| josch | noooooooooooo | 00:00 |
|---|---|---|
| josch | https://mister-muffin.de/p/8zbJ.png | 00:00 |
| josch | never reboot a running system, kids... :( | 00:00 |
| - mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 00:05 | |
| + tomenzgg (~tomenzgg@24.148.84.215) | 00:12 | |
| - tomenzgg (QUIT: Changing host) (~tomenzgg@24.148.84.215) | 00:12 | |
| + tomenzgg (~tomenzgg@user/tomenzgg) | 00:12 | |
| - vyvv (QUIT: Quit: Leaving) (~vyvv@user/vyvv) | 00:15 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 00:16 | |
| minute | josch: anything in dmesg? do you perchance have a custon mesa build that is thwarted by apparmor? | 00:26 |
| minute | josch: on friday i struggled a while with testing mesa in /usr/local/lib, because apparmor prevented certain gtk apps to load stuff from there | 00:28 |
| josch | minute: that might exactly be it! | 00:28 |
| josch | minute: my apt logs say that i installed apparmor a few days ago as a dependency of another package | 00:29 |
| josch | lets remove it | 00:29 |
| josch | nope, wasn't it. and no, nothing in dmesg and no custom mesa | 00:34 |
| josch | output on the console: https://mister-muffin.de/p/Iiwu.txt | 00:36 |
| josch | hrm... i also upgraded from linux 7.1 to 7.2 without rebooting | 00:41 |
| minute | josch: removing apparmor doesn't disable it | 00:41 |
| josch | oh | 00:41 |
| josch | i even removed with --purge | 00:41 |
| minute | josch: i had to put apparmor=0 in my commandline | 00:41 |
| josch | o0 | 00:41 |
| josch | as an environment variable? | 00:42 |
| minute | no | 00:42 |
| minute | kernel | 00:42 |
| josch | ah okay, lets try that | 00:42 |
| josch | oh no what's this: Could not connect to source.mnt.re:443 (152.53.105.8), connection timed out | 00:42 |
| minute | hm? | 00:43 |
| minute | hm yeah seems down, weird | 00:43 |
| minute | (tried on my phone, walking home atm) | 00:44 |
| josch | nope, apparmor=0 in /proc/cmdline but didn't do it | 00:45 |
| josch | it works with MESA_GL_VERSION_OVERRIDE=3.3 MESA_GLSL_VERSION_OVERRIDE=330 | 00:46 |
| josch | but that wasn't necessary before | 00:46 |
| - tomenzgg (QUIT: Ping timeout: 252 seconds) (~tomenzgg@user/tomenzgg) | 00:53 | |
| minute | josch: ah ok. which opengl + es versions do you normally get in glxinfo, without the overrides? | 01:08 |
| minute | josch: i don't believe it's a kernel thing then in any case, more likely the game changed requirements or the mesa/etnaviv gallium (userspace part of the driver) changed something. | 01:09 |
| josch | Max GLES[23] profile version: 3.1 | 01:09 |
| minute | josch: ah i wrote etnaviv for some reason lol | 01:10 |
| minute | josch: buuut you're on rk3588 right | 01:10 |
| josch | yes | 01:10 |
| minute | josch: so you should have vulkan | 01:10 |
| minute | josch: but the log says no vulkan | 01:10 |
| josch | yes | 01:10 |
| minute | josch: what does vkcube say? | 01:10 |
| josch | Selected GPU 0: Mali-G610 MC4, type: IntegratedGpu | 01:11 |
| josch | spins the cube | 01:11 |
| minute | ok nice | 01:11 |
| minute | > Unable to open device : No such file or directory (VK_ERROR_INCOMPATIBLE_DRIVER) | 01:11 |
| minute | before we strace that, what kind of environment does the game run in? and is it the arm64 build? | 01:12 |
| josch | yes arm64 build on gnome with debian unstable | 01:12 |
| josch | the odd thing is that it stopped working after a reboot | 01:13 |
| minute | josch: ok like, directly, not steam etc... then it's not a matter of a library in a container not fitting to the host | 01:13 |
| josch | i had a pretty long uptime of maybe 2 weeks where i installed a bunch of things in the meantime | 01:13 |
| josch | yes, directly without steam | 01:13 |
| minute | josch: next Q: you do have a custom build of mesa in /usr/local/lib ... ? | 01:13 |
| josch | /usr/local/lib | 01:13 |
| josch | is empty | 01:13 |
| minute | ah no custom build | 01:14 |
| minute | then i would do: strace -e trace=open,openat factorio 2>&1 | 01:15 |
| minute | and look at failed file opens before that error happens | 01:15 |
