| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 00:16 | |
| + pomel0 (~pomel0@user/pomel0) | 00:16 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 00:17 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@98.19.128.69) | 00:59 | |
| - chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 01:27 | |
| tookmund | @minute: curious about the gunyah hypervisor limitation, does this mean any OS running on the Quasar will be running in a hypervisor? | 01:35 |
|---|---|---|
| + bkeys (~Thunderbi@98.19.128.69) | 01:40 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@98.19.128.69) | 01:52 | |
| + bkeys (~Thunderbi@98.19.128.69) | 01:52 | |
| - bkeys (QUIT: Client Quit) (~Thunderbi@98.19.128.69) | 01:54 | |
| + bkeys1 (~Thunderbi@98.19.128.69) | 01:54 | |
| * bkeys1 -> bkeys | 01:57 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@98.19.128.69) | 02:00 | |
| + bkeys (~Thunderbi@h69.128.19.98.dynamic.ip.windstream.net) | 02:00 | |
| + kfx (~kfx@grendel.sciops.net) | 02:22 | |
| kfx | my reform rk3588 hangs during high i/o when the debian image is booted from sd card. anyone else see this? | 02:23 |
| - mjw (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 02:40 | |
| elb | I have not seen that, but mine did hang during high I/O with NVMe at one point; I changed to a different NVMe drive and it hasn't done it since | 02:46 |
| elb | that doesn't _feel_ related, but you might try a different SD card and see what happens | 02:46 |
| kfx | weirdly mine doesn't do it on nvme (and I stressed it pretty hard to make sure), but maybe I'l try some different SD cards | 02:46 |
| - paperManu_ (QUIT: Ping timeout: 264 seconds) (~paperManu@146.71.9.156) | 03:26 | |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@146.71.9.156) | 03:28 | |
| elb | mine doesn't do it with the NVMe I have in it now, but it did with the first one I tried | 03:31 |
| + sknebel (~quassel@v22016013254630973.happysrv.de) | 03:35 | |
| jfred | there have been a few of us with similar sounding issues recently. I wonder if there was a kernel regression | 03:54 |
| sigrid | kfx: is there anything on the serial? | 03:55 |
| sigrid | and also I wonder if alpine also has the same behavior | 03:56 |
| jfred | kfx: if you boot from nvme but do a lot of I/O to SD, can you reproduce it then? | 03:56 |
| jfred | ACTION should dig out a USB serial cable, I know I have one somewhere... | 03:56 |
| - bkeys (QUIT: Ping timeout: 246 seconds) (~Thunderbi@h69.128.19.98.dynamic.ip.windstream.net) | 04:12 | |
| kfx | jfred: will try | 04:16 |
| kfx | sigrid: not in my experience. alpine seems to run fine on the sd card | 04:16 |
| kfx | but under debian, just apt upgrade is enough to bring the pain | 04:16 |
| sigrid | interesting | 04:17 |
| + bkeys (~Thunderbi@98.19.128.69) | 04:17 | |
| - bkeys (QUIT: Remote host closed the connection) (~Thunderbi@98.19.128.69) | 04:17 | |
| + bkeys (~Thunderbi@98.19.128.69) | 04:20 | |
| jfred | hrm, I found a serial cable but I just get garbled text at boot and nothing after. as if I had the baud rate set wrong, but I'm almost certain I don't... been a while since I've had to connect to this thing via serial, heh | 04:31 |
| kfx | now using dd(1) to put a new system image on the sd card, no crash yet | 04:32 |
| kfx | no crash, but it did spontaneously reboot :/ | 04:37 |
| kfx | nothing over serial, everything normal and then wham, early boot output | 04:38 |
| jfred | ah yeah I've had that happen too | 04:39 |
| jfred | interesting that there's nothing over serial when that happens | 04:39 |
| kfx | ok, yeah. that time it hung. no serial output, just stops responding to input, no output, no ping from outside | 04:59 |
| swivel | anything from sysrq? | 05:00 |
| kfx | nope :/ | 05:01 |
| jfred | oh, I did have the baud rate set wrong for rk3588. needs to be 1500000, which I think is faster than my serial adapter supports anyways. ah well. | 05:32 |
| - voltaire28 (QUIT: Ping timeout: 260 seconds) (~jlafon@28.162.2.93.rev.sfr.net) | 06:02 | |
| - lanodan (QUIT: Quit: WeeChat 4.5.2) (~lanodan@2a01:e0a:d6:9930::35) | 06:14 | |
| - digitalrane (QUIT: Ping timeout: 250 seconds) (~digitalra@user/digitalrane) | 06:58 | |
| + digitalrane (~digitalra@user/digitalrane) | 07:03 | |
| + chomwitt (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 07:18 | |
| + voltaire28 (~jlafon@28.162.2.93.rev.sfr.net) | 07:20 | |
| + lanodan (~lanodan@2a01:e0a:d6:9930::35) | 07:39 | |
