| jfred | Huh. There's an M.2 Oculink adapter mentioned here... could it be possible to connect a Pocket Reform to an eGPU this way? https://www.theverge.com/2023/6/2/23747057/gpd-g1-dock-crowdfunding-announce-oculink | 00:59 |
|---|---|---|
| jfred | you'd have to route it to the outside of the case somehow of course | 01:00 |
| jfred | I'm sure the CPU would be a huge bottleneck there but it'd be kinda hilarious | 01:00 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 01:24 | |
| + mtm- (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 01:26 | |
| - mtm (QUIT: Remote host closed the connection) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:00 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:01 | |
| - mtm- (QUIT: Ping timeout: 265 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:03 | |
| - S0rin (QUIT: Ping timeout: 250 seconds) (~S0rin@user/s0rin) | 02:19 | |
| + S0rin (~S0rin@user/s0rin) | 02:22 | |
| AbortRetryFail | I've salvaged dozens of 18650 cells from old laptop batteries. They're all li-ion though not lifepo4 | 03:22 |
| dozens | dozens you say?? | 03:22 |
| AbortRetryFail | yeah, often the "high capacity" packs have 9 or 12 cells in them, when they are "dead" it's just one of the cells is gone bad | 03:23 |
| AbortRetryFail | "18650" is just the dimensions of the cell. 18mm diameter x 65mm length | 03:24 |
| AbortRetryFail | lol dammit, i am looking at weechat colors here and didn't realize i hit you with a nick hilight, sorry :) | 03:25 |
| dozens | happens all the time ;) | 03:36 |
| - nsc (QUIT: Ping timeout: 240 seconds) (~nicolas@236-98-142-46.pool.kielnet.net) | 03:39 | |
| + nsc (~nicolas@236-98-142-46.pool.kielnet.net) | 03:40 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 03:56 | |
| + mtm- (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:10 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 04:38 | |
| - eery (QUIT: Remote host closed the connection) (~eery@77.137.65.251) | 06:19 | |
| + bgs (~bgs@212-85-160-171.dynamic.telemach.net) | 07:15 | |
| + reading (~reading@174.164.41.109) | 07:56 | |
| reading | greetings, i have a few questions about the lpc firmware update process | 07:57 |
| ex-parrot | ask away reading | 07:58 |
| reading | first off, is there an artifact in the CI pipeline somewhere for the firmware.bin? ive poked around here but didnt find anything: https://source.mnt.re/reform/reform/-/tree/master/reform2-lpc-fw | 07:58 |
| reading | secondly, im getting a build error when i try to make it myself which i didnt get around a month ago and was wondering if anyone can help me debug. i see there havent been any checkins since it worked last time, so the issue is on my end. but i blew away my dev environment and am not sure what's different this time around | 08:00 |
| reading | so i was hoping for a quick download of the firmware.bin from somewhere in the repository in case it takes a while to debug my build environment. | 08:03 |
| - bgs (QUIT: Remote host closed the connection) (~bgs@212-85-160-171.dynamic.telemach.net) | 08:07 | |
| reading | as for the build error, the reform2-lpc-fw/tools/lpcrc makes fine first, but when i then try to make from reform2-lpc-fw i get the following error: | 08:12 |
| reading | LINKING bin/firmware.elf (cortex-m0 -O2 CFG_BRD_REFORM2) | 08:12 |
| reading | /usr/lib/gcc/arm-none-eabi/12.2.1/../../../arm-none-eabi/bin/ld: cannot find -lm: No such file or directory | 08:12 |
| reading | ah, gotta run for now but ill check the irc logs later in case someone happens to have any tips. thanks | 08:28 |
| - reading (PART: !!unknown attribute: msg!!) (~reading@174.164.41.109) | 08:28 | |
| - mjw (QUIT: Ping timeout: 248 seconds) (~mjw@gnu.wildebeest.org) | 13:00 | |
| * mark_ -> mjw | 13:52 | |
| - mtm- (QUIT: Ping timeout: 250 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:02 | |
| - mtm (QUIT: Ping timeout: 250 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:02 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 14:10 | |
| + robin (~robin@user/terpri) | 14:11 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 14:14 | |
| + robin (~robin@user/terpri) | 14:14 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 14:16 | |
| + robin (~robin@user/terpri) | 14:16 | |
| - robin (QUIT: Remote host closed the connection) (~robin@user/terpri) | 14:19 | |
| + robin (~robin@user/terpri) | 14:19 | |
| - robin (QUIT: Read error: Connection reset by peer) (~robin@user/terpri) | 14:30 | |
| + robin (~robin@user/terpri) | 15:03 | |
