IRC log for #openezx on 20070628

00:34.03*** join/#openezx agd5f (n=alex@c-69-143-17-173.hsd1.va.comcast.net)
02:39.51*** join/#openezx AlexZhang (n=alex@58.61.45.63)
02:50.10drWyrMhi Alex.
02:55.00*** join/#openezx AlexZhan1 (n=alex@58.61.45.63)
02:55.29AlexZhan1hi Daniel
02:55.57AlexZhan1I am working on mounting mmc part2
02:57.23*** join/#openezx AlexZhan1 (n=alex@58.61.45.63)
02:58.44*** join/#openezx AlexZhan1 (n=alexzhan@58.61.45.63)
02:59.13drWyrMgood! any success?
03:00.24AlexZhan1;-), I just telnet to original firmware, and then mount my TF card part2
03:01.17AlexZhan1I want to install your busybox to my phone, and then using dmesg to see the boot log.
03:02.55drWyrMhave you built the part2 image with OpenEmbedded?
03:05.15AlexZhan1no, I just download the compiled busybox from openezx.org
03:05.36AlexZhan1I cannot build OE, I got an error.
03:07.11AlexZhan1even my compiler for linux-2.6.21 was downloaded from Internet
03:07.31drWyrMahh, you are working on the orig FW. On a780 i cat /proc/kmsg to see kernel messages.
03:08.44drWyrMwe use OE toolchain to compile .21. You compiled gcc yourself?
03:09.28AlexZhan1no, I downloaded gcc binary from Internet.
03:09.38AlexZhan1but it works. ;)
03:10.33AlexZhan1from here http://www.codesourcery.com/gnu_toolchains/arm/download.html
03:11.28AlexZhan1?
03:12.39AlexZhan1Daniel, what is mux_cli?
03:14.13drWyrMmux_cli is a usb demultiplexer to comunicate with the BP.
03:15.35AlexZhan1I lsmod on orig FW, just find two module are loaded, mux_cli and paniclogger
03:15.52AlexZhan1so I think MMC driver is not a module, do you think so?
03:16.43drWyrMif you have a card working on the slot then it is buit-in.
03:17.56drWyrMi think it is one of drivers/mmc/bvd*.c
03:17.58AlexZhan1yes, my card is working.
03:19.04AlexZhan1bvd-ezx_mmc.c?
03:19.07drWyrMbvd-ezx_mmc.c
03:19.12drWyrM:)
03:19.20AlexZhan1oh! I think so. ;)
03:19.42AlexZhan1and do you have an e680?
03:19.57drWyrMno, an a780.
03:20.18AlexZhan1who works on e680?
03:20.41drWyrMe680 is almost the same hardware as a780.
03:20.46AlexZhan1because i find a file : bvd-e680-mmc.c.
03:21.08AlexZhan1but a780 's mmc driver is a module
03:21.23drWyrMwe had people working on it before, but right now i think that nobody is working on e680.
03:21.23AlexZhan1seems e680 is using bvd-e680-mmc.c?
03:23.06drWyrMno, e680 kernel is on a different tree.
03:23.27drWyrMsame tree as a780.
03:25.46AlexZhan1oh! ok
03:26.19drWyrManyway, e680 mmc driver is already working on .21 :)
03:28.28AlexZhan1after I modify the mmc driver of A1200, how can I know if it is working for me?
03:29.51AlexZhan1from the boot message.
03:31.19drWyrMyes, you will see a message like <6>mmcblk0: mmc0:bffc SU02G 1985024KiB
03:32.47AlexZhan1thanks. and another question, is it possible to root to nfs  now?
03:33.04AlexZhan1if no, just because USB does work?
03:33.34drWyrMno. Its because there is no EOC support.
03:34.18AlexZhan1EOC?
03:34.27drWyrMthe mini usb port on a1200 support USB, audio, and serial connections, and the Emu One Chip controls the port behaviour.
03:34.50drWyrMemu = Enchanced Mini USB
03:34.51AlexZhan1oh! I got
03:35.09AlexZhan1so, we cannot control the usb mode right now?
03:35.34AlexZhan1just because no EOC support. right.
03:35.39AlexZhan1right?
03:35.45drWyrMit exists on a780/e680 too, but on a1200/e2/e6 it is different.
03:35.51drWyrMright
03:36.52AlexZhan1oh! it may be another headache work! O:-)
03:38.06drWyrMmore fun. :)
03:42.50AlexZhan1ok, back to MMC driver!
03:43.19AlexZhan1any hints? I am reading the bvd-ezx-mmc.c
03:45.59drWyrMcompare 2.4 and 2.6 (a1200-mci.patch)
03:46.19drWyrMmmc detect irq number is wrong on 2.6
03:47.02drWyrMcompare the gpio pins configration
03:47.16AlexZhan1just mmc pins?
