IRC log for #maemo on 20150716

00:00.32KotCzarnyi meant, can i get that node working without oss emulation module
00:03.31KotCzarnyALSA lib pcm.c:2144:(snd_pcm_open_noupdate) Unknown PCM pcm.!default
00:03.36KotCzarnyfun.
00:04.39DocScrutinizer05the ! is illegal char I'd guess
00:04.52DocScrutinizer05it's a special meaning char not part of name
00:05.28KotCzarnyumkay, something is working
00:05.41KotCzarnyi even get some feedback when tapping on the mic
00:05.50DocScrutinizer05honestly, an empty device string is same like "default"
00:05.56DocScrutinizer05afaik
00:07.34DocScrutinizer05afk, cya
00:07.39KotCzarnythx, and gn
00:10.37KotCzarnyfun. device is still opened after program closed (and i did close both streams)
00:14.41DocScrutinizer05you need to free them too iirc
00:14.59DocScrutinizer05just closing isn't enough
00:15.15KotCzarnysure, but now that app closed, how do i free them?
00:15.23DocScrutinizer05or rather, closing might actually fail when you didn't free them beforehand
00:15.44DocScrutinizer05good question, prolly reload the als kernel module
00:15.47DocScrutinizer05dunno
00:15.49KotCzarnyhehe
00:15.59KotCzarnyreboot then
00:16.24DocScrutinizer05what makes you think they are still open?
00:16.57KotCzarny[ 3808.617187] EAC mode: play enabled, rec enabled
00:17.09KotCzarnyand no 'disabled' line in dmesg later on
00:17.16KotCzarnyalso, can't run app second time
00:17.22DocScrutinizer05hmm
00:18.22DocScrutinizer05no idea. actually that type alsa_dsp is new to me
00:18.39KotCzarnytold you, it's hairy
00:18.59DocScrutinizer05must be the kernel interface to ... dsp ;-)
00:19.02KotCzarnypity they couldn't write proper kernel device
00:19.27KotCzarnythat would keep to standards and failsafe from user hackings
00:20.27KotCzarnyyeah, it fails at snd_pcm_open
00:20.36KotCzarnyso i can't even reset it without the reboot
00:21.06DocScrutinizer05OpenMAX IL
00:21.20KotCzarnyyeah, 'open' 'max'
00:21.26KotCzarnyboth lies
00:21.31DocScrutinizer05http://wiki.maemo.org/Documentation/Maemo_5_Developer_Guide/Architecture/Multimedia_Domain
00:21.40DocScrutinizer05it's prolly a TI thing
00:21.52DocScrutinizer05DSP-bridge
00:22.07KotCzarnybut you are reading n900's docs
00:22.42DocScrutinizer05yes, but the OMAP2 isn't that much different in that regard
00:22.54KotCzarnyit is, driver is worse
00:22.55DocScrutinizer05afaik it also has a DSP
00:24.05DocScrutinizer05and I guess alsa_dsp is ALSA's interface to OpenMAX IL or whatever the thing is called on OMAP2
00:24.40KotCzarnyok, so how do i close the bastard properly?
00:24.45KotCzarnysnd_pcm_drop ?
00:24.52KotCzarnythen snd_pcm_close?
00:25.06DocScrutinizer05unload the kernel module omap-dsp or whatever it's called
00:25.09KotCzarnynope
00:25.11KotCzarnybuiltin
00:25.15DocScrutinizer05aah
00:25.30DocScrutinizer05check source for aplay
00:25.41DocScrutinizer05it's the reference inplementation
00:28.46KotCzarnyi think it just uses snd_pcm_close
00:29.03KotCzarnyunless there is something to error handling
00:30.36DocScrutinizer05as mentioned you need to free stuff before close will succeed
00:31.06KotCzarnyok, got disabled msg
00:31.53KotCzarnyoh, fun, it stays opened when closing app via ctrl-c
00:33.34DocScrutinizer05http://privatepaste.com/99b5d4da1f
00:35.30DocScrutinizer05from /home/jr/Desktop/twinkle-1.1/src/audio/audio_device.cpp
00:36.03*** join/#maemo beford (~beford@unaffiliated/beford)
00:36.37KotCzarnynah, it's just that author of the app didnt add snd stop func at all close cases
00:38.01DocScrutinizer05sorry, can't help any further. That's the end of my alsa 'expertise'
00:38.15KotCzarnyno worries, it almost works
00:38.30KotCzarnystill have to hack at buffers (clicking etc)
00:38.58DocScrutinizer05twinkle-1.1/src/audio/audio_device.cpp is about as much as I know about ALSA
00:39.04DocScrutinizer05nice code
00:39.30DocScrutinizer05I found the // Without the snd_pcm_hw_free, snd_pcm_close sometimes fails.
00:40.05KotCzarnyi free hw just after init
00:40.09DocScrutinizer05since nobody but me bothers to check error return code of a close()
00:40.14KotCzarnyor not
00:40.31*** join/#maemo dreamer (~henk@melange.soleus.nu)
00:42.43DocScrutinizer05and yes, ALSA docs suck in that they don't exlain when and how to use snd_pcm_hw_free() and *close() etc
00:47.40DocScrutinizer05get twinkle-1.*/src/audio/audio_device.cpp it's probably pretty instructive particularly when you like OSS
00:48.00KotCzarnyi'm tired a bit
00:48.08KotCzarnyso for now i just install signal handler
00:49.16DocScrutinizer05http://privatepaste.com/ece01709c6
00:49.47KotCzarnyi wouldnt call it simple interface
00:50.13KotCzarnybut in my case it was the missing signal handler
00:50.28KotCzarnyi'm amazed that kernel doesnt close the device on it's own...
00:50.56KotCzarnyi guess that's the reason of n8x0's audio being flaky (sometimes works, sometimes not)
00:53.08KotCzarnycan i catch -KILL ?
01:10.24ZetaRKotCzarny: SIGKILL cannot be caught.
01:10.42KotCzarnyyeah. pity
01:11.44ZetaRWell, it does give you a way to kill a program that is hanging and refusing to respond to other signals.
01:12.28KotCzarnyyeah, still, if i will ever have to kill this app via -KILL, audio device will be blocked
01:14.07ZetaRThat... sounds like really poor design.
