| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@129.110.242.224) | 00:23 | |
| + jacobk (~quassel@utdpat242016.utdallas.edu) | 00:35 | |
| - mjw (QUIT: Ping timeout: 258 seconds) (~mjw@gnu.wildebeest.org) | 00:41 | |
| - jacobk (QUIT: Ping timeout: 240 seconds) (~quassel@utdpat242016.utdallas.edu) | 00:45 | |
| + vagrantc (~vagrant@2a00:79e1:abc:133:77d:8d2b:ed76:33d) | 00:48 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 00:54 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2a00:79e1:abc:133:77d:8d2b:ed76:33d) | 01:21 | |
| - jacobk (QUIT: Ping timeout: 240 seconds) (~quassel@utdpat241106.utdallas.edu) | 01:28 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 01:29 | |
| - mtm (QUIT: Ping timeout: 245 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:02 | |
| - ajr (QUIT: Quit: Connection closed for inactivity) (uid609314@user/ajr) | 02:21 | |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@utdpat241106.utdallas.edu) | 02:37 | |
| + reform29030 (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 03:33 | |
| reform29030 | minute: no worries, i got kvm to work on ls1028a -- it just is done differently compared to imx8mq :) | 03:34 |
|---|---|---|
| reform29030 | pandora: blender support depends on opengl support of the som | 03:35 |
| reform29030 | pandora: starting with 2.80, blender requires 3.2+, so 2.79b is the last one that works on reform with imx8mq | 03:36 |
| reform29030 | pandora: we tried to forward port old blender for a while but gave up doing that a while ago | 03:36 |
| - reform29030 (QUIT: Quit: Leaving) (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 03:39 | |
| - nsc (QUIT: Ping timeout: 255 seconds) (~nicolas@34-49-142-46.pool.kielnet.net) | 04:00 | |
| + nsc (~nicolas@201-98-142-46.pool.kielnet.net) | 04:01 | |
| + reform32580 (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 04:02 | |
| reform32580 | minute: sysimage-v4 has a /debian directory in /. It comes from firedecor_0.1~2023-10-10-git7daf1ec_arm64.deb | 04:02 |
| reform32580 | minute: https://mntre.com/modularity.html#table used to work but now the #table anchor is gone somehow? i found that useful to point people to | 04:06 |
| reform32580 | minute: also shop pages like https://shop.mntre.com/products/mnt-reform-ls1028a-module link to it | 04:06 |
| reform32580 | pandora: according to that table, you'll need ls1028a with an external pcie gpu (probably in a rack reform?) for blender support as blender needs opengles 3.2 and all boards have opengles 3.1 maximum -- no idea whether hacks exist that make it work anyways | 04:08 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:09 | |
| - reform32580 (QUIT: Quit: Leaving) (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 04:09 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 04:33 | |
| pandora | oh I also saw that opengl dropped development in 2017 | 04:35 |
| pandora | the latest blender seem to even need opengl 4.3 | 04:38 |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@utdpat241106.utdallas.edu) | 06:04 | |
| + jacobk (~quassel@64.189.201.150) | 06:28 | |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 07:00 | |
| + reform3033 (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 07:26 | |
| reform3033 | df -h /tmp | 07:26 |
| reform3033 | Filesystem Size Used Avail Use% Mounted on | 07:26 |
| reform3033 | tmpfs 16G 13G 3.9G 77% /tmp | 07:27 |
| reform3033 | oh the joy! | 07:27 |
| - marty_mcfly88 (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~quassel@065-026-065-202.biz.spectrum.com) | 07:46 | |
| + marty_mcfly88 (~quassel@065-026-065-202.biz.spectrum.com) | 07:47 | |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@64.189.201.150) | 08:01 | |
| - reform3033 (QUIT: Quit: Leaving) (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 08:14 | |
