| 00:00.31 | *** join/#htc-linux riotz (riotz@unaffiliated/riotz) |
| 00:06.12 | WisTilt2 | mgross029: did you try booting that kernel again by chance and see if hci0 shows up when you try to enable BT? everything looks like its working but userland doesn't think its enabled still. |
| 00:24.13 | *** join/#htc-linux furtardo (~mks@nat/yahoo/x-axpvzvueusxtuivf) |
| 00:38.28 | *** join/#htc-linux mitsutaka (~mitsutaka@rt.miraclelinux.com) |
| 00:56.10 | d3tul3 | WisTilt2, i have accel working in GB with .39 |
| 00:57.49 | WisTilt2 | ? |
| 00:58.06 | d3tul3 | g-sensor |
| 00:58.14 | d3tul3 | don't tell me that's already working for you somehow |
| 00:59.13 | WisTilt2 | ah, nice. im on my last couple of hairs to pull out with this damn BT. everything appears to be setup and working but userland still thinks it didn't get enabled. |
| 01:00.32 | *** join/#htc-linux riotz (riotz@u.cant.drop.my.firewall.made.with.packet-filter.org) |
| 01:00.33 | *** join/#htc-linux riotz (riotz@unaffiliated/riotz) |
| 01:12.09 | WisTilt2 | going home and regroup on this BT later. look forward to the g-sensor's d3tul3 good job. later all. |
| 01:12.57 | d3tul3 | WisTilt2 it was a one-liner in userland, not quite equal to the BT task |
| 01:16.18 | *** join/#htc-linux Regulator (~duffman@unaffiliated/regulator) |
| 01:20.21 | *** join/#htc-linux hardwalker (~hardwalke@122-117-115-146.HINET-IP.hinet.net) |
| 01:24.48 | *** join/#htc-linux AstainHellbring (AstainHell@unaffiliated/astainhellbring) |
| 01:27.26 | arrrghhh | d3tul3, that's some fine work there |
| 01:32.59 | d3tul3 | arrrghhh, all i see these days is [acl] pinging you somewhere |
| 01:33.17 | d3tul3 | nand taking over your life? |
| 01:33.31 | arrrghhh | lol |
| 01:33.36 | arrrghhh | pretty much man |
| 01:33.47 | arrrghhh | i haven't really been on WinMo/HaRET for a while now |
| 01:34.04 | arrrghhh | i came back to test some things for a day... then back to NAND. |
| 01:34.31 | arrrghhh | jONpRY, lol |
| 01:34.42 | arrrghhh | just thinking of you. always wonder why you hate it so. |
| 01:34.46 | arrrghhh | all the risk is gone now |
| 01:35.01 | arrrghhh | the dirty days of mapping are gone |
| 01:35.18 | jONpRY | cruise control |
| 01:35.19 | arrrghhh | well, at least i think they are. wistilt2 seems to want to do uboot and completely rid the phone of the winmo bootloader. |
| 01:35.52 | arrrghhh | which sounds cool, i just worry about returning to winmo. |
| 01:35.57 | arrrghhh | seems there would be no return to winmo... |
| 01:36.30 | jONpRY | might be able to boot winmo |
| 01:36.44 | arrrghhh | well |
| 01:36.47 | arrrghhh | the only reason i care? |
| 01:36.51 | arrrghhh | is warranty work |
| 01:37.00 | arrrghhh | i pay $7/mo for that insurance poop |
| 01:37.11 | jONpRY | i'm sure you can flash it back |
| 01:37.14 | arrrghhh | and if i claim it stolen, i gotta shell out $100 |
| 01:37.16 | arrrghhh | how tho? |
| 01:37.26 | jONpRY | spl thingy |
| 01:37.36 | arrrghhh | i thought wistilt2 wanted to rid the phone of that. |
| 01:37.49 | arrrghhh | eliminating the whole nbh flasher doohickey |
| 01:37.53 | jONpRY | no i don't think so |
| 01:38.00 | arrrghhh | hrm |
| 01:38.02 | jONpRY | just getting rid of tinboot |
| 01:38.10 | arrrghhh | maybe i don't understand it. quite possible. |
| 01:38.11 | jONpRY | and the msflash that makes it work |
| 01:39.15 | d3tul3 | jONpRY, made any progress on QDL? |
| 01:39.24 | jONpRY | my understanding is that if you could get msflash loaded and running. it will boot anything loaded on the second nbh partition |
| 01:39.27 | jONpRY | dur no |
| 01:41.02 | jONpRY | <PROTECTED> |
| 01:41.15 | arrrghhh | lol |
| 01:41.19 | arrrghhh | i'm not a masochist |
| 01:41.30 | arrrghhh | i just want the ability to jump ship back to stock if the need ever arises |
| 01:41.49 | arrrghhh | dunno what they'd even give me at this point |
| 01:41.51 | arrrghhh | probably not a TP2 |
| 01:41.55 | jONpRY | there is no need to get rid of spl as phone without radio is lamo |
| 01:42.15 | arrrghhh | oh, yea i forgot about that... |
| 01:42.28 | arrrghhh | well, now i'm just more confused. oh well. |
| 01:44.21 | jONpRY | thats android for ya |
| 01:44.28 | arrrghhh | heh |
| 01:44.39 | arrrghhh | part of the problem is i'm a simpleton, but that's another story. |
| 01:45.24 | jONpRY | hmmm |
| 01:45.42 | jONpRY | maybe you need cruise control |
| 01:45.53 | arrrghhh | hahaha |
| 02:01.32 | *** join/#htc-linux riotz (riotz@unaffiliated/riotz) |
| 02:01.37 | *** join/#htc-linux Cass (~Cass@188-220-34-222.zone11.bethere.co.uk) |
| 02:12.14 | *** join/#htc-linux TheXev (~user@93.sub-174-255-65.myvzw.com) |
| 02:18.27 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com) |
| 02:37.02 | *** join/#htc-linux TheXev (~user@93.sub-174-255-65.myvzw.com) |
| 02:38.07 | *** join/#htc-linux WisTilt2 (~wisgreg@wireless248.wirelesstcp.net) |
| 02:41.37 | arrrghhh | WisTilt2, how goes it sir |
| 02:42.03 | WisTilt2 | hey arrrghhh:) you've been quiet lately. keeping you busy at work i take it? |
| 02:42.26 | arrrghhh | yea, the call center project has really picked up |
| 02:42.37 | arrrghhh | i've also been on the dark side for... jeez over a month now i think. |
| 02:42.50 | WisTilt2 | dark side as in nand? |
| 02:42.54 | arrrghhh | yessir |
| 02:43.26 | WisTilt2 | lol. well ill be there myself when we get everything working with our u-boot setup. winmo no mo |
| 02:44.38 | WisTilt2 | had dinner and some wine so going to take another stab at this pita BT problem |
| 02:45.06 | arrrghhh | yea, i've been reading about the saga |
| 02:46.56 | WisTilt2 | so you running nand on your 400 or you get another device? |
| 02:49.19 | arrrghhh | same old 400 |
| 02:50.03 | arrrghhh | so uboot - are you doing that instead of LK/tinboot? or are you actually replacing the winmo bootloader where you flash nbh's? |
| 02:50.56 | WisTilt2 | winmo is totally gone. with the whole flash unlocked there's a lot of room to stick just about anything in there. |
| 02:51.18 | arrrghhh | could you revert to winpoo? |
| 02:51.29 | arrrghhh | only reason i'd want to would be warranty stuff. i still pay for that :P |
| 02:51.56 | WisTilt2 | should be able to flash that all back in. |
| 02:52.10 | arrrghhh | cool |
| 02:52.13 | arrrghhh | so you want a 400? |
| 02:52.22 | arrrghhh | i saw you might for dumping/testing purposes |
