| - mjw (QUIT: Remote host closed the connection) (~mark@gnu.wildebeest.org) | 00:00 | |
| + mjw (~mark@gnu.wildebeest.org) | 00:00 | |
| - ajr (QUIT: Quit: WeeChat 3.6) (~ajr@user/ajr) | 00:10 | |
| - Gooberpatrol66 (QUIT: Quit: Leaving) (~Gooberpat@user/gooberpatrol66) | 00:30 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 00:59 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 01:06 | |
| - pinoaffe (QUIT: Quit: Bridge terminating on SIGTERM) (~pinoaffep@2a01:4f9:c010:a00b:1337:1337:11be:10) | 01:53 | |
| + pinoaffe (~pinoaffep@2a01:4f9:c010:a00b:1337:1337:11be:10) | 01:54 | |
| - mtm (QUIT: Ping timeout: 252 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 02:02 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 03:55 | |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 04:08 | |
| - chartreuse (QUIT: Quit: leaving) (~chartreus@S0106f0f249dfd9c3.cg.shawcable.net) | 05:10 | |
| + chartreuse (~chartreus@S0106f0f249dfd9c3.cg.shawcable.net) | 05:25 | |
| + qwer (~qwer@78-80-28-214.customers.tmcz.cz) | 05:29 | |
| + fsx (~fsx@durian.61924.nl) | 09:12 | |
| - qwer (QUIT: Ping timeout: 240 seconds) (~qwer@78-80-28-214.customers.tmcz.cz) | 10:46 | |
| + qwer (~qwer@37-48-2-248.nat.epc.tmcz.cz) | 11:02 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by wielaard!~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440))) (~mark@gnu.wildebeest.org) | 11:30 | |
| * wielaard -> mjw | 11:30 | |
| + mark__ (~mark@gnu.wildebeest.org) | 11:30 | |
| - mtm (QUIT: Ping timeout: 244 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 14:03 | |
| flowy1 | minute: congrats on the LS1028A progress! what are your subjective impressions in regards to web browsing using this SOM? noticeably snappier? | 14:35 |
|---|---|---|
| * flowy1 -> flowy | 14:36 | |
| flowy | and re: GPU performance, are there subjectively noticeable differences doing normal desktop stuff i.e. not gaming? | 14:38 |
| - buckket (QUIT: Remote host closed the connection) (~buckket@pdp8.buckket.org) | 14:39 | |
| + buckket (~buckket@pdp8.buckket.org) | 14:42 | |
| flowy | and theoretically, if a carrier board were designed, it would be able to take any module that the reform could? | 14:43 |
| flowy | FYI, i'm eager about the module upgrade, but given the price i think for me some kind of reform motherboard would be a dealbreaker. mITX would be amazing. then i'd know that i could put the old module to some use | 14:44 |
| C-Keen | hm, after the kernel upgrade it seems that the sound init works more reliable during boot. or is that coincidence? | 14:45 |
| C-Keen | (from v1 to v3 image I mean) | 14:45 |
| flowy | or if i decided the LS1028A didn't fit my reform usage so well, i could swap back and still have some other use for it | 14:45 |
| flowy | C-Keen: i never experienced that but i'd imagine it could have been kind of racy, and that the usage of kernel modules in sysimagev3 has dodged the race issue | 14:48 |
| C-Keen | flowy: yeah it probably is. | 14:51 |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 15:07 | |
| - kitty4 (QUIT: Ping timeout: 248 seconds) (~kitty@096-039-147-043.res.spectrum.com) | 15:19 | |
| - mark__ (QUIT: Ping timeout: 268 seconds) (~mark@gnu.wildebeest.org) | 15:19 | |
| + kitty4 (~kitty@096-039-147-043.res.spectrum.com) | 15:23 | |
| + mark__ (~mark@gnu.wildebeest.org) | 15:23 | |
| - kitty4 (QUIT: Ping timeout: 268 seconds) (~kitty@096-039-147-043.res.spectrum.com) | 16:02 | |
| minute | flowy: yep, i want to make an mITX carrier. | 16:25 |
| flowy | minute: niice | 16:45 |
| minute | https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.19 | 16:57 |
| minute | commit 2518f226c60d8e04d18ba4295500a5b0b8ac7659 has etnaviv stuff | 16:58 |
| minute | etnaviv: | 16:58 |
| minute | - fencing improvements | 16:58 |
| minute | - fix address space collisions | 16:58 |
| minute | - cleanup MMU reference handling | 16:58 |
| minute | and commit f8122500a039abeabfff41b0ad8b6a2c94c1107d | 16:58 |
| minute | fix address space collisions in some edge cases when userspace is | 16:58 |
