00:14 <bluecmd[m]> Anyone have any ideas how to solve this issue: On the vEdge the root disk is usually /dev/sda2 - but it can be any sdX as it is quite common that if a user has another USB disk inserted that other disk will be sda. Inside OpenWrt I can find the correct device name by looking at /sys/bus/usb/devices/1-1/1-1:1.0/host0/target0:0:0/0:0:0:0/block - but how can I make the kernel understand to boot this? Right now we hardcode

00:14 <bluecmd[m]> root=/dev/sda2 in our images which is not good.

00:15 <bluecmd[m]> In other distributions I would have added logic to find the root partition inside the initramfs, but from what I can tell that does not seem to be "the OpenWrt way" in that I can't find other platforms doing this already

06:23 <enyc> I guess posting on relevant forum thread helps!

06:28 <Znevna> o0

06:44 <KanjiMonster> bluecmd[m]: see how some other targets set the PARTUUID of the boot partition (which uniquely identifies it regardless which sdX it is)

06:52 <KGB-1> https://tests.reproducible-builds.org/openwrt/openwrt_sunxi.html has been updated. (0% images and 99.9% packages reproducible in our current test framework.)

07:11 <RoganDawes> hi flks. I am trying to get a firmware image for the Wink Hub v1, an imx28-based device. I have managed to build a kernel and initramfs, which has got me a shell. I have got the ubi* utils in the initramfs, but appear stuck getting a handle on the actual underlying flash - i.e. the MTD. I appear to have mtd support built in, as /proc/mtd exists. Unfortunately, it is empty. I have verified that

07:12 <RoganDawes> the partition table is in the device tree, under /sys/firmware/devicetree/base/apb@80000000/apbh@80000000/nand-controller@8000c000

07:12 <RoganDawes> any suggestions as to what I should do next?

07:15 <RoganDawes> I *don't* see any reference to the partitions or mtd in teh kernel dmesg, though.

07:17 <PaulFertser> RoganDawes: probably the flash device itself isn't, or the flash controller.

07:17 <PaulFertser> RoganDawes: I'd start with checking the flash controller in /sys/bus/platform/drivers , to see if the right driver is present and whether it's bound.

07:19 <RoganDawes> PaulFertser: yeah, I see nothing that references mtd or gpmi

07:20 <PaulFertser> RoganDawes: presence of the driver there would tell if you have it enabled in .config or not.

07:20 <RoganDawes> understood, thanks!

07:20 <PaulFertser> RoganDawes: for quick tests you can do "make kernel_menuconfig" and enable whatever you need with =y manually.

07:21 <RoganDawes> makes sense, thanks

07:21 <PaulFertser> "git grep <compatible-string>" in the Linux sources would tell you which driver is supposed to handle the device.

07:22 <RoganDawes> great, thanks

08:58 <Christophe[m]1> Is https://forum.openwrt.org/t/-/135378/9 the only attempt porting openwrt to the visionfive 2 board?

14:19 <KGB-1> https://tests.reproducible-builds.org/openwrt/openwrt_x86.html has been updated. (100.0% images and 100.0% packages reproducible in our current test framework.)

14:46 <mark22k> Hello, maybe a stupid question: How do I re-generate the translation template (and if necessary how do I adapt the old translation file) for a LuCi app? For example, if I inserted new string in the code and removed old ones. Is there a pre-made script for this or do I have to create the translation file by hand. By translation file I mean the .po

14:46 <mark22k> and .pot files.

19:33 <csharper2005> mesh between mt7602 and mt7985. 40 MHz BW doesn't work, only 20 MHz. Is it a known issue?

19:33 <csharper2005> latest snapshot on both devices