| NanoCodeBug | this dts entry was based on the reference one for the som. | 00:53 |
|---|---|---|
| NanoCodeBug | i did learn that mapping to spidev changed at some point in 5.0+ kernel, which might have caused issues | 00:56 |
| cinap_lenrek | minute: doesnt the lpc also have a unidirectional serial connection to the imx8? | 00:57 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 00:57 | |
| minute | cinap_lenrek: yeah | 01:20 |
| + mark_ (~mark@gnu.wildebeest.org) | 01:25 | |
| - mjw (QUIT: Ping timeout: 246 seconds) (~mark@gnu.wildebeest.org) | 01:25 | |
| - mark_ (QUIT: Quit: Leaving) (~mark@gnu.wildebeest.org) | 01:59 | |
| - Nulo (QUIT: *.net *.split) (~Nulo@user/nulo) | 02:00 | |
| - Gooberpatrol66 (QUIT: *.net *.split) (~Gooberpat@user/gooberpatrol66) | 02:00 | |
| - bluerise (QUIT: *.net *.split) (~bluerise@user/bluerise) | 02:00 | |
| - buckket (QUIT: *.net *.split) (~buckket@pdp8.buckket.org) | 02:00 | |
| - grubman (QUIT: *.net *.split) (~ufotofu@149.248.16.17) | 02:00 | |
| - ezequielg (QUIT: *.net *.split) (sid363064@id-363064.uxbridge.irccloud.com) | 02:00 | |
| + bluerise (~bluerise@p5b0acfb5.dip0.t-ipconnect.de) | 02:00 | |
| + Gooberpatrol66 (~Gooberpat@2603-8080-4540-7cfb-0000-0000-0000-113a.res6.spectrum.com) | 02:00 | |
| + buckket (~buckket@pdp8.buckket.org) | 02:00 | |
| + grubman (~ufotofu@149.248.16.17) | 02:00 | |
| + ezequielg (sid363064@id-363064.uxbridge.irccloud.com) | 02:00 | |
| - Gooberpatrol66 (QUIT: Changing host) (~Gooberpat@2603-8080-4540-7cfb-0000-0000-0000-113a.res6.spectrum.com) | 02:01 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 02:01 | |
| + Nulo (~Nulo@user/nulo) | 02:01 | |
| - mtm (QUIT: Ping timeout: 272 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 02:04 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 02:14 | |
| - scops (QUIT: Ping timeout: 260 seconds) (~scopstchn@2001:470:69fc:105::8da) | 02:15 | |
| + scops (~scopstchn@2001:470:69fc:105::8da) | 02:19 | |
| bkeys | Man I hate docker so much | 02:40 |
| bkeys | Normally software takes 1 command to install, docker supposedly makes it so "easy" | 02:41 |
| bkeys | https://docs.funkwhale.audio/stable/installation/index.html | 02:41 |
| bkeys | How the heck does it make this software easier to install? This is terrible | 02:41 |
| bkeys | </rant> | 02:41 |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:587:dc1c:800:5b3a:daea:252e:ac17) | 02:44 | |
| klardotsh | bkeys: IMO it makes orchestrating complexity easier, but doesn't make _installing_ software that is otherwise a relatively easy fit for traditional unix package managers much easier. it's a shame folks tend to brand it as such. | 02:45 |
| klardotsh | my day job would be horrible trying to manage everything through sysvinit or systemd unit files, dpkg/apt/rpm/whatever, manual iptables forwarding, ruby version locking, etc. etc. etc. | 02:46 |
| klardotsh | but most stuff... isn't in that realm | 02:46 |
| bkeys | My experience with it has been awful, I have spent days on and off trying to install funkwhale so my wife can maybe drop apple music but it forces me to use docker | 02:46 |
| bkeys | I'd have this up in less than a half hour if it weren't for docker | 02:46 |
| klardotsh | then again, a service like funkwhale that seems to document 5-6 required services just to do anything seems to be EXACTLY the type of thing Docker (and its ilk) + Compose, Swarm, Nomad, whatever, excels at and I would refuse to touch in almost any other system (except maybe NixOS) | 02:47 |
| bkeys | It's a ppc64le server, 80% of docker packages don't even exist for my architecture | 02:47 |
| flowy | bkeys: have you seen navidrome? funkwhale looked too complex for me to try. | 02:47 |
| bkeys | I already have navidrome at music.bkeys.org but it streams like crap | 02:47 |
| flowy | hm. using the web app? | 02:48 |
| bkeys | Music will stutter and sometimes just stop playing all together | 02:48 |
| bkeys | Phone app | 02:48 |
| flowy | which phone app | 02:48 |
| bkeys | I want my wife to stop using apple music, I have the music hoarded on a drive I just need a way to stream it | 02:48 |
| bkeys | I am using Subtracks which feels like a really solid app | 02:48 |
| bkeys | iPhone doesn't have nearly as good apps as android does; but even a guy like me who doesn't demand a super smooth experience found it unusable | 02:49 |
| bkeys | I'd be playing music while I ride my bike and it would just cut out for minutes at a time. | 02:49 |
| klardotsh | huh. my friend ran a navidrome server I was able to use just fine with dsub on android. wonder if that's a client-side issue. have you tried running a PC client of some sort to test? | 02:50 |
