| arrrghhh | i wonder if this is the complete audio death bug | 00:00 |
|---|---|---|
| arrrghhh | that would be sa-weet. | 00:00 |
| d3tul3 | that whole sequence is off patching to mCurSndDevice (-1) | 00:00 |
| d3tul3 | er i should've tested that lib before i uploaded it i guess | 00:01 |
| arrrghhh | no test like the forums | 00:01 |
| d3tul3 | note how we always set Mic=1 in set_audio_path :) | 00:04 |
| arrrghhh | classy | 00:05 |
| arrrghhh | whatever works! | 00:05 |
| *** rpierce99 has quit IRC | 00:08 | |
| arrrghhh | oh well, i'm off to bed. cya. | 00:10 |
| *** d3tul3 has quit IRC | 00:14 | |
| *** frustro has left #xdandroid | 03:04 | |
| *** ke- has joined #xdandroid | 03:25 | |
| *** kalemas has joined #xdandroid | 04:06 | |
| *** hardwalker has quit IRC | 05:25 | |
| *** Searover has joined #xdandroid | 06:39 | |
| *** d3tul3 has joined #xdandroid | 07:05 | |
| *** Searover has quit IRC | 07:31 | |
| *** Searover has joined #xdandroid | 07:31 | |
| *** Searover has quit IRC | 07:41 | |
| *** Searover has joined #xdandroid | 07:41 | |
| *** Searover_ has joined #xdandroid | 07:59 | |
| *** Searover has quit IRC | 08:01 | |
| *** Searover_ has quit IRC | 08:01 | |
| *** kalemas has left #xdandroid | 08:39 | |
| *** helicopter88 has joined #xdandroid | 08:49 | |
| *** rneese has quit IRC | 08:54 | |
| *** rneese has joined #xdandroid | 08:54 | |
| rneese | mornig | 08:55 |
| rneese | arrrghhh how oes the update | 08:55 |
| rneese | oes/goes | 08:56 |
| *** mgross029 has joined #xdandroid | 09:00 | |
| *** fishhead2567 has joined #xdandroid | 09:29 | |
| *** [7] has quit IRC | 09:38 | |
| *** TheSeven has joined #xdandroid | 09:38 | |
| xdandroid | Donation received! Thanks, Adam Goodberg, who says: You've turned a Windows 6.5 HTC Pure (a co-worker hand-me-down) into a real gem. Thank you! | 09:41 |
| *** ke- has quit IRC | 09:47 | |
| *** ke- has joined #xdandroid | 09:48 | |
| *** rpierce99 has joined #xdandroid | 09:56 | |
| *** raymonddull has joined #xdandroid | 10:02 | |
| *** raymonddull has quit IRC | 10:27 | |
| *** rpierce99 has quit IRC | 10:35 | |
| *** rpierce99 has joined #xdandroid | 10:36 | |
| d3tul3 | ok i think i have a handle on what's causing the audio crash | 10:42 |
| rneese | why a handle and not a knobe or a grip | 10:45 |
| rneese | lol | 10:45 |
| *** helicopter88 has quit IRC | 10:57 | |
| *** rneese has quit IRC | 11:54 | |
| *** rneese has joined #xdandroid | 11:58 | |
| *** d3tul3 has quit IRC | 12:17 | |
| *** Bruce_Wayne has joined #xdandroid | 12:31 | |
| *** Bruce_Wayne has quit IRC | 12:39 | |
| *** rneese_ has joined #xdandroid | 12:40 | |
| *** |Jeroen| has joined #xdandroid | 12:43 | |
| *** rneese has quit IRC | 12:44 | |
| *** rneese_ is now known as rneese | 12:52 | |
| *** as_df has joined #xdandroid | 13:10 | |
| *** rpierce99 has quit IRC | 13:12 | |
| *** rneese has quit IRC | 13:13 | |
| *** rpierce99 has joined #xdandroid | 13:13 | |
| *** rneese has joined #xdandroid | 13:15 | |
| *** raymonddull has joined #xdandroid | 13:17 | |
| *** rneese has quit IRC | 13:20 | |
| *** BadSector has joined #xdandroid | 13:20 | |
| *** rneese has joined #xdandroid | 13:21 | |
| detule | hi BadSector | 13:22 |
| BadSector | Hello detule | 13:22 |
| detule | so you can reproduce this consistently? | 13:23 |
| BadSector | yes, I have only tried this 4 to 5 times.. but every time its the same effect | 13:24 |
| BadSector | everything works great until i ignore an incoming call... | 13:24 |
| detule | alright, you have GetLogs installed | 13:25 |
| BadSector | I'm currently in progress of getting that installed.. on the fourm i had just wanted to confirm what the other guy was saying.. and figured getting set up for logs might help you all figure this out :) ... so in progess, but working at same time.. lol | 13:26 |
| detule | ok...when you have it set up, try the latest 3.1 kernel from the autobuild (i just patched something this morning)....if you still hit the crash, collect bugs, post them, and then we can see if we need to patch the lib as well | 13:27 |
| as_df | i seem to have a worse problem with the ring output on RHOD 400 -- attempting to ring will never work but sometimes crash the phone or alarm clock | 13:28 |
| as_df | but i'm on the original library. should i first try an updated library before dumping logs? | 13:28 |
| BadSector | ok, i will get busy on that now detule.. will let you know when i'm there | 13:28 |
| detule | BadSector, i have an idea of ONE manner in which acoustic is misbehaving, not sure if this is causing the all_out audio crash | 13:29 |
| detule | as_df, can you get audio otherwise | 13:29 |
| rneese | whats the link to the new 3.1 kernel | 13:30 |
| detule | rneese, stay out of this | 13:30 |
| as_df | detule: sms in and preview of ringtone works until i first get called. after i get called all back audio fails | 13:30 |
| detule | as_df, post logs before you try anything, not sure what your issue is, it doesn't quite fall within spec of the other issues....it seems there is a set of cdma rhods that fail to ring on incoming calls but i don't think it causes complete failure of audio after the fact | 13:32 |
| *** helicopter88 has joined #xdandroid | 13:36 | |
| as_df | "sifting cinders"'s post in the thread is my exact issue and pfc_m_drake has problems previewing ringtones after a ring fail as well | 13:39 |
| detule | *cinder in his post talks about his ringer working on fresh boot, you are describing something else....call it an exercise in collecting logs.... | 13:43 |
