| ex-parrot | josch: nice! problem solved, for now :) | 00:08 |
|---|---|---|
| gsora | xz is a really nice piece of software. | 00:52 |
| - cobra (QUIT: Ping timeout: 240 seconds) (~cobra@user/Cobra) | 01:35 | |
| - xktr (QUIT: Ping timeout: 256 seconds) (~xktr@user/xktr) | 01:48 | |
| + xktr (~xktr@user/xktr) | 01:48 | |
| - mtm (QUIT: Ping timeout: 252 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:04 | |
| + cobra (~cobra@user/Cobra) | 02:11 | |
| - nsc (QUIT: Ping timeout: 255 seconds) (~nicolas@i5C74DD7E.versanet.de) | 03:15 | |
| + nsc (~nicolas@23-99-142-46.pool.kielnet.net) | 03:17 | |
| - murphnj (QUIT: Remote host closed the connection) (~murph@user/murphnj) | 03:21 | |
| - chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@ppp-2-87-3-184.home.otenet.gr) | 03:35 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:10 | |
| josch | gsora: i think quite a few people would disagree with you. :) You are aware about the controversies around the suitability of xz as a long term archival format? :) | 04:23 |
| josch | indeed in the past xz was blowing up with the -T0 option. I remember a testsuite of mine which, when executed on a 192 core arm64 ampere system, would routinely take down the whole VM stack as it was eating memory for breakfast on 192 xz threads -- apparantly that issue got fixed now | 04:26 |
| josch | the practical problem with this is reproducibility. The xz file format differs depending on whether you build with or without multithreated compression and if there is not enough ram, xz might decide to only compress with a single thread in which case the output would look different compared to when it was done on a machine with more ram | 04:27 |
| AbortRetryFail | hehe, multithreated | 04:34 |
| josch | whoops :) | 04:37 |
| josch | ex-parrot, pandora: thanks to your motivation i finally managed to start fixing reform.d.n: https://salsa.debian.org/reform-team/orchestrator/-/commit/57cfd9eae81a964d7d6fc788092f00dd62dbc04e | 06:39 |
| josch | ex-parrot: this should fix everything except debian-installer where i started to fix things in this commit: https://salsa.debian.org/reform-team/reform-debian-installer/-/commit/04a23072c85abc976b697831815d96404d95883d | 06:40 |
| josch | and the HTML pages still need to be fixed | 06:40 |
| josch | but that part is easy :) | 06:40 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@37.4.230.225) | 07:28 | |
| + akira (~akira@37.4.230.225) | 07:29 | |
| josch | minute: reform-flash-uboot is broken on the pocket reform because in 2024-03-15 we decided to name the u-boot binaries not flash.bin but after their dtb basepath like meson-g12b-bananapi-cm4-mnt-pocket-reform-flash.bin | 07:37 |
| josch | minute: i'm going to fix the affected u-boot repos and create new tags to align everything with what we agreed upon back in march, okay? | 07:38 |
| + chomwitt (~chomwitt@2a02:587:7a27:1100:1ac0:4dff:fedb:a3f1) | 08:15 | |
| josch | minute: and now that linux isn't broken anymore i'll also merke your a311d on pocket and rk3588 MRs | 08:30 |
| josch | minute: same goes for your a311d MR against reform-tools (wifi power fixes) | 08:31 |
| josch | in related news: reform-tools 1.45 was just released | 08:31 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@37.4.230.225) | 09:10 | |
| + akira (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 09:20 | |
| chartreu1e | Well after way way too long ignoring it and not getting around to it, finally replaced the mosfet again (and coil to be safe) and really hoping it doesn't blow again and just works :/ | 09:31 |
| chartreu1e | Just pulliong up the schematics again and looking for anything that would kill Q17 again before plugging it in, as I don't remember everything I checked last year | 09:31 |
| * chartreu1e -> chartreuse | 09:33 | |
| chartreuse | It lives! I have no clue why it died last time (after the initial fault/repair) but it powered up for the first time in probably a year | 09:38 |
