| Age | Commit message (Collapse) | Author |
|---|
|
This replaces dm_remove_devices_flags() calls in all boot
implementations to ensure non vital devices are consistently removed
first. All boot implementation except arch/arm/lib/bootm.c currently
just call dm_remove_devices_flags(DM_REMOVE_ACTIVE_ALL). This can result
in crashes when dependencies between devices exists. The driver model's
design document describes DM_FLAG_VITAL as "indicates that the device is
'vital' to the operation of other devices". Device removal at boot
should follow this.
Instead of adding dm_remove_devices_flags() with (DM_REMOVE_ACTIVE_ALL |
DM_REMOVE_NON_VITAL) everywhere add dm_remove_devices_active() which
does this.
Fixes a NULL pointer deref in the apple dart IOMMU driver during EFI
boot. The xhci-pci (driver which depends on the IOMMU to work) removes
its mapping on removal. This explodes when the IOMMU device was removed
first.
dm_remove_devices_flags() is kept since it is used for testing of
device_remove() calls in dm.
Signed-off-by: Janne Grunau <[email protected]>
|
|
All the uclass functions for finding a device end up creating a uclass
if it doesn't exist. Add a function which instead returns NULL in this
case.
This is useful when in the 'unbind' path, since we don't want to undo
any unbinding which has already happened.
Signed-off-by: Simon Glass <[email protected]>
|
|
People complained that enabling (SPL_)DM_WARN was now totally unusable
due to the amount of messages printed on the console.
Let's downgrade the log level of some messages that are clearly not on
the error path.
Note that there's one pr_debug in there, because it is followed by
pr_cont so it made sense to reuse the same family of functions.
Reported-by: Alexander Dahl <[email protected]>
Fixes: 6afdb1585112 ("dm: core: migrate debug() messages to use dm_warn")
Signed-off-by: Quentin Schulz <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Patrick Rudolph <[email protected]> says:
Based on the existing work done by Simon Glass this series adds
support for booting aarch64 devices using ACPI only.
As first target QEMU SBSA support is added, which relies on ACPI
only to boot an OS. As secondary target the Raspberry Pi4 was used,
which is broadly available and allows easy testing of the proposed
solution.
The series is split into ACPI cleanups and code movements, adding
Arm specific ACPI tables and finally SoC and mainboard related
changes to boot a Linux on the QEMU SBSA and RPi4. Currently only the
mandatory ACPI tables are supported, allowing to boot into Linux
without errors.
The QEMU SBSA support is feature complete and provides the same
functionality as the EDK2 implementation.
The changes were tested on real hardware as well on QEMU v9.0:
qemu-system-aarch64 -machine sbsa-ref -nographic -cpu cortex-a57 \
-pflash secure-world.rom \
-pflash unsecure-world.rom
qemu-system-aarch64 -machine raspi4b -kernel u-boot.bin -cpu cortex-a72 \
-smp 4 -m 2G -drive file=raspbian.img,format=raw,index=0 \
-dtb bcm2711-rpi-4-b.dtb -nographic
Tested against FWTS V24.03.00.
Known issues:
- The QEMU rpi4 support is currently limited as it doesn't emulate PCI,
USB or ethernet devices!
- The SMP bringup doesn't work on RPi4, but works in QEMU (Possibly
cache related).
- PCI on RPI4 isn't working on real hardware since the pcie_brcmstb
Linux kernel module doesn't support ACPI yet.
Link: https://lore.kernel.org/r/[email protected]
|
|
Add a new method to acpi_ops to let drivers fill out ACPI MADT.
The code is unused for now until drivers implement the new ops.
TEST: Booted on QEMU sbsa using driver model generated MADT.
Signed-off-by: Patrick Rudolph <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Cc: Simon Glass <[email protected]>
|
|
Simon Glass <[email protected]> says:
When the SPL build-phase was first created it was designed to solve a
particular problem (the need to init SDRAM so that U-Boot proper could
be loaded). It has since expanded to become an important part of U-Boot,
with three phases now present: TPL, VPL and SPL
Due to this history, the term 'SPL' is used to mean both a particular
phase (the one before U-Boot proper) and all the non-proper phases.
This has become confusing.
For a similar reason CONFIG_SPL_BUILD is set to 'y' for all 'SPL'
phases, not just SPL. So code which can only be compiled for actual SPL,
for example, must use something like this:
#if defined(CONFIG_SPL_BUILD) && !defined(CONFIG_TPL_BUILD)
In Makefiles we have similar issues. SPL_ has been used as a variable
which expands to either SPL_ or nothing, to chose between options like
CONFIG_BLK and CONFIG_SPL_BLK. When TPL appeared, a new SPL_TPL variable
was created which expanded to 'SPL_', 'TPL_' or nothing. Later it was
updated to support 'VPL_' as well.
