09:16 <srinik> aka_[m]: I see that all of the afe commands are failing since start, which sounds like some issue with firmware..

09:50 <aka_[m]> srinik: well I tried to add OSR clock but it didn't help

09:51 <aka_[m]> Much

09:51 <aka_[m]> I set multimedias to quin yet q6 routing was throwing me errors about not enabled backend dai

09:52 <aka_[m]> At some point it started throwing some errors about not being able to set amic links

09:55 <srinik> aka_[m]: I think its important to understand if "all" the APR commands are failing or not, if they are failing then there is an issue with fw, if not we can debug this further. Can I send you a debug patch that dumps APR packets.

09:56 <aka_[m]> Sure

09:56 <aka_[m]> I have some logs from downstream too

09:58 <aka_[m]> https://pastebin.com/C85awDc1

10:01 <aka_[m]> https://github.com/boffik/kernel-s2/blob/master/sound/soc/msm/msm8952.c#L117

10:01 <aka_[m]> I have added osr clock same way like ibit? Is set inside apq8016_sbc.c

10:06 <aka_[m]> Angelo was back porting stuff for 8956 on top of 4.9 maybe something here might be helpful?

10:06 <aka_[m]> https://github.com/sonyxperiadev/kernel/commits/aosp/LE.UM.2.3.2.r1.4/techpack/audio/dsp

10:06 <aka_[m]> Or forward porting not sure, whatever

11:08 <srinik> aka_[m]: Thanks, in downstream I can see all the apr commands successful. can you try this https://pastebin.com/t2yz5hhV debug patch and w can also add similar debug in downstream to understand the sequence..

11:29 <calebccff> Does anyone have any insight in Qualcomm's naming scheme for their charger drivers? Older PMICs use "smbb" (or switch mode battery (charger and) buck (converter?)), then pmi8998 uses "smb2", and pm8150 uses smb5

11:30 <calebccff> is is sensible to reflect these names upstream? This is already the case for the smbb charger

11:30 <calebccff> Or would it be better to just rename them to match the PMIC they appeared with "pmi8998_charger" for example

11:32 <calebccff> context: https://lore.kernel.org/linux-arm-msm/20220401202643.877609-1-caleb.connolly@linaro.org/T/#m7c74bfd7151aa268c4b14cf50391454be0975ddf

11:47 <Mis012[m]> pmi8998 FLAT has `SCHG8998_CHGR_BASE BASE 0x00001000 schg8998_chgraddr 31:0`

11:48 <z3ntu> Just a another datapoint, on pm7250b (used on sm6350/sm7225) also has qcom,qpnp-smb5 as charger and qcom,qpnp-qg as fuel gauge

11:48 <z3ntu> there's also "smblite" used on pm2250

11:48 <calebccff> yeah the fact that they went from smb2 to smb5 on the /next/ gen pmic is odd

11:49 <z3ntu> and that pmic is used on "scuba" SoC (which apparently is qcm2290)

11:52 <Mis012[m]> next we have `SCHG8998_OTG_BASE BASE 0x00001100 schg8998_otgaddr 31:0`

12:51 <konradybcio> At least the previous gen was just "smbcharger" (pmi8994/8996)

12:56 <konradybcio> Was the fuel gauge series accepted after all?

13:30 <calebccff> konradybcio: i don't think that's been sent?

14:00 <Tooniis[m]> <calebccff> "Or would it be better to just..." <- smbcharger appears in at least 5 PMICs

14:00 <konradybcio> calebccff: im not sure either, you have been touching many pmic related thingies, i might have confused it :p

16:44 <aka_[m]> https://pastebin.com/5r50ugC7

16:44 <aka_[m]> srinik im a bit too dumb to understand but it seems it executes some commands bit it fails on 100e5

16:47 <aka_[m]> mainline:... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/aQjocBMfPuoIrQgnhewVhVgz)

16:54 <srinik> aka_[m]: I can see that DSP is responding to APR commands that is good sign, but AFE_PORT_START command is timing out which could be possibly due to clocks not enabled for that interface.

17:03 <aka_[m]> this is kinda sad

17:03 <aka_[m]> because i set ibit/osr clocks

17:04 <aka_[m]> and q6afe-clocks supposed to set rest

17:04 <aka_[m]> but it fails in big loop of retries which also appear on 8953 but after it sorts out

17:04 <aka_[m]> if i leave apq8016_sbc.c disabled after some time clk_summary shows q6afe-clocks