| chartreuse | I guess now I have to catch up with a lot of changes, and make sure my batteries are somewhat balanced | 09:39 |
| josch | chartreuse: congratulations!!! \o/ | 10:13 |
| josch | chartreuse: will you start your system from scratch or try to upgrade your existing system? | 10:13 |
| chartreuse | I might try and upgrade it, though I'm not really sure how much has changed in the system image since then. Probably would be easiest to start again, though dunno | 10:16 |
| josch | chartreuse: if you can tell me the version of kernel and reform-tools then we can get an idea | 10:17 |
| chartreuse | I figure the majority of changes to the main image are related to the different cpu boards now available. | 10:17 |
| chartreuse | Was just trying to find where I put the screws for the bottom case lid >_>, I'll just start it up to confirm then | 10:18 |
| chartreuse | Alright I'm on 6.1.0-reform2-arm64 (sounds like the stock one and not one of my custom ones I was tinkering with), reform tools is 1.20. And on System Image v3 2023-04-19 | 10:20 |
| josch | chartreuse: from where do you get the system image version? | 10:21 |
| josch | chartreuse: what is your version of base-files? | 10:22 |
| chartreuse | The system image version was shown in the motd after logging in on the console. base-files is 12.4 | 10:23 |
| josch | oooh? are you by any chance running debian bookworm from reform.debian.net? | 10:25 |
| chartreuse | I'm not, I don't think that existed when I did this. It was a fresh install of the v3 system image from sometime last year or two, after my v2 image was getting a bit dated and figured it'd be easier to do a full upgrade. It's that image copied to nvme and booting from nvme | 10:26 |
| josch | i see | 10:27 |
| chartreuse | I do have the relatively newer uboot stuff that can boot straight without the card inserted | 10:27 |
| josch | but it looks like when you want to go the upgrade route instead of replacing your system, you have the choice of either keeping unstable or switching to debian bookworm | 10:27 |
| chartreuse | The base-files might be newer as I was keeping it up to date with apt after the system image | 10:27 |
| chartreuse | Ah we've settled on stable now? Yeah I'm still on a 12.4 unstable/sid | 10:28 |
| josch | chartreuse: no, debian bookworm is inofficial | 10:28 |
| josch | *unofficial | 10:28 |
| josch | i'm building all the reform stuff for bookworm and provide system images and an apt repo for it | 10:28 |
| josch | the official MNT images and repo are still tracking unstable | 10:29 |
| chartreuse | Makes sense yeah, Going stable wouldn't be terrible though haven't had much bad luck with unstable so far. Especially with how bookworm-backports is being kept fairly up to date on a lot of packages | 10:29 |
| chartreuse | I'll give a apt update apt upgrade a try, worst case I just flash a new image. I already backed up the files back when I put it away | 10:32 |
| chartreuse | Might just start from scratch to clean things up a bit more, since I was messing with box86/64 stuff so it's a multiarch setup | 10:33 |
| josch | chartreuse: if you do that, would you mind trying out the system images i'm just building? :) | 10:35 |
| chartreuse | I could try them on the sd card if you want some quick tests. Or are you looking for more a long term testing thing? | 10:35 |
| josch | just if things boot :) | 10:36 |
| josch | and i don't have an imx8mq anymore | 10:36 |
| josch | and we switched to the v15 cadence patch stack | 10:36 |
| josch | and i hope this didn't break anything | 10:36 |
| chartreuse | Sure, the one from https://reform.debian.net/images/ or something newer? | 10:36 |
| josch | yes, the one that will fall out here in about 30-40 minutes: https://source.mnt.re/reform/reform-system-image/-/jobs/4768 | 10:37 |
| josch | this is also with new u-boot binaries | 10:37 |
