| + cinap_lenrek (~cinap_len@188.26.222.244) | 00:02 | |
| cinap_lenrek | \o/ | 00:02 |
|---|---|---|
| josch | hello :) | 00:03 |
| cinap_lenrek | the weinachtsmann came buy and sent me a ls1028 board for the reform ;) | 00:03 |
| josch | another one!! :D | 00:03 |
| josch | did you try it out already? | 00:03 |
| cinap_lenrek | well played! ;) | 00:03 |
| cinap_lenrek | josh: no time yet, but i will | 00:04 |
| cinap_lenrek | i got the 4k pages reference manual for the SoC anyway | 00:04 |
| cinap_lenrek | https://felloff.net/doc/comp/spec/LS1028ARM.pdf | 00:05 |
| cinap_lenrek | the fucking displayport TX controller for sure will be a challange | 00:06 |
| cinap_lenrek | poco a poco | 00:06 |
| josch | wow literally 4k+ pages o0 | 00:06 |
| cinap_lenrek | yep | 00:06 |
| minute | cinap_lenrek: great to hear! | 00:08 |
| cinap_lenrek | i'm a bit drunk today | 00:09 |
| cinap_lenrek | but just wanted to say, you all rock | 00:09 |
| cinap_lenrek | thank you so much! | 00:10 |
| cinap_lenrek | keep up the great work | 00:10 |
| cinap_lenrek | all of you! | 00:10 |
| josch | same to you :) | 00:10 |
| minute | cinap_lenrek: cheers! | 00:20 |
| cinap_lenrek | minute: you too :) | 00:20 |
| cinap_lenrek | how's it going? | 00:20 |
| cinap_lenrek | i wasted a bit of company time today to read the manual... | 00:22 |
| cinap_lenrek | i was quite surprised that the ethernet controller appears to be a pcie root complex | 00:22 |
| cinap_lenrek | clearly, without the actual signaling, but taking advantage of the configuration infrastucture | 00:23 |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 00:23 | |
| minute | cinap_lenrek: i guess those are the parts from qoriq heritage | 00:25 |
| cinap_lenrek | of course | 00:25 |
| cinap_lenrek | dude, qoriq is my day job currently ;) | 00:25 |
| cinap_lenrek | so i had alot of deja-vu moments reading the manual | 00:25 |
| cinap_lenrek | unfortunately, i dont think this will be the end | 00:26 |
| minute | cinap_lenrek: oh ok! | 00:26 |
| cinap_lenrek | all the socs are still kind of a chaos | 00:26 |
| cinap_lenrek | dont worry, i'll definitely get everything working | 00:27 |
| minute | nice | 00:27 |
| cinap_lenrek | but i glady miss the pc spirit in all of it | 00:27 |
| cinap_lenrek | pci is an amazing technology if you think about it | 00:27 |
| minute | just one word of advice: don't run it without the heatsink | 00:28 |
| cinap_lenrek | in the abstract sense | 00:28 |
| cinap_lenrek | lie forget about the actual physical bus | 00:28 |
| minute | (i think i managed to burn one of 'em) | 00:28 |
| minute | cinap_lenrek: i see | 00:28 |
| bkeys | minute: Did you look at this link? | 00:28 |
| cinap_lenrek | but it has exactly wahat is required for being independent of the underlying architecture | 00:29 |
| bkeys | https://arace.tech/products/milk-v-mars-cm?variant=42351990046900 | 00:29 |
| cinap_lenrek | i/o has converged to memory mapped devices | 00:29 |
| cinap_lenrek | its all transactions on a unified single address space | 00:29 |
| cinap_lenrek | and pci gives you the flexibility to put the devices anywhere in the address space you want | 00:30 |
| cinap_lenrek | so its pretty genetic | 00:30 |
| cinap_lenrek | MSI interrupts solve the freaking chaos which is out of band interrupt signaling | 00:31 |
| cinap_lenrek | i have no idea why ARM isnt pushing for a standard to organize a bit more the axi bus | 00:31 |
| minute | bkeys: yep. is there a reference manual for that chip? | 00:31 |
| cinap_lenrek | its really not much that is needed to reach PCI level funtionality | 00:32 |
| cinap_lenrek | like forget about the signaling and serdes stuff | 00:32 |
| minute | cinap_lenrek: ah, you mean something like auto discovering devices? | 00:32 |
| bkeys | minute: I am not certain. It says it uses the 2nd HDMI for display | 00:32 |
| cinap_lenrek | the actual benefit is really about a standartized OS interface for managing address space and device identification | 00:32 |
| minute | cinap_lenrek: i only have experience with axi from zynq, and there i had to assign everything to fixed places in a memory map | 00:33 |
| bkeys | I am willing to take the risk of preordering it and doing testing, but I'm not certain if it could work even in theory, but if it did it would be a big deal, you could have a SoM with 8GB of RAM, riscv for less than $100 | 00:33 |
| cinap_lenrek | minute: yeah, been there | 00:33 |
| minute | cinap_lenrek: yeah, it would be a lot better than this devicetree stuff | 00:33 |
