2026-08-26.log

minuteok, the video ports (vps) don't directly have something to do with my problem. it's more likely the 4 "cluster" assignments (those are like layers)00:01
+ f_underscore (~f_undersc@189.113.240.53)00:06
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)00:17
- lidstah (QUIT: Ping timeout: 265 seconds) (~lidstah@gateway/tor-sasl/lidstah)00:22
+ lidstah (~lidstah@gateway/tor-sasl/lidstah)00:23
- Esi (QUIT: Server closed connection) (sid14426@helmsley.irccloud.com)00:29
+ Esi (sid14426@id-14426.helmsley.irccloud.com)00:30
- f_underscore (QUIT: Ping timeout: 243 seconds) (~f_undersc@189.113.240.53)00:35
+ f_underscore (~f_undersc@189.113.240.53)00:37
minuteyep. the 3 outputs (dsi, hdmi, edp->usb-c) work in arbitrary plugging order when i patch the driver so that it avoids cluster1-3 and uses cluster0 for hdmi and esmart0+1 for the rest... so the cluster setup code in the vop2 is totally borked somehow00:52
- lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah)01:01
+ robin (~robin@user/terpri)01:03
+ lidstah (~lidstah@gateway/tor-sasl/lidstah)01:05
+ bkeys (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)01:06
- f_underscore (QUIT: Ping timeout: 270 seconds) (~f_undersc@189.113.240.53)01:17
- mjw (QUIT: Ping timeout: 249 seconds) (~mjw@gnu.wildebeest.org)01:21
+ simba (~simba@static-151-2-60-95.ipcom.comunitel.net)01:27
- bkeys (QUIT: Ping timeout: 264 seconds) (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)01:31
- simba (QUIT: Ping timeout: 259 seconds) (~simba@static-151-2-60-95.ipcom.comunitel.net)01:31
- onepict (QUIT: Server closed connection) (~quassel@user/tisiphone)01:33
+ tisiphone (~quassel@ganymede-0.gladserv.net)01:33
- pasta161 (QUIT: Ping timeout: 262 seconds) (~pasta161@user/xha)01:39
+ pasta161 (~pasta161@user/xha)01:56
- a3f (QUIT: Read error: Connection reset by peer) (~a.fatoum@flummi.grey.stw.pengutronix.de)02:12
+ a3f (~a.fatoum@flummi.grey.stw.pengutronix.de)02:14
- vagrantc (QUIT: Ping timeout: 274 seconds) (~vagrant@2600:3c01:e000:21:7:77:0:50)02:21
- SavagePeanut (QUIT: Server closed connection) (59eaa45ac7@irc.cheogram.com)02:21
+ SavagePeanut (59eaa45ac7@irc.cheogram.com)02:22
- yuu (QUIT: Server closed connection) (sid267332@id-267332.ilkley.irccloud.com)02:29
+ yuu (sid267332@id-267332.ilkley.irccloud.com)02:29
- AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-169-191-86.subs.proxad.net)02:31
- svp (QUIT: Server closed connection) (~svp@host-79-57-105-252.business.telecomitalia.it)02:36
+ svp (~svp@host-79-57-105-252.business.telecomitalia.it)02:47
minutethis build works for me to get 3 displays, with working usb-c dp + hdmi hotplug, on pocket reform with mb2.0 https://source.mnt.re/reform/reform-debian-packages/-/jobs/2401302:56
- paperManu_ (QUIT: Ping timeout: 257 seconds) (~paperManu@173.206.254.242)03:09
+ bkeys (~Thunderbi@71.31.254.246)03:24
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@71.31.254.246)03:37
+ bkeys1 (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)03:38
* bkeys1 -> bkeys03:40
- paperManu (QUIT: Ping timeout: 245 seconds) (~paperManu@173.206.254.242)03:41
- laumann (QUIT: Server closed connection) (31d44fbb95@gentoo/developer/laumann)03:59
+ laumann (31d44fbb95@gentoo/developer/laumann)03:59
- lanodan (QUIT: Server closed connection) (~lanodan@2a01:e0a:11f3:9920::35)04:41
+ lanodan (~lanodan@2a01:e0a:11f3:9920::35)04:41
+ robin_ (~robin@user/terpri)04:54
- robin (QUIT: Ping timeout: 271 seconds) (~robin@user/terpri)04:58
- bkeys (QUIT: Ping timeout: 248 seconds) (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)05:02
+ manis (01a66df340@185.72.67.185)06:10
- wakest_ (QUIT: Server closed connection) (b9de55a967@2001:bc8:1210:2cd8::6b5)06:11
+ wakest_ (b9de55a967@2001:bc8:1210:2cd8::6b5)06:11
+ simba (~simba@static-151-2-60-95.ipcom.comunitel.net)06:32
- simba (QUIT: Ping timeout: 253 seconds) (~simba@static-151-2-60-95.ipcom.comunitel.net)06:36
- jameshjacks0njr_ (QUIT: Server closed connection) (sid381910@user/jameshjacks0njr/x-7942062)07:00
+ jameshjacks0njr_ (sid381910@user/jameshjacks0njr/x-7942062)07:00
- lidstah (QUIT: Ping timeout: 259 seconds) (~lidstah@gateway/tor-sasl/lidstah)07:36
joschcrazy o007:37
+ lidstah (~lidstah@gateway/tor-sasl/lidstah)07:38
- lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah)07:38
+ lidstah (~lidstah@gateway/tor-sasl/lidstah)07:38
aloo_shu?08:06
joschi mean: given the struggle minute had to go to yesterday, they are probably the first with that setup, wow08:12
- jedavies (QUIT: Server closed connection) (~jedavies@plan9.info)08:43
+ jedavies (~jedavies@plan9.info)08:43
MartiniMoelol, 161 people in this channel09:02
- josch (QUIT: Server closed connection) (~josch@mister-muffin.de)09:13
+ josch (~josch@mister-muffin.de)09:14
* ChanServ changed mode (+o, josch)09:14
+ simba (~simba@static-151-2-60-95.ipcom.comunitel.net)09:37
- simba (QUIT: Ping timeout: 260 seconds) (~simba@static-151-2-60-95.ipcom.comunitel.net)09:45
- whereiseveryone (QUIT: Server closed connection) (206ba86c98@2a03:6000:1812:100::2e4)09:51
+ whereiseveryone (206ba86c98@2a03:6000:1812:100::2e4)09:51
rick_woaaawww to the 3 diplay setup <309:53
- tisiphone (QUIT: Changing host) (~quassel@ganymede-0.gladserv.net)10:04
+ tisiphone (~quassel@user/tisiphone)10:04
a3fstephano, jfyi, barebox v2026.08.0 now also properly supports NVMEs with 4K sectors.10:16
+ simba (~simba@static-151-2-60-95.ipcom.comunitel.net)10:17
- simba (QUIT: Ping timeout: 244 seconds) (~simba@static-151-2-60-95.ipcom.comunitel.net)10:21
termosi see people mention some cellular modems in that pocket reform in the wild thread, which ones are they using? i can't tell10:23
- buckket (QUIT: Server closed connection) (~buckket@vps.buckket.org)10:30
+ buckket (~buckket@vps.buckket.org)10:30
+ mjw (~mjw@gnu.wildebeest.org)10:34
- loon (QUIT: Server closed connection) (~luna@user/loon)10:34
+ loon (~luna@user/loon)10:34
- dodo (QUIT: Server closed connection) (~dodo@user/dodo)10:56
+ dodo (~dodo@user/dodo)10:56
+ simba (~simba@static-151-2-60-95.ipcom.comunitel.net)11:16
- simba (QUIT: Ping timeout: 256 seconds) (~simba@static-151-2-60-95.ipcom.comunitel.net)11:21
+ f_underscore (~f_undersc@189.113.241.238)11:42
- lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah)12:08
+ lidstah (~lidstah@gateway/tor-sasl/lidstah)12:09
- lidstah (QUIT: Remote host closed the connection) (~lidstah@gateway/tor-sasl/lidstah)12:22
+ lidstah (~lidstah@gateway/tor-sasl/lidstah)12:26
- f_underscore (QUIT: Ping timeout: 266 seconds) (~f_undersc@189.113.241.238)12:27
+ paperManu (~paperManu@173.206.254.242)12:57
+ f_underscore (~f_undersc@189.113.241.238)12:58
- f_underscore (QUIT: Ping timeout: 255 seconds) (~f_undersc@189.113.241.238)13:05
minutei tested my new kernel build on classic reform w/rk3588, everything appears to work well with the reshuffled vop2 planes... gonna release later today, need to remove the devmem patch from the stack13:44
minutemeanwhile my pocket's usb hub perma-crashed or sth13:44
+ paperManu_ (~paperManu@173.206.254.242)14:07
minutenow it has recovered, after leaving it off for a while14:09
- mjw (QUIT: Ping timeout: 243 seconds) (~mjw@gnu.wildebeest.org)14:22
joschtermos: quectel em06-e14:32
joschminute: you are really something -- buying the new pocket reform motherboard is still on my wishlist. I will have more money again next month. XD14:33
* Guest9924 -> mjw14:34
joschoh MNT August 2026 Newsletter is out!14:37
- tretinha (QUIT: Server closed connection) (3a571d9f43@2a03:6000:1812:100::1151)14:51
+ tretinha (3a571d9f43@2a03:6000:1812:100::1151)14:51
minuteyeah a rare newsletter15:12
minutejosch: hehe. yes, i'm happy that dp alt-mode is approaching usability. the signal quality over a plain usbc cable isn't good enough yet for my dell monitor @ 1440p though. i'll probably have to ship a redriver flex cable at some point. but with a usb-c -> displayport cable it works better15:14
minutebut also, i'm using only 2 lanes, will try with 4 lanes also... that should probably reduce the required bandwidth15:14
minutei could also lock it to 1080p for now so that people won't run into black or glitchy screens. hmm, maybe this could be a devicetree setting (max pixelclock on eDP), so people could experiment with that without rebuilding the kernel.15:16
minutei already introduced a devicetree setting for the number of lanes15:16
+ bkeys (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)15:34