| bkeys | I can't ask my wife to cut off her apple music subscription if it does that on a regular basis | 02:50 |
| flowy | bkeys: yeah... i'm honestly considering starting an iOS subsonic client as my next hobby project | 02:50 |
| flowy | play:sub is acceptable though | 02:50 |
| flowy | once you've figured out the bugs | 02:51 |
| bkeys | My wife needs it more than I do and she is not as willing as I am to accept bad user experience | 02:51 |
| bkeys | Thankfully we use Kodi and a hoard of videos and have cut out netflix and all the other streaming services, which saves good money. | 02:52 |
| flowy | amperfy might be worth a try. it looks good | 02:52 |
| bkeys | Yeah, it'd be nice if I could use it but it's moreso for my wife | 02:52 |
| flowy | i would guess if your navidrome is up to date, the problems are not likely to be coming from there | 02:53 |
| flowy | it's getting pretty mature | 02:53 |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 02:59 | |
| bkeys | I just pulled up subtracks on my phone and yeah it's already breaking | 03:00 |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 03:00 | |
| bkeys | I found d sub on android but it says it requires google play services and a donation; I don't mind donating even if I end up not using it but not if it uses google play services | 03:01 |
| bkeys | Funkwhale looks really nice and cushy it's so unfortunate that they force the usage of docker | 03:01 |
| klardotsh | thing is: objectively they don't. they recommend it for reasons I agree with (that's the type of app where it pays off), but they link to packages for Arch, NixOS, and I think I saw a couple others, if you want to DIY it up without Docker | 03:03 |
| klardotsh | (now, it having so many moving parts? that's an architectural discussion with potential merit :) ) | 03:04 |
| bkeys | I'd be okay if I could follow their instructions and have docker actually work; it just never does | 03:11 |
| bkeys | Like there are these images that the funkwhale depends on, I try to run the docker up -d it mentions in their instructions and it just errors out so I have to manually install them | 03:12 |
| bkeys | Even then it still doesn't work | 03:12 |
| bkeys | If I had to guess celeryworker and celerybeat don't have images for ppc64le | 03:13 |
| - klardotsh (QUIT: Ping timeout: 255 seconds) (~klardotsh@172.56.104.172) | 03:21 | |
| - arminweigl (QUIT: Ping timeout: 255 seconds) (~arminweig@sourcehut/user/arminweigl) | 03:42 | |
| + arminweigl (~arminweig@sourcehut/user/arminweigl) | 03:43 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:20) | 03:47 | |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 04:10 | |
| + chomwitt (~chomwitt@2a02:587:dc1c:800:5c71:ba1b:727c:3afe) | 08:24 | |
| + MajorBiscuit (~MajorBisc@86-88-79-148.fixed.kpn.net) | 08:30 | |
| + Major_Biscuit (~MajorBisc@c-001-009-008.client.tudelft.eduvpn.nl) | 08:31 | |
| - MajorBiscuit (QUIT: Ping timeout: 276 seconds) (~MajorBisc@86-88-79-148.fixed.kpn.net) | 08:36 | |
| + chartreuse (~chartreus@S0106f0f249dfd9c3.cg.shawcable.net) | 09:08 | |
| - bluerise (QUIT: Changing host) (~bluerise@p5b0acfb5.dip0.t-ipconnect.de) | 10:23 | |
| + bluerise (~bluerise@user/bluerise) | 10:23 | |
| Boostisbetter | Is hibernation something that could be implemented with the Reform? For me this seems like the easiest solution to the battery draining issue. | 10:57 |
| Boostisbetter | Granted the status of every Reform user is not known. It would be better to only use hibernation when you are using an NVME drive and not just the SD card. | 10:58 |
| Boostisbetter | Still I would LOVE it to be implemented on the Reform. It will be even more helpful on the Pocket Reform as it is even more portable than the Reform itself. | 10:58 |
| + Christoph_ (~Christoph@p4fe73e14.dip0.t-ipconnect.de) | 11:47 | |
| - tinybronca[m] (QUIT: Ping timeout: 248 seconds) (~tinybronc@2001:470:69fc:105::2:1af6) | 12:04 | |
| - scops (QUIT: Ping timeout: 264 seconds) (~scopstchn@2001:470:69fc:105::8da) | 12:06 | |
| + scops (~scopstchn@2001:470:69fc:105::8da) | 12:30 | |
| minute | Boostisbetter: how does hibernation solve battery drain of turned-off computer? | 12:31 |
| + tinybronca[m] (~tinybronc@2001:470:69fc:105::2:1af6) | 12:31 | |
| - iank (QUIT: Quit: ZNC 1.8.2+deb2+b1 - https://znc.in) (~iank@fsf/staff/iank) | 12:32 | |
| + iank (~iank@fsf/staff/iank) | 12:32 | |
| Boostisbetter | minute: it doesn't. The computer being completely shutdown is the best solution for power savings. However, I keep my Reform constantly on but in standby. This means I have to constantly keep track of battery status, which is no problem. I do this because it is a pain to constantly have to reopen things, etc. | 12:44 |
