Andrew Goodbody [Wed, 23 Jul 2025 15:54:07 +0000 (16:54 +0100)]
clk: versaclock: return value calculated by ERR_PTR
In versaclock_get_name -ENOMEM is passed to ERR_PTR but nothing is
done with the value that this calculates which is obviously not the
intention of the code. This is confirmed by the code around where this
function is called.
Instead return the value from ERR_PTR.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Andrew Goodbody [Tue, 30 Sep 2025 15:52:24 +0000 (16:52 +0100)]
usb: musb-new: Cannot test unsigned member to be negative
You cannot test an unsigned member of a struct for being negative, the
test will always fail. Instead assign the return value of
fdtdec_get_int, which returns an int, to a temporary variable declared
as an int, so that it can be tested for being negative before being
assigned to the unsigned struct member.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Andrew Goodbody [Tue, 30 Sep 2025 15:52:23 +0000 (16:52 +0100)]
usb: musb-new: Limit check array index before use
epnum is used as an index into an array. The limit check for this index
should be performed before using it to access an element in the array to
prevent possible bounds overrun.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org> Reviewed-by: Mattijs Korpershoek <mkorpershoek@kernel.org>
Andrew Goodbody [Tue, 30 Sep 2025 15:52:22 +0000 (16:52 +0100)]
usb: musb-new: Null check before dereference
A null check for the variable 'data' was introduced before dereferencing
it for set_phy_power but other uses were not so protected. Add the null
check for other dereferences of 'data'.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Andrew Goodbody [Tue, 30 Sep 2025 10:34:44 +0000 (11:34 +0100)]
usb: xhci: exynos: variable node should be signed
The variable node is assigned to the return value of a function that
returns an int. It is tested for being negative and then passed as an
argument to a function that takes an int. So 'node' should not be
declared as unsigned. Correct it.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Andrew Goodbody [Tue, 30 Sep 2025 16:06:44 +0000 (17:06 +0100)]
usb: ulpi: Incorrect operator used
Combining two bits into a mask to be used so that the same write code
can be used to set or reset bits in a register clearly needs to use the
binary 'or' operator, not the binary 'and'. Fix it.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Andrew Goodbody [Tue, 30 Sep 2025 09:56:02 +0000 (10:56 +0100)]
usb: ohci-hcd: Null check lurb_priv before dereference
When a variable needs a null check before it is dreferenced ensure that
this is done even in the case of assignment on declaration. This was not
happening for lurb_priv so correct it.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Andrew Goodbody [Thu, 25 Sep 2025 11:56:44 +0000 (12:56 +0100)]
usb: fsl-dt-fixup: Return an error code on error
fsl_fdt_fixup_usb_erratum uses strcmp to detect an error but then
returns 'err' without it being set to an error. Calling code may not
detect that an error occurred leading to a silent failure. Instead just
return -EINVAL.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Andrew Goodbody [Mon, 29 Sep 2025 16:53:11 +0000 (17:53 +0100)]
usb: ehci: exynos: variable node should be signed
THe variable node is assigned to the return value of a function that
returns an int. It is tested for being negative and then passed as an
argument to a function that takes an int. So 'node' should not be
declared as unsigned. Correct it.
This issue was found by Smatch.
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org> Reviewed-by: Marek Vasut <marek.vasut@mailbox.org>
E Shattow [Tue, 21 Oct 2025 23:09:59 +0000 (16:09 -0700)]
configs: starfive: Add Milk-V Mars CM and Mars CM Lite to visionfive2
Add Milk-V Mars CM and Mars CM Lite to visionfive2.
These boards were previously supported and then removed in the transition
to OF_UPSTREAM. The dts have landed in the for-next queue so let's add the
boards again.
Signed-off-by: E Shattow <e@freeshell.de> Reviewed-by: Hal Feng <hal.feng@starfivetech.com>
E Shattow [Wed, 15 Oct 2025 10:22:53 +0000 (03:22 -0700)]
riscv: dts: starfive: add Milk-V Mars CM Lite system-on-module
Milk-V Mars CM Lite is a System-on-Module based on the Milk-V Mars CM
without the onboard eMMC storage component populated and configured
instead for SD3.0 Card Slot on that interface via 100-pin connector.