| + eery (~eery@77.137.65.251) | 16:04 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:08 | |
| josch | reading: if you are writing the logs later, let me write you now even though you are offline :) | 16:10 |
| josch | vagrantc is working on making the firmware binaries available as a Debian package: https://salsa.debian.org/vagrant/mnt-reform-firmware | 16:11 |
| + mtm- (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:11 | |
| josch | on which platform are you building this and what command are you running? | 16:11 |
| josch | i can also build (and sign with my gpg key) a lpc firmware binary for you if you'd like | 16:12 |
| + mark_ (~mjw@gnu.wildebeest.org) | 16:30 | |
| - mark_ (QUIT: Ping timeout: 265 seconds) (~mjw@gnu.wildebeest.org) | 17:11 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 17:44 | |
| josch | vagrantc: earlier a user was interested in prebuilt reform lpc firmware files -- how far is your Debian packaging of those? | 17:55 |
| - XYZ (QUIT: Remote host closed the connection) (~XYZ@89-24-41-33.nat.epc.tmcz.cz) | 18:17 | |
| + XYZ (~XYZ@89-24-41-33.nat.epc.tmcz.cz) | 18:26 | |
| vagrantc | josch: i've only built them for myself ... but it worked last i tried :) | 18:53 |
| vagrantc | josch: the whole upstream monorepo thing makes it kind of mildly frustrating to maintain over time ... but not a total deal breaker | 18:54 |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:54 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:00 | |
| * mark_ -> mjw | 19:00 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:01 | |
| - robin (QUIT: Quit: Leaving) (~robin@user/terpri) | 19:04 | |
| josch | minute: can we have git tags for the lpc and keyboard firmware? this is of course a bit weird because of the mono-repo but it would be nice if one could say "i'm running version X of the firmware" | 19:21 |
| + robin (~robin@user/terpri) | 19:23 | |
| AbortRetryFail | I'm glad I was not the only one confused about the LPC firmware versions. | 19:27 |
| minute | josch: yes, wondering how they should be formatted. iso dates? | 19:28 |
| josch | minute: i guess that depends on whether you want to have something similar to semver or just rolling releases? | 19:34 |
| josch | ideally, that version should also be displayed when pressing circle+s | 19:34 |
| josch | the current firmware has been saying 20220221 for a while | 19:35 |
| josch | that would be FW_STRING3 in reform2-lpc-fw/src/boards/reform2/board_reform2.c | 19:36 |
| josch | and KBD_FW_REV in reform2-keyboard-fw/constants.h | 19:36 |
| josch | i guess you should go with just an iso8601 based version string because semver doesn't make much sense in a mono-repo, right? | 19:37 |
| - eschaton (QUIT: Ping timeout: 248 seconds) (~eschaton@li541-49.members.linode.com) | 19:52 | |
| - jnerula (QUIT: Ping timeout: 240 seconds) (~jnerula@li1009-93.members.linode.com) | 19:53 | |
| + jnerula (~jnerula@li1009-93.members.linode.com) | 19:54 | |
| + eschaton (~eschaton@li541-49.members.linode.com) | 19:55 | |
| + nocko (~nock@user/nocko) | 20:39 | |
| + reading (~reading@174.164.41.109) | 20:40 | |
| AbortRetryFail | or change the FW_STRING3 so it gets populated at build time with the git short hash | 20:47 |
| minute | josch: yes, rolling makes more sense | 20:55 |
| minute | and it also tells you how old that is at a glance | 20:55 |
| reading | josch: im running debian unstable on a reform =). and was following the instructions from nanocodebug on july '22 in this thread https://community.mnt.re/t/lpc-spi-firmware-and-driver-prototype/1030/18 | 20:56 |
| reading | the first make in reform2-lpc-fw/tools/lpcrc worked but the second make in reform2-lpc-fw is the one that failed | 20:56 |
| reading | and actually, just searching through the history of that repository i may have just answered my own question. for some reason i thought this had been merged back upstream into the main reform tree. so i was just following these instructions but inside a clone of the main reform/reform | 21:04 |
| reading | hum. but the lpc driver was definitely been merged back into reform-tools. so does that mean the reform2-lpc-fw in reform/reform is out of sync with the NanoCodeBug/reform-lpc-fw-spi fork? | 21:13 |
| reading | hopefully that makes sense. im only now trying to ramp up on the codebase and git organization, so i may be totally confused =p | 21:17 |
| reading | let me go try buildling the firmware from nanocodebug's fork and see if that works. (ill keep checking the irc logs, thanks for responding btw) | 21:19 |
