]> git.kernelconcepts.de Git - karo-tx-uboot.git/log
karo-tx-uboot.git
8 years agosunxi: configs/sunxi-common.h: Enable CONFIG_CMD_PART
Hans de Goede [Tue, 3 Feb 2015 23:43:36 +0000 (00:43 +0100)]
sunxi: configs/sunxi-common.h: Enable CONFIG_CMD_PART

The recent changes to config_distro_bootcmd.h require CONFIG_CMD_PART to be
defined, as the default bootcmd not uses the "part" command.

This fixes sunxi boards not booting with v2015.04-rc1.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
8 years agoAdd linux/compiler-gcc5.h to fix builds with gcc5
Hans de Goede [Sat, 7 Feb 2015 21:52:40 +0000 (22:52 +0100)]
Add linux/compiler-gcc5.h to fix builds with gcc5

Add linux/compiler-gcc5/h from the kernel sources at:

commit 5631b8fba640a4ab2f8a954f63a603fa34eda96b
Author: Steven Noonan <steven@uplinklabs.net>
Date:   Sat Oct 25 15:09:42 2014 -0700

    compiler/gcc4+: Remove inaccurate comment about 'asm goto' miscompiles

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
8 years agoimx:mx6 set normal APS and standby PFM mode
Peng Fan [Tue, 27 Jan 2015 02:14:04 +0000 (10:14 +0800)]
imx:mx6 set normal APS and standby PFM mode

To normal mode, use APS switching mode.
To standy mode, use PFM switching mode.

Signed-off-by: Peng Fan <Peng.Fan@freescale.com>
Acked-by: Przemyslaw Marczak <p.marczak@samsung.com>
8 years agopmic: fix missing SWITCH_SIZE
Stefano Babic [Wed, 11 Feb 2015 11:35:46 +0000 (12:35 +0100)]
pmic: fix missing SWITCH_SIZE

Applying ccbb18713b279f1326479cc10664d247206e9e76,
the define disappeared. Fix it.

Signed-off-by: Stefano Babic <sbabic@denx.de>
8 years agopmic:pfuze implement pmic_mode_init
Peng Fan [Tue, 27 Jan 2015 02:14:03 +0000 (10:14 +0800)]
pmic:pfuze implement pmic_mode_init

This patch is to implement pmic_mode_init function, and add prototype
in header file.

This function is to set switching mode for pmic buck regulators to
improve system efficiency.

Mode:
OFF: The regulator is switched off and the output voltage is discharged.
PFM: In this mode, the regulator is always in PFM mode, which
     is useful at light loads for optimized efficiency.
PWM: In this mode, the regulator is always in PWM mode operation
     regardless of load conditions.
APS: In this mode, the regulator moves automatically between
     pulse skipping mode and PWM mode depending on load conditions.

Signed-off-by: Peng Fan <Peng.Fan@freescale.com>
Acked-by: Przemyslaw Marczak <p.marczak@samsung.com>
8 years agolcd: split splash code into its own function
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:32 +0000 (13:32 +0200)]
lcd: split splash code into its own function

lcd_logo() currently performs tasks well beyond just displaying the logo.
It has code which displays splash image, it has logic which determines
when the different display features are displayed, and it is coupled with
the lcd console because it holds the responsibility of returning the
lcd console base address.

Make lcd_logo() just about the logo by:
* Moving splash image display code into a dedicated function
* Moving the logic regarding when various features are displayed to
  lcd_clear() (which is arguably not the correct name for housing such
  code either, but it is currently the most fitting location code wise)
* Move the responsibility of setting the console base address to
  lcd_clear() too.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: dt: extract simplefb support
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:31 +0000 (13:32 +0200)]
lcd: dt: extract simplefb support

We now have api functions that can support compiling simplefb code as its own
module. Since this code is not part of the display functionality, extract it
to its own file.

Raspberry Pi is updated to accommodate the changes.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
Cc: Stephen Warren <swarren@wwwdotorg.org>
8 years agolcd: rename bitmap_plot to better represent its functionality
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:30 +0000 (13:32 +0200)]
lcd: rename bitmap_plot to better represent its functionality

The name "bitmap_plot" is misleading because it implies that this is a generic
function capable of dealing with any bitmap, but its implementation only works
with the logo data.

Rename the function to better reflect this.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: various cleanups
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:29 +0000 (13:32 +0200)]
lcd: various cleanups

This cleanup mostly focuses on removing unnecessary whitespace and comments
which are superfluous and/or do not conform to the coding style.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: remove unused includes
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:28 +0000 (13:32 +0200)]
lcd: remove unused includes

Remove unused includes.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: introduce lcd_set_cmap
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:27 +0000 (13:32 +0200)]
lcd: introduce lcd_set_cmap

Reduce the lcd_display_bitmap #ifdef complexity by extracting Atmel-specific
code for setting cmap for bitmap images into a new function lcd_set_cmap().
A default version is implemented with the remainder of the code.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: logo: move generic cmap setting to lcd_logo_set_cmap()
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:26 +0000 (13:32 +0200)]
lcd: logo: move generic cmap setting to lcd_logo_set_cmap()

Get rid of platform-specific #ifdefs in bitmap_plot() by moving the generic
case of setting cmap into the weak lcd_logo_set_cmap().