| minute | kfx: jfred: for me there's a swiotlb problem with the sd card driver on 6.17.8. this is not the most recent kernel though | 09:32 |
| minute | i should probably try to upgrade | 09:33 |
| - nsc (QUIT: Ping timeout: 244 seconds) (~nicolas@i5C74DC74.versanet.de) | 09:55 | |
| + nsc (~nicolas@i5C74DC74.versanet.de) | 09:57 | |
| josch | minute: we got flash-kernel with support for 6.18 uploaded yesterday and reform-tools with support for 6.18 was uploaded the day before that. We still need to have a Breaks relationship in the kernel package to make it impossible to install 6.18 while either flash-kernel or reform-tools are still too old to support it but I have a patch in the linux6.18 branch doing exactly that in case you would like | 10:34 |
| josch | me to press the merge button now instead of waiting for an upload of a kernel with this fix to Debian unstable. | 10:35 |
| minute | josch: ohhh ok! please merge away | 10:39 |
| minute | running next already 1 hour 36 minutes on 0% battery. need to work on that gauge code :D | 11:02 |
| josch | i did a last test and it boots on rk3588, imx8mq and a311d, so merge button was pressed and it should be in the repo in an hour | 11:02 |
| minute | josch: exciting | 11:02 |
| josch | debian experimental already has 6.19 :) | 11:02 |
| minute | ohhhhh | 11:02 |
| minute | we should integrate qcs6490 + qcs8550 dtb soon in reform-debian-packages | 11:02 |
| minute | qcs6490 can already boot with display on pocket reform thanks to kcxt | 11:03 |
| josch | yes, right now reform-system-images has a hack to ignore quasar because the dtb is missing | 11:03 |
| minute | woops | 11:03 |
| josch | (because the latest reform-tools release has the machine.conf files for it) | 11:03 |
| josch | (which also still need some fixing but better than nothing) | 11:04 |
| josch | gordon1: okay, so i would've liked to use the diamond shaped SK8707-01-005 because it's thinner but the mounting holes are to the side of the stem and there is no place on the pcb to attach these. | 11:05 |
| josch | errr... sorry i meant SK8707-01-008 which is the diamond one | 11:06 |
| kcxt | minute: speaking of DT, i gave up with overlays in the end (way more annoying than i hoped), I have a dts for the pocket and the reform2 with a common dtsi, seemed to work well for 6490 so 8550 will need the same changes | 11:06 |
| josch | but if you move the two holes you have in the pcb right now by about half a mm down, then the existing mounting holes of SK8707-01-005 will likely fit | 11:07 |
| josch | i would still like to have more than 2 screws attaching the two boards together so maybe we can have a few more holes in the keyboard pcb -- i'll make a drawing for you | 11:08 |
| gordon1 | can't find 008 anywhere | 11:08 |
| gordon1 | oh, nvm, can't type | 11:08 |
| josch | i liked 008 more because it's thinner and we are very space constrained but your idea of making the hole bigger and letting the thick part of the stem sit flat with the pcb top surface should work i think | 11:09 |
| gordon1 | yeah, i think 008 should work, let me check | 11:10 |
| josch | i think 008 could be mounted if it were first mounted to some mounting plate/pcb *below* it and then that plate/pcb would be attached to the keyboard. So indirectly attaching 008 would work but would make it of course a bit more fragile mechanically speaking. To attach 008 directly to the keyboard pbc, the mounting holes sit too close to the holes for the choc pins i think. But let me know if i made a | 11:13 |
| josch | mistake. | 11:13 |
| minute | kcxt: ah very nice | 11:19 |
| minute | kcxt: i just pushed a slightly cleaned up (just commented out stuff removed etc) qcs6490 dts to a reform-debian-packages branch | 11:19 |
| minute | kcxt: josch: https://source.mnt.re/reform/reform-debian-packages/-/commit/7b90396574ad96a6ee1e4a8a958fe5b437f103c5 | 11:20 |
| minute | kcxt: josch: that sounds like we'll also have to put our first .dtsi in there soon :D | 11:21 |
| gordon1 | josch: https://tinystash.undef.im/il/5PiDZD8JLESuTKCR485DYASaWXY2auU9aRXJ9pLmryYbtJqGk4JUowf6Gxitrk3ghvHAqYD94qXjw6Hc7uHcik38.png | 11:21 |
| gordon1 | so the holes line up well with switches, sort of, but because those are at the same line, you either cutting into a switch with bigger stick hole below, or has no space for screw heads | 11:22 |
| kcxt | minute: yeah i made quite a lot of changes to refactor everything and split out the reform2/pocket specific DTS parts | 11:23 |
