Danct12 changed the topic of #msm8937-mainline to: Boot Linux on your MSM8917/37/40 and QM215 mobile! | GitHub: https://github.com/msm89x7-mainline | Logs: https://oftc.irclog.whitequark.org/msm8937-mainline
travmurav[m] has joined #msm8937-mainline
<hacker420[m]> ok I seem to have sudo back
<hacker420[m]> did some zapping and reinstalls
<hacker420[m]> also made dedicated linux and device packages for mainline
<hacker420[m]> to minimize fuckery with downstream
Daanct12 has joined #msm8937-mainline
<hacker420[m]> hm
<hacker420[m]> so if I build the linux package with --envkernel sudo breaks
<hacker420[m]> what
<hacker420[m]> if I use a commit from my github kernel fork outside of envkernel it's fine
<hacker420[m]> do I have some kernel config option that breaks it or what
<hacker420[m]> ok I removed regulators for iris from msm8917.dtsi and it works now
<hacker420[m]> ???
<hacker420[m]> I did git reset hard on the cloned kernel
<hacker420[m]> whatever, it seems to work now
<hacker420[m]> god that usb port is just horrible
<hacker420[m]> lanik123: what are these hacks for?
luka177[m] has quit [Quit: Client limit exceeded: 20000]
bstaja[m] has joined #msm8937-mainline
<hacker420[m]> so right now I have not working wifi
<hacker420[m]> or bt
<hacker420[m]> I feel like the regulators have nothing to do with it
<hacker420[m]> because it works on lenovo
<hacker420[m]> * on lenovo k5
<barni2000[m]> What about reserved-memory sections?
<barni2000[m]> Are you sure you have wcn3620?
<hacker420[m]> what else could I have on 8937
<barni2000[m]> Maybe you can try 3660b compatible
<barni2000[m]> Idk, i just guessing
<hacker420[m]> barni2000[m]: guess I could screw around with those
<barni2000[m]> Check the remoteproc nodes also
<hacker420[m]> downstream
<hacker420[m]> mainline
<barni2000[m]> Downstream is disabled
<barni2000[m]> Check where it has been enabled
<hacker420[m]> no
<hacker420[m]> mainline is disabled
<hacker420[m]> and I enable it back in the DTS
<barni2000[m]> Oh sry i miss read
<barni2000[m]> Btw wcn3620 will need a commit for not crashing
<hacker420[m]> hm
<hacker420[m]> what commit
<barni2000[m]> A moment
<barni2000[m]> Nvm i think you should have if the fork based on msm8916-mainline
<hacker420[m]> link it anyway
<hacker420[m]> just to be sure
<barni2000[m]> But it is only needed when it is already working
<barni2000[m]> Without it the kernel will be panicing
<barni2000[m]> s/panicing/panicking/
<hacker420[m]> yeah the fix is there
<barni2000[m]> Is there any debug print in the module?
<barni2000[m]> Maybe worth to enable it
<hacker420[m]> there is debugfs support
<hacker420[m]> but it doesn't output anything interesting
<barni2000[m]> As far as i know downstream can write out the wcn register valu for check what you have
<barni2000[m]> *log out
<barni2000[m]> Check the iris reg value
<barni2000[m]> Wifi/Bt
<barni2000[m]> I think it could 3660b
<barni2000[m]> * it could be 3660b
<hacker420[m]> I'll see late
<hacker420[m]> * I'll see later
Daanct12 has quit [Quit: WeeChat 4.1.1]
<hacker420[m]> hm
<hacker420[m]> could I check it in schematics?
<hacker420[m]> <barni2000[m]> "I think it could 3660b" <- Huh
<hacker420[m]> if the schematics are correct, it is
<hacker420[m]> how do I modify the compatible only for my device though?
<hacker420[m]> something like that?
<hacker420[m]> hm now
<hacker420[m]> did that overwrite the whole iris node
<hacker420[m]> or is something else at fault
<jojo_autoboy[m]> hmm i wonder what wcn my land uses
<jojo_autoboy[m]> took it apart but i'm somehow unable to find the chip
<jojo_autoboy[m]> all i see is a "wtr2965"
<hacker420[m]> jojo_autoboy[m]: > <@jojo_autoboy:matrix.org> hmm i wonder what wcn my land uses
<hacker420[m]> > took it apart but i'm somehow unable to find the chip
<hacker420[m]> you checked under the metal shields?
