- Replace `armbian.list` with `armbian.sources`. This holds the same
information in a newer format, deb822.
- Replace HTTP with HTTPS for Armbian repositories.
- python3-pip implies a very old setuptools (which is system-wide and takes precedence)
- python3-dev implies python3-distutils (which is long deprecated)
- get-pip.py allows us to version pip in requirements.txt just like everything else
- in the end this fixes the conundrum with pylibfdt / dtschema on Jammy
- and, finally, the setuptools we specify in requirements.txt will be actually used
* rock-3a - fix formatting of gpio-line-names
properly format gpio-line-names, following the example from Trevor Woerner for his contribution to the radxa-zero-3 dt.
Also fix a wrong pin number
* fix applying
The extcon-usbc-virtual-pd driver requests the detection GPIO with the flag GPIOD_OUT_LOW, which leads to a failure in gpiod_to_irq and consequently causes the driver loading to fail.
This issue can be fixed by changing GPIOD_OUT_LOW to either GPIOD_IN or GPIOD_ASIS.
This patch opts to change GPIOD_OUT_LOW to GPIOD_ASIS, as this is a less invasive approach. For devices intending to use this driver, it's also necessary to set the pinctrl property of the detection GPIO to input mode in the dts file.
- as Armbian u-boot is possibly deployed to SPI (replacing Petitboot),
try to make as compatible as possible; eg enable squashfs support
- in this case, `setexpr` and `fileenv` are used by HAOS
- found `fileenv` here 049e1a1949
Signed-off-by: Ricardo Pardini <ricardo@pardini.net>
This reverts commit a3dac432fc.
USB patch for 6.12.y needs rework - its a bit more work as code has been moved around kernel tree. TBD after 25.2 release.
Follow change from radxa 738cda44532c324e07296c95751e3083fa95f282.
Change with command:
sed -i "/CONFIG_OPTEE/d" patch/u-boot/legacy/u-boot-radxa-rk35xx/defconfig/*
- turns out `pahole` for `vmlinux` can take multiple gigabytes of RAM to run successfully
- I can't simply decide based on available RAM, as that would make .config hashes mismatch
- thus, introduce:
- default is to enable BTF; if on low-ram host, error out unless KERNEL_BTF=yes is passed
- if KERNEL_BTF=no is passed, the BTF debug info is always disabled
- if KERNEL_BTF=yes is passed, then a warning is produced, but BTF is still enabled
- the magic number "6451 MiB" was determined empirically (and is probably bs)
- switch MOTD download to new central location https://github.armbian.com/
- move cronjob for MOTD download from weekly to daily
- change mord defaults, remove clear from defaults
* Add MXQ target. Copy HDMI fix from odroid-c1.
* meson8, MXQ: add boot from usb support, configurable dtb
* MXQ: remove boot logo
Built-in U-BOOT is used, so the logo will not be displayed anyway.
* meson: kernel update: legacy -> 6.6, current -> 6.12
* Change Odroid C1 and Onecloud to community supported as build now passes
---------
Co-authored-by: Igor Pecovnik <igor@armbian.com>
Scanning accesspoints requires an activated wifi device.
Networkd does not activate it (only NetworkManager does so).
As a result, firstlogin is not able to configure and use wifi on minimal
images, as only Server and Desktop images have NetworkManeger installed.
Activating the wifi device using "ip link set ${WIFI_DEVICE} up" before
scanning fixes this.
(And can safely be done - there is no problem activating it multiple
times, as it is the case with NetworkManager installed.)
- Replace creation of `armbian-config.list` with
`armbian-config.sources`. This holds the same information in a newer
format, deb822. This does not affect `armbian.list`.
See also: https://github.com/armbian/configng/pull/407
Overlay for testing purposes:
There is a strange interaction between that ethernet adapter and the rtc
(it runs better -but still not well- with the ethernet adapter enabled).
Contains rewritten dts, based on decompiled dtb from original Klipad50
(as similiar as possible to original).
Kernel defconfig taken from MKSPI (replaced "mkspi" with "mksklipad50")
While syncing with upstream Linux helios64 dts for 6.9 I missed
the diff for the helios64 status led label (upstream was
helios64:green:status instead of Armbian helios64::status) and
thus broke armbian helios64-heartbeat-led.service.
