| - GNUmoon (QUIT: Ping timeout: 240 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 00:06 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 00:07 | |
| - yewscion (QUIT: Remote host closed the connection) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 00:50 | |
| + yewscion (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 00:51 | |
| - mjw (QUIT: Ping timeout: 245 seconds) (~mjw@gnu.wildebeest.org) | 01:12 | |
| - mtm (QUIT: Ping timeout: 272 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:02 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 02:49 | |
| - nsc (QUIT: Ping timeout: 245 seconds) (~nicolas@131-98-142-46.pool.kielnet.net) | 04:00 | |
| + nsc (~nicolas@94-98-142-46.pool.kielnet.net) | 04:02 | |
| abortretryfail | josch - i can try to debug it now. Sorry I didn't see your message sooner. | 04:06 |
|---|---|---|
| abortretryfail | Hm, maybe I would have better luck in the morning. I am UTC-4 time zone. | 04:07 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:09 | |
| josch | abortretryfail: do you have a usb uart adapter to connect directly to u-boot via serial? | 04:49 |
| josch | i'm at utc+2 but with irregular sleep schedule thanks to baby ;) | 04:49 |
| - klardotsh (QUIT: Ping timeout: 272 seconds) (~klardotsh@98.97.114.50) | 04:52 | |
| abortretryfail | hmm, i don't think I do | 04:53 |
| josch | abortretryfail: is your u-boot new enough that you have display support? | 04:53 |
| abortretryfail | yeah, i had display support and i could work at the prompt. | 04:54 |
| abortretryfail | I already "fixed" it by copying the old (non-hdmi) .dtb file from the previous kernel into the place where the new one was expecting it in /boot and it boots | 04:54 |
| abortretryfail | but i suspect that's not the *right way* :) | 04:54 |
| josch | okay, so you had it configured for hdmi before it broke? | 04:55 |
| josch | via reform-display-config? | 04:55 |
| abortretryfail | honestly, I don't recall | 04:55 |
| abortretryfail | It might've been | 04:55 |
| josch | what kernel are you running now? | 04:56 |
| abortretryfail | just a sec, lemme grab it | 05:07 |
| abortretryfail | 6.3.0-2-reform2-arm64 #1 SMP Debian 6.3.11-1+reform20230618T175433Z1 | 05:08 |
| abortretryfail | hopefully I didn't typo that | 05:08 |
| josch | you are not online with you reform? | 05:09 |
| abortretryfail | I can be if necessary | 05:09 |
| abortretryfail | I'm using another PC at the moment for IRC | 05:09 |
| josch | that would make copypasting easier but it's not required | 05:09 |
| josch | what does this output: | 05:10 |
| josch | readlink /boot/dtb-6.3.0-2-reform2-arm64 | 05:10 |
| abortretryfail | dtbs/6.3.0-2-reform2-arm64/freescale/imx8mq-mnt-reform2-hdmi.dtb | 05:11 |
| josch | good! | 05:11 |
| josch | your boot.scr should use the /boot/dtb-6.3.0-2-reform2-arm64 symlink to figure out the dtb | 05:12 |
| josch | as you can see it points to the hdmi dtb | 05:12 |
| abortretryfail | yeah, i wonder if it got fixed the right way in a later update. It was pointing to the not-hdmi one which was missing | 05:13 |
| abortretryfail | reform-check still says E: your currently loaded dtb is not the one referenced by flash-kernel | 05:13 |
| abortretryfail | maybe the hdmi one is the that was missing and i copied from the older kernel and I just forgot | 05:13 |
| abortretryfail | let me check the dates on the files. | 05:14 |
| josch | hm... it's hard to get to the bottom of this problem if you are not sure anymore what exactly happened :/ | 05:14 |
| abortretryfail | yep, i completely agree | 05:14 |
| abortretryfail | i was at my mom's house just trying to get my laptop working. Probably would have been prudent to take an image of the SD card first. | 05:15 |
| abortretryfail | in that folder I have imx8mq-mnt-reform2.dtb from jul 10 | 05:18 |
| abortretryfail | then i have imx8mq-mnt-reform2-hdmi.dtb and imx8mq-mnt-reform2-hdmi.dtb.bak , both identical from jul 11 | 05:19 |