Link to Milk-V Mars CM Lite schematics: https://github.com/milkv-mars/mars-files/tree/main/Mars-CM_Hardware_Schematices
Link to StarFive JH7110 Technical Reference Manual: https://doc-en.rvspace.org/JH7110/TRM/index.html
Link to Raspberry Pi CM4IO datasheet: https://datasheets.raspberrypi.com/cm4io/cm4io-datasheet.pdf
Add the devicetree file to make use of StarFive JH7110 common supported
features PMIC, EEPROM, UART, I2C, GPIO, PCIe, QSPI Flash, PWM, and
Ethernet. Also configure the eMMC interface mmc0 for SD Card use and
configure the common SD Card interface mmc1 for onboard SDIO BT+WiFi.
E Shattow [Wed, 15 Oct 2025 10:22:51 +0000 (03:22 -0700)]
riscv: dts: starfive: add Milk-V Mars CM system-on-module
Milk-V Mars CM is a System-on-Module based on the StarFive VisionFive 2
board and Radxa CM3 System-on-Module compatible with the Raspberry Pi
CM4IO Classic IO Board.
Mars CM SoM features:
- StarFive JH7110 System on Chip with RV64GC up to 1.5GHz
- AXP15060 Power Management Unit
- LPDDR4 2GB / 4GB / 8GB DRAM memory
- BL24C04F 4K bits (512 x 8) EEPROM
- GigaDevice 25LQ128EWIG QSPI NOR Flash 16M or SoC ROM UART loader for
boot (selectable by GPIO)
- eMMC5.0 8GB / 16GB / 32GB flash storage onboard
- AP6256 via SDIO 2.0 onboard wireless connectivity WiFi 5 + Bluetooth
5.2 (optional, present in models with WiFi feature)
- 1x Motorcomm YT8531C Gigabit Ethernet PHY
- IMG BXE-4-32 Integrated GPU with 3D Acceleration:
- H.264 & H.265 4K@60fps Decoding
- H.265 1080p@30fps Encoding
- JPEG encoder / decoder
Additional features available via 2x 100-pin connectors for CM4IO Board:
- 1x HDMI 2.0
- 1x MIPI DSI (4-lanes)
- 1x 2CH Audio out (via GPIO)
- 1x MIPI CSI (2x2-lanes or 1x4-lanes)
- 1x USB 2.0
- 1x PCIe 1-lane Host, Gen 2 (5Gbps)
- Up to 28x GPIO, supporting 3.3V
- UART x6
- PWM x8
- I2C x7
- SPI
- I2S
Link to Milk-V Mars CM schematics: https://github.com/milkv-mars/mars-files/tree/main/Mars-CM_Hardware_Schematices
Link to StarFive JH7110 Technical Reference Manual: https://doc-en.rvspace.org/JH7110/TRM/index.html
Link to Raspberry Pi CM4IO datasheet: https://datasheets.raspberrypi.com/cm4io/cm4io-datasheet.pdf
Add the devicetree file to make use of StarFive JH7110 common supported
features PMIC, EEPROM, UART, I2C, GPIO, eMMC, PCIe, QSPI Flash, PWM, and
Ethernet. Also configure the common SD Card interface mmc1 for onboard
SDIO BT+WiFi.
E Shattow [Wed, 15 Oct 2025 10:22:48 +0000 (03:22 -0700)]
riscv: dts: starfive: jh7110-common: drop mmc post-power-on-delay-ms
Drop post-power-on-delay-ms from mmc0 mmc1 interfaces. There is no
known reason for these properties to continue, testing appears to be fine
without them [1].
E Shattow [Wed, 15 Oct 2025 10:22:47 +0000 (03:22 -0700)]
riscv: dts: starfive: jh7110-common: drop no-mmc property from mmc1
Relax no-mmc restriction on mmc1 for jh7110 boards. The restriction is
only needed to block use of commands that would cause a device to
malfunction, which by testing and observation [1] is not any problem.
E Shattow [Wed, 15 Oct 2025 10:22:44 +0000 (03:22 -0700)]
riscv: dts: starfive: jh7110-common: drop no-sdio property from mmc1
Relax no-sdio restriction on mmc1 for jh7110 boards. Property was
introduced for StarFive VisionFive2 dts to configure mmc1 for SD Card
but this is not necessary, the restriction is only needed to block use of
commands that would cause a device to malfunction.
Jimmy Ho [Mon, 13 Oct 2025 02:24:25 +0000 (10:24 +0800)]
timer: sifive_clint: Add GHRTv2 compaible string
The current sifive_clint driver can fully support GHRTv2 clint.
