| - klardotsh (QUIT: Ping timeout: 260 seconds) (~klardotsh@c-67-170-115-80.hsd1.wa.comcast.net) | 00:07 | |
| minute | josch: any idea what went wrong here? https://source.mnt.re/reform/reform-system-image/-/jobs/3914 | 00:37 |
|---|---|---|
| minute | ah i see | 00:37 |
| minute | uboot blob url | 00:37 |
| minute | running once again :3 | 00:47 |
| + klardotsh (~klardotsh@c-67-170-115-80.hsd1.wa.comcast.net) | 00:57 | |
| - klardotsh (QUIT: Quit: zzz) (~klardotsh@c-67-170-115-80.hsd1.wa.comcast.net) | 01:12 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:26 | |
| - mtm (QUIT: Ping timeout: 255 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:04 | |
| - cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra) | 02:14 | |
| + cobra (~cobra@user/Cobra) | 03:01 | |
| - colinsane (QUIT: Read error: Connection reset by peer) (~colinunin@97-113-159-4.tukw.qwest.net) | 03:29 | |
| + colinsane (~colinunin@97-113-159-4.tukw.qwest.net) | 03:30 | |
| - nsc (QUIT: Ping timeout: 260 seconds) (~nicolas@117-97-142-46.pool.kielnet.net) | 03:32 | |
| - colinsane (QUIT: Client Quit) (~colinunin@97-113-159-4.tukw.qwest.net) | 03:32 | |
| + nsc (~nicolas@24-48-142-46.pool.kielnet.net) | 03:33 | |
| + colinsane (~colinunin@97-113-159-4.tukw.qwest.net) | 03:35 | |
| - mjw (QUIT: Ping timeout: 255 seconds) (~mjw@gnu.wildebeest.org) | 03:37 | |
| - aperezdc (QUIT: Ping timeout: 255 seconds) (~aperezdc@2a03:6000:6e61:633::43) | 03:40 | |
| + aperezdc (~aperezdc@46.23.89.43) | 03:41 | |
| - mesaoptimizer (QUIT: Read error: Connection reset by peer) (~mesaoptim@user/PapuaHardyNet) | 03:58 | |
| - cmahns (QUIT: Ping timeout: 260 seconds) (8fe824803c@2a03:6000:1812:100::10cd) | 03:59 | |
| + cmahns (8fe824803c@2a03:6000:1812:100::10cd) | 04:02 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:09 | |
| - Gooberpatrol66 (QUIT: Remote host closed the connection) (~Gooberpat@user/gooberpatrol66) | 04:56 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 06:44 | |
| - Gooberpatrol66 (QUIT: Remote host closed the connection) (~Gooberpat@user/gooberpatrol66) | 07:43 | |
| - kuno (QUIT: Ping timeout: 268 seconds) (sid299442@id-299442.helmsley.irccloud.com) | 07:46 | |
| + kuno (sid299442@id-299442.helmsley.irccloud.com) | 07:49 | |
| - hairu (QUIT: Remote host closed the connection) (m-uotkmd@user/hairu) | 08:16 | |
| + hairu (m-uotkmd@user/hairu) | 08:18 | |
| * Guest2704 -> mjw | 10:12 | |
| + mesaoptimizer (~mesaoptim@user/PapuaHardyNet) | 10:56 | |
| - amk (QUIT: Ping timeout: 272 seconds) (~amk@user/amk) | 11:57 | |
| + amk (~amk@user/amk) | 11:59 | |
| + chomwitt (~chomwitt@2a02:587:7a17:f700:1ac0:4dff:fedb:a3f1) | 12:31 | |
| - chomwitt (QUIT: Ping timeout: 255 seconds) (~chomwitt@2a02:587:7a17:f700:1ac0:4dff:fedb:a3f1) | 13:18 | |
| minute | does anyone have time to test this image version on a LS1028A (on another SD card)? https://source.mnt.re/reform/reform-system-image/-/jobs/3919/artifacts/browse | 13:39 |
| josch | yes | 13:41 |
| josch | on it | 13:41 |
| josch | what do you expect to break/fail? | 13:42 |
| josch | anything i should specifically test? | 13:42 |
| + chomwitt (~chomwitt@2a02:587:7a17:f700:1ac0:4dff:fedb:a3f1) | 13:49 | |
| minute | josch: does it boot into the setup wizard? does the wizard do the job? does it then go to tuigreet? can you login with the account you created? (select sway desktop in the wizard, wayfire currently has no window decorations) | 13:54 |
| minute | for me all of these work fine on pocket reform with imx8mp, which is great, display is rotated correctly everywhere too | 13:54 |
| minute | i will test on imx8mq + a311d | 13:54 |
| josch | minute: sorry, i don't have a second display so i cannot test gui stuff | 13:56 |
| josch | and my PCIe gpus never worked with wayland | 13:57 |
| minute | josch: oh ok | 13:57 |
| minute | josch: so you can test if you can still login normally then :D | 13:57 |
| minute | setup wizard should not fire on serial console, only on tty1 | 13:57 |
| josch | that i can do but that's not as useful :) | 13:57 |