| josch | so just a "it boots" is already a win for me :) | 10:37 |
| josch | i can only test on a311d and ls1028a | 10:38 |
| chartreuse | Alright | 10:39 |
| chartreuse | Hmm lost wifi after updating (ath9k card) and doesn't show up in lspci or dmesg, odd | 11:18 |
| chartreuse | And even modprobe ath9k isn't finding it... odd | 11:19 |
| josch | do you have anything interesting in dmesg? | 11:19 |
| chartreuse | Just doing a cold boot again, but only thing that originally caught my eye was something about firmware for the imx dma controller not found falling back to device or such... | 11:21 |
| chartreuse | And huh a cold boot fixed it, must have been something with the reboot after updating | 11:21 |
| josch | interesting -- there is a similar issue with a311d but there the wifi setup is completely different | 11:21 |
| josch | great, https://source.mnt.re/reform/reform-system-image/-/jobs/4768 succeeded and now is only waiting to finish uploading the artifacts | 11:22 |
| josch | you will be able to download system images in another ~10 minutes | 11:23 |
| chartreuse | Yeah just saw that it was finally building after queueing for nearly 40 minutes | 11:24 |
| josch | i'm uploading stuff across the board for the past 8 hours or so :) | 11:24 |
| josch | i felt motivated to finally fix reform.d.n for the pocket | 11:25 |
| chartreuse | Just noticed that the runner is on Ampere, is it just for convienece that it's an Arm64 architecture? | 11:26 |
| chartreuse | Alright downloading the 8mq image | 11:33 |
| chartreuse | Just realized I have it flashed to emmc so I need to go and try and flick that switch again | 11:36 |
| josch | chartreuse: why? | 11:38 |
| josch | chartreuse: your u-boot on emmc will happily load the rest from sd-card | 11:38 |
| chartreuse | So it boots from the nvme without needing the sd card present. So I can use the sdcard slot for camera cards and such without needing to keep the card with the laptop | 11:39 |
| chartreuse | Oh okay | 11:39 |
| chartreuse | I thought you were wanting to test the uboot on the card | 11:39 |
| josch | if possible, yes | 11:39 |
| josch | but i'll not ask you to remove your heatsink :) | 11:39 |
| josch | that's too much effort :D | 11:39 |
| chartreuse | I just put it back on with fresh paste earlier, but I think I can reach that switch with a fine tool from the side | 11:40 |
| josch | you really don't have to :) | 11:40 |
| josch | i'll be happy with testing everything above u-boot | 11:41 |
| pandora | @josch: it started to build the image | 11:49 |
| pandora | But then died after a few hours | 11:49 |
| pandora | And killed waybar and vscode | 11:50 |
| pandora | I guess it ran out of memory | 11:50 |
| pandora | The log with 83k lines seems too much for mister-muffin | 11:50 |
| chartreuse | Is there a way from the uboot console to boot the sdcard instead of nvme, or do I just have to use the tool from linux to set that | 11:51 |
| chartreuse | Oh wait I guess it just booted it this time since it said starting mnt setup wizard | 11:52 |
| chartreuse | Oh this is fancy | 11:52 |
| chartreuse | josch: I'd say that's working fine, booted to the install wizard, I was able to create the accounts and then log in. What's the TUI based login/destop manager that is being used? | 11:54 |
| josch | chartreuse: tuigreet | 11:55 |
| josch | https://github.com/apognu/tuigreet | 11:55 |
| josch | chartreuse: thanks a lot for testing! | 11:56 |
| chartreuse | Looking through dmesg I do get the simple-framebuffer cannot reserve video memory at... message but I think that comes up on the other image too. As well as imx-bus-devfreq 32700000.interconnect: Missing property fsl,ddrc, skip scaling DRAM | 12:00 |
| - iank (QUIT: Quit: ZNC 1.8.2+deb2+b1 - https://znc.in) (~iank@fsf/staff/iank) | 12:01 | |
| + iank (~iank@fsf/staff/iank) | 12:01 | |
