2022-06-16.log

vkoskiv_Any known issues at the moment or should I try to look for one? No repro steps yet. I think it has to do with me having it on a charger, turning it on and then leaving it for a while. Not sure.00:00
vkoskiv_I'll try clicking the kbd reset with a toothpick00:00
vkoskiv_I can see it in there from the gap00:00
vkoskiv_reset button didn't do a thing. Last time pulling charger and batteries fixed it00:01
vkoskiv_Just noticed - haven't looked at schematic yet, two legs on my LTC power measurement chip are bridget. On the not-pin-1 side.00:02
+ Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66)00:03
+ patrick_ (224fa09e8b@2604:bf00:561:2000::217)00:03
+ conky_ (5fb0fe5593@2604:bf00:561:2000::10b)00:03
+ yankcrim- (~nick@agrajag.dischord.org)00:04
+ jfred_ (sid534649@libera/sponsor/jfred)00:04
+ nocko_ (sid501219@id-501219.helmsley.irccloud.com)00:04
+ eschaton_ (~eschaton@li541-49.members.linode.com)00:04
+ jfred-linode_ (quassel@libera/sponsor/jfred)00:05
+ Asmadeus_ (~asmadeus@240b:13:8c80:d300:e:98c:8000:d300)00:06
+ sbp_ (~sbp@2001:19f0:6800:1102:5400:ff:fe11:39a1)00:06
+ leonardo1 (~leonardo@user/leonardo)00:06
+ sinedeo_ (~sinedeo@2a01:4f9:c010:2837::1)00:06
+ jnerula_ (~jnerula@li1009-93.members.linode.com)00:07
+ swivel_ (~swivel@shells.gnugeneration.com)00:07
- eschaton (QUIT: Ping timeout: 250 seconds) (eschaton@2600:3c01::f03c:91ff:fefd:5d92)00:07
- nocko (QUIT: Ping timeout: 250 seconds) (sid501219@user/nocko)00:07
- scops (QUIT: Ping timeout: 250 seconds) (~scopstchn@2001:470:69fc:105::8da)00:07
- patrick (QUIT: Ping timeout: 250 seconds) (224fa09e8b@fsf/member/patrick)00:07
- leonardo (QUIT: Ping timeout: 250 seconds) (~leonardo@user/leonardo)00:07
- Asmadeus (QUIT: Ping timeout: 250 seconds) (~asmadeus@240b:13:8c80:d300:e:98c:8000:d300)00:07
- swivel (QUIT: Ping timeout: 250 seconds) (~swivel@shells.gnugeneration.com)00:07
- jfred (QUIT: Ping timeout: 250 seconds) (sid534649@libera/sponsor/jfred)00:07
- conky (QUIT: Ping timeout: 250 seconds) (5fb0fe5593@2604:bf00:561:2000::10b)00:07
- sinedeo (QUIT: Ping timeout: 250 seconds) (~sinedeo@2a01:4f9:c010:2837::1)00:07
- jfred-linode (QUIT: Ping timeout: 250 seconds) (quassel@libera/sponsor/jfred)00:07
- jnerula (QUIT: Ping timeout: 250 seconds) (~jnerula@li1009-93.members.linode.com)00:07
- Gooberpatrol66 (QUIT: Ping timeout: 250 seconds) (~Gooberpat@user/gooberpatrol66)00:07
- sbp (QUIT: Ping timeout: 250 seconds) (~sbp@apache/doge/sbp)00:07
- yankcrime (QUIT: Ping timeout: 250 seconds) (~nick@agrajag.dischord.org)00:07
* conky_ -> conky00:07
* jfred_ -> jfred00:07
* leonardo1 -> leonardo00:07
+ scops (~scopstchn@2001:470:69fc:105::8da)00:18
- mjw (QUIT: Quit: Leaving) (~mark@gnu.wildebeest.org)00:22
* Asmadeus_ -> Asmadeus00:23
- S0rin (QUIT: Ping timeout: 276 seconds) (~S0rin@user/s0rin)00:24
+ S0rin (~S0rin@user/s0rin)00:28
- Gooberpatrol_66 (QUIT: Quit: Leaving) (~Gooberpat@user/gooberpatrol66)00:33
* sbp_ -> sbp00:36
- sbp (QUIT: Changing host) (~sbp@2001:19f0:6800:1102:5400:ff:fe11:39a1)00:36
+ sbp (~sbp@apache/doge/sbp)00:36
joschvkoskiv_: sounds like this: https://community.mnt.re/t/power-issues-and-lpc-wakeup/59600:45
joschI've had the same effect two days ago for the first time -- only plugging the batteries fixed it00:45
minuteone person had a proposed patch for it00:53
* nocko_ -> nocko00:58
joschsince i don't understand https://source.mnt.re/reform/reform/-/merge_requests/26 and since this problem happened to me so rarely, i wanted to wait until it got official blessing by getting merged before trying this out -- i might change my mind in case this happens to me more often :)01:11
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)01:13
- Gooberpatrol66 (QUIT: Ping timeout: 258 seconds) (~Gooberpat@user/gooberpatrol66)01:45
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)01:52
- GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon)02:00
+ GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon)02:01
