ChanServ 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 | Bridged to #msm8937-mainline:kde.org on Matrix
<barni2000[m]> 8917 clocks also different from downstream maybe i should check 3.18 also
<NekoCWD[m]> Night camss session :D
<barni2000[m]> i guess it is different rounding but idk why
<barni2000[m]> i am stopping now, maybe tomorrow
<NekoCWD[m]> btw we need to add `.link_entities = camss_link_entities` to msm8917_resources.
<NekoCWD[m]> Good night
<barni2000[m]> i have forgot to add it at rebase
<barni2000[m]> i have uploaded a fixup
veynamer[m] has joined #msm8937-mainline
<veynamer[m]> Hello, when will the Riva add support for the Goodix touchscreen? I have this model on hand, why doesn't the touchscreen work
<barni2000[m]> because not trivial to support 2 touch
<barni2000[m]> i have both
<barni2000[m]> when i have started to port i only have focaltech variant
<barni2000[m]> <veynamer[m]> "Hello, when will the Riva add..." <- for statistic which display you have?
<NekoCWD[m]> Writing this from my prada :3
<barni2000[m]> it seems land lk2nd is working now without workarounds
<barni2000[m]> i would update the kernel now so NekoCWD if you want agassi and prada in pmos send your PRs
<barni2000[m]> or they can be merged in 6.13
<NekoCWD[m]> I will send pr now
<NekoCWD[m]> barni2000which branch i shoud use?
<barni2000[m]> msm89x7/6.12.5
<barni2000[m]> can you add dmesg.txt also and squash your commits
<barni2000[m]> maybe sound stuff should be separated pr
<barni2000[m]> and flash leds
<barni2000[m]> you should open PR for panel drivers in the panel driver repo
<NekoCWD[m]> For huawei-agassi I wrote the driver myself
<barni2000[m]> NekoCWD[m]: in that case you should add it in the PR
<barni2000[m]> * in the device PR
<NekoCWD[m]> https://github.com/msm89x7-mainline/linux/pull/7 PR for flash led nodes
<NekoCWD[m]> *
<barni2000[m]> you can move prada sound and flash commits also in those PRs
<barni2000[m]> i will merge audio later
<barni2000[m]> no i will generate the panels
<barni2000[m]> * no, i
<barni2000[m]> you should only send the device dt (and docs)
hacker420[m] has joined #msm8937-mainline
<hacker420[m]> god who pinged me
<barni2000[m]> when?
<barni2000[m]> maybe it was me ages ago
<hacker420[m]> barni2000[m]: Just now
<hacker420[m]> Probably cuz Cedric in message
<hacker420[m]> anyway, any fun developments since I last looked?
<NekoCWD[m]> hacker420[m]: Flashlight and sound nodes
<barni2000[m]> yes kernel is merged in pmos and generic device package
<hacker420[m]> NekoCWD[m]: working?
<NekoCWD[m]> hacker420[m]: Yea
<hacker420[m]> Indeed a fun development
<NekoCWD[m]> Listenning music on my prada :3
<hacker420[m]> barni2000[m]: with the charging driver?
<barni2000[m]> i will merge charging driver after kernel update at pmos
<NekoCWD[m]> barni2000 is it ok? https://github.com/msm89x7-mainline/linux/pull/5
<hacker420[m]> barni2000[m]: should probably upload my DTS some day
<NekoCWD[m]> git rebase --interactive is my favourite command now :D
<hacker420[m]> hm ping me when that charger driver and battery stuff are merged in
<hacker420[m]> then I probably can just toss in my dts
<barni2000[m]> NekoCWD[m]: yes i like it also, it cause addiction
<barni2000[m]> 1. i should not be the code owner of prada, it is your work even if you copied land dt (dts will be same if you start it from scratch)
<barni2000[m]> 2. please squash your commits
<NekoCWD[m]> barni2000[m]: > <@barni2000:matrix.org> 1. i should not be the code owner of prada, it is your work even if you copied land dt (dts will be same if you start it from scratch)... (full message at <https://matrix.org/oftc/media/v1/media/download/AbFYUMAEyuoHPaSrKRIbnlO5iB9NuSyeCf8E-IuMNzYsKe8dEJ6qw5V4EV7d9S_AmRHFAwcHnQ44f2PeeCCu81VCeUiZKrnQAG1hdHJpeC5vcmcvZmZzaFlRcmJWa1JZQkZMTkdjTGlnaXpE>)
<barni2000[m]> 2. yep
<NekoCWD[m]> Fine
<NekoCWD[m]> barni2000You will make a pr to pmaports, please add my panels into initfs modules
<barni2000[m]> oh and don't forgot to sign-off your commit
<barni2000[m]> NekoCWD[m]: i will
<barni2000[m]> every kernel commit should signed-off
<NekoCWD[m]> Oh, gpg-sign != sign-off
<barni2000[m]> git commit -s
<NekoCWD[m]> Thank you
<NekoCWD[m]> Is SPDX-License-Identifier: BSD-3-Clause right?