This commit restores the Armbian label helios64::status to let
helios64-heartbeat-led.service starts.
Fixes: cbaf67f00 ("Sync helios64 to its updated 6.9 dts")
Signed-off-by: Alban Browaeys <alban.browaeys@gmail.com>
- this allows us to drop a bunch of code needed for 2025.01
- UMS works out of the box (simple dr_mode = peripheral change enough)
- USB3 Type-A port works out of the box
- SPI works out of the box
This makes:
W: Possible missing firmware /lib/firmware/regulatory.db for built-in driver cfg80211
W: Possible missing firmware /lib/firmware/regulatory.db.p7s for built-in driver cfg80211
go away.
Since v5.11, Linux upstream commit 8b5c2b45b8f0 disabled the internal pull-down
for the strobe line explicitly instead of leaving it in its default
state (the opposite of Rockchip Kernel which force enable this intrnal
pulldown by default) thus causing I/O errors in HS400 mode for various
eMMC modules (the ones that did not have an external pulldown on the
strobe line likely).
Discussed upstream and the choice has been made to keep this
behavior as it has been there for too long and reverting might break
other boards http://lists.infradead.org/pipermail/linux-rockchip/2024-April/046416.html
HS400 for helios64 was disabled in commit 5f2e23e90
"Disabled hs400 on Helios64 (hopefully only temporarily) (#3126)"
but for 5.10.63 (ie before the above commit disabled internal pulldown).
Back then I had bisected the emmc I/O issue to a double init of the emmc
regulator introduced while adding deferred support to the regulator driver
core, ie commit 06653ebc0ad2 "regulator: core: resolve supply for boot-on/always-on regulators"
in 5.10.60. But due to the new default for the strobe line internal pulldown
introduced in 5.11, when this double voltage regulator init issue was fixed likely by
8a866d527ac0 "regulator: core: Resolve supply name earlier to prevent double-init"
in Linux 6.1.
* add nanopi-r3s.csc board config
* Nanopi R3S - add uboot patch
* Nanopi R3S - add dts
* fix bogus patch header
* Migrate to new patch series
Source & credits https://patchwork.ozlabs.org/project/uboot/list/?series=438193
I still need to figure out how to convert a patch series from patchwork in a way so the build framework doesn't complain about headers and magic and whatever. So a few hundreds of header lines have been replaced with something generic.
Applies and builds fine now against current stable v2024.10
* Create board-nanopi-r3s-fix-leds.patch
Thx https://github.com/armbian/build/pull/7556#issuecomment-2615110100
* Rename board-nanopi-r3s-fix-leds.patch to xx-board-nanopi-r3s-fix-leds.patch
* add new patch
- post_family_tweaks_bsp_ is called during the build of armbian-bsp-cli-xxx
- the cli package is built separately on the build server
- comparisons to build parameters such as DESKTOP_ENVIRONMENT do not work
Signed-off-by: Martin Schmiedel <Martin.Schmiedel@tq-group.com>
Now that upstream U-Boot has support for building K3 boot artifacts
we can drop using the ti-u-boot fork. Switch to v2025.01 and
adjust file names and add new dependencies as needed.
Option "--connect-timeout" used to determine external IP covers only
connection phase. Curl may "hans" in a different phase and this makes
login to the system impossible if MOTD "header" module enabled.
Using "--max-time" is more reliable and prevents hanging for hours.
Fixes issue https://github.com/armbian/build/issues/7753
Signed-off-by: Uladzimir Bely <wiselord1983@gmail.com>
- also LEDs, networking, more supported compression methods...
- copypasta reigns supreme. refactor is coming, just not today
- also hack the u-boot dtsi for consistent-with-Linux-mainline-DT mmc/sd numbering
- also LEDs, networking, more supported compression methods...
- copypasta reigns supreme. refactor is coming, just not today
- also hack the u-boot dtsi for consistent-with-Linux-mainline-DT mmc/sd numbering
- this was previously a separate patch in 2024.10
In the current main branch, workflow update-tools includes the content
of the upstream release notes in the created GitHub Issue. This
triggers a notification to every upstream contributor mentioned in the
release note. This also creates a cross-reference link to our GitHub
Issue (reporting that our dependency is not up-to-date) in every
upstream issue and pull request. The information that we need to
update our dependency is not relavant to the upstream contributors,
issues, and pull requests at all. We should avoid this situation.
