IRC log for #openezx on 20081116

00:01.27stefan_schmidttmzt: congrats, first git commit arrived
00:01.32stefan_schmidtgoes to bed
00:01.35stefan_schmidtNight guys
01:49.18*** join/#openezx sabrod_ (n=mac@lns-bzn-38-82-253-69-160.adsl.proxad.net)
04:38.34WyrMtmzt: I added you to our cia.vc filter.
04:39.00WyrMyour next commits will show on #openezx.
05:21.23tmztWyrM: hey, I don't get any PCAP interrupts now
05:23.29tmztI added request_gpio to motoq_init
05:41.03WyrMtmzt: it was gpio 1, right?
05:42.20tmztyes
05:42.26tmztgit is current
05:42.40tmztI mean I pushed my current code
05:43.01WyrMare you getting spi I/O?
05:43.28WyrMcan you see the register values on sysfs regs?
05:43.31tmztI can dump/write regs but adc reads crash kernel/lockup machine
05:43.41tmztno meesages just locked
05:43.56WyrMok, let me look at motoq.c
05:53.14CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * ra153d8701b 10openezx/.topdeps: update dependency
05:53.15CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r062d8ff72c 10openezx/drivers/rtc/rtc-pcap.c: remove unedeed mask mangling
05:53.15CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * ra0f72510d2 10openezx/ (.topdeps .topmsg): create
05:53.16CIA-29OpenEZX: 03guiming zhuo <gmzhuo@gmail.com> 07ezx/current * re78594f4d4 10openezx/drivers/rtc/ (Kconfig Makefile rtc-pcap.c): add pcap-rtc driver.
05:53.21CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * ra312bb3793 10openezx/drivers/rtc/rtc-pcap.c: add comment
05:53.24CIA-29OpenEZX: 03gmzhuo <gmzhuo@localhost.localdomain> 07ezx/current * rb348474af3 10openezx/drivers/rtc/rtc-pcap.c: fix drivers/rtc/rtc-pcap.c style error
05:53.27CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r9b16a3287e 10openezx/sound/soc/pxa/Kconfig: remove dependency on pxa_ezx for motoq
05:53.30CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r923a7a9718 10openezx/drivers/rtc/rtc-pcap.c: scripts/Lindent
05:53.35CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r6caf1449a8 10openezx/drivers/rtc/rtc-pcap.c: remove uneeded mask mangling again
05:53.38CIA-29OpenEZX: 03gmzhuo <gmzhuo@localhost.localdomain> 07ezx/current * rbfab348b07 10openezx/drivers/rtc/rtc-pcap.c: fix compile error.
05:53.43CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r1d366956ed 10openezx/sound/soc/pxa/ezx.c: add MODULE_LICENSE("GPL");
05:53.48CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r15d9587685 10openezx/arch/arm/mach-pxa/ezx.c: minor changes on keypads as per Stefan´s mail.
05:53.51CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * ra7672037c7 10openezx/drivers/rtc/rtc-pcap.c: remove non-sense cast
05:53.54CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r56716c6dc4 10openezx/drivers/rtc/rtc-pcap.c: pcap event is not irq
05:53.59CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r611001d865 10openezx/drivers/rtc/rtc-pcap.c: no need to mask on the assignment, the value would never overflow.
05:54.02CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * rb6a8df14e4 10openezx/.topdeps: add dependency on pcap_rtc
05:54.05CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * rb6ae44657a 10openezx/drivers/rtc/rtc-pcap.c: fix pcap event release
05:54.10CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r6a1fb72fe9 10openezx/arch/arm/configs/ezx_defconfig: Automatic defconfig update
05:54.13CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * re09964dc06 10openezx/drivers/rtc/rtc-pcap.c: cosmetic
05:54.16CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r48815bd292 10openezx/ (.topdeps .topmsg): create
05:54.23CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r26b38c2584 10openezx/drivers/rtc/rtc-pcap.c:
05:54.26CIA-29OpenEZX: remove uneeded function depth caused by bad copynpaste habit
05:54.28CIA-29OpenEZX: remove dubious hz event for adjusting the rtc clock based on system time automatically.
05:59.59WyrMargh, cia showed a lot of unrelated changes, and not the 2 relevant commits.