| minute | kcxt: i should probably give you write access to reform-debian-packages so you can copy those files there? | 11:23 |
| gordon1 | so yeah i guess you need a mounting solution sandwiched in between | 11:24 |
| josch | gordon1: in that image you kept the center position of the circle and just increased the radius, right? | 11:27 |
| josch | maybe using countersunk screw heads will allow for the switches to sit on top of them | 11:28 |
| gordon1 | josch: that is 2mm hole instead of 5mm for reference https://tinystash.undef.im/il/5tu6hX4gb29A3yUejwRERL6AbCxAUN2vKD1rKs8a4JoPhx7EJDd324kCV7ekNWQRVKewYgZN3YegRUYuUaEQnBw1.png, i moved it a bit down (0.35mm) to get clearence between screw holes and choc pins | 11:28 |
| gordon1 | i'm not sure that countersunk heads won't interfere with pins above | 11:31 |
| josch | gordon1: ah you moved it down, that's why the holes to the side fit as they do. Yes, me saying that it doesn't fit was assuming that you would not move the center hole. | 11:31 |
| kcxt | minute: pushed my updated DT stuff, worth noting that the same DTS won't work on 6.17 and 6.18 due to some changes in sc7280.dtsi (unless you copy that file too)... https://source.mnt.re/reform/reform-qcs6490-uboot/-/commit/21f2a04837afddc2e7a6383d5b696b2c1c0e9a8e | 11:31 |
| josch | gordon1: indeed you are probably right | 11:31 |
| josch | gordon1: given all that would it not be easier to use SK8707-01-005 instead? Then the position of the center hole can remain where it was and just the two holes to the side have to be moved down by 0.5 mm. | 11:32 |
| gordon1 | josch: every hole has keepout area (cyan circle in this pic), just moved it so the pin's ring is outside it https://tinystash.undef.im/il/4u6dWXf7oFEsgxRkuV1VQuFYojVb2xdeD1r4dYtLTp4VoqRWrBDizfaKKUrxKVkr3mCjNLnMV3jPWfutxzLqFZQH.png | 11:32 |
| kcxt | minute: Something i'd really like to do on the U-Boot side is find a way to determine if what mainboard we're attached too, maybe some GPIO that's pulled high/low only on one? | 11:33 |
| gordon1 | aren't 005 screws at the same distance as 008? | 11:34 |
| gordon1 | or do you intend to use only two of the four holes? | 11:34 |
| josch | gordon1: yes, but there are four mounting holes and two of them are not at the same level as the center | 11:34 |
| josch | gordon1: so i had thought to use the two bottom mounting holes of the four in SK8707-01-005, those which are not at the same height as the center | 11:35 |
| gordon1 | i'm not sure if that is mechanically sound, you will be creating momentum when you apply a force to the stick | 11:35 |
| gordon1 | hard to judge the forces, needs testing | 11:36 |
| minute | kcxt: i2c could be a way if not too complicated | 11:36 |
| minute | kcxt: the boards have different stuff on the 2 i2c busses... | 11:36 |
| josch | yes, but that's why i was talking about one or two more holes in the pcb further above so that you can add a sandwhich layer below everything and mount that to the keyboard pcb as well | 11:36 |
| kcxt | minute: hmm alright, should be doable! | 11:37 |
| minute | kcxt: but for that we need that custom xbl build :0 | 11:37 |
| minute | kcxt: or bodge the hw to put everything on one i2c... | 11:37 |
| gordon1 | josch: but yeah, that works well on the pcb https://tinystash.undef.im/il/3vw4tq3QC7zzXWZfWPLpLS3xhQiMJmFYpmXCWxb9gEgvp79UtdpfC2e3SaF5kzaJAz8T49rGf5hQ9PU1CM4L4ntV.png | 11:39 |
| josch | gordon1: i see your argument with the momentum if the things is not mounted with the holes directly to the sides of the center but the further down we move the center hole the more difficult it also is to find space for the stem itself. | 11:40 |
| josch | gordon1: yes, this is what i had in mind | 11:40 |
| gordon1 | tbh that would probably work | 11:41 |
| josch | and then i'd 3d print a thin case to go below all of that and screw that case into the keyboard as well to add more support or something | 11:41 |
| gordon1 | one extra hole on top somehow would make it ideal | 11:41 |
| josch | gordon1: the unit has these castellated pins. Maybe it can be soldered to yet another pcb of ours and that can then be fixed to the keyboard? | 11:42 |
| josch | some sort of additional pcb is needed anyways for the pull-up resistors mentioned in the pdf | 11:43 |
| gordon1 | i wouldn't apply any stress on those pads,that's sure way to rip them off | 11:43 |
