| pandora | bdw anyone got a good source for an ix industrial to rj45 ethernet cable? All i could find was like mouser with 40 bucks + 20 for shipping | 00:10 |
|---|---|---|
| - jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@2600:1700:9e1e:7800:2c94:5a14:bbc3:b280) | 00:17 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a2b:1eff:c85d:fd50:5b36:418) | 00:28 | |
| + akira (~akira@37.4.230.225) | 00:28 | |
| minute | pandora: unfortunately they are expensive, but i've been thinking to make our own adapters, which would be cheaper | 00:34 |
| minute | now giving a311d module a spin in my pocket reform | 00:38 |
| minute | oh niice, display comes on | 00:39 |
| grimmware | \o/ | 00:39 |
| minute | (flashed system image 4529 to microsd card and booted that) | 00:39 |
| minute | nice, haven't seen pocket setup wizard on a311d yet | 00:40 |
| minute | so cool, i expected it had maybe bitrotted a bit, but no, works fine so far | 00:41 |
| minute | afaik i still have to fix sound (but after vacation) | 00:41 |
| grimmware | I never quite know what to do with myself when stuff works | 00:44 |
| minute | haha | 00:44 |
| minute | yeah, today i got a haircut and bleach and hair color by holo_memory and during that time i figured out how to run katakis on vice64 on the pocket, and once it worked, i was really hooked | 00:45 |
| minute | and finally just played through the game... :D | 00:45 |
| minute | ok, a311d in pocket reform is bliss | 00:45 |
| minute | speeeed | 00:46 |
| grimmware | Hey look here man I can’t justify buying a processor module until o have the laptop | 00:47 |
| grimmware | *I | 00:47 |
| minute | ah, luckily there is a thing that doesn't work, nvme is not detected :D | 00:47 |
| minute | grimmware: haha sure! | 00:47 |
| minute | the imx8mplus is cool too | 00:47 |
| grimmware | I’m kinda waiting to see how much the rcore destroys the battery when you get that working :P | 00:48 |
| minute | doesn't destroy it much | 00:48 |
| minute | ah you mean in the pocket! | 00:48 |
| minute | yeah, curious too | 00:48 |
| grimmware | The temptation to have a 32GB memory pocket reform is high. Needlessly powerful. | 00:49 |
| minute | :D | 00:50 |
| josch | grimmware: if stuff just works, it must be because somehow you tested it wrongly (my own experience) ;) | 00:51 |
| grimmware | josch: facts. | 00:51 |
| minute | ah yeah, rcm4 has a design bug where pcie reset line is not pulled up | 00:52 |
| minute | so in pocket with a311d probably only nvmes work that don't need reset pullup | 00:52 |
| grimmware | I’ve just started playing around with an SDR so I’m hoping after I move house to take my Pocket on walks to the Thames to try and track boats. | 00:53 |
| minute | well, the rcm4 adapter board itself is pretty cheap, we can probably replace it if necessary | 00:53 |
| minute | grimmware: neat | 00:53 |
| grimmware | Oh nice my window will be facing London City Airport too | 00:53 |
| minute | can confirm that this wifi fix for a311d works indeed https://community.mnt.re/t/a311d-wifi-issues-firmware-failed-to-leave-lps-state-disconnects/2112/2 | 01:01 |
| minute | so this should be shipped by default | 01:01 |
| josch | minute: "wifi.powersave = 2" makes things better but not perfect | 01:02 |
| minute | josch: hm! | 01:02 |
| josch | also, our kernel not being bootable might turn out to be a binutils bug fixed here: https://sourceware.org/git/?p=binutils-gdb.git;a=commitdiff;h=961befd69308895bf1dc39737d1598246dc296a8 | 01:03 |
| minute | wow @ binutils btw | 01:03 |
| josch | i hope only few people made their reforms non-bootable | 01:03 |
| minute | josch: yeah, i think we pulled it pretty quickly | 01:03 |
| josch | running unstable, we are just the first affected by problems like this | 01:03 |
| josch | minute: i have a workaround building our stuff with old binutils as a MR in case you want to go that route | 01:04 |
| josch | otherwise we can just wait for the fix to get uploaded to unstable | 01:04 |
| minute | josch: do you think it will take long for the fix to arrive? | 01:05 |
| josch | i have absolutely no clue | 01:06 |
| josch | people in #debian-arm are still making sure that this is the right fix | 01:06 |
| josch | but it's being handled now which makes me happy because it means my job is done XD | 01:06 |
| grimmware | This is like, the opposite of the vendored android problem | 01:07 |
| josch | in what way? | 01:08 |
| grimmware | You can’t get timely updates on a vendor branded Android release whereas the reform image is so fresh it’s picking up breaking changes in unstable :D | 01:09 |
| grimmware | To clarify, that is a problem I would much rather have :) | 01:10 |
| josch | indeed -- the idea is of course: if nobody is using "unstable" and running into bugs, then those bugs will end up in "stable" | 01:10 |
| josch | yes, it absolutely is a trade-off between what kind of problem you rather have :) | 01:10 |
| josch | grimmware: also, if you don't like unstable, you can try reform.debian.net | 01:10 |
| grimmware | Oh I’ll be running unstable :) | 01:12 |
| grimmware | Right, I gotta go to sleep. Catch you all later! | 01:13 |
| josch | good night! | 01:14 |
| minute | oh nice, sound works now @ a311d pocket, and lpc too | 01:24 |