| - reading (QUIT: Quit: Client closed) (~reading@174.164.41.109) | 21:19 | |
| josch | AbortRetryFail: this makes it harder to build without the git but from release tarballs instead -- in Debian you don't have the upstream git at hand but a tarball instead | 21:30 |
| AbortRetryFail | hm, good point | 21:30 |
| minute | reading: i have recently built the fw on my computer and didn't experience your problem. let me try again | 21:32 |
| minute | reading: i am on master branch of https://source.mnt.re/reform/reform | 21:33 |
| minute | reading: arm-none-eabi-gcc (15:12.2.rel1-1) 12.2.1 20221205 | 21:34 |
| minute | reading: you probably need the package libnewlib-arm-none-eabi, maybe it's not installed? | 21:35 |
| josch | reading: to build all the reform firmware you need the following packages: avr-libc gcc-avr gcc-arm-none-eabi libnewlib-arm-none-eabi | 21:37 |
| josch | I also just successfully built from git a55fa397d18cfaa630f67bcf96dcb4b6300476b1 | 21:37 |
| josch | reading: also, if you are looking to upgrade your lpc firmware to get the battery status updates you do not need to build any fork | 21:40 |
| josch | i would like to tell you what firmware i'm running but the versioning is missing ;) | 21:41 |
| josch | minute: would you like a merge request to get the versioning thing going? | 21:41 |
| josch | minute: would you like some gitlabci thing to build the firmware so that it gets published as an artifact? | 21:41 |
| josch | hrm... but it would be nice if, when building any firmware from mnt.re git, the version would include the git hash | 21:54 |
| josch | something similar to git describe --always --dirty --tags | 21:55 |
| + reading (~reading@174.164.41.109) | 21:56 | |
| josch | vagrantc: do you know an easy way to do that such that this does not interfer with the packaging git? Maybe d/rules passes the version as read from dpkg-parsechangelog to the build to override this? | 21:56 |
| reading | minute/josch: thank you. that was it i was just missing those packages | 21:58 |
| vagrantc | josch: not quite following ... you want a firmware-specific version? | 22:02 |
| vagrantc | josch: and each firmware in theory would have it's own version ... | 22:03 |
| vagrantc | or you just use the git head ... | 22:03 |
| reading | are the full build insructions online or in the repo somewhere? id like to make sure im not missing anything else for the future | 22:03 |
| josch | vagrantc: minute said that they would just do a date-based tagging for the repo | 22:04 |
| josch | vagrantc: it would be nice if the built firmware included that tag | 22:04 |
| josch | now it would be simple to just edit the respective headers with the tag name for each commit that one puts the tag on | 22:04 |
| josch | the disadvantage of that would be that if one builds from git between two tags, this information would not be reflected in the built firmware files | 22:05 |
| vagrantc | josch: could either pass configure options to override or maybe environment variables. .. | 22:05 |
| vagrantc | i mean, in theory ... would require code changes to make it actually work | 22:05 |
| josch | as such it would be nicer if the built generated that version on the fly, for example by running "git describe --always --dirty --tags" | 22:05 |
| josch | (which is what i saw other projects run) | 22:06 |
| josch | but that would break the Debian packaging as it would pick up the tags from the packaging git | 22:06 |
| vagrantc | right | 22:06 |
| josch | so i was wondering what you are thinking about overriding that information with an argument to make, environment variable or some such | 22:07 |
| vagrantc | also, the newer trackball2 firmware requires things not in debian yet | 22:07 |
| josch | that variable could automatically be filled from the output of dpkg-parsechangelog | 22:08 |
| josch | vagrantc: do you have a trackball2? | 22:08 |
| vagrantc | no | 22:08 |
| josch | me neither, so it's not a problem ;) | 22:08 |
| vagrantc | but someone surely does... | 22:08 |
| vagrantc | and i guess they can fix it if they want it. :) | 22:08 |
| josch | well, it's certainly not a blocker | 22:09 |
| vagrantc | josch: i do not see anything wrong with passing variables or configure or make arguments or whatever for specific firmware versions | 22:09 |
| josch | oh, the pico-sdk? | 22:09 |
| josch | vagrantc: okay, then i'll propose doing it like that to minute and then send you a merge request as well once that is done (unless you want to set this up yourself) | 22:10 |