| minute | for me it was access denied | 01:15 |
| josch | https://mister-muffin.de/p/jjV6.txt | 01:18 |
| josch | i only get some ENOENT for /etc/drirc | 01:18 |
| josch | oh wait | 01:18 |
| minute | josch: another thing that comes to mind: it's unknown to me if the game ever worked with vulkan for you. but i realized that gl 3.1 was the norm without PAN_MESA_DEBUG=gl3 which used to give 3.3. do you not have that variable anymore? | 01:19 |
| josch | nope that wasn't it either | 01:19 |
| josch | yeah, i should've collected a log when it was working earlier today :( | 01:19 |
| minute | the bad_match comes first and that's when it tries a GL 3.3 context probably | 01:20 |
| josch | i don't have PAN_MESA_DEBUG in my environment | 01:20 |
| minute | and then the vulkan error strangely comes from the intel driver | 01:20 |
| josch | but it's set in /etc/profile.d/reform-mali.sh | 01:20 |
| minute | josch: sooo... it's set or not set? | 01:20 |
| josch | the same starts fine with PAN_MESA_DEBUG=gl3 | 01:20 |
| josch | it was not set | 01:20 |
| minute | josch: so reform-mali.sh is not working? | 01:20 |
| josch | at least it's not set in gnome terminal | 01:20 |
| josch | yes, but no idea why it stopped | 01:20 |
| minute | ok but that's the explanation then at least | 01:21 |
| josch | yes, it's a step further | 01:21 |
| josch | i'll see if i can build a fresh system image tomorrow and investigate whether gnome has the variable there or not | 01:21 |
| minute | source was down because it's pummeled by bots | 01:22 |
| josch | because it is still exported in /etc/profile.d/reform-mali.sh | 01:22 |
| josch | this should be in reform-check -- warn if the variable is gone from the environment | 01:22 |
| minute | josch: do you have it in the linux console? | 01:23 |
| josch | found it | 01:28 |
| josch | i switched my shell to fish | 01:28 |
| josch | turns out that it ignores /etc/profile.d | 01:28 |
| minute | eek | 01:29 |
| minute | meanwhile source.mnt.re is now blocking 29 unique IPs per second (!) at the moment | 01:29 |
| minute | really nasty attack wave | 01:31 |
| josch | o0 | 01:33 |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@71.31.254.246) | 01:37 | |
| - dd88 (QUIT: Ping timeout: 258 seconds) (~dd88@2601:646:8683:7820:9c6c:50a1:f48:db35) | 01:57 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 02:00 | |
| - dd88 (QUIT: Ping timeout: 257 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 02:07 | |
| + bkeys (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net) | 02:35 | |
| + paperManu_ (~paperManu@198.58.136.247) | 02:48 | |
| minute | cool, panfrost developers don't want llm commits https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/45124 | 02:55 |
| + dd88 (~dd88@2601:646:8683:7820:9c6c:50a1:f48:db35) | 02:57 | |
| - paperManu_ (QUIT: Ping timeout: 271 seconds) (~paperManu@198.58.136.247) | 03:01 | |
| - dd88 (QUIT: Ping timeout: 262 seconds) (~dd88@2601:646:8683:7820:9c6c:50a1:f48:db35) | 03:02 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 03:04 | |
| - paperManu (QUIT: Ping timeout: 260 seconds) (~paperManu@198.58.136.247) | 03:32 | |
| + paperManu (~paperManu@198.58.136.247) | 03:41 | |
| - paperManu (QUIT: Ping timeout: 254 seconds) (~paperManu@198.58.136.247) | 03:57 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:58 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:58 | |
| + paperManu (~paperManu@79.127.134.34) | 03:59 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:06 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:06 | |
| - dd88 (QUIT: Ping timeout: 248 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 04:17 | |
| - paperManu (QUIT: Read error: Connection reset by peer) (~paperManu@79.127.134.34) | 04:30 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 04:47 | |
| - Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66) | 05:00 | |
| + spew (~spew@user/spew) | 05:42 | |
| - dd88 (QUIT: Ping timeout: 246 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 06:28 | |
