| staticbunny | and its probably typical internal BS. 90% of the stuff you hear is negative which only accounts for 1% of all activity. Social media really will be the thing that destroys us | 00:10 |
|---|---|---|
| staticbunny | *internet BS | 00:10 |
| josch | minute: anything else you'd like to put into the next reform-tools rolease? | 00:17 |
| minute | josch: none that i can think of at this very moment! | 00:24 |
| - nsc (QUIT: Ping timeout: 264 seconds) (~nicolas@71-97-142-46.pool.kielnet.net) | 00:26 | |
| - staticbunny (QUIT: Quit: My MacBook Air has gone to sleep. ZZZzzz…) (~textual@76-223-253-78.lightspeed.frokca.sbcglobal.net) | 00:27 | |
| + nsc (~nicolas@83-98-142-46.pool.kielnet.net) | 00:28 | |
| + staticbunny (~textual@76-223-253-78.lightspeed.frokca.sbcglobal.net) | 00:45 | |
| - staticbunny (QUIT: Ping timeout: 260 seconds) (~textual@76-223-253-78.lightspeed.frokca.sbcglobal.net) | 00:54 | |
| amospalla | hramrach: thank you (: | 00:57 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:12 | |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 01:21 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 01:21 | |
| - paperManu (QUIT: Ping timeout: 252 seconds) (~paperManu@198.16.214.40) | 02:19 | |
| + jacobk (~quassel@47-186-105-237.dlls.tx.frontiernet.net) | 02:25 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:02 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:03 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:03 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 03:03 | |
| midfavila | SASKATCHEWAN ELECTION LET'S GOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO | 03:24 |
| midfavila | wait | 03:24 |
| midfavila | wrong channel nvm | 03:24 |
| midfavila | meant that for ##Canada lmao | 03:24 |
| midfavila | now to resign in shame | 03:25 |
| - midfavila (PART: !!unknown attribute: msg!!) (midfavila@freeshell.org) | 03:25 | |
| - mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org) | 04:18 | |
| - jn (QUIT: Ping timeout: 252 seconds) (~quassel@user/jn/x-3390946) | 04:21 | |
| + jn (~quassel@2001-4dd4-501d-0-20d-b9ff-fe49-15fc.ipv6dyn.netcologne.de) | 04:21 | |
| - jn (QUIT: Changing host) (~quassel@2001-4dd4-501d-0-20d-b9ff-fe49-15fc.ipv6dyn.netcologne.de) | 04:21 | |
| + jn (~quassel@user/jn/x-3390946) | 04:21 | |
| - spew (QUIT: Quit: spew) (~spew@155.133.15.121) | 04:56 | |
| - cobra (QUIT: Ping timeout: 245 seconds) (~cobra@user/Cobra) | 07:02 | |
| + cobra (~cobra@user/Cobra) | 08:38 | |
| - cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra) | 08:54 | |
| + cobra (~cobra@user/Cobra) | 08:59 | |
| amospalla | My pocket now boots with a big font, wonderful. The change that sets the kernel boot option to use a bigger font will arrive to current Debian/stable? | 09:36 |
| amospalla | If not I'll document that in the forums. | 09:38 |
| josch | amospalla: i'll need to fix the kernel build on reform.d.n but after that's done i see no reason why this change should not also be available to bookworm users | 09:43 |
| amospalla | josch: thank you very much. Yeah, no need then if will arrive. | 09:45 |
| amospalla | I am a bit lost about things that get down into Debian/stable or not. | 09:45 |
| josch | amospalla: i'd love to have more contributors! :) | 09:50 |
| amospalla | josch: knowing bash, python, and linux os in general, what kind of issues could I contribute on? Do you mantain | 09:55 |
| amospalla | a list of issues of things to be done? | 09:56 |
| hramrach | the change cannot be rolled out through upgrades. The canonical place to set the argument is in u-boot, and that does not get upgraded | 09:57 |
| josch | amospalla: you can reproduce the current issue on your own machine by running: | 09:58 |
| josch | sh -xc 'BASESUITE=bookworm-backports; . ./setup.sh; cd linux; . ./build.sh' | 09:58 |
| hramrach | Anyone who already has a pocket they should apply this workaround if they care about font size during early boot | 09:58 |