03:49.36drWyrMyes, the mmc driver should not touch pins not related to mmc. :)
03:51.12AlexZhan1I read the bvd-ezx-mmc.c, I think because A1200 dont support hot plug mmc( dont know if A780/e680 supported?). so it use MMC_DATA3 to detect if a mmc card is on the slot.
03:51.43drWyrMa780/e680 support it.
03:51.49AlexZhan1/* Check DAT3 for triflash card detection */
03:51.49AlexZhan1gpioval = GPLR(GPIO_MMC_DATA3) & GPIO_bit(GPIO_MMC_DATA3);
03:51.49AlexZhan1DEBUG(2, "check if triflash exist %lx\n", gpioval);
03:51.49AlexZhan1
03:51.49AlexZhan1/* if triflash DAT3 is high, there is triflash card    */
03:51.49AlexZhan1if( GPLR(GPIO_MMC_DATA3) & GPIO_bit(GPIO_MMC_DATA3) )
03:51.51AlexZhan1<PROTECTED>
03:51.53AlexZhan1else
03:51.55AlexZhan1return 1;
03:53.14drWyrMhum...
03:53.21AlexZhan1so, can we forget the detect pin, just assume the card is always on the slot?
03:57.43drWyrMi dont know. take a look on pxamci.c on 2.6 if the interrupt routine passed to a1200_mci_init() is required to use the card.
04:00.26AlexZhan1oh, let me see...
04:01.08drWyrMpxamci.c line 511 and 408. I think it is not required.
04:03.44AlexZhan1ok, I add some printk there?
04:03.59drWyrMmaybe the pcap driver is not working, and it is not powering on the card.
04:04.01drWyrMwait.
04:05.16AlexZhan1yes, maybe
04:05.24drWyrMarch/arm/mach-pxa/ezx-pcap.c line 30 change to #if 1
04:06.18AlexZhan1ok
04:07.31AlexZhan1and then compile?
04:11.01drWyrMyes, this will dump all read/write operations on pcap registers to the framebuffer. If you see any inconsistence in then, the driver is not working properly.
04:12.03drWyrMeg, always reading 00000000, reading weird values just after writing some different data to the same register..
04:13.37AlexZhan1:-[I dont understand what's the meaning of these values...
04:14.57*** join/#openezx jamesmck (n=james@gimel.esc.cam.ac.uk) [NETSPLIT VICTIM]
04:14.57*** join/#openezx Hopscotch (n=cd@touchdown.maintech.de) [NETSPLIT VICTIM]
04:14.57*** join/#openezx TommyMT (n=tk@touchdown.maintech.de) [NETSPLIT VICTIM]
04:14.57*** join/#openezx CIA-24 (n=CIA@208.69.182.149) [NETSPLIT VICTIM]
04:15.01drWyrMdont care about the meaning of them now. just assure that it is working.
04:15.20drWyrMor look at power_ic.h :)
04:17.12AlexZhan1ok, but how can I know if the pcap driver is powering to the card?
04:20.12drWyrMyou have to change the same bits as the power_ic_periph_set_flash_card_on() does on bvd-ezx_mmc.c line 304
04:21.00AlexZhan1oh, let me see.
04:21.55drWyrMAlex, i have to go sleep. its 1:25AM here :)
04:22.09drWyrMwork tomorrow morning :/
04:22.13drWyrMgood night.
04:23.27AlexZhan1oh, sweet dream, thanks for you guide!
04:24.20drWyrMreport any success, and thank you for your work.
04:24.25drWyrM:) bye.
04:24.38AlexZhan1bye!