| vkoskiv | Curious, suddenly got this error on boot, wifi disabled: https://mister-muffin.de/p/pbmC.txt | 08:53 |
| vkoskiv | I don't think I ran an apt upgrade or anything | 08:54 |
| vkoskiv | I'll try to run one, see if that helps | 08:54 |
| vkoskiv | I ran an apt upgrade. Still that same error, but wifi works now :D | 09:02 |
| vkoskiv | On the previous boot, it wasn't even showing up in rfkill | 09:03 |
| vkoskiv | I'll trawl the log to see if I can spot the actual cause | 09:03 |
| + reform12882 (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 09:10 | |
| reform12882 | vkoskiv: don't scare me like that! :D when you wrote about an error on boot i feared i had created another bug that breaks your machine :D | 09:11 |
| reform12882 | vkoskiv: the muffin paste bug you found was just a case of -ENOSPC btw :) | 09:12 |
| reform12882 | as you already found out it is fixed now | 09:12 |
| reform12882 | on debian you can also use the pastebinit tool to send stuff to a pastebin from a terminal | 09:13 |
| reform12882 | minute: where can i read my cpu temp on ls1028a? | 09:17 |
| reform12882 | /sys/devices/virtual/thermal/thermal_zone0/temp says 84000 | 09:17 |
| reform12882 | sorry for coming off and on all the time -- i'm on ls1028a via serial and am reading the channel log via http :) | 09:18 |
| + mjw (~mjw@gnu.wildebeest.org) | 09:51 | |
| - mjw (QUIT: Ping timeout: 258 seconds) (~mjw@gnu.wildebeest.org) | 10:15 | |
| * mark_ -> mjw | 10:20 | |
| minute | reform12882: i just use "sensors" | 10:26 |
| minute | reform12882: that's quite a high temp. heatsink is on? | 10:26 |
| + mtm- (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 10:28 | |
| - mtm (QUIT: Ping timeout: 240 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 10:30 | |
| pandora | is it correct that the GC7000 has full mesa support? | 10:42 |
| + mark_ (~mjw@gnu.wildebeest.org) | 12:05 | |
| reform12882 | minute: yup. maybe i didnt seat it very well. idle temp is 67°C | 12:11 |
| reform12882 | okay, i tested the reform-tools changes and ran some benchmarks. everything went well. maybe i should carefully attach the display and run some more things now... | 12:12 |
| reform12882 | wish me luck with the cable! | 12:12 |
| - reform12882 (QUIT: Quit: Leaving) (~josch@ip1f12fe3b.dynamic.kabel-deutschland.de) | 12:13 | |
| josch | and finally back to imx8mq... | 13:03 |
| josch | phew... | 13:03 |
| josch | swapped ls1028a in and out four times now | 13:03 |
| josch | everything still works, gladly | 13:03 |
| minute | josch: cool, i'm happy to hear this. module yield is low so far, so i hope yours has a long life | 13:14 |
| - iank (QUIT: Quit: ZNC 1.8.2+deb2+b1 - https://znc.in) (~iank@fsf/staff/iank) | 13:25 | |
| + iank (~iank@fsf/staff/iank) | 13:27 | |
| josch | minute: the ls1028a really seems to have been a problem child from start to finish :( | 13:31 |
| - mark_ (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org) | 13:31 | |
| josch | i think the heatsink was just a bit off last time -- it's difficult to install properly because it's really hard to determine whether the screws are completely in or just very hard to turn further | 13:32 |
| Boostisbetter | josch: was the performance improvement very obvious? | 13:35 |
| josch | Boostisbetter: depends on what you are doing :) | 13:41 |
| josch | Boostisbetter: for example, here are some benchmarks of my SSD on ls1028a: https://community.mnt.re/t/does-drive-encryption-cause-the-reform-to-slowdown/871/10 | 13:56 |
| - mtm- (QUIT: Ping timeout: 245 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:03 | |
| josch | i also learned: | 14:04 |
| josch | the LS1028A is not happy with my PLA casemods | 14:05 |
| josch | ah the other way round | 14:05 |