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: mpc823: move mpc823-specific lcd_logo_set_cmap code to mpc8xx_lcd.c
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:25 +0000 (13:32 +0200)]
lcd: mpc823: move mpc823-specific lcd_logo_set_cmap code to mpc8xx_lcd.c

Reduce the bitmap_plot #ifdef complexity by extracting MPC823-specific code for
setting cmap into its own implementation of lcd_logo_set_cmap(), implemented in
mpc8xx_lcd.c. In the MPC823 implementation, ARRAY_SIZE(bmp_logo_palette) is
switched for BMP_LOGO_COLORS to avoid having to include bmp_logo_data.h, which
would cause a compilation error because the logo data and palette arrays would
be defined twice.

This is a step towards cleaning bitmap_plot() of platform-specific code.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: atmel: introduce lcd_logo_set_cmap
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:24 +0000 (13:32 +0200)]
lcd: atmel: introduce lcd_logo_set_cmap

Reduce the bitmap_plot #ifdef complexity by extracting Atmel-specific code for
setting cmap into a new function lcd_logo_set_cmap(), which is implemented in
atmel_lcdfb driver and defined as part of common/lcd.c api with a weak dummy
version. In the Atmel implementation, ARRAY_SIZE(bmp_logo_palette) is
switched for BMP_LOGO_COLORS to avoid having to include bmp_logo_data.h, which
would cause a compilation error because the logo data and palette arrays would
be defined twice.

This is a step towards cleaning bitmap_plot() of platform-specific code.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: mpc8xx: move mpc823-specific fb_put_byte to mpc8xx_lcd.c
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:23 +0000 (13:32 +0200)]
lcd: mpc8xx: move mpc823-specific fb_put_byte to mpc8xx_lcd.c

Reduce the amount of platform-specific code in common/lcd.c by moving MPC823
implementation of fb_put_byte() to mpc8xx_lcd.c. Since we must also have a
default implementation for everybody else, make the remainder of the code
into a weak function.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: atmel: move atmel-specific fb_put_word to atmel_lcdfb
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:22 +0000 (13:32 +0200)]
lcd: atmel: move atmel-specific fb_put_word to atmel_lcdfb

Reduce the amount of platform-specific code in common/lcd.c by moving Atmel
implementation of fb_put_word() to atmel_lcdfb.c. Since we must also have a
default implementation for everybody else, make the remainder of the code
into a weak function.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: split configuration_get_cmap
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:21 +0000 (13:32 +0200)]
lcd: split configuration_get_cmap

configuration_get_cmap() is multiple platform-specific functions stuffed into
one function. Split it into multiple versions, and move each version to the
appropriate driver to reduce the #ifdef complexity.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agolcd: move platform-specific structs to their own headers
Nikita Kiryanov [Tue, 3 Feb 2015 11:32:20 +0000 (13:32 +0200)]
lcd: move platform-specific structs to their own headers

common/lcd code is full of platform-specific code and definitions, which
ideally should reside with the respective driver code. Take a step towards that
goal by moving platform-specific structs from lcd.h to their own header files.

The structs for the generic case (the #else for all the platform-specific
cases) is retained in lcd.h as the default case.

Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Bo Shen <voice.shen@atmel.com>
Tested-by: Josh Wu <josh.wu@atmel.com>
Cc: Bo Shen <voice.shen@atmel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Anatolij Gustschin <agust@denx.de>
8 years agomx53loco: Fix boot hang during reboot stress test
Fabio Estevam [Mon, 9 Feb 2015 09:29:33 +0000 (07:29 -0200)]
mx53loco: Fix boot hang during reboot stress test

Currently by running the following test:

=> setenv bootcmd reset
=> save
=> reset

, we observe a hang after approximately 20-30 minutes of stress reboot test.

Investigation of this issue revealed that when a single DDR chip select is used,
the hang does not happen. It only happens when the two chip selects are active.

MX53 reference manual states at "28.6.2 Memory ZQ calibration sequence":

"The controller must keep the memory lines quiet (except for CK) for the ZQ
calibration time as defined in the Jedec (512 cycles for ZQCL after reset, 256
for other ZQCL and 64 for ZQCS)."

According to the SDE_0 and SDE_1 bit descriptions from register ESDCTL_ESDCTL:

"Writing 1 to SDE0 or SDE1 will initiate power up delays as JEDEC defines.
Power up delays are a function of the configured memory type (DDR2/DDR3/LPDDR2)"

So make sure to activate one chip select at time (CS0 first and then CS1 later),
so that the required JEDEC delay is respected for each chip select.

