2025-06-12.log

+ bkeys (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)00:03
zehadoes -bios work with ovmf?00:05
zehai always pass something like -drive if=pflash,format=raw,unit=0,file.filename=/usr/share/OVMF/OVMF_CODE_4M.fd,file.locking=off,readonly=on00:05
minutehmm weird, the image i downloaded has arm64 stuff on /boot00:06
minutezeha: well, the UEFI menu is there and functional00:06
zehaok00:07
zehagood to know00:08
minutei guess the amd64/efi stuff wasn't enabled in this pipeline https://source.mnt.re/reform/reform-system-image/-/merge_requests/13300:08
minutewell, closing this chapter for today00:08
minutejosch: can you/i run a pipeline in CI that'll make the amd64 image?00:09
minutewell, this doesn't seem to be accurate https://status.mastodon.social/00:10
minutei'm getting lots of 503 and > 504 Gateway Time-out00:10
sigridnvm, I have MB 2.0 :') good thing I looked before doing anything00:18
- Gooberpatrol66 (QUIT: Remote host closed the connection) (~Gooberpat@user/gooberpatrol66)00:35
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)00:39
minutesigrid: the build makes one for 2.0 as well!00:53
sigridyeah I am about to flash it00:58
- jacqueline (QUIT: Remote host closed the connection) (~jacquelin@user/jacqueline)01:05
+ jacqueline (~jacquelin@user/jacqueline)01:06
- vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50)01:08
sigriddoesn't seem like lpc wants to show up with any of the micro usb cables I use01:14
+ L29Ah (~L29Ah@wikipedia/L29Ah)01:19
- chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:85f:9a0d:f00:1ac0:4dff:fedb:a3f1)01:26
kfxsigrid: it took me several tries -- I had to leave the software running, toggle the switch a few times, and then hold down the reset button a bit...01:30
- AnimaInvicta (PART: !!unknown attribute: msg!!) (~AnimaInvi@88-120-179-216.subs.proxad.net)01:31
minutesigrid: you have to connect the usb cable only after resetting the lpc into programming mode01:32
sigridkfx: all while usb was connected?01:32
minutesigrid: otherwise it won't work01:32
sigridyeah that's what I am doing exactly01:32
sigridabsolutely nothing happens. tried three cables01:32
sigridpower jack in, LPCP to on, reset, connect one end of the usb, then the other to the laptop -> nothing01:33
minutesigrid: that's super weird... except if all those cables are power-only cables01:33
minutesigrid: the procedure sounds good01:33
sigridit could be a combination of power only and/or broken cables01:33
sigridguess I have to stock up on a couple before they completely disappear01:33
minutethe curse of microusb :/01:33
+ ndufresne4 (~ndufresne@apple.collaboradmins.com)01:45
+ paperManu (~paperManu@107.159.213.145)02:08
- nsc (QUIT: Ping timeout: 265 seconds) (~nicolas@i5C74DD0A.versanet.de)03:14
+ nsc (~nicolas@66-99-142-46.pool.kielnet.net)03:16
- paperManu (QUIT: Ping timeout: 248 seconds) (~paperManu@107.159.213.145)03:21
- mjw (QUIT: Ping timeout: 268 seconds) (~mjw@gnu.wildebeest.org)04:29
- RandyK (QUIT: Ping timeout: 244 seconds) (~RandyK@user/randyk)06:27
+ RandyK (~RandyK@user/randyk)06:28
+ chomwitt (~chomwitt@2a02:85f:9a0d:f00:1ac0:4dff:fedb:a3f1)06:39
+ murph_nj (~murph@pool-108-35-93-154.nwrknj.fios.verizon.net)07:30
- cobra (QUIT: Ping timeout: 272 seconds) (~cobra@user/Cobra)07:30
- murphnj (QUIT: Ping timeout: 248 seconds) (~murph@user/murphnj)07:30
+ cobra (~cobra@user/Cobra)07:33
+ josch-pocket (~josch-poc@ip-185-104-138-25.ptr.icomera.net)08:05
josch-pocketminute: the reform-system-image project does not have the x86 runner assigned to it, so it cannot build my latest changes08:06
josch-pocketminute: i pushed something that should build it but i don't think i have the privileges to add the x86 runner?08:06
- chomwitt (QUIT: Ping timeout: 260 seconds) (~chomwitt@2a02:85f:9a0d:f00:1ac0:4dff:fedb:a3f1)08:07
+ josch-pocket_ (~josch-poc@ip-185-104-138-25.ptr.icomera.net)08:23
- josch-pocket (QUIT: Read error: Connection reset by peer) (~josch-poc@ip-185-104-138-25.ptr.icomera.net)08:23
* josch-pocket_ -> josch-pocket08:23
- austriancoder (QUIT: *.net *.split) (sid152545@id-152545.hampstead.irccloud.com)08:25
- aphistic (QUIT: *.net *.split) (sid347194@id-347194.ilkley.irccloud.com)08:25
- ezequielg (QUIT: *.net *.split) (sid363064@id-363064.uxbridge.irccloud.com)08:25
+ austriancoder (sid152545@id-152545.hampstead.irccloud.com)08:30
+ aphistic (sid347194@id-347194.ilkley.irccloud.com)08:30
+ ezequielg (sid363064@id-363064.uxbridge.irccloud.com)08:30
+ Core1951 (~aakast@80-62-117-191-mobile.dk.customer.tdc.net)08:34
- aakast (QUIT: Ping timeout: 244 seconds) (~aakast@net-93-65-57-194.cust.vodafonedsl.it)08:36
+ aakast (~aakast@net-93-65-57-194.cust.vodafonedsl.it)08:41
- Core1951 (QUIT: Read error: Connection reset by peer) (~aakast@80-62-117-191-mobile.dk.customer.tdc.net)08:41