06:00.13WyrMtmzt: http://git.openezx.org/?p=openezx.git;a=shortlog;h=refs/heads/motoq/mach
06:02.40tmztthat looks right, let me test it
06:06.25tmztWyrM: how do I pull this on my branch?
06:06.36*** join/#openezx the_mgt_ (n=ich@dslb-082-083-239-030.pools.arcor-ip.net)
06:06.53WyrMgit checkout motoq/mach && git pull origin motoq/mach
06:07.13WyrM&& tg update
06:07.39WyrMaltough the tg update is not needed in this case, because the change was to motoq/mach branch
06:07.53WyrMif it was to a deeper branch, the tg update would be needed.
06:12.40CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r8e46cf1eac 10openezx/.topdeps: insert ezx/mach/pcap_rtc before ezx/mach/udc
06:12.50CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * raac987b08e 10openezx/drivers/usb/core/hub.c: use old_scheme_first for usb enumeration
06:13.02CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * re5d13205aa 10openezx/ (2 files in 2 dirs): change bp_rdy interrupt to rising and add a function to wake up bp
06:13.21CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r843bcebcef 10openezx/arch/arm/mach-pxa/motoq.c: gpio1 conflict: remove usb_vbus gpio, change pcap irq pin on mfp-table to 1
06:13.39CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r8d6a06b496 10openezx/.topdeps: remove dependency on ezx/pcap_rtc
06:13.52CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * re7d487535c 10openezx/arch/arm/mach-pxa/motoq.c: fix initial cs state
06:14.03CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r9a9bf78377 10openezx/drivers/char/ts0710_mux_usb.c: add ezx_wake_bp hook
06:16.44WyrMtmzt, motoq/machtype should be below motoq/mach on the stack
06:17.25tmztI'm not using it, I was trying to figure out topgit but I didn't mean to push it
06:17.53tmztwhen I created it I wasn't on master but on motoq/mach
06:18.02tmztso it made it dep on motoq/mach
06:18.19tmztusb client is not working
06:18.59tmztand no PCAP interrupts in /proc/interrupts
06:20.12WyrMwyrm@brutus:~/devel/linux-2.6$ git push openezx :motoq/machtype
06:20.13WyrMTo git@git.openezx.org:openezx.git
06:20.14WyrM<PROTECTED>
06:20.14WyrMwyrm@brutus:~/devel/linux-2.6$ git push openezx :refs/top-bases/motoq/machtype
06:20.14WyrMTo git@git.openezx.org:openezx.git
06:20.14WyrM<PROTECTED>
06:20.37WyrMplease, remote it locally to not automatically create it again.
06:20.56WyrMs/remote/remove/
06:21.20tmzttg delete -f ?
06:21.44WyrMalso, check if you don't have the v4l patch on your tree.
06:22.09tmztthis is the tree I pulled yesterday, I don't know where that came from
06:22.52WyrMyeah, some developer is running git remote --populate origin automatically before updating from our tree.
06:23.21tmztit looks like motoq/machtype is gone
06:23.26WyrMand is overriding all branch heads, and creating deleted branchs again.
06:23.32tmztit's not in git branch
06:24.09WyrMit probably still shows on tg summary if you don't deleted it locally.
06:24.41tmztmotoq/machtype is not in tg summary -t but v4l is
06:24.53WyrMplease remove it.
06:26.44tmzttg delete v4l/prolink-pixelview-mpeg-8000gt-fix-gpio
06:26.45tmzttg: fatal: invalid branch name: v4l/prolink-pixelview-mpeg-8000gt-fix-gpio
06:27.23WyrMrm -f `find .git -name prolink-pixelview-mpeg-8000gt-fix-gpio` will do the trick
06:29.03tmztgone on mine
06:30.10tmztany idea on PCAP?
06:30.43WyrMpcap what?
06:30.46WyrMirq?
06:30.50tmztyes
06:31.06WyrMdidn't my commit solve the issue?
06:31.11tmztno
06:31.25WyrMany error on dmesg?
06:35.35tmztno
06:40.02tmztdo we know that pcap cs is working? regs wouldn't work without it, right?
06:40.13WyrMright
06:40.28WyrMwould show 000000 or ffffff for all registers.
06:40.34tmztwould regs have the isr?
06:40.57WyrMyes.
06:41.05tmztwhere is it?