| josch | it is really heartwarming to see more and more projects taking a clear stance (that's what i thought could happen when i helped draft the Debian GR shortly after projects like codeberg published their own policy but oh well here we are...) | 06:38 |
| josch | and reform-check is going to have a test for the environment variables exported by /etc/profile.d/reform-*.sh scripts in its next release -- lets hope that next time somebody has my problem, debugging this is going to be quicker :) | 06:40 |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 07:25 | |
| - dd88 (QUIT: Ping timeout: 263 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 07:30 | |
| + mjw (~mjw@gnu.wildebeest.org) | 07:50 | |
| - mjw (QUIT: Ping timeout: 251 seconds) (~mjw@gnu.wildebeest.org) | 08:13 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 08:31 | |
| - dd88 (QUIT: Ping timeout: 265 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 08:36 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 09:38 | |
| - dd88 (QUIT: Ping timeout: 266 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 09:42 | |
| + mjw (~mjw@gnu.wildebeest.org) | 09:44 | |
| minute | josch: nice! | 11:25 |
| gordon1 | messed up pcie impedance calculations, well hope it's going to be fine for 4mm long traces | 11:31 |
| bremner | any idea what is going on with my nvme (pocket reform): https://paste.debian.net/hidden/eff1c750 | 11:46 |
| gordon1 | bremner: dmesg | tail -50 ; nvme smart-log /dev/nvme0 ; nvme error-log /dev/nvme0 pls | 11:50 |
| bremner | dmesg is full of yelling about firmware | 11:51 |
| gordon1 | nvme list? | 11:51 |
| bremner | nothing under nvme list | 11:52 |
| gordon1 | but you have /dev/nvme0, don't you? maybe it just needs sudo? | 11:52 |
| gordon1 | do you have nvme in lspci? | 11:53 |
| bremner | welp. nvme smart-log /dev/nvme0 says dev/nvme0: Resource temporarily unavailable. Maybe because anonther partition is in use? | 11:53 |
| bremner | yes, nvme shows under lspci, and nvme0n1 looks ok in lsblk | 11:55 |
| gordon1 | bremner: no, mount should not interfere with nvme smart-log | 11:57 |
| gordon1 | you have to show dmesg, can't tell anything without that | 11:57 |
| bremner | https://paste.debian.net/hidden/6cdf73bd | 11:58 |
| gordon1 | oh, yeah that's unrelated to nvme | 11:59 |
| bremner | ok, stopped reform-power-daemon, let's try again | 11:59 |
| gordon1 | whatever has happened - happened quite a bit before | 11:59 |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 12:01 | |
| bremner | now I get several reps of "Buffer I/O error on dev nvme0n1p1, logical block 0, async page read" followed by nvme nvme0: Identify namespace failed (-5) | 12:01 |
| gordon1 | bremner: you don't have rootfs on nvme, do you? if not you can try to do the following: find device number (first column) in lspci, then do echo 1 >/sys/bus/pci/devices/<device number>/rescan, if that does not help do echo 1 >.../remove and then echo 1 >/sys/bus/pci/rescan | 12:01 |
| gordon1 | bremner: when you do that - try nvme smart-log and error-log first before you try your mkfs again | 12:02 |
| gordon1 | and check dmesg | 12:02 |
| jah | minute: hi there, I'm the one who have issue with running my pocket reform on battery and have a v2 charging board | 12:03 |
| jah | I did receive the new batteries, but before putting them in place, I did measure last time the two stock battery, and weirdly I got: 0.0v and 7.1v | 12:03 |
| jah | then I swap the battery, and when I measure I get 0.0v on both batteries (and the pocket does not boot)... now if I put back the old one, I still measure 0.0 | 12:03 |
| jah | I did try my multimeter on an other DC power supply I have around and it works fine... any idea what could be the issue then? | 12:03 |
| - dd88 (QUIT: Ping timeout: 246 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 12:05 | |
| bremner | gordon1: uh, it came back as nvme1 | 12:07 |
| gordon1 | bremner: sure, but does it work? | 12:08 |