| 02:54.02 | WisTilt2 | acl said he already has all that cdma stuff mapped. dont know what all he has though. we dumped the gsm flash so we have the full image, just need to identify each boundary but should be all that difficult. |
| 02:54.41 | arrrghhh | it's working quite well with what he's setup |
| 02:54.44 | arrrghhh | lk/tinboot |
| 02:54.45 | arrrghhh | whatever the hekc |
| 02:54.47 | arrrghhh | heck* |
| 02:54.47 | WisTilt2 | there's a ton of free space on the flash but as it comes it is locked pretty tight |
| 02:54.48 | arrrghhh | lol |
| 02:54.57 | arrrghhh | interesting |
| 02:56.07 | arrrghhh | ah well, i'm gonna go eat some dinner. i'll probably be back on later.... good luck with BT! |
| 02:56.41 | WisTilt2 | on our device we dev'd we have 4 partitions setup so we are testing with the same way we did that. so far we booted .39 np but lots to change so it will be awhile im sure. |
| 02:56.59 | WisTilt2 | ok, nice talking to you. we'll be on awhile until BT gets the best of me. |
| 02:57.07 | arrrghhh | cool |
| 02:57.25 | arrrghhh | yea, ACL said he did change a lot of stuff for .27 to boot... |
| 02:57.34 | arrrghhh | i'm sure you've seen his tree tho |
| 02:57.53 | WisTilt2 | nope probably should see what he's done |
| 02:58.20 | arrrghhh | yea he's got prox working well |
| 02:58.29 | arrrghhh | i guess he wrote a new ls driver... |
| 02:58.36 | arrrghhh | changed a lot of crap in the panel |
| 02:58.37 | WisTilt2 | problem we will have is without any winmo, nothing at all gets setup so we have to do it all. |
| 02:58.42 | arrrghhh | yup |
| 02:58.48 | arrrghhh | no more cheating :P |
| 02:59.13 | arrrghhh | well checkout his tree for sure |
| 02:59.20 | arrrghhh | should help, he's done a lot of the dirty work |
| 02:59.49 | arrrghhh | http://gitorious.org/~acl/linux-on-qualcomm-s-msm/linux-msm-rhod-nand |
| 03:00.07 | WisTilt2 | ok don't eat too much or you'll want to just sleep |
| 03:00.14 | arrrghhh | k, i'm out for reals this time |
| 03:00.16 | arrrghhh | hahahaha |
| 03:00.18 | arrrghhh | ;) |
| 03:04.27 | WisTilt2 | d3tule3: before i waste tonight on BT, do you know if they actually have it working on .27 or .35 with GB? i know it works on .35 with froyo np, i've used alex's .35 acoustic tree and works great but it didn't work at that time on GB. |
| 03:05.14 | d3tul3 | WisTilt2, I could be wrong about this mgross029 is the expert on BT w/ 35, but i think emwe has BT working but not doing data transfer perhaps? |
| 03:06.30 | WisTilt2 | i just want to make sure someone has it working on GB as far as pairing, which i have verified on .39, and audio working. data transfers i can deal with later. |
| 03:06.45 | *** join/#htc-linux mgross029_ (~mgross@cpe-107-10-11-218.neo.res.rr.com) |
| 03:06.54 | d3tul3 | there's the expert |
| 03:07.15 | d3tul3 | (sorry i don't know for a fact one way or another) |
| 03:07.59 | WisTilt2 | mgross029 any luck with that kernel after you locked up? everything looks like it should be working but userland still thinks it is disabled. frustrating |
| 03:10.18 | d3tul3 | jONpRY, ping |
| 03:10.31 | *** join/#htc-linux Bry8Star_ (~Bry8Star@gateway/tor-sasl/bry8star) |
| 03:10.33 | jONpRY | yo |
| 03:10.39 | d3tul3 | what do you think git@gitorious.org:~detule/linux-msm-rhod/detules-linux-msm-rhod.git ? |
| 03:10.51 | d3tul3 | uh https://gitorious.org/~detule/linux-msm-rhod/detules-linux-msm-rhod |
| 03:11.58 | jONpRY | um confused. what is this? a clone? |
| 03:12.14 | d3tul3 | yeah i have it up to 3.0.9 locally in my tree |
| 03:12.40 | jONpRY | oic. you applied the incremental patches? |
| 03:13.36 | jONpRY | or cherry picked it somehow? |
| 03:13.41 | jONpRY | does it work? |
| 03:13.41 | d3tul3 | i couldn't find some of the incremental patches so i git format-patch 3.0 -> 3.01 ; 3.01->3.02 |
| 03:13.45 | d3tul3 | they all applied cleanly |
| 03:14.02 | d3tul3 | compiling now |
| 03:14.10 | jONpRY | yeah that how you have to do the stuff after rc |
| 03:14.24 | jONpRY | you got the rc to final patches? |
| 03:14.40 | d3tul3 | you had the rc to 3.0.0 final |
| 03:14.51 | d3tul3 | so i got it current to 3.0.9 |
| 03:15.34 | jONpRY | hrm. then 3.0 final didn't work too good |
| 03:16.15 | d3tul3 | hopefully 3.0.9 will be better |
| 03:16.15 | jONpRY | this is good though. might as well pull in that stuff |
| 03:19.57 | jONpRY | going to upgrade 39? |
| 03:22.56 | d3tul3 | yeah i guess that's next |
| 03:23.10 | d3tul3 | kind of done hacking away at microp* so i need new things to keep me busy |
| 03:24.22 | jONpRY | getting patches into 3.0 could be good |
| 03:25.32 | d3tul3 | btw whatever you and WisTilt2 did to scbs has improved screen operation back to normal |
| 03:26.12 | jONpRY | good |
| 03:31.41 | jONpRY | i'm having a terrible time with prypad usb atm |
| 03:37.07 | d3tul3 | hey these commit messages "sync with .39 tree" how exactly did you do that? |
| 03:37.50 | jONpRY | different methods |
| 03:38.45 | jONpRY | easiest thing is to take a group of commits that are all to the mach folder and just copy it |
| 03:38.56 | jONpRY | for other make a patch and try to apply it |
| 03:40.03 | d3tul3 | ok |
| 03:40.15 | d3tul3 | well it finished building |
| 03:47.49 | *** join/#htc-linux surge (surge@pool-72-88-82-28.bflony.fios.verizon.net) |
| 03:52.48 | d3tul3 | it boots fine though it needs some love |
| 03:56.59 | *** join/#htc-linux surgex (surge@pool-72-88-82-28.bflony.fios.verizon.net) |
| 03:58.48 | jONpRY | like missing patch love? |
| 04:02.31 | d3tul3 | yeah |
| 04:02.43 | d3tul3 | actually it's doing well |
| 04:03.02 | d3tul3 | i'll format-patch most of the stuff from .39 |
| 04:03.07 | d3tul3 | and see if it will apply cleanly |
| 04:03.08 | jONpRY | it was ok back in the day, but it would SR once a day or so |
| 04:13.36 | *** join/#htc-linux bzo (~chatzilla@c-174-62-79-238.hsd1.ca.comcast.net) |
| 04:13.42 | arrrghhh | WisTilt2, looks like you never got a response |
| 04:13.49 | arrrghhh | .35 + GB, BT works fine. |
| 04:13.55 | arrrghhh | there's a bug where you actually have to have it on... |
| 04:14.03 | arrrghhh | system won't sleep otherwise, lol |
| 04:14.10 | *** join/#htc-linux Rob2222 (~Miranda@p4FFF121C.dip.t-dialin.net) |
| 04:20.00 | *** join/#htc-linux Bry8Star (~Bry8Star@gateway/tor-sasl/bry8star) |
| 04:21.16 | WisTilt2 | thanks arrrghhh. do audio path works in a call via BT then also? |
| 04:21.26 | arrrghhh | emwe said yes |
| 04:21.30 | arrrghhh | i haven't tested a BT headset |