| josch | abortretryfail: you are using your reform with stable or unstable? | 05:21 |
| abortretryfail | unstable, it seems | 05:21 |
| abortretryfail | trixie / sid | 05:21 |
| josch | another idea for next time is to have u-boot-menu installed which will give you a boot menu | 05:22 |
| josch | this is of course only useful if display and keyboard work in u-boot (which was not the case a few months ago and that's why u-boot-menu is not enabled unconditionally) | 05:22 |
| josch | if you are using unstable, an advantage of u-boot-menu is that if you install a new kernel and it fails to boot, you can reboot and choose your old kernel | 05:23 |
| abortretryfail | well, when it broke, I was able to use my keyboard and display in u-boot and poke around at the prompt, but the 'help' command wasn't particularly useful without a pager | 05:23 |
| abortretryfail | i'll go grab that and see if the menu works, i would have liked that functionality. :) | 05:24 |
| abortretryfail | I think I need to show the LPC some love too. the battery % always says ???% | 05:24 |
| josch | abortretryfail: even with an old lpc version, the ???% goes away at the point where it was fully charged once | 05:26 |
| josch | you only need a new lpc version if you want to see the battery status in your OS | 05:26 |
| abortretryfail | Yeah, it doesn't on this one. | 05:26 |
| abortretryfail | like, it does once it goes fully charged, but if i put the machine away and don't use it for a while when i turn it on next it will be ???% again | 05:27 |
| josch | oh that shouldn't happen | 05:27 |
| josch | the battery status usually gets reset to ???% once the battery board gets disconnected | 05:27 |
| abortretryfail | System status gives me: "MREF2LPCR420230208normaL,1668,430,175 | 05:27 |
| abortretryfail | it also sometimes takes a few smacks of the O key to get it to wake up so i can hit 1 and power on | 05:28 |
| josch | there is an easy solution to that one: to save power, the lpc is not "on" continously, so if the reform has been off for a while, you need to hold the key for about a second | 05:29 |
| abortretryfail | Yeah, I figured it was something like that. Not a big deal | 05:30 |
| josch | because it will only check the key status every once in a while to prevent draining the battery once it's off | 05:30 |
| abortretryfail | I'm going to head to bed. Will tinker with this thing a bit more tomorrow. | 05:33 |
| josch | okay, good night! :) | 05:34 |
| + klardotsh (~klardotsh@98.97.114.50) | 05:43 | |
| + jacobk (~quassel@47-186-110-219.dlls.tx.frontiernet.net) | 06:00 | |
| - yewscion (QUIT: Remote host closed the connection) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 06:52 | |
| + yewscion (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 06:53 | |
| + inhji (~inhji@2a01:4f8:1c0c:42e4::1) | 08:25 | |
| - klardotsh (QUIT: Ping timeout: 272 seconds) (~klardotsh@98.97.114.50) | 09:04 | |
| + mjw (~mjw@gnu.wildebeest.org) | 10:42 | |
| * rah_ -> ah | 11:41 | |
| * ah -> rah | 11:41 | |
| - mjw (QUIT: Remote host closed the connection) (~mjw@gnu.wildebeest.org) | 12:08 | |
| * mark_ -> mjw | 12:42 | |
| - mlarkin (QUIT: Ping timeout: 245 seconds) (~mlarkin@047-036-115-056.res.spectrum.com) | 12:45 | |
| + mlarkin (~mlarkin@047-036-115-056.res.spectrum.com) | 12:47 | |
| sevan | josch: picking up from where I left off on the switch to the stable Debian release. I notice in the article you wrote that it states " If you haven’t upgraded your system since the release of Debian bookworm, then you can easily turn your Debian unstable into a Debian stable (codename bookworm) system", I'm curious what adventures lie ahead (I've been keeping the system up to date fairly regularly) :) | 13:38 |
| josch | sevan: you mean the adventures with unstable? | 13:39 |