<barni2000[m]> yes
<barni2000[m]> msm8937 and msm8940 BSD-3
<barni2000[m]> 8917 GPL-2.0-only
<barni2000[m]> lanik123: why mainline vfe clocks are differs from diwnstream?
<barni2000[m]> s/diwnstream/downstream/
<barni2000[m]> NekoCWD: did you verified everything?
<barni2000[m]> s/did/have/
<NekoCWD[m]> barni2000[m]: Yes on prada
<NekoCWD[m]> barni2000which nodes i shoud remove? I don't understand
<barni2000[m]> cmd and data
<barni2000[m]> default value is 0x10 = 16 so no need to set to 12
<NekoCWD[m]> Ok, done. Thank you
<barni2000[m]> i have find 1 more issue 😿
<barni2000[m]> s/find/found/
<barni2000[m]> no external-pfet and qcom,cabc is missing
<NekoCWD[m]> what is cabc?
<barni2000[m]> i have never seen any differences with and without it
<barni2000[m]> but downstream it has this
<NekoCWD[m]> I haven't strict checked working nodes :)
<barni2000[m]> others should be fine ig but you should check the regulators
<barni2000[m]> after i am going to merge it
<NekoCWD[m]> let's assume that the sensors have the correct regulators
<barni2000[m]> you cannot check that easily :D
<barni2000[m]> let me know when i can merge it
<NekoCWD[m]> All regulators correct
<NekoCWD[m]> barni2000you can merge prada
<NekoCWD[m]> barni2000need i add prada to dt-bindings?
<barni2000[m]> no need you only need to add it if you send it upstream
<barni2000[m]> but 8937 should be send it first
<barni2000[m]> my plan is 8937+land 8940+santoni
<barni2000[m]> i am waiting for rpm clocks to be merged atm
<NekoCWD[m]> agassi also ready to merge
<barni2000[m]> not yet
<NekoCWD[m]> Fine
<barni2000[m]> 1 more thing can you please move out panel driver from msm89x7-generated
<barni2000[m]> only generated drivers should be there
<barni2000[m]> and i will need to know which drivers should be enabled in the config
<NekoCWD[m]> barni2000moved panel
<lanik123[m]> <barni2000[m]> "lanik123: why mainline vfe..." <- What do you mean? gcc freq table?
<barni2000[m]> yes
<NekoCWD[m]> <barni2000[m]> "and i will need to know which..." <- panels i think and atmel ts
<barni2000[m]> and what about agassi touch?
<barni2000[m]> <barni2000[m]> "F( 50000000, gpll0,..." <- this
<NekoCWD[m]> barni2000[m]: CONFIG_TOUCHSCREEN_CYTTSP5=m
<barni2000[m]> ty
<NekoCWD[m]> Also add panels to modules-initfs please. I want to see pmos logo :D
<barni2000[m]> i will
<lanik123[m]> barni2000[m]: You are about diff like 177780000 and 177777778?
<lanik123[m]> s/diff/diffs/
<barni2000[m]> yes
<barni2000[m]> ig different rounding
<barni2000[m]> can it caused any issue or is it fine
<lanik123[m]> barni2000[m]: Because I was taking this frequency from the newer gcc from 4.19 kernel. QCOM used limited rounding of frequencies in msm framework due to limitations. In QCOM driver there is no problem with this and they started to write frequency 1 in 1 with datasheet xd
<barni2000[m]> NekoCWD: oh your sign-offs are not fine totally `Signed-off-by: NekoCWD <` it should contain your name
<barni2000[m]> lanik123[m]: ok thanks for the calrification
<barni2000[m]> s/calrification/clarification/
<lanik123[m]> barni2000[m]: If needed, you can freely change frequencies to the same as in 4.9
<barni2000[m]> i will send it upstream in the near future, i don't think it is needed i have just need to use same freqs in camss.c
<lanik123[m]> Btw, cpr worked with msm8917 and I fixed the psci problem
<barni2000[m]> you can send a PR or i can cherry-pick it
<lanik123[m]> lanik123[m]: Not sure about 8937 yet, cpr driver was made for cpu with 1 cluster. It might need some additionally modification
<barni2000[m]> they should be sent to upstream btw
<lanik123[m]> barni2000[m]: Ok, I'll send them later. Not sure about stability of these changes now
<barni2000[m]> does psci a big change?
<lanik123[m]> barni2000[m]: Only dts
<barni2000[m]> PSCI-0.2?