- josch-pocket (QUIT: Quit: josch-pocket) (~josch-poc@ip-185-104-138-25.ptr.icomera.net)08:58
+ josch-pocket (~josch-poc@ip-185-104-138-25.ptr.icomera.net)08:58
+ gustav28 (~gustav@c-78-82-55-148.bbcust.telenor.se)09:44
- josch-pocket (QUIT: Ping timeout: 248 seconds) (~josch-poc@ip-185-104-138-25.ptr.icomera.net)09:56
+ josch-pocket (~josch-poc@185.238.219.89)10:28
+ arminweigl_ (~arminweig@sourcehut/user/arminweigl)10:38
- arminweigl (QUIT: Ping timeout: 260 seconds) (~arminweig@sourcehut/user/arminweigl)10:38
* arminweigl_ -> arminweigl10:38
+ Core1951 (~aakast@80-62-117-191-mobile.dk.customer.tdc.net)10:40
- aakast (QUIT: Ping timeout: 268 seconds) (~aakast@net-93-65-57-194.cust.vodafonedsl.it)10:42
bremnerwhat do people recommend for temperature monitor10:44
bremner(on the pocket reform)10:44
bremnerI can look in /sys, but I wondered if there is something with some simple logging. GUI or CLI doesn't matter to me10:45
minutejosch-pocket: ahh ok, i can do that of course10:50
minutebremner: watch -n 1 sensors >> logfile10:51
bremnerthanks10:53
minutenot sure if sensors has some formatting options to just output one line etc10:54
josch-pocketbremner: i have the temperature plugin enabled in waybar with this config:"temperature": { "format": "{temperatureC}°C" }, 10:54
gsoraoh, panvk is vulkan 1.2 conformant now10:58
bremnerthanks. I guess I need to start actually configuring things.  I got distracted part way through my "deploy new machine via ssh" process 10:58
gsorark3588 keeps on delivering :^) gotta understand how toplay with the thing tho10:58
josch-pocketbremner: what automation are you using for automatic deployment?10:59
bremnerjosch-pocket: consfigurator10:59
josch-pocketah sean's thing - that was on my list to check out, thanks!11:00
bremnerno problem, happy to discuss if you get stuck. 11:00
bremneror drop by #consfigurator; it's small and quiet but sean is usually around11:01
+ andreas-e (~Andreas@bou73-h01-176-169-244-250.dsl.sta.abo.bbox.fr)11:03
- andreas-e (QUIT: Client Quit) (~Andreas@bou73-h01-176-169-244-250.dsl.sta.abo.bbox.fr)11:04
josch-pocket👍11:04
gsoraoh, installing mesa-vulkan-drivers makes vulkan magically appear11:16
gsorahttps://files.catbox.moe/ed1rsx here's the output of vulkaninfo, mali gpu is there!11:18
+ arminweigl_ (~arminweig@sourcehut/user/arminweigl)11:19
- arminweigl (QUIT: Ping timeout: 244 seconds) (~arminweig@sourcehut/user/arminweigl)11:20
- josch-pocket (QUIT: Quit: josch-pocket) (~josch-poc@185.238.219.89)11:21
- arminweigl_ (QUIT: Ping timeout: 248 seconds) (~arminweig@sourcehut/user/arminweigl)11:24
gsoratesting vulkan from a rust gpu-accelerated program, makes a world of difference!11:25
+ arminweigl (~arminweig@sourcehut/user/arminweigl)11:36
- Ar|stote|is (QUIT: Quit: No Ping reply in 180 seconds.) (~linx@149.210.5.33)12:17
+ paperManu (~paperManu@107.159.213.145)12:21
+ Ar|stote|is (~linx@149.210.5.33)12:22
+ mjw (~mjw@gnu.wildebeest.org)12:51
- wiedi_ (QUIT: Remote host closed the connection) (~wiedi@ip5f581498.dynamic.kabel-deutschland.de)12:58
bremnerIs there some recommended reading / best practices about using mmc vs using nvme (on a pocket reform). Related, is there good reason not to use luks / FDE?13:05
zehai use luks on nvme and i'm somewhat sus' of the speed, but havent done any crosschecking13:07
bremnerthat's on rk3588?13:09
zehayes13:10
+ wiedi (~wiedi@ip5f581498.dynamic.kabel-deutschland.de)13:11
+ arminweigl_ (~arminweig@sourcehut/user/arminweigl)13:35
- arminweigl (QUIT: Ping timeout: 260 seconds) (~arminweig@sourcehut/user/arminweigl)13:35
* arminweigl_ -> arminweigl13:35
gsoraLUKS is not as fast as the OPAL encryption, but you can trust it more - plain luks is usually slower as well13:38
- Core1951 (QUIT: Ping timeout: 244 seconds) (~aakast@80-62-117-191-mobile.dk.customer.tdc.net)13:44
+ aakast (~aakast@87-49-45-53-mobile.dk.customer.tdc.net)13:56
joschzeha: I preformed MBit/s benchmarks on imx8mq and ls1028 with luks here: https://community.mnt.re/t/does-drive-encryption-cause-the-reform-to-slowdown/87114:33
- paperManu (QUIT: Ping timeout: 244 seconds) (~paperManu@107.159.213.145)14:37
* mjw -> Guest737614:46
- Guest7376 (QUIT: Killed (erbium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org)14:46
* Guest1879 -> mjw14:46
+ Guest7376 (~mjw@gnu.wildebeest.org)14:47
- RandyK (QUIT: Remote host closed the connection) (~RandyK@user/randyk)15:00
- Ar|stote|is (QUIT: Quit: No Ping reply in 180 seconds.) (~linx@149.210.5.33)15:00
+ RandyK (~RandyK@user/randyk)15:00
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)15:03
+ Ar|stote|is (~linx@149.210.5.33)15:05
gsorathis is my encrypted samsung 990 pro on rk3588 + OPAL: https://files.catbox.moe/sni4lu.png15:12
+ bkeys (~Thunderbi@38-146-94-247.echocast.zone)15:14