With this change applied the board has gone through three days of reboot stress
test without any hang.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Stefano Babic <sbabic@denx.de>
8 years agoarm: mxs: Add 'Wait for JTAG user' if booted in JTAG mode
Graeme Russ [Sun, 25 Jan 2015 01:07:53 +0000 (12:07 +1100)]
arm: mxs: Add 'Wait for JTAG user' if booted in JTAG mode

When booting in JTAG mode, there is no way to use soft break-points, and
no way of knowing when SPL has finished executing (so the user can issue
a 'halt' command to load u-boot.bin for example)

Add a debug output and simple loop to stop execution at the completion of
the SPL initialisation as a pseudo break-point when booting in JTAG mode

Signed-off-by: Graeme Russ <gruss@tss-engineering.com>
8 years agoarm: mxs: Enable booting of mx28 without battery
Graeme Russ [Sun, 25 Jan 2015 01:07:52 +0000 (12:07 +1100)]
arm: mxs: Enable booting of mx28 without battery

Section 4.1.2 of Freescale Application Note AN4199 describes the
configuration required to operate the mx28 from a 5V source without a
battery.

This patch changes the behaviour of the dropout control of the DC-DC
converter (refer to section 11.12.9 of the mx28 Application Processor
Reference Manual - Document Number: MCIMX28RM, Rev 2, 08/2013) to the
following:
 - Always use 4P2 Linear Regulator if CONFIG_SYS_MXS_VDD5V_ONLY is defined
 - Switch between 4P2 Linear Regulator and Battery, using whichever has
   the highest voltage if CONFIG_SYS_MXS_VDD5V_ONLY isnot set (this is
   the same as the pre-patch behaviour)

Signed-off-by: Graeme Russ <gruss@tss-engineering.com>
Signed-off-by: Damien Gotfroi <dgotfroi@greenwatch.be>
8 years agoarm: mxs: Add debug outputs and comments to mxs SPL source files
Graeme Russ [Sun, 25 Jan 2015 01:07:51 +0000 (12:07 +1100)]
arm: mxs: Add debug outputs and comments to mxs SPL source files

It is difficult to track down fail to boot issues in the mxs SPL.
Implement the following to make it easier:
 - Add debug outputs to allow tracing of SPL progress in order to track
where failure to boot occurs. DEUBUG and CONFIG_SPL_SERIAL_SUPPORT must
be defined to enable debug output in SPL
 - Add TODO comments where it is not clear if the code is doing what it
is meant to be doing, even tough the board boots properly (these comments
refer to existing code, not to any code added by this patch)

Signed-off-by: Graeme Russ <gruss@tss-engineering.com>
8 years agoimx: mx6: Fixed AIPS3 base address issue
Ye.Li [Wed, 14 Jan 2015 09:18:12 +0000 (17:18 +0800)]
imx: mx6: Fixed AIPS3 base address issue

Should use AIPS3 configuration address 0x0227C000 to set AIPS3,
not the AIPS3 base address.
Additional, replace AIPS1_BASE_ADDR to AIPS3_ARB_BASE_ADDR to align with
AIPS1 and AIPS2, and resolve the AIPS3_ARB_BASE_ADDR undefine problem.

Signed-off-by: Ye.Li <B37916@freescale.com>
8 years agoimx: imximage: Add QuadSPI boot support
Ye.Li [Tue, 13 Jan 2015 07:53:06 +0000 (15:53 +0800)]
imx: imximage: Add QuadSPI boot support

Add QuadSPI boot support to imximage tool.

Note: The QuadSPI configuration parameters at offset 0x400 are not
included in this patch. Need other tools to generate the parameters
part.

Signed-off-by: Ye.Li <B37916@freescale.com>
8 years agoimx: mx6qsabreauto: Change to use common GPMI IO clock function
Ye.Li [Mon, 12 Jan 2015 09:37:13 +0000 (17:37 +0800)]
imx: mx6qsabreauto: Change to use common GPMI IO clock function

Since a clock function setup_gpmi_io_clk is implemented for GPMI
IO clock settings, change to use this common function in GPMI setup.

Signed-off-by: Ye.Li <B37916@freescale.com>
Acked-by: Stefano Babic <sbabic@denx.de>
8 years agopmic:pfuz100 add switch mode and more registers
Peng Fan [Thu, 8 Jan 2015 13:00:36 +0000 (21:00 +0800)]
pmic:pfuz100 add switch mode and more registers

Add more pfuze register offset.
And switch mode definition.

Signed-off-by: Peng Fan <Peng.Fan@freescale.com>
8 years agopmic:pfuze add macro for setting voltage
Peng Fan [Fri, 9 Jan 2015 08:59:41 +0000 (16:59 +0800)]
pmic:pfuze add macro for setting voltage

"#define PFUZE100_SW1ABC_SETP(x) ((x - 3000) / 250)"
This macro is for configuring SW1A/B/C Output Voltage easily.

