Simon Glass [Mon, 25 Apr 2022 05:31:25 +0000 (23:31 -0600)]
bootstd: Add setup for the bootflow tests
We need to create a disk image with a partition table and a DOS-format
filesystem containing a few files. Provide a fallback binary for CI since
it does not seem able to detect the loopback partitions.
Add this to a dm_init test so that it happens when needed.
Simon Glass [Mon, 25 Apr 2022 05:31:22 +0000 (23:31 -0600)]
bootstd: Add an implementation of script boot
Add a bootmeth driver which handles distro boot from a disk via a U-Boot
script, so we can boot a bootflow using this commonly used mechanism. This
is required by Armbian, for example.
Simon Glass [Mon, 25 Apr 2022 05:31:21 +0000 (23:31 -0600)]
bootstd: sandbox: Add a hostfs bootdev
It is helpful to be able to try out bootstd on sandbox, using host files.
This is easier than using a block device, which must have a filesystem,
partition table, etc.
Add a new driver which provides this feature. For now it is not used in
tests, but it is likely to be useful.
Add notes in the devicetree also, but don't disturb the tests.
Simon Glass [Mon, 25 Apr 2022 05:31:19 +0000 (23:31 -0600)]
bootstd: Add an implementation of EFI bootmgr
Add a bootmeth driver which handles EFI boot manager, using EFI_LOADER.
In effect, this provides the same functionality as the 'bootefi bootmgr'
command and shares the same code. But the interface into it is via a
bootmeth, so it does not require any special scripts, etc.
For now this requires the 'bootefi' command be enabled. Future work may
tidy this up so that it can be used without CONFIG_CMDLINE being enabled.
Simon Glass [Mon, 25 Apr 2022 05:31:18 +0000 (23:31 -0600)]
bootstd: Add a system bootdev for strange boot methods
Some boot methods don't act on a single bootdev but instead do their own
thing. An example is EFI bootmgr which scan various devices using its own
logic. Add a bootdev to handle this.
Simon Glass [Mon, 25 Apr 2022 05:31:17 +0000 (23:31 -0600)]
bootstd: Add an implementation of EFI boot
Add a bootmeth driver which handles EFI boot, using EFI_LOADER.
In effect, this provides the same functionality as the 'bootefi' command
and shares the same code. But the interface into it is via a bootmeth,
so it does not require any special scripts, etc.
For now this requires the 'bootefi' command be enabled. Future work may
tidy this up so that it can be used without CONFIG_CMDLINE being enabled.
There was much discussion about whether this is needed, but it seems
that it is, at least for now.
Simon Glass [Mon, 25 Apr 2022 05:31:07 +0000 (23:31 -0600)]
bootstd: Add the bootdev uclass
A 'bootdev' is a device which can be used to boot an operating system.
It is a child of the media device (e.g. MMC) which handles reading files
from that device, such as a bootflow file.
Add a uclass for bootdev and the various helpers needed to make it
work. Also add a binding file, empty for now.
Simon Glass [Mon, 25 Apr 2022 05:31:05 +0000 (23:31 -0600)]
bootstd: Add the concept of a bootflow
A bootflow encapsulates the process used to boot an operating system.
It typically has a control file (such as extlinux.conf) and information
about which 'bootdev' it came from.
Add the header file for this first, since it is needed by all other
files.
Simon Glass [Mon, 25 Apr 2022 05:31:04 +0000 (23:31 -0600)]
fs: Add a function to set the filesystem type
When sandbox is used with hostfs we won't have a block device, but still
must set up the filesystem type before any filesystem operation, such as
loading a file. Add a function to handle this.
Simon Glass [Mon, 25 Apr 2022 05:31:00 +0000 (23:31 -0600)]
dm: core: Allow finding a uclass device by partial name
In some cases two devices are related and the only way to tell is to
check that the names partially patch. Add a way to check this without
needing to create a new string for the comparison.
Fix the comment for device_find_child_by_namelen() while we are here.
