Andreas Färber [Wed, 9 Oct 2019 14:03:55 +0000 (16:03 +0200)]
configs: Add Khadas VIM3 defconfig
Derived from odroid-n2_defconfig and README.odroid-n2.
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: Andreas Färber <afaerber@suse.de> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Neil Armstrong [Fri, 30 Aug 2019 12:09:24 +0000 (14:09 +0200)]
video: meson: sync with linux drm-misc tree
Synchronize the Amlogic Meson Video driver back with the latest
DRM misc tree, adding G12A platform support, from the latest commit: 528a25d040bc ("drm: meson: use match data to detect vpu compatibility")
The sync includes the following changes from Linux adapted to U-Boot:
- Add support for VIC alternate timings
- Switch PLL to 5.94GHz base for 297Mhz pixel clock
- Add registers for G12A SoC
- Add G12A Support for VPP setup
- Add G12A Support for VIU setup
- Add G12A support for OSD1 Plane
- Add G12A support for plane handling in CRTC driver
- Add G12A support for CVBS Encoder
- Add G12A Video Clock setup
- Add G12A support for the DW-HDMI Glue
- fix G12A HDMI PLL settings for 4K60 1000/1001 variations
- fix primary plane disabling
- fix G12A primary plane disabling
- mask value when writing bits relaxed
- crtc: drv: vpp: viu: venc: use proper macros instead of magic constants
- global clean-up
- add macro used to enable HDMI PLL
- venc: set the correct macrovision max amplitude value
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Reviewed-by: Anatolij Gustschin <agust@denx.de>
Daniel Drake [Tue, 6 Aug 2019 08:10:42 +0000 (16:10 +0800)]
board: amlogic: document alternative libretech-cc installation methods
As already documented in this README, several binaries must be
glued together in order to boot the device.
Extend the documentation to cover the prebuilt binaries
(saving you the hassle of installing ancient cross-compilers),
and also mention the open source replacements for the encryption
tool (which is especially useful if you want to avoid requiring
32-bit x86 binaries in your build system).
Signed-off-by: Daniel Drake <drake@endlessm.com> Cc: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Tom Rini [Thu, 17 Oct 2019 11:26:16 +0000 (07:26 -0400)]
Merge tag 'for-v2020.01' of https://gitlab.denx.de/u-boot/custodians/u-boot-ubi
ubi enhancements for 2020.01
- provide a way for skipping crc checks ported from linux, and add an
U-Boot command to set this flag on already installed systems.
- fix redundand environment management
Tom Rini [Wed, 16 Oct 2019 22:10:31 +0000 (18:10 -0400)]
Merge tag 'dm-pull-15oct19' of https://gitlab.denx.de/u-boot/custodians/u-boot-dm
binman enhancements:
- Dropping some test Elf files and building them from source instead
- Refactoring of x86 16-bit entries
- Support for SPL symbols within sections
- Handle the 'notes' sections and hidden symbols in recent binutils
- Improved error reporting with a tool fails
libfdt and documentation fixes
vboot required-key test
driver model power-domain controls
patman Message-Id enhancement
Tom Rini [Wed, 16 Oct 2019 22:10:01 +0000 (18:10 -0400)]
Merge branch '2019-10-15-master-imports'
- Split travis jobs again so we don't have failures due to excessive
build times.
- Android image updates
- cscope bugfix, update ARM maintainers entry, drop H2200 platform.
Tom Rini [Mon, 14 Oct 2019 22:12:59 +0000 (18:12 -0400)]
travis: Split and rename xilinx ARM in to two jobs
Split the ARMv7 and AArch64 platforms into separate jobs, to avoid them
taking too long to build overall. Also rename them from "Xilinx" to
"Zynq*" to reflect slightly better what is being built and to pull in a
few more board matches.
Reviewed-by: Michal Simek <michal.simek@xilinx.com> Signed-off-by: Tom Rini <trini@konsulko.com>
`find -L` tests properties, but name, and cscope can't cope with symbolic
links (a lot of bugs in upstream were simple closed as kinda invalid).