| Boostisbetter | But by being able to hibernate I can save my session, and resume it again later, without wasting battery power. | 12:44 |
| Boostisbetter | Granted for many people it doesn't really make sense, because they don't really care about preserving a session. | 12:45 |
| Boostisbetter | Doing so would provide a convenience that would save power and wear and tear though. | 12:45 |
| - iank (QUIT: Quit: ZNC 1.8.2+deb2+b1 - https://znc.in) (~iank@fsf/staff/iank) | 13:25 | |
| + iank (~iank@fsf/staff/iank) | 13:27 | |
| minute | Boostisbetter: yeah, hibernation does theoretically work but there was some snag on resuming also | 13:48 |
| sknebel | minute: pocketreform article also at the top of HN | 13:49 |
| minute | yeah, but old (updated though) article | 13:50 |
| josch | Boostisbetter: hibernation is currently disabled on systemd level. After enabling it and trying it I found that it (similar to suspend) only sometimes comes back successfully: https://mntre.com/reform-irc-logs/2022-06-20.log.html#t23:43:06 | 13:57 |
| minute | i wonder if related to the wakeup issue. | 14:00 |
| minute | also, we should explore if we can use upstream TF-A instead of boundary/nxp TF-A | 14:00 |
| minute | currently we have a patch that adapts sleep/wake code in the kernel to nxp TF_A behavior | 14:01 |
| minute | that is because back then, sleep/wake with upstream code did not work at all | 14:01 |
| minute | but i don't know if the situation has changed perhaps | 14:01 |
| - mtm (QUIT: Ping timeout: 246 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 14:02 | |
| minute | another related TF-A hack https://github.com/varigit/meta-variscite-imx/commit/ac9ba789ec9323de1087adfc8b0b2d8a47e2893b | 14:03 |
| minute | ah, this hack is already in imx-atf | 14:04 |
| josch | i'd be willing to shell out a few hundred bucks to get suspend and hibernate working -- maybe somebody can set up a code bounty thing or similar if it's just about somebody needing to get compensated for the time necessary to figure this out? | 14:05 |
| minute | from my POV it's cool to post a bounty on community.mnt.re. maybe others would chime in | 14:06 |
| + kuno (sid299442@id-299442.helmsley.irccloud.com) | 14:11 | |
| * wielaard -> mjw | 14:18 | |
| minute | josch: purism have similar issues https://source.puri.sm/Librem5/linux/-/merge_requests/600 | 14:24 |
| minute | https://source.puri.sm/Librem5/linux/-/issues/428 | 14:24 |
| josch | well purism already got 600 EUR from me and even now three years later i got no hardware -- so in principle purism already got "a few hundred bucks" from me without giving me anything in return, maybe they should use it to fix suspend for us ;) | 14:29 |
| minute | well well :D | 14:31 |
| mjw | ACTION just got his librem, so they do eventually ship hardware | 14:52 |
| mjw | I ordered October 2017 though, so it does take 4+ years... | 14:54 |
| minute | mjw: how do you like it? does suspend/wake work? | 15:06 |
| mjw | minute, it is sturdy, like the mnt reform, it also seems as fast as my mnt reform, and it gets as hot as my mnt reform (which isn;t too hot and fine for the laptop, but somewhat too hot imho to keep it in my pocket) | 15:10 |
| mjw | minute, I don't think it does suspend/wake or has very good power settings, I can use if for a couple of hours on full battery, but I really need to plug it in whenever I can. | 15:12 |
| minute | mjw: so you can't put it in standby? | 15:27 |
| mjw | let me see | 15:27 |
| minute | i mean, i'm outside right now with an android phone | 15:29 |
| mjw | you people assume I am technical, I really am not. Looking at the settings | 15:29 |
| minute | i type some messages and press power button and have it in my pocket | 15:29 |
| minute | no i mean just from a user perspective | 15:29 |
| mjw | I can turn on automatic suspend, which pops up a warning message that it is experimental | 15:29 |
| minute | i assume you can "lock" the phone and it kinda sleeps then? | 15:29 |
| minute | isn't it like a normal smartphone in this regard? | 15:30 |
| mjw | ACTION doesn't know what a normal smartphone is. My previous phone was a very simple nokia which they sell to seniors because it has big buttons. | 15:30 |
| mjw | I do own a pinephone now, which is somewhat similar to the librem. | 15:31 |
| minute | oh | 15:31 |
| minute | you never used an iphone or an android phone? | 15:31 |
| mjw | It doesn't get that hot and because I got the keyboard with it (which has a large battery) it works for ~24 hours, which is nice. | 15:32 |