bkeysjosch: So I think I want to revert to uboot on my rk3588, at least until the edk2 situation gets better. How would I go about flashing it? It doesn't seem like I can do so within Linux15:15
grimmwarebattery rebalancing looks to have gone pretty well so far - I'm now seeing that the sysctl thinks the batteries are at 1% but it just keeps on trucking15:15
joschbkeys: you can use reform-flash-uboot15:16
joschsudo reform-flash-uboot emmc15:16
bkeysHow would that work if the mainline kernel I am booting doesn't see the emmc?15:17
bkeysIs there a way to force booting from SD card?15:17
joschwait what?15:17
joschoh15:17
bkeysSo I have edk2, and I boot it up and it doesn't see the emmc15:17
joschyou are mainline without collabora patches?15:17
grimmwareO_O15:17
bkeysYeah, but I think the reason it doesn't see it is the boot firmware15:18
joschhm i cannot tell you :/15:18
bkeysFor some reason I can only ever boot in ACPI mode and I'm told if I can boot DT it all goes away; but boot never succeeds in DT15:18
grimmwarebkeys: you probably want to put a system image on an SD card, boot from that and then use reform-flash-uboot15:19
grimmwareseems like the quickest way around your immediate problem15:19
joschgsora: I just ran gnome-disks benchmark on my rk3588 pocket reform (you are rk3588 classic reform?) and i get around 430 MB/s (raw reads without encryption)15:19
joschgrimmware: bkeys cannot do that because they don't have u-boot on emmc15:20
bkeysgrimmware: It would, but is there a way I can force booting from SD card?15:20
bkeyssystem image needs uboot not edk2 afaik15:20
bkeysI need a way to bypass edk2 entirely15:20
joschbkeys: earlier today, minute repeated that rk3588 will load u-boot from sd-card and not emmc if it fails to boot from emmc and not otherwise15:20
grimmwareoh shit right, yeah I'll be quiet15:20
gsorajosch: try with kdiskmark, you'll get more precise results IMO15:21
gsorathat said, my encryption is drive-level, which would explain the numbers15:21
joschgsora: you are on rk3588 classic reform?15:21
joschgsora: the results might be more precise but they'll still be in the same ballpark :)15:22
bkeysjosch: Yeah but how will I make it purposefully fail to boot from emmc?15:22
joschbkeys: sorry, i honestly don't know :(15:22
gsorajosch: yup15:22
gsorato both questions!15:22
joschbkeys: if you like you could be the guinea pig for my Debian efi boot image15:22
bkeysNo problem, I'm anticipating a slow day at my bike shop so I'll give progress updates as progress happens15:22
bkeysjosch: As in test my edk2?15:22
gsoraooo you own a bike shop? cool15:23
bkeysI mean in the end of the day I like a graphical UEFI so I think maybe I want to try more with this edk2 before nuking it; and seeing about device tree15:23
joschbkeys: then your edk2 might be booting an image which maybe can do more? who knows. Maybe it doesn't work at all. I only tested via qemu so far.15:23
bkeysgsora: Yeah I got hit by the elon musk layoffs and decided I didn't want to write software professionally anymore. Finding a new job was hard and it was always a dream to me to own a bike shop and I started one with a box of parts and now I'm a Cannondale dealer, and I sell escooters and other stuff15:24
bkeysI never worked for musk, just the wave of layoffs he started15:24
gsoraha, oddly enough that was my aim after my last job went belly up15:25
gsorai just can't deal with customers :D15:25
joschminute: i'm catching up with backlog right now. Yesterday you said that the fixed pocket reform backplate is the one with "no black color inside" but mine has both of its sides on the inside black -- does that mean I have one of the old backplates? My pocket shipped after you found and corrected the issue I think, so I am a bit confused.15:25
bkeysgsora: Yeah I like this a lot better, it's less money but I have freedom and tech is like when I was a teenager again, I can just have fun and do things in obscure ways just for the sake of obscurity I don't have to justify it to anyone15:27
gsoraagreed 100%! i could talk hours about my experiments with mineral oil brakes :^)15:28
zehajosch: on the inside side there should be at least a few cm of visible pcb base material on the left and right15:30
joschbkeys: your job sounds awesome -- i 100% understand the appeal to do nothing with computers for a living but i'm too much of a chicken to switch professions XD15:30
joschzeha: oh phew i just misremembered, probably from the photos on teh forum15:31
bkeysjosch: Yeah I felt the same way, I was making a lot of money and working on paying my house off and I thought "Yeah, once I have no house payment I'll quit" but in reality I was never going to quit15:31
bkeysThank you for the nice words15:31
zeha(cant show you a picture as i'm already on the train to brno)15:31
