2024-02-19.log

hramrachThere used to be something specialized for css, now it's greasemonkey00:04
hramrachThe derk mode extension tries to do something generic, not sure how well00:05
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50)00:05
wakestwas using this firefox invert colors plugin pretty successfully for a long time https://addons.mozilla.org/en-US/firefox/addon/invert-colors/01:24
- jacobk (QUIT: Ping timeout: 240 seconds) (~quassel@utdpat241106.utdallas.edu)01:33
- Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149.210.4.126)01:36
jackhillshould I worry about learning about FIT, or is that an "advanced" topic that I can safely put off until later (I'm going throught th u-boot documentation)01:40
- mjw (QUIT: Ping timeout: 256 seconds) (~mjw@gnu.wildebeest.org)01:40
+ Ar|stote|is (~linx@149.210.4.126)01:48
+ mrdaught (~mrdaught@50.237.127.178)01:52
- vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50)02:08
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)02:10
+ mrdaught (~mrdaught@50.237.127.178)02:17
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)02:19
+ mrdaught (~mrdaught@50.237.127.178)02:29
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)02:31
+ mrdaught (~mrdaught@50.237.127.178)02:48
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)02:52
+ mrdaught (~mrdaught@50.237.127.178)03:11
- mrdaught (QUIT: Ping timeout: 264 seconds) (~mrdaught@50.237.127.178)03:15
jackhilljosch: how does u-boot know where to look for an extlinux.conf?03:24
- nsc (QUIT: Ping timeout: 252 seconds) (~nicolas@103-99-142-46.pool.kielnet.net)03:48
+ nsc (~nicolas@62-97-142-46.pool.kielnet.net)03:49
+ mrdaught (~mrdaught@50.237.127.178)04:11
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)04:19
- Bertl (QUIT: Ping timeout: 260 seconds) (herbert@IRC.13thfloor.at)04:47
+ jacobk (~quassel@129.110.242.173)05:37
+ mrdaught (~mrdaught@50.237.127.178)06:12
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)06:12
joschjackhill: it's the distro_boot mechanism which checks the devices u-boot knows about in a pre-defined order06:20
joschjackhill: the order is essentially set at compile-time06:20
joschfor the reform u-boot blobs, it's usb, sd-card, emmc06:21
- jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@129.110.242.173)06:54
+ jacobk (~quassel@129.110.242.173)07:04
elektron07:05
- jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@129.110.242.173)07:09
+ jacobk (~quassel@129.110.242.173)07:35
- bkeys (QUIT: Remote host closed the connection) (~Thunderbi@45.134.140.153)07:53
+ mrdaught (~mrdaught@50.237.127.178)08:08
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)08:27
+ mrdaught (~mrdaught@50.237.127.178)08:29
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)08:30
- jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@129.110.242.173)08:47
+ jacobk (~quassel@64.189.201.150)09:19
+ mrdaught (~mrdaught@50.237.127.178)09:54
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)09:55
+ mrdaught (~mrdaught@50.237.127.178)10:23
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)10:24
+ mjw (~mjw@gnu.wildebeest.org)10:34
- Ar|stote|is (QUIT: Ping timeout: 255 seconds) (~linx@149.210.4.126)10:40
+ Ar|stote|is (~linx@149.210.4.29)10:45
hramrachand you can override that in u-boot environment. the 'distroboot' is a script from the enevironment that u-boot automatically executes, if you change it in the environment something else gets executed11:33
- mjw (QUIT: Ping timeout: 255 seconds) (~mjw@gnu.wildebeest.org)11:51
* Guest4857 -> mjw12:15
minutei did some experiments with different wifi cards on ls1028a and there is an MSI issue12:51
minutei have an intel card working now12:51
minutebut if you disable MSI globally via pci=nomsi kernel option, that prevents ethernet from working on ls1028a, because it internally uses PCIe+MSI for that12:53
minutebut i found that you can disable MSI per PCIe bus!12:53
minuteso you can do:12:53
minuteecho 0 > /sys/devices/platform/soc/3400000.pcie/pci0001:00/0001:00:00.0/msi_bus12:55
minuteyou can unload the card:12:55
minuteecho 1 > /sys/devices/platform/soc/3400000.pcie/pci0001:00/0001:00:00.0/0001:01:00.0/remove12:55
