tilpner has quit [Remote host closed the connection]
tilpner has joined #nixos-aarch64
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<sphalerite>
thefloweringash: oh no, now I need one!
<sphalerite>
samueldr: did you ever encounter problems with the kernel freezing when vctrl-regulator is probed on rk3399?
<sphalerite>
(or thefloweringash for that matter)
<thefloweringash>
I haven't spent a lot of time on my rk3399 recently. Last time I tried I think I was about to boot successfully about 50% of the time.
<sphalerite>
that's probably an issue with the MMC
zupo has joined #nixos-aarch64
zupo has quit [Client Quit]
<thefloweringash>
words cannot express how disappointed I am with microSD cards, and how pleased I am to run the honeycomb from an intel nvme ssd
<gchristensen>
thefloweringash: pretty good price, too! wow!
<clever>
ive had several sd cards fail, and 2 uSD readers die just from being put into a pocket
<clever>
and last week, i discovered that every single uSD card in the house has vanished
<thefloweringash>
gchristensen: the lx2k boards? intel ssds? :-)
<gchristensen>
lx2k :)
Thra11 has joined #nixos-aarch64
<thefloweringash>
I got the "developer"-ish edition. The errata list is long and mildly frustrating.
<gchristensen>
ouch
t184256 has left #nixos-aarch64 ["Disconnected: Replaced by new connection"]
t184256 has joined #nixos-aarch64
<thefloweringash>
I think the new terminal stuff in nix 2.3 means I can't see the progress of my chromium build (or any ninja build)
<Thra11>
thefloweringash: because it chops off the [123/2345] progress bit?
<thefloweringash>
it seems to prevent it from showing at all, don't really know why
<Thra11>
I wonder if the nix output could be changed to always keep the nix progress overview as the bottom line, but allow the full build output to spam up the screen
<clever>
there is a mechanism in place, to report which phase the build is on
<clever>
let me find it...
<thefloweringash>
I'm using nix-build (ie, nix 1 commands) here, I'm guessing nix never considers the ninja status line complete so never prints it
<clever>
basically, you write json blobs to $NIX_LOG_FD
<thefloweringash>
I'm sure it's in the build phase, somewhere between 1 and 34000 targets complete
<clever>
but, you could add a hook, to make ninja report its progress directly to nix
<clever>
and then nix can integrate that into its own progress report
<samueldr>
haven't had that much RK3399 stuff running... most of it is in a state of waiting for something to be done
<Thra11>
I'm running rk3399 stuff :)
<samueldr>
for my tablet I need to figure out a good strategy for logging-in / locking it
<samueldr>
no keyboard!
<samueldr>
sphalerite: your coreboot thingy, did you flash through the suzy-qable?
<Thra11>
samueldr: Do you know what state gfx u-boot for PBP is in? I couldn't work out whether it should work/is being worked on/isn't being worked on by anybody yet.
<samueldr>
Thra11: no idea, AFAICT no one working on it
<samueldr>
that's one of the things I'm waiting on for the PBP, though not that much... since I intend to dogfood the u-boot stuff from mobile-nixos
<Thra11>
That explains why I couldn't find any sign of anyone working on it.
<samueldr>
I could be wrong, and it's possible that people are working on it without publicly stating so, too
<Thra11>
Kernel seems nice and stable now (since last few changes wifi has stopped crashing and battery info works)
<samueldr>
"Kernel", which, and on PBP?
<samueldr>
wondering how's the u-boot support with the gru platform
<samueldr>
and more specifically scarlet/dumo
<samueldr>
though at the same time, it should be left running depthcharge to dogfood that platform
<thefloweringash>
clever: is there precedent for integrating build systems into nix progress output? I'm trying to imagine what it would look like amongst the mysterious (1/2/3) triples
<clever>
thefloweringash: not sure, something that could be discussed in an issue on nix itself
<Thra11>
samueldr: Yes PBP. kernel built from tsys' 5.4.0-rc6 source tree
<samueldr>
great, I was thinking about finally setting it up
<sphalerite>
samueldr: yes, I did
<samueldr>
do you have links to relevant docs? I haven't searched recently, but in the past it was hard finding coherent info
<sphalerite>
samueldr: not really, but I'll write up what I did
<samueldr>
thanks
<sphalerite>
samueldr: pushed docs to the repo :)
<samueldr>
great
<samueldr>
and the few specific words at least are good ways to find more info, like now I know it's raiden_debug_spi
<samueldr>
and from your repo's sources I assume it's only in their fork :)
<samueldr>
hmmmm I think I'm hitting a serious bug with *something* with ext4 with the pixel 2 kernel
<samueldr>
I thought I saw a fix, but it's already in the kernel tree :/
<danielrf[m]>
just wondering if you've encountered this
<samueldr>
no, but that's not exactly surprising either
<samueldr>
msm is for qualcomm
<samueldr>
wondering what ispif is though
<danielrf[m]>
OK i'll begin debugging the next thing :)
<samueldr>
camera stuff
<samueldr>
most likely
<samueldr>
fun thing is, you also need to compare the same verbosity for the kernel as Android boots with, and if that kernel has the same messages on Android!
<samueldr>
I have seem OEM kernels that were sooooo chatty about buggy stuff
<danielrf[m]>
true--do you happen to know if a slow serial terminal will slow down the whole boot process?
<danielrf[m]>
just because it was sooo many messages :D
<samueldr>
pretty sure it does slow down the boot process
<samueldr>
oh
<samueldr>
I misread
<samueldr>
you can't change the speed really AFAIK
<samueldr>
but you can use something like --logfile for picocom
<samueldr>
which will log everything... to a fail :)
<danielrf[m]>
no, I didn't mean to change away from 115200
<samueldr>
oh, I actually read fine the first time then :)
<samueldr>
yeah, pretty sure it does slow down the boot process
<danielrf[m]>
it was just outputting a message every 0.000020 seconds
<samueldr>
oof
<danielrf[m]>
yeah--must be stuck in some relatively tight loop
<samueldr>
you could try to (temporarily) disable the option in your kernel build
<samueldr>
if it's well implemented behind an option
<samueldr>
you have a good string to grep for in the kernel source :)
<danielrf[m]>
yep. thanks for the help. i'll let you know if I make any more progress on this
<samueldr>
please do :)
<danielrf[m]>
looking forward to the day when all my devices will be running nixos!
<samueldr>
so glad to have pushed forward with those different devices at first, I can already see issues that would have passed me by