| - mjw (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org) | 00:39 | |
| - Gooberpatrol_66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66) | 01:01 | |
| + Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66) | 01:01 | |
| - AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net) | 01:06 | |
| sigrid | kfx: I've located my ft2232h board so I can get more involved in rk3588 alpining process | 01:24 |
|---|---|---|
| - chomwitt (QUIT: Ping timeout: 276 seconds) (~chomwitt@ppp178059008065.access.hol.gr) | 01:27 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:30 | |
| kfx | sigrid: ok, I've got the salient facts together | 01:31 |
| kop316 | So even with the most up to date keyboard firmware, the batteries do not charge with several chargers. What seems to happen is that at first, it gets the right current, but the battery voltage goes up to 4.0 or 4.2 V, then the current goes back down to almost nothing. Has anyone else seen or experienced this? | 01:34 |
| zeha | josch: nice | 01:54 |
| zeha | josch: haha the mouse trail | 01:55 |
| kfx | sigrid: http://okturing.com/src/25673/body and once this current build finishes I'll push my kernel build aport thing to git somewhere | 02:05 |
| kfx | sigrid: with these instructions and that kernel you hopefully won't need serial output | 02:12 |
| sigrid | noice, thank you | 02:25 |
| kfx | shit, I missed a spot. libpgiod is in the community repo so you have to enable that too | 02:38 |
| kfx | updated the toot at least, but the okturing post is ruined forever. | 02:40 |
| minute | josch: i don't know if you know it, but i saw casey use this tool today on/for the qcs6490 and i was perplexed how mega fast it is compared to update-initramfs https://gitlab.postmarketos.org/postmarketOS/postmarketos-mkinitfs | 02:56 |
| minute | they also told me about https://www.freedesktop.org/software/systemd/man/latest/systemd-repart.html | 02:57 |
| - L29Ah (QUIT: Read error: Connection timed out) (~L29Ah@wikipedia/L29Ah) | 03:19 | |
| + L29Ah (~L29Ah@wikipedia/L29Ah) | 03:20 | |
| - paperManu (QUIT: Ping timeout: 248 seconds) (~paperManu@72.10.128.164) | 03:48 | |
| + rpm12 (~rpm@pool-72-73-23-92.clppva.fios.verizon.net) | 04:12 | |
| - rpm12 (QUIT: Quit: Client closed) (~rpm@pool-72-73-23-92.clppva.fios.verizon.net) | 04:19 | |
| + rpm9 (~rpm@pool-72-73-23-92.clppva.fios.verizon.net) | 04:19 | |
| - rpm9 (PART: !!unknown attribute: msg!!) (~rpm@pool-72-73-23-92.clppva.fios.verizon.net) | 04:20 | |
| - cwebber (QUIT: Quit: crossing the threshold of Lugonu's domain) (~Christine@user/cwebber) | 04:25 | |
| + cwebber (~Christine@user/cwebber) | 04:32 | |
| josch | minute: i know of no tool slower than mkinitramfs. The trade-off is, that by switching initramfs generators we get improved speed on user's computers and we pay with our developer time when something goes wrong and with user's frustration when they have to debug a thing that they are not familiar with. I'm not in love with mkinitramfs either but personally, I rather maintain what is the default so that | 05:58 |
| josch | I don't have to allocate more time than necessary on it. | 05:59 |
| josch | minute: systemd-repart only changes the partition table itself. To also grow the filesystem you need systemd-growfs and we can switch to that once somebody teaches me how to make that grow the root filesystem which it is running from. | 05:59 |
| josch | The current solution grows the rootfs in the initramfs *before* it gets mounted. | 06:00 |
| - robin (QUIT: Ping timeout: 276 seconds) (~robin@user/terpri) | 06:50 | |
