| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 00:37 | |
| - chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 01:14 | |
| + chrcav (~chrcav@user/chrcav) | 01:22 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 01:23 | |
| + mjw (~mjw@gnu.wildebeest.org) | 01:26 | |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 01:47 | |
| + jogu_ (~jogu@user/jogu) | 03:10 | |
| + cwebber- (~Christine@user/cwebber) | 03:13 | |
| - jogu (QUIT: *.net *.split) (~jogu@user/jogu) | 03:17 | |
| - cwebber (QUIT: *.net *.split) (~Christine@user/cwebber) | 03:17 | |
| - cmahns (QUIT: *.net *.split) (8fe824803c@2a03:6000:1812:100::10cd) | 03:17 | |
| - theesm1 (QUIT: *.net *.split) (2cbdf4b38a@2a03:6000:1812:100::11c8) | 03:17 | |
| - larme (QUIT: *.net *.split) (22544863b6@user/larme) | 03:17 | |
| * voltagex_ -> voltagex | 03:17 | |
| - paperManu (QUIT: Ping timeout: 252 seconds) (~paperManu@107.159.15.124) | 03:33 | |
| - op_4 (QUIT: Remote host closed the connection) (~tslil@user/op-4/x-9116473) | 04:05 | |
| + op_4 (~tslil@user/op-4/x-9116473) | 04:05 | |
| - robin (QUIT: Ping timeout: 246 seconds) (~robin@user/terpri) | 04:27 | |
| + larme (22544863b6@user/larme) | 06:56 | |
| + cmahns (8fe824803c@2a03:6000:1812:100::10cd) | 06:56 | |
| + theesm1 (2cbdf4b38a@2a03:6000:1812:100::11c8) | 06:56 | |
| + gustav25 (~gustav@c-78-82-52-240.bbcust.telenor.se) | 10:02 | |
| + potash (~potash@user/foghorn) | 11:09 | |
| - svp (QUIT: Quit: Gateway shutdown) (~svp@host-79-7-240-189.business.telecomitalia.it) | 12:07 | |
| + svp (~svp@host-79-7-240-189.business.telecomitalia.it) | 12:09 | |
| + mjw (~mjw@gnu.wildebeest.org) | 12:52 | |
| - amospalla (QUIT: Quit: WeeChat 4.7.1) (~jordi@user/amospalla) | 12:59 | |
| + paperManu (~paperManu@107.159.15.124) | 13:31 | |
| * cwebber- -> cwebber | 13:43 | |
| cwebber | minute: you must be tapping into some secret power lines running through the universe lately with the stuff you've been doing. it's hard to keep up! | 13:45 |
|---|---|---|
| cwebber | but as you know, getting Blender running is one of the big things that's keeping the Reform from being my main device, and it looks like there are now two paths? | 13:45 |
| cwebber | the MNT Quasar and the external GPU | 13:46 |
| cwebber | is the Quasar also still looking like a likely path for modern Blender usage? | 13:46 |
| cwebber | regardless, I am really interested in the external GPU thing | 13:46 |
| cwebber | it's surprising to me that it works at all for some reason | 13:46 |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 14:08 | |
| + amospalla (~jordi@user/amospalla) | 14:10 | |
| qbit | apparently 'halt -p' should not be used <.< | 14:13 |
| qbit | turned off my reform that way (few weeks ago) and now the batteries are super low (trying to recover them atm) | 14:14 |
| f_ | josch: might be for some secureboot stuff, can't remember | 14:15 |
| f_ | but it's completely irrelevant for you | 14:15 |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 15:15 | |
| minute | cwebber: yes, 2 proven paths and 1 unproven (panthor compute shader emulation on rk3588) | 15:25 |
| minute | cwebber: i've done a quick test of blender the first time casey and me brought up a qcs6490 devkit a few months ago | 15:26 |
| minute | cwebber: it worked fine but crashed the gpu at some point, maybe it's been fixed but will def. be fixed | 15:26 |
| minute | cwebber: but with external amdgpu on rk3588 it runs much better than i ever expected | 15:26 |
| + pomel0 (~pomel0@user/pomel0) | 15:46 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 16:33 | |
| - Gooberpatrol_66 (QUIT: Ping timeout: 246 seconds) (~Gooberpat@user/gooberpatrol66) | 16:49 | |
| cwebber | minute: innnnnteresting! | 17:27 |
| cwebber | minute: how hard was it to set up? | 17:27 |
| josch | f_: thank you, then i'll prepare a patch for gxlimg to allow making these 10 bytes reproducible by respecting $SOURCE_DATE_EPOCH | 17:33 |
| - chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 17:49 | |
| jfred | ACTION is curious if some other CAD programs might also run on the Quasar that don't run on rk3588 | 18:58 |
| + chomwitt (~chomwitt@2a02:85f:9a5f:900:1ac0:4dff:fedb:a3f1) | 19:23 | |
| + paperManu_ (~paperManu@107.159.15.124) | 19:55 | |
| minute | cwebber: the hard part was finding a working oculink dock (minisforum deg-1 did _not_ work), and it was only possible because someone recently made a kernel patchset to work around pcie controller limitations. i put these in a branch of reform-debian-packages and installed the kernel package from the CI build once it succeeded. | 20:48 |
| minute | i originally found out about those patches because i regularly skim new radxa forum posts. | 20:49 |
| cwebber | minute: innnnnteresting | 21:46 |
| minute | before i delve into "tlp" i wrote a prototype reform-power-daemon to get a baseline/feeling | 22:11 |
| cwebber | :O | 22:11 |
| - gustav25 (QUIT: Quit: Quit) (~gustav@c-78-82-52-240.bbcust.telenor.se) | 22:15 | |
| - mjw (QUIT: Ping timeout: 264 seconds) (~mjw@gnu.wildebeest.org) | 22:52 | |
| minute | josch: ch: https://source.mnt.re/reform/reform-tools/-/merge_requests/149 | 23:00 |
| minute | and cc elb | 23:02 |
| minute | hm. i wonder if we want /usr/lib/NetworkManager/conf.d/default-wifi-powersave-off.conf with rk3588 / pcie wifi | 23:04 |
| minute | great, my tool can reproducibly hang the gpu (?) | 23:07 |
| minute | (and the whole soc) | 23:07 |
| minute | hmm, seems ok on normal charger and after removing tuned* | 23:31 |
| minute | will try again on monitor | 23:31 |
| minute | doesn't happen when just un/plugging usb/charge cable from monitor | 23:32 |
| minute | now with hdmi... | 23:32 |
| minute | ok, hang when hdmi is connected and unplugging usb/power cable | 23:34 |
| minute | oh i wonder if this is a race in the panel driver | 23:47 |
| minute | yeah looks like it | 23:50 |
| minute | yeah | 23:50 |
| minute | yeah looks like it | 23:50 |
| minute | new bug found: if the pocket reform panel is turned off (dpms), and one issues "brightnessctl s 50%" (example), then the whole system freezes | 23:53 |
| + pomel0 (~pomel0@user/pomel0) | 23:59 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!