| cinap_lenrek | minute: but in the end its just bullshit work | 00:33 |
| minute | yep | 00:33 |
| cinap_lenrek | minute: someone has todo the extra work to maintain the devicetree files for a specific setup to make it all happen | 00:34 |
| bkeys | minute: I g2g soon, but they have emails at the bottom of the link, their Development Department could likely get you a reference manual | 00:34 |
| bkeys | I can send an email | 00:34 |
| cinap_lenrek | bkeys: hahaha | 00:34 |
| minute | bkeys: i was hoping someone other than me would get into this thing | 00:34 |
| cinap_lenrek | bkeys: the milk is indeed a interesting machine | 00:34 |
| minute | i kind of had enough bringups for a while | 00:35 |
| bkeys | minute: I can be that person | 00:35 |
| minute | and i have yet to tackle rk3588 | 00:35 |
| minute | bkeys: yes please | 00:35 |
| bkeys | Is the fact that it uses the 2nd HDMI port a non starter? | 00:35 |
| minute | bkeys: it's good | 00:36 |
| minute | bkeys: see https://mastodon.social/@mntmn/111184030822276309 | 00:36 |
| bkeys | Okay so using the 2nd HDMI port might be the standard | 00:36 |
| minute | it's luck maybe | 00:37 |
| bkeys | In that case it should theoretically be able to work; I'm willing to take the risk and I'll place the order | 00:37 |
| minute | i routed the second hdmi to fpc cable for internal display | 00:37 |
| bkeys | Hopefully it will come around the same time my keyboard does | 00:37 |
| cinap_lenrek | https://felloff.net/doc/comp/spec/HDMISpecification14a.pdf | 00:38 |
| cinap_lenrek | https://felloff.net/doc/comp/spec/HDMISpecification2.0.pdf | 00:38 |
| bkeys | minute: Do you see any reason to get the non-wifi version over the wifi version? | 00:38 |
| cinap_lenrek | good luck! | 00:38 |
| cinap_lenrek | pirate while you can | 00:38 |
| bkeys | I'm not sure if wifi on other SoMs has caused any seriuos issues; I know sometimes they block pcie lines | 00:38 |
| bkeys | minute: So I'm reading a document on it now, it might use the 1st HDMI port? Would that ruin it? | 00:41 |
| minute | bkeys: does it have only 1 display? no mipi dsi? | 00:42 |
| + ethulhu (ethulhu@nora.ethulhu.co.uk) | 00:46 | |
| - chomwitt (QUIT: Ping timeout: 240 seconds) (~chomwitt@athedsl-368454.home.otenet.gr) | 00:56 | |
| - rah (QUIT: Ping timeout: 272 seconds) (rah@verain.settrans.net) | 00:56 | |
| + rah (rah@verain.settrans.net) | 00:57 | |
| bkeys | minute: it has mipi dsi and hdmi | 00:58 |
| bkeys | I g2g now, if the SoM has a chance at working according to you I will place the order and when you send me back my keyboard and the like I'll start looking at it right away | 01:00 |
| bkeys | Cause a riscv reform would be big news | 01:00 |
| cinap_lenrek | danger robinson :) | 01:04 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 01:23 | |
| - ethulhu (QUIT: Excess Flood) (ethulhu@nora.ethulhu.co.uk) | 01:59 | |
| - mtm (QUIT: Ping timeout: 272 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:04 | |
| - jacobk (QUIT: Ping timeout: 272 seconds) (~quassel@utdpat241106.utdallas.edu) | 02:21 | |
| + ethulhu (ethulhu@nora.ethulhu.co.uk) | 02:48 | |
| - nsc (QUIT: Ping timeout: 248 seconds) (~nicolas@132-48-142-46.pool.kielnet.net) | 03:06 | |
| + nsc (~nicolas@84-97-142-46.pool.kielnet.net) | 03:08 | |
| - mjw (QUIT: Ping timeout: 255 seconds) (~mjw@gnu.wildebeest.org) | 03:15 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:09 | |
| + eery (~eery@139.55.3.210) | 04:40 | |
| - eery (QUIT: Ping timeout: 240 seconds) (~eery@139.55.3.210) | 05:06 | |
| + eery (~eery@h210.3.55.139.dynamic.ip.windstream.net) | 05:11 | |
| - violet (QUIT: Remote host closed the connection) (~vi@user/violet) | 05:14 | |
| - eery (QUIT: Ping timeout: 272 seconds) (~eery@h210.3.55.139.dynamic.ip.windstream.net) | 05:16 | |
| + violet (~vi@user/violet) | 05:25 | |
| + jacobk (~quassel@129.110.242.173) | 06:04 | |
| - jacobk (QUIT: Ping timeout: 272 seconds) (~quassel@129.110.242.173) | 06:26 | |
| + jacobk (~quassel@64.189.201.150) | 06:42 | |
| - jacobk (QUIT: Ping timeout: 255 seconds) (~quassel@64.189.201.150) | 08:29 | |
| erle | cinap_lenrek i vaguely remember your name. are you a plan 9 or 9front user? | 09:41 |
| + chomwitt (~chomwitt@2a02:587:7a24:b000:1ac0:4dff:fedb:a3f1) | 10:07 | |
| - kensanata (QUIT: Quit: OK) (~alex@user/kensanata) | 10:51 | |
| + kensanata (~alex@user/kensanata) | 10:51 | |
| + mjw (~mjw@gnu.wildebeest.org) | 12:01 | |
| vkoskiv | I was looking at boot logs and saw: | 12:03 |
