| josch | huh... building packages for four architectures is not as bad as i feared | 00:12 |
|---|---|---|
| josch | running reform-debian-packages usually takes 52-55 minutes | 00:12 |
| josch | building armhf, arm64, i386 and amd64 takes 1:21 hours | 00:13 |
| josch | main reason is that building i386 and amd64 happens in parallel to armhf and arm64 and that things like the kernel and arch:all packages are only built on arm64 but not on the other arches | 00:14 |
| minute | nice | 00:27 |
| minute | on the same machine or on both arm+x86 builders in parallel? | 00:27 |
| josch | i386 and amd64 happen on the intel box and do not build arch:all package nor are kernel images built (so that's really fast -- kernel goes brrrrt in 6 minutes) | 00:28 |
| josch | armhf and arm64 share the ampere vps and the builds there happen in parallel to the builds on the intel box | 00:29 |
| josch | (hope that included the answer to your question :D) | 00:30 |
| minute | awesome, debian's chromium 134 is here and it includes the patch to fix rk3588/panfrost in chromium wayland mode | 00:41 |
| minute | it's very fast | 00:42 |
| minute | josch: very cool, thanks for explaining | 00:42 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:11 | |
| - Gooberpatrol_66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66) | 01:18 | |
| + Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66) | 01:19 | |
| + robin_ (~robin@user/terpri) | 01:21 | |
| - robin (QUIT: Ping timeout: 246 seconds) (~robin@user/terpri) | 01:23 | |
| + robin__ (~robin@user/terpri) | 01:48 | |
| - robin_ (QUIT: Ping timeout: 252 seconds) (~robin@user/terpri) | 01:51 | |
| ^alex | hmm | 01:51 |
| ^alex | unplugging from the charger still shows `AC` in the log from the sysctl | 01:52 |
| - paperManu (QUIT: Ping timeout: 260 seconds) (~paperManu@107.159.71.33) | 02:06 | |
| minute | ^alex: known bug afaik | 02:36 |
| minute | blender 4.5 https://mastodon.social/@mntmn/114169521234955503 | 02:40 |
| + bkeys1 (~Thunderbi@173.186.16.211) | 03:39 | |
| - bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net) | 03:40 | |
| * bkeys1 -> bkeys | 03:40 | |
| - xktr (QUIT: Quit: leaving) (~xktr@user/xktr) | 04:03 | |
| - op_4 (QUIT: Remote host closed the connection) (~tslil@user/op-4/x-9116473) | 04:05 | |
| + op_4 (~tslil@user/op-4/x-9116473) | 04:05 | |
| - sts-q (QUIT: Ping timeout: 248 seconds) (~sts-q@2a11:fb80:3c2:5d01:e95d:3f09:e49d:b0ec) | 04:29 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 04:31 | |
| - Gooberpatrol_66 (QUIT: Ping timeout: 248 seconds) (~Gooberpat@user/gooberpatrol66) | 04:32 | |
| - Gooberpatrol66 (QUIT: Client Quit) (~Gooberpat@user/gooberpatrol66) | 04:33 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 04:33 | |
| - Gooberpatrol66 (QUIT: Client Quit) (~Gooberpat@user/gooberpatrol66) | 04:34 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 04:34 | |
| + sts-q (~sts-q@89.58.73.210) | 04:36 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:56 | |
| - mtm (QUIT: Ping timeout: 260 seconds) (~textual@47.202.75.129) | 06:03 | |
| + mtm (~textual@47.202.75.129) | 06:06 | |
| josch | minute: you had the problem with multi-arch:same version skews for the packages the MNT CI patches right? Can you try out adding this repository to your sources.list to see if upgrading from there fixing things for you? | 07:08 |
| josch | deb [trusted=yes] https://source.mnt.re/reform/reform-debian-packages/-/jobs/8002/artifacts/raw/repo reform main | 07:09 |
| + xktr (~xktr@user/xktr) | 07:17 | |
| josch | okay, shaved off 7 minutes by building the chroots natively in paralle on both runners | 09:20 |
| + gustav28 (~gustav@c-78-82-53-248.bbcust.telenor.se) | 10:02 | |
| - jacobk_ (QUIT: Ping timeout: 260 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 10:04 | |
| + jacobk (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 10:04 | |
| - sts-q (QUIT: Remote host closed the connection) (~sts-q@89.58.73.210) | 10:17 | |
| josch | and a bit more by using the gitlab 'depends' to not wait for all jobs of the setup stage to finish before the first job of the build stage can start | 10:44 |