| 04:21.36 | arrrghhh | both of us had data xfer issues tho... |
| 04:21.39 | WisTilt2 | ok, guess i'll keep digging:) |
| 04:21.50 | arrrghhh | i could pair, but not transfer anything between phones |
| 04:22.10 | WisTilt2 | but with a BT headset someone has had audio? |
| 04:22.29 | arrrghhh | emwe has, yes. |
| 04:22.38 | arrrghhh | i can test if you wish, i haz a headset here |
| 04:23.42 | WisTilt2 | if emwe has verified it we'll assume it works. i just wanted to make sure GB wasn't the actual problem before i spend more time in the kernel |
| 04:24.24 | WisTilt2 | ive only got pairing working, just no audio either way so im thinking something in acoustic |
| 04:24.49 | arrrghhh | 2011-11-09 19:11 emwe using bt headset also ok for me. |
| 04:24.51 | arrrghhh | :D |
| 04:25.03 | arrrghhh | WisTilt2, that's interesting |
| 04:25.12 | arrrghhh | people had hit-or-miss issues with headsets before headsets |
| 04:25.21 | arrrghhh | some said they worked fine, others would get no audio routing... |
| 04:26.00 | WisTilt2 | i dont have an actual bt headset, only bt built into my truck but it works great in winmo and also on alex's .27 acoustic on froyo. |
| 04:26.12 | arrrghhh | hum |
| 04:26.23 | arrrghhh | did you try .27 + GB? |
| 04:26.25 | arrrghhh | just to rule it out |
| 04:26.37 | WisTilt2 | no, might need to do that before i continue |
| 04:26.47 | arrrghhh | might save you some time |
| 04:27.48 | WisTilt2 | doesn't .27 have the old acoustic naming in it though so wouldn't work with GB? alex's .27 has it renamed. |
| 04:28.14 | arrrghhh | hrm |
| 04:28.31 | arrrghhh | yea, i'm not sure if there would be a functional .27 + GB setup |
| 04:28.40 | arrrghhh | let me see if i can drag something up from mah email, 1 sec |
| 04:29.52 | *** join/#htc-linux YataroX (~user@211.sub-174-252-199.myvzw.com) |
| 04:30.07 | arrrghhh | hrm |
| 04:30.21 | arrrghhh | mgross029 i see tested the newest gb build on .27... |
| 04:30.31 | arrrghhh | but had some issues with audio |
| 04:30.41 | arrrghhh | i don't know what .27 kernel he's using tho |
| 04:31.31 | WisTilt2 | i may wait on this until tomorrow and check with emwe. |
| 04:33.22 | arrrghhh | i hear pmem needs some lovin :P |
| 04:34.15 | WisTilt2 | yeah something haywire there. seeing all kinds of overlaps but randomly and seems to be acoustic related. |
| 04:34.50 | arrrghhh | weird |
| 04:35.04 | WisTilt2 | scbs is flying though. nice to have that puppy back |
| 04:35.08 | arrrghhh | would that explain a lot of the oddities with acoustic? |
| 04:35.42 | arrrghhh | yea, i miss SCBS. still trying to get jONpRY on board with NAND. or maybe I can just con ACL into shoehorning it in :P |
| 04:35.48 | WisTilt2 | possibly but dont know if its only happening in .39 since i dont run anything else anymore. been on .39 for months now. |
| 04:35.56 | arrrghhh | nice |
| 04:36.05 | arrrghhh | newer than all the native devices, lol |
| 04:36.20 | WisTilt2 | scbs is so advanced over anything else imo |
| 04:36.36 | arrrghhh | yea, i don't know what else is thar |
| 04:36.45 | arrrghhh | 'cept the broken ass "model" that is used by default |
| 04:37.41 | jONpRY | its not good? |
| 04:37.59 | arrrghhh | no, i'm saying w/o SCBS |
| 04:38.03 | arrrghhh | whatever camro did with his blackstone |
| 04:38.15 | arrrghhh | i think you said it's basically based on his phone batteries model or some such |
| 04:38.24 | arrrghhh | which is obviously broken to begin with |
| 04:38.29 | WisTilt2 | he made a good attempt, better than what we had before but scbs is lightyears above it |
| 04:38.37 | arrrghhh | not to mention no accounting for peripherals. |
| 04:38.38 | arrrghhh | for sure |
| 04:38.43 | jONpRY | just sucks power |
| 04:38.56 | jONpRY | we need to move it to arm9, ;- |
| 04:39.09 | WisTilt2 | yeah that's the ticket. |
| 04:40.04 | WisTilt2 | jonpry you were right about A1010. it broke speakerphone mic but still worked with handset mic. |
| 04:40.52 | jONpRY | i think its possible to make it dual mic |
| 04:41.11 | arrrghhh | i thought that was the point in speakerphone |
| 04:41.20 | arrrghhh | one mic helps with noise cancelling, the other is the mic |
| 04:41.52 | jONpRY | sounds right |
| 04:42.10 | WisTilt2 | i still dont think both are actually working but the handset mic is sensitive enough in speakerphone mode to pick up nicely. |
| 04:46.34 | d3tul3 | oi you guys didn't make it easy with these early patches |
| 04:46.54 | jONpRY | no probably not |
| 04:47.55 | *** join/#htc-linux Andreyxxl[HD2EU] (~Andreyxxl@94.52.236.39) |
| 04:48.03 | *** join/#htc-linux TheXev (~user@211.sub-174-252-199.myvzw.com) |
| 04:50.14 | jONpRY | lots of reverting stuff |
| 04:50.50 | d3tul3 | your commits contain all this binaries |
| 04:52.17 | jONpRY | oh yeah. never did figure out how to not do that |
| 04:52.46 | WisTilt2 | i finally did- the beauty of .gitignore |
| 04:53.12 | WisTilt2 | spent a week playing with git so i didn't have to keep asking how the hell to do things:) |
| 04:53.22 | jONpRY | if your ever wondering why it is that devices never support host mode. its because the silicon is broken, the drivers are broken, and the cables and devices are broken |
| 04:54.16 | WisTilt2 | lol |
| 04:54.33 | WisTilt2 | or its running any form of windows |
| 04:54.53 | jONpRY | winmo |
| 05:13.03 | WisTilt2 | jonpry: re scbs power... it only draws 20ma with the daemon and 18ma without. thats still not bad for the gain of better battery management |
| 05:13.47 | jONpRY | weren't there ports of says 3 days vs 4 days of life? |
| 05:14.02 | WisTilt2 | if you can poll a couple times a minute or even at minute intervals that will be lower |
| 05:14.36 | jONpRY | yeah thats possible |
| 05:14.46 | jONpRY | have we ever messed with idle collapse? |
| 05:15.44 | WisTilt2 | i actually have idle collapse in the kernel now, just not pushed. |
| 05:16.02 | jONpRY | does it work? |
| 05:16.25 | WisTilt2 | so far it seems to. been running it since yesterday. |
| 05:16.31 | jONpRY | whoa |
| 05:16.41 | jONpRY | saves juice? |
| 05:17.16 | jONpRY | nobody will ever use 27 again |
| 05:17.32 | WisTilt2 | i changed the default modes both to 1 instead of 0 and enabled the idle collapse. hasn't blown up yet. i need to do some longer tests to see how much we gained. |
| 05:17.42 | WisTilt2 | .27? whats that? |
| 05:17.45 | jONpRY | this is a game changer :) |
| 05:18.27 | WisTilt2 | i love .39 especially since all the fixes lately. very stable, quick, and everything important is working with not much more to fix. |
