| Age | Commit message (Collapse) | Author |
|---|
|
There currently are multiple allocation API's in the LMB module. There
are a couple of API's for allocating memory(lmb_alloc() and
lmb_alloc_base()), and then there are two for requesting a reservation
for a particular memory region (lmb_reserve() and
lmb_alloc_addr()). Introduce a single API lmb_alloc_mem() which will
cater to all types of allocation requests and replace lmb_reserve()
and lmb_alloc_addr() with the new API.
Moreover, the lmb_reserve() API is pretty similar to the
lmb_alloc_addr() API, with the one difference being that the
lmb_reserve() API allows for reserving any address passed to it --
the address need not be part of the LMB memory map. The
lmb_alloc_addr() does check that the address being requested is
actually part of the LMB memory map.
There is no need to support reserving memory regions which are outside
the LMB memory map. Remove the lmb_reserve() API functionality and use
the functionality provided by lmb_alloc_addr() instead. The
lmb_alloc_addr() will check if the requested address is part of the
LMB memory map and return an error if not.
Signed-off-by: Sughosh Ganu <[email protected]>
Acked-by: Ilias Apalodimas <[email protected]>
|
|
Some androidboot image have invalid kernel/ramdisk load addr,
force to ignore those value and use loadaddr instead.
There is a concern on exposing the whole memory to image loading is
dangerous. Also, since it's not always possible to change the load
addr by repacking the boot.img (mainly due to AVB signature mismatch),
we need a way to use kernel_addr_r and ramdisk_addr_r.
Suggested-by: Casey Connolly <[email protected]>
Reviewed-by: Neil Armstrong <[email protected]>
Tested-by: Mattijs Korpershoek <[email protected]>
Signed-off-by: George Chan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Casey Connolly <[email protected]>
|
|
Marek Vasut <[email protected]> says:
Rename the environment related variables and add ENV_ prefix, so that
all configuration options which are related to environment would have
an CONFIG_ENV_ prefix. No functional change.
Link: https://lore.kernel.org/r/[email protected]
|
|
Rename the variable and add ENV_ prefix, so that all configuration
options which are related to environment would have an CONFIG_ENV_
prefix. No functional change.
Also rename USE_DEFAULT_ENV_FILE to USE_ENV_DEFAULT_ENV_TEXT_FILE .
Signed-off-by: Marek Vasut <[email protected]>
|
|
There is no such thing as struct bootmeth, it's probably a typo. This
issue doesn't affect the execution as it's a pointer, and pointer sizes
are the same for all data types. But it can be confusing, so make it
struct udevice, as it should be.
Fixes: a950d31abe98 ("bootstd: Add the bootmeth uclass and helpers")
Signed-off-by: Sam Protsenko <[email protected]>
|
|
Specifying efi_mgr in 'bootmeths' environment variable leads to NULL
pointer dereference when 'bootflow scan' is executed, with call trace
like this:
priv->fake_dev // NULL pointer dereference
.read_bootflow = efi_mgr_read_bootflow()
bootmeth_get_bootflow()
bootflow_check()
bootflow_scan_first()
do_bootflow_scan()
'bootflow scan -l'
That happens because in case when 'bootmeths' env var is defined the
bootmeth_efi_mgr driver is not probed, and the memory for its private
data isn't allocated by .priv_auto. In case when 'bootmeths' env var is
not defined, the std->bootmeth_count is 0, and the execution flow in
bootmeth_setup_iter_order() takes "no ordering" path, which in turn runs
uclass_get_device_by_seq() -> ... -> device_probe(), so issue isn't
present there. But when 'bootmeths' is defined and contains efi_mgr, the
std->bootmeth_count > 0, so bootmeth_setup_iter_order() follows the "we
have an ordering" path, where devices are not probed. In other words:
'bootmeths' defined 'bootmeths' not defined
--------------------------------------------------------
priv == NULL priv != NULL
^ ^
| device_alloc_priv()
no probe device_of_to_plat()
^ device_probe()
| uclass_get_device_tail()
dev = order[i] uclass_get_device_by_seq()
^ ^
| have an ordering | no ordering
+----------------+---------------+
|
bootmeth_setup_iter_order()
bootflow_scan_first()
do_bootflow_scan()
Add an explicit device_probe() call in "we have an ordering" case to fix
the issue.
Fixes: c627cfc14c08 ("bootstd: Allow scanning for global bootmeths separately")
Signed-off-by: Sam Protsenko <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Free memory allocated for 'order' (array of bootmeths) on error paths in
bootmeth_setup_iter_order() function.