To work around the problem, exclude symbolic links from the cscope.files.
Note, it's done in two pass to speed up the process (`-exec realpath ...`
approach is not portable and introduces a 3x delay).
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
U-Boot cannot be built for h2200_defconfig with CONFIG_DM=y.
The maintainer Lukasz Dalek suggested to remove the board.
https://lists.denx.de/pipermail/u-boot/2019-August/380685.html
Cc: Lukasz Dalek <luk0104@gmail.com> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
[trini: As this is the last non-toradex PXA board, update travis too] Signed-off-by: Tom Rini <trini@konsulko.com>
Philippe Reynes [Mon, 14 Oct 2019 17:41:06 +0000 (19:41 +0200)]
ubi: env: fix redundand management
We check (with a #if defined()) the config ENV_UBI_VOLUME_REDUND
to know if there is a redundant env. But this config is a string
and is always defined with env is in ubi, so we always consider
that a redundand env is used.
To fix this issue, I've added a hidden flag ENV_UBI_IS_VOLUME_REDUND
that is true when ENV_UBI_VOLUME_REDUND is not "". Then, I check
this flag in the code, instead of the string ENV_UBI_VOLUME_REDUND.
hs: fixed typo s/condider/consider
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com> Reviewed-by: Heiko Schocher <hs@denx.de>
Stefan Roese [Tue, 17 Sep 2019 07:17:53 +0000 (09:17 +0200)]
ubi: Add "skipcheck" command to set/clear this bit in the UBI volume hdr
U-Boot now supports the "skip_check" flag to optionally skip the CRC
check at open time. Currently its only possible to set this bit upon
UBI volume creation. But it might be very useful to also set this bit
on already installed systems (e.g. field upgrade) to make also use of
the boot-time decrease on those systems.
This patch now adds a new "ubi" command "ubi skipcheck" to set or clear
this bit in the UBI volume header:
=> ubi skipcheck rootfs0 on
Setting skip_check on volume rootfs0
BTW: This saves approx. 10 seconds Linux bootup time on a MT7688 based
target with 128MiB of SPI NAND.
Signed-off-by: Stefan Roese <sr@denx.de> Reviewed-by: Heiko Schocher <hs@denx.de> Cc: Quentin Schulz <quentin.schulz@bootlin.com> Cc: Boris Brezillon <boris.brezillon@collabora.com> Cc: Heiko Schocher <hs@denx.de> Cc: Andreas Dannenberg <dannenberg@ti.com>
Some users of static UBI volumes implement their own integrity check,
thus making the volume CRC check done at open time useless. For
instance, this is the case when one use the ubiblock + dm-verity +
squashfs combination, where dm-verity already checks integrity of the
block device but this time at the block granularity instead of verifying
the whole volume.
Skipping this test drastically improves the boot-time.
Adapted to U-Boot by Stefan Roese.
Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com> Signed-off-by: Stefan Roese <sr@denx.de> Reviewed-by: Heiko Schocher <hs@denx.de> Cc: Quentin Schulz <quentin.schulz@bootlin.com> Cc: Boris Brezillon <boris.brezillon@bootlin.com>
Patrick Delaunay [Mon, 30 Sep 2019 08:19:13 +0000 (10:19 +0200)]
dm: Tidy up dump output when there are many devices
At present the 'Index' column of 'dm tree' assumes there is
two digits, this patch increase it to 3 digits.
It also aligns output of 'dm uclass', assuming the same 3 digits index.
The boards with CONFIG_PINCTRL_FULL activated have one pinconfig
by pin configuration, so they can have more than 100 devices
pinconfig (for example with stm32mp157c-ev1 board we have
106 pinconfig node).
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com> Reviewed-by: Simon Glass <sjg@chromium.org>
In the Linux pinctrl binding, the pin configuration nodes don't need to
be direct children of the pin controller device (may be grandchildren for
example). This behavior is managed with the pinconfig u-class which
recursively bind all the sub-node of the pin controller.