| minute | what do you do when you want to put librem in your pocket? does the display just stay on? | 15:32 |
| mjw | no, although my children do have and make fun of me. | 15:32 |
| mjw | No the screen turns off. But it always keeps running. That is why it is somewhat hot. | 15:33 |
| minute | ah. | 15:33 |
| mjw | It is not unbearable. Just, aha, that is why the battery is draining so fast. | 15:35 |
| mjw | Basically I thought I wanted a "smart" phone. But now that I have seen the MNT Pocket Reform I think what I really want is that! :) | 15:36 |
| minute | ha! | 15:43 |
| mjw | Actually it is nice to be able to make phone calls. But both the pinephone and the librem are not great at making phone calls. It works, but it echos or audio quality is meh. I'll often end the call shouting I'LL SENT YOU AN EMAIL! | 15:52 |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 16:01 | |
| minute | mjw: phonecalls without a headset? | 16:19 |
| mjw | headset? | 16:21 |
| mjw | never tried that, but it does have a headset plug. Maybe I should try that. | 16:21 |
| mjw | ACTION is a bad phone person and a terrible computer user. But... o well. | 16:22 |
| josch | mjw: that reminds me a lot of my experience with the openmoko freerunner | 16:22 |
| flowy | minute: so yeah. it seems i can actually reliably hang my new keyboard if i shutdown overnight. 4/4 nights, also upgraded the LPC in between for good measure. newest kb firmware. i'm thinking my next steps will be swapping the old keyboard back in, and if the problem disappears, i'll figure out exactly which firmware it has and flash that to the new keyboard | 16:53 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 16:53 | |
| flowy | if that doesn't help maybe i'll use this as an excuse to finally start learning how to debug electronics, maybe there's a difference between the boards that can be found | 16:54 |
| flowy | my old keyboard _never_ hung and i shutdown every night. | 16:54 |
| minute | flowy: ok, the new one was from a standalone? does it have matte black soldermask? | 16:59 |
| minute | hm: # echo 0 > /sys/power/pm_async | 17:00 |
| minute | This option, which is normally enabled, causes Linux to suspend and resume all the devices in parallel to save time. Disabling async suspend/resume can be used to check if the suspend/hibernate failure is caused by some unknown device dependency issue. This can also be used to tune driver suspend/resume latency. | 17:00 |
| NanoCodeBug | hmm yeah I've having a keyboard resume issue too, i flashed the latest firmware for it a few days ago, they keyboard refuses to wakeup. disconnecting the usb power to it resets it, but the reset button does not | 17:21 |
| flowy | NanoCodeBug: really? for me i can take the bezel off and hit the kb reset button. you're giving a solid push? | 17:23 |
| NanoCodeBug | yus it makes a click and everything | 17:24 |
| NanoCodeBug | i know its not the lpc either, triggering a reset of that doesn't bring the kb oled back | 17:24 |
| minute | yeah, the keyboard doesn't need the lpc to work | 17:30 |
| flowy | NanoCodeBug: reminds me of this post: https://community.mnt.re/t/power-issues-and-lpc-wakeup/596/37 | 17:30 |
| minute | i might sound like a broken clock, but does this workaround work for you? https://source.mnt.re/reform/reform/-/merge_requests/26 | 17:31 |
| minute | broken record i mean | 17:31 |
| minute | idioms are hard | 17:31 |
| flowy | NanoCodeBug: does your reset button normally have issues? i.e. under normal circumstances | 17:32 |
| NanoCodeBug | neat! i'll flash that and see what happens | 17:32 |
| NanoCodeBug | no, it works fine for entering program mode | 17:32 |
| flowy | minute: i think both my boards have black masks? but i'm not sure exactly how to tell or what you mean :D | 17:33 |
| flowy | yes the new one is from standalone | 17:33 |
| flowy | and the old one came with my reform which is dated march 2021 | 17:34 |
| minute | flowy: there's a newer board that has matte instead of shiny black mask | 17:36 |
| minute | flowy: also it has a arduino style ISP header (but only the holes for it) and the old one does not have that | 17:36 |
| minute | holes/pads | 17:37 |
| flowy | ok i'll take it out and compare | 17:37 |
| flowy | later tonight | 17:38 |
| flowy | from the bit that is peeking out over the motherboard, from the backside, it looks matte | 17:38 |
| flowy | ah yeah i see the holes too | 17:41 |
| flowy | my old board is shiny and doesn't have pads | 17:41 |
| flowy | i guess my new board is the newer board | 17:42 |