| vagrantc | josch: if you're motivated to do it, please go for it. :) | 22:10 |
| vagrantc | happy to see merge requests as long as i understand them :) | 22:10 |
| josch | it will be a simple one | 22:11 |
| josch | i'm just annoyed that pressing circle+s doesn't show me the firmware version i'm running | 22:11 |
| josch | and i'm annoyed that it already happened multiple times that people come online here and ask how they should compile the firmware -- i don't think they should have to do that | 22:12 |
| josch | so i want to fix this :) | 22:12 |
| vagrantc | good. :) | 22:13 |
| - eery (QUIT: Ping timeout: 240 seconds) (~eery@77.137.65.251) | 22:17 | |
| reading | josch: thanks for the help, i should have just come here first instead of searching the forum =). im hoping to poke around the codebase some more, are there full build instructions somewhere i can walk through? (like in case the needed packages change at some point) | 22:19 |
| josch | minute: this pipeline is stuck again: https://source.mnt.re/josch/reform/-/pipelines/953 | 22:19 |
| josch | minute: maybe you have to manually enable the CI for my fork? | 22:20 |
| josch | reading: they are here for the keyboard: https://mntre.com/reform2/handbook/parts.html#keyboard-firmware | 22:21 |
| josch | hrm... the lpc section of the handbook misses compilation instructions: https://mntre.com/reform2/handbook/parts.html#flashing-the-firmware | 22:22 |
| reading | josch: yea thats what i was following initially but the package list isnt there | 22:22 |
| reading | heh, right | 22:22 |
| josch | reading: I used this for the package list: https://salsa.debian.org/vagrant/mnt-reform-firmware/-/blob/wip/debian/latest/debian/control | 22:23 |
| reading | i only picked up the (partial/incorrect) list from forum searching | 22:23 |
| josch | if all goes well the problem should soon be fixed by a) downloadable firmware from gitlab artifacts and b) installable firmware via a signed Debian package | 22:24 |
| reading | reading: ahh very nice. yes that's perfect thank you. im new to linux development, so sorry if this was something obvious i should have known =) | 22:26 |
| reading | err, that was for josch obviously. i only talk to myself in my head | 22:27 |
| josch | reading: no worries, if nobody complains and puts their finger at a problem, things don't get fixed, so thank you! :) | 22:27 |
| minute | josch: yes, a MR to get the versioning going, for sure | 22:30 |
| minute | josch: and gitlabci artifact: yes pleaes | 22:31 |
| minute | please | 22:31 |
| reading | well, im off to flash away. thanks again for the help! | 22:31 |
| minute | reading: enjoy! | 22:31 |
| minute | i have just tested external PCIe (via special hdmi cable/adapter) on the LS1028A for the first time, works | 22:32 |
| - reading (QUIT: Quit: Client closed) (~reading@174.164.41.109) | 22:32 | |
| josch | nice!! @ LS1028A hdmi+pci :) | 22:33 |
| josch | minute: do you have an idea why my pipeline is stuck? | 22:33 |
| minute | josch: lets see, maybe it's on the wrong builder | 22:35 |
| minute | reform does not have a builder yet! | 22:37 |
| josch | it didn't have a .gitlab-ci.yml yet, so that makes sense :) | 22:37 |
| minute | josch: the builder is now assigned to your fork and to the main repo | 22:38 |
| josch | thank you!! :) | 22:39 |
| minute | it's doing something https://source.mnt.re/josch/reform/-/jobs/1243 | 22:39 |
| josch | nice :) | 22:39 |
| josch | well that was easy | 22:40 |
| josch | succeeded on the first try | 22:40 |
| minute | nice! | 22:40 |
| josch | reading left a few minutes too early XD | 22:40 |
| minute | wow, all 3 at once! | 22:41 |
| josch | lpc is missing | 22:42 |
| josch | so there was a problem | 22:42 |
| josch | pushed the fix now | 22:42 |
| josch | seems to work now -- feel free to merge https://source.mnt.re/reform/reform/-/merge_requests/41 | 22:45 |
| josch | this also documents the needed dependencies and it documents the commands needed to compile everything | 22:45 |
| minute | josch: perfect, thanks! good work! | 22:52 |
| - S0rin (QUIT: Ping timeout: 248 seconds) (~S0rin@user/s0rin) | 22:53 | |
| + S0rin (~S0rin@user/s0rin) | 22:54 | |
| josch | reading: in case you read this from the log, there are now officially built firmware files here: https://source.mnt.re/reform/reform/-/jobs/1245/artifacts/browse | 23:10 |
| - S0rin (QUIT: Ping timeout: 256 seconds) (~S0rin@user/s0rin) | 23:26 | |
| + S0rin (~S0rin@user/s0rin) | 23:27 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!