| gordon1 | however again, there are quite few big ones there, so maybe it's gonna be enough force to keep them | 11:44 |
| kcxt | minute: oh yeah xbl is probably the next thing i should look into | 11:44 |
| gordon1 | maybe i am overthinking it and just two screws will be more than enough | 11:44 |
| gordon1 | it's a steel plate ffs | 11:45 |
| josch | :) | 11:45 |
| josch | yes, but only M2 screws :) | 11:45 |
| gordon1 | yeah, but just 1.6mm of screw for it to flex | 11:47 |
| minute | kcxt: ok cool, tell me if you need anything. and then pcie iommu mysteries? btw how is the qcs6490 pocket? nice? | 11:47 |
| josch | gordon1: in your last picture, is the center circle still at x=133.6 y=99.15 ? | 11:48 |
| gordon1 | yes, i moved it back | 11:48 |
| kcxt | minute: for sure, it'll be a few days before i can spend more time on it probably | 11:48 |
| josch | gordon1: i like it like that, can you push that version to the git branch? | 11:48 |
| josch | i'd like to have a look and visualize it a bit more | 11:49 |
| gordon1 | i need to clean it up a bit, but sure | 11:49 |
| josch | no stress :) | 11:49 |
| josch | thank you! | 11:49 |
| josch | minute: the trackstick thing is coming along well -- we found a company to source them easily from :) | 11:50 |
| kcxt | minute: the pocket is nice! still getting used to the keyboard heh. I need to figure out how to get the panel working consistently and figure out suspend. For PCIe in the end i copied the hack that the rb3gen2 has to get stuff working over a pcie bridge and that worked for wifi at least xD | 11:50 |
| kcxt | minute: how are the nvme drives connected? | 11:51 |
| minute | kcxt: oh wow @ wifi | 11:55 |
| minute | kcxt: i think nvme should be on pcie1 but let me double check | 11:56 |
| [tj] | minute: your pcb renders always have 3d models, how on earth to you archive that with kicad? | 11:56 |
| kcxt | theres teo nvme slots on pocket right? | 11:56 |
| minute | kcxt: no only one, the other is usb only, for wwan | 11:56 |
| minute | kcxt: and there's a third slot which is pcie+usb for wifi (which you got working i guess) | 11:56 |
| minute | kcxt: the next has 2 nvmes (one is microsd express at the moment tho) | 11:57 |
| minute | kcxt: sorry, pcie1 isn't actually used on the pocket | 11:58 |
| minute | kcxt: everything should be on pcie0 | 11:58 |
| minute | kcxt: nvme on pocket should also show up through the pcie switch. | 12:00 |
| minute | kcxt: and the pcie ethernet macphy but i think it's borked on the qcs6490 physically (not sure how yet). it shows up on the qcs8550 | 12:01 |
| - chomwitt (QUIT: Ping timeout: 265 seconds) (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 12:02 | |
| gordon1 | josch: done | 12:04 |
| minute | current estimate of reform next runtime with normal/medium heavy remote work tasks over wifi + some video or music on speaker is around 4 hours, but it's still running, so lets see | 12:13 |
| ch | (intel) wifi on my pocket seems to disconnect now each night. thats a new issue somehow | 12:17 |
| josch | minute: that compares well to the classic reform with rk3588 which survived 4.5 hours for me yesterday | 12:20 |
| josch | gordon1: thank you! I overlayed the pcb with the drawing from the pdf here: https://mister-muffin.de/p/B6uO.png | 12:20 |
| minute | ch: weird. any evidence? | 12:20 |
| josch | gordon1: if i add some 3d printed support below that, then adding more mounting holes (green dots) would be nice -- what do you think? | 12:20 |
| gordon1 | josch: i would advice to start with just the stock steel plate and see how it goes and not to overengineer it from the beginning | 12:22 |
| josch | fine by me :) | 12:22 |
| gordon1 | just looking at it now with the actual drawing i feel it should be enough | 12:22 |
| josch | i was just thinking that if it turns out that more support would be necesary, then better have the holes in place already instead of having to manufacture another pcb | 12:23 |
| gordon1 | those holes are quite close to the stick itself so i think it should be quite fine | 12:23 |
| josch | okay! | 12:24 |
| gordon1 | oh that's not the worst idea | 12:24 |
| gordon1 | i thought you want to test it with 3d printed mockup first | 12:24 |
| josch | yes, i was just throwing in an idea. No need to do anything now. | 12:25 |
| josch | i also want to read this chat history later and not forget about thoughts i had :) | 12:25 |