jfredI'm very much looking forward to dp alt-mode :)15:51
f_me too :>15:52
jfredI've been docking my pocket with a USB-C dock and the microhdmi port for display out, but the dock itself does display out too. Single cable docking will be very nice15:53
jfredThough I have... perhaps too many things hanging off that USB-C hub, haha. I've had occasional USB dropouts, not sure yet if it's more related to the hub or the pocket15:54
jfredEr, USB hub, not USB-C. USB-C dock -> 7-port hub ^^;15:55
+ kfx (~kfx@grendel.sciops.net)16:03
kfxHello, currently we make a kernel package for Alpine Linux by importing the Reform patches and using the normal aports build system16:04
minutejfred: which make of dock?16:04
minutekfx: nice!16:04
kfxOnce merge request 195 lands, how do we extract the patches into regular patch files again?  I do not have a phd in git and I can't tell what's going on in there16:04
jfredminute: Lenovo. I believe it's a ThinkPad USB-C dock gen 216:05
minutejfred: thx! looks like i gotta collect some docks, and/or think about a good way of collecting PD logs from people16:08
minutekfx: i think josch can answer that. but git has some commands for converting arbitrary commit ranges to patch stacks.16:09
f_kfx: which reform som?16:11
kfxrk358816:11
f_ah ok16:11
jfredminute: I've got a decent collection of docks and would be glad to collect logs if you figure out something that'll get you useful info16:12
jfredincluding some monitors at the office with built-in docks that I know can be fiddly even with our company machines, haha16:13
minutejfred: normally the output of "tio /dev/ttyACM0" from the point of plugging it in until the output has somewhat settled16:28
minutejfred: one text file per model, with modelname as the filename would be mega awesome :D16:29
minutejfred: cherry on top would be `sudo dmesg -w` output from the point of plugging in... then we can see the kernel side of thigns16:29
jfredminute: I'll give it a go tonight for the docks I can find nearby :)16:31
minutejfred: actually... do you have mb2.0 yet? :D because otherwise it makes only limited sense :016:31
jfredoh - no, not yet haha16:31
jfred(I have an order in for it but I'd guess it's waiting on something else in the order)16:33
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)16:34
+ bkeys (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)16:34
jfredoh, or maybe it is the next mainboard batch I'm waiting for16:36
minuteyep16:41
+ simba (~simba@static-151-2-60-95.ipcom.comunitel.net)16:52
- mesaoptimizer0 (QUIT: Server closed connection) (~user@user/PapuaHardyNet)16:54
+ mesaoptimizer0 (~user@user/PapuaHardyNet)16:55
- simba (QUIT: Ping timeout: 247 seconds) (~simba@static-151-2-60-95.ipcom.comunitel.net)16:56
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)17:06
+ bkeys1 (~Thunderbi@h246.254.31.71.dynamic.ip.windstream.net)17:06
* bkeys1 -> bkeys17:08
- dook (QUIT: Server closed connection) (c9db4a6425@2a03:6000:1812:100::1222)17:19
+ dook (c9db4a6425@2a03:6000:1812:100::1222)17:19
- glob (QUIT: Server closed connection) (sid272774@user/gl0b)17:29
+ glob (sid272774@user/gl0b)17:30
+ jogu (~jogu@user/jogu)17:30
- Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66)17:46
+ wielaard (~mjw@gnu.wildebeest.org)17:56
- Asmadeus (QUIT: Server closed connection) (~asmadeus@user/meow/Asmadeus)18:05
+ Asmadeus (~asmadeus@user/meow/Asmadeus)18:06
- theesm (QUIT: Server closed connection) (~theesm@guixtheplanet.de)18:12
+ theesm (~theesm@guixtheplanet.de)18:12
stephanoa3f: I saw, thank you for all that work! I've been following that since June's patch set. Getting Barebox running on the MNT platforms makes for some very fun, hackable hardware. :)18:13
+ AnimaInvicta (~AnimaInvi@88-169-191-86.subs.proxad.net)18:19
- AnimaInvicta (QUIT: Read error: Connection reset by peer) (~AnimaInvi@88-169-191-86.subs.proxad.net)18:25
lidstahminute: congrats for the 3 displays working! seems to have been a lot of work from what I read in the backlog :)18:25
minutelidstah: yeah very complicated19:11
minutei did some hacking and got the club3d csv-2553 hub working as a pd power supply, hdmi output and usb-hub/card-reader/ethernet thing19:12
