uvos | Wizzup: http://uvos.xyz/maserati/stockinfo/call | 00:21 |
---|---|---|
Wizzup | great | 00:21 |
Wizzup | interesting to see 0006 for 081c | 00:22 |
uvos | which one? | 00:22 |
uvos | handset has 0x1 | 00:23 |
uvos | as expected | 00:23 |
uvos | ah speakerphone | 00:23 |
uvos | right | 00:23 |
uvos | yeah thats the real value for it | 00:23 |
uvos | 0x0 is invalid | 00:23 |
uvos | it should never be 0, its a mux 0 makes no sense | 00:24 |
Wizzup | I've seen it at 0002 | 00:25 |
uvos | Wizzup: right android is enabling the left speaker too | 00:27 |
uvos | even though it dosent exist | 00:27 |
uvos | this is harmless | 00:27 |
uvos | i think its because the code base is shared and xt912 has 2 speakers | 00:28 |
uvos | also mz6xx | 00:28 |
uvos | anyhow muteing the mic in android changes CPCAP_REG_TXI | 00:29 |
Wizzup | I see | 00:29 |
uvos | maybe check what it is in leste | 00:29 |
Wizzup | when not in calls: | 00:30 |
Wizzup | # cat /sys/kernel/debug/regmap/spi0.0/registers | grep 0814 | 00:30 |
Wizzup | 0814: 0084 | 00:30 |
Wizzup | compared to 0cc0 in android? | 00:31 |
uvos | 0cc6 | 00:31 |
uvos | and 0cc0 muted | 00:31 |
Wizzup | right | 00:32 |
Wizzup | hm 0828 is also diff for headset vs speakerphone | 00:33 |
Wizzup | there is also CPCAP_REG_TXMP | 00:33 |
uvos | ok | 00:34 |
Wizzup | its mic gain | 00:34 |
Wizzup | I guess? | 00:34 |
uvos | so the bit its setting in 0814 is ATXINEN | 00:35 |
uvos | that makes sense (mic amp enable) | 00:35 |
uvos | but we have that set | 00:35 |
uvos | and "ATXOUTEN Reserved for output TXOUT enable, currently not used" | 00:35 |
uvos | i gues this has maybe a different meaning in cpcap | 00:36 |
uvos | than mc13783 | 00:36 |
Wizzup | yeah | 00:36 |
Wizzup | hm CPCAP_REG_TXI also changed for headset and speakerphone | 00:37 |
uvos | im sure they change all the gains and such | 00:38 |
uvos | i would just place a call on leste | 00:38 |
uvos | and then sync up the headset register values with android untill it works | 00:38 |
uvos | thats basicly what i did | 00:38 |
uvos | and there was just one bit or so that was really nessecary that was missing | 00:39 |
uvos | s/headset/handset | 00:39 |
uvos | (besides 081c) | 00:39 |
Wizzup | ok | 00:40 |
Wizzup | I think the left/right being set to the right mic also helps with echo cancel somehow? | 00:43 |
uvos | no idea | 00:43 |
Wizzup | I think I synced up most in the 0x8* range | 00:47 |
Wizzup | echo 0804 60df > /sys/kernel/debug/regmap/spi0.0/registers | 00:47 |
Wizzup | this one I think | 00:47 |
Wizzup | from 60cf to 60df | 00:48 |
Wizzup | at least that seems to make the difference | 00:48 |
Wizzup | speakerphone also works with that | 00:49 |
Wizzup | interesting | 00:49 |
uvos | speakerphone with 081c at 0 has really wierd behavior | 00:49 |
uvos | if you mess with the other regs | 00:49 |
Wizzup | anyway maybe try this: | 00:50 |
Wizzup | echo 0804 60df > /sys/kernel/debug/regmap/spi0.0/registers | 00:50 |
uvos | CPCAP_REG_CC | 00:50 |
Wizzup | codec | 00:50 |
uvos | so what bit makes the difference? | 00:50 |
Wizzup | 60cf vs 60df | 00:50 |
uvos | so bit 5 | 00:51 |
Wizzup | yes | 00:51 |
Wizzup | sorry it took me a bit to remember bin() in python | 00:52 |
