| elb | (I'm looking for the exact wording now) | 00:00 |
|---|---|---|
| elb | You can tell gcc not to assume that, though | 00:00 |
| erle | josch do you have any more tips to get a … stable … imgx8mq system? the random lockups kinda annoy me. | 00:04 |
| josch | erle: do you only get these lock-ups after resuming from suspend? | 00:05 |
| elb | ahh here: "The empty list in a function declarator that is not part of a definition of that function specifies that no information about the number or types of the parameters is supplied." | 00:06 |
| elb | C99 declares the intention to obsolete this, but certainly allows it | 00:07 |
| erle | josch not only, but mostly. sometimes i get them when programs do something. like the first time i started flare-game. | 00:07 |
| erle | josch or on a specific youtube video | 00:07 |
| erle | and also sometimes the reform just hangs on boot right after systemd says something about udev | 00:08 |
| josch | okay, your lock-ups might be completely different from the ones i had which usually appeared when the system was idle over night | 00:08 |
| erle | nah, mine are either suspend-wake or intense graphical activity (but never minetest) | 00:08 |
| elb | looking at the commit in question, I would use `bool gfx_init(void);` if I wanted to be fully explicit | 00:14 |
| elb | and it looks to me (surprisingly) that the option which causes gcc to automatically warn for this, -Wstrict-prototypes, is NOT part of -Wall and (more surprisingly) neither is it in -Wextra | 00:14 |
| josch | yes, thaht's what i also observed | 00:16 |
| - mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 00:21 | |
| + shtrophic (~m-hrdsqi@user/shtrophic) | 00:31 | |
| minute | erle: iirc the hangs can be related to the dcss display pipeline. try to switch to dual display mode. this selects an alternative device tree dtb on imx8mq which uses lcdif instead of dcss. | 00:50 |
| minute | erle: sudo reform-display-config --dual (or dual, don't remember) | 00:51 |
| erle | minute i did that last night and it broke suspend/resume (computer woke up again, but locked up) | 00:51 |
| erle | like, i saw a single cursor blinking, then just a black screen | 00:51 |
| minute | erle: sorry, i don't work on imx8mq support anymore | 00:51 |
| erle | well who does? | 00:51 |
| minute | noone | 00:51 |
| erle | funny, i play with that guy on oysterity anarchy. i'll ask him then. | 00:52 |
| minute | erle: you could migrate to 9front | 00:52 |
| minute | erle: it probably has better mnt reform imx8mq support than linux by now ^^ | 00:53 |
| erle | i don't really want to | 00:53 |
| erle | i mean apart from the random hangups the thing does what i need to | 00:53 |
| erle | i just want it to be as stable as an old thinkpad | 00:53 |
| josch | i'm still rebasing that cadence hdmi patch stack on every kernel release... | 00:53 |
| minute | or get a newer processor module (suspend is broken on all of them though) | 00:53 |
| minute | the next one that can probably do suspend is the qcs6490 module that's coming soon | 00:54 |
| erle | nah, if i get a new laptop, it will be a thinkpad again probably. shitty keyboard, but that's why i have 2 of the mnt keyboards. | 00:55 |
| erle | although i hope i can find some way that i don't have to do that | 00:56 |
| erle | (i.e. i need to find a way to avoid random hangs) | 00:56 |
| minute | why a new laptop? | 00:57 |
| josch | erle: maybe you can try the dual display again at the cost of suspend just to see if it fixes the hangs you have or not? | 00:57 |
| erle | josch i will! | 00:58 |
| minute | josch: we could just change the single display dts to use lcdif as well | 00:58 |
| minute | as i suspect hdmi could cause suspend to break | 00:58 |
| josch | i'd say it's too late to change things on a platform where not much active development happens anymore | 00:59 |
| minute | it's a trivial change. i can do it tomorrow, at least in a branch | 00:59 |
| erle | i will assist as much as i can, given i have a vested interest in this working, just tell me what to do | 00:59 |
| minute | well, lets see. | 00:59 |