| josch | minute: we can get more parallelization by cross-building for armhf on the intel xeon box. But that comes at the cost of cross-compilation sometimes being broken in unstable when new gcc or glibc gets uploaded. What do you prefer? | 10:45 |
| josch | we are at 1:12 hours right now (~54 minutes before) | 10:45 |
| - gordon1 (QUIT: Remote host closed the connection) (~gordon1@citadel.niflheim.info) | 10:53 | |
| + gordon1 (~gordon1@citadel.niflheim.info) | 10:54 | |
| - L29Ah (QUIT: Ping timeout: 260 seconds) (~L29Ah@wikipedia/L29Ah) | 11:16 | |
| josch | minute: an even bigger speed-up would be to get pocket-reform-handbook into debian. Could you cut a new tag, maybe called 2025-03-16 for https://source.mnt.re/reform/pocket-reform-handbook/ please? | 12:02 |
| josch | the last tag is from august 2024 | 12:02 |
| minute | josch, thank you, will do it in the next hour and also test the mesa upgrade on a reform next | 12:10 |
| josch | \o/ | 12:12 |
| josch | nice :) | 12:12 |
| josch | as we learned recently, the median time a package spends waiting in NEW is less than 48 hours | 12:12 |
| josch | so if i upload today it should be in Debian in 2 days instead of being stuck there for 9 months like a certain other reform-related package of mine, right, right? XD | 12:13 |
| - MyNetAz (QUIT: Remote host closed the connection) (~MyNetAz@user/MyNetAz) | 12:17 | |
| + MyNetAz (~MyNetAz@user/MyNetAz) | 12:28 | |
| minute | josch: upgrading mesa from your ci job works at first (it pulls in arm64 and armhf debs) but then there are issues with "trying to overwrite shared .../changelog.Debian.gz which is different from other instances of package..." | 12:46 |
| josch | minute: between which two packages please? | 12:47 |
| minute | http://dump.mntmn.com/20250316-custom-mesa-upgrade-log.txt | 12:48 |
| josch | perfect, thank you | 12:48 |
| josch | ooooh got it! | 12:54 |
| + paperManu (~paperManu@107.159.71.33) | 12:55 | |
| minute | :0 | 13:02 |
| josch | new pipeline running | 13:02 |
| josch | this was a bug that was also present before but nobody noticed :) | 13:02 |
| josch | (the dch utility does not honour SOURCE_DATE_EPOCH and thus has to be run inside faketime) | 13:03 |
| ch | or sed? :) | 13:04 |
| - MyNetAz (QUIT: Remote host closed the connection) (~MyNetAz@user/MyNetAz) | 13:58 | |
| minute | josch: btw is the pocket-reform-handbook already ok for debian licensing wise or do we have to port something over from reform-handbook still? idk if i did all the homework for both handbooks | 14:01 |
| minute | josch: pocket-reform-handbook does not have a README with licenses. i'll add | 14:07 |
| BoostisBetter | minute: any chance the pocket could get the webcam that the Next is getting? | 14:13 |
| minute | BoostisBetter: good point | 14:13 |
| minute | camera(s) on pocket would be cool at some point | 14:14 |
| BoostisBetter | I would mind it at the bottom of the display bezel. Not sure where there is room for the webcam PCB | 14:14 |
| BoostisBetter | I wouldn't mind it at the bottom of the display bezel. Not sure where there is room for the webcam PCB | 14:14 |
| BoostisBetter | though | 14:15 |
| ch | slide-out on the side! (dunno) | 14:15 |
| BoostisBetter | ch: that would be awesome! | 14:15 |
| BoostisBetter | Would remind me of the HP Omnibooks and the pop out mouse | 14:15 |
| minute | i was talking to someone at hackaday berlin yesterday and we had this idea of making a cartridge slot for the left side of pocket | 14:17 |
| minute | with usb signals inside | 14:18 |
| minute | so that people could make any kind of i/o cartridges and share them, without having to open the device | 14:18 |
| ch | ohh | 14:18 |
| minute | a camera could be like that | 14:18 |
| minute | (below the display we currently have the speaker) | 14:19 |
| BoostisBetter | minute: that would be seriously awesome! | 14:19 |
| minute | for this we would need to identify some semi rugged connector that would receive the cartridge connection inside. maybe just usb-a, or some card edge receptacle that's not too thick (a la old game consoles, c64, but needs to be more compact) | 14:20 |
| BoostisBetter | minute: why not just use pogo pins? | 14:20 |
| minute | https://www.samtec.com/rugged-power/high-speed/micro-edge-card/ | 14:21 |
| minute | BoostisBetter: with card edge the card could be held in place by the socket | 14:21 |