06:41.06WyrMbut I don't think this is the issue.
06:41.25WyrMunless you changed ezx-pcap.c code, this code was tested already
06:41.33tmzthaven't changed that
06:42.05WyrMthen it is something else, you are not getting the irq on gpio1
06:43.58tmztok, PCAP is in /proc/interrupts, 0:  0 GPIO-1 PCAP
06:45.51tmztcan I see the config = 0 in regs?
06:46.01WyrMno
06:46.48tmztwhat is the ezx_pcap_register_event in pcap_probe for?
06:46.55WyrMadc
07:00.05tmztso if I'm not getting interrupts it
07:00.11WyrMtmzt: are the values on /sys/bus/spi/.../regs ok?
07:00.23tmztit's either the alt function, which you set as gpio
07:00.27tmztyes, they look right
07:00.40WyrMwhat alt function?
07:01.31tmztgpio1_gpio
07:02.20WyrMits not alt function :)
07:02.36WyrMbut this is ok now, I don't see a reason for it to not work.
07:02.46tmzt<PROTECTED>
07:03.10tmztthat's from my dump from ce
07:08.09tmztis linux clearing the edge-detect?
07:10.45tmztWyrM: it doesn't look like the camera is registed if the CONFIG is not set
07:10.59tmztso the gpio_ changes aren't getting called
07:12.51tmztset_irq_type(pdata->irq, IRQ_TYPE_EDGE_RISING); set_irq_wake(pdata->irq, 1);
07:12.56tmztthat's not it then
07:17.53WyrMI don't see a reason for it to not work.
07:18.05WyrMare you sure that your tree is up-to-date?
07:18.19tmztyou mean pushed?
07:18.26tmztI just pulled your changes
07:18.35tmztthere in the file
07:19.18tmztezx/local/pcap
07:19.18tmztezx/pcap_leds
07:19.18tmztezx/asoc
07:19.28tmztthat's what you put there, is something missing?
07:20.12WyrMgit checkout motoq/mach && tg info shows Up-to-date?
07:20.28tmztyes
07:20.52tmztI've only pulled this branch though
07:21.00WyrMnp
07:21.01tmztgit pull origin motoq/mach
07:24.20CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * rdca752529a 10openezx/arch/arm/mach-pxa/motoq.c: this really should be the first thing on _init()
07:25.49tmztshould that make a difference with this issue?
07:26.50WyrMno.
07:26.57tmzt<PROTECTED>
07:27.23tmztjust checking mfp
07:28.13tmztGPIO28_SSP1_SFRM    MFP_CFG_IN(GPIO28, AF3)
07:28.31WyrMthis don't make a difference.
07:28.33tmztwhat was that used for in spi?
07:28.53WyrMCS
07:29.04tmztoh
07:29.22WyrMwe can control it via the hardware controller or via pxa2xx-spi's cs_control
07:29.59WyrMif you get correct spi I/O certainly CS is not the problem.
07:30.17tmztgpio_request(28, "PCAP CS"); gpio_direction_output(28, 0);
07:30.17WyrMwhat do you have on registers 0, 1, 2 and 3 of pcap?
07:31.02tmzt00 00480404
07:31.10tmzt<PROTECTED>
07:31.23tmzt02 00620980
07:31.35tmzt03 00100000
07:32.06WyrMecho "0 0" > regs
07:32.25WyrMand cat /proc/interrupts
07:32.33WyrMsee if it is 1 now.
07:32.40WyrMerr.
07:32.47tmztno
07:32.54WyrMecho "0 ffffffff" > regs
07:33.46tmztno
07:34.07WyrMwhat is reg 0 after the fffff write?
07:34.48WyrM00000004?
07:34.51tmzt00 00000004
07:34.57WyrMgood
07:35.13WyrMso, pcap is generating the interrupts and setting isr as expected.
07:35.39WyrMwe are back to gpio1 again, the interrupts are no arriving to pxa.
07:35.41tmztis gpio1 set as input anywhere?
07:36.00WyrMGPIO1_GPIO defaults to input
07:36.36WyrMbut you can set the direction to input on motoq_init() to be sure. :)
07:37.22tmzt<PROTECTED>
07:37.41WyrMhave you unplugged usb?
07:37.46WyrMor reboot?