01:14.19KotCzarnyon the nokia/ti, yes
01:14.42ZetaRYou should immediately free resources when the blocking program gets killed.
01:17.35*** join/#maemo Humpelstilzchen (erik@f054027165.adsl.alicedsl.de)
01:18.59ZetaRUnfortunately, I don't really know much about OMAP and ALSA, so I can't really help you out.
01:19.30KotCzarnynah, as i've said, it was a missing signal handler, it's fixed now (to a point)
01:19.49KotCzarnydo you know c?
01:20.34ZetaRYeah, but not real experienced.
01:20.43KotCzarny<PROTECTED>
01:20.46KotCzarnywhat does it do?
01:22.20ZetaRHm, let me check K&R. Not sure what the >> is for.
01:22.30KotCzarny>> is bit shift
01:22.36KotCzarnybut is >>= special?
01:23.03ZetaROh, well that sounds like you have an array and you are bit shifting and assigning a 1.
01:23.16KotCzarnyo.o
01:23.29KotCzarnyi wonder what the author meant by it..
01:23.38KotCzarnybut for shift you need param
01:23.46KotCzarnyas in var>>2 etc
01:25.43ZetaRHmm, maybe it is some other thing then.
01:30.26ZetaRAccording to K&R, most of the binary operators have a corresponding assignment operator, which is what >>= is. So that line should be equivalent to "process_buffer[i] = process_buffer[i] >> 1;"
01:30.42KotCzarnyuhum
01:31.03ZetaR"expr1 op = expr2" is equivalent to "expr1 = (expr1) op (expr2)"
01:31.41KotCzarnyi still don't know what app author wanted to achieve, because it only shifts by 1 bit then overwrites it next
01:32.59ZetaRUnfortunately, K&R does not contain the necessary rituals for divining the intentions of other programmers.
01:33.35KotCzarnyhttp://pastebin.com/raw.php?i=miq1Ty4j
01:33.39KotCzarnythis is the part of the code
01:33.44KotCzarnyignore  <--->
01:34.02KotCzarnyprocess_buffer is audio data, 2ch, 16bit samples
01:34.37ZetaRThat doesn't overwrite it.
01:34.39KotCzarnyif i read it correctly, it overwrites one channel with the other shifted by one bit
01:35.17ZetaROh, I thought you meant it overwrites it uselessly.
01:36.20KotCzarnyprocess_buffer data type is short
01:37.15ZetaRSo it shifts the samples in the array by one, and bitshifts the samples by one?
01:37.32KotCzarnyit's late, i might be wrong/blind
01:37.38ZetaRSorry, I need to go AFK for ~30min.
01:37.43KotCzarnyno worries
01:43.21*** join/#maemo LauRoman|Alt (~LauRoman@5-14-58-90.residential.rdsnet.ro)
01:49.30*** join/#maemo pcfe (~pcfe@2a02:8109:8080:fa4:beee:7bff:fed9:8f1b)
01:49.32*** join/#maemo pcfe (~pcfe@redhat/pcfe)
02:01.09DocScrutinizer05hey, that divides by two (>>1) and copies left into right chan
02:01.40KotCzarnyyeah, that's why i don't understand it
02:02.04DocScrutinizer05which makes sense since stereo->mono later on adds both channels and thus compensates for the divide by 2 aka rightshift
02:02.19KotCzarnyhuh?
02:02.50KotCzarnyso it's primitive stereo->mono code?
02:02.53DocScrutinizer05it's prolly mono isgnal in left chan of a stereo stream
02:03.40DocScrutinizer05this code copies left chan into right chan so it's balanced stereo, and compensates for the volume increase that causes
02:04.03KotCzarnywhat do you mean by 'balanced stereo'
02:04.06DocScrutinizer05it's rather mono->stereo
02:05.35DocScrutinizer05input is L?L?L?, output is (L/2)(L/2)(L/2)(L/2)(L/2)(L/2)
02:05.50DocScrutinizer05? is probably 0
02:05.58DocScrutinizer05or bogus data
02:06.19KotCzarnyi could understand that it overwrites right chan with quited left chan
02:06.29DocScrutinizer05a correct stereo input would be LRLRLR
02:07.22DocScrutinizer05looks like what I'd do to a headset mic input
02:07.46KotCzarnyunless your mic input is really stereo
02:07.47DocScrutinizer05which per definition is on left channel when input is stereo
02:07.57KotCzarnyor if input is line-in
02:08.08DocScrutinizer05never seen a headset with stereo mic :-)
02:08.34KotCzarnythis app comes from desktop
02:08.50KotCzarnybut ok, thanks for helping me understand the author
02:09.21DocScrutinizer05it copies left into right channel and attenuates both by 50%
02:09.26KotCzarnyyeah
02:09.42KotCzarnynow i have to tidy up the block sizes he uses
02:14.01ZetaRIt seems that DocScrutinizer05 knows the correct divining rituals we were missing. :)
02:14.20KotCzarnyor just had his powernap
02:14.21KotCzarny;)
02:15.49ZetaRI think it is something like 4am where he is.
02:18.04KotCzarnyon another note, im so gonna steal many of gtkgep effects and add them into oscp
02:18.07KotCzarnyMUAHAHAHAHA
02:21.59ZetaRSomeone just turned into Mr. Hyde.
02:22.23KotCzarnyanyway, thanks, and cya, time for my powernap
02:22.36ZetaRCya.