| josch | i'm going to integrate your changes into my mockup, do a new print and wait for the trackstick module to arrive and then share with you how it worked :) | 12:26 |
| gordon1 | looking at the overlayed drawing, it looks as if they designed it to fit the keyboard, didn't they? :D | 12:27 |
| ch | https://paste.debian.net/hidden/0de07f88 looks like the rekeying stops working at some point? weird | 12:28 |
| josch | gordon1: the mounting hole distance nicely fits yes. Lets see how it looks like in my mock-up. :) | 12:29 |
| gordon1 | yeah, it is for more standard 19mm key spacing i guess, reform keyboard is tiny bit tighter | 12:29 |
| josch | i recently had a closer look at the key spacing. It's nicely regular for the top four rows. It's a bit off for the fifth row and a total mess for the sixth row. Nothing you can do about it mathematically speaking but it triggered some OCD sensor in me and now i cannot unsee the different spacing. XD | 12:31 |
| + mjw (~mjw@gnu.wildebeest.org) | 12:31 | |
| gordon1 | ortholinearity to the rescue! | 12:31 |
| josch | indeed! | 12:31 |
| josch | I was looking at the key spacing because I wanted to have a more accurate rainbow effect and for that I needed the precise location of each LED so that i can compute the correct rainbow color for each LED as it animates. | 12:32 |
| josch | Here is my current POC: https://paste.debian.net/hidden/028cdcae | 12:32 |
| josch | But I should probably instead hack the firmware and should put the effect into it instead. The rp2040 is probably mostly idle anyways... | 12:33 |
| gordon1 | oh, can you control the backlight from userspace? | 12:34 |
| gordon1 | i need to make a notifications and layout indication based on backlight | 12:34 |
| + paperManu (~paperManu@146.71.9.156) | 12:35 | |
| gordon1 | oh boy i have too many hidraws without way of distinguishing those | 12:36 |
| josch | gordon1: there are symlinks, look for /dev/input/by-id/usb-MNT_Research_MNT_Reform_Keyboard_4.0_*-hidraw | 12:38 |
| gordon1 | no, there aren't, i don't have udev | 12:38 |
| gordon1 | i need to add those into my mdev config | 12:38 |
| minute | josch: now also approaching 4.5h here | 12:48 |
| - nsc (PART: !!unknown attribute: msg!!) (~nicolas@i5C74DC74.versanet.de) | 12:49 | |
| josch | gordon1: this is how it looks for udev: https://sources.debian.org/src/systemd/259-1/rules.d/60-persistent-hidraw.rules | 12:52 |
| gordon1 | yeah, i don't think mdev has ID_PATH, but it does have usb vid/pid so i think i'll stick with those | 12:54 |
| josch | gordon1: are you using reform-tools? Is such a file something reform-tools could ship for mdev? | 12:57 |
| gordon1 | no, i'm not using reform-tools, i sorta hacked my own scripts | 12:58 |
| gordon1 | josch: not sure it might be helpful, mdev has single massive config file, so maybe just a snipped of code in readme? | 13:00 |
| gordon1 | tbh not sure if anyone uses mdev on reform but me | 13:01 |
| gordon1 | *snippet | 13:04 |
| gordon1 | (don't get the wrong impression that reform-tools weren't useful tho pls, they were of immense help) | 13:05 |
| + chomwitt (~chomwitt@2a02:85f:9a0a:f200:1ac0:4dff:fedb:a3f1) | 13:19 | |
| josch | gordon1: no worries, just wanted to offer in case it would help your setup :) | 13:22 |
| - paperManu (QUIT: Ping timeout: 252 seconds) (~paperManu@146.71.9.156) | 14:03 | |
| + johl (~johl@wikidata/Jens-Ohlig) | 14:18 | |
| - pomel0 (QUIT: Ping timeout: 246 seconds) (~pomel0@user/pomel0) | 14:20 | |
| + pomel0 (~pomel0@user/pomel0) | 14:21 | |
| gordon1 | SUBSYSTEM=hidraw;.* root:input 660 @[[ $(realpath /sys/class/hidraw/$MDEV) == *:1209:6D02.* ]] && ln -sf $MDEV kbd_hidraw | 14:43 |
| gordon1 | ACTION feels dirty | 14:43 |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 14:47 | |
| - janaa (QUIT: Ping timeout: 260 seconds) (~jana@kochab.uberspace.de) | 14:52 | |
| - pomel0 (QUIT: Read error: Connection reset by peer) (~pomel0@user/pomel0) | 14:52 | |
| + pomel0 (~pomel0@user/pomel0) | 14:52 | |
| - angelwood (QUIT: Remote host closed the connection) (~angelwood@user/angelwood) | 15:11 | |
| + angelwood (~angelwood@user/angelwood) | 15:12 | |
| - pomel0 (QUIT: Ping timeout: 246 seconds) (~pomel0@user/pomel0) | 15:22 | |
| + pomel0 (~pomel0@user/pomel0) | 15:24 | |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@98.19.128.69) | 15:28 | |
| - pomel0 (QUIT: Ping timeout: 244 seconds) (~pomel0@user/pomel0) | 15:28 | |