In the first place, we do not need to include a copy of the upstream
release note. We can just put a link to the upstream release note.
Currently, we anyway hide the content in the <details> tag and require
the reader to press the button. The user experience is not so
different from pressing a link and visiting the upstream release note.
- we need to run the hooks _before_ running oldconfig, otherwise it is too late
- drop `kernel_config_check_and_repair()` since it won't ever detect anything now
* Refactor armbian-config into install as external package
- https://github.armbian.com/configng repository is always present
- package can still be removed from the install list
Since this tool will have deeper integration with Armbian, we should probably drop possibility to remove the tool from the list.
* Enable armbian-config as extension
- `post_repo_customize_image`: runs after repos have been enabled
- `post_armbian_repo_customize_image`: same, but only if Armbian repo is enabled
- both run after apt update, so packages can be directly installed from repos
- we used to have a bare-dt for odroid-m1 that got removed as it landed upstream
- but mainline doesn't have the nodes we use for the overlays
- at least i2c3 (which is aliased to i2c0) and uart1 breaks without them
- we used to have a bare-dt for odroid-m1 that got removed as it landed upstream
- but mainline doesn't have the nodes we use for the overlays
- at least i2c3 (which is aliased to i2c0) and uart1 breaks without them
The latest 'parted' package that is shipped with Ubuntu noble does not
allow to set the 'lba' flag on MBR/DOS partitions and fails with an
error. Use 'sfdisk' instead for the same.
Signed-off-by: Andrew Davis <afd@ti.com>
Signed-off-by: Beleswar Padhi <b-padhi@ti.com>
NCM has advantages over RNDIS and postmarketOS has toggled
to it 2 years ago. And I have tested it can work with common
linux and window10.
I also made some changes to let this script run inside intramfs.
And the systemd service can run well after rootfs is loaded.
- in f3f10a9517 the 16k-page stuff was dropped
- and that was the only reason for a separate kernel
- thus drop the change and unify; 2 less kernels to build and same result
- with DEBUG=yes, allows us to see what is leftover in directory during builds
- with new cleaning in separate commit, there shouldn't be any leftovers _at all_
- with this, every u-boot build will be a full rebuild.
- still, building u-boot is pretty fast
- and it is better to be a bit slower and actually produce correct results
- incl some indicative fdt/extlinux/gpt info
- include full config and defconfig for each target
- mark the old .config file as legacy (it was always the last-target's config)
- fix: savedefconfig after build, not before (as target_map might change it)
- fix: some very old u-boots (2011, odroidc1) do not have `savedefconfig` at all, so make it optional
- fix: same for `.config` -- very old u-boots worked different (`boards.cfg`?)
The patch changing pcie2 phy driver of rk356x is merged to v6.13-rc5 but
the dts patch in the same series is not merged, which will break pcie2
on rk356x boards.
This patch should be soon deleted since I have reported it upstream.
The files included in u-boot are rarely changed or do not
change at all. Compare bit by bit the memory region in which
the file will be located and the file itself. If they match,
then skip the recording.
This saves the life of the memory chip.
- refactor `prepare_host_binfmt_qemu()` out of `prepare_host_noninteractive()` and into `rootfs/qemu-static.sh`
- further split into more functions and return early to avoid deep nesting
- implement force import and load of qemu-arm for non-armhf capable arm64 hosts (incl magic numbers)
- enhance `deploy_qemu_binary_to_chroot()` & `undeploy_qemu_binary_from_chroot()`;
- add 2nd param "caller" for better logging/tracking
- does sanity-check and preserve existing binary if it exists
- explicitly deploy/undeploy for the 3 cases:
- image: moved undeploy from `post_debootstrap_tweaks()` into image build proper for consistency
- rootfs: was leaving trash behind (since post_debootstrap_tweaks never ran for rootfs), now properly undeploys
- initrd: was already fine, just added caller info
- added `arch-test` host dependency
- ensure `arch-test ${ARCH}` works during prepare-host
- > tl,dr: "can build 32-bit armv7 armhf using Apple silicon; can use rootfs cache cross-arch reliably"
- this affects `rewrite-kernel-config` and `kernel-config` CLIs
- this does not actually change the .config's we have, that will happen later / gradually as configs are rewritten
- no more `.defconfig` files are produced (they were in gitignore)
In order for the cleanup to be correct for tools, we need to pass
the VMLINUX_BTF variable, which contains the real path to the vmlinux
file we just compiled.