| erle | for now i just change back to the dual-display thing | 01:00 |
| josch | same here -- next week is the new kernel release that i have to test on imx8mq | 01:00 |
| minute | to derive a single display lcdif dts, it would be enough to copy the dual display dts and remove/disable the hdmi node(s) | 01:01 |
| erle | minute i need a new laptop because the multiple thinkpads i have all are either >15 years old or broken like “WLAN has very high jitter with modern kernel”, like this one: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1115862 | 01:02 |
| erle | my fav idea would though to make the reform work, because it has the best keyboard i have ever used | 01:02 |
| minute | i see, what i mean to say is that the reform can be cheaply upgraded to better processors (except the rk3588 which is comparably expensive) | 01:03 |
| josch | heh today i observed that the rk3588 runs the tests of one of my software packages faster (only takes 3.5 hours instead of 5) than my old thinkpad that i had before :D | 01:03 |
| erle | yeah but you said yourself that this breaks suspend/resume | 01:03 |
| josch | erle: some folks are working on hibernate (but don't hold your breath) | 01:03 |
| erle | also the reform is *very* sturdy and apart from the “loose screws” thing it's probably tankier than most thinkpads | 01:03 |
| erle | meanwhile, on newer thinkpads the frame deforms so much over time that keys get stuck | 01:04 |
| minute | i still want to make at least hibernate work on rk3588, and for qcs6490 i lined up a collaboration with linaro to make sure that suspend works | 01:05 |
| erle | weird, i always thought hibernation on linux was a thing that did not need much hardware-specific details, just “write the RAM into swap” and the reverse. seems i was wrong. | 01:05 |
| minute | erle: the screw stuff was solved a longer time ago, with these https://shop.mntre.com/products/m2-x-5-mm-phillips-countersunk-screws?taxon_id=13 (nylok like) | 01:05 |
| erle | i just wanted to be specific ;) | 01:06 |
| erle | but thanks anyway | 01:06 |
| minute | erle: well, that's why i think hibernate is possible to fix on rk3588 but not sure about suspend to ram | 01:06 |
| minute | i mean, i'm sure it's possible to fix that too, with enough time and money | 01:07 |
| erle | hmm, question. can i boot from USB? or only SD card? i have more spare USB sticks than SD cards. | 01:07 |
| minute | yes @ usb | 01:08 |
| erle | ok thx | 01:08 |
| josch | yes, even with imx8mq | 01:08 |
| josch | erle: modern u-boot releases for the reform should order usb boot after sd-card and emmc | 01:08 |
| erle | josch btw thanks for https://reform.debian.net | 01:12 |
| josch | you are welcome -- i'm just scratching my own itch :) | 01:13 |
| - chomwitt (QUIT: Ping timeout: 265 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 01:13 | |
| josch | it is not without conceptual problems though -- often people report issues without mentioning that they do not use the official MNT debian packages which makes debugging harder | 01:16 |
| josch | this is made even worse by the fact that reform.d.n is at the top of search engine results when one searches for reform debian system images... | 01:17 |
| josch | and then there is the problem that without having people experience issues in unstable, the problems will not get fixed before the next stable release | 01:17 |
| josch | for example, i have some mesa bugs in trixie that i didn't notice because i was running mesa from bookworm before... | 01:18 |
| josch | a.ustriancoder helped me debug them and now i'm running sway with mesa from git... | 01:18 |
| erle | i was just able to suspend-resume with the dual setup! so it does not 100% break suspend-resume. | 01:20 |
| erle | it is, however, not reliable | 01:20 |
| erle | josch you have imx8mq as well, right? | 01:20 |
| josch | yes | 01:20 |
| erle | josch if you can spare the time, please send me a screenshot from how glxgears looks with mesa from git | 01:21 |
| josch | erle: sure thing! can you ping me about that? i'll have forgotten tomorrow :) | 01:21 |
| josch | (and now it's too late) | 01:21 |