Add the compatible of GHRTv2 clint, sifive,clint2, to sifive_clint_ids
list.
Signed-off-by: Vincent Chen <vincent.chen@sifive.com> Signed-off-by: Wayling Chen <wayling.chen@sifive.com> Signed-off-by: Max Hsu <max.hsu@sifive.com> Signed-off-by: Jimmy Ho <jimmy.ho@sifive.com>
No big features this time, mostly a collection of patches that have been
lying around for a bit. There are some DT updates, for those SoCs that
do not use OF_UPSTREAM yet, hopefully that's the last time we need to do
this exercise. And that's offset by switching over two more SoCs to
OF_UPSTREAM. Two new boards get a defconfig, and some improvements for
the sun8i-emac Ethernet driver. Finally a patch that fixes occassional
DRAM size misdetection for new A523 boards.
There are a few outstanding patches that just wait for getting some
details confirmed, which I might send then later.
CI passed, and I tested this briefly on affected boards.
Andre Przywara [Fri, 24 Oct 2025 00:30:00 +0000 (01:30 +0100)]
sunxi: dts: arm: update devicetree files from Linux kernel tree
Sync the kernel devicetree source files for the Allwinner SoCs with
32-bit cores that do not use OF_UPSTREAM yet. The files were taken
from a v6.18-rc1 tree.
To maintain ABI compatibility with existing LTS kernels, one change
moving some IP blocks to the r_intc interrupt controller is excluded.
This effectively reverts Linux commits 994e5818392c and 9fdef3c3d8c2.
This commit also adds a new board devicetree for the A33 Vstar board,
plus one DT overlay for the OrangePi Zero interface board.
This update should not impact any existing U-Boot functionality.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Andre Przywara [Fri, 24 Oct 2025 00:08:38 +0000 (01:08 +0100)]
sunxi: dts: arm64: update devicetree files from Linux kernel tree
Sync the devicetree files from the official Linux kernel tree, v6.18-rc1.
This is covering Allwinner SoCs with 64-bit ARM cores.
The bulk is cosmetic changes: board model name changes, DT node renames,
whitespace fixes.
The actual changes are not dramatic: the CPU cores get their caches
described properly, some A64 video clocks get fixed, some A64 boards
describe the header pins for the WiFi module, the Pinephone adds an
alternative magnetometer used on some board revisions.
On the H5 side the microSD slots get marked as having no write-protect
detection, and the NanoPi Neo Plus2 board describes its regulators better.
The H6 boards switch from RSB to I2C for their PMIC connection.
As before, this omits the non-backwards compatible changes to the R_INTC
controller, to remain compatible with older kernels.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Andre Przywara [Sat, 25 Oct 2025 18:03:07 +0000 (19:03 +0100)]
sunxi: switch Allwinner A80 boards to OF_UPSTREAM
In contrast to some other Allwinner SoCs, there is no difference between
the DTs for the Allwinner A80 SoCs (sun9i) between the U-Boot and the
Linux kernel repository.
Remove the old copies of the A80 related .dts and .dtsi files, and switch
most of sun9i boards over to use OF_UPSTREAM.
There is the Sunchip CX-A99 (family of) board(s) for which we don't have
a DT in the kernel tree. Keep the .dts file in the legacy U-Boot DT
directory, and let the board's defconfig opt out of OF_UPSTREAM.
Signed-off-by: Andre Przywara <andre.przywara@arm.com> Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Andre Przywara [Sat, 25 Oct 2025 17:57:27 +0000 (18:57 +0100)]
sunxi: switch the Allwinner T113 SoC to OF_UPSTREAM
In contrast to some other Allwinner SoCs, there is no difference between
the DTs for the Allwinner T113-s3 SoC (sun20i) between the U-Boot and the
Linux kernel repository.
Remove the old copies of the T113-s3 related .dts and .dtsi files, and
switch the whole SoC (represented by just one board) over to use
OF_UPSTREAM.
Signed-off-by: Andre Przywara <andre.przywara@arm.com> Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
net: sun8i-emac: Add support for active-low leds with internal PHY
A device-tree property is already defined to indicate that the internal
PHY should be used with active-low leds, which corresponds to a
specific bit in the dedicated syscon register.
Add support for setting this bit when the property is present.