But for some binding (when pin configuration is only children of pin
controller) that is not necessary. U-Boot can save memory and reduce
the number of pinconf instance when this feature is deactivated
(for arch stm32mp for example for SPL).
This patch allows to control this feature with a new option
CONFIG_PINCONF_RECURSIVE when it is possible for each individual
pin controller device.
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Fixed CONFIG_IF_ENABLED() condition, added __maybe_unused: Signed-off-by: Simon Glass <sjg@chromium.org>
dm: core: device: switch off power domain after device removal
The power domain associated with a device is enabled when probing,
but currently the domain remains enabled when the device is removed.
Some boards started to disable power domains for selected devices
via custom board_quiesce_devices(), but it doesn't work in many
cases, i. e. because devices still can be accessed later in
.remove() callback on behalf of dm_remove_devices_flags().
Utilize the DM core to power off the device power domain, but add a
device flag to be able to selectively let the power domain enabled
after device removal. This might be required for devices that must
remain enabled when booting OS, i. e. serial console for debug
output, etc.
dm: core: Allow for not controlling the power-domain by DM framework
In some remoteproc cases, enabling the power domain of the core will
start running the core. In such cases image should be loaded before
enabling the power domain. But the current DM framework enables the
power-domain by default during probe. This is causing the remotecore
to start and crash as there is no valid image loaded.
In order to avoid this introduce a DM flag that doesn't allow for
enabling/disabling the power-domain by DM framework.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com> Reviewed-by: Simon Glass <sjg@chromium.org>
dm: device: Fix typo in the non-DEVRES version of devm_kmalloc_array()
When DEVRES is not set, devm_kmalloc_array() is spelled
devm_kmaloc_array() (with one 'l' only).
Fixing it so that the name is the same with and without DEVRES.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Douglas Anderson [Fri, 27 Sep 2019 16:23:56 +0000 (09:23 -0700)]
patman: Use the Change-Id, version, and prefix in the Message-Id
As per the centithread on ksummit-discuss [1], there are folks who
feel that if a Change-Id is present in a developer's local commit that
said Change-Id could be interesting to include in upstream posts.
Specifically if two commits are posted with the same Change-Id there's
a reasonable chance that they are either the same commit or a newer
version of the same commit. Specifically this is because that's how
gerrit has trained people to work.
There is much angst about Change-Id in upstream Linux, but one thing
that seems safe and non-controversial is to include the Change-Id as
part of the string of crud that makes up a Message-Id.
Let's give that a try.
In theory (if there is enough adoption) this could help a tool more
reliably find various versions of a commit. This actually might work
pretty well for U-Boot where (I believe) quite a number of developers
use patman, so there could be critical mass (assuming that enough of
these people also use a git hook that adds Change-Id to their
commits). I was able to find this git hook by searching for "gerrit
change id git hook" in my favorite search engine.
In theory one could imagine something like this could be integrated
into other tools, possibly even git-send-email. Getting it into
patman seems like a sane first step, though.
NOTE: this patch is being posted using a patman containing this patch,
so you should be able to see the Message-Id of this patch and see that
it contains my local Change-Id, which ends in 2b9 if you want to
check.
rsa: Return immediately if required-key verification fails
Currently, if image verification with a required key fails, rsa_verify()
code tries to find another key to verify the FIT image. This however, is
not the intended behavior as the documentation says that required keys
"must be verified for the image / configuration to be considered valid".
This patch fixes the issue by making rsa_verify() return immediately if
the verification of a required key fails.
dm: pinctrl: Skip not associated gpio phandle and rise a warning message
Skip not associated gpio phandle let register the other gpios on a group.
We need anyway to send out a warning to the user to fix their uboot-board.dtsi.
Thhe handle id can be found inside the decompiled dtb
Peng Fan [Tue, 17 Sep 2019 09:29:19 +0000 (09:29 +0000)]
power: domain: add dev_power_domain_on
Add this new API to power on multiple domains attached
to a device.
Signed-off-by: Peng Fan <peng.fan@nxp.com> Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Changed to static inline and added a condition into C file: Signed-off-by: Simon Glass <sjg@chromium.org>
When CONFIG_OF_PRIOR_STAGE is enabled, this workaround was needed
before device_bind_common assigned request numbers sequentially in the
absence of aliases.