| 05:18.39 | bzo | WisTilt2: is that 20ma draw with or without idle collapse? |
| 05:18.45 | WisTilt2 | without |
| 05:18.56 | WisTilt2 | ill measure it tomorrow with it |
| 05:19.07 | bzo | wow, that could be nice |
| 05:19.37 | WisTilt2 | since ive seen 8ma in winmo i'll bet we'll get down there eventually |
| 05:20.40 | bzo | in theory you can standby for weeks if you turn of the radio |
| 05:21.27 | WisTilt2 | idle collapse the radios on though |
| 05:21.41 | jONpRY | nook color has something like that going on 60 days standby |
| 05:22.29 | WisTilt2 | i did notice one thing that may or may not be caused by idle collapse. seems with it, it takes a little longer to wake things up on incoming calls. |
| 05:22.34 | bzo | for a more apples comparison, the vogue reportedly will standby for weeks |
| 05:23.23 | bzo | jONpRY: any progress on standby for the touchpad? |
| 05:24.26 | jONpRY | we can turn on collapse but there are problems with the sound driver |
| 05:25.00 | bzo | are those the issues assigned to you? :P |
| 05:25.12 | jONpRY | i think so :p |
| 05:25.46 | bzo | it feels like standby is not great in webos either |
| 05:25.48 | jONpRY | we get qdsp6 watchdog bites which result in reboot/death within a few seconds |
| 05:26.42 | jONpRY | there was some talk of trying to effect a Q6 reboot, but there is no reason q6 should be dying |
| 05:28.02 | WisTilt2 | do either of you know what time IDLE_SPIN_TIME is in? default is 80000 and thats where i left it. |
| 05:28.39 | bzo | ns? |
| 05:29.55 | jONpRY | the big 20000*** number is 20 seconds |
| 05:30.24 | WisTilt2 | ok that makes sense. 20million is big and wondered about that |
| 05:31.47 | jONpRY | 8x60 is nutso with collapse. it has this non modem kind of collapse that it will use for sub millisecond sleeps |
| 05:32.00 | WisTilt2 | bzo before you disappear. dont know if you read my comment earlier today but camera started working last night, strangely after i made a call. i rebooted and tried it again and jpeg crashed like before. i think these pmem overlaps we keep seeing are the problem. |
| 05:32.33 | WisTilt2 | has to be acoustic related somehow |
| 05:32.40 | bzo | WisTilt2: yes read it, acoustic does seem suspect |
| 05:33.24 | WisTilt2 | the address are similar that keep overlapping but dont really know how to trace that back easily. |
| 05:34.15 | bzo | the qdsp code seems to track its own pmem allocations, maybe there is a bug in how acoustic is allocating |
| 05:35.20 | WisTilt2 | detule told me to disable acoustic. that fixed the camera totally. i think you're right, im going to try sticking the camera alloc up in smi2 where i found that 10mb open space. |
| 05:35.39 | WisTilt2 | on the 300 anyway, different on the cdma's |
| 05:35.55 | bzo | that was my experience as well in disabling acoustic |
| 05:36.18 | bzo | not sure that it is clashing in the camera pmem block though |
| 05:36.38 | bzo | might be the general pmem region, but I guess that should be easy enough to figure from the clash messages |
| 05:38.22 | WisTilt2 | should be fun to figure out |
| 05:53.43 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-81-126.dynamic.mnet-online.de) |
| 06:03.43 | d3tul3 | git am -3 ftw |
| 06:07.45 | *** join/#htc-linux mitsutaka (~mitsutaka@p2004-ipbf2206marunouchi.tokyo.ocn.ne.jp) |
| 06:08.06 | jONpRY | whats that do? |
| 06:08.25 | d3tul3 | merge resolution |
| 06:10.03 | jONpRY | so its coming along? |
| 06:10.33 | d3tul3 | almost there, it just compiled mach-msm with the patches in, stuck on suspend.c something with the last patch or two |
| 06:10.48 | d3tul3 | kernel/power/suspend.c:140:2: error: implicit declaration of function 'sysdev_suspend' |
| 06:11.59 | jONpRY | i don't think i added any calls to such a function |
| 06:12.20 | d3tul3 | i am looking at your patch and i guess you did |
| 06:13.54 | d3tul3 | they droped it 39->3.0 |
| 06:13.58 | d3tul3 | can you believe that |
| 06:14.18 | d3tul3 | find me a substitute function:) or I guess i can revert the scbs patches |
| 06:14.47 | jONpRY | change to syscore_suspend(); |
| 06:15.00 | jONpRY | resume needs a change as well |
| 06:15.36 | jONpRY | er |
| 06:17.21 | jONpRY | changes definitely need to be made, but sysdev_suspend is still in the kernel |
| 06:17.28 | jONpRY | at least in the 3.0-final i have |
| 06:17.49 | jONpRY | guess i will pull |
| 06:19.48 | jONpRY | d3tul3, nm i was on wrong branch, change both to syscore |
| 06:19.58 | d3tul3 | with void arg? |
| 06:20.41 | jONpRY | right |
| 06:21.25 | d3tul3 | k, past it, let me see if it will snag somewhere else |
| 06:38.36 | bzo | jONpRY: it seems like you guys had to do a little hacking on pmem to get it working back in june |
| 06:39.36 | jONpRY | hmm. link? |
| 06:39.49 | bzo | http://irclog.netripper.com/htc-linux/2011/6/5/ |
| 06:40.14 | bzo | and I'm seeing in source a check was disabled in pmem_open() |
| 06:41.09 | bzo | when I re-enable it I get the same error you reported back then |
| 06:41.21 | bzo | so I'm wondering if there is still some problem with the mapping |
| 06:41.56 | jONpRY | you can't enable that check :p |
| 06:42.30 | jONpRY | all it does is make sure nobody opens a pmem segment twice |
| 06:42.59 | jONpRY | but the mechanism they use is a hack because 39 fills up the priv_data pointer with something, and it assumes it would normally be null |
| 06:43.37 | jONpRY | unless your really doing something broken though the situation its checking for can't happen |
| 06:43.56 | bzo | hmm, perhaps there is some other 39 specific behavior that is screwing pmem somewhere |
| 06:44.05 | bzo | we're getting clashes of overlapping regions |
| 06:44.16 | jONpRY | i don't understand what means |
| 06:44.33 | jONpRY | don't you specify the start and end for all pmem regions? |
| 06:44.48 | bzo | I mean individual chucks used within a pmem region |
| 06:45.18 | bzo | a chunk A is getting allocated, then another chunk B is getting memory allocated that overlaps with A |
| 06:45.23 | bzo | and failing a sanity check |
| 06:45.34 | jONpRY | like the virtual addresses are wrong? |
| 06:45.47 | bzo | dunno, maybe |
| 06:46.05 | jONpRY | what sanity check fails where? |
| 06:46.40 | bzo | it's in the qdsp5 code |
| 06:46.55 | bzo | it has lists of pmem it is using, then cross references with other chunks |
| 06:47.13 | bzo | if you grep "clash" in the qdsp5, you can see where we are observing it |