| + robin (~robin@user/terpri) | 06:56 | |
| + chomwitt (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 08:02 | |
| + tuesdaaay (~ruby@159.196.168.242) | 09:08 | |
| - tuesdaaay (QUIT: Remote host closed the connection) (~ruby@159.196.168.242) | 09:08 | |
| - L29Ah (QUIT: Ping timeout: 268 seconds) (~L29Ah@wikipedia/L29Ah) | 09:10 | |
| josch | minute: oh thank you for correcting my community post title typo! :D | 09:14 |
| + mjw (~mjw@gnu.wildebeest.org) | 09:24 | |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 09:25 | |
| josch | grimmware: thanks to your recent suggestion I now created this topic and pinned it globally (will expire automatically in 3 months): https://community.mnt.re/t/system-image-v5-gnome-default-configuration-changes-reform-tools-1-73-or-later/3521 | 09:54 |
| josch | minute: I just learned this about discourse: via pinning it's possible to make a thread appear globally independent of its category. So if we'd like to make an announcement that affects all users, we should probably use the "pin" functionality more so that it does not end up in "uncategorized" (because it's not specific to either classic or pocket) where maybe fewer people read it. | 09:59 |
| + L29Ah (~L29Ah@wikipedia/L29Ah) | 10:05 | |
| - L29Ah (QUIT: Read error: Connection reset by peer) (~L29Ah@wikipedia/L29Ah) | 10:47 | |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 11:00 | |
| + buckket (~buckket@vps.buckket.org) | 11:01 | |
| - mjw (QUIT: Ping timeout: 248 seconds) (~mjw@gnu.wildebeest.org) | 11:22 | |
| * Guest3019 -> mjw | 11:36 | |
| + reform22922 (~kanata@fp73a2d6f3.ap.nuro.jp) | 11:48 | |
| - reform22922 (QUIT: Client Quit) (~kanata@fp73a2d6f3.ap.nuro.jp) | 11:49 | |
| + paperManu (~paperManu@72.10.128.164) | 12:13 | |
| bremner | small moment of panic: after halting the PocRe, I didn't power off, so when I came back the next day, powering on did nothing | 12:21 |
| minute | ohh | 12:25 |
| + CephonAltera (~cephonalt@user/CephonAltera) | 12:26 | |
| CephonAltera | Hey, just wanna check, is it safe to update my system despite this? https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1102690 | 12:27 |
| + chomwitt (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 13:05 | |
| minute | josch: alright @ mkinitramfs | 13:06 |
| minute | josch: about systemd-repart, it sounded like it can generate image files with partitions etc | 13:06 |
| - CephonAltera (QUIT: Ping timeout: 260 seconds) (~cephonalt@user/CephonAltera) | 13:29 | |
| minute | my private gitlab token is expired and system image redirect is not updating, fixing that now | 13:43 |
| + L29Ah (~L29Ah@wikipedia/L29Ah) | 14:31 | |
| bremner | hmm. my Pocket Reform wifi woes definitely seem to be related to the AP. I set up a temporary OpenWRT access point and 2000 pings later it seems clean, no dropped packets or delays longer than 100ms | 14:33 |
| bremner | so I guess that's good news? No big surprise if my ISP supplied AP is terrible... | 14:33 |
| minute | nice | 14:35 |
| * funderscore -> f_ | 14:45 | |
| * f_ -> funderscore | 14:45 | |
| bkeys | josch: https://download.copr.fedorainfracloud.org/results/bkeys/kernel-reform/fedora-42-aarch64/09189685-kernel/builder-live.log.gz | 14:55 |
| bkeys | I keep getting these failed builds, if you look at the bottom it says "Error: arch/arm64/boot/dts/rockchip/rk3588-mnt-reform2-dsi.dts:241.1-8 Label or path pd_gpu not found" do you know what's going on at all? It's all kernel 6.14.11 | 14:55 |
| + cephon (~anko@mm-15-235-126-178.vitebsk.dynamic.pppoe.byfly.by) | 15:05 | |