| BadSector | i'll be back | 13:56 |
| *** BadSector has left #xdandroid | 13:56 | |
| detule | back in 15 | 13:59 |
| *** BadSector has joined #xdandroid | 14:06 | |
| *** raymonddull has quit IRC | 14:10 | |
| *** |Jeroen| has quit IRC | 14:10 | |
| as_df | GetLogs.apk is taking quite a long time on "Gathering and uploading logs"... | 14:12 |
| as_df | http://pastebin.com/USiisrkW I got the relevant bits of the logcat though | 14:12 |
| BadSector | detule: first call no ring, didn't do anything to it (did not answer or ignore)... second call got ring... next call phone has stopped responding.. ie looked to be locked.. no lights, pwr button did nothing...waited a bit and well after call was done it popped up showing the call incoming again with a quick ring and phone started to respond again.. | 14:14 |
| BadSector | getting logs from this now | 14:14 |
| BadSector | actually, let me test one more call | 14:14 |
| BadSector | i see call, but no ring now | 14:14 |
| BadSector | getting logs | 14:14 |
| rpierce99 | as_df: if it hangs too long on that step, it probably hung trying to get root permissions from superuser, did you get a superuser prompt? | 14:15 |
| as_df | rpierce99: yes, and the process was running as superuser | 14:16 |
| as_df | had to su to kill it | 14:16 |
| as_df | detule: those logs are two calls. neither rang the ringtone, but the phone crashed on the second one. the second one is also the only one with W/AudioTrack( 1272): obtainBuffer timed out (is the CPU pegged?) which makes me think that's a symptom and not a cause | 14:17 |
| detule | as_df need getlogs output, there's stuff missing in dmesg that i need | 14:18 |
| detule | yes obtainbuffer is a symptom | 14:18 |
| as_df | detule: i can paste the entire dmesg | 14:18 |
| as_df | well, pastebin | 14:18 |
| as_df | i put the dmesg from the second call below the adb logcat | 14:19 |
| detule | there's stuff missing | 14:19 |
| as_df | hmmm | 14:19 |
| as_df | http://pastebin.com/raw.php?i=6LBqvzGi entire dmesg | 14:20 |
| *** raymonddull has joined #xdandroid | 14:21 | |
| detule | that's better but i don't mean to sound like a repeating record but need getlogs output, among other things it also contains the radio log which is useful when dealing with incoming calls | 14:22 |
| as_df | yeah, getlogs is crashing | 14:22 |
| as_df | i'll emulate a getlogs call | 14:22 |
| rpierce99 | the commands that getlogs runs are listed in the post, but it's much easier if you just get it working, if you can get me the stack trace from the crash i can build you a new version | 14:23 |
| detule | ^listen to the man | 14:24 |
| arrrghhh | lol | 14:25 |
| arrrghhh | as_df, you have the newest getlogs? | 14:25 |
| arrrghhh | there was an older apk that didn't play nice with gb. | 14:25 |
| as_df | rpierce99: it just pegs the cpu | 14:25 |
| rpierce99 | that's true, i assumed he was running the bundled | 14:25 |
| as_df | there's a bundled getlogs? i got it from the thread | 14:25 |
| rpierce99 | it's in androidapps | 14:25 |
| arrrghhh | AndroidApps\Other | 14:25 |
| rpierce99 | and depending on when you pulled it from the thread and what post you pulled it from you could have gotten the old one | 14:26 |
| arrrghhh | bundled version should be the same as what's on the thread (AFAIK) | 14:26 |
| arrrghhh | true. | 14:26 |
| detule | BadSector, this with the bundled libaudio.so or did you change it to one I posted in the thread? | 14:27 |
| rpierce99 | that's an interesting feature idea, somehow allow for a list of files to be input which spits out the ls -l output or md5 hash on those files | 14:28 |
| *** corein has joined #xdandroid | 14:29 | |
| detule | md5sum system/lib ? :) | 14:29 |
| arrrghhh | oh yea | 14:30 |
| arrrghhh | that'll run REAL fast | 14:30 |
| arrrghhh | lol | 14:30 |
| helicopter88 | let me run it! | 14:31 |
| as_df | rpierce99: updated getlogs. OOMs now | 14:31 |
| as_df | http://pastebin.com/jZH6xebJ | 14:32 |
| helicopter88 | doesn't take a lot | 14:32 |
| helicopter88 | 3-5 seconds | 14:32 |
| arrrghhh | i guess that's not bad | 14:32 |
| arrrghhh | i was thinking it would md5sum each file, but i guess it md5's the entire directory. | 14:32 |
| helicopter88 | (partitions + overlock) | 14:32 |
| as_df | arrrghhh: what's the distinction there? same number of bytes go in, random seeks are fast on flash | 14:32 |
| arrrghhh | eh? | 14:33 |
| as_df | arrrghhh: to "md5sum an entire directory" you need to read all the files in there, so it would take roughly the same amount of time to md5sum each file individually (minus outputting each individual md5sum) | 14:33 |
| BadSector | detule: .. i did not add the libaudio.so that was previously on the fourm... i just was using the 3.1? sorry, possibly didn't understand what you wanted me to do.. this was a straight test using the 3.1 | 14:33 |
| arrrghhh | as_df, i dunno. | 14:33 |
| detule | BadSector, you did great, let me get this straight this log you posted, just the ringer died, did you test playing sounds say elsewhere? | 14:34 |
| arrrghhh | <rneese> whats the link to the new 3.1 kernel <---- holyfuckingshit read the forums. | 14:34 |
| BadSector | no, tested no other sounds... i possibly did get a text message in between that i did get a sound for | 14:35 |
| detule | excellent | 14:35 |
| helicopter88 | arrghhh,takes 3.19 seconds | 14:36 |
| detule | ok try that 3.1 kernel with the lib in here http://forum.xda-developers.com/showpost.php?p=21180563&postcount=84 | 14:36 |
| arrrghhh | i guess not bad. | 14:36 |