| erle | i'll try | 01:22 |
| josch | erle: you filed this as a mesa bug, no? if you link me the bug i can have it open in a tab as a reminder | 01:23 |
| erle | josch i have not yet because i have not yet setup mesa from git | 01:23 |
| erle | i'll try to remind myself to do it :P | 01:24 |
| erle | then you can chime in with “i have this issue” or “i don't have this issue” | 01:24 |
| josch | feel free to ping me if you need help with building mesa -- it's quite easy | 01:24 |
| erle | so far i have only ever built mesa demos | 01:25 |
| erle | to prove to some people that on a potato pc you can do shadows and reflections | 01:25 |
| josch | 6.16.8 was uploaded to unstable yesterday -- it does not break the patch stack and reform-debian-packages pipeline is running | 01:29 |
| josch | system image with unreleased reform-setup-wizard and unreleased reform-tools is here: https://source.mnt.re/josch/reform-system-image/-/jobs/13241/artifacts/browsehttps://source.mnt.re/josch/reform-system-image/-/jobs/13241/artifacts/browse | 01:29 |
| josch | ups | 01:29 |
| josch | i'll try that out tomorrow | 01:30 |
| josch | if i do it on imx8mq, i catch two birds at once :) | 01:30 |
| erle | so i now suspect that the single/dual thing may not affect suspend/resume hangs at all. with both configurations i was able to successfully and unsuccessfully resume. | 01:30 |
| erle | also, to distinguish between a full hang and just a display hang i play music ;) | 01:31 |
| erle | good night! | 01:32 |
| elb | speaking of playing music I couldn't get music to come out of my headphones the other day, but I suspect it was PEBKAC | 01:32 |
| elb | the speaker worked fine | 01:32 |
| elb | ugh every time I go to shop.mnt.re I want to get the v2 rcore and wifi slot headset board, I had thought my pocket would come with them and it didn't | 01:33 |
| elb | but I don't really need it and I feel like buying extra parts I don't need for the laptop I got so it never has to become ewaste is counter-productive ;-) | 01:34 |
| erle | elb according to https://lenowo.org/ nothing has to become ewaste! check this out: https://lenowo.org/viewtopic.php?t=28 | 01:38 |
| erle | > After the recent hacker news "invasion", I have now determined that the page can handle up to 1536 users before running out of RAM, meaning that the IP camera surprisingly is fully sufficient for its purpose. | 01:39 |
| erle | ;) | 01:39 |
| erle | and someone else replies: | 01:39 |
| erle | > I have a homebrew linux computer based on an AT91SAM harvested from a vape | 01:39 |
| erle | life … finds a way | 01:39 |
| elb | yeah but in this case I'd have a carrier board and a headphone board left over, but no motherboard or SoM | 01:40 |
| erle | rewire the speakers? xD | 01:40 |
| erle | alternatively … do you know jabra? | 01:41 |
| erle | they make headphones that come with a usb sound card (as far as i can tell) | 01:41 |
| erle | and the heapdhones break a lot | 01:41 |
| erle | because they contain some kind of load-bearing cable | 01:41 |
| erle | so because my previous employer simply replaced them i now have a bunch of grimy usb sound card dongles | 01:42 |
| elb | no, I mean, the pocket reform I have already has a carrier board and it already has a headphone board, and what I keep looking at is a replacement carrier and headphone board that I _don't need_ | 01:42 |
| erle | if you know anyone who used to have a jabra headphone, maybe that person has a jabra usb dongle for you | 01:42 |
| erle | oh | 01:42 |
| erle | i guess i should go to sleep so my reading comprehension goes up again | 01:43 |
| erle | good night! | 01:43 |
| elb | good night :-) | 01:45 |
| elb | no I definitely don't have any headphone audio | 01:47 |
| elb | hmmm the plug is plugged in on both ends and I get a little bit of crackle in the headphones when I plug _them_ in, so there's _something_ there | 01:51 |
| elb | wow, that wifi chip gets hot | 01:51 |
| elb | ok looks like I have to change audio output via the audio configuration tab, not output devices | 01:56 |
| + emdevt (~emdevt@2001:7e8:faf6:f101:70f9:10a4:805e:9fe1) | 02:58 | |
