| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 00:00 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 00:00 | |
| + bkeys (~Thunderbi@173.16.175.75) | 00:05 | |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 00:18 | |
| - mtm (QUIT: Ping timeout: 265 seconds) (~textual@47.202.75.129) | 01:04 | |
| + mtm (~textual@47.202.75.129) | 01:06 | |
| ch | there is an assembly | 01:11 |
|---|---|---|
| ch | people should be back tomorrow | 01:11 |
| ch | in hall H | 01:11 |
| minute | BoostisBetter: works on rk3588 with gl override to 3.3 and glsl override to 330 | 01:24 |
| mhoye | For future reference, if your machine suddenly decides it doesn't have a wireless card in it for totally inexplicable and hard-to-diagnose reasons, step one is "find NetworkManager.state, delete it and reboot". | 01:31 |
| + jacobk (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 01:46 | |
| - jacobk (QUIT: Ping timeout: 276 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 02:03 | |
| + jacobk (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 02:10 | |
| - aloo_shu (QUIT: Ping timeout: 246 seconds) (~aloo_shu@90.166.98.198) | 03:04 | |
| + aloo_shu (~aloo_shu@90.166.98.198) | 03:14 | |
| - nsc (QUIT: Ping timeout: 265 seconds) (~nicolas@i5C74DC72.versanet.de) | 03:19 | |
| + nsc (~nicolas@153-96-142-46.pool.kielnet.net) | 03:21 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@173.16.175.75) | 03:27 | |
| - paperManu (QUIT: Ping timeout: 244 seconds) (~paperManu@107.159.243.8) | 03:34 | |
| - op_4 (QUIT: Remote host closed the connection) (~tslil@user/op-4/x-9116473) | 04:05 | |
| + op_4 (~tslil@user/op-4/x-9116473) | 04:05 | |
| - mjw (QUIT: Ping timeout: 244 seconds) (~mjw@gnu.wildebeest.org) | 04:33 | |
| - jacobk (QUIT: Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) (~quassel@47-186-65-73.dlls.tx.frontiernet.net) | 05:49 | |
| josch | https://media.ccc.de/v/38c3-7-years-later-why-and-how-to-make-portable-open-hardware-computers | 05:53 |
| josch | it says "0 views" but i swear i watched it... :( | 05:54 |
| - cobra_ (QUIT: Ping timeout: 244 seconds) (~cobra@user/Cobra) | 07:08 | |
| + cobra (~cobra@user/Cobra) | 07:10 | |
| - mesaoptimizer (QUIT: Quit: WeeChat 4.0.4) (~mesa@user/PapuaHardyNet) | 10:01 | |
| - aperezdc (QUIT: Remote host closed the connection) (~aperezdc@2a03:6000:6e61:633::43) | 10:02 | |
| + gustav28 (~gustav@c-78-82-52-175.bbcust.telenor.se) | 10:02 | |
| + aperezdc (~aperezdc@2a03:6000:6e61:633::43) | 10:40 | |
| + mesaoptimizer (~mesa@user/PapuaHardyNet) | 10:41 | |
| minute | ah > This is just a stream recording, the final release of this talk is still being worked on. | 11:36 |
| + pandora (uid585533@id-585533.ilkley.irccloud.com) | 11:36 | |
| josch | then cool of them to release early for those who don't want to wait :D | 11:44 |
| - nsc (QUIT: Ping timeout: 244 seconds) (~nicolas@153-96-142-46.pool.kielnet.net) | 11:57 | |
| minute | josch: yeah totally | 12:04 |
| minute | i'm gonna have lunch with an old friend from scholl and then head back to congress between 14-15h | 12:05 |
| - mtm (QUIT: Ping timeout: 252 seconds) (~textual@47.202.75.129) | 13:02 | |
| + mtm (~textual@47.202.75.129) | 13:06 | |
| + nsc (~nicolas@153-96-142-46.pool.kielnet.net) | 13:16 | |
| ch | cool. i'll be at the non-secret secret debian meetup from 14:37 on, but will be back in H after that | 13:18 |
| josch | 13:37 UTC, nice :) | 13:21 |
| + glu_ (~glu@user/glu) | 13:45 | |
| - pandora (QUIT: Quit: Connection closed for inactivity) (uid585533@id-585533.ilkley.irccloud.com) | 13:46 | |