| BadSector | detule: ... it was very strange... if you scroll up I described exactly what happened... phone kind of locked up.. would not respond at all.. but then woke back up and worked | 14:36 |
| BadSector | ok detule... it will be a few mins.. have to do some office work lol | 14:36 |
| arrrghhh | helicopter88, just would be difficult if we md5 the entire directory to then pinpoint what's different... but i guess if anything's different that's a red flag. | 14:36 |
| detule | BadSector, there's an outstanding bug where on certain incoming calls it is very slow to bring up the inCall screen | 14:36 |
| arrrghhh | detule, you should fix that. | 14:36 |
| BadSector | ok | 14:37 |
| detule | i think WisTilt2 is on that | 14:37 |
| arrrghhh | haha | 14:37 |
| arrrghhh | i think it's another uniquely CDMA issue. | 14:37 |
| rpierce99 | as_df: download the version from post #1 in the thread, it replaces the usage of replaceAll in that stack trace with just replace( which might not have such a problem with long logs | 14:40 |
| as_df | rpierce99: that one just spins at 100% cpu | 14:40 |
| rpierce99 | when did you download it? | 14:40 |
| as_df | just now | 14:40 |
| as_df | well, today | 14:40 |
| rpierce99 | ok, i updated it jan 5 | 14:41 |
| as_df | -rw-r--r-- 1 patrick patrick 21625 Jan 10 12:37 /home/patrick/Downloads/GetLogs.apk | 14:41 |
| arrrghhh | md5sum it as_df | 14:42 |
| arrrghhh | i'll md5 mine | 14:42 |
| rpierce99 | hm, never had any problems with the logs being too long and running out of memory but i guess if your phone has been up a long time and your logs are long, and maybe your system is low on memory, there might be a problem there | 14:42 |
| as_df | 424eb5ef15dae195e336001790368596 /home/patrick/Downloads/GetLogs.apk | 14:42 |
| rpierce99 | is that the bundled or post 1? | 14:42 |
| as_df | rpierce99: the phone's only been up for like 20 mins | 14:42 |
| as_df | post 1 | 14:42 |
| as_df | i haven't actually tried bundled | 14:42 |
| rpierce99 | the log you gave me wasn't from post 1 | 14:43 |
| as_df | right | 14:43 |
| as_df | that one was from later in the thread | 14:43 |
| as_df | post 1 was the original "just does nothing for 10 minutes while pegging the CPU" problem | 14:43 |
| arrrghhh | as_df, same md5 as the bundled version. | 14:44 |
| rpierce99 | post 1 should be 243f8b1db7826d1f2ef671c5f53d0a5c | 14:44 |
| rpierce99 | hm its not | 14:45 |
| arrrghhh | lol | 14:46 |
| rpierce99 | the one i have on my hard drive is 243... | 14:46 |
| arrrghhh | odd | 14:46 |
| arrrghhh | you holding out on us? | 14:47 |
| rpierce99 | didn't think so | 14:47 |
| arrrghhh | although, the bundled getlogs works fine for me. | 14:47 |
| detule | i never mask any strings so i don't have these problems | 14:47 |
| rpierce99 | weird, the one on my hard drive says updated dec 19 | 14:47 |
| as_df | i didn't mask any strings either :/ | 14:47 |
| rpierce99 | it masks some things automatically | 14:48 |
| rpierce99 | i guess i could make that optional | 14:48 |
| rpierce99 | or i could do the file operations is chunks | 14:48 |
| rpierce99 | so i don't have to read the whole log into memory | 14:48 |
| arrrghhh | i figured the logs would fill the buffers before it gave your app any trouble. | 14:49 |
| rpierce99 | as did i | 14:49 |
| arrrghhh | some of the automatic masks are really quite nice. | 14:49 |
| arrrghhh | one of the main reasons i really like using the app. that, and overall laziness. | 14:49 |
| detule | why is it allocating 10MB | 14:49 |
| detule | that's huge, or are the logs really that big | 14:49 |
| rpierce99 | as_df does just a file work? | 14:50 |
| as_df | i disabled facebook's uploadmanager to free up some ram and rebooted | 14:50 |
| as_df | it works now | 14:50 |
| as_df | http://pastebin.com/AZh5E9z6 is the log from one phone call detule | 14:51 |
| rpierce99 | i doubt we have 10mb of buffers set aside for logs between logcat and dmesg | 14:51 |
| as_df | after the phone call all speaker operations fail | 14:51 |
| arrrghhh | i blame facebook. | 14:51 |
| arrrghhh | yea 10mb of text is gigantor. | 14:51 |
| rpierce99 | but i would assume dalvik tries to get memory in chunks larger than it needs | 14:51 |
| rpierce99 | so as not to have to go back for more | 14:51 |
| as_df | yeah, amortized O(1), but that still means the request was on order 5 MB | 14:53 |
| detule | k as_df try 3.1 and then post the log | 14:55 |
| rpierce99 | I allocate a stringbuilder, fill it with lines from the log output until the logs are over, and then allocate a string from the .toString() method of the stringbuilder, then run replace( on that string several times to perform the masking | 14:56 |
| rpierce99 | because i never destroy the stringbuilder i'm guessing i'm using twice the necessary space | 14:57 |
| rpierce99 | i don't know how efficient the internals of the replace method are | 14:57 |
| *** rneese has quit IRC | 15:00 | |
| detule | hm...life wold be easier if i could recreate this on my device....i am on a build that's a few commits behind from gbx0b perhaps those make a difference | 15:00 |
| *** rneese has joined #xdandroid | 15:05 | |
| as_df | detule: ringtone still not functional but doesn't break after failing to ring anymore | 15:06 |
| as_df | logs generating | 15:06 |
| detule | ok try this as well http://forum.xda-developers.com/showpost.php?p=21180563&postcount=84 | 15:07 |
| *** Siftingcinders has joined #xdandroid | 15:08 | |