| minute | aha! | 17:44 |
| flowy | i was also having the weirdest problem during first attempts at re-assembly after upgrading the keyboard and some otehr stuff where not only was the system unresponsive but the white LEDs were turning on immediately upon receiving battery power. but no boot. no idea if that was related, or even how i fixed it really. | 17:46 |
| flowy | i thought it was related to the OLED screens not being properly connected but i probably inadvertantly did something else like some other shaky connector | 17:46 |
| flowy | i don't particularly mind the keyboard hanging right now because i'm at home. but next time i travel it would be pretty inconvenient so i wanna figure it out. got some weeks to slowly debug. i'll try that patch after trying this other stuff i've mentioned first | 17:48 |
| flowy | i'd like to see if my old keyboard's firmware solves it | 17:49 |
| flowy | it would be interesting | 17:49 |
| minute | flowy: did you already desolder the 3v3 ldo? | 17:50 |
| flowy | minute: ooh should i have done that :D | 17:51 |
| flowy | i only reflashed. i don't know much about electronics | 17:51 |
| minute | well, it might cause the problem. a few days ago i think i wrote here about the difference of standalone vs internal version | 17:52 |
| flowy | yeah, you helped me clarify why i couldn't bring up my old keyboard externally. | 17:53 |
| flowy | ok so you're not sure that would do something weird but it might | 17:53 |
| flowy | it's another thing to try | 17:53 |
| flowy | i guess it would be nice to transplant it anyways because i have all the parts for a working standalone and i could give it to someone or whatever. | 17:54 |
| flowy | ok so debug path forwards is: try old firmware, desolder 3v3 ldo, then that patch | 17:56 |
| flowy | it will just take a while because my only way of reproducing the bug is by leaving it overnight haha | 17:56 |
| flowy | maybe i should see if some hours does it too. | 17:57 |
| - mjw (QUIT: Quit: Leaving) (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 17:59 | |
| + mjw (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 18:00 | |
| minute | flowy: in any case having the LDO there is not good. this way, you connect 2 slightly different 3v3 sources together. | 18:01 |
| flowy | sorry for my kind of naive electronics questions but, can the keyboard somehow retain some state on its own (disconnected) for some time when it's in sleep mode? like, there must be a capacitor or two right? | 18:01 |
| minute | flowy: not impossible | 18:01 |
| flowy | interesting... i wonder if this could explain the weird issues i was experiencing at first. | 18:02 |
| flowy | hope i didn't fry anything. maybe i'll just do that next then, desoldering the ldo. | 18:04 |
| NanoCodeBug | keyboard fw flashed! | 18:04 |
| + mark_ (~mark@gnu.wildebeest.org) | 19:11 | |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 19:14 | |
| - Major_Biscuit (QUIT: Ping timeout: 255 seconds) (~MajorBisc@c-001-009-008.client.tudelft.eduvpn.nl) | 19:21 | |
| kfx | it's amazing how consistently the internet's response to mnt computers boils down to one group posting "finally" and another group posting "this is not a macbook, I am outraged" | 19:42 |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mark@gnu.wildebeest.org))) (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 19:50 | |
| * mark_ -> mjw | 19:50 | |
| + wielaard (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 19:50 | |
| Boostisbetter | kfx, what do you mean? | 19:58 |
| minute | kfx: very normal | 20:01 |
| minute | i think this is called "polarizing" ^^ | 20:02 |
| flowy | unfortunately, i think open hardware might remain fringe and polarising unless very bad things continue happening in the world, making open hardware an eventual necessity (curbing e-waste, encouraging local technological innovation etc.) | 20:15 |
| flowy | Boostisbetter: a good example would be the comment thread at https://arstechnica.com/gadgets/2022/01/review-mnt-reform-laptop-has-fully-open-hardware-and-software-for-better-or-worse/, trigger warning lol | 20:25 |
| Boostisbetter | oh yeah I read that. | 20:25 |
| Boostisbetter | The problem is that the reviewer looked at from the wrong angle. | 20:26 |
| Boostisbetter | For one thing all that you get with the Reform is not something you get with other laptops. | 20:26 |
| Boostisbetter | I mean Framework, which is great, is still not the same. The Reform is built like a tank, and I'm scared to squish the Framework laptop too hard. | 20:26 |
| flowy | i think the comments would have been equally polarised regardless of the content of the review | 20:26 |