This series starts a change in terminology and usage to resolve the
above issues:
- The word 'xPL' is used instead of 'SPL' to mean a non-proper build
- A new CONFIG_XPL_BUILD define indicates that the current build is an
'xPL' build
- The existing CONFIG_SPL_BUILD is changed to mean SPL; it is not now
defined for TPL and VPL phases
- The existing SPL_ Makefile variable is renamed to SPL_
- The existing SPL_TPL Makefile variable is renamed to PHASE_
It should be noted that xpl_phase() can generally be used instead of
the above CONFIGs without a code-space or run-time penalty.
This series does not attempt to convert all of U-Boot to use this new
terminology but it makes a start. In particular, renaming spl.h and
common/spl seems like a bridge too far at this point.
The series is fully bisectable. It has also been checked to ensure there
are no code-size changes on any commit.
|
|
Use PHASE_ as the symbol to select a particular XPL build. This means
that SPL_TPL_ is no-longer set.
Update the comment in bootstage to refer to this symbol, instead of
SPL_
Signed-off-by: Simon Glass <[email protected]>
|
|
Use XPL_ as the symbol to indicate an SPL build. This means that SPL_ is
no-longer set.
Signed-off-by: Simon Glass <[email protected]>
|
|
Use the new symbol to refer to any 'SPL' build, including TPL and VPL
Signed-off-by: Simon Glass <[email protected]>
|
|
Christian Marangi <[email protected]> says:
This series is a reworked version of the previous seried:
misc: introduce STATUS LED activity function
This series port and expand the legacy concept of LED boot from
the legacy Status LED API to new LED API.
One thing that many device need is a way to communicate to the
user that the device is actually doing something.
This is especially useful for recovery steps where an
user (for example) insert an USB drive, keep a button pressed
and the device autorecover.
There is currently no way to signal the user externally that
the bootloader is processing/recoverying aside from setting
a LED on.
A solid LED on is not enough and won't actually signal any
kind of progress.
Solution is the good old blinking LED but uboot doesn't
suggest (and support) interrupts and almost all the LED
are usually GPIO LED that doesn't support HW blink.
Additional Kconfg are also introduced to set the LED boot and
activity. Those are referenced by label.
A documentation for old and these new LED API is created.
|
|
Implement ofnode_options helpers to read options in /options/u-boot to
adapt to the new way to declare options as described in [1].
[1] dtschema/schemas/options/u-boot.yaml
Signed-off-by: Christian Marangi <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Add an option so that this feature can be enabled in TPL for boards
which need it.
Signed-off-by: Simon Glass <[email protected]>
|
|
Enabling CONFIG_DM_SEQ_ALIAS enables code which relies on
`trailing_strtol()` - which is only linked in SPL when CONFIG_SPL_STRTO
is enabled.
CONFIG_SPL_STRTO is not enabled by default - to ensure this function is
available in SPL, CONFIG_SPL_DM_SEQ_ALIAS should select
CONFIG_SPL_STRTO.
Signed-off-by: Oliver Gaskell <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Jonas Karlman <[email protected]> says:
Use of fdt_parent_offset() is very expensive as detailed by the function
documentation:
NOTE: This function is expensive, as it must scan the device tree
structure from the start to nodeoffset, *twice*.
This series remove multiple calls to fdt_parent_offset() or
ofnode_get_parent() when instead a single call can be made and the
returned value can be reused.
This series help reduce boot time by around:
- ~137ms on a Radxa ROCK Pi 4 (RK3399)
- ~33ms on a Radxa ZERO 3W (RK3566)
|
|
Until a live tree is used in U-Boot proper after relocation, use of
ofnode_get_parent() will trigger a call to the very expensive
fdt_parent_offset(), as detailed by the function documentation:
NOTE: This function is expensive, as it must scan the device tree
structure from the start to nodeoffset, *twice*.
Re-use the returned value from a single call instead of having to make
multiple calls for same node.
Signed-off-by: Jonas Karlman <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Until a live tree is used in U-Boot proper after relocation, use of
ofnode_get_parent() will trigger a call to the very expensive
fdt_parent_offset() as detailed by the function documentation:
NOTE: This function is expensive, as it must scan the device tree
structure from the start to nodeoffset, *twice*.
Re-use the returned value from a single call instead of having to make
multiple calls for same node.
Signed-off-by: Jonas Karlman <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Use of fdt_parent_offset() is very expensive as detailed by the function
documentation:
NOTE: This function is expensive, as it must scan the device tree
structure from the start to nodeoffset, *twice*.
Re-use the returned value from a single call instead of having to make
multiple calls for same nodeoffset.
