josch | khm: you are not planning on using the standard Debian image that comes with the reform, right? | 00:00 |
---|---|---|
khm | I am. I'm planning on using the debian image on the emmc, and I'm hoping to upstream changes to the boot script such that if a plan 9 kernel is detected it presents a menu | 00:01 |
khm | so that I can run the debian image or boot into plan 9 without fiddling with hardware | 00:02 |
josch | the debian boot.scr has nothing to do with upstream u-boot | 00:03 |
josch | it is generated by flash-kernel | 00:03 |
khm | yes, I was hoping to upstream it to *you* :) | 00:03 |
khm | the current way people do this is by putting a custom boot.scr in the emmc, but that wouldn't survive an update on the debian side | 00:04 |
josch | vagrantc: something for you to look at ^ | 00:04 |
josch | khm: why would you put a custom boot.scr onto the emmc? | 00:04 |
khm | in order to boot plan 9 | 00:07 |
josch | just throw whatever you need into /etc/flash-kernel/preboot.d | 00:07 |
josch | then flash-kernel will pick that up and put it into the boot.scr it generates | 00:07 |
josch | and would survive the updates | 00:07 |
khm | I don't see a way to invoke a bootmenu via preboot.d, since it inserts before the kernels are detected | 00:15 |
josch | that's the beauty: you would not do kernel detection at all so you do not have to re-implement that | 00:16 |
josch | you only would need to do whatever is needed to check whether plan9 is there or not | 00:16 |
josch | and then ask the user whether they want to boot plan9 or continue to linux | 00:16 |
khm | I'll give it a shot | 00:17 |
josch | and the other thing you could look into is to utilize extlinux.conf to boot plan9 -- i do not know if that can be done or not but extlinux.conf has the advantage of being declarative instead of turing-complete shell | 00:22 |
josch | and you get a menu for free | 00:22 |
khm | and, just so I'm sure: reform's u-boot does not support nvme, correct? | 00:26 |
josch | khm: i heard it should work with upstream u-boot on imx8mq | 00:27 |
josch | bluerise might know more | 00:27 |
josch | khm: but the u-boot that MNT ships do not support it unless i missed this | 00:27 |
- natalie (QUIT: Remote host closed the connection) (~natalie@user/natalie) | 00:31 | |
khm | no, that's consistent with what I see | 00:33 |
+ natalie (~natalie@user/natalie) | 00:33 | |
khm | woops, our u-boot doesn't support bootmenu (CONFIG_CMD_BOOTMENU is not set) ... probably should've checked that before I got started! | 00:49 |
khm | ok, got it working with extlinux.conf, with the provisio that u-boot-menu generates entries that won't boot -- you must add an `fdt` line so it can find the device tree | 01:31 |
josch | the fdt line is needed for linux as well | 01:34 |
josch | here is how mine looks like: https://paste.debian.net/hidden/8f4d09ff/ | 01:34 |
josch | how does an entry for 9front look? | 01:34 |
khm | like this, but it's not really fully functional right now http://okturing.com/src/19984/body | 01:56 |
khm | when not using the extlinux.conf method, we do this in boot.scr http://okturing.com/src/19985/body | 01:57 |
khm | with the extlinux.conf version, 9front doesn't find the plan9.ini, so I'll need to fix that somehow... | 01:58 |
josch | or maybe change ${ramdisk_addr_r} to 0x40010000 so that plan9.ini gets loaded into the same location as it does with your boot.scr | 02:02 |
josch | u-boot doc/develop/distro.rst lists the variables that extlinux uses | 02:02 |
josch | <= bed | 02:02 |
- cobra (QUIT: Ping timeout: 246 seconds) (~cobra@user/Cobra) | 02:18 | |
- vagrantc (QUIT: Ping timeout: 268 seconds) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 02:22 | |
- Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.16.81) | 03:16 | |
+ Ar|stote|is (~linx@149.210.16.81) | 03:16 | |
- nsc (QUIT: Ping timeout: 268 seconds) (~nicolas@150-98-142-46.pool.kielnet.net) | 03:17 | |
+ nsc (~nicolas@181-96-142-46.pool.kielnet.net) | 03:18 | |
