swivel | mntmn: is the opengl support for the imx8 incomplete or something? | 00:08 |
---|---|---|
mntmn | swivel: sure | 00:09 |
mntmn | i understand now that etnaviv has a strong focus on GLES and wayland | 00:10 |
mntmn | X11 or Xwayland or Desktop GL don't seem to be a priority | 00:10 |
mntmn | so those have bugs | 00:11 |
swivel | ah, that's a bummer | 00:11 |
mntmn | at least GTK3, QT and SDL work pretty well | 00:11 |
mntmn | GTK2 stuff can be pretty broken | 00:12 |
mntmn | Non-ES GL apps are hit and miss | 00:12 |
mntmn | but GC7000 is also still pretty new, kernel development wise | 00:12 |
mntmn | so things are still under heavy construction | 00:12 |
swivel | damn, gtk2 stuff not working right suggests core xorg stuff is broken, no? | 00:13 |
adjtm_ | swivel, https://mesamatrix.net/ | 00:14 |
adjtm_ | it can also have bugs of the features it already supports | 00:15 |
swivel | adjtm_: ouch, thanks | 00:15 |
mntmn | swivel: Xorg doesn't work at all with etnaviv currently, at least on imx8m | 00:15 |
mntmn | i've completely migrated to sway + xwayland | 00:15 |
swivel | mntmn: ack | 00:16 |
adjtm_ | mntmn, have you tested other wayland compositors in addition to sway and weston like mutter or kwin? | 00:17 |
swivel | there's also practically nothing from 2019 in the etnaviv column :( | 00:17 |
adjtm_ | swivel, that's only for opengl 3.0+ and opengles 3.0+ | 00:18 |
adjtm_ | everything done for opengl 2 doesn't appear there | 00:19 |
adjtm_ | same for opengles 2 | 00:19 |
swivel | is the opengl 2 story good? | 00:19 |
mntmn | good enough for blender 2.79 for example | 00:23 |
mntmn | adjtm_: haven't tested kwin yet, mutter / gnome 3 worked last time i tested it | 00:23 |
mntmn | i think kwin should work well | 00:23 |
swivel | does solvespace work? | 00:24 |
mntmn | didn't test that one yet, but i can! | 00:29 |
swivel | cool, I use that much more than blender :) | 00:32 |
mntmn | solvespace works correctly, but it could use more fps hmm | 00:44 |
mntmn | i wonder how they're doing the rendering | 00:44 |
swivel | hmm, i haven't scrutinized its implementation, it seems fine on my old thinkpads but does get bogged down with complex models | 00:46 |
swivel | mntmn: what version? | 00:47 |
mntmn | llvmpipe is faster than etnaviv on solvespace | 00:48 |
mntmn | like, with LIBGL_ALWAYS_SOFTWARE=1 | 00:49 |
mntmn | swivel: solvespace 2.3+repack1-3+b2 from debian | 00:49 |
swivel | ouch | 01:03 |
swivel | re: llvmpipe being faster | 01:03 |
mntmn | well yeah these are the kind of little things we have to track down later | 01:13 |
mntmn | but open source makes that possible at least | 01:13 |
mntmn | first reform 2 rev 2 board just went into the infrared oven | 01:14 |
swivel | woot, that's gotta be a milestone | 01:16 |
Jookia | mntmn: if theres extra space in the oven is it possible to slide like a tray of cookies in too | 01:23 |
mntmn | Jookia: that's probably unhealthy :D | 01:43 |
chomwitt | ACTION checking http://dump.mntmn.com/reform2-rev2-ibom.html | 01:46 |
mntmn | a bit sad that traces are not included in ibom | 01:47 |
mntmn | so you can't see the connectivity | 01:47 |
mntmn | it's the first time i used SMD solderable screw threads by wuerth | 01:48 |
mntmn | they're pretty cool | 01:48 |
mntmn | plan is to mount the heatsink with those | 01:49 |
- Jookia (QUIT: Remote host closed the connection) (~Jookia@gateway/tor-sasl/jookia) | 10:37 | |
+ Jookia (~Jookia@gateway/tor-sasl/jookia) | 10:49 | |
- B[] (QUIT: Ping timeout: 276 seconds) (~Thunderbi@122-61-176-31-fibre.sparkbb.co.nz) | 13:35 | |
- synaption[m] (QUIT: Remote host closed the connection) (synaptionm@gateway/shell/matrix.org/x-bgqykfekzxtsbxyi) | 13:38 | |
- MilkManzJourDadd (QUIT: Read error: Connection reset by peer) (milkmanz1@gateway/shell/matrix.org/x-tokambmdgvbpeaom) | 13:38 | |
- jfred[m] (QUIT: Write error: Connection reset by peer) (jonterracr@gateway/shell/matrix.org/x-wekomjvcjcndodrz) | 13:39 | |
- jryans (QUIT: Read error: Connection reset by peer) (jryansmatr@gateway/shell/matrix.org/x-yeftwqhxvhiltoqh) | 13:39 | |
- anjannath (QUIT: Read error: Connection reset by peer) (anjannathm@gateway/shell/matrix.org/x-afueboubsbnhnbxn) | 13:39 | |
mntmn | reform2 rev2 mostly works | 14:51 |
+ MilkManzJourDadd (milkmanz1@gateway/shell/matrix.org/x-zwhqbnxxvfiapmpl) | 14:55 | |
+ jryans (jryansmatr@gateway/shell/matrix.org/x-yrmcinxqxskuvdbi) | 14:55 | |
+ jfred[m] (jonterracr@gateway/shell/matrix.org/x-nfxnfggvfatygcmj) | 14:55 | |
+ synaption[m] (synaptionm@gateway/shell/matrix.org/x-mevzakktzrvmkyax) | 14:55 | |
+ anjannath (anjannathm@gateway/shell/matrix.org/x-glkcbrnuruuvndzx) | 14:55 | |
rvense | \o/ | 14:58 |
adjtm_ | mntmn, have you ever used a computer that works completely? | 16:36 |
mntmn | adjtm_: hmm, commodore 64 maybe | 16:36 |
mntmn | but it stops as soon as you deal with floppy drives etc :D | 16:37 |
mntmn | adjtm_: good point! | 16:37 |
adjtm_ | my new laptop also mostly works, except the fingerprint reader and bugs with wayland version of mutter (or gnome shell) which has bugs | 16:49 |
adjtm_ | and in addition of all the uefi blobs it requieres proprietary firmware for wifi and probably other | 16:50 |
mntmn | haha yeah | 16:51 |
mntmn | i made a new discovery. PCIe doesn't work at all with the mainline kernel on this board. it doesn't matter if there is an external clock or not. with the vendor kernel, both ports work, same hardware | 16:52 |
- buckket (QUIT: Quit: buckket) (~buckket@2001:41d0:a:6733::1) | 19:20 | |
+ buckket (~buckket@uncloaked.net) | 19:27 | |
- Jookia (QUIT: Remote host closed the connection) (~Jookia@gateway/tor-sasl/jookia) | 22:40 | |
+ Jookia (~Jookia@gateway/tor-sasl/jookia) | 22:42 | |
+ erlehmann (~erlehmann@x2f7fe56.dyn.telefonica.de) | 23:44 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!