| josch | i can see whether the process exists | 13:58 |
| josch | *check | 13:58 |
| minute | process won't exist i thinki | 13:58 |
| josch | should it not auto-boot into sway with setup wizard? | 13:58 |
| josch | even without a display attached | 13:58 |
| josch | or does it notice that | 13:58 |
| josch | minute: in the past you said that your ls1028a + pcie-gfx worked with an nvidia kelper gpu. If you can find out the exact card you have, maybe i'm able to reproduce your setup. | 14:01 |
| minute | josch: zotac gt710 1gb 64bit ddr3 | 14:02 |
| minute | https://www.zotac.com/de/product/graphics_card/geforce%C2%AE-gt-710-1gb | 14:03 |
| - mtm (QUIT: Ping timeout: 252 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:03 | |
| josch | perfect, thank you! | 14:03 |
| josch | it's kinda scary that right now on serial, you one only gets | 14:04 |
| josch | Starting kernel ... | 14:04 |
| josch | and then long silence until the login prompt comes up | 14:04 |
| minute | ah, you can change that by doing setenv bootargs loglevel=7 | 14:04 |
| josch | but yes, i'm able to log in via serial with that image | 14:04 |
| minute | on the display you should get systemd messages (if you have a display) | 14:05 |
| josch | ah right yes, that's useful | 14:05 |
| minute | josch: ok cool | 14:05 |
| minute | josch: thanks for testing! | 14:05 |
| minute | so now the system image for pocket has reached what i would call beta state | 14:05 |
| Sario | \o/ | 14:05 |
| minute | there are some critical fixes i need to still do before release: | 14:05 |
| minute | - initial sound config for the tlv chip/card on pocket (right now it starts with some alsa stuff turned off, so silent) | 14:07 |
| minute | - fix weird display init glitch (driver hacking) | 14:07 |
| minute | - implement lpc spi compatible interface for turning it off in rp2040 (or alternatively bring back usb kbd turnoff as a fallback) | 14:07 |
| josch | minute: nice, i found that exact card on kleinanzeigen for 20 bucks -- lets see if i can make it work with that one :) | 14:16 |
| minute | josch: great! | 14:16 |
| minute | josch: system image works fine on 8mq and setup wizard is snappy! but lpc poweroff doesn't work (unrelated, but needs to be addressed in general) | 14:16 |
| josch | yeah i've seen the issue on the forum but i don't spot anything weird in the command output the user has sent | 14:17 |
| josch | i hope it's not a kernel regression... :( | 14:17 |
| minute | image also works great on a311d | 14:20 |
| minute | font awesome is still missing! | 14:20 |
| minute | josch: can you add fonts-font-awesome as a recommended to reform-tools? | 14:21 |
| minute | and i will add it to system-image | 14:21 |
| josch | minute: both is done | 14:21 |
| minute | oh ok thanks | 14:22 |
| josch | minute: you can check "apt-cache show reform-tools" | 14:22 |
| minute | about lpc turnoff: this thing bugs me in general, because as we know it used to cut power too early | 14:22 |
| josch | it should be version 1.38 and have fonts-font-awesome in the Recommends | 14:22 |
| josch | yes | 14:22 |
| minute | and i don't like that one can not reliably turn off the system without a special driver | 14:22 |
| josch | +1 | 14:23 |
| minute | josch: hmm how come fonts-font-awesome is not preinstalled on the image? did i run the image creation too early (last midnight)? different branch? | 14:23 |
| josch | minute: can you confirm that you have reform-tools 1.38 installed and that it has the Recommends on fonts-font-awesome? | 14:24 |
| minute | josch: my suggestion: we remove poweroff stuff from lpc driver and we bring back ye olde systemd poweroff hook, and just send the hid message to all hid usb devices in the system. and in the keyboard firmware i'll implement a delay that turns off after a few seconds of delay | 14:25 |
| minute | josch: ok checking | 14:25 |
| josch | minute: your choice how you solve it :) | 14:26 |
| minute | ok | 14:26 |