| sevan | adventures in taking an up to date (untill last week) back to stable | 13:40 |
| josch | oh well | 13:40 |
| sevan | :) | 13:40 |
| josch | it will probably work | 13:40 |
| josch | but you'll end up with a weird mix of packages in their stable versions and unstable | 13:41 |
| josch | i would not recommend downgrading unless you know what you are doing | 13:41 |
| josch | one problem this might cause is, if there are security updates to the version in stable | 13:41 |
| josch | which you will never get because you installed the (newer) unstable version | 13:41 |
| sevan | ah, understood | 13:41 |
| sevan | I guess I should get a new SD card and install the stable release on there | 13:42 |
| sevan | (not a problem) | 13:42 |
| josch | there is not yet a full stable system image | 13:42 |
| josch | because of https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1039472 | 13:43 |
| josch | but the rescue image works | 13:43 |
| sevan | understood, so fetch a bookworm rescue image & use that install the system on a fresh SD card? | 13:44 |
| josch | yes, you can do that | 13:44 |
| josch | you would be the first to test that image :) | 13:45 |
| josch | so i'd be eager for feedback | 13:45 |
| sevan | shit! I mean, wonderful! ;) | 13:45 |
| josch | i didn't announce reform.d.n yet because i'm waiting for a new upload of the debian-keyring package | 13:45 |
| sevan | ack | 13:46 |
| sevan | ok, I need to get a new SD card. I'm not familiar with the image/release process for debian but I guess, fetch stock rescue image, add reform u-boot to it, boot? | 13:47 |
| josch | the rescue image contains u-boot | 13:47 |
| josch | let me write a few words on the procedure | 13:48 |
| sevan | <3 thanks. | 13:48 |
| - mtm (QUIT: Ping timeout: 240 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:03 | |
| + oomono (uid328183@id-328183.tinside.irccloud.com) | 14:04 | |
| josch | sevan: https://reform.debian.net/images#how-to-flash-to-sd-card | 14:14 |
| josch | does that help? | 14:14 |
| sevan | *clicks link* | 14:17 |
| sevan | heh, "The disk image contains the bootloader (U-Boot) as well as a partition table, so it has to be written to the SD-card directly instead of being copied to a mounted filesystem or written onto an existing partition" rejigged my thinking. I keep thinking about it in terms of x86 hardware and going through an installation process, whereas you are just booting a prepped image (there's no install process | 14:20 |
| sevan | necessarily). | 14:20 |
| + ming_ (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 14:22 | |
| - yewscion (QUIT: Read error: Connection reset by peer) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 14:22 | |
| sevan | josch: worth compressing the images? | 14:24 |
| josch | ah indeed that could be done | 14:24 |
| sevan | the full system image link works on the page but I'm downloading the rescue image | 14:35 |
| sevan | josch: is there a CDN facility for debian to mirror the images? | 14:42 |
| + mark_ (~mjw@gnu.wildebeest.org) | 14:53 | |
| josch | sevan: these are not official Debian images so i cannot make use of the Debian cdn | 14:59 |
| sevan | ah, ok | 15:00 |
| josch | one reason that the images will remain unofficial is, that the license of the non-free binary blobs does not permit redistribution | 15:09 |
| josch | so what i'm doing with reform.d.n i'm doing on my own legal risk | 15:09 |
| sevan | ack :) | 15:10 |
| sevan | writing reform2-imx8mq-rescue-system.img to an SD card | 15:21 |
| - qbit (QUIT: Quit: WeeChat 3.8) (~qbit@2602:ff16:3:0:1:3a0:0:1) | 15:25 | |
| sevan | booting... | 15:29 |
| sevan | josch: worked, I'm at the root prompt | 15:30 |
| + qbit (~qbit@h.suah.dev) | 15:31 | |
| - mark_ (QUIT: Ping timeout: 272 seconds) (~mjw@gnu.wildebeest.org) | 15:32 | |
| sevan | josch: should the image be set to resize itself automatically on boot (just ran apt update and it failed as fs is full - it's on a 32GB SD card) | 15:36 |
