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
YuzheSong[m] has joined #msm8937-mainline
<dr_shui[m]> there seems like xiaomi-ugg have a usable wcnss at https://wiki.postmarketos.org/wiki/Xiaomi_Redmi_Note_5A
<dr_shui[m]> same msm8940 as xiaomi-santoni
<dr_shui[m]> but that's weird that i can't found ugg's .dts
hitechshell[m] has joined #msm8937-mainline
<lanik123[m]> <barni2000[m]> "it is missing" <- I didn't add it because I wasn't sure it was correct
<lanik123[m]> <barni2000[m]> "his fork should be rebased i..." <- Yep, I'll do it after I get a working display on sdm439
<hacker420[m]> huh another issue
<hacker420[m]> <lanik123[m]> "I didn't add it because I wasn't..." <- well on danct's fork it works fine
<dr_shui[m]> Are you still working on santoni? Danct12
<hacker420[m]> <lanik123[m]> "Yep, I'll do it after I get a..." <- hang on
<hacker420[m]> isn't it basically a 430 with higher clocks
<hacker420[m]> or is there some other difference
<lanik123[m]> hacker420[m]: sdm439 use own 12nm pll driver
<hacker420[m]> ah
<hacker420[m]> so that's where the difference lies
<hacker420[m]> also mind helping me with that display driver?
<hacker420[m]> without any regulators and wled it didn't do anything on screen
<f_> Danct12: Change the topic here
<hacker420[m]> <hacker420[m]> "```..." <- > <@hacker420:kde.org> ```... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/zmDLfLYAwsElyFmfPXGiWYIt>)
<lanik123[m]> hacker420[m]: It should work without regulators, maybe problem more deeper
<hacker420[m]> well
<hacker420[m]> I get a completely black screen
<hacker420[m]> with the regulators and wled I get backlight but the probe is deferred
<lanik123[m]> hacker420[m]: Without any backlight
<hacker420[m]> yes
<hacker420[m]> completely off
<hacker420[m]> I assume it should at least display the kernel log
<f_[mtrx]> test
<hacker420[m]> without needing to get to the DE
<hacker420[m]> f_[mtrx]: pong
<f_[mtrx]> broken :(
<f_[mtrx]> Did you get rid of /topic hacker420? Also why is there no op in here?
<f_[mtrx]> Finally, who's f__???
<hacker420[m]> f_[mtrx]: no I didn't
<lanik123[m]> hacker420[m]: This is a problem in PM or defconfig, I don't have any ideas how to fix it yet
<lanik123[m]> lanik123[m]: Same with pwm on msm8917
<hacker420[m]> well any idea how to run it without wled and regulators
<f_[mtrx]> then gomuks is completly broken.
<hacker420[m]> unless I am missing something in defconfig
<lanik123[m]> hacker420[m]: Probably not at all, it depends on how phone PM part is made
<hacker420[m]> would schematics help anything here
<lanik123[m]> No, you already know that you cannot get working bl without regulators xd
<hacker420[m]> lol
<hacker420[m]> so guess I wait until you figure out why the probe gets deferred on enabled regulators
<hacker420[m]> maybe it's the same thing as 8916
<hacker420[m]> where you need wcnss up
<lanik123[m]> hacker420[m]: I can try to find solution in msm8953 or msm8916 mainline rep. I don't think we're the first who faced this problem.
<hacker420[m]> > Note: Due to weird dependencies issues in the DRM MSM driver, you need to make WiFi/Bluetooth work to have the DRM MSM driver finish loading successfully. (WCNSS takes a while to load, so it will trigger attempting to probe DRM MSM again after it fails to load on its own due to dependencies issues between the DRM driver and the panel driver...)
<lanik123[m]> hacker420[m]: lol what
<hacker420[m]> lanik123[m]: ask qualcomm
<travmurav[m]> ^ is not an issue if you explicitly load modules in order: panel, msm
<hacker420[m]> hm what if I have both built in
<travmurav[m]> tough luck /s
<hacker420[m]> do I switch them to modules then
<travmurav[m]> i think there is a bootparam to extend the deferral timeout
<hacker420[m]> might test that
<travmurav[m]> coudl try that
<travmurav[m]> deferred_probe_timeout=30 I htink
<hacker420[m]> yes
<lanik123[m]> lanik123[m]: As I remember it was wled driver have deferred probe before drm
<travmurav[m]> travmurav[m]: though the default for this seems to have a long story of being changed around so who knows...