| bremner | oof. then went away again | 12:08 |
| gordon1 | okay, then dmesg should have something interesting | 12:08 |
| bremner | https://paste.debian.net/hidden/bfafbe40 | 12:09 |
| bremner | I wonder if the device is not seated properly or something | 12:09 |
| bremner | I guess I should try rebooting before opening it up | 12:09 |
| gordon1 | yeah no harm in opening and trying to re-seat it, also check if its temperature is unreasonable | 12:10 |
| gordon1 | bremner: did it work before or is it new fresh nvme? | 12:11 |
| bremner | it worked enough to set up a partition with luks, but more or less unused | 12:12 |
| gordon1 | you can do a diagnostic run with pci=nomsi and/or pcie_aspm.policy=performance on kernel command line to see if that solves the issue | 12:13 |
| gordon1 | but my bet would be it's probably a badly seated nvme | 12:13 |
| bremner | hopefully this attempt to boot off of nvme falls back to mmc after a bit | 12:15 |
| bremner | I see "Starting kernel ..." as the last line on the serial console | 12:16 |
| gordon1 | is it trying to boot from nvme? | 12:17 |
| bremner | yeah, that is the motivation here. I am part way through install guix system to the nvme. | 12:17 |
| bremner | weirdly it seemed to be actually getting the the initrd and dtb ok | 12:18 |
| gordon1 | right, well, can you ask it not to in u-boot? | 12:18 |
| gordon1 | you have serial console, i guess you can jump in the u-boot console, can't you? | 12:18 |
| bremner | I can try | 12:18 |
| bremner | ok, I am at the => prompt | 12:19 |
| bremner | ok, phew. Apparently C-c out of the nvme menu works | 12:20 |
| gordon1 | yeah no familiar with pocket, i assume it should be the same as classic, so you can try bootflow cmd but first you need to remove nvme from bootopts | 12:21 |
| gordon1 | boot_targets, that's the name of the var | 12:21 |
| gordon1 | but you found another workaround i guess | 12:21 |
| bremner | yeah. Not sure how reproducible, but I am booted now | 12:22 |
| bremner | nvme smart-log and nvme error-log are working again | 12:24 |
| bremner | nothing obvious in either. temp is 42C, doesn't sound so bad | 12:25 |
| bremner | 85 unsafe shutdowns | 12:25 |
| minute | jah: how did you measure them exactly? 7v can't be right | 12:26 |
| gordon1 | bremner: leave it for few minutes without touching and see if it drops from the bus again, if so that is probably ASPM issue | 12:27 |
| bremner | ok | 12:28 |
| jah | minute: so I connect them to the charging board, and then I put the pins of the multimeters to the connector side which is on the board | 12:28 |
| jah | the pin of the multimeter are too big to put them directly in the connector attached to the batteries | 12:29 |
| minute | gordon1: about 4mm traces, as far as i know that falls under "lumped circuit" or something | 12:30 |
| gordon1 | yeah probably will be fine | 12:30 |
| gordon1 | did so much of this calculation multiple times to make sure i didn't mess it up, so in the end i did mess it up by selecting 4 layers instead of two | 12:31 |
| minute | jah: so you measured on the two pins of a single connector? or did you measure from minus of one to plus of the other? | 12:31 |
| minute | jah: or did you measure from ground to plus? | 12:32 |
| minute | gordon1: well 4 layers is probably better, if the inner layers are GND :D | 12:33 |
| gordon1 | no no, it's the other way around, i have 2 layer board, but selected 4 layers in the calculator by accident | 12:33 |
| minute | gordon1: uh oh! do you have gnd on the back then? solid gnd under the pairs? | 12:34 |
| jah | minute: does that beautiful drawing answer your question? https://imgur.com/a/ePLlmI1 | 12:34 |
| minute | gordon1: btw i'm pretty sure the diff pairs on the classic mb are off as well | 12:34 |
| minute | jah: hm that looks good, not sure how you could get 7v that way | 12:35 |
| gordon1 | minute: well, i mean it's complicated, https://tinystash.undef.im/il/5EBiFinv2gKuPgZYLkYKpX6yEHQg23jmkxpx942DKRE7zsR4kUFkMuzcebdhdtzd1doeNQ8ghpqbtw9i9bZWRwNT.png but i'm pretty sure it's going to be fine just judging by the fact that those are so short | 12:35 |