| Boostisbetter | But the other thing, is that while they keep talking about how slow it is, they never stop to admit how capable it is. How it does nearly everything that you could want from a computer. | 20:27 |
| flowy | looks like ars is also now featuring the pocket reform PR | 20:27 |
| Boostisbetter | The populace are conditioned to look at benchmarks and only go with what is the fastest. | 20:27 |
| Boostisbetter | Notwithstanding that most computer don't need the processing power they have. | 20:28 |
| Boostisbetter | the comments on that Pocket aren't nearly as bad or numerous yet | 20:29 |
| flowy | there's also just a lot of tech tribalism in general | 20:29 |
| Boostisbetter | I hope minute doesn't really pay these kinds of things too much attention. Built it and they will come. | 20:29 |
| Boostisbetter | I can't wait for the Pocket. | 20:29 |
| flowy | Boostisbetter: oh really ;) | 20:29 |
| Boostisbetter | Yeah, I am basically foaming at the bit for one. | 20:30 |
| flowy | haha | 20:31 |
| nocko | Introspective analysis isn't a socialized value in the global north. Most folks seem content to lash out a hundred times against something different before they'd give a minute of thought to their own actions. | 20:31 |
| nocko | "If we're all doing it it cannot be wrong, anyone who causes me the slightest doubt must be {dumb,crazy,&c}." | 20:32 |
| Boostisbetter | The thing is, the Reform doesn't need to appeal to everyone. | 20:33 |
| Boostisbetter | To expect it to is not really logical. | 20:33 |
| Boostisbetter | It is different, but that difference is for a reason. | 20:33 |
| Boostisbetter | I do think people using their Reform out in public will be the best thing for it. | 20:33 |
| Boostisbetter | People seeing it be a laptop that people use is all you need. | 20:34 |
| flowy | i need to fix mine first so that unscrewing the bezel and poking the reset button with a chopstick is not what people are seeing ;) | 20:38 |
| sknebel | heh, reminds me of the last months of my X220 thinkpad. the fan had a bearing failure and was really rattling | 20:38 |
| Boostisbetter | Yeah, I kind of baby my electronics. I have not had any issues like that. I have taken the Reform with me to work and on vacation several times. I am an introvert though, and the idea of me sitting in a cafe with it is not going to happen. Hahaha | 20:38 |
| sknebel | but on power-on the bios checks if the fan is spinning up, so I could just disconnect it | 20:39 |
| sknebel | *couldn't | 20:39 |
| Boostisbetter | sknebel, wrong chat i am afraid | 20:39 |
| sknebel | nope | 20:39 |
| Boostisbetter | oh sorry, I didn't see your x220 thing | 20:39 |
| Boostisbetter | my bad | 20:39 |
| sknebel | so I had the keyboard screws removed, and every time I got it out and turned it on somewhere in public it was "get it out, lift keyboard, remove stick from fan, turn on, let fan spin up, put finger on fan to stop it, put stick back in to block it, cloe keyboard, use" :D | 20:40 |
| Boostisbetter | My two favorite things about the Reform are the solid construction and the way the lid slots into the case which is better for the display and also act as a strain relief on the hinges when in transit. | 20:41 |
| sknebel | it was surprisingly usable in fanless mode. just couldn't watch 1080p videos anymore | 20:41 |
| Boostisbetter | I still have a X220T | 20:41 |
| Boostisbetter | I always hated how much bigger and bloated the case design on that was to the x220 itself. | 20:41 |
| sknebel | that was a bit weird, yeah | 20:42 |
| flowy | sknebel: quite analogous to starting up an old car heh | 20:44 |
| Boostisbetter | and while I know that it is tricky I do think once we can get steam working on here, that is going to be nice. | 21:03 |
| Boostisbetter | Not crucial thoughj | 21:03 |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 21:04 | |
| flowy | it would be cooler if valve started targeting arm. it's not even such an out there idea, considering the culture at valve and what they've already done with linux | 21:15 |
| flowy | with the popularity of using pis for retro gaming and console emulation i could imagine this being profitable at a niche level. i.e. mini DIY steam machines | 21:16 |
| minute | i wonder if steam will be rewritten at some point to be less of a rube goldberg machine | 21:17 |
| flowy | i read that valve has a somewhat anarchist management structure where workers are free to start their own projects or join the ones they deem most stimulating | 21:17 |
| minute | at least it was like that, yeah | 21:17 |