| vkoskiv | Oct 06 12:42:01 reform kernel: KASLR disabled due to lack of seed | 12:03 |
| vkoskiv | Probably not ideal? | 12:04 |
| vkoskiv | I wanted to try suspend on A311D just to see what happens. It certainly did go to 100mA for a bit, but obviously didn't go well from there. | 12:04 |
| vkoskiv | I'd really love to make myself useful and investigate suspend on A311D | 12:05 |
| minute | vkoskiv: i think this kaslr thing is around forever? | 12:17 |
| vkoskiv | Yeah, I do recall seeing it on imx8 too I think | 12:18 |
| minute | vkoskiv: cool! serial debugging it is, and maybe ask for some pointers in #linux-amlogic, people are helpful | 12:18 |
| vkoskiv | Found a typo in /usr/sbin/reform-standby, the error saying it doesn't support this hardware says 'reform-flash-rescue' instead of 'reform-standby' | 12:19 |
| cinap_lenrek | erle: yes | 12:19 |
| vkoskiv | I was confused why in the log it was trying to flash the rescue partition when suspending, but it's just probably a copypaste goof | 12:19 |
| vkoskiv | I'll start with setting up serial, I think I have the needed hw for it | 12:19 |
| vkoskiv | Then I'd have to modify that standby script to detect A311D and then just try and see what works | 12:20 |
| vkoskiv | Probably best to do all this on a fresh SD card instead of my main nvme though :D | 12:21 |
| - mjw (QUIT: Ping timeout: 255 seconds) (~mjw@gnu.wildebeest.org) | 12:25 | |
| minute | vkoskiv: yep! try with as little peripherals as possible. also, hook up your uart adapter to SER2, not 1z | 12:30 |
| minute | not 1 | 12:30 |
| vkoskiv | Fairly certain my flipper zero toy can work as a UART adapter | 12:35 |
| vkoskiv | Finally some actual use for the thing beyond tomfoolery :D | 12:36 |
| abortretryfail | erle: no i don't have a specific example, but if there is a way to do it in Sway, that's where it'll be. | 12:43 |
| abortretryfail | actually, re-reading that after sleeping on it i wonder if that's even true. | 12:43 |
| - pandora (QUIT: Quit: Connection closed for inactivity) (uid585533@id-585533.ilkley.irccloud.com) | 13:04 | |
| abortretryfail | sorry, man 5 sway. See the "CRITERIA" section at the bottom. you can do things like swaymsg [con_id=foo] move position 300 24 | 13:09 |
| abortretryfail | I knew there was a way to do it i just forgot because there's never any reason to do it. | 13:09 |
| - Boostisbetter (QUIT: Ping timeout: 272 seconds) (4a410829d7@irc.cheogram.com) | 13:12 | |
| erle | yeah okay so it's not possible to spawn a window | 13:19 |
| erle | just to move it | 13:19 |
| erle | ig this provides some flexibility, but ultimately is not what i want (i want hints, i will just have to wait) | 13:20 |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 13:33 | |
| minute | erle: sway is the wrong wm for positioning windows in specific places. maybe check out wayfire | 14:02 |
| - mtm (QUIT: Ping timeout: 260 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:02 | |
| minute | erle: also with wayfire you can just write a little decoration plugin that puts cats everywhere you want | 14:05 |
| erle | lol | 14:06 |
| erle | never heard of wayfire, is this the thing that is like compiz fusion? | 14:06 |
| erle | like i have seen some things where people made wayland windows very weird | 14:07 |
| erle | like skewing them and so on | 14:07 |
| erle | “this is the backside of my terminal” | 14:07 |
| erle | it reminded me of compiz, but obviously it could not be compiz | 14:07 |
| - Boostisbetter (QUIT: Ping timeout: 252 seconds) (4a410829d7@irc.cheogram.com) | 14:31 | |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 14:34 | |
| minute | erle: it is in the spirit of compiz. we ship a customized version in system image v4. | 14:39 |
| Boostisbetter | minute: how does one try out wayfire with the v4 system image? | 14:45 |
| minute | Boostisbetter: one types "wayfire" | 14:58 |
| minute | Boostisbetter: if you have upgraded an old system you'll need to do some tweaks though... do you have a fresh v4 sd card / system? | 14:59 |
| minute | Boostisbetter: IIRC the most important pieces are: `apt install wayfire firedecor papirus-icon-theme` and `cp /etc/skel/.config/wayfire.ini ~/`. if you also want the new waybar setup, `cp /etc/skel/waybar/* ~/.config/waybar/`. i wrote these now from memory so might have missed something | 15:01 |
| Boostisbetter | ok, need to look into it. Not sure what the advantages over Sway are. | 15:02 |
| erle | positioning windows in specific places | 15:02 |