| josch | my 3d printed case mods are not happy with the heat from the LS1028A | 14:05 |
| josch | lesson learned: next time print everything in PET instead | 14:05 |
| vkoskiv | As opposed to ABS? ABS has a higher glass transition temperature than PLA | 14:11 |
| vkoskiv | Oh, I misread PET as PLA | 14:11 |
| vkoskiv | PLA is very melty :D | 14:12 |
| josch | yup, it even died in the sun for me already :D | 14:14 |
| minute | josch: yes, the ls1028a project is the hardest/most bad-luck-ridden we've had so far | 14:19 |
| josch | i hope happier days arrive now with pocket reform development :) | 14:20 |
| minute | i hope so! i'm not off the hook with ls though | 14:20 |
| minute | will need to tell some people we can't ship their devices now, again | 14:20 |
| minute | and lets hope that from the full batch (for which i'm still missing the socs) enough will be ok to cover orders | 14:21 |
| josch | uff :/ | 14:21 |
| josch | and it's not even like the efforts would pay off by delivering the does-it-all platform that is better than the options :/ | 14:22 |
| josch | ah sadly i found out that indeed the cpu is not fast enough to decode h264 1080p@60 -- but no big deal as i want to use it with a dedicated gpu in the rack reform anyways | 14:23 |
| - marty_mcfly88 (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~quassel@065-026-065-202.biz.spectrum.com) | 14:31 | |
| + marty_mcfly88 (~quassel@2603-6000-8c01-0f3e-7dac-14c7-19c1-fd5c.res6.spectrum.com) | 14:32 | |
| josch | Boostisbetter: as a benchmark, take a software i maintain which creates Debian chroots as an example. Running that software to create a chroot on imx8mq takes 46 seconds. On ls1028a it's just 29 seconds. This is a massive boost but the reason for that is, that creating a chroot is a process that only utilizes a single core and becomes much faster with /tmp mounted as a tmpfs (which i'm not doing on | 14:47 |
| josch | imx8mq due to limited ram). | 14:47 |
| josch | so as the maintainer of this software with a testsuite that runs that program 300+ times, i get a massive speed boost | 14:48 |
| josch | it would probably be even faster on a311d | 14:48 |
| josch | i still have 55 lines of notes to go through from my ls1028a tests -- more benchmarks results will be coming :) | 14:49 |
| vkoskiv | I could let you SSH into my reform with a fresh SD card to test :D | 14:49 |
| josch | vkoskiv: you could just "sudo apt install mmdebstrap" and then run "mmdebstrap --variant=apt unstable /dev/null" and see how fast it is :) | 14:50 |
| josch | vkoskiv: another benchmark that i did was to compile linux as done by reform-debian-packages -- if your machine has around six hours of free-time that's also something you could try :) | 15:02 |
| erle | josch thanks for the eremit suggestion | 15:02 |
| - GNUmoon (QUIT: Ping timeout: 252 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 15:09 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 15:15 | |
| * truck -> lastebil | 16:07 | |
| * lastebil -> truck | 16:07 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:09 | |
| + IchikaZou (~IchikaZou@36-231-67-151.dynamic-ip.hinet.net) | 16:19 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 16:26 | |
| - GNUmoon (QUIT: Ping timeout: 252 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 16:32 | |
| - S0rin (QUIT: Ping timeout: 255 seconds) (~S0rin@user/s0rin) | 16:33 | |
| + S0rin (~S0rin@user/s0rin) | 16:34 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 16:37 | |
| - mark_ (QUIT: Ping timeout: 245 seconds) (~mjw@gnu.wildebeest.org) | 16:43 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 16:51 | |
| - ex-parrot (QUIT: Read error: Connection reset by peer) (~fincham@user/ex-parrot) | 17:05 | |
| + ex-parrot (~fincham@user/ex-parrot) | 17:05 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:24 | |