| josch | maybe there is a magic number of seconds which allows all writes to be finished | 14:26 |
| - chomwitt (QUIT: Ping timeout: 255 seconds) (~chomwitt@2a02:587:7a17:f700:1ac0:4dff:fedb:a3f1) | 14:27 | |
| minute | josch: reform-tools 1.38 is installed and does recommend the font | 14:27 |
| josch | minute: your problem is "The source branch is 1 commit behind the target branch." in https://source.mnt.re/reform/reform-system-image/-/merge_requests/92 | 14:27 |
| minute | josch: hmm, a "sync" before sending the command might also help | 14:27 |
| josch | the missing commit is the one installing fonts-font-awesome in reform-system-image main | 14:27 |
| minute | josch: ah thanks! because i'm always unlucky with rebasing i'll merge main into that branch... unless you object | 14:28 |
| josch | go ahead :) | 14:29 |
| josch | whatever is easier for you | 14:29 |
| minute | ok done | 14:30 |
| josch | minute: what i did locally to do the rebase was to just run "git rebase main" while being in the setup-and-login branch and that succeeded without problems | 14:46 |
| josch | i'm now investigating how we can make the created disk images "sparse" -- I posted my question on superuser.com here in case anybody knows the answer: https://superuser.com/questions/1837708/is-there-a-tool-like-dd-but-respect-sparse-ness-of-input-file | 14:47 |
| + mark_ (~mjw@gnu.wildebeest.org) | 15:43 | |
| minute | josch: ok so i just tested poweroff on a311d with the same family of system images, and there it works via lpc | 16:03 |
| minute | so maybe it's only an issue on imx8mq? | 16:03 |
| josch | huh | 16:05 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:09 | |
| - mark_ (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org) | 16:48 | |
| minute | josch: how can we prebuild + include the qcacld2-dkms driver? | 16:50 |
| minute | (on imx8mp) | 16:50 |
| minute | josch: asking if you have a recipe in mind that i could implement, presumably on system image? | 16:50 |
| josch | minute: depends... i have it uploaded to NEW where it has been sitting for 5 days now: https://ftp-master.debian.org/new.html | 17:06 |
| josch | minute: so at some point this will get processed and then we just apt install it | 17:06 |
| + chomwitt (~chomwitt@2a02:587:7a17:f700:1ac0:4dff:fedb:a3f1) | 17:06 | |
| josch | but there is of course no way to tell how long that is going to take | 17:06 |
| minute | josch: yeah that's too long for me unfortunately (i am under pressure to ship pockets) | 17:07 |
| minute | josch: so i need to build this in our infra somehow | 17:07 |
| josch | not necessarily | 17:07 |
| minute | aha? | 17:07 |
| josch | because the package is Architecture:all | 17:07 |
| minute | well, i need an arm64 binary package derived from this | 17:07 |
| josch | so what has been built on salsa.debian.org on amd64 will be installable on arm64 as well | 17:07 |
| minute | it takes too long to build on pocket | 17:08 |
| josch | no | 17:08 |
| josch | oh | 17:08 |
| josch | wait | 17:08 |
| josch | you don't want to build the dkms module the normal way as part of reform-system-image? | 17:08 |
| minute | well either that or pre build it in it's own repo | 17:08 |
| josch | i would've suggested, you wget the .deb from here: https://salsa.debian.org/debian/ezurio-qcacld-2.0-dkms/-/jobs/5511321/artifacts/browse/debian/output/ | 17:08 |
| minute | firstly building as part of RSI is fine | 17:08 |
| josch | as part of mkimage.sh and install that | 17:09 |
| minute | aha! | 17:09 |
| minute | but how are kernel updates handled? | 17:09 |
| minute | if someone updates to a newer kernel on pocket, it's unacceptable that the device would need to rebuild this thing | 17:10 |
| minute | i think it needs to be built in sync with reform-debian-packages | 17:10 |
| minute | so it would always fit to the kernel | 17:10 |
| josch | then you have to set up your own thing and not use the debian dkms packaging infrastructure | 17:10 |
| minute | oh ok | 17:10 |
| josch | the normal way to do this is to build the source on the host | 17:10 |
