Add support for moving from TPL->VPL->SPL so that the VPL build can fit
properly into the boot flow.
Use #ifdefs to avoid creating unwanted symbols which Binman would then
try (and perhaps fail) to provide.
Add debugging to indicate the next phase.
Signed-off-by: Simon Glass <sjg@chromium.org>
Simon Glass <sjg@chromium.org> says:
This includes various patches towards implementing the VBE abrec
bootmeth in U-Boot. It mostly focuses on introducing a relocating
SPL-loader so that VBE can run in the limited amount of SRAM available
on many devices.
Another minor new feature is support in VBE for specifying the image
phase when loading from a FIT. This allows a single FIT to include
images for several boot phases, thus simplifying image-creation.
One lingering niggle in this series is that it has a different code path
for sandbox, since it does not support the relocating jump. It should be
possible to resolve this with additional work, but I have not attempted
this so far.
For v2, I have split the first patch into 5 pieces, to make it easier to
see the code-size impact, plus added a few tweaks to reduce code size.
Again, only MMC is supported so far.
Looking ahead, series G will have some more plumbing and H some rk3399
pieces. That should be enough to complete these feature.
Here is a run in my lab, with the VBE ABrec bootmeth. You can see that
VPL runs before memory is set up. SPL sets up memory and can be upgraded
in the field reliably.
$ ub-int vbe
Building U-Boot in sourcedir for rk3399-generic
Bootstrapping U-Boot from dir /tmp/b/rk3399-generic
Writing U-Boot using method rockchip
U-Boot TPL 2025.01-rc3-00345-gdfbdbf1eb56c-dirty (Jan 08 2025 - 10:47:58)
Trying to boot from vbe_abrec
load: Firefly-RK3399 Board
Using 'config-3' configuration
Trying 'image-vpl' firmware subimage
Using 'config-3' configuration
Trying 'fdt-3' fdt subimage
U-Boot VPL 2025.01-rc3-00345-gdfbdbf1eb56c-dirty (Jan 08 2025 - 10:47:58)
Trying to boot from vbe_abrec
load: Firefly-RK3399 Board
Starting with empty state
VBE: Firmware pick A at 800000
Using 'config-3' configuration
Trying 'spl' firmware subimage
Using 'config-3' configuration
Trying 'fdt-3' fdt subimage
Channel 0: DDR3, 800MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
Channel 1: DDR3, 800MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
256B stride
U-Boot SPL 2025.01-rc3-00345-gdfbdbf1eb56c-dirty (Jan 08 2025 - 10:47:58 -0700)
Trying to boot from vbe_abrec
load: Firefly-RK3399 Board
VBE: Firmware pick A at 900000
load_simple_fit: Skip load 'atf-5': image size is 0!
Relocating bloblist ff8eff00 to 100000: done
ns16550_serial serial@ff1a0000: pinctrl_select_state_full: uclass_get_device_by_phandle_id: err=-19
U-Boot 2025.01-rc3-00345-gdfbdbf1eb56c-dirty (Jan 08 2025 - 10:47:58 -0700)
SoC: Rockchip rk3399
Reset cause: POR
Model: Firefly-RK3399 Board
DRAM: 4 GiB (effective 3.9 GiB)
Core: 314 devices, 33 uclasses, devicetree: separate
MMC: mmc@fe310000: 3, mmc@fe320000: 1, mmc@fe330000: 0
Loading Environment from SPIFlash... Invalid bus 0 (err=-19)
*** Warning - spi_flash_probe_bus_cs() failed, using default environment
In: serial,usbkbd
Out: serial,vidconsole
Err: serial,vidconsole
Model: Firefly-RK3399 Board
Net: PMIC: RK808
eth0: ethernet@fe300000
starting USB...
Bus usb@fe380000: USB EHCI 1.00
Bus usb@fe3a0000: USB OHCI 1.0
Bus usb@fe3c0000: USB EHCI 1.00
Bus usb@fe3e0000: USB OHCI 1.0
Bus usb@fe900000: Register 2000140 NbrPorts 2
Starting the controller
USB XHCI 1.10
scanning bus usb@fe380000 for devices... 1 USB Device(s) found
scanning bus usb@fe3a0000 for devices... 1 USB Device(s) found
scanning bus usb@fe3c0000 for devices... 2 USB Device(s) found
scanning bus usb@fe3e0000 for devices... 1 USB Device(s) found
scanning bus usb@fe900000 for devices... 1 USB Device(s) found
scanning usb for storage devices... 0 Storage Device(s) found
Hit any key to stop autoboot: 0
Link: https://lore.kernel.org/r/20250116012723.2820301-1-sjg@chromium.org
This is fairly easy to use. The SPL loader sets up some fields in the
spl_image_info struct and calls spl_reloc_prepare(). When SPL is ready
to do the jump it must call spl_reloc_jump() instead of jump_to_image().
Add this logic.
Signed-off-by: Simon Glass <sjg@chromium.org>
This function will be used by the relocating jumper too, so add a
typedef to the header file to avoid mismatches.