minute(or probably rmmod)12:55
minuteand rescan:12:56
minuteecho 1 > /sys/devices/platform/soc/3400000.pcie/pci0001:00/0001:00:00.0/rescan12:56
minuteso now i can use iwlwifi + ls1028a native ethernet together12:56
joschhuh12:57
minuteoh wow https://community.nxp.com/t5/Layerscape/Can-t-get-MSI-interrupts-on-ls1046a-rdb/td-p/135434012:58
minuterelated!12:58
minutebecause i got these errors with msi enabled for the card:12:59
minute[   12.346286] OF: /soc/pcie@3400000: no iommu-map translation for id 0x100 on (null)12:59
minute[   12.354245] iwlwifi 0001:01:00.0: enabling device (0000 -> 0002)12:59
minute[   12.408278] OF: /soc/pcie@3400000: no msi-map translation for id 0x100 on /interrupt-controller@6000000/msi-controller@602000012:59
minute> The problem should be u-boot PCIe fixup didn't configure msi-map for this card12:59
minuteso uhhhm what fixup is that12:59
minuteplot thickens https://www.mail-archive.com/barebox@lists.infradead.org/msg47672.html13:01
minuteapparently u-boot has to map DMA stream ids for pcie devices and put them in the device tree?!13:05
+ mrdaught (~mrdaught@50.237.127.178)13:32
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)13:32
senzillaminute: I am looking into supporting an effort around porting Tianocore EDK2 (native UEFI) for the iCore-3588Q. I have been deploying quite few arm64 + EDK2 systems in the last few years, including ARM ServerReady (SBSA) systems. There is a lot of porting effort on-going here: https://github.com/edk2-porting/edk2-rk358814:03
minutesenzilla: oh, that sounds great14:04
minutesenzilla: how can i help?14:04
senzillaminute: The ultimate end goal is obviously to be able to run MNT Reform with EDK2. But for the initial port for iCore-3588Q using the reference carrier board AIO-3588Q. Any thoughts on this?14:04
minutesenzilla: sounds good. i have multiple options on the adapter for driving the display in descending order of preference: eDP, DSI, HDMI14:05
minutei see in the list that manual eDP config is possible?14:05
minutethat would be great14:06
minutesenzilla: otherwise, my adapter is modeled after what i had working on AIO14:06
senzillaI have not looked into the detail of the display driver, but it's good to know your preference here!14:06
senzillaminute: Ah cool, you used the AIO board for some initial work?14:07
minutesenzilla: yes, see here https://files.mastodon.social/media_attachments/files/111/649/125/777/931/758/original/a91dc4344c7bfa89.jpg14:08
senzillasweet! I missed that post somehow :)14:08
minutefor reference, the post url is https://mastodon.social/@mntmn/11164913113994994814:08
senzillathanks!14:08
senzillastarred! :D14:09
minutesenzilla: in this proto, i'm using an HDMI to eDP adapter to drive the display14:10
senzillaminute: I am in touch with the lead developer for the EDK2 porting effort, and they are interested in working on this. So I am about to buy and donate a AIO-3588Q board to the lead developer of the EDK2 porting effort, explicitly for this purpose.14:12
senzillaminute: you will be able to follow along here if you are interested: https://github.com/edk2-porting/edk2-rk3588/issues/12414:12
senzillaWhile I intend to run OpenBSD on this in the future, you might be able to run Windows on the MNT Reform after this :P14:16
senzillaSorry :D14:16
minutesenzilla: haha, well, that might be interesting to some14:19
minuteit would certainly be a strange, but nevertheless interesting milestone14:19
minutesenzilla: also, extremely cool @ your support, thanks a ton!14:20
senzillaminute: well, thank you for building an awesome platform for this! :)14:20
senzillathe background for me here is that I want to build that server / network appliance I was talking about last week (the one with all the PCIe lanes :)) . And ideally I want it to be classified as "ARM ServerReady" (SBSA) for which native UEFI and ACPI is needed14:22
senzillaThe UEFI shim that U-Boot provides does not suffice in this case...14:23
- aphistic (QUIT: Ping timeout: 246 seconds) (sid347194@id-347194.ilkley.irccloud.com)14:35
+ aphistic (sid347194@id-347194.ilkley.irccloud.com)14:35
- violet (QUIT: Ping timeout: 260 seconds) (~vi@user/violet)14:46
