IRC log for #openezx on 20090623

00:12.10*** join/#openezx gDD (n=gDD@unaffiliated/gdd)
00:29.50*** join/#openezx huzhangliandroid (n=huzhangl@59.175.215.66)
00:56.36*** join/#openezx _Andrew (n=andrew@unaffiliated/andrew/x-643872)
00:59.10*** join/#openezx sabro1 (n=sab@mic92-8-82-234-143-184.fbx.proxad.net)
01:02.49*** join/#openezx johanbr (n=j@blk-7-129-240.eastlink.ca)
01:32.27*** join/#openezx huzhangliandroid (n=huzhangl@59.175.215.66)
02:37.52*** join/#openezx sabrod (n=sab@mic92-8-82-234-143-184.fbx.proxad.net)
03:11.15*** join/#openezx arne__ (i=nobody@dslc-082-082-068-044.pools.arcor-ip.net)
03:45.46*** join/#openezx alvin (n=huzhangl@59.175.215.66)
04:01.53*** join/#openezx parmaster_ (i=par@69.60.134.179)
04:11.30*** join/#openezx alvin (n=huzhangl@59.175.215.66)
04:13.58*** join/#openezx anais (n=huzhangl@59.175.215.66)
05:05.25*** join/#openezx alvin (n=huzhangl@59.175.215.66)
05:11.19*** join/#openezx sabrod (n=sab@mic92-8-82-234-143-184.fbx.proxad.net)
05:22.07*** join/#openezx gDD|mobilE (n=gDD@unaffiliated/gdd)
06:23.26*** join/#openezx l403 (n=l@85.132.159.239)
06:53.15alvinao2 hi
07:10.39*** join/#openezx tmzt_ (n=tmzt@adsl-99-164-52-98.dsl.akrnoh.sbcglobal.net)
07:16.15Hopscotchgood morning
07:24.17alvinI2C_BOARD_INFO("tea5767", 0x81),
07:24.25alvinhi all
07:24.44alvinwhat devie is tea5767?
07:37.42*** join/#openezx anais (n=huzhangl@59.175.215.66)
07:51.16*** join/#openezx florian (n=fuchs@95.89.44.245)
08:13.41bartvgoogle says it is a fm radio chip
08:23.01*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
09:09.24*** join/#openezx alvin (n=huzhangl@59.175.215.66)
09:44.26*** join/#openezx Sleep_Walker (n=Sleep@nat/novell/x-6f119d3d1ae7de31)
09:47.27*** join/#openezx anais (n=huzhangl@59.175.215.66)
12:25.56*** join/#openezx stefan_schmidt (n=stefan@c120.apm.etc.tu-bs.de)
12:43.56*** join/#openezx gmzhuo (n=gmzhuo@125.70.175.213)
13:08.30gmzhuoseriously panic
13:08.32*** join/#openezx lclaudio (n=lclaudio@201-40-140-35.pltce701.dsl.brasiltelecom.net.br)
13:18.55*** join/#openezx johanbr (n=j@blk-7-129-106.eastlink.ca)
13:46.26ao2gmzhuo, what's happened?
13:47.13gmzhuoso many oops msg,too fast to see
13:49.55WyrMhi
13:50.23gmzhuoHi WyrM
13:50.30WyrMgmzhuo: OpenEZX?
13:50.37gmzhuoyes
13:50.47WyrMgmzhuo: dump the mem with gen-blob
13:50.54WyrMand look at the kernel msgs. :)
13:51.24gmzhuofrom where to where?
13:51.25WyrMare you using the default .config?
13:51.44WyrMgmzhuo: 30MB starting at 0xa0000000
13:51.57WyrMI will test it here too.
13:57.56gmzhuono usefull info
13:58.54WyrM?
13:59.02WyrMI can see the messages here.
13:59.17gmzhuowhere it is?
13:59.36WyrMI dump the memory, and then "strings memdump | less"
13:59.37WyrM:)
14:00.08*** join/#openezx l404 (n=l@85.132.159.239)
14:01.17WyrMhttp://www.pastebin.ca/1471140
14:02.21gmzhuofrom PCAP?
14:02.37WyrMfrom pcap probe.
14:02.50WyrMseems that NR_IRQS was not increased correctly.
14:03.05*** join/#openezx lclaudio_ (n=lclaudio@201-11-243-106.pltce701.dsl.brasiltelecom.net.br)
14:03.28mickeylhi guys
14:03.37mickeylWyrM: asoc killer! :D
14:03.42mickeylwhen can we expect it back?
14:03.53WyrMmickeyl: don't know :(
14:04.10WyrMmickeyl: I will ask for advice from ASoC people.
