| + eibachd (~eibachd@p200300dcf7353e005380a75b8a7edcb7.dip0.t-ipconnect.de) | 00:00 | |
| - mjw (QUIT: Ping timeout: 268 seconds) (~mjw@213.246.238.195) | 00:22 | |
| - S0rin (QUIT: Ping timeout: 240 seconds) (~S0rin@user/s0rin) | 00:32 | |
| + S0rin (~S0rin@user/s0rin) | 00:34 | |
| - erle (QUIT: Ping timeout: 264 seconds) (~erle@89.205.129.254) | 00:45 | |
| - S0rin (QUIT: Ping timeout: 264 seconds) (~S0rin@user/s0rin) | 00:57 | |
| + erle (~erle@89.205.129.81) | 00:59 | |
| + S0rin (~S0rin@user/s0rin) | 01:02 | |
| - S0rin (QUIT: Ping timeout: 260 seconds) (~S0rin@user/s0rin) | 01:08 | |
| + S0rin (~S0rin@user/s0rin) | 01:09 | |
| - S0rin (QUIT: Ping timeout: 252 seconds) (~S0rin@user/s0rin) | 01:13 | |
| + S0rin (~S0rin@user/s0rin) | 01:14 | |
| + elektron (~elektron@dom0.halo.nu) | 01:15 | |
| elektron | howdy, dusting off my reform from a while back... it's running bookworm/sid | 01:16 |
|---|---|---|
| elektron | system image version 2 2021-06-08 is what the motd says... | 01:17 |
| minute | elektron: dusty indeed! | 01:17 |
| elektron | so before i apt-get upgrade it, think i should change the suite to bookwork innstead of sid? | 01:17 |
| minute | were on sysimage v4 now ^^ | 01:17 |
| minute | we're | 01:17 |
| elektron | :-o | 01:18 |
| minute | elektron: is it already very customized? otherwise i would recommend to download a fresh image and start from that | 01:18 |
| elektron | I could just backup my homedir... | 01:19 |
| minute | yeah, recommended | 01:19 |
| elektron | Is the system image much customized from debiann? | 01:19 |
| minute | upgrading by hand from v2 to v3+ will be more challenging | 01:19 |
| minute | not very much, but the integration with uboot is | 01:19 |
| minute | before v3 it wasn't very clean | 01:19 |
| elektron | right.. okay... I'll do the latest image then ans start from there. thanks | 01:20 |
| minute | hope it'll run smoothly! | 01:20 |
| swivel | https://halo.nu/@pengaru/111870356232094258 | 01:23 |
| minute | oh, what was that | 01:25 |
| - jacobk (QUIT: Ping timeout: 240 seconds) (~quassel@utdpat241106.utdallas.edu) | 01:41 | |
| elektron | I had some things i was putting off... 1) dead 18650s and new protected battery boards to install. | 01:53 |
| elektron | 2) trying to reflow the keyboard because the 'n' key acts funny. | 01:53 |
| elektron | and 3) I pulled the barrel plug port for the DC in off the mainboard and had to repair that... | 01:54 |
| elektron | ACTION is kinda hard on stuff... | 01:54 |
| - chartreuse (QUIT: Remote host closed the connection) (~chartreus@2604:3d09:c73:9e10:d0:9258:37d9:b3df) | 01:54 | |
| digitalrayne | keyboards must be punished for all the problems they have caused in the world | 01:54 |
| + chartreuse (~chartreus@S0106908d78501d1d.cg.shawcable.net) | 01:54 | |
| swivel | careful, soon enough they'll be sentient and demand reparations | 01:57 |
| digitalrayne | as would be their right, all hail our new keyboard overlords | 01:59 |
| digitalrayne | i've just found I can decompile existing dtbs into dtbs but if anyone has those dtbs used to build the modified reform2 dtb for OpenBSD i'd still be interested | 02:00 |
| digitalrayne | there's an interesting panic on boot I'm trying to debug | 02:01 |
| minute | digitalrayne: i think they are by bluerise | 02:33 |