| - Boostisbetter (QUIT: Ping timeout: 255 seconds) (4a410829d7@irc.cheogram.com) | 18:01 | |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 18:01 | |
| minute | josch: interesting how sata performance on ls1028a is better than nvme on imx8mq | 18:07 |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:19 | |
| - IchikaZou (QUIT: Remote host closed the connection) (~IchikaZou@36-231-67-151.dynamic-ip.hinet.net) | 18:24 | |
| + vagrantc (~vagrant@2a00:79e1:abc:133:77d:8d2b:ed76:33d) | 18:24 | |
| josch | oh then i made an interesting benchmark \o/ | 18:26 |
| josch | for me, either of those results is more than enough for what i'm doing with the machine :D | 18:26 |
| + jacobk (~quassel@64.189.201.150) | 18:31 | |
| - stites (QUIT: Ping timeout: 252 seconds) (~stites@130.44.147.204) | 18:52 | |
| + stites (~stites@2607:fb91:dc8:e0e2:5656:70eb:8498:35ff) | 18:52 | |
| abortretryfail | hmm... where does /usr/lib/systemd/sleep.conf.d/reform-sleep.conf come from? dpkg-query doesn't think its from any package. | 18:53 |
| abortretryfail | does that get baked into the system image somehow? | 18:54 |
| josch | abortretryfail: what is the output of | 18:54 |
| josch | dpkg-query --search /usr/lib/systemd/sleep.conf.d/reform-sleep.conf | 18:54 |
| abortretryfail | that's exactly what i did. | 18:54 |
| abortretryfail | dpkg-query: no path found matching pattern /lib/systemd/sleep.conf.d/reform-sleep.conf | 18:55 |
| abortretryfail | that's the file that explicitly disables hibernate. I changed it on my box while I try to get hibernate working. | 18:55 |
| josch | huh | 18:55 |
| abortretryfail | right now it looks like encrypted swap on the nvme is a Problem, but it might be possible to hibernate a SD card booted Reform | 18:55 |
| josch | ooooh | 18:56 |
| abortretryfail | I need to make a SD card with the base system image and a large enough swap partition and try | 18:56 |
| josch | i can tell you what is going on | 18:56 |
| josch | it's the merged-/usr mess we are going through right now | 18:56 |
| josch | so you didn't run the command i told you | 18:56 |
| josch | you ran it with /lib/systemd/sleep.conf.d/reform-sleep.conf | 18:56 |
| josch | not with /usr/lib/systemd/sleep.conf.d/reform-sleep.conf | 18:56 |
| abortretryfail | oh god lol | 18:56 |
| abortretryfail | sorry | 18:56 |
| josch | dpkg knows about the latter not the former | 18:56 |
| josch | but due to the symlink they are the same file | 18:57 |
| abortretryfail | reform-tools owns that yeah. | 18:57 |
| Boostisbetter | Hmm, I am not encrypted at all on my Reform. I wonder if it would work for me. | 18:57 |
| Boostisbetter | Any chance you could detail in the thread what you have tried doing to get it working? I know that grub needs to be update to resume from the swap partition. | 18:58 |
| abortretryfail | Boostisbetter: - maybe. You'll need a kernel cmdline to actually resume from power off. There is no grub. It'll have to be u-boot stuff | 18:58 |
| josch | abortretryfail: how did you find that? did it cause any problem for you? | 18:59 |
| abortretryfail | how did i find which part? | 18:59 |
| Boostisbetter | abortretryfail: ok, sounds good. u-boot is pretty nice, and i would think it would support that. I trust you have already set that up with u-boot? | 18:59 |
| abortretryfail | no, I'm not messing with the bootloader until I get a successful test resume from swap | 18:59 |
| josch | abortretryfail: you were asking where it came from -- was it just because you used dpkg-query and it didn't find it? | 19:00 |
| josch | or was there another problem with this? | 19:00 |
| abortretryfail | you can do echo test_resume > /sys/power/disk then systemctl hibernate to try | 19:00 |
| abortretryfail | josch: I was looking at the symlink and didnt realize it was a symlink | 19:00 |