| josch | sevan: yes it should. There is a patch doing that here: https://source.mnt.re/reform/reform-system-image/-/merge_requests/43 | 15:42 |
| sevan | thanks for the pointer | 15:43 |
| sevan | looks like there's some regression with reform-windowmaker (I did get it working on unstable, need to find the commit that fixed it then) | 15:44 |
| sevan | ah, reform-windowmaker shell function is there, but package is not installed :) | 15:48 |
| sevan | assumed it was because sway is. | 15:48 |
| minute | i believe we will sunset that stuff soon (reform-windowmaker) | 15:58 |
| sevan | ack | 16:00 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:09 | |
| + GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 16:09 | |
| - GNUmoon (QUIT: Ping timeout: 240 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 16:11 | |
| sevan | josch: with latest version of reform-check it warns that the default /etc/flash-kernel/preboot.d/00reform2_preboot file exists, overriding the one in /usr/share, so delete it. You delete it, and then /etc/flash-kernel/preboot.d/00reform2_preboot gets listed in "the following files differ from how they are shipped by reform-tools" we should just stop shipping the default file in /etc/flash-kernel? | 16:14 |
| sevan | https://source.mnt.re/reform/reform-tools/-/blob/main/bin/reform-check#L160 | 16:17 |
| - kensanata (QUIT: Quit: OK) (~alex@user/kensanata) | 16:18 | |
| + kensanata (~alex@user/kensanata) | 16:20 | |
| minute | plomlompom: hast du gerade internet? ich habe das shop-sql-script gefixt, aber ich weiß nicht wie daraus eine mengenmeldung gemacht wird. hast du dein tooling dafür online? | 16:20 |
| minute | woops, wrong channel | 16:20 |
| minute | ACTION dealing with all the coworkers enjoying their vacations | 16:22 |
| sevan | :) | 16:23 |
| - jacobk (QUIT: Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) (~quassel@47-186-110-219.dlls.tx.frontiernet.net) | 16:46 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 16:50 | |
| - mark_ (QUIT: Ping timeout: 250 seconds) (~mjw@gnu.wildebeest.org) | 16:54 | |
| josch | sevan: if you delete /etc/flash-kernel/preboot.d/00reform2_preboot, then how can it complain about differing? | 17:21 |
| josch | sevan: reform-tools doesn't ship /etc/flash-kernel/preboot.d/00reform2_preboot anymore | 17:22 |
| josch | sevan: can you rebase https://source.mnt.re/reform/reform-tools/-/merge_requests/43 on the latest main branch? | 17:23 |
| sevan | josch: there's a separate section about files shipped that differ (let me point to the source, one moment) | 17:26 |
| sevan | josch: re-basing, that's interesting, I used the web IDE to edit the file on gitlab itself to raise that change | 17:27 |
| sevan | s/re-basing/re-rebasing | 17:27 |
| sevan | ah, I must have an out of date clone of the tree | 17:28 |
| sevan | josch: if the file exists, the block at L160 executes, if you delete it, the block at L201 come into effect https://source.mnt.re/reform/reform-tools/-/blob/main/bin/reform-check#L201 | 17:30 |
| sevan | hmm, did the fingerprint on source.mnt.re change recently? | 17:39 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:48 | |
| + murphnj (~murph@user/murphnj) | 17:49 | |
| josch | sevan: yes, the server was moved and the ssh private key changed | 17:51 |
| sevan | understood, thanks. | 17:52 |
| sevan | josch: raised a new merge request | 18:03 |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:15 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:17 | |
| * mark_ -> mjw | 18:17 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:18 | |
| - GNUmoon2 (QUIT: Ping timeout: 240 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 18:37 | |
| - ming_ (QUIT: Remote host closed the connection) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 18:37 | |
| + ming_ (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 18:37 | |
| + GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 18:39 | |
| josch | sevan: are you a native english speaker? (i'm not) but https://stackoverflow.com/questions/8651562/overwrite-or-override seems to suggest that overwriting is correct in this context | 19:06 |
| abortretryfail | josch: thanks for the help yesterday. I got the u-boot-menu installed and working this morning. | 19:34 |
| - ming_ (QUIT: Ping timeout: 272 seconds) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 19:35 | |
| + yewscion (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 19:35 | |
| josch | abortretryfail: now the question is whether your setup still suffers from the problem you experienced before or not... :/ | 19:36 |
| abortretryfail | yeah, not knowing what caused it in the first place sucks. | 19:37 |
| josch | oh actually we can try out soon! | 19:38 |
| josch | 6.1.38-1 was uploaded to bookworm updates | 19:38 |
| josch | let me re-run the pipeline... | 19:38 |
| josch | abortretryfail: you have 6.1.37, right? | 19:39 |
| abortretryfail | 6.3.11 running trixie/sid | 19:42 |
| abortretryfail | I can go check for updates and see if there's a new kernel | 19:43 |
| + klardotsh (~klardotsh@98.97.115.59) | 19:43 | |
| abortretryfail | wonder if this is just blood on the edge | 19:43 |
| abortretryfail | 6.4.1 is available, let's update and see if it breaks again :) | 19:45 |
| josch | abortretryfail: ah okay you are running unstable | 19:56 |
| josch | abortretryfail: yes, since a kernel upgrade broke it for you last time, maybe this happens again with this upgrade | 19:56 |
| abortretryfail | I think it shipped with unstable. I never tried to change it. | 19:56 |
| josch | abortretryfail: yes, the default is unstable | 19:58 |
| abortretryfail | 6.4.1 booted fine and reform-check no longer complains about the .dtb | 20:05 |
| josch | huh | 20:05 |
| josch | minute: but you still experience the problem unless you perform some manual intervention, right? | 20:05 |
| josch | abortretryfail: and /proc/device-tree/model shows that you have the HDMI dtb loaded as you expect, right? | 20:06 |
| abortretryfail | MNT Reform 2 HDMI | 20:06 |
| josch | cool! | 20:06 |
| abortretryfail | What's the purpose of having 2 different dtbs anyway? is there some benefit to be had by disabling HDMI? | 20:09 |
| minute | josch: not at the moment as i have migrated to a311d | 20:10 |
| minute | abortretryfail: higher refresh rate on the internal display and no throttling of pcie | 20:10 |
| josch | abortretryfail: grep -B3 "Machine: MNT Reform 2 HDMI" /usr/share/flash-kernel/db/all.db | 20:10 |
| abortretryfail | minute: ahh that's actually worthwhile. | 20:11 |
| abortretryfail | I don't have a non-HDMI dtb for this kernel. I wonder if it will break if i try to switch. Let's see | 20:12 |
| abortretryfail | i did reform-display-config single and it put the right .dtb in and rebooted without issue. | 20:21 |
| abortretryfail | proc/device-tree/model now just says MNT Reform 2 | 20:21 |
| abortretryfail | I wonder if the problem I had only occurred during the bookworm/sid -> trixie/sid upgrade transition | 20:24 |
| minute | sounds good | 20:29 |
| abortretryfail | yep, works for me. Now on to my next task: migrate to this NVMe SSD rescued from a Steam deck upgrade. :) | 20:44 |
| murph[m] | Things seem a little quiet on the pocket reform front. Any news? Things progressing as planned? | 21:51 |
| dozens | i was just noticing that we haven't had a blog post in a month or two | 21:54 |
| noam | Is there a different uboot image needed to boot off of emmc vs uSD? Or, if I dd if=uSD of=eMMC, can I just flip the switch on the mobo to boot off the emmc? | 21:54 |
| noam | ...mostly asking because I lost my boot SD card and already have uboot flashed to emmc, but it doesn't boot at present :P | 21:54 |
| sigrid_ | it's the same u-boot | 21:56 |