- svp (QUIT: Ping timeout: 246 seconds) (sid537750@id-537750.uxbridge.irccloud.com)14:47
+ svp (sid537750@uxbridge.irccloud.com)14:54
- yuu (QUIT: Ping timeout: 260 seconds) (sid267332@id-267332.ilkley.irccloud.com)14:55
+ yuu (sid267332@id-267332.ilkley.irccloud.com)14:55
- yuu (QUIT: Ping timeout: 256 seconds) (sid267332@id-267332.ilkley.irccloud.com)15:03
+ yuu (sid267332@id-267332.ilkley.irccloud.com)15:03
+ violet (~vi@user/violet)15:03
- brennen (QUIT: Ping timeout: 268 seconds) (~brennen@user/brennen)15:12
minutesenzilla: i see, that makes sense15:14
- yuu (QUIT: Ping timeout: 246 seconds) (sid267332@id-267332.ilkley.irccloud.com)15:23
+ brennen (~brennen@user/brennen)15:27
+ yuu (sid267332@id-267332.ilkley.irccloud.com)15:42
- jackhill (QUIT: Ping timeout: 268 seconds) (~jackhill@kalessin.dragonsnail.net)16:20
+ mrdaught (~mrdaught@50.237.127.178)16:24
+ jackhill (~jackhill@kalessin.dragonsnail.net)16:25
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)16:28
+ Bertl (herbert@IRC.13thfloor.at)16:31
jackhilljosch: hramrach: thanks!16:48
jackhillany chance of there being another run of the i.MX8MPlus module?16:54
- jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@64.189.201.150)17:03
joschjackhill: you mean for the big reform?18:04
- bleb (QUIT: Ping timeout: 264 seconds) (~cm@user/bleb)18:06
jackhilljosch: yeah18:06
+ bleb (~cm@user/bleb)18:07
elektronACTION looks their way...18:07
joschif I remember correctly (but minute can correct me) the plan was to not offer the imx8mp modules until the pocket reforms have shipped to make sure there are enough imx8mp for the pocket first. After that, you will probably be able to buy an imx8mp from MNT but in the past minute also has said that to not make the configuration matrix too large, the big reform with imx8mplus might not get official 18:08
joschsupport.18:09
jackhillI really like the simple low power cores of the original module, as well as USB3 and dual display. But more RAM and the potential for hardware video encode of the Plus make it the most attractive of all the upgrades currently available.18:09
joschi would expect that if the community (you?) send patches, then having support for big reform + imx8mplus in the MNT provided kernel will not be a problem18:11
jackhillcool, I'll wait for confirmation, but that sounds reasonable and probably makes sense for me. Had the reform on the shelf for a little bit, but getting back into it now (so much fun!). Sort of unofficial like the rpi module? When it was origially announced I thought I should hold off on upgrades until I was actively working with reform.18:11
jackhilljosch: cool, coordinating on kernel, device tree, and u-boot (if needed) soudns reasonable. I hope to run not-debian on it anyway, so am not too worried about full official image. I assume much of the work is already done for pocket?18:13
joschabout the official image: if somebody vouches to maintain the kernel patches, i will volunteer to make reform-tools support the configuration18:14
joschthis of course does not help you because you'll not be running debian but maybe it does help others :)18:14
joschit should not differ much from what has to be added anyways for the pocket reform, i expect18:15
jackhillI'm probably not (yet!) knowledgeable enough to be solely responsible for kernel work, but happy to learn, test, and report problems.18:15
- colinsane (QUIT: Ping timeout: 264 seconds) (~colinunin@97-113-159-4.tukw.qwest.net)18:22
+ colinsane (~colinunin@97-113-159-4.tukw.qwest.net)18:24
+ jacobk (~quassel@utdpat242016.utdallas.edu)18:26
- jacobk (QUIT: Ping timeout: 240 seconds) (~quassel@utdpat242016.utdallas.edu)18:34
+ jacobk (~quassel@129.110.241.106)18:39
+ mark_ (~mjw@gnu.wildebeest.org)19:14
- jacobk (QUIT: Ping timeout: 268 seconds) (~quassel@129.110.241.106)19:51
+ jacobk (~quassel@utdpat241106.utdallas.edu)19:56
+ mrdaught (~mrdaught@50.237.127.178)20:02
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)20:04
+ mrdaught (~mrdaught@50.237.127.178)20:10
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@50.237.127.178)20:14
- elektron (QUIT: Quit: Changing server) (~elektron@dom0.halo.nu)20:17
+ elektron (~elektron@dom0.halo.nu)20:22