| josch | okay :) | 19:00 |
| abortretryfail | Boostisbetter: also if you test, make sure /sys/power/resume has the major:minor number of your swap device (check lsblk) first. | 19:02 |
| - stites (QUIT: Read error: Connection reset by peer) (~stites@2607:fb91:dc8:e0e2:5656:70eb:8498:35ff) | 19:33 | |
| + stites (~stites@130.44.147.204) | 19:33 | |
| - jacobk (QUIT: Ping timeout: 248 seconds) (~quassel@64.189.201.150) | 19:38 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2a00:79e1:abc:133:77d:8d2b:ed76:33d) | 19:52 | |
| + jacobk (~quassel@129.110.242.224) | 19:59 | |
| abortretryfail | well, I got it to hibernate and *mostly* recover from it. the nvme blows up and it gets stuck trying to read thermal sensors after resume now. | 20:06 |
| abortretryfail | maybe i will mess with this more later. I need to get real work done. | 20:07 |
| Boostisbetter | abortretryfail: I think you are real close. Which is awesome! Hope real work doesn't drain ya too much! Looking forwarding to seeing if you get nay futher. | 20:17 |
| abortretryfail | I think I'm real close to running out of easy stuff. | 20:18 |
| abortretryfail | nvme and the sensors thing is getting into hardware hairyness that smells like uncharted waters for embedded ARM boards. | 20:18 |
| - mtm (QUIT: Remote host closed the connection) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 20:21 | |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@129.110.242.224) | 20:23 | |
| + jacobk (~quassel@utdpat241106.utdallas.edu) | 20:27 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.32.204) | 20:31 | |
| + Ar|stote|is (~linx@149.210.32.204) | 20:32 | |
| minute | josch: i downloaded the latest system image for imx8mq and am testing it in QA with a reform to ship, and the display turns off in linux (it's there in uboot) | 20:36 |
| minute | wat > [ 20.447101] platform 30830000.spi: deferred probe pending | 20:38 |
| minute | ah, now maintenance mode (on serial) | 20:38 |
| minute | ls: cannot access '/dev/fb*': No such file or directory | 20:39 |
| minute | [ 8.637458] imx-dcss 32e00000.display-controller: [drm] *ERROR* fbdev-dma: Failed to setup generic emulation (ret=-12) | 20:39 |
| minute | [ 8.648933] imx-dcss 32e00000.display-controller: [drm] *ERROR* fbdev-dma: Failed to setup generic emulation (ret=-12) | 20:39 |
| minute | this seems new | 20:39 |
| minute | josch: new kernel shenanigans? | 20:39 |
| minute | oh shit what | 20:41 |
| minute | this module claims 2GB ram | 20:41 |
| minute | that's the explanation then. cma=512M@3G can't work on this module | 20:42 |
| minute | maybe we should do cma=512M@1G | 20:42 |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 21:06 | |
| * mark_ -> mjw | 21:06 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 21:07 | |
| vkoskiv | josch: 31.96s on first cold run for that mmdebstrap cmd | 21:15 |
| vkoskiv | minute: Is that on the ls1028? | 21:17 |
| vkoskiv | josch: Ran same command with hyperfine, 10 runs, mean 31.312s +- 0.122s | 21:22 |
| vkoskiv | Range as 31.004s...31.382s | 21:22 |
| vkoskiv | User: 19.667s, System 13.530s | 21:22 |
| vkoskiv | I guess the first manual cold run populated some caches, if the slowest run here was much faster than it. | 21:23 |
| minute | vkoskiv: no, imx8mq | 21:32 |
| vkoskiv | > uptime 5:45 | 21:55 |
| vkoskiv | Just ticked over to 10% on the gauge there | 21:55 |
| vkoskiv | Reading azoteq docs, good tunes on wireless headphones. | 21:57 |
| minute | vkoskiv: 5:45h runtime on batteries? | 22:04 |
| vkoskiv | Not 100% sure of that figure, I don't remember if I had it on the wire for a bit at the start | 22:04 |
| minute | ah ok :D | 22:04 |
| vkoskiv | But most of the time, it's been on the battery.\ | 22:04 |