02:30.00*** join/#maemo RedW (~redw@89-73-179-171.dynamic.chello.pl)
03:01.48*** join/#maemo lxp (~lxp@unaffiliated/lxp)
03:19.33*** join/#maemo FlameReaper (~assassin@175.137.196.26)
03:21.03*** join/#maemo phlixi_ (~phlixi@ppp-188-174-185-241.dynamic.mnet-online.de)
03:58.14*** join/#maemo timeless (sid4015@firefox/developer/timeless)
04:10.05*** join/#maemo FlameReaper (~assassin@175.137.196.26)
04:14.48*** join/#maemo darkschneider2 (~gab@93-32-47-36.ip31.fastwebnet.it)
04:23.13*** join/#maemo FlameReaper (~assassin@175.137.196.26)
04:33.49*** join/#maemo infobot (ibot@69-58-76-73.ut.vivintwireless.net)
04:33.49*** topic/#maemo is Welcome to #maemo http://maemo.org/intro/ | N9/harmattan related questions please in #harmattan ! | "firmware" see ~tabletsdev ~combined ~emmc ~flasher sdk: ~sb | http://wiki.maemo.org/Community_Council | Source: http://wiki.maemo.org/Sources | Chanlog: http://mg.pov.lt/maemo-irclog | #maemo-ssu is where the (few) devels and maintainers of the maemo-future meet
04:40.08*** join/#maemo rm_work (rm_work@Maemo/community/cssu/rm-you)
04:46.35*** join/#maemo XDS2010 (sid1218@gateway/web/irccloud.com/x-guinigznnvihuegq)
06:53.49*** join/#maemo Haudegen (~quassel@85.124.51.57)
06:58.26*** join/#maemo Humpelst1lzchen (erik@x55b1c1e5.dyn.telefonica.de)
07:09.13*** join/#maemo Humpelstilzchen (erik@x55b16573.dyn.telefonica.de)
07:55.21*** part/#maemo L29Ah (~L29Ah@195.19.225.242)
08:09.40*** join/#maemo bugzy (~bugzy@2602:306:3bf8:b1f0:ae22:bff:fe50:c418)
08:23.46*** join/#maemo L29Ah (~L29Ah@195.19.225.242)
08:25.25*** join/#maemo LjL (~ljl@unaffiliated/ljl)
08:33.37*** part/#maemo L29Ah (~L29Ah@195.19.225.242)
08:35.57*** join/#maemo louisdk (~louisdk@94.144.63.228)
08:40.09*** join/#maemo L29Ah (~L29Ah@195.19.225.242)
08:41.30*** join/#maemo phlixi (~phlixi@ppp-188-174-185-241.dynamic.mnet-online.de)
08:42.32*** join/#maemo phlixi (~phlixi@ppp-188-174-185-241.dynamic.mnet-online.de)
08:50.34*** part/#maemo L29Ah (~L29Ah@195.19.225.242)
08:52.48KotCzarnystupid alsa
08:53.31*** join/#maemo BCMM (~BCMM@unaffiliated/bcmm)
08:54.11*** join/#maemo dreamer (~henk@melange.soleus.nu)
09:01.07*** join/#maemo xorly (~xorly@ip-86-49-15-121.net.upcbroadband.cz)
09:23.12*** join/#maemo silviof (~silviof@unaffiliated/silviof)
09:32.47*** join/#maemo L29Ah (~L29Ah@195.19.225.242)
09:37.33*** join/#maemo zGrr (~grr@185.13.106.74)
09:37.46zGrrmoin :)
09:40.39*** join/#maemo dreamer (~henk@melange.soleus.nu)
10:01.12*** join/#maemo aloril (~aloril@dsl-tkubrasgw2-54f80e-20.dhcp.inet.fi)
10:05.18*** join/#maemo infobot (ibot@69-58-76-73.ut.vivintwireless.net)
10:05.18*** topic/#maemo is Welcome to #maemo http://maemo.org/intro/ | N9/harmattan related questions please in #harmattan ! | "firmware" see ~tabletsdev ~combined ~emmc ~flasher sdk: ~sb | http://wiki.maemo.org/Community_Council | Source: http://wiki.maemo.org/Sources | Chanlog: http://mg.pov.lt/maemo-irclog | #maemo-ssu is where the (few) devels and maintainers of the maemo-future meet
10:08.19*** join/#maemo dreamer (~henk@melange.soleus.nu)
10:14.43*** join/#maemo louisdk (~louisdk@94.144.63.228)
10:25.02*** join/#maemo obsed (~obesd@unaffiliated/obesd)
10:34.51*** join/#maemo louisdk (~louisdk@94.144.63.228)
10:40.12*** join/#maemo dreamer (~henk@melange.soleus.nu)
11:10.39*** join/#maemo Haudegen (~quassel@85.124.51.57)
11:10.58*** join/#maemo louisdk (~louisdk@94.144.63.228)
11:12.38*** join/#maemo erlehmann (~erlehmann@f055122010.adsl.alicedsl.de)
11:13.02*** join/#maemo vakkov (~vakkov@84.54.169.227)
11:18.27KotCzarnyhehehe, fun
11:18.59KotCzarnyalsa lib example tool latency.c also leaves device enabled (ie. blocked)
11:29.04*** join/#maemo dhbiker (~dhbiker@APN-123-193-192-gprs.simobil.net)
12:20.20*** join/#maemo louisdk (~louisdk@94.144.63.228)
12:25.22*** join/#maemo BCMM (~BCMM@unaffiliated/bcmm)
12:48.38*** part/#maemo L29Ah (~L29Ah@195.19.225.242)
12:54.09*** join/#maemo LauRoman|Alt (~LauRoman@5-14-58-90.residential.rdsnet.ro)
12:57.27*** join/#maemo louisdk (~louisdk@94.144.63.228)
13:06.08*** join/#maemo FlameReaper (~assassin@175.137.196.26)
13:33.52*** join/#maemo L29Ah (~L29Ah@195.19.225.242)
13:44.56*** join/#maemo dreamer (~henk@melange.soleus.nu)
13:51.43KotCzarnyugh, n8x0's mic/recording is limited to 8kHz
13:52.09KotCzarnyand diablo is missing gst-plugins-'bad'
13:52.17KotCzarnycastrated all the way
14:10.43kerioi reckon i'll switch to android soon
14:10.53kerio:(
14:10.57KotCzarnygot a new toy/
14:12.06DocScrutinizer05KotCzarny: prolly N810 indeed is pretty fsckdup re audio
14:12.13KotCzarnyyeah
14:12.29KotCzarnyplaying works, mostly
14:12.35KotCzarnyrecording.. well
14:13.13KotCzarnywith the alsa code i made it working, but i run into underruns or some mismatched buffers problems
14:13.21keriomight get to use a samsung galaxy grand prime
14:13.26KotCzarnydo you have a moment to help me with the code?