Signed-off-by: Thomas Fitzsimmons <fitzsim@fitzsim.org> Cc: Bin Meng <bmeng.cn@gmail.com> Cc: Simon Glass <sjg@chromium.org>
For CONFIG_OF_PRIOR_STAGE, in the absence of a device tree alias for a
given device, use the next request number for that type of device.
This allows aliases to be used when they're available, while still
allowing unaliased devices to be probed.
Signed-off-by: Thomas Fitzsimmons <fitzsim@fitzsim.org> Cc: Bin Meng <bmeng.cn@gmail.com> Cc: Simon Glass <sjg@chromium.org>
The reset function sets the pin to 0 then 1 but if the pin is marked
ACTIVE_LOW in the DT it gets inverted and leaves the TPM in reset.
Let the gpio driver take care of the reset polarity.
Default address cells value on the livetree access function
returns the wrong value. Fix this so that the value returned
corresponds to the device tree specification.
The commit "libfdt: fdt_address_cells() and fdt_size_cells()" introduced
a bug as it consolidated code between the helpers for getting
be 0, and is frequently found so in practice for /cpus. IEEE1275 only
requires implementations to handle 1..4 for #address-cells, although one
could make a case for #address-cells == #size-cells == 0 being used to
represent a bridge with a single port.
While we're there, it's not totally obvious that the existing implicit
cast of a u32 to int will give the correct results according to strict C,
although it does work in practice. Straighten that up to cast only after
we've made our range checks.
This is based on upstream commit: b8d6eca ("libfdt: Allow #size-cells of 0")
but misses the test cases,as we don't implement them in U-Boot.
libfdt: return correct value if #size-cells property is not present
According to the device tree specification, the default value for
was not present.
This patch also makes fdt_address_cells() and fdt_size_cells() conform
to the behaviour documented in libfdt.h. The defaults are only returned
if fdt_getprop() returns -FDT_ERR_NOTFOUND, otherwise the actual error
is returned.
This is based on upstream commit: aa7254d ("libfdt: return correct value if #size-cells property is not present")
but misses the test case part, as we don't implement them in U-Boot.
Add internal fdt_cells() to avoid copy and paste. Fix typo in
fdt_size_cells() documentation comment.
This is based in upstream commit: c12b2b0 ("libfdt: fdt_address_cells() and fdt_size_cells()")
but misses the test cases, as we don't implement them in U-Boot.
Signed-off-by: Matthias Brugger <mbrugger@suse.com> Reviewed-by: Simon Glass <sjg@chromium.org>
AKASHI Takahiro [Tue, 27 Aug 2019 08:17:03 +0000 (17:17 +0900)]
sandbox: fix cpu property in test.dts for pytest
When I tried to run some new efi tests with pytest, efi_smbios_register()
triggered a segmentation fault. Here is the location where it happened:
efi_init_obj_list()
efi_smbios_register()
write_smbios_table()
smbios_write_type4()
smbios_write_type4_dm()
where dev_get_parent_platdata() should return a pointer to struct
cpu_platdata, but it is actually NULL because any cpu device on
sandbox is attached to "root_driver."
With this patch, this issue will be fixed by moving all the definitions
of cpus under "cpus" node so that they have a "cpu_bus" parent.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
In the Linux pinctrl binding, the pin configuration nodes don't need to
be direct children of the pin controller device (may be grandchildren for
example). This behavior is managed with the pinconfig u-class which
recursively bind all the sub-node of the pin controller.
But for some binding (when pin configuration is only children of pin
controller) that is not necessary. U-Boot can save memory and reduce
the number of pinconf instance when this feature is deactivated
(for arch stm32mp for example for SPL).
This patch allows to control this feature with a new option
CONFIG_PINCONF_RECURSIVE when it is possible for each individual
pin controller device.