| as_df | completely unrelated: also my phone seems to tend to get stuck on 1x. i need to flip to airplane mode and then back again | 15:08 |
| detule | see if arrrghhh can hook you up with the VZ radio | 15:08 |
| as_df | http://pastebin.com/tcGJQVCS is the 3.1 without changes | 15:09 |
| as_df | but i'm on sprint? or is this not vz-specific | 15:09 |
| detule | ok no more obtainBuffer that's good | 15:10 |
| rpierce99 | i'm using the sprint radio, but arrrghhh is on sprint and uses the verizon radio with no issues, it's just a cdma radio, verizon just has the newest | 15:10 |
| Siftingcinders | I've never really had an issue with the Sprint radio either | 15:11 |
| as_df | i get stuck in 1x if i ever use it :/ | 15:12 |
| detule | i used to get erratic 1x behavior on the sprint radio, perhaps it's placebo but with the VZ radio it doesn't do that (as much) | 15:12 |
| Siftingcinders | Hay detule, I made it | 15:14 |
| detule | hey thanks for dropping by on the party | 15:14 |
| Siftingcinders | one issue I just thought of, I'm using my phone as a hotspot for this, so if I restart I lose Internet :/ | 15:15 |
| detule | i was going to have you try the 3.1 kernel but I am not even sure if the hotspot feature works with that | 15:16 |
| as_df | detule: works after the first call | 15:18 |
| Siftingcinders | Well, I have the zimage and modules already to go, I just need to boot into it from froyo | 15:18 |
| as_df | detule: the first call also is really buggy | 15:18 |
| as_df | won't unlock the screen | 15:18 |
| detule | disregard the first call, i think i am not init-ing the audio components correctly in 3.1....play a sound through settings->sound first | 15:19 |
| as_df | okay rebooting to start from clean slate | 15:19 |
| detule | Siftingcinders, i made some changes this morning, make sure you have the latest one from the autobuild | 15:20 |
| Siftingcinders | ah | 15:20 |
| as_df | first call still fails | 15:26 |
| as_df | second+ calls seem to succeed | 15:27 |
| Siftingcinders | detule: in 3.1, do you want me to try to crash the audio, then get logs? | 15:28 |
| as_df | including after decline and accept. log is generating | 15:29 |
| detule | Siftingcinders, sure | 15:29 |
| detule | as_df, so baring that first call everything else is functional? | 15:29 |
| as_df | yeah | 15:29 |
| as_df | i can't get it to break | 15:29 |
| detule | alright | 15:30 |
| as_df | http://pastebin.com/CUim7UNM is the log. was there a combination of kernel and libaudio problems? | 15:30 |
| as_df | i'll miss the 3.5mm jack on 3.1, but i am much happier with a phone that doesn't crash when people call it | 15:30 |
| detule | it'll make it to .27 just need to hunt down emwe | 15:31 |
| detule | the obtainbuffer issue is userland | 15:32 |
| *** raymonddull has quit IRC | 15:33 | |
| detule | i added some robustness in the kernel to check whether userland is asking for something stupid, and this seems to have resolved it | 15:33 |
| detule | the issue with no ringtone, but everything else functions is a mystery | 15:33 |
| detule | (to me) | 15:33 |
| detule | rpierce99, and arrrghhh also have 400s and they don't see this | 15:34 |
| detule | i on the other hand have never had the ringer work on an incoming call with the stock libacoustic | 15:34 |
| detule | i thought my device was broken | 15:34 |
| as_df | pfc_m_drake didn't have the problem solved with new libaudio and 3.1 in the thread though | 15:37 |
| detule | i patched 3.1 this morning | 15:37 |
| as_df | ah | 15:38 |
| detule | it might still end up not being the final solution, but glad I didn't waste your time at least | 15:39 |
| Siftingcinders | I'm on 400, and i only had the ringer work during one boot about halfway through... I can't remember what was happening that made it start working ( if anything) | 15:39 |
| Siftingcinders | Okay, I'll get back to you if I find anything detule | 15:42 |
| detule | sure, thanks | 15:42 |
| *** Siftingcinders has quit IRC | 15:43 | |
| as_df | is there a git repository for your 3.1, detule? | 15:44 |
| as_df | (where does the autobuild build from?) | 15:44 |
| detule | .39 3.0 and 3.1 are built from https://gitorious.org/linux-msm-rhod/linux-msm-rhod | 15:44 |
| arrrghhh | as_df, all our poop is open. | 15:46 |
| arrrghhh | except for the poop that is closed, by usually qualcomm. | 15:46 |
| detule | as_df, btw it's not mine | 15:47 |
| mgross029 | detule: kexec enabled in 3.1? | 15:48 |
| as_df | arrrghhh: also what's this about a VZ radio? i'm having issues with getting stuck on 1x on my sprint phone | 15:48 |
| arrrghhh | as_df, so put the vz radio on | 15:48 |
| arrrghhh | i'm on it, works great. | 15:48 |
| arrrghhh | it's pretty much necessary on NAND. just helpful on HaRET | 15:48 |
| detule | mgross029, yes | 15:48 |
| *** BadSector has left #xdandroid | 15:48 | |
| mgross029 | Cool! Thanks! | 15:48 |
| mgross029 | Checking out your new sound tweaks. | 15:49 |
| detule | you never had any issues with the ringer on your rhod500 | 15:49 |
| detule | they will likely not do anything for you | 15:49 |
| arrrghhh | lol | 15:50 |
| as_df | arrrghhh: where do i get the vz radio? | 15:50 |
| arrrghhh | as_df, the forums...? | 15:50 |
| arrrghhh | the radio thread would be a good place to look. i'm guessing you want a hand-fed link? | 15:50 |
| *** BadSector has joined #xdandroid | 15:52 | |
| *** BadSector has left #xdandroid | 15:52 | |
| mgross029 | detule, Actually I do have issues from time to time. Ringer just doesn't ring sometimes or becomes in audible. Just turned on vibrate always to avoid. | 15:52 |