| minute | using softpin and cleans up the MMU reference handling a bit. | 16:58 |
| minute | actually a bunch more. we should try to upgrade to 5.19 asap cc josch | 16:59 |
| * qbit -> qbit_ | 17:01 | |
| * qbit_ -> qqbit | 17:02 | |
| * qqbit -> qbqit | 17:02 | |
| * qbqit -> qbit | 17:02 | |
| * qbit -> qbit_m | 17:08 | |
| * qbit_m -> qbit | 17:09 | |
| eery | huh, sounds like that'd help a lot with the out-of-memory errors with etnaviv | 17:11 |
| vkoskiv_ | I modified the kbd fw a bit to ignore OLED draw commands when the menu is being interacted with. Short demo: https://youtu.be/6sKAFdyoz1I | 17:31 |
| vkoskiv_ | MR coming shortly. | 17:31 |
| vkoskiv_ | I think this make sense as a default, the menus are more important than whatever the OS is trying to draw on there. | 17:34 |
| Kooda | minute: I was prompted to upgrade to at least 5.19 to fix my Mesa 22.2.0 problems | 17:35 |
| josch | minute: 5.19 was uploaded to Debian unstable a few hours ago so I can start rebasing the patches this weekend. | 17:36 |
| Kooda | Awesome <3 | 17:37 |
| josch | Kooda: did you attempt to rebase the patches already? | 17:37 |
| Kooda | I haven’t, no | 17:37 |
| josch | in the past multiple people did the rebasing independently from each other which would be nice to avoid ;) | 17:37 |
| Kooda | I’m very unfamiliar with these parts so I was afraid so mess something up. Especially since some PLL stuff seem to have been merged, but are slightly different | 17:38 |
| minute | josch: amazing, thanks! | 17:38 |
| josch | minute: do we still need the ffmpeg patch? | 17:45 |
| josch | https://github.com/martinetd/FFmpeg/tree/v4l2-request has not been updated for ffmpeg 5.1 and doesn't apply anymore | 17:47 |
| minute | josch: afaik we agreed to drop it for now | 17:49 |
| - S0rin (QUIT: Ping timeout: 252 seconds) (~S0rin@user/s0rin) | 18:05 | |
| * cwebber` -> cwebber | 18:05 | |
| - cwebber (QUIT: Changing host) (~user@2600:6c64:7e3f:fdb0::af5) | 18:05 | |
| + cwebber (~user@user/cwebber) | 18:05 | |
| vkoskiv_ | I added a MR for the above feature: https://source.mnt.re/reform/reform/-/merge_requests/31 | 18:07 |
| - mark__ (QUIT: Ping timeout: 244 seconds) (~mark@gnu.wildebeest.org) | 18:10 | |
| + mark__ (~mark@gnu.wildebeest.org) | 18:11 | |
| + S0rin (~S0rin@user/s0rin) | 18:18 | |
| - S0rin (QUIT: Ping timeout: 268 seconds) (~S0rin@user/s0rin) | 18:24 | |
| - sknebel (QUIT: *.net *.split) (~quassel@v22016013254630973.happysrv.de) | 18:28 | |
| - josch (QUIT: *.net *.split) (~josch@mister-muffin.de) | 18:28 | |
| - jomo (QUIT: *.net *.split) (~jomo@user/jomo) | 18:28 | |
| - austriancoder (QUIT: *.net *.split) (sid152545@id-152545.hampstead.irccloud.com) | 18:28 | |
| - aphistic (QUIT: *.net *.split) (sid347194@id-347194.ilkley.irccloud.com) | 18:28 | |
| + josch (~josch@mister-muffin.de) | 18:28 | |
| + austriancoder (sid152545@id-152545.hampstead.irccloud.com) | 18:28 | |
| + aphistic (sid347194@id-347194.ilkley.irccloud.com) | 18:28 | |
| josch | okay, so I rebased the patches on top of 5.19 so that they apply cleanly again | 18:28 |
| + sknebel (~quassel@v22016013254630973.happysrv.de) | 18:28 | |
| josch | i temporarily disabled building reform-handbook because there is a latex error that i have to investigate | 18:29 |
| + jomo (~jomo@user/jomo) | 18:29 | |
| josch | but while the patches apply again, i have no idea whether i did the right things, maybe somebody could have a look especially at the changes to | 18:29 |
| josch | mnt3004-MNT-Reform-imx8mq-add-PHY_27M-clock.patch: https://source.mnt.re/reform/reform-debian-packages/-/commit/136cc425afbe95f78d7b2a77e62ae8a72165d038 | 18:30 |
| minute | josch: checking | 18:32 |
| minute | josch: why the clock name change to hdmi_phy_27m ? | 18:34 |
| - qbit (QUIT: Quit: WeeChat 3.6) (~qbit@h.suah.dev) | 18:35 | |
| minute | hmm i see, digging a bit | 18:36 |
| + S0rin (~S0rin@user/s0rin) | 18:37 | |
| vkoskiv_ | I messed up the MR, need to fix it sometime | 18:38 |
| minute | https://github.com/torvalds/linux/commit/cdc86e473b353c8a026a337ee9fb9e1fbbe2276b | 18:38 |