| minute | n8 grimmware | 01:24 |
| minute | btw i was thinking that later we could introduce our own semi-stable repo, where there's just a delay between the direct unstable repo and this semi-stable one | 01:25 |
| minute | i.e. a month or two weeks or so | 01:25 |
| minute | i would always run on the real unstable (and hopefully some volunteers) and catch the early bugs | 01:26 |
| josch | minute: debian linux kernel devs wanted to upload 6.9 to unstable today | 01:27 |
| josch | minute: had the reform users not caught this, then this upload would've happened | 01:27 |
| josch | now it's on hold until the problem is fixed | 01:27 |
| josch | so many users running unstable is indeed useful | 01:27 |
| minute | wow | 01:29 |
| minute | yeah, it is | 01:29 |
| josch | situations like this happened in the past too | 01:30 |
| josch | just not as bad as "the whole system becomes unbootable" :) | 01:30 |
| minute | yeah "D | 01:32 |
| josch | mostly it was stuff where packages failed to build or failed to install | 01:32 |
| minute | i added a new MR for a311d/pocket, but stopped the pipeline because it doesn't make sense right now, right? https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/44 | 01:32 |
| josch | yes, the build result will be unbootable | 01:34 |
| minute | ok... but i'm happy to report that a311d will be nicely usable in pocket once we have builds again :D | 01:37 |
| josch | \o/ | 01:37 |
| josch | pandora: can you confirm that the ls /opt/reform-qcacld2/qcacld2-*.ko can be replaced using this find command? | 01:43 |
| josch | find /opt/reform-qcacld2/qcacld2-*.ko -maxdepth 1 -type f -print0 | sort -Vzr | 01:43 |
| josch | using ls has the problem that ls output is meant for humans, so ls does all sorts of funny things for filnames that include more than a-z | 01:44 |
| josch | https://www.shellcheck.net/wiki/SC2012 | 01:44 |
| josch | it also allows using \x00 as the delimiter | 01:45 |
| josch | (technically, filenames can include newline characters) | 01:45 |
| minute | dual mnt reforms :0 https://tooting.ch/@vimja/112668406067498207 | 01:45 |
| - ggoes (QUIT: Ping timeout: 268 seconds) (~gregf@fsf/staff/ggoes) | 01:46 | |
| - chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:587:7a27:9000:1ac0:4dff:fedb:a3f1) | 01:47 | |
| josch | pandora: oh actually, since you want to loop over these files, that can be done better... sec... | 01:47 |
| josch | minute: see, *that* is the setup i need for debian-installer work! :D | 01:48 |
| josch | though i will fail at vimja's crazy display top XD | 01:48 |
| minute | :D | 01:51 |
| minute | purple pocket at teardown2024 https://chaos.social/@ishotjr/112668612095307998 | 01:52 |
| minute | josch: why does one need ls or find, because of the date sorting? | 01:52 |
| josch | because of version sorting | 01:53 |
| josch | version sort is differet from alphanumeric sort | 01:53 |
| + ggoes (~gregf@fsf/staff/ggoes) | 01:56 | |
| ch | any recommendations for a 4G/LTE modem for pocket reform in EU? | 01:57 |
| + cow321_ (~deflated8@user/meow/deflated8837) | 01:57 | |
| josch | ch: Quectel EM06-E | 01:58 |
| - cow321 (QUIT: Ping timeout: 268 seconds) (~deflated8@user/meow/deflated8837) | 01:59 | |
| ch | thanks | 02:00 |
| ch | crowdsupply hates my creditcard. guess thats gonna be a phone call tomorrow | 02:00 |
| - mtm (QUIT: Ping timeout: 255 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:03 | |
| - cow321_ (QUIT: Ping timeout: 268 seconds) (~deflated8@user/meow/deflated8837) | 02:03 | |
| josch | pandora: i have an even better idea -- how about using "dpkg-query -l 'reform-qcacld2-*-mnt-reform-arm64'" and then using dpkg-query -L to get the path to the .ko? | 02:04 |
| josch | this has the advantage that the output of dpkg-query -l is guaranteed to be sorted by version | 02:05 |
| josch | and it's an even better sort because GNU version sort is not exactly the same as Debian version sort | 02:05 |
| josch | and the resulting package names are guaranteed to only cover certain characters, so it's much easier to stuff that into a shell loop | 02:05 |
| josch | pandora: what do you think? | 02:05 |
| - ggoes (QUIT: Ping timeout: 268 seconds) (~gregf@fsf/staff/ggoes) | 02:10 | |
| + ggoes (~gregf@fsf/staff/ggoes) | 02:11 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 02:14 | |
| violet | "rcm4 has a design bug where pcie reset line is not pulled up" | 02:19 |
| violet | hmm, maybe this is why optane was not working with my soquartz in the mnt | 02:19 |
| - cow321 (QUIT: Ping timeout: 255 seconds) (~deflated8@user/meow/deflated8837) | 02:20 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 02:25 | |
| minute | violet: hmm could be, but you use mpcie to m2 adapter with reset pullup fix, no? | 02:27 |
| violet | ah if the included adapter fixes that then yeah i suppose so | 02:28 |
| violet | was not aware it was anything other than a passive connector adapter! | 02:29 |
| + cow321_ (~deflated8@user/meow/deflated8837) | 02:40 | |
| - cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra) | 02:41 | |
| - cow321 (QUIT: Ping timeout: 252 seconds) (~deflated8@user/meow/deflated8837) | 02:42 | |
| + cobra (~cobra@user/Cobra) | 02:43 | |