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Simon Glass [Sat, 24 Aug 2019 13:23:12 +0000 (07:23 -0600)]
binman: Write symbol info before image inclusion
At present the symbol information is written to binaries just before
binman exits. This is fine for entries within sections since the section
contents is calculated when it is needed, so the updated symbol values are
included in the image that is written.
However some binaries are inside entries which have already generated
their contents and do not notice that the entries have changed (e.g. Intel
IFWI).
Move the symbol writing earlier to cope with this.
Simon Glass [Sat, 24 Aug 2019 13:23:11 +0000 (07:23 -0600)]
binman: Support writing symbols into entries within an IFWI
The Intel IFWI (Integrated Firmware Image) is effectively a section with
other entries inside it. Support writing symbol information into entries
within it.
Simon Glass [Sat, 24 Aug 2019 13:23:04 +0000 (07:23 -0600)]
binman: Increase size of TPL and SPL test data
At present these are large enough to hold 20 bytes of symbol data. Add
four more bytes so we can add another test.
Unfortunately at present this involves changing a few test files to make
room. We could adjust the test files to not specify sizes for entries.
Then we could make the tests check the actual sizes. But for now, leave it
as it is, since the effort is minor.
Simon Glass [Sat, 24 Aug 2019 13:23:03 +0000 (07:23 -0600)]
binman: Avoid needing the section size in advance
Entries which include a section and need to obtain its contents call
GetData(), as with any other entry. But the current implementation of this
method in entry_Section requires the size of the section to be known. If
it is unknown, an error is produced, since size is None:
TypeError: can't multiply sequence by non-int of type 'NoneType'
There is no need to know the size in advance since the code can be
adjusted to build up the section piece by piece, instead of patching each
entry into an existing bytearray.
Simon Glass [Sat, 24 Aug 2019 13:23:00 +0000 (07:23 -0600)]
binman: Allow symbols to be resolved inside sections
At present we only support symbols inside binaries which are at the top
level of an image. This restrictions seems unreasonable since more complex
images may want to group binaries within different sections.
Relax the restriction, adding a new _SetupTplElf() helper function.
Simon Glass [Sat, 24 Aug 2019 13:22:54 +0000 (07:22 -0600)]
binman: Use the Makefile for u_boot_ucode_ptr
Remove this file from git and instead build it using the Makefile.
Update tools.GetInputFilename() to support reading files from an absolute
path, so that we can read the Elf test files easily. Also make sure that
the temp directory is report in ELF tests as this was commented out.
Simon Glass [Sat, 24 Aug 2019 13:22:53 +0000 (07:22 -0600)]
binman: Use the Makefile to build ELF test files
At present the ELF test files are checked into the U-Boot tree. This is
covenient since the files never change and can be used on non-x86
platforms. However it is not good practice to check in binaries and in
this case it does not seem essential.
Update the binman test-file Makefile to support having source in a
different directory. Adjust binman to run it to build bss_data, as a
start. We can add other files as needed.
Simon Glass [Sat, 24 Aug 2019 13:22:51 +0000 (07:22 -0600)]
binman: Fix IFWI output when using an Intel FIT image
At present this entry does not work correctly when a FIT image is used as
the input. It updates the FIT instead of the output image. The test passed
because the FIT image happened to have the right data already.
Simon Glass [Sat, 24 Aug 2019 13:22:50 +0000 (07:22 -0600)]
binman: Add support for Intel FIT
A Firmware Image Table (FIT) is a data structure defined by Intel which
contains information about various things needed by the SoC, such as
microcode.
Add support for this entry as well as the pointer to it. The contents of
FIT are fixed at present. Future work is needed to support adding
microcode, etc.
Simon Glass [Sat, 24 Aug 2019 13:22:49 +0000 (07:22 -0600)]
binman: x86: Separate out 16-bit reset and init code
At present these two sections of code are linked together into a single
2KB chunk in a single file. Some Intel SoCs like to have a FIT (Firmware
Interface Table) in the ROM and the pointer for this needs to go at
0xffffffc0 which is in the middle of these two sections.
Make use of the new 'reset' entry and change the existing 16-bit entry to
include just the 16-bit data.
