| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 00:11 | |
| - mundane-1emantic (QUIT: Remote host closed the connection) (~user@192-184-203-82.fiber.dynamic.sonic.net) | 00:12 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 00:14 | |
| - murphnj (QUIT: Remote host closed the connection) (~murph@user/murphnj) | 00:21 | |
| + murphnj (~murph@user/murphnj) | 00:22 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 00:26 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 00:30 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 00:36 | |
| + nocko (~nock@user/nocko) | 00:49 | |
| + mundane-1emantic (~user@142.147.89.217) | 00:50 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 00:57 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 01:01 | |
| - murphnj (QUIT: Quit: Leaving) (~murph@user/murphnj) | 01:03 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 01:06 | |
| josch | minute: another idea for wasting less space on many platform-specific disk images is to ship a universal image which makes use of the same mechanism that the debian-installer sd-card makes use of, namely u-boot setting the ${fdtfile} variable: | 01:07 |
|---|---|---|
| josch | https://source.mnt.re/reform/reform-system-image/-/merge_requests/104 | 01:07 |
| josch | the image created by this MR will boot on all platforms that have u-boot on eMMC | 01:07 |
| josch | it does not come with its own u-boot as that would make the image device-specific | 01:07 |
| josch | so if you flash the image generated by the CI pipeline for this MR to an sd-card, it should boot fine on the pocket-reform | 01:08 |
| josch | it will likely not boot on a311d as that one usually does not have u-boot on emmc | 01:08 |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 01:09 | |
| josch | if you want to make that image boot on a311d you can write the image to an sd-card and then write u-boot to the beginning (for example by using "reform-flash-uboot sd") | 01:09 |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 01:14 | |
| * jackhill -> KM4MBG | 01:15 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 01:17 | |
| mundane-1emantic | i'm confused a bit by the gpu driver setup for the ls1028a -- it looks like mali_dp and etnaviv are both loaded, but only etnaviv uses gpu_sched. trying to modprobe them and enable dynamic printk's hasn't led anywhere yet, so I assume they're both supposed to be doing something, but I dont know what. | 01:22 |
| mundane-1emantic | and like, cards on the table, i'm sitting at 50% probability that the display is just that i mechanically or electrically messed up the SoM circuit board or the motherboard somehow during my tinkering, and the drivers dont matter at all | 01:23 |
| mundane-1emantic | still trying to figure out if either of those drivers _believe_ to be correctly driving a edp display, if possible | 01:24 |
| jn | i think malidp is a display controller, not a GPU | 01:24 |
| jn | that's probably why (from a very quick glance; i'm not a GPU driver expert) | 01:25 |
| ^alex | minute, the gitlab is 500ing when we're trying to make the merge request for the pico-sdk-ification | 01:25 |
| ^alex | relatedly,, we just switched to the pico-sdk's tinyusb on our pocket keyboard :3 | 01:26 |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 01:27 | |
| mundane-1emantic | jn: thanks, that makes sense. | 01:29 |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 01:31 | |
| ^alex | actually hmm | 01:34 |
| ^alex | this might be a bad idea b/c the submodule in the pico-sdk is old | 01:34 |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 01:38 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 01:40 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:45 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 01:54 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 01:58 | |
| - mtm (QUIT: Ping timeout: 245 seconds) (~textual@c-71-228-84-213.hsd1.fl.comcast.net) | 02:04 | |
| + mtm (~textual@c-71-228-84-213.hsd1.fl.comcast.net) | 02:06 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 02:09 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 02:12 | |
