RSS Bot

Description Update odroidxu4-current kernel to 5.4.246 How Has This Been Tested? [x] Reboot of my Odroid HC1 Checklist: [x] My code follows the style guidelines of this project [x] I have performed a self-review of my own code [ ] I have commented my code, particularly in hard-to-understand areas [ ] I have made corresponding changes to the documentation [x] My changes generate no new warnings [ ] Any dependent changes have been merged and published in downstream modules View the full article

Contrary to traditional software development, data is more important than code in machine learning. Building a high-performing model requires using reliable, precisely labelled data but poor-quality data is not always obvious. View the full article

Backport updates from Linux 6.3. In my testing this has no ill effects on SDIO 8822CS and fixes USB 8821CU dmesg | grep rtw [ 8.684999] rtw_8821cu 4-1:1.0: Firmware version 24.11.0, H2C version 12 [ 9.426555] usbcore: registered new interface driver rtw_8821cu lsmod | grep rtw88 rtw88_8821cu 16384 0 rtw88_8821c 81920 1 rtw88_8821cu rtw88_usb 20480 1 rtw88_8821cu rtw88_core 131072 2 rtw88_usb,rtw88_8821c mac80211 458752 2 rtw88_core,rtw88_usb cfg80211 331776 2 rtw88_core,mac80211 Checklist: [ ] My code follows the style guidelines of this project [ ] I have performed a self-review of my own code [ ] I have commented my code, particularly in hard-to-understand areas [ ] I have made corresponding changes to the documentation [X] My changes generate no new warnings [ ] Any dependent changes have been merged and published in downstream modules View the full article

wireguard can be built successful on rk3566/rk3568/rk3588 socs. so we can remove these configurations. tested built orangepi5-plus and rock-3a. tested module load on orangepi5-plus. View the full article

Description Update odroidxu4-current kernel to 5.4.245 How Has This Been Tested? [x] Reboot of my Odroid HC1 Checklist: [x] My code follows the style guidelines of this project [x] I have performed a self-review of my own code [ ] I have commented my code, particularly in hard-to-understand areas [ ] I have made corresponding changes to the documentation [x] My changes generate no new warnings [ ] Any dependent changes have been merged and published in downstream modules View the full article

orangepi5-plus.wip: initial Orange Pi 5 plus support using Xunlong vendor u-boot kernel dts is waiting for https://github.com/armbian/linux-rockchip/pull/38 to be merge i had tested on my opi5 plus(jammy/cli), it can boot from emmc, both 2 eth ports and nvme is working fine. View the full article

Description The 32-bit allwinner boards are known to hang with 6.1.30 kernel. On investigation, I was able to narrow the issue down to net-phy-Support-yt8531c.patch and upstream change for lib/mapple tree for some reason these two changes don't seem to be playing along very well. I am trying to get the reason for it, but so far even after enabling low level debugging, hung task detection, earlyprintk and countless such debugging option, I was not able to get more output from the kernel or to get the kernel into kgdb mode. This merge request is raised as a workaround for fixing the hang issue so that other task depending on the same can be unblocked. Jira reference number AR-1764 How Has This Been Tested? [X] Created an image with the change and booted on Nanopiduo2 Checklist: [X] My code follows the style guidelines of this project [X] I have performed a self-review of my own code [X] I have commented my code, particularly in hard-to-understand areas [X] have made corresponding changes to the documentation [X] My changes generate no new warnings [X] Any dependent changes have been merged and published in downstream modules View the full article

Description Requirement for commit c83b11cf1486ade72cfb92447e9212e654b43d8f so iw command is available on the default images. Checklist: [ ] My code follows the style guidelines of this project [x] I have performed a self-review of my own code [ ] I have commented my code, particularly in hard-to-understand areas [ ] I have made corresponding changes to the documentation [ ] My changes generate no new warnings [ ] Any dependent changes have been merged and published in downstream modules View the full article

Remove stmmac change from net-phy-Support-yt8531c.patch as its causing 32-bit allwinner boards to hang on boot Description The 32-bit allwinner boards are known to hang with 6.1.30 kernel. On investigation, I was able to narrow the issue down to the above mentioned patch. The patch is supposed to add support for yt8531c, but it was also modifying stmmac ethernet driver to add a fix for realtek chip. This doesn't make sense to me and removing that code solved the issue. Jira reference number AR-1764 How Has This Been Tested? Created an bookworm current image for nanopi duo2 with and without stmmac part removed. [X] Removing the stmmac part of the patch results into board booting successfully [X] Restoring the stmmac part of the patch causes the board to hang Checklist: [X] My code follows the style guidelines of this project [X] I have performed a self-review of my own code [X] I have commented my code, particularly in hard-to-understand areas [X] I have made corresponding changes to the documentation [X] My changes generate no new warnings [X] Any dependent changes have been merged and published in downstream modules View the full article

Description Mainline kernel 6.3 and above implements yt8531 driver. This pull request removes the legacy driver patch and fix the device tree for Orange Pi 4 LTS and other devices to use that driver. Note: this PR has a "dependency" upon this other PR (https://github.com/armbian/build/pull/5290) - these two PRs are unrelated, but the PR-5290 bug does not allow me to easily test without having conflicts later. This is just a note, everything should go on flawlessy once PR-5290 is merged. Jira reference number AR-1774 How Has This Been Tested? [x] Tested dtb on a live Orange Pi 4 LTS system [ ] Tested on Orange Pi R1 LTS (@schwar3kat) Checklist: [x] My code follows the style guidelines of this project [x] I have performed a self-review of my own code [x] I have commented my code, particularly in hard-to-understand areas [x] I have made corresponding changes to the documentation [x] My changes generate no new warnings [ ] Any dependent changes have been merged and published in downstream modules View the full article

