oweals/u-boot.git
8 years agovideo: Remove legacy VESA and coreboot framebuffer drivers
Bin Meng [Sun, 9 Oct 2016 11:14:19 +0000 (04:14 -0700)]
video: Remove legacy VESA and coreboot framebuffer drivers

Now that all x86 boards have been converted to DM video, drop the
legacy drivers.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agox86: coreboot: Convert to use DM coreboot video driver
Bin Meng [Sun, 9 Oct 2016 11:14:18 +0000 (04:14 -0700)]
x86: coreboot: Convert to use DM coreboot video driver

This converts coreboot to use DM framebuffer driver.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agodm: video: Don't do anything in alloc_fb() when plat->size is zero
Bin Meng [Sun, 9 Oct 2016 11:14:17 +0000 (04:14 -0700)]
dm: video: Don't do anything in alloc_fb() when plat->size is zero

With DM VESA driver on x86 boards, plat->base/size/align are all
zeroes and starting address passed to alloc_fb() happens to be 1MB
aligned, so this routine does not trigger any issue. On QEMU with
U-Boot as coreboot payload, the starting address is within 1MB
range (eg: 0x7fb0000), thus causes failure in video_post_bind().

Actually if plat->size is zero, it makes no sense to do anything
in this routine. Add such check there.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agodm: video: Add driver for coreboot framebuffer device
Bin Meng [Sun, 9 Oct 2016 11:14:16 +0000 (04:14 -0700)]
dm: video: Add driver for coreboot framebuffer device

This adds a DM driver for coreboot framebuffer device.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agovbe: Make vbe_setup_video_priv() public
Bin Meng [Sun, 9 Oct 2016 11:14:15 +0000 (04:14 -0700)]
vbe: Make vbe_setup_video_priv() public

vbe_setup_video_priv() might be useful to other drivers.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agox86: doc: Correct qfw command example
Bin Meng [Sun, 9 Oct 2016 11:14:14 +0000 (04:14 -0700)]
x86: doc: Correct qfw command example

The kernel load address for zboot should be 0x1000000.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agox86: Convert to use DM VESA video driver
Bin Meng [Sun, 9 Oct 2016 11:14:13 +0000 (04:14 -0700)]
x86: Convert to use DM VESA video driver

At present only chromebook boards are converted to DM video. Other
x86 boards are still using the legacy cfb_console driver. This
switches to use DM version drivers.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agodm: video: Output verbose information in vbe_setup_video()
Bin Meng [Sun, 9 Oct 2016 11:14:12 +0000 (04:14 -0700)]
dm: video: Output verbose information in vbe_setup_video()

With DM conversion, information like "Video: 1024x768x16" is not
shown anymore. Now add these verbose output back.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agodm: video: Add driver for VESA-compatible device
Bin Meng [Sun, 9 Oct 2016 11:14:11 +0000 (04:14 -0700)]
dm: video: Add driver for VESA-compatible device

This adds a DM driver for VESA-compatible device.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agox86: doc: Document coreboot framebuffer driver issue on QEMU
Bin Meng [Sun, 9 Oct 2016 11:14:10 +0000 (04:14 -0700)]
x86: doc: Document coreboot framebuffer driver issue on QEMU

For some unknown reason, coreboot framebuffer driver never works on
QEMU since day 1. It seems the driver only works on real hardware.
Document this issue.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agox86: mrccache: Fix error handling in mrccache_get_region()
Simon Glass [Mon, 26 Sep 2016 03:33:40 +0000 (21:33 -0600)]
x86: mrccache: Fix error handling in mrccache_get_region()

This should return normal errors, not device-tree errors. Fix it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Drop unused init_helper functions
Simon Glass [Mon, 26 Sep 2016 03:33:38 +0000 (21:33 -0600)]
x86: Drop unused init_helper functions

Drop init_bd_struct_r() which is no-longer used. Also drop the declaration
for init_func_spi() since this is now handled by generic board init.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: ivybridge: Tidy up enable_clock_gating() for 64-bit
Simon Glass [Mon, 26 Sep 2016 03:33:35 +0000 (21:33 -0600)]
x86: ivybridge: Tidy up enable_clock_gating() for 64-bit

Fix the hex case and remove unused brackets. Use ~0U instead of ~0UL to
allow compilation on 64-bit machines.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: ivybridge: Fix PCH power setup
Simon Glass [Mon, 26 Sep 2016 03:33:34 +0000 (21:33 -0600)]
x86: ivybridge: Fix PCH power setup

At present pch_power_options() has the arguments to writel() around the
wrong way. Fix this and update it to compile on 64-bit machines.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Don't export interrupt handlers with x86_64
Simon Glass [Mon, 26 Sep 2016 03:33:31 +0000 (21:33 -0600)]
x86: Don't export interrupt handlers with x86_64

We don't have a way of adjusting these at present so it is best to refuse to
export these functions. This can be implemented later if the API is required.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: i2c: Fix cast of address to 32-bit value
Simon Glass [Mon, 26 Sep 2016 03:33:30 +0000 (21:33 -0600)]
x86: i2c: Fix cast of address to 32-bit value