The vmlinux file itself is not involved in cleaning,
but the Makefile checks for its presence and cleaning
is aborted if it is not found.
Here, a folder named patches.addon must exist in order for
users to add new patches. These patches will be applied but
they are awaiting more careful consideration.
Let's just sort the sequence a little bit.
This driver will request firmware by the first compatiable string from
devicetree, which is usually a device specific string not in the driver
code. If this driver is built as module, it can't get loaded because
only the first compatible string is used to probe driver. There is no
such issue if this driver is built in kernel.
`ROCKCHIP_THUNDER_BOOT_DEFER_FREE_MEMBLOCK` should depend
on `ROCKCHIP_THUNDER_BOOT`, and it creates an artificial
threads-max limit that is not suitable for generic linux.
Signed-off-by: nyanmisaka <nst799610810@gmail.com>
* Correcting the Perf unit's interrupt source number to 66 as stated in the Allwinner A10User manual to resolve conflict with Uart 2 which actually uses interrupt source number 3
* rename to all lower case and add in sun4i-emac dma support. Note that support only exist in the driver for the rx channel
* Ensure that sun4i patches actually implemented
Signed-off-by: Ryan L <ryestar101@gmail.com>
* Apply performance monitoring unit IRQ number patch to uboot
* Add patch for resolving the perf unit irq number to the edge kernel
Signed-off-by: Ryan L <ryestar101@gmail.com>
* Add sun4i emac ethernet DMA support to the edge kernel
Signed-off-by: Ryan L <ryestar101@gmail.com>
* Append missing .patch so that our patch is found and works properly
Signed-off-by: Ryan L <ryestar101@gmail.com>
* Apply EMAC DMA support and PMU fix to edge kernel
---------
Signed-off-by: Ryan L <ryestar101@gmail.com>
armbian-config uses debconf-apt-progress which is part of debconf.
libtext-iconv-perl is needed to prevent the following error, when
uninstalling a package through debconf-apt-progress:
Can't locate object method "new" via package "Text::Iconv" (perhaps you forgot to load "Text::Iconv"?) at /usr/share/perl5/Debconf/Encoding.pm line 65, <GEN2> line 2.
dpkg: error processing package nfs-common (--configure):
After that, the script successfully loads the DTB file from
two possible paths and overlays. If the file does not exist,
then the default file will be uploaded.
Use the DTB name from the u-boot default for the third
iteration if the provided DTB name was not found. Also
delete the duplicate of the previous iteration.
In the second iteration, we check the path that does not
contain the vendor folder. But this will also be done if
the real path contains the vendor's folder and the vendor's
name is contained in the DTB file name as
fdtfile=allwinner/sun50i-*-bananapi-*.dtb.
However, overlays will not be available.
Remove "allwinner/" from all board configurations that
use the boot script mechanism so that the script works
correctly.
Signed-off-by: The-going <48602507+The-going@users.noreply.github.com>
Inform the user about the actual file that will be uploaded
on the first iteration. Inform the user that it cannot be
downloaded only once if the first iteration was unsuccessful.
Set the expected default path to /boot/dtb/allwinner and
start analyzing if the DTB file is not found in this path.
The folder with the vendor's name is now part of the path,
not part of the file name. This greatly facilitates the analysis
of various possible options in a simple u-boot shell.
- we've some smelly stuff in write_uboot_platform for some families that we'd rather catch early
- implement small syntax fixes in setup_write_uboot_platform
- In line with 35352fa288 (bump to rkr4.1)
- "Copypasta from rockchip-rk3588.conf family file -- we _really_ gotta find a better way!"
- I'm not proud of having done it this way, but now it broke (two vendor debs are produced in different versions) so lets fix it
this commit reverts commit ca2dc35e555e7043de585f4e46123d8fbd2b5a21
from mainline kernel because rk3288 requires clock gating to
assert irq on usb device plug in and thus proceed with
reset-on-wake quirk
GenBook is a rk3588 based laptop from coolpi.
The armbian can boot from a usb disk, with pre installed
mainline u-boot on SPI Nor flash.