| ch | other people do usb-c, no? *hides* | 14:21 |
| minute | pogo could be an option sure | 14:21 |
| minute | ch: haha yeah. also an option | 14:21 |
| ch | (i can see how thats not a good plan) | 14:22 |
| minute | but then you need a usb-c connector on the cartridge, with card edge it doesn't need a connector ;D | 14:22 |
| minute | (card edge means the pcb plugs directly into a socket) | 14:22 |
| minute | example https://www.samtec.com/products/mec1-ra | 14:23 |
| minute | https://suddendocs.samtec.com/ebrochures/mecx-ebrochure.pdf | 14:23 |
| minute | josch: also, reform-handbook has a bunch of size reduction things you did, that you could potentially port over to pocket-reform-handbook (if they're not in there yet) | 14:25 |
| ch | are there pcie lanes left that could be exposed there btw? | 14:26 |
| minute | ch: iirc rk3588 has an in-pocket unused 2nd pcie controller | 14:28 |
| minute | pocket motherboard update would be an option (incl 4x nvme lanes) | 14:30 |
| ch | mhhhh | 14:32 |
| minute | luckily the motherboard is not that expensive/complex (unlike reform) | 14:32 |
| minute | but so far we still only made one (huge) batch! | 14:33 |
| minute | but soon it's time to gather all kinds of fixes that have to be done for a second rev | 14:33 |
| minute | there's also an 8 inch display option but it doesn't fit in the current case ;/ would need complete case redesign incl flipping the hinges | 14:34 |
| minute | worth it, but not something done quickly | 14:34 |
| minute | and there's open questions about the charger + also the battery size could be expanded | 14:34 |
| minute | yesterday with elen we were talking about that if the mps FA should be inconclusive we could port the new charger chip from the next | 14:35 |
| minute | josch: looks like i don't have my private gpg key at home so can't sign the tag now | 14:41 |
| minute | josch: so i made a unsigned one on pocket-reform-handbook | 14:42 |
| + MyNetAz (~MyNetAz@user/MyNetAz) | 14:43 | |
| josch | minute: thanks to grimmware, the size reduction is in pocket-reform-handbook as well \o/ https://source.mnt.re/reform/pocket-reform-handbook/-/commit/5ef7991e44529c7d998dec17d9ad3c64c417185d | 14:48 |
| minute | ah noice | 14:48 |
| josch | and thank you for the README, looks good! | 14:49 |
| josch | unsigned git tag works just as well | 14:50 |
| josch | importing it now... | 14:50 |
| josch | ch: luckily the problem is already solved with dch in Trixie which has a --date option since https://salsa.debian.org/debian/devscripts/-/merge_requests/357 | 14:54 |
| ch | josch: ah, i -thought- there was an option for it :) | 14:55 |
| josch | ch: it is undocumented in the man page... have to open another MR for that... | 14:56 |
| ch | :/ | 14:56 |
| ch | would be nice if it also took SOURCE_DATE_EPOCH | 14:57 |
| josch | ch: do we have a volunteer for a new MR? :) | 14:58 |
| ch | did i already say i don't speak perl? :) | 14:58 |
| BoostisBetter | minute: btw, I think the issue I was having with the KB on the pocket were only when the amospala suspend script changed the led states. If you disable that then the kb is stable and doesn't reset hardly ever. | 14:58 |
| minute | mega coincidental find https://github.com/VoronDesign/Voron-Tap/issues/42 | 15:06 |
| minute | this sounds a bit like our mps issue (it might not be though). the solution is interesting, a polyfuse that has a 1-6 ohm resistance during normal operation, which acts as dampening to prevent voltage spikes | 15:06 |
| grimmware | josch: wow I remember poking around in the repo but I do *not* remember making that MR | 15:22 |
| ch | interesting | 15:23 |
| josch | grimmware: happens to me all the time too :) | 15:25 |
| grimmware | I would much rather have been useful and not remembered it than the other way round for sure | 15:28 |
| minute | https://www.analog.com/media/en/technical-documentation/application-notes/an88f.pdf | 15:29 |
| minute | cool old paper from linear about "Ceramic Input Capacitors Can Cause Overvoltage Transients" | 15:29 |
| josch | you people all have weird hobbies for a sunday ;) | 15:30 |
| minute | lol that is true | 15:30 |
| minute | i guess i'm compensating for too many meetings and a big noisy event this week | 15:30 |