| - paperManu (QUIT: Ping timeout: 250 seconds) (~paperManu@198.16.179.107) | 03:26 | |
| + paperManu (~paperManu@198.16.179.107) | 03:30 | |
| - paperManu (QUIT: Ping timeout: 255 seconds) (~paperManu@198.16.179.107) | 04:31 | |
| - skipwich (QUIT: Remote host closed the connection) (~skipwich@user/skipwich) | 05:01 | |
| + skipwich (~skipwich@user/skipwich) | 05:01 | |
| - robin (QUIT: Read error: Connection reset by peer) (~robin@user/terpri) | 06:11 | |
| + robin (~robin@user/terpri) | 06:11 | |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 07:21 | |
| + guest12345 (~guest1234@gidzit.org) | 08:08 | |
| - guest12345 (QUIT: Remote host closed the connection) (~guest1234@gidzit.org) | 08:30 | |
| - Guest4696 (QUIT: Ping timeout: 248 seconds) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 09:17 | |
| + mjw (~mjw@gnu.wildebeest.org) | 09:57 | |
| + andreas-e (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net) | 10:06 | |
| minute | ok, still got "reset by watchdog" this morning, so there's something else besides that fixed memory leak. | 10:14 |
| josch | oh :( | 11:15 |
| josch | was it a reboot? how did you identify that it was due to "reset by watchdog"? | 11:16 |
| - RandyK (QUIT: Remote host closed the connection) (~RandyK@user/randyk) | 11:23 | |
| + RandyK (~RandyK@user/randyk) | 11:23 | |
| + MartiniMoe (~moe@p200300eecf1ede1f60e8e526224b38e9.dip0.t-ipconnect.de) | 12:30 | |
| - mjw (QUIT: Ping timeout: 258 seconds) (~mjw@gnu.wildebeest.org) | 12:31 | |
| + paperManu (~paperManu@198.16.179.107) | 12:33 | |
| - MartiniMoe (QUIT: Ping timeout: 248 seconds) (~moe@p200300eecf1ede1f60e8e526224b38e9.dip0.t-ipconnect.de) | 12:39 | |
| minute | josch: the OLED says "Reset by watchdog." | 12:55 |
| + gustav2 (~gustav@c-78-82-55-162.bbcust.telenor.se) | 13:03 | |
| + thelounge5700 (~thelounge@148.168.138.88.rev.sfr.net) | 13:14 | |
| - TheProtoHobo (QUIT: Quit: leaving) (~dfbarth@mobile-166-175-60-240.mycingular.net) | 13:17 | |
| - iank_ (QUIT: Quit: ZNC 1.8.2+deb2+deb11u1 - https://znc.in) (~iank@fsf/staff/iank) | 13:25 | |
| + iank (~iank@fsf/staff/iank) | 13:29 | |
| + mjw (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 13:29 | |
| josch | hrm... i don't remember having seen that when my pocket rebooted the last time | 13:43 |
| [tj] | minute: oh I get that constantly | 13:45 |
| [tj] | and then all typing is backgrounded with a screen saying "press xanndy to boot" | 13:46 |
| minute | [tj]: what | 13:47 |
| [tj] | whatever the instructions are on the old to turn on the pocket | 13:48 |
| [tj] | led | 13:48 |
| [tj] | oled | 13:48 |
| minute | [tj]: ok, can you try my new build that fixes the memory leaks at least https://source.mnt.re/reform/pocket-reform/-/jobs/13221/artifacts/browse/pocket-reform-keyboard-fw/pocket-hid/build/ | 13:48 |
| minute | josch: it doesn't have to do with the reboots. that's a separate issue. | 13:49 |
| minute | josch: it's only a reset of the keyboard itself | 13:49 |
| minute | josch: the reboots probably have to do with system controller fw | 13:49 |
| josch | oh no :( | 13:51 |
| josch | then my issue is a different one -- i do not remember watchdog resets anymore with the latest firmware... | 13:51 |
| josch | in more positive news, waydroid is running well on rk3588 and i was able to install signal and too good to go without issues: https://mister-muffin.de/p/Tqfb.png (screenshot of waydroid) | 13:56 |
| [tj] | minute: sure I'll try later today when I'm done work | 13:56 |
| josch | i also managed to install the app of our kindergarten and snooping the traffic with mitmproxy revealed that it's just a browser for a typo3 website, so no need for an app there :) | 13:56 |
| [tj] | other than the message on the oled my pocket reform experience is really solid | 13:56 |
| josch | just our local cargo-bike sharing app refuses to install :( | 13:56 |
| [tj] | otherwise I'm kind of numb to bugs these days | 13:57 |
| - paperManu (QUIT: Ping timeout: 245 seconds) (~paperManu@198.16.179.107) | 14:10 | |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 14:48 | |
