2025-01-10.log

- simba (QUIT: Ping timeout: 252 seconds) (~simba@ppp-185-38-55-90.pool9.filstalnetz.de)00:01
noamminute: any eta on v3 classic mobo? :P00:13
minutenoam: yeah i just put it into my personal reform yesterday for intense testing00:44
- chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:587:7a09:1500:1ac0:4dff:fedb:a3f1)00:48
noam:O YES00:51
- mtm (QUIT: Ping timeout: 265 seconds) (~textual@47.202.75.129)01:03
+ mtm (~textual@47.202.75.129)01:05
* robin__ -> robin01:11
+ wickedshell (~wickedshe@2601:8c0:800:4baa:db9b:f115:99d5:fda)01:17
- vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50)01:52
- cobra (QUIT: Ping timeout: 265 seconds) (~cobra@user/Cobra)02:20
- aloo_shu (QUIT: Ping timeout: 260 seconds) (~aloo_shu@85.51.17.195)02:37
+ aloo_shu (~aloo_shu@90.166.99.64)02:39
- nsc (QUIT: Ping timeout: 260 seconds) (~nicolas@i5C74DED4.versanet.de)03:06
+ nsc (~nicolas@i5C74DC6E.versanet.de)03:08
- wickedshell (QUIT: Ping timeout: 248 seconds) (~wickedshe@2601:8c0:800:4baa:db9b:f115:99d5:fda)03:08
+ wickedshell (~wickedshe@2601:8c0:800:4baa:444a:1b9b:da9d:99ca)03:10
- wickedshell (QUIT: Read error: Connection reset by peer) (~wickedshe@2601:8c0:800:4baa:444a:1b9b:da9d:99ca)03:20
- mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org)03:24
- paperManu (QUIT: Ping timeout: 252 seconds) (~paperManu@107.159.243.8)04:40
+ wickedshell (~wickedshe@2601:8c0:800:4baa:444a:1b9b:da9d:99ca)05:29
- Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.7.160)05:33
+ Ar|stote|is (~linx@149.210.7.160)05:34
- Ar|stote|is (QUIT: Client Quit) (~linx@149.210.7.160)05:36
+ Ar|stote|is (~linx@149.210.7.160)05:36
- aloo_shu (QUIT: Ping timeout: 264 seconds) (~aloo_shu@90.166.99.64)07:25
+ aloo_shu (~aloo_shu@90.166.98.2)07:28
+ cobra (~cobra@user/Cobra)07:31
+ chomwitt (~chomwitt@2a02:587:7a09:1500:1ac0:4dff:fedb:a3f1)07:42
+ simba (~simba@ppp-185-145-220-148.pool13.filstalnetz.de)08:08
amospallaminute: with the latest keyboard firmware my pocket battery takes more or less the same time to empty.08:48
amospallaProbably the keyboard controller takes less energy, but what eats the battery is the main som.08:50
- aloo_shu (QUIT: Ping timeout: 260 seconds) (~aloo_shu@90.166.98.2)09:16
+ aloo_shu (~aloo_shu@90.166.99.23)09:18
- Ar|stote|is (QUIT: Ping timeout: 252 seconds) (~linx@149.210.7.160)09:26
+ Ar|stote|is (~linx@149.210.18.130)09:31
minuteamospalla: alright09:39
- jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@47-186-65-73.dlls.tx.frontiernet.net)10:04
+ jacobk (~quassel@47-186-65-73.dlls.tx.frontiernet.net)10:04
gsoraany chance of getting my reform 2 at fosdem? :v jkjk10:35
- natalie (QUIT: Quit: quit) (~natalie@user/natalie)10:55
+ natalie (~natalie@user/natalie)10:59
- chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:587:7a09:1500:1ac0:4dff:fedb:a3f1)11:19
BoostisBettergsora: where is fosdem this year?11:19
gsoraguess in the place of last year? not sure, first time there11:20
- aloo_shu (QUIT: Ping timeout: 265 seconds) (~aloo_shu@90.166.99.23)11:20
joschBoostisBetter: fosdem is always in 11:21
joschBrussels11:22
joschwow, that was a super delayed paste...11:22
+ aloo_shu (~aloo_shu@90.166.99.125)11:22
[tj]I love how difficult it is to find where fosdem is on their website11:23
gsorait's implicit :P11:23
gsora"FOSDEM takes place at the ULB Solbosch Campus, Brussels, Belgium, Europe, Earth." 11:24
gsorait's on their website 11:24
joschthey forgot to say where Earth is11:24
vkoskivand which multiverse11:24
[tj]but it isn't on about or practical11:24
[tj]but below the fold on the home page11:25
[tj]https://usercontent.irccloud-cdn.com/file/1Xuwm06z/image.png11:25