- mtm (QUIT: Ping timeout: 256 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net)02:02
AsmadeusI guess the warning to not fiddle with wdt from interrupt context sort of make sense, but that sleep being required (according to forum post) is quite weird...02:04
- Christoph_ (QUIT: Remote host closed the connection) (~Christoph@p54bf616b.dip0.t-ipconnect.de)02:18
* swivel_ -> swivel03:16
- klardotsh (QUIT: Ping timeout: 240 seconds) (~klardotsh@172.56.104.60)03:39
+ bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com)03:44
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@static-198-54-135-69.cust.tzulo.com)03:57
+ bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com)03:57
+ mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net)04:08
flowyminute: oh, that's cool. i would order a reform trad layout keyboard, for sure! i'm not a fan of the current stagger, however i am definitely a fan of making compromises so that we actually have nice things and they don't simply remain pipe dreams forever :)04:26
flowyi tried fps gaming with the standalone keyboard and the stagger actually causes finger pain on WASD keys (arrow formation...). it really illustrates the ergonomic compromise04:28
flowydirect diagonal combos can be painful basically04:29
flowyanyways i still like the reform keyboard04:50
flowyit's got character04:50
- bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com)04:55
- vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20)06:50
+ littlebo1eep (~alMalsamo@gateway/tor-sasl/almalsamo)07:34
- littlebo1eep (QUIT: Ping timeout: 240 seconds) (~alMalsamo@gateway/tor-sasl/almalsamo)08:09
+ MajorBiscuit (~MajorBisc@c-001-020-042.client.tudelft.eduvpn.nl)09:11
Boostisbetterhave you ever used and orthogonal layout keyboard? The Moonlander I use on my desktop is the most comfortable keyboard I have ever used. 10:11
BoostisbetterI find the keyboard on the Reform to be very good though. 10:11
Boostisbetterone of my favorite parts of the keyboards is the display. I love that you can check battery status and have an on and off switch that are all independent from the OS. 10:14
* mark -> mjw11:32
+ Christoph_ (~Christoph@p54bf6030.dip0.t-ipconnect.de)12:24
- mjw (QUIT: Quit: Leaving) (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440)12:48
- MajorBiscuit (QUIT: Ping timeout: 272 seconds) (~MajorBisc@c-001-020-042.client.tudelft.eduvpn.nl)13:01
+ MajorBiscuit (~MajorBisc@2a02-a461-129d-1-193d-75d8-745d-e91e.fixed6.kpn.net)13:03
+ mjw (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440)13:06
flowyi've only tried ortholinear layouts briefly. i appreciate the aesthetic, but i'm not sold yet on the ergonomics. i've only seen anecdotal claims that swing either way in regards to whether they help RSI. i'm wary of investing energy into learning another layout if the benefits are unclear13:35
flowyi suffer from light RSI and had a couple bad periods, fortunately i found various things that work for me before going deep into alternative keyboard layouts.13:39
flowybut moonlander and other keyboards from that company look very cool!13:41
flowyi'm not particularly interested in the pocket reform personally, again for ergonomics. i'm wary of devices that small, particularly for eye strain. but i guess i don't have a particular use for it. the reform's size is pretty good for me as a functional portable computer. could just be a bit slimmer13:44
flowyi'm a big fan of high DPI screens + big fonts :)13:45
flowyfor me more pixels mean better type, not more space13:45
flowythe pocket reform sounds really cool though and i wish i did have some use for it heh13:47
flowylots of time to think of a reason13:47
- mtm (QUIT: Ping timeout: 240 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net)14:02
- MajorBiscuit (QUIT: Ping timeout: 248 seconds) (~MajorBisc@2a02-a461-129d-1-193d-75d8-745d-e91e.fixed6.kpn.net)14:03
+ MajorBiscuit (~MajorBisc@c-001-030-042.client.tudelft.eduvpn.nl)14:05
