00:09 <Habbie> DWFREED

00:09 <Habbie> [ 0.000000] Linux version 6.1.69 (peter@syracuse) (mips64-openwrt-linux-musl-gcc (OpenWrt GCC 12.3.0 r24685-4a2ff73177) 12.3.0, GNU ld (GNU Binutils) 2.40.0) #0 SMP Sat Dec 23 11:52:06 2023

00:09 <Habbie> several happy lines later

00:09 <Habbie> [ 8.368044] 1180000000800.serial: ttyS0 at MMIO 0x1180000000800 (irq = 34, base_baud = 31250000) is a OCTEON�

00:09 <Habbie> and silence :D

00:09 <Habbie> i just remembered that besides 'bootelf' there was also 'bootoctlinux - Boot from a linux ELF image in memory'

00:24 <Habbie> lol [ 0.000000] Unknown kernel command line parameters "bootoctlinux earlyprintk", will be passed to user space.

00:56 <Habbie> oh, the poor ending at ttyS0 was just linux 'fixing' the baudrate to 115200 which is not what i was using to talk to uboot (9600)

00:56 <Habbie> now I die after [ 69.095473] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:

01:01 <Habbie> well, not die, it happens every few minutes, plus 50 lines :)

01:01 <slh> still sounds like progress ;)

01:01 <Habbie> it does!

01:01 <Habbie> i was really stuck on not having any kernel messages

01:11 <Habbie> so, turns out the kernel does not need the dts for working serial

01:28 <Habbie> [ 9.453413] Core0: Cache error exception:

01:28 <Habbie> exciting!

01:29 <Habbie> the internet tells me these are memory errors

01:30 <KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_lantiq.html has been updated. (95.9% images and 100.0% packages reproducible in our current test framework.)

03:38 <neggles> Habbie: oh god why octeon

03:38 <neggles> oh. juniper srx210. good luck...

03:41 <neggles> you will not be getting much further than where you are with an ER-L devicetree

03:41 <neggles> these chips are cursed

07:59 <dwfreed> Habbie: \o/

12:12 <Habbie> neggles, i get about as far with or without a tree at all; looks like the UART is found without one just fine

12:13 <Habbie> i can most likely improve (reduce) the time to crash by increasing my serial baud rate :D

12:25 <neggles> habbie: sort of... the initial/early boot console is handled by firmware routines that u-boot sets up

12:25 <neggles> octeon's haunted

12:26 <neggles> even uart doesn't work without one of the accelerators initialized

12:35 <Habbie> right, but linux sees ttyS0 and ttyS1 without a DTS

12:37 <Habbie> "sadly" it's xmas now and i won't have much time to continue

16:28 <skorpy[m]> We have scheduled an OpenWrt meetup at 37c3. Currently: Day 4 14:00 - 16:00 Saal E

16:28 <skorpy[m]> https://events.ccc.de/congress/2023/hub/en/event/openwrt-meetup_8kwr/

20:19 <swalker> updated openwrt/upstream, https://sdwalker.github.io/uscan/index.html

20:47 <KGB-2> 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.)