joschthe inside of my display plate is *not* black but dark green-ish on the sides15:31
zehathat should be a fixed one then15:31
bkeysThe horrible economy is what made me realize "You know what? If I can't find a new job and I'll be broke soon, I mine as well chase my dream instead of a stupid job"15:31
joschminute: false alarm, sorry, all good :)15:31
bkeysAnd it just happened to work out15:31
joschnice success story, thank you for sharing! :)15:32
grimmwarebkeys: that's really nice to hear <3 super stoked for you15:33
spewbkeys: this is very inspiring15:33
bkeysspew, grimmware: I'm glad I can be an inspiration c:15:33
bkeysWhat's great is I also sell a lot of escooters, and even though this is America and we worship cars, the town here isn't that bad but I'm actively changing the streets15:34
gsoraif i entered a bike shop whose owner is hacking away on a reform i would be puzzled *and* happy15:34
bkeysI got multiple places and the city to put up bike racks, and I'm pushing for a bike lane now because the old guard of the town is really tired of people riding on the sidewalk (even though they blame me). In reality a lot of people in my hometown are impoverished and escooters and bikes are their only option15:35
bkeysSo I'm glad I can increase the heat on the establishment to try to push them into putting up more bike lanes so these people have somewhere other than the sidewalk to ride15:35
zehain theory maskrom mode should allow you to overwrite emmc ... see https://community.mnt.re/t/rk3588-maskrom-boot/3286/2 and the links from there15:41
gsorain rk maskrom mode you can do pretty much whatever you want yeah, you have to load ram training software first though15:42
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@38-146-94-247.echocast.zone)16:09
+ bkeys (~Thunderbi@38-146-94-247.echocast.zone)16:09
+ bkeys1 (~Thunderbi@66.110.201.50)16:11
- bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50)16:11
minutejosch: hmmmm let me double check16:14
- bkeys (QUIT: Ping timeout: 252 seconds) (~Thunderbi@38-146-94-247.echocast.zone)16:14
minutejosch: ah, now i read the false alarm message16:14
+ bkeys (~Thunderbi@66.110.201.50)16:14
minutebkeys: i haven't tried any of this yet, but in general, EDK2/EFI will load anything you put in the EFI folder in a boot/ESP partition, no? so i think you just need the right tool there, like EFI shell, to be able to specify the correct dtb and load linux manually16:16
minutebkeys: then, from this loaded linux, you can erase the emmc16:16
minutebkeys: i also found this tool, but didn't have time yet to check if that's a good solution https://github.com/TravMurav/dtbloader?tab=readme-ov-file#usage16:17
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)16:19
+ bkeys (~Thunderbi@66.110.201.50)16:19
minutebkeys: did you see my messages?16:19
minuteasking because you're popping in and out here16:19
bkeysYes I did16:19
minuteok cool16:19
bkeysI pop in and out because right now I have to mooch wifi from the movie theatre and a car will drive by and destroy my connection16:20
minutebkeys: i will cook lunch now in the mnt kitchen and pop in the edk2 v1.1 sd card into reform next and see if i can see sth16:23
bkeysAlright, I'm downloading a minimal fedora rawhide image and am gonna see if that boots on DT16:23
bkeysIt's 6.16 so it should have the proper devicetree stuff mainlined16:24
bkeysIt's weird for me it's the opposite of what the edk2 image documents say, I can't boot absolutely anything in DT ACPI is the only thing that works16:24
minutebkeys: did you understand that you need to manually choose a dtb though? because fedora can't know that you have a reform16:26
minuteit thinks you have firefly aio16:26
bkeysYeah I'll just have to be walked through it16:26
bkeysThis is a realm I am completely ignorant of16:26
bkeysThat's right I needed to see if I hook it up to external display if then I can see output or not16:26
minutebkeys: i have no idea how to do it either, yet16:27
minutei just booted into edk2 on my reform next and it shows up on the internal display via hdmi converter16:27
bkeysYes edk2 will boot via internal display16:29
bkeysI'm wondering if Fedora, thinking it's firefly loads up the external display post edk216:29
bkeysCause maybe that explains why the screen is blank16:29
minutei had to connect an external keyboard but now i can navigate the edk2 menu16:32
minuteunder acpi/device tree there is an option DTB override16:32
minutebkeys: i typed in \dtb in that field16:34
minuteand popped in my regular microsd with reform system image... it seems to boot 16:35
minuteyes, booted to gnome, and devicetree is correct. BUT it could be that edk2 just respects extlinux.conf16:36
minutethat would make things even easier...16:37