| josch | amospalla: what hramrach said is technically correct, but you'll need an upgrade of reform-tools to be able to use reform-flash-uboot to upgrade to the latest u-boot version | 09:59 |
| josch | oh no even that... | 10:00 |
| josch | we currently have EMMC_BOOT=false for the pocket reform | 10:00 |
| amospalla | I see. | 10:00 |
| josch | minute: should that be changed? should users other than expert users be able to easily upgrade their imx8mp u-boot, for example for the fontsize fix or for the ${fdtfile} fix? | 10:00 |
| josch | hramrach: if i remember correctly, there were other attempts at fixing the fontsize -- should those be undone? | 10:03 |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@47-186-105-237.dlls.tx.frontiernet.net) | 10:04 | |
| + jacobk (~quassel@47-186-105-237.dlls.tx.frontiernet.net) | 10:04 | |
| amospalla | With the kernel boot argument there is no need for /etc/initramfs-tools/hooks,scripts with setfont and the font itself. | 10:06 |
| amospalla | The kernel itself has the font embedded and already applies it. | 10:07 |
| amospalla | Is there any central faq or wiki where it can noted down that one can set kernel boot arguments through /etc/default/flash-kernel ? It is a useful thing and something Pocket (or other) users may benefit knowing of. | 10:10 |
| hramrach | josch: yes, the early font setting does not work reliably on rockchip | 10:13 |
| hramrach | https://source.mnt.re/reform/reform-tools/-/merge_requests/91 | 10:17 |
| josch | hramrach: perfect, thank you! | 10:19 |
| josch | amospalla: i can think of 3 options: a) open an issue or merge request against reform-handbook.git so that it gets included there b) create a wiki post on the community forum c) create a new wiki article at https://source.mnt.re/reform/reform-debian-packages/-/wikis/home | 10:24 |
| ch | community forum wiki might be best? dunno if the handbook has/wants debian-specific things? | 10:56 |
| josch | ch: but how would you organize these bits of information using the existing forum software? how do the wiki articles become easily findable? | 11:02 |
| josch | posts like this one for example: https://community.mnt.re/t/howto-upgrade-an-existing-mnt-reform-system-on-encrypted-nvme-for-rk3588/2630/1 | 11:04 |
| josch | i guess in the end the solution will win for which there are people using and maintaining it | 11:09 |
| josch | i don't find the gitlab wiki ideal either... | 11:09 |
| grimmware | I think the issue is that both are high friction and unintuitive for most people | 11:13 |
| grimmware | so information is going to continue to go in discourse and get lost over time | 11:13 |
| amospalla | I can think of a wikified thread in the forums that everybody can edit and pinned, resembling a F.A.Q. with one liners like this one (kernel boot args). | 11:15 |
| amospalla | Or other discrete bits of information, or links to other sources of information. | 11:15 |
| grimmware | without worrying too much about who would do it for a second, I do wonder whether a community maintained wiki is a sensible idea. | 11:15 |
| josch | if you ask yourself: suppose you wonder where to set kernel boot arguemnts -- where would you try to find this information? | 11:16 |
| josch | for example right now, /usr/share/flash-kernel/ubootenv.d/00reform2_ubootenv has a commend explaining the LINUX_KERNEL_CMDLINE bit which was the topic yesterday | 11:16 |
| josch | does anybody read that? | 11:17 |
| grimmware | if someone in the community was willing to set up, host and admin a mediawiki (for example) which could then be either linked in a canonical place or even pointed to by a mnt domain | 11:17 |
| grimmware | yeah like I read most of this shit but it's because I'm a hyper obsessed weirdo but even then there are bits that I skip over and more importantly don't recall because they may not be relevant to me at the time | 11:17 |