Signed-off-by: Peng Fan <Peng.Fan@freescale.com>
8 years agomicroblaze: spl: Add LISTS to linker script
Michal Simek [Tue, 3 Feb 2015 14:09:52 +0000 (15:09 +0100)]
microblaze: spl: Add LISTS to linker script

This is required for driver model.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: spl: Do not call mem_malloc_init and use early alloc
Michal Simek [Tue, 3 Feb 2015 15:24:48 +0000 (16:24 +0100)]
microblaze: spl: Do not call mem_malloc_init and use early alloc

This patch has some parts connected together:
- Use _gd in bss section which is automatically cleared
  Location at SPL_MALLOC_END wasn't cleared at all
- Use MALLOC_F_LEN(early alloc) instead of FULL MALLOC
  (mem_malloc_init is not called at all)
- Simplify malloc and stack init.
  At the end of SPL addr is malloc area and below is stack

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Add support for CONFIG_SYS_MALLOC_F_LEN
Michal Simek [Fri, 30 Jan 2015 14:46:43 +0000 (15:46 +0100)]
microblaze: Add support for CONFIG_SYS_MALLOC_F_LEN

Create space for dm_init where calloc is called
and malloc_base has to be initialized.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Do not use CONFIG_SYS_GENERIC_GLOBAL_DATA
Michal Simek [Fri, 30 Jan 2015 14:45:02 +0000 (15:45 +0100)]
microblaze: Do not use CONFIG_SYS_GENERIC_GLOBAL_DATA

Because it is not compatible with DM where
malloc_base has to be available early and init
has to be done in ASM.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agobdinfo: Show information about fdt blob via bdinfo
Michal Simek [Tue, 27 Jan 2015 14:41:08 +0000 (15:41 +0100)]
bdinfo: Show information about fdt blob via bdinfo

Microblaze target supports both OF and !OF cases
and from log is not clear which version is running.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Speedup code copy
Michal Simek [Tue, 27 Jan 2015 14:10:37 +0000 (15:10 +0100)]
microblaze: Speedup code copy

Remove one instruction in the loop which speedup
code copying.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Move architecture to use generic board init
Michal Simek [Thu, 8 May 2014 14:08:44 +0000 (16:08 +0200)]
microblaze: Move architecture to use generic board init

Compile code with -fPIC to get GOT. Do not build SPL
with fPIC because it increasing SPL size for nothing.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Enable SPL_NOR support when FLASH_BASE is setup
Michal Simek [Tue, 27 Jan 2015 13:25:38 +0000 (14:25 +0100)]
microblaze: Enable SPL_NOR support when FLASH_BASE is setup

Simplify SPL NOR init.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Fix gd_t address which is placed at the end of BRAM
Michal Simek [Tue, 27 Jan 2015 13:06:14 +0000 (14:06 +0100)]
microblaze: Fix gd_t address which is placed at the end of BRAM

Setup gd from ASM to be availalbe for board_init_r.
Setting it up in spl_board_init is too late when
MALLOC is used.
Space for gd is located behind MALLOC area at the end of BRAM.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Remove unused asm label
Michal Simek [Tue, 27 Jan 2015 11:46:07 +0000 (12:46 +0100)]
microblaze: Remove unused asm label

It is not used at all that's why remove it.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Use standard interrupt_init() function
Michal Simek [Tue, 27 Jan 2015 11:44:12 +0000 (12:44 +0100)]
microblaze: Use standard interrupt_init() function

Do not use microblaze specific interrupt init function.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Remove unneeded data section adding from DTB
Michal Simek [Mon, 26 Jan 2015 14:52:48 +0000 (15:52 +0100)]
microblaze: Remove unneeded data section adding from DTB

DTB is added to rodata section:
  [ 2] .rodata           PROGBITS        84c5b60c 05c60c 00c618 00   A
0   0  4
  [ 3] .dtb.init.rodata  PROGBITS        84c67c30 068c30 003c80 00   A
0   0 16
  [ 4] .rela.dyn         RELA            84c6b8b0 06c8b0 000534 0c   A
0   0  4
  [ 5] .data             PROGBITS        84c6bde4 06cde4 001536 00  WA
0   0 16

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Add debug message about enabling interrupts
Michal Simek [Mon, 26 Jan 2015 14:25:32 +0000 (15:25 +0100)]
microblaze: Add debug message about enabling interrupts

Add one more debug message about enabling global interrupts.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Fix coding style
Michal Simek [Mon, 26 Jan 2015 13:39:22 +0000 (14:39 +0100)]
microblaze: Fix coding style

No functional changes just to pass checkpatch.pl.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Remove DEBUG_INT macro and use debug() instead
Michal Simek [Mon, 26 Jan 2015 13:37:52 +0000 (14:37 +0100)]
microblaze: Remove DEBUG_INT macro and use debug() instead

Do not use specific macros for debugging.
Also remove compilation warning:
w+../arch/microblaze/cpu/interrupts.c: In function 'interrupt_handler':
w+../arch/microblaze/cpu/interrupts.c:153:2: warning: format '%x'
expects argument of type 'unsigned int', but argument 2 has type 'void
(*)(void *)' [-Wformat]
w+../arch/microblaze/cpu/interrupts.c:153:2: warning: format '%x'
expects argument of type 'unsigned int', but argument 4 has type 'void
*' [-Wformat]

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Fix coding style in exception.c
Michal Simek [Mon, 26 Jan 2015 13:36:13 +0000 (14:36 +0100)]
microblaze: Fix coding style in exception.c

Just coding style cleanup - no functional changes.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Show return address from exception
Michal Simek [Mon, 26 Jan 2015 13:32:23 +0000 (14:32 +0100)]
microblaze: Show return address from exception

Show also return address from exception which should
suggest where the problem is.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agomicroblaze: Fix stack usage in interrupt handler
Michal Simek [Mon, 26 Jan 2015 13:24:08 +0000 (14:24 +0100)]
microblaze: Fix stack usage in interrupt handler

Do not save registers below r1 stack pointer because
it is not checked by stack undeflow is not able to detect
it.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
8 years agocommon/board_r: manual relocation for cmd table
Andreas Bießmann [Mon, 19 Jan 2015 23:29:05 +0000 (00:29 +0100)]
common/board_r: manual relocation for cmd table

This is required for architectures still need manual relocation like avr32, mk68
and others.

Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Michal Simek <michal.simek@xilinx.com>
8 years agocommon: Move dram_init() declaration to common location
Michal Simek [Fri, 30 Jan 2015 09:51:46 +0000 (10:51 +0100)]
common: Move dram_init() declaration to common location

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoarc: build libgcc in U-Boot
Alexey Brodkin [Tue, 3 Feb 2015 10:58:20 +0000 (13:58 +0300)]
arc: build libgcc in U-Boot

This way we may have very limited set of functions implemented so we
save some space.

Also it allows us to build U-Boot for any ARC core with the same one
toolchain because we don't rely on pre-built libgcc.

For example:
 * we may use little-endian toolchain but build U-Boot for ether
endianess
 * we may use non-multilibbed uClibc toolchain but build U-Boot for
whatever ARC CPU flavour that current GCC supports

Private libgcc built from generic C implementation contributes only 144
bytes to .text section so we don't see significant degradation of size:
--->8---
$ arc-linux-size u-boot.libgcc-prebuilt
   text    data     bss     dec     hex filename
 222217   24912  214820  461949   70c7d u-boot.libgcc-prebuilt

$ arc-linux-size u-boot.libgcc-private
   text    data     bss     dec     hex filename
 222361   24912  214820  462093   70d0d u-boot.libgcc-private
--->8---

Also I don't notice visible performance degradation compared to
pre-built libgcc (where at least "*div*" functions are had-written in
assembly) on typical operations of downloading 10Mb uImage over TFTP and
bootm.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: move CPU flags selection to the main "config.mk"
Alexey Brodkin [Tue, 3 Feb 2015 10:58:19 +0000 (13:58 +0300)]
arc: move CPU flags selection to the main "config.mk"

As a preparation to ARCv2 port submission we're moving CPU slection
flags to a common location.
Also it will allow us to have more flexible CPU specification, not only
ISA version but CPU family as well checking CONFIG_ARC_CPU_xxx.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: move SYS_MONITOR_BASE setup in Konfig
Alexey Brodkin [Tue, 3 Feb 2015 10:58:18 +0000 (13:58 +0300)]
arc: move SYS_MONITOR_BASE setup in Konfig

Following SPARK ARC now has SYS_MONITOR_BASE setup via Kconfig.
This makes "include/configs/*.h" cleaner and more flexible.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: hard-code CONFIG_ARCH_EARLY_INIT_R in asm/config.h
Alexey Brodkin [Tue, 3 Feb 2015 10:58:17 +0000 (13:58 +0300)]
arc: hard-code CONFIG_ARCH_EARLY_INIT_R in asm/config.h

Common arch_early_init_r() is used in "arc/lib/cpu.c" for all ARC boards
so there's no sense in separate per-board definitions.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: get rid of useless CONFIG_SKIP_LOWLEVEL_INIT
Alexey Brodkin [Tue, 3 Feb 2015 10:58:16 +0000 (13:58 +0300)]
arc: get rid of useless CONFIG_SKIP_LOWLEVEL_INIT

Currently there's nothing related to really low-level init on ARC so
CONFIG_SKIP_LOWLEVEL_INIT definition makes no sense.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: hard-code CONFIG_SYS_GENERIC_BOARD into asm/config.h
Alexey Brodkin [Tue, 3 Feb 2015 10:58:15 +0000 (13:58 +0300)]
arc: hard-code CONFIG_SYS_GENERIC_BOARD into asm/config.h

There're no other options for ARC except "generic board" so ther's no
point to define CONFIG_SYS_GENERIC_BOARD per board.

We now have it set fo all ARC boards.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: add selection of endianess in Kconfig
Alexey Brodkin [Tue, 3 Feb 2015 10:58:14 +0000 (13:58 +0300)]
arc: add selection of endianess in Kconfig

This change allows to keep board description clean and minimalistic.
This is especially helpful if one board may house different CPUs with
different features.

It is applicable to both FPGA-based boards or those that have CPUs
mounted on interchnagable daughter-boards.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: select cache settings via menuconfig
Alexey Brodkin [Tue, 3 Feb 2015 10:58:13 +0000 (13:58 +0300)]
arc: select cache settings via menuconfig

This change allows to keep board description clean and minimalistic.
This is especially helpful if one board may house different CPUs with
different features.