- skipwich (QUIT: Quit: DISCONNECT) (~skipwich@user/skipwich) | 05:02 | |
+ skipwich (~skipwich@user/skipwich) | 05:05 | |
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:18 | |
- vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:39 | |
+ cobra (~cobra@user/Cobra) | 05:48 | |
+ chomwitt (~chomwitt@2a02:587:7a09:6800:1ac0:4dff:fedb:a3f1) | 06:13 | |
- GNUmoon2 (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 06:34 | |
+ GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 06:35 | |
- chomwitt (QUIT: Ping timeout: 255 seconds) (~chomwitt@2a02:587:7a09:6800:1ac0:4dff:fedb:a3f1) | 07:30 | |
josch | minute: https://source.mnt.re/reform/reform-debian-packages/-/issues is the best place to collect reform kernel-related issues? | 07:54 |
+ jaume (~user@user/jaume) | 08:06 | |
- GNUmoon2 (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 08:07 | |
+ GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 08:08 | |
minute | josch: sure, why not | 09:38 |
josch | from time to time somebody comes along and asks "i want to help, what is there to do?" and even if chances are very little, i wanted to start collect at least some of the long-standing issues | 10:10 |
josch | like flaky usb on ls1028a | 10:10 |
josch | as over time you get used to the status quo and then at least i forget about all the bugs and todo items for which we implemented workarounds... | 10:10 |
minute | josch: true | 10:12 |
- nybble (QUIT: Ping timeout: 268 seconds) (~sprang@user/csprng) | 10:16 | |
- GNUmoon2 (QUIT: Read error: Connection reset by peer) (~GNUmoon@gateway/tor-sasl/gnumoon) | 10:52 | |
+ GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 10:53 | |
- buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 10:54 | |
+ buckket (~buckket@vps.buckket.org) | 10:54 | |
- mjw (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org) | 12:08 | |
- jaume (QUIT: Remote host closed the connection) (~user@user/jaume) | 12:20 | |
* Guest3713 -> mjw | 12:23 | |
- ndufresne (QUIT: Quit: Ping timeout (120 seconds)) (~ndufresne@apple.collaboradmins.com) | 12:34 | |
+ ndufresne (~ndufresne@apple.collaboradmins.com) | 12:34 | |
- sir-photch (QUIT: Remote host closed the connection) (~m-hy5poy@2a01:4f8:1c1b:4579::1) | 12:37 | |
+ sir-photch (~m-hy5poy@2a01:4f8:1c1b:4579::1) | 12:42 | |
+ pr8dan (~pr8dan@pub151248192164.dh-hfc.datazug.ch) | 13:45 | |
- pr8dan (QUIT: Changing host) (~pr8dan@pub151248192164.dh-hfc.datazug.ch) | 13:45 | |
+ pr8dan (~pr8dan@user/pr8dan) | 13:45 | |
pr8dan | fyi, with apt-get update I am getting a "N:Missing Signed-by in the sources.list(5) entry for https:/mntre.com/reform-debian-repo" | 13:52 |
josch | pr8dan: the reason for that is that the repository is indeed not signed at all: https://community.mnt.re/t/strategy-for-apt-upgrade/2010/7 | 14:02 |
minute | https://lore.kernel.org/linux-arm-kernel/20240506124836.3621528-3-heiko@sntech.de/ | 14:42 |
- pr8dan (QUIT: Ping timeout: 260 seconds) (~pr8dan@user/pr8dan) | 14:42 | |
+ mark_ (~mjw@gnu.wildebeest.org) | 14:49 | |
josch | the rk3588 being able to power the display directly via mipi dsi without going through a hdmi adapter would be so hype! | 14:54 |
+ chomwitt (~chomwitt@2a02:587:7a09:6800:1ac0:4dff:fedb:a3f1) | 14:57 | |
- mark_ (QUIT: Ping timeout: 260 seconds) (~mjw@gnu.wildebeest.org) | 15:21 | |
- chomwitt (QUIT: Ping timeout: 272 seconds) (~chomwitt@2a02:587:7a09:6800:1ac0:4dff:fedb:a3f1) | 15:32 | |
mjw | plomlompom, minute, sorry to be a broken record, but if you could update support ticket #3488 that would be really appreciated. I am fine with you saying it is too much work to do right away, but having some estimate whether it will be days, weeks, months, would be nice. | 15:34 |
+ andreas-e (~Andreas@2001:861:c4:f2f0::c64) | 15:46 | |
+ pr8dan (~pr8dan@user/pr8dan) | 16:30 | |
pr8dan | @josch: thx | 16:31 |
minute | josch: not really because the display is eDP.... :D | 16:35 |
minute | josch: but it would be nice in pocket reform of course | 16:35 |
