ChanServ changed the topic of #linux-msm to:
irungentoo has quit [Quit: ZNC 1.8.2+deb3.1 - https://znc.in]
irungentoo has joined #linux-msm
marvin24_ has joined #linux-msm
marvin24 has quit [Ping timeout: 480 seconds]
<aka_[m]> aklimov: mclk should be internal
<aka_[m]> I guess
<aka_[m]> Will give a spin
<aka_[m]> Tho not sure part of rx1 you wrote
<aka_[m]> Setting rx1 into mm1 produces a lot of cracking around and some fail at the end
<aka_[m]> Going to try muxing rx0 to int2_1 for sure
<aka_[m]> As usual will try in 7h
jhovold has joined #linux-msm
srinik has joined #linux-msm
pespin has joined #linux-msm
<aka_[m]> aklimov: ds playback tinymix
<aka_[m]> muxing both INT0 and INT2 into RX still results in no playback on aux
<aka_[m]> guess i will have tinymix soon from mainline
<aka_[m]> uh tinymix calls for card parameter which does not happen on android
<aka_[m]> tinymix contents from mainline:
xtex has quit [Read error: Connection reset by peer]
xtex has joined #linux-msm
srinik has quit [Ping timeout: 480 seconds]
<aka_[m]> <aklimov> "aka_: sorry for delays from my..." <- i don't really understand interconnection between these but it looks like there are few "ports" like INT0 INT1 INT2 and these each can have muxed-in RXs
<aka_[m]> But now how mapping is between rx-macro INTs<->Soundwire<->WCD
<aka_[m]> qcom,swr-port-mapping seems to contain something but if there is rx0-rx5 then there could be 6 ports exposed to soundwire
<aka_[m]> tho by checking INT stuff it seems RXes are configured well, INT0 takes RX0 and INT2 takes RX1
<aka_[m]> So if i playback on RX1 shouldn't it enable CLK?
<aka_[m]> RX0 does
srinik has joined #linux-msm
<aklimov> aka_[m]: rx_macro could have 6 ports but maybe not all of are connected, just 5 are used
<aklimov> aka_[m]: which clk? rx-macro internal clk?
<aka_[m]> CLK_EN inside LPASS_RX_CDC_RXy_RX_PATH_CTL
<aka_[m]> 0x0a600480
<aka_[m]> its only enabled on RX0
<aklimov> if i am reading it correct, offset 0x480 in rxmacro, then downstream shows that content of that reg is 0x4 during playback and it is the same with mainline kernel
<aka_[m]> disturbing kinda
<aklimov> interesting, my dumps don't have 0x24 there
<aka_[m]> you can see RX0 and RX1 which are muxed into INT0 and INT1
<aklimov> aka_[m]: did limiting number of channels to 1 help? i think i tried it and there no changes
<aka_[m]> nope no change
<aklimov> i should probably join matrix
<aka_[m]> tho in my ds tinymix it shows single channel
<aka_[m]> probably
<aka_[m]> i guess you get links to msgs
<aka_[m]> so on headphones i get RX0 and RX1 CLK enabled and both channels work
<aka_[m]> on earpiece+speaker its RX0 and RX1 with CLK enabled
<aka_[m]> *RX2
<aklimov> the 0x500 on mainline -- i have 0x24 there
<aka_[m]> would be good to know how its on DS playback
<aka_[m]> Headphone and earpiece+speaker path
<aklimov> the same as with downstream kernel
<aklimov> yeah, i don't have it on qrb4210, only speaker output
<aka_[m]> so on DS during aux output RX1 is no clk?
<aka_[m]> and tinymix has RX INT2_1 MIX1 INP0 on RX1?
<aklimov> nah, my downstream commands are
<aklimov> # tinymix set "RX_CDC_DMA_RX_1 Channels" "One"
<aklimov> # tinymix set "RX_MACRO RX2 MUX" "AIF2_PB"
<aklimov> # tinymix set "RX INT2_1 MIX1 INP0" "RX2"
<aklimov> # tinymix set "AUX_RDAC Switch" 1
<aklimov> # tinymix set "SpkrMono WSA_RDAC" "Switch"
<aklimov> # tinymix set "RX_CDC_DMA_RX_1 Audio Mixer MultiMedia1" 1
<aklimov> # tinymix set 'RX_RX2 Digital Volume' 80
<aklimov> it should go into rx2/AIF2_PB
<aklimov> from CDC (dsp) RX_1 side
<aklimov> and if I am not mad, it is rx3 on wcd side
<aklimov> each RX<number> is "local" to each device
<aka_[m]> so we have it a bit different
<aklimov> heh
<aklimov> as far as i understood, people say that one can use any combination of CDC RX<number> <--> RX<number> on rxmacro side
<aklimov> there are just some "best" practices/convention people try to follow
<aka_[m]> aklimov: something like this for me?
<aka_[m]> in first line RX1 could be also INT1
<aklimov> yes!
<aklimov> Looks good
<aklimov> aka_[m]: so there are ports from rx macro to soundwire and ports on wcd side and this mapping should be set correctly?
<aka_[m]> no idea tbh how it works with ports
<aka_[m]> RX also have ports based on dts
<aka_[m]> and these seems to be same for wcd
<aka_[m]> so mapping is 1-1
<aklimov> aka_[m]: and within each port there are "channels" like hph_l, hph_r, clsh, comp_l, comp_r, lo, dsd_r, dsd_l ?
<aka_[m]> probably
<aka_[m]> each port can prob have 2 channels
<aka_[m]> maybe im connecting wrong ports somehow
<aklimov> yeah, i am not sure that it is all correct here too
<aklimov> but what i don't understand that i was told that people verified everything via RX0 to RX0/AIF1/INP2_1 on rx macro and to RX3 on wcd, it works
<aka_[m]> maybe they have some expensive setups
<aklimov> maybe time to read downstream DTs and sources to understand how channels/ports map
<aklimov> aka_[m]: oh, i need to share some debug data with you, i forgot
<aka_[m]> a39.skl@gmail.com
<aklimov> yeah, i saw it in ucm files
srinik has quit [Remote host closed the connection]
srinik has joined #linux-msm
pespin has quit []
jhovold has quit [Quit: WeeChat 4.3.2]
srinik has quit [Ping timeout: 480 seconds]
jhovold has joined #linux-msm
jhovold has quit [Ping timeout: 480 seconds]