14:04.44WyrMon how we should deal with baseband.
14:04.55WyrMI think they will point us to mainstone code.
14:04.56mickeylhmm
14:05.03mickeylwhat did you not like in the previous system?
14:05.08mickeyldidn't it work ok?
14:05.16WyrMwhich requires an open() from userspace to setup the bp audio.
14:05.17mickeyl[mixer control as router]
14:05.22mickeylah, that
14:05.32WyrMmickeyl: because on EZX its not just routing.
14:05.56mickeylyes, we probably need some more mixer controls
14:05.59WyrMon OpenMoko the baseband audio is connected to an analog input on the codec, right?
14:06.06mickeylyes
14:06.14*** join/#openezx stefan_schmidt (n=stefan@c120.apm.etc.tu-bs.de)
14:06.24mickeylafaict as long as everything is mappable through mixer controls, I'm fine.
14:06.37mickeylif we need more than that, well... so be it. need to hide it somewhere htne
14:06.41WyrMon ezx it is connected to the same SSP port as everything else.
14:07.08WyrMso a mixer control is not the best solution. :/
14:07.38mickeylhmm
14:08.01WyrMand we need to provide an alsa/pcm device so we can for example, record gsm audio, etc..
14:08.09mickeylright
14:08.39mickeylmixer controls + more pcm devices
14:08.44mickeyl:)
14:14.07WyrMgmzhuo: now we need CONFIG_PXA_HAVE_BOARD_IRQS
14:16.27gmzhuowhere it can be set?
14:16.36WyrMlooking at it now.
14:17.32gmzhuoshould add it when select ezx system
14:18.00WyrMprobably :)
14:18.21WyrMim regenerating cscope db, will look at it in a minute
14:29.29*** join/#openezx johanbr (n=j@JBrannlund2.MathStat.Dal.Ca)
14:31.30gmzhuoyes,it works after have b0ard irqs set
14:33.02*** part/#openezx lclaudio_ (n=lclaudio@201-11-243-106.pltce701.dsl.brasiltelecom.net.br)
14:33.35CIA-16OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r362c79b548 10openezx/arch/arm/mach-pxa/Kconfig: add CONFIG_PXA_HAVE_BOARD_IRQS
14:33.35CIA-16OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r682d901110 10openezx/sound/soc/pxa/pxa-ssp.c: continue asoc leftj/i2s work
14:33.37CIA-16OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r74c7cfecf6 10openezx/arch/arm/mach-pxa/Kconfig: add CONFIG_PXA_HAVE_BOARD_IRQS
14:34.53CIA-16OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * ra04618848b 10openezx/arch/arm/configs/ezx_defconfig: update defconfig
14:35.44*** join/#openezx gDD (n=gDD@unaffiliated/gdd)
14:43.56ao2adding clean URLs to gitweb configuration, so commit links from git-notify can work
14:45.07ao2I will break gitweb for a bit
14:48.12*** join/#openezx sabrod (n=sab@mic92-8-82-234-143-184.fbx.proxad.net)
14:51.52WyrMao2: thanks! :)
15:20.10*** join/#openezx gDD|mobilE (n=gDD@unaffiliated/gdd)
15:34.32gmzhuono mmc again
15:40.47WyrMgmzhuo: it is working here.
15:40.50WyrMon a1200.
15:41.06WyrMcheck the kernel messages for MMC or pcap/spi errors.
15:41.30WyrMs/spi/regulator/
15:45.40*** join/#openezx pH5 (n=ph5@e178210146.adsl.alicedsl.de)
15:50.08gmzhuomaybe it's my config fault,works now
15:53.49WyrMgood :)
15:54.52WyrMshould I send pcap subdrivers now? or wait until Samuel accepts the pcap fixes?
16:17.05gmzhuowhich fixes?
16:18.24gmzhuoadc-locking?irq-handler?
16:18.41pH5WyrM: why does the keypad driver need to know the pcap internal irq number?
16:19.23WyrMpH5: Because we need to check if it's a key press or release.
16:20.09WyrMso we need to know which pcap bit caused the interrupt.
16:20.33gmzhuoif the fix not relate the pcap interface for subdrivers,then subdrivers can be send
16:20.41WyrMor, we could use one irq handler for each input event.
16:20.46ao2ok, now gitweb should be ok again, please let me know if there is any regression with the new configuration
16:21.59pH5WyrM: where can I see the keypad driver?
16:22.26pH5the other fixes look like they don't influence the interface to the subdrivers?
16:22.37WyrMpH5: http://www.pastebin.ca/1471323
16:23.03WyrMpH5: the set_ts_bits() is needed for our touchscreen.
