| - cobra (QUIT: Read error: Connection reset by peer) (~cobra@user/Cobra) | 01:39 | |
| + cobra_ (~cobra@user/Cobra) | 01:40 | |
| - GNUmoon (QUIT: Ping timeout: 252 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 01:41 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:41 | |
| - pomel0 (QUIT: Read error: Connection reset by peer) (~pomel0@user/pomel0) | 02:26 | |
| + pomel0 (~pomel0@user/pomel0) | 02:26 | |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@107.159.15.124) | 02:38 | |
| - spew (QUIT: Ping timeout: 244 seconds) (~spew@user/spew) | 02:45 | |
| + spew (~spew@user/spew) | 02:50 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 03:02 | |
| - paperManu_ (QUIT: Ping timeout: 240 seconds) (~paperManu@107.159.15.124) | 04:06 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 04:18 | |
| - spew (QUIT: Quit: nyaa~) (~spew@user/spew) | 04:41 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.3.83) | 05:38 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:39 | |
| - Ar|stote|is (QUIT: Client Quit) (~linx@149.210.3.83) | 05:39 | |
| + Ar|stote|is (~linx@149.210.3.83) | 05:39 | |
| - digitalrane (QUIT: *.net *.split) (~digitalra@user/digitalrane) | 06:39 | |
| + digitalrane (~digitalra@user/digitalrane) | 06:40 | |
| - amospalla (QUIT: Remote host closed the connection) (~jordi@user/amospalla) | 06:57 | |
| + amospalla (~jordi@user/amospalla) | 06:58 | |
| - Ar|stote|is (QUIT: Ping timeout: 240 seconds) (~linx@149.210.3.83) | 08:09 | |
| + Ar|stote|is (~linx@149.210.3.83) | 08:15 | |
| - kxtells (QUIT: Ping timeout: 252 seconds) (~kxtells@2a0c:5a82:a505:6300:62ac:2198:5988:93c) | 08:40 | |
| - Ar|stote|is (QUIT: Ping timeout: 244 seconds) (~linx@149.210.3.83) | 08:51 | |
| + Ar|stote|is (~linx@149.210.3.83) | 09:00 | |
| + kxtells (~kxtells@84.123.7.93.dyn.user.ono.com) | 09:23 | |
| - nybble (QUIT: Ping timeout: 260 seconds) (~sprang@user/csprng) | 09:57 | |
| - mlarkin (QUIT: Ping timeout: 260 seconds) (~mlarkin@syn-076-081-194-027.biz.spectrum.com) | 10:10 | |
| - kxtells (QUIT: Changing host) (~kxtells@84.123.7.93.dyn.user.ono.com) | 10:14 | |
| + kxtells (~kxtells@user/kxtells) | 10:14 | |
| + paperManu (~paperManu@107.159.15.124) | 12:39 | |
| josch | my rk3588 classic reform with latest system image just stopped here: https://mister-muffin.de/p/NqdT.jpg | 12:59 |
|---|---|---|
| josch | display is on but no reaction to keyboard or mouse or on serial | 12:59 |
| josch | anybody had this as well? | 12:59 |
| + gustav25 (~gustav@c-78-82-52-152.bbcust.telenor.se) | 13:02 | |
| + mjw (~mjw@gnu.wildebeest.org) | 13:19 | |
| minute | josch: hmmm random hang | 13:21 |
| - iank (QUIT: Quit: ZNC 1.8.2+deb2+deb11u1 - https://znc.in) (~iank@fsf/staff/iank) | 13:25 | |
| josch | but i don't want to have reservations to switch my main machine to the rk3588 XD /o\ | 13:25 |
| + iank (~iank@fsf/staff/iank) | 13:28 | |
| gordon1 | josch: did you have serial connected during that? it usually dumps panic stacktrace there | 13:33 |
| + jahkosha (~jahkosha@user/jahkosha) | 13:33 | |
| * jah -> Guest7614 | 13:33 | |
| - Guest7614 (QUIT: Killed (lithium.libera.chat (Nickname regained by services))) (~jahkosha@user/jahkosha) | 13:33 | |
| * jahkosha -> jah | 13:33 | |
| josch | nope, i only connected it afterwards -- i wish i had | 13:39 |
| chorc | got my v2 charging board, thanks MNT! quick question: so far no known issues with pop and smoke, right? | 13:49 |
| - paperManu (QUIT: Ping timeout: 245 seconds) (~paperManu@107.159.15.124) | 14:10 | |
| gsora | josch: i'm doing some testing on hdmi & linux 6.18, so far i managed to have 4k60hz on my hdmi 2.1 monitor by using linux-rockchip and this patchset: https://lore.kernel.org/linux-rockchip/20251203-dw-hdmi-qp-scramb-v1-0-836fe7401a69@collabora.com/ | 14:24 |
| gsora | i'd love to help on making reform's 6.18 kernel work with my monitor, lmk if i can test it in any way! | 14:25 |
| gsora | this is all incredibly exciting, using reform as my daily driver for desktop usage would be so cool | 14:27 |