minuteit's a bit of a lottery though because if the connection starts with pocket in power SRC mode, it doesn't work all the way because something goes wrong when the dock wants to do a power role swap19:13
minutebut, getting there19:13
- chorc (QUIT: Ping timeout: 259 seconds) (~chorc@user/chorc)19:18
+ chorc (~chorc@user/chorc)19:18
+ Guest299528 (~ndufresne@transit.collaboradmins.com)19:34
- Guest29952 (QUIT: Ping timeout: 265 seconds) (~ndufresne@transit.collaboradmins.com)19:36
* Guest299528 -> Guest2995219:36
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)19:39
+ tomenzgg (~tomenzgg@24.148.84.215)19:44
- tomenzgg (QUIT: Changing host) (~tomenzgg@24.148.84.215)19:44
+ tomenzgg (~tomenzgg@user/tomenzgg)19:44
- arjan (QUIT: Server closed connection) (~arjan@user/arjan)19:52
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)20:27
- Gooberpatrol66 (QUIT: Ping timeout: 246 seconds) (~Gooberpat@user/gooberpatrol66)20:47
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)20:49
chorclatest kernel update enables DP-1 output for me even though I dont have it connected; disabled it manually in wlrandr20:52
chorctermos: I have quectel em060k-gl20:54
chorcI wasn't happy about reception I had with it and my antennas, but it mightve been vodafone; I've switched to o2 today and it works much better20:57
- mjw (QUIT: Killed (uranium.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:4600:5952:3a9:6e0e:555a)20:58
* wielaard -> mjw20:58
+ Guest9510 (~mjw@2001:1c06:2486:4600:5952:3a9:6e0e:555a)20:59
joschkfx: the answer to your question is in MR 195 which you are referring to: https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/195/diffs?file_path=linux%2Fbuild.sh#line_788c7af95_A65921:06
joschkfx: ouch, i was told by other distro maintainers that they disliked us shipping a bunch of patch files and preferred a full linux git with our commits on top instead21:07
joschstephano: was it arch which preferred it that way?21:07
joschor was it guix (ping vagrantc )21:09
joschi should keep some notes about who is on which OS...21:09
vagrantcjosch: well ... heh.21:13
vagrantci use both approaches depending on what i'm doing ... i have something working with the patches-as-files now21:13
vagrantcbut anytime i try to rebase against a new upstream version, i want to have them in git21:14
vagrantcmight be able to just directly use git repositories though21:14
vagrantcthe linux(-libre) kernel workflow in guix is ... an adventure21:14
joschvagrantc: you are in luck because we are switching to git now :)21:14
vagrantceither luck, or hairpulling, or both :)21:15
joschvagrantc: you saw the debian kernel packaging? ;)21:15
vagrantcheh.21:15
joschvagrantc: it took me an afternoon with ben hutchings in berlin to get things into their current state :)21:16
vagrantcjosch: nice!21:16
joschyeah, that was cool :)21:16
kfxjosch: the current way we can just copy the patch files.  I'm not clear on this; it looks like we have to keep a whole linux source tree in git to run this?21:16
joschkfx: "we" being...?21:16
vagrantcjosch: the tricky part with guix is applying all the linux-libre machinery ... some sed and awk scripts and python ... which i am not sure how to apply to a git repository (it currently uses upstream tarballs)21:17
kfxagain, I apologize for not understanding advanced git flagellation; I don't understand all of the interrelations between tips and branches and tags and etc21:17
joschkfx: welcome to the club -- i haven't done these kind of git gymnastics before MR 195 either :D21:18
kfxjosch: currently in aports we just pull down a git tarball and apply patch files, so "we" is the people maintaining the alpine kernel package21:18
joschand soon you have to get the "patch files" from somewhere else because things are in git now?21:18
kfxright, figuring out how to do that is where I am now21:19
vagrantcjosch: my main request would be to keep old branches alive indefinitely, if at all possible, although softwareheritage.org generally does a good job of archiving as a fallback21:19
vagrantcjosch: worst case, I can probably generate the patch files from git and keep doing it the clumsy way i am doing now :)21:19
joschkfx: the debian linux packaging requires patch files as well. So i recommend you take a closer look at mr 195 because in there are the command i propose to use to turn the linux git into patch files debian needs.21:20