Signed-off-by: Paul Kocialkowski <contact@paulk.fr> Reviewed-by: Andre Przywara <andre.przywara@arm.com>
James Hilliard [Fri, 11 Apr 2025 01:10:25 +0000 (19:10 -0600)]
Makefile: pass KEYDIR when set to sunxi-spl.bin mkimage
Currently we pass this for u-boot-spl.kwb targets, however when
building sunxi-spl.bin in the TOC0 format we may also need to
specify a KEYDIR, as such we should also pass this when set
to mkimage for the sunxi-spl.bin target.
Signed-off-by: James Hilliard <james.hilliard1@gmail.com> Reviewed-by: Andre Przywara <andre.przywara@arm.com>
Andre Przywara [Mon, 20 Oct 2025 14:26:22 +0000 (15:26 +0100)]
sunxi: a523: Re-use H6/H616 DRAM size detection method
The H6 and H616 DRAM initialisation code recently gained a joint and
improved size detection routine, which helped to avoid the occasional
DRAM size misdetection.
Use the same code for the A523 DRAM initialisation as well, which
suffers from the same misdetection at times.
Chen-Yu Tsai [Mon, 20 Oct 2025 16:19:30 +0000 (00:19 +0800)]
sunxi: a523: Add OrangePi 4A defconfig
The OrangePi 4A is a typical Raspberry Pi model B sized development
board from Xunlong designed around an Allwinner T527 SoC.
The board has the following features:
- Allwinner T527 SoC
- AXP717B + AXP323 PMICs
- Up to 4GB LPDDR4 DRAM
- micro SD slot
- optional eMMC module
- M.2 slot for PCIe 2.0 x1
- 16 MB SPI-NOR flash
- 4x USB 2.0 type-A ports (one can be used in gadget mode)
- 1x Gigabit ethernet w/ Motorcomm PHY (through yet to be supported GMAC200)
- 3.5mm audio jack via internal audio codec
- HDMI 2.0 output
- eDP, MIPI CSI (2-lane and 4-lane) and MIPI DSI (4-lane) connectors
- USB type-C port purely for power
- AP6256 (Broadcom BCM4345) WiFi 5.0 + BT 5.0
- unsoldered headers for ADC and an additional USB 2.0 host port
- 40-pin GPIO header
Add defconfig for it.
Signed-off-by: Chen-Yu Tsai <wens@kernel.org> Reviewed-by: Andre Przywara <andre.przywara@arm.com>
* develop: virtio: Fix qemu example (true/false -> on/off)
* separate read and write command documentation
* usage: Add general rule for `$?`
* askenv: Reword and remove return value
* seama: Reword return value section
* usage: Use glob for all commands
* Fix typos and formatting
Tom Rini [Fri, 24 Oct 2025 16:02:59 +0000 (10:02 -0600)]
doc: seama: Reword return value section
With the addition of general text about how the return value is handled,
reference that while retaining the additional information about setting
$seama_image_size
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de> Signed-off-by: Tom Rini <trini@konsulko.com>
Tom Rini [Fri, 24 Oct 2025 16:02:58 +0000 (10:02 -0600)]
doc: askenv: Reword and remove return value
With the addition of general text about how the return value is handled,
remove the examples from the askenv documentation as they are all normal
expected results.
Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
Tom Rini [Fri, 24 Oct 2025 16:02:57 +0000 (10:02 -0600)]
doc: usage: Add general rule for `$?`
For nearly all commands in U-Boot the '?' variable is handled the same
way with 0 meaning success, 1 meaning any failure. Explain this in the
general rules section of the cmdline documentation (with a link to a
counter example) and then remove the redundant wording from most
commands. We retain a section about the return value in a number of
places where we are doing something such as always returning a specific
value or we have useful additional information to go along with the
normal return codes.
Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Wolfgang Wallner [Fri, 24 Oct 2025 15:11:58 +0000 (17:11 +0200)]
doc: develop: Change formatting to make binman doc more readable
Change the formatting of binman.rst so that the compiled HTML output
becomes more readable. Changes include enumerations and the escaping of
arguments starting with a double dash (e.g. --debug).
Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Wolfgang Wallner <wolfgang.wallner@br-automation.com>
Casey Connolly [Thu, 23 Oct 2025 14:26:46 +0000 (16:26 +0200)]
efi_loader: efi_console: support editable input fields
When editing eficonfig "optional data" (typically cmdline arguments)
it's useful to be able to edit the string rather than having to re-type
the entire thing. Implement support for editing buffers to make this a
whole lot nicer. Specifically, add support for moving the cursor with
the arrow keys and End key as well as deleting backwards with the delete
key.
doc: separate read and write command documentation
* Avoid two step navigation to get to the description of the write command.