<jojo_autoboy[m]> duh
<jojo_autoboy[m]> schematics don't mention the wcn model ether
<hacker420[m]> weird
<hacker420[m]> integrated in SoC?
<hacker420[m]> or somehow renamed
<hacker420[m]> mmm lovely fw vulns
<hacker420[m]> jojo_autoboy[m]: you tried to search on IC sites for a wifi chip for that model
<jojo_autoboy[m]> yeah tried that
<jojo_autoboy[m]> mentions WCN 3615
<jojo_autoboy[m]> hm and this
<jojo_autoboy[m]> oh wait i think i found where the chip is
<hacker420[m]> this also lists 3615
<jojo_autoboy[m]> sigh
<hacker420[m]> nice metal bro
<jojo_autoboy[m]> wait what
<jojo_autoboy[m]> so it's not under there
<jojo_autoboy[m]> where the hell is it
<jojo_autoboy[m]> oh ok now i see where it must be
<jojo_autoboy[m]> hmmmmmm
<jojo_autoboy[m]> yeah i have no clue on what this uses
<hacker420[m]> try that wcnss reg thing
<hacker420[m]> <barni2000[m]> "https://wiki.postmarketos.org/..."; <- ^
<jojo_autoboy[m]> darn twrp doesn't give that info
<hacker420[m]> yes because it has no wifi :p
<hacker420[m]> so if I put 3660b in the iris compatible it times out
<barni2000[m]> <hacker420[m]> "&pronto {..." <- > <@hacker420:kde.org> ```... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/VyijALhDwKhJvnWoZgvXZgtq>)
<hacker420[m]> I replicated what msm8916 does
<hacker420[m]> in 8917 dtsi
<barni2000[m]> wcn3615 should work with wcn3620 xiaomi ysl and onclite has 3615
<hacker420[m]> maybe I should remove that compatible?
<barni2000[m]> Yes and use &wcnss_iris
<hacker420[m]> nope
<hacker420[m]> maybe I really do need the regulators
<hacker420[m]> lemme add those
<hacker420[m]> Hah
<hacker420[m]> think I got it
<hacker420[m]> wifi connected
<hacker420[m]> works
<hacker420[m]> pings fine
<hacker420[m]> was also able to connect to my ouya controller
<jojo_autoboy[m]> cursed
<hacker420[m]> yes
<hacker420[m]> it happily receives events, checked via evtest
<hacker420[m]> hello sirs
<hacker420[m]> gomuks works nicely
<hacker420[m]> and this screenshot is proof my controller is recognized
<hacker420[m]> need to fix volume buttons
<Danct12[m]> nice
<hacker420[m]> now waiting for lanik to upstream his gpu stuff
<hacker420[m]> then I'll try bringing up the panel
<jojo_autoboy[m]> i'm waiting for pm8937 charging/fg stuff to be implemented
<hacker420[m]> someone ping me when the stuff needed for proper panel support lands
<jojo_autoboy[m]> heh land
<hacker420[m]> yes
<hacker420[m]> wait until land support lands in mainline
<f_> Danct12: you have +o
illen[m] has quit [Quit: Client limit exceeded: 20000]
<lanik123[m]> lol what
<lanik123[m]> They got root without any modifications on Pixel 7
<lanik123[m]> <hacker420[m]> "https://github.com/Lanik123/..."; <- This hack for old wlan fw(like on rolex) to get compatibility with mainline wcnss
<lanik123[m]> <hacker420[m]> "https://github.com/Lanik123/..."; <- This hack fix tx/rx error log spam in dmesg
<lanik123[m]> I'm not sure they are needed on the new kernel version
<Danct12[m]> the second half has been upstreamed
Danct12[m] has quit [Ping timeout: 480 seconds]
jojo_autoboy[m] has quit [Ping timeout: 480 seconds]
travmurav[m] has quit [Ping timeout: 480 seconds]
lanik123[m] has quit [Ping timeout: 480 seconds]
bstaja[m] has quit [Ping timeout: 480 seconds]
barni2000[m] has quit [Ping timeout: 480 seconds]
minecrell[m] has quit [Ping timeout: 480 seconds]
hacker420[m] has quit [Ping timeout: 480 seconds]
lanik123[m] has joined #msm8937-mainline
Danct12[m] has joined #msm8937-mainline
travmurav[m] has joined #msm8937-mainline
barni2000[m] has joined #msm8937-mainline
minecrell[m] has joined #msm8937-mainline