| minute | who is now highlighted and could hopefully chime in | 02:33 |
| digitalrayne | thank you! | 02:33 |
| minute | if i remember correctly that dtb just turns something off that crashes. maybe pcie? | 02:40 |
| digitalrayne | yeah it should just be nvme getting turned off | 02:40 |
| minute | digitalrayne: so perhaps you can use the regular dts and change the pcie nodes to disabled? | 02:42 |
| digitalrayne | i was going to compare the regular dts and the modified dts to see if only the nvme was disabled - as there is a page-table mapping related panic on boot now with OpenBSD, way before the pcie or nvme would be initialised, not even 7.2 boots anymore (or 7.4, the latest) | 02:44 |
| digitalrayne | i can definitely do the comparison by decompiling the regular one though i just thought if someone had the dts handy it would save me some time :) | 02:44 |
| minute | oh | 02:44 |
| minute | i don't have it right now, hopefully bluerise can jump in at some point... but i'm grateful you're looking into this | 02:45 |
| digitalrayne | it's a good way for me to learn more about the page table mapping system in OpenBSD for sure, and about device trees too | 02:46 |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 03:14 | |
| - erle (QUIT: Quit: Just say no, then Putin can not legally invade your nation without your consent.) (~erle@89.205.129.81) | 03:22 | |
| digitalrayne | found a comment in the irc logs that the OpenBSD dtb is the mainline linux dtb with pcie disabled, great memory minute :) | 03:27 |
| digitalrayne | https://mntre.com/reform-irc-logs/2023-01-18.log.html | 03:27 |
| + mjw (~mjw@213.246.238.195) | 04:15 | |
| - S0rin (QUIT: Ping timeout: 264 seconds) (~S0rin@user/s0rin) | 04:22 | |
| - nsc (QUIT: Ping timeout: 264 seconds) (~nicolas@75-98-142-46.pool.kielnet.net) | 04:28 | |
| + nsc (~nicolas@150-121-142-46.pool.kielnet.net) | 04:30 | |
| + S0rin (~S0rin@user/s0rin) | 04:37 | |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@213.246.238.195) | 04:43 | |
| swivel | minute: is it possible to use a reform2 included keyboard as a usb-c standalone as-is or do you need to flash it? I see a usb-c receptacle on it, but trying to plug in doesn't seem to do any HID activity on the host | 05:32 |
| - jacobk (QUIT: Ping timeout: 240 seconds) (~quassel@utdpat241106.utdallas.edu) | 05:51 | |
| + jacobk (~quassel@129.110.242.16) | 06:03 | |
| - jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@129.110.242.16) | 06:36 | |
| swivel | found the cause for the double keypresses: https://halo.nu/@pengaru/111871776796261637 | 07:25 |
| josch | swivel: same thing happened to one of my switches | 07:58 |
| josch | i only found the debris after already having desoldered the switch | 07:59 |
| josch | now that i know where it's hiding, maybe next time i can get it out just by removing some keycaps | 07:59 |
| josch | hramrach: thank you for your explanations yesterday. Then I'm looking forward to a future where dtbs for the reform become stable. As you see with the changes required for openbsd above, we are sadly not there yet. :( | 08:02 |
| - eibachd (QUIT: Ping timeout: 260 seconds) (~eibachd@p200300dcf7353e005380a75b8a7edcb7.dip0.t-ipconnect.de) | 08:09 | |