| minute | Boostisbetter: the most important thing is it's more of a traditional, floating windows (with border snapping) thing | 15:02 |
| minute | it will feel more familiar if you come from windows, mac, kde, gnome 2, fvwm, openbox, ... | 15:03 |
| minute | also recommended: `apt install dunst` and `cp -Rv /etc/skel/.config/dunst ~/` | 15:04 |
| minute | (for nice looking notifications) | 15:04 |
| + mjw (~mjw@gnu.wildebeest.org) | 15:06 | |
| Boostisbetter | But you can still snap things to a similar grid layout like in Sway? Is it light weight like Sway? | 15:25 |
| - stites (QUIT: Ping timeout: 260 seconds) (~stites@130.44.147.204) | 15:29 | |
| minute | Boostisbetter: it is based on the same compositor library wlroots. maybe slightly heavier. | 15:29 |
| + stites (~stites@2607:fb91:dc3:545:4935:e9c4:fad4:f984) | 15:29 | |
| Boostisbetter | minute: thank you. I will look into, but I'm pretty content with sway. | 15:31 |
| Boostisbetter | I like it a lot on the Reform. | 15:31 |
| + aliosablack (~chomwitt@2a02:587:7a24:b000:1ac0:4dff:fedb:a3f1) | 15:32 | |
| - Gooberpatrol66 (QUIT: Read error: Connection reset by peer) (~Gooberpat@user/gooberpatrol66) | 15:33 | |
| + Gooberpatrol_66 (~Gooberpat@2603-8080-4540-7cfb-0000-0000-0000-113a.res6.spectrum.com) | 15:33 | |
| - Gooberpatrol_66 (QUIT: Changing host) (~Gooberpat@2603-8080-4540-7cfb-0000-0000-0000-113a.res6.spectrum.com) | 15:33 | |
| + Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66) | 15:33 | |
| - chomwitt (QUIT: Ping timeout: 272 seconds) (~chomwitt@2a02:587:7a24:b000:1ac0:4dff:fedb:a3f1) | 15:34 | |
| sevan | following on from WD firmware the other day, found this post on updating WD firmware on Linux https://community.frame.work/t/western-digital-drive-update-guide-without-windows-wd-dashboard/20616/5 | 15:36 |
| sevan | haven't tried it yet. | 15:36 |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@2607:fb91:dc3:545:4935:e9c4:fad4:f984) | 15:46 | |
| + stites (~stites@155.33.134.29) | 15:46 | |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@155.33.134.29) | 15:47 | |
| + stites (~stites@155.33.134.29) | 15:47 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 15:59 | |
| + jacobk (~quassel@64.189.201.150) | 16:06 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:08 | |
| + IchikaZou (~IchikaZou@36-231-76-229.dynamic-ip.hinet.net) | 16:14 | |
| Boostisbetter | sevan: I have. It works. | 16:19 |
| sevan | Boostisbetter: that's great to hear | 16:20 |
| Boostisbetter | sevan: yeah I used to run Ubuntu LTS on my Framework 13, but I got to a point where it was just easier for work and all to use Windows with WSL. Most of the applications I'm running on Windows (that has been locked down) are through WSL, but for me it is the best of both worlds. | 16:23 |
| sevan | I actually want to try this out on my reform as I have a WD Black installed in it. | 16:24 |
| abortretryfail | Heh, I still have not tried Wayfire yet. | 16:25 |
| * mark_ -> mjw | 16:28 | |
| IchikaZou | What is maximum storage size support by pocket reform m.2 slot? | 16:29 |
| sknebel | probably "the maximum you can buy" | 16:31 |
| abortretryfail | anything that fits in the slot? power consumption might be a limiting factor | 16:31 |
| IchikaZou | I got a 4TB ssd and my laptop can’t detect it, so might ask weather pocket reform would support it before return it. | 16:33 |
| IchikaZou | Some laptop seem to not support up to 4TB ssd, only 2TB. | 16:35 |
| abortretryfail | weird. does it say anything in dmesg or lspci about it? | 16:35 |
| - aliosablack (QUIT: Ping timeout: 258 seconds) (~chomwitt@2a02:587:7a24:b000:1ac0:4dff:fedb:a3f1) | 16:36 | |
| IchikaZou | Nope. It can even detect by bios, but that ssd can be see on my desktop from bios and lsblk. | 16:37 |
| IchikaZou | It can’t even detect by bios | 16:37 |
| abortretryfail | oh i see, you have something that is not a reform and it doesn't detect it | 16:39 |
| IchikaZou | I haven’t got my reform yet, the ssd only can’t be read by my old dell | 16:40 |
| abortretryfail | If you boot the old dell into Linux, can you see it in lspci? it could just be the old dell's bios has no nvme support? | 16:41 |
| abortretryfail | or some kind of weird vendor-locking and that slot is only intended for a wwan card or something (I have a HP that is like this) | 16:41 |
| IchikaZou | I asked on dell forums and got the answers the some old dell models that only support up to 2TB | 16:42 |
| abortretryfail | right, so the BIOS doesn't know what to do with it, and you couldn't use it as a boot device. | 16:43 |