Simon Glass [Mon, 25 Apr 2022 05:30:59 +0000 (23:30 -0600)]
dm: core: Rename and fix uclass_get_by_name_len()
It seems that namelen is more common in U-Boot. Rename this function to
fit in better. Also fix a bug where it breaks the operation of
uclass_get_by_name() and add a test.
Signed-off-by: Simon Glass <sjg@chromium.org> Reported-by: Patrick Delaunay <patrick.delaunay@foss.st.com> Reported-by: Tim Harvey <tharvey@gateworks.com>
Simon Glass [Mon, 25 Apr 2022 05:30:58 +0000 (23:30 -0600)]
lib: Add a way to find the postiion of a trailing number
At present it is not possible to find out which part of the string is the
number part and which is before it. Add a new variant which provides this
feature, so we can separate the two in the caller.
Simon Glass [Mon, 25 Apr 2022 05:30:57 +0000 (23:30 -0600)]
lib: Fix a few bugs in trailing_strtoln()
At present this has a minor bug in that it reads the byte before the
start of the string, if it is empty. Also it doesn't handle a
non-numeric prefix which is only one character long.
Fix these bugs with a reworked implementation. Add a test for the second
case. The first one is hard to test.
Tom Rini [Sat, 23 Apr 2022 12:06:28 +0000 (08:06 -0400)]
Merge branch '2022-04-22-assorted-updates'
- Add "-q" to fdt addr and use it in distro_bootcmd to make the user
experience less scary reading in normal try/fail cases.
- Let the adc update an environment variable like many other commands do
- Fix TPL SEPARATE_BSS check when locating DTB
- Allow ":" in PXE file names again
- Two Apple M1 fixes
Mark Kettenis [Tue, 19 Apr 2022 19:20:31 +0000 (21:20 +0200)]
arm: apple: Point stdout-path to framebuffer when keyboard present
Unless you have a spare Apple Silicon machine, getting access to
the serial port on Apple Silicon machines requires special
hardware. Given that most machines come with a built-in screen
the framebuffer is likely to be the most convenient output device
for most users. While U-Boot will output to both serial and
framebuffer, OSes might not. Therefore set stdout-path to point
at /chosen/framebuffer when a keyboard is connected to the machine.
This behaviour can be overridden by setting the "stdout" variable
in the U-Boot environment. I addition to that keep the serial
console as the default when running under the m1n1 hypervisor.
The m1n1 hypervisor virtualizes the serial port such that it
can be easily accessed from any other machine with a USB port.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org> Reviewed-by: Janne Grunau <j@jannau.net> Tested-by: Janne Grunau <j@jannau.net>
Mark Kettenis [Tue, 19 Apr 2022 19:10:29 +0000 (21:10 +0200)]
arm: apple: Don't clear framebuffer
Enable CONFIG_NO_FB_CLEAR to preserve the Asahi logo. Since that
logo is drawn on a black background also enable
CONFIG_SYS_WHITE_ON_BLACK such that text printed by U-Boot is still
visible.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Lyle Franklin [Sat, 16 Apr 2022 15:36:43 +0000 (11:36 -0400)]
Allow colon in PXE bootfile URLs
- U-boot's PXE flow supports prefixing your bootfile name with an
IP address to fetch from a server other than the DHCP server,
e.g. `hostIPaddr:bootfilename`:
https://github.com/u-boot/u-boot/commit/a93907c43f847f076dd0e34ee3b69b5e8e6d0d29
- However, this breaks bootfile paths which contain a colon, e.g.
`f0:ad:4e:10:1b:87/7/pxelinux.cfg/default`
- This patch checks whether the `hostIPaddr` prefix is a valid
IP address before overriding the serverIP otherwise the whole
bootfile path is preserved
Signed-off-by: Lyle Franklin <lylejfranklin@gmail.com>
The documentation for CONFIG_IS_ENABLED() in include/linux/kconfig.h
implies that we will get the correct behaviour, but the actual behaviour
differs such that this condition is now always false.
This stopped TPL being able to load the device tree blob at least on the
ROCKPro64 board (RK3399 SoC), since the wrong device tree location was
chosen.