| amospalla | As an example, coming myself from intel pc, I have no idea what flash-kernel is, I would open a wiki thread or come here and ask (what I did). | 11:18 |
| josch | in my case, i find myself just grepping the irc history for information very often... | 11:18 |
| grimmware | ^ which, again, most people don't have access to so mostly the search function is "ask in irc and josch will either remember or grep it for you" | 11:19 |
| josch | haha | 11:20 |
| josch | wget -nc -r --no-parent https://mntre.com/reform-irc-logs/ | 11:20 |
| josch | this should also be in a wiki page ;) | 11:20 |
| grimmware | ACTION eyeballs AWS OpenSearch | 11:26 |
| grimmware | I don't think we'd ever hit the limit for the free tier for IRC logs | 11:26 |
| + pandora (uid585533@id-585533.ilkley.irccloud.com) | 11:27 | |
| grimmware | that's a pretty big hammer though, I'm pretty sure someone could whip up a javascript-based search for that. | 11:27 |
| grimmware | infrastructure is my hammer, everything looks like a nail | 11:28 |
| grimmware | I'm also very reticent to let bullshit like AWS and lovely things like MNT touch :P | 11:28 |
| amospalla | If the wikified wiki faq is something to try I'd be glad to help mantaining it. | 11:30 |
| josch | amospalla: if you need something changed in the forum, feel free to ping me, i should have the required privileges | 11:32 |
| amospalla | josch: thank you. | 11:33 |
| amospalla | What is the logic of a wikified thread? Who can edit it, needs approvals? | 11:33 |
| josch | amospalla: i wikified this one try to edit it! https://community.mnt.re/t/migrating-encrypted-nvme-to-different-som-imx8mq-to-a311d/1783 | 11:35 |
| sknebel | https://meta.discourse.org/t/editing-and-creating-wiki-posts/30801 | 11:35 |
| josch | <= afk | 11:37 |
| amospalla | yeah, I could edit it. | 11:37 |
| amospalla | sknebel: thank you. | 11:39 |
| amospalla | I understand once a user is trusted on the forums can edit the wiki post and it mantains a history. | 11:39 |
| + paperManu (~paperManu@198.16.214.40) | 11:49 | |
| hramrach | josch: yes, I read /boot/boot.scr which includes that file. But I knew that u-boot scripts exist to start with, that's pretty obscure knowledge already | 11:52 |
| hramrach | grimmware: there is a wiki that uses git backend (unlike mediawiki) which makes the maintenance easier | 11:53 |
| hramrach | one such wiki is ikiwiki but others possibly exist | 11:53 |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 11:56 | |
| hramrach | The discourse is already set up os that's easier so long as we do not hit some limitation of the platform | 11:56 |
| + buckket (~buckket@vps.buckket.org) | 11:57 | |
| grimmware | as long as people can find and edit the content they need to I'm happy tbh, I think we should see how it plays out | 12:02 |
| ch | josch: agree, whatever people use/maintain in the end; i imagine whatever is easiest to edit will happen | 12:03 |
| ch | josch: traditionally i would have said gitlab is the worst, because that wont get indexed by search engines. but search engines have already become useless | 12:03 |
| grimmware | once there are a couple of wiki posts it might be worth making one that's an index of wiki posts and pinning it | 12:03 |
| + mjw (~mjw@gnu.wildebeest.org) | 12:21 | |
| + andreas-e (~Andreas@2001:861:c4:f2f0::c64) | 12:42 | |
| + gustav28 (~gustav@c-78-82-52-4.bbcust.telenor.se) | 13:02 | |
| - andreas-e (QUIT: Quit: Leaving) (~Andreas@2001:861:c4:f2f0::c64) | 13:11 | |
| josch | grimmware: right but does discourse allow for some tree-like structure? | 13:25 |
| josch | for example there is a pane on the left-hand-side of https://meta.discourse.org/t/editing-and-creating-wiki-posts/30801 and that looks nice and i'd like to have that to structure wiki pages for the reform discourse | 13:26 |
| hramrach | I think structure is conterproductive in wiki | 13:28 |
| hramrach | Add links to related pages, not creat inflexible trees | 13:30 |