vkoskiv_I'm a fan of the stock reform keyboard. I didn't find it difficult to adjust to the layout. The only thing that still occasionally trips me up is hitting return when I meant to hit ', and hitting the up arrow when going for rshift.14:28
vkoskiv_The first issue is just the ansi/horizontal return, which I'm not used to. All euro layouts I've used before have a vertical return.14:29
vkoskiv_The only thing I don't like is the deck flex in the bottom centre of the board. I'm thinking about modeling a little bracket that would grab onto the edge of the PCB and brace against the acrylic bottom cover. Been a while since I've used CAD, though.14:29
vkoskiv_Like an F-shape bracket, but with a wide base that then sits against the acrylic cover when closed. Possibly with some interface material to prevent scratching over time.14:30
flowyyeah that's a good idea14:38
vkoskiv_I have... *a* 3D printer. It's a bit of a crusty one, I built it in 2016.14:46
+ bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com)14:52
vkoskiv_Another firmware idea: When hitting circle+0, if the system is on, first send acpi shutdown signal to linux, and then only force shutdown after some timeout interval (tbd)15:44
vkoskiv_Would that make sense? Seems more gentle than a forced shutdown. Then there could be a key combo to force the shutdown15:45
vkoskiv_Alternatively, just a quick "you sure?" check for the shutdown cmd. Would be simpler.15:45
vkoskiv_What is gsd-xsettings? Some gnome settings daemon? Do I need it? It's using up a bunch of ram.16:07
+ mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net)16:09
slkfx: vkoskiv_: i eventually made my image board single user (me) after repeated and persistent invasions16:13
vkoskiv_Yeah. Mine's just so unknown that I'm the only semi-active user16:15
vkoskiv_I can disable registrations if I want. Or make them approve-only.16:16
vkoskiv_I don't understand why anything from gnome-settings-daemon is running when I'm not using gnome. I'm just running sway.16:16
vkoskiv_I'm comparing this to my arch setup, which had none of that stuff running. I'm used to it taking up <100MB of memory when booted up with no sway session running.16:17
vkoskiv_82MB of RAM seems like a ton for a single settings daemon to be using, IMO :D16:17
vkoskiv_I'd love to know what it's doing with that memory.16:18
* vkoskiv_ -> vkoskiv16:21
Boostisbetterjust had a weird wifi bug for the first time. When I resumed it came back up, but wifi did not. When looking through the connection wifi wasn't even an option. I slept the Reform again and woke it shortly thereafter and wifi was back. I guess the wifi card just took a little too long on resume and the kernel ignored it. 16:42
vkoskivSpeaking of resume, it still fails 100% of the time for me16:44
vkoskivI just get 'no response from lpc' every time16:44
Boostisbetteroh, since upgrading I have had really reliable standby and resume action. 16:45
BoostisbetterIt works REALLY well. 16:45
BoostisbetterSo much so that when it occasionally does not, I have no problem just quickly setting everything back up in my session. 16:45
vkoskivI don't think it's a software issue in my case, the firmware probably doesn't even send the wake up signal to the system16:51
vkoskivI'm running the latest firmware on both kbd and lpc though, so not sure why my particular case doesn't work16:51
vkoskivI'm also running that little patch I made, but that doesn't affect this.16:52
- GNUmoon (QUIT: Ping timeout: 240 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon)17:04
+ GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon)17:18
- mjw (QUIT: Quit: Leaving) (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440)17:21
+ mjw (~mark@gnu.wildebeest.org)17:51
- MajorBiscuit (QUIT: Quit: WeeChat 3.5) (~MajorBisc@c-001-030-042.client.tudelft.eduvpn.nl)18:34