| minute | jah: well maybe if one cell is a shunt | 12:35 |
| gordon1 | it's just now it is there and i already ordered | 12:35 |
| jah | minute: maybe the multimeter was tripping, I can not repuce that measurement | 12:36 |
| minute | jah: and the charger is powered and tries to apply 7v | 12:36 |
| jah | reproduce* | 12:36 |
| minute | maybe it was there only for a moment | 12:36 |
| gordon1 | and it will bug me constantly until i change it | 12:36 |
| jah | minute: it was not connected to power supply for days | 12:36 |
| jah | minute: actually, maybe I should connect to the power supply to try to boot wit the batteries? | 12:36 |
| jah | the new batteries* | 12:36 |
| jah | I thought they should be charged already | 12:36 |
| minute | gordon1: haha oh right, 2 layers with card gives a nice gnd plane split antenna there | 12:37 |
| minute | 2 layers with card edge i mean | 12:37 |
| gordon1 | minute: may i draw your attention to the yesterday's question about keyboard diodes again? | 12:38 |
| minute | jah: new batteries have the correct polarity yes? just double checking | 12:38 |
| minute | gordon1: sure | 12:38 |
| gordon1 | minute: i noticed that keyboard4 classic diodes mention part number 1N4007WS which is in SOD-323 but the footprint is 0603 (hand solder version), is that intentional or did you replace diodes with some other part? | 12:38 |
| gordon1 | they do do fit but it's a bit tight | 12:39 |
| minute | gordon1: that's ancient times but iirc sod323 fits that footprint as well probably? | 12:39 |
| minute | gordon1: i probably sometimes use different diodes | 12:39 |
| minute | the specs of those are not very important | 12:39 |
| minute | generic small signal diodes | 12:39 |
| gordon1 | right, i see | 12:40 |
| gordon1 | i just sorta threw the BOM at lcsc and assumed it should work | 12:40 |
| gordon1 | looks like it does but it made me a bit nervous :) | 12:41 |
| gordon1 | minute: https://tinystash.undef.im/il/2DAZorRCMxY5ik5m9sneiPRoM71h4xwSeQyU4NQ5dDdfcCwnYqhgu6nfQodicXSLcz7U1He3XbypuKdhrgsYzjT2.png | 12:42 |
| + paperManu (~paperManu@198.58.136.247) | 12:47 | |
| minute | gordon1: woah that's exciting | 12:48 |
| gordon1 | minute: can't quite find bigger than 1x1 keycaps that i like so had to add this trench with some indication leds in the middle | 12:49 |
| minute | gordon1: it reminds me of the olkb design | 12:49 |
| minute | it had a similar trench | 12:49 |
| gordon1 | right, i feel better about it then! | 12:50 |
| minute | gordon1: this one https://www.reddit.com/r/olkb/comments/ocf32y/ortholinear_laptop_making_a_keyboard_kit_for_the/ | 12:50 |
| minute | was unfortunately never published | 12:50 |
| gordon1 | oh yeah i remember that picture, well i promise to publish mine on source.mnt.re if it works! | 12:51 |
| gordon1 | it should mostly be ready for jlc assembly order, except the fpc connectors on the bottom probably need to be hand-soldered | 12:52 |
| ianthetechie | Ooh an ortholinear layout would be really nice. I’m spoiled on my Kinesis but even a flat ortho layout would be great. | 12:57 |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 12:58 | |
| gordon1 | just a caveat, i did remove backlight here since i plan to use it with blank keycaps, but if there is a demand i can try to make some time and put it back | 12:59 |
| jah | minute: I did the red of the new batteries to the +, but how can I check if their polarities are correct? | 13:01 |
| jah | connect the red* | 13:01 |
| - dd88 (QUIT: Ping timeout: 258 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 13:03 | |
| minute | jah: red to plus is good | 13:04 |
| jah | minute: I'll try putting the new batteries with the power plug this time and see if it boots | 13:04 |
| minute | jah: some models have red and black swapped vs our original batteries, that's only why i was wondering | 13:04 |
| jah | minute: okay let me compare the wiring | 13:05 |