<lanik123[m]> barni2000[m]: Nope, I find more simple solution. We can delete all psci node and specify idle states directly in cpu nodes
<lanik123[m]> lanik123[m]: Like this one
<barni2000[m]> btw lk2nd don't need all 3 id `qcom,msm-id = <303 0x0>, <308 0x0>, <309 0x0>;`
<lanik123[m]> barni2000[m]: I know, I don't like to use lk2nd and this is purely for my convenience xd
<barni2000[m]> 8917 is enough for lk
<barni2000[m]> i have just say less is enough for the bootloader
<lanik123[m]> barni2000[m]: Hm, ok. ty
<barni2000[m]> does model and compatible changes are necessary ig it should boot without it
<barni2000[m]> but your tree i don't review it :D
<lanik123[m]> barni2000[m]: Now I'm using your tree, so you can forget about that old horror xd
<lanik123[m]> * your tree as base, so
<NekoCWD[m]> barni2000[Flash PR](https://github.com/msm89x7-mainline/linux/pull/7) ready.
<NekoCWD[m]> Ah, or not
<NekoCWD[m]> Now ready
<barni2000[m]> i will merge it later
<barni2000[m]> maybe you can challange it as an RFC at upstream
<NekoCWD[m]> Will try, thank you
<barni2000[m]> <lanik123[m]> "Now I'm using your tree, so..." <- rolex will conflict with rolex :D
<lanik123[m]> <barni2000[m]> "rolex will conflict with rolex..." <- Ah, your about it. I will rebase my changes later
<NekoCWD[m]> I'm writing ucm for my prada.
<NekoCWD[m]> All works fine, but if i switch audio configuration, audio routing stops :D
<NekoCWD[m]> alsa ucm + qcom audio issues. AGAIN
<NekoCWD[m]> <barni2000[m]> "https://gitlab.postmarketos.org..." <- Xiaomi redmi 4 is mido. Xiaomi 4 Standart is prada
<barni2000[m]> no
<NekoCWD[m]> fr?
<barni2000[m]> Xiaomi Redmi Note 4 is mido
<NekoCWD[m]> Ahh
<NekoCWD[m]> Sorry
<barni2000[m]> Redmi 4x is Redmi 4 in india
<barni2000[m]> it seems there is an msm8953 prada
<M0xCAFEBABE[m]> what
<M0xCAFEBABE[m]> it's more likely markw lol
<barni2000[m]> yes it should be markw
<M0xCAFEBABE[m]> barni2000[m]: fun fact: there's a MTK variant of Redmi Note 4
<barni2000[m]> yes like note 3
<barni2000[m]> but those are electrical wastes
<NekoCWD[m]> [ucm for prada](https://github.com/NekoCWD/alsa-ucm-conf )
<barni2000[m]> i will fork msm8953 repo and you can send your pr there
<NekoCWD[m]> PulseAudio configuration switching kinda laggy. I will make "UCM reload" button for sxmo :D
<barni2000[m]> pulseaudio is a mess, i should let a lot of workaround in configs when i have made msm8953 configs
<NekoCWD[m]> pa also need configs?
<barni2000[m]> no just it does not supports ucm fully
<barni2000[m]> last time i cannot set mono to mono and i could not use transitions
<M0xCAFEBABE[m]> Standart -> Standard ..?
<barni2000[m]> Redmi 4?
<NekoCWD[m]> idk, on gsmarea it was standart
<barni2000[m]> phonedb seems better a bit
<NekoCWD[m]> barni2000[m]: Updated
<NekoCWD[m]> barni2000can we add msm.prefer_mdp5=false to cmdline? mdp5 glitches sometines
<barni2000[m]> i can invert the the flag
<barni2000[m]> STK has around 40fps on prada
<NekoCWD[m]> What about touchscreen? What panel variant do you have?
<M0xCAFEBABE[m]> both are atmel
<barni2000[m]> ctc
<M0xCAFEBABE[m]> one is mostly dead
<M0xCAFEBABE[m]> * one of them is mostly
<NekoCWD[m]> barni2000[m]: Me too
<barni2000[m]> yes one of them has some issues but it works better a bit on linux then on google linux (android)
<barni2000[m]> maybe is should heat it a bit
<barni2000[m]> which sensors was added?
<NekoCWD[m]> Prada design >>> enchilada (oneplus 6)
<barni2000[m]> ltr seems not works with phosh :(
<NekoCWD[m]> ltr?
<barni2000[m]> ltr559
<NekoCWD[m]> proximity sensor. You mean auto-brightness or screen off in calls?
<barni2000[m]> phosh thinks there is no sensor
<barni2000[m]> i can get the values in sysfs so maybe just feedbackd missing something
<NekoCWD[m]> btw, I've copied sensors from markw. Maybe there are something in 8953 generic package
<barni2000[m]> i have not added any sensor related stuff for msm8953
<barni2000[m]> soc package only contains udev rule for haptic