| vkoskiv | I'd really appreciate a nice suspend on reform, but I feel bad just shouting from the rooftops for *someone* to do it :D | 21:17 |
| vkoskiv | Usually I'd like to tackle the problem myself, but the whole linux ecosystem is way beyond my skillset, I think. | 21:18 |
| kfx | minute: I'm sure fixing steam is on the to-do list just after half-life 3 | 21:19 |
| Boostisbetter | I think the fact that Purism on their Librem 5 have yet to perfect suspend just speaks to the difficulty of it all. We are so close on the Reform though | 21:19 |
| cinap_lenrek | what kind of suspend? | 21:20 |
| minute | well, someone could read purisms discussion and patches about this | 21:20 |
| minute | and check if they apply to our situation | 21:20 |
| cinap_lenrek | like only refresh the dram and the cores being off? | 21:20 |
| minute | cinap_lenrek: suspend to ram | 21:20 |
| minute | cinap_lenrek: yeah | 21:20 |
| cinap_lenrek | ah | 21:21 |
| cinap_lenrek | isnt that handled by PSCI? | 21:21 |
| minute | it's kind of wonky at the moment | 21:21 |
| minute | well, psci ~and~ calling into wonky tf-a code | 21:21 |
| cinap_lenrek | i guess PSCI wont cover the peripherals | 21:21 |
| vkoskiv | I just want the same workflow as on my previous machine. No lid sensor, so I can just hit some keycombo to put it to sleep, and then wake it up again when I'm ready | 21:21 |
| vkoskiv | I want to (almost) never reboot | 21:21 |
| minute | vkoskiv: sure, i want that too | 21:21 |
| cinap_lenrek | minute: how does it wake up? | 21:22 |
| minute | it will be my one of my next topics after reviewing and merging the battery status driver | 21:22 |
| vkoskiv | Oooh battery status is already in the about to merge phase, exciting! | 21:22 |
| cinap_lenrek | haha, or is the M4 still up? | 21:22 |
| minute | cinap_lenrek: can be done by UART from lpc or by gpio from LPC | 21:22 |
| cinap_lenrek | i see, by an interrupt? | 21:23 |
| minute | the imx8m arm things can wake by peripheral interrupt yep | 21:23 |
| minute | since 5.18 the waking doesn't work at all anymore i think? at least not for me | 21:23 |
| minute | tried it earlier today | 21:23 |
| minute | cinap_lenrek: tf-a code will then get the dram into working state again | 21:24 |
| minute | and then it returns to the kernel | 21:24 |
| minute | we could probably do some printf debugging in tf-a... | 21:25 |
| minute | assuming the cores come back up | 21:25 |
| minute | there's also jtag but i never bothered to get a jlink and wire it up | 21:26 |
| NanoCodeBug | ah i have a j-link, i could help with debugging that | 21:47 |
| minute | cool! | 21:51 |
| NanoCodeBug | just pulled up the schematic, J19, i already have a cable for that jtag connector, noice | 21:54 |
| - bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 21:54 | |
| minute | NanoCodeBug: oh great. i hope that connector is actually populated, but should be IIRC | 21:59 |
| minute | hmmmm there was a snag IIRC i think 2 signals are shared with ethernet LEDs | 21:59 |
| minute | might need some resistor de-soldering on the SOM :( | 22:00 |
| NanoCodeBug | minute: yes, the header is populated, i should double check it matches the j-link pinout first though | 22:01 |
| - grubman (QUIT: *.net *.split) (~ufotofu@149.248.16.17) | 22:01 | |
| - jfred (QUIT: *.net *.split) (sid534649@libera/sponsor/jfred) | 22:01 | |
| - qbit (QUIT: *.net *.split) (~qbit@h.suah.dev) | 22:01 | |
| - hl (QUIT: *.net *.split) (~hl@user/hl) | 22:01 | |
| - skalk (QUIT: *.net *.split) (~skalk@vond.sysret.de) | 22:01 | |
| - ggoes (QUIT: *.net *.split) (~gregf@fsf/staff/ggoes) | 22:01 | |
| - scops (QUIT: *.net *.split) (~scopstchn@2001:470:69fc:105::8da) | 22:01 | |
| - tinybronca[m] (QUIT: *.net *.split) (~tinybronc@2001:470:69fc:105::2:1af6) | 22:01 | |
| - chomwitt (QUIT: *.net *.split) (~chomwitt@2a02:587:dc1c:800:5c71:ba1b:727c:3afe) | 22:01 | |
| - robin (QUIT: *.net *.split) (~robin@user/terpri) | 22:01 | |
| - sinedeo_ (QUIT: *.net *.split) (~sinedeo@2a01:4f9:c010:2837::1) | 22:01 | |
| - jfred-linode_ (QUIT: *.net *.split) (quassel@libera/sponsor/jfred) | 22:01 | |
| - conky (QUIT: *.net *.split) (5fb0fe5593@2604:bf00:561:2000::10b) | 22:01 | |
| - patrick_ (QUIT: *.net *.split) (224fa09e8b@2604:bf00:561:2000::217) | 22:01 | |
| - frank2 (QUIT: *.net *.split) (~frank2@juicy.frank2.net) | 22:01 | |
| - _E (QUIT: *.net *.split) (~e@bsd.moe) | 22:01 | |
| - buckket (QUIT: *.net *.split) (~buckket@pdp8.buckket.org) | 22:01 | |
| - sbates (QUIT: *.net *.split) (sid451432@id-451432.ilkley.irccloud.com) | 22:01 | |
| - flowy (QUIT: *.net *.split) (~flowy@2a01:4f8:c0c:1a8f::1) | 22:01 | |
