#nixos-aarch64 on 2020-10-24

2019-09-27 17:31 gchristensen changed the topic of #nixos-aarch64 to: Get access to the NixOS Community aarch64 build box: https://github.com/nix-community/aarch64-build-box ... todo: https://hydra.nixos.org/jobset/nixpkgs/trunk#tabs-jobs build box status: https://status.nixos.org/grafana/d/5LANB9pZk/per-instance-metrics?orgId=1&refresh=30s&var-instance=147.75.77.190

00:42 monk has left #nixos-aarch64 ["Error from remote client"]

00:49 monk has joined #nixos-aarch64

01:01 monk has left #nixos-aarch64 ["Error from remote client"]

01:02 justanotheruser has quit [Ping timeout: 260 seconds]

01:08 justanotheruser has joined #nixos-aarch64

01:17 veleiro has joined #nixos-aarch64

01:22 monk has joined #nixos-aarch64

01:26 h0m1 has quit [Ping timeout: 246 seconds]

01:28 h0m1 has joined #nixos-aarch64

02:01 jdnixx-M1 has joined #nixos-aarch64

02:15 h0m1 has quit [Ping timeout: 272 seconds]

02:17 h0m1 has joined #nixos-aarch64

02:48 orivej has joined #nixos-aarch64

02:55 <colemickens> cool. uboot+extlinux I guess automatically tries subsequent entries until one boots

02:56 <samueldr> yeah

03:20 <colemickens> my naive shot at mainline with uboot on rpi4 was apparently unbootable.

03:21 <samueldr> care to explain what you tried?

03:22 <samueldr> could maybe get you thinking in rubber ducky ways :)

03:22 <samueldr> but maybe first: started from a known good defconfig?

03:22 <samueldr> (or outright kernel config)

03:25 monk has left #nixos-aarch64 ["Error from remote client"]

03:32 monk has joined #nixos-aarch64

03:36 <colemickens> the only thing I did was change the kernel from _rpi to _5_9. it was a very naive attemt

03:41 monk has left #nixos-aarch64 ["Error from remote client"]

03:42 <samueldr> right

03:42 <samueldr> there might be kernel options that are new and not turned on by our setup

03:42 <samueldr> or did the kernel outright not start?

03:42 <samueldr> I guess it didn't if u-boot continued through

03:43 <colemickens> that was my assumption, but I don't have uart/hdmi right now, so I can't confirm that.

03:43 <samueldr> ah

04:03 orivej has quit [Ping timeout: 240 seconds]

04:03 monk has joined #nixos-aarch64

04:46 zupo has joined #nixos-aarch64

04:50 zarel_ has joined #nixos-aarch64

04:51 zarel has quit [Ping timeout: 260 seconds]

05:34 rajivr has joined #nixos-aarch64

06:02 <veleiro> hello

06:07 monk has left #nixos-aarch64 ["Error from remote client"]

06:08 cole-h has quit [Ping timeout: 258 seconds]

06:20 monk has joined #nixos-aarch64

06:54 h0m1 has quit [Quit: WeeChat 2.9]

06:55 h0m1 has joined #nixos-aarch64

07:23 h0m1 has quit [Quit: WeeChat 2.9]

07:23 h0m1 has joined #nixos-aarch64

07:41 orivej has joined #nixos-aarch64

08:09 orivej has quit [Ping timeout: 260 seconds]

08:32 orivej has joined #nixos-aarch64

08:39 NekomimiScience has quit [Ping timeout: 272 seconds]

08:42 NekomimiScience has joined #nixos-aarch64

08:44 feepo has joined #nixos-aarch64

08:45 feepo has quit [Excess Flood]

08:45 feepo has joined #nixos-aarch64

08:46 <patagonicus> Had to fix the same test failures for gnutls as well (slightly different patch file since they are in a different directory), but now my system built successfully at 630f19b3ef9cebef1b85bc5c8c6e1fe2a3c6aa01. :)

08:47 claudiii has joined #nixos-aarch64

09:13 zupo has quit [Ping timeout: 256 seconds]

09:15 zupo has joined #nixos-aarch64

09:45 orivej has quit [Ping timeout: 256 seconds]

10:03 monk has left #nixos-aarch64 ["Error from remote client"]

10:12 monk has joined #nixos-aarch64

10:13 alpernebbi has joined #nixos-aarch64

10:27 alp has joined #nixos-aarch64

10:47 alp has quit [Ping timeout: 272 seconds]

10:58 heywoodlh has quit [Quit: ZNC 1.8.2 - https://znc.in]

10:58 heywoodlh has joined #nixos-aarch64

11:19 mvnetbiz_9 has joined #nixos-aarch64

11:23 mvnetbiz_9 is now known as mvnetbiz_

11:30 alp has joined #nixos-aarch64

11:47 Darkmatter66 has joined #nixos-aarch64

11:49 Darkmatter66_ has quit [Ping timeout: 260 seconds]