[tj]which doesn't look like it scrolls for me11:25
gsorai see your point11:26
+ mjw (~mjw@gnu.wildebeest.org)11:35
- aloo_shu (QUIT: Ping timeout: 264 seconds) (~aloo_shu@90.166.99.125)11:54
+ aloo_shu (~aloo_shu@85.51.17.16)11:56
- mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org)12:12
* Guest5032 -> mjw12:19
minutestill not sure if i wanna go to fosdem12:26
vkoskivI'll likely be at Revision again this year, they announced the dates: 18.-21.4.12:27
minutenice12:28
gsoratoo much chaos?12:31
BoostisBetterThe issue for me is that Brussels is my least favorite city to go to. Traffic is insane, and just junk is just so overpriced and not very good. 12:32
+ paperManu (~paperManu@107.159.243.8)12:53
+ gustav28 (~gustav@c-78-82-52-175.bbcust.telenor.se)13:02
- mtm (QUIT: Ping timeout: 276 seconds) (~textual@47.202.75.129)13:03
+ mtm (~textual@47.202.75.129)13:06
- aloo_shu (QUIT: Ping timeout: 252 seconds) (~aloo_shu@85.51.17.16)13:43
+ aloo_shu (~aloo_shu@90.166.98.214)13:45
BoostisBetterdoes anyone here know of a good UPS that has user replacable batteries that you can get in Germany? I need to replace an UPS. 13:49
BoostisBetterbut I hate how they all are note user replacable .13:49
BoostisBetterbut I hate how they all are note user replacable.13:49
chapc batteries were never a problem14:18
BoostisBetterthe models I got just weren't made with replacable batteries in mind14:31
BoostisBetterI have two here and no idea how to get replacable batteries for them14:32
- aloo_shu (QUIT: Ping timeout: 244 seconds) (~aloo_shu@90.166.98.214)14:33
+ chomwitt (~chomwitt@2a02:587:7a09:1500:1ac0:4dff:fedb:a3f1)14:35
+ aloo_shu (~aloo_shu@85.51.17.83)14:36
BoostisBetterminute: was there a newer version of your keyboard firmware already? I ask because the one I am using is getting worse and worse with disconnects. Now the keyboards OLED is working but it just seems like it is not connected to the mainboard. 14:54
BoostisBetterReseting the keyboard around 5 times got it back.14:54
BoostisBetterIs there a recommended time one should hold down the reset button on the keyboard RP2040?14:55
chUPS, please make a decision on when you intend to deliver15:01
ch'wednesday', 'no, monday!', 'no, friday afternoon!', 'no, friday morning!', 'no, friday by end of day'15:02
joschi have so many "funny" UPS stories with MNT shipments...15:05
chmostly i'm waiting for it to say 'delivered to shop'15:05
+ bkeys (~Thunderbi@38-146-94-247.echocast.zone)15:12
minutemaybe they can fix all that with AI15:56
cwebberalso fyi minute, you'll probably see it, but looks like I'm not the only one who ran into the "purple flash" issue on the pocket https://community.mnt.re/t/purple-flash-and-then-screen-doesnt-work-while-booting/2914/516:03
cwebberso hopefully debugging mine will help uncover a general issue that can be resolved16:03
cwebberit's off and on its way as of yesterday morning btw!16:03
minutecwebber: ahh awesome16:10
minuteBoostisBetter: sounds weird. is that still with psuspend or independent? also i don't remember which version you were using. can you tell me the versions displayed on System Status OLED screen?16:14
BoostisBetterminute: sure one second16:16
minuteBoostisBetter: there is one more thing. i recently discovered we never did a usb hub reset and i added one to the imx8mp uboot now. so you could also update your uboot to this one when it's done building: https://source.mnt.re/reform/reform-imx8mp-uboot/-/jobs/732716:17
BoostisBetterminute: PREFHID2024041616:17
BoostisBetterminute: but I am pretty sure this is the same version as initial. The dates on the firmware didn't seem to update16:17