| - mjw (QUIT: Quit: Leaving) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 15:15 | |
| + mjw (~mjw@gnu.wildebeest.org) | 15:29 | |
| ch | nice @ cppcheck fixes | 15:58 |
| minute | josch: oh nice, how did you install waydroid? | 16:08 |
| josch | minute: i do not trust the random android binary blobs so everything is happening inside qemu. Here is my POC: https://paste.debian.net/hidden/70c46c6e/ | 16:09 |
| josch | i'll blog about this once i'm happy with the script which needs mitmproxy packaged in debian | 16:09 |
| josch | mitmproxy was kicked out of debian trixie and i didn't notice until now :( | 16:09 |
| josch | a handful of rust crates need to get packaged before the new version can get uploaded to unstable | 16:10 |
| josch | oh and on rk3588 i need to somehow figure out why there are graphics glitches | 16:11 |
| josch | (works fine on a311d and imx8mq) | 16:11 |
| erle | josch imx8mq regarding “random” hangs, any workload that triggered it for you? | 16:15 |
| josch | erle: nope | 16:15 |
| erle | josch also i want to remind you to post a screenshot of glxgears with mesa git on the vivante GPU | 16:15 |
| josch | erle: you are too late, i'm already in the process of building mesa :) | 16:17 |
| josch | erle: https://mister-muffin.de/p/xVJU.png | 16:29 |
| - jogu (QUIT: Remote host closed the connection) (~jogu@user/jogu) | 16:30 | |
| + jogu (~jogu@user/jogu) | 16:31 | |
| + jogu_ (~jogu@user/jogu) | 16:32 | |
| - jogu (QUIT: Ping timeout: 265 seconds) (~jogu@user/jogu) | 16:36 | |
| + jogu (~jogu@user/jogu) | 16:39 | |
| - jogu_ (QUIT: Ping timeout: 244 seconds) (~jogu@user/jogu) | 16:39 | |
| - emdevt (QUIT: Remote host closed the connection) (~emdevt@2001:7e8:faf6:f101:70f9:10a4:805e:9fe1) | 16:41 | |
| erle | josch given that you have built recent mesa just now and it looks broken, do you want to file the bug? | 16:44 |
| erle | josch also thank you | 16:44 |
| josch | erle: i would appreciate if you could take care of it -- i have too many projects already as it is | 16:45 |
| erle | i'll try | 16:49 |
| - erle (QUIT: Quit: K-lined) (~erle@user/erle) | 16:49 | |
| josch | erle: when you file it, you can @-mention me as another affected user, i'm @josch | 16:49 |
| josch | ah too late | 16:50 |
| josch | in unrelated news, if you want to to have a rk3588 classic reform tomorrow, there is one sold in Berlin, Germany: https://www.kleinanzeigen.de/s-anzeige/3195881956 | 16:50 |
| - gsora (QUIT: Ping timeout: 258 seconds) (~gsora@user/gsora) | 16:58 | |
| + gsora (~gsora@user/gsora) | 17:01 | |
| + MartiniMoe (~moe@p200300eecf1ede1f1f8f638cacc020bb.dip0.t-ipconnect.de) | 17:58 | |
| mhoye | re:powersave, thanks all | 17:59 |
| - andreas-e (QUIT: Quit: Leaving) (~Andreas@2a02-8434-b6a3-e901-facc-8e87-8e54-890d.rev.sfr.net) | 18:03 | |
| - thelounge5700 (QUIT: Quit: The Lounge - https://thelounge.chat) (~thelounge@148.168.138.88.rev.sfr.net) | 18:18 | |
| - chomwitt (QUIT: Ping timeout: 265 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 18:21 | |
| + AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net) | 18:27 | |
| - swivel (QUIT: Ping timeout: 260 seconds) (~swivel@shells.gnugeneration.com) | 18:32 | |
| + swivel (~swivel@shells.gnugeneration.com) | 18:33 | |
| + synnfynn (~synnfynn@user/synnfynn) | 18:51 | |
| + antti- (~antti@user/antti) | 20:15 | |
| - antti (QUIT: Ping timeout: 265 seconds) (~antti@user/antti) | 20:16 | |
| * antti- -> antti | 20:30 | |
| + guest12345 (~guest1234@gidzit.org) | 20:49 | |
| - guest12345 (QUIT: Remote host closed the connection) (~guest1234@gidzit.org) | 21:00 | |
| + testtest (~testtest@gidzit.org) | 21:00 | |
| - testtest (QUIT: Remote host closed the connection) (~testtest@gidzit.org) | 21:00 | |
| mjw | https://www.crowdsupply.com/mnt/mnt-reform-next/updates/september-update | 21:08 |
| + chomwitt (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 21:10 | |
| + reform19974 (~gsamu@209.216.99.9) | 21:34 | |
| - reform19974 (QUIT: Client Quit) (~gsamu@209.216.99.9) | 21:35 | |
| timide_ | Thank you for the update <3 | 21:38 |
| - gustav2 (QUIT: Quit: Quit) (~gustav@c-78-82-55-162.bbcust.telenor.se) | 22:15 | |