Signed-off-by: Andy Yan <andyshrk@163.com>
- name:Update ${{ matrix.tool.VERSION_VAR}} in script
# @TODO Make sure that the version is actually higher, not lower (the 'latest' tag does not neccessarily mean that the version is higher!)
@@ -106,13 +99,6 @@ jobs:
body:|
Bump [${{ matrix.tool.PROJECT_NAME}}](https://github.com/${{ matrix.tool.PROJECT_NAME }}) from ${{ steps.get-version-current.outputs.version_current }} to ${{ steps.get-version-latest.outputs.version_latest }} by bumping `${{ matrix.tool.VERSION_VAR}}` in `${{ matrix.tool.VAR_FILE}}`.
<details><summary><b>Release notes</b></summary>
<p><em>Sourced from <a href="https://github.com/${{ matrix.tool.PROJECT_NAME }}/releases">${{ matrix.tool.PROJECT_NAME }}'s releases</a>.
<br>Please note that this only shows the release notes for the latest release.</em></p>
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
# Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go
UEFI_EDK2_BOARD_ID="nanopc-cm3588-nas" # This _only_ used for uefi-edk2-rk3588 extension; cm3588-nas was introduced in v0.12 of edk2-porting/edk2-rk3588
function post_family_tweaks__cm3588_nas_udev_naming_audios() {
display_alert "$BOARD" "Renaming CM3588 audio interfaces to human-readable form" "info"
@@ -42,17 +43,73 @@ function post_family_config__cm3588_nas_use_mainline_uboot() {
declare -g BOOTDELAY=1 # Wait for UART interrupt to enter UMS/RockUSB mode etc
declare -g BOOTSOURCE="https://github.com/u-boot/u-boot.git" # We ❤️ Mainline U-Boot
declare -g BOOTBRANCH="tag:v2024.10"
declare -g BOOTPATCHDIR="v2024.10"
# Don't set BOOTDIR, allow shared U-Boot source directory for disk space efficiency
declare -g BOOTBRANCH="tag:v2025.01"
declare -g BOOTPATCHDIR="v2025.01"
declare -g BOOTDIR="u-boot-${BOARD}" # do not share u-boot directory
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # Disable stuff from rockchip64_common; we're using binman here which does all the work
# Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
# Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
# Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
@@ -51,10 +51,44 @@ function post_family_config_branch_edge__nanopct6_use_mainline_uboot() {
}
}
# U-boot 2025.04+ can detect and set fdtfile automatically across T6 and T6-LTS boards.
# So if using mainline u-boot, unset BOOT_FDT_FILE to let u-boot handle it.
# That way, both variants can boot from the same image; lets keep the -lts board file for vendor kernel/u-boot.
function post_family_config_branch_edge__t6_and_t6_lts_auto_dtb_name_via_uboot_detection() {
unset BOOT_FDT_FILE
}
function pre_config_uboot_target__nanoptc6_patch_uboot_dtsi_for_ums() {
[[ "${BRANCH}" != "edge" ]] && return 0
display_alert "u-boot for ${BOARD}" "u-boot: add to u-boot dtsi for UMS" "info" # avoid a patch, just append to the dtsi file
cat <<- EOD >> arch/arm/dts/rk3588-nanopc-t6-u-boot.dtsi # Append to the t6 u-boot dtsi file with stuff for enabling gadget/otg/peripheral mode
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
# Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go
# The upstream DTs (kernel 6.13 / u-boot 2025.01) are in flux and different
# let's just patch to normalize the SD and eMMC order as in mainline Linux DT: https://github.com/torvalds/linux/blob/master/arch/arm64/boot/dts/rockchip/rk3588s-nanopi-r6.dtsi#L14-L15
function pre_config_uboot_target__r6c_patch_uboot_dtsi_for_sd_emmc_order() {
display_alert "u-boot for ${BOARD}" "u-boot: add to u-boot dtsi for SD=mmc0 and eMMC=mmc1" "info"
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
# Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go
# The upstream DTs (kernel 6.13 / u-boot 2025.01) are in flux and different
# let's just patch to normalize the SD and eMMC order as in mainline Linux DT: https://github.com/torvalds/linux/blob/master/arch/arm64/boot/dts/rockchip/rk3588s-nanopi-r6.dtsi#L14-L15
function pre_config_uboot_target__r6s_patch_uboot_dtsi_for_sd_emmc_order() {
display_alert "u-boot for ${BOARD}" "u-boot: add to u-boot dtsi for SD=mmc0 and eMMC=mmc1" "info"
BOOTPATCHDIR="v2025.01" # but all changes are done in this board file, not patches
BOOTCONFIG="odroid-m1-rk3568_defconfig"
BOOTDIR="u-boot-${BOARD}" # do not share u-boot directory
@@ -21,10 +21,6 @@ BOOTDIR="u-boot-${BOARD}" # do not share u-boot directory
# The overlays for this board are prefixed by 'rockchip-rk3568-hk' (see for example patch/kernel/archive/rockchip64-6.6/overlay/rockchip-rk3568-hk-i2c0.dts)