Signed-off-by: Jonas Karlman <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Currently uclass index is shown in DM tree dump which ignores alias
sequence numbering. The result could be confusing since these 2 numbers
could be different. Show device sequence number instead as it's more
meaningful.
Also update documentation to use sequence number.
Signed-off-by: Zixun LI <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
The linux kernel has the list_count_nodes() API functions which is
used for counting nodes of a list. This has now been imported in
U-Boot as part of an earlier commit. Use this function and drop the
list_count_items().
Signed-off-by: Sughosh Ganu <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Ilias Apalodimas <[email protected]>
|
|
Qualcomm platforms may have drivers that bind to reserved memory nodes
(cmd-db [1] and smem [2]) which are relevant to U-Boot. Include
/reserved-memory in dm_extended_scan() so that these will be handled
correctly.
[1]: https://www.kernel.org/doc/Documentation/devicetree/bindings/reserved-memory/qcom%2Ccmd-db.yaml
[2]: https://www.kernel.org/doc/Documentation/devicetree/bindings/soc/qcom/qcom%2Csmem.yaml
Reviewed-by: Simon Glass <[email protected]>
Signed-off-by: Caleb Connolly <[email protected]>
|
|
Drop all duplicate newlines. No functional change.
Signed-off-by: Marek Vasut <[email protected]>
|
|
This part of driver model is a little subtle, so add some more comments
to promote better understanding.
Signed-off-by: Simon Glass <[email protected]>
|
|
It should read "in SPL" and not "wuth SPL".
Signed-off-by: Quentin Schulz <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Prior to that, seeing the debug() messages required to enable DM_DEBUG
which defines DEBUG (and then _DEBUG) which in turn makes failing
assert() calls reset U-Boot which isn't necessarily what is desired.
Instead, let's migrate to dm_warn which is using log_debug when unset or
log_warn when set.
While at it, reword the DM_DEBUG symbol in Kconfig to explain what it
now actually does.
Signed-off-by: Quentin Schulz <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
outp always point to an unsigned type in ofnode_read_u* functions but
the format specifier is currently always using signed type.
This is an issue since the signed type can only contain half of the
unsigned type values above 0.
However, this now breaks another usecase. Indeed,
ofnode_read_s32_default is actually passing an s32 but it'll be printed
as a u32 instead. But since the function is called u32, it makes more
sense to have it print an unsigned value.
This was discovered because arm,smc-id = <0x82000010>; on RK3588S is
above the max signed value and therefore would return a negative signed
decimal value instead of its proper unsigned one.
Fixes: fa12dfa08a7b ("dm: core: support reading a single indexed u64 value")
Fixes: 4bb7075c830c ("dm: core: support reading a single indexed u32 value")
Fixes: 7e5196c409f1 ("dm: core: Add ofnode function to read a 64-bit int")
Fixes: 9e51204527dc ("dm: core: Add operations on device tree references")
Signed-off-by: Quentin Schulz <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
A driver can have multiple compatible. When the id->compatible matches
for that driver, the first compatible supported by the driver is
currently returned, which gives the following confusing message:
- found match at 'rk3588_syscon': 'rockchip,rk3588-sys-grf' matches 'rockchip,rk3588-pmugrf'
Considering that the compatible passed in argument is necessarily the
one that exactly matched to enter this code path, there's no need to do
some elaborate logic, just print the driver name and the compatible
passed in argument.
Fixes: d3e773613b6d ("dm: core: Use U-Boot logging instead of pr_debug()")
Signed-off-by: Quentin Schulz <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Prepare v2024.07-rc5
|
|
%s/prbing/probing/
Signed-off-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Quentin Schulz <[email protected]>
Reviewed-by: Igor Opaniuk <[email protected]>
|
|
As part of bringing the master branch back in to next, we need to allow
for all of these changes to exist here.
Reported-by: Jonas Karlman <[email protected]>
Signed-off-by: Tom Rini <[email protected]>
|
|
When bringing in the series 'arm: dts: am62-beagleplay: Fix Beagleplay
Ethernet"' I failed to notice that b4 noticed it was based on next and
so took that as the base commit and merged that part of next to master.
This reverts commit c8ffd1356d42223cbb8c86280a083cc3c93e6426, reversing
changes made to 2ee6f3a5f7550de3599faef9704e166e5dcace35.
Reported-by: Jonas Karlman <[email protected]>
Signed-off-by: Tom Rini <[email protected]>
|
|
Remove <common.h> from this driver directory and when needed
add missing include files directly.
Signed-off-by: Tom Rini <[email protected]>
|
|
During address translation #address/size-cells props are expected to
apply to child nodes. However, devfdt_get_addr_index() incorrectly use
the parent offset of the udevice parent instead of getting the offset of
the node parent. This will work in most cases when there is only one
udevice tied to a node.
