oldgalileo has quit [Remote host closed the connection]
oldgalileo has joined #tegra
marvin24 has quit [Ping timeout: 252 seconds]
marvin24_ has joined #tegra
oldgalileo has quit [Remote host closed the connection]
oldgalileo has joined #tegra
oldgalileo has quit [Remote host closed the connection]
oldgalileo has joined #tegra
gouchi has joined #tegra
<DavidHeidelberg>
what is best approach: bridge binding now has `port@0` and `port@1` inside the main node, not the `ports`, should I write DTS according to that, or send patch for moving ports to the `ports` node inside?
oldgalileo has quit [Remote host closed the connection]
oldgalileo has joined #tegra
<DavidHeidelberg>
digetx quick check if you recall, grouper-common has rt5640 audio-codec, but no interrupt assigned, in original txt file there was interrupt marked as required, other DTS(i) has it, should it be really required, since audio worked just fine on Nexus
oldgalileo has quit [Remote host closed the connection]
oldgalileo has joined #tegra
gouchi has quit [Remote host closed the connection]
gouchi has joined #tegra
<digetx>
DavidHeidelberg: it's valid to have single port without the 'ports' node, perhaps you can remove @0 if it's the problem
<DavidHeidelberg>
(I added second port, since it has two ports)
<DavidHeidelberg>
... not sure if connected
<digetx>
there should the ports node in glide
<digetx>
interrupt is controversial on n7, it should be routed to tegra in accordance to schematics, but I couldn't validate it because it doesn't work, so maybe it's not
<digetx>
you can make interrupt optional
<DavidHeidelberg>
ok, thanks. I'll check sii9234 driver if it's able to accept ports { port@} instead of just port
<digetx>
the glide developer is active, he got merged regulator driver used by glide to upstream recently, you may ask him to test it
oldgalileo has quit [Remote host closed the connection]
<DavidHeidelberg>
digetx great, thanks! I'll do.
oldgalileo has joined #tegra
<DavidHeidelberg>
after some fine-tuning of CI, I'm currectly around 600 lines of important warnings
<DavidHeidelberg>
digetx I asked Sergey if he can test the Glide change and send MR
<DavidHeidelberg>
for dtbs job, now CI produces Critical warnings, where are omited not converted txt -> yaml and shows only warnings related to existing stuff
gouchi has quit [Remote host closed the connection]
gouchi has joined #tegra
<digetx>
kwizart: the question is who needs those patches in the first place, IIUC tagr wants to have some tests, but IGT may be more suitable for that
<digetx>
it's not an option for unstable userspace to use upstream libdrm because always there is something that needs to be changed
<DavidHeidelberg>
digetx what you think about just this? Since I realized, that not all warnings are printed to github warnings in Summary, what about keep there numbers and just compare with previous build?
<digetx>
better to turn warning into error and fix it; if there are too mane warnings for now, then fix all warnings and enable errors after
<digetx>
too many warnings for now
<DavidHeidelberg>
yes, if I run it on tegra_defconfig with all "missing yaml", it's still about 3000 even with maaaany patches
<DavidHeidelberg>
on grate_defconfig without "missing dt-schema" it's about 530 now
<DavidHeidelberg>
I think that'll be first reachable goal
<DavidHeidelberg>
then I'd switch warnings (without missing yaml warnings) to errors and missing yamls into warnings
<DavidHeidelberg>
still it's hard to figure out, if someone introduced new lines
<DavidHeidelberg>
I could eventually use GH action cache to save outputs and then diff against new, but I think that's more complexity than sending patches for getting to 0
<DavidHeidelberg>
jonasschwoebel are u ok with adding compatible = "thine,thc63lvd103", "lvds-encoder"; ?