11:54 alp has quit [Ping timeout: 272 seconds]

11:58 monk has left #nixos-aarch64 ["Error from remote client"]

12:15 dstzd has joined #nixos-aarch64

12:15 dstzd has quit [Changing host]

12:17 dstzd has quit [Client Quit]

12:18 dstzd has joined #nixos-aarch64

12:18 mvnetbiz_ has quit [Ping timeout: 260 seconds]

12:18 dstzd has quit [Client Quit]

12:21 dstzd has joined #nixos-aarch64

12:21 dstzd has quit [Changing host]

12:23 monk has joined #nixos-aarch64

12:27 dstzd has joined #nixos-aarch64

12:27 dstzd has quit [Changing host]

12:38 mvnetbiz_9 has joined #nixos-aarch64

12:47 mvnetbiz_9 has quit [Quit: Bye!]

13:18 superherointj has joined #nixos-aarch64

13:27 <das_j> clever: Do you know if the partition types are relevant for the pi?

13:55 orivej has joined #nixos-aarch64

13:58 <das_j> nvm, it's c

14:03 cole-h has joined #nixos-aarch64

14:11 zupo has quit [Ping timeout: 265 seconds]

14:14 zupo has joined #nixos-aarch64

14:24 <das_j> umm rpi4 stage 1, there are neither sd* devices nor mmc* devices. What module am I missing?

14:29 monk has left #nixos-aarch64 ["Error from remote client"]

14:41 monk has joined #nixos-aarch64

14:51 alp has joined #nixos-aarch64

15:12 superherointj has quit [Quit: Leaving]

15:40 hmpffff has joined #nixos-aarch64

15:51 zupo has quit [Ping timeout: 258 seconds]

15:52 zupo has joined #nixos-aarch64

16:08 zupo has quit [Ping timeout: 260 seconds]

16:16 zupo has joined #nixos-aarch64

16:30 h0m1 has quit [Quit: WeeChat 2.9]

16:31 h0m1 has joined #nixos-aarch64

16:48 <clever> das_j: there is a small list of types it will accept, for the pi4

16:53 monk has left #nixos-aarch64 ["Error from remote client"]

16:54 monk has joined #nixos-aarch64

17:02 monk has left #nixos-aarch64 ["Error from remote client"]

17:18 qyliss has quit [Quit: bye]

17:19 veleiro has quit [Ping timeout: 264 seconds]

17:22 qyliss has joined #nixos-aarch64

17:22 monk has joined #nixos-aarch64

17:42 ryantrinkle1 has quit [Quit: Leaving.]

17:43 ryantrinkle has joined #nixos-aarch64

17:46 alp has quit [Ping timeout: 272 seconds]

17:49 monk has left #nixos-aarch64 ["Error from remote client"]

17:55 monk has joined #nixos-aarch64

18:25 alp has joined #nixos-aarch64

18:27 <das_j> clever: Yeah, the boot hangs when you have a type 83 partition (even when booting from sd. stage -2 doesn't even start)

18:27 <das_j> s/sd/network/

18:30 <clever> das_j: yeah, ive also seen that a typecode 0x80 (ext4?) causes it to halt, rather then skip

18:30 <das_j> not sure if bug or feature ;)

18:30 <clever> id say its a bug

18:30 <clever> because an empty fat32 tagged 0xc, makes it skip

18:48 Raito_Bezarius has joined #nixos-aarch64

18:49 <Raito_Bezarius> Hi there, I'm trying to install NixOS on a Pine A64-LTS

18:49 <Raito_Bezarius> I'm wondering how to use mmcblkXbootY partitions to store the bootloader there

18:49 <Raito_Bezarius> Also, I'm getting test failures while installing mtd-utils, did anyone succeed to build them?

18:50 <Raito_Bezarius> The wiki seems not to say anything about separated boot partitions of the form mmcblkXbootY

18:54 monk has left #nixos-aarch64 ["Error from remote client"]

18:57 <samueldr> Raito_Bezarius: simply put, Allwinner A64 can't use the "boot" hardware partitions of eMMC

18:57 <Raito_Bezarius> even using U-Boot on the SPI flash?

18:57 <samueldr> that would be a bit different then, I assumed you wanted to install u-boot to them :)

18:57 <samueldr> to the "boot" hardware partition

18:58 <Raito_Bezarius> Hm, you're right

18:58 <Raito_Bezarius> But I'm okay with U-Boot on the SPI flash

18:58 <samueldr> let me rephrase my previous assertion: the boot flow of the allwinner a64 SoC itself will not use the boot hw partitions on eMMC

18:58 <Raito_Bezarius> I guess it makes the boot partitions useless

18:58 <Raito_Bezarius> So I'll just use the non-boot partitions I suppose

18:58 <samueldr> but I guess that if u-boot can, you could do funny stuff in u-boot with them

18:58 <Raito_Bezarius> :D

18:58 <Raito_Bezarius> u-boot the u-boot

18:59 <samueldr> maybe something like a recovery OS stashed in there!