Fixes: c627cfc14c08 ("bootstd: Allow scanning for global bootmeths separately")
Fixes: 10d16faa436c ("bootstd: Detect empty bootmeth")
Signed-off-by: Sam Protsenko <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Martin Schwan <[email protected]> says:
This series implements a new bootmeth for RAUC A/B systems. RAUC (Robust
Auto Update Controller) is a lightweight update client, providing "Safe
and Secure OTA Updates for Embedded Linux". See the following links for
more information about RAUC:
https://rauc.io/
https://rauc.readthedocs.io/en/latest/
PHYTEC uses RAUC in its Yocto based distribution "Ampliphy" as the
default way of updating embedded devices based on PHYTEC hardware. So
far, the logic selecting the correct partitions and files to boot was
being implemented in the U-Boot environment. While this is a
straightforward way to do it, adding and supporting new platforms became
somewhat tedious and is platform-specific. The introduction of U-Boot's
"Standard Boot" provided a convincing alternative, promising a simpler
and more portable way of booting, even for RAUC systems. This led me to
implement a new bootmeth supporting RAUC A/B systems. Note, that this
new bootmeth is not proprietary to PHYTEC products and is designed to
work on other hardware with a RAUC A/B system, too.
The bootmeth currently only supports symmetric A/B partitioning layouts.
E.g. A/rescue is not (yet) supported. The partition indexes and default
slot tries can be specified via configuration options.
For now, the bootmeth_rauc uses a similar approach for loading the
Kernel and device tree as the bootmeth_script, in that it requires a FIT
containing a U-Boot script loading the desired distro. It could be
possible to support booting without a script and load the Kernel and DT
directly with this bootmeth, but I found the script method to be very
flexible for now, in letting the distro decide what to load.
The bootmeth_rauc was tested on a phyBOARD-Pollux i.MX8M Plus [1] with
BSP-Yocto-Ampliphy-i.MX8MP-PD24.1.2 [2].
Supported boot devices are currently only MMC devices, but it should be
possible to add SPI flashes in the future.
To test this patch stack with PHYTEC's phyBOARD-Pollux i.MX8M Plus
board, you need to adjust the boot files to include the boot.scr.uimg
containing the distro's boot script and set "optargs" to "${raucargs}"
in it. Also disable any legacyboot in the U-Boot environment and simply
boot with Standard Boot:
bootmeth order rauc
bootflow scan -lb
[1]: https://www.phytec.eu/en/produkte/single-board-computer/phyboard-pollux/
[2]: https://download.phytec.de/Software/Linux/BSP-Yocto-i.MX8MP/BSP-Yocto-Ampliphy-i.MX8MP-PD24.1.2/
Link: https://lore.kernel.org/r/[email protected]
[trini: Don't enable by default]
|
|
Add a bootmeth driver which supports booting A/B system with RAUC as
their update client.
Signed-off-by: Martin Schwan <[email protected]>
Tested-by: Wadim Egorov <[email protected]>
|
|
Although if we don't setup an initrd, there could be a stale initrd
setting from the previous boot firmware in the live device tree. So,
make sure there is no setting left if we don't want an initrd.
This can happen when booting on a Raspberry Pi. The boot firmware can
happily load an initrd before us and configuring the addresses in the
live device tree we get handed over.
Especially the setting `auto_initramfs` in config.txt is dangerous.
When enabled (default), the firmware tries to be smart and looks for
initramfs files.
Signed-off-by: Richard Weinberger <[email protected]>
|
|
Add support for generating STM32 Image V2.0, which is used by STM32MP13xx.
The image header layout is similar to STM32MP15xx STM32 Image V1.0, but is
different enough to justify duplicate functions to generate the v2 image.
This code at least attempts to align the V1 and V2 image handling where
possible.
Signed-off-by: Marek Vasut <[email protected]>
|
|
Prepare v2025.07-rc4
|
|
This reverts commit 14a4792a71db3561bea065415ac1f2ac69ef32b5 as
discussed in [1].
[1] https://lists.denx.de/pipermail/u-boot/2025-May/590841.html
Signed-off-by: Mayuresh Chitale <[email protected]>
Reviewed-by: Leo Yu-Chi Liang <[email protected]>
|
|
Commit 21e7fa0e3ac5 ("image: android: handle ramdisk default address")
changed the default behavior for header versions less than or equal to 2.
The ramdisk address (img_data.ramdisk_ptr) is only assigned to *rd_data
if the physical load address (img_data.ramdisk_addr) is equal to 0 or
the Android default ramdisk address.