It is applicable to both FPGA-based boards or those that have CPUs
mounted on interchnagable daughter-boards.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: define and use PTAG AUX regs for MMUv3 only
Alexey Brodkin [Tue, 3 Feb 2015 10:58:12 +0000 (13:58 +0300)]
arc: define and use PTAG AUX regs for MMUv3 only

DC_PTAG and IC_PTAG registers only exist in MMUv3.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: add more flavours of ARC700 series CPU
Alexey Brodkin [Tue, 3 Feb 2015 10:58:11 +0000 (13:58 +0300)]
arc: add more flavours of ARC700 series CPU

Now we may select a particular version of ARC700:
 * ARC750D or
 * ARC770D

It allows more flexible (or more fine tuned) configuration of U-Boot.
Before that change we relied on minimal configuration but now we may
use specific features of each CPU.

Moreover allows us to escape manual selection of options that
exist in both CPUs but may have say different version like MMUv2 in
ARC750D vs MMUv3 in ARC770D.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: remove CPU hard-coded selection from board description in include/configs
Alexey Brodkin [Tue, 3 Feb 2015 10:58:10 +0000 (13:58 +0300)]
arc: remove CPU hard-coded selection from board description in include/configs

With switch to Kconfig we only need very board-specific descriptions in
include/configs.

CPU selection is performed with either defconfig or manually via
menuconfig.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
8 years agoarc: memcmp - fix zero-delay loop utilization
Igor Guryanov [Tue, 3 Feb 2015 10:58:09 +0000 (13:58 +0300)]
arc: memcmp - fix zero-delay loop utilization

It's prohibited to put branch instruction in the very end of zero-delay
loop. On execution this causes "Illegal instruction" exception.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Igor Guryanov <guryanov@synopsys.com>
8 years agoarm, at91: taurus remove MACH_TYPE definitions in config file
Heiko Schocher [Wed, 21 Jan 2015 08:22:20 +0000 (09:22 +0100)]
arm, at91: taurus remove MACH_TYPE definitions in config file

remove MACH_TYPE definitions in config file, as they come from
the defconfig.

Signed-off-by: Heiko Schocher <hs@denx.de>
8 years agoarm, at91: add reset controller status register
Heiko Schocher [Wed, 21 Jan 2015 07:42:53 +0000 (08:42 +0100)]
arm, at91: add reset controller status register

add reset controller status register

Signed-off-by: Heiko Schocher <hs@denx.de>
Acked-by: Bo Shen <voice.shen@atmel.com>
8 years agoarm, at91, taurus: enable WDT
Heiko Schocher [Wed, 21 Jan 2015 07:38:23 +0000 (08:38 +0100)]
arm, at91, taurus: enable WDT

enable WDT for the taurus board.

Signed-off-by: Heiko Schocher <hs@denx.de>
8 years agoarm, at91, wdt: make timeout configurable
Heiko Schocher [Wed, 21 Jan 2015 07:38:22 +0000 (08:38 +0100)]
arm, at91, wdt: make timeout configurable

make the HW WDT timeout configurable through the define
CONFIG_AT91_HW_WDT_TIMEOUT.

Signed-off-by: Heiko Schocher <hs@denx.de>
8 years agocommon/board_f: add at91 wdt
Heiko Schocher [Wed, 21 Jan 2015 07:38:21 +0000 (08:38 +0100)]
common/board_f: add at91 wdt

call hw_watchdog_init() also if CONFIG_AT91SAM9_WATCHDOG
is used.

Signed-off-by: Heiko Schocher <hs@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoarm, at91, wdt: do not disable WDT in SPL
Heiko Schocher [Wed, 21 Jan 2015 07:38:20 +0000 (08:38 +0100)]
arm, at91, wdt: do not disable WDT in SPL

if CONFIG_AT91SAM9_WATCHDOG is set, do not disable WDT in
SPL

Signed-off-by: Heiko Schocher <hs@denx.de>
8 years agoARM: at91: at91sam9x5: save environment to a FAT file in MMC card
Wu, Josh [Tue, 20 Jan 2015 02:33:33 +0000 (10:33 +0800)]
ARM: at91: at91sam9x5: save environment to a FAT file in MMC card

This patch will save U-Boot environment as a file: uboot.env, in FAT partition
instead of saving it in raw sector of MMC card.

This make us easier to manage the environment file.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: at91: sama5d3xek: save enviroment as a FAT file in MMC card
Wu, Josh [Tue, 20 Jan 2015 02:33:32 +0000 (10:33 +0800)]
ARM: at91: sama5d3xek: save enviroment as a FAT file in MMC card

This patch will save U-Boot environment as a file: uboot.env, in FAT partition
instead of in raw sector of MMC card.

This make us easier to manage the environment file.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: at91: sama5d3_xplained: save environment in a FAT file in MMC card
Wu, Josh [Tue, 20 Jan 2015 02:33:31 +0000 (10:33 +0800)]
ARM: at91: sama5d3_xplained: save environment in a FAT file in MMC card

This patch will save U-Boot environment as a file: uboot.env, in FAT partition
instead of saving it in raw sector of MMC card.