07:38.10tmztI plugged in usb, the connection might have been unplugged at the other end before
07:38.20tmztbut /proc/interrupts is still 0
07:38.47WyrMso, at least now you know that pcap is being used as the transceiver too ;)
07:39.14tmztif it was plugged in when booting? that's what worked before
07:42.52WyrMhum.
07:43.06WyrMtry to echo "1 0" > regs
07:43.33WyrMand echo "0 ffffffff" > regs
07:43.34WyrMsee it it rises the interrupt count
07:43.38tmztok, I boot with usb unplugged and get 00 00000004
07:44.31WyrMit looks like the adcdone interrupt is not being unmasked.
07:44.32tmztyes, 2
07:45.35tmztbut ezx_probe is being called, or how else would PCAP be in /proc/interrupts?
07:45.50WyrMezx_probe?
07:46.06tmztezx_pcap_probe
07:46.08tmztsorry
07:48.40tmzt<PROTECTED>
07:48.51tmzt<PROTECTED>
07:48.51WyrMare you using .config = 0?
07:48.59tmztyes
07:49.21WyrMcat adc_battery &
07:49.28WyrMcat regs | head -1
07:50.44tmzt<PROTECTED>
07:51.09tmzt<PROTECTED>
07:51.11WyrMand cat adc_battery freeze?
07:51.16tmztno
07:51.23WyrMit showed a value?
07:51.35WyrMor it is still running on backgroud?
07:51.44tmztnow it did, but didn't lock up
07:52.04WyrMit showed a value?
07:52.21tmztnot still running is ps aux
07:52.29tmztno value either time
07:52.31WyrMthen it is freeze. :)
07:53.01tmztwait, two instances are still running
07:54.07WyrMit will never exit, you will wait years.
07:54.27tmzt<PROTECTED>
07:54.55WyrMecho "0 ffffffff" > regs may work
07:55.09WyrMor may also cause a panic. :)
07:55.26*** join/#openezx tombhadAC (n=tombhad@pD95DF856.dip.t-dialin.net)
07:55.28tmztdoes this make more sense now?
07:55.42WyrMno.
07:55.45WyrMI mean.
07:55.59WyrMI know what the problem is, but I don't know what causes it :)
07:56.04tmztwhat is supposed to unmask those?
07:56.19tmzt<PROTECTED>
07:56.40tmztI mean irq_handler
07:56.43WyrMlooking better, it is unmasked.
07:57.06WyrM01 007bfffe
07:57.21WyrMPCAP_IRQ_ADCDONE (1 << 0)
07:57.53tmztwhat is this: queue_work(pcap.workqueue, &pcap.work);
07:58.05WyrMa workqueue.
07:58.16tmztit calls pcap_work ?
07:58.24WyrMyes
07:59.24WyrMtry to move /* mask/ack all PCAP interrupts */ (and the 2 following lines) on ezx-pcap.c to after the  set_irq_wake(pdata->irq, 1);
08:00.13tmztis processor sleeping?
08:00.19tmztsomething in my ~/.config ?
08:00.23WyrMno
08:01.48WyrMno, thinking better it would not solve the issue.
08:02.39tmztwhy does ezx-pcap request adc on probe?
08:02.57WyrMbecause it is handled on the same driver :)
08:04.05WyrMadc is not the issue.
08:04.24WyrMit is working 00480404 -> 004880405
08:04.41WyrM0x1 is the adcdone interrupt.
08:05.54WyrMhave you tried to boot with the usb unplugged, and then use the adc?
08:06.11tmztyes, usb was unplugged that time
08:06.36WyrMI mean, not unplug after linux is running.
08:06.55WyrMboot linux with it unplugged, and plugging after linux is running.
08:07.07tmztyes, booted without usb plugged in
08:07.46WyrMand then catting adc_battery (without touching regs).
08:07.55WyrMhave you tried? still lock-up?
08:09.52tmztbooting without usb now
08:11.43tmztno ints even when plugging in usb
08:12.14tmztadc_battery reads
08:12.17tmzt578
08:12.28tmztadc is working
08:12.38WyrMyeah, what I expected :)
08:12.38tmztnot 9 ints
08:12.42tmztnow
08:12.49tmztboot with usb?
08:13.02WyrMnot really.