| detule | as_df, still around | 16:10 |
| as_df | detule: yep | 16:10 |
| detule | you in airplane mode in winmo? | 16:10 |
| as_df | detule: yep | 16:11 |
| detule | ok | 16:12 |
| detule | nevermind | 16:12 |
| *** f00bar80 has joined #xdandroid | 16:20 | |
| f00bar80 | no news about a stable GB release soon ? | 16:20 |
| helicopter88 | you can use the acoustic one | 16:22 |
| f00bar80 | helicopter88, is it considered stable ? | 16:30 |
| rpierce99 | as stable as anything else this project has ever released | 16:30 |
| helicopter88 | It works very well | 16:30 |
| f00bar80 | what about Iceacream sandwitch | 16:35 |
| rpierce99 | very unlikely | 16:36 |
| detule | as_df this is from your log right before the first ring 01-10 14:25:55.141 E/RingtoneManager( 1385): Failed to open ringtone null | 16:39 |
| as_df | which log is that | 16:39 |
| detule | CUim7UNM | 16:40 |
| arrrghhh | rpierce99> [01/10 14:30] as stable as anything else this project has ever released | 16:40 |
| arrrghhh | this | 16:40 |
| as_df | there was one log where i did screw that up but the ringtone has been fixed | 16:40 |
| arrrghhh | f00bar80, ICS may happen, but it won't be anytime soon. | 16:40 |
| as_df | hrm, that one still has that? | 16:40 |
| arrrghhh | considering how many native devices have it? | 16:40 |
| as_df | that is odd | 16:40 |
| detule | it's whatever latest log you sent me, you have a stock ringtone or an mp3? | 16:40 |
| as_df | a mp3 | 16:40 |
| arrrghhh | as_df, please test with built in rings. | 16:40 |
| detule | it's possible that mediascanner hasn't finished scanning your directories yet | 16:41 |
| arrrghhh | you're adding layers of complexity here that are not necessary. | 16:41 |
| as_df | hm, okay, switching to stock | 16:41 |
| arrrghhh | just to level the testing field as it were. | 16:42 |
| f00bar80 | is there a way to send sms from terminal ? | 16:43 |
| *** mgross029 has quit IRC | 16:45 | |
| detule | also give it a healthy minute or two with the screen on before you call the device | 16:46 |
| detule | i have no idea what that error is about | 16:46 |
| as_df | yeah changing to stock ringtone still gives ringtone null. now rebooting again but not calling as soon as powerup | 16:50 |
| arrrghhh | f00bar80, doubtful. that sounds painful. | 16:50 |
| arrrghhh | f00bar80, that's really a question for native devices, not necessarily about the port. just sayin you should probably search around there first if you have questions like that which apply to all of android. | 16:51 |
| f00bar80 | arrrghhh, i've tried am start -a android.intent.action.SENDTO -d sms:012222333 --es sms_body HELLO --ez exit_on_sent true , but this didn't send it | 16:57 |
| arrrghhh | why would you do such horrible things? | 16:58 |
| arrrghhh | there's these app thingys that make sending messages pretty easy. there's even a touchscreen interface built just for it! | 16:58 |
| as_df | detule: works | 17:00 |
| as_df | with both mp3 and stock ringtone | 17:01 |
| as_df | (letting the phone 'warm up' instead of calling as soon as i see the lock screen) | 17:01 |
| f00bar80 | just asking , kind of coding an automation script ... the code above .. how to trigger the view's onClick/onEditorAction automatically | 17:01 |
| arrrghhh | f00bar80, really a native Android problem/question. | 17:02 |
| arrrghhh | i'd search/bug those guys... | 17:02 |
| arrrghhh | as_df, were you calling immediately? the phone isn't up immediately. | 17:02 |
| detule | as_df no more ringtone null? | 17:02 |
| arrrghhh | i usually wait until the service bars go green before doing anything | 17:02 |
| as_df | arrrghhh: i was calling as soon as i saw the lockscreen and the service provider string | 17:03 |
| as_df | detule: no more ringtone null | 17:03 |
| detule | alright done my deed for the day, thanks for testing i am out | 17:03 |
| arrrghhh | detule, thanks dude! | 17:03 |
| as_df | thanks! | 17:03 |
| detule | np | 17:03 |
| *** helicopter88 has quit IRC | 17:06 | |
| *** siftingcinders has joined #xdandroid | 17:29 | |
| *** ke- has quit IRC | 17:31 | |
| siftingcinders | detule, the audio does not crash. also I got a ringer on the first call | 17:34 |
| detule | cool thanks for stopping by | 17:35 |
| siftingcinders | And second call | 17:36 |
| siftingcinders | do you one logs? | 17:37 |
| siftingcinders | want* | 17:37 |
| detule | not if everything is working fine | 17:39 |
| siftingcinders | One issue, speaker phone audio is dead the first time you turn it on in a call. If you toggle it off and on again, it works. | 17:42 |
| siftingcinders | I had sortof the same issue in .27 | 17:43 |
| detule | this on every call? | 17:44 |
| siftingcinders | on the 2 i tried | 17:44 |
| detule | wait let me get this straight, you get a phone call it rings fine, you answer it, routes fine to the headset, you press the speaker soft key nothing comes out? | 17:46 |
| as_df | detule: just tried that sequence right now | 17:47 |
| as_df | works fine | 17:47 |
| as_df | well, i initiated the call | 17:47 |
| arrrghhh | yea that's a new one siftingcinders | 17:47 |
| as_df | hrrrm i'll try the receiving | 17:47 |
| arrrghhh | speakerphone has "always" worked. | 17:47 |
| as_df | detule: other direction does not work | 17:48 |
| as_df | you have to toggle speakerphone twice | 17:48 |
| arrrghhh | orly | 17:48 |
| arrrghhh | that's new. | 17:48 |
| as_df | okay | 17:49 |
| as_df | you don't have to toggle twice | 17:49 |
| as_df | waiting five seconds | 17:49 |
| as_df | then turning on the speakerphone works | 17:49 |