| abortretryfail | The Reform doesn't exactly have a "BIOS" | 16:43 |
| IchikaZou | I see, pocket reform is my first arm machine so there are some stuff I don’t understand | 16:44 |
| + pandora (uid585533@id-585533.ilkley.irccloud.com) | 16:46 | |
| IchikaZou | Guess I don’t need to return that ssd. | 16:48 |
| abortretryfail | try booting the dell off USB into linux and see if the ssd shows up in lspci and lsblk | 16:49 |
| + aliosablack (~chomwitt@2a02:587:7a24:b000:1ac0:4dff:fedb:a3f1) | 17:15 | |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@155.33.134.29) | 17:43 | |
| + stites (~stites@155.33.134.29) | 17:44 | |
| - aliosablack (QUIT: Ping timeout: 240 seconds) (~chomwitt@2a02:587:7a24:b000:1ac0:4dff:fedb:a3f1) | 17:49 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 17:49 | |
| + murphnj (~murph@user/murphnj) | 17:55 | |
| minute | i think 4TB SSD should just work but still need to try one myself. | 18:00 |
| abortretryfail | It's all down to whether the Linux kernel supports it, right? | 18:03 |
| + eery (~eery@h210.3.55.139.dynamic.ip.windstream.net) | 18:18 | |
| - jacobk (QUIT: Ping timeout: 272 seconds) (~quassel@64.189.201.150) | 18:44 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 18:55 | |
| - IchikaZou (QUIT: Remote host closed the connection) (~IchikaZou@36-231-76-229.dynamic-ip.hinet.net) | 19:02 | |
| - violet (QUIT: Ping timeout: 272 seconds) (~vi@user/violet) | 19:13 | |
| + violet (~vi@user/violet) | 19:15 | |
| - jacobk (QUIT: Ping timeout: 272 seconds) (~quassel@utdpat241106.utdallas.edu) | 19:30 | |
| + jacobk (~quassel@129.110.242.224) | 19:33 | |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@155.33.134.29) | 19:38 | |
| + stites (~stites@155.33.134.29) | 19:39 | |
| eery | Does anyone know if the current A311D firmware/kernel support switching to HDMI output? | 19:41 |
| - stites (QUIT: Ping timeout: 255 seconds) (~stites@155.33.134.29) | 19:43 | |
| + stites (~stites@2607:fb91:dcf:cc9b:5aee:fc9:75af:2def) | 19:43 | |
| - murphnj (QUIT: Quit: Leaving) (~murph@user/murphnj) | 19:47 | |
| - mlarkin (QUIT: Ping timeout: 240 seconds) (~mlarkin@047-036-115-056.res.spectrum.com) | 19:51 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 19:54 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:57 | |
| * mark_ -> mjw | 19:57 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:57 | |
| abortretryfail | uhh darn... apt upgrade on sid made my reform not boot. currently in chroot trying to fix it. | 20:00 |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@2607:fb91:dcf:cc9b:5aee:fc9:75af:2def) | 20:00 | |
| + stites (~stites@130.44.147.204) | 20:00 | |
| + mlarkin (~mlarkin@047-036-115-056.res.spectrum.com) | 20:03 | |
| abortretryfail | libpython3.11-testsuite fails to config, but i dont see what this has to do with booting... | 20:04 |
| - stites (QUIT: Ping timeout: 272 seconds) (~stites@130.44.147.204) | 20:15 | |
| + stites (~stites@130.44.147.204) | 20:18 | |
| Boostisbetter | abortretryfail: are you using the normal imx8 SoC? | 20:21 |
| Boostisbetter | I'm on sid trixie and I have been updating a lot and have not had any issues. | 20:22 |
| abortretryfail | yeah, normal imx8 encrypted nvme root | 20:22 |
| abortretryfail | booted just fine off the latest system image v4 sdcard | 20:23 |
| abortretryfail | brb gonna reboot and try again | 20:23 |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.32.204) | 20:23 | |
| Boostisbetter | abortretryfail: ahh ok, I'm not encrypted, maybe that matters. | 20:24 |
| eery | The python thing sounds like a problem we ran into at work with updating sid a month or two ago | 20:24 |
| + Ar|stote|is (~linx@149.210.32.204) | 20:25 | |
| Boostisbetter | honestly i've been very surprised with how stable sid has been in general | 20:36 |
| Boostisbetter | but I don't feel the need to update too much. | 20:36 |
| - eery (QUIT: Remote host closed the connection) (~eery@h210.3.55.139.dynamic.ip.windstream.net) | 20:36 | |
| Boostisbetter | I kind of have the system working and resuming from suspend everytime perfectly. I don't want to mess that up. Hahahaha | 20:37 |
| Boostisbetter | flatpaks for the win. | 20:37 |
| + eery (~eery@h210.3.55.139.dynamic.ip.windstream.net) | 20:47 | |
| - eery (QUIT: Remote host closed the connection) (~eery@h210.3.55.139.dynamic.ip.windstream.net) | 20:55 | |
| - Boostisbetter (QUIT: Ping timeout: 240 seconds) (4a410829d7@irc.cheogram.com) | 20:57 | |