/* Ramdisk can be used in-place, use current ptr */
if (img_data.ramdisk_addr == 0 ||
img_data.ramdisk_addr == ANDROID_IMAGE_DEFAULT_RAMDISK_ADDR) {
*rd_data = img_data.ramdisk_ptr;
} else {
ramdisk_ptr = img_data.ramdisk_addr;
*rd_data = ramdisk_ptr;
memcpy((void *)(ramdisk_ptr), (void *)img_data.ramdisk_ptr,
img_data.ramdisk_size);
}
When the img_data.ramdisk_addr and the img_data.kernel_addr are the same
*rd_data needs to be assigned to the ramdisk address (ramdisk_ptr), not
the physical address (ramdisk_addr).
As a result of the current behavior, we can no longer boot a kernel on
the Renesas R-Car S4 board.
Add an additional check to the if clause so that the ramdisk address is
assigned when the kernel address and the ramdisk address are the same,
restoring the previous default behavior.
Fixes: 21e7fa0e3ac5 ("image: android: handle ramdisk default address")
Signed-off-by: Eddie Kovsky <[email protected]>
Reviewed-by: Mattijs Korpershoek <[email protected]>
Tested-by: Mattijs Korpershoek <[email protected]> # khadas vim3
|
|
When building with only bootz command, without bootm, images pointer
inside bootm_info structure is not initialized. And since this structure
is stored in stack, the generated error is kind of random, but most of
the time this will generate: "ramdisk - allocation error".
Also, after analysis, this problem could occur with the command booti,
if the command bootm is disabled.
Currently bootm_init() is called by: do_bootz(), do_bootm(), do_booti()
and by do_stm32prog(). And all of these commands execute bootm_run_states()
which access the images pointer stored into bootm_info structure.
So, to fix this issue, just do the assignment unconditionally.
Fixes: c2211ff65136 ("bootm: Add more fields to bootm_info")
Signed-off-by: Benjamin ROBIN <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
|
|
The expo now has all that is needed to apply a suitable theme, so drop
this unnecessary code. Any further tweaks can be added to the generic
expo code.
Signed-off-by: Simon Glass <[email protected]>
|
|
Apply the theme, calculate dimensions, highlight the menu and arrange
the scene correctly, so that everything is ready to go when the expo
is rendered.
Signed-off-by: Simon Glass <[email protected]>
|
|
Some callers may wish to add items later as they are discovered. Split
the setup code into its own function, to permit this.
Signed-off-by: Simon Glass <[email protected]>
|
|
Rather than returning a bootflow, return the index of the bootflow. This
will allow callers to do their own translation to bootflows or some
other data structure.
Also return a special code when the user tries to move the pointer, so
that the caller can cancel the boot-menu timeout, if this is in use.
Signed-off-by: Simon Glass <[email protected]>
|
|
The menu is currently quite basic. Make use of some recently added
features in expo, to:
- Show proper prompts
- Highlight the current item
- Centre text
- Use multi-line text instead of two independent lines
- Put a box around the items
Signed-off-by: Simon Glass <[email protected]>
|
|
It is often necessary to centre objects within the display area. Add a
special position value to indicate this.
Signed-off-by: Simon Glass <[email protected]>
|
|
Within tests it is useful to be able to control rendering of the expo.
Drop the automatic call to expo_render() within expo_poll() and adjust
its callers to handle this instead.
Signed-off-by: Simon Glass <[email protected]>
|
|
If the user does not quickly provide some input, exit so that other
things can be done, such as searching for the next OS.
Signed-off-by: Simon Glass <[email protected]>
|
|
Expo normally uses a pointer to show the current item. Add support for
highlighting as well, since this makes it easier for the user to see the
current item.
Signed-off-by: Simon Glass <[email protected]>
|
|
Provide some forward declarations for types used in this file, to keep
the LSP happy and avoid errors if the caller happens not to include the
required header.
Signed-off-by: Simon Glass <[email protected]>
|
|
The gap is intended to separate options which are on different bootdevs,
but most of the time there is only one. So drop this gap.
Keep the code around in case it becomes useful to have it in the style,
or based on some other mechanism.
Signed-off-by: Simon Glass <[email protected]>
|
|
For USB devices we use the string 'usb_mass_storage' which is a bit
verbose. Use 'USB' instead.
Signed-off-by: Simon Glass <[email protected]>
|
|
It is useful to be able to edit text, e.g. to allow the user to edit the
environment or the command-line arguments for the OS.
Add the beginnings of an implementation. Future work is needed to finish
this: keypress handling and scrolling. For now it just displays the
text.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add support for left, right and centred alignment for text, in the
horizontal dimension.
Also support top, bottom and centred in the vertical dimension, for the
text object as a whole.