| siftingcinders | ohh I always try within that 5 second window | 17:49 |
| as_df | did this work before? i didn't really use the speakerphone much before | 17:50 |
| detule | i can't get it to misbehave 5 seconds or not | 17:50 |
| siftingcinders | http://pastebin.com/bCNtfBUW --Last call I turn speakerphone on after one or 2 seconds | 17:51 |
| as_df | huh that is really weird | 17:52 |
| as_df | if i turn the speakerphone on immediately, the speaker actually works for a tiny bit | 17:52 |
| as_df | and then cuts out | 17:52 |
| siftingcinders | me too | 17:52 |
| arrrghhh | son of a biscuit. | 17:53 |
| arrrghhh | you guys are right -_- | 17:53 |
| arrrghhh | i don't remember this ever being an issue, must be an acoustic thing... | 17:53 |
| detule | arrrghhh, which libs are you running stock all the way? | 17:54 |
| *** f00bar80 has quit IRC | 17:57 | |
| arrrghhh | yessir | 17:58 |
| arrrghhh | stock libs, 3.1 | 17:58 |
| arrrghhh | probably an "old" 3.1 at this point too :P | 17:58 |
| detule | that's helpful older than yesterday? | 17:58 |
| arrrghhh | sorry | 17:58 |
| arrrghhh | uhm... lemme look | 17:58 |
| arrrghhh | it was yesterday i downloaded it (i think) | 17:58 |
| arrrghhh | modules-3.1.6-g50f5d39.tar.gz | 17:59 |
| detule | cool that saved me some head scratching | 17:59 |
| arrrghhh | i'm guessing i should update? | 18:01 |
| rpierce99 | detule is just glad it's not in his new changes | 18:01 |
| arrrghhh | lol | 18:01 |
| *** siftingcinders has quit IRC | 18:01 | |
| detule | he's got it right, the last two commits were acoustic related and you are older than that so i didn't break anything | 18:01 |
| detule | i can't reproduce it... | 18:02 |
| detule | like ten times out of ten | 18:02 |
| detule | speaker works every time | 18:02 |
| rpierce99 | it sounds like the accepting of the phone call and routing audio takes some time, but you can jump the line with the speakerphone button and reroute the audio, but once the main answer process catches up and executes its audio routing it sets it back to regular mode | 18:02 |
| *** Siftingcinders has joined #xdandroid | 18:03 | |
| *** fishhead2567 has quit IRC | 18:03 | |
| Siftingcinders | Want me to do anything else detule? | 18:07 |
| as_df | (joking) solution: add a 2 second lag to the speakerphone button | 18:07 |
| rpierce99 | as_df: hey that's how we fix shit around here | 18:08 |
| detule | Siftingcinders, if you can get me a log with a single phone call where you experience the issue that would be helpful | 18:09 |
| as_df | it'll work if there's a good way to look at the call timer and only do that when it's the start of the call | 18:10 |
| rpierce99 | solving a race condition by adding an artificial delay is never the right way to approach it, but then again neither is always setting mic = 1 :) | 18:13 |
| detule | there shouldn't be a race with powering on the speaker, at least in 3.1 the speaker power calls are on a workqueue | 18:15 |
| rpierce99 | well it sounds like the normal "answer call" routine sets speaker off, but does so 5 seconds after the call is answered | 18:16 |
| rpierce99 | that could be completely false though | 18:16 |
| detule | i can | 18:17 |
| detule | t reporduce it therefore it's not happening :) | 18:17 |
| rpierce99 | i've used that line several times over the last week or so on a project for work :) | 18:18 |
| rpierce99 | in my case it's more like "only you can reproduce it, no one else can, so it's your computer, not the code" | 18:18 |
| Siftingcinders | detule: should i do it on a clean data.image | 18:18 |
| detule | no need | 18:19 |
| detule | just reboot, receive call speaker on (test if it's working) speaker off speaker on speaker off end call collect log | 18:20 |
| *** corein has quit IRC | 18:20 | |
| detule | so you fellas can reproduce this 100% of the time? | 18:21 |
| as_df | yeah | 18:23 |
| as_df | someone took the other phone that i was testing with back | 18:23 |
| as_df | so i haven't dumped a log yet | 18:23 |
| detule | my money is on this https://gitorious.org/xdandroid/hardware_msm7k/commit/789281cf4ce98d29df64563fb62bc5ca523b0318 that's about the only commit that i don't have in my system | 18:27 |
| Siftingcinders | http://pastebin.com/XjYdqgHG -- called voicemail, speakerphone on ( audio cut out after a second), speakerphone off, on ( works), end call | 18:39 |
| Siftingcinders | ahh no ringer on this boot | 18:43 |
| *** mes__ has joined #xdandroid | 18:43 | |
| mes__ | any one know why su doesn't seem to work anymore? the one in /system/xbin/su ? | 18:47 |
| Siftingcinders | detule, pm me if you need more done! | 18:48 |
| *** Siftingcinders has quit IRC | 18:48 | |
| detule | it's ok thanks it's not a kernel issue userland is actually asking it to stop | 18:49 |
| detule | no idea why i almost want to pull libaudiopolicy from my device to see if it will rectify the issue on yours | 18:49 |
| as_df | i'm having issues getting wifi up in 3.1. insmod: can't insert '/lib/modules/bcm4329.ko': invalid parameter with freshly extracted modules-3.1.6-g038bab8.tar.gz | 18:53 |
| mes__ | I get "Stopped (signal) su" Bus error su | 18:54 |
| as_df | dmesg doesn't seem to have anything that immediately tells me any issue http://pastebin.com/hjcFeJYd | 18:55 |
| detule | "strings /lin/modules/bcm4329.ko | grep magic" | 18:56 |
| detule | "strings /lib/modules/bcm4329.ko | grep magic" | 18:56 |
| rpierce99 | i always love looking for magic in my files | 18:58 |
| as_df | vermagic=3.1.6-g038bab8 preempt mod_unload ARMv6 | 19:00 |