| josch | this is also painful for other large dkms drivers like zfs, for example | 17:11 |
| minute | ok i see | 17:11 |
| minute | so then i make a .deb that has a .ko and the firmware in it | 17:11 |
| josch | probably, yes | 17:11 |
| josch | and somehow you have to make multiple .deb | 17:11 |
| josch | one for each new kernel version | 17:11 |
| josch | or ship every .ko for each kernel in the same .de | 17:12 |
| josch | .deb | 17:12 |
| josch | and come up with a way how to manage dependencies with what the user has installed | 17:12 |
| minute | oof | 17:12 |
| josch | i cannot help you with that as i have not seen this being done elsewhere | 17:12 |
| josch | maybe it's easily possible but i don't know | 17:12 |
| josch | minute: did you already run the building on the pocket and timed it? how long does it take? | 17:13 |
| minute | josch: ok i can time it for real now... but the code is *huge* | 17:13 |
| josch | minute: also, did you have a look at htop while it was building and made sure that it is utilizing all cores? | 17:13 |
| minute | 100000s of lines of code | 17:13 |
| josch | yes, it takes a very long time to compile on a311d as well | 17:14 |
| minute | you see... | 17:14 |
| josch | i have the dkms package installed right now | 17:14 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:14 | |
| minute | so it takes at least double that on pocket | 17:14 |
| minute | anyway, will time it and look at the core utilization | 17:15 |
| minute | also not sure _why_ it is so big, i wonder if there are unnecessary features | 17:15 |
| minute | that could be turned off... | 17:15 |
| josch | i just had a look, it's using all six cores of my a311d | 17:16 |
| josch | (i'm timing it right now as well) | 17:16 |
| josch | ah, wrong kernel version | 17:17 |
| josch | i cannot test it i'm afraid | 17:17 |
| josch | i'm still on kernel 6.5 because i'm running bookworm | 17:17 |
| josch | and because 6.6 locks up my system around five minutes after boot... didn't investigate yet | 17:19 |
| minute | one good thing: it can easily be cross built | 17:22 |
| minute | one just needs the kernel source tree from reform-debian-packages | 17:22 |
| minute | on i9-9900k: real0m28.393s | 17:24 |
| minute | command used for building: KBUILD_MODPOST_WARN=1 ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu- MODNAME=qcacld2 KERNEL_SRC=/home/minute/src/mref/reform-debian-packages/linux/linux make | 17:24 |
| josch | minute: pay somebody to fix ath10k to work with the QCA9377? | 17:29 |
| josch | a quick internet search also yields results like https://github.com/pabilbado/ath10k-fix-QCA9377 | 17:30 |
| minute | josch: https://www.mail-archive.com/search?l=ath10k@lists.infradead.org&q=subject:%22Re%5C%3A+Problems+with+QCA9377+SDIO+on+NXP+i.MX8MPlus+SOM%22&o=newest | 17:31 |
| + jacobk (~quassel@64.189.201.150) | 17:36 | |
| - jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@64.189.201.150) | 17:56 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:24 | |
| q66 | a crossbuilt kernel has the limitation that you cannot compile modules for it on the target device | 18:25 |
| q66 | because the source tree or kernel headers tree created from it are specific to the host arch they were set up on | 18:26 |
| q66 | there is no workaround | 18:26 |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 18:28 | |
| minute | q66: i don't mean to crossbuild the kernel, only an out of tree module | 18:30 |
| q66 | how do you do that without previously having crossbuilt the kernel? | 18:36 |
| minute | also we do have source and header packages for the kernel | 18:43 |
| minute | so not sure what you mean | 18:43 |
| * mjw -> Guest2951 | 18:47 | |
| - Guest2951 (QUIT: Killed (tungsten.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:47 | |
| * mark_ -> mjw | 18:47 | |
| + Guest2951 (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:48 | |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:587:7a17:f700:1ac0:4dff:fedb:a3f1) | 18:54 | |