14:13.49keriospecs are kinda crap anyway
14:13.54kerioi dunno
14:14.05KotCzarnythat was to the doc
14:14.14kerioi would've liked to buy a neo900 but it costs eleventy billion euros
14:15.47*** join/#maemo Ras_Older (blank@iddqd.fi)
14:19.50*** join/#maemo bugzy (~bugzy@2602:306:3bf8:b1f0:ae22:bff:fe50:c418)
14:38.56*** part/#maemo L29Ah (~L29Ah@195.19.225.242)
14:44.15*** join/#maemo dreamer (~henk@melange.soleus.nu)
15:02.45*** join/#maemo dreamer (~henk@melange.soleus.nu)
15:22.22*** join/#maemo kerio (kerio@duckbot.xd.cm)
15:30.18KotCzarnylol, n800 was codenamed gagarin
15:31.27KotCzarny770 was codenamed sputnik and n900 rover
15:33.14KotCzarnyn9 'lankku' and n-gage 'starship'
15:33.50*** join/#maemo Pali (~pali@Maemo/community/contributor/Pali)
15:34.40bencoh:)
15:35.13bencoh"rover" is pretty much accurate for n900
15:35.28KotCzarnyrove, rove, rove your boat?
15:36.40KotCzarnyis there usb-audio kernel module for n8x0 ?
16:06.08*** join/#maemo endso (~mypc@S0106306023d554b3.no.shawcable.net)
16:24.11*** join/#maemo teotwaki (~teotwaki@Maemo/community/contributor/crashanddie)
16:38.30*** join/#maemo Haudegen (~quassel@85.124.51.57)
16:58.53KotCzarny[16610.820312] usbcore: registered new interface driver snd-usb-audio
16:58.55KotCzarny:)
17:00.09KotCzarnypity alsa oss emulation doesnt register the card
17:01.40KotCzarnyso, anyone with n8x0, usb gender bender cable, usb audio card and wanting to test usb audio?
17:04.42bencohdont you have a DAC on n8x0 ?
17:04.54KotCzarnymore like CAC
17:05.01KotCzarnycrap audio converter
17:05.39KotCzarnybencoh, unless you point me to some code that can record good quality audio for n8x0
17:07.45bencohrecord ? I thought you wanted to play
17:07.55KotCzarnyrecord, apply effect, then play
17:08.07KotCzarnyplaying part works fine
17:12.08KotCzarnynice, nfs module
17:14.20*** join/#maemo L29Ah (~L29Ah@195.19.225.242)
17:14.56*** join/#maemo Pali (~pali@Maemo/community/contributor/Pali)
17:21.33endsoI unfortunately don't have the right equip. to test external audio w/ n8x0
17:21.33endsoat least not at this time...
17:21.34endsousb audio card eh?
17:21.34KotCzarnyyeah, and 100mA max, or with external power hub
17:23.06endsohm...I'll look
17:25.32APicB-)
17:30.31*** join/#maemo aloril (~aloril@dsl-tkubrasgw2-54f80e-20.dhcp.inet.fi)
17:34.34*** join/#maemo florian (~fuchs@Maemo/community/contributor/florian)
17:37.34vectisI've got a set of usb speakers if thats any help? usb sound, but no mic input obviously
17:37.52endsodug through a couple of boxes of adapters, cables, plugs and interfaces I have for audio for comp and devices....no I don't have a usb audio card handy..
17:38.16KotCzarnyvectis, if your usb speakers have minijack too, then they use usb only for power
17:38.56vectisno, it's all on the usb cable.
17:39.08KotCzarnyand do you have n8x0 ?
17:39.22vectisyep :)
17:39.27vectisn810
17:39.40KotCzarnyand usb adapter to connect them to n810?
17:40.29vectisyer, got the cables. Just looking for a powered hub
17:43.08vectisJust looking for the last piece of the puzzle.......... the power supply
17:46.00*** join/#maemo BCMM (~BCMM@unaffiliated/bcmm)
17:48.03*** join/#maemo troulouliou_div2 (~troulouli@unaffiliated/troulouliou-div2/x-0271439)
17:50.08vectis[78860.308593] usb 1-1.4: Product: USB Audio [78860.308593] usb 1-1.4: Manufacturer: C-Media INC. [78860.308593] usb 1-1.4: device v0d8c p0001 is not supported
17:50.31KotCzarnylsmod |grep usb-audio
17:50.54KotCzarnyor lsmod|grep usb_audio
17:51.16vectisnothing
17:51.21KotCzarnyhttps://transfer.sh/19OFqz/n810-extra.tar.gz
17:51.24KotCzarnyhere are modules
17:51.36KotCzarnybuilt for n800, but n810's kernel is almost identical
17:57.26vectishmm, insmod: can't open 'snd_#####.ko': No such file or directory thats for all of them
17:57.43KotCzarnydmesg?
17:57.46*** join/#maemo DerHat (~chatzilla@dyndsl-092-252-046-115.ewe-ip-backbone.de)
17:57.52DerHatHi
17:57.59KotCzarnyalso, are you in the directory with .ko files?
17:59.04*** join/#maemo futpib (~futpib@176.104.204.3)
17:59.13DerHati soldered my n900 usb port again. looks like my battery is dead, no matter because i got more of them!
17:59.25KotCzarnyalso, i've made a script that will load modules in correct order
18:00.05DerHatbut is there any known way to fix with lower capacity or something?
18:00.42*** join/#maemo shentey (~shentey@x55b34223.dyn.telefonica.de)
18:03.45freemangordonDerHat: the only known fix for a battery with lowered capacity is a *new* battery :)
18:05.06DerHatok, i got 2 dead batteries :/
18:05.41KotCzarny:)
18:05.44DerHatn900 needs so much time!