18:59 <Raito_Bezarius> Indeed, that'd be interesting :-)

18:59 <samueldr> but yeah, you're limited to SPI -> eMMC -> SD as far as the SoC will boot from

19:00 <Raito_Bezarius> At the moment, I'm trying to flash the NOR SPI from Linux mainline, but mtd-utils won't get compiled using NixOS, is this known?

19:00 <samueldr> not by me

19:00 <Raito_Bezarius> I'm reading: https://github.com/ayufan-rock64/linux-package/blob/master/root-rock64/usr/local/sbin/rock64_reset_emmc.sh to try to see if I can do something similar without mtd-utils

19:00 <samueldr> and at the time I was looking into that I used u-boot to flash

19:00 <samueldr> a few years back

19:01 <samueldr> my notes: https://nixos.wiki/wiki/NixOS_on_ARM/PINE_A64-LTS#SPI_NOR_flash

19:01 <samueldr> false

19:01 <samueldr> I flashed from FEL lol

19:01 <samueldr> I *think* mainline could do it too, but it might be missing configuration options

19:02 <Raito_Bezarius> Ha

19:02 <Raito_Bezarius> I'm hunting for the rock64 write_to_spi_flash script

19:02 <Raito_Bezarius> https://gitlab.com/ayufan-repos/rock64/linux-build/-/blob/master/recipes/flash-spi.md

19:03 <Raito_Bezarius> which seems to perform something similar to what I want

19:03 <samueldr> though I'd like to see installation from linux instructions, as I think it'd be better than installing from FEL or U-Boot

19:03 <Raito_Bezarius> (the 4. part the script rock64_XXXX)

19:07 <Raito_Bezarius> I'll try mtdutils 2.1.2 rather than 2.1.1

19:14 orivej has quit [Ping timeout: 265 seconds]

19:15 <Raito_Bezarius> okay, it didn't work, I'll just disable doCheck

19:34 monk has joined #nixos-aarch64

19:49 cole-h has quit [Ping timeout: 272 seconds]

19:56 alpernebbi has quit [Quit: alpernebbi]

20:00 orivej has joined #nixos-aarch64

20:01 jumper149 has joined #nixos-aarch64

20:09 zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]

20:10 zupo has joined #nixos-aarch64

20:12 zupo_ has joined #nixos-aarch64

20:14 alp has quit [Ping timeout: 272 seconds]

20:15 zupo has quit [Ping timeout: 240 seconds]

20:21 orivej has quit [Ping timeout: 258 seconds]

20:29 justanotheruser has quit [Ping timeout: 260 seconds]

20:39 <Raito_Bezarius> samueldr: did you do something specific when installing nixos?

20:39 <Raito_Bezarius> I followed the wiki

20:39 <samueldr> years ago, not really

20:39 <Raito_Bezarius> and have U-Boot on the SPI (from the Linux installation, I have found a way)

20:39 <Raito_Bezarius> But the U-Boot is complaining about not having any partition table on the eMMC

20:39 <samueldr> it worked *enough* with the mainline kernel IIRC

20:39 <samueldr> at all?

20:40 <Raito_Bezarius> yes, but maybe because it's me who is provoking the stuff :D by using btrfs on /dev/mmcblk1

20:40 <samueldr> which booting scheme are you using? UEFI or Extlinux?

20:40 <Raito_Bezarius> Extlinux atm

20:40 <Raito_Bezarius> But I'm okay to go UEFI and I prefer UEFI

20:40 <samueldr> if you're using extlinux, remember to set the partition as bootable!

20:40 <samueldr> UEFI it has to be a real ESP

20:40 <Raito_Bezarius> hm

20:40 <samueldr> the partition with the extlinux files

20:40 <samueldr> which the default assumptions are the rootfs

20:40 <Raito_Bezarius> and it cannot mmc1blkboot0 right?

20:41 <Raito_Bezarius> for the esp?

20:41 <samueldr> nope

20:41 <Raito_Bezarius> okay

20:41 <Raito_Bezarius> understood the issue, thanks samueldr !