This gives a build warning on 64-bit x86. Fix it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Heiko Schocher <hs@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Correct address casts in interrupt code
Simon Glass [Mon, 26 Sep 2016 03:33:25 +0000 (21:33 -0600)]
x86: Correct address casts in interrupt code

We should cast an address to unsigned long, not u32.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Correct address casts in cpu code
Simon Glass [Mon, 26 Sep 2016 03:33:24 +0000 (21:33 -0600)]
x86: Correct address casts in cpu code

We should cast an address to unsigned long, not u32.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Allow interrupts to be disabled in 64-bit mode
Simon Glass [Mon, 26 Sep 2016 03:33:23 +0000 (21:33 -0600)]
x86: Allow interrupts to be disabled in 64-bit mode

Update the code to support both 32-bit and 64-bit modes.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agousb: pci: Fix cast for 64-bit compilation
Simon Glass [Mon, 26 Sep 2016 03:33:21 +0000 (21:33 -0600)]
usb: pci: Fix cast for 64-bit compilation

Fix a cast that causes warnings on 64-bit machines.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agotpm: Tidy up use of size_t
Simon Glass [Mon, 26 Sep 2016 03:33:20 +0000 (21:33 -0600)]
tpm: Tidy up use of size_t

We should consistently use %z with size_t, and avoid passing a uint32_t as
a size_t value. Fix these issues to avoid warnings on 64-bit machines.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agortc: Use CONFIG_X86 instead of __I386__
Simon Glass [Mon, 26 Sep 2016 03:33:11 +0000 (21:33 -0600)]
rtc: Use CONFIG_X86 instead of __I386__

For 64-bit x86, __I386__ should perhaps not be defined. It is not clear from
the definition, but let's use CONFIG_X86 to be sure.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agoelf: Add the Elf64_Rela type
Simon Glass [Mon, 26 Sep 2016 03:33:09 +0000 (21:33 -0600)]
elf: Add the Elf64_Rela type

Add this so that we can support 64-bit relocation on x86.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agoboard_f: Drop the extra fdtdec_prepare_fdt()
Simon Glass [Mon, 26 Sep 2016 03:33:07 +0000 (21:33 -0600)]
board_f: Drop the extra fdtdec_prepare_fdt()

This is already called earlier, from fdtdec_setup(), so drop this unnecessary
call from the init sequence.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agobios_emulator: Fix cast for 64-bit compilation
Simon Glass [Mon, 26 Sep 2016 03:33:06 +0000 (21:33 -0600)]
bios_emulator: Fix cast for 64-bit compilation

Fix a cast that causes warnings on 64-bit machines.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agoAdd _image_binary_end section declaration
Simon Glass [Mon, 26 Sep 2016 03:33:05 +0000 (21:33 -0600)]
Add _image_binary_end section declaration

This is used in some link scripts, so add a declaration for it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agodm: x86: Move link to use driver model for video
Simon Glass [Thu, 6 Oct 2016 02:42:20 +0000 (20:42 -0600)]
dm: x86: Move link to use driver model for video

Update the configuration to use the new driver. Drop the existing plumbing
code and unused header files.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agodm: x86: Move samus to use new driver model support
Simon Glass [Thu, 6 Oct 2016 02:42:19 +0000 (20:42 -0600)]
dm: x86: Move samus to use new driver model support

Update the samus driver to avoid the direct call to the video BIOS setup.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Adjust config to support DM_VIDEO
Simon Glass [Thu, 6 Oct 2016 02:42:18 +0000 (20:42 -0600)]
x86: Adjust config to support DM_VIDEO

Update the common configuration so that it works correctly when
CONFIG_DM_VIDEO is enabled. This involves dropping the legacy CONFIG_VIDEO
option and changing the stdio device from "vga" to "vidconsole".

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agodm: x86: video: Add a driver-model driver for ivybridge graphics
Simon Glass [Thu, 6 Oct 2016 02:42:15 +0000 (20:42 -0600)]
dm: x86: video: Add a driver-model driver for ivybridge graphics

At present we use the legacy vesa driver for graphics. Add a driver which
supports driver model. This can be probed only when needed, removing the
need to start up the display if it is not used.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agodm: video: Add driver-model support to vesa graphics
Simon Glass [Thu, 6 Oct 2016 02:42:17 +0000 (20:42 -0600)]
dm: video: Add driver-model support to vesa graphics

Provide a function to run the Vesa BIOS for a given PCI device and obtain
the resulting configuration (e.g. display size) for use by the video
uclass. This makes it easier to write a video driver that uses vesa and
supports driver model.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agodm: stdio: Allow lazy probing of video devices
Simon Glass [Thu, 6 Oct 2016 02:42:16 +0000 (20:42 -0600)]
dm: stdio: Allow lazy probing of video devices

At present all video devices are probed on start-up. It would be better to
probe a device only when it is needed. This can happen if it is referenced
in the stdout environment variable, for example.