| - natalie (QUIT: *.net *.split) (~natalie@user/natalie) | 22:01 | |
| - piroko (QUIT: *.net *.split) (~piroko@104.225.216.16) | 22:01 | |
| - C-Keen (QUIT: *.net *.split) (cckeen@pestilenz.org) | 22:01 | |
| NanoCodeBug | don't want to fry the expensive debugger, i've already done that before | 22:01 |
| NanoCodeBug | ah, yeah i see the eth part, i'll probably just cut the connections in a spare cable. | 22:02 |
| minute | i think bluerise hit this problem before | 22:03 |
| minute | NanoCodeBug: here are the SOM schematics https://dump.mntmn.com/SCH_Nit8MQ_SOM_2r8e_REV10_20190228.pdf | 22:04 |
| minute | NanoCodeBug: on sheet 8 you see RN1 and RN2... i believe only RN1 is populated and you need to move it over to RN2 to use jtag | 22:05 |
| NanoCodeBug | huh, fascinating. yeah RN2 is not populated | 22:13 |
| + aliosablack (~chomwitt@2a02:587:dc1c:800:5c71:ba1b:727c:3afe) | 22:14 | |
| + tinybronca[m] (~tinybronc@2001:470:69fc:105::2:1af6) | 22:14 | |
| + scops (~scopstchn@2001:470:69fc:105::8da) | 22:14 | |
| + chomwitt (~chomwitt@2a02:587:dc1c:800:5c71:ba1b:727c:3afe) | 22:14 | |
| + grubman (~ufotofu@149.248.16.17) | 22:14 | |
| + buckket (~buckket@pdp8.buckket.org) | 22:14 | |
| + robin (~robin@user/terpri) | 22:14 | |
| + jfred (sid534649@libera/sponsor/jfred) | 22:14 | |
| + qbit (~qbit@h.suah.dev) | 22:14 | |
| + sbates (sid451432@id-451432.ilkley.irccloud.com) | 22:14 | |
| + hl (~hl@user/hl) | 22:14 | |
| + flowy (~flowy@2a01:4f8:c0c:1a8f::1) | 22:14 | |
| + skalk (~skalk@vond.sysret.de) | 22:14 | |
| + sinedeo_ (~sinedeo@2a01:4f9:c010:2837::1) | 22:14 | |
| + jfred-linode_ (quassel@libera/sponsor/jfred) | 22:14 | |
| + conky (5fb0fe5593@2604:bf00:561:2000::10b) | 22:14 | |
| + patrick_ (224fa09e8b@2604:bf00:561:2000::217) | 22:14 | |
| + ggoes (~gregf@fsf/staff/ggoes) | 22:14 | |
| + frank2 (~frank2@juicy.frank2.net) | 22:14 | |
| + _E (~e@bsd.moe) | 22:14 | |
| + natalie (~natalie@user/natalie) | 22:14 | |
| + piroko (~piroko@104.225.216.16) | 22:14 | |
| + C-Keen (cckeen@pestilenz.org) | 22:14 | |
| - tinybronca[m] (QUIT: Max SendQ exceeded) (~tinybronc@2001:470:69fc:105::2:1af6) | 22:14 | |
| - chomwitt (QUIT: Max SendQ exceeded) (~chomwitt@2a02:587:dc1c:800:5c71:ba1b:727c:3afe) | 22:14 | |
| minute | NanoCodeBug: do you have soldering equipment+skills to move it? | 22:17 |
| + tinybronca[m] (~tinybronc@2001:470:69fc:105::2:1af6) | 22:19 | |
| NanoCodeBug | I've done 805 sized hot air and hot plate soldering before. the resistor net isn't too small. I think I can move it | 22:19 |
| minute | NanoCodeBug: awesome | 22:22 |
| minute | NanoCodeBug: on the motherboard, you will probably have to desolder 0-ohms R146, R147, R116. this is easy with a soldering iron+solder blob | 22:24 |
| minute | then, JTAG should theoretically work | 22:25 |
| NanoCodeBug | yup I see them, makes sense, that could interfere since they share lines | 22:25 |
| minute | yeah | 22:26 |
| NanoCodeBug | minute: hmm i'm not familiar with jtag_mod, it looks like it sets a security state that can enable/disable jtag entirely? | 22:35 |
| minute | here's a mirror of the reference manual https://dump.mntmn.com/IMX8MDQLQRMref2.pdf | 22:40 |
| minute | lets see | 22:40 |
| minute | > 4.12.1.5 JTAG topology | 22:41 |
| minute | > There is only one JTAG on the chip, and two JTAG modes are supported. Select via the | 22:41 |
| minute | JTAG_MOD pin. | 22:41 |
| minute | > Debug mode: JTAG_MOD == 0, DAP is the only TAP controller in the daisy chain. SJC will be attached to JTAG-AP of DAP. | 22:41 |
| minute | > Test mode: JTAG_MOD == 1, SJC is the only TAP controller in the daisy chain. 1149.1-compliant, and support 1149.6 AC coupled test | 22:42 |
| minute | i think you want debug mode? | 22:42 |
| NanoCodeBug | nice, that matches the snippit i saw in the nxp forums. pretty sure I want 0 | 22:42 |
| NanoCodeBug | in the schematic it looks like its bridged to ground by default via R48, trying to find if its populated on the som now | 22:44 |
| minute | ok cool | 22:44 |
| NanoCodeBug | nice it is, lol its right next to RN2, i'm here looking at the other side with a magnifying glass | 22:46 |
| minute | hehe | 22:50 |
| minute | so boundary devices probably used jtag in the lab | 22:50 |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 23:43 | |
| - Christoph_ (QUIT: Remote host closed the connection) (~Christoph@p4fe73e14.dip0.t-ipconnect.de) | 23:48 | |
| - vkoskiv (QUIT: Ping timeout: 272 seconds) (~vkoskiv@89-166-62-97.bb.dnainternet.fi) | 23:49 | |
| + vkoskiv (~vkoskiv@89-166-62-97.bb.dnainternet.fi) | 23:56 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!