| - cow321_ (QUIT: Ping timeout: 268 seconds) (~deflated8@user/meow/deflated8837) | 02:46 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 02:57 | |
| - cow321 (QUIT: Ping timeout: 246 seconds) (~deflated8@user/meow/deflated8837) | 03:02 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 03:14 | |
| noam | you've put the soquartz into the Reform?? | 03:20 |
| - nsc (QUIT: Ping timeout: 268 seconds) (~nicolas@i5C74DC29.versanet.de) | 03:22 | |
| + nsc (~nicolas@64-97-142-46.pool.kielnet.net) | 03:23 | |
| - Gooberpatrol66 (QUIT: Remote host closed the connection) (~Gooberpat@user/gooberpatrol66) | 03:49 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:09 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 04:40 | |
| - cow321 (QUIT: Remote host closed the connection) (~deflated8@user/meow/deflated8837) | 04:44 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 04:44 | |
| - cow321 (QUIT: Remote host closed the connection) (~deflated8@user/meow/deflated8837) | 04:46 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 04:48 | |
| pandora | @josch: oh replacing ls with find is acutally a really good idea. I forgot how much ls is doing to special characters. I didn't know that \x00 is possible :D | 05:00 |
| pandora | or newlines | 05:00 |
| pandora | does "dpkg-query -l 'reform-qcacld2-*-mnt-reform-arm64'" give you every available ko file in /opt/reform-qcacld2/? | 05:05 |
| pandora | i thought it only gives u installed packages | 05:06 |
| pandora | but hmm yeah ... since qcacld2 is a deb package | 05:14 |
| pandora | it should find all of them | 05:15 |
| pandora | that could work indeed | 05:17 |
| - cow321 (QUIT: Remote host closed the connection) (~deflated8@user/meow/deflated8837) | 05:29 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 05:29 | |
| - cow321 (QUIT: Remote host closed the connection) (~deflated8@user/meow/deflated8837) | 05:35 | |
| - mtm (QUIT: Ping timeout: 252 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 05:42 | |
| - cobra (QUIT: Remote host closed the connection) (~cobra@user/Cobra) | 06:26 | |
| + cobra (~cobra@user/Cobra) | 06:27 | |
| - akira (QUIT: Ping timeout: 264 seconds) (~akira@37.4.230.225) | 07:45 | |
| + akira (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 07:47 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 08:00 | |
| + akira (~akira@185.111.34.19) | 08:01 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@185.111.34.19) | 08:05 | |
| + akira (~akira@185.111.34.19) | 08:06 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@185.111.34.19) | 08:20 | |
| + akira (~akira@185.111.34.19) | 08:20 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@185.111.34.19) | 08:29 | |
| + akira (~akira@185.111.34.19) | 08:29 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 08:44 | |
| + chomwitt (~chomwitt@2a02:587:7a24:dd00:1ac0:4dff:fedb:a3f1) | 08:53 | |
| + reform10850 (~ning@64.186.9.81) | 09:00 | |
| - reform10850 (QUIT: Client Quit) (~ning@64.186.9.81) | 09:01 | |
| - mtm (QUIT: Ping timeout: 260 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 09:14 | |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:587:7a24:dd00:1ac0:4dff:fedb:a3f1) | 09:19 | |
| + mjw (~mjw@gnu.wildebeest.org) | 09:23 | |
| - akira (QUIT: Ping timeout: 268 seconds) (~akira@185.111.34.19) | 09:36 | |
| + akira (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 09:37 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 09:38 | |
| + akira (~akira@185.111.34.19) | 09:38 | |
| - nybble (QUIT: Quit: WeeChat 4.2.2) (~sprang@user/csprng) | 09:46 | |
| ex-parrot | josch: how does 'no u-boot' and 'none of the above' differ in this debconf question? Choices: no u-boot, ${MY_CHOICES}, none of the above | 09:56 |
| ex-parrot | it seems like they're equivalent functionality-wise to me? | 09:56 |
| josch | thanks, let me have a look | 09:57 |
| ex-parrot | unless I've misunderstood something | 09:57 |
| ex-parrot | I am again working on my notes and this was one of the questions :) | 09:57 |
| ex-parrot | was also comparing how that question operates vs the "where do you want grub installed" question that people might? be more familiar with e.g. https://i.sstatic.net/Mu4GP.png | 09:58 |
| ex-parrot | it might make most sense to either have a multi-select box like the grub one, or have a "do not install u-boot" as the final option, to mirror the "do you want a bootloader?" question which has "do not install a boot loader" as the final choice | 09:58 |
| ex-parrot | hope that makes some kind of sense | 09:59 |
| josch | about u-boot: yes, only the values "eMMC" and "SD-card" let something happen -- any other value is ignored | 09:59 |
| josch | ex-parrot: this is a very good suggestion, i'm leaving a FIXME in the code | 10:00 |
| ex-parrot | is one of the noop choices redundant then? | 10:00 |
| josch | and of course patches welcome :) | 10:00 |
| ex-parrot | sweet :) | 10:00 |
| ex-parrot | happy to have a swing at it once I've finished these notes | 10:00 |
| josch | re redundancy: yes, everything that is not "eMMC" or "SD-card" does the same thing: nothing | 10:00 |
| ex-parrot | yep, got it | 10:00 |
| ex-parrot | so we could prob remove one of them? that's what I am trying to be sure of | 10:01 |
| ex-parrot | one of the noops that is | 10:01 |