| jah | minute: they are wired similarly (old and new batteries) | 13:05 |
| jah | minute: okay my bad, I just put the new battery and I realize that if I don't put the pin of the multimeter right on the connector it does not measure anything (I was putting in the board, not on the connector itself), now the ready of the new battery is both ~3.9 | 13:09 |
| jah | it should be booting then, even without pluggin to external power supply ... but it was not earlier, I will try again now | 13:10 |
| jah | ACTION feels ashamed | 13:14 |
| jah | minute: I think I had forgotten to turn the lateral switch on last time I tried, no it works perfectly \o/ | 13:14 |
| jah | ACTION do the reform pocket dance | 13:14 |
| jah | minute: sorry for the trouble, and thanks so much for your support | 13:14 |
| jah | I still have one question though, I want to get back on hacking the keyboard firmware... so far I did some small changes and it went well, but now I wan to do more significant change I fear that I might break something around the hyper+enter logic, if that happen, am I totally screwed or is there a way to recover? maybe with an external keyboard or some jumper placement inside? | 13:15 |
| + siviq (~siviq@user/siviq) | 13:18 | |
| + svp (~svp@host-79-57-105-252.business.telecomitalia.it) | 13:25 | |
| gordon1 | jah: i don't have a pocket and don't know much about how it is done there, but my guess you would need one of these https://shop.mntre.com/products/mnt-pocket-reform-keyboard-usb-c-adapter?taxon_id=13 | 13:26 |
| jah | gordon1: yeah, that looks like exactly what I need indeed! thank you :) | 13:27 |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 13:47 | |
| - dd88 (QUIT: Ping timeout: 250 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 13:56 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 14:15 | |
| - dd88 (QUIT: Ping timeout: 268 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 14:19 | |
| - paperManu (QUIT: Ping timeout: 278 seconds) (~paperManu@198.58.136.247) | 14:44 | |
| - mjw (QUIT: Ping timeout: 261 seconds) (~mjw@gnu.wildebeest.org) | 14:54 | |
| + mjw (~mjw@2001:1c00:f482:400:e8e0:45d8:6dac:b7b7) | 15:02 | |
| jogu | jah: There's physical buttons on the PCB to force a reset / force the keyboard into flashing mode. You'd only need that PCB if you have no way of getting the pocket back on at all and need to use another computer to flash back to stock firmware | 15:03 |
| + riabenko (~riabenko@2a01:d0:333::831) | 15:03 | |
| + mark_ (~mjw@2001:1c00:f482:400:c901:9e4f:ea0c:d6f4) | 15:06 | |
| - mjw (QUIT: Killed (iridium.libera.chat (Nickname regained by services))) (~mjw@2001:1c00:f482:400:e8e0:45d8:6dac:b7b7) | 15:06 | |
| * mark_ -> mjw | 15:06 | |
| + Guest9086 (~mjw@2001:1c00:f482:400:e8e0:45d8:6dac:b7b7) | 15:06 | |
| + paperManu (~paperManu@modemcable065.144-200-24.mc.videotron.ca) | 15:07 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 15:10 | |
| - dd88 (QUIT: Ping timeout: 254 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 15:14 | |
| - Guest29952 (QUIT: Quit: The Lounge - https://thelounge.chat) (~ndufresne@transit.collaboradmins.com) | 15:22 | |
| minute | yay, my mesa MR got merged https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/45130 | 15:29 |
| minute | jah: no problem, i'm glad it works now! | 15:29 |
| minute | jah: yes the adapter gordon1 linked was originally made for exactly this case: hacking on the keyboard with no regrets | 15:30 |
| siviq | Is the adapter compatible with the standalone keyboard? Or it doesn't really need it. | 15:32 |
| - qbit (QUIT: Remote host closed the connection) (~qbit@user/qbit) | 15:33 | |
| + qbit (~qbit@user/qbit) | 15:36 | |
| minute | siviq: it's integrated in the standalone keyboard. | 15:38 |
| minute | (the design is actually a bit smaller now than what's on the shop picture) | 15:39 |
| minute | but same functionality | 15:39 |
| siviq | minute perfect, as I am certain I will need it. :) | 15:40 |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 15:50 | |