| noam | Cool, thanks :) | 21:57 |
| noam | the chipset has no fallback for when the selected method is unavailable, though? :/ | 21:57 |
| noam | Would be kinda nice if in the next revision, that switch is controlled by the system controller instead of a physical switch TBH | 21:57 |
| noam | so that you can switch the boot device through the keyboard firmware | 21:57 |
| - yewscion (QUIT: Remote host closed the connection) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 21:58 | |
| sigrid_ | you'd probably need to wire lpc on that specific switch | 21:58 |
| sigrid_ | not sure you can do that easily | 21:58 |
| sigrid_ | other boards might not even have one to begin with | 21:59 |
| sigrid_ | also from security pov maybe you DON'T want to do it that easily | 21:59 |
| - oomono (QUIT: Quit: Connection closed for inactivity) (uid328183@id-328183.tinside.irccloud.com) | 22:03 | |
| abortretryfail | would be nice to not have to unscrew the heatsink to get to it. | 22:07 |
| abortretryfail | but that's simple enough to solder on a remote switch | 22:08 |
| noam | Ahhh right, it's a switch on the module itself, not the mobo. Cool. | 22:10 |
| sigrid_ | drilling a hole on the heatsink could help | 22:12 |
| sigrid_ | likewise, I think all switches should be accessible through a small hole | 22:13 |
| sigrid_ | then maybe a tooth pick is enough to do the job. or a needle | 22:13 |
| noam | to confirm, it's the small switch by the bottom right corner? | 22:14 |
| noam | (orientation: laptop on table with batteries near me) | 22:14 |
| josch | noam: maybe this helps: https://source.mnt.re/reform/reform-handbook/-/issues/2 | 22:14 |
| josch | personally, i have u-boot on my sd-card (so that i can do experiments easily), /boot on emmc and / on nvme | 22:15 |
| noam | Yep, that's the one I thought it was, thanks! | 22:16 |
| sevan | josch: I am, but I can still be bad at it :) It might be applicable but what does the implementation do, does it source the file in /usr/share, and if the files in /etc exists, it sources that after or does it check if the file in /etc exists and sources that, otherwise sources the file in /usr/share. If it's the latter, the file in /etc overrides. I prefer overrides as it means it takes precedence. | 22:16 |
| sevan | Overwrite means there's a copy process to me. | 22:16 |
| noam | josch: yeah, my experiments are all with the *kernel* though :P | 22:16 |
| noam | So as long as u-boot prioritizes uSD over eMMC, I'm good | 22:16 |
| noam | I can do testing as needed | 22:16 |
| noam | And, in normal operation, I can just leave the data card in there | 22:16 |
| josch | noam: u-boot does prioritize sd over emmc independent on whether u-boot from sd or emmc is used | 22:17 |
| abortretryfail | sigrid_: a hole in the heatsink is a good idea. I might do that to mine. | 22:17 |
| sigrid_ | best to drill with heatsink off the board | 22:18 |
| abortretryfail | :) | 22:18 |
| noam | heh | 22:18 |
| noam | josch: I know, that's my point ;) | 22:18 |
| josch | sevan: you are right! | 22:18 |
| noam | I don't plan on updating u-boot *ever* unless there's a compelling reason | 22:18 |
| josch | noam: do you have display support in your u-boot? | 22:19 |
| noam | yes | 22:19 |
| josch | sevan: i approved your MR -- now it's up to minute to press the merge button | 22:19 |
| abortretryfail | ACTION still needs to learn how to inkscape and fix the plans for the I/O covers. | 22:19 |
| josch | noam: then the only other important upgrade i can think of (unless we discover another bug) would be switching from the current vendor u-boot to mainline u-boot | 22:20 |
| sevan | josch: thanks :) | 22:20 |
| josch | noam: mainline u-boot would come with support for directly booting from nvme | 22:20 |
| noam | Not interested in that :P | 22:21 |
| noam | booting from eMMC makes life easier for me tbh | 22:21 |
| josch | cool! | 22:21 |