The issues causing this behaviour with CONFIG_IS_ENABLED() are:
1. The documentation implies that CONFIG_SPL_BUILD =>
CONFIG_SPL_<option> is considered before the TPL equivalent.
Actually, the TPL options have higher priority - see definition of
_CONFIG_PREFIX.
2. The documentation implies a fallthrough, eg. if CONFIG_SPL_BUILD is
defined but the CONFIG_SPL_<option> is not, then it will proceed to
check if CONFIG_TPL_BUILD
Actually, if CONFIG_TPL_BUILD is defined, then it stops there
and CONFIG_SPL_BUILD is not considered - see definition of
_CONFIG_PREFIX.
During TPL build, at least for the ROCKPro64, both CONFIG_TPL_BUILD
and CONFIG_SPL_BUILD are defined, but because of the above, only TPL
options are considered. Since there is no CONFIG_TPL_SEPARATE_BSS,
this fails.
Fixes: 5064a35a96 ("fdt: Correct condition for SEPARATE_BSS") Signed-off-by: Andrew Abbott <andrew@mirx.dev>
Sean Anderson [Wed, 23 Mar 2022 18:26:09 +0000 (14:26 -0400)]
power-domain: Fix use of uninitialized value in dev_power_domain_ctrl
If dev_count_phandle_with_args returns 0 or another error, then pd will never
have been initialized by power_domain_get_by_index. Avoid comparing against
pd.dev in this situation.
Fixes: efbabe6645 ("power-domain: fix hang in endless loop on i.MX8") Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
Michal Simek [Wed, 9 Mar 2022 09:05:45 +0000 (10:05 +0100)]
usb: dwc3: Add support for usb3-phy PHY configuration
When usb3-phy label is found, PHY driver is called and serdes line is
initialized. This is preparation for serdes/psgtr driver to configure GT
lines based on description in DT.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Jim Liu [Fri, 1 Apr 2022 09:59:39 +0000 (17:59 +0800)]
wdt: nuvoton: Add support for Nuvoton
Add watchdog controller driver for NPCM7xx/npcm8xx
the wdt design of npcm750 and npcm845 is the same.
so the driver can work on npcm750 and npcm845.
about npcm845 wdt dtsi i will followed kernel dts name
to use nuvoton,npcm750-wdt.
Signed-off-by: Jim Liu <JJLIU0@nuvoton.com> Reviewed-by: Stefan Roese <sr@denx.de> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Thu, 21 Apr 2022 23:55:38 +0000 (19:55 -0400)]
Merge branch '2022-04-21-further-cleanups'
- Fix SPL_SYS_MALLOC_SIMPLE and non-SPL_FRAMEWORK boards (a large number
of PowerPC platforms)
- Remove duplication of crc16 functionality
- Migrate COUNTER_FREQUENCY to CONFIG_COUNTER_FREQUENCY and have it in
Kconfig
Currently the space between kernel_addr_r and the fdt_addr_r is only 32MB.
To have enough space to load kernel images bigger than 32MB change the
variables to a feasible value.
The new environment variables layout is based on the scheme from
"include/configs/ti_armv7_common.h".
The CONFIG_SYS_LOAD_ADDR value is set to 0x42000000. With that we have
the same value as for the kernel_addr_r.
Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com> Reviewed-by: Peng Fan <peng.fan@nxp.com>
In arch/arm/mach-imx/imx8m/soc.c there's an implementation of
board_fix_fdt() introduced by commit 3efdcc5fc335. Remove the
redundant one to avoid failed to build from source when enabling
CONFIG_OF_BOARD_FIXUP.
Peng Fan [Wed, 13 Apr 2022 09:47:19 +0000 (17:47 +0800)]
arch: arm: Kconfig: set default COUNTER_FREQUENCY
Set default COUNTER_FREQUENCY according to config header file
under include/configs/
i.MX6UL/ULL/7D/8QM/8QXP all has system counter frequency run at 8MHz,
so set default value for them.