| + eibachd (~eibachd@p200300dcf7353e01aa3ea7f72375e622.dip0.t-ipconnect.de) | 08:09 | |
| + mjw (~mjw@213.246.238.195) | 08:34 | |
| - mjw (QUIT: Ping timeout: 276 seconds) (~mjw@213.246.238.195) | 08:53 | |
| + mjw (mjw@151.216.131.124) | 10:02 | |
| - eibachd (QUIT: Ping timeout: 268 seconds) (~eibachd@p200300dcf7353e01aa3ea7f72375e622.dip0.t-ipconnect.de) | 10:08 | |
| + eibachd (~eibachd@p200300dcf7353e00626db73a6e5d2c4c.dip0.t-ipconnect.de) | 10:08 | |
| + erle (~erle@89.205.129.81) | 11:04 | |
| - mjw (QUIT: Ping timeout: 252 seconds) (mjw@151.216.131.124) | 11:23 | |
| - eibachd (QUIT: Ping timeout: 268 seconds) (~eibachd@p200300dcf7353e00626db73a6e5d2c4c.dip0.t-ipconnect.de) | 11:30 | |
| + eibachd (~eibachd@p200300dcf7353e007764c47b8ab063f2.dip0.t-ipconnect.de) | 11:31 | |
| minute | swivel: it depends on the keyboard version. for v2 you would need to solder in a little ldo (there's an empty footprint). kbd v3 has a "standalone" dip switch | 11:46 |
| minute | swivel: (or for v2, alternatively, supply 3v3 through the uart port) | 11:46 |
| + mjw (mjw@151.216.131.124) | 12:00 | |
| - mjw (QUIT: Ping timeout: 260 seconds) (mjw@151.216.131.124) | 12:29 | |
| - eibachd (QUIT: Ping timeout: 246 seconds) (~eibachd@p200300dcf7353e007764c47b8ab063f2.dip0.t-ipconnect.de) | 13:06 | |
| + eibachd (~eibachd@p200300dcf7353e00c824442314ec2325.dip0.t-ipconnect.de) | 13:07 | |
| hramrach | josch: the change for OpenBSD is trivial, though. It merely disables a device, it does not alter the hardware description in any way. You can easily apply such change in u-boot either as overlay or cook up something as a command with its dynamic DT modification functions already in place. | 13:19 |
| hramrach | aargh, on my stale distribution I cannot open kicad files because they are made by newer version .. that does compile but fails tests | 14:26 |
| + bkeys (~Thunderbi@45.134.140.153) | 14:59 | |
| + pandora (uid585533@2a03:5180:f:3::8:ef3d) | 15:17 | |
| + jacobk (~quassel@64.189.201.150) | 16:41 | |
| minute | so who's gonna hack it into their reform https://linuxgizmos.com/new-aetina-mxm-gpu-series-with-nvidia-ada-technology/ | 16:46 |
| minute | smallest one has 60W TDP though ^^ | 16:46 |
| hramrach | is MXM a module standard of some kind? | 16:56 |
| hramrach | Apparently it is Mobile PCI Express Module as well as a former k-pop band | 16:58 |
| hramrach | it's used by the SMARC modules discussed previously (maybe on the web forum) | 16:59 |
| - eibachd (QUIT: Ping timeout: 256 seconds) (~eibachd@p200300dcf7353e00c824442314ec2325.dip0.t-ipconnect.de) | 16:59 | |
| + eibachd (~eibachd@p200300dcf7353e0139c4e000b0b07938.dip0.t-ipconnect.de) | 17:00 | |
| swivel | minute: thanks | 17:26 |
| swivel | elektron: ^^ | 17:27 |
| - flowy (QUIT: Quit: WeeChat 2.3) (~flowy@2a01:4f8:c0c:1a8f::1) | 17:28 | |
| minute | hramrach: smarc modules use the same connector. mxm is mostly for mobile gpus but it's not super standardized | 17:39 |
| hramrach | there is a log of optional pins, some GPIO and OEM pins | 17:43 |
| - eibachd (QUIT: Ping timeout: 264 seconds) (~eibachd@p200300dcf7353e0139c4e000b0b07938.dip0.t-ipconnect.de) | 17:44 | |
| + eibachd (~eibachd@p200300dcf7353e0062ae60a5b279e04b.dip0.t-ipconnect.de) | 17:44 | |