| - glu (QUIT: Ping timeout: 252 seconds) (~glu@user/glu) | 13:46 | |
| * glu_ -> glu | 13:46 | |
| ch | so, will uboot default to reading extlinux config? | 13:52 |
| + mjw (~mjw@gnu.wildebeest.org) | 14:03 | |
| ch | apparently not | 14:09 |
| + paperManu (~paperManu@107.159.243.8) | 14:10 | |
| + Guest17 (~Guest17@79.117.213.185) | 14:18 | |
| - Guest17 (QUIT: Client Quit) (~Guest17@79.117.213.185) | 14:19 | |
| BoostisBetter | minute: Could you pretty please get me a link to the original sysctry and lpc firmware? Despite my efforts the LPC crashed again. | 14:39 |
| BoostisBetter | I would really like to get back on the firmware previous to the current. | 14:39 |
| + Guest38 (~Guest38@1F2EC924.catv.pool.telekom.hu) | 15:48 | |
| - Guest38 (QUIT: Client Quit) (~Guest38@1F2EC924.catv.pool.telekom.hu) | 15:49 | |
| ch | which fw do you have now? | 16:04 |
| ch | just so its clear what you're looking for | 16:05 |
| BoostisBetter | ch: the aug release of this year but it shows in the status page of the lpc that it is 30 Sept | 16:08 |
| BoostisBetter | ch: the one that is currently available is the one that took the charge starting delay from 5 seconds to 1 second | 16:10 |
| BoostisBetter | ch: but somehow this version gets into a state where the LPC is just not responsible. You have power cycle the pocket to get it work again. | 16:11 |
| ch | ok, i cant find anything that would claim 30 Sept | 16:14 |
| ch | probably means i don't know enough | 16:14 |
| BoostisBetter | ch: the version you get from here is what I have currently. https://source.mnt.re/reform/pocket-reform/-/tree/main/pocket-reform-sysctl-fw?ref_type=heads | 16:16 |
| ch | so, sysctl not lpc | 16:17 |
| BoostisBetter | yeah, I think I have the terminolgy wrong. I just interface with the sysctrl via the lpc | 16:20 |
| josch | oh, compiling the pocket-reform-sysctl-fw also requires pico-extras on top of the sdk | 16:22 |
| ch | yeah | 16:22 |
| ch | (current head shouldnt iirc) | 16:23 |
| BoostisBetter | pico-extras is not available | 16:23 |
| ch | you want the specific version from git anyway | 16:23 |
| josch | ch: nah, still requires pico/sleep.h which is only in extras | 16:25 |
| ch | i meant, extras from git, too | 16:25 |
| josch | ah okay :) | 16:25 |
| ch | ah the current version? | 16:25 |
| ch | like, HEAD? | 16:26 |
| ch | sorry, talking past each other i guess | 16:26 |
| josch | pocket-reform-sysctl-fw/inc/sysctl.h needs pico/sleep.h | 16:26 |
| josch | i guess, but there is no problem :) | 16:26 |
| ch | ah yes | 16:26 |
| ch | i think the #include is useless | 16:27 |
| ch | yeah | 16:28 |
| josch | and i need to package the new pico-sdk -- the old one does not have hardware_sleep | 16:30 |
| ch | thats in extras | 16:30 |
| ch | hold on | 16:31 |
| josch | it... links fine if i just remove that from CMakeLists.txt... | 16:31 |
| ch | the light here in the hacklight is so bad, i cant read anything | 16:33 |
| ch | https://source.mnt.re/reform/pocket-reform/-/merge_requests/22 this should so | 16:33 |
| ch | do* | 16:33 |
| josch | thanks, that's what i had locally to make it build without extras now as well | 16:34 |
| ch | i'll quickly flash this to check | 16:35 |
| BoostisBetter | if I had to guess the idle power improvements are probably the culprit. That is just a guess. I don't see how anything else would cause the sysctrl to become unreachable | 16:35 |
| ch | also nonworking ipv4, grr | 16:37 |
| ch | overlay networking it is | 16:37 |