| gordon1 | on an unrelated note, 2mm pin pitch female pin header sockets fit perfectly and hold nicely in 2mm jst (like the usb ones on the reform mb) | 15:32 |
| + pomel0 (~pomel0@user/pomel0) | 15:35 | |
| josch | gordon1: do you mean dupont plugs? | 15:37 |
| gordon1 | yep, but 2mm ones not 2.54mm | 15:38 |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 15:39 | |
| + pomel0 (~pomel0@user/pomel0) | 15:40 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 15:51 | |
| elb | hmmm for some reason my battery seems to be suffering today | 15:52 |
| elb | two hours of uptime and I'm down to 30% | 15:53 |
| gsora | tangential: whats your "shutdown" battery life like, reform classic users? mine seems to be completely drained after 2-3 days | 15:59 |
| [tj] | off draw is about half a watt | 16:00 |
| elb | pocket here, but mine is also a few days | 16:01 |
| elb | rk3588 | 16:01 |
| [tj] | oh classic I don't know about | 16:01 |
| gsora | 3588 here as well | 16:02 |
| gsora | not a big deal ofc | 16:03 |
| + pomel0 (~pomel0@user/pomel0) | 16:17 | |
| - pomel0 (QUIT: Ping timeout: 244 seconds) (~pomel0@user/pomel0) | 16:22 | |
| + bkeys (~Thunderbi@172.58.3.186) | 16:22 | |
| + pomel0 (~pomel0@user/pomel0) | 16:24 | |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 16:24 | |
| + pomel0 (~pomel0@user/pomel0) | 16:33 | |
| - pomel0 (QUIT: Read error: Connection reset by peer) (~pomel0@user/pomel0) | 16:37 | |
| rick_ | it seems i need more coffee. i just tried to boot my reform via my external usb keyboard with hitting ctrl+enter and wondered for a sec why nothing happens XDD | 16:38 |
| rick_ | stupid muscle memory XDD | 16:38 |
| + pomel0 (~pomel0@user/pomel0) | 16:38 | |
| - mjw (QUIT: Killed (zirconium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org) | 16:52 | |
| * Guest5349 -> mjw | 16:52 | |
| + Guest6263 (~mjw@gnu.wildebeest.org) | 16:53 | |
| rick_ | elb isn't that normal? especially with fedi open in my firefox i get at max 3 hours out of my pocket. so for me 30% left in two hours are 'normal' | 17:05 |
| rick_ | sadly escpecially firefow is powerhungry.. so when i try to save battery i close firefox before i lock the screen XD | 17:06 |
| cwebber | meow | 17:08 |
| rick_ | mauu xD | 17:09 |
| josch | rick_: i send it SIGSTOP with pkill :) | 17:18 |
| rick_ | isn't that kinda the same? :3 | 17:19 |
| + spew (~spew@user/spew) | 17:24 | |
| josch | hm? no | 17:26 |
| - qbit (QUIT: Remote host closed the connection) (~qbit@user/qbit) | 17:30 | |
| + qbit (~qbit@user/qbit) | 17:30 | |
| + spew_ (~spew@user/spew) | 17:31 | |
| jfred | Ubuntu Touch SIGSTOPs backgrounded apps and it seems to work really well for battery life (along with whatever other power management stuff they have) | 17:32 |
| - spew (QUIT: Ping timeout: 245 seconds) (~spew@user/spew) | 17:33 | |
| josch | yes, the problem with that approach is, then you cannot anymore paste stuff from the application you have sent sigstop to | 17:33 |
| josch | and once you sent sigcont, then the paste will happen | 17:33 |
| - johl (QUIT: Remote host closed the connection) (~johl@wikidata/Jens-Ohlig) | 17:34 | |
| jfred | ooh, right, because clipboard access involves both apps | 17:34 |
| jfred | if the window manager is involved it could briefly SIGCONT the backgrounded app when you paste, right? | 17:34 |
| * spew_ -> spew | 17:34 | |
| elb | rick_: it varies for me, but yeah firefox + element hurts | 17:37 |
| rick_ | josch ohh i totally forgot about that.. that ou can stop and start programms.. nice idea i should use that too. thanks! | 17:37 |
| rick_ | elb i use fluffly.. should check that next time how much it uses | 17:38 |
| - paperManu (QUIT: Ping timeout: 246 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 17:40 | |
| rick_ | oh yeah, this works nice. i should integrate that in my swaylock command ^^ | 17:41 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:49 | |
| elb | oh yeah that's clever, sigsto on lock ... | 17:49 |
| - spew (QUIT: Quit: nyaa~) (~spew@user/spew) | 17:50 | |
| gsora | gnome at 150% scaling looks awesome on the reform classic screen | 18:19 |
| gsora | s/150/125/g | 18:19 |
| jfred | sigstop on lock is an interesting idea, though I wonder what the best way to select processes to stop would be... | 18:25 |
| jfred | don't want to accidentally sigstop swaylock :) | 18:25 |
| L29Ah | killall -SIGSTOP firefox qutebrowser torbrowser palemoon | 18:26 |
