guillaume_g has quit [Quit: Konversation terminated!]
pbergin has quit [Remote host closed the connection]
prabhakarlad has joined #u-boot
ilunev has joined #u-boot
redbrain has quit [Read error: Connection reset by peer]
redbrain has joined #u-boot
<ilunev>
hello! I'm trying to run Yocto, meta-odroid on c4, hardkernel linux 5.10 and u-boot 2015.01. It seems like u-boot is looking for reserved-memory/linux,cma node but it's absent in meson64_odroidc4.dtb; apparently, it's a relict from 4.9 kernel. Boot logs stops after booti command, nothing happens. U-boot fails its rsvmem command before that.
akaWolf has quit [Ping timeout: 248 seconds]
<mps>
repeating my question I asked two days ago: I don't understand how the BOOT_RETRY works with BOOT_RETRY_TIME=60? I enabled it but then u-boot doesn't make pause in prompt to select kernel and just boot default
<mps>
could someone help me to understand how to 'fix' this
<kallisti5[m]>
Running into needing it again 😆 Not exposing the relocated address makes gdb debugging a pain
<kallisti5[m]>
Heinrich mentioned the output potentially interrupting EFI menus.. would a simple change from log_info to EFI_PRINT address his concerns?
<marex>
mps: do you have bootdelay set to -2 maybe ?
<marex>
kallisti5[m]: EFI -> talk to xypron
tre has quit [Remote host closed the connection]
<mps>
marex: hmm, I think no, lets look
<mps>
marex: CONFIG_BOOTDELAY=2
akaWolf has joined #u-boot
<mps>
marex: this boot delay works, but not timeout in extlinux.conf
<marex>
mps: and the bootretry variable is set ?
<marex>
(in your u-boot env, use printenv to check)
<mps>
I have few kernels set in extlinux.conf and want to select them
<mps>
marex: I have BOOT_RETRY_TIME=60 in .config
<marex>
but isnt boot retry something like ... reset after N seconds ?
<marex>
its to prevent the system from being stuck in u-boot shell, no?
<mps>
marex: as I understand yes
<j`ey>
if the boot fails.. it retries, in your case the boot doesnt fail, so it never retries or waits
<mps>
I have a board on which u-boot sometimes doesn't detect emmc card
<marex>
due to HW bug ?
<marex>
mmc dev X || reset ; mmc rescan || reset ?
<mps>
marex: I have no idea, but could bad emmc on the board
<marex>
if mmc dev X fails, you should have $? set, so you can handle the fault in u-boot shell
<mps>
j`ey: when u-boot doesn't detect emmc after some time trying netboot it stops at u-boot prompt
<mps>
and I type 'boot' on prompt and it boots then
<marex>
use bootcounter and altbootcmd ?
mmu_man has quit [Ping timeout: 256 seconds]
<mps>
marex: I need this board on remote site and don't like idea to drive 50-60 kilometers just to reboot in case it is stuck
<marex>
mps: enable WDT and write your bootscript such that it always hits 'reset' if it fails ?
<marex>
mps: the boot retry is likely also a good idea
<mps>
I thought that BOOT_RETRY and BOOT_RETRY_TIME could be workaround
<marex>
bootcmd is the default boot command , that will run after bootdelay elapsed the first N times, the N+1th time the altbootcmd will run and that can be some sort of fallback
<marex>
N is set in bootcounter limit
<mps>
marex: could altbotcmd be set in extlinux.conf
<marex>
it should be part of u-boot env
<marex>
if you load extlinux.conf from possibly failing emmc, it is already too late
<mps>
ah, yes
<mps>
stupid thing is that board reads and load u-boot from card and after that says card couldn't be found
<mps>
that will hard because it happens just sometimes
<mps>
about once in week or two
<mps>
two weeks*
josem has quit [Ping timeout: 252 seconds]
GNUtoo has quit [Ping timeout: 240 seconds]
GNUtoo has joined #u-boot
<xypron>
kallisti5[m]: kallisti5 I think writing the memory address in cmd/bootefi.c with log_debug() is the best choice. You then can switch the message on and of with the log command. But I would not like to write messages while an EFI application is running.
<xypron>
kallisti5 The next stage bootloader can write the entry address to the kernel itself.
torez has joined #u-boot
monstr has quit [Remote host closed the connection]
<Rusty_Almighty>
Anyone know how to get the HUSH cli to get a DHCP address without attempting to TFTP load a file.
<Rusty_Almighty>
?
___nick___ has quit [Ping timeout: 248 seconds]
vagrantc has joined #u-boot
<Tartarus>
setenv autoload no, Rusty_Almighty
workman76 has quit [Remote host closed the connection]
torez has quit [Quit: torez]
<Rusty_Almighty>
@tartarus: Thank you. I'm going to go test this now.
sbach has quit [Read error: Connection reset by peer]
sbach has joined #u-boot
littlebobeep has quit [Ping timeout: 240 seconds]
Rusty_Almighty has quit [Quit: Leaving.]
Rusty_Almighty has joined #u-boot
<Rusty_Almighty>
I have one more question, is there a utility to convert a script in a text file into a properly formatted bootcmd or maybe a regular expression to convert a string for the setenv command to delay evaluation of variables until the variable is ran as part of the command?
<Rusty_Almighty>
Or am I going about setting the bootcmd all wrong?
<Tartarus>
There are many solutions
<Tartarus>
You can write a script, and run it past mkimage to make what's often called a "boot.scr" file
<Tartarus>
Or you can modify the default environment, and build the board, flash that
<Rusty_Almighty>
@Tartarus thank you for the response again. I'm looking for something I can quickly use in setenv for rapid prototyping.