Alignment is not yet implemented for images. It has no meaning for
menus. A textline object uses a text object internally, so alignment
is supported there.
Provide some documentation to explain how objects are positioned.
Signed-off-by: Simon Glass <[email protected]>
|
|
It is useful to be able to draw a box around elements in the menu. Add
support for an unfilled box with a selectable thickness.
Note that there is no support for selecting the colour for any expo
objects yet.
Signed-off-by: Simon Glass <[email protected]>
|
|
Allow this setting to be controlled from the theme.
Signed-off-by: Simon Glass <[email protected]>
|
|
Use the measurement info to write each line of text separately, thus
respecting word-wrapping and newlines.
Fix up the comment for scene_obj_render() while we are here.
Since a lineedit does not support alignment, add a special case to just
display the text if there is no measurement. This happens assuming the
lineedit is initially empty.
Signed-off-by: Simon Glass <[email protected]>
|
|
In some cases dynamic text is needed, e.g. for a menu countdown. Add a
function which handles this, allowing the caller to take control of the
text that is shown on each render.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add an early return if there is no string. Move all declarations to the
top of the function.
Signed-off-by: Simon Glass <[email protected]>
|
|
The code to render text is quite long, so put it in its own function.
Signed-off-by: Simon Glass <[email protected]>
|
|
In preparation for adding more text types, refactor the common fields
into a new structure. This will allow common code to be used.
Signed-off-by: Simon Glass <[email protected]>
|
|
At present labels are lined up vertically. Do the same with keys and
descriptions, since it looks nicer.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add a function to allow the size of an object to be set independently
of its position.
Also add a function to permit the object's bounding box to be set
independently of its dimensions.
Signed-off-by: Simon Glass <[email protected]>
|
|
In some cases it is useful to obtain more than just two bounding boxes
from a menu, e.g. to line up all descriptions vertically.
Use an array to obtain bounding-box information and calculate it
separately for each item.
Signed-off-by: Simon Glass <[email protected]>
|
|
At present each object has a width and height and the bounding box is
implicit in that.
This is not flexible enough to handle objects which are larger than
their contents might need. For example, when centring a text object we
might want to have it stretch across the whole width of the display even
if the text itself does not need that much space.
Create a new 'dimensions' field and convert the existing width/height
into x1/y1 coordinates.
Signed-off-by: Simon Glass <[email protected]>
|
|
It is more convenient to put strings in an abuf so they can easily be
resized. Adjust the struct accordingly.
Signed-off-by: Simon Glass <[email protected]>
|
|
These coordinates are the top left values, so rename them to x0 and y0
in preparation for changing the width and height to x1 and y1
Signed-off-by: Simon Glass <[email protected]>
|
|
At present we assume that each object is a simple box and that it fills
the whole box.
This is quite limiting for text objects, which we may want to centre
within the box. We need a position within the box where drawing starts.
Rename the scene_dim struct to indicate that it is a bounding box.
Signed-off-by: Simon Glass <[email protected]>
|
|
Refactor bootflow_menu_add_all() to call a new bootflow_menu_add() to
add each of its bootflows.
Move the last_bootdev value into struct menu_priv to make this work.
Signed-off-by: Simon Glass <[email protected]>
|
|
Split the iteration piece of this function into bootflow_menu_add_all()
so that it is possible for the caller to be in control of adding items
to the menu.
Move the expo_destroy() call into the caller.
Signed-off-by: Simon Glass <[email protected]>
|
|
Split the starting piece of this function into bootflow_menu_start()
and the polling part into bootflow_menu_poll() so that it is possible
for the caller to be in control of the event loop.
Move the expo_destroy() call into the caller.
Signed-off-by: Simon Glass <[email protected]>
|
|
At present when the user moves to a new menu item, the menu code
handles this immediately. This means it is not possible to use an expo
action to effect the same change.
Update this so that EXPOACT_POINT_ITEM is handled in cedit_do_action().
Expand the test to cover this.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add functions to allow a caller to find out the current menu item and to
select a different one.
Update the event handling so that an attempt to change the current item
(e.g. by pressing the up-arrow key) is reported to the caller, since
this may be used to cancel an autoboot timeout.
Signed-off-by: Simon Glass <[email protected]>
|
|
At present this function locates it own video device. Pass it in to
provide more flexibility.
Signed-off-by: Simon Glass <[email protected]>
|
|
Refactor the action-processing code into a new cedit_do_action()
function so we can call it from a test. Check moving to a new field and
opening the menu, to ensure that rendering is correct.
Signed-off-by: Simon Glass <[email protected]>
|