| - cobra (QUIT: Ping timeout: 276 seconds) (~cobra@user/Cobra) | 03:25 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 03:32 | |
| + cobra (~cobra@user/Cobra) | 03:33 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 03:34 | |
| ^alex | signs of progress: we're bricking our keyboard in ways that we can recover from just by pressing the reset button | 04:00 |
| ^alex | no `prog` dance | 04:01 |
| ^alex | however it's still hanging when trying to come out of sleep, gonna have to figure out some way to put debug-probe pins on | 04:01 |
| - cobra (QUIT: Quit: ZNC 1.8.2 - https://znc.in) (~cobra@user/Cobra) | 04:01 | |
| + cobra (~cobra@user/Cobra) | 04:06 | |
| - cobra (QUIT: Remote host closed the connection) (~cobra@user/Cobra) | 04:07 | |
| + cobra (~cobra@user/Cobra) | 04:09 | |
| - bkeys (QUIT: Remote host closed the connection) (~Thunderbi@45.134.140.153) | 04:42 | |
| * KM4MBG -> jackhill | 04:54 | |
| mundane-1emantic | okay so i swapped back to the nitrogen som and the display works fine out of the box. i think i might have somehow damaged my ls1028a :/ maybe worth trying mixing/matching the adapters and ribbon cables to make sure none of those is the cause | 05:01 |
| mundane-1emantic | i'll probably just setup the imx8 for defcon and then get back to debugging when home from vegas i think | 05:12 |
| mundane-1emantic | sad that i messed it up, but excited that having access to the schematics might mean i have a chance of diagnosing and repairing whatever it is i messed up | 05:13 |
| - _justin_kelly9 (QUIT: Quit: Ping timeout (120 seconds)) (~justinkel@user/justin-kelly/x-6011154) | 05:48 | |
| + _justin_kelly3 (~justinkel@user/justin-kelly/x-6011154) | 05:48 | |
| - _justin_kelly (QUIT: Quit: Ping timeout (120 seconds)) (~justinkel@user/justin-kelly/x-6011154) | 05:50 | |
| * _justin_kelly3 -> _justin_kelly | 05:50 | |
| + _justin_kelly8 (~justinkel@user/justin-kelly/x-6011154) | 05:50 | |
| - skipwich (QUIT: Read error: Connection reset by peer) (~skipwich@user/skipwich) | 08:17 | |
| + skipwich (~skipwich@user/skipwich) | 08:19 | |
| minute | andy piper's review of pocket reform linked on hn https://news.ycombinator.com/item?id=41176817 | 09:59 |
| - khm (QUIT: Ping timeout: 245 seconds) (~kfx@wopr.sciops.net) | 10:25 | |
| - _justin_kelly (QUIT: Quit: The Lounge - https://thelounge.chat) (~justinkel@user/justin-kelly/x-6011154) | 10:31 | |
| * _justin_kelly8 -> _justin_kelly | 10:31 | |
| - _justin_kelly (QUIT: Quit: The Lounge - https://thelounge.chat) (~justinkel@user/justin-kelly/x-6011154) | 10:32 | |
| josch | somehow i was preparing myself for tons of toxic comments but then was pleasantly surprised by the positivity :) | 10:33 |
| + _justin_kelly (~justinkel@user/justin-kelly/x-6011154) | 10:38 | |
| kensanata | On my pocket reform, I ran reform-check, which told me that my uboot was not the latest and told me to use reform-flash-uboot if I wanted to fix this. That script seems to have two typos in it, however? I'm pretty clueless, though. In two instances it uses $MMC_BOOT instead of $EMMC_BOOT. | 11:24 |
| kensanata | The "/usr/share/reform-tools/machines/MNT Pocket Reform with i.MX8MP Module.conf" file only sets EMMC_BOOT, as far as I can tell. | 11:25 |
| kensanata | The end result when running it without arguments is "/usr/sbin/reform-flash-uboot: 59: MMC_BOOT: parameter not set" | 11:25 |
| josch | kensanata: you found this problem: https://community.mnt.re/t/updating-uboot/2348 | 11:31 |
| josch | kensanata: this will be fixed with the next release of reform-tools | 11:31 |
| josch | since you are now the second person who run into this, i guess i should do a release today | 11:31 |
| josch | kensanata: the important bit is: do *not* flash u-boot to emmc on the pocket reform | 11:32 |
| josch | there is a bug in the u-boot offset which will make your pocket unbootable if you fix the other bug in reform-flash-uboot | 11:32 |
| kensanata | Argh! | 11:32 |
| kensanata | Thanks for the warning. | 11:32 |
| kensanata | Ah, so this is what andi piper means when he writes: "I also managed to temporarily brick the machine by corrupting the uboot in flash…" | 11:38 |