| amospalla | I don't see any reference to TOC or indexes on discourse wiki posts. | 13:31 |
| amospalla | That means, a FAQ would be a long list of text. | 13:31 |
| hramrach | that's what FAQ always is | 13:32 |
| - pandora (QUIT: Quit: Connection closed for inactivity) (uid585533@id-585533.ilkley.irccloud.com) | 13:37 | |
| grimmware | I really don't think structure is counterproductive, especially if there's already search anyway. That's just a spicy opinion though, I agree that related links should be done wherever possible. | 13:48 |
| grimmware | minute: also got a spare AsiaRF, will send it back once I'm all set up. | 13:49 |
| [tj] | can I send minute money and grimmware can bring it to my house next week? | 13:49 |
| grimmware | oh that might work too | 13:49 |
| - CandidCadaver (QUIT: Ping timeout: 252 seconds) (~aloo_shu@90.166.98.89) | 14:00 | |
| grimmware | what's the currently recommended upgrade path for pocket rcore when you're using encrypted nvme? I'm kinda just assuming it's "take a dump of your home directory, replace the SoM, run the migration script and then try to reinstall all of your stuff"? | 14:00 |
| grimmware | unless I want to try to make josch's instructions for migration to a311d work | 14:01 |
| grimmware | actually that doesn't sound *too* difficult | 14:08 |
| + CandidCadaver (~aloo_shu@90.166.98.89) | 14:10 | |
| grimmware | yeah fuck it I'll give that a try and worst comes to the worst whatever I do won't fuck with the bootability of emmc or sd card so I can always just dump my home directory, get a package list and reinstall | 14:15 |
| grimmware | and there I was thinking I'd get some work done today | 14:16 |
| grimmware | hmm, actually I can probably skip the SD restore image because the image is already on the emmc right? I think a311d only boots from SD or nvme hence the extra step, right? | 14:17 |
| - Aard (QUIT: Read error: Connection reset by peer) (~bwachter@edna-edison.lart.info) | 14:58 | |
| + aard_ (~bwachter@edna-edison.lart.info) | 14:58 | |
| - mjw (QUIT: Killed (iridium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org) | 15:29 | |
| * Guest7761 -> mjw | 15:29 | |
| ch | heres what i did: i booted from the rcore emmc, did a minimal setup using the wizard, fully upgraded that install. then put a new image onto sd card, booted from sd, luksOpened the encrypted NVMe, chrooted into that, updated /etc/fstab (mmcblk* changes), update-initramfs -u -kall, rebooted without sd card, and that just worked | 15:40 |
| ch | all iirc | 15:40 |
| gsora | did you use OPAL encryption by chance? | 15:41 |
| josch | grimmware: it's funny that you earlier suggested that most people just ask josch to grep irc for them because just today i pasted this link only 10 minutes before your suggestion and that's what you wanted to have, no? https://community.mnt.re/t/howto-upgrade-an-existing-mnt-reform-system-on-encrypted-nvme-for-rk3588/2630/1 | 16:04 |
| grimmware | oh god damn I am the worst | 16:25 |
| grimmware | thank you | 16:25 |
| grimmware | you know what the worst part is? I've definitely read this before | 16:27 |
| grimmware | and favourited it | 16:28 |
| josch | you see, that's what grep is for -- so that your brain is free to do more useful activities | 16:30 |
| josch | like trying to get factorio to run on a311d... | 16:31 |
| grimmware | josch: I am eternally thankful for your patience. | 16:43 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:57 | |
| * aard_ -> aard | 18:18 | |
| + mrbcmorris (~mrbcmorri@1513413-static.lxtnkya3.metronetinc.net) | 18:29 | |
| - mrbcmorris (QUIT: Changing host) (~mrbcmorri@1513413-static.lxtnkya3.metronetinc.net) | 18:30 | |
| + mrbcmorris (~mrbcmorri@user/mrbcmorris) | 18:30 | |
| + spew (~spew@201.141.99.170) | 18:42 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:56 | |
| minute | ch: how do you like the rk3588? | 18:56 |
| * mjw -> Guest1169 | 18:59 | |
