09:13 <brads> Trying to get the pcm5122 driver connected to i2s pins (tdm_b) on Odroid C4, the i2c part is connecting and I have defined the pins in tdmif_b section (tried with mclk and without as it supports LR clock sync) but I cannot get the pins to dance at all. Am I likey to need to "TOACODEC" component to make this work? Im trying to work in in conjuction with HDMI and at the moment devicetree

09:13 <brads> https://pastebin.com/Z0eQMvA6 but I have played around quiet a bit with different settings

09:15 <brads> maybe I need to do something like for sei610? https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/arch/arm64/boot/dts/amlogic/meson-sm1-sei610.dts?h=v5.9.1

09:16 <brads> but unsure if I can run both HDMI and the i2s pins both on tdm_b?

09:58 <narmstrong> brads: no you can have both on the same tdm

10:11 <brads> ok I will test some more changes

13:01 <narmstrong> brads: so 1) HDMI should be moved to TDM_A, and you should add your device to TDM_B

13:56 <brads> narmstrong: ok thanks I will give that a try again (I did yesterday actually but it appeared those pins on TDM_A were reserved for spi, ill take a closer look

13:56 <narmstrong> brads: I'm also trying on my side

14:10 <brads> narmstrong: thanks, this is my latest version of the dts https://pastebin.com/YBC8VZnr Im not sure I have the right clock assigned to the pcm5122 but I don't think that matters at this stage as I cannot seems to get output on the tdm_b pins

14:12 <brads> I suspect I have some wrong pads configure for mclk to in the tdmif_b defintion

14:12 <narmstrong> it should be AUD_CLKID_TDM_MCLK_PAD0

14:13 <narmstrong> and you'll need to play with dai-tdm-slot-tx-mask

14:14 <brads> ok

14:45 <brads> narmstrong: im getting much closer now, hdmi is moved to TDM_A but now have to have a closer look at the masks for TDM_B and make sure I can configure the mixers properly in alsamixer as im seeing " fe.dai-link-0: ASoC: no backend DAIs enabled for fe.dai-link-0" trying to use the i2c. Thanks for you help :)

14:46 <brads> s/i2c/i2s/

17:24 <narmstrong> brads: this could help you set the mixers https://usercontent.irccloud-cdn.com/file/y0cSFNAx/SM1%20ALSA%20Mixers%20(1).png

17:55 <jbrunet> brads: here is an example connecting 8 mics to the 40pins header of vim3. You should be able to figure out how to adapt this to your use case on the oc4

17:55 <jbrunet> https://pastebin.com/YtJjjCy9

17:59 <jbrunet> And no, you don't need to bother with acodec, this is the internal dac of the sm1.

18:04 <jbrunet> "fe.dai-link-0: ASoC: no backend DAIs enabled for fe.dai-link-0" - This means that you did not setup a valid route between and an actual output ... Asla cannot consume the data and is nicely asking you to do something about it

18:05 <jbrunet> "but unsure if I can run both HDMI and the i2s pins both on tdm_b?" : you can.

18:07 <jbrunet> either both run from TDM_B: hdmi from internal pins, your codec from the physical ones

18:07 <jbrunet> Or: TDM_A for HDMI, TDM_B for your codec, both connected to the same FRDDR

18:46 <narmstrong> I tried on my side with a WM8731 on the GPIOAO tdm_b pins on the C4 J7 connector, the mixers are ok, but I have nothing on the pins at all (no sclk, no lrclk, no data...), but the pinctrl is ok

21:47 <brads> narmstrong, jbrunet: Thanks for the detials, ill be looking at this again later on. So far I haven't seen anything on the pins with an Oscilloscope. I still have some more things to try (ie the masks) and I will double check the C4 schematics / datasheet re the pins and also the aml firmware / uboot im currently using. JTAG or somthing maybe.

21:48 <brads> Seems to be pulling some pins high on boot that I did not expect so maybe a werid configuration there, I can probably start checking registers and the bsp kernel to see what is already set.