| - sevan (QUIT: Ping timeout: 255 seconds) (~sevan@user/venture37) | 18:54 | |
| minute | ok, lets try if this works... https://source.mnt.re/reform/qcacld2/-/jobs/3931/artifacts/browse | 19:12 |
| + minutepocket (~reform@p509936c9.dip0.t-ipconnect.de) | 19:12 | |
| minute | minutepocket: https://source.mnt.re/reform/qcacld2/-/jobs/3931/artifacts/browse | 19:12 |
| minute | uh oh, we need to blocklist ath10k_sdio | 19:15 |
| minute | but the driver works :# | 19:20 |
| + minutepo1ket (~reform@p509936c9.dip0.t-ipconnect.de) | 19:21 | |
| minutepo1ket | typing this via that driver | 19:21 |
| - minutepocket (QUIT: Ping timeout: 268 seconds) (~reform@p509936c9.dip0.t-ipconnect.de) | 19:24 | |
| q66 | minute: kernel headers packages contain binaries that need to be runnable on HOST_CC machine, so a headers package that was built natively should not be usable for module crosscompilation | 19:25 |
| q66 | unless you happen to be real lucky | 19:26 |
| minute | ok, i didn't need the kernel header package at all though | 19:26 |
| q66 | how do you build a module without kernel headers? | 19:26 |
| minute | with the kernel source package, which is guess includes the headers... | 19:26 |
| minute | https://source.mnt.re/reform/qcacld2/-/blob/boundary-CNSS.LEA.NRT_3.1/.gitlab-ci.yml?ref_type=heads | 19:26 |
| minute | s/is guess/i guess | 19:27 |
| q66 | kernel source package is just a bigger version of the headers package | 19:27 |
| q66 | either can be used for building modules | 19:27 |
| minute | gotcha | 19:27 |
| minute | in any case i got a working .ko | 19:27 |
| minute | now just need to package it + the firmware in a nice .deb | 19:27 |
| q66 | same thing applies though | 19:27 |
| q66 | so you were probably lucky enough that this particular build does not invoke the binaries | 19:27 |
| minute | nice | 19:28 |
| q66 | don't expect it to work generically though | 19:28 |
| minute | what binaries are those? | 19:28 |
| q66 | there is a bunch of them, don't recall the names off the top of my head | 19:28 |
| q66 | i can look when I'm home | 19:28 |
| q66 | (i'm just going back from work right now) | 19:29 |
| + sevan (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 19:29 | |
| - sevan (QUIT: Changing host) (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 19:30 | |
| + sevan (~sevan@user/venture37) | 19:30 | |
| - minutepo1ket (QUIT: Ping timeout: 264 seconds) (~reform@p509936c9.dip0.t-ipconnect.de) | 19:44 | |
| + minutepocket (~reform@p509936c9.dip0.t-ipconnect.de) | 19:51 | |
| - minutepocket (QUIT: Ping timeout: 255 seconds) (~reform@p509936c9.dip0.t-ipconnect.de) | 19:58 | |
| + minutepocket (~reform@p509936c9.dip0.t-ipconnect.de) | 19:58 | |
| - Twodisbetter (QUIT: Quit: Gateway shutdown) (2cc0e4ea1c@irc.cheogram.com) | 20:08 | |
| - f_[xmpp] (QUIT: Quit: Gateway shutdown) (fffdb90022@fases/developer/funderscore) | 20:08 | |
| q66 | minute: located one of my aarch64 kernel build logs | 20:15 |
| q66 | for instance these https://gist.github.com/q66/bbbfeb601b51477643911141257d00e7 | 20:15 |
| q66 | iirc at least modpost is used during module builds, i'm not 100% sure about the others | 20:17 |
| josch | q66: the kernel that minute is using is probably from the Debian package? cross building the debian package produces packages with the same content as building them natively. | 20:21 |
| minute | yes | 20:21 |
| q66 | josch: not sure i understand | 20:21 |
| q66 | when you crossbuild a kernel, it needs to build some intermediate binaries with hostcc, because it runs them | 20:22 |
| q66 | those binaries are also present in the target source tree | 20:22 |
| minute | woop woop https://source.mnt.re/reform/qcacld2/-/jobs/3934/artifacts/browse | 20:22 |
| josch | q66: you are not using gnu terminology for the architectures, no? | 20:22 |
| minute | there is a deb, lets see if that works | 20:22 |
| q66 | so when you crossbuild a kernel, you get your host-arch version of those, otherwise you get a target-arch version of those | 20:22 |