16:23.23WyrMthe other two are really fixes to correct bugs.
16:26.12pH5WyrM: I think you should provide the pcap_keys driver with the actual irqs via irq resources.
16:26.29pH5instead of using that pcap_to_irq mechanism.
16:27.17pH5also, you might want to hook up PCAP_IRQ_HS via alsa/asoc, they have grown some jack detection infrastructure lately
16:27.42WyrMpH5: I have code for this on ASoC too.
16:28.22WyrMpH5: jack insert is old stuff, and mickeyl requested a jack insert event on input subsystem.
16:28.48WyrMI haven't checked alsa lately, to see if it reports jack insertion/removal on input subsystem.
16:29.21WyrMas for using resource, I also tought about this.
16:29.40WyrMbut I don't want to hardcode the resources on the core driver, like twl4030.
16:30.08WyrMand it would make huge platform data for pcap on the board file.
16:30.32pH5ah, ok. I don't know either if alsa can report to input subsys yet. It should though, eventually.
16:31.16WyrMpcap_to_irq/irq_to_pcap is what generates the smaller code, and relativelly to pcap irq_base all the irqs are fixed anyway. so... ;)
16:31.17pH5WyrM: you could just create a structure with the three IRQ resources in pcap and then let mfd add the pcap irq_base to that.
16:31.56WyrMpH5: heh, +RTC +TS +USB at least.
16:33.22WyrMyou mean, a struct resource inside the core driver? or on the board file?
16:36.28pH5in the core driver, like asic3 for ds1wm.
16:37.24pH5but I see you'd still need the irq_to_pcap to check for the edge direction.
16:37.33WyrMyep.
16:37.33pH5or is pstat level information?
16:37.44WyrMlevel.
16:37.50pH5maybe you could just fit this into a gpiochip?
16:37.51WyrMbut not only because of this.
16:38.11WyrMI also want to _not_ register some subdrivers depending on the board file.
16:38.26WyrMso I do not want to move the subdrivers table to the core driver.
16:38.42WyrMon some phones we don't have touchscreen, or we can't access RTC registers.
16:39.15WyrMon some phones PCAP is used to charge battery, so we use pcap_battery subdriver, on others we have another chip for this.
16:39.21pH5hm, that's right. I can only disable the code in asic3, not optimize it away.
16:43.44WyrMpH5: (gpiochip) it would only fit the pcap_keys driver.
16:43.57WyrMfor all other irqs its not really gpio.
16:44.55WyrMsome irqs are an ADC threshold (USB), for TS and RTC pstat has no meaning.
16:45.26WyrMand PSTAT is read only. :)
16:47.27WyrMsee the ATLAS datasheet, it's an evolution from PCAP2 PMIC.
16:47.28WyrMhttp://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MC13783&webpageId=1098824671538699514881&nodeId=01435974404881&fromPage=tax
16:48.11pH5ok, that was just a thought :)
16:48.55WyrMheh, np ;)
16:49.00pH5I think I agree your current method is the least complicated, and there's no use to make anything more generic.
16:49.43pH5generally, I'd suggest to submit the subsystem patches right now, and hint that they depend on stuff pending for the mfd tree.
16:49.59pH5although I don't know how that applies right now, the merge window being open and all that.
16:50.14pH5s/subsystem/subdevice/
16:50.51WyrMyeah, I want the subdevices for .32.
16:51.44WyrMbut Im afraid that Dmitry may want to rush it for .31 and merge it before Samuel merges the fixes.
16:52.25WyrMthe RTC and the regulator has no dependencies on the fixes, so they are OK. Only the input subsystem drivers need to wait.
17:17.29*** join/#openezx tombhadAC (n=tombhad@pD95DD701.dip.t-dialin.net)
17:22.05*** join/#openezx pH5_ (n=ph5@e178224017.adsl.alicedsl.de)
18:11.54*** join/#openezx WyrM (n=wyrm@unaffiliated/wyrm)
18:24.33*** join/#openezx WyrM (n=wyrm@unaffiliated/wyrm)
18:42.00*** join/#openezx alvin (n=huzhangl@59.175.215.66)
19:40.21*** join/#openezx marex (n=marex@vasut.kolej.mff.cuni.cz)
19:59.26*** join/#openezx florian (n=fuchs@95.89.44.245)
20:07.03*** join/#openezx stefan_schmidt (n=stefan@p5B036E1C.dip.t-dialin.net)
20:38.24*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
23:45.12*** join/#openezx ecloud_work (n=srutledg@gw.autoxray.com)
23:56.53*** join/#openezx johanbr (n=j@blk-7-129-106.eastlink.ca)

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