| + chartreuse (~chartreus@node-1w7jr9ql2247t9azo97sze19g.ipv6.telus.net) | 01:17 | |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 01:30 | |
| + ggoes (~gregf@fsf/staff/ggoes) | 01:31 | |
| - ggoes (QUIT: Client Quit) (~gregf@fsf/staff/ggoes) | 01:32 | |
| + ggoes (~gregf@fsf/staff/ggoes) | 01:33 | |
| - ggoes (QUIT: Ping timeout: 256 seconds) (~gregf@fsf/staff/ggoes) | 01:47 | |
| - mtm (QUIT: Ping timeout: 276 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 02:04 | |
| - GNUmoon (QUIT: Ping timeout: 268 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 02:05 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 02:11 | |
| - GNUmoon (QUIT: Ping timeout: 268 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 02:18 | |
| - bluerise (QUIT: Changing host) (~bluerise@p5b211827.dip0.t-ipconnect.de) | 02:21 | |
| + bluerise (~bluerise@user/bluerise) | 02:21 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 02:25 | |
| + ggoes (~gregf@fsf/staff/ggoes) | 02:42 | |
| - ggoes (QUIT: Client Quit) (~gregf@fsf/staff/ggoes) | 02:42 | |
| + ggoes (~gregf@fsf/staff/ggoes) | 03:05 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 03:13 | |
| cinap_lenrek | minute: how many watt do you need for powering the reform2? | 03:14 |
|---|---|---|
| cinap_lenrek | wonder if one could power/charge it thru PoE+/AT 25W? | 03:17 |
| bkeys | I believe we talked about this before cause I asked the same thing; and I believe the answer was yes | 03:46 |
| cinap_lenrek | because the system controller shows only miliamps @8*3.3V? | 03:47 |
| cinap_lenrek | tho not sure if the display is accurate or if anything can exceed it? | 03:47 |
| cinap_lenrek | i only remember that the reform takes a variable voltage range | 03:48 |
| cinap_lenrek | (on barrel jack) | 03:49 |
| cinap_lenrek | but thanks bkeys! | 03:54 |
| cinap_lenrek | good night. | 03:54 |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 04:09 | |
| - bkeys (QUIT: Ping timeout: 244 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 04:36 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 04:54 | |
| josch | cinap_lenrek: I had success with charging my reform with USB 2 (not USB-C) from 80% to 100% but due to the ampere limitations of USB2 the process is *very* slow: https://community.mnt.re/t/will-we-be-able-to-usb-power-mnt-reform/241/12 | 06:08 |
| - erle (QUIT: Ping timeout: 240 seconds) (~erle@ip5f5af7e0.dynamic.kabel-deutschland.de) | 08:12 | |
| + chomwitt (~chomwitt@2a02:587:dc0d:4a00:c5c0:a118:2df4:ea86) | 10:19 | |
| + erle (~erle@ip5f5af7e0.dynamic.kabel-deutschland.de) | 11:29 | |
| minute | josch: hmm > (com.github.rafostar.Clapper:3662): Gjs-CRITICAL **: 11:37:25.392: JS ERROR: Error: GSettings key use-playbin3 not found in schema com.github.rafostar.Clapper | 11:37 |
| minute | > clapper/unstable,now 0.5.2-2 arm64 [installed] | 11:38 |
| josch | minute: you just start clapper and that's the output? on a rescue system or a full one? | 12:05 |
| minute | josch: full system, i installed clapper and have latest reform-tools, yeah | 12:07 |
| josch | okay, i'll try to reproduce | 12:07 |
| minute | josch: i mean, it is not a regular downloaded sysimage v3 | 12:07 |
| minute | but my system that i brought to kind of the same state over 2 years | 12:08 |
| minute | maybe there's an implicit dependency on something? | 12:08 |
| minute | gstreamer version perhaps? | 12:09 |
| sigrid | looks more like it can't load settimg | 12:11 |
| sigrid | *ngs | 12:11 |
| sigrid | maybe try removing files with settings and see if it helps | 12:12 |
| minute | sigrid: ah! | 12:13 |
| minute | also, this didn't make it onto my system https://source.mnt.re/reform/reform-tools/-/commit/2e0cff9b45d45237822a1d27e40dfd74af9476d0 | 12:13 |
| minute | `gsettings list-keys com.github.rafostar.Clapper` does not include use-playbin3 for me | 12:15 |
| minute | josch: i have old clapper residue in /usr/local, that might be it | 12:16 |