| 06:47.21 | jONpRY | well qdsp5 is totally unaware of the virtual mappings created by pmem.c |
| 06:47.32 | jONpRY | as in the pmem driver |
| 06:47.38 | bzo | right, it just takes what is given to it |
| 06:48.00 | WisTilt2 | i checked those virtual addresses a dozen times. something is happening with pmem_access_prot not happening. i see several pmem_mmap mapping same locations to similar areas. |
| 06:48.45 | d3tul3 | that access_prot dmesg spam *could* be unrelated |
| 06:48.48 | WisTilt2 | let me paste what im talking about. maybe im chasing nothing here |
| 06:49.28 | WisTilt2 | [ 4278.969757] pmem_access_prot: did not set access prot: 0, 1, 0 |
| 06:49.28 | WisTilt2 | [ 4278.969818] pmem_mmap: mmaping 0x24E00000 x 0x2000000 to: 0x45014000 |
| 06:49.28 | WisTilt2 | [ 4279.813659] pmem_access_prot: did not set access prot: 0, 1, 0 |
| 06:49.28 | WisTilt2 | [ 4279.813781] pmem_mmap: mmaping 0x24E00000 x 0x2000000 to: 0x45510000 |
| 06:49.28 | WisTilt2 | [ 4279.976409] pmem_access_prot: did not set access prot: 0, 1, 0 |
| 06:49.29 | WisTilt2 | [ 4279.976470] pmem_mmap: mmaping 0x24E00000 x 0x2000000 to: 0x45BC0000 |
| 06:49.45 | WisTilt2 | mapping same 0x24e00000 to different addresses? |
| 06:50.08 | *** join/#htc-linux TheXev (~user@211.sub-174-252-199.myvzw.com) |
| 06:50.16 | d3tul3 | could be that happens on .27 as well, we just don't have that printk in place over there |
| 06:50.37 | jONpRY | there also not seperated enough |
| 06:51.25 | jONpRY | assuming 0x200000 is the size |
| 06:51.49 | WisTilt2 | it is from what i can tell |
| 06:52.23 | WisTilt2 | this is coming from staging/dream/pmem.c |
| 06:54.09 | WisTilt2 | these are causing the clashes down the road when camera or acoustic related things start |
| 06:55.34 | WisTilt2 | i have to get out of here for tonight. let me know if you guys find anything. btw, with idle collapse im seeing under 1% battery drain per hour over the past several hours. ill measure actual current tomorrow. nite fella. |
| 06:59.33 | jONpRY | maybe check io_remap_pfn_range, the call is all funky because something are PFN and others not. so its possible the length is being >> 13'd or something stupid |
| 06:59.34 | *** join/#htc-linux YataroX (~user@4.sub-174-252-213.myvzw.com) |
| 07:00.50 | *** join/#htc-linux kiozen (~kiozen@p54BB73F3.dip.t-dialin.net) |
| 07:01.06 | jONpRY | d3tul3 it working? |
| 07:01.59 | d3tul3 | jONpRY, sure |
| 07:02.22 | d3tul3 | i'll push these out tomorrow |
| 07:02.25 | d3tul3 | to my repo |
| 07:02.31 | jONpRY | as good as 39? |
| 07:02.43 | d3tul3 | touch to say,need some time on it |
| 07:03.08 | d3tul3 | s/touch/tough/ |
| 07:03.11 | bzo | jONpRY: do you remember why you commented out that stuff in pmem_access_prot() ? |
| 07:04.53 | jONpRY | i had to use different functions to achieve the same effect |
| 07:05.07 | jONpRY | at this point all the different prots are supported |
| 07:05.46 | bzo | ok |
| 07:06.12 | jONpRY | it used to be a weak symbols kind of thing |
| 07:20.22 | d3tul3 | jONpRY, all the patching is done and up in my clone |
| 07:20.26 | d3tul3 | i am out for the night later folks |
| 07:25.04 | *** join/#htc-linux dobrin (~dobrin@85.91.150.26) |
| 07:55.55 | *** join/#htc-linux rzk (~rzk@95-25-22-100.broadband.corbina.ru) |
| 08:01.49 | *** join/#htc-linux YataroX (~user@160.sub-174-252-213.myvzw.com) |
| 08:05.10 | *** join/#htc-linux rob_w (~bob@ppp-93-104-169-233.dynamic.mnet-online.de) |
| 08:22.11 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
| 08:25.19 | *** join/#htc-linux BabelO (~fcr@AMontpellier-553-1-44-52.w90-48.abo.wanadoo.fr) |
| 08:25.19 | *** join/#htc-linux BabelO (~fcr@unaffiliated/babelo) |
| 09:16.03 | *** join/#htc-linux gauner1986 (~Miranda@87.253.171.208) |
| 09:21.42 | *** join/#htc-linux ychavan (ychavan@nat/redhat/x-wdlnyfyvvyoxpudj) |
| 09:33.42 | *** join/#htc-linux NYL (~NYL@109.96.36.166) |
| 10:13.13 | *** join/#htc-linux paulk (~paulk@lib33-1-82-233-88-171.fbx.proxad.net) |
| 10:23.09 | *** join/#htc-linux mgross__ (~mgross@cpe-107-10-11-218.neo.res.rr.com) |
| 10:23.24 | *** join/#htc-linux rzk (~rzk@95-25-119-53.broadband.corbina.ru) |
| 10:25.44 | *** join/#htc-linux Bry8Star_ (~Bry8Star@gateway/tor-sasl/bry8star) |
| 10:26.44 | *** join/#htc-linux mgross029 (~mgross@cpe-107-10-11-218.neo.res.rr.com) |
| 10:45.54 | *** join/#htc-linux lamikr (lamikr@nat/nokia/x-kixveivljavdyzhh) |
| 11:03.19 | *** join/#htc-linux LordDeath (~LordDeath@cable-81-173-166-124.netcologne.de) |
| 11:15.09 | *** join/#htc-linux joplayer (~jonathan@2a01:e34:ef80:50:21d:60ff:feec:b36c) |
| 11:43.25 | *** join/#htc-linux stroughtonsmith_ (~steven@86-43-221-235-dynamic.b-ras2.bbh.dublin.eircom.net) |
| 11:45.48 | *** join/#htc-linux NYL (~NYL@109.96.36.166) |
| 11:57.32 | *** join/#htc-linux MethoS- (~clemens@134.102.106.250) |
| 12:12.59 | *** join/#htc-linux EdLin (~EdLin@securabit/listener/edlin) |
| 12:40.11 | *** join/#htc-linux Mavy-bnc (mavfree@91.196.169.2) |
| 12:47.11 | *** join/#htc-linux kiozen (~kiozen@p54BB73F3.dip.t-dialin.net) |
| 12:49.29 | *** join/#htc-linux kiozen (~kiozen@p54BB73F3.dip.t-dialin.net) |
| 12:54.58 | *** join/#htc-linux kiozen (~kiozen@p54BB73F3.dip.t-dialin.net) |
| 12:55.49 | *** join/#htc-linux kiozen (~kiozen@p54BB73F3.dip.t-dialin.net) |
| 12:56.29 | *** join/#htc-linux Andreyxxl1 (~Andreyxxl@94.52.236.39) |
| 13:37.24 | *** join/#htc-linux mgross029 (~mgross@cpe-107-10-11-218.neo.res.rr.com) |
| 13:47.07 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:cdad:f553:ce94:42a) |
| 14:26.54 | *** join/#htc-linux Cotulla (~myfakemai@nat100-255-205-109.tvoe.tv) |
| 14:45.37 | *** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:cdad:f553:ce94:42a) |
| 14:52.21 | *** join/#htc-linux kiozen (~kiozen@p54BB73F3.dip.t-dialin.net) |
| 15:03.27 | *** join/#htc-linux vinceweis (433575ca@gateway/web/freenode/ip.67.53.117.202) |
| 15:18.30 | *** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring) |
| 15:50.10 | *** join/#htc-linux rob_w (~bob@ppp-93-104-169-233.dynamic.mnet-online.de) |
| 15:58.44 | *** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com) |
| 16:12.39 | *** join/#htc-linux avinashhm (~avinash-h@192.91.66.189) |
| 16:26.41 | *** join/#htc-linux mitsutaka (~mitsutaka@p2004-ipbf2206marunouchi.tokyo.ocn.ne.jp) |
| 16:30.50 | detule | where does one go for upstream code in "drivers/staging/android" these days? |