jfredI have been using https://stacked-git.github.io/ recently for managing a series of local patches on top of an upstream project and have been enjoying it21:20
vagrantcjosch: another approach would be to base the guix kernel package for mnt/reform off of whatever debian packages y'all come up with :)21:21
joschessentially, it's just a "git clone --branch=mnt-vX.Y.Z --depth=116 https://..." and then "git format-patch vX.Y.Z...HEAD"21:21
vagrantc(someone recently submitted a linux-debian package to guix based on the patchs debian applies)21:21
joschjfred: i have considered that but am trying with bare git for now to avoid everyone from having to install and learn yet another tool just to manage the patch stack21:22
jfredyeah, fair enough. it definitely has a learning curve (that I'm not entirely past yet myself)21:22
joschkfx, vagrantc, stephano: you are now in my new list of who is running which OS -- welcome! I now know who to ping in the future. :)21:24
vagrantcin theory i am running debian also, but i botched my /boot partition when i last experimented with upstream u-boot and never got around to fixing it :)21:25
kfxI don't think we have the resources to automate this stuff21:25
kfxprobably just gonna stick with the patches we have and port them forward21:26
joschkfx: theoretically you could also extract the patch files from the debian source package after our own machinery created it from the linux git21:28
joschthey can be found in ./debian/patches/reform21:28
kfxthat might be an easier plan, thanks.  unpacking a deb is more achievable than understanding git21:29
joschkfx: It's a source package, not a deb. So it's just a tarball.21:31
joschFor the latest kernel: curl https://mntre.com/reform-debian-repo/pool/main/l/linux/linux_7.1.10-1%2Breform20260826T130932Z.debian.tar.xz | tar -tJv | grep debian/patches/reform21:31
- paperManu_ (QUIT: Ping timeout: 249 seconds) (~paperManu@173.206.254.242)21:39
minutechorc: oh huh @ dp-1, do you have anything connected?21:45
minutechorc: to the usb-c charging port, i mean?21:45
stephanojosch: for arch, i've been pulling the reform-debian-packages repo and using the patches from there. the main issue is when additional patches are required. then it is either patch-the-patch, or rewrite the patch (a kind of override). Yocto solves the problem of layered patches quite elegently. My solution is hacky at best. Simply pulling a vendor kernel with patches already applied would be21:45
stephanoideal.21:46
chorcminute: nope, nothing connected, not even a charger21:51
minutechorc: ok that's weird, i wonder why it would show as connected... could i get a sudo dmesg?21:52
minutechorc: and it's with sway or...?21:54
chorcminute: it's with sway, but even before wayland boots I can see it has another screen detected (main boot screen does that weird 90 degrees turn, when I'm entering the password for FDE)21:56
+ arjan (~arjan@user/arjan)21:56
minutechorc: ok, do you use a non-standard dtb maybe?21:56
minutehmm no that wouldn't make no sense either... chorc: in any case dmesg would be appreciated, i should see a bunch of debug output for analogix dp21:57
chorcminute: here you go, dmesg https://pastebin.com/pjeYyHe721:58
minutechorc: thank you! lets see21:59
chorcminute: I don't recall changing anything about DTB, this is pretty stock MNT forky distro, just with some apps22:00
minuteok...22:01
minuteah dang22:02
+ paperManu_ (~paperManu@142.169.16.214)22:12
+ simba (~simba@static-151-2-60-95.ipcom.comunitel.net)22:46
- simba (QUIT: Ping timeout: 250 seconds) (~simba@static-151-2-60-95.ipcom.comunitel.net)22:50
- paperManu_ (QUIT: Ping timeout: 243 seconds) (~paperManu@142.169.16.214)23:32
+ simba (~simba@95.60.2.151)23:34
minutechorc: could you give me the output of: cat /sys/kernel/debug/gpio | grep hpd23:51
minuteah yes, when i put mntsc in bootsel mode, i get "hi" on this gpio as well23:52
minuteand i get an unknown DP-1 display then m)))23:53
minutewhy doesn't the pulldown work...23:54
minuteah, can try moar drv_level23:58
minutehehe, up to > pcfg-pull-down-drv-level-1523:59

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