Add support for this by searching for a suitable device when needed, probing
it, and finding the stdio device it creates.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: video: Fix typo in broadwell Kconfig
Simon Glass [Thu, 6 Oct 2016 02:42:14 +0000 (20:42 -0600)]
x86: video: Fix typo in broadwell Kconfig

'enabled' should be 'enables'. Fix it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agodm: core: Add a function to get a uclass name
Simon Glass [Thu, 6 Oct 2016 02:42:13 +0000 (20:42 -0600)]
dm: core: Add a function to get a uclass name

It is useful in debug() statements to display the name of the uclass for a
device. Add a simple function to provide this.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agolist: Add list_last_entry() to find the last entry
Simon Glass [Thu, 6 Oct 2016 02:42:12 +0000 (20:42 -0600)]
list: Add list_last_entry() to find the last entry

We have list_first_entry() but in some cases it is useful to find the last
item added to the list. Add a macro for this.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agoFix return value in trailing_strtoln()
Simon Glass [Thu, 6 Oct 2016 02:42:11 +0000 (20:42 -0600)]
Fix return value in trailing_strtoln()

This function should return -1 if there is no trailing integer in the
string. Instead it returns 0. Fix it by checking for this condition at the
start.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Add an accelerated memmove() function
Simon Glass [Thu, 6 Oct 2016 02:42:10 +0000 (20:42 -0600)]
x86: Add an accelerated memmove() function

Bring in a faster memmove() from Linux 4.7. This speeds up scrolling on the
display.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
8 years agoRevert "x86: broadwell: gpio: Remove the codes to set up pin control"
Simon Glass [Thu, 6 Oct 2016 02:42:09 +0000 (20:42 -0600)]
Revert "x86: broadwell: gpio: Remove the codes to set up pin control"

This makes the assumption that setting up pinctrl in cpu_init_r() is safe.
On samus we need GPIOs before relocation in order to support power control.
This commit fixes the following message on boot:

   initcall sequence ffe5c6f4 failed at call ffe01d3d (err=-1)
   ### ERROR ### Please RESET the board ###

In any case it seems better to leave init to driver model, so that it can
pick up the GPIO driver when it needs it. Since pinctrl is a dependency of
the GPIO driver, we may as well put the dependency there and avoid these
problems.

This reverts commit 9769e05bcf79939bad23a719982dd1f85a110f3c.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: Fix Linux v4.7+ zimage booting (update bootparam.h)
Stefan Roese [Fri, 30 Sep 2016 07:15:59 +0000 (09:15 +0200)]
x86: Fix Linux v4.7+ zimage booting (update bootparam.h)

Booting Linux kernel v4.7+ does not work since Linux kernel commit 974f221c
"x86/boot: Move compressed kernel to the end of the decompression buffer".

This patch adds the latest version of the setup_header struct, adding
"init_size" which is needed since this commit referenced above. With this
patch, booting Linux v4.8-rc8 does work again on x86 boards.

Signed-off-by: Stefan Roese <sr@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: baytrail: Add 2nd eMMC controller to the PCI probe list
Stefan Roese [Mon, 26 Sep 2016 08:18:07 +0000 (10:18 +0200)]
x86: baytrail: Add 2nd eMMC controller to the PCI probe list

With this addition, the eMMC device available on the congatec and DFI
BayTrail SoM is detected correctly.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
8 years agox86: conga-qeval20-qa3: Add README to explain the console UART options
Stefan Roese [Fri, 9 Sep 2016 05:45:49 +0000 (07:45 +0200)]
x86: conga-qeval20-qa3: Add README to explain the console UART options

This patch adds a small README to explain the 2 defconfig files and its
usage for the different console UART options.

Signed-off-by: Stefan Roese <sr@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Cc: Simon Glass <sjg@chromium.org>
8 years agoMerge branch 'master' of git://www.denx.de/git/u-boot-imx
Tom Rini [Sat, 8 Oct 2016 02:02:10 +0000 (22:02 -0400)]
Merge branch 'master' of git://denx.de/git/u-boot-imx

8 years agoARM: AM437X: Add Silicon ID support
Lokesh Vutla [Tue, 4 Oct 2016 04:04:50 +0000 (09:34 +0530)]
ARM: AM437X: Add Silicon ID support

Add silicon ID code for AM437x silicon. This can be used to print
the cpu info using CONFIG_DISPLAY_CPUINFO.
Also printing "CPU :" along with cpu name in order to be consistent
with other OMAP platforms.

Reviewed-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
8 years agospl: saveenv: adding saveenv support in SPL
B, Ravi [Wed, 28 Sep 2016 09:16:18 +0000 (14:46 +0530)]
spl: saveenv: adding saveenv support in SPL

By default saveenv option is not supported for SPL. This patch
enable the support for save environment variable for SPL build.

Enable save environment support in SPL after setenv. By default
the saveenv option is not provided in SPL, but some boards need
this support in 'Falcon' boot, where SPL need to boot from
different images based on environment variable set by OS. For
example OS may set "reboot_image" environment variable to
"recovery" inorder to boot recovery image by SPL. The SPL read
"reboot_image" and act accordingly and change the reboot_image
to default mode using setenv and save the environemnt.