| josch | gsora: my work on 6.18 is here: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/155 | 14:29 |
| josch | unfortunately, 6.18 changed the location of the dtbs, so a new reform-tools release is needed before | 14:29 |
| josch | that merge request patches reform-tools (and flash-kernel, which is also affected) because of this | 14:30 |
| josch | without those patches, your machine will become unbootable with 6.18 from Debian | 14:30 |
| josch | gsora: i can throw your patch into that MR if you lik | 14:30 |
| gsora | oh? i've been manually building 6.19/6.18 kernels and it's booting up fine | 14:30 |
| gsora | i'm also manually editing extlinux.conf, maybe it's that | 14:31 |
| gsora | i'll pull that mr, compile a clean kernel and see if it still works | 14:31 |
| gsora | if it does, you can merge them | 14:31 |
| josch | gsora: but you have been building the kernels without reform-tools integration, right? | 14:35 |
| josch | you are building a non-Debian kernel and you handle copying the dtbs yourself | 14:35 |
| gsora | yes, straight up scp-ing Image around | 14:35 |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 14:42 | |
| + mlarkin (~mlarkin@syn-076-081-194-027.biz.spectrum.com) | 14:42 | |
| gsora | thought there was some other thing related to 6.18 i should've looked out for | 14:45 |
| gsora | hm, some of the patches I linked above are clashing with some of the patches in the mr | 14:50 |
| gsora | 3/4 of the WIP drm patches are older versions, can be discarded, the last one doesnt apply cleanly sadly | 15:11 |
| gsora | good news, that last patch wasn't needed | 15:37 |
| gsora | or at least, it works without it | 15:39 |
| + potash1 (~potash@user/foghorn) | 15:43 | |
| gsora | confirmed that https://lore.kernel.org/linux-rockchip/20251203-dw-hdmi-qp-scramb-v1-0-836fe7401a69@collabora.com/ is required for my screen to work - those are the patches that I had to delete from your branch josch in order for it to apply: https://tangled.org/strings/did:plc:6ll5xi67lyuyovt6fiv4fnjo/3mabgpounem22 | 15:57 |
| gsora | last patch in that patchset still doesn't apply cleanly, i omitted it but works otherwise | 15:57 |
| josch | gsora: i have no clue what i'm doing so i'm not (again) removing patches from the collabora patch stack | 16:05 |
| josch | gsora: last time, minute had to clean up after me | 16:05 |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 16:09 | |
| * Guest5752 -> mjw | 16:12 | |
| gsora | oof, sure, I understand | 16:12 |
| gsora | to be honest, not really having a clue myself, i'm just a guy with 24 fast threads and a good internet connection | 16:12 |
| josch | rk3588 is the fastest cpu in my house right now :) | 16:13 |
| josch | no wait, it's probably the cpu in my partner's smartphone lol | 16:13 |
| - aloo_shu (QUIT: Ping timeout: 250 seconds) (~aloo_shu@85.51.18.34) | 16:21 | |
| + aloo_shu (~aloo_shu@85.51.17.47) | 16:27 | |
| minute | gsora: can you share your finished patchstack (the one that works) later? then i can review it for the 6.18 mr | 17:02 |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 17:28 | |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 17:29 | |
| gsora | i'll try! so far i'm manually deleting patches that overlap and using b4 to apply ones from lore | 17:33 |
| josch | gsora: if you want a patch stack with patches which are already cleanly applying on top of each other, take this one: https://gitlab.collabora.com/hardware-enablement/rockchip-3588/linux/-/tree/rockchip-devel?ref_type=heads | 17:41 |
| josch | gsora: the patches you showed above are in there | 17:42 |
| josch | and if something from MR155 is in the way, figure out the corresponding patch from the current patch stack in rockchip-devel and update the patch | 17:42 |
| gsora | yup, i did that | 17:46 |
| gsora | some of them were just out of date | 17:46 |
| josch | sigrid: i've just built ffmpeg and mpv with the patch stack you linked for v4l2request. Do you have anything special in your ~/.config/mpv/mpv.conf? I'm getting [vo/gpu/libplacebo] vk->CreateSwapchainKHR(...): VK_ERROR_OUT_OF_HOST_MEMORY (../src/vulkan/swapchain.c:622) | 17:46 |