| minute | yeah | 12:23 |
| minute | i had to nuke stuff in /usr/local/share, esp. /usr/share/glib-2.0/schemas/ | 12:23 |
| minute | works perfectly now | 12:25 |
| - S0rin (QUIT: Ping timeout: 255 seconds) (~S0rin@user/s0rin) | 12:36 | |
| josch | ah phew, nice :) | 12:45 |
| josch | i was getting worried because this was my first Debian package that I created 100% from scratch on the reform and tested it there as well | 12:46 |
| + S0rin (~S0rin@user/s0rin) | 12:48 | |
| minute | josch: sorry for the false alarm! | 12:51 |
| josch | no worries, i'm happy the package is getting some testing and it is totally possible that i was missing some dependencies, maybe there still are some :) | 12:52 |
| - S0rin (QUIT: Ping timeout: 244 seconds) (~S0rin@user/s0rin) | 13:33 | |
| * lastebil_ -> lastebil | 13:42 | |
| + S0rin (~S0rin@user/s0rin) | 13:46 | |
| - mtm (QUIT: Ping timeout: 240 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 14:03 | |
| cinap_lenrek | minute: sigrid reported high cpu activity for our hid driver | 14:23 |
| cinap_lenrek | just found it seems the mouse continuously responds with hid report even tho nobody's touching the mouse | 14:24 |
| minute | cinap_lenrek: ah | 14:24 |
| minute | makes sense | 14:24 |
| cinap_lenrek | this is not the case for the keyboard | 14:24 |
| minute | also makes sense | 14:24 |
| cinap_lenrek | only the trackball | 14:24 |
| cinap_lenrek | it might not be a big issue | 14:24 |
| minute | they're based on different usb hid code | 14:24 |
| minute | well, could be a nice optimization | 14:25 |
| minute | in Mouse.c | 14:25 |
| minute | PR welcome ;) | 14:25 |
| cinap_lenrek | the hid driver is smart enougth to diff the state and not send the event upstream if nothing changed | 14:25 |
| cinap_lenrek | yeah | 14:25 |
| cinap_lenrek | minute: okay :D | 14:25 |
| cinap_lenrek | minute: i'll take a look :) | 14:25 |
| cinap_lenrek | oh | 14:31 |
| cinap_lenrek | the return value is to force to send a report | 14:32 |
| cinap_lenrek | oh | 14:35 |
| cinap_lenrek | they do a memcmp | 14:35 |
| cinap_lenrek | thats perfect | 14:35 |
| cinap_lenrek | so the library saves the previous report descriptor | 14:35 |
| cinap_lenrek | and if we return false, it'll send it if it differs from previous one | 14:36 |
| minute | ah | 14:36 |
| + Christoph_ (~Christoph@p54bf6443.dip0.t-ipconnect.de) | 14:38 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 14:53 | |
| minute | since a recent update, gtk2 apps are incredibly slow to update... wonder why | 15:11 |
| minute | UI elements i mean | 15:12 |
| Boostisbetter | minute: I don't think it is necessary and this IRC channel is doing great, but have you ever thought about doing a Discord group? | 15:12 |
| Boostisbetter | or do you think that would open the floodgates of the internet and would this all too unmanageable? | 15:12 |
| minute | Boostisbetter: discord is a walled garden, i'm not running things on that | 15:13 |
| minute | me being on twitter is already v questionable ^^ | 15:14 |
| cinap_lenrek | minute: works. | 15:14 |
| cinap_lenrek | minute: it still polls, but much slower | 15:14 |
| cinap_lenrek | minute: all you need todo is return false instead of true | 15:15 |
| minute | cinap_lenrek: great | 15:15 |
| Boostisbetter | minute: ok totally agree. Just curious | 15:15 |
| Boostisbetter | minute: twitter feels like a cesspool of negativity. You have to try real hard while on it to not be brought down. | 15:16 |
| sigrid | cinap_lenrek: how oftem does it send a report now? | 15:16 |
| sigrid | *often | 15:16 |
| Boostisbetter | minute: and sadly Mastodon has a lot of the same kind of garbage going on with it as well. | 15:16 |
| minute | Boostisbetter: depends who you follow and/or mute etc but yeah | 15:16 |
| Boostisbetter | true, I do my best to block the negativity but sometimes it feels like a full time job. I then have to ask myself, what am I really getting from them. | 15:17 |