08:13.18tmztit doesn't look like g_ether is built in
08:13.24WyrMmore like something wrong with haret/ce.
08:14.01WyrMpcap is rising an interrupt before linux runs, and this is getting unhandled somehow.
08:14.02tmztbuilt as module
08:14.05tmztlike what?
08:14.22tmztthe home screen reads adc_battery/battcury a lot
08:14.35WyrMalotough the /* mask/ack all PCAP interrupts */ should be able to fully reset the hardware irq muxer.
08:15.43tmztharet clears interrupts, but should start with timer/ost
08:15.55tmztthe other zImage works also
08:16.29WyrMno, haret would not clear interrupts on a interrupt expander.
08:16.51tmztright, I mean pxa interrupts so that nothing could be reading adc
08:18.08WyrMgood, so now you know how to workaround this, with the cable plugged in at boot gpio1 will not work as interrupt source for some misterious reason.
08:18.23WyrMis adc_* stable now?
08:18.36tmztit worked on the other zImage, I could use usb network if cable was plugged at boot
08:19.00WyrMvery likely you can with this zImage also.
08:19.06tmztI haven't asked for ADC twice
08:19.17WyrMthe problem is probably intermitent, as it depends on a very tight timing.
08:19.21tmztwell, g_ether isn't built in
08:19.35tmztit's doing that every time now
08:19.40tmztit worked a few times before
08:20.01tmztI have 39 ints on gpio1
08:20.07tmzthaven't tried that many times
08:20.32WyrMsome stuff generates more than one int.
08:21.56*** join/#openezx LoneStar (n=xinzhen@114.92.179.21)
08:25.28tmztI need to copy /proc/config.gz from the other zImage and try that I think, something is just not right
08:26.00tmztit's built from the motoq.c in the tg patch in the old tree
08:26.32tmztexcept for those mfp modes which are now commented out
08:26.51tmztthat was before I tried soc/camera
08:27.21tmztok, usb works after modprobe g_ether, it sees the resets
08:28.01tmzt<PROTECTED>
08:28.26tmztit can't be doing vbus detects without PCAP interrupt???
08:28.37WyrMyes, you shouldn't get udc interrupts unless you register the events. its correct.
08:29.04WyrMby default, motoq should only get adcdone interrupts.
08:29.26WyrMis adc_* working?
08:29.29tmztoh, it's calling udc_connect on registering the gadget driver
08:30.55WyrMis adc working?
08:31.11tmzthaven't asked for ADC, didn't hang
08:31.36tmztthat's reported to dmesg, so no output from cat
08:32.29WyrMif you cat adc_battery it freezes again? or shows some value?
08:33.14tmztno
08:33.26tmztjust the haven't asked for ADC twice then exits
08:33.43tmztthe line in /proc/interrupts isn't GPIO-1 it's GPIO-l
08:33.55tmzt<PROTECTED>
08:34.32WyrM<PROTECTED>
08:34.37WyrMrc5 is out
08:35.42*** join/#openezx borman999 (n=quassel@ppp91-76-105-228.pppoe.mtu-net.ru)
08:38.28tmztstill something not being cleared?
08:40.11WyrMdon't know. I still don't know if your adc is working..
08:40.26tmztcharger id and battery are still printing the message but also printing a value now
08:40.39WyrMthe message?
08:40.51WyrMon dmesg, I haven't asked for adc...  bla bla bla?
08:40.56tmztyes
08:41.36WyrMso, you get more than one adc result every time you ask.
08:41.37WyrMweird.
08:42.35tmztnot clearing fast enough?
08:43.29WyrMhum, I don't think so. I program the adc to do a sinlge conversion. there shouldn't be a second adcdone interrupt
08:50.43tmztthere's something else, I was doing the printk's before the writes in my ezx-pcap.c
08:51.18tmztnot this one, the old one
09:27.33tmzt+CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ=y
09:49.28*** join/#openezx l403 (n=l@85.132.159.239)
09:51.28*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
10:00.54*** join/#openezx borman999 (n=quassel@ppp91-76-213-22.pppoe.mtu-net.ru)
10:10.37*** join/#openezx l403 (n=l@85.132.159.239)
10:23.41*** join/#openezx tmzt_ (n=tmzt@adsl-99-155-213-174.dsl.akrnoh.sbcglobal.net)
11:59.27*** join/#openezx gmzhuo (n=gmzhuo@118.113.160.73)
12:25.29*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
12:31.34gmzhuoHi,ao2
12:31.40ao2hi gmzhuo
12:33.30gmzhuoWyrM,are you here?
