| - GNUmoon (QUIT: Ping timeout: 255 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 00:13 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 00:16 | |
| bkeys | PLEASE NO | 00:34 |
|---|---|---|
| bkeys | Okay? | 00:36 |
| bkeys | Okay no worries | 00:40 |
| minute | bkeys: hm? | 00:46 |
| minute | bkeys: what happened? | 00:47 |
| bkeys | Between the two plugs where the battery boards hook up to the carrier board, it looks like a see through bead of silicone | 00:50 |
| bkeys | I was worried something melted | 00:50 |
| bkeys | But it's booting up fine, no sounds or smells and all batteries report to be there | 00:50 |
| bkeys | It's great, no more worrying about buying another set of batteries | 00:51 |
| - mtm (QUIT: Ping timeout: 272 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 01:03 | |
| - natalie (QUIT: Quit: quit) (~natalie@user/natalie) | 01:30 | |
| + natalie (~natalie@user/natalie) | 01:34 | |
| - mjw (QUIT: Ping timeout: 264 seconds) (~wielaard@gnu.wildebeest.org) | 02:36 | |
| bkeys | Is there a graphical display manager that works on the Reform? | 02:57 |
| eery[m] | gdm works fine (on fedora 37) | 02:59 |
| bkeys | How do you get F37 on here? | 03:00 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 03:09 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 04:59 | |
| + cinap_lenrek (~cinap_len@ns3076381.ip-147-135-136.eu) | 05:00 | |
| cinap_lenrek | hola! | 05:00 |
| cinap_lenrek | mainscreen turn on! :) | 05:00 |
| cinap_lenrek | lcdif working in uboot on the reform | 05:00 |
| cinap_lenrek | is here any magic required to get the usb keyboard working? | 05:07 |
| - klardotsh (QUIT: Ping timeout: 260 seconds) (~klardotsh@2607:fb90:ec1f:d57c:52eb:71ff:fefa:beaf) | 05:19 | |
| + klardotsh (~klardotsh@172.56.104.104) | 05:21 | |
| cinap_lenrek | hm | 05:36 |
| cinap_lenrek | usb tree does not show the keyboard :( | 05:36 |
| + bgs (~bgs@212-85-160-171.dynamic.telemach.net) | 07:05 | |
| - GNUmoon (QUIT: Ping timeout: 255 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 07:12 | |
| cinap_lenrek | usb_find_and_bind_driver: Match found: usb_kbd | 07:23 |
| cinap_lenrek | Missing uclass for driver usb_kbd | 07:23 |
| cinap_lenrek | ib/errno_str.c:ERRNO_MSG(EPFNOSUPPORT, "Protocol family not supported"), | 07:27 |
| cinap_lenrek | include/linux/errno.h:#defineEPFNOSUPPORT96/* Protocol family not supported */ | 07:27 |
| cinap_lenrek | yea | 07:27 |
| cinap_lenrek | seems the usb_kbd.c driver is fucked | 07:27 |
| cinap_lenrek | what a mess | 07:27 |
| josch | cinap_lenrek: but you got uboot to turn the screen on!! do you have a patch? :D | 07:38 |
| cinap_lenrek | here we go! | 07:44 |
| cinap_lenrek | got usb keyboard to work now | 07:45 |
| cinap_lenrek | need to enable uclass drivers | 07:45 |
| cinap_lenrek | josch: i will release patch soon | 07:45 |
| cinap_lenrek | i spend like 2 days trying to figure out uboots crazy mess | 07:45 |
| cinap_lenrek | then gave up and just ported my plan9 lcdif driver over | 07:46 |
| cinap_lenrek | i removed the fbpanel.c shit | 07:46 |
| cinap_lenrek | https://felloff.net/usr/cinap_lenrek/lcdif.c | 07:47 |
| cinap_lenrek | i'm stunned that there is absolutely no infrastructure for like setting timer frequencies | 07:48 |
| - bgs (QUIT: Remote host closed the connection) (~bgs@212-85-160-171.dynamic.telemach.net) | 07:48 | |
| cinap_lenrek | everyone just hammers registers | 07:49 |
| cinap_lenrek | i'll clean it up tomorrow | 08:02 |
| cinap_lenrek | i would like to figure out if i can make the framebuffer address fixed | 08:04 |
| cinap_lenrek | thats makes it very easy for anyone wanting to port some os | 08:04 |