| vkoskiv | I found a servo with a dupont connector that might be suitable to create an adapter cable for UART | 21:00 |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 21:03 | |
| + eery (~eery@139.55.3.210) | 21:05 | |
| - Boostisbetter (QUIT: Ping timeout: 272 seconds) (4a410829d7@irc.cheogram.com) | 21:11 | |
| vkoskiv | Is there a recommended way to poke at /dev/hidraw devices like the keyboard without running as root? | 21:15 |
| vkoskiv | My current approach is to just run as root just while I open the file, then drop root right away, and carry on | 21:16 |
| vkoskiv | But I don't want to write code that runs as root at all if I can avoid it | 21:16 |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 21:18 | |
| - eery (QUIT: Remote host closed the connection) (~eery@139.55.3.210) | 21:20 | |
| Zaba | add a udev rule to set permissions on the /dev/hidraw file? | 21:35 |
| vkoskiv | Is that an acceptable thing to do in a debian package? | 21:39 |
| abortretryfail | probably not | 21:43 |
| abortretryfail | does anyone know what the heck this message means? | 21:44 |
| abortretryfail | mxsfb 30320000,lcd-controller: error -ENODEV: Cannot connect bridge | 21:44 |
| abortretryfail | DEBUG: ti_sn65dsi86_suspend skipped. | 21:45 |
| - klardotsh (QUIT: Ping timeout: 240 seconds) (~klardotsh@c-67-170-115-80.hsd1.wa.comcast.net) | 21:55 | |
| abortretryfail | oh that's the adapter that connects the internal display. weird. | 21:57 |
| minute | abortretryfail: is this a fresh update from today? | 22:04 |
| abortretryfail | yes, this morning | 22:04 |
| abortretryfail | i did apt update; apt upgrade; the broken python package seems to be a red herring. I removed it | 22:04 |
| abortretryfail | weird thing is, all of the kernels i have in the uboot menu say something similar about the edp bridge | 22:05 |
| minute | abortretryfail: unfortunately i'm totally out of energy for today/this week so i can't immediately help, but the first thing i'd do is try to image an sd card from https://mnt.re/system-image and see if that wokrs | 22:05 |
| minute | works | 22:05 |
| minute | ah, uboot menu... haven't tried that | 22:05 |
| abortretryfail | i have a system image booted and it works fine, so the hardware is OK | 22:05 |
| minute | abortretryfail: you're getting this error on serial, i suppose? | 22:05 |
| abortretryfail | no, thats the last thing it writes to the LCD before it just stops. | 22:06 |
| abortretryfail | i dont have a ttl serial console handy | 22:06 |
| minute | ah, well if it can write to the display this shouldn't be an issue then, probably can ignore it | 22:06 |
| abortretryfail | i can ignore it but the machine is still useless in that state :) | 22:06 |
| minute | it would be useful to see the effective kernel commandline | 22:06 |
| minute | maybe you're bitten by the recent problem that the decryption question is on the serial console | 22:07 |
| minute | because the console= bootargs are the wrong way around or similar | 22:07 |
| minute | ah, you can try replacing boot.scr with the known good one: | 22:07 |
| abortretryfail | ugh, maybe? i'll check that thanks | 22:07 |
| minute | abortretryfail: https://community.mnt.re/t/reform-won-t-finish-booting/1675/13 | 22:07 |
| minute | download this file, rename to boot.scr and put on the first partition of your sd card | 22:08 |
| minute | but i'm not sure how it interacts with u-boot-menu, how did you set that up? | 22:08 |
| minute | i haven't personally tested the menu | 22:08 |
| abortretryfail | i dont remember, that was a couple months ago | 22:08 |
| minute | ok, sounds like that might be related | 22:09 |
| abortretryfail | can i just edit the boot.scr file or is the binary gibberish in the top going to get upset? | 22:09 |
| minute | you cannot | 22:09 |
| abortretryfail | stinking ARM ecosystem i swear... :) | 22:09 |
| minute | you will have to convert it to a text file first and then recreate the gibberish with mkimage | 22:09 |
| minute | well, uboot is to blame here ;( | 22:09 |
| abortretryfail | yeah, that's what i mean | 22:09 |
| minute | abortretryfail: mkimage -c none -A arm -T script -d boot.script boot.scr | 22:10 |
| minute | where boot.script is your plaintext "source" of boot.scr without the first binary line | 22:10 |
| minute | ok, will head home, this week has been really exhausting, good luck! | 22:11 |
| vkoskiv | I have crafted the world's ugliest UART adapter cable :D | 22:14 |
| vkoskiv | dupont -> 3x pins I can stick into my flipper zero's GPIO | 22:14 |