| josch | yes, removing either would work | 10:01 |
| josch | or by making it a multi-select and then the user can also select nothing | 10:01 |
| ex-parrot | cool | 10:01 |
| ex-parrot | I might have a go at that | 10:01 |
| josch | ex-parrot: the relevant code is this: https://salsa.debian.org/reform-team/reform-debian-installer/-/blob/main/run.sh?ref_type=heads#L237 | 10:02 |
| josch | MY_CHOICES has to be filled depending on the module | 10:02 |
| ex-parrot | yah, I was just reading through :) | 10:03 |
| ex-parrot | that cpio filtering is some dark magic / geniu | 10:03 |
| ex-parrot | s | 10:03 |
| josch | ideally, we'd use /usr/share/reform-tools/machines/*.conf to find out what is supported and what not, so maybe the best patch would look for that in the rootfs and retrieve the variables EMMC_BOOT and SD_BOOT for the current board | 10:03 |
| josch | ex-parrot: yes, the cpio filtering is the "never touch again" kind | 10:04 |
| ex-parrot | unfortunately it seems I deleted my working flash-kernel setup from my last install instead of backing it up | 10:19 |
| ex-parrot | so I'm just going to re-make that now :) | 10:19 |
| - akira (QUIT: Ping timeout: 256 seconds) (~akira@185.111.34.19) | 10:37 | |
| + akira (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 10:38 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 10:39 | |
| + akira (~akira@185.111.34.19) | 10:39 | |
| + andreas-e (~Andreas@2001:861:c4:f2f0::c64) | 10:48 | |
| josch | minute: you were interested whether the bluetooth adapter from handheld scientific would work with the reform keyboard. The answer is yes: https://mister-muffin.de/p/rp6H.jpg | 10:59 |
| josch | minute: it even works with a usb hub in the middle and a reform trackball connected to the hub | 10:59 |
| josch | and then i get both keyboard and mouse via bluetooth | 10:59 |
| josch | now i need to find a usb powerbank on kleinanzeigen and then the whole setup will become wireless \o/ | 11:00 |
| minute | :0 | 11:03 |
| minute | ohh so it's powering via some "local" usb? | 11:03 |
| josch | it just needs 5V into the usb port at the left-hand-side of the white device in the bottom right of the photo | 11:05 |
| ex-parrot | josch: I put notes up here https://lab.hotplate.co.nz/fincham/computers/reform/ in particular you might be interested in the `flash-kernel` thing. I am really not sure that it's "better" or "the right thing to do", I am presenting it merely as an option that might be helpful | 11:05 |
| josch | minute: if you attach the thing into a computer, you can configure it, program it etc | 11:06 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@185.111.34.19) | 11:06 | |
| ex-parrot | I am gonna reformulate these notes at some point and add some stuff about the WIP lpc poweroff patch too | 11:06 |
| + akira (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 11:06 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 11:06 | |
| + akira (~akira@185.111.34.19) | 11:07 | |
| ex-parrot | just going to add a table of contents meanwhile... one sec | 11:07 |
| josch | ex-parrot: correction to your text: the images from MNT and those on reform.d.n mainly differ by what the base suite is not by how vanilla they are | 11:08 |
| josch | ex-parrot: the modifications (done by reform-system-image scripts and reform-tools) are nearly the same independent on whether you use Debian unstable from the MNT repos or Debian stable from reform.d.n | 11:09 |
| ex-parrot | ah yeah, I'll clarify that | 11:09 |
| ex-parrot | I was thinking more about the stock sway install, fonts, etc | 11:09 |
| ex-parrot | and base applications | 11:09 |
| josch | ex-parrot: those should also be the same | 11:09 |
| josch | ex-parrot: reform.d.n uses the same scripts as are used on the MNT gitlab | 11:10 |
| ex-parrot | hmmmmm ok, I might have misunderstood | 11:10 |
| josch | ex-parrot: in fact, Lukas permitted all my changes that make reform.d.n possible to be carried by the official MNT scripts | 11:10 |
| ex-parrot | the installer images just install a vanilla Debian though? | 11:10 |
| josch | so conveniently, the "only" thing that reform.d.n does is to run the same scripts but with different arguments | 11:10 |
| josch | ex-parrot: vanilla plus reform-tools plus reform.d.n repo and patched kernel | 11:11 |
| ex-parrot | ok cool, I'll clarify this in my notes, thanks :) | 11:11 |
| ex-parrot | I am mainly interested in the installer not the prebuilt images, I didn't realise they'd be ~ effectively the MNT shipped images | 11:11 |
| josch | yes, just different base (unstable versus stable) | 11:12 |
| ex-parrot | cool, I'll make that clear | 11:13 |
| ex-parrot | these pages are opted out of google so there's little chance of misleading someone who stumbles on them anyway :P | 11:13 |
| josch | this is also why this is not really a fork | 11:13 |
| ex-parrot | yeah makes sense | 11:13 |
| ex-parrot | personally I just want an empty Debian install w/ the reform-* tools, so the installer is ideal for me | 11:13 |
| ex-parrot | bbs | 11:14 |
| ex-parrot | (having "stable" images is also very cool though, not to diminish your hard work on this!!!) | 11:14 |
| josch | yes, with those requirements, the debian-installer is definitely the better choice | 11:14 |
