frieder has quit [Remote host closed the connection]
<marex>
broonie: have you been at fosdem ?
<broonie>
not this year
<marex>
whew, at least I didn't miss saying hi to you
<broonie>
:)
mckoan is now known as mckoan|away
prabhakarlad has quit [Quit: Client closed]
apritzel has quit [Ping timeout: 248 seconds]
<hays>
can anyone help me discern what might be going wrong with this serial console output? https://h0c.us/v/sf84
prabhakarlad has joined #u-boot
<hays>
spl: partition error
<hays>
idbloader.img flashing at 32K offset and u-boot at 8M offset
<marex>
hays: I think one problem is almost 6 years old version of U-Boot, time to update ?
<hays>
marex: lol I'd like to
<hays>
but onfortunately this board is not mainlined yet
<hays>
I am following what I think are the same steps as the vendor's build script
<marex>
(do I need to spell it out?)
<hays>
not sure--are you saying you can't help on an old u-boot?
vagrantc has joined #u-boot
<marex>
it is very hard to provide support for ancient downstream vendor fork of U-Boot
<marex>
if you need that, ask the vendor for help, it was their decision to not cooperate with upstream, so I don't see why upstream should sponsor them and provide support for any of their downstream hacks
<marex>
(and yes, you might be in a crappy situation because of poor vendor)
<hays>
yeah. ok. well they will tell me to use the script. thanks anyway.
<sjg1>
Tartarus: I sent the splc series and there is a size delta for each commit in the pastebin. It doesn't include the patches brought in from splb, but I don't expect any delta with those
<mps>
peach-pi_defconfig for some reason set exynos-5420.dtb as default dtb instead of exynos5800-peach-pi.dtb. where I could search why this happens?
<mps>
sorry, exynos-smdk5420.dtb
<sjg1>
mps: It is OK to change it...the DT is added in the packaging step so it doesn't matter
<mps>
sjg1: but u-boot trying to load exynos-smdk5420.dtb instead of exynos5800-peach-pi.dtb with 'fdtdir' directive
<sjg1>
mps: It could be in include/config_distro_bootcmd.h - you could try added fdtfile to the environment. You'll need to change the header inclusion though, or perhaps move the whole env to a text environment
<mps>
sjg1: I solved problem by using 'fdt /dtbs/exynos5800-peach-pi.dtb' in extlinux.conf
<sjg1>
mps: OK good
<mps>
but I'm thinking why u-boot doesn't select proper one with fdtdir directive
<mps>
on all other board I use 'fdtdir' instead of 'fdt' in extlinux.conf
<Tartarus>
Well, it depends on what logic the peach-pi has to know the right one
<Tartarus>
Most commonly someone will write "findfdt" as a script in the environment, and run that to set fdtfile dynamically
<Tartarus>
and do "run findfdt;run distro_bootcmd" as the bootcommand
ldevulder has quit [Read error: Connection reset by peer]
ldevulder has joined #u-boot
<mps>
Tartarus: I don't understand why u-boot change this from expected one
<mps>
maybe because smdk5420 is base for peach-pi
<sjg1>
mps: Yes, they all use the same files apart from the DT. That was the point
<mps>
hm, looks like board file for peach-pi doesn't exits in board/samsung dir
<mps>
ok, now I understand
<mps>
btw, I found workaround for bug with peach-pi printing pressed key character, reported about 8-10 months ago, https://tpaste.us/nkZv
<mps>
simply commented this annoying printf
<Tartarus>
Ah, a change over time here, uh
<Tartarus>
This rings a bell
BobBeck has joined #u-boot
<Tartarus>
mps: can you bisect to when this changed?
MWelchUK1 has joined #u-boot
<mps>
I tried earlier and look like it is a long ago, years
<mps>
but I forgot where I stopped with bisect
<mps>
maybe could find some time on weekend and try again
<mps>
git log -p show this commit f3424c554c572761f35b00988deb8ed35d1870e3
<mps>
looks like it is there from beginning
<Tartarus>
Is this one of the things were we need a seemingly random driver to get the logic to come out right elsewhere?
MWelchUK1 has quit [Ping timeout: 246 seconds]
BobBeck has quit [Ping timeout: 268 seconds]
<mps>
to my inexperienced eyes this look related only to cros_ec