20:41 <samueldr> *blkboot* are "special" in many ways

20:41 <Raito_Bezarius> it seems like so

20:41 <Raito_Bezarius> i found the anti write tech to be neat

20:41 <samueldr> to reduce confusion, unless the SoC tells you to use them, assume they don't exist :(

20:42 <Raito_Bezarius> samueldr++

20:42 <{^_^}> samueldr's karma got increased to 290

20:42 <Raito_Bezarius> Yeah, makes sense samueldr

20:42 <samueldr> though, I'm not saying "don't use them", but that's a big YMMV :)

20:43 <Raito_Bezarius> Of course

20:44 monk has left #nixos-aarch64 ["Error from remote client"]

20:45 <samueldr> almost there :) https://stuff.samueldr.com/screenshots/2020/10/20201024164441_48p4dvsdq4siqdkylsh.png

20:45 <Raito_Bezarius> it's showing a black screen for me? :p samueldr

20:47 monk has joined #nixos-aarch64

20:47 <samueldr> nah, that's a transparent png because I didn't look into the issue with my screenshot script

20:47 * samueldr re-shoots

20:48 <samueldr> https://stuff.samueldr.com/screenshots/2020/10/20201024164800_4fmpq2i9ydo4qbzw0mp.png

20:49 <Raito_Bezarius> oh <3

20:49 <Raito_Bezarius> looks really nice

20:49 <Raito_Bezarius> I'm seriously interested into buying a PinePhone just for Mobile NixOS

20:50 <Raito_Bezarius> but that'll be when I'll have time to do some relevant and useful contributions

20:51 <samueldr> unless you have some kind of plan, don't "for Mobile NixOS"

20:51 <samueldr> at least not right now

20:51 <Raito_Bezarius> I'm not planning to use this device as a daily driver but rather as an experimental device for adding stuff to Mobile NixOS

20:51 <Raito_Bezarius> Though, I have no real plan

20:51 <Raito_Bezarius> I'm just curious

20:52 <samueldr> yeah, if you want to just add stuff to Mobile NixOS before using it, it can be done through the QEMU VM, that's how I'm doing half of the development for boot stuff that's generic

20:59 <Raito_Bezarius> hmm, i was wondering what was the state regarding LTE and stuff like that

20:59 <Raito_Bezarius> But, true

21:00 <samueldr> on the Pinephone, just implemented LTE last week

21:00 <samueldr> well, verified data

21:00 <samueldr> I don't have calls/sms on my plan

21:01 <samueldr> (and I know there are further issues to be solved upstream too)

21:01 <Raito_Bezarius> upstream == on pine side?

21:01 <samueldr> on the pine community side

21:01 <samueldr> meaning that it's not issues I expect would be unique to NixOS

21:01 <Raito_Bezarius> I see

21:02 <samueldr> but I was able to use data, only "issue" was that the current ModemManager doesn't have an in-depth APN database like Android does

21:02 <samueldr> just now I'm thinking and wondering if it can be imported from AOSP in some manner

21:03 <Raito_Bezarius> wouldn't most mobile ISP nowadays send the APN info in some way automatically?

21:03 monk has left #nixos-aarch64 ["Error from remote client"]

21:05 * samueldr shrugs

21:05 <samueldr> if it does, ModemManager didn't seem aware

21:07 <samueldr> CAUTION: browsers might not like rendering 35577 lines of text: https://android.googlesource.com/device/sample/+/master/etc/apns-full-conf.xml

21:07 <samueldr> https://android.googlesource.com/device/sample/+log/master/etc/apns-full-conf.xml

21:07 <samueldr> considering all this

21:07 <samueldr> I'd wager that they don't

21:09 <samueldr> https://source.android.com/devices/tech/config/update

21:09 <samueldr> >> Carriers can update their Access Point Name (APN) information and their carrier-specific configuration settings (CarrierConfig) in the Android Open Source Project (AOSP).

21:09 <samueldr> so uh, yeah

21:11 tilpner has quit [Remote host closed the connection]

21:12 tilpner has joined #nixos-aarch64

21:14 zupo_ has quit [Ping timeout: 260 seconds]

21:16 zupo has joined #nixos-aarch64

21:17 tilpner has quit [Ping timeout: 240 seconds]

21:19 zupo has quit [Client Quit]

21:23 monk has joined #nixos-aarch64

21:29 justanotheruser has joined #nixos-aarch64

21:29 tilpner has joined #nixos-aarch64

21:38 justanotheruser has quit [Ping timeout: 260 seconds]

21:54 alp has joined #nixos-aarch64

21:57 monk has left #nixos-aarch64 ["Error from remote client"]

21:58 monk has joined #nixos-aarch64

22:00 <Raito_Bezarius> :D

22:13 Darkmatter66 has quit [Ping timeout: 260 seconds]

22:14 Darkmatter66 has joined #nixos-aarch64

22:17 jumper149 has quit [Ping timeout: 246 seconds]

22:19 jumper149 has joined #nixos-aarch64

22:42 hmpffff has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]

22:46 justanotheruser has joined #nixos-aarch64

23:02 alp has quit [Ping timeout: 272 seconds]