Signed-off-by: Ravi Babu <ravibabu@ti.com>
Reviewed-by: Simon Glass <sig@chromium.org>
change in v1:
- dropped SUPPORT, use CONFIG_SPL_SAVEENV
- updates the comments in mmc_private.h

8 years agocros_ec: Honor the google,remote-bus dt property
Moritz Fischer [Tue, 27 Sep 2016 22:42:07 +0000 (15:42 -0700)]
cros_ec: Honor the google,remote-bus dt property

Boards where ECs that use a I2C port != 0 specify this in the
devicetree file via the google,remote-bus property.
Previously this was ignored and hardcoded to port 0.

Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Heiko Schocher <hs@denx.de>
Cc: u-boot@lists.denx.de
Acked-by: Simon Glass <sjg@chromium.org>
8 years agoboard: ti: dra7xx: complex definitions should be protected with parentheses
Mugunthan V N [Tue, 27 Sep 2016 07:31:42 +0000 (13:01 +0530)]
board: ti: dra7xx: complex definitions should be protected with parentheses

As a standard practice complex definitions should be protected
with parentheses, as it might fail when used in a complex if
statements.

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoARM: dts: dra72: add rev C evm support
Mugunthan V N [Tue, 27 Sep 2016 07:31:41 +0000 (13:01 +0530)]
ARM: dts: dra72: add rev C evm support

Add DTS support for dra72 evm Rev C which has the following
changes
* Two ethernet ports now instead of the single one in rev B.
* DP83867 ethernet phy instead of DP838865.

Cc: Vignesh R <vigneshr@ti.com>
Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoarm: Add return value argument to longjmp
Alexander Graf [Tue, 27 Sep 2016 07:30:32 +0000 (09:30 +0200)]
arm: Add return value argument to longjmp

The normal longjmp command allows for a caller to pass the return value
of the setjmp() invocation. This patch adds that semantic to the arm
implementation of it and adjusts the efi_loader call respectively.

Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoenv: tool: add command line option to input lockfile path
B, Ravi [Mon, 26 Sep 2016 12:54:08 +0000 (18:24 +0530)]
env: tool: add command line option to input lockfile path

The default lockname is set to /var/lock. This limits the
usage of this application where OS uses different lockfile
location parameter.
For example, In case of android, the default lock
path location is /data.
Hence by providing the command line option to input lockfile
path will be useful to reuse the tool across multiple
operating system.

usage: ./fw_printenv -l <lockfile path>

Signed-off-by: Ravi Babu <ravibabu@ti.com>
8 years agodra7x: dfu: qspi: increase the qspi spl partition to 256K
B, Ravi [Mon, 26 Sep 2016 12:51:13 +0000 (18:21 +0530)]
dra7x: dfu: qspi: increase the qspi spl partition to 256K

The SPL size for dra7x platform increased beyond 64K,
increasing the size to 256K to cater for future enhancement.

Signed-off-by: Ravi Babu <ravibabu@ti.com>
8 years agodra7xx: config: cleanup: moved to kconfig for CONFIG_SPL_ENV_SUPPORT
B, Ravi [Mon, 26 Sep 2016 12:50:33 +0000 (18:20 +0530)]
dra7xx: config: cleanup: moved to kconfig for CONFIG_SPL_ENV_SUPPORT

removing CONFIG_SPL_ENV_SUPPORT defined in header files
due to moved to kconfig option for CONFIG_SPL_ENV_SUPPORT

Signed-off-by: Ravi Babu <ravibabu@ti.com>
8 years agoRevert "ns16650: Make sure we have CONFIG_CLK set before using infrastructure"
Masahiro Yamada [Mon, 26 Sep 2016 11:45:27 +0000 (20:45 +0900)]
Revert "ns16650: Make sure we have CONFIG_CLK set before using infrastructure"

This reverts commit 82f5279b0cd99a9163d34cfe926d0316d9dc0d37.

The build failure of k2*evm boards was fixed in a different way by
the previous commit.  It is nasty to patch generic drivers around
with #ifdef CONFIG_CLK just for the KeyStone's matter.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoARM: keystone: rename clk_get_rate() to ks_clk_get_rate()
Masahiro Yamada [Mon, 26 Sep 2016 11:45:26 +0000 (20:45 +0900)]
ARM: keystone: rename clk_get_rate() to ks_clk_get_rate()

The KeyStone platform has its own clk_get_rate() but its prototype
is different from that of the common-clk (clk-uclass) framework.

Prefix the KeyStone specific implementation with ks_ in order to
avoid name-space conflict.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoboard: ge: bx50v3: Pass video bootargs for b850v3
Ken Lin [Fri, 7 Oct 2016 14:26:56 +0000 (10:26 -0400)]
board: ge: bx50v3: Pass video bootargs for b850v3

Due to clock source restrictions on i.MX6, certain pixel clock rates can
not be supported. Hence default the resolution/frame rate during boot to a
supported value by passing video bootargs 1024x768@60 for
HDMI (Display Port1) and LVDS (Display Port2) on B850v3.