| cinap_lenrek | also the clocks are not programmed correctly yet | 08:05 |
| cinap_lenrek | i'm parsing the edid to get the correct pixel clock | 08:05 |
| cinap_lenrek | but i have to calculate the dividers | 08:06 |
| cinap_lenrek | which obviously in uboot theres no code for this | 08:06 |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 08:14 | |
| - natalie (QUIT: Remote host closed the connection) (~natalie@user/natalie) | 08:21 | |
| Boostisbetter | minute: it feels like the crowd funding for the pocket is imminent. | 08:50 |
| + natalie- (~natalie@user/natalie) | 09:32 | |
| + MajorBiscuit (~MajorBisc@c-001-028-012.client.tudelft.eduvpn.nl) | 09:52 | |
| minute | cinap_lenrek: amazing work! can't wait to test it! | 11:01 |
| minute | cinap_lenrek: also i very much agree with your chosen strategy | 11:01 |
| minute | Boostisbetter: yes, we're working on campaign content | 11:05 |
| + mjw (~wielaard@gnu.wildebeest.org) | 11:22 | |
| - mjw (QUIT: Ping timeout: 268 seconds) (~wielaard@gnu.wildebeest.org) | 12:16 | |
| - klardotsh (QUIT: Ping timeout: 265 seconds) (~klardotsh@172.56.104.104) | 12:17 | |
| Boostisbetter | 👍 | 12:20 |
| + klardotsh (~klardotsh@172.56.105.180) | 12:23 | |
| minute | Boostisbetter: i'm also working on the imx8mplus display bringup. last night it turned on for the first time, but i still have some encoding issue | 12:48 |
| minute | (before i was working with the imx8mq, which has a different DSI IP) | 12:48 |
| * wielaard -> mjw | 13:03 | |
| - mtm (QUIT: Ping timeout: 260 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 13:04 | |
| Boostisbetter | minute: exciting stuff! Do you want to wait for that to be ironed out before launching? Hopefully not. 😊 | 14:01 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 15:09 | |
| + wholesomedonut (~wholesome@76.129.124.232) | 16:30 | |
| - wholesomedonut (QUIT: Quit: Leaving) (~wholesome@76.129.124.232) | 16:34 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 17:06 | |
| - ec0 (QUIT: Ping timeout: 246 seconds) (~ec0@vps-446f4f39.vps.ovh.ca) | 17:08 | |
| * bleb_ -> bleb | 17:19 | |
| + ec0 (~ec0@vps-446f4f39.vps.ovh.ca) | 17:34 | |
| + bgs (~bgs@212-85-160-171.dynamic.telemach.net) | 17:54 | |
| minute | Boostisbetter: yes, but it's almost solved | 18:47 |
| Boostisbetter | minute, wunderbar! | 18:48 |
| - MajorBiscuit (QUIT: Ping timeout: 260 seconds) (~MajorBisc@c-001-028-012.client.tudelft.eduvpn.nl) | 18:53 | |
| cinap_lenrek | minute: https://felloff.net/usr/cinap_lenrek/lcdif.patch | 19:01 |
| cinap_lenrek | minute: i have not tested if this breaks linux boot tho | 19:01 |
| cinap_lenrek | i dont have linux sdcard at hand right now | 19:02 |
| cinap_lenrek | i need to build the system image | 19:02 |
| cinap_lenrek | https://felloff.net/usr/cinap_lenrek/flash.bin | 19:02 |
| cinap_lenrek | you probably want me to clone the uboot in gitlab, no | 19:03 |
| cinap_lenrek | and make a branch and then you can pull it? | 19:03 |
| cinap_lenrek | i'm really bad with git :( | 19:03 |
| sigrid | cinap_lenrek: a small not important thing: CTRL1_CUR_FRAME_DONE_IRQ_EN isn't needed to be set | 19:03 |
| sigrid | since you removed blanking | 19:04 |
| cinap_lenrek | right | 19:04 |
| sigrid | guess that goes for all irq stuff | 19:04 |
| cinap_lenrek | makes sense | 19:04 |
| cinap_lenrek | tho uboot has no irq's enabled anyway | 19:04 |
| sigrid | yep | 19:04 |
| cinap_lenrek | but yeah, should not set that flag :) | 19:04 |
| sigrid | idk if it's worth to add lcdif on/off as commands to uboot | 19:04 |
| cinap_lenrek | sigrid: i dont think so | 19:05 |