| josch | ex-parrot: i got no such vibe from you :) | 11:15 |
| josch | ex-parrot: ah okay i now understand how you do the flash-kernel magic | 11:18 |
| josch | i was not familiar with the Boot-Device: option in all.db -- seems only five boards are making use of it | 11:19 |
| josch | this sounds like a useful solution | 11:19 |
| minute | good morning! | 11:37 |
| minute | lol. pocket with a311d didn't boot up today after lid screwed back on etc | 11:37 |
| minute | took off the lid, connected uart... now it boots | 11:37 |
| minute | meh :D | 11:37 |
| josch | classic :) | 11:38 |
| ex-parrot | josch: yeah I think it might be tidier than forcing /boot to be on emmc? I am really not sure though | 11:38 |
| minute | i mean, it did half-boot before i guess, backlight came on but no text output | 11:38 |
| ex-parrot | annoyingly you have to specify the other lines too otherwise it tries to create e.g. "/boot/whatever" on the emmc | 11:38 |
| ex-parrot | I'll mention that also in the notes | 11:38 |
| minute | i will now try migrating to eMMC on a311d pocket :0 :0 | 11:38 |
| josch | ex-parrot: the problem with that solution is, that then you have to tell users of the debian-installer: please make sure *not* to use your eMMC during installation or thereafter -- what if users want to use debian-installer to install their system to eMMC? | 11:38 |
| josch | minute: wait, is that not dangerous? | 11:39 |
| josch | oh nevermind, my /boot is also on eMMC apparently XD | 11:39 |
| minute | haha | 11:39 |
| minute | josch: do you have separate partition for /boot ? | 11:40 |
| josch | yes | 11:40 |
| josch | don't we all since sysimage-v3 | 11:40 |
| minute | sorry, i meant on the emmc | 11:40 |
| minute | my emmc is unpartitioned yet | 11:40 |
| ex-parrot | josch: yeah I agree, it makes the default case work but does still leave that issue. one solution might be to use like "vmlinuz-boot" or something for the filename so it won't collide even if someone is using /boot on emmc. I'll have to test that | 11:40 |
| minute | reform-migrate --help gives some parted commands that only suggest to create a single partition | 11:40 |
| minute | or is that an older version? i foggily remember we talked about this a while ago | 11:41 |
| josch | aha, one sec... | 11:41 |
| minute | also, cc violet i totally forgot yesterday that i already made an R2 version of RCM4 with the pcie reset situation fixed | 11:41 |
| minute | but the one i grabbed from the office is R1 | 11:41 |
| josch | minute: according to git blame i wrote those lines in 2022 XD | 11:42 |
| minute | aha, so that's not updated yet i reckon | 11:43 |
| josch | minute: i'm a bit confused though... this is only for the rootfs, not for /boot | 11:43 |
| minute | ideally ideally it would have the option to create those partitions | 11:43 |
| minute | i'm also confused josch | 11:43 |
| josch | you want reform-migrate be the tool to create partitions? | 11:43 |
| minute | ah, normally reform-migrate doesn't move /boot, right | 11:44 |
| minute | sorry | 11:44 |
| minute | so there's no simple process atm to move completely to emmc | 11:44 |
| josch | i think the best we have is reform-flash-rescue | 11:44 |
| minute | (i realize it's not often needed) | 11:44 |
| minute | ahh right | 11:45 |
| minute | does that use bmaptool and then go in to tweak fstab etc? | 11:45 |
| josch | it does use dd and then calls reform-boot-config at the end which will tweak fstab | 11:45 |
| ex-parrot | josch: fixed up notes to clarify the difference between the complete images and the installer, am out of time for tonight though | 11:45 |
| ex-parrot | see you all perhaps on the weekend when I have some more hours :) | 11:45 |
| minute | josch: aha i see, so that's mostly what i want i guess | 11:45 |
| minute | ex-parrot: cheers! | 11:46 |
| josch | it does not use bmap tool as a) it is not part of the system image and b) we didn't know bmaptool existed when this script was written :) | 11:46 |
| minute | josch: very true | 11:46 |
| josch | so it's only piping curl or wget output into gzip -d into dd | 11:46 |
| minute | josch: i have grown to really like bmaptool a lot | 11:46 |
| josch | same here! | 11:47 |
| minute | imho we can ship it as standard | 11:47 |
| ex-parrot | huh bmaptool looks really handy | 11:47 |
| ex-parrot | ACTION makes a note to investigate further | 11:47 |
| josch | minute: probably, it only takes 164 kB of disk space and just depends on python | 11:48 |
| josch | ex-parrot: it's nice because suddenly your flashing process takes 30% less time | 11:48 |
| josch | and it makes sure that you do not accidentally flash to something important | 11:48 |
| josch | and it downloads and de-compresses for you without temporary files :D | 11:49 |
| ex-parrot | yeah all of that looks great | 11:49 |
| ex-parrot | I can imagine a bunch of uses for this in my so-called every-day life.... | 11:49 |
| josch | according to its maintainer though, the tool is not very actively developed anymore upstream | 11:49 |
| josch | but it's just a bunch of python so it's probably not too hard to keep it alive even if upstream stops caring about it | 11:49 |
| ex-parrot | yeah seems doable | 11:50 |