| ch | quick test is ok | 16:39 |
| ch | not really surprising :) | 16:40 |
| ch | josch: do you happen to know if i can force uboot on sdcard to boot using extlinux.conf thats also on sdcard? | 16:47 |
| ch | i've tried deleting boot.scr, but it ended up finding boot.scr on the emmc :( | 16:48 |
| josch | if it doesn't boot extlinux.conf on sd-card in preference to boot.scr on emmc, then that's a bug | 16:49 |
| josch | ch: can you connect to u-boot via serial? | 16:49 |
| ch | no, i dont have a cable/adapter with me | 16:50 |
| josch | you go out with your reform *without* an UART adapter? :D | 16:50 |
| ch | maybe minute has one, will find out later | 16:50 |
| ch | i packed a ton of things, but i forgot the uart adapter | 16:50 |
| josch | should've reminded you when i saw your message on fedi ;) | 16:50 |
| ch | yeah | 16:50 |
| ch | also i need to wire it up somehow so i dont need to unscrew everything to get to the uart | 16:51 |
| josch | maybe you didn't put extlinux.conf in the right place or maybe it's not having good content? | 16:51 |
| ch | its possible | 16:51 |
| josch | did you craft the extlinux.conf yourself | 16:51 |
| ch | i have it in /extlinux/extlinux.conf on /dev/mmcblk1p1 | 16:51 |
| ch | unfortunately yes | 16:51 |
| josch | the location is correct | 16:52 |
| ch | https://paste.debian.net/hidden/96657950/ i imagine sth in there is wrong | 16:52 |
| josch | intentionally without initrd? | 16:52 |
| ch | yes | 16:52 |
| josch | assuming that /Image and /dtb exist, this looks okay at a glance | 16:53 |
| josch | you could try flashing debian-installer to your sd-card and see what happens? | 16:53 |
| ch | tbh, that Image is at least 30M too small | 16:53 |
| josch | because d-i also uses extlinux.conf | 16:53 |
| josch | maybe you are trying to boot 1994 linux? :) | 16:54 |
| ch | does it need uboot on the sdcard in the first place? | 16:54 |
| josch | no | 16:54 |
| josch | the imx8mplus will only read from emmc | 16:55 |
| ch | whats the correct d-i file that i'm looking for? | 16:55 |
| josch | https://reform.debian.net/d-i/pocket-reform-system-imx8mp-bpo.img.xz | 16:55 |
| josch | those work at least with imx8mq | 16:55 |
| josch | i can even flash them on a usb-stick and they boot me into d-i | 16:56 |
| ch | that would be very convenient | 16:56 |
| josch | but i don't know if imx8mplus u-boot has usb-boot enabled | 16:56 |
| josch | also, at least imx8mq is *very* picky about which usb-stick it likes | 16:56 |
| ch | ;) | 16:57 |
| josch | i had a 25% success rate | 16:57 |
| ch | rk3588 ;) | 16:57 |
| ch | right | 16:57 |
| josch | sorry, i was talking about imx8mplus for all my answers above | 16:57 |
| ch | sorry, should have said so earlier | 16:57 |
| josch | it's hard to keep in mind who has which module in their unit XD | 16:57 |
| josch | ugh... i really need to write down the u-boot preferences of each module some day | 16:58 |
| josch | ah wait | 16:59 |
| josch | past me was smart and wrote this commit message: | 16:59 |
| josch | "rk3588 will prefer u-boot from sd-card if present" | 16:59 |
| ch | also wondering if it will find the usbstick behind the ti usb hub | 16:59 |
| ch | doesnt seem to care about it anyway | 16:59 |
| minute | never fear, i is here | 17:00 |
| ch | :) | 17:00 |
| ch | \o\ | 17:00 |
| ch | \o/ | 17:00 |
| josch | minute: this channel is better with you in it :) | 17:00 |
| BoostisBetter | minute: this channel wouldn't exsist without you. 🙂 | 17:04 |
| ch | ok, one step further | 17:05 |