| - MartiniMoe (QUIT: Remote host closed the connection) (~moe@p200300eecf1ede1f1f8f638cacc020bb.dip0.t-ipconnect.de) | 22:28 | |
| minute | interesting https://github.com/raspberrypi/pico-sdk/issues/1471 | 22:36 |
| ch | pff | 22:56 |
| ch | we should fix pocket-reform-sysctl-fw/build.sh to use the same pico-sdk as the ci build | 23:00 |
| ch | same for kbd | 23:00 |
| - paperManu (QUIT: Ping timeout: 250 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 23:01 | |
| + kfx (~kfx@wopr.sciops.net) | 23:03 | |
| josch | okay, this has been a problem for too long -- i can submit a MR doing this | 23:06 |
| josch | i'm thinking about either calling the build.sh scripts from .gitlab-ci.yml or to remove the build.sh scripts and put all the "magic" (it's just calling cmake..) into .gitlab-ci.yml | 23:07 |
| josch | minute: preferences? | 23:07 |
| ch | my 2c: the cmake files should set PICO_SDK_PATH themselves correctly (to match install-fw-dependencies.sh) and then the build.sh scripts can stay or not but dont change the build result | 23:10 |
| josch | ch: are you saying the CMakeLists.txt should just hardcode the path that was set up by install-fw-dependencies.sh? | 23:14 |
| ch | yeah why not. could just say ../../pico-sdk or whatever the path is | 23:14 |
| ch | (might need to be "${CMAKE_CURRENT_LIST_DIR}/../../blah" | 23:15 |
| ch | ) | 23:15 |
| josch | how about if (NOT DEFINED ENV{PICO_SDK_PATH}) set(PICO_SDK_PATH "${CMAKE_SOURCE_DIR}/../pico-sdk") | 23:17 |
| + paperManu (~paperManu@198.16.179.107) | 23:17 | |
| ch | i would skip the if part | 23:20 |
| ch | but yeah | 23:20 |
| josch | ch: i'd leave it in because users familiar with pico-sdk expect to be able to use that environment variable to set the location | 23:21 |
| ch | yeah and i'd be the opposite | 23:21 |
| ch | "no you get to build with -this- sdk, and not something nobody else uses" | 23:21 |
| josch | *shrug* i'd say minute should decide and then i do that :) | 23:22 |
| minute | josch: please don't nuke my dear build.sh :D but in general, first we need to decide which sdk version to build against, and where the source of truth is for that. thinking this way, the CI doesn't seem like the correct place, and the build.sh seems like a better place | 23:37 |
| minute | and if you want to build against something else you can just not use build.sh | 23:38 |
| minute | summarizing, build.sh could become the source of truth for official builds in each fw directory, and dictate which sdk version is used | 23:38 |
| minute | and CI calls that. | 23:39 |
| josch | nice, that's what i'm doing right now | 23:39 |
| ch | ./install-fw-dependencies.sh could be the source | 23:39 |
| ch | having both the truth source and the build instructions in the same file is annoying | 23:39 |
| minute | ch: ok, and should build.sh call this or no? | 23:42 |
| minute | ch: how should the build script check/know that the correct sdk version was cloned? | 23:43 |
| ch | i think its fine to have a text in README saying "run ./install-fw-dependencies.sh once and then you can use build.sh or whatever you like" | 23:43 |
| ch | other projects also say things like 'install these things before you can run ./configure' | 23:44 |
| josch | maybe we are talking a little bit past each other. Code is better than words, so I created https://source.mnt.re/reform/pocket-reform/-/merge_requests/52 -- if you want to change something, tell me where and I change it :) | 23:45 |
| josch | (maybe this is buggy, CI is running now for the first time) | 23:45 |
| josch | it is not running | 23:45 |
| josch | minute: can you please enable the CI runner for https://source.mnt.re/josch/pocket-reform/ | 23:46 |
| - chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:85f:9a74:ce00:1ac0:4dff:fedb:a3f1) | 23:50 | |
| minute | josch: runner assigned! | 23:54 |
| josch | thank you! | 23:55 |
| josch | does anybody still need wayfire? | 23:55 |
| josch | it's currently QA maintained in Debian and there is a new upstream version | 23:55 |
| josch | i can work on it if somebody needs it... | 23:55 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