| sigrid | but maybe framebuffer address exported as a var? | 19:05 |
| cinap_lenrek | i'm removing the fbpanel command completely because i couldnt figure out all that magic | 19:05 |
| cinap_lenrek | sigrid: yes! | 19:05 |
| cinap_lenrek | sigrid: or even better, put on fixed address | 19:05 |
| cinap_lenrek | or both! | 19:05 |
| sigrid | yep | 19:06 |
| cinap_lenrek | i'm fine if anyone wants todo that work | 19:06 |
| cinap_lenrek | as i said, i'm horrible with git AND uboot | 19:06 |
| cinap_lenrek | i hate it like a plague | 19:06 |
| cinap_lenrek | sigrid: ok, updated the patch | 19:11 |
| minute | cinap_lenrek: i can try to merge it myself :3 | 19:12 |
| cinap_lenrek | nah, you'r busy man | 19:12 |
| cinap_lenrek | i just have to get my shit together | 19:12 |
| cinap_lenrek | and also need to test linux | 19:12 |
| sigrid | + sprintf(lcdif.modeIdent, "%dx%dx%d", mode.hactive, mode.vactive, 24); | 19:16 |
| sigrid | some spurious space char it seems | 19:16 |
| sigrid | otherwise lgtm | 19:16 |
| cinap_lenrek | will fix | 19:17 |
| cinap_lenrek | fixed. | 19:19 |
| cinap_lenrek | haha | 19:27 |
| cinap_lenrek | can we get rid of the silly logo somehow? | 19:27 |
| sigrid | would probably make sense to replace it with mntre one | 19:30 |
| cinap_lenrek | found it | 19:37 |
| sigrid | some config option? | 19:37 |
| cinap_lenrek | sigrid: yeah :D | 19:37 |
| cinap_lenrek | sigrid: its in include/configs/nitrogen8m_som.h | 19:37 |
| sigrid | oh, it's a logo AND a splash screen | 19:38 |
| sigrid | maybe uboot has video support? | 19:38 |
| cinap_lenrek | uboot is completely insane | 19:38 |
| cinap_lenrek | no sane functions for anything, but fucking spash screens | 19:38 |
| cinap_lenrek | ok | 19:40 |
| cinap_lenrek | where where we | 19:40 |
| cinap_lenrek | right | 19:40 |
| cinap_lenrek | export video mode and framebuffer address? | 19:40 |
| cinap_lenrek | its also baffeling why thats the job of the graphics driver | 19:40 |
| sigrid | yeah. guess that would be enough for some uboot scripts to point kernels at the right things | 19:40 |
| cinap_lenrek | sigrid: in uboots global data struct, it has a fb_base | 20:04 |
| bkeys | When is the Reform meet up this February supposed to be? | 20:04 |
| cinap_lenrek | theres also a function to reserve framebuffer somehow | 20:05 |
| cinap_lenrek | in common board_f | 20:05 |
| cinap_lenrek | sigrid: i think setting gd->fb_base correctly would be very usefull | 20:06 |
| cinap_lenrek | sigrid: as some of te image loaders can pass a framebuffer address | 20:06 |
| cinap_lenrek | our uboot also only seems to support the first blank up to 4GB | 20:07 |
| cinap_lenrek | like from 40000000-100000000 | 20:07 |
| cinap_lenrek | we could probably reserve the fb from there | 20:07 |
| cinap_lenrek | sigrid: bdinfo also shows the FB base | 20:08 |
| sigrid | maybe that's what uboot provides to efi | 20:08 |
| cinap_lenrek | yes | 20:08 |
| cinap_lenrek | what pisses me off is that nothing is in any way consistent | 20:09 |
| cinap_lenrek | like you cant even find one example where you could follow the structure | 20:10 |
| cinap_lenrek | its all a complete mess | 20:10 |
| cinap_lenrek | ok | 20:10 |
| cinap_lenrek | i'll get some food :) | 20:10 |
| minute | highlight bluerise | 20:13 |
| minute | cinap_lenrek: sigrid: i plan to integrate a cool MNT Reform logo of course | 20:14 |
| cinap_lenrek | minute: :D | 20:14 |
| cinap_lenrek | we should make like a picture frame with the mntreform zigzag resistor pattern | 20:14 |
| sigrid | yay | 20:14 |
| + markw (~wielaard@gnu.wildebeest.org) | 20:20 | |