Signed-off-by: Ken Lin <ken.lin@advantech.com.tw>
Signed-off-by: Akshay Bhat <akshay.bhat@timesys.com>
8 years agoARM: keystone: remove declaration of unused functions
Masahiro Yamada [Mon, 26 Sep 2016 11:45:25 +0000 (20:45 +0900)]
ARM: keystone: remove declaration of unused functions

These two functions are neither defined nor referenced.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agokbuild: generate u-boot.cfg as a byproduct of include/autoconf.mk
Masahiro Yamada [Mon, 26 Sep 2016 04:05:01 +0000 (13:05 +0900)]
kbuild: generate u-boot.cfg as a byproduct of include/autoconf.mk

Our build system still parses ad-hoc CONFIG options in header files
and generates include/autoconf.mk so that Makefiles can reference
them.  This gimmick was introduced in the pre-Kconfig days and will
be kept until Kconfig migration is completed.

The include/autoconf.mk is generated like follows:

  [1] Preprocess include/common.h with -DDO_DEPS_ONLY and
      retrieve macros into include/autoconf.mk.tmp
  [2] Reformat include/autoconf.mk.dep into include/autoconf.mk
      with tools/scripts/define2mk.sed script
  [3] Remove include/autoconf.mk.tmp

Here, include/autoconf.mk.tmp is similar to u-boot.cfg, which is
also generated by preprocessing include/config.h with -DDO_DEPS_ONLY.
In other words, there is much overlap among include/autoconf.mk and
u-boot.cfg build rules.

So, the idea is to split the build rule of include/autoconf.mk
into two stages.  The first preprocesses headers into u-boot.cfg.
The second parses the u-boot.cfg into include/autoconf.mk.  The
build rules of u-boot.cfg in Makefile and spl/Makefile will be gone.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agokbuild: make dependencies in scripts/Makefile.autoconf more readable
Masahiro Yamada [Mon, 26 Sep 2016 04:05:00 +0000 (13:05 +0900)]
kbuild: make dependencies in scripts/Makefile.autoconf more readable

I do not remember why I wrote the code like this, but let's make it
a bit more readable.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agokbuild: move no_new_adhoc_configs_check to "all" target command
Masahiro Yamada [Mon, 26 Sep 2016 04:04:59 +0000 (13:04 +0900)]
kbuild: move no_new_adhoc_configs_check to "all" target command

I am going to move the build rule of u-boot.cfg.  Before that,
no_new_adhoc_configs_check must be tweaked to not depend on it.

The ad-hoc option check can be done at the end of build, along
with other checks.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agocheck-config: fix wrong comment about how to build whitelist
Masahiro Yamada [Mon, 26 Sep 2016 04:04:58 +0000 (13:04 +0900)]
check-config: fix wrong comment about how to build whitelist

The command suggested in this comment block is wrong; it would not
rip off CONFIG options that had already been converted to Kconfig.

Instead, we should use the scripts/build-whitelist.sh tool.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoconfig_whitelist: remove bogus options
Masahiro Yamada [Mon, 26 Sep 2016 02:52:30 +0000 (11:52 +0900)]
config_whitelist: remove bogus options

These are not CONFIG options (detected by my eyes).

CONFIG_SPL_BUILD and CONFIG_TPL_BUILD are build options defined only
for building SPL and TPL, respectively.

The others are just mentioned in comment blocks.

Now, scripts/build-whitelist.sh never picks up new options.  Once
we kill these false ones, they will never revive.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoconfig_whitelist: sync by tool
Masahiro Yamada [Mon, 26 Sep 2016 02:52:29 +0000 (11:52 +0900)]
config_whitelist: sync by tool

It is a good practice to drop an option from the whitelist when we
convert it to Kconfig, but we may sometimes forget to do that.

So, it might be a good idea to sync the whitelist from time to time.

This commit was generated by:
  scripts/build-whitelist.sh

Looks like we had a bit progress...

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agobuild-whitelist: do not add new options to whitelist when update
Masahiro Yamada [Mon, 26 Sep 2016 02:52:28 +0000 (11:52 +0900)]
build-whitelist: do not add new options to whitelist when update

If somebody adds references to new CONFIG options in source files,
they will be added in the whitelist when we sync it.  (For example,
if we run scripts/build-whitelist.sh against commit 42f75050667b,
new options CONFIG_SPL_DFU_SUPPORT and CONFIG_USB_XHCI_UNIPHIER will
appear in the list.)

In order to make steady progress of Kconfig migration, we want to
only decrease whitelist options, but never increase.

So, when we update the whitelist, we should create a temporary list,
then take the intersection of the temporary one and the current one.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agovexpress: disable cci ace slave ports when booting in non-sec/hyp mode
Sudeep Holla [Fri, 23 Sep 2016 16:38:39 +0000 (17:38 +0100)]
vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting
to nonsec booting") added support to check if the firmware on TC2  is
configured appropriately before booting in nonsec/hyp mode.

However when booting in non-secure/hyp mode, CCI control must be done in
secure firmware and can't  be done in non-secure/hyp mode. In order to
ensure that, this patch disables the cci slave port inteface so that it
is not accessed at all.