* Add missing index entries
* Correct formatting of the examples.
* Describe configuration and return value.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Benjamin Hahn [Tue, 21 Oct 2025 14:34:17 +0000 (16:34 +0200)]
bootstd: make it possible to use tftp for netboot with standardboot
Add the option to load the bootscript with the tftp command (static IP)
instead of the dhcp command (dynamic IP). For this a new function
tftpb_run similar to dhcp_run, is needed. The selection of which command
to use can be done with the ip_dyn environment variable, which can be
set to yes or no. The ip_dyn variable was chosen as it is already in use
on the imx platforms.
Also edit the bootstd doc.
Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Benjamin Hahn <B.Hahn@phytec.de>
Peng Fan [Fri, 17 Oct 2025 09:32:34 +0000 (17:32 +0800)]
firmware: scmi: Add i.MX95 SCMI CPU Protocol
This protocol allows an agent to start, stop a CPU or set reset vector.
It is used to manage auxiliary CPUs in an LM (e.g. additional cores in an
AP cluster).
Signed-off-by: Peng Fan <peng.fan@nxp.com> Reviewed-by: Alice Guo <alice.guo@nxp.com>
phy: ti: phy-j721e-wiz: Allow reinitialization when SERDES is pre-configured
Move the SERDES configuration check after clock and reset initialization
and change it from a hard failure to a skip of WIZ initialization. This
allows the driver to probe successfully when the SERDES has been
pre-configured by a previous boot stage (e.g., ROM or SPL).
This approach aligns with how the Linux kernel handles pre-configured
SERDES, where the driver gracefully skips reinitialization rather than
failing to probe.
Add Logical Machine Management(LMM) protocol which is intended for boot,
shutdown, and reset of other logical machines (LM). It is usually used to
allow one LM to manager another used as an offload or accelerator engine.
Following Linux Kernel, created a separate folder for holding vendor
protocol drivers.
Signed-off-by: Peng Fan <peng.fan@nxp.com> Reviewed-by: Alice Guo <alice.guo@nxp.com>
Peng Fan [Fri, 17 Oct 2025 09:32:32 +0000 (17:32 +0800)]
firmware: scmi: Support probe vendor ID 0x80 and 0x82
Preparing to add i.MX LMM and CPU protocol driver, support probe SCMI
vendor ID 0x80(i.MX SCMI LMM ID) and 0x82(i.MX SCMI CPU ID). And use
Kconfig option to support conditional compilation.
Signed-off-by: Peng Fan <peng.fan@nxp.com> Reviewed-by: Alice Guo <alice.guo@nxp.com>
Tom Rini [Tue, 21 Oct 2025 20:37:01 +0000 (14:37 -0600)]
env: fat, ubi: Fix gd->env_valid for the first write
As resolved and explained in detail in commit e589d5822cac ("env: spi:
Fix gd->env_valid for the first write") and archived discussion there is
a corner case where we don't do the right thing with redundant
environments. This same exact check was present in the mmc code and
resolved with commit 813a0df27a8a ("env: Invert gd->env_valid for
env_mmc_save") and in the discussion of that patch, I noted that both
fat and ubi (and at the time, sf) were doing the same thing. Take the
time now to correct fat and ubi environment.
Peng Fan [Fri, 17 Oct 2025 09:32:31 +0000 (17:32 +0800)]
firmware: scmi: Cleanup the SCMI MISC ID
SCMI_IMX_PROTOCOL_ID_MISC was never used, so drop it. And move
SCMI_PROTOCOL_ID_IMX_MISC out of enum scmi_std_protocol to
scmi_nxp_protocols.h, because it is i.MX specific and following Linux
Kernel style, use macro definition.
Signed-off-by: Peng Fan <peng.fan@nxp.com> Reviewed-by: Alice Guo <alice.guo@nxp.com>
Peng Fan [Fri, 17 Oct 2025 09:32:30 +0000 (17:32 +0800)]
firmware: scmi: Conditionally compile protocol support
Add conditional compilation for SCMI protocol support in scmi_get_protocol()
and scmi_add_protocol() based on corresponding Kconfig options. This ensures
that only the enabled protocols are compiled and accessed, and reducing binary
size.