05:23.54*** join/#openezx koen (n=koen@dominion.kabel.utwente.nl)
05:24.38*** join/#openezx Fry-kun (n=kostya@h-66-166-233-66.snvacaid.covad.net)
05:29.42*** join/#openezx pleemans (n=peter@leemanpe.ugent.be)
05:30.22*** join/#openezx goxboxlive (n=goxboxli@141.80-202-161.nextgentel.com)
06:38.57*** join/#openezx james__ (n=james@gimel.esc.cam.ac.uk)
06:43.30*** join/#openezx pH5 (n=ph5@p5485FD2D.dip.t-dialin.net)
07:07.06Hopscotchgood morning
07:20.29*** join/#openezx james__ (n=james@gimel.esc.cam.ac.uk)
07:45.32*** join/#openezx dion (n=dion@inhex.net)
07:50.40*** join/#openezx florian (n=fuchs@217.146.132.69)
08:26.26*** join/#openezx ade|desk (n=adavey@194.200.143.249)
09:07.52*** join/#openezx z72ka (n=hermanj@r5af183.net.upc.cz)
09:38.45*** join/#openezx file (n=file@neutrino-114-86.joshua-colp.com)
09:49.20*** join/#openezx LapTop006 (n=laptop00@sparc006.chriskaine.com.au) [NETSPLIT VICTIM]
09:51.41*** join/#openezx file (n=file@neutrino-114-86.joshua-colp.com)
10:05.16*** join/#openezx goxboxlive (n=goxboxli@141.80-202-161.nextgentel.com)
10:21.34*** part/#openezx AlexZhan1 (n=alexzhan@58.61.45.63)
10:25.23*** join/#openezx |dion| (n=dion@inhex.net)
11:43.32*** join/#openezx pleemans (n=peter@leemanpe.ugent.be)
12:38.28*** join/#openezx blobster (i=enne@mozart.cc.tut.fi)
12:45.09*** join/#openezx drWyrM (i=c88e3027@gateway/web/cgi-irc/ircatwork.com/x-fbd96ff8e2389d3f)
12:45.21drWyrMgood morning.
12:45.48koenhey drWyrM
13:01.01*** join/#openezx dion (n=dion@inhex.net)
13:07.43slapinhi, all!
13:07.47slapinpH5, hi
13:07.57slapinpH5, anything interesting yet?
13:11.10pH5slapin: I have massaged the pxa-camera code a bit to better fit with CodingStyle and started a small ov96xx.ko platform driver that can enable the clocks / gpios, detect the ov96xx revision and dump all registers (not more until now).
13:11.14pH5no pictures yet :)
13:12.06slapinpH5, please push it!!!
13:14.03pH5slapin: ok, five minutes..
13:31.03drWyrMYes, all this camera progress is great! :)
13:32.07drWyrMhow is the pxa CI driver going? anything working already? we have other camera sensors to support for EZX (mt9m111 and adcm2700).
13:33.42pH5I don't understand the i2c part yet. It seems the attach_adapter callback is called twice and the second probe times out.
13:36.22pH5drWyrM: are those chips also clocked by the CI MCLK?
13:37.04pH5what is even more interesting, are there any users of those chips that do not use the pxa CI?　(omap, samsung, etc?)
13:37.31slapinpH5, yes, pxa270 has 2 busses
13:37.43slapinpH5, and it is aclled for both of them
13:37.47slapin*called
13:38.03slapinpH5, that's ok for most cases...
13:39.02drWyrMpH5: i think so. i only have an a780 with mt9m111. And the driver is very similar to ov9640 on 2.4. pxa CI and i2c.
13:39.03pH5ah, sounds kind of stupid that the i2c core tries to look for our chip on the poweri2c.
13:46.59*** join/#openezx drath_ (i=vmaster@p5B07F2DE.dip.t-dialin.net)
13:51.56pH5slapin: pushed, again. I've cleaned out the currently unused code from ov96xx
13:57.10slapinpH5, thanks!
13:58.05pH5slapin: I intend to work on wiring up the ioctls the V4L2 way next.
13:58.51slapinpH5, look at *ioctl* file in video driver, I think you could want to reuse some code from it.
14:00.00koenpushed to where?
14:03.23pH5koen: slapin's git repo at ossfans.org
14:03.33pH5slapin: do you have a public git server or gitweb set up?
14:04.40slapinpH5, I think I'll setup it by weekend.
14:04.53slapinit will be readonly git://
14:05.42pH5slapin: so we have until the weekend to make that driver work? ;)
14:06.14slapinpH5, it don't have to be working, you know :)
14:08.16*** join/#openezx AlexZhang (n=alexzhan@61.141.182.80)
14:46.43drWyrMFry-kun: ping
15:03.28drWyrMFry-kun: AlexZhang found that the PCAP driver is not working properly on a1200, and it is required to power up the TF card. So, ezx_ssp.c and ezx-pcap.c is currently where you should be working.
15:29.17drWyrMFry-kun: pcap driver is fixed., CS on a1200 is active high (active low on a780)
15:29.49drWyrMi will commit it to svn asap.
15:55.07*** join/#openezx dion (n=dion@inhex.net)
15:59.27*** join/#openezx file (n=file@asterisk/developer-and-muffin-lover/file)
15:59.27*** join/#openezx Hopscotch (n=cd@touchdown.maintech.de) [NETSPLIT VICTIM]
15:59.27*** join/#openezx TommyMT (n=tk@touchdown.maintech.de) [NETSPLIT VICTIM]
15:59.27*** join/#openezx CIA-24 (n=CIA@208.69.182.149) [NETSPLIT VICTIM]
16:11.23Fry-kunmorning
16:11.51Fry-kunooh, so TF will work now?