SUNXI/EXYNOS/ROCKCHIP_RK3128/ROCKCHIP_RK3288/ROCKCHIP_RK322X/ROCKCHIP_RK3036
at 24MHz. ARCH_LX2160A at 25MHz
ARCH_ZYNQMP at 100MHz
fs: ext4: Use CRC-16 implementation from linux/crc16.h
Implementation in linux/crc16.h provides standard CRC-16 algorithm with
polynomial x^16 + x^15 + x^2 + 1. Use it and remove duplicate ext4 CRC-16
specific code.
Signed-off-by: Pali Rohár <pali@kernel.org> Reviewed-by: Stefan Roese <sr@denx.de>
crc16-ccitt: Rename file with CRC-16-CCITT implementation to crc16-ccitt.c
U-Boot CRC-16 implementation uses polynomial x^16 + x^12 + x^5 + 1 which is
not standard CRC-16 algorithm, but it is known as CRC-16-CCITT. Rename file
crc16.c to crc16-ccitt.c to reduce confusion.
Signed-off-by: Pali Rohár <pali@kernel.org> Reviewed-by: Stefan Roese <sr@denx.de>
Inspection showed that gd->malloc_limit is zero. And it is because
generally SPL_FRAMEWORK initialize SPL's gd->malloc_limit. But when
SPL_FRAMEWORK is not enabled then in most cases nobody initialize
gd->malloc_limit and so SPL malloc simple does not work.
So disable SPL_SYS_MALLOC_SIMPLE by default when SPL_FRAMEWORK is not
enabled. SPL_SYS_MALLOC_SIMPLE can be disabled only by setting
SPL_SYS_MALLOC_F_LEN to zero. So do it.
This change fixes SPL error "alloc space exhausted" on P2020 board.
Tom Rini [Thu, 21 Apr 2022 14:56:58 +0000 (10:56 -0400)]
Merge branch '2022-04-21-assorted-improvements'
- For the environment, add a arch_env_get_location hook and make use of
it on some NXP platforms (so that boards can override SoCs).
- Remove some unused squashfs code from SPL
- Resync am335x beaglebone related DTS files
- Enable SPL_SEPARATE_BSS if SPL_BSS_START_ADDR is used
Enable DM_SERIAL for both U_Boot and the SPL. The uart4 and its pinmux
are already marked with u-boot,dm-spl but we need to move the call to
preloader_console_init() after spl_init() to avoid a board hang
as dm can't be used until after spl_init().
Remove the manual config of the UART pinmux now that it is no longer
needed.
Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com> Cc: Tim Harvey <tharvey@gateworks.com> Reviewed-by: Fabio Estevam <festevam@denx.de> Tested-by: Ariel D'Alessandro <ariel.dalessandro@collabora.com>
Sean Anderson [Tue, 12 Apr 2022 14:59:04 +0000 (10:59 -0400)]
treewide: Enable SPL_SEPARATE_BSS if SPL_BSS_START_ADDR is used
If .bss does not immediately follow the end of the image, then
CONFIG_SPL_SEPARATE_BSS must be selected. Typically, the location of bss
is specified by using CONFIG_SPL_BSS_START_ADDR in a linker script. On
these arches, CONFIG_SPL_SEPARATE_BSS should be enabled. If there is an
option to use an alternate boot script (e.g. CONFIG_SPL_LDSCRIPT is just
a default), just imply. If there is not, select.
Signed-off-by: Sean Anderson <sean.anderson@seco.com>
Paul Barker [Mon, 11 Apr 2022 15:42:01 +0000 (15:42 +0000)]
board: ti: am335x: Add support for BBE Extended WiFi
The Sancloud BeagleBone Enhanced Extended WiFi (BBE Extended WiFi) has
its own devicetree file and the board can be identified by the 2nd
letter of the config string within the common EEPROM.