18:05.50DerHatmy hobby: maemo n900
18:05.51DerHat:D
18:06.08KotCzarnyjust buy some chinese one for a quick-fix, then hunt for brand new original one
18:06.56freemangordonnot original, but powercell ;)
18:06.59freemangordon~powercell
18:07.08freemangordonhmm
18:07.08KotCzarnyoriginal powercell
18:07.14freemangordonyeah
18:08.10*** join/#maemo troulouliou_div2 (~troulouli@unaffiliated/troulouliou-div2/x-0271439)
18:08.56DerHatmaybe one of my batteries is getting a zombie if i overvoltage or just wait some hrs
18:10.14vectisnothing doing. I extracted the tar.gz and cd to the folder. ran the script and I get the same "No such file or directory" 8 times (run as su)
18:10.50KotCzarnyhmm
18:10.52freemangordonhmm, google knows nothing about "powercell bl5j"
18:11.28*** join/#maemo troulouliou_div2 (~troulouli@unaffiliated/troulouliou-div2/x-0271439)
18:11.49KotCzarnyvectis: weird
18:12.14KotCzarnyaaah. ok script fails because i've used lsmod output, silly me
18:12.27KotCzarnydo ls, then copypaste the name
18:12.47vectisgrep usb-audio now waits at the promt, returned nothing before
18:13.50KotCzarnyvectis: try: insmod snd-rawmidi.ko
18:14.42vectisthat worked
18:14.55KotCzarnyk, then change all _ into - in the script and just run
18:15.14KotCzarnyor load by hand, whichever is faster
18:16.21vectisha ha, just spotted that! gimme five.......
18:16.29*** join/#maemo sparetire_ (~sparetire@unaffiliated/sparetire)
18:17.36freemangordonDerHat: http://www.ebay.co.uk/itm/PolarCell-Battery-Nokia-X6-00-5230-5800-N900-C3-00-Bl-5j-Xpress-Music-Xm-/331336409344?pt=LH_DefaultDomain_3&hash=item4d25306100
18:17.53freemangordonthat one (1500 mAh)
18:18.09KotCzarnyfreemangordon.. thats polarcell, not powercell
18:18.12KotCzarny~polarcell
18:18.12infoboti heard polarcell is http://www.ebay.co.uk/itm/390402445382 BL-5J Replacement
18:18.24freemangordonoh, stupid, blind me :D
18:18.24KotCzarnytypo day or what?
18:18.30freemangordonyeah, typo day
18:19.17*** join/#maemo shentey (~shentey@x55b34223.dyn.telefonica.de)
18:19.24bencoh:)
18:19.25freemangordonI even disassembled my second device to read how *exactly* it is written
18:19.26DerHat5,55 W/h
18:19.48bencohpolarcell <3
18:20.00freemangordonyeah, but still I googled for powercell :D
18:20.27bencoheven though the "last charge" slowly decreases over time
18:20.40KotCzarnypity
18:20.42freemangordonwell, that's normal afaik
18:20.49bencoh1492 when I bought it, 1475 as of now
18:20.53KotCzarnyyeah, li-ion fun
18:20.55freemangordonnot that bad
18:20.55bencoh(6 months later iirc)
18:21.02vectisok, the lights are on. usbcore: registered new interface driver snd-usb-audio
18:21.03bencohyeah, not that bad
18:21.25KotCzarnyvectis: woo-hoo, now, i wonder what's the alsa id of this card
18:21.49bencohdaily use, charge every two days (roughly) from ~50% to 100%
18:22.01vectisjust tell me what buttons to press :D
18:22.04L29Ah> W/h
18:22.06KotCzarnyyou can try: mplayer -ao alsa=hw:1,0
18:22.14KotCzarnyyou can try: mplayer -ao alsa=hw:1,0 some.mp3
18:22.52KotCzarnyor: mplayer -ao alsa:device=hw:1,0 some.mp3
18:22.55freemangordonbencoh: how do you manage to go through more than one day with a single charge? I guess your n900 is not your only device.
18:24.20vectishmmm, i'll have to install mplayer. I'll get back to you asap. I need coffee.........
18:24.50DerHatyes, you
18:24.53KotCzarnyi've just checked, proper syntax is: mplayer -ao alsa:device=hw=1.0 some.mp3
18:25.03vectisoakies
18:25.21KotCzarnyyou may try different hw combinations
18:25.30KotCzarnydon't know which id it got assigned
18:26.01bencohfreemangordon: laptop at work and at home, 3G only when needed (web or tethering)
18:26.38freemangordondo you have IM and/or email accounts?
18:26.59bencohssh/irssi over gprs
18:27.03KotCzarnyvectis: you may check /proc/asound/cards
18:27.21bencoh(and mutt over ssh for mails :)
18:28.32bencohand I dont need to send tcp keepalives so it doesnt need to wake the cpu too often
18:28.51bencoh(basically when I use it or when I receive data to display on screen/terminal)
18:29.09freemangordonbencoh: weird, could be the cells quality, but I can't make more than a day with gtalk and fb chat and gmail/exchange accounts over 2G
18:29.56bencohor they keep waking it up
18:30.18bencohkeepalives at tcp or application level
18:30.31KotCzarnyif you set them to 120s it's not that bad
18:30.41bencohhow does fb chat work btw ? http/json api ?
18:30.48dhbikerhi... where can i get images for OS reflash
18:30.48KotCzarnybut if server doesnt' disconnect you for idling, no need to
18:30.56KotCzarny~flashing
18:30.56infobotit has been said that maemo-flashing is http://wiki.maemo.org/Updating_the_tablet_firmware, or - on linux PC - download&extract http://maemo.cloud-7.de/maemo5/patches_n_tools/maemo-my-private-workdir.tgz, cd into it, do sudo ./flash-it-all.sh
18:30.57freemangordonbencoh: xmpp?
18:31.13bencohfreemangordon: oh, didnt know it still worked with fb
18:31.18KotCzarny~combined
18:31.18infobotmethinks combined is the rootfs fiasco image of maemo. For N900 latest (PR1.3.1) see http://nds2.fds-fire.nokia.com/fdp/interface/FiRe/2011/9/--FID--A0A22MVWFVFAM/--LID--FiRe1317015685654/RX-51_2009SE_21.2011.38-1_PR_COMBINED_MR0_ARM.bin, or http://galif.eu/nokia/
18:31.21bencohI thought they closed that a long time ago
18:31.37dhbikersweet ty KotCzarny
18:31.44KotCzarny~fiasco
18:31.44infobotL4-compatible real-time microkernel capable of running Linux in usermode. URL: http://os.inf.tu-dresden.de/fiasco/
18:31.49KotCzarny:)
18:31.51KotCzarnyhehe
18:31.54freemangordonbencoh: suprisingly, it still works, albeit st is officially closed :)
18:32.07freemangordons/st/it
18:33.18*** join/#maemo arcean (~arcean@nat3.finemedia.pl)
18:33.52bencohoh :)
18:34.00bencohfreemangordon: http://tools.ietf.org/html/rfc6120#section-4.6 (xmpp keepalive)
18:36.38*** join/#maemo arossdotme-nolog (~zxy@79-69-196-192.dynamic.dsl.as9105.com)
18:38.02freemangordonbencoh: so you think both fb and gtalk send keepalives?