minuteBoostisBetter: ah, so it's before i cleaned the version display up16:18
minuteBoostisBetter: then please update to this: https://source.mnt.re/reform/pocket-reform/-/jobs/7316/artifacts/browse16:18
BoostisBetterminute: vielen Dank!16:19
chdelivered to shop!16:24
minuteaudio is wrong on rk3588 pocket!16:30
minuteand nobody noticed it yet?16:31
minutethere is some bit error16:31
minuteif you listen to a sweep test there's lots of aliasing16:31
chi always see tons of errors in dmesg that seem audio related16:31
minuteyeah those i think are ok16:31
chbut i dont really use audio :>16:31
minuteaudio sweep sounds perfect on rk3588 classic reform 3.016:33
minuteso it's a problem with rk3588 <-> tlv audio chip16:33
minutewhy oh why did they discontinue the wm8960...16:33
chthanks for 'ci: add pocket-hid.cab'16:34
minuteha16:35
minuteyou're more than welcome!16:35
minutei guess soon we could roll out the first firmwares via the real fwupd?16:35
BoostisBetterminute: should the update of the firmware be near instant? 16:37
chi hope so! upstream fixed the bug i reported when updating from a real fwupd on scsi-free hw (like reform)16:37
ch(so we should a bit for them to stamp a release and it to land in debian, but we can also post a 'be sure' workaround for now)16:38
joschch: what's the problem with scsi-free hw?16:42
joschch: we can also apply a patch to fwupd and distribute a patched version in the MNT repos very easily16:43
- aloo_shu (QUIT: Ping timeout: 248 seconds) (~aloo_shu@85.51.17.83)16:43
BoostisBetterminute: was finally able to get the newer kb firmware installed. It was being a real pain for some reason. 16:57
BoostisBetterminute: firmware now reads: PREF1HIDs2ab0d3a16:58
+ aloo_shu (~aloo_shu@85.51.17.83)17:02
- aloo_shu (QUIT: Ping timeout: 244 seconds) (~aloo_shu@85.51.17.83)17:06
- amospalla (QUIT: Quit: WeeChat 4.4.4) (~jordi@user/amospalla)17:07
+ aloo_shu (~aloo_shu@90.166.193.236)17:09
minuteBoostisBetter: nice17:13
[tj]crowdsupply now has a banner saying "Due to severe weather, order processing and shipping will be delayed."17:18
+ amospalla (~jordi@user/amospalla)17:19
ericsfragahey hum. be patient, he says to himself...17:20
chjosch: without sd_mod loaded when fwupd.service starts, it will reset the rp2040 into bootrom mode, but then cannot talk to it - https://github.com/fwupd/fwupd/issues/826517:29
chjosch: so i guess adding a fwupd.service dropin file in reform-tools could also work17:30
josch"I tried sticking strace into the fwupd.service unit, which apparently causes it to succeed." wow, classic18:15
chyeah that was annoying to track down18:15
joschoh wow18:16
joschi understand now what's happening18:16
joschthe usb device is the first scsi device18:16
joschch: but where would reform-tools have to install fwupd.service to override it? I'd like to avoid installing anything into /etc18:18
chjosch: no, just the dropin file should be fine18:25
chimo it was a bad idea to use scsi over usb for disks18:26
chi understand it can be useful but eh18:26
joschch: i never used the systemd dropin mechanism. I'd have to put the Wants and After lines in the [Unit] section into a /lib/systemd/system/fwupd.d/pull8270.conf and that's it?18:29
chlet me check18:30
joschand fwupd is not cross-buildable so my approach to patch it is no that trivial: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/7818:30
minuteryukazou: about your upgrade kit order inkl sinbon adapter, this adapter is confirmed to ship to us on jan 20th so the order is waiting for that, wifi card (here next week) and internal dsi cable (should be here any day, maybe disrupted by US weather issues)18:32
