| doctorhoo | mntmn: is SysRq configured in the keyboard fw by default and on which key is it in that case? | 00:04 |
|---|---|---|
| mntmn | doctorhoo: haven't looked into how that works! | 00:05 |
| - mtm (QUIT: Ping timeout: 256 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 01:03 | |
| doctorhoo | for anyone interested in testing the no_console_suspend kernel parameter and whether it helps with suspend issues: you can enable that option at runtime by running the command | 01:03 |
| doctorhoo | echo N | sudo tee /sys/module/printk/parameters/console_suspend | 01:03 |
| mntmn | doctorhoo: perfect, thanks! | 01:04 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:21:21:0:100e) | 01:42 | |
| - nsc (QUIT: Ping timeout: 256 seconds) (~nicolas@i5C7445E1.versanet.de) | 03:09 | |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 03:09 | |
| + nsc (~nicolas@i5C744B37.versanet.de) | 03:11 | |
| * nsc -> Guest6673 | 03:12 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:21:21:0:100e) | 04:49 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149-210-116-38.mobile.ren.cosmote.net) | 05:14 | |
| + Ar|stote|is (~linx@149-210-116-38.mobile.ren.cosmote.net) | 05:14 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:21:21:0:100e) | 05:17 | |
| + MajorBiscuit (~MajorBisc@86-88-79-148.fixed.kpn.net) | 08:37 | |
| - MajorBiscuit (QUIT: Client Quit) (~MajorBisc@86-88-79-148.fixed.kpn.net) | 08:42 | |
| + MajorBiscuit (~MajorBisc@86-88-79-148.fixed.kpn.net) | 08:42 | |
| + mjw (~mark@gnu.wildebeest.org) | 11:04 | |
| + reform29697 (~ckeen@2001:a62:146e:ff01:6f0:21ff:fe91:8a6) | 11:29 | |
| reform29697 | hi! | 11:29 |
| reform29697 | After an upgrade of the debian sid packages I get a linker error when starting waybar. Is there a way to rebuild the packaged binaries from the image in /usr/local/bin? | 11:30 |
| doctorhoo | reform29697: I got the waybar working again by installing the corresponding package via apt and changing the path in the sway config from /usr/local/bin to /usr/bin -- I saw no issues with that so far. | 11:38 |
| - mjw (QUIT: Quit: Leaving) (~mark@gnu.wildebeest.org) | 11:42 | |
| reform29697 | doctorhoo: ah ok I was just wondering whether those contain patches that aren't in upstream | 11:44 |
| doctorhoo | I believe this was mostly about the version being shipped with debian used to be too old. That is no longer the case, and therefore the move to stock packages is on the way: https://source.mnt.re/reform/reform-system-image/-/issues/15 | 11:48 |
| reform29697 | another thing: the rofi reform-theme does not show the hilight on the selection anymore. Is that a known problem too? | 11:49 |
| doctorhoo | In the reform-system-image repository, you find several mkuserland scripts that compile those packages if you are curious or want to reproduce a step. | 11:50 |
| doctorhoo | reform29697: now that you mention it, I think I have seen that too. | 11:51 |
| reform29697 | doctorhoo: for now I have changed it to a theme in /usr/share/rofi/themes... | 11:58 |
| reform29697 | following upstream debian here is more sustainable I guess | 11:59 |
| + Christoph_ (~Christoph@p54bf6199.dip0.t-ipconnect.de) | 12:07 | |
| + Major_Biscuit (~MajorBisc@c-001-029-007.client.tudelft.eduvpn.nl) | 12:32 | |
| - MajorBiscuit (QUIT: Ping timeout: 240 seconds) (~MajorBisc@86-88-79-148.fixed.kpn.net) | 12:35 | |
| - reform29697 (QUIT: Ping timeout: 268 seconds) (~ckeen@2001:a62:146e:ff01:6f0:21ff:fe91:8a6) | 12:52 | |
| + reform29697 (~ckeen@aftr-82-135-86-75.dynamic.mnet-online.de) | 12:53 | |
| - mtm (QUIT: Ping timeout: 256 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 13:03 | |
| - reform29697 (QUIT: Quit: Leaving) (~ckeen@aftr-82-135-86-75.dynamic.mnet-online.de) | 13:11 | |
| mntmn | this is a known problem and easy to fix: > the rofi reform-theme does not show the hilight on the selection anymore | 14:00 |
| + mjw (~mark@gnu.wildebeest.org) | 14:06 | |
| - qbit (QUIT: Quit: WeeChat 3.3) (~qbit@ns2.suah.dev) | 14:13 | |
| josch | Okay, I feel super stupid. I cannot get the serial connection to work. I connected GND, RX, TX of my FTDI232 set to 3.3V https://mister-muffin.de/p/M8dC.jpg to the S1/J18 of the Reform https://mister-muffin.de/p/t7No.jpg and ran `sudo screen /dev/ttyUSB0 115200` but nothing shows up once the reform boots. Any ideas what I could've messed up? Yes, RX is connected to TX and TX to RX. | 14:39 |
| bluerise | Maybe you need to disable hardware flow control? Though input should work regardless, I thought | 14:45 |
| bluerise | Wiring looks fine, baudrate is fine | 14:45 |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 15:09 | |