| minute | Boostisbetter: but i did not mean the content, more like the fact that it is a commercial, closed, ad-fueled thing | 15:17 |
| Boostisbetter | I guess twitter has helped me know about MNT stuff, so there is that. | 15:17 |
| cinap_lenrek | sigrid: it should honour the poll interval of the endpoint | 15:18 |
| Boostisbetter | minute: yeah I know what you mean. I hate that Framework insist on using it, but i think for their purposes they are trying to reach as many people as possible and that is why they justify it. Nirav is a super Linux open source dude himself, and heavily supports it on the Framework laptop. | 15:18 |
| cinap_lenrek | err sorry | 15:19 |
| cinap_lenrek | report interval | 15:19 |
| cinap_lenrek | its a setting the host can override with a control request | 15:19 |
| cinap_lenrek | i think default is 500ms? | 15:19 |
| cinap_lenrek | i should measure it exactly | 15:19 |
| cinap_lenrek | lemme see | 15:20 |
| minute | hmm gtk2 can be made a lot faster by changing to an ancient theme like redmond... so i guess it's glamor drawcalls that are slow | 15:24 |
| - leonardo (QUIT: Ping timeout: 248 seconds) (~leonardo@user/leonardo) | 15:26 | |
| minute | oh yeah that is the problem | 15:28 |
| minute | setting XWAYLAND_NO_GLAMOR=1 in my profile makes gtk2 or x11 apps 10x faster | 15:28 |
| minute | gargh | 15:28 |
| minute | so software rendering is much much faster than the weird gpu 2d rendering that is shoehorned in there | 15:29 |
| cinap_lenrek | it alternates between 100ms and 30ms | 15:30 |
| cinap_lenrek | ah | 15:32 |
| cinap_lenrek | but we override the idel time because of raspi work arround | 15:33 |
| sigrid | on raspi the usb front falls off if it's too long? | 15:33 |
| cinap_lenrek | should probably do this conditionally | 15:33 |
| cinap_lenrek | sigrid: the controller looses fucking events | 15:33 |
| cinap_lenrek | if you use usb ethernet AND hid at the same time | 15:33 |
| sigrid | ah right | 15:34 |
| cinap_lenrek | this is especially bad for keyboard | 15:34 |
| cinap_lenrek | its just a clusterfuck | 15:34 |
| cinap_lenrek | lemme remove the hack | 15:34 |
| cinap_lenrek | yep | 15:35 |
| cinap_lenrek | now its 500ms | 15:35 |
| sigrid | nice! | 15:35 |
| cinap_lenrek | so my patch works | 15:35 |
| cinap_lenrek | minute: http://felloff.net/usr/cinap_lenrek/Mouse.c.diff | 15:39 |
| cinap_lenrek | ignore the other stuff | 15:39 |
| cinap_lenrek | that was just autism | 15:39 |
| cinap_lenrek | the real change is just return ture -> false | 15:39 |
| sigrid | I wonder if that issue could cause the trackpoint failing to work on linux resume from suspend | 15:40 |
| cinap_lenrek | no | 15:40 |
| cinap_lenrek | i dont think so | 15:40 |
| sigrid | I should probably read the firmware code instead of imagining things | 15:41 |
| cinap_lenrek | sigrid: the controller should still poll according to the endpoints poll interval | 15:42 |
| cinap_lenrek | and then the library decides if it returns data or not on each poll | 15:42 |
| cinap_lenrek | on each poll, we sample the sensor and buttons and generate a report | 15:42 |
| cinap_lenrek | if we return false, the library will compare the generated report blob with the previous one it send | 15:43 |
| cinap_lenrek | and and sends a nak if it hasnt changed | 15:43 |
| sigrid | yeah, that's how understood it works | 15:43 |
| cinap_lenrek | and theres a idle timer inside the device as well that we can set | 15:44 |
| cinap_lenrek | how do i make pull request? | 15:52 |
| sigrid | I think you need to create an account | 15:54 |
| cinap_lenrek | pfffff | 15:54 |
| cinap_lenrek | nope. | 15:54 |
| sigrid | I am sure minute can apply the patch without though | 15:55 |
| cinap_lenrek | yeah | 15:55 |
| cinap_lenrek | http://felloff.net/usr/cinap_lenrek/0001-trackball-dont-force-send-report.patch | 16:03 |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 16:09 | |