| rick_ | jfred ill write a small script where i put all my main progs (or the ones which i use the most) in a list. the script executes SIGSTOP on these programms, invokes swaylock and then.. i just have to find out a good method to dedect when swaylock unlooks the screen again ^^v | 18:27 |
| L29Ah | in X you can just put it in your script that runs your screen locker, after the locker process dies | 18:29 |
| rick_ | yeah,, i could do sth like that too _think_ just check periodically if swaylock still runs | 18:30 |
| - gianmarcogg03 (QUIT: Ping timeout: 250 seconds) (~quassel@user/gianmarcogg03) | 18:33 | |
| + gianmarcogg03 (~quassel@user/gianmarcogg03) | 18:33 | |
| rick_ | nice it works, thanks for the tips ^^ | 18:52 |
| lidstah | Hi all :) | 19:05 |
| lidstah | finally got this raspberry pi nvme, it was in UPS limbo since a week… | 19:07 |
| minute | lidstah: nice. does it work? | 19:07 |
| lidstah | minute: well… I don't have my reform classic yet - per the FAQ, should be here around mid february/mid march | 19:08 |
| lidstah | btw, just backlogged almost a week of logs, and saw you had a surgery recently, I wish you a good recovery! | 19:09 |
| lidstah | (and I hope it's the proper way to wish recovery to someone in english) | 19:11 |
| minute | lidstah: thank you! definitely proper! | 19:16 |
| lidstah | you're welcome! I hope everything went well, and that you're able to get proper rest! | 19:17 |
| lidstah | originally I had a spare kingston nv3 nvme, but I used it in one of my proxmox nodes to replace a samsung evo970 which was way too hot for my tastes (controller at 110°C,NANDs between 70-80°C, while the nv3 is between 45-50°C, controller at max 70°C during stress) | 19:19 |
| lidstah | and well, didn't have another spare to put in the reform when it'll arrive, and due to the actual prices… the raspberry pi nvme at 40 bucks for the 256GB model was a grab (there was only 4 in stock when I ordered it, the next day there was no more in stock) | 19:21 |
| minute | lidstah: everything went well and i'm alternating between resting and looking at work stuff and irc a little bit. | 19:22 |
| minute | hmm i updated to the current linux-image and display doesn't come up on the next. i wonder if my current dts is in there | 19:23 |
| + spew (~spew@user/spew) | 19:26 | |
| lidstah | glad to hear that everything went well and that you're able to rest while still hacking at stuff :) | 19:26 |
| rick_ | omg i over enigneered it.. i don't even have to activle check if swaylock is running.. cause when i invoke swaylog within my script, it just waits for swaylock to 'finfish' and then executes the rest of the bash script lol | 19:26 |
| minute | ah, this wasn't merged yet woops https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/154 | 19:27 |
| - voltaire28 (QUIT: Ping timeout: 260 seconds) (~jlafon@28.162.2.93.rev.sfr.net) | 19:32 | |
| - pomel0 (QUIT: Ping timeout: 240 seconds) (~pomel0@user/pomel0) | 19:33 | |
| + pomel0 (~pomel0@user/pomel0) | 19:33 | |
| minute | ha, useful side effect of having set up tailscale: can still ssh into the displayless laptop and fix it, without needing serial cable stuff | 19:33 |
| - pomel0 (QUIT: Remote host closed the connection) (~pomel0@user/pomel0) | 19:39 | |
| lidstah | ahah, indeed tailscale is great for this (although I'm a netbird user, both @home and at work - similar use cases), as long as the machine boots and has access to the overlay network, it's manageable :) | 19:42 |
| lidstah | openziti is interesting too, especially the part when a client can access resources in a domain (e.g *.mydomain.tld) without having any clues of said resources network topology, but less user-friendly on the setup side of things imho | 19:44 |
| minute | yes, i want to migrate to netbird later :3 | 19:45 |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@172.58.3.186) | 19:46 | |
| + bkeys (~Thunderbi@172.58.3.186) | 19:46 | |
| minute | ah lol josch you already picked everything from next-dp right? | 19:47 |
| josch | minute: which things exactly do you mean? | 19:47 |
| lidstah | netbird is great indeed. Easy to setup and maintain, easy SSO setup (as long as it talks OIDC), easy user setup too, open-source server and clients (iirc, only android and ios clients are closed-source but I might be wrong here) | 19:47 |
| minute | my MR shows 0 changes now :D | 19:47 |
| josch | ah yes, i cherry picked a bunch of stuff which you earlier said you want added | 19:47 |