This make us easier to manage the environment file.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: at91: mmc portA support is only for at91sam9g20ek_2mmc board
Wu, Josh [Mon, 19 Jan 2015 07:25:56 +0000 (15:25 +0800)]
ARM: at91: mmc portA support is only for at91sam9g20ek_2mmc board

Current the MMC support will enable MCI port A, Which is only exist
for 2mmc board.
So by default we need to disable MMC (port A) support. And only enable
it for 2mmc board. Otherwise, dataflash won't work in at91sam9260ek board
as MMC has confliction with Dataflash in the CLK pin.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
8 years agomtd: atmel_nand: according to pmecc version to perform 0xff page correction
Wu, Josh [Fri, 16 Jan 2015 03:54:46 +0000 (11:54 +0800)]
mtd: atmel_nand: according to pmecc version to perform 0xff page correction

As the PMECC hardware has different version. In SAMA5D4 chip, the PMECC ip
can generate 0xff pmecc ECC value for all 0xff sector.

According to this, add PMECC version check, if it's SAMA5D4 then we always
let PMECC hardware to correct it.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Bo Shen <voice.shen@atmel.com>
Acked-by: Andreas Bießmann <andreas.devel@googlemail.com>
8 years agoARM: atmel: cleanup: remove at91cap9 related code
Bo Shen [Fri, 16 Jan 2015 02:55:46 +0000 (10:55 +0800)]
ARM: atmel: cleanup: remove at91cap9 related code

As the at91cap9adk board is removed by commit: b5508344
(ARM: remove broken "at91cap9adk" board), so the at91cap9
code is not used anymore, and also the document for
at91cap9 can not be found on www.atmel.com, so remove the
at91cap9 related code.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
Acked-by: Andreas Bießmann <andreas.devel@googlemail.com>
8 years agoARM: atmel: sama5d4_xplained: enable spl support
Bo Shen [Mon, 15 Dec 2014 05:24:39 +0000 (13:24 +0800)]
ARM: atmel: sama5d4_xplained: enable spl support

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: sama5d4ek: enable SPL support
Bo Shen [Mon, 15 Dec 2014 05:24:38 +0000 (13:24 +0800)]
ARM: atmel: sama5d4ek: enable SPL support

The sama5d4ek support boot up from NAND flash, SD/MMC card and
also the SPI flash.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: sama5d4: build related file when enable SPL
Bo Shen [Mon, 15 Dec 2014 05:24:37 +0000 (13:24 +0800)]
ARM: atmel: sama5d4: build related file when enable SPL

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: sama5d4: can access DDR in interleave mode
Bo Shen [Mon, 15 Dec 2014 05:24:36 +0000 (13:24 +0800)]
ARM: atmel: sama5d4: can access DDR in interleave mode

The SAMAA5D4 SoC can access DDR in interleave mode.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: sama5d4: add interrupt redirect function
Bo Shen [Mon, 15 Dec 2014 05:24:35 +0000 (13:24 +0800)]
ARM: atmel: sama5d4: add interrupt redirect function

Signed-off-by: Bo Shen <voice.shen@atmel.com>
[fix subject]
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
8 years agoARM: atmel: sama5d4: add bus matrix init function
Bo Shen [Mon, 15 Dec 2014 05:24:34 +0000 (13:24 +0800)]
ARM: atmel: sama5d4: add bus matrix init function

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: sama5d4: add matrix1 base addr definition
Bo Shen [Mon, 15 Dec 2014 05:24:33 +0000 (13:24 +0800)]
ARM: atmel: sama5d4: add matrix1 base addr definition

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: spl: can not disable osc for sama5d4
Bo Shen [Mon, 15 Dec 2014 05:24:32 +0000 (13:24 +0800)]
ARM: atmel: spl: can not disable osc for sama5d4

The SAMA5D4 SoC on chip rc oscillator can not be disabled.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: spl: add saic to aic redirect function
Bo Shen [Mon, 15 Dec 2014 05:24:31 +0000 (13:24 +0800)]
ARM: atmel: spl: add saic to aic redirect function

Some SoC need to redirect the saic to aic to make the interrupt to
work, here add a weak function to be replaced by real function.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: spl: add weak bus matrix init function
Bo Shen [Mon, 15 Dec 2014 05:24:30 +0000 (13:24 +0800)]
ARM: atmel: spl: add weak bus matrix init function

Some SoC need to configure the bus matrix, add an weak function
to be replace by real function.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agoARM: atmel: sama5: add sfr register header file
Bo Shen [Mon, 15 Dec 2014 05:24:29 +0000 (13:24 +0800)]
ARM: atmel: sama5: add sfr register header file

The SFR (special function registers) can be shared bwteen
sama5d3 and sama5d4 soc.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
[whitespace adoptions for 80 char compliance]
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
8 years agoARM: atmel: sama5: add bus matrix header file
Bo Shen [Mon, 15 Dec 2014 05:24:28 +0000 (13:24 +0800)]
ARM: atmel: sama5: add bus matrix header file

This matrix header file can be shared between sama5d3 and sama5d4 soc.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
[whitespace adaptions for 80 char compliance]
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
8 years agoARM: atmel: clock: make it possible to configure HMX32
Bo Shen [Mon, 15 Dec 2014 05:24:27 +0000 (13:24 +0800)]
ARM: atmel: clock: make it possible to configure HMX32

Signed-off-by: Bo Shen <voice.shen@atmel.com>
8 years agokwbimage: Make the Makefile pass in CONFIG_SYS_SPI_U_BOOT_OFFS
Tom Rini [Sat, 7 Feb 2015 12:03:00 +0000 (07:03 -0500)]
kwbimage: Make the Makefile pass in CONFIG_SYS_SPI_U_BOOT_OFFS

We can't use config.h directly as some platforms include headers that
aren't safe to use in normal Linux userland.

