 21:47:20 +0000 (18:47 -0300)]
mx6cuboxi: Use more standard namings for fdt variables

README file suggests to use 'fdtfile' for the dtb file name and
'fdt_addr_r' for the dtb address in RAM, so do as suggested.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
9 years agomx6cuboxi: Differentiate Cubox-i and Hummingboard
Fabio Estevam [Sat, 25 Apr 2015 21:47:19 +0000 (18:47 -0300)]
mx6cuboxi: Differentiate Cubox-i and Hummingboard

Introduce is_hummingboard() function that reads GPIOs that can distinguish
between Cubox-i and Hummingboard.

Print the board name accordingly.

Based on a patch from Rabeeh Khoury.

Signed-off-by: Rabeeh Khoury <rabeeh@solid-run.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
9 years agomx6cuboxi: Introduce multi-SoC support
Fabio Estevam [Sat, 25 Apr 2015 21:47:18 +0000 (18:47 -0300)]
mx6cuboxi: Introduce multi-SoC support

Cubox-i and Hummingboard support several MX6 SoCs: mx6solo, mx6dual-lite,
mx6dual and mx6quad. Add support for the different SoC/memory sizes
combinations.

DDR initialization values were extracted from Solid-run internal U-boot.

Tested on a CuBox-i4Pro, HummingBoard-i2eX and HummingBoard-i1.

Signed-off-by: Rabeeh Khoury <rabeeh@solid-run.com>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
9 years agomx6cuboxi: Prepare for multi SoC support
Fabio Estevam [Sat, 25 Apr 2015 21:47:17 +0000 (18:47 -0300)]
mx6cuboxi: Prepare for multi SoC support

Cubox-i and Hummingboard support several MX6 SoCs: mx6solo, mx6dual-lite,
mx6dual and mx6quad.

Use IOMUX_PADS() macro in order to prepare for the multi-SoC support.
Also pass 'MX6QDL' in the defconfig to indicate it.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
9 years agomx6cuboxi: Fix the defconfig name
Fabio Estevam [Sat, 25 Apr 2015 21:47:16 +0000 (18:47 -0300)]
mx6cuboxi: Fix the defconfig name

The correct name of the defconfig file is 'mx6cuboxi_defconfig'.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
9 years agosocfpga: implement arria V socdk SPI flash config in dts
Pavel Machek [Sat, 25 Apr 2015 19:36:16 +0000 (21:36 +0200)]
socfpga: implement arria V socdk SPI flash config in dts

Arria V SocDK has same QSPI and SPI flash configuration as Socrates. Add
support for it.

Signed-off-by: Pavel Machek <pavel@denx.de>
9 years agoMerge branch 'master' of git://git.denx.de/u-boot-fdt
Tom Rini [Fri, 24 Apr 2015 19:07:19 +0000 (15:07 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-fdt

9 years agoMerge branch 'master' of git://git.denx.de/u-boot-mips
Tom Rini [Fri, 24 Apr 2015 17:44:28 +0000 (13:44 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-mips

9 years agoMerge branch 'master' of git://git.denx.de/u-boot-fsl-qoriq
Tom Rini [Fri, 24 Apr 2015 17:43:24 +0000 (13:43 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-fsl-qoriq

9 years agoMIPS: implement device-tree handover to Linux kernel
Daniel Schwierzeck [Sun, 22 Feb 2015 15:58:30 +0000 (16:58 +0100)]
MIPS: implement device-tree handover to Linux kernel

Add device-tree handover to Linux kernel conforming with MIPS UHI [1].
Register $a0 will be set to the reserved value -2. The address of
the device-tree blob will be stored as KSEG0 address in $a1. $a2 and
$a3 are set to zero.

[1] http://prplfoundation.org/wiki/MIPS_documentation

Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
9 years agofdt: Fix handling of paths with options in them
Hans de Goede [Mon, 20 Apr 2015 09:13:37 +0000 (11:13 +0200)]
fdt: Fix handling of paths with options in them

After syncing the sunxi dts files with the upstream kernel dm/fdt sunxi
builds would no longer boot.