bkeysminute: Do you get hardware acceleration?16:37
bkeysIt's weird cause my edk has absolutely no issue with my internal keyboard16:38
minuteyes i do16:38
minutebkeys: well, i guess that's reform next related16:38
minutebkeys: ok so i think edk2 just found my custom extlinux.conf and that's why everything worked16:39
minutebkeys: do you have a microsd card with the latest mnt reform system image?16:39
bkeysI have plenty let me see if any are in the shop16:39
bkeysAlso do you have a link to the system image handy?16:40
bkeysI can start the download of htat16:40
bkeysminute: 16:40
minutebkeys: mnt.re/system-image16:41
minutehttps://source.mnt.re/reform/reform-system-image/-/jobs/10736/artifacts/file/reform-system-rk3588.img.gz16:42
bkeysThat's gonna take a minute to download16:43
bkeysNo pun intended16:43
bkeysYes I have an SD card16:44
sigriddoes anyone know if edk2 with dsi mode possible?16:46
sigridor is it currently hdmi converter only16:47
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50)16:47
+ bkeys (~Thunderbi@66.110.201.50)16:47
minutesigrid: no idea. not out of the box. in any case someone would need to get the actual device tree for reform into edk216:48
sigridwhich edk2 repo/branch is it, btw?16:49
+ bkeys1 (~Thunderbi@38-146-94-247.echocast.zone)16:50
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@66.110.201.50)16:52
* bkeys1 -> bkeys16:52
minutesigrid: https://github.com/edk2-porting/edk2-rk358816:52
sigridthanks16:53
bkeysminute: With your system image are you able to see the emmc?16:53
minutebkeys: why i shouldn't be able to? it's using the correct device tree16:54
bkeysIt doesn't on my Fedora image which makes sense kinda; I'm just wondering16:54
bkeysSo if edk loaded the system image, but doesn't the system image have uboot on the front of the SD card?16:54
minuteuboot is not loaded because edk2 is loaded from emmc16:55
minutebkeys: lets not try to solve 2 different problems at the same time. lets first get your computer into working state again16:55
bkeysAlright, I'm downloading the system image, it says about 15 minutes16:55
minutebkeys: ping me again when you got it unpacked and dded on microsd16:56
bkeysWill do16:56
bkeysminute: Curious question, so when you put \dtb into the override and it read the dtb from the SD card; does it assume the SD card or do you know if it can read it from any device?16:57
minutebkeys: i chose the sd card in the boot manager. but! i made a mistake, i see right now that it wants to reset after saving the config, so i probably booted with uboot. let me do it again16:59
bkeysDang it, I only have the internal SD card reader of the Reform17:06
bkeysI do have a USB hard disk17:07
- RandyK (QUIT: Remote host closed the connection) (~RandyK@user/randyk)17:09
+ RandyK (~RandyK@user/randyk)17:09
+ bkeys1 (~Thunderbi@66.110.201.50)17:09
- bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50)17:10
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@38-146-94-247.echocast.zone)17:11
+ bkeys (~Thunderbi@66.110.201.50)17:12
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)17:12
+ bkeys1 (~Thunderbi@66.110.201.50)17:13
* bkeys1 -> bkeys17:15
minutemy usb card reader isn't seen by edk2 so not sure how to continue17:16
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)17:16
+ bkeys (~Thunderbi@66.110.201.50)17:16
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)17:19
+ bkeys1 (~Thunderbi@66.110.201.50)17:19
* bkeys1 -> bkeys17:21
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)17:24
+ bkeys1 (~Thunderbi@66.110.201.50)17:24
minuteinteresting https://www.kernel.org/doc/Documentation/efi-stub.txt17:25
gsoraa dog just appeared on my oled screen!17:25
- bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50)17:26
+ bkeys (~Thunderbi@66.110.201.50)17:26
grimmwareThat's Tina!17:27
bkeysminute: Okay so I have the system image on an SD card17:28
gsoraminute: that's what i'm working on but in rust, in my spare time :^)17:28
bkeysI tried to boot via SD card and it just went back to edk2 after a whie of silence17:29
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)17:29
minutebkeys: ok, i figured out how to load our linux kernel from sd card using uefi shell, but need to check how to pass the correct fdt17:29
minutebkeys: yep, it doesn't work automatically for some reason17:29
minutewhat i already figured:17:29
minutein uefi shell you can enter "map" to get a list of drive mappings17:29
minutefor example FS2: is the boot partition of an sd card for me17:30
minutethen it works a bit like DOS17:30
minuteyou can type dir or ls and get the list of files17:30
minuteand you can type vmlinu TAB to tab complete the kernel, and press enter to load it. but for me, that results in a black screen presumably because i have to point it to the right dtb17:31