18:38.17bencoheither the server or the client, yes
18:38.39bencohmost likely the client
18:38.57freemangordonhmm, that might explain the "no more than a day" case
18:39.01bencohI think they cut the session if they dont receive it
18:39.40KotCzarnyhttp://stackoverflow.com/questions/7967585/how-does-facebook-chat-tcp-connection-keep-alive
18:40.11bencoh(hmm I usually dont let ssh run all day long btw)
18:40.12freemangordonwell, but on the other hand - I don;t believe such keepalives are send that often
18:40.26bencohI'd say one every minute
18:40.34freemangordonwhich is not that much
18:40.42bencohhm yeah, not that much
18:40.45KotCzarnyfreemangordon, let it idle for 1h, and compare /proc/*/net/* counters
18:41.37bencohbut I enabled it for ssh once (shitty internet uplink with tcp timeout) and battery decreased more quickly
18:41.39freemangordonor rather run tcpdump for 10 minutes :)
18:41.44bencohyeah :)
18:41.48KotCzarnythat would work too
18:42.10freemangordonbencoh: define "more quickly"
18:42.41bencohunfortunately I didnt keep track of the numbers
18:42.57freemangordonwell, subjectively
18:43.02KotCzarnyor: freemangordon, don't use facebook for one day
18:43.03KotCzarny;)
18:43.06bencohbut it was like "oh hey, 20% just disappeared in 2h" or something like that
18:43.12freemangordonoh
18:43.46bencoh(ssh from my computer to the n900)
18:44.05bencoh(wireless)
18:44.28freemangordonKotCzarny: i will have to disable both fb and gtalk
18:44.39KotCzarnyfreemangordon: do it
18:44.41KotCzarny:)
18:45.22freemangordonthough, I am going to WOA in 10 days, I guess I can use the time I'll be there to disconnect from the social networks and chats :)
18:45.22bencohfreemangordon: irssi+bitlbee (or the other xmpp/irc gateway .. forgot the name) over ssh ? 0:)
18:45.52bencohoh you're actually dropping fb/gtalk for real forever ?
18:45.57freemangordonno
18:46.26freemangordonjust 4 days while I'll be in Germany :)
18:46.31bencoh:)
18:46.40freemangordon~woa
18:46.56bencoh~google woa
18:47.03bencoh~search woa
18:47.05bencohmeh :)
18:47.16freemangordon~WOA is Wacken Open Air
18:47.16infobotfreemangordon: okay
18:47.39bencohhaha
18:48.33freemangordonanyway, with ~80000 people on a couple of cells, I doubt there will be much of a connectivity :D
18:49.11bencohhaha
18:49.21KotCzarnymicrowavefest
18:49.22bencohyeah, best to just switch to offline mode
18:49.50bencohKotCzarny: ear damages werent enough 0:)
18:50.08freemangordonwell, not offline, but disconnected from internet
18:50.39*** join/#maemo _rd (~rd@p4FD1FECE.dip0.t-ipconnect.de)
18:50.41freemangordonI guess there'll be no need to charge
18:51.03bencohyeah, it should last 5 days with no connectivity
18:51.24bencohhmm, maybe not, that's if you keep it in your pocket and dont really use it
18:51.50freemangordonwill see
18:52.24bencohand .... shit ... I forgot mtr running since yesterday ... again. :D
18:52.52bencohthat's the one battery killer I always forget about
18:52.53vectis~/MyDocs/.sounds $ mplayer -ao alsa:device=hw=1.0 Moby-In_My_Heart.mp3  ->>  alsa-lib: pcm.c:2144:(snd_pcm_open_noupdate) Unknown PCM hw:1,0
18:52.57freemangordonmtr is?
18:53.24bencohhttps://en.wikipedia.org/wiki/MTR_(software) (traceroute)
18:53.31freemangordonoh
18:53.44KotCzarnyvectis: can you show /proc/asound/cards
18:53.50bencohmuch better than traceroute, really handy
18:53.54KotCzarnyyeah
18:54.04KotCzarnybencoh, but they return different results sometimes
18:54.10bencohKotCzarny: indeed
18:54.56bencohand ... with slow/lossy networks and strange uplinks/backbones it can make it easier to get a tcp/ssh session open ... dont ask me why
18:55.23KotCzarnybecause it keeps celltower awaken
18:55.29KotCzarnyie. not freeing timeslot
18:55.35vectis1 [Audio          ]: USB-Audio - USB Audio                       C-Media INC. USB Audio at usb-musb_hdrc-1.4, full speed
18:55.35bencohmaybe
18:55.38KotCzarnysimilar efect with ping -s8 8.8.8.8
18:55.48freemangordonKotCzarny: on 2G? I doubt
18:56.04freemangordonthere are no dedicated slots for data on 2G iirc
18:56.11bencohI suspect it involves some terrible routing table management
18:56.13KotCzarnyfreemangordon: tested and tried, ping every 1s and it's much better for ssh
18:56.28freemangordonKotCzarny: again, on 2G or 3G?
18:56.48bencohs/routing table/FIB (forwarding information base/table) on some routers/l3-switches
18:56.51freemangordonKotCzarny: also, remember there is iphbd running on n900 ;)
18:57.09bencohiphbd ?
18:57.18freemangordon~iphb
18:57.26freemangordon~iphbd
18:57.27KotCzarnyfreemangordon: no matter, when i need interactivity i just run ping in the background (-s8 to keep is small datawise)
18:57.49KotCzarnyvectis: /proc/asound/devices ?
18:58.31freemangordonbencoh: IP heartbeat daemon or somesuch
18:58.35bencohwow, I never noticed it running
18:58.52vectis<PROTECTED>
18:59.08KotCzarnyhmm, it should work as 1.0 then
18:59.35freemangordonafaik it delays data transmission until a timeout has expired or there is enough data to be send, thus preventiong radios from waking up too often
18:59.41freemangordonbencoh: ^^^
19:00.04bencohhmm
19:00.11bencohit might be just that then
19:00.20bencohsounds a bit silly but ...