chjosch: this works https://paste.debian.net/hidden/8041c03c/ 18:32
chjosch: note the exact path18:32
chjosch: well, s/override/pull8270/18:33
joschah it's fwupd.service.d18:33
joschthank you, lets put this into the next reform-tools release18:33
joschi wanted to make one anyways so that minute can try out reform-emmc-bootstrap18:34
chnice18:34
+ mark_ (~mjw@gnu.wildebeest.org)18:53
- aloo_shu (QUIT: Ping timeout: 246 seconds) (~aloo_shu@90.166.193.236)18:55
+ aloo_shu (~aloo_shu@90.166.99.83)18:58
ryukazouminute: Thanks for the update. Very anticipate for the rk3588 upgrade, it should made pocket reform more daily drivable.18:59
minuteok i finally fixed the audio quality issue on rk3588 pocket reform and reform next19:00
joscheasy/hard fix?19:01
minutevery easy, just needed to remove 1 line in dts19:05
minutejosch: i'm inclined to merge this to test with motherboard 3.019:10
minutejosch: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/6/diffs19:10
minutejosch: ah i'm silly, i would need to add this to the rk3588 version of the devicetree to actually test this :D19:12
joschminute: you probably mean this MR instead? https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/79 But there seem to be a whitespace only change? And I guess the change to rk3588-reform2 is still not in19:24
amospallajosch: my pocket is now running on NVME and I have my old system backed up. Let's try your script.19:27
+ Savasten (~kvirc@64.39.201.193)19:27
minutejosch: ah yeah i messed that up, thx19:29
- aloo_shu (QUIT: Ping timeout: 248 seconds) (~aloo_shu@90.166.99.83)19:29
amospallawhere is it located? Should I break something, like deleting the partition eMMC table?19:29
joschamospalla: do you want to wait until the next reform-tools version built?19:29
joschamospalla: you don't need to break anything -- it will overwrite your emmc19:30
joschone second, i'm preparing the next release right now19:30
amospallaI'm ok with whatever you prefer, it is your script, also, I'm on Debian/Stable, will it arrive to it too?19:30
minuteuh oh https://source.mnt.re/reform/reform-debian-packages/-/jobs/7344#L187419:31
minutemore rk3588 merged eh19:32
+ aloo_shu (~aloo_shu@85.51.18.102)19:32
joschminute: i have that fixed -- let me push that to main19:32
minuteoh nice19:32
minutethx19:32
joschminute: done. Upstream applied 0063-phy-rockchip-samsung-hdptx-Set-drvdata-before-enabli.patch in 9d23e48654620fdccfcc74cc2cef04eaf7353d07 that's why the patch fails19:34
joschamospalla: please test reform-tools 1.65 from this pipeline: https://source.mnt.re/reform/reform-tools/-/jobs/735919:35
josch(once it succeeds)19:35
amospallajosch: will it install into Debian/Stable?19:36
joschmaybe :)19:36
joschif it does not, please tell me19:36
joschi can rebuild it for you for stable as well19:37
amospallaAnyway, I'm only needing a script of it right?19:37
joschyes, you can also grab the sbin/reform-emmc-bootstrap script from the MR19:38
amospallaIt has no help, it runs straight.19:41
joschexactly19:42
joschthere is also no man page yet19:42
joschand it needs a "this is wiping your emmc, you you really want to proceed?" message19:47
joschthere are fixme comments about that in the script19:47
amospallajosch: the script is mounting current live root volume into a tmp location, and chrooting into it.19:47
josch"live root volume"?19:47
joschit should be mounting a rootfs from nvme19:48
amospallaI'm running my pocket from nvme. I run the script, and it mounts my current root (on nvme): `mount /dev/nvme0n1p1 /tmp/reform-emmc-bootstrap.kpbaqRfqg3`19:49
joschoh :D19:49
joschbut you cannot run it from nvme19:49