Simon Glass [Sat, 24 Aug 2019 13:22:48 +0000 (07:22 -0600)]
binman: Add support for an x86 'reset' section
At present binman has a single entry type for the 16-bit code code needed
to start up an x86 processor. This entry is intended to include both the
reset vector itself as well as the code to move to 32-bit mode.
However this is not very flexible since in some cases other data needs to
be included at the top of the SPI flash, in between these two pieces. For
example Intel requires that a FIT (Firmware Image Table) pointer be placed
0x40 bytes before the end of the ROM.
To deal with this, add a new reset entry for just the reset vector. A
subsequent change will adjust the existing 'start16' entry.
Simon Glass [Sat, 24 Aug 2019 13:22:47 +0000 (07:22 -0600)]
binman: Correct use of 'replace' in IFWI tests
At present the Intel IFWI entry uses 'replace' without the 'ifwi-' prefix.
This is a fairly generic name which might conflict with the main Entry
base class at some point, if more features are added. Add a prefix.
Simon Glass [Sat, 24 Aug 2019 13:22:43 +0000 (07:22 -0600)]
binman: Use cls instead of self for class methods
It is more common to use the name 'cls' for the class object of a class
method, to distinguish it from normal methods, which use 'self' Update the
binman tests accordingly.
Simon Glass [Sat, 24 Aug 2019 13:22:42 +0000 (07:22 -0600)]
patman: Update command.Run() to handle failure better
At present tools are not expected to fail. If they do an exception is
raised but there is no detail about what went wrong. This makes it hard
to debug if something does actually go wrong.
Fix this by outputting both stderr and stdout on failure.
Sam Protsenko [Thu, 15 Aug 2019 17:25:07 +0000 (20:25 +0300)]
image: android: Support boot image v1 and v2
In new versions of Android Boot Image next fields are added to the
header (and corresponding payloads are added to the image itself):
- v1: Recovery DTBO
- v2: DTB
Account for these new fields in next functions:
1. android_image_get_end(): as v1 and v2 have new payloads in the
image, the calculation of image end address should be extended
correspondingly; this is used e.g. by "bootm" command when booting
the kernel from Android Boot Image
2. android_print_contents(): new fields values in v1 and v2 should be
printed; the result of this function can be seen e.g. in "iminfo"
command
This commit doesn't add the means for working with new payloads in v1
and v2 images (it will be done in further commits), it only updates
existing functions w.r.t. changes in boot image v1/v2.
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Sam Protsenko [Fri, 9 Aug 2019 12:31:29 +0000 (15:31 +0300)]
image: android: Fix possible build errors
As android_image.h uses types like u32, we need to include corresponding
headers in place. Otherwise the user will be forced to include those in
C file, or next build error can occur:
include/android_image.h:32:5: error: unknown type name 'u32'
u32 kernel_size; /* size in bytes */
Include required headers for data types used. While at it, remove
typedef struct, which is prohibited by kernel coding style, and fix the
comment.
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
- logicpd pinmux
- i.MX7ULP: imx_ddr_size
- fixes Toradex i.MX6/i.MX7
- pico-imx7d
- tpc70 converted to DM
- New Board: meerkat96
- add HAB version command
- i.MX8 :
imx8: Jump from alias to OCRAM address at SPL init
imx8qm/qxp: Set SPL TEXT base to OCRAM base
Tom Rini [Mon, 14 Oct 2019 11:28:32 +0000 (07:28 -0400)]
Merge tag 'mmc-10-10-2019' of https://gitlab.denx.de/u-boot/custodians/u-boot-mmc
- Add sdhci driver for Broadcom iProc platform
- Add a driver callback for power-cycle for mmc
- Implement host_power_cycle callback for stm32_sdmmc2
- spl: dm_mmc: Initialize only the required mmc device
Use imx_ddr_size() to retrieve the total DDR size instead
of passing a hardcoded memory size.
imx_ddr_size() calculates the memory size based on the actual
MMDC registers values and is useful to detect misconfigurations,
so switch to this more robust approach.
projects / u-boot.git / log