| - pomel0 (QUIT: Ping timeout: 264 seconds) (~pomel0@user/pomel0) | 17:47 | |
| + pomel0 (~pomel0@user/pomel0) | 17:47 | |
| sigrid | josch: just hwdec=v4l2request really | 17:48 |
| josch | minute: about my linux 6.18 MR: i am successfully running the packages created by the MNT gitlab CI for this MR on rk3588 but as usual: i cannot test hdmi here, somebody else has to do that :) | 17:48 |
| sigrid | do you have a video in question available? | 17:48 |
| josch | sigrid: i'm just playing the 1080p@60 version of big buck bunny: https://download.blender.org/demo/movies/BBB/ | 17:51 |
| sigrid | works with no issues for me | 17:54 |
| sigrid | and you have the kernel patch stack for hardware decoding also applied? | 17:54 |
| josch | "the kernel patch stack"? :D | 17:58 |
| josch | is it in the collabora stack? | 17:59 |
| - kxtells (QUIT: Ping timeout: 264 seconds) (~kxtells@user/kxtells) | 17:59 | |
| sigrid | not really. https://gitlab.collabora.com/detlev/linux/-/commits/add-vdpu381-and-383-to-rkvdec-v6 | 18:00 |
| sigrid | I'm building 6.18.2 with this one (had an older version before) to try out | 18:01 |
| sigrid | this is what I did so far, maybe you'll find it useful: https://git.sr.ht/~ft/aports/commit/0bd75367c0cfe0ba38bab93f0713bbcb4775bdf8 | 18:04 |
| sigrid | a couple patches needed small changes to apply | 18:04 |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 18:05 | |
| josch | sigrid: thank you! i'll try this out with 6.18 then :) | 18:08 |
| + pomel0 (~pomel0@user/pomel0) | 18:08 | |
| josch | i'll just use your patch stack -- the original by casanova is against 6.19-rc1 :D | 18:18 |
| sigrid | yeah, that's sensible as long as it works with 6.18 which I still have to test :) | 18:19 |
| - lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah) | 18:24 | |
| + lidstah (~lidstah@gateway/tor-sasl/lidstah) | 18:24 | |
| + nybble (~sprang@user/csprng) | 18:39 | |
| + spew (~spew@user/spew) | 18:40 | |
| + wielaard (~mjw@gnu.wildebeest.org) | 18:49 | |
| * mjw -> Guest6376 | 18:49 | |
| - Guest6376 (QUIT: Killed (copper.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 18:49 | |
| * wielaard -> mjw | 18:49 | |
| + Guest6376 (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 18:49 | |
| kfx | sigrid: wrapping up a proper guide to installing alpine. last step is figuring out how to modify the devicetree to tell u-boot to load /boot from nvme instead of emmc | 19:06 |
| - cli (QUIT: Ping timeout: 244 seconds) (~m-vsauiy@user/cli) | 19:06 | |
| + cli (~m-vsauiy@user/cli) | 19:09 | |
| josch | kfx: isn't it the u-boot distro-boot command which dictates the order in which devices are tried? | 19:11 |
| sigrid | kfx: awesome! | 19:12 |
| sigrid | josch: 6.18.2 works great. video plays fine | 19:12 |
| kfx | josch: I'm trying to figure a way to tell u-boot to load /boot from the nvme without disturbing the debian installation onf the emmc | 19:19 |
| kfx | I'm also a little confused because bootdev list | 19:20 |
| kfx | shows mmc, then I run bootdev hunt and bootdev list shows the nvme as well | 19:20 |
| kfx | so there might need to be an extra step to get it to probe the nvme | 19:20 |
| sigrid | I created a dummy /boot on nvme for that future but ofc had to use the emmc one so far | 19:42 |
| spew | sigrid, on my rk3588 I could not get your alpine image to boot off the card | 19:51 |
| spew | I assumed it was some difference in the motherboard but I haven't tried to look into it further | 19:51 |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 19:56 | |
| + pomel0 (~pomel0@user/pomel0) | 19:56 | |
| sigrid | spew: that's weird. how do you know it did not boot? | 19:56 |
| sigrid | were you running with or without the hdmi adapter? | 19:56 |
| spew | I'm not sure whether the newest motheboard has the hdmi adapter | 20:01 |
| sigrid | it's a separate device that you insert yourself | 20:01 |
| spew | then no | 20:01 |
| sigrid | interesting. well, there still needs to be a proper alpine image not this hack I built :( | 20:02 |