| josch | yes, andi piper fixed the typo in reform-flash-uboot and then ran "reform-flash-uboot emmc" which then wrote u-boot to the wrong offset on emmc | 11:45 |
| josch | kensanata: okay, i just merged the reform-tools 1.48 branch into main | 11:53 |
| josch | unfortunately, we currently cannot build it and have it published in the repo because linux 6.10 broke the creation of extra meta packages | 11:54 |
| josch | hrm... maybe we can trick the CI by just re-running the build_custom job of the last successful run... | 11:58 |
| kensanata | Actually, for my case it might be more important to just change reform-check so that it doesn't suggest flashing uboot? | 12:16 |
| kensanata | In the thread with Andy you said: "updating u-boot should only be done by people who really, really want to and know what they are doing" – and that's not me. :D | 12:16 |
| gsora | lots of positive feedback about mnt products on the pocket reform hn thread :o | 12:23 |
| josch | kensanata: this suggestion not showing up in reform-check is what will happen with reform-tools 1.48 | 12:25 |
| josch | kensanata: in '/usr/share/reform-tools/machines/MNT Pocket Reform with i.MX8MP Module.conf' EMMC_BOOT will be set to false | 12:26 |
| josch | and doing so will stop reform-flash-uboot to allow flashing on the pocket and it will stop the suggestion from showing up in reform-check | 12:26 |
| kensanata | Nice. | 12:55 |
| josch | it worked! | 13:03 |
| josch | kensanata: try "apt update && apt upgrade" | 13:03 |
| josch | that should install you reform-tools 1.48 | 13:03 |
| josch | if you then run reform-check, the suggestion should be gun | 13:03 |
| josch | if you try running "reform-flash-uboot emmc" it should throw an error | 13:03 |
| + larme (22544863b6@user/larme) | 13:19 | |
| - mtm (QUIT: Ping timeout: 276 seconds) (~textual@c-71-228-84-213.hsd1.fl.comcast.net) | 14:03 | |
| + mtm (~textual@c-71-228-84-213.hsd1.fl.comcast.net) | 14:06 | |
| + bkeys (~Thunderbi@45.134.140.153) | 14:32 | |
| - laumann (QUIT: Remote host closed the connection) (~quassel@user/laumann) | 14:52 | |
| + laumann_ (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 14:53 | |
| minute | gsora: yes, i was pleasantly surprised that things have developed in the last years concerning our reputation and grown userbase | 15:08 |
| gsora | great to see :D | 15:21 |
| josch | minute: do you want to push more things onto https://source.mnt.re/reform/reform-debian-packages/-/merge_requests/49 or shall i take care to rebase and merge it into the linux6.10 branch? | 15:37 |
| minute | josch: haha i always get error 500 when looking at that. i don't have anything else to push, if you could take care of it i'd be very thankful! | 15:42 |
| josch | great, will do :) | 15:44 |
| josch | usually pressing f5 a couple of times will make it work at some point | 15:44 |
| josch | i guess the same happened when you tried reading my work on "generic-no-uboot" system images? :) | 15:44 |
| + xuochi (~xuochi@138.sub-75-236-210.myvzw.com) | 15:55 | |
| - ex-parrot (QUIT: Quit: _b) (~fincham@user/ex-parrot) | 16:46 | |
| + ex-parrot (~fincham@user/ex-parrot) | 16:46 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:53 | |
| josch | vagrantc: what is the policy for accepting changes to db/all.db of flash-kernel? We have accumulated this diff here: https://paste.debian.net/hidden/7a4081c2/ Must a device tree be in the upstream kernel first? | 18:02 |
| vagrantc | josch: historically the results are mixed ... depends on who is reviewing :/ | 18:05 |
| vagrantc | josch: i would say there is a preference for upstream, but not absolutely required ... there are many platforms that predate anything being upstream | 18:06 |
| josch | vagrantc: and am I reading https://sources.debian.org/src/flash-kernel/3.108/functions/#L660 correctly that a dtb is always installed with a symlink from the basename to the basename prefixed with the dirname? Is that done for legacy reasons? I see ${fdtfile} in u-boot being prefixed with the dirname on all our u-boot except for the boundary vendor u-boot for imx8mq | 18:07 |