| minute | josch: thanks for that! | 19:48 |
| minute | josch: that means i can directly try that package now on the next | 19:48 |
| josch | sweet | 19:48 |
| minute | lidstah: ah nice at SSO, coincidentally i recently set up authentik | 19:49 |
| minute | (for grist, but can now use it for other stuff as well of course) | 19:49 |
| minute | josch: yesss 6.18,3 works on the next, display is up again | 19:57 |
| josch | \o/ | 20:00 |
| minute | aha, and swiotlb problem of the sd card is indeed fixed! | 20:00 |
| + paperManu (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 20:25 | |
| lidstah | minute: sorry, was making food for the family here, authentik is great too :D (using it at work, at a local not-for-profit ISP and at home :)) - although sometimes they tend to push new versions a bit too soon imho. the authenticating proxy functionnality (and the ldap one too) are great for stuff without authentication (pg-operator-webui for example) | 20:27 |
| minute | nice | 20:31 |
| lidstah | I like it a lot, there's a nice documentation (with example for many services, from discourse to netbox), and it's been easy to maintain both with docker and kubernetes (official helm chart) here. As long as you keep daily (and tested!) backups of its database, even in the case of a major hiccup it's easy to get back online and runnin'. | 20:36 |
| lidstah | it's also able to use SAML, which can be handy for some saml-reliant applications | 20:37 |
| + voltaire28 (~jlafon@28.162.2.93.rev.sfr.net) | 20:38 | |
| - mjw (QUIT: Killed (tantalum.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 20:39 | |
| * Guest6263 -> mjw | 20:39 | |
| + Guest149 (~mjw@2001:1c06:2486:a800:a09a:fc1c:5a8:e74d) | 20:39 | |
| + pomel0 (~pomel0@user/pomel0) | 20:42 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@172.58.3.186) | 20:46 | |
| + bkeys (~Thunderbi@172.58.3.186) | 20:46 | |
| + bkeys1 (~Thunderbi@172.58.3.186) | 20:49 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@172.58.3.186) | 20:49 | |
| * bkeys1 -> bkeys | 20:50 | |
| - bkeys (QUIT: Ping timeout: 240 seconds) (~Thunderbi@172.58.3.186) | 20:54 | |
| + siviq (~siviq@user/siviq) | 20:54 | |
| - siviq (QUIT: Quit: Client closed) (~siviq@user/siviq) | 21:10 | |
| - cli (QUIT: Ping timeout: 264 seconds) (~m-vsauiy@user/cli) | 21:18 | |
| - voltaire28 (QUIT: Quit: WeeChat 4.6.3) (~jlafon@28.162.2.93.rev.sfr.net) | 21:27 | |
| + voltaire28 (~jlafon@28.162.2.93.rev.sfr.net) | 21:28 | |
| + bkeys (~Thunderbi@172.58.2.40) | 21:58 | |
| - aelius (QUIT: Remote host closed the connection) (~aelius@user/aelius) | 22:24 | |
| + aelius (~aelius@user/aelius) | 22:24 | |
| - aelius (QUIT: Remote host closed the connection) (~aelius@user/aelius) | 22:25 | |
| + aelius (~aelius@user/aelius) | 22:26 | |
| - spew (QUIT: Quit: nyaa~) (~spew@user/spew) | 22:35 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@172.58.2.40) | 22:53 | |
| + bkeys1 (~Thunderbi@172.58.2.40) | 22:54 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 22:54 | |
| - mjw (QUIT: Read error: Connection reset by peer) (~mjw@gnu.wildebeest.org) | 22:54 | |
| * bkeys1 -> bkeys | 22:56 | |
| - paperManu (QUIT: Ping timeout: 264 seconds) (~paperManu@modemcable141.205-200-24.mc.videotron.ca) | 22:57 | |
| minute | lots of stuff still going on in panvk land https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/?label_name%5B%5D=panvk | 22:58 |
| + wielaard (~mjw@gnu.wildebeest.org) | 23:01 | |
| - mark_ (QUIT: Ping timeout: 240 seconds) (~mjw@gnu.wildebeest.org) | 23:02 | |
| + paperManu (~paperManu@146.71.9.156) | 23:13 | |
| - cow321 (QUIT: Ping timeout: 244 seconds) (~deflated8@user/meow/deflated8837) | 23:13 | |
| + bkeys1 (~Thunderbi@2607:fb90:e34a:16:27e0:4a9a:a546:1d59) | 23:23 | |
| - bkeys (QUIT: Ping timeout: 246 seconds) (~Thunderbi@172.58.2.40) | 23:25 | |
| * bkeys1 -> bkeys | 23:26 | |
| - bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@2607:fb90:e34a:16:27e0:4a9a:a546:1d59) | 23:29 | |
| + bkeys1 (~Thunderbi@2607:fb90:e34a:16:27e0:4a9a:a546:1d59) | 23:29 | |
| * bkeys1 -> bkeys | 23:32 | |
| + paperManu_ (~paperManu@146.71.9.156) | 23:34 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 23:35 | |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@2607:fb90:e34a:16:27e0:4a9a:a546:1d59) | 23:43 | |
| gsora | we love to see it! | 23:52 |
| + bkeys (~Thunderbi@134.22.115.162) | 23:53 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