| - Guest1169 (QUIT: Killed (zinc.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:59 | |
| * mark_ -> mjw | 18:59 | |
| + Guest1169 (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:59 | |
| grimmware | minute: I'm following the rcore pocket nvme migration instructions and (other than some other errata that I'll share) I'm installing the kernel and getting "Unsupported platform 'MNT Pocket Reform with RCORE RK3588 Module'." from update-initramfs | 19:44 |
| grimmware | flash-kernel is allegedly at the newest version tho | 19:52 |
| + irimi1 (47a15995ae@2a03:6000:1812:100::12aa) | 19:53 | |
| ch | haven't used it much unfortunately :( but its a lot better than the imx8mp | 19:56 |
| ch | grimmware: i'd check if reform-tools is current (but thats just a wild guess) | 19:59 |
| minute | grimmware: i would suggest the same as ch, try apt updating and apt install reform-tools and also the latest kernel image | 20:08 |
| grimmware | hmm no dice | 20:11 |
| grimmware | I gotta go eat then I'll come back to it | 20:11 |
| minute | grimmware: this is chrooted into an old install? | 20:13 |
| digitalrane | even though talking to josch is very nice, most irc bouncers also have a module for logging to disk which makes for very convenenient josch-free grepping as well | 20:16 |
| digitalrane | i spent some time battling the boundary uboot fork to get DSI activated and rge device tree activated and i haven't been anle to get uboot to try and activate it on display enumeration on imx8mp pocket yet | 20:18 |
| digitalrane | *the | 20:18 |
| - amospalla (QUIT: Ping timeout: 252 seconds) (~jordi@user/amospalla) | 20:21 | |
| grimmware | minute: yeah | 20:35 |
| minute | grimmware: you can try reform-check, it might be old files from /etc shadowing the new stuff under /usr, i.e. the machines db for flash-kernel | 20:46 |
| grimmware | `apt reinstall flash-kernel` looks like it might have worked | 21:06 |
| minute | oh, looks like we're on 6.11.5 now | 21:08 |
| grimmware | Hmm, so when I rebooted it hung on systemd-udevd so I rebooted again and it’s booted off of the sd card again | 21:10 |
| grimmware | On the .4 kernel | 21:11 |
| grimmware | Gonna redo some of the install steps I think | 21:11 |
| minute | security PSA https://mastodon.social/@mntmn/113392534680330930 (yesterday's issue) | 21:13 |
| grimmware | Okay now it’s just not trying to boot the new kernel at all | 21:17 |
| grimmware | Wait wtf | 21:18 |
| minute | ugh | 21:18 |
| - spew (QUIT: Ping timeout: 246 seconds) (~spew@201.141.99.170) | 21:18 | |
| grimmware | The kernel is missing from /boot? | 21:18 |
| hramrach | the file mode of the devil | 21:19 |
| grimmware | Oh shiiit I see what I’ve done | 21:19 |
| minute | grimmware: what have you done! | 21:20 |
| grimmware | Wrong partition | 21:20 |
| grimmware | emmc instead of sd | 21:21 |
| grimmware | Actually should that work if I take the sd out? | 21:21 |
| grimmware | One quick way to find out I guess | 21:22 |
| grimmware | lol yeah totally does | 21:22 |
| minute | oh | 21:23 |
| grimmware | So I had 2 problems, I needed to reinstall flash-kernel because the db didn’t have the entry in and I also chose the wrong /boot (the emmc) to install on which, thankfully, is actually where I wanted it installed in the long run anyway | 21:24 |
| minute | phew | 21:25 |
| grimmware | fuck me this thing is a lot faster | 21:25 |
| grimmware | much better wifi performance as well | 21:25 |
| hramrach | hm, was the WiFi on iMX8MP resolved somehow? | 21:28 |
| grimmware | hramrach: probably more "sufficiently mitigated" | 21:28 |
| hramrach | It still crashes occasionally for me | 21:28 |
| grimmware | oh I never had it crash | 21:29 |
| grimmware | hramrach: did you add a thermal pad and migrate to SSD? | 21:30 |
| vkoskiv | Wonder what the deal is with DRAM prices. I was looking at 4x16GB DDR4, and the cheapest prices for used stuff on ebay are ~same as retail. Huh?! | 21:30 |