Signed-off-by: Paul Barker <paul.barker@sancloud.com>
Paul Barker [Mon, 11 Apr 2022 15:42:00 +0000 (15:42 +0000)]
board: ti: am335x: Add BBE Lite support
The Sancloud BeagleBone Enhanced Lite (BBE Lite) has its own devicetree
file and the board can be identified by the 2nd letter of the config
string within the common EEPROM.
Signed-off-by: Paul Barker <paul.barker@sancloud.com>
Paul Barker [Mon, 11 Apr 2022 15:41:59 +0000 (15:41 +0000)]
arm: dts: Resync dts for BeagleBone and derivatives
Sync BeagleBone dts files & TPS dtsi files with Linux v5.17 and include
the SanCloud BBE Extended WiFi dts added in v5.18-rc1. Also pull in
changes to am33xx-l4.dtsi needed to support the BeagleBone Blue.
The change to use the cpsw switch driver (commit c477358e66a3 in Linux)
is excluded from the sync as u-boot does not recognise the new
compatible string.
Signed-off-by: Paul Barker <paul.barker@sancloud.com>
Marek Vasut [Wed, 6 Apr 2022 00:21:34 +0000 (02:21 +0200)]
ARM: imx: imx8m: env: Switch to arch_env_get_location()
Implement arch_env_get_location() instead of env_get_location(), so that
the env_get_location() can be implemented on board level and override the
arch_env_get_location() architecture defaults.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Adam Ford <aford173@gmail.com> Cc: Fabio Estevam <festevam@denx.de> Cc: Marek Behún <marek.behun@nic.cz> Cc: Peng Fan <peng.fan@nxp.com> Cc: Simon Glass <sjg@chromium.org> Cc: Stefano Babic <sbabic@denx.de> Cc: Tim Harvey <tharvey@gateworks.com> Cc: Tom Rini <trini@konsulko.com> Cc: Tommaso Merciai <tomm.merciai@gmail.com> Cc: Vladimir Oltean <vladimir.oltean@nxp.com>
Marek Vasut [Wed, 6 Apr 2022 00:21:33 +0000 (02:21 +0200)]
armv8: layerscape: env: Switch to arch_env_get_location()
Implement arch_env_get_location() instead of env_get_location(), so that
the env_get_location() can be implemented on board level and override the
arch_env_get_location() architecture defaults.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Adam Ford <aford173@gmail.com> Cc: Fabio Estevam <festevam@denx.de> Cc: Marek Behún <marek.behun@nic.cz> Cc: Peng Fan <peng.fan@nxp.com> Cc: Simon Glass <sjg@chromium.org> Cc: Stefano Babic <sbabic@denx.de> Cc: Tim Harvey <tharvey@gateworks.com> Cc: Tom Rini <trini@konsulko.com> Cc: Tommaso Merciai <tomm.merciai@gmail.com> Cc: Vladimir Oltean <vladimir.oltean@nxp.com>
Currently there is only one way to override desired environment location,
by implementing env_get_location(). This is increasingly being conflated
both on board level and architecture level, which leads to a problem on
boards where this function is already implemented on architecture level,
since those boards have no way to override this environment location on
board level anymore.
Implement arch_env_get_location() function which is architecture specific
and should only ever be implemented in architecture code. This function
has lower priority than env_get_location(), which should only ever be
implemented in board code, and which overrides the arch_env_get_location()
architecture environment selection.
This way, architecture can define its default environment chooser, while
board can now override it as needed at all times.
There is no functional change, since env_get_location() simply returns
arch_env_get_location(), and arch_env_get_location() implements the
current env_get_location() default content.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Adam Ford <aford173@gmail.com> Cc: Fabio Estevam <festevam@denx.de> Cc: Marek Behún <marek.behun@nic.cz> Cc: Peng Fan <peng.fan@nxp.com> Cc: Simon Glass <sjg@chromium.org> Cc: Stefano Babic <sbabic@denx.de> Cc: Tim Harvey <tharvey@gateworks.com> Cc: Tom Rini <trini@konsulko.com> Cc: Tommaso Merciai <tomm.merciai@gmail.com> Cc: Vladimir Oltean <vladimir.oltean@nxp.com>