| bluerise | mntmn: congrats, that was quick! guess the pins on the connector will be a bit harder to test | 15:21 |
| Boostisbetter | So what does your guys' typical workflow look like on the Reform. I normally work across 5 workspaces, and have 7 programs at once. | 15:33 |
| + qbit (~qbit@ns2.suah.dev) | 15:37 | |
| Boostisbetter | I just recently started taking my Reform to work with me, where I work on while on my lunch break. My day is about 9 hours plus about 40 minutes of driving round trip. I have been seeing remaining battery at that point at around 20-30% This is mainly standby time with about 1-2 hours of use in between. | 15:46 |
| Boostisbetter | Anybody else bringing there's places? What kind of battery life are you seeing? | 15:46 |
| - doctorhoo (QUIT: Ping timeout: 240 seconds) (~hanno@194-18-252-127-no2005.tbcn.telia.com) | 16:30 | |
| technomancy | I don't plan on taking mine places until I can resume from suspend | 16:53 |
| technomancy | I haven't measured specifically how long mine lasts but a couple days ago when the power went out for 2-3 hours it was enough to drain mine from 100% to zero with the screen off | 16:56 |
| sigrid | mine was >5h of usual work with heavy browsing, ~25% battery left after that | 16:58 |
| sigrid | should be noted I am mostly using it as a terminal to a way more powerful machine. just a bit of actualy compilation on reform itself | 16:58 |
| sigrid | s/actualy/actual/ | 16:58 |
| sigrid | today I am trying the same thing, but with exact start time of measurement known, and will go closer to 0 | 17:01 |
| mntmn | josch: tried to swap rx/tx anyway? | 17:18 |
| josch | mntmn: I now did. It didn't change anything. I also tried a different cable. This is probably just something very stupid... | 17:43 |
| - Major_Biscuit (QUIT: Quit: WeeChat 3.3) (~MajorBisc@c-001-029-007.client.tudelft.eduvpn.nl) | 18:40 | |
| + doctorhoo (~hanno@194-18-252-127-no2005.tbcn.telia.com) | 18:58 | |
| Guest6673 | I'm using my reform as "thin client" for work, using RDP via openconnect vpn tunnelled by ssh. Didn't measure battery life, but it does hold for about 5-7 hours. | 19:25 |
| Guest6673 | About battery life with real local workflow, I can't tell anything yet. | 19:28 |
| Boostisbetter | technomancy, it does for the most part. Ruff's script and integrating it with systemd has improved it dramatically. | 19:44 |
| Boostisbetter | I have supended over 70 times at this point and only had 1 time that it didn't resume. | 19:45 |
| Boostisbetter | I wouldn't bring it to work with me if it WASN'T suspending properly first. | 19:45 |
| Boostisbetter | when suspended the battery lasts just under 12 hours. | 19:45 |
| Boostisbetter | that was the little informal test I did | 19:45 |
| + illo (~illo@user/illo) | 19:49 | |
| + reform19145 (~ng@2a02:8084:903:d400:6f0:21ff:fe91:3e7) | 19:59 | |
| - reform19145 (QUIT: Client Quit) (~ng@2a02:8084:903:d400:6f0:21ff:fe91:3e7) | 20:03 | |
| technomancy | Boostisbetter: that sounds awesome; where can I find this script? | 20:05 |
| technomancy | wait, is this one of the things that requires replacing the OS? | 20:05 |
| Boostisbetter | ruff, posted about it on the standby thread on the MNT forum. | 20:05 |
| Boostisbetter | and no, you don't replace the OS at all. | 20:06 |
| technomancy | ah ok cool; I heard one other way to improve resume was to switch to alpine but I wasn't too keen on that | 20:06 |
| Boostisbetter | it is a modified version of the the reform-stanby script, that is then integrated with systemd | 20:06 |
| Boostisbetter | so that you can trigger it with a keybinding in sway | 20:06 |
| Boostisbetter | it works excellent | 20:06 |
| Boostisbetter | I mean it is the reason that I am actually taking it with me to work these days. | 20:21 |
| josch | mntmn: so I just connected the same FTDI232 adapter with the same cables to a raspberry pi board and ran the same screen command. I can clearly see the pi booting in screen. When connecting the GND/TX/RX to the reform with an otherwise equal setup I see nothing. So the FTDI is fine, the cables are fine and the screen command is fine. But when replacing the raspberry pi by the reform I don't see when | 20:21 |
| josch | booting. I'm using the latest rescue image from https://source.mnt.re/reform/reform-system-image/-/pipelines on my SD card. | 20:21 |
| bluerise | but the system is booting/ | 20:29 |
| bluerise | ? | 20:29 |
| bluerise | U-Boot definitely always prints something on the UART lines | 20:29 |
| josch | yes, the system boots up and i can log in as root | 20:29 |
| josch | is the kernel also printing on uart? | 20:30 |
| bluerise | not sure, I'm not using Linux, but U-Boot definitely prints | 20:30 |
| josch | :/ | 20:30 |