| hramrach | only added a thermal pad. Does migrating to SSD help? | 21:30 |
| vkoskiv | 10 year old RAM standard. And somehow DDR5 is cheaper? | 21:30 |
| hramrach | the current RAM tends to be cheaper, or that one before when the current one is very fresh | 21:31 |
| grimmware | hramrach: anecdotally, yes | 21:31 |
| hramrach | anecdotally, adding thermal pad works .. until it does not. The difference is not so big that I can be even 100% sure it did something. It's not like it became stable | 21:32 |
| hramrach | but it does seem like under load the wifi is more likely to drop | 21:33 |
| grimmware | yeah I found that once I migrated to nvme it was suddenly stable. Reception sucked but it was stable. I think it's the difference in the thermal load when you migrate disk access away from the emmc on the SoM | 21:33 |
| minute | makes sense | 21:34 |
| hramrach | but if moving to NVMe works then it might be bus problem, both use MMC, right? | 21:34 |
| minute | interesting idea | 21:34 |
| minute | the wifi is sdio yeah, different controller but knowing NXP it could be some weird QoS issue | 21:35 |
| minute | i've never thought of that | 21:35 |
| hramrach | there is the thing that when the wifi stops working reloading the driver fixes it | 21:35 |
| minute | a311d(bpi) also had bizarre wifi issues that were mitigated by increasing its sdio freq | 21:36 |
| hramrach | which means something got into a bad state, and diont the initialization again, reloading firmware, etc does something that fixes it | 21:36 |
| minute | hramrach: yep makes a lot of sense | 21:36 |
| grimmware | tfw you start a conversation and then you realise the big kids are finishing it hahah | 21:37 |
| - jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@47-186-105-237.dlls.tx.frontiernet.net) | 21:40 | |
| josch | minute: small update from my side: i think the permission problem of apt sources.list is fixed now but in the process of fixing it i found more problems | 21:49 |
| josch | specifically mkimage.sh now runs this easy to remember one-liner: | 21:49 |
| josch | ind "." -xdev \! -path ./dev/full \! -path ./dev/null \! -path ./dev/random \! -path ./dev/tty \! -path ./dev/urandom \! -path ./dev/zero \( \! -type c \) \( \! -type l \) \( \! -type d -o \! -perm /o+t \) \( \( -gid 0 -perm /g+w \) -o -perm /o+w \) -print0 | 21:49 |
| josch | essentially, trying to find more files affected by the same problem | 21:50 |
| josch | and mkimage.sh as run by the CI has a few more files with a similar symptom but when run locally, there is no problem | 21:50 |
| josch | i've been investigating this since yesterday but it seems i'll need more time to get to the bottom of this... | 21:51 |
| grimmware | minute: this thing runs like an absolute dream <3 | 21:51 |
| grimmware | I can't believe I just upgraded the cpu in my laptop | 21:52 |
| grimmware | 18 year old me would be overwhelmed with all of this | 21:52 |
| josch | grimmware: congratulations! :) | 21:52 |
| josch | maybe somebody [tm] can now do some battery runtime comparisons with rk3588 | 21:53 |
| josch | just 2 hours ago there was another forum post asking for some data on this | 21:53 |
| josch | i think this has been a long sought staticstic for potential rk3588 buyers :) | 21:53 |
| grimmware | what sort of load do you reckon it should be under? | 21:53 |
| josch | grimmware: you are on the pocket? | 21:54 |
| grimmware | yeah | 21:54 |
| josch | when jeff geerling tried benchmarking the battery runtime of the classic reform, this video was played in fullscreen in firefox: https://www.youtube.com/watch?v=_DA0Jr4WH-4 | 21:56 |
| josch | at full display brightness | 21:56 |
| josch | a311d reform with fresh eremit cells lasts 4:45 with that | 21:56 |
| josch | make sure to disable upower first or otherwise it might shut off your reform before the batteries are truly empty | 22:04 |