+ bkeys (~Thunderbi@66.110.201.50)17:32
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)17:35
+ bkeys (~Thunderbi@66.110.201.50)17:37
minutewow, this think has a full text editor17:38
minuteand hex editor17:38
gsorathe uefi?17:38
minuteyes17:39
minuteuefi shell of edk217:39
minutealso can download and boot stuff from http17:39
gsorayeah the http thing i knew, wicked stuff17:39
gsorait also scares me17:39
gsorathe attack surface looks gargantuan17:39
minutehttps://uefi.org/sites/default/files/resources/UEFI_Shell_Spec_2_0.pdf17:40
minuteahhhh17:41
minuteok, got it to work17:43
minutethis is pretty awesome tbh17:43
minuteok so on arm64 and amd64 the kernel is also an efi executable17:43
minuteand you can just pass it commandline params17:43
gsorawith an efistub yeah17:44
minutelike dtb=... and initrd=... and kernel options17:44
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)17:44
minuteand then it just loads those and boots happily17:44
+ bkeys1 (~Thunderbi@66.110.201.50)17:44
minuteso this way i was able to load our system image from edk2 uefi shell17:44
* bkeys1 -> bkeys17:47
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)17:47
+ bkeys (~Thunderbi@66.110.201.50)17:47
minutebkeys: i made you a recipe https://mastodon.social/@mntmn/11467118166701757317:51
minutebkeys: happy hacking17:51
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)17:52
+ bkeys (~Thunderbi@66.110.201.50)17:52
minuteok, back to other topics then...17:56
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50)17:57
+ bkeys1 (~Thunderbi@66.110.201.50)17:57
* bkeys1 -> bkeys18:00
bkeysSorry about not existing18:04
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50)18:05
+ bkeys (~Thunderbi@66.110.201.50)18:05
bkeysminute: So you were able to boot the system image without issue on edk2?18:05
bkeysSorry about not existing18:05
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)18:07
+ bkeys (~Thunderbi@66.110.201.50)18:07
minutebkeys: yes18:08
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)18:12
+ bkeys (~Thunderbi@66.110.201.50)18:12
sigridgot two micro usb cables that work. updated lpc fw, driver works properly now18:14
+ AnimaInvicta (~AnimaInvi@88-120-179-216.subs.proxad.net)18:14
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)18:15
+ bkeys (~Thunderbi@66.110.201.50)18:15
- bkeys (QUIT: Remote host closed the connection) (~Thunderbi@66.110.201.50)18:17
+ bkeys (~Thunderbi@66.110.201.50)18:18
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)18:20
+ bkeys1 (~Thunderbi@66.110.201.50)18:20
* bkeys1 -> bkeys18:23
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)18:23
+ bkeys (~Thunderbi@66.110.201.50)18:23
- Ar|stote|is (QUIT: Quit: No Ping reply in 180 seconds.) (~linx@149.210.5.33)18:25
- bkeys (QUIT: Remote host closed the connection) (~Thunderbi@66.110.201.50)18:28
+ bkeys (~Thunderbi@66.110.201.50)18:28
minutesigrid: oof nice to hear something positive today18:29
+ Ar|stote|is (~linx@149.210.5.33)18:30
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)18:32
+ reform6040 (~john@83.24.27.204.ipv4.supernova.orange.pl)18:32
+ bkeys (~Thunderbi@66.110.201.50)18:32
- reform6040 (QUIT: Client Quit) (~john@83.24.27.204.ipv4.supernova.orange.pl)18:34
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)18:35
+ bkeys (~Thunderbi@66.110.201.50)18:35
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)18:39
+ bkeys1 (~Thunderbi@66.110.201.50)18:39
- bkeys1 (QUIT: Ping timeout: 260 seconds) (~Thunderbi@66.110.201.50)18:43
joschminute: bad day? :(19:08
joschminute: could you also assign the amd64 runner to reform-system-image? i'm currently building the images on rk3588 and rk3588 is fast but emulating amd64 is still slooooooow19:09
josch(aka i am still finding bugs, so good you are not testing this yet yourself :D)19:12
+ bkeys (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)19:19
bkeysOkay I'm on break and have a reliable connection19:20
bkeysminute: So mind expounding on your steps to boot system image via edk2 when you have a moment?19:20
sigridminute: sorry but it also jumps to random numbers - 44%, then 0%19:21
sigridthen back to 100%19:21
sigridI think the driver has a bug19:21
joschsigrid: rk3588 classic reform? there was a recent bug with the baud rate spotted a few days ago. Are you running the very latest lpc version from git?19:25
sigridyeah19:26
sigridI built it yesterday19:26
+ chomwitt (~chomwitt@2a02:85f:9a08:a800:1ac0:4dff:fedb:a3f1)19:32
+ mark_ (~mjw@gnu.wildebeest.org)19:34
- RandyK (QUIT: Remote host closed the connection) (~RandyK@user/randyk)19:37
+ RandyK (~RandyK@user/randyk)19:37
- Guest7376 (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org)19:37
+ Core1951 (~aakast@80-62-117-167-mobile.dk.customer.tdc.net)20:01