| vkoskiv_ | Git being git again. | 18:38 |
| vkoskiv_ | I rebased it and it rebased things from master to master. | 18:38 |
| minute | vkoskiv_: hey, does my battery status update code work? i didn't have a chance to test it :D | 18:38 |
| + qbit (~qbit@h.suah.dev) | 18:39 | |
| minute | josch: i think your patch is fine, just gitlab shows it strangely. this looks good to me https://source.mnt.re/reform/reform-debian-packages/-/blob/136cc425afbe95f78d7b2a77e62ae8a72165d038/linux/patches/mnt3004-MNT-Reform-imx8mq-add-PHY_27M-clock.patch | 18:43 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 18:49 | |
| - style- (QUIT: Quit: fatal error(); /bed/sleep/drea.ms) (the@bot.st) | 18:54 | |
| + ajr (~ajr@user/ajr) | 19:14 | |
| - smoothduck (QUIT: Quit: Connection closed for inactivity) (sid37490@id-37490.helmsley.irccloud.com) | 19:42 | |
| minute | josch: i've attempted to fix a problem on the build server, rerunning your pipeline | 20:01 |
| sigrid | minute: is there a way to turn off the power on the lcd panel using SW? | 20:06 |
| sigrid | looks like only 3.3v aux pwr is the one that powers it, but idk if it's possible to toggle it | 20:06 |
| minute | sigrid: hmm i think not currently, but you could either add that as a command to the keyboard fw or just send a black bitmap | 20:06 |
| minute | (via HID) | 20:07 |
| sigrid | wait. I'm talking about eDP :) | 20:07 |
| minute | ah sorry | 20:07 |
| minute | lol | 20:07 |
| minute | i think it should be possible to put it to sleep with dpms | 20:08 |
| minute | and there's a gpio to turn off the backlight | 20:08 |
| minute | maybe there's an i2c or mipi dcs command in the datasheet of the mipi->edp converter? | 20:09 |
| sigrid | there's an aux channel over i2c through the bridge | 20:13 |
| sigrid | the one to get edid. guess I could use that? | 20:13 |
| minute | i think so | 20:14 |
| sigrid | ok, that's a good pointer, thanks :) | 20:15 |
| minute | related, 8.3.3.2 Suspend Mode | 20:18 |
| minute | it says that asserting ~EN pin will lead to the biggest power savings... that's EDP_RESETn in the schematic. it's a gpio | 20:19 |
| minute | see 8.4.3 Power Down Sequence | 20:20 |
| sigrid | rn I just toggle VSTREAM_ENABLE on (un)blank | 20:22 |
| sigrid | I started with turning off lcdif dotclock altogether and made it work, but the panel was starting to produce really weird flickering at the top of the screen | 20:23 |
| sigrid | I did not find ways to solve it | 20:23 |
| sigrid | even turning the power off (or booting into linux) would not get rid of the flickering | 20:25 |
| sigrid | so not sure it's even related to lcdif or the bridge | 20:26 |
| sigrid | keeping backlight and pwm on would not help either | 20:26 |
| minute | the panel will retain the flickering yeah | 20:44 |
| minute | you can cause some long term weird oscillation in these panels | 20:45 |
| minute | only letting them lying around turned off for a while will make it go away | 20:45 |
| minute | (or maybe turned on with the correct timing) | 20:45 |
| vkoskiv_ | I really don't understand how Gitlab merge requests work. | 20:47 |
| vkoskiv_ | It shows completely wrong info on the MR page right now. | 20:47 |
| vkoskiv_ | It claims I'm 25 commits behind, which simply isn't the case | 20:48 |
| vkoskiv_ | I thought I messed it up, but I redid it and it still shows wrong info. | 20:48 |
| vkoskiv_ | My branch is one commit *ahead* master, not 25 behind. I choose to believe Git over GitLab. | 20:49 |
| vkoskiv_ | It's also showing a ton of changes I never did. | 20:49 |
| vkoskiv_ | Bizarre. I'll try a 3rd time, there is something messed up on my local copy too now. | 20:51 |
| josch | okay, there is a build failure of 5.19... investigating... | 20:52 |
| vkoskiv_ | Yeah, git won't pull the latest changes from master, it claims the latest commit is from august 22nd, which is also false. | 20:53 |
| vkoskiv_ | I'll just delete the repo and clone a fresh copy. Must be corrupted. | 20:53 |
| josch | but first getting my baby to bed -- here is the log in case others are faster than me: https://source.mnt.re/reform/reform-debian-packages/-/jobs/885/raw | 20:53 |