| 16:31.11 | jONpRY | depends on what |
| 16:32.14 | detule | lowmemorykiller |
| 16:32.18 | detule | and mm/ashmem.c |
| 16:33.18 | detule | in 3.0 they've re-written how the slab shrinker interacts with the shrinkers (ashmem, lowmemorykiller)....with the old drivers, the shrinkers don't work |
| 16:33.48 | detule | it's kind of funny that the problem is in the same place as in .39 though different problem |
| 16:36.28 | detule | i have an idea what the issue is and could try to patch these but i would rather see if google has conformed at this point |
| 16:41.03 | jONpRY | there is no gitweb that i know of for current android source |
| 16:41.23 | jONpRY | and omapzoom is down |
| 16:41.48 | jONpRY | i think CAF has a 3.0 kernel |
| 16:42.26 | jONpRY | dur it appears to be non android |
| 16:43.37 | *** join/#htc-linux rzk (~rzk@95-25-119-53.broadband.corbina.ru) |
| 16:44.08 | *** join/#htc-linux Dave888 (~Dave888@197.1.245.102) |
| 16:44.35 | *** join/#htc-linux paulk_ (~paulk@lib33-1-82-233-88-171.fbx.proxad.net) |
| 16:44.39 | Cotulla | hi |
| 16:44.42 | Cotulla | ONRY |
| 16:44.45 | *** join/#htc-linux mgross029 (c0234f46@gateway/web/freenode/ip.192.35.79.70) |
| 16:45.04 | *** join/#htc-linux WisTilt2 (~wisgreg@wireless251.wirelesstcp.net) |
| 16:45.45 | jONpRY | detule, https://github.com/android/kernel_common/tree/android-3.0 |
| 16:46.08 | Cotulla | 3.0 4.0 lol |
| 16:46.20 | Cotulla | they like big numbers |
| 16:46.51 | jONpRY | me too |
| 16:49.24 | mgross029 | WisTilt2: Pulled the latest and still can't enable bt on .39 |
| 16:50.22 | WisTilt2 | are you seeing hci0 after you try turning it on? also check interrupts and see if bt_host_wake is firing at all. |
| 16:52.13 | mgross029 | Let me check forgot to do that. damn. :p |
| 16:52.43 | detule | jONpRY, thanks that's exactly what i needed |
| 16:52.43 | jONpRY | can you do hcitool scan and stuff? |
| 16:53.01 | mgross029 | arrrghhh: I did test with the .27 kernel that emwe sent out in email, last friday iirc. |
| 16:53.33 | mgross029 | jONpRY: I can yes. |
| 16:54.00 | WisTilt2 | everything appears to be working but userland doesn't see it as enabled |
| 16:55.29 | WisTilt2 | jonpry: you and bzo come up with any more ideas with pmem? |
| 16:58.12 | mgross029 | WisTilt2: no hci0 present. hciconfig is empty too |
| 16:58.18 | WisTilt2 | mgross029 when you say you pulled the latest... you mean you dl'd the kernel pack on my local server for testing bt right? |
| 16:58.34 | WisTilt2 | git tree has no BT pushed yet |
| 16:59.15 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
| 16:59.23 | mgross029 | I will double check but I thought that was the latest yesterday. Did you update since then? |
| 17:00.44 | mgross029 | WisTilt2: g42d1a47 correct? |
| 17:00.45 | WisTilt2 | i just uploaded the one im using so try it. dont know which one you dl'd |
| 17:01.35 | WisTilt2 | g9c618ee |
| 17:02.20 | mgross029 | Hmm That isn't what I'm seeing out there. Maybe my browser is caching it. Let me clear. |
| 17:03.31 | WisTilt2 | i dont think the one you have had all the BT stuff in it. the one i just put up is current and you should see hci0 |
| 17:03.33 | mgross029 | Still seeing this on the site. Current Kernel Pack: kernel-pack-htc-msm-2.6.39.3-g42d1a47-dirty-20111116_113601.zip |
| 17:03.59 | WisTilt2 | lol. you didn't dl from the link i pm'd you? |
| 17:04.16 | WisTilt2 | you're pulling off my autobuild not the private server:) |
| 17:04.19 | jONpRY | detule any crashes yet? iirc i was able to cause crashing by starting and stopping birds a couple of times |
| 17:04.45 | detule | that's because the shrinkers aren't working it loads up all ram and dies |
| 17:04.56 | jONpRY | oh |
| 17:05.02 | detule | btw, the suspend.c syscore functions are blowing up sleep in 3.0 |
| 17:05.41 | jONpRY | it turns out that .39 should have both syscore and sysdev |
| 17:05.56 | detule | here's the pastebin http://pastebin.com/FuC0dppv |
| 17:06.28 | detule | google has already updated lowmemorykiller i am in the process of patching ashmem, we'll see if this makes them engage properly |
| 17:09.33 | jONpRY | thats not a crash, just a WARN() |
| 17:09.56 | WisTilt2 | i'm sure this is a dumb question but why do we have pmem_size set to 32mb? how was that size determined just curious. |
| 17:09.57 | detule | yeah but it makes sleep a painful affair as in 10second resumes |
| 17:10.09 | jONpRY | in scbs stuffs. it must not like us accessing ktime during sleep or something. probably have to patch ktime |
| 17:10.29 | detule | booting 3.0 without the scbs daemon running resuming works fine |
| 17:14.54 | jONpRY | just comment out kernel/time/timekeeping.c line 245 |
| 17:15.43 | detule | you think it's the printk that's overwhelming resume much like before? |
| 17:17.57 | jONpRY | probably |
| 17:18.35 | jONpRY | needs to be done because we support ktime stuffs during sleep |
| 17:18.47 | detule | oh yeah it's humming along with the lowmemorykiller + ashmem patch |
| 17:18.52 | detule | can't bring the available memory down |
| 17:19.17 | detule | might be time for you to fire up that rhod:) |
| 17:22.33 | mgross029 | WisTilt2: same results as ^^^ |
| 17:23.30 | WisTilt2 | mgross029: after you try to enable it you are not seeing hci0 anywhere? what about bt host wake irq, still 0? |
| 17:24.08 | WisTilt2 | hci0 should be at sys/class/bluetooth for sure |
| 17:25.35 | WisTilt2 | im sure you already tried doing a find / -name hci0 on the device? |
| 17:25.45 | mgross029 | I will check after I get back to Android. Ran hciattach and got the bdaddr the 1(special characters) and option enable_hci and the phone locked |
| 17:26.44 | WisTilt2 | if you could try enabling it then paste your cat /proc/interrupts so i can take a look |
| 17:26.58 | mgross029 | Nothing for hci0 anywhere so far. Let me get back in and I'll look in sys |
| 17:27.05 | mgross029 | Sure |
| 17:27.20 | WisTilt2 | just do the find, it may be in several places like im seeing also |
| 17:28.22 | mgross029 | Ok. |
| 17:28.28 | *** part/#htc-linux Dave888 (~Dave888@197.1.245.102) |
| 17:33.31 | mgross029 | WisTilt2: [ 284.266815] microp_set_color_led_state: color=GREEN bluetooth rfkill state ON. This was in dmesg from last time. I've never seen this before in any of the kernels |
| 17:33.52 | mgross029 | find found nothing for hci0 |
| 17:36.07 | WisTilt2 | let me build another for you to test if you have the time right now? |