12:57.21*** join/#openezx l403 (n=l@85.132.159.239)
13:07.31*** join/#openezx johanbr (n=j@blk-7-128-3.eastlink.ca)
13:55.22*** join/#openezx l403 (n=l@85.132.159.239)
15:25.39*** join/#openezx jamesmck (n=james@gimel.esc.cam.ac.uk)
16:36.23WyrMao2: see "[PATCH] [ARM] pxa: fix I2C controller device being registered twice on Akita" on arm-linux.
16:56.15CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * rac8dfdc7fe 10openezx/drivers/leds/leds-pcap.c: fix checkpatch.pl errors
17:36.30CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r7bb5f98e7c 10openezx/arch/arm/mach-pxa/ezx.c: checkpatch.pl
17:42.24*** join/#openezx Wyr1 (n=wyrm@189.13.222.20)
17:43.58ao2WyrM, let me give a look at the akita init
17:56.01ao2WyrM, as I thought, they call pxa_set_i2c_info(NULL); also in common_init()
17:56.32ao2so it is actually _explicily_ called twice, this is not true for us
17:57.16WyrMao2: good, thanks :)
18:20.56CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r6b92276e57 10openezx/drivers/rtc/rtc-pcap.c: move to ezx-pcap.h
18:21.07CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * rf0d2f11a84 10openezx/include/linux/mfd/ezx-pcap.h: move from ezx/pcap_rtc
18:21.07CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r4870f09cc7 10openezx/drivers/rtc/rtc-pcap.c: cleanup
18:29.08ao2folks, sabrod_ offered us another A910 for development. Who wants it? If I got it I would play with the external smartpanel screen, the one I have have the external screen broken. But if anyone else is interested, just tell.
18:33.18WyrMsabrod_: thanks again! :)
18:33.32WyrMao2: take it.
18:34.09WyrMao2: btw, how is connected the 2nd screen on a910?
18:35.02ao2I guess via pxafb smartpanel interface? never checked.
18:45.17*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
19:33.50*** join/#openezx johanbr (n=j@JBrannlund2.MathStat.Dal.Ca)
20:49.04ao2wow ts0701 driver is very verbose now
20:50.22*** join/#openezx Kensan (n=ken@gw.ptr-80-238-204-38.customer.ch.netstream.com)
20:50.39ao2_mux_usb that is
20:51.24WyrMoh, sorry, my fault I left debug enabled.
20:52.04ao2nothing to be sorry about, really, it's just that I didn't expect so much output :)
21:32.00*** join/#openezx l403 (n=l@85.132.159.239)
21:40.05CIA-29OpenEZX: 03Antonio Ospite <ao2@openezx.org> 07ezx/current * ree7c490208 10openezx/arch/arm/mach-pxa/ezx.c:
21:40.05CIA-29OpenEZX: Follow the scheme suggested by Android for some buttons.
21:40.05CIA-29OpenEZX: Use KEY_SEND for the green phone button
21:40.06CIA-29OpenEZX: Use KEY_BACK for the cancel button, it might be used also to close apps
21:40.06CIA-29OpenEZX: Use KEY_END for the red phone button
21:53.10CIA-29OpenEZX: 03Antonio Ospite <ao2@openezx.org> 07ezx/current * rd9f867032e 10openezx/arch/arm/mach-pxa/ezx.c: Convert leading blanks to tab, make checkpatch.pl happy
21:53.21ao2WyrM, should I add the gpio_keys_button (A780 and A910 flip) patches to ezx/mach/gpio? I think we can submit that too.