| as_df | __mod_vermagic5 | 19:00 |
| detule | no idea works fine here | 19:02 |
| rpierce99 | could be arrrghhhs firmware junk doesn't work :) | 19:03 |
| detule | as_df, try that audio bug with http://dl.dropbox.com/u/38520332/libaudiopolicy.so in system/lib | 19:06 |
| detule | also rpierce99 is suggesting to check whether md5sum /system/etc/wifi/bcm432x/bcm4325-rtecdc.bin is bca2cb274070423a7ea332c1e8cee115 | 19:07 |
| as_df | no, still same symptoms | 19:14 |
| as_df | md5sum matches | 19:15 |
| detule | as_df, still same audio with that *.so | 19:15 |
| as_df | detule: yeah | 19:23 |
| *** pfc_m_drake has joined #xdandroid | 19:42 | |
| *** pfc_m_drake has quit IRC | 19:48 | |
| mes__ | I see, su isn't su anymore, ie. doesn't work from cli | 19:50 |
| *** pfc_m_drake has joined #xdandroid | 19:50 | |
| *** hardwalker has joined #xdandroid | 19:51 | |
| arrrghhh | mes__, that's some issue you created all on your own. | 19:55 |
| arrrghhh | SU/the app works great here. | 19:55 |
| arrrghhh | ...new data.img? at least to test the theory. | 19:55 |
| mes__ | from adb? | 19:56 |
| arrrghhh | rpierce99, my firmware logic is flawless :P | 19:56 |
| pfc_m_drake | Hey guys - Detule asked me to drop by and help him with some audio testing. | 19:56 |
| arrrghhh | mes__, from everywhere. adb, apps, doesn't matter. | 19:56 |
| arrrghhh | pfc_m_drake, i *think* he got all the demons out. detule ? | 19:56 |
| arrrghhh | they were still hashing some acoustic problems about 45 mins ago. | 19:56 |
| mes__ | so you can login with adb and su from the commandline? | 19:57 |
| arrrghhh | mes__, good lord | 19:57 |
| arrrghhh | YES | 19:57 |
| arrrghhh | YES I CAN | 19:57 |
| arrrghhh | IT WORKS | 19:57 |
| arrrghhh | IT WORKS GREAT | 19:57 |
| arrrghhh | does that clear it up? :P | 19:57 |
| arrrghhh | sometimes SU doesn't pop up immediately. i have had a few times where i had to try the command again. | 19:58 |
| arrrghhh | usually that's after a reboot and i'm being impatient. | 19:58 |
| pfc_m_drake | Ok, thanks arrrghhh. I'll hang around for a while in case he still wants/needs something. As long as that's ok with you guys...I have to confess that I don't get to hang with the 'cool kids' very much in real life :P | 19:58 |
| arrrghhh | lol | 19:58 |
| arrrghhh | pfc_m_drake, pretty sure none of us are cool kids :P | 19:58 |
| arrrghhh | but you are more than welcome to hang out. | 19:58 |
| mes__ | pop up? I have only an xterm on the laptop running adb, but I will try with the GB build you posted a few days ago | 19:59 |
| arrrghhh | ... | 19:59 |
| arrrghhh | mes__, i really honestly have no freaking clue what you're talking about, but SU works great. has since forever. | 20:00 |
| arrrghhh | there was a problem on 2.3.7, but that was never publicly released. it was fixed a looong time ago. | 20:00 |
| arrrghhh | but i'm goin home, cya guys later. | 20:01 |
| mes__ | connect usb, type /android-sdk-linux_86/platform-tools/adb shell, gives $ | 20:01 |
| pfc_m_drake | Later, arrrghhh. | 20:01 |
| rpierce99 | mes__: our build is secured, the phone needs you to tap "allow" in the superuser app to give adbd permission to run as root | 20:02 |
| mes__ | yes I see that now, normally I don't touch the phone when I use adb | 20:02 |
| *** d3tul3 has joined #xdandroid | 20:12 | |
| d3tul3 | pfc_m_drake, thanks for stopping by we beat acoustic to death earlier today | 20:12 |
| mes__ | is there some way to stop the network update check in Superuser? | 20:13 |
| pfc_m_drake | No problem detule. Sorry I couldn't make it until now. Real life jobs and all...I'm sure you know how that goes. | 20:13 |
| pfc_m_drake | But I'm glad to hear you licked the problem. Honestly I felt like kind of a pain in the ass for bringing it up. | 20:14 |
| pfc_m_drake | So I'm glad you guys knocked it out. | 20:14 |
| *** mes__ has quit IRC | 20:22 | |
| d3tul3 | not sure if we knocked it out, but we made some progress, thanks to some good log collecting by yourself and some other folks in that thread | 20:25 |
| pfc_m_drake | No problem. If there's anything else you need from me, just shoot me a PM. Happy to help however I can...you guys do awesome work! | 20:26 |
| d3tul3 | cool thanks for being available…i really don't know what i am doing with this just futzing around, unfortunately the guy who wrote the acoustic code is not around much these days | 20:27 |
| pfc_m_drake | Sorry to hear that. But yeah - I'm available and I'll lend whatever hand I can. I'm going to take off for now, but feel free to look me up whenever. Catch you all later. | 20:31 |
| *** pfc_m_drake has quit IRC | 20:31 | |
| d3tul3 | as_df, around? | 20:51 |
| d3tul3 | or arrrghhh | 20:51 |
| arrrghhh | wat up playa | 20:52 |
| d3tul3 | hey you were seeing that problem with the speaker too | 20:52 |
| d3tul3 | have a minute to try a lib? | 20:52 |
| arrrghhh | sure | 20:53 |
| d3tul3 | ihttp://dl.dropbox.com/u/38520332/libhtc_acoustic.so | 20:53 |
| d3tul3 | i have to stop looking at acoustic | 20:53 |
| d3tul3 | bugs will never stop | 20:53 |
| arrrghhh | going blind? | 20:53 |
| arrrghhh | lol | 20:53 |
| d3tul3 | it's the gift that keeps on giving | 20:53 |
| arrrghhh | indeed | 20:53 |
| arrrghhh | you saw emwe get sucke din | 20:53 |
| arrrghhh | sucked in* | 20:53 |
| d3tul3 | i keep thnking "this is the last" | 20:54 |
| arrrghhh | hah. funny guy. | 20:54 |
| arrrghhh | acoustic does seem close tho. this speakerphone one is odd, i swear i've never experienced it before. | 20:55 |