| minute | need to go to the nerd hole sometimes :D | 15:30 |
| minute | btw there's another rabbit hole: i currently have the IVO M125NWF6 R0 display in my classic reform, which is kind of a drop in replacement for the older display, and i only now noticed that the spec on panelook says 120Hz | 15:31 |
| minute | i tried messing with swaymsg output modeline earlier and the panel accepts almost any combination of hsync/vsync porches and pixel clock, but something starts glitching around 70hz, so it might be wrong @ 120hz or the stdp2600 doesn't do the right thing (or needs more DP output lanes) | 15:32 |
| minute | hm yeah curiously it says 4 lanes on panelook | 15:37 |
| josch | pocket-reform-handbook_2025-03-16+dfsg-1_both.changes is NEW | 16:51 |
| josch | lets see if we get lucky this time :) | 16:51 |
| josch | minute: if you replace job 8002 by job 8106 in your sources.list, you should be able to co-install the mesa packages | 16:53 |
| josch | minute: you might also want to give an update in https://community.mnt.re/t/no-response-on-order-for-3-months/3199/1 | 16:54 |
| josch | again sorry to ping you on a sunday | 16:54 |
| - chrcav (QUIT: Ping timeout: 268 seconds) (~chrcav@user/chrcav) | 16:58 | |
| + chrcav (~chrcav@user/chrcav) | 17:10 | |
| + L29Ah (~L29Ah@wikipedia/L29Ah) | 17:28 | |
| - L29Ah (QUIT: Ping timeout: 244 seconds) (~L29Ah@wikipedia/L29Ah) | 17:44 | |
| - mtm (QUIT: Ping timeout: 260 seconds) (~textual@47.202.75.129) | 18:03 | |
| + mtm (~textual@47.202.75.129) | 18:06 | |
| - xktr (QUIT: Quit: leaving) (~xktr@user/xktr) | 18:15 | |
| + L29Ah (~L29Ah@wikipedia/L29Ah) | 18:16 | |
| + Ar|stote|is (~linx@149.210.35.228) | 18:28 | |
| + xktr (~xktr@user/xktr) | 18:41 | |
| + hanno (~hanno@h-98-128-180-142.NA.cust.bahnhof.se) | 19:32 | |
| hanno | Hej! I just tried to connect a headset to my Reform2. While I could change the audio output to the headset just fine, there was no option in pasystray to set an input. Don't see any candidate via alsamixer either. Any clues on how to configure the mic? | 19:34 |
| - Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66) | 19:55 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 19:55 | |
| + Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66) | 20:22 | |
| - Gooberpatrol66 (QUIT: Ping timeout: 248 seconds) (~Gooberpat@user/gooberpatrol66) | 20:22 | |
| - digitalrane (QUIT: Ping timeout: 250 seconds) (~digitalra@user/digitalrane) | 20:34 | |
| + digitalrane (~digitalra@user/digitalrane) | 20:47 | |
| hramrach | that sounds like the card does not have an input, either at all or some device description is missing, or you missed it in the audio settings | 21:05 |
| hramrach | I use pavucontrol to see the card configutaion, it's difficult to select the correct view in alsamixer | 21:06 |
| hanno | hramrach: thx! Actually, the systray application does launch pavucontrol, I just realized. There is says "no input devices available" when I select "show all except monitors". I am running the A311D and have not messed with the system image in a meaningful way I believe (only recently reinstalled the system). Any suggestions where I could start looking? It's been a long time since I messed with audio on debian, it usually works w/o me | 21:11 |
| hanno | messing with it :) | 21:11 |
| - Gooberpatrol_66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66) | 21:12 | |
| - jacobk (QUIT: Ping timeout: 276 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 22:07 | |
| + jacobk (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 22:07 | |
| - MyNetAz (QUIT: Remote host closed the connection) (~MyNetAz@user/MyNetAz) | 22:11 | |
| - hanno (QUIT: Ping timeout: 260 seconds) (~hanno@h-98-128-180-142.NA.cust.bahnhof.se) | 22:12 | |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-53-248.bbcust.telenor.se) | 22:15 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 22:16 | |
| + MyNetAz (~MyNetAz@user/MyNetAz) | 22:36 | |
| + jacobk_ (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 23:10 | |
| - jacobk (QUIT: Ping timeout: 276 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 23:10 | |
| - MyNetAz (QUIT: Remote host closed the connection) (~MyNetAz@user/MyNetAz) | 23:30 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!