| chartreuse | Also mxsfb 30320000.lcd-controller: error -ENODEV: Cannot connect to bridge. And, [drm:drm_bridge_attach] *ERROR* failed to attach bridge /soc@0/bus@30800000/dsi@30a00000 to encoded None-31: -517 | 12:02 |
| chartreuse | I think those are all common to other reform images though and the system seems to work fine | 12:02 |
| josch | in 2022, Lukas said to that message "ah, that's just noise" | 12:03 |
| chartreuse | Figured as much, just thought I'd mention the errors spotted in. the Simplefb one comes with a stack trace so it stands out, but I believe that's normal on the other image too | 12:04 |
| chartreuse | ACTION commits some kind of sin by gaffer taping the bottom panel on until he finds the container with the screws in it | 12:21 |
| josch | i bought a box with 100 M2 screws in it as mine constantly got lost :D | 12:31 |
| chartreuse | Alright wifi broke again... I did have the sd card in the time it did work, but it definitly seemed to have booted the nvme root since it had the newer kernel... odd | 12:31 |
| chartreuse | Yeah, just harder to get small metric hardware in Canada it seems. I do have some M2 but they're pan head rather than counter sunk | 12:31 |
| josch | ah okay | 12:34 |
| josch | maybe it makes you feel better to learn that even in metric europe, getting M2 screws is quite the pain :D | 12:34 |
| josch | too small for most places to have them, so you have to order them online... | 12:35 |
| chartreuse | It's like getting #4 and smaller imperial screws. I do have a store where I can get #4 (slightly smaller than M3), but stuff like #3 to #1 are quite specialty | 12:36 |
| gsora | josch: just backlogged the xz conversation. Just to double check, you think xz is not suitable for long term archival because of its instability (of the file format)? | 13:06 |
| - mtm (QUIT: Ping timeout: 264 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:02 | |
| josch | gsora: no, personally i think that the problems that do exist are too small to be relevant in practice | 14:21 |
| josch | ex-parrot, pandora: reform.d.n status report: the repository is working again and linux 6.1 as well as 6.7 builds again: https://reform.debian.net/debian/pool/main/l/linux/ | 15:46 |
| josch | next challenge is the system image because it requires the reform-setup-wizard, reform-qcacld2 and tuigreet to be uploaded to backports | 15:46 |
| josch | the problem with tuigreet is, that it is a rust package and backporting rust can be challenging... | 15:47 |
| josch | the problem with reform-setup-wizard is, that we still have autopkgtest failures on loong64, riscv64 and s390x which need to be looked at: https://ci.debian.net/packages/r/reform-setup-wizard/ | 15:48 |
| josch | so if anybody wants to help with these items, please do :) | 15:48 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:08 | |
| + mtm- (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:23 | |
| - mtm (QUIT: Ping timeout: 264 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:24 | |
| - sevan (QUIT: Changing host) (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 16:27 | |
| + sevan (~sevan@user/venture37) | 16:27 | |
| sknebel | preparing for a very wide range of new SOMs? ;) | 16:34 |
| josch | sknebel: would you not call the existing eight reform configuration a "wide range"? :) | 16:40 |
| sknebel | I feel like s390x would be a noticable expansion :P | 16:40 |
| josch | oh | 16:41 |
| josch | that's what you mean | 16:41 |
| josch | the reform-setup-wizard is not just useful for reform products | 16:42 |
| josch | it being in debian means that it can also be used by others who want a first-boot wizard | 16:42 |
| josch | and s390x is a debian release architecture | 16:42 |
| sknebel | yeah, was just joking because the idea seemed funny | 16:53 |
| sevan | been so preocupied with wishing for a riscv reform that I never stopped to wonder how about a MIPS variant reform (loong) | 16:57 |