| ch | now the q is whats the correct root= | 17:05 |
| josch | ch: oh what was the problem? | 17:06 |
| ch | apparently truncated kernel image | 17:07 |
| ch | i find it funny that uboot then just ignores extlinux.conf and boots whatever comes next | 17:07 |
| ch | right now linux aborts with VS: Unable to mount root fs on unknown-block(0,0) | 17:08 |
| ch | which is weird | 17:08 |
| ch | more frankenstein and it boots | 17:16 |
| + colinsane (~colinunin@97-113-135-40.tukw.qwest.net) | 17:19 | |
| ch | so, what was my original plan | 17:24 |
| ch | bad news: with the display panel from 6.11 the display doesnt turn on at all | 17:46 |
| ch | [ 258.807138] rk_gmac-dwmac fe1b0000.ethernet end0: __stmmac_open: Cannot attach to PHY (error: -19) | 17:55 |
| ch | [ 276.745591] rk_gmac-dwmac fe1b0000.ethernet end0: Register MEM_TYPE_PAGE_POOL RxQ-0 | 17:55 |
| ch | [ 276.746067] rk_gmac-dwmac fe1b0000.ethernet end0: Register MEM_TYPE_PAGE_POOL RxQ-1 | 17:55 |
| ch | gah | 17:55 |
| - mjw (QUIT: Killed (osmium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org) | 18:35 | |
| * Guest1607 -> mjw | 18:35 | |
| + Guest5145 (~mjw@gnu.wildebeest.org) | 18:35 | |
| BoostisBetter | minute: I can't really see, but I think there are times when you can see your desktop and can see the MNT Sway layout, but are you using a Reform to do your talk at 38c3? | 19:06 |
| BoostisBetter | minute: I am about half way through, and I just wanted to thank you for giving it. Was educational and good to hear your perspective! | 19:06 |
| BoostisBetter | minute: you and I are very similar about obsessions with UMPCs | 19:07 |
| BoostisBetter | small world!!! | 19:07 |
| * mjw -> Guest1908 | 19:15 | |
| - Guest1908 (QUIT: Killed (copper.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 19:15 | |
| * Guest5145 -> mjw | 19:15 | |
| + Guest1908 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c) | 19:15 | |
| BoostisBetter | minute: You answered the laptop question in your talk. Very good work once again! I think the progression of the MNT catalogue really shows how far you and the team have come along. Things are really only getting better and better. | 19:15 |
| BoostisBetter | minute: there was a question about how much the open side of the catalog was a draw versus design and functionality. I can say that I was someone who was drawn in by the design over the open nature of it all. I don't really care about FOSS in general, although I think it is good. I really think you nailed the design. It shows a true passion to computing that is not evident in all of the modern la | 19:21 |
| BoostisBetter | ptops that all look the same. I think your design team is absolutely fantastic. | 19:21 |
| ch | the rk dsi2 phy driver seems funny | 19:29 |
| ch | if (dsi2->mode_flags & MIPI_DSI_MODE_VIDEO_NO_HFP) | 19:29 |
| ch | val |= BLK_HFP_HS_EN; | 19:30 |
| ch | if (NO) ENABLE; | 19:30 |
| ch | maybe the names are wrong, but this seems silly | 19:30 |
| - sterni (QUIT: ) (~quassel@user/sterni) | 19:32 | |
| + sterni (~quassel@user/sterni) | 19:34 | |
| BoostisBetter | curious thing. If you download the artifact for the sysctrl update from releases, it will show a 0730 date for the firmware. | 20:12 |
| BoostisBetter | However it you get it from this artifact: https://source.mnt.re/reform/pocket-reform/-/jobs/6894/artifacts/download?file_type=archive | 20:12 |
| BoostisBetter | Then after updating it will then show 0830 as it should in the status screen of the LPC. I am hopeful this will improve the LPC (Sysctrl) from crashing. | 20:13 |