Cc: Jon Medhurst <tixy@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Acked-by: Jon Medhurst <tixy@linaro.org>
Tested-by: Jon Medhurst <tixy@linaro.org>
8 years agocommon/console.c: ensure GD_FLG_SILENT is set or cleared
Chris Packham [Fri, 23 Sep 2016 03:59:43 +0000 (15:59 +1200)]
common/console.c: ensure GD_FLG_SILENT is set or cleared

When CONFIG_SILENT_CONSOLE is defined and the default environment has
silent=1 it is not possible for a user to make the console un-silent if
the environment is not available when console_init_f() is called (for
example because the environment is in SPI).

Add a new helper function console_update_silent() and call it from both
console_init_f() and console_init_r().

Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoarm: dra7xx: Move fastboot options to defconfig
Semen Protsenko [Thu, 22 Sep 2016 18:52:07 +0000 (21:52 +0300)]
arm: dra7xx: Move fastboot options to defconfig

Now that fastboot options are available in Kconfig, we can migrate them
from DRA7 header to corresponding DRA7 defconfigs.

Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
8 years agoARM: vf610: use strcpy for soc environment variable
Stefan Agner [Wed, 28 Sep 2016 09:29:28 +0000 (11:29 +0200)]
ARM: vf610: use strcpy for soc environment variable

To create the soc environment variable we concatenate two strings
on the stack. So far, strcat has been used for the first string as
well as for the second string. Since the variable on the stack is
not initialized, the first strcat may not start using the first
entry in the character array. This then could lead to an buffer
overflow on the stack.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
8 years agoconfigs: enable device tree for Colibri iMX7
Stefan Agner [Wed, 5 Oct 2016 22:27:12 +0000 (15:27 -0700)]
configs: enable device tree for Colibri iMX7

Enable device tree configuration and specify default device tree
for Toradex Colibri iMX7.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
8 years agocolibri_imx7: use Ricoh RN5T567 to reboot the board
Stefan Agner [Wed, 5 Oct 2016 22:27:11 +0000 (15:27 -0700)]
colibri_imx7: use Ricoh RN5T567 to reboot the board

Use the external PMIC Ricoh RN5T567 to reliably restart the system.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
8 years agoarm: dts: imx7: add Ricoh RN5T567 PMIC node
Stefan Agner [Wed, 5 Oct 2016 22:27:10 +0000 (15:27 -0700)]
arm: dts: imx7: add Ricoh RN5T567 PMIC node

Add device tree node for Ricoh RN5T567. Currently we do not need
the individual DC/DC converters or LDO's (and they are also not
yet supported by the driver).

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
8 years agopower: pmic: add Ricoh RN5T567 PMIC support
Stefan Agner [Wed, 5 Oct 2016 22:27:09 +0000 (15:27 -0700)]
power: pmic: add Ricoh RN5T567 PMIC support

Add device model enabled PMIC driver for Ricoh RN5T567 PMIC used
on Colibri iMX7.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agocolibri_imx7: remove legancy UART platform data
Stefan Agner [Wed, 5 Oct 2016 22:27:08 +0000 (15:27 -0700)]
colibri_imx7: remove legancy UART platform data

We now use device tree to provide SoC data to the UART driver, there
is no need for the legancy UART platform data.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
8 years agocolibri_imx7: remove legancy I2C support
Stefan Agner [Wed, 5 Oct 2016 22:27:07 +0000 (15:27 -0700)]
colibri_imx7: remove legancy I2C support

Remove legancy I2C config and code in favor of upcomming DM/DT
enable I2C support.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
8 years agoarm: dts: imx7: add basic i.MX 7/Colibri iMX7 device tree
Stefan Agner [Wed, 5 Oct 2016 22:27:06 +0000 (15:27 -0700)]
arm: dts: imx7: add basic i.MX 7/Colibri iMX7 device tree

Add base device for NXP i.MX 7Solo/7Dual. The two SoC are very
similar and hence can share the same device tree for boot loaders
purpose.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoarm: dts: imx7: add pinctrl defines
Stefan Agner [Wed, 5 Oct 2016 22:27:05 +0000 (15:27 -0700)]
arm: dts: imx7: add pinctrl defines

Add pinctrl defines for NXP i.MX 7Solo/7Dual SoC. The pinctrl format
is compatible to the Linux kernel, hence this file is a simple copy
from the Linux kernel (commit 97f5c1817b7e).

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
8 years agopinctrl: imx: do not announce driver initialization
Stefan Agner [Wed, 5 Oct 2016 22:27:04 +0000 (15:27 -0700)]
pinctrl: imx: do not announce driver initialization

It is not usual that drivers announce when they have been initialized.
use dev_dbg to announce device initialization.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agodm: imx: serial: support device tree
Stefan Agner [Wed, 5 Oct 2016 22:27:03 +0000 (15:27 -0700)]
dm: imx: serial: support device tree

Support instatiation through device tree. Also parse the fsl,dte-mode
property to determine whether DTE mode shall be used.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoscripts: Add script to extract default environment
Lukasz Majewski [Sat, 17 Sep 2016 04:57:39 +0000 (06:57 +0200)]
scripts: Add script to extract default environment

This script looks for env_common.o object file and extracts from it default
u-boot environment, which is afterwards printed on standard output.