Signed-off-by: Peng Fan <peng.fan@nxp.com> Reviewed-by: Alice Guo <alice.guo@nxp.com>
Ilias Apalodimas [Wed, 22 Oct 2025 12:23:15 +0000 (15:23 +0300)]
makefile: Fix recursive makes
Since the Kbuild bump to 5.1 and specifically
commit af1a9935708a ("kbuild: make -r/-R effective in top Makefile for old Make versions")
the recursion rules have changed.
'make O=/output/dir/' as well as './test/py/test.py --bd sandbox --build'
is working fine but anything that recursively called the makefile was
failing e.g 'make tests', 'make qcheck' etc, which calls a bash script
that ends up calling the makefile again.
The reason is that the internal 'sub_make_done' variable of the makefile
was set after the first pass and the output dir was never evaluated
properly. Reset the variable value if we are executing any of these
tests.
Fixes: af1a9935708a ("kbuild: make -r/-R effective in top Makefile for old Make versions") Reported-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org> Test-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Tom Rini [Wed, 22 Oct 2025 20:17:16 +0000 (14:17 -0600)]
Merge patch series "boot: Support priority for global bootmeths"
Simon Glass <sjg@chromium.org> says:
At present global bootmeths always run first, before all other
bootmeths. Optimisations in the code take advantage of this, putting
them at the end, so they can be used once and then forgotten.
In some cases it is useful to run global bootmeths later in the boot.
For example, the EFI-bootmgr bootmeth may itself scan devices and the
network, so running it first can hold up the boot significantly for
boards not actually relying on EFI-bootmgr to boot.
This series introduces a new field in global bootmeths which indicates
the priority, using the same scheme as is used with bootdev hunters.
Thus it is possible to insert the EFI-bootmgr bootmeth just before the
hunter for network bootdevs is invoked.
Despite the simplicity of the concept and the relatively small series,
this is a fairly significant enhancement. It is also quite tricky to
implement, largely due to the way the original code was written, with
global bootmeths being a small, size-optimised add-on to the original
bootstd implementation.
For now we only allow each global bootmeth to run at most once, but this
implementation is written in a way that we could relax that if needed.
Then the bootmeth itself could decide whether to run at any particular
point in the bootflow iteration.
Simon Glass [Wed, 15 Oct 2025 15:44:14 +0000 (16:44 +0100)]
boot: Run the EFI bootmgr just before network devices
At present the EFI bootmgr scans all devices in the system before
deciding which one to boot. Ideally it would use the bootstd iterator
for this, but in the meantime, give it a lower priority, so it runs
just before the network devices.
Note that if there are no hunted network devices hunted, then it will
run at the end, after all bootdevs are exhausted. In other words, it
will always run.
Simon Glass [Wed, 15 Oct 2025 15:44:13 +0000 (16:44 +0100)]
boot: Run global bootmeths after all bootdevs are exhausted
When there are no more bootdevs we should still go through the global
bootmeths, since some may not have yet been used, if their priority has
not yet come up.
Add a final check for this at the end of the iterator.
Update the documentation to match the new behaviour of global bootmeths.
Simon Glass [Wed, 15 Oct 2025 15:44:12 +0000 (16:44 +0100)]
boot: Don't change the method count after global bootmeths
At present before scanning global bootmeths, the iterator sets the
method count to the index of the first global bootmeth. Now that we
support scanning the global bootmeths multiple times, we must leave this
count alone.
Check against have_global and first_glob_method instead.
Simon Glass [Wed, 15 Oct 2025 15:44:11 +0000 (16:44 +0100)]
boot: Implement a priority for global bootmeths
Allow bootmeths to select when they want to run, using the bootdev
priority. Provide a new bootmeth_glob_allowed() function which checks if
a bootmeth is ready to use.
Fix a comment in bootflow_system() which is a test for global bootmeths.
Simon Glass [Wed, 15 Oct 2025 15:44:10 +0000 (16:44 +0100)]
boot: Only run global bootmeths once each
Use the methods_done flags to make sure that each global bootmeth is
only used once. For now this has no effect, since they are all processed
at the start.
Simon Glass [Wed, 15 Oct 2025 15:44:09 +0000 (16:44 +0100)]
boot: Support rescanning the global bootmeths
Add the logic to scan through the global bootmeths for every new
bootdev, in preparation for allowing global bootmeths to select where in
the hunter ordering they go.
Use a new bootmeth_glob_allowed() function to check if a bootmeth is
allowed, ensuring that each can run at most once.