19:00.50bencohI guess a poor single tcp syn just wouldnt have any chance alone
19:01.01bencohis it opensource ?
19:01.05freemangordonno :D
19:01.09bencohshit :D
19:01.12freemangordonbut there is dbus iface
19:01.17bencohhaha
19:01.23freemangordonhttp://wiki.maemo.org/N900_dbus#com.nokia.iphbd
19:01.36bencohthx
19:01.37freemangordonlook at the header
19:01.51bencohhttp://www.cncmods.net/files/dbus/libiphb.h woooh
19:02.29bencohhmm
19:02.35freemangordonthere is also a kernel module (FOSS) iirc
19:02.42bencohlooks like you explicitly need to use it
19:02.45bencohoh
19:03.31freemangordonhttps://github.com/nemomobile/libiphb/tree/master/src
19:04.30freemangordonseems it has been open-sourced after all
19:04.37bencohneat
19:05.11bencohwonder if telepathy/rtcom-messaging makes use of it
19:06.13freemangordonI'd bet on that
19:06.19KotCzarnyvectis, btw. is there /dev/dsp file? (ignore /dev/dsptask*)
19:07.48freemangordonbencoh: seem nemomobile iphbd has some differences compared to maemo, but I wonder if I should arm IDA and RE it, should be pretty much easy task with all those headers and source code around
19:08.49KotCzarnyfreemangordon: do it
19:09.00vectisyer, dsp1
19:09.12KotCzarnyvectis: http://pastebin.com/raw.php?i=ikt2FUUq
19:09.22KotCzarnyput contents into ~/.asoundrc
19:09.30KotCzarnythen try mplayer -ao alsa some.mp3
19:12.21freemangordonKotCzarny: on the other hand - why noone else want to do RE work? It is not *that* hard!
19:12.33freemangordon*wants
19:12.37bencohfreemangordon: it doesnt look like a critical part of maemo, but well if you have some time :)
19:13.08KotCzarnyfreemangordon: it is for someone who doesn't have tools/knowledge/experience
19:13.18freemangordonlooks for volunteers to do RE job, with my help :)
19:13.32KotCzarnyfreemangordon: make a wiki thread
19:13.37KotCzarnyi mean tmo thread
19:13.46freemangordonKotCzarny: hmm, might work
19:14.13*** join/#maemo eMHa__ (~mh@176.0.112.166)
19:14.16KotCzarnyon another note, my usb audio card claims to suck 500mA, yet, usb meter only shows ~30mA
19:14.31KotCzarnynow i need some gender bender for my n800
19:14.48freemangordonKotCzarny: connect headphones to it and raise the volume up
19:15.05freemangordonlow-resistance ones
19:15.07KotCzarnyi've connected it to stereo box
19:15.34KotCzarnybut i know that it can eat more
19:15.41freemangordonthat is why only 30 mA
19:15.53KotCzarnystill, nice to know i can connect it to n800 and use as a grabber card
19:16.07KotCzarnywithout external power
19:16.24freemangordondoesn't it require CPU resources?
19:16.38KotCzarnydont know, what for?
19:17.05KotCzarnyi mean, apart from data moving
19:18.26vectishmm, do I need to reboot?
19:18.32KotCzarnyvectis, nope
19:18.48freemangordonno idea, I am under impression USB is not the most effective when it comes to low CPU usage
19:18.53vectisstill no go...........
19:19.03KotCzarnyvectis, same error msg?
19:19.25vectisyer
19:19.25bencohfreemangordon: I'm writing assembly on a DSP at $NEWJOB (2 weeks) so I might be of some help with RE-ing stuff and asm after some time :)
19:19.39KotCzarnyfreemangordon: yeah, and it's usb1, still, working solution is better than non working low power solution
19:20.17KotCzarnybencoh: can we has dsp based equalizer on n8x0/n900?
19:20.18KotCzarny:)
19:20.19freemangordonbencoh: C64+? O.o
19:20.28bencohfreemangordon: hm no
19:20.57freemangordonwait, wait, why assembly?
19:21.16bencohbecause ... there is no C compiler for our target :D
19:21.25freemangordonomg :D
19:21.51bencoh(small/specific embedded)
19:21.53freemangordonwhat an exotic piece of silicon is that?
19:22.17freemangordonif not under NDA ofc :)
19:22.37KotCzarnyvectis: don't forget to remove that .asoundrc file, it can interfere with the system
19:22.59bencohI'm afraid it is (full specs arent public and it's not in our product description either) :(
19:23.09*** join/#maemo SpeedEvil (~quassel@tor/regular/SpeedEvil)
19:23.23bencohbut that's one for sound processing
19:25.16KotCzarnyvectis: mplayer -ao oss:/dev/dsp1 some.mp3 ?
19:27.49vectisKotCzarny: :( nope
19:27.55KotCzarnysame msg?
19:28.37vectisdoesn't mention alsa this time but much the same
19:28.59KotCzarnymplayer -ao oss:/dev/dsp1 some.mp3 >debug.txt 2>&1
19:29.08KotCzarnythen can you share the file somehow?
19:31.29*** join/#maemo sq-one (~sq-one@HSI-KBW-134-3-167-183.hsi14.kabel-badenwuerttemberg.de)
19:32.12bencohhttp://pastebin.notk.org/pastebin.php?show=d4663d931
19:32.37bencohfunny maemo-laucher depends on it
19:32.45KotCzarnycalls home?
19:34.49KotCzarnyvectis: cat /proc/asound/pcm ?