| minute | hmm, i don't really see what the actual error/problem is in the log | 21:05 |
| vkoskiv_ | Fixed my MR, it should be good to go now. | 21:08 |
| vkoskiv_ | For whatever reason, it would no longer pull new changes from upstream master branch. | 21:08 |
| vkoskiv_ | Just reset my local copy and now it makes more sense. | 21:08 |
| vkoskiv_ | I'm working on a neat little project, and this little change will go really nicely with that. | 21:09 |
| vkoskiv_ | But not absolutely required. | 21:09 |
| sigrid | minute: that is good to know (re panel) | 21:14 |
| vkoskiv_ | minute: Battery page does indeed update automatically now. Very neat! | 21:15 |
| minute | vkoskiv_: cool! i'm not sure if i thought of every case though. like, does it stop updating when you press the circle key again? | 21:17 |
| minute | (hmm, should) | 21:17 |
| vkoskiv_ | And I rebased my changes on top of that, fwiw. | 21:17 |
| vkoskiv_ | Let me check | 21:17 |
| vkoskiv_ | Well, circle takes me to the menu | 21:17 |
| vkoskiv_ | So yeah | 21:17 |
| minute | ok cool | 21:17 |
| vkoskiv_ | If you're wondering that it will update back to the battery view after pressing circle, then no | 21:18 |
| vkoskiv_ | It works as I would expect, at least. | 21:18 |
| minute | cool, thanks for testing! | 21:18 |
| vkoskiv_ | My changes to ignore OLED draw commands when using the menus are small | 21:18 |
| vkoskiv_ | A youtube demo video is in the MR description | 21:18 |
| vkoskiv_ | Not strictly required, but if you can merge it soon-ish, it'll go really neatly with this other thing I'm working on | 21:19 |
| vkoskiv_ | I really like this auto-refreshing battery menu. | 21:19 |
| vkoskiv_ | Ideally I'd like to shuffe some of the menu code a bit to make adding new views a bit easier. | 21:20 |
| vkoskiv_ | And to add an 'are you sure?' prompt to some of the options. | 21:20 |
| vkoskiv_ | Right now Circle+X doesn't show up in the menu, but if accidentally hit, borks the keyboard with no obvious way to bring it back (dfu-programmer) | 21:21 |
| minute | vkoskiv_: yeah, an are you sure prompt is missing | 21:21 |
| vkoskiv_ | So hitting that would ideally bring up a lil confirmation prompt. Same for LPC reset and *maybe* power off | 21:21 |
| minute | true, true | 21:24 |
| vkoskiv_ | minute: Does the Reform camera have a mic as well? | 21:31 |
| vkoskiv_ | Also, are you looking for firmware maintainers? :D | 21:33 |
| minute | vkoskiv_: no, it doesn't have a mic, as we think it's better to use a headset (wired or bt) | 21:36 |
| minute | vkoskiv_: yes, looking for fw maintainers. | 21:37 |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark__!~mark@gnu.wildebeest.org))) (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 22:49 | |
| * mark__ -> mjw | 22:49 | |
| + wielaard (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 22:50 | |
| vkoskiv_ | I can't say I'm an expert in embedded, but I know C moderately well. I | 22:56 |
| vkoskiv_ | I'd love to take a stab at maintaining some bit at some point. Validating and merging MRs, maintaining a stable branch, stuff like that. | 22:56 |
| minute | vkoskiv_: great, lets do it | 22:57 |
| vkoskiv_ | Perhaps the keyboard firmware? I don't know if I know enough to touch code that controls battery charging. | 22:58 |
| minute | yep, sounds good, we can start with that | 23:03 |
| vkoskiv_ | Cool! I guess I'd ping you here to ask if a suggested feature is okay to go in, then test & merge it if you give the OK? | 23:09 |
| * vkoskiv_ -> vkoskiv | 23:10 | |
| minute | vkoskiv: sounds good | 23:11 |
| vkoskiv | I'm also @vkoskiv on the gitlab | 23:13 |
| minute | vkoskiv: i'll assign you as a developer in the morning! | 23:13 |
| vkoskiv | Thanks! | 23:15 |
| vkoskiv | I've tested both my MRs extensively, so I might merge both if you're okay with the changes. The disconnected battery startup MR I'll have to test. | 23:16 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 23:49 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!