22:12.15CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r38226d8de3 10openezx/ (.topdeps .topmsg arch/arm/mach-pxa/ezx.c): import from openezx.git
22:12.17CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r07c28d6019 10openezx/arch/arm/mach-pxa/ezx.c: register spi board info too
22:12.18CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r5e72a41806 10openezx/ (.topdeps .topmsg arch/arm/mach-pxa/ezx.c): import from openezx.git
22:12.19CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r7d993284bd 10openezx/ (.topdeps .topmsg arch/arm/mach-pxa/ezx.c): import from openezx.git
22:12.20CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r1bc2dc7d57 10openezx/arch/arm/mach-pxa/ezx.c: syntax error
22:12.23CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r9a1b2fd5ee 10openezx/arch/arm/mach-pxa/ezx.c: syntax error
22:12.24WyrMargh
22:12.26CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r5523a143ad 10openezx/ (.topdeps .topmsg arch/arm/mach-pxa/ezx.c): import from openezx.git
22:12.30CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r3a74cb010d 10openezx/arch/arm/mach-pxa/include/mach/mfp-pxa2xx.h: add gpio3 and gpio4 defines
22:12.35CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * re4cb7e6ef8 10openezx/arch/arm/mach-pxa/ezx.c: fix headers
22:12.40CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r4c5582ce50 10openezx/drivers/input/touchscreen/pcap_ts.c: remove unused headers
22:12.43CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * re16f0c64b7 10openezx/ (.topdeps .topmsg arch/arm/mach-pxa/ezx.c): import from openezx.git
22:12.45CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r1050e23221 10openezx/ (5 files in 2 dirs): import from openezx.git
22:12.49CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r80b0c64cab 10openezx/ (.topdeps .topmsg arch/arm/mach-pxa/ezx.c): import from openezx.git
22:12.52CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * re8d1be4711 10openezx/ (.topdeps .topmsg arch/arm/mach-pxa/ezx.c): import from openezx.git
22:12.57CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r697636408a 10openezx/ (5 files in 2 dirs): import from openezx.git
22:12.59*** part/#openezx alphaone (n=alphaone@2001:638:602:af01:0:0:0:1)
22:13.02CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r2b8a1271c8 10openezx/arch/arm/mach-pxa/ezx.c: fix headers
22:13.05CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r6c80f77ad4 10openezx/ (.topdeps .topmsg arch/arm/configs/ezx_defconfig): update defconfig
22:13.10CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * rf72a07540d 10openezx/ (.topdeps .topmsg): fake branch with all patches applied
22:13.12CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * rd062059306 10openezx/arch/arm/configs/ezx_defconfig: update defconfig
22:13.16CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * rffa83362b6 10openezx/arch/arm/configs/ezx_defconfig: Automatic defconfig update
22:13.20CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r91616b9622 10openezx/arch/arm/configs/ezx_defconfig: Automatic defconfig update
22:13.23CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r17accb9a7c 10openezx/arch/arm/configs/ezx_defconfig: updated defconfig
22:13.26CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r602e0c49a0 10openezx/arch/arm/configs/ezx_defconfig: update defconfig
22:13.31CIA-29OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07motoq/mach * r173b53d69c 10openezx/.topdeps: update dependency
22:13.34CIA-29OpenEZX: 03Stefan Schmidt <stefan@datenfreihafen.org> 07motoq/mach * r7cdb380d97 10openezx/arch/arm/configs/ezx_defconfig: ezx_defconfig: Switch CONFIG_USB_ETH from module to built in. Allow easier development
22:14.25WyrMI pulled ezx/current changes inside motoq/mach. :/
22:14.44WyrMao2: yes, ok for me.
22:16.31ao2WyrM, if I can help... maybe pushing motoq/mach only? Is that possible?
22:17.30WyrMao2: just don't pull now, let me try to rewind the motoq/mach branch
22:17.55ao2ok
22:18.20*** join/#openezx neolynx (n=u@155-252-221-213-pool.dsl.fcom.ch)
22:23.17*** join/#openezx tmzt (n=tmzt@adsl-99-51-209-163.dsl.akrnoh.sbcglobal.net)
22:23.26WyrM<PROTECTED>
22:23.36WyrMrewinded, hope nobody pulled my changes ;)
22:24.01tmztI keep getting knocked off here
22:24.07*** join/#openezx flo_lap (n=fuchs@e179103029.adsl.alicedsl.de)
22:26.20tmztWyrM: I'm running cat adc_battery repeatedly (while loop) on the old zImage, it's working fine.  It also prints debugging information for each write because of code added to ezx-pcap.c
22:26.45tmztnext I'm going to try adding a udelay instead of the debugging and use the same .config file on the new tree
22:27.46WyrMtmzt: any of the "maybe BP" messages?
