Ricardo Pardini
c51b973217
drop UBOOT_USE_GCC, KERNEL_USE_GCC, ATF_USE_GCC completely
...
- also `CRUST_USE_GCC`
- those don't serve any purpose and cause confusion
2026-01-14 20:08:21 +01:00
Ricardo Pardini
6889c8c0a4
drop find_toolchains, SKIP_EXTERNAL_TOOLCHAINS and all $toolchain PATH injections
...
- we've had SKIP_EXTERNAL_TOOLCHAINS=yes for ~5 years now
- drop all usages, mostly through `find_toolchains()`
- drop all manual PATH env injections (we've centralized if ever needed)
2026-01-14 20:08:21 +01:00
EvilOlaf
e145d689c2
mainline: bump to rc5
2026-01-14 13:24:10 +01:00
Viacheslav Bocharov
43a786c327
JetHub: Update JetHub D2 uboot to 2025.04
...
Signed-off-by: Viacheslav Bocharov <adeep@lexina.in>
2026-01-13 09:00:36 +03:00
EvilOlaf
77dc5f7154
sunxi: switch current and edge back to auto bumping
2026-01-12 19:19:14 +01:00
Ricardo Pardini
1b1e24e4d7
genio: add alsa-ucm-conf + ucm2 config for working audio in userspace
...
- from Collabora's Debian 13 impl
2026-01-12 19:18:57 +01:00
TuAFBogey
b954ad893f
Fix copyright year to 2026
2026-01-12 13:18:53 +01:00
TuAFBogey
9b12e0b978
Address final review: declare -g consistency, function keyword, move compile_optee inside hook, use echo in write_uboot_platform
2026-01-12 13:18:53 +01:00
TuAFBogey
60650cf015
Address rpardini review round 2: major cleanup
...
- Use SERIALCON instead of systemd override for serial console
- Remove fdtfile from bootenv (use BOOT_FDT_FILE instead)
- Remove ATF_COMPILER (arm64 default)
- Inline OPTEE variables into fetch_sources_tools hook
- Remove empty family_tweaks() and family_tweaks_bsp()
- Use only debug mode for ATF build
- Build fiptool in atf_custom_postprocess
- Move header creation inline into uboot_custom_postprocess
- Use <<- heredoc for Python script
2026-01-12 13:18:53 +01:00
TuAFBogey
c9f8f3caef
Address rpardini review: remove BTF patches, use full kernel config, use fetch_from_repo for OP-TEE
2026-01-12 13:18:53 +01:00
TuAFBogey
f24801893c
Use EXTRA_BUILD_DEPS for OP-TEE dependencies instead of runtime apt-get
2026-01-12 13:18:53 +01:00
TuAFBogey
9ecd83082e
fix: address review comments
...
- Rename board config to .csc (no maintainer)
- Remove duplicate board name from comment
- Change verbosity from 7 to 1 (default)
- Remove ATF patch, use ATF_SKIP_LDFLAGS_WL instead
- Move CONFIG_RD_* to kernel config (fix artifact caching)
- Fix networking comment (chrony -> NetworkManager)
- Replace -j12 with -j8
- Add fdt addr/resize after DTB reload on overlay error
2026-01-12 13:18:53 +01:00
TuAFBogey
c6e8c5b680
board: add Nuvoton MA35D1 NuMaker IoT board support
...
Add support for Nuvoton MA35D1 SoC family:
- ARM Cortex-A35 dual-core @ 800MHz + Cortex-M4 heterogeneous
- NuMaker IoT MA35D16F90 board (512MB DDR)
- Vendor kernel 5.10.140 from Nuvoton GitHub
- TF-A (BL2/BL31) + OP-TEE (BL32) + U-Boot (BL33) boot chain
- SD card boot support
Vendor kernel modules (pending):
Nuvoton provides closed-source Verisilicon licensed drivers:
- dcultrafb.ko (DCUltra display controller)
- ma35d1-vc8000.ko (H.264 video codec)
- galcore.ko (GPU driver)
These modules require exact kernel vermagic match:
5.10.140-vendor-nuvoton-ma35d1 SMP mod_unload aarch64
Nuvoton is preparing custom builds of these modules for Armbian.