| ex-parrot | I sometimes write Python code for money, I could probably manage most things if necessary | 11:50 |
| minute | hmm, gnome-disks fails to create partitons on a311d emmc with weird error messages | 11:50 |
| ex-parrot | anyway, I'm actually locking my PC now :) | 11:50 |
| minute | gnome-disks is buggy somehow, parted just works | 11:55 |
| minute | ok, reform-migrating to emmc... | 11:57 |
| minute | i'm extremely curious what happens if i flash uboot to emmc :D | 11:57 |
| josch | either it works or there goes the rest of your vacation to unbrick your a311d? :) | 11:58 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@185.111.34.19) | 12:00 | |
| + akira (~akira@185.111.34.19) | 12:00 | |
| minute | strange, update-initramfs -u doesn't do anything | 12:13 |
| minute | ah, i need to mount /boot i ghuess | 12:13 |
| minute | yup :D | 12:13 |
| minute | josch: lol! | 12:13 |
| minute | it's interesting to do these things at home without all the tools at the company | 12:14 |
| minute | and reform as my primary computer | 12:14 |
| minute | i do have a working system with /boot and / on emmc now, will try to flash uboot | 12:16 |
| minute | josch: aha, uboot loads from emmc, but is confused and wants to do network boot | 12:18 |
| - andreas-e (QUIT: Quit: Leaving) (~Andreas@2001:861:c4:f2f0::c64) | 12:20 | |
| minute | the a311d boot script doesn't try to boot from mmc dev 1 (which is the emmc) | 12:27 |
| minute | but it can still boot from sd card | 12:27 |
| minute | i have a workaround for non-loading wifi on a311d | 12:32 |
| minute | https://community.mnt.re/t/a311d-wifi-issues-firmware-failed-to-leave-lps-state-disconnects/2112/5 | 12:39 |
| minute | hmm strange, we use mainline uboot for a311d but that doesn't try emmc boot?! | 12:45 |
| + chomwitt (~chomwitt@ppp-2-84-146-12.home.otenet.gr) | 12:58 | |
| minute | josch: if we can fix mainline uboot for a311d's boot device list, we can remove the special casing that forbids flashing uboot on emmc | 13:21 |
| minute | i wonder why so many things want to update the linux image | 13:37 |
| minute | i can't install emacs-pgtk because it wants to install linux-image-6.8.12-mnt-reform-arm64 | 13:37 |
| minute | so we'll need to merge josch's binutils workaround | 13:38 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@185.111.34.19) | 14:02 | |
| + akira (~akira@185.111.34.19) | 14:03 | |
| - mjw (QUIT: Killed (erbium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org) | 14:17 | |
| * Guest7521 -> mjw | 14:17 | |
| + Guest3973 (~mjw@gnu.wildebeest.org) | 14:17 | |
| anzu | Hi, people who use the A311D soc, how is your wifi and bluetooth performance? I have noticed that bluetooth headphones are pretty much unusable because of bad connection whenever you move your head. And wifi is really unstable when the router is one floor below, where all other devices would have no problems at all | 14:27 |
| anzu | I wonder if that's normal behaviour if that's my setup only | 14:27 |
| minute | anzu: hmm, i've had good experience with bluetooth, but wifi is limited: i get 30mbit/s download max, and you have to make a config file to disable power saving mode | 14:28 |
| minute | anzu: what kind of antenna setup do you have? | 14:29 |
| minute | anzu: i'm testing it now with my pocket reform where i've just installed the a311d module. streaming music over wifi and to my bluetooth headphones (bose quietcomfort) | 14:33 |
| minute | anzu: i get dropouts when i move a few meters away from the pocket reform but that's because i have only one antenna connected to the module atm | 14:33 |
| + murph_nj (~murph@ool-457bb02e.dyn.optonline.net) | 14:35 | |
| anzu | Actually, now that you mention it, I did not install the antenna holders at all. They are lying around | 14:35 |
| minute | anzu: and no antenna connected? | 14:36 |
| - murphnj (QUIT: Ping timeout: 246 seconds) (~murph@user/murphnj) | 14:37 | |
| anzu | minute: I don't think so | 14:40 |
| minute | anzu: well, that explains it :D | 14:40 |
| minute | impressive that it works without antennas at all | 14:40 |
| anzu | :D | 14:40 |
| josch | minute: could you run your apt-get install emacs-pgtk with a few more debugging options and paste the output somewhere? these options: -oDebug::pkgProblemResolver=true -oDebug::pkgDepCache::Marker=1 -oDebug::pkgDepCache::AutoInstall=1 | 14:41 |
| josch | minute: i made the same experience with various wireless cards | 14:41 |
| josch | not only wifi and bt but also umts modems | 14:41 |
| anzu | I've been using ethernet at home and never used my headphones living alone. Now I travel with the reform and have noticed these things, I think I remember where I placed the antenna. | 14:41 |
| josch | minute: do you have other things that want to update the linux image? i'd like to have a small list of them (i'm collecting those in a code comment) | 14:43 |
| minute | josch: my guess is that it is related to gcc | 14:45 |
| josch | minute: gcc should not relate to the kernel though -- i'm actually curious, so if you have some time and could send me the output that would help me connect this :) | 14:46 |
| minute | josch: http://dump.mntmn.com/emacs-wants-a-kernel.txt | 14:47 |
| minute | ah, it is libc | 14:47 |
| minute | libc6-dev | 14:47 |