| bluerise | but I think kernel will also print and provide login prompt | 20:30 |
| josch | yeah, the u-boot bootargs seem to suggest that it does: console=ttymxc0,115200 console=tty1 | 20:32 |
| sigrid | linux does print on uart as well | 20:37 |
| sigrid | my only issue with uart was that I couldn't stop uboot from automatically booting the kernel, which was fixed by replacing the usb cable of the serial interface | 20:38 |
| sigrid | otherwise it worked just fine right away. I used bus pirate v4 | 20:39 |
| - illo (QUIT: Quit: WeeChat 3.3) (~illo@user/illo) | 20:44 | |
| mntmn | josch: maybe there is a problem with the GND pin on the S1 header, can you check underneath? i had one board where it wasn't soldered properly | 21:35 |
| mntmn | sorry i'm a bit distracted by bringing up the first rpi cm4 based reform | 21:36 |
| josch | mntmn: don't worry about not replying quickly -- i'm also very exciting about a rpi cm4 option so i'm happy that you spend time on it :) | 21:37 |
| josch | and there is no time pressure here -- thanks for the hint, i'll check it out | 21:37 |
| josch | I cannot see anything wrong with the soldering of S1 and S2 on the bottom of the board. | 22:11 |
| + vagrantc (~vagrant@2600:3c01:e000:21:21:21:0:100e) | 22:23 | |
| sigrid | so, exactly 6h without charger with heavy firefox browsing, music playing (files mounted over sshfs), some video watching, some c++ compilation → 6% charge left | 22:33 |
| Boostisbetter | That is pretty good. Firefox is a hog especially when heavily used. speakers use a lot of power as well, and keeping the CPU pegged is power intensive. 6 hours is great, actually. | 22:35 |
| vagrantc | curious what the cell voltages are at 6% | 22:36 |
| Boostisbetter | me too. Typically speaking it isn't good for Lipo to get that low. | 22:37 |
| vagrantc | these are lifepo4, which are very different | 22:37 |
| sigrid | vagrantc: 3.1, a few 3.0 | 22:38 |
| vagrantc | but still, don't want them dropping below 2.5v if you can help it | 22:38 |
| vagrantc | that's not bad at all | 22:38 |
| sigrid | it was mostly very very equal | 22:38 |
| sigrid | already charged it to 15%, allegedly | 22:38 |
| technomancy | I don't really trust the percentage readout because it seems to always go immediately to 74% within a minute or two of removing the power supply | 22:45 |
| technomancy | is there anything I can do to improve that calibration? | 22:46 |
| mntmn | technomancy: sounds like a bug | 22:46 |
| mntmn | technomancy: you can file a bug with your observations against the reform repo | 22:46 |
| technomancy | sure; I'll see if I can gather a bit more data | 22:47 |
| technomancy | I figured it was just something that would calibrate over time initially but it doesn't seem to have changed much | 22:47 |
| vagrantc | i do notice it typically drawing a lot more power during the firstbit of charge | 22:49 |
| vagrantc | each cell can maybe take 1-1.5 amps during charge? or is it configured for less to increase battery life expectancy | 22:51 |
| mntmn | that sounds about right | 22:59 |
| - Ar|stote|is (QUIT: Read error: Connection reset by peer) (~linx@149-210-116-38.mobile.ren.cosmote.net) | 22:59 | |
| + Ar|stote|is (~linx@149-210-116-38.mobile.ren.cosmote.net) | 23:00 | |
| vagrantc | that would be ~25-38W of charging (8 cells at 3.2 volts at 1-1.5 amps) ... that would suggest from complete discharge maybe 1-2 hours ? but i've definitely noticed the charger spiking up to 60w early on in the charge | 23:07 |
| mntmn | ok mouser official purchase order is there | 23:08 |
| vkoskiv_ | *excitement noises* | 23:09 |
| vkoskiv_ | I am probably going to add a reform-specific feature to my raytracer where it displays render progress and status on the oled display | 23:09 |
| mntmn | nice :D | 23:11 |
| vagrantc | so the CM4 module is actually working? that's pretty exciting for various forms of upgradability ... | 23:11 |
| mntmn | yes, even if it did not mechanically fit | 23:11 |
| mntmn | (i bodged it with some wires) | 23:11 |
| vagrantc | oh no! | 23:12 |
| * jackhill -> KM4MBG | 23:12 | |
| mntmn | yeah it was a silly mistake, i ordered the wrong pcb thickness | 23:12 |
| mntmn | but we soldered one anyway to do basic tests | 23:12 |
| mntmn | and actually i came quite far | 23:12 |
| * KM4MBG -> jackhill | 23:12 | |
| dj-death | mntmn: does the system feel snappier? (the CPU cores are faster right?) | 23:21 |
| mntmn | dj-death: yep, but it's io bound as i don't have an nvme in it | 23:23 |
| mntmn | i tried the cpu overclocked to 2ghz also | 23:23 |
| mntmn | and switched to the kms driver instead of fkms to get smooth wayland | 23:23 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