| vkoskiv | Three hasty solder joints, individually wrapped in kapton + all three wrapped in more kapton | 22:14 |
| vkoskiv | It should work | 22:14 |
| abortretryfail | minute: thanks for pointin me in the right direction. Have a nice weekend. | 22:16 |
| abortretryfail | vkoskiv: !!! I dont have a TTL UART adapter but I *do* have a Flipper Zero. I totally forgot it could do that. Thanks for the reminder! | 22:16 |
| vkoskiv | I guess the flipper shows up as /dev/ttyACM0 | 22:18 |
| vkoskiv | Now I have to re-remember how to show serial links, I guess screen | 22:18 |
| vkoskiv | 115200 baud I'll assume | 22:19 |
| vkoskiv | Yeah! it's working | 22:22 |
| vkoskiv | Alright, I'll de-jank the setup a bit, then I can mess around with this | 22:22 |
| abortretryfail | I need to figure out how to edit the kernel cmdline from the u-boot prompt | 22:23 |
| abortretryfail | basically the only thing you ever need to do with your bootloader and u-boot makes it hard to do | 22:24 |
| vkoskiv | This reminds me that I haven't done anything with the 'hack the planet' connector yet. Have any of you? | 22:24 |
| abortretryfail | what are you going to add to the LPC? | 22:25 |
| vkoskiv | I don't know yet | 22:37 |
| vkoskiv | At the moment I'm setting up to see what happens when I go to suspend A311D | 22:38 |
| vkoskiv | I'll modify the existing suspend script I guess, and go from there | 22:38 |
| abortretryfail | minute: u-boot-menu appears to be the culprit. I renamed /boot/extlinux/extlinux.conf and its able to boot and prompts for cryptsetup on the console correctly. | 22:40 |
| abortretryfail | for now i might just uninstall that... | 22:40 |
| Boostisbetter | abortretryfail, | 22:44 |
| Boostisbetter | abortretryfail, whew! Glad you got it working! | 22:44 |
| vkoskiv | keyboard also works on uart, that means I hooked the pins up correctly | 22:45 |
| vkoskiv | Very neat | 22:45 |
| + aliosablack (~chomwitt@2a02:587:7a24:b000:1ac0:4dff:fedb:a3f1) | 22:48 | |
| vkoskiv | For the curious, here's me running suspend with nothing modified: https://mister-muffin.de/p/mIU-.txt | 22:55 |
| - jacobk (QUIT: Ping timeout: 255 seconds) (~quassel@129.110.242.224) | 22:56 | |
| abortretryfail | while logged in via the serial console? | 22:58 |
| vkoskiv | yeah | 22:59 |
| vkoskiv | Since I lose display + input on the thing | 22:59 |
| abortretryfail | I need to wire that up and see if i can get hibernate to work | 23:02 |
| vkoskiv | Yeah, hibernate is IMO even more valuable than suspend | 23:02 |
| vkoskiv | That was my MO for all the years I owned Macs. Super convenient to be able to turn it "off", and then all the state is there when I turn it back on | 23:03 |
| abortretryfail | its weird, i cant find good examples of working hibernate with Linux on ARM anywhere. I went digging last night | 23:03 |
| vkoskiv | That's maybe an idea for the hack the planet connector. Some tiny reed switch to detect the lid closing :D | 23:03 |
| abortretryfail | worst default behavior ever, sleep on lid close :) | 23:04 |
| abortretryfail | would be nice to have it kill the backlight though. that's a good idea | 23:04 |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 23:05 | |
| minute | vkoskiv: i recommend "tio" instead of screen. thank me later | 23:05 |
| vkoskiv | Ah, yeah, I always thought screen was just clunky to use | 23:06 |
| vkoskiv | I'll give it a go after I'm done reading these logs :D | 23:06 |
| vkoskiv | I have to put it into "copy mode" to be able to even scroll. And then surprise! The default scrollback is 1024 lines, because I probably have what, 32MB of RAM? | 23:06 |
| minute | yes, exactly | 23:06 |
| minute | tio doesn't have this issue | 23:06 |
| minute | and tolerates disconnection of the adapter etc | 23:07 |
| vkoskiv | the earliest ddr training stuff comes from before u-boot, right? | 23:07 |
| minute | yep | 23:07 |
| minute | from arm trusted firmware | 23:07 |
| vkoskiv | Ah, I see. And that runs the DDR training thing, is that a blob on this SoM too? | 23:08 |
| minute | yes, funnily it's the same blob | 23:09 |
| abortretryfail | woah that auto-reconnect feature is killer | 23:09 |
| minute | (they licensed the same controller ip as imx8m) | 23:09 |
| vkoskiv | Heh | 23:10 |
| minute | vkoskiv: looks like suspend/resume kind of just works?! | 23:10 |
| vkoskiv | I mean, the display doesn't come back up, that's what I'm investigating | 23:10 |
| vkoskiv | The panel backlight does turn on | 23:10 |
| vkoskiv | But no picture | 23:10 |
| vkoskiv | And it comes back up from suspend right away, I didn't wake it | 23:10 |