Signed-off-by: Tom Rini <trini@ti.com>
8 years agox86: Use tab instead of space to indent in PCIE_ECAM_BASE
Bin Meng [Mon, 2 Feb 2015 13:25:09 +0000 (21:25 +0800)]
x86: Use tab instead of space to indent in PCIE_ECAM_BASE

Space is used before 'default' in PCIE_ECAM_BASE in arch/x86/Kconfig
so it looks misaligned. Replace the space with tab to indent.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agobootstage: Fix typos in the comment
Bin Meng [Mon, 2 Feb 2015 13:25:08 +0000 (21:25 +0800)]
bootstage: Fix typos in the comment

There are two typos in the comment block in bootstage.h, fix them.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: Add Intel Galileo instructions in README.x86
Bin Meng [Wed, 4 Feb 2015 08:26:14 +0000 (16:26 +0800)]
x86: Add Intel Galileo instructions in README.x86

Add some instructions about building U-Boot for Intel Galileo board.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: Add SD/MMC support to quark/galileo
Bin Meng [Wed, 4 Feb 2015 08:26:13 +0000 (16:26 +0800)]
x86: Add SD/MMC support to quark/galileo

Intel Galileo board has a microSD slot which is routed from Quark SoC
SDIO controller. Enable SD/MMC support so that we can use an SD card.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: Add SPI support to quark/galileo
Bin Meng [Wed, 4 Feb 2015 08:26:12 +0000 (16:26 +0800)]
x86: Add SPI support to quark/galileo

The Quark SoC contains a legacy SPI controller in the legacy bridge
which is ICH7 compatible. Like Tunnel Creek and BayTrail, the BIOS
control register offset in the ICH SPI driver is wrong for the Quark
SoC too, unprotect_spi_flash() is added to enable the flash write.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: pci: Add pci ids for Quark SoC
Bin Meng [Wed, 4 Feb 2015 08:26:11 +0000 (16:26 +0800)]
x86: pci: Add pci ids for Quark SoC

Add pci ids for Intel Quark SoC.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: galileo: Add GPIO support
Bin Meng [Wed, 4 Feb 2015 08:26:10 +0000 (16:26 +0800)]
x86: galileo: Add GPIO support

Quark SoC has a legacy GPIO block in the legacy bridge (D0:F31),
which is just the same one found in other x86 chipset. Since we
programmed the GPIO register block base address, we should be
able to enable the GPIO support on Intel Galileo board.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: quark: Initialize non-standard BARs
Bin Meng [Wed, 4 Feb 2015 08:26:09 +0000 (16:26 +0800)]
x86: quark: Initialize non-standard BARs

Quark SoC has some non-standard BARs (excluding PCI standard BARs)
which need be initialized with suggested values. This includes GPIO,
WDT, RCBA, PCIe ECAM and some ACPI register block base addresses.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: quark: Call MRC in dram_init()
Bin Meng [Thu, 5 Feb 2015 15:42:28 +0000 (23:42 +0800)]
x86: quark: Call MRC in dram_init()

Now that we have added Quark MRC codes, call MRC in dram_init() so
that DRAM can be initialized on a Quark based board.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agodt-bindings: Add Intel Quark MRC bindings
Bin Meng [Thu, 5 Feb 2015 15:42:27 +0000 (23:42 +0800)]
dt-bindings: Add Intel Quark MRC bindings

Add standard dt-bindings macros to be used by Intel Quark MRC node.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agofdtdec: Add compatible id and string for Intel Quark MRC
Bin Meng [Thu, 5 Feb 2015 15:42:26 +0000 (23:42 +0800)]
fdtdec: Add compatible id and string for Intel Quark MRC

Add COMPAT_INTEL_QRK_MRC and "intel,quark-mrc" so that fdtdec can
decode Intel Quark MRC node.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: quark: Enable the Memory Reference Code build
Bin Meng [Thu, 5 Feb 2015 15:42:25 +0000 (23:42 +0800)]
x86: quark: Enable the Memory Reference Code build

Turn on the Memory Reference code build in the quark Makefile.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: quark: Add System Memory Controller support
Bin Meng [Thu, 5 Feb 2015 15:42:24 +0000 (23:42 +0800)]
x86: quark: Add System Memory Controller support

The codes are actually doing the memory initialization stuff.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: quark: Add utility codes needed for MRC
Bin Meng [Thu, 5 Feb 2015 15:42:23 +0000 (23:42 +0800)]
x86: quark: Add utility codes needed for MRC

Add various utility codes needed for Quark MRC.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
8 years agox86: quark: Add Memory Reference Code (MRC) main routines
Bin Meng [Thu, 5 Feb 2015 15:42:22 +0000 (23:42 +0800)]
x86: quark: Add Memory Reference Code (MRC) main routines

Add the main routines for Quark Memory Reference Code (MRC).

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>