Signed-off-by: Simon Glass <sjg@chromium.org>
Rather than doing autoprobe within the driver model code, move it out to
the board-init code. This makes it clear that it is a separate step from
binding devices.
For now this is always done twice, before and after relocation, but we
should discuss whether it might be possible to drop the post-relocation
probe.
For boards with SPL, the autoprobe is still done there as well.
Note that with this change, autoprobe happens after the
EVT_DM_POST_INIT_R/F events are sent, rather than before.
Link: https://lore.kernel.org/u-boot/20240626235717.272219-1-marex@denx.de/
Signed-off-by: Simon Glass <sjg@chromium.org>
The variable 'ret' is defined twice, which is not intended. This may
have been a local merge error.
Signed-off-by: Simon Glass <sjg@chromium.org>
Fixes: 2eefeb6d893 ("spl: Report a loader failure")
The current check looks only at SPL, but TPL or VPL might have a
different setting. Update the condition.
Signed-off-by: Simon Glass <sjg@chromium.org>
The current check looks only at SPL, but TPL or VPL might have a
different setting. Update the condition.
Signed-off-by: Simon Glass <sjg@chromium.org>
If a loader returns an error code it is silently ignored. Show a message
to at least provide some feedback to the user.
Signed-off-by: Simon Glass <sjg@chromium.org>
This reverts commit 1fdf53ace1, reversing
changes made to e5aef1bbf1.
I had missed that this caused too much size growth on rcar3_salvator-x.
Signed-off-by: Tom Rini <trini@konsulko.com>
The variable 'ret' is defined twice, which is not intended. This may
have been a local merge error.
Signed-off-by: Simon Glass <sjg@chromium.org>
Fixes: 2eefeb6d893 ("spl: Report a loader failure")
The current check looks only at SPL, but TPL or VPL might have a
different setting. Update the condition.
Signed-off-by: Simon Glass <sjg@chromium.org>
The current check looks only at SPL, but TPL or VPL might have a
different setting. Update the condition.
Signed-off-by: Simon Glass <sjg@chromium.org>
If a loader returns an error code it is silently ignored. Show a message
to at least provide some feedback to the user.
Signed-off-by: Simon Glass <sjg@chromium.org>
The call to malloc() is a bit strange. The naming of the arguments
suggests that an address is passed, but in fact it is a pointer, at
least in the board_init_r() function and SPL equivalent.
Update it to work as described. Add a function comment as well.
Note that this does adjustment does not extend into the malloc()
implementation itself, apart from changing mem_malloc_init(), since
there are lots of casts and pointers and integers are used
interchangeably.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Sughosh Ganu <sughosh.ganu@linaro.org> says:
This is a follow-up from an earlier RFC series [1] for making the LMB
and EFI memory allocations work together. This is a non-rfc version
with only the LMB part of the patches, for making the LMB memory map
global and persistent.
This is part one of a set of patches which aim to have the LMB and EFI
memory allocations work together. This requires making the LMB memory
map global and persistent, instead of having local, caller specific
maps. This is being done keeping in mind the usage of LMB memory by
platforms where the same memory region can be used to load multiple
different images. What is not allowed is to overwrite memory that has
been allocated by the other module, currently the EFI memory
module. This is being achieved by introducing a new flag,
LMB_NOOVERWRITE, which represents memory which cannot be re-requested
once allocated.
The data structures (alloced lists) required for maintaining the LMB
map are initialised during board init. The LMB module is enabled by
default for the main U-Boot image, while it needs to be enabled for
SPL. This version also uses a stack implementation, as suggested by
Simon Glass to temporarily store the lmb structure instance which is
used during normal operation when running lmb tests. This does away
with the need to run the lmb tests separately.
The tests have been tweaked where needed because of these changes.
The second part of the patches, to be sent subsequently, would work on
having the EFI allocations work with the LMB API's.
[1] - https://lore.kernel.org/u-boot/20240704073544.670249-1-sughosh.ganu@linaro.org/T/#t
Notes:
1) These patches are on next, as the alist patches have been
applied to that branch.
2) I have tested the boot on the ST DK2 board, but it would be good to
get a T-b/R-b from the ST maintainers.
3) It will be good to test these changes on a PowerPC platform
(ideally an 85xx, as I do not have one).
The spl_board_init() function on sandbox invokes the unit
tests. Invoking the tests should be done once the rest of the system
has been initialised. Call the spl_board_init() function at the very
end, once the rest of the initilisation functions have been called,
including the setting up of the LMB memory map.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
The memory map maintained by the LMB module is now persistent and
global. This memory map is being maintained through the alloced list
structure which can be extended at runtime -- there is one list for
the available memory, and one for the used memory. Allocate and
initialise these lists during the board init.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Initialize DRAM size in SPL stage since networking requires DDR
to be initialized.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: Chintan Vankar <c-vankar@ti.com>
The early malloc region is normally quite small and is certainly less
than 4GB, so use a 32-bit value for the limit and pointer. Update the
comments for clarity while we are here.