joschhow did you boot that?19:49
joschthis script is for when your emmc is borked and you cannot boot anyomre19:49
joschthen you put in a rescue sd-card and run that script19:49
joschbut good point, i need to guard against this19:49
amospallawell, you told me that the tool only recreated /boot, and was designed to people using a nvme19:49
amospallaoh, I see19:50
amospallathen you didn't say that to me, but I imagined it hehe19:50
joschsorry XD19:50
amospallano, sorry me19:50
joschindeed i should've added that --help output19:50
joschi'd say you should flash a rescue sd-card, then run reform-setup-encrypted-nvme to install to nvme19:51
amospallaI put an nvme and migrated. It has been a good experience, now I understand a bit better how it boots.19:51
joschonce everything works, pop in the rescue sd-card again and run the script19:51
amospallabut I already migrated into nvme, what do you mean?19:52
joschokay, good19:52
joschdoes it still boot or did the script damage something?19:52
amospallaoh yes, I see, now I can try your tool from sd-card.19:52
joschif it damaged your /boot on emmc, then good i guess :D19:52
amospallahahaha. I have only installed a nvme disk and migrated by `reform-migrate --emmc /dev/nvme...`, and here I am running from the nvme instead of the emmc.19:53
amospallaNow I'll try to use your tool from a sd-card.19:54
amospallaYou meant, the tool is meant to be used using nvme (that is why it mounts my nvme) when there is no /boot, so forcibily if you are using this tool you are booting from an sd-card.19:57
amospallanow I get it19:57
amospallahhihi19:57
- Savasten (QUIT: Ping timeout: 260 seconds) (~kvirc@64.39.201.193)19:57
joschamospalla: let me give you more context19:58
joschhttps://community.mnt.re/t/migrating-encrypted-nvme-to-different-som-imx8mq-to-a311d/178319:58
joschhttps://community.mnt.re/t/migration-from-i-mx8mplus-to-a311d/231719:58
joschhttps://source.mnt.re/reform/reform-system-image/-/wikis/Migration19:59
joschso this is for people who "loose" their /boot on emmc because they switched their SoM19:59
joschinstead of them having to manually fiddle with commands, lets have a tool that automates it19:59
+ wielaard (~mjw@gnu.wildebeest.org)20:01
amospallaI understand.20:01
- mark_ (QUIT: Ping timeout: 248 seconds) (~mjw@gnu.wildebeest.org)20:03
* mjw -> Guest818420:12
- Guest8184 (QUIT: Killed (osmium.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)20:12
* wielaard -> mjw20:12
+ Guest8184 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)20:12
+ Savasten (~kvirc@64.39.201.193)20:16
joschi think i want to support the case where the tool is called with the OS running from nvme20:17
joschuse-case: the user booted the system with /boot from sd-card and wants to move their /boot partition to emmc20:17
joschamospalla: i'm now afk for ~1 hour20:18
amospallapok20:18
amospallaok trying your tool now20:19
amospallawill post20:19
amospallawas updating my old sd-card and solving dependencies, and each run made initramfs to rebuild for a loong time hihi20:19
amospallajosch: yay, it booted my nvme again, it works :)20:36
- hairu (QUIT: Remote host closed the connection) (m-uotkmd@user/hairu)20:37
+ hairu (m-uotkmd@user/hairu)20:38
- Savasten (QUIT: Ping timeout: 252 seconds) (~kvirc@64.39.201.193)20:49
+ Savasten (~kvirc@64.39.201.193)20:54
- Savasten (QUIT: Client Quit) (~kvirc@64.39.201.193)20:56
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)21:50
- Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66)22:08
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)22:09
minutedamn > [  684.808645] dma-pl330 fea10000.dma-controller: pl330_update:1678 Unexpected!22:10