| minute | > as Recommends of libgcc-14-dev | 14:47 |
| minute | but --no-install-recommends turns these into depends :D | 14:48 |
| minute | josch: http://dump.mntmn.com/emacs-needs-a-kernel.txt | 14:49 |
| josch | thank you! though i'm staring at that output and it still does not make sense to me. I do not see the connection between linux-libc-dev and linux-image-6.8.12-mnt-reform-arm64 in their dependencies... hrm... | 14:57 |
| minute | josch: i think linux-libc-dev is a kernel package? | 15:02 |
| minute | josch: it has "Version: 6.8.12-1+reform20240605T104210Z" here | 15:02 |
| minute | https://mntre.com/reform-debian-repo/pool/main/l/linux/linux-libc-dev_6.8.12-1%2Breform20240605T104210Z_all.deb | 15:03 |
| ch | IIRC apt tries to keep binaries built from the same source at the same version | 15:03 |
| minute | ch: interesting | 15:03 |
| josch | ch: that is news to me! | 15:03 |
| josch | ch: is that a recent thing? | 15:04 |
| ch | josch: possible that its new in trixie, i remember juliank talking about it on irc | 15:04 |
| josch | interesting... i guess one would need the --no-upgrade option then and that option does exist in bookworm | 15:06 |
| josch | ch: thank you for that -- TIL | 15:06 |
| minute | josch: i tried with --no-upgrade but it still wants to pull in the kernel | 15:09 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 15:19 | |
| anzu | For not having any antenna installed, the wifi actually works excellent | 15:49 |
| anzu | I was able to use it in the office and train with no issues | 15:50 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@185.111.34.19) | 15:53 | |
| + akira (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 15:53 | |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 15:54 | |
| + akira (~akira@185.111.34.19) | 15:54 | |
| - akira (QUIT: Ping timeout: 256 seconds) (~akira@185.111.34.19) | 15:59 | |
| + akira (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 16:00 | |
| Twodisbetter | In the Reform or pocket Reform? | 16:04 |
| - ndufresne (QUIT: Quit: The Lounge - https://thelounge.chat) (~ndufresne@apple.collaboradmins.com) | 16:04 | |
| minute | Twodisbetter: what is the question about? | 16:04 |
| Twodisbetter | minute: anzu: was commenting on wifi performance without the antenna connected. | 16:05 |
| - akira (QUIT: Read error: Connection reset by peer) (~akira@2a01:599:a1d:76c4:2175:8bf9:96d7:1937) | 16:16 | |
| + akira (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 16:16 | |
| anzu | Twodisbetter: in the reform | 16:19 |
| anzu | With A311D | 16:19 |
| minute | need some volunteers for testing this kernel package :D https://source.mnt.re/reform/reform-debian-packages/-/jobs/4716/artifacts/file/repo/pool/main/l/linux/linux-image-6.8.12-mnt-reform-arm64_6.8.12-1+reform20240624T113923Z_arm64.deb | 16:44 |
| minute | i will test on my a311d pocket reform | 16:44 |
| minute | ok, this kernel works for me on a311d pocket, and i have sound and battery status, because the build also includes my updated dtb cc josch | 16:56 |
| minute | trying to load the old qcacld2 module freezes insmod though | 17:01 |
| minute | (lol, strike that--of course there's no qcacld2 on a311d) | 17:02 |
| minute | josch: some a311d wifi mitigations: https://source.mnt.re/reform/reform-tools/-/merge_requests/74 | 17:15 |
| minute | lol, i'm on vacation and working on this stuff all day :D maybe that's enough for now | 17:15 |
| - murph_nj (QUIT: Remote host closed the connection) (~murph@ool-457bb02e.dyn.optonline.net) | 17:18 | |
| mtm | hmm, if Feierabend is quitting time what would the equivalent be for a work-free vacation? Feierferien or some such? | 17:21 |
| + murphnj (~murph@user/murphnj) | 17:30 | |
| - lexik_ (QUIT: Read error: Connection reset by peer) (~lexik@93.185.97.218) | 17:40 | |
| + lexik (~lexik@93.185.97.218) | 17:41 | |
| minute | mtm: Urlaub :D | 17:46 |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:26 | |
| - Guest3973 (QUIT: Remote host closed the connection) (~mjw@gnu.wildebeest.org) | 18:26 | |
| Twodisbetter | anzu: ok, awesome hear. So do you think you will be getting the antenna anyway? | 18:40 |
| - xktr (QUIT: Ping timeout: 240 seconds) (~xktr@user/xktr) | 18:42 | |
| + xktr (~xktr@user/xktr) | 18:44 | |
| anzu | Twodisbetter: Yes, for example visiting my parents, when the router is one floor below me, it already became really unstable. And bluetooth is very unstable as well. I already have the Molex 208482-0100 antenna, I'll connect it when I'm back home :) | 18:48 |
| anzu | I think there's no good reason to use the wifi without an antenna | 18:50 |
| Twodisbetter | very cool. In my own house, I have a mesh network with much overlap and without the antenna connectivity was very spotty. I was surprised by that, but there are number of reasons for that. Happy to hear things are working for you! | 19:46 |
| * mjw -> Guest8770 | 20:28 | |
| - Guest8770 (QUIT: Killed (calcium.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:28 | |
| * mark_ -> mjw | 20:28 | |
| + Guest8770 (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 20:28 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 20:32 | |