- aakast (QUIT: Ping timeout: 276 seconds) (~aakast@87-49-45-53-mobile.dk.customer.tdc.net)20:02
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)20:07
+ bkeys (~Thunderbi@66.110.201.50)20:12
* mjw -> Guest44820:13
- Guest448 (QUIT: Killed (silver.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)20:13
* mark_ -> mjw20:13
+ Guest448 (~mjw@2001:1c06:2486:a800:7602:5eff:dc71:a72c)20:14
minutei just launched the first version of https://support.mnt.re/20:18
minutebut not linked on the website yet20:18
minutebkeys: i sent you a link to a recipe of the steps earlier, didn't see it?20:19
bkeysminute: I don't have a link no20:19
minutebkeys: >  17:51:45 <@minute>bkeys: i made you a recipe https://mastodon.social/@mntmn/11467118166701757320:20
bkeysminute: So you don't think I need uboot to get a nicely booted Linux, even if it's not Fedora right now?20:21
bkeysminute: So do you know how to scroll in the UEFI shell?20:24
minutebkeys: what do you mean "think", i tried it20:25
minutebkeys: you can scroll by using page up and page down20:26
minutebkeys: which should by hyper+up / hyper+down20:26
bkeysI suppose you know then20:26
- bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@66.110.201.50)20:26
+ bkeys (~Thunderbi@66.110.201.50)20:26
minutesigrid: please find the file "firmware" in some subdirectory containing "spi1.1" (or so) under "/sys" and cat this file 100 times in a loop. is the output consistent? if not, can i get a copy?20:31
minutesigrid: does "dmesg | grep lpc" mention an API version?20:32
- bkeys (QUIT: Ping timeout: 265 seconds) (~Thunderbi@66.110.201.50)20:35
+ bkeys (~Thunderbi@66.110.201.50)20:38
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@66.110.201.50)20:43
+ bkeys1 (~Thunderbi@66.110.201.50)20:43
- bkeys1 (QUIT: Client Quit) (~Thunderbi@66.110.201.50)20:45
- Core1951 (QUIT: Ping timeout: 260 seconds) (~aakast@80-62-117-167-mobile.dk.customer.tdc.net)20:48
+ aakast (~aakast@87-49-42-18-mobile.dk.customer.tdc.net)20:49
minuteok, i'm away20:50
+ bkeys (~Thunderbi@66.110.201.50)20:51
bkeysminute: i got the system image booting from USB disk you are awesome your guide worked like a charm20:51
bkeysNow I can use this as a base to boot other kernels20:51
+ potash5 (~potash@user/foghorn)20:52
- potash (QUIT: Ping timeout: 276 seconds) (~potash@user/foghorn)20:52
- potash5 (QUIT: Client Quit) (~potash@user/foghorn)20:54
+ potash (~potash@user/foghorn)20:55
+ Core1951 (~aakast@87-49-42-140-mobile.dk.customer.tdc.net)20:56
- aakast (QUIT: Ping timeout: 260 seconds) (~aakast@87-49-42-18-mobile.dk.customer.tdc.net)20:58
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@66.110.201.50)20:58
+ aakast (~aakast@80-62-117-175-mobile.dk.customer.tdc.net)20:59
- Core1951 (QUIT: Ping timeout: 252 seconds) (~aakast@87-49-42-140-mobile.dk.customer.tdc.net)21:00
sigridminute: nope. it missed some of the chunks. I remember seeing this with 9front as well and it was caused by not having the timings in the driver correct. timings being delays in this case21:02
sigriddmesg doesn't show API version21:03
+ bkeys (~Thunderbi@66.110.201.50)21:06
- Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66)21:07
- bkeys (QUIT: Ping timeout: 245 seconds) (~Thunderbi@66.110.201.50)21:10
+ bkeys (~Thunderbi@66.110.201.50)21:16
- bkeys (QUIT: Client Quit) (~Thunderbi@66.110.201.50)21:17
+ bkeys (~Thunderbi@66.110.201.50)21:17
- aakast (QUIT: Ping timeout: 252 seconds) (~aakast@80-62-117-175-mobile.dk.customer.tdc.net)21:18
+ aakast (~aakast@80-62-117-206-mobile.dk.customer.tdc.net)21:22
+ bkeys1 (~Thunderbi@66.110.201.50)21:23
- bkeys (QUIT: Ping timeout: 276 seconds) (~Thunderbi@66.110.201.50)21:24
+ bkeys (~Thunderbi@38-146-94-247.echocast.zone)21:25
- L29Ah (QUIT: Ping timeout: 272 seconds) (~L29Ah@wikipedia/L29Ah)21:27
- xha (QUIT: Quit: WeeChat 4.6.3) (~xha@user/xha)21:27
- bkeys1 (QUIT: Ping timeout: 260 seconds) (~Thunderbi@66.110.201.50)21:28
- potash (QUIT: Ping timeout: 260 seconds) (~potash@user/foghorn)21:28
+ potash5 (~potash@user/foghorn)21:33
- potash5 (QUIT: Read error: Connection reset by peer) (~potash@user/foghorn)21:37
minutejosch: i've now clicked several times around mindlessly in the wrong admin panels in gitlab to find the runners21:37
minutespoiler: it's not settings -> ci/cd -> runners21:38
minuteinstead, it's ci/cd -> runners21:38
joscho021:38
joschwow, brilliant :/21:38
minutesigrid: ok sounds like you don't have the updated driver somehow21:38
minutejosch: not good when brain not working at 100% :D21:39
- bkeys (QUIT: Ping timeout: 272 seconds) (~Thunderbi@38-146-94-247.echocast.zone)21:39
joschminute: good news though is that i verified that the thing boots with the qemu invocation you have above21:39
minutejosch: oh great21:39