- elektron (QUIT: Client Quit) (~elektron@dom0.halo.nu)20:23
+ elektron (~elektron@dom0.halo.nu)20:23
minutejosch: thoughts? https://source.mnt.re/reform/reform/-/merge_requests/6720:31
+ mrdaught (~mrdaught@2607:fb90:9fab:971c:b1fa:f50a:5696:9a4)20:41
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9fab:971c:b1fa:f50a:5696:9a4)20:43
+ grimmware (uid638840@id-638840.hampstead.irccloud.com)21:07
grimmwarehello :)21:08
minutehi!21:13
- jacobk (QUIT: Ping timeout: 260 seconds) (~quassel@utdpat241106.utdallas.edu)21:18
+ mrdaught (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)21:18
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)21:20
grimmwareminute: how's it going?21:26
joschminute: i think ultimately i'd like a firmware downloader and flasher to be integrated into reform-tools21:27
joschminute: but maybe having it there is a good first step?21:27
+ mrdaught (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)21:35
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)21:42
minutejosch: i agree! but also it's a bit of a special case because you have to run this on another computer that might not be a reform21:44
minutegrimmware: good good! working on mnt reform handbook 2nd edition today21:44
minutei connected my almost empty reform to a lab supply. it charges with 77W21:46
minutewoops21:46
grimmwareI'm so excited to have a computer with an actual handbook21:46
minutenice :321:46
grimmwareThe last time I used a handbook for a computer it was for an Apple II that my dad found in the attic (with no recollection of ever owning one)21:47
grimmwareI took the handbook to a hacker camp and built a serial -> keyboard interface with a microcontroller and then took it home and it just worked with the actual machine21:47
minutewow, nice21:48
+ mrdaught (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)22:01
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)22:01
+ mrdaught (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)22:18
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)22:20
joschminute: clearly they should just buy a second reform then ;)22:26
joschmaybe download-fw.sh should be put into flash.sh? It's just a few lines22:27
+ mrdaught (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)22:35
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)22:36
digitalrayneTIL: UEFI for phones (after wondering what EDK2 is)22:46
digitalraynetruly we are living in the future22:46
minutejosch: sure, if you think that's good practice lets integrate them22:46
joschminute: i think it's a balance. Many tiny scripts where it's not clear which script does what versus one huge mega script that is too big and too complex to be maintainable. The true path is probably somewhere in the middle.22:48
+ vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20)23:00
vagrantchrm. heard a startling spark-like sound and then noticed one of the cells on the mnt/reform was reading 0v ... maybe time to finally put the upgraded battery boards in :/23:01
vagrantcno immediately obvious damage or shorts23:01
joschanother screw came loose in mine yesterday. This time from the trackball module. About the only screws that i hadn't fixed with loctite yet...23:02
joschi managed to get it out while it was only rolling around on the bottom acrylic before it was able to touch anything23:03
minuteuh oh23:03
joschthe screws of the trackball are riiiight next to the batter contacts -- a miracle :)23:03
minutewe should offer those nylon dispensed screws in a bag maybe23:03
minutewe already use them in production now23:03
joschi'd put some in my next order -- never used that kind before23:04
joschon the other hand, given how much i open mine up and put it back together, it's kinda expected that things break every once in a while :)23:04
joscha luxery problem to have in a way, compared to hardware that one never takes apart23:05
+ mrdaught (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)23:05
vagrantcreading the voltage on the 0v cell was up to 0.67v or so by the time i got around to it ... not sure if i should try to replace that individual cell or recharge it individually and then put it all back together or what23:06