| - dd88 (QUIT: Ping timeout: 241 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 15:54 | |
| - xktr (QUIT: Ping timeout: 249 seconds) (~xktr@user/xktr) | 16:00 | |
| + xktr (~xktr@user/xktr) | 16:01 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 16:08 | |
| + siviq (~siviq@user/siviq) | 16:08 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 16:31 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 16:51 | |
| - dd88 (QUIT: Ping timeout: 240 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 16:55 | |
| - riabenko (QUIT: Quit: riabenko) (~riabenko@2a01:d0:333::831) | 17:09 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 17:34 | |
| - aperezdc (QUIT: Remote host closed the connection) (~aperezdc@2a03:6000:6e61:633::43) | 17:37 | |
| - dd88 (QUIT: Ping timeout: 244 seconds) (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 17:38 | |
| + dd88 (~dd88@2601:646:8683:7820:84af:ca3a:c8cf:b9c4) | 17:50 | |
| minute | this will be interesting to tune https://docs.qualcomm.com/doc/80-70015-10/topic/18-customize.html | 17:53 |
| jfred | o.o wow that's a lot of low-level tunables | 17:57 |
| + riabenko (~riabenko@lab.riabenko.com) | 17:58 | |
| + aperezdc (~aperezdc@2a03:6000:6e61:633::43) | 18:00 | |
| - Guest9086 (QUIT: Ping timeout: 246 seconds) (~mjw@2001:1c00:f482:400:e8e0:45d8:6dac:b7b7) | 18:22 | |
| + Guest9086 (~mjw@2001:1c00:f482:400:2ddb:78dd:775f:43b8) | 18:26 | |
| - riabenko (QUIT: Quit: riabenko) (~riabenko@lab.riabenko.com) | 18:38 | |
| jah | jogu: I see, that is very good to know thank you! | 18:44 |
| * mjw -> Guest9357 | 18:57 | |
| * Guest9086 -> mjw | 18:57 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 19:15 | |
| + riabenko (~riabenko@lab.riabenko.com) | 19:43 | |
| - f_ (QUIT: Ping timeout: 261 seconds) (16abab341f@postmarketOS/funderscore) | 20:15 | |
| + f_ (16abab341f@postmarketOS/funderscore) | 20:17 | |
| - f_ (QUIT: Ping timeout: 246 seconds) (16abab341f@postmarketOS/funderscore) | 20:22 | |
| + f_ (16abab341f@postmarketOS/funderscore) | 20:22 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 20:23 | |
| * mjw -> Guest3996 | 20:24 | |
| * mark_ -> mjw | 20:24 | |
| - f_ (QUIT: Ping timeout: 240 seconds) (16abab341f@postmarketOS/funderscore) | 20:26 | |
| - Guest3996 (QUIT: Ping timeout: 240 seconds) (~mjw@2001:1c00:f482:400:2ddb:78dd:775f:43b8) | 20:26 | |
| jah | hey folks, just wanted to report that I can not access http://source.mnt.re/ | 20:35 |
| jah | it looks like the connection timeout from here | 20:35 |
| + f_ (16abab341f@postmarketOS/funderscore) | 20:37 | |
| - jah (QUIT: Ping timeout: 243 seconds) (~jahkosha@user/jahkosha) | 20:45 | |
| - buckket (QUIT: Ping timeout: 274 seconds) (~buckket@vps.buckket.org) | 20:45 | |
| + buckket (~buckket@vps.buckket.org) | 20:46 | |
| + jahkosha (~jahkosha@user/jahkosha) | 20:47 | |
| * jahkosha -> jah | 20:47 | |
| - jah (QUIT: Ping timeout: 263 seconds) (~jahkosha@user/jahkosha) | 20:57 | |
| + jahkosha (~jahkosha@user/jahkosha) | 20:58 | |
| * jahkosha -> jah | 20:59 | |
| minute | hm, restarting gitlab | 21:19 |
| - spew (QUIT: Quit: nyaa~) (~spew@user/spew) | 21:24 | |
| - paperManu (QUIT: Quit: WeeChat 4.9.5) (~paperManu@modemcable065.144-200-24.mc.videotron.ca) | 21:30 | |
| + paperManu (~paperManu@modemcable065.144-200-24.mc.videotron.ca) | 21:34 | |
| - gianmarcogg03 (QUIT: Ping timeout: 251 seconds) (~quassel@user/gianmarcogg03) | 21:44 | |
| + gianmarcogg03 (~quassel@user/gianmarcogg03) | 21:46 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 22:04 | |
| - TadeusTaD (QUIT: Remote host closed the connection) (tadeustad@user/TadeusTaD) | 22:07 | |
| - riabenko (QUIT: Quit: riabenko) (~riabenko@lab.riabenko.com) | 22:10 | |
| - cellmoose (QUIT: Ping timeout: 269 seconds) (cellmoose@tilde.green) | 22:12 | |
| + TadeusTaD (tadeustad@psifactor.pl) | 22:13 | |
| - TadeusTaD (QUIT: Changing host) (tadeustad@psifactor.pl) | 22:13 | |
| + TadeusTaD (tadeustad@user/TadeusTaD) | 22:13 | |