| noam | kernel and config in eMMC, NVMe is fully encrypted | 22:21 |
| noam | FDE is a lot easier with an unencrypted disk to load the kernel from :P | 22:21 |
| josch | yes | 22:23 |
| josch | my goal in the distant future is to have /boot unencrypted as the first partition on nvme | 22:23 |
| sevan | josch: with u-boot on emmc? | 22:23 |
| josch | the advantage would be that then i can switch my SoM without bothing about missing or outdated /boot on emmc which would be specific to the SoM | 22:24 |
| josch | sevan: yes | 22:24 |
| josch | because u-boot is device specific | 22:24 |
| josch | so it belongs on emmc | 22:24 |
| josch | but ideally my /boot should be able to contain kernel and initrd and dtbs that allow it to run on multiple platforms | 22:25 |
| noam | how do you get the heatsink back on without knocking down the screw stands O_o | 22:25 |
| sevan | just been thinking about that, I want to go root on zfs. I currently have everything on sd card, emmc has been blank since day-1 | 22:25 |
| * sigrid_ -> sigrid | 22:25 | |
| - sigrid (QUIT: Quit: Reconnecting) (~sigrid@ftrv.se) | 22:25 | |
| + sigrid (~sigrid@ftrv.se) | 22:25 | |
| josch | noam: with a lot of patience -- i don't know of any trick | 22:26 |
| sigrid | noam: pour some beer on it so it sticks | 22:26 |
| sevan | :) | 22:26 |
| noam | i think it'd make more sense to pour beer into myself | 22:26 |
| noam | so that I can stick to trying even when it fails | 22:27 |
| sigrid | only if it's a good beer | 22:27 |
| sevan | then lick the heatsink | 22:27 |
| noam | of course, I just got them in immediately after saying that :P | 22:27 |
| minute | ok nice, webrender actually works on a311d | 22:27 |
| minute | (in firefox) | 22:27 |
| vkoskiv | What's webrender? | 22:27 |
| minute | sevan: i heard my name! what shall i merge | 22:27 |
| minute | vkoskiv: the gpu accelerated rendering/compositing engine of firefox | 22:27 |
| sevan | minute: please :) | 22:27 |
| minute | josch: i meant josch :D which repo? | 22:28 |
| josch | minute: https://source.mnt.re/reform/reform-tools/-/merge_requests/44 | 22:28 |
| minute | i just watched a bit of a 1080p @ 60fps video in youtube on firefox | 22:28 |
| minute | (sw rendered i believe) | 22:29 |
| josch | wow | 22:29 |
| josch | minute wants me to abandon my ls1028a and switch to the a311d as well ;) | 22:30 |
| minute | nooo ;DD | 22:30 |
| minute | cannibalization effects | 22:30 |
| minute | i'm sure you'll need those 16GB ram! | 22:31 |
| josch | haha :) | 22:31 |
| josch | i'm very excited to see whether the ls1028a lets me do everything the imx8mq does | 22:31 |
| josch | lets see! | 22:31 |
| minute | yes :3 | 22:33 |
| + yewscion (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 22:33 | |
| noam | ...it's not booting after I flipped the switch and put everything back together :( | 22:34 |
| noam | ...or if it is, it's using a u-boot without display, but I don't think that's the case | 22:34 |
| noam | So now I have to *undo* that, flash uboot to a random SD card, boot that, flash it from there to eMMC, and do this _AGAIN_ | 22:35 |
| abortretryfail | always test before you put all the screws back in | 22:35 |
| noam | I'd rather put the screws in and take them back out if something went wrong than try and run it without a secure bottom :P | 22:36 |
| josch | noam: how did you flash u-boot to your emmc? | 22:36 |
| sevan | doesn't the system still try to boot from an sd card if you flip the switch on the module? | 22:36 |
| josch | noam: do you have an usb uart adapter for serial? | 22:36 |
| noam | sevan: you mean, if emmc isn't bootable? | 22:36 |
| noam | josch: I flashed it via dd if=SD of=eMMC count=~8MiB | 22:36 |
| noam | [pseudocommand there ofc] | 22:37 |
| josch | noam: did you specify the correct offset? | 22:37 |
| abortretryfail | sevan:mine prefers the sd card | 22:37 |
| noam | josch: ...it uses a different offset for eMMC and for uSD? | 22:37 |