| + leonardo (~leonardo@user/leonardo) | 16:32 | |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 18:23 | |
| - Christoph_ (QUIT: Remote host closed the connection) (~Christoph@p54bf6443.dip0.t-ipconnect.de) | 19:29 | |
| + Christoph_ (~Christoph@p54bf6443.dip0.t-ipconnect.de) | 19:48 | |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 20:05 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 20:06 | |
| flowy | minute: uhh WTH! thank you for the XWAYLAND_NO_GLAMOR=1 tip!! | 20:49 |
| Boostisbetter | flowy: What does that do, and where would you insert that? Your Sway config file? | 20:50 |
| flowy | this single-handedly made certain apps useable. i hadn't bothered investigating why i couldn't use gimp, for example. i just used a different workstation instead. | 20:50 |
| flowy | it feels like more than a 10x improvement for certain tasks... some things just hung. | 20:50 |
| Boostisbetter | I imagine it just turns off visual effects, etc. | 20:50 |
| flowy | no, nothing looks different | 20:50 |
| flowy | i just put 'export XWAYLAND_NO_GLAMOR=1' into my ~/.profile | 20:51 |
| flowy | make sure to logout then back in | 20:51 |
| josch | maybe this should go into https://source.mnt.re/reform/reform-tools/-/blob/main/etc/profile.d/reform.sh | 20:51 |
| flowy | flicking that on and off changes gimp from being useable to unuseable honestly | 20:52 |
| flowy | yeah, if there isn't a better root cause to sort out, this definitely needs to be defaulted in sysimagev3 | 20:53 |
| flowy | but i don't really understand what it is | 20:53 |
| flowy | i thought it was really weird that certain apps weren't working. | 20:53 |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 21:00 | |
| - Christoph_ (QUIT: Ping timeout: 240 seconds) (~Christoph@p54bf6443.dip0.t-ipconnect.de) | 22:11 | |
| grubman | is there a supported OS list anywhere/ | 22:33 |
| grubman | also what is the clear plastic(?) piece that is included with the sdcard ? | 22:46 |
| Boostisbetter | grubman: that is a wifi antenna mount | 22:48 |
| Boostisbetter | I think | 22:48 |
| grubman | hm ok | 22:50 |
| grubman | looking at the mcpie mount, it doesn't look like it supports half-height models? like an atheros ar5b93 | 22:50 |
| grubman | no biggie. i'll just use a USB NIC | 22:51 |
| grubman | that plastic piece has no screw holes in it. it's about 2" long and 1" wide. | 22:55 |
| cinap_lenrek | cpu1: 1400MHz ARM Cortex A53 | 23:02 |
| cinap_lenrek | TURRRRBO MODE | 23:02 |
| josch | grubman: https://mntre.com/media/reform_md/mnt-reform2-diy-manual-r1.pdf you put that around the UI1/UI2 headers | 23:04 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 23:10 | |
| - bluerise (QUIT: Ping timeout: 244 seconds) (~bluerise@user/bluerise) | 23:19 | |
| - chomwitt (QUIT: Ping timeout: 240 seconds) (~chomwitt@2a02:587:dc0d:4a00:c5c0:a118:2df4:ea86) | 23:22 | |
| minute | flowy: i kinda know what it is | 23:24 |
| minute | flowy: glamor is some code in xwayland that tries to use OpenGL to accelerate 2d graphics operations | 23:26 |
| minute | flowy: but it goes wrong somehow in a recent version (have not yet bisected since when), so it's a regression | 23:26 |
| minute | flowy: so the GPU method is 10 times slower than the software method | 23:26 |
| minute | flowy: a bad side effect of turning it of is that you also lose OpenGL in X11 applications | 23:26 |
| minute | because unfortunately the people who wrote this kind of hard-bound GLX and glamor together | 23:27 |
| minute | i wish this was not so | 23:27 |
| minute | because glamor is like my number one enemy | 23:27 |
| minute | with this super slow version our patch also doesn't make a difference josch | 23:33 |
| grubman | josch: i see. thanks. | 23:38 |
| + bluerise (~bluerise@p5b211827.dip0.t-ipconnect.de) | 23:51 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!