| hramrach | it is standardized to some extent but not sure how useful the standard is in practice if you waht to do more than connect a GPU over PCIe and pull a DP output out of it | 17:45 |
| + pr8dan (~pr8dan@pub151248192164.dh-hfc.datazug.ch) | 18:16 | |
| - erle (QUIT: Ping timeout: 276 seconds) (~erle@89.205.129.81) | 18:47 | |
| bluerise | digitalrayne: what's up? | 18:48 |
| bluerise | I'm running a new U-Boot https://github.com/bluerise/u-boot/commits/mnt/ | 18:50 |
| bluerise | and the DTB that's part of that U-Boot, none extra | 18:50 |
| minute | bluerise: apparently there's a kernel panic when booting openbsd nowadays? | 18:50 |
| minute | (i don't have an imx r/n so can't check myself) | 18:51 |
| bluerise | Well there are a few problems. First of all OpenBSD is using 99% mainline bindings, apart from PCIe, because PCIe was not upstreamed yet | 18:51 |
| bluerise | so for PCIe I have a patch in U-Boot | 18:52 |
| bluerise | or maybe not | 18:52 |
| minute | aha? | 18:52 |
| bluerise | here's only the hack for the U-Boot PCIe code https://github.com/bluerise/u-boot/commit/dd59fc2858ca9ffe35e6111a8f35e2e43e23597e | 18:53 |
| bluerise | digitalrayne | found a comment in the irc logs that the OpenBSD dtb is the mainline linux dtb with pcie disabled, great memory minute :) | 18:55 |
| bluerise | but yeah, that's what matters | 18:55 |
| minute | bluerise: thx | 18:55 |
| bluerise | it shouldn't necessarily be with PCIe disabled | 18:56 |
| bluerise | because mine uses PCIe for NVMe (u-boot already boots kernels from NVMe) + PCIe for WiFi | 18:56 |
| + Guest10 (~Guest10@88.97.17.88) | 18:58 | |
| - Guest10 (QUIT: Client Quit) (~Guest10@88.97.17.88) | 18:59 | |
| + Guest10 (~Guest10@88.97.17.88) | 18:59 | |
| + erle (~erle@89.205.133.5) | 19:00 | |
| - Guest10 (QUIT: Ping timeout: 250 seconds) (~Guest10@88.97.17.88) | 19:16 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 20:00 | |
| - xktr (QUIT: Ping timeout: 246 seconds) (~xktr@user/xktr) | 20:19 | |
| + xktr (~xktr@user/xktr) | 20:21 | |
| + Guest10 (~Guest10@88.97.17.88) | 20:27 | |
| digitalrayne | bluerise: thanks, i'll give the u-boot on that branch a try and see if I get the same issue. I tried the latest boundary u-boot/mnt u-boot and tested 7.2 and 7.4 and got a panic during page table mapping | 20:46 |
| digitalrayne | the panic was mentioned on the openbsd mailing list recently too I found, panic: pmap_pte_insert: invalid cache mode just after ef0 is initialised | 20:46 |
| digitalrayne | https://marc.info/?l=openbsd-bugs&m=170189272731225&w=2 | 20:53 |
| digitalrayne | openbsd-bugs sorry | 20:53 |
| + mjw (~mjw@213.246.238.195) | 22:23 | |
| - pr8dan (QUIT: Remote host closed the connection) (~pr8dan@pub151248192164.dh-hfc.datazug.ch) | 23:12 | |
| bluerise | yeah, that's because the u-boot is just way too old | 23:25 |
| bluerise | > U-Boot 2018.07 | 23:25 |
| digitalrayne | ok - thank you, that makes sense. i took a look at the page table mapping code last week and had wondered if the information being populated by uboot was causing the issue, it seemed like the cache type data in memory wasn't matching a cache type flag openbsd knew about | 23:36 |
| - Guest10 (QUIT: Ping timeout: 250 seconds) (~Guest10@88.97.17.88) | 23:59 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!