bkeysjosch: So something I have been thinking about is, it shouldn't matter that my Reform doesn't boot debian since I am trying to boot Fedora anyway. But I realized, that the image I am booting does not have any reference to distroboot, do you have any clue why that is?20:00
bkeysDoes your image have your distroboot changes?20:01
+ klardotsh (~klardotsh@98.97.119.207)20:11
joschbkeys: no, you have to build your own uboot from the main branch20:12
joschbkeys: the current sysimage-v3 uses the v3 tag of u-boot and the distroboot stuff came afterwards20:12
joschit might still matter that you cannot boot debian because this hints at that something doesn't work as you expect it to which will make booting fedora harder because you don't have a working setup to start from20:13
joschbkeys: I verified that this one works: wget -O flash.bin https://source.mnt.re/reform/reform-boundary-uboot/-/jobs/artifacts/master/raw/flash.bin\?job\=build20:39
- TadeusTaD (QUIT: Quit: Stay Cheeki Breeki) (tadeustad@psifactor.pl)21:06
+ TadeusTaD (tadeustad@2a01:4f8:13b:288c::2)21:28
bkeysOkay well that at least makes sense then.21:40
bkeysjosch: Do you have a full system image somewhere that does use the distroboot stuff? I agree, I'd like to at least have a working system.21:41
joschbkeys: you can just combine the latest working sysimage-v3 with more recent u-boot -- would you like me to do that for you or tell you how to do it?21:42
josch(the first version would have the advantage that i could test the exact same image to make sure that it boots for me)21:43
bkeysMy guess is that it basically is that the first 8 megabytes (or so) already have the non-distroboot uboot on there, I flash this and the rest of the debian image should "just work"21:44
bkeysBasically I dd the flash.bin you just sent to the front of the SD card21:44
bkeysHopefully it works...21:44
joschyou have to dd it with the correct offset21:45
joschdd if=flash.bin of=/dev/mmcblk0 conv=notrunc bs=1k seek=3321:45
bkeysBooting...21:51
bkeysNothing on the screen so far.21:51
joschbkeys: did you manage to get hold of a serial adapter or a hdmi screen?21:53
bkeysNo it's in the other room; let me go get it21:53
bkeysjosch: Bad Linux ARM64 Image Magic!22:22
bkeysThere is no reference to distroboot in the printenv either22:23
joschu-boot tells you that?22:23
bkeysYes22:23
joschone sec, lets make sure we are both using the same thing -- i'll prepare something22:23
bkeysOh wait... I am on emmc, I was on SD at the beginning of this convo22:24
bkeysI mean I didn't have it working then either22:24
bkeysNeither of them reference uboot cause I tried it on both22:25
bkeys*neither of them reference distroboot22:27
joschbkeys: you are printing your environment by accessing your u-boot via serial?22:30
- erle (QUIT: Ping timeout: 240 seconds) (~erle@ip5f5af7b3.dynamic.kabel-deutschland.de)22:30
bkeysYes22:32
bkeyssudo minicom -D /dev/ttyUSB022:32
bkeysThen I run22:32
bkeys=> printenv22:32
bkeysERROR: invalid device tree22:33
bkeysERROR: failed to install device tree22:33
joschnice, having serial will make this much easier :)22:33
bkeysThat is the error I get when I boot22:33
bkeysFrom SD card22:33
joschcan you show me the full log from u-boot?22:33
bkeysWhich it's trying to read from mmc 0 and the SD card is mmc 1 so I see why it says that22:33
bkeysSure22:33
bkeysI see what's happening but I have no clue how that is true22:35
bkeyshttps://paste.centos.org/view/a82a23f722:36
bkeysjosch: It still has all my changes from when I tried to load Fedora on here a few days ago?22:36
bkeysBut I have flashed this thing so many times22:36
bkeysAnd i know it is loading Uboot from the SD card, because if I remove the SD card I get no serial output22:36
bkeysI am gonna grab a totally fresh SD card22:37
joschbkeys: this is definitely the wrong uboot22:38
joschthis is ancient uboot from before sysimage-v322:38
kfxdon't uboot-shame bkeys 22:40
joschsorry XD22:40
bkeysI agree haha22:42
joschhere is what I do to produce a working sysimage-v3 sd-card with most recent u-boot including distro_bootcmd:22:43