22:27.57tmztno
22:28.04WyrMgood!
22:28.13WyrMI want to remove that code.
22:28.28tmztI haven't tested this on the new tree though, this is for baseline
22:28.39WyrMplease, test and report.
22:29.18tmzt<PROTECTED>
22:30.13WyrMwe usually get much more on touchscreen phones.
22:30.14WyrM:)
22:34.22ao2bye, see you tomorrow
22:34.28WyrMao2: bye!
22:39.31*** join/#openezx tombhadAC (n=tombhad@pD95DCD79.dip.t-dialin.net)
23:02.20tmztWyrM: it seems to have worked
23:02.51tmzt+       udelay(100);
23:03.11tmztit slows down the cat noticibly but returns a value without the error message
23:03.48tmztok, the first two reads were slower
23:04.26WyrMwhich error msg?
23:04.31WyrMand udelay where?
23:04.55tmztbefore the ezx_pcap_putget(value) in both ezx_pcap_read and write
23:05.11tmzterror message was the I havent requested ADC...
23:05.42tmztrunning the while loop now and it's reading adc_battery fine
23:06.29tmztalso, windows is very slow about updating the battery meter and doesn't even try until 20seconds after it starts but that might be when the home screen plugin is initialized
23:06.48tmztuntil then, it says dead battery and doesn't detect a charger
23:09.09tmztcould it be that it needs a delay between reading and writing the mask and this effectively gives it that?
23:09.21WyrMtmzt: pcap adc is used on touchscreen for ezx phones, im certain that this delay is not needed :)
23:09.46tmztok, I'll reduce it and try again
23:14.21WyrM<PROTECTED>
23:14.34WyrMno error msg.
23:15.16tmztthere could certainly be an alternate explaination but I don't know what it would be
23:15.47tmztwe still don't know how it could have gotten two interrupts on one read
23:16.04WyrMI can even use the touchscreen while catting the adc_* files on a loop.
23:17.50tmztafter writing fffffffff to 01 and 0 to 00 cat adc_battery locks up
23:18.01tmztI mean to cat regs
23:18.05WyrMof course.
23:18.24tmztwanted to see if usb could be used if booted without it plugged in
23:18.28WyrMyou are fiddling whit the interrupt controller registers with the interrupt handler running.
23:18.59WyrMit can, but not by manipulating irq. :)
23:19.25tmztit thought that's what you said to do yesterday
23:19.29tmztrebooting anyway
23:19.36WyrMbut not for usb.
23:21.28WyrMwant to get usb working?
23:22.52tmztwhen I get the regressions on newtree fixed, adc is the only one left
23:34.05tmztok, udelay(50) works without usb plugged in at boot, but gives "I havent requested ADC" with it plugged in at boot
23:34.39tmztso something else is going on
23:37.59tmztWyrM: what do you need, regs when it's plugged in or not?
23:38.08WyrMremove all udelay, its really not the correct solution.
23:38.22tmztok
23:43.36WyrMdo you get a error msg at boot?
23:44.07WyrMor only when you try the first cat adc_battery, or every time you cat adc_battery?
23:45.35tmztevery time
23:46.09WyrMand you get a sane value?
23:46.15WyrMor a freeze?
23:46.52tmztwas getting a sane value
23:47.01tmztonly froze when I wrote the isr
23:47.14tmztregister
23:47.33tmztok, I've taken out the udelay, will try booting that
23:48.06tmztwith or without usb?
23:48.16WyrMwithout.
23:52.57tmztok, booted
23:55.12WyrMany error at boot?
23:55.19WyrMand whent catting adc_battery?
23:55.57tmztno errors at boot
23:56.44tmztworks
23:57.01WyrMno error messages at all?
23:57.04tmztno
23:57.14WyrMgood, now plug usb
23:57.31tmztok
23:57.34WyrMand check if adc still works.
23:57.40tmztno
23:57.54tmztstill has a value but also the havent asked for ADC
23:58.16tmztunplugging while in the while loop starts working again
23:58.27tmztplugging it starts printing the error
23:58.57tmztthat works consistently
23:59.05WyrMerr, can you try it without a loop?
23:59.16tmztyes
23:59.43tmztsame thing
23:59.56tmztand the error is always printed twice

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