Description As per subject, implements same fix as https://github.com/armbian/build/pull/5216, but for kernel 6.3 How Has This Been Tested? [x] Patch applies [x] Bare dtb tested on live system Checklist: [x] My code follows the style guidelines of this project [x] I have performed a self-review of my own code [x] I have commented my code, particularly in hard-to-understand areas [x] I have made corresponding changes to the documentation [x] My changes generate no new warnings [x] Any dependent changes have been merged and published in downstream modules View the full article

pipeline: repo-reprepro: run clearvanished and check before adding to repo pipeline: repo-reprepro: run clearvanished and check before adding to repo View the full article

repo: split image apt .list deployment into image-early and image-late repo: split image apt .list deployment into image-early and image-late split image apt .list deployment into image-early (after rootfs extract, before building) and image-late (after building) rationale: having the repo enabled during the image build can cause 'apt upgrade' and others to do inconsistent stuff depending on the (possibly random) state of the repo View the full article

prepare-host: check_host_has_enough_disk_space(): don't check if dir does not exist; fixes #5246 prepare-host: check_host_has_enough_disk_space(): don't check if dir does not exist; fixes #5246 View the full article

distro-agnostic: don't assume armbian.list exists when seding for beta distro-agnostic: don't assume armbian.list exists when seding for beta View the full article

uboot: introduce hook post_config_uboot_target uboot: introduce hook post_config_uboot_target runs after other configuration changes, but before actually compiling a target useful to hack into .config for a specific target View the full article

rk3568-odroid edge: .config, remove ARCH_TEGRA and ARCH_SUNXI, since this is Rockchip kernel rk3568-odroid edge: update config, no actual changes rk3568-odroid edge: .config, remove ARCH_TEGRA and ARCH_SUNXI, since this is Rockchip kernel View the full article

BREAKING CHANGE: patches failing to apply now break the build. fixes #4958 BREAKING CHANGE: patches failing to apply now break the build. fixes #4958 also break on legacy process_patch_file() failure, remove EXIT_PATCHING_ERROR View the full article

Upstream link: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/drivers/net/phy/motorcomm.c?h=v6.3.5&id=4ac94f728a588e7096dd5010cd7141a309ea7805 How Has This Been Tested? No test preformed. Checklist: [ ] My code follows the style guidelines of this project [ ] I have performed a self-review of my own code [ ] I have commented my code, particularly in hard-to-understand areas [ ] I have made corresponding changes to the documentation [ ] My changes generate no new warnings [ ] Any dependent changes have been merged and published in downstream modules View the full article

Lets us put down the hammer. Cleaned 6.1, 6.3 and added 6.2 patch set. Simplified function: rtw88_driver This PR was created in order to ease the pain here https://github.com/armbian/build/pull/5276 and add support for the current edge branch 6.2. Checklist: [ ] My code follows the style guidelines of this project [ ] I have performed a self-review of my own code [ ] I have commented my code, particularly in hard-to-understand areas [ ] I have made corresponding changes to the documentation [X] My changes generate no new warnings [ ] Any dependent changes have been merged and published in downstream modules View the full article

Description The 32-bit Allwinner dtb files are under /boot/dtb and not under /boot/dtb/allwinner as expected by u-boot because of the current board config. Fixed dtb location by removing the allwinner/ prefix for nanopi-r1 and nanopi duo2 boards. This issue seems to have already reported on the forums as well - https://forum.armbian.com/topic/28117-nanopi-duo-2-hang-on-boot/ How Has This Been Tested? Tested by creating edge image with the changes included in the pull request. The stable image still doesn't boot because of the ongoing issue with the 6.1.30 kernel image, but 6.2.16 kernel from edge image works [x] Created edge image and booted on nanopi duo2 Checklist: [x] My code follows the style guidelines of this project [x] I have performed a self-review of my own code [x] I have commented my code, particularly in hard-to-understand areas [x] I have made corresponding changes to the documentation [x] My changes generate no new warnings [x] Any dependent changes have been merged and published in downstream modules View the full article

driver_rtl8723cs: completely (not partially) disable for bcm2711 and d1 families I think Igor half-disabled this for the release. No use having half of this View the full article

rock-5a: fix broken u-boot build + move from family to board file rock-5a: fix broken u-boot build by forcing build of u-boot.dtb before u-boot.itb rock-5a: move board-specific code from family file to board file with post_family_config_branch_legacy hook View the full article

For @pyavitz @adeepn appreciation, most patches failed to apply on 6.3, and a few on 6.2. View the full article

rpardini's bag-of-stuff incl fixes - early June/'23 I'm back at it, fixing and breaking stuff. Fixing: prepare-host: check_host_has_enough_disk_space(): don't check if dir does not exist; fixes #5246 distro-agnostic: don't assume armbian.list exists when seding for beta pipeline: repo-reprepro: run clearvanished & check before adding to repo Breaking: BREAKING CHANGE: patches failing to apply now break the build. fixes #4958 also break on legacy process_patch_file() failure, remove EXIT_PATCHING_ERROR repo: split image apt .list deployment into image-early (after rootfs extract, before building) and image-late (after building) rationale: having the repo enabled during the image build can cause 'apt upgrade' and others to do inconsistent stuff depending on the (possibly random) state of the repo Other stuff armbian-gaming customize launcher for @NicoD-SBC (v2) add image suffix -gaming uboot: introduce hook post_config_uboot_target, runs after other configuration changes, but before actually compiling a target View the full article