| vagrantc | josch: depends on the platform, but typically anything new includes the dirname ... older 32-bit arm platforms may not include the dirname. it's so inconsistant the easiest thing was to always include both | 18:11 |
| vagrantc | josch: although, that is not apparently what you asked ... :) | 18:11 |
| josch | vagrantc: okay, thank you. As I recently hinted, I'm currently replicating a bit of flash-kernel to have a device-neutral system image, so i was reading a bit of flash-kernel code. :) | 18:12 |
| josch | vagrantc: do you think i should just update https://salsa.debian.org/installer-team/flash-kernel/-/merge_requests/48 with the current diff and adjust the MR title accordingly? | 18:12 |
| vagrantc | josch: my apologies and condolences! i try to read as little ofd that as i can get away with :) | 18:12 |
| josch | fear not -- you have no idea what other code i am staring at for my $dayjob :D | 18:13 |
| vagrantc | and glad of it. | 18:13 |
| minute | josch: sorry, lots of administrative/coordination work here atm | 18:14 |
| josch | minute: take your time -- nothing urgent at all :) | 18:19 |
| josch | minute: in other news i found a workaround for the 6.10 regression so we can merge that soon after i built and tested some system images with it | 18:20 |
| minute | josch: niiiice | 18:34 |
| + andypiper (~andypiper@89.36.117.58) | 19:05 | |
| minute | hmm https://lore.kernel.org/all/20240122235814.831343487@linuxfoundation.org/ | 19:10 |
| minute | but i guess this is in our kernel by now... | 19:10 |
| minute | interesting https://lore.kernel.org/linux-wireless/20240724050501.7550-1-pkshih@realtek.com/ | 19:17 |
| josch | minute: yes it is as 53ee0b3b99edc since v6.8-rc7 | 19:18 |
| minute | currently checking if there's any hope to fix 8822cs | 19:18 |
| josch | that's the realtek bluetooth? | 19:18 |
| minute | wifi | 19:19 |
| minute | also bluetooth, but that doesn't have known issues | 19:19 |
| minute | (on a311d) | 19:19 |
| josch | oh the wifi connection problems | 19:19 |
| josch | connection drops rather | 19:19 |
| minute | yeah. for me the connection doesn't really drop, it just stalls | 19:20 |
| josch | right | 19:20 |
| josch | yeah, that would be very nifty to have fixed | 19:21 |
| vkoskiv | Yeah, Linux thinks there is still a connection, but no packets go through | 19:21 |
| minute | exactly | 19:21 |
| minute | i see that the usb siblings of this chip have a problem with tx aggregation https://lore.kernel.org/linux-wireless/cb46ea35-7e59-4742-9c1f-01ceeaad36fb@gmail.com/ | 19:21 |
| josch | right, i'm seeing the same even after the powersafe workaround (just less often) | 19:21 |
| minute | josch: same | 19:21 |
| josch | the randomness makes this super hard to reproduce... | 19:22 |
| minute | also, the usb version rtw8821cu has issues with power https://lore.kernel.org/linux-wireless/20240527173454.459264-3-marcin.slusarz@gmail.com/ | 19:22 |
| josch | minute: i do not envy you somehow having to put priority on one issue over the other :D | 19:22 |
| minute | josch: yeah... | 19:22 |
| - andypiper (QUIT: Ping timeout: 252 seconds) (~andypiper@89.36.117.58) | 19:49 | |
| + gustav28 (~gustav@c-2336524e.019-141-67626730.bbcust.telenor.se) | 20:02 | |
| josch | for future-me searching the IRC logs, here is u-boot printenv output of a311d: https://mister-muffin.de/p/KdIj.txt | 20:12 |
| ch | could stick all printenv collections into git somewhere | 20:16 |
| josch | i was thinking about that, yes | 20:22 |
| josch | every once in a while i want to fix something and i find myself wondering: is what i see there what i should be seeing or is this odd? | 20:22 |
| josch | would be nice to be able to compare to a known good output | 20:22 |
| mhoye | Unrelated to everything, but here's a console font https://github.com/mhoye/syncraati | 20:29 |
| + andypiper (~andypiper@45.148.12.75) | 20:30 | |
| dozens | whoa | 20:30 |
| mhoye | If you want something with a certain stylistic furtureness to it, I mean. | 20:30 |
| dozens | this has been my terminal font for several years now | 20:33 |
| dozens | https://dtinth.github.io/comic-mono-font/ | 20:33 |
| josch | wow | 20:38 |