Usage example:
get_default_envs.sh > u-boot-env-default.txt

The generated text file can be used as input for mkenvimage.

Signed-off-by: Lukasz Majewski <l.majewski@majess.pl>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoARM: Add register defines for am33xx ePWM registers
tomas.melin@vaisala.com [Fri, 16 Sep 2016 10:21:39 +0000 (10:21 +0000)]
ARM: Add register defines for am33xx ePWM registers

Register definitions needed for configuring the
ePWM module.

Signed-off-by: Tomas Melin <tomas.melin@vaisala.com>
8 years agodoc: typo fix addess -> address
Jelle van der Waa [Wed, 14 Sep 2016 19:54:53 +0000 (21:54 +0200)]
doc: typo fix addess -> address

Signed-off-by: Jelle van der Waa <jelle@vdwaa.nl>
8 years agoARM: OMAP5+: Override switch_to_hypervisor function
Keerthy [Wed, 14 Sep 2016 05:13:33 +0000 (10:43 +0530)]
ARM: OMAP5+: Override switch_to_hypervisor function

Override the switch_to_hypervisor function to switch cpu to hypervisor
mode using the available ROM code hook early in the boot phase before
the boot loader checks for HYP mode.

Based on the work done by Jonathan Bergsagel jbergsagel@ti.com.

Cc: beagleboard-x15@googlegroups.com
Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoARM: Introduce function to switch to hypervisor mode
Keerthy [Wed, 14 Sep 2016 05:13:32 +0000 (10:43 +0530)]
ARM: Introduce function to switch to hypervisor mode

On some of the SoCs one cannot enable hypervisor mode directly from the
u-boot because the ROM code puts the chip to supervisor mode after it
jumps to boot loader. Hence introduce a weak function which can be
overridden based on the SoC type and switch to hypervisor mode in a
custom way.

Cc: beagleboard-x15@googlegroups.com
Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoconfigs: dra7xx_evm_defconfig: Enable LPAE mode
Keerthy [Wed, 14 Sep 2016 05:13:31 +0000 (10:43 +0530)]
configs: dra7xx_evm_defconfig: Enable LPAE mode

Enable Linear Physical Address Extension mode which is a
prerequisite for hypervisor mode.

Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoconfigs: am57xx_evm_defconfig: Enable LPAE mode
Keerthy [Wed, 14 Sep 2016 05:13:30 +0000 (10:43 +0530)]
configs: am57xx_evm_defconfig: Enable LPAE mode

Enable Linear Physical Address Extension mode which is a
prerequisite for hypervisor mode.

Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoomap: Set appropriate cache configuration for LPAE and non-LAPE cases
Keerthy [Wed, 14 Sep 2016 05:13:29 +0000 (10:43 +0530)]
omap: Set appropriate cache configuration for LPAE and non-LAPE cases

Cache configuration methods is different for LPAE and non-LPAE cases.
Hence the bits and the interpretaion is different for two cases.
In case of non-LPAE mode short descriptor format is used and we need
to set Cache and Buffer bits.

In the case of LPAE the cache configuration happens via MAIR0 lookup.

Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoomap: Remove hardcoding of mmu section shift to 20
Keerthy [Wed, 14 Sep 2016 05:13:28 +0000 (10:43 +0530)]
omap: Remove hardcoding of mmu section shift to 20

As of now the mmu section shift is hardcoded to 20 but with LPAE
coming into picture this can be different. Hence replacing 20 with
MMU_SECTION_SHIFT macro.

Signed-off-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
8 years agoPWM: Correct misspellings of "module" in context of PWM
Robert P. J. Day [Tue, 13 Sep 2016 12:35:18 +0000 (08:35 -0400)]
PWM: Correct misspellings of "module" in context of PWM

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Acked-by: Heiko Schocher <hs@denx.de>
8 years agoARM64: Add support for some of atomic64 operations
Adam Oleksy [Tue, 13 Sep 2016 06:40:58 +0000 (08:40 +0200)]
ARM64: Add support for some of atomic64 operations

These functions are needed in UBI/UBIFS on ZynqMP platform (ARM64).

Signed-off-by: Adam Oleksy <adam.oleksy@nokia.com>
Cc: Albert Aribaud <albert.u.boot@aribaud.net>
8 years agocmd/onenand.c: block align warning
Ladislav Michl [Tue, 13 Sep 2016 05:40:00 +0000 (07:40 +0200)]
cmd/onenand.c: block align warning

An attempt to write non block aligned data fails silently, add warning and
set result.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
8 years agosearch.h: Numerous grammatical fixes, comment updates
Robert P. J. Day [Fri, 9 Sep 2016 10:22:10 +0000 (06:22 -0400)]
search.h: Numerous grammatical fixes, comment updates

Tweaks (no functional changes) to include/search.h, including:

 * use standard multiple inclusion check
 * fix spelling mistakes
 * have comments match actual names in function prototypes
 * remove obsolete reference to "do_apply"
 * replace "hashing table" with "hash table"

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
8 years agoVarious, accumulated typos collected from around the tree.
Robert P. J. Day [Wed, 7 Sep 2016 18:27:59 +0000 (14:27 -0400)]
Various, accumulated typos collected from around the tree.