| josch | cinap_lenrek: i got display with uboot *and* usb keyboard! nice!! :D | 20:27 |
| josch | i do not have linux booting though ;) | 20:27 |
| josch | (screen is on but stays blank) | 20:29 |
| josch | (maybe linux boots but i cannot check with uart right now because i don't have a second computer at hand) | 20:29 |
| minute | bkeys: hit up https://mastodon.social/@bnys@lasersword.club for the meetup | 20:33 |
| minute | according to bluerise we can probably get openbsd going now, too | 20:36 |
| - mjw (QUIT: Killed (NickServ (GHOST command used by markw!~wielaard@gnu.wildebeest.org))) (~mjw_@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:39 | |
| * markw -> mjw | 20:39 | |
| + wielaard (~mjw_@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:39 | |
| bluerise | hello hello | 20:39 |
| bluerise | is there a branch? | 20:40 |
| bluerise | does it work automatically or do I need to set something on the CLI? | 20:40 |
| sigrid | https://felloff.net/usr/cinap_lenrek/lcdif.patch | 20:40 |
| + bkeys1 (~Thunderbi@157.201.98.1) | 20:40 | |
| bluerise | that's on top of which branch? the regular mnt reform one? | 20:41 |
| minute | i think it is | 20:41 |
| minute | it does apply cleanly on top of https://source.mnt.re/reform/reform-boundary-uboot at least | 20:42 |
| bluerise | Cool | 20:43 |
| minute | there's a also a secret version of RiscOS that just needs this patch to be usable IIRC | 20:51 |
| vagrantc | am i reading correctly that there is some work towards a u-boot that can output on the screen for the mnt/reform? :) | 20:59 |
| sigrid | yes | 20:59 |
| vagrantc | ACTION claps | 20:59 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 21:39 | |
| + ajr (~ajr@user/ajr) | 21:40 | |
| - bkeys1 (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@157.201.98.1) | 21:54 | |
| cinap_lenrek | FB base = 0xFE540000 | 21:56 |
| cinap_lenrek | here we go | 21:56 |
| cinap_lenrek | updated the patch | 21:57 |
| cinap_lenrek | the driver now, if gd->fb_base is set, uses that for the framebuffer | 21:57 |
| cinap_lenrek | otherwise allocates it and sets it | 21:57 |
| cinap_lenrek | i cannot figure out the maze that sets gd->fb_base | 21:57 |
| cinap_lenrek | CONFIG_FB_ADDR seems to only apply when we have CONFIG_LCD | 21:58 |
| cinap_lenrek | which implies that we also populate some global panel_info structure | 21:58 |
| minute | oof | 21:58 |
| cinap_lenrek | but who the fuck knows | 21:58 |
| cinap_lenrek | it is like 5 difference cases | 21:58 |
| cinap_lenrek | randomly excluding ARM for no reason | 21:58 |
| minute | i also wanted to find that address once to dump it... | 21:58 |
| cinap_lenrek | i dont understand any of this mess | 21:58 |
| minute | yeah | 22:00 |
| cinap_lenrek | and i dont want to touch common/board_f.c | 22:01 |
| cinap_lenrek | maybe the right thing todo is to implement CONFIG_LCD | 22:01 |
| cinap_lenrek | which also seems to populate device tree with the framebuffer info | 22:01 |
| cinap_lenrek | but jesus crist | 22:02 |
| cinap_lenrek | how many framebuffer info structures does uboot have? | 22:02 |
| cinap_lenrek | anyway | 22:03 |
| c-keen[m] | As many as you can cast | 22:04 |
| cinap_lenrek | the idea would have been to move te framebuffer rigt below 4GB | 22:04 |
| cinap_lenrek | otherwise, you'r wasting like 30 MB from the top | 22:04 |
| cinap_lenrek | i mean, its not rocket science for an os to recover the space after the framebuffer | 22:04 |
| cinap_lenrek | but it seems ugly | 22:04 |
| cinap_lenrek | another option might be to move the framebuffer at the beginning of dram | 22:05 |
| cinap_lenrek | thats what 9front does | 22:05 |