| 17:36.34 | mgross029 | Sure. Just eating lunch. :p |
| 17:46.54 | *** join/#htc-linux |Jeroen| (~jeroen@d5152B25B.access.telenet.be) |
| 17:53.36 | mgross029 | After testing an incomming call my dmesg is getting blasted with this. http://pastebin.com/xydnBzwb |
| 17:54.46 | WisTilt2 | what rootfs and system image you using? |
| 17:55.44 | mgross029 | The latest. |
| 17:56.20 | WisTilt2 | latest from where? should be emwe's rootfs-unsecure and his latest GB image? |
| 17:56.45 | mgross029 | correct |
| 17:58.24 | mgross029 | I just built a new gb image after emwe pushed the kexec into the tree. |
| 18:00.01 | Cotulla | lol my poor |
| 18:08.43 | *** join/#htc-linux raymonddull (~raymonddu@c-98-209-145-76.hsd1.mi.comcast.net) |
| 18:10.29 | WisTilt2 | mgross029: try this kernel pack |
| 18:10.57 | WisTilt2 | same link/name |
| 18:11.10 | mgross029 | Yep. Pulling it down now |
| 18:11.14 | *** part/#htc-linux Cotulla (~myfakemai@nat100-255-205-109.tvoe.tv) |
| 18:11.52 | WisTilt2 | if you dont get hci0 showing up after enabling BT then im stumped. i need to see your /proc/interrupts also after you enable if you could. |
| 18:12.25 | mgross029 | Ok. I will post from previous too sec. |
| 18:13.05 | WisTilt2 | jonpry you still around? |
| 18:14.08 | mgross029 | Interrupts from previous kernel http://pastebin.com/UgcWUx4U |
| 18:14.33 | WisTilt2 | well bt host_wake fired |
| 18:15.02 | WisTilt2 | im puzzled why hci0 isn't anywhere. hopefully this build it will |
| 18:15.25 | mgross029 | Yeah it was wierd. I was refreshing after trying to enable again and it did not increment |
| 18:16.16 | WisTilt2 | i think once you try to enable BT the rfkill off never happens again since userland thinks its still disabled so all you see is rfkill on everytime you enable it. |
| 18:16.58 | WisTilt2 | but host_wake doesn't fire again since it thinks its already awake i guess |
| 18:18.28 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-81-126.dynamic.mnet-online.de) |
| 18:18.47 | mgross029 | Ok. |
| 18:20.59 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
| 18:26.36 | jONpRY | hi WisTilt2 |
| 18:28.56 | WisTilt2 | you and that crazy name you got going on. you got a minute to take a look at something in pm.c? i dont think the way this is setup that idle_collapse is actually able to happen. |
| 18:29.27 | jONpRY | sure |
| 18:29.54 | jONpRY | cruise control is aWeSoMe |
| 18:30.15 | WisTilt2 | if you look starting at line 283, i dont think the msm_arch_idle ever gets called unless we're in apps_sleep |
| 18:30.25 | WisTilt2 | and what is cruise control |
| 18:30.38 | jONpRY | capslock is cruise control for cool |
| 18:30.53 | WisTilt2 | GUESS IM BEHIND IN THE TIMES |
| 18:31.02 | jONpRY | i was too |
| 18:36.32 | detule | jONpRY, merge 3.0.9 into linux-msm-rhod? |
| 18:38.19 | jONpRY | sounds good |
| 18:39.21 | jONpRY | WisTilt2, does arch_idle need to be called? |
| 18:39.39 | jONpRY | iirc idle collapse was effectively disabled by that 20 second thing in Kconfig |
| 18:39.41 | WisTilt2 | yes i believe so. |
| 18:40.17 | WisTilt2 | i enabled it in config and set its mode = 1 for power collapse. it was in suspend by default |
| 18:40.42 | WisTilt2 | i just dont see how it does anything without arch_idle ever being called though. |
| 18:40.55 | jONpRY | yeah but that setting has no effect with the timer settings all fubar |
| 18:41.02 | jONpRY | msm_arch_idle is just WFI |
| 18:41.36 | jONpRY | see what happens is arch_idle definitely gets called |
| 18:42.49 | jONpRY | the it checks this: if (sleep_time < msm_pm_idle_sleep_min_time || !allow_sleep) { |
| 18:42.57 | WisTilt2 | i stuck a printk in arch_idle and it is definitely getting called from somewhere bigtime. filled dmesg quickly. so i guess its working but we need to shorten that 20s time then? |
| 18:43.23 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
| 18:43.31 | jONpRY | but this is no good: int allow_sleep = |
| 18:43.31 | jONpRY | msm_pm_idle_sleep_mode < MSM_PM_SLEEP_MODE_WAIT_FOR_INTERRUPT && |
| 18:44.08 | jONpRY | er i guess COLLAPSE < WFI |
| 18:44.44 | jONpRY | yeah so the idle loop will always have something to do in less than 20 s. so the comparison fails |
| 18:46.04 | mgross029 | WisTilt2: I cannot enable bt but I am seeing some information in hciconfig now. So it looks like the UART is configured bu the BD Address is empty. |
| 18:46.24 | WisTilt2 | that should always test true since we're setting sleep mode to 1 though, which is < WFI |
| 18:46.53 | WisTilt2 | mgross029: bdaddr should have the BT mac addy? mine does. are you seeing hci0 now? |
| 18:46.56 | mgross029 | I lost my usb connectivity because of my POS XP workstation so I can't pull logs right now. Only Term emulator at the moment |
| 18:47.15 | jONpRY | yeah allow sleep is fine. i was confused. byt sleep_min_time needs to be way way smaller |
| 18:47.59 | WisTilt2 | yeah im dropping it to 1s for kicks and added printks throughout the logic to see where its going. this should be interesting |
| 18:48.15 | jONpRY | i don't think thats low enough |
| 18:48.29 | jONpRY | could potentially need to be sub 10 millisecond |
| 18:48.35 | mgross029 | WisTilt2: hciconfig shows 00:00:00:00:00:00. Just tried hciattach and I'm locked. |
| 18:49.11 | WisTilt2 | ill try 5ms |
| 18:50.24 | WisTilt2 | mgross029: my hciconfig shows 00:23:76:A5:E5:6A |
| 18:50.31 | WisTilt2 | strange |
| 18:50.53 | mgross029 | Yeah I agree. That is strange |
| 18:51.34 | *** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring) |
| 18:57.24 | detule | detule pushed 871 commits to linux-msm-rhod/linux-msm-rhod:htc-msm-3.0.0-rc1 |
| 18:57.59 | jONpRY | so why didn't 39 crash running angry birds when oom was broke? |
| 18:58.22 | detule | there's another android shrinker ashmem |
| 18:58.27 | detule | ashmem was also defunct in 3.0 |
| 18:58.44 | jONpRY | oh |
| 19:13.40 | *** join/#htc-linux balans (~server@82-170-217-205.ip.telfort.nl) |
| 19:16.33 | *** join/#htc-linux hasi (hasi@mausi.hasicon.org) |
| 19:19.29 | WisTilt2 | jONpRY: as i look at this, if this time is in ns then 20000000 would be 20ms not 20 seconds |
| 19:22.15 | jONpRY | thats possible |
| 19:30.45 | detule | i have to say, 3.0 is winning me over |
| 19:32.44 | jONpRY | thats good news. would be cool to switch to 3.0 and start a 3.1 |
| 19:33.38 | detule | i think you need to get your rhod210 for that, i have no idea what you two did in the early parts to get it off the ground |