| spew | when I try booting the device with a card inserted that has the normal system image, it boots that image | 20:02 |
| spew | but when I tried the same with yours, it just boots off the disk | 20:02 |
| spew | all I noticed was that the disk label types were different between your image and the mnt ones, but then I haven't looked into it more | 20:03 |
| sigrid | do you have a serial? | 20:05 |
| spew | no | 20:05 |
| sigrid | :C | 20:05 |
| spew | it's okay I'll figure it out eventually | 20:06 |
| kfx | sigrid: the guide I've made lets you start with the stock debian installation | 20:30 |
| kfx | since alpine-chroot-install is in debian's repository | 20:30 |
| kfx | there's some dancing you have to do to set up repos and select the right kernel | 20:30 |
| kfx | but you don't need anything but what mnt ships to get started | 20:30 |
| kfx | but I'll also work up an image builder | 20:31 |
| spew | kfx: where's the guide please? | 20:33 |
| kfx | it's not up yet, because I'm still fucking with the last piece (/boot) | 20:36 |
| kfx | I'll post a skeleton quickly | 20:36 |
| kfx | here you go spew http://okturing.com/src/16712/body | 20:51 |
| josch | kfx: can setup-alpine be run unattended? If yes, this could be run in a pipeline on MNT gitlab semi-regularly, creating alpine images by MNT | 20:59 |
| josch | kfx: i'm regularly creating 13+ GB artifacts as a byproduct of my experiments. I heard alpine is small, so it shouldn't be a burden to the gitlab. :) | 21:01 |
| - se6astian (QUIT: ) (~quassel@2a03:b0c0:3:f0:0:1:9e0a:7000) | 21:03 | |
| + se6astian (~quassel@2a03:b0c0:3:f0:0:1:9e0a:7000) | 21:04 | |
| kfx | yes, setup-alpine can take an answerfile or env vars, has a quiet mode and a noninteractive mode | 21:05 |
| josch | kfx: turn your instructions into a shellscript :) | 21:05 |
| kfx | I already have :) | 21:05 |
| kfx | I have scripts for building this into an sd card image | 21:08 |
| kfx | but I don't want to push anything until I figure out the bootflow problem | 21:08 |
| josch | make sure not to let the perfect be the enemy of the good :) | 21:12 |
| kfx | if that were the case I'd be porting openbsd! | 21:13 |
| josch | :D | 21:13 |
| kfx | but no, there's a solution here if I can find it | 21:13 |
| kfx | is there a reason nvme isn't in the default boot device priorities at all? | 21:13 |
| josch | kfx: i remember that there were experiments with a311d to let it boot from nvme in addition to emmc and sd-card: https://source.mnt.re/reform/reform-a311d-uboot/-/merge_requests/3 | 21:19 |
| kfx | I think I'm gonna try to patch u-boot to try sd card first, then usb, then nvme, then emmc | 21:24 |
| elb | ooh pocket keyboard v2 | 22:06 |
| elb | I need to get back to unifying firmware :-\ | 22:06 |
| elb | the semester is _almost over_ for me, maybe I can dedicate some real time to that over break | 22:06 |
| - gustav25 (QUIT: Quit: Quit) (~gustav@c-78-82-52-152.bbcust.telenor.se) | 22:30 | |
| - cli (QUIT: Ping timeout: 265 seconds) (~m-vsauiy@user/cli) | 22:32 | |
| + cli (~m-vsauiy@user/cli) | 22:37 | |
| josch | sigrid: i think hardware decoding works now -- i can watch 3840x2160@60 big buck bunny with 10% cpu usage :) | 23:01 |
| sigrid | nice! | 23:02 |
| josch | unfortunately i still am getting VK_ERROR_OUT_OF_HOST_MEMORY after a while | 23:02 |
| sigrid | interesting. maybe there is a leak somewhere (mpv?) | 23:02 |
| josch | indeed but your patches are definitely an improvement for the MR, thank you! :) | 23:03 |
| sigrid | np | 23:03 |
| - paperManu (QUIT: Ping timeout: 240 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 23:22 | |
| josch | waaaaait | 23:31 |
| josch | sigrid: you are *probably* not using gnome, right? | 23:32 |
| + paperManu (~paperManu@107.159.15.124) | 23:37 | |
| sigrid | nope | 23:42 |
| josch | it works fine in sway :) | 23:43 |
| sigrid | huh | 23:47 |
| sigrid | linux graphics stack spaghetti monster | 23:48 |
| - kensanata (QUIT: Quit: OK) (~alex@user/kensanata) | 23:58 | |
| + kensanata (~alex@user/kensanata) | 23:59 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