sigrid | luckily there is programming manual for rk3588 which describes how to drive display via various configurations, just one being hdmi | 16:50 |
sigrid | kinda hoping I will be able to figure it out when the module comes ^__^"""" | 16:51 |
- pr8dan (QUIT: Ping timeout: 256 seconds) (~pr8dan@user/pr8dan) | 17:13 | |
+ chomwitt (~chomwitt@2a02:587:7a09:6800:1ac0:4dff:fedb:a3f1) | 17:41 | |
- chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:587:7a09:6800:1ac0:4dff:fedb:a3f1) | 18:06 | |
+ mark_ (~mjw@gnu.wildebeest.org) | 18:06 | |
* mjw -> Guest9859 | 18:23 | |
- Guest9859 (QUIT: Killed (tungsten.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:23 | |
* mark_ -> mjw | 18:23 | |
+ Guest9859 (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:23 | |
+ amk (~amk@user/amk) | 18:27 | |
- amk (QUIT: Remote host closed the connection) (~amk@user/amk) | 18:34 | |
+ amk (~amk@user/amk) | 18:35 | |
+ nybble (~sprang@user/csprng) | 18:35 | |
- amk (QUIT: Remote host closed the connection) (~amk@user/amk) | 18:41 | |
+ amk (~amk@user/amk) | 18:41 | |
- amk (QUIT: Remote host closed the connection) (~amk@user/amk) | 18:46 | |
+ amk (~amk@user/amk) | 18:46 | |
+ pr8dan (~pr8dan@user/pr8dan) | 19:10 | |
minute | mjw: hey, it's already in my list > take care of mjw's reform (package is here), ticket 3488 | 19:36 |
minute | mjw: i'm gonna get to it in the coming week. i had to prioritize some critical stuff first | 19:37 |
bluerise | josch, khm: upstream U-Boot doesn't yet support Reform with NVMe | 19:37 |
bluerise | they upstreamed i.MX8MP PCIe support, which is similar to MQ and can probably be adjusted for the MQ | 19:37 |
bluerise | but even upstream Linux doesn't actually support both PCIe ports on the MQ... | 19:38 |
bluerise | (because of some ext/non-ext clock things) | 19:38 |
minute | it needs a tiny patch... that will never get accepted upstream for unknown reasons | 19:38 |
bluerise | haha | 19:38 |
bluerise | It's sad that Linux is the upstream for DT bindings | 19:38 |
bluerise | but at least there's one place with active discussion/feedback/bikeshedding | 19:39 |
minute | yeah | 19:40 |
mjw | minute, thanks | 19:40 |
Twodisbetter | minute: do you know if the rk3385 already works in the pcket reform? | 19:41 |
mjw | minute, to make it better, I also need a way to pay for your time/repairs. I only paid for the parts and shipping. So please do update the ticket with how I can get more money to you :) | 19:44 |
khm | bluerise: does upstream u-boot otherwise work? like video out, emmc/sd support, and so forth? would be nice to use bootmenu | 19:48 |
bluerise | without LCD | 19:49 |
bluerise | did have a working tree here https://github.com/bluerise/u-boot/commits/mnt/ | 19:49 |
minute | Twodisbetter: don't know yet because i couldn't test DSI yet (lack of drivers), but there's something on lkml | 19:59 |
Twodisbetter | minute: ok. Can you imagine that beast in the pocket? Would be amazing! | 20:00 |
- pr8dan (QUIT: Ping timeout: 240 seconds) (~pr8dan@user/pr8dan) | 20:02 | |
+ gustav28 (~gustav@c-0c35524e.019-141-67626730.bbcust.telenor.se) | 20:02 | |
khm | bluerise: thanks | 20:03 |
minute | Twodisbetter: yes it will be awesome (if not too hot lol) | 20:03 |
Twodisbetter | Yeah, I really can't wait to get the pocket! Great work and please thank the team for their hardwork as well! | 20:04 |
- amk (QUIT: Remote host closed the connection) (~amk@user/amk) | 20:08 | |
+ amk (~amk@user/amk) | 20:08 | |
+ pr8dan (~pr8dan@user/pr8dan) | 20:09 | |
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 20:26 | |
- pr8dan (QUIT: Remote host closed the connection) (~pr8dan@user/pr8dan) | 22:02 | |
- andreas-e (QUIT: Quit: Leaving) (~Andreas@2001:861:c4:f2f0::c64) | 22:05 | |
- gustav28 (QUIT: Quit: Quit) (~gustav@c-0c35524e.019-141-67626730.bbcust.telenor.se) | 22:15 | |
- robin (QUIT: Quit: Leaving) (~robin@user/terpri) | 22:56 | |
+ robin (~robin@user/terpri) | 23:05 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!