| minute | vkoskiv: probably i didn't nuke enough power off/reset/suspend things in the sn65dsi86 driver then | 23:11 |
| vkoskiv | I haven't modified the sleep/suspend script yet, but I think right now that script just exits | 23:11 |
| minute | vkoskiv: i guess you have the debug uart as a wakeup source | 23:11 |
| minute | so i would look into that first | 23:11 |
| vkoskiv | It happened before I hooked up UART too | 23:11 |
| vkoskiv | But I'll check | 23:11 |
| minute | hmm ok | 23:12 |
| minute | there's some way to list all wakeup sources in the system | 23:12 |
| vkoskiv | Are those ti_sn65 errors relevant to the display not coming back perhaps? | 23:12 |
| vkoskiv | That's the eDP adapter chip, is it not? | 23:12 |
| minute | vkoskiv: yes. as i said, not enough nuking of power off stuff in the driver | 23:13 |
| vkoskiv | Ah, okay. | 23:13 |
| vkoskiv | So I just flashed my SD card with the system image, I wonder what the most ergonomic solution is for testing things out? | 23:14 |
| minute | this is what i did so far https://source.mnt.re/reform/reform-debian-packages/-/blob/main/linux/patches/meson-g12b-bananapi-cm4-mnt-reform2/0018-sn65dsi86-never-turn-off.patch?ref_type=heads | 23:14 |
| vkoskiv | I guess I can just dev on the SD card, can't I? | 23:14 |
| minute | sure can | 23:14 |
| minute | vkoskiv: grow the partition first | 23:14 |
| vkoskiv | I'll grok this patch. Though is this in a module or do I have to cross-compile the whole kernel? | 23:15 |
| minute | module | 23:15 |
| vkoskiv | Cool, so should be quick to iterate, then | 23:15 |
| minute | but you have to set up a kernel build | 23:15 |
| vkoskiv | Yeah | 23:15 |
| minute | i recommend getting the config from /boot | 23:16 |
| vkoskiv | How likely is the SD card fs to corrupt from repeated hard shutdowns from suspends gone awry | 23:16 |
| minute | and clone roughly the same kernel version | 23:16 |
| minute | put the config as .config in linux git | 23:16 |
| vkoskiv | Will do, yeah, I remember doing that some time back on another system | 23:16 |
| vkoskiv | Then make oldconfig, right? | 23:17 |
| minute | and before building apply all patches | 23:17 |
| vkoskiv | Also that, yeah | 23:17 |
| minute | i never used make oldconfig... it should just work with make | 23:17 |
| vkoskiv | Do I have to build the whole thing once? Or just build the module I want | 23:17 |
| minute | yes build everything once i think | 23:17 |
| minute | use a big computer:3 | 23:17 |
| vkoskiv | Hmm, so a cross-compiler toolchain build on big PC, then transfer it to the SD card? | 23:18 |
| minute | i think the goal is to have sn65dsi86 to never reset, power off, reinit etc | 23:18 |
| vkoskiv | Yeah | 23:19 |
| minute | vkoskiv: you can copy the .ko over the network | 23:19 |
| vkoskiv | Ah, that works. | 23:19 |
| vkoskiv | should be easy enough to set up cross-compilation, I'll get going on that | 23:19 |
| minute | cool | 23:19 |
| minute | another approach would be to fix all kernel bugs that happen when unloading the modules | 23:19 |
| minute | so one could reload it... | 23:20 |
| vkoskiv | Yeah, are there many? | 23:20 |
| minute | idk how many, but there are oopses | 23:20 |
| vkoskiv | These are just 3rd party modules? Or do you mean reform-specific ones? | 23:20 |
| minute | not 3rd party at all | 23:20 |
| minute | mainline linux | 23:20 |
| minute | but nobody tested them thoroughly i guess | 23:21 |
| vkoskiv | Ah, the reform things are in mainline too now? | 23:21 |
| minute | no | 23:21 |
| minute | but there are no reform specific modules | 23:21 |
| vkoskiv | I know the device tree is in there | 23:21 |
| minute | well ok | 23:21 |
| vkoskiv | The lpc driver is out of tree? | 23:21 |
| minute | yep | 23:21 |
| - stites (QUIT: Ping timeout: 272 seconds) (~stites@130.44.147.204) | 23:45 | |
| + stites (~stites@2607:fb91:dcd:e247:3117:f970:e6:e12c) | 23:46 | |
| vkoskiv | Heh, just resizing the volume somehow borked /boot on the SD card | 23:54 |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@2607:fb91:dcd:e247:3117:f970:e6:e12c) | 23:55 | |
| vkoskiv | Weird, it looks okay, mounts fine | 23:55 |
| + stites (~stites@2603:3005:b69:4100:ffbe:3c4e:c9c0:ad22) | 23:56 | |
| vkoskiv | Maybe bad contact. fsck'd it, now it boots | 23:56 |
| josch | vkoskiv: thank you for the typo you found in /usr/sbin/reform-standby -- i locally fixed the same thing in other scripts as well | 23:58 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!