| ^alex | andypiper! ahoy :) we were trying to comment on your page but the comment engine just displays a spinner when we post. but. the Pocket doesn't take 18650s, it takes a particular kind of flat cell | 20:39 |
| mundane-1emantic | i went through the new setup after swapping in the nitrogen8 som back in, and wow this is a very cool streamlined experience. graphical environment without any configuration, and scripts to setup encrypted NVME. it worked great for me, thanks to all the folks that have been improving it. | 20:46 |
| mundane-1emantic | also if theres gonna be any kind of reform meetup or jam at DEFCON, i'd love to chat w/ folks about how they've been hacking on theirs | 20:46 |
| josch | mundane-1emantic: sorry to hear that you have issues with your ls1028a though :( | 20:48 |
| minute | mundane-1emantic: glad to hear about the imx8mq and sorry about ls1028a. not sure what is going wrong there. you could send the module & adapter(s) & cable to us so i can take a look. | 20:51 |
| minute | i ordered a few of these https://asiarf.com/wp-content/uploads/2023/08/M27612-BU3_V1-1P-1.pdf | 20:52 |
| minute | one of the rare b-key wifi cards, to test in pocket's wwan slot | 20:53 |
| andypiper | ^alex: oh! Thank you (and sorry about the commenting thing...) - I shall fix! | 20:57 |
| andypiper | (also I am glad to see generally positive things being posted on HN etc about MNT) | 20:58 |
| ^alex | also hope you're doing well since the last time we were in the same chat | 20:58 |
| mundane-1emantic | minute: thanks for the idea! I'm going to try harder to figure it out after i'm back from vegas. I've definitely got more things to test out first. | 21:20 |
| Twodisbetter | minute: Do you have an idea of where most sales are going? I mean country wise? | 21:21 |
| Twodisbetter | I would imagine Germany, but I'm curious. | 21:21 |
| * andypiper -> andypiper[afk] | 21:50 | |
| * andypiper[afk] -> andypiper | 21:59 | |
| xuochi | Twodisbetter: I would think that the USA is probably the largest purchaser of MNT hardware. | 22:11 |
| xuochi | Twodisbetter: this is based upon the price of the hardware and the tendency of people in the USA to just build up insane amounts of debt, while people in Europe tend to be a bit more fiscally reserved. | 22:12 |
| Twodisbetter | xuichi: I think that is certianly part of it. I would also say that Americans in general have more spending cash than Europeans. Compare the rate software engineers are paid in Europe compared to the US, to see what I mean. | 22:13 |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-2336524e.019-141-67626730.bbcust.telenor.se) | 22:15 | |
| + gustav28 (~gustav@c-2336524e.019-141-67626730.bbcust.telenor.se) | 22:28 | |
| xuochi | Twodisbetter: sure. slightly more. there are a ton of things that erode US spending power tho. For example, most Americans must own a car, and those cars are huge due to size/weight to fuel effiency regulations. | 22:32 |
| xuochi | this then costs the average American anywhere from $300/mo to 1100/mo. | 22:33 |
| xuochi | plus the mandatory insurance, maintenance, etc | 22:33 |
| Twodisbetter | xuichi: having lived in Germany for going on 14 years now, I feel I have a good grasp on the difference bettwen Germany and the US. | 22:33 |
| Twodisbetter | depending on your job, your employer is the one handling your health insurance. | 22:33 |
| Twodisbetter | Couple the fact that one generally makes more money there, the other dynamic is that the US economy is geared for convience and spending. | 22:35 |
| Twodisbetter | Americans are much more willing to buy things. | 22:35 |
| Twodisbetter | and try things. | 22:35 |
| Twodisbetter | Anyway kind of off topic for this channel so I'll desist. | 22:36 |
| xuochi | Twodisbetter: Du hast teilweise recht, aber wir sind vom Thema abgekommen, ja. | 22:38 |
| - mundane-1emantic (QUIT: Quit: Lost terminal) (~user@142.147.89.217) | 22:48 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-150-69.tukw.qwest.net) | 23:28 | |
| + colinsane (~colinunin@97-113-150-69.tukw.qwest.net) | 23:32 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!