| minute | josch: i have reenabled the debian repo and the fixed kernel package is there, too. i will now try to upgrade to it on pocket | 21:02 |
| - akira (QUIT: Ping timeout: 268 seconds) (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 21:06 | |
| + akira (~akira@ip2504e6e1.dynamic.kabel-deutschland.de) | 21:09 | |
| minute | > Linux pocki3000 6.8.12-mnt-reform-arm64 #1 SMP Debian 6.8.12-1+reform20240624T113923Z (2024-06-24) aarch64 GNU/Linux | 21:23 |
| - laumann (QUIT: Changing host) (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 21:53 | |
| + laumann (~quassel@user/laumann) | 21:53 | |
| - chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@ppp-2-84-146-12.home.otenet.gr) | 22:27 | |
| + chomwitt (~chomwitt@2a02:587:7a1a:7400:1ac0:4dff:fedb:a3f1) | 22:29 | |
| josch | minute: yay!! \o/ | 22:47 |
| - chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:587:7a1a:7400:1ac0:4dff:fedb:a3f1) | 22:48 | |
| josch | About linux-libc-dev upgrade upgrading your kernel, this can be disabled by setting APT::Get::Upgrade-By-Source-Package to false | 22:49 |
| + chomwitt (~chomwitt@2a02:587:7a08:700:1ac0:4dff:fedb:a3f1) | 22:50 | |
| pandora | @minute: gonna try the kernel update on the default pocket reform via apt | 23:09 |
| pandora | it still warns about bugs | 23:22 |
| josch | pandora: can you show what you mean? | 23:22 |
| + reform30366 (~user@business-90-187-186-49.pool2.vodafone-ip.de) | 23:24 | |
| reform30366 | https://mister-muffin.de/p/g23f.txt | 23:24 |
| pandora | just doing apt update and then apt upgrade on a fresh system | 23:25 |
| josch | pandora: to make use of Debian unstable less scary, the package apt-listbugs is installed | 23:26 |
| josch | pandora: that package, whenever you install or upgrade packages, will check if doing so might introduce release critical bugs in your system | 23:27 |
| josch | pandora: this is a safety mechanism to avoid downloading and installing packages with known bugs that are marked as "release critical", meaning bugs so severe, that debian will not release with them | 23:27 |
| josch | pandora: if you are interested and do not want to look up details, i can explain to you all these shown bugs except for firefox and ruby | 23:28 |
| ch | i guess the call to action "Restart APT session to enable" is not very clear? in this case that means 'run apt upgrade a second time'(?) | 23:29 |
| pandora | the question would be how someone with an mnt know if those bugs affect their system or not? | 23:30 |
| pandora | and what the proper way of handling would be? put those packages in question on hold? | 23:30 |
| josch | pandora: these bugs are not reform specific -- they require knowledge about debian as an OS to understand them | 23:31 |
| josch | pandora: if you find this scary (i understand and i think it is scary) then you might want to consider running debian stable instead | 23:31 |
| pandora | no i find it interesting. i kinda wanna toy around with the mnt | 23:31 |
| ch | the 7th to last line says > firefox ruby-xmlrpc binutils login will be pinned. Restart APT session to enable | 23:31 |
| pandora | i have my macbook for a stable environment | 23:32 |
| ch | don't need to manually hold the packages | 23:32 |
| - chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:587:7a08:700:1ac0:4dff:fedb:a3f1) | 23:33 | |
| pandora | oh that means those packages wont get upgrade at the moment | 23:33 |
| + chomwitt (~chomwitt@2a02:587:7a10:8c00:1ac0:4dff:fedb:a3f1) | 23:34 | |
| pandora | so thats more like a warning and if i run apt again it will upgrade everything except the pinned packages | 23:34 |
| josch | try it out :) | 23:34 |
| - reform30366 (QUIT: Quit: reform30366) (~user@business-90-187-186-49.pool2.vodafone-ip.de) | 23:35 | |
| + reform12304 (~user@business-90-187-186-49.pool2.vodafone-ip.de) | 23:48 | |
| reform12304 | https://mister-muffin.de/p/8wMR.txt | 23:48 |
| pandora | not enough space :D | 23:48 |
| josch | pandora: are you running this from the sd-card with default partitioning? | 23:49 |
| pandora | yeah ... gonna extend the filesystem rn | 23:49 |
| josch | pandora: i wonder how you run it so that you get this warning: "WARNING: apt does not have a stable CLI interface. Use with caution in scripts." | 23:49 |
| - reform12304 (QUIT: Client Quit) (~user@business-90-187-186-49.pool2.vodafone-ip.de) | 23:50 | |
| josch | ah because you are piping it into a logfile? | 23:50 |
| pandora | yeah exactly | 23:50 |
| pandora | if something goes wrong | 23:50 |
| pandora | that i have a log | 23:50 |
| pandora | it is easier to find issues and share it with people :D | 23:51 |
| pandora | so ... it is upgrading rn | 23:51 |
| ch | i think nowadays there are automatic logs in /var/log/apt - might be easier (but check!) | 23:52 |
| pandora | yeah .... i should check them. Never sure if they log everything i see on the screen | 23:52 |
| josch | yes in /var/log/apt/term.log | 23:52 |
| pandora | ah ok... then i do unnecessary things LD | 23:53 |
| josch | nah, it's totally fine to do it the way you do it i think | 23:54 |
| - chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:587:7a10:8c00:1ac0:4dff:fedb:a3f1) | 23:58 | |
| pandora | hmm | 23:58 |
| pandora | it seems it is stuck or so | 23:58 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