Signed-off-by: TuAFBogey <uguvenen@gmail.com>
2026-01-12 13:18:53 +01:00
Werner
19446b0d1f
sunxi: bump current and edge to latest minor ( #9229 )
2026-01-12 08:30:24 +01:00
Ricardo Pardini
9e7f6dc264
rockchip64: patch ATF 2.13 to build on gcc-14+ (ref rk3399's m0)
...
- move patches into version folder (`atf-rockchip` -> `atf-rockchip/v2.13`)
- also prepares patches for v2.14 since I had them anyway
- keeps using v2.13 for now, someone can test v2.14 later and bump; I've no rk3399
- new patches:
- add `-no-pie` to rk3399's m0 build LDFLAGS (since upstream doesn't honor M0_LDFLAGS)
- bump PMSURAM_RSIZE from 8k to 16k to avoid overflows - confirmed working by Amazingfate
- build tested OK on jammy/bookworm/trixie/noble/resolute
- resolute's gcc 15 and it still builds
2026-01-10 11:13:47 +01:00
Igor Pecovnik
d7a4c1b3ab
Fix broken patch due to upstream changes, drop edge kernel as its not used
2026-01-08 21:38:09 +01:00
EvilOlaf
e38d028214
sunxi-current: bump to 6.12.64
2026-01-08 17:58:43 +01:00
juanesf
2c025ef929
sun55iw3.conf bump u-boot to v2026.01 (final)
...
Based on (https://github.com/armbian/build/pull/9179 )
2026-01-08 17:58:17 +01:00
EvilOlaf
176f78fdc4
Let's do edge only for now
2026-01-08 15:38:23 +01:00
EvilOlaf
ba7701a991
sunxi: bump edge to 6.18.4 current to 6.12.64
2026-01-08 15:38:23 +01:00
Ricardo Pardini
91b2454496
genio: export (from the image) bins needed for flashing
...
- so regular armbian/build users can get them in `output/images`
2026-01-08 12:30:24 +01:00
Ricardo Pardini
8c727d0f3a
genio: implement complete OPTEE/u-boot/TF-A (x2)/FIP build
...
- genio: implemented complete bootloader build according to Collabora's scripts
- main difference from other similar stack is that TF-A builds final FIP, not u-boot
- thus I first disable ATF build initially, then build it in hook
- alternatively: we could build our own FIP, but that would diverge too much
- declare functions inside hook so hashing does all the versioning automatically
- write_uboot_platform() is a no-op; don't write bootloader to image
- ensure bootloader _builds_ across bookworm/trixie/jammy/noble/resolute
- keep in mind original code is for bookworm (tested to build and work on trixie)
- need to juggle the `-Wl,` prefix on LDFLAGS
- this is not mainline ATF and it does "stupid with flags"
- genio: u-boot: always clean optee & first atf build before building
- so one can switch build host with `DOCKER_ARMBIAN_BASE_IMAGE` (UFS...)
- genio: u-boot: using `armbian/mtkbin` repo
- outputs fip.img & lk.bin directly to output(/images)
- disk images do _not_ contain any bootloader
2026-01-08 12:30:24 +01:00
Ricardo Pardini
f60c1dcf12
genio: collabora: bump to Collabora's 6.19-rc4
2026-01-08 12:09:02 +01:00
Ricardo Pardini
11c1424b97
genio: collabora: drop KERNELPATCHDIR override, be standard
...
- patches should go into `patch/kernel/archive/genio-6.19` as per default
2026-01-08 12:09:02 +01:00
Ricardo Pardini
9109d4897c
genio: Collabora's kernel is at 6.19-rc1; use SHA1 instead of branch
...