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f8e:55c3:1069:19b1:a9c7:25a9)23:06
vagrantcalthough regardless, time to find where i put the upgraded boards :)23:07
vagrantcand upgraded trackball ... been actually using it more and it could stand a little improvement with motion23:07
vagrantcACTION rummages around for the manual23:09
joschvagrantc: how do you like the white switches of trackball v2?23:11
vagrantcdon't know yet, it's still sitting in the package ...23:12
vagrantcor is this actually the trackball v1, just with the ballbearing upgrade?23:12
vagrantcjust looks like a replacement cup, so i'm guessing that is the only difference23:13
vagrantcok, so i pull all cells out one-by-one ... any opinion on which order?23:13
joschfirst disconnect the boards from the mainboard and then start from the inside so that you can get under them with your finger more easily i think?23:14
vagrantcoh, that's contrary to the listed instructions23:15
joschoh?23:15
joschwhat do those say?23:15
vagrantcor at least the ifixit instructions23:15
vagrantchttps://www.ifixit.com/Guide/MNT+Reform+Battery+Boards+Replacement/15918923:15
vagrantcmaybe i should stick to the officiallly official instructions rather than just referenced ones ? :)23:16
joschuh interesting23:16
joschi think the last thing minute reminded me of that the first thing to do after opening the acrylic is to disconnect the battery boards23:16
vagrantci'll trust you on that :)23:17
vagrantcthe battery board cables have a locking connector?23:18
joschi don't think so, i just pull on them23:18
vagrantcwow, these clips sure hold things in nice and strong.23:20
+ mrdaught (~mrdaught@2607:fb90:9f10:1b48:8dc9:97e8:fd7d:e9e4)23:22
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f10:1b48:8dc9:97e8:fd7d:e9e4)23:24
vagrantccell 8 seems to have been very dead ... i plugged it into a charger and it is not responding at all23:26
vagrantcshuld look up the replacement trackball cup instructions ... maybe do that while the battery boards are disconnected23:28
joschi would only replace the trackball module with the battery boards removed because the screws are so close to the battery clips which means one is very likely to touch the battery clips with the (metal) screw driver...23:31
vagrantci don't need to replace the module, just the cup ... but yeah23:31
vagrantcnot finding any obvbious instructions for the cup replacement23:32
joschoh your reform is that old that it has one of the very old cups?23:33
minutevagrantc: they absolutely have a locking connector23:33
minutedon't just rip em out. the way is to squeeze both ends on the side basically23:34
joscho023:34
joschwell, i have done it wrong all the time...23:34
minutehttps://www.molex.com/en-us/products/connectors/wire-to-board-connectors/pico-lock-connectors23:35
vagrantcminute: point taken for next time :)23:35
joschthey really can be pulled out easily though... :/23:36
joschcan't believe i never realized...23:36
minutehttps://youtu.be/OyDk9BFLknE?si=o6Is0pXQi0FduS5323:37
joschvagrantc: you see it's a miracle my reform is still alive after all this mistreatment XD23:40
vagrantchrm. struggling to get the screws into the new cup ... probably should be dong this on a well lit clean surface, not just hovering over my lap or something23:40
vagrantcjosch: a stellar review right there!\23:41
joschvagrantc: i think you were offline when i had the fireworks when removig the keyboard bezel without having disconnected the battery boards first. Batteries were fine but my wifi card couldn't connect to any wifi anymore.23:42
joschvagrantc: that's when minute told me that the rule is: "whenever you open it up, disconnect the battery boards first"23:43
joschnow i realize i did even that wrongly XD23:43
+ mrdaught (~mrdaught@2607:fb90:9f10:1b48:8dc9:97e8:fd7d:e9e4)23:43
+ Nixkernal_ (~quassel@2a02:1210:1613:e600:cd68:645a:f17c:8f85)23:46
- mrdaught (QUIT: Read error: Connection reset by peer) (~mrdaught@2607:fb90:9f10:1b48:8dc9:97e8:fd7d:e9e4)23:46
- Nixkernal (QUIT: Ping timeout: 255 seconds) (~quassel@2a02:1210:1613:e600:fdd7:bb07:fde0:c3d0)23:46

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!