| vkoskiv | Wanted to drain it down. I'll do a measurement tomorrow | 22:05 |
| vkoskiv | Or maybe I'll set it to echo the uptime to a file in a loop or something | 22:05 |
| minute | vkoskiv: just don't forget `sync` in that script :D | 22:07 |
| minute | (did that a few times) | 22:07 |
| vkoskiv | Yeah, can have an oopsie with that | 22:09 |
| vkoskiv | Although the last time my unit did a panic shutdown, it looked like a safe-ish one? | 22:10 |
| vkoskiv | Dropped to a console, bunch of text, etc. | 22:10 |
| vkoskiv | I think that was linux initiating it, it was when the fuel gauge was doing the weird thing | 22:10 |
| josch | vkoskiv: are you using an apt mirror in the network or on your local disk? | 22:52 |
| vkoskiv | I don't know, I just invoked the command you showed | 22:54 |
| josch | ah okay | 22:54 |
| josch | then it probably took some seconds downloading packages | 22:54 |
| josch | i benchmarked the command with my apt mirror proxied from the local drive | 22:54 |
| josch | so that my network speed would not become part of the benchmark | 22:54 |
| josch | if it took the same time *including* the downloading, then a311d is massively faster than ls1028a for this task :) | 22:56 |
| vkoskiv | Curious to hear opinions, this is what my OLED looks like now: https://www.youtube.com/watch?v=67x9bpDkjyc | 22:57 |
| sigrid | looks cool | 23:00 |
| vkoskiv | the lissajous gets very excited and spinny when the CPU gets busy :^) | 23:00 |
| sigrid | that reminded me to make oled brightness configurable via raw hid. maybe it's already there | 23:00 |
| vkoskiv | I did implement that, but it doesn't get that dim in my experience | 23:01 |
| vkoskiv | I also tried out syncing it with the kbd backlight, but I prefer to have mine on the lowest brightness all the time, it's already plenty bright | 23:01 |
| josch | vkoskiv: looks super impressive! | 23:02 |
| josch | i'd still fear burn-in effects on my oled though :( | 23:02 |
| vkoskiv | Yeah, I'll try to work out something for that | 23:02 |
| vkoskiv | The frames aren't really necessary, mostly just for testing | 23:03 |
| vkoskiv | And I could shift things around every so often, down, right, up, left | 23:03 |
| vkoskiv | Something like that | 23:03 |
| vkoskiv | But in my experience, even if the OLED does get burned in, you only notice it if you look really closely, or the display is all white | 23:04 |
| vkoskiv | More hacking to follow tomorrow | 23:04 |
| josch | i like the idea of moving some of the information i currently have in my waybar to the oled | 23:04 |
| Boostisbetter | that would be pretty cool. Get a bit more use out of it. I kind of wish the OLED could show things like battery percentage in real time, and keep updating. | 23:10 |
| josch | Boostisbetter: it already does that | 23:11 |
| josch | the display you get when pressing circle, b is updated every few seconds | 23:11 |
| josch | another benchmark (the mmdebstrap testsuite): 522m on imx8mq and 402m on ls1028a -- that's ~9 hours versus ~7 hours | 23:31 |
| - natalie- (QUIT: Read error: Connection reset by peer) (~natalie@user/natalie) | 23:35 | |
| josch | vkoskiv: i'm now running the reform-debian-packages linux build -- i'd love to see how long that takes on a311d as well :) | 23:37 |
| josch | my existing logs suggest that compiling linux on ls1028a is just as fast as on imx8mq which would make sense because in contrast to the other workloads i've tested, building linux benefits a lot from having more cores available | 23:39 |
| + natalie (~natalie@user/natalie) | 23:40 | |
| - Boostisbetter (QUIT: Ping timeout: 255 seconds) (4a410829d7@irc.cheogram.com) | 23:58 | |
| + natalie- (~natalie@user/natalie) | 23:58 | |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 23:59 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!