Wizzup | so is that CPCAP_BIT_AUD_LOWPWR_SPEED ? | 00:53 |
Wizzup | wait nvm | 00:53 |
uvos | CPCAP_BIT_MIC1_CDC_EN | 00:54 |
uvos | right | 00:54 |
Wizzup | yeah 0 indexed | 00:54 |
uvos | that makes sense :) | 00:54 |
Wizzup | ACD left/right ? | 00:54 |
uvos | so do we set CPCAP_BIT_MIC1_CDC_EN when in hifi mode? | 00:54 |
Wizzup | adc* | 00:54 |
uvos | or is this a indipendant bug | 00:54 |
Wizzup | we always set it to 0 at least in cpcap.c | 00:55 |
uvos | ok | 00:55 |
uvos | so its just a oversigt then | 00:55 |
uvos | let me check what android dose | 00:56 |
Wizzup | you mean the old kernel src? | 00:56 |
uvos | whilest recording ti cpu | 00:56 |
Wizzup | ok | 00:56 |
Wizzup | if you're on it, maybe you can also check out the headphone setup, although I guess that maybe just works | 00:56 |
Wizzup | let me try that | 00:56 |
uvos | i dont have a headset with mic | 00:57 |
uvos | and android can tell | 00:57 |
uvos | (ie it will use the internal mic in this case) | 00:57 |
Wizzup | right | 00:58 |
uvos | i dont get what register CPCAP_REG_CC is in mc13783 datasheet | 01:00 |
uvos | (also why the heck dose the datasheet not give register values) | 01:00 |
Wizzup | is this the same ic? | 01:00 |
uvos | this annoys me every time | 01:00 |
uvos | no | 01:01 |
uvos | but some hw blocks are the same | 01:01 |
uvos | the audio block is one of those | 01:01 |
Wizzup | can't reproduce it right now, I think I'm too tired, but that bit was the last one I changed and it made the diff | 01:09 |
uvos | Wizzup: ok | 01:13 |
uvos | i added recording to cpu register state | 01:13 |
Wizzup | is that in handset? | 01:18 |
Wizzup | or speakerphone? | 01:18 |
Wizzup | looks like a big difference in regs :p | 01:19 |
Wizzup | most things except for 0x0814 | 01:19 |
uvos | thats recording from mic1 | 01:27 |
uvos | and no audio output | 01:27 |
Wizzup | ok | 01:30 |
buZz | xt921? | 10:54 |
buZz | oh 912* | 10:55 |
buZz | looks nice but where's the qwerty! :D | 10:55 |
Wizzup | is that the razr? | 10:56 |
Wizzup | right | 10:56 |
buZz | yeah, Droid Razr | 10:59 |
buZz | Also known as DROID HD | 10:59 |
buZz | hmmmm https://www.ebay.com/itm/194077446917 | 11:03 |
buZz | next wednesday getting a pinephone :P | 11:04 |
buZz | want to see how many things that are in 'trick' or 'things to try' on various pages we could move to 'leste wide' pages | 11:04 |
dreamer | oooh, nice dock | 11:08 |
buZz | dreamer: so wish we could find these cheaper :) | 11:08 |
dreamer | hmhm. wish that seller I found that claimed to have a bunch was actually true :( | 11:12 |
Wizzup | I have one here but it doesn't seem to work | 11:15 |
buZz | Wizzup: oh? not even for charging? | 11:25 |
Wizzup | buZz: no, but maybe the psu is dead | 12:42 |
Wizzup | (I didn't bother to try to replace it yet, but iirc I measured it and it worked) | 12:42 |
* enyc meows :O | 23:58 | |
enyc | sorted out so much in my life but not found my n900 charger lately ;/ | 23:58 |
enyc | Curiosu how general interest and so-on going up... | 23:59 |
enyc | whose taking up leste, not just those into older n900 etc devices, anyhow? | 23:59 |
enyc | is there a user-survey of any form? | 23:59 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!