Signed-off-by: Simon Glass <sjg@chromium.org>
We don't need a full word for this boolean value. Convert it into a flag
to save space in global_data.
Reviewed-by: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
Define spl_set_header_raw_uboot() always so we can drop the last #ifdef
in this function.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
This function has a number of unnecessary #ifdefs so remove them.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
These should have a CONFIG_ prefix. Add it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Fixes: 7a0d88076b ("Add in the ability to load and boot an uncompr...")
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
Specify the FIT and include information about each loaded image, as
required by the UPL handoff.
Write the UPL handoff into the bloblist before jumping to the next phase.
Control this using a runtime flag to avoid conflicting with other
handoff mechanisms.
Signed-off-by: Simon Glass <sjg@chromium.org>
Some architectures use spl_board_init() in their SoC specific
implementation. Board developers should be able to add board specific
implementation via spl_board_init(). Hence, introduce a spl_soc_init()
method which is called right before spl_board_init() for SoC
specific implementation.
Signed-off-by: Lukas Funke <lukas.funke@weidmueller.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20240327121153.2455126-2-lukas.funke-oss@weidmueller.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
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 <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
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 c8ffd1356d, reversing
changes made to 2ee6f3a5f7.
Reported-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
This follows the example of RISC-V where <asm/global_data.h> includes
<asm/u-boot.h> directly as "gd" includes a reference to bd_info already
and so the first must include the second anyhow. We then remove
<asm/u-boot.h> from all of the places which include references to "gd"
an so have <asm/global_data.h> already.
Signed-off-by: Tom Rini <trini@konsulko.com>
When `gd` is relocated during `spl_relocate_stack_gd()` the
doubly-linked circular list in the `log_head` member is broken.
The last element of the list should point back to the initial
`list_head`, but as the initial `list_head` is moved the pointer becomes
stale. As a result the loop in `log_dispatch` would never finish.
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Add an API which enforces framebuffer reservation from end of RAM.
This is done so that next stage can directly skip this region before
carrying out further reservations.
Signed-off-by: Devarsh Thakkar <devarsht@ti.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Implementers of SPL_LOAD_IMAGE_METHOD have to correctly determine what
type of image is being loaded and then call the appropriate image load
function correctly. This is tricky, because some image load functions
expect the whole image to already be loaded (CONFIG_SPL_LOAD_FIT_FULL),
some will load the image automatically using spl_load_info.read()
(CONFIG_SPL_LOAD_FIT/CONFIG_SPL_LOAD_IMX_CONTAINER), and some just parse
the header and expect the caller to do the actual loading afterwards
(legacy/raw images). Load methods often only support a subset of the
above methods, meaning that not all image types can be used with all
load methods. Further, the code to invoke these functions is
duplicated between different load functions.
To address this problem, this commit introduces a "spl_load" function.
It aims to handle image detection and correct invocation of each of the
parse/load functions.
Although this function generally results in a size reduction with
several users, it tends to bloat boards with only a single user.
This is generally because programmers open-coding the contents of this
function can make optimizations based on the specific loader. For
example, NOR flash is memory-mapped, so it never bothers calling
load->read. The compiler can't really make these optimizations across
translation units. LTO solves this, but it is only available on some
arches. To address this, perform "pseudo-LTO" by inlining spl_load when
there are one or fewer users. At the moment, there are no users, so
define SPL_LOAD_USERS to be 0.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The purpose of SHOW_ERRORS is to print extra information. Make it depend
on LIBCOMMON to avoid having to check for two configs.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
All "physical" addresses in SPL must be converted to virtual addresses
before access in order for sandbox to work. Add some calls to map_sysmem in
appropriate places. We do not generally call unmap_sysmem, since we need
the image memory to still be mapped when we jump to the image. This doesn't
matter at the moment since unmap_sysmem is a no-op.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
spl_board_prepare_for_boot() is not called before jumping/invoking atf,
optee, opensbi or linux images.
Jump to image at the end of board_init_r() to fix this.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Simon Glass <sjg@chromium.org>
The bloblist should not be finalised until the image is fully set up.
This allows any final handoff information to be included in the bloblist.
Signed-off-by: Simon Glass <sjg@chromium.org>
For some reason this code was put in the main spl.c file. Move it out
to the FIT implementation where it belongs.
Signed-off-by: Simon Glass <sjg@chromium.org>
This feature has some helpers in its header file so that its functions
resolve to nothing when the feature is disabled. Add a few more and use
these to simplify the code.
With this there are no more #ifdefs in board_init_r()
Signed-off-by: Simon Glass <sjg@chromium.org>
Use the new SPL/TPL/VPL_SYS_MALLOC_F symbols to determine whether the
malloc pool exists.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <sean.anderson@seco.com>
This code is pretty ugly, with many #ifdefs
There are quite a lot of IH_OS_U_BOOT values so the compiler struggles
to create a jump table here. Also, most of the options are normally
disabled.
Change it to an else...if construct instead. Add an accessor for the
spl_image field behind an #ifdef to avoid needing #ifdef in the C code.
Signed-off-by: Simon Glass <sjg@chromium.org>