bkeysBut I don't know why it won't update no matter what22:43
joschwget https://source.mnt.re/reform/reform-system-image/-/jobs/793/artifacts/raw/reform2-imx8mq/reform-rescue-system.img.gz22:43
joschzcat reform-rescue-system.img.gz | sudo dd of=/dev/mmcblk0 bs=10M status=progress22:43
joschwget https://source.mnt.re/reform/reform-boundary-uboot/-/jobs/809/artifacts/raw/flash.bin22:43
joschsudo dd if=flash.bin of=/dev/mmcblk0 conv=notrunc bs=1k seek=3322:43
joschI just did this for my sd-card and it boots from it as expected22:43
joschif that does not work for you, we really have to look elsewhere for the problem22:44
bkeysI will do those things exactly, and if it doesn't update i will just get a new SD card22:46
joschi assume you are still using the one that was shipped with your reform?22:46
bkeysYes22:46
joschokay, good22:47
joschminute: recently I was talking about switching the CI backend now that creating a sysimage-v3 doesn't require root privileges anymore. To test my theory, I just attempted running mkimage.sh on the Debian gitlab CI and this failed because the docker CI runner doesn't support binfmt_misc to emulate arm64 via qemu. I have no idea about docker or gitlab CI but I guess this is something to keep in mind when 22:56
joschchoosing another CI runner. The internet says binfmt_misc can work with docker if run with --privileged.22:56
bkeysjosch: Okay I did your list of commands, and if that doesn't work I will use a whole new SD card23:04
bkeysjosch: Same stuff, no changes to printenv23:05
bkeysI'm gonna blame the SD card at the moment23:05
klardotshjosch: I've done GitLab CI to x-compile aarch64 docker images before (used binfmt_misc under the hood), worked great. some good reading in the space: https://blog.56k.cloud/building-docker-multi-arch-images-with-gitlab-cicd/23:06
+ tadeus_brick (tadeustad@psifactor.pl)23:10
joschklardotsh: as a noob in that area, this article confuses me. Is this article not suggesting to build a docker image for each architecture you need?23:11
bkeysOkay, I put in a totally new SD card and now my printenv is printing out nothing23:14
bkeysI wish this thing would just work...23:15
joschbkeys: this is super odd... if your cpu module would somehow get u-boot from emmc you would at least see some output on serial23:15
bkeysIf I try to run something like ls mmc 0 it reboots23:15
bkeysYes23:15
bkeysI get serial output on this new SD card23:16
joschhow can you run something like ls mmc 0 if it prints nothing?23:16
bkeysprintenv prints nothing23:16
joschah so you are in uboot?23:16
bkeysYes23:16
bkeysBut if I run ls mmc 0 it resets23:16
joschyour uboot output looks like it comes from emmc23:17
joschit cannot come from your sd-card23:17
joschhow certain are you, that your dip-switch is indeed set to ON?23:17
bkeysWell if I remove the SD card I get no serial output at all23:17
joschi don't know why that happens but the u-boot output you see cannot come from the u-boot version i linked to above23:18
joschdoes your dip switch look like this: https://source.mnt.re/reform/reform-handbook/uploads/6e7a46b06df43e788f705b783bd46c15/17e7457e179e68e3c74ab15fb2b47e3ab8f2945d.jpeg23:19
bkeysIt's on ON23:20
bkeysSo no23:21
joschdamn :/23:21
joschi'm out of ideas then23:21
joschi have no explanation for why you see what you are seeing23:21
joschsorry :(23:21
bkeysMaybe the best thing to do would be send it back? I know you guys are really busy as is23:23
bkeysBut I already have the battery issue, I just don't know what to do at this point23:23
bkeysIt'd be nice to have it repaired and maybe having it in your guys's hands would help23:23
joschminute: ^23:24
bkeysI'd be willing to pay for shipping both ways23:24
minutebkeys: sure, please contact support@mntre.com23:43
+ erle (~erle@ip5f5af7b3.dynamic.kabel-deutschland.de)23:55
- bkeys (QUIT: Ping timeout: 276 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com)23:58

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!