minute(with latest kernel from today on rk3588, after testing a bunch auf audio. audio stopped working at that moment)22:10
- gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-52-175.bbcust.telenor.se)22:15
- aloo_shu (QUIT: Ping timeout: 260 seconds) (~aloo_shu@85.51.18.102)22:17
minuteok that's gone after a reboot and now audio is perfect22:18
+ aloo_shu (~aloo_shu@90.166.98.45)22:19
BoostisBetterminute: that is a relief! Aweosme to hear! Also nice update on the Reform Next update! 22:23
minutethx! now finally some time for some relaxing22:27
BoostisBetterminute: most excellent!22:31
joschamospalla: nice!! \o/22:36
amospallathank you :)22:37
minutejosch: thanks for making this tool btw, will test next week!22:38
amospallahaving this tool makes me feel more comfortable in case I break something :) :)22:38
joschamospalla: okay, then next step is making everything nice, writing help text, man page etc22:38
joschminute: sure thing! It bugged me that this was still so manual. If you have any suggestion from what you read here already, i'm all ears. :)22:39
amospallaif I can suggest something, using strings like $(strings /proc/device-tree/model) instead of using cat makes bash not complain about null strings. It doesn't affect the behaviour because it just ignores them, but avoids bash complaining on stderr.22:40
joschminute: oh in your recent message to fedi you suggest to "sudo apt install linux-image-6.12.9-mnt-reform-arm64" -- this is not the best suggestion because this will tell apt to mark the kernel as manually installed, so now apt will never suggest to auto-remove it22:41
minutejosch: ok, what's the right command to upgrade only the kernel package? i'll edit22:41
minuteahhh linux-image-mnt-reform-arm6422:42
minuteyes?22:42
minutenext time i'll remember22:42
joschprobably: sudo apt --only-upgrade install linux-image-mnt-reform-arm6422:42
joschyes22:42
joschupgrading the meta package will pull in the new real package if necessary22:43
minutewhy the --only-upgrade?22:43
joschhrm... actually...22:44
joschsorry, i'm confirming...22:44
joschsorry, without --only-upgrade22:44
joschyour post right now is correct i think22:44
joschhave a nice weekend _o/22:44
minuteok awesome22:45
minutethanks josch, you too!22:45
- colinsane (QUIT: Quit: bye) (~colinunin@97-113-95-212.tukw.qwest.net)22:47
BoostisBetteramospalla: I see that you added a dev/input change variable in the psuspend.conf, but how do I determine what dev/input/event I need to use? Using evtest I see this: /dev/input/event0:MNT Pocket Reform Input 1.0 Keyboard22:56
BoostisBetteramospalla: so how would I use that in the psuspend.conf to get the keyboard picked up again?22:56
amospallaBoostisBetter: I use a regular expression now to look for the keyboard, so it should work on any past and probably future string.22:58
BoostisBetteramospalla22:58
BoostisBetteramospalla: what do you mean?22:58
amospallajust upgrade and restart psuspend22:58
amospallait will work22:58
BoostisBetteramospalla: roger! Gracias!;22:58
joschamospalla: how do i see the problem where bash complains about something? The script is not written using bash but dash23:00
amospallajosch: ohhh hehehe sorry, yeah, my fault, I ran it manually with bash -x to have a verbose output 23:02
amospallayes it uses /bin/sh23:03
+ colinsane (~colinunin@97-113-95-212.tukw.qwest.net)23:03
amospallaI use bash for my scripts and I mechanically do bash things ^^23:04
- colinsane (QUIT: Ping timeout: 248 seconds) (~colinunin@97-113-95-212.tukw.qwest.net)23:10

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