On e.g. an Orange Pi R1 Plus TLS board the dwc3-generic (parent) and
dwc3-generic-host (child) udevice is tied to the same node.
In that case both the offset and parent values end up being the same.
As a result, the #address/size-cells props intended for child nodes
incorrectly gets applied to the node itself resulting in wrong addr
being returned, 0x0 instead of 0xff600000.
The following can be seen on console:
dwc3-generic-host usb@ff600000: this is not a DesignWare USB3 DRD Core
dwc3-generic-host usb@ff600000: failed to initialize core
Fix this by using the offset of the parent node and not the offset to
the node the parent udevice is tied to.
Signed-off-by: Jonas Karlman <[email protected]>
|
|
Check the header before starting to use it, since this could provide
very confusing later, when ofnode calls start to fail.
Signed-off-by: Simon Glass <[email protected]>
|
|
Prepare v2024.01-rc3
|
|
This does not existing anymore. Update SPL_DM_STATS to use the correct
Kconfig option, which is SPL_DM
Signed-off-by: Simon Glass <[email protected]>
|
|
CONFIG_DM_WARN has a text indicating that these messages should only
provided when debugging. This implies that the setting must be default no.
We should still create debug messages.
Reported-by: Andre Przywara <[email protected]>
Signed-off-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
The baudrate configured in .config is taken by default by serial. If
change of baudrate is required then the .config needs to changed and
u-boot recompilation is required or the u-boot environment needs to be
updated.
To avoid this, support is added to fetch the baudrate directly from the
device tree file and update.
The serial, prints the log with the configured baudrate in the dtb.
The commit c4df0f6f315c ("arm: mvebu: Espressobin: Set default value for
$fdtfile env variable") is taken as reference for changing the default
environment variable.
The default environment stores the default baudrate value, When default
baudrate and dtb baudrate are not same glitches are seen on the serial.
So, the environment also needs to be updated with the dtb baudrate to
avoid the glitches on the serial.
Also add test to cover this new function.
Signed-off-by: Algapally Santosh Sagar <[email protected]>
Signed-off-by: Venkatesh Yadav Abbarapu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Michal Simek <[email protected]>
|
|
lib/acpi/acpigen.o is only compiled into SPL when SPL_ACPIGEN is enabled.
Update several files which reference these functions accordingly.
Signed-off-by: Sean Anderson <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Sandbox unit tests in U-Boot proper load a test device tree to have some
devices to work with. In order to do the same in SPL, we must enable
SPL_OF_REAL. However, we already have SPL_OF_PLATDATA enabled. When
generating platdata from a devicetree, it is expected that we will not need
devicetree access functions (even though SPL_OF_CONTROL is enabled). This
expectation does not hold for sandbox, so allow user control of
SPL_OF_REAL.
There are several places in the tree where conditions involving OF_PLATDATA
or OF_REAL no longer function correctly when both of these options can be
selected at the same time. Adjust these conditions accordingly.
Signed-off-by: Sean Anderson <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Same as dev_read_addr_name[_size](), but returns a pointer, cast
through map_sysmem().
Signed-off-by: Matthias Schiffer <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Checking for the error cast to fdt_addr_t is rather awkward - IS_ERR()
can be used, but it's not really made to be used on fdt_addr_t, which
may not even be the same size as a native pointer.
Most places in U-Boot only check for FDT_ADDR_T_NONE; let's adjust the
error return to match the expectation.
Signed-off-by: Matthias Schiffer <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
The PCI helpers read only the base address for a PCI region. In some cases
the size is needed as well, e.g. to pass along to a driver which needs to
know the size of its register area.
Update the functions to allow the size to be returned. For serial, record
the information and provided it with the serial_info() call.
A limitation still exists in that the size is not available when OF_LIVE
is enabled, so take account of that in the tests.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add support for writing a single 64-bit value into a property.
Repurpose the existing tests to handle this case too.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add functions to write a boolean property. This involves deleting it if
the value is false.
Add a new ofnode_has_property() as well. Add a comment about the behaviour
of of_read_property() when the property value is empty.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add a way to flatten a devicetree into binary form. For livetree this
involves generating the devicetree using fdt_property() and other calls.
For flattree it simply involves providing the buffer containing the tree.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add a function to delete a node in an existing tree.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add a function to copy a node to another place under a new name. This is
useful at least for testing, since copying a test node with existing
properties is easier than writing the code to generate it all afresh.
Signed-off-by: Simon Glass <[email protected]>
|
|
Provide a function to create a new, empty tree.
Signed-off-by: Simon Glass <[email protected]>
|
|
Follow the order used by memcpy() as it may be less confusing.
Signed-off-by: Simon Glass <[email protected]>
|
|
There are copying errors in the help. Fix these.
Signed-off-by: Simon Glass <[email protected]>
|