| BoostisBetter | does anyone have the USBDEV address of the keyboard RP2040? I want to update the firmware there. | 20:31 |
| ch | https://source.mnt.re/reform/pocket-reform/-/blob/main/pocket-reform-keyboard-fw/pocket-hid/flash.sh?ref_type=heads | 20:32 |
| ch | apparently something on bus 1 | 20:32 |
| BoostisBetter | yeah, but without the address it is not possible to address it and get it to boot in bootsel mode | 20:36 |
| ch | i think you cant anyway | 20:36 |
| ch | you need to press the button | 20:36 |
| BoostisBetter | How is it possible to do that on the Sysctrl then that is also a 2040? | 20:36 |
| ch | different firmware | 20:37 |
| BoostisBetter | gotcha | 20:37 |
| ch | sysctl fw exposes a reset API, the keyboard fw does not (currently) | 20:37 |
| BoostisBetter | ahh so to update the keyboard we need to open the bottom plate and push the bootsel button | 20:38 |
| ch | the manual has the correct path documented | 20:39 |
| ch | i dont know it offhand | 20:39 |
| BoostisBetter | thanks, checking | 20:41 |
| BoostisBetter | don;t see any mention of it, but I am going to see if the newer sysctrl firmware fixes the issue I was having. | 20:44 |
| BoostisBetter | amospalla: I'm getting file system errors again with psuspend | 21:05 |
| BoostisBetter | amospalla: kernel: EXT4-fs (dm-2): ext4_do_writepages: jbd2_start: 4096 pages, ino 31720156; err -30 | 21:06 |
| BoostisBetter | these start happening after resuming from psuspend | 21:06 |
| BoostisBetter | ch: the aug release of this year | 21:37 |
| BoostisBetter | but it shows in the status of the lpc 30 Sept | 21:37 |
| josch | 95% Funded | 22:13 |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-52-175.bbcust.telenor.se) | 22:15 | |
| - aloo_shu (QUIT: Quit: bbl migrating) (~aloo_shu@90.166.98.198) | 22:16 | |
| - colinsane (QUIT: Ping timeout: 272 seconds) (~colinunin@97-113-135-40.tukw.qwest.net) | 22:24 | |
| BoostisBetter | yep and in barely 2 days time. | 22:34 |
| + colinsane (~colinunin@97-113-84-53.tukw.qwest.net) | 22:38 | |
| - antti (QUIT: Ping timeout: 252 seconds) (~antti@user/antti) | 22:46 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 22:47 | |
| + antti (~antti@user/antti) | 22:49 | |
| vagrantc | curious if there is a power use comparison with rk3588 16GB vs. 32GB ram ... i would assume more ram leads to some degree of higher idle power draw ... obviously in situations where you *need* more ram it would be important | 23:03 |
| vagrantc | e.g. swapping out to even NVMe is probably going to draw more power than actaul ram... | 23:04 |
| + aloo_shu (~aloo_shu@90.166.98.198) | 23:04 | |
| sevan | vagrantc: would measuring the draw from power outlet be a way of finding out, having disconnected the batteries from the system so that it is running solely from PSU? | 23:06 |
| sevan | I ask as I have a meansuring device + a 32GB RAM module that's not installed yet. | 23:07 |
| josch | vagrantc: as far as i'm aware, there is not yet a measurement or benchmark of rk3588 power consumption in the reform | 23:17 |
| josch | even less so a comparison of 16 GB vs. 32 GB | 23:18 |
| josch | sevan: you have a rk3588 in classic or pocket reform? | 23:18 |
| sevan | josch: not installed in the classic | 23:19 |
| josch | sevan: if it's not installed yet, maybe just measure the battery runtime with a reproducible workload before and after? | 23:19 |
| sevan | josch: still on the original CPU module, got side tracked with the v4 keyboard firmware. | 23:19 |