- so it does not keep changing beneath us as happened with 6.18 -> 6.19
2026-01-08 12:09:02 +01:00
Ricardo Pardini
0771c181b4
genio: don't force INSTALL_HEADERS=yes for collabora branch
...
- while it is nice having them, and also a good test of being able to
build the headers "board-side", enabling it by default should be a
global change, and not specific to genio
2026-01-08 12:09:02 +01:00
tabris
1e7201fc06
kernel filogic - use extension_hook_opt_out for ntsync
2026-01-08 12:07:56 +01:00
tabris
abde4ddc94
filogic kernel family - update comments around armbian_kernel_config__enable_various_filesystems to use extension_hook_opt_out
2026-01-08 12:07:56 +01:00
Ricardo Pardini
0f5db61567
u-boot: bump u-boot to v2026.01 (final) for boards that had -rcX
...
- cm3588-nas: u-boot: bump v2026.01-rc2 -> v2026.01-rc
- khadas-vim3l: u-boot: bump v2026.01-rc2 -> v2026.01
- mekotronics-458-4x4: : bump v2026.01-rcX -> v2026.01
- nanopct6: u-boot: bump v2026.01-rc2 -> v2026.01
- nanopi-r6c: u-boot: bump v2026.01-rc2 -> v2026.01
- nanopi-r76s: u-boot: bump v2026.01-rcX -> v2026.01
- odroidhc4: u-boot: bump v2026.01-rc2 -> v2026.01
- odroidn2: u-boot: bump v2026.01-rcX -> v2026.01
- rock-5b: u-boot: bump v2026.01-rc2 -> v2026.01
- this is tedious. time for `UBOOT_MAJOR_MINOR`?
2026-01-08 12:07:20 +01:00
Ricardo Pardini
df970b8bee
mvebu64: re-revitalize espressobin - can build (under noble) again
...
- try to squash all shellcheck warnings
- drop toolchain stuff we don't have anymore
- then it becomes obvious there was a spurious `/` in `ATF1`
- requires & works with `compile_atf()`'s LDFLAGS handling in https://github.com/armbian/build/pull/9162
- build-tested on noble (default Docker image)
- does NOT build on trixie+:
- and apparently never will: https://lists.trustedfirmware.org/archives/list/tf-a@lists.trustedfirmware.org/thread/A2NLSTWUNTSH3WAGOW4LGATKYYCTZXXZ/
- the whole thing has been dropped from TF-A CI, so it's effectively junk now
- also fixes `macchiatobin-doubleshot`, but that is .eos anyway
- that builds on noble & trixie
- Fixes: 81ad50f257
2026-01-08 12:06:19 +01:00
Ricardo Pardini
d3ddbb0c31
rk35xx-legacy: opt-out of driver_rtl8723cs so it builds again
...
- legacy is slated for removal and only useful to flash the RTL8169 MAC EEPROM
2026-01-07 13:02:43 +01:00
Ricardo Pardini
64581d1b39
sun55iw3-syterkit: opt-out of driver_rtl8723cs so it builds again
2026-01-07 13:02:43 +01:00
tabris
7ea0983321
framework and kernel config hooks - split FS modules into a separate hook function to allow customization
2026-01-07 11:14:30 +01:00
Ricardo Pardini
1bc8f20f77
mainline-kernel: bump 6.19 to 6.19-rc4
2026-01-06 21:47:54 +01:00
Ricardo Pardini
bacd8dff88
rockchip64_common: bump edge 6.18 -> 6.19, current 6.12 -> 6.18
2026-01-06 21:47:54 +01:00
Paolo Sabatino
329ec0626a
rockchip: bump current kernel to v6.18
2026-01-06 18:13:52 +01:00
Werner
4648add145
radxa-cubie-a5e: maint and fixes ( #9152 )
...