| noam | That explains it. | 22:37 |
| noam | I assumed it was the same offset | 22:37 |
| sevan | noam: I thought the switch meant that it would try booting from the eMMC if an SD card was not found. The other (default) setting is it tries the sd card & only the sd card | 22:38 |
| noam | ...okay, I dunno | 22:38 |
| sevan | abortretryfail: mine too, I haven't flipped the switch. | 22:38 |
| sigrid | it's also a protected "partition" | 22:38 |
| josch | noam: and did you run echo 0 > /sys/class/block/mmcblk0boot0/force_ro | 22:38 |
| noam | ^ can someone answer for sure what, exactly, the behavior of the switch it? | 22:38 |
| noam | josch: I'm not on Linux | 22:38 |
| noam | so: no | 22:38 |
| abortretryfail | i flipped the switch so it boots from emmc, but i think what is happening is the bootloader on the emmc is booting the kernel and whatnot from the sd card first. | 22:38 |
| abortretryfail | idk, still learning about it though | 22:39 |
| josch | noam: the behaviour of the switch should be explained here https://source.mnt.re/reform/reform-handbook/-/issues/2 if something is not clear, please point it out so that i can fix the text | 22:39 |
| sigrid | noam: did you follow http://wiki.9front.org/mnt-reform | 22:39 |
| sevan | abortretryfail: ah, ok. | 22:39 |
| noam | josch: yeah, that matches what I thought | 22:39 |
| noam | Switch determines device to boot from, there is no fallback | 22:39 |
| noam | sigrid: nope, I always forget we *have* a wiki tbh | 22:40 |
| noam | > The other two are “boot” partitions, where U-Boot resides (/dev/sdO1 specifically) | 22:40 |
| noam | oops. | 22:40 |
| noam | Thanks :P | 22:40 |
| noam | found a trick for the heatsink replacement | 22:46 |
| noam | Don't let the bumpers fall off in the first place | 22:46 |
| abortretryfail | there are bumpers? | 22:46 |
| noam | The whatchamacallems | 22:48 |
| noam | The tubes | 22:48 |
| sigrid | spacers, perhaps | 22:48 |
| noam | That go between the screws and the, uh - | 22:48 |
| noam | yeah | 22:48 |
| noam | spacers :P | 22:48 |
| josch | from mars! | 22:48 |
| abortretryfail | I didn't notice any, hope I didn't lose them already... | 22:48 |
| noam | Anything better to clean the acrylic with than water? | 22:48 |
| noam | It's got fingerprints all over it :( | 22:49 |
| sigrid | they are hard to miss when then start moving, so it must be all good abortretryfail | 22:49 |
| sigrid | *they | 22:49 |
| noam | ^ yeah... | 22:49 |
| abortretryfail | definitely don't use acetone | 22:49 |
| noam | ...that dissolves acrylic, right? :P | 22:49 |
| abortretryfail | yep | 22:49 |
| sigrid | maybe to some artistic effect it can be used | 22:49 |
| minute | the spacing guild | 22:49 |
| noam | Ah yeah, doesn't quite dissolve it, discolors it? | 22:50 |
| abortretryfail | acetone+stencil might be interesting | 22:50 |
| abortretryfail | oh it will dissolve it if you use enough and give it time lol | 22:50 |
| noam | Sure, but | 22:50 |
| noam | That's like saying water will dissolve stone if you use it enough and give it enough time ;) | 22:50 |
| noam | Albeit not quite the same timescales | 22:50 |
| minute | also don't use IPA on acrylic, it will make the acrylic brittle | 22:50 |
| abortretryfail | yeah, it will do so on a human time scale | 22:50 |
| minute | you can use a microfiber cloth for example | 22:51 |
| noam | I have a 4G SD card [not micro] that's old but probably works, but I can't tell what's on it because it won't mount as any file system type... did I put a tarball on it, maybe?? | 22:53 |
| noam | nope. Weird | 22:54 |
| noam | Ah well, it's uboot now. | 22:54 |
| noam | ahaha and literally less than ten seconds after flashing it, I found the one I was using before. | 22:55 |
| - mjw (QUIT: Ping timeout: 250 seconds) (~mjw@gnu.wildebeest.org) | 23:36 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!