| minute | bkeys: sounds like pd_gpu (power domain for the gpu) is missing in the rk3588.dtsi / rk3588-base.dtsi, maybe it was renamed or they have an old version? | 15:07 |
| bkeys | No I don't think I added those files at all | 15:07 |
| bkeys | minute: Where do I get those files from? | 15:08 |
| minute | bkeys: no these are part of linux | 15:08 |
| minute | bkeys: don't know how to phrase this better, so forgive me, but i think it would be really worth your time to learn a bit more about devicetree in linux, i think it'll lead to many "aha" moments | 15:09 |
| bkeys | No offense taken, it probably would | 15:09 |
| minute | bkeys: for example, if you look at this file, you will see some includes at the top: https://github.com/torvalds/linux/blob/master/arch/arm64/boot/dts/rockchip/rk3588-mnt-reform2.dts | 15:10 |
| minute | bkeys: > #include "rk3588-firefly-icore-3588q.dtsi" | 15:10 |
| minute | bkeys: so it includes this https://github.com/torvalds/linux/blob/master/arch/arm64/boot/dts/rockchip/rk3588-firefly-icore-3588q.dtsi | 15:10 |
| minute | bkeys: and this in turn has > #include "rk3588.dtsi" | 15:11 |
| minute | bkeys: "dtsi" means dts include | 15:11 |
| minute | bkeys: this is so that different devices can share the same "base" devicetree files that specify all the hardware that is common amongst those machines | 15:11 |
| minute | bkeys: like, the rk3588 chip itself is common, and doesn't need to be redeclared in every dts | 15:12 |
| bkeys | Yeah I mean I have done C++ development for over a decade so I get the idea of .dtsi files | 15:12 |
| minute | bkeys: rk3588.dtsi includes rk3588-base.dtsi and rk3588-extra.dtsi. why extra? because the rk3588s (note the "s") is a stripped down version that doesn't heave the "extra" features | 15:12 |
| minute | bkeys: like, the rk3588 is the big one and has more pcie, more hdmi and other stuff vs the rk3588s, but they share most of the base features. so all those are in rk3588-base.dtsi. | 15:13 |
| minute | bkeys: you can see pd_gpu declared here in line 882 https://github.com/torvalds/linux/blob/master/arch/arm64/boot/dts/rockchip/rk3588-base.dtsi | 15:15 |
| bkeys | https://github.com/torvalds/linux/blob/v6.14/arch/arm64/boot/dts/rockchip/rk3588-base.dtsi | 15:15 |
| minute | bkeys: if you switch to blame mode you can see it was added 4 months ago https://github.com/torvalds/linux/blame/master/arch/arm64/boot/dts/rockchip/rk3588-base.dtsi#L882 | 15:15 |
| bkeys | I guess this is the version of it I am dealing with | 15:15 |
| minute | bkeys: right. it seems to be not there yet in 6.14 | 15:16 |
| bkeys | I'll just bump the kernel version then | 15:16 |
| bkeys | 6.15 seems to have it | 15:17 |
| minute | sounds sensible... 6.14 is old hat now i guess | 15:17 |
| bkeys | There is 6.16 which would be path of least resistance but then I think I have to incorporate a ton of patches | 15:40 |
| bkeys | Because Collabara kernel isn't at 6.16 | 15:40 |
| bkeys | josch, minute: Do you have any clue what patches are required for kernel 6.16? IIRC the system image is at 6.16 | 15:43 |
| minute | bkeys: we don't have a 6.16 patchstack yet afaik. | 15:43 |
| bkeys | ACTION nods | 15:43 |
| bkeys | I think from a packaging standpoint I could get a tarball of the system image kernel where it's patched and ready to go and drop it into the Fedora infrastructure | 15:43 |
| bkeys | I'm guessing josch has scripts to create said tarball | 15:44 |
| bkeys | That would remove some duplicate effort | 15:44 |
| minute | bkeys: oh wait a second. i noticed your compile issue is from -dsi.dts. but why are you building that one? | 15:44 |