| vagrantc | josch: does the mnt reform kernel building process produce a debian source package (.dsc) ... ? | 22:26 |
| vagrantc | if it did ... i could in theory grab the patches out of the .debian.tar | 22:28 |
| minute | gitlab is feeling healthier now i think | 22:31 |
| vagrantc | ACTION sends a get well card | 22:32 |
| - lidstah (QUIT: Ping timeout: 243 seconds) (~lidstah@gateway/tor-sasl/lidstah) | 22:43 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 22:57 | |
| lidstah | meh, ssd crapped out on an old chap (small celeron nuc from 2015) which ran my irc vm… | 22:58 |
| lidstah | and hi all | 22:58 |
| lidstah | fsck and it's up again | 22:58 |
| lidstah | smart seems legit, not the first time it happens with this machine (with another SSD back in the days which I still have around, might as well give it a proper smart test) | 23:02 |
| - paperManu (QUIT: Ping timeout: 274 seconds) (~paperManu@modemcable065.144-200-24.mc.videotron.ca) | 23:06 | |
| minute | wb lidstah | 23:07 |
| lidstah | hey minute :) | 23:09 |
| lidstah | sorry for the semi-rant ^^' | 23:09 |
| + cellmoose (cellmoose@tilde.green) | 23:12 | |
| josch | vagrantc: the reform-debian-packages pipeline uses sbuild to build packages and sbuild cannot build something of which it cannot build a .dsc because the .dsc is the transport format to copy source packages into the chroot -- so the question is more about: does the machinery discard that dsc or does it get copied into the repository | 23:13 |
| josch | vagrantc: and unfortunately it only includes the arch-specific .changes file which sbuild spits out and sbuild is run with --no-source-only-changes | 23:15 |
| josch | lidstah: don't forget backing that system up regularly before the ssd gives up for good! | 23:17 |
| vagrantc | josch: hmmm... will have to scheme up a different more effortful approach.... :) | 23:17 |
| vagrantc | josch: the problem i'm having is guix can check out git branches from the build system, but it throws away the history ... which makes this tricky to generate the kernel patches from git history :) | 23:18 |
| ^alex | who's Colonel Patches? | 23:21 |
| + paperManu (~paperManu@198.58.136.247) | 23:21 | |
| lidstah | josch: don't worry, the VMs are not on the local storage of this machine (and they're backed up, proxmox backup server is my friend :D). Might as well retire it from my proxmox cluster and use it as an Haiku box for the fun. I do keep my old celeron boxes in my proxmox cluster by pure nostalgy (they were parts of my first homelab virtualization cluster) - and hi :) | 23:22 |
| vagrantc | josch: even if i could generate the patches from history, it would be an expensive git checkout to get the entire kernel history | 23:23 |
| vagrantc | (even a shallow checkout is a bit expensive) | 23:24 |
| josch | vagrantc: who says that this cannot change? Let me try to see what happens if the build process also produces a linux_source.changes | 23:24 |
| vagrantc | josch: oh :) | 23:24 |
| josch | vagrantc: the linux/build.sh script does a shallow clone for that reason | 23:24 |
| vagrantc | josch: wait, you can change software? :) | 23:25 |
| ^alex | vagrantc, you jest, but we once worked at a place where someone said "you can't change that! we've modified it!" | 23:25 |
| minute | wow, linux 7.2 and 7.3 don't even have a GPU node yet for sm8750 | 23:26 |
| minute | https://lists.freedesktop.org/archives/dri-devel/2026-August/587659.html | 23:27 |
| minute | > | 23:27 |
| minute | This series add support for Adreno 830 GPU present in Snapdragon SM8750 chipset, which is almost 2 yr old now | 23:27 |
| josch | oh sounds like the endless cadence patch stack :) | 23:29 |
| + paperManu_ (~paperManu@142.169.16.123) | 23:50 | |
| josch | vagrantc: i see an orig.tar.xz here: https://mntre.com/reform-debian-repo/pool/main/l/linux/ | 23:57 |
| josch | vagrantc: the repo has a Sources.gz which includes src:linux | 23:57 |
| vagrantc | josch: promising :) | 23:59 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