For now this has no actual effect, since the global bootmeths are
unconditionally processed at the start, with iter->methods_done being
updated to include all of them. Therefore when scanning again, no
unprocessed global bootmeths will be found.
Simon Glass [Wed, 15 Oct 2025 15:44:07 +0000 (16:44 +0100)]
boot: Add a flag for whether there are global bootmeths
The current 'doing_global' refers to being in the state of processing
global bootmeths. Since global bootmeths are currently used once at the
start, it becomes false once the last global bootmeth has been used.
In preparation for allowing bootmeths to run at other points in the
bootstd interation, add a new 'have_global' flag which tracks whether
there are any global bootmeths in the method_order[] list. It is set up
when iteration starts. Unlike doing_global which resets back to false
after the global bootmeths have been handled, once have_global is set to
true, it remains true for the entire iteration process. This provides a
quick check as to whether global-bootmeth processing is needed.
Simon Glass [Wed, 15 Oct 2025 15:44:06 +0000 (16:44 +0100)]
boot: Update first_glob_method when dropping a bootmeth
For now we only support dropping non-global bootmeths from the
iteration. Update first_glob_method in that case and add a few checks
that things are correct.
Simon Glass [Wed, 15 Oct 2025 15:44:04 +0000 (16:44 +0100)]
boot: Try all bootmeths on the final partition
At present, normally when one bootmeth fails on a partition, we move on
and try the next bootmeth. However, this was not the case for the final
partition due to a bug. Rework the logic so that all partitions are
treated the same.
Tom Rini [Wed, 22 Oct 2025 18:23:48 +0000 (12:23 -0600)]
Merge patch series "Add support for dynamic MMU configuration"
Anshul Dalal <anshuld@ti.com> says:
In U-Boot, TI only provides a single memory map for all k3 platforms, this
does not scale for devices where atf and optee lie outside the range 0x80000000
- 0x80080000 and 0x9e780000 - 0xa0000000 respectively.
There are also issues for devices with < 2GiB of memory (eg am62SiP with 512MiB
of RAM) as the maximum size for the first DRAM bank is hardcoded to 2GiB in the
current memory map. Furthermore the second DRAM bank is mapped even for devices
that only have a single bank.
Therefore this patch set adds the required functionality to create the MMU table
at runtime based on the device-tree.
The patch set has been build tested on all effected platforms but boot-tested
only on TI's K3 EVMs, the beagleplay and phytec's phycore-am6* platforms.
The following effected boards have not been boot tested:
- verdin-am62
- iot2050
Anshul Dalal [Fri, 17 Oct 2025 13:15:32 +0000 (18:45 +0530)]
mach-k3: add carveouts for TFA and optee
K3 platforms have reserved memory regions for TFA and OPTEE which should
be unmapped for U-Boot. While other "no-map" memory regions like the
memory pools for remote cores should not be unmapped to allow U-Boot to
load firmware during remoteproc.
Therefore this patch adds the necessary fdt fixups to properly set the
load address for TFA/OPTEE and unmaps both by mmu_unmap_reserved_mem.
Anshul Dalal [Fri, 17 Oct 2025 13:15:31 +0000 (18:45 +0530)]
mach-k3: add reserved memory fixups for next boot stage
The boot flow in K3 platforms requires the subsequent boot stages to be
aware of the memory reserved for previous boot binaries. In the regular
boot sequence of:
Both A-core SPL and U-Boot should be made aware of the memory reserved
for ATF and OP-TEE from their device-tree. Currently this information is
absent.
Therefore this patch adds the reserved-memory regions as part of
spl_perform_arch_fixups for the next stage's DT. This is called during
both R5 and A-core SPL which fixes up the DT for their respective next
stages: A-core SPL and U-Boot proper.
Anshul Dalal [Fri, 17 Oct 2025 13:15:30 +0000 (18:45 +0530)]
spl: split spl_board_fixups to arch/board specific
The current spl_board_fixups API allows for modification of spl_image
before the SPL jumps to it. This can be used to modify the DT for the
next boot stage, however the current API only allows either the machine
arch or the board to use it.
This limits the utility of the API as there might be certain fixups that
should be applied to all boards sharing the same machine architecture
with others being board specific.
For TI's K3 specifically, this prevents us from performing architecture
level fixups since a lot of TI boards are already making use of the
spl_board_fixups API.
Therefore this patch splits the API into two to allow both board and the
architecture specific fixups. The order is kept as arch then board to
give board specific fixups the precedence.