<hacker420[m]> <hacker420[m]> "```..." <- > <@hacker420:kde.org> ```... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/fCkTNZcZGygnRoYBvCfMCcvy>)
<travmurav[m]> seems like it was "forever -> 30 sec -> 10 sec -> 0 sec -> 30 sec -> forever" so who knows xD
<hacker420[m]> lol
<hacker420[m]> guess could be worth a shot
<hacker420[m]> but what do I set it to
<travmurav[m]> but fwiw we ship msm as a module on 8916
<travmurav[m]> can have panels buil-in if it's easier
<travmurav[m]> probably 30 sec should be enough
<travmurav[m]> I think cros sets it to 30 on modern qcom devices
<lanik123[m]> s/have//
<lanik123[m]> s/I/U/
<hacker420[m]> well 2 minutes in and nothing
<hacker420[m]> think I might try the modules idea
<hacker420[m]> building msm and panel as module
<hacker420[m]> <travmurav[m]> "^ is not an issue if you..." <- how do I set them to load in order?
<hacker420[m]> cmdline?
<travmurav[m]> hacker420: in pmOS there are special files for modprobe, and modules are loaded in-order
<hacker420[m]> ah
<hacker420[m]> you have an example?
<travmurav[m]> (it was deviceinfo prop for a long time but they moved them out)
<barni2000[m]> <hacker420[m]> "[ 0.000000] OF: reserved mem:..." <- > <@hacker420:kde.org> ```... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/OJvFdoZUnzwCwjFGzxQLrfeq>)
<hacker420[m]> in a bit let me first test the modules
<hacker420[m]> something like this? travmurav
<travmurav[m]> yep
<hacker420[m]> well so far now I am stuck at the lk2nd screen
<hacker420[m]> oh
<hacker420[m]> just now blanked out with backlight
<hacker420[m]> maybe I need to somehow pipe the dmesg out to the screen?
<hacker420[m]> for now it's still blank with backlight on
<hitechshell[m]> IMHO, if you bringup simple framebuffer, you can display dmesg on it
<hitechshell[m]> deviceinfo_kernel_cmdline="PMOS_NO_OUTPUT_REDIRECT PMOS_NOSPLASH clk_ignore_unused pd_ignore_unused earlycon console=tty0"
<hitechshell[m]> and add framebuffer node to device tree
<hacker420[m]> hitechshell[m]: well yes
<hacker420[m]> that I already know
<hacker420[m]> the thing is I am trying to bring up the actual panel now
<hacker420[m]> with the driver
<hitechshell[m]> sorry, i not yet ready all messages)
<hitechshell[m]> s/ready/read/
<hacker420[m]> hm
<hacker420[m]> any idea how someone else can install the kernel package into the rootfs that's built with pmbootstrap
<hacker420[m]> I don't mean sideloading to a running device
<jojo_autoboy[m]> chroot?
f_[mtrx] has quit []
f_[mtrx] has joined #msm8937-mainline
<hacker420[m]> lanik123: why did you disable the modem here?
<hacker420[m]> <barni2000[m]> "> <@hacker420:kde.org> ```..." <- nope
<hacker420[m]> instantly dies
<hacker420[m]> wait
<hacker420[m]> nope
<hacker420[m]> it does not like having it removed
<barni2000[m]> wait i will check in downstream
<hacker420[m]> well removing it doesn't work so...
<barni2000[m]> i don't find the qseecom in downstream only the other-ext
<barni2000[m]> why ramoops using the same adress like watchdog?
<hacker420[m]> because it has the same address afaik
<barni2000[m]> strang
<barni2000[m]> * strange
<barni2000[m]> what happens if you only keep the removed-dma-pools ranges from msm8937.dtsi the reserved-memory part in downstream cedric dt is strange for me a bit
<barni2000[m]> s/cedric/motorol-common.dtsi/, s/dt//
<hacker420[m]> barni2000[m]: hm?
<barni2000[m]> did you find this header moto-mem-reserve.h?
<hacker420[m]> yes