| bkeys | Ignorance | 15:44 |
| minute | bkeys: afaik you have the hdmi version | 15:44 |
| minute | bkeys: so you don't need that file | 15:44 |
| bkeys | Yeah but I was getting the same error from other .dts files | 15:44 |
| minute | bkeys: ok. | 15:45 |
| bkeys | https://download.copr.fedorainfracloud.org/results/bkeys/kernel-reform/fedora-42-aarch64/09189396-kernel/builder-live.log.gz | 15:45 |
| bkeys | Like in this build it was the Reform Next doing the same thing | 15:45 |
| minute | bkeys: one sec | 15:46 |
| minute | bkeys: let me check why we are referencing pd_gpu directly in the dts | 15:46 |
| bkeys | Alright | 15:46 |
| minute | lol ok https://source.mnt.re/reform/reform-debian-packages/-/blob/main/linux/rk3588-mnt-reform2.dts?ref_type=heads#L241 | 15:47 |
| minute | bkeys: just delete that | 15:47 |
| minute | bkeys: delete the following from all our rk3588 dts: | 15:47 |
| minute | &pd_gpu { | 15:47 |
| minute | status = "okay"; | 15:47 |
| minute | }; | 15:47 |
| minute | bkeys: copy that? i have to go now | 15:54 |
| - chomwitt (QUIT: Ping timeout: 252 seconds) (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 16:00 | |
| bkeys | minute: Yeah I had to play with my kids for a bit | 16:15 |
| + mark_ (~mjw@gnu.wildebeest.org) | 16:19 | |
| - bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net) | 16:48 | |
| + bkeys (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net) | 16:51 | |
| + ericsfraga (~user@2a00:23cc:b464:9901::99e) | 16:55 | |
| + AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net) | 17:44 | |
| bkeys | Alright I got a patched and dts kernel build going for 6.15; hopefully it doesn't fail | 17:46 |
| bkeys | Aaaaaand it failed, sigh. At least it failed for different reasons now | 17:50 |
| - L29Ah (QUIT: Ping timeout: 244 seconds) (~L29Ah@wikipedia/L29Ah) | 17:50 | |
| + chomwitt (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 18:02 | |
| - ericsfraga (QUIT: Quit: ERC 5.6.1-git (IRC client for GNU Emacs 31.0.50)) (~user@2a00:23cc:b464:9901::99e) | 18:14 | |
| * funderscore -> f_ | 18:29 | |
| + L29Ah (~L29Ah@wikipedia/L29Ah) | 19:37 | |
| - chomwitt (QUIT: Ping timeout: 276 seconds) (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 20:12 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 21:03 | |
| + josch-pocket (~josch-poc@194.213.108.6) | 21:08 | |
| minute | josch-pocket: just FYI i updated a pocket reform with a311d here that was on kernel 6.12 on sd card, and used setup-encrypted-disk to move to nvme, and then used reform-flash-uboot emmc, and then updated the full-desktop to migrate to gdm and gnome, and everything worked mega smoothly | 21:27 |
| - AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net) | 22:08 | |
| + chomwitt (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 22:45 | |
| - chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:85f:9a3f:9300:1ac0:4dff:fedb:a3f1) | 23:00 | |
| josch-pocket | minute: oh wow!! :D | 23:22 |
| josch-pocket | i must admit that while reading the message i kept waiting for the bug report description :D | 23:26 |
| josch-pocket | thanks to its battery life i used my pocket reform like others use their smart phone today | 23:55 |
| josch-pocket | i just had it in my backpack and looked stuff up when i needed to and then stuffed it back without switching it off completely | 23:55 |
| josch-pocket | (i was not connected to irc to spare you the frequent disconnects while connection drops during the train ride) | 23:56 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!