| 19:34.06 | jONpRY | the 39 port was ugly because we were so far behind |
| 19:34.18 | jONpRY | but i probably put like a days work into 3.0 |
| 19:34.50 | detule | need to figure out what you did outside of /mach-msm |
| 19:35.18 | detule | i am sure quite a bit |
| 19:36.47 | jONpRY | most of it was not a big deal |
| 19:48.01 | jONpRY | just a lot of hulala to find the one liner needed |
| 20:03.37 | WisTilt2 | jONpRY: pm.c line 388 t2 is zero no? line 404 t1 - t2 will be negative since t2 doesn't get a value until later at line 458. or am i missing something |
| 20:04.36 | WisTilt2 | nm im backwards |
| 20:05.11 | WisTilt2 | 5ms didn't crash anything but i need a way to measure how much time we spent in idle_collapse to know |
| 20:05.57 | jONpRY | it should put some stuff in pm_stats |
| 20:06.33 | WisTilt2 | yeah that's where im trying to setup a printk to show the idle time so that sound reasonable to you? |
| 20:07.49 | jONpRY | the proc entry |
| 20:07.50 | WisTilt2 | looks like i can printk the whole stat table |
| 20:07.59 | jONpRY | should tell you how much time is spent in what sleep state |
| 20:08.06 | jONpRY | and it makes buckets |
| 20:08.42 | jONpRY | <PROTECTED> |
| 20:09.01 | WisTilt2 | :) guess i should have looked there first |
| 20:09.08 | WisTilt2 | yeah that has what i need thanks |
| 20:09.43 | WisTilt2 | tons of idle requests but no idle sleeps at all |
| 20:10.07 | WisTilt2 | looks to be idle about 95% of the time |
| 20:11.13 | WisTilt2 | we dont have an idle collapse stat table |
| 20:15.05 | mgross029 | WisTilt2: /sys/class/bluetooth is there but not hci0. Just an fyi. |
| 20:17.07 | WisTilt2 | mgross029: was this after you tried to enable BT from userland? |
| 20:17.15 | WisTilt2 | it has to be spawning that |
| 20:17.49 | mgross029 | Ah so dynamically created when enabled correct? |
| 20:17.58 | WisTilt2 | yep |
| 20:18.13 | mgross029 | Was not able to enable so that was probably why then. |
| 20:18.23 | WisTilt2 | you'll see rfkill ON then hci0 appears |
| 20:18.37 | WisTilt2 | the check mark wont come on but hci0 should spawn |
| 20:19.14 | WisTilt2 | userland still doesn't get notified that we actually turned it on for whatever reason but the bt radio does come up |
| 20:19.17 | *** join/#htc-linux dr1337 (78931828@gateway/web/freenode/ip.120.147.24.40) |
| 20:19.28 | WisTilt2 | bdaddr should have your mac also after you do this |
| 20:19.44 | *** join/#htc-linux GNUtoo|laptop (~gnutoo@host246-111-dynamic.48-82-r.retail.telecomitalia.it) |
| 20:20.08 | mgross029 | Hmm ok. I'll try more a little later then if I have some time. Have to get back to rl. :p |
| 20:20.25 | mgross029 | Thanks |
| 20:20.33 | WisTilt2 | ok np, let me know, this is driving me buggy |
| 20:23.58 | mgross029 | Same here. I was seeing bdaddr before now I can't and hciattach keeps killing my phone. Very strange |
| 20:24.53 | *** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl) |
| 20:32.28 | dr1337 | hey is anyone working on msm7k from ICS? |
| 20:33.02 | *** join/#htc-linux Alex[sp3dev] (~alexander@178.176.173.54) |
| 20:49.51 | *** join/#htc-linux jONpRY (~jon@unaffiliated/jonpry) |
| 20:59.36 | WisTilt2 | jONpRY i'll have idle stuff working nicely shortly. problem was static variables were only int so that sleep_min_time always was zero and nothing ever tested true where it needed to. i have all kinds of things happening now during idle. |
| 21:00.17 | jONpRY | hasn't even blown up yet? |
| 21:00.48 | WisTilt2 | nope but im blowing out dmesg with all these printk's in the idle loop so removing all that now. |
| 21:15.54 | jONpRY | all irq's pretty much have to have wake enabled |
| 21:35.38 | *** part/#htc-linux Alex[sp3dev] (~alexander@178.176.173.54) |
| 21:50.28 | *** join/#htc-linux MN-- (~yaaic@5e0dd1f9.bb.sky.com) |
| 21:57.31 | detule | WisTilt2, when you get a chance can you check on the 3.0 autobuild |
| 21:58.51 | WisTilt2 | you mean mine? is it not building 3.0 correctly? |
| 22:00.22 | detule | no, didn't you need to change something about the way you/it's build 39 |
| 22:00.36 | detule | perhaps it's running into the same issues with the 3.0 build script |
| 22:00.47 | WisTilt2 | yeah the config. ill fix it hang on |
| 22:02.27 | WisTilt2 | ok give it a few mins and it should build it |
| 22:02.35 | detule | thanks |
| 22:03.31 | WisTilt2 | gotta run an errand bbl |
| 22:18.07 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:18.07 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:18.23 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:18.23 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:18.42 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:18.42 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:19.02 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:19.02 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:19.19 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:19.39 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:19.59 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:19.59 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:20.19 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:20.19 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:20.39 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:21.04 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:21.22 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:21.22 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:23.55 | *** join/#htc-linux Regulator (~duffman@unaffiliated/regulator) |
| 22:29.36 | jONpRY | oh joy |
| 22:29.59 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:30.19 | *** join/#htc-linux LTxda (~anon@c-98-194-107-10.hsd1.tx.comcast.net) |
| 22:30.19 | *** join/#htc-linux LTxda (~anon@unaffiliated/ltxda) |
| 22:35.36 | *** join/#htc-linux avinashhm (~avinash-h@192.94.92.11) |
| 23:02.07 | *** join/#htc-linux XirXes (~XirXes@67-2-15-211.slkc.qwest.net) |
| 23:17.22 | *** join/#htc-linux swc|666 (~gecko@unaffiliated/swc666/x-4934821) |
| 23:24.34 | *** join/#htc-linux d3tul3 (~detule@pool-96-234-141-27.bltmmd.east.verizon.net) |
| 23:39.52 | *** join/#htc-linux Bry8Star (~Bry8Star@gateway/tor-sasl/bry8star) |
| 23:44.35 | *** join/#htc-linux BabelO (~fcr@AMontpellier-553-1-44-52.w90-48.abo.wanadoo.fr) |
| 23:44.35 | *** join/#htc-linux BabelO (~fcr@unaffiliated/babelo) |
| 23:59.20 | *** join/#htc-linux arif-ali (~arif-ali@94-192-24-56.zone6.bethere.co.uk) |