| josch | (which is what i think happened to jeff when, in the video, the reform suddenly initiated a soft shut-down after around 2.5 hours) | 22:05 |
| grimmware | hmm, I don't appear to be getting any sound any mroe | 22:10 |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-52-4.bbcust.telenor.se) | 22:15 | |
| minute | josch: that's a nice oneliner! | 22:26 |
| minute | grimmware: very happy to hear about the upgrade feelings! | 22:26 |
| minute | grimmware: check alsamixer, f6, rk3588-tlv320... | 22:26 |
| minute | grimmware: there's a bunch of toggles that need to be on, maybe reform-hw-setup missed some. like "Output Left From Left DAC", "Output Right From Right DAC" and all the Speaker things. (toggle with 'm' key) | 22:27 |
| josch | minute: i think i now understand what is happening. git only tracks the executable bit of files, not the other permissions bits. And it seems that unless you have FF_DISABLE_UMASK_FOR_DOCKER_EXECUTOR set for your gitlab runners, the git repo is cloned with umask 0000 -- so all files will be world read/writable by default | 22:51 |
| josch | minute: this means that the files mkimage.sh copies into the chroot already have the wrong permission bits and all need to be fixed up | 22:51 |
| minute | josch: urgh | 22:55 |
| + spew (~spew@201.141.99.170) | 22:56 | |
| grimmware | minute: that's fixed it! | 22:59 |
| grimmware | I listened to HEALTH as my test track :3 | 23:01 |
| CandidCadaver | quick, alsactl store and an offering of incense | 23:07 |
| - spew (QUIT: Ping timeout: 248 seconds) (~spew@201.141.99.170) | 23:07 | |
| CandidCadaver | in an ideal world, that ought to persist the card's state | 23:08 |
| minute | grimmware: nice | 23:08 |
| + spew (~spew@201.141.99.170) | 23:09 | |
| hramrach | Is there any reason to not use a general purpose firmware like QMK for the keyboard? | 23:12 |
| hramrach | there is reform-specific functionality but that's small part of the code | 23:14 |
| minute | hramrach: when i started the actual keyboard part seemed kind of tiny compared to that other stuff | 23:14 |
| minute | and i have no personal experience or history with qmk | 23:15 |
| minute | i figured that people who like qmk make that work | 23:16 |
| grimmware | yeah I've done a bunch with qmk and found that I couldn't be bothered to try to figure out how to implement the pocket keyboard for it because the layout was easy enough to modify | 23:19 |
| grimmware | I'd maybe consider implementing tap vs hold functionality so I could put hyper on the space key or something but I don't want it enough to do it right now | 23:20 |
| grimmware | need to polish off the accelerometer work and write it up | 23:20 |
| hramrach | the point is that these general firmware tools get stuff like layout configurator, can be built for multiple platforms (consider unified firmware for keyboerd 3 and 4), etc. | 23:24 |
| hramrach | that is a wheel that somebody already invented and polished to be reasonably round | 23:25 |
| hramrach | I think there are other general purpose keyboard firmware builders besides QMK (other than its forks) I just don't remember them | 23:27 |
| grimmware | minute: by the way, I don't know if you caught it earlier but I also got a second AsiaRF in the box and I'm going to see [tj] next week and he wants one so we were hoping I could just give it to him and he could just send you the money for it | 23:34 |
| grimmware | saves on postage and hassle for everyone | 23:34 |
| minute | hramrach: sure, feel free to do a qmk fork | 23:34 |
| minute | grimmware: oh yeah for sure, much appreciated | 23:42 |
| - josch (QUIT: Ping timeout: 265 seconds) (~josch@mister-muffin.de) | 23:55 | |
| + josch (~josch@mister-muffin.de) | 23:55 | |
| - Gooberpatrol_66 (QUIT: Remote host closed the connection) (~Gooberpat@user/gooberpatrol66) | 23:58 | |
| + Gooberpatrol_66 (~Gooberpat@user/gooberpatrol66) | 23:59 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!