* radxa-.cubie-a5e: bump atf branch to v4 and fix fdt file name
* bump u-boot to rc5
because why not?
* potential fix for corrupted images
2026-01-03 14:19:06 +01:00
Jianfeng Liu
d2f6c47f6d
Revert "uefi-loong64: *temporary* hold uefi-loong64 edge at 6.18"
...
This reverts commit e0f38cb9be .
2025-12-30 10:28:26 +01:00
Ricardo Pardini
61a5d307ce
uefi(-all): legacy 6.6 -> 6.12; current/cloud 6.12 -> 6.18
2025-12-29 16:33:29 +01:00
Ricardo Pardini
e0f38cb9be
uefi-loong64: *temporary* hold uefi-loong64 edge at 6.18
...
- while amazingfate works 6.19 separately
2025-12-29 16:33:29 +01:00
Ricardo Pardini
f4a269f095
uefi_common: bump edge 6.18 -> 6.19 (x86/arm64/risc64/loongarch64)
2025-12-29 16:33:29 +01:00
Ricardo Pardini
91ea87dacb
meson64_common: bump current 6.12 -> 6.18
2025-12-29 12:50:02 +01:00
Ricardo Pardini
fc62cbaf1d
mainline-kernel: bump 6.19 to 6.19-rc3
2025-12-29 12:50:02 +01:00
Ricardo Pardini
186e5b4375
mainline-kernel: bump 6.19 to 6.19-rc2
2025-12-29 12:50:02 +01:00
Ricardo Pardini
6467d998a4
meson64_common: meson64 bump edge to 6.19
2025-12-29 12:50:02 +01:00
Ricardo Pardini
1a63a5a22c
Revert "meson64: bump edge to 6.19; current 6.12 -> 6.18 ( #9069 )" (wrong squash)
...
This reverts commit f1b5acc9a1 .
2025-12-29 12:50:02 +01:00
Werner
f1b5acc9a1
meson64: bump edge to 6.19; current 6.12 -> 6.18 ( #9069 )
...
* meson64-6.19: copy 6.18 patches into 6.19
* meson64_common: meson64 bump edge to 6.19
* meson64-6.19: rework patches for 6.19-rc1
- fix jethome
- fix socinfo patch, re-enable another
- seems like no longer needed, upstream code has changed
- disable more breaking patches. Buildable state
- remove upstreamed patch
- adjust patching_config kernel versions
* mainline-kernel: bump 6.19 to 6.19-rc2
* meson64-6.19: rewrite patches against 6.19-rc2
* meson64-6.19: use llm suggestion for overlay README
* mainline-kernel: bump 6.19 to 6.19-rc3
* meson64-6.19: pick dwc-MPS-slowness-fix patch that missed the bus from 6.18
* meson64-6.19: rewrite against 6.19-rc3
* meson64_common: bump `current` 6.12 -> 6.18
* meson64-6.18: rewrite against 6.18.2
---------
Co-authored-by: Ricardo Pardini <ricardo@pardini.net>
2025-12-29 12:30:26 +01:00
Ricardo Pardini
1a4d46d3d7
mekotronics(all): bump generic mainline u-boot to v2026.01-rc5
...
- also enable CONFIG_ROCKCHIP_MASKROM_IMAGE for quick development
- fix conflicting vendor hook
- only hack the vendor u-boot down to next-dev-v2024.03 for vendor/legacy
2025-12-28 12:38:02 +01:00
Ricardo Pardini
d40033d2a0
mekotronics: u-boot: generic mainline u-boot 2025.01
...
- this is hidden under `MEKO_USE_MAINLINE_UBOOT=yes`
- also disable PLYMOUTH then
2025-12-28 12:38:02 +01:00
Mecid
b62d0ae4f1
Genio: Change INSTALL_ARMBIAN_FIRMWARE to 'yes' ( #9134 )
2025-12-28 12:05:11 +01:00