| josch | build architecture: the architecture you are building on | 20:22 |
| josch | host architectuer: the architecture you are building for | 20:22 |
| josch | target architecture: the architecture your compiler produces | 20:22 |
| q66 | in this case host architecture == what you rae building on, target architecture == what you are building for | 20:22 |
| minute | minutepocket: https://source.mnt.re/reform/qcacld2/-/jobs/3934/artifacts/browse | 20:22 |
| + Twodisbetter (2cc0e4ea1c@irc.cheogram.com) | 20:23 | |
| josch | q66: the debian package build builds some stuff twice -- once for the build and once for the host architecture | 20:23 |
| q66 | the kernel afaik uses the same terminology, because "hostcc" in kernel means compiler for the arch you are running the build on | 20:23 |
| josch | good | 20:23 |
| minute | ugh, gitlab ci makes all directories 0777 somehow | 20:23 |
| josch | the debian linux kernel build builds some utilities with hostcc (those end up in the produced binary) and some with the naitive cc (those are executed during the build) | 20:24 |
| q66 | the whole build/host/target terminology of gnu is confusing and unintuitive | 20:24 |
| josch | +1 | 20:24 |
| josch | unfortunately we are stuck with it | 20:24 |
| minutepocket | woops > reform-qcacld2 depends on linux-image-6.7.9-reform2-2+reform20240327t174618z; | 20:25 |
| - Twodisbetter (QUIT: Client Quit) (2cc0e4ea1c@irc.cheogram.com) | 20:25 | |
| q66 | josch: anyway to crossbuild the kernel you end up with a source tree with specific binaries for the host(build) architecture; in theory (maybe in practice too?) you could later rebuild them for the package with the target(host) compiler to get a similar package to non-cross-build | 20:25 |
| q66 | but then you end up with the opposite problem when cross-building moduls | 20:26 |
| q66 | *modules | 20:26 |
| q66 | because cross-building modules assumes a kernel source/headers tree that was crossbuilt | 20:26 |
| q66 | this is a sort of unfortunate limitation/bad design of the kernel build system | 20:27 |
| minute | ok, my hacky .deb works. i just need to fix the permission problems or migrate to sbuild or so | 20:27 |
| q66 | basically when cross-building modules => you are expected to have a cross-built source tree, when native-building modules => you are expected to have a natively-built source tree | 20:28 |
| q66 | so i'm interested in what kind of magic is used to make it work when you mix them, if it's not just luck | 20:28 |
| - minutepocket (QUIT: Remote host closed the connection) (~reform@p509936c9.dip0.t-ipconnect.de) | 20:28 | |
| minute | josch: how horrible is this for you? https://source.mnt.re/reform/qcacld2/-/blob/binary-deb/.gitlab-ci.yml | 20:29 |
| minute | (on a scale of 1-10) | 20:29 |
| q66 | ah, you are calling modules_prepare | 20:30 |
| q66 | this will rebuild the binaries | 20:30 |
| q66 | for your environment | 20:30 |
| q66 | though it's not guaranteed to always work when your tree is already "dirty" | 20:31 |
| josch | minute: i'll be afk until ~22:00 unfortunately | 20:31 |
| q66 | i guess in this case it does | 20:31 |
| josch | (bringing the little one to bed) | 20:31 |
| josch | will have a look then | 20:31 |
| minute | josch: sure thing! looking fwd to your feedback | 20:31 |
| - sevan (QUIT: Ping timeout: 255 seconds) (~sevan@user/venture37) | 20:45 | |
| + f_[xmpp] (fffdb90022@fases/developer/funderscore) | 20:46 | |
| + Twodisbetter (2cc0e4ea1c@irc.cheogram.com) | 20:59 | |
| + chomwitt (~chomwitt@2a02:587:7a17:f700:1ac0:4dff:fedb:a3f1) | 21:18 | |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@utdpat241106.utdallas.edu) | 21:29 | |
| + sevan (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 21:47 | |
| - sevan (QUIT: Changing host) (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 21:47 | |
| + sevan (~sevan@user/venture37) | 21:47 | |
| josch | minute: you are distributing the resulting .deb via the mntre.com repo? | 22:19 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!