The problem is that stdout-path is now set like this in the upstream dts
files: stdout-path = "serial0:115200n8". The use of options in of-paths,
either after an alias name, or after a full path, e.g. stdout-path =
"/soc@01c00000/serial@01c28000:115200", is standard of usage, but something
which the u-boot dts code so far did not handle.

This commit fixes this, adding support for both path formats.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Simon Glass <sjg@chromium.org>
9 years agosocfpga: implement socdk SPI flash config in dts
Pavel Machek [Thu, 23 Apr 2015 07:14:01 +0000 (09:14 +0200)]
socfpga: implement socdk SPI flash config in dts

SocDK has same QSPI and SPI flash configuration as Socrates. Add
support for it.

Signed-off-by: Pavel Machek <pavel@denx.de>
9 years agoarmv8/fsl-lsch3: Implement workaround for I2C erratum A009203
York Sun [Mon, 23 Mar 2015 17:41:35 +0000 (10:41 -0700)]
armv8/fsl-lsch3: Implement workaround for I2C erratum A009203

This erratum requires setting GLITCH_EN bit in debug register to
enable digital filter to improve clock stability.

Signed-off-by: York Sun <yorksun@freescale.com>
CC: Heiko Schocher <hs@denx.de>
9 years agols2085a: esdhc: Add esdhc support for ls2085a
Yangbo Lu [Sat, 21 Mar 2015 02:28:31 +0000 (19:28 -0700)]
ls2085a: esdhc: Add esdhc support for ls2085a

This patch adds esdhc support for ls2085a.

Signed-off-by: Yangbo Lu <yangbo.lu@freescale.com>
Signed-off-by: York Sun <yorksun@freescale.com>
9 years agoarmv8/ls2085ardb: Enable NAND SPL support
Scott Wood [Tue, 24 Mar 2015 20:25:03 +0000 (13:25 -0700)]
armv8/ls2085ardb: Enable NAND SPL support

Enable NAND boot support using SPL framework. To boot from
NAND, either use DIP switches on board, or "qixis_reset nand"
command. Details of forming NAND image can be found in README.

Signed-off-by: Scott Wood <scottwood@freescale.com>
[York Sun: Remove +S from defconfig after commit 252ed872]
Signed-off-by: York Sun <yorksun@freescale.com>
9 years agofreescale/qixis: Add support for booting from NAND
Scott Wood [Sat, 21 Mar 2015 02:28:29 +0000 (19:28 -0700)]
freescale/qixis: Add support for booting from NAND

Use "qixis_reset nand" to reset the board to boot from NAND.

Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: York Sun <yorksun@freescale.com>
9 years agoarmv8/ls2085aqds: NAND boot support
Scott Wood [Tue, 24 Mar 2015 20:25:02 +0000 (13:25 -0700)]
armv8/ls2085aqds: NAND boot support

This adds NAND boot support for LS2085AQDS, using SPL framework.
Details of forming NAND image can be found in README.

Signed-off-by: Scott Wood <scottwood@freescale.com>
[York Sun: Remove +S from defconfig after commit 252ed872]
Signed-off-by: York Sun <yorksun@freescale.com>
9 years agodriver/ifc: Add 64KB page support
Jaiprakash Singh [Sat, 21 Mar 2015 02:28:27 +0000 (19:28 -0700)]
driver/ifc: Add 64KB page support

IFC has two register pages.Till IFC version 1.4 each
register page is 4KB each.But IFC ver 2.0 register page
size is 64KB each.IFC regiters structure is break into
two viz FCM and RUNTIME.FCM(Flash control machine) registers
are defined in PAGE0 and controls IFC generic functionality.
RUNTIME registers are defined in PAGE1 and controls NAND and
GPCM funcinality.

FCM and RUNTIME structures defination is common for IFC
version 1.4 and 2.0.

Signed-off-by: Jaiprakash Singh <b44839@freescale.com>
Signed-off-by: York Sun <yorksun@freescale.com>