Fix various misspellings of:

 * deprecated
 * partition
 * preceding,preceded
 * preparation
 * its versus it's
 * export
 * existing
 * scenario
 * redundant
 * remaining
 * value
 * architecture

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Reviewed-by: Jagan Teki <jteki@openedev.com>
Reviewed-by: Stefan Roese <sr@denx.de>
8 years agoARMv8/sec-firmware: fix a compile error
Hou Zhiqiang [Tue, 6 Sep 2016 06:23:07 +0000 (14:23 +0800)]
ARMv8/sec-firmware: fix a compile error

When enabled sec firmware framework, but lack of definition of
the marco SEC_FIRMWARE_FIT_IMAGE, SEC_FIRMEWARE_FIT_CNF_NAME
and SEC_FIRMWARE_TARGET_EL, there will be some build errors,
so give a default definition.

Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
8 years agoARM: Respect CONFIG_SPL_STACK define in lowlevel_init.S
Siarhei Siamashka [Mon, 5 Sep 2016 03:36:10 +0000 (06:36 +0300)]
ARM: Respect CONFIG_SPL_STACK define in lowlevel_init.S

The SPL and U-Boot proper may use different initial stack
locations, which are configured via CONFIG_SPL_STACK and
CONFIG_SYS_INIT_SP_ADDR defines. The lowlevel_init.S
code needs to handle this in the same way as crt0.S

Without this fix, setting the U-Boot stack location to some
place, which is not safely accessible by the SPL (such as
the DRAM), causes a very early SPL deadlock.

Signed-off-by: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
8 years agoSuspected Spam: Do not open attachements![PATCH 4/6] tools/env: flash_write_buf:...
Andreas Fenkart [Mon, 29 Aug 2016 21:17:00 +0000 (23:17 +0200)]
Suspected Spam: Do not open attachements![PATCH 4/6] tools/env: flash_write_buf: enforce offset to be start of environment

This allows to take advantage of the environment being block aligned.
This is not a new constraint. Writes always start at the begin of the
environment, since the header with CRC/length as there.
Every environment modification requires updating the header

Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
8 years agotools/env: lookup dev_type directly from flash_read_buf/flash_write_buf
Andreas Fenkart [Mon, 29 Aug 2016 21:16:59 +0000 (23:16 +0200)]
tools/env: lookup dev_type directly from flash_read_buf/flash_write_buf

flash_write_buf already looks up size/offset/#sector from struct
envdev_s. It can look up mtd_type as well. Same applies to
flash_read_buf. Makes the interface simpler

Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
8 years agotools/env: pass bad block offset by value
Andreas Fenkart [Mon, 29 Aug 2016 21:16:58 +0000 (23:16 +0200)]
tools/env: pass bad block offset by value

the offset is not modified by linux ioctl call
see mtd_ioctl{drivers/mtd/mtdchar.c}
Makes the interface less ambiguous, since the caller can
now exclude a modification of blockstart

Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
8 years agotools/env: factor out environment_end function
Andreas Fenkart [Mon, 29 Aug 2016 21:16:57 +0000 (23:16 +0200)]
tools/env: factor out environment_end function

instead of adhoc computation of the environment end,
use a function with a proper name

Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
8 years agogunzip: cache-align write buffer memory
Clemens Gruber [Mon, 29 Aug 2016 15:10:36 +0000 (17:10 +0200)]
gunzip: cache-align write buffer memory

When using gzwrite to eMMC on an i.MX6Q board, the following warning
occurs repeatedly:
CACHE: Misaligned operation at range [4fd633184fe63318]

This patch cache-aligns the memory allocation for the gzwrite writebuf,
therefore avoiding the misaligned dcache flush and the warning from
check_cache_range.

Signed-off-by: Clemens Gruber <clemens.gruber@pqgruber.com>
Reviewed-by: Eric Nelson <eric@nelint.com>
Reviewed-by: Stefan Agner <stefan.agner@toradex.com>
8 years agoREADME: Fix CONFIG_SYS_NAND_MAX_DEVICE typo
Simon Glass [Mon, 3 Oct 2016 00:01:11 +0000 (18:01 -0600)]
README: Fix CONFIG_SYS_NAND_MAX_DEVICE typo

This should be CONFIG_SYS_MAX_NAND_DEVICE. Fix it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Scott Wood <oss@buserror.net>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: York Sun <york.sun@nxp.com>
8 years agoREADME: Drop CONFIG_MPC8349ADS
Simon Glass [Mon, 3 Oct 2016 00:01:10 +0000 (18:01 -0600)]
README: Drop CONFIG_MPC8349ADS

This option is not used now.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: York Sun <york.sun@nxp.com>
8 years agoREADME: Drop README.imx31
Simon Glass [Mon, 3 Oct 2016 00:01:09 +0000 (18:01 -0600)]
README: Drop README.imx31

The only content of this file is CONFIG options which are no-longer present
in U-Boot. Drop it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: York Sun <york.sun@nxp.com>