| sevan | josch: yak shaving with dead battery cells. I have new cells but also not installed yet + I think the v4 firmware lacks anything regarding battery so I'll just end up killing the new cells? | 23:21 |
| josch | sevan: how should the keyboard firmware kill the batteries? | 23:21 |
| josch | charging happens via lpc and you have the protected battery board, no? | 23:21 |
| sevan | there was handling in the original firmware for when the battery is low which is currently missing in the v4 firmware. | 23:22 |
| sevan | I do have the protected battery board. | 23:23 |
| + reform26176 (~ckeen@2001:67c:20a1:1121:b0fa:6051:5b28:4f91) | 23:26 | |
| reform26176 | hey there! my reform's waybar battery status diverges from the percentage value shown in my LCD display. Where should I look for the cause of this? | 23:27 |
| BoostisBetter | reform26176: perhaps the polling rate of the waybar is just different. How big of a difference are we talking about? | 23:27 |
| josch | reform26176: classic reform? everything upgraded? including firmware? | 23:29 |
| reform26176 | reform2, maybe I need to update the LPC firmware... it is at MREF2LPC320220621 | 23:30 |
| reform26176 | the difference is not fixed, it starts at 5% and increases as the battery is drained | 23:30 |
| josch | reform26176: system is fully upgraded, yes? | 23:31 |
| josch | i ask because waybar changed how it keeps track of the batter percentage | 23:32 |
| reform26176 | ii waybar 0.11.0-2+b2 arm64 Highly customizable Wayland bar for Sway and Wlroots based compositors | 23:33 |
| josch | okay, good | 23:34 |
| reform26176 | I had to change the battery name to BAT0 in my config | 23:34 |
| josch | yes, that means that you recently also upgraded the dkms module | 23:34 |
| reform26176 | crrect | 23:34 |
| reform26176 | do I need to reflash the LPC controller firmware? | 23:34 |
| josch | reform26176: motherboard 2.0 i guess? | 23:34 |
| reform26176 | yes, with updated battery modules | 23:35 |
| josch | okay, i can try and reproduce your problem tomorrow | 23:35 |
| reform26176 | thanks! | 23:35 |
| reform26176 | you do not happen to be at 38c3 by chance? | 23:35 |
| josch | sadly not :( | 23:35 |
| josch | likely next year but that doesn't help today :) | 23:36 |
| reform26176 | oh well looking forward to meeting you in person :) | 23:36 |
| reform26176 | I will check in with a more stable irc connection tomorrow | 23:37 |
| josch | okay! | 23:37 |
| josch | reform26176: when you upgrade your lpc firmware, please use the artifacts built by this MR: https://source.mnt.re/reform/reform/-/merge_requests/79 | 23:37 |
| josch | if you just take the latest version, you will run into this bug: https://community.mnt.re/t/new-batteries-not-charging/2695 | 23:38 |
| vagrantc | sevan: mostly just a curiosity ... someone convinced me to get less ram on another computer once, but mostly through blind acceptance rather than actually running the numbers :) | 23:38 |
| vagrantc | i've got the rk3588 in my mnt/reform now and ... it does eat through battery pretty fast ... does not help that i have not paid attention and let it get down to 0% a few times | 23:39 |
| BoostisBetter | yeah, I would imagine the power draw would double between the imx8 and the rk3588 | 23:39 |
| vagrantc | although it also never seems to lag anymore like the earlier imx8mq module :) | 23:39 |
| reform26176 | josch: thanks for the heads up | 23:41 |
| reform26176 | aye | 23:43 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!