# A better equivalent to patching a defconfig, do changes to .config via code.
# For UMS/RockUSB to work in u-boot, &usb_host0_xhci { dr_mode = "otg" } is required. See 0002-board-rockchip-ODROID-M1-override-kernel-DT-for-xhci-otg-dr_mode.patch
function post_config_uboot_target__extra_configs_for_odroid-m1() {
@@ -119,6 +137,13 @@ function post_config_uboot_target__extra_configs_for_odroid-m1() {
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
# Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go
# Note: whitespace/newlines are significant; BOOT_SUPPORT_SPI & BOOT_SPI_RKSPI_LOADER influence the postprocess step that runs for _every_ target and produces rkspi_loader.img
# Note: whitespace/newlines are significant; BOOT_SUPPORT_SPI & BOOT_SPI_RKSPI_LOADER influence the postprocess step that runs for _every_ target and produces rkspi_loader.img
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
@@ -35,3 +36,28 @@ function post_family_config_branch_edge__rock-3a_use_mainline_uboot() {
# Note: whitespace/newlines are significant; BOOT_SUPPORT_SPI & BOOT_SPI_RKSPI_LOADER influence the postprocess step that runs for _every_ target and produces rkspi_loader.img
unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already
&usbdrd_dwc3_0 { status = "okay"; dr_mode = "otg"; };
UBOOT_BOARD_DTSI_OTG
}
# "rockchip-common: boot SD card first, then NVMe, then SATA, then USB, then mmc"
# On rockpro64, mmc0 is the eMMC, mmc1 is the SD card slot https://github.com/torvalds/linux/blob/master/arch/arm64/boot/dts/rockchip/rk3399-rockpro64.dtsi#L15-L16
function pre_config_uboot_target__rockpro64_patch_rockchip_common_boot_order() {
declare -a rockchip_uboot_targets=("mmc1" "nvme" "scsi" "usb" "mmc0" "pxe" "dhcp" "spi") # for future make-this-generic delight
display_alert "u-boot for ${BOARD}/${BRANCH}" "u-boot: adjust boot order to '${rockchip_uboot_targets[*]}'" "info"
sed -i -e "s/#define BOOT_TARGETS.*/#define BOOT_TARGETS \"${rockchip_uboot_targets[*]}\"/" include/configs/rockchip-common.h
# declare -g UEFI_MOUNT_POINT_SKIP_FSTAB="yes" # If we leave the /boot/efi in fstab, systemd hangs waiting for it; FIXED: by adding mtools pkg
enable_extension "grub-with-dtb" # important, puts the whole DTB handling in place.
@@ -29,8 +27,8 @@ enable_extension "grub-with-dtb" # important, puts the whole DTB handling in pla
declare -g BOARD_FIRMWARE_INSTALL="-full"
function post_family_config_branch_sc8280xp__jhovolds_wip_kernel() {
declare -g KERNEL_MAJOR_MINOR="6.12" # Major and minor versions of this kernel.
declare -g KERNELBRANCH='branch:wip/sc8280xp-6.12' # @TODO: this is down to 27 patches, from hundreds back in the day. Considering merging this with default arm64 uefi kernel
declare -g KERNEL_MAJOR_MINOR="6.13" # Major and minor versions of this kernel.
declare -g KERNELBRANCH='branch:wip/sc8280xp-6.13' # @TODO: this is down to 23 patches, from hundreds back in the day. Considering merging this with default arm64 uefi kernel
declare -g LINUXCONFIG="linux-${ARCH}-${BRANCH}" # for this board: linux-arm64-sc8280xp
display_alert "Set up jhovold's kernel ${KERNELBRANCH} for" "${BOARD}" "info"
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.