| arrrghhh | a dog. getting a bah mitzvah. | 20:55 |
| arrrghhh | really? | 20:55 |
| arrrghhh | bat mitzvah? | 20:56 |
| arrrghhh | whatever | 20:56 |
| rpierce99 | d3tul3: you know this is how hyc ended up spending months on the RIL right? lol | 20:57 |
| arrrghhh | ugh. whenever sync kicks in the phone slows to an absolute crawl. | 20:57 |
| arrrghhh | rpierce99, mainly because there was just stubs for cdma. he wrote that crap from nothing. | 20:57 |
| arrrghhh | i'm still amazed he went that extra mile on mms. really was above and beyond. | 20:58 |
| d3tul3 | i can't replicate that speaker bug for the life of me | 20:59 |
| arrrghhh | oh is this the "outdated" lib from your build? | 20:59 |
| d3tul3 | no, that was libaudiopolicy which apparently doesn't matter | 20:59 |
| d3tul3 | i just saw something that could be wrong in libhtc_acoustic | 20:59 |
| d3tul3 | here i can lay it out | 20:59 |
| arrrghhh | lol ok | 20:59 |
| d3tul3 | the problem is basically lines 902 and 903 in this log that they sent me earlier today http://pastebin.com/XjYdqgHG | 21:00 |
| arrrghhh | damnit | 21:01 |
| arrrghhh | that 'i' was screwing me up | 21:01 |
| d3tul3 | you see it's settling on routing to device 13 (speaker_mic), but set_audio_path still says Speaker = 0 | 21:01 |
| arrrghhh | weird. | 21:01 |
| d3tul3 | there's quite a few things that happen between those two lines | 21:02 |
| d3tul3 | i just picked one spot where i think it's failing | 21:02 |
| rpierce99 | ah the dartboard method, good call :) | 21:02 |
| d3tul3 | in particular | 21:02 |
| d3tul3 | i changed this line https://gitorious.org/xdandroid/hardware_msm7k/blobs/gingerbread/libacoustic/libacoustic.c#line1616 | 21:03 |
| d3tul3 | - if ( (device_out == SND_DEVICE_SPEAKER) || (device_out == SND_DEVICE_PLAYBACK_HANDSFREE) ) { | 21:03 |
| d3tul3 | + if ( (device_out==SND_DEVICE_SPEAKER_MIC) || (device_out == SND_DEVICE_SPEAKER) || (device_out == SND_DEVICE_PLAYBACK_HANDSFREE) ) { | 21:03 |
| d3tul3 | - if ( (device_out == SND_DEVICE_SPEAKER) || (device_out == SND_DEVICE_PLAYBACK_HANDSFREE) ) { | 21:03 |
| d3tul3 | - if ( (device_out == SND_DEVICE_SPEAKER) || (device_out == SND_DEVICE_PLAYBACK_HANDSFREE) ) { | 21:03 |
| d3tul3 | + if ( (device_out==SND_DEVICE_SPEAKER_MIC) || (device_out == SND_DEVICE_SPEAKER) || (device_out == SND_DEVICE_PLAYBACK_HANDSFREE) ) { | 21:03 |
| d3tul3 | uh too many pastes | 21:03 |
| d3tul3 | just the first minus plus | 21:03 |
| rpierce99 | so you added SND_DEVICE_SPEAKER_MIC | 21:04 |
| d3tul3 | yes, that seems obviously wrong | 21:04 |
| rpierce99 | if device_out is snd_device_speaker_mic it would hit the else block and speaker and headset would both be false | 21:04 |
| d3tul3 | right which seems like a bug to me | 21:05 |
| rpierce99 | line 1632 echoes the device_out value, are you seeing it equal SND_DEVICE_SPEAKER_MICs value? | 21:06 |
| arrrghhh | chmod to 755? | 21:06 |
| d3tul3 | yeah that's like 903 in the pastebin | 21:06 |
| d3tul3 | it echoes 13 | 21:06 |
| d3tul3 | arrrghhh, 777 to be sure :) | 21:07 |
| arrrghhh | oh | 21:07 |
| arrrghhh | 644 | 21:07 |
| arrrghhh | 777? | 21:07 |
| arrrghhh | lol | 21:07 |
| arrrghhh | i was just matching the other lib, which is 644... | 21:07 |
| rpierce99 | and SND_DEVICE_SPEAKER_MIC is 13? | 21:07 |
| d3tul3 | yes | 21:07 |
| rpierce99 | sounds good to me | 21:07 |
| rpierce99 | not that my approval means anything | 21:07 |
| arrrghhh | alrighty. gonna rebootski and try this out | 21:08 |
| d3tul3 | arrrghhh, matching the other lib is a good idea | 21:08 |
| arrrghhh | oO i should go to the newest 3.1 | 21:08 |
| arrrghhh | anything else i should mangle? | 21:09 |
| d3tul3 | not for the purposes of testing that speaker bug i guess | 21:09 |
| arrrghhh | k | 21:09 |
| d3tul3 | probably don't need to update the kernel either | 21:09 |
| arrrghhh | well i'm going to anyways. | 21:09 |
| arrrghhh | so there. | 21:09 |
| arrrghhh | :P | 21:09 |
| arrrghhh | damn, reboot hung. | 21:10 |
| d3tul3 | in 3.1 power-off is reboot | 21:11 |
| d3tul3 | :) | 21:11 |
| arrrghhh | reboot is... freeze? | 21:11 |
| d3tul3 | i guess | 21:11 |
| arrrghhh | lol | 21:11 |
| arrrghhh | oh, wasn't that necessary for kexec for some odd rason? | 21:11 |
| arrrghhh | reason* | 21:11 |
| d3tul3 | arrrghhh, what's the word | 21:22 |
| arrrghhh | it just went green | 21:22 |
| arrrghhh | so testing commencing | 21:23 |
| arrrghhh | seems to work fine | 21:23 |
| arrrghhh | speaker went on @ 1-2s | 21:23 |
| arrrghhh | t'was functional | 21:24 |
| d3tul3 | i'll take it | 21:24 |
| arrrghhh | last time it failed on an outbound call | 21:25 |
| arrrghhh | so i figured another outbound call would be a valid test. | 21:25 |
| d3tul3 | yeah this log that guy sent me was outbound | 21:25 |
| arrrghhh | well crap man, good job. this acoustic code might be flaky, but for good reason. the logic is wrong :P | 21:25 |
| arrrghhh | so the only thing i really want to see done with acoustic is the csv's in the system image, and perhaps the overall volume tweaked higher... which i guess the csv's can help with? | 21:26 |
| arrrghhh | i guess that's 2 things. | 21:26 |
| arrrghhh | i should stop guesing. | 21:26 |
| arrrghhh | or typing... | 21:26 |
| *** BulitPruf has joined #xdandroid | 22:19 | |
| *** TheSeven has quit IRC | 22:23 | |
| *** [7] has joined #xdandroid | 22:23 | |
| *** rneese has left #xdandroid | 23:30 | |
| *** rneese has joined #xdandroid | 23:34 | |
| *** d3tul3 has quit IRC | 23:55 | |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!