19:36.16bencohhmm there is a kernel module indeed
19:36.27Luke-Jrwow, Maemo is pre-NPTL :o
19:36.31vectishttp://pastebin.com/4shP0bVA
19:36.36vectis01-00: USB Audio : USB Audio : playback 1
19:38.25*** join/#maemo chem|st (~mayor@svn.unix-ag.uni-kl.de)
19:45.24KotCzarnyhrm, found my usb micro to usb cable. pity n800 uses usb mini o.o
19:45.57freemangordonLuke-Jr: hmm? 2.6.28
20:00.05*** join/#maemo _rd (~rd@p4FD1FECE.dip0.t-ipconnect.de)
20:01.59KotCzarnyunrelated but nice: http://alsa.opensrc.org/TwoCardsAsOne
20:07.04*** join/#maemo BitEvil (~quassel@tor/regular/SpeedEvil)
20:13.36Luke-Jrfreemangordon: yeah, NPTL appears to need 2.6.32 :x
20:15.18bencoh:]
20:16.27freemangordonhmm, I thought that all 2.6.x kernels have NPTL
20:19.20freemangordonman pthreads says the same
20:19.26freemangordonLuke-Jr: ^^^
20:20.39Luke-Jrfreemangordon: well, glibc 2.20 apparently begs to differ and won't work with <2.6.32 :/
20:21.41freemangordonafaik it is bacause the kernel misses other things that glibc needs
20:21.45freemangordon*because
20:22.20*** join/#maemo KotCzarny (~ariel@bgg35.neoplus.adsl.tpnet.pl)
20:22.49L29Ahi removed the check and it crashes :3
20:23.14L29Ahthe previous version works afair
20:23.15Luke-Jr2.19 crashes for me :/
20:23.57freemangordonbut why did you decide it is NPTL that is the problem?
20:26.53Luke-Jr"You need a kernel of at least 2.6.32 for NPTL support!"
20:28.27freemangordonwell, that sounds like 2.20 requires some features in 2.6.32 to enable NPTL, but doesn't sound like there is no NPTL support in 2.6.28
20:29.38*** join/#maemo florian (~fuchs@Maemo/community/contributor/florian)
20:35.37*** join/#maemo shentey (~shentey@x55b34234.dyn.telefonica.de)
20:40.08Luke-Jrwell, the main problem is that 2.20 won't work with 2.6.28
20:41.16*** join/#maemo florian (~fuchs@Maemo/community/contributor/florian)
20:45.11bencohI thought there were some patches out there for 2.19 ?
20:51.32*** join/#maemo _rd (~rd@p4FD1FECE.dip0.t-ipconnect.de)
21:13.16KotCzarny2.20 > 2.19
22:14.19*** join/#maemo xorly (~xorly@ip-86-49-15-121.net.upcbroadband.cz)
22:26.27*** join/#maemo {-0-} (~bob@cpc2-bolt15-2-0-cust323.10-3.cable.virginm.net)
22:59.34*** join/#maemo erlehmann_ (~erlehmann@82.113.98.44)
23:01.34Vajbheh, all that audio talk here made me think, is there frequency response measurement software for n900?
23:01.51KotCzarnymost likely yes
23:01.55Vajbtho i guess n900 mic quality wouldn't be enough
23:03.11Vajbalso i just noticed that yappari has disconnected and wont connect again
23:03.55Vajbjust checking to see if there is an update, but ham is taking her time....as usually
23:04.54Vajbbtw KotCzarny, most likely yes for ap or insufficient mic quality?
23:05.12KotCzarnyfor fft display/finder
23:06.01KotCzarnyhttp://maemo.org/downloads/product/Maemo5/tuner/
23:06.11KotCzarnyhmm
23:06.16KotCzarnywrong way, scratch that
23:10.54KotCzarnyhmm
23:13.41Vajbcool, update fixed yappari
23:14.02KotCzarnyhttp://talk.maemo.org/showthread.php?p=1343610
23:17.43bencohI wouldnt worry too much about it not being "good enoug" (well, it's probably terrible anyway :p) ... people do fft/audio measurements with an iphone/ipad these days
23:18.39KotCzarnyyou can always connect external mic of better quality
23:19.20bencohor do a clean calibration (white noise / pink noise) to get the response curve of the mic
23:19.40KotCzarnythat's assuming it's linear over volume/freq
23:20.04bencoherr not pink
23:20.21bencohyeah ... you can also try it at different levels :)
23:20.45*** join/#maemo till- (~till@vispillo.org)
23:20.49KotCzarnyhttp://gaydenko.com/qloud/
23:21.46bencoh"The sourceforge.net website is temporarily in static offline mode" wtf
23:21.49bencoh(for awt)
23:21.51bencohqwt*
23:21.57KotCzarnyyeah, sf has it's ups and downs
23:22.51*** join/#maemo amospalla (~amospalla@unaffiliated/amospalla)
23:24.12KotCzarnyanyone with qt experience to compile/port qloud ?
23:27.25KotCzarnyhttp://www.roomeqwizard.com/
23:27.40*** join/#maemo Oksana (~chatzilla@Maemo/community/council/Wikiwide)
23:30.23KotCzarnyhttp://www.minidsp.com/products/acoustic-measurement/umik-1
23:33.48*** join/#maemo pentanol (~Unknown@unaffiliated/pentanol)
23:43.46KotCzarnyhttp://realtraps.com/art_measuring.htm
23:51.38DocScrutinizer05for frequency response you easily can calibrate the most crappy microphone you could find, as long as it does cover the relevant frequency range at all
23:52.15DocScrutinizer05ooh bencoh beat me to it
23:53.17vectisdoes alsa oss emulation work for n8x0? need it for this to work: http://stian.cubic.org/project-ocp.php
23:54.35KotCzarnyvectis: if you talk about internal audio card of n8x0, nope
23:54.54vectis:(
23:55.07KotCzarnyyou've checked and seen only /dev/dsp1
23:55.24vectisyer sorry, it's late
23:55.53KotCzarnyand if you need player for weird formats, have you checked oscp?
23:56.39KotCzarnyit uses uade2 and xmp libs
23:56.50vectisit was the graphic eq that caught my eye
23:56.54erlehmannhey
23:57.20erlehmanndid someone of you fix a n900 with a broken screen and found data from the previous owner on it?
23:57.20erlehmannif so, thanks!
23:57.43erlehmannbencoh slashdot is also in static offline mode
23:57.48erlehmannseems this thing is huge!
23:58.37KotCzarnyvectis: you mean visualization or audio processor?
23:58.59vectisvisualization
23:59.30KotCzarnyi shall merge oscp and osc one day
23:59.59erlehmannWHOM OF YOU IS GOING TO BE ON THE CCC CAMP?

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.