joschminute: i would've 100% expected the setting to be in settings -> ci/cd -> runners as well21:39
+ potash (~potash@user/foghorn)21:39
minutejosch: ok, project assigned to runner21:40
joschand already errored out -- fixing...21:41
joschbut it's running on the right runner, thank you!21:41
sigridminute: is the updated driver available with apt update/apt upgrade?21:43
+ L29Ah (~L29Ah@wikipedia/L29Ah)21:44
- potash (QUIT: Ping timeout: 248 seconds) (~potash@user/foghorn)21:45
+ potash (~potash@user/foghorn)21:46
minutejosch: cool! now that sigrid is asking, IIRC you did make a new reform-tools release with the stuff from !128 no? https://source.mnt.re/reform/reform-tools/activity21:47
minuteah i think that's in experimental21:47
minutebut we have 1.73 in our repo https://mntre.com/reform-debian-repo/pool/main/r/reform-tools/21:48
minutesigrid: do you have reform2-lpc-dkms_1.73?21:48
joschminute: no, i did not make a new reform-tools release21:49
joschwould you like me to make one now?21:50
minutejosch: yeah, that would be great!21:51
joschokay, on it...21:51
minutejosch: sorry, i might be confusing things. i mostly care about reform2-lpc-dkms. but they're built together, right?21:52
joschminute: yes, it's bundled together21:52
joschminute: we can also split the lpc module out and package it sepearately if you like21:52
minutejosch: no it's fine21:52
minutesigrid: that's the explanation then, sorry, i thought that the new lpc driver was already released.21:54
minutesigrid: if you want to test it immediately, that's quite easy, just `make` here https://source.mnt.re/reform/reform-tools/-/tree/main/lpc?ref_type=heads21:54
minutesigrid: and then rmmod reform2_lpc; insmod ./reform2_lpc.ko 21:55
+ bkeys (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)22:00
minutei pushed the pocket system controller firmware update to fwupd stable, so it should be available for most people now (or whenever the mirrors are synced)22:04
joschminute: oh no i know what went wrong -- i totally missed your message from 3 days ago: https://mntre.com/reform-irc-logs/2025-06-09.log.html#t17:21:1322:08
minutejosch: aww ok!22:08
+ bkeys1 (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)22:09
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)22:12
* bkeys1 -> bkeys22:12
- gustav28 (QUIT: Quit: Quit) (~gustav@c-78-82-55-148.bbcust.telenor.se)22:15
joschminute: reform-tools 1.74-1 uploaded to experimental22:15
minutejosch: noice, thanks a ton22:15
joschif my pocket randomly, with the larger battery packes, while sitting on the table without being touched, switches off and reboots at full battery and while plugged in, would you still suspect a loose connection?22:16
joschminute: this was your release, most changes are by you :)22:16
+ bkeys1 (~Thunderbi@173.186.16.211)22:17
- bkeys (QUIT: Ping timeout: 260 seconds) (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)22:18
* bkeys1 -> bkeys22:18
+ Gooberpatrol66 (~Gooberpat@user/gooberpatrol66)22:30
sigridminute: can confirm latest fixes the issue. thank you!22:31
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)22:35
- bkeys (QUIT: Read error: Connection reset by peer) (~Thunderbi@173.186.16.211)22:44
+ bkeys (~Thunderbi@h211.16.186.173.dynamic.ip.windstream.net)22:45
minutesigrid: great, that's a relief22:59
minutejosch: well, certainly not normal @ reboot. a bad connection can also be just too high resistance so that the voltage drops under load.23:00
+ cow321_ (~deflated8@user/meow/deflated8837)23:07
- aakast (QUIT: Read error: Connection reset by peer) (~aakast@80-62-117-206-mobile.dk.customer.tdc.net)23:07
+ aakast (~aakast@net-93-65-57-194.cust.vodafonedsl.it)23:07
joschokay, then i'll re-do the JST-PH connector -- that one seemed suspicious as it was sitting way too tightly....23:08
- cow321 (QUIT: Ping timeout: 268 seconds) (~deflated8@user/meow/deflated8837)23:09
* cow321_ -> cow32123:09
joschminute: here is your first amd64 efi image for you to try out at your leisure: https://source.mnt.re/reform/reform-system-image/-/jobs/10868/artifacts/browse23:11
minutejosch: niiiice23:23
bkeysI am booted into the system image via edk2 and it doesn't see my NVME drive -_-23:43
bkeysI was hoping to drop the DTB file on the nvme drive which has Fedora on it and I'd be smooth sailing in theory, but no there's always gotta be a last stumbling block23:43
bkeysI don't have any nvme adapters either23:44
joschand my power-off problems might not've been the larger cells after all -- i put the original cells back in and am still getting the sudden power off issue. I now reseated the green and black cable to the charger board just to be sure...23:51
joschafter that, i'll bridge pins 2 and 3 again to find out whether this is charger board related or not23:51

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!