| sevan | these would be essential for hanging out in coffee shops imo | 16:58 |
| + wakest (m-a7d6fe@45.77.48.108) | 17:56 | |
| * bluerise_ -> bluerise | 18:16 | |
| [tj] | the olimex imx8mp som doesn't want to boot today | 18:17 |
| - gsora (QUIT: Remote host closed the connection) (~gsora@user/gsora) | 19:09 | |
| + gsora (~gsora@user/gsora) | 19:11 | |
| noam | minute: ended up showing the store link to at least five or six people by the end of it, and explaining it to three times that ;) | 19:16 |
| noam | I joked that I should bill you for the advertising lmfao | 19:16 |
| noam | (no, obviously not :) | 19:16 |
| [tj] | noam: mnt should get paid a fee for making you look cool :D | 19:17 |
| noam | Mmmmm | 19:17 |
| noam | I dunno | 19:17 |
| noam | At least one engineer at Systems Distributed went "Wait, you're _who_?? Can I get a picture?!" _before_ being showed the Reform :P | 19:17 |
| noam | So I think I'm already cool???? Apparently??? | 19:17 |
| noam | (How the hell did they know me?!) | 19:17 |
| [tj] | ah, everyone knows noam | 19:19 |
| [tj] | one of those cool dudes bowie sang about | 19:19 |
| noam | (That's rhetorical, I do know the answer, I'm just still confused at it) | 19:20 |
| - chomwitt (QUIT: Ping timeout: 264 seconds) (~chomwitt@2a02:587:7a27:1100:1ac0:4dff:fedb:a3f1) | 19:23 | |
| + chomwitt (~chomwitt@2a02:587:7a10:d00:1ac0:4dff:fedb:a3f1) | 19:25 | |
| + erle (~erle@user/erle) | 20:24 | |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:587:7a10:d00:1ac0:4dff:fedb:a3f1) | 20:27 | |
| - sevan (QUIT: Ping timeout: 255 seconds) (~sevan@user/venture37) | 20:35 | |
| + reform24917 (~ozel@2a02:8071:886:560:f267:c311:8e6b:246c) | 21:02 | |
| * reform24917 -> ozel | 21:04 | |
| ozel | iHi! should the mic input on the headset connector work out of the box? having troubles here | 21:05 |
| ozel | i'm on pocket reform | 21:05 |
| + sevan (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 21:17 | |
| - sevan (QUIT: Changing host) (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 21:17 | |
| + sevan (~sevan@user/venture37) | 21:17 | |
| - ozel (QUIT: Remote host closed the connection) (~ozel@2a02:8071:886:560:f267:c311:8e6b:246c) | 21:27 | |
| + ozel (~ozel@ip-176-199-211-075.um44.pools.vodafone-ip.de) | 21:51 | |
| ozel | ok, figured it out. this was missing: amixer -c 1 cset name='ADC Capture Switch' on | 21:52 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 22:01 | |
| + akira (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 22:02 | |
| + dominicm (45ee0d70e9@2a03:6000:1812:100::3e6) | 22:03 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 22:03 | |
| + akira (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 22:04 | |
| - akira (QUIT: Ping timeout: 246 seconds) (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 22:08 | |
| - ozel (QUIT: Ping timeout: 250 seconds) (~ozel@ip-176-199-211-075.um44.pools.vodafone-ip.de) | 22:09 | |
| + akira (~akira@2a01:599:a0c:db70:ec4c:47c2:506b:273f) | 22:09 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a0c:db70:ec4c:47c2:506b:273f) | 22:12 | |
| + akira (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 22:12 | |
| - gsora (QUIT: Remote host closed the connection) (~gsora@user/gsora) | 22:33 | |
| + gsora (~gsora@user/gsora) | 22:33 | |
| josch | ex-parrot: OSError: [Errno 28] No space left on device | 23:08 |
| josch | 40 G are not enough :) | 23:08 |
| + jacobk (~quassel@2600:1700:9e1e:7800:6c:fff2:2f4c:4383) | 23:14 | |
| minute | sevan: @riscv reform why not tinker with milk-v cm4 thingie then? | 23:27 |
| minute | noam: awesome, thank you, much appreciated | 23:30 |
| - akira (QUIT: Ping timeout: 268 seconds) (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 23:45 | |
| + akira (~akira@37.4.230.225) | 23:46 | |
| - erle (QUIT: Quit: Democracy must always be better armed than tyranny.) (~erle@user/erle) | 23:59 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!