Anshul Dalal [Fri, 17 Oct 2025 13:15:29 +0000 (18:45 +0530)]
arm: armv8: mmu: add mmu_unmap_reserved_mem
For armv8, U-Boot uses a static map defined as 'mem_map' for configuring
the MMU's page tables, done by mmu_setup.
Though this works well for simpler platforms, it makes creating runtime
carveouts by modifying the static array at runtime exceedingly complex
like in mach-snapdragon/board.c.
Creation of such carveouts are much better handled by APIs such as
mmu_change_region_attr once the page tables are configured. Usually such
carveouts are configured via the device-tree's reserved-memory node
which provides the address and size for the carveout.
Therefore this patch adds mmu_unmap_reserved_mem which acts as a wrapper
over mmu_change_region_attr, helping unmap a reserved-memory region.
Anshul Dalal [Fri, 17 Oct 2025 13:15:27 +0000 (18:45 +0530)]
arm: armv8: mmu: add mem_map_from_dram_banks
For armv8, U-Boot uses a static map defined as 'mem_map' for configuring
the MMU as part of mmu_setup.
But since the exact configuration of memory banks might not be known at
build time, many platforms such as imx9, versal2 etc. utilize
gd->bd->bi_dram to configure the static map at runtime.
Therefore this patch adds a new API mem_map_from_dram_banks that
modifies the static map in a similar way. Allowing the caller to map all
dram banks by just passing the index to last entry in their mem_map and
it's length.
Anshul Dalal [Fri, 17 Oct 2025 13:15:26 +0000 (18:45 +0530)]
arm: armv8: invalidate dcache entries on dcache_enable
In dcache_enable, currently the dcache entries are only invalidated when
the MMU is not enabled. This causes issues when dcache_enable is called
with the MMU already configured, in such cases the existing dcache
entries are not flushed which might result in un-expected behavior.
This patch invalidates the cache entries on every call of dcache_enable
before enabling dcache (by setting CR_C). This makes dcache_enable
behave similar to icache_enable as well.
Anshul Dalal [Fri, 17 Oct 2025 13:15:25 +0000 (18:45 +0530)]
arm: armv8: mmu: export mmu_setup
The mmu_setup function configures the page tables based on the board
supplied mem_map struct array. It is called implicitly as part of
dcache_enable but this limits us to only be able to use APIs such as
mmu_change_region_attr only after caches are enabled.
This might lead to speculative accesses before we can unmap a region
that is marked as cacheable in the static memory map.
Therefore this patch exports the mmu_setup function in mmu.h allowing
users to have more control over when the mmu is configured.
For K3 specifically this allows for the following configuration sequence
as part of enable_caches:
Anshul Dalal [Fri, 17 Oct 2025 13:15:24 +0000 (18:45 +0530)]
mach-k3: use custom enable_cache
U-Boot's provided enable_caches enforces the following sequence:
icache_enable -> mmu_setup (as part of dcache_enable) -> dcache_enable
Whereas for K3 devices, we would like to add entries to the provided
static array (k3_mem_map) as per gd->bd->bi_dram and then call mmu_setup
to configure the MMU but also create carveouts for TFA/TEE before we
enable caches to prevent speculative accesses to the region. Thus the
following desired sequence:
Anshul Dalal [Fri, 17 Oct 2025 13:15:23 +0000 (18:45 +0530)]
mach-k3: use minimal memory map for all K3
The K3 family of SoCs encompasses a wide variety of devices with varying
DDR configurations and memory carveout requirements, the current static
memory map provides basic support for TI EVMs but does not scale well
for newer platforms (such as AM62SiP with 512MiB of RAM).
Therefore this patch replaces the existing memory map with a minimal
one, that could be more easily modified at runtime.
net:
- airoha: improvements
- Tighten a few more driver dependencies
- designware: fix bitbang init error
- phy: Make driver overloading get_phy_id depend on !COMPILE_TEST
- phy: add paged PHY register accessors
- make dhcp_run() common for NET and NET_LWIP
- dwc_eth_ops: Correct check for FDT_64BIT
- mediatek: mt7988: various fixup + MDIO detach
- phy: aquantia: switch to use phy_get_ofnode(), fix bindings typo
net-legacy:
- bootp: Prevent buffer overflow to avoid leaking the RAM content
- tftp: make TFTP ports unconditionally configurable
misc:
- uthreads: Make use of CONFIG_IS_ENABLED consistently
projects / thirdparty / u-boot.git / log