| cinap_lenrek | we reserve the first MB of dram for some uncaced memory | 22:05 |
| cinap_lenrek | given the crazy imx memory map... | 22:05 |
| cinap_lenrek | but i dont have the load addresses of linux at hand | 22:06 |
| cinap_lenrek | would suck if that collides with the kernel load address | 22:06 |
| cinap_lenrek | yeah, no | 22:08 |
| cinap_lenrek | thats a bad idea | 22:08 |
| cinap_lenrek | i see the scripts use that memory area | 22:08 |
| cinap_lenrek | otg_upgradeu=run usbnetwork; tftp 40020000 net_upgradeu.scr && source 40020000 | 22:08 |
| minute | yeah | 22:09 |
| cinap_lenrek | so at best you'd want to move the framebuffer right before 4GB | 22:09 |
| minute | i'm currently on imx8mp, but Moving Image from 0x40480000 to 0x40600000, end=426f0000 | 22:09 |
| cinap_lenrek | then below that is uboot | 22:09 |
| cinap_lenrek | which can be recovered by the os | 22:09 |
| minute | sounds good | 22:11 |
| minute | (iff one has 4GB of memory) | 22:11 |
| cinap_lenrek | rigt | 22:11 |
| cinap_lenrek | what i *REALLY* ment is end of the dram bank that is below 4GB ;) | 22:12 |
| minute | ^^ | 22:12 |
| minute | josch: do you remember what happened when you tested greetd? was it unable to do anything after login? | 23:03 |
| josch | minute: greetd had a bad bug that immediately logged you out after installation and then didn't let anybody log in: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1025872 | 23:05 |
| josch | that bug should be fixed though | 23:06 |
| josch | i didn't try it out but the diff looks like it should be correct | 23:06 |
| josch | i'm using greetd on my reform together with agreety | 23:06 |
| minute | josch: oh wow ok | 23:06 |
| minute | i am trying it with tuigreet but it just restarts after logging in | 23:06 |
| josch | minute: what's in your /etc/greetd/config.toml? | 23:08 |
| minute | ah, this seems to be a probelm with tuigreet perhaps | 23:08 |
| minute | with agreety, i can get into a shell | 23:08 |
| josch | okay, good | 23:08 |
| josch | i'm using this: command = "/usr/sbin/agreety --cmd sway" | 23:09 |
| minute | with tuigreet, this doesn't even work: command = "/usr/local/bin/tuigreet --cmd /bin/sh" | 23:09 |
| minute | perhaps tuigreet executable has to be blessed somehow? | 23:10 |
| josch | i do not have any immediate idea :/ | 23:13 |
| c-keen[m] | <minute> "josch: do you remember what..." <- I am running greetd with wlgreet successfully | 23:13 |
| josch | c-keen[m]: my main problems were with greetd not with wlgreet | 23:13 |
| josch | c-keen[m]: wlgreet worked but it has no way change the keyboard layout in the login screen | 23:14 |
| c-keen[m] | Oh ok | 23:14 |
| minute | lol the tuigreet executable is 76 MB | 23:14 |
| minute | probably because i built a debug build | 23:15 |
| minute | > reform greetd[11629]: error: check_children: greeter exited without creating a session | 23:17 |
| minute | ok anyway, too late to debug this today | 23:17 |
| c-keen[m] | Is it using a VT that is already busy? | 23:23 |
| minute | hmm, not sure... also, i wonder why it works with agreety with otherwise identical settings | 23:26 |
| - cinap_lenrek (PART: !!unknown attribute: msg!!) (~cinap_len@ns3076381.ip-147-135-136.eu) | 23:41 | |
| + cinap_lenrek (~cinap_len@ns3076381.ip-147-135-136.eu) | 23:45 | |
| - cinap_lenrek (QUIT: Read error: Connection reset by peer) (~cinap_len@ns3076381.ip-147-135-136.eu) | 23:45 | |
| + cinap_lenrek (~cinap_len@ns3076381.ip-147-135-136.eu) | 23:46 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 23:48 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!