16:14.44drWyrMno.
16:14.47drWyrM:)
16:15.34Fry-kunaw :(
16:15.56Fry-kuni don't see any svn updates, btw
16:16.11Fry-kunyou didn't commit the pcap fix yet?
16:16.21drWyrMhe found that the PCAP driver was reading just 0000 from any registers, and we fixed that. You have power up the card by setting a pcap register, so it was never going to work.
16:16.53Fry-kuni see
16:32.09*** join/#openezx TimRiker (i=timr@205.233.55.201)
16:36.16*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
16:40.42*** part/#openezx ade|desk (n=adavey@194.200.143.249)
16:42.53ao2hi
16:43.02Fry-kunhello
16:46.53drWyrMhi Antonio.
16:49.16ao2drWyrM, I played a little bit with hciattach :) Ah, and I noted that if we do "hciconfig hci0 up && hciconfig hci0 down" the bt devive lose its bdaddr
16:49.50ao2swap 'up' and 'down' in the above line, of course
16:50.40*** join/#openezx LuitvD (n=luitvd@beigetower/luitvd)
16:54.06drWyrMhumm...
16:55.39drWyrMnever tried that. But this should not be an issue, If you want to turn BT off properly, you have to kill hciattach, and turn it off with the EBTP command.
16:56.03drWyrMyou would have to hciattach again, so you can set bdaddr again too :)
16:57.25ao2sure
16:58.16drWyrMi was thinking yesterday about the function of the motorola patches to the bcm2035 firmware...
16:58.23drWyrMi think that they are audio related.
16:59.57drWyrMso, in the end we may need to patch the FW again to get bluetooth audio working... :)
17:01.13z72kaao2: I tested it yesterday.. this is not problem, if hciattach is running its not recomended to shutting down hci device :)
17:01.47ao2drWyrM, well, with no working audio driver that is really not an issue :P
17:01.50z72kaao2: I mean, for using is only bluetooth start and stop.. thats all we need
17:01.53ao2z72ka, I see
17:02.09drWyrMits just a guess, but it is very likely that they are audio related. bcm2035 audio ssp port is conected together with PXA, neptune and PCAP, all on the same SSP.
17:02.32drWyrMon a1200 it is diferent (but also the chip, is a bcm2045).
17:02.51drWyrMao2: right, no audio, no bt audio related issues ;)
17:03.14ao2ah drWyrM, where the motorola soft take the btaddr value from?
17:03.26ao2bdaddr
17:03.28drWyrMcommandline.
17:03.55drWyrMit is called by tapisrv, i dont know where tapisrv gets it.
17:05.04drWyrMmay be from BP, there are some small BP code groups, that looks like serial numbers.
17:06.41ao2our bdaddr are like 00:15:A8:xx:xx:xx , the first three bytes are motorola id, I guess
17:06.55z72kadrWyrM: We have bdaddr probably in small ROM.. Marex dumped it and found (by then) half of bdaddr :)
17:07.00ao2the last three are derived from some BP query?
17:08.08z72kaao2: Hmmm... and Palm has 00:07:e0: ....
17:08.33drWyrMlater, we should work on bcm2035 power management support.
17:09.33drWyrMthere is a pin to wake up ap on BT events, in EZX i think we dont need to mess with the BT_RESET pin, we can cycle power with the EBTP command.
17:11.43z72kadrWyrM: Hmm.. on Palm is BT_WAKE gpio still unknown...
17:30.55*** join/#openezx LoneStar (n=xinzhen@218.81.28.68)
18:16.20*** join/#openezx agd5f (n=deuchera@65.242.105.130)
19:00.59*** join/#openezx WM (n=wyrm@201.2.68.83)
19:10.07*** join/#openezx TimRiker (i=timr@205.233.55.201)
19:19.10*** join/#openezx florian (n=fuchs@e179096021.adsl.alicedsl.de)
19:37.33*** join/#openezx Fry-kun (n=kostya@h-67-100-89-239.snfccasy.covad.net)
20:12.26*** join/#openezx koen (n=koen@dominion.kabel.utwente.nl)
21:20.29*** join/#openezx Fry-ku1 (n=Fry-kun@63.82.6.66)
21:38.05*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
22:10.24*** join/#openezx laurens (n=laurens@62.58.149.208)
22:29.25*** part/#openezx agd5f (n=deuchera@65.242.105.130)
22:32.19*** join/#openezx Shoragan (n=shoragan@datenfreihafen.org)
22:33.33*** join/#openezx alphaone (n=daniel@datenfreihafen.org)
23:10.16*** part/#openezx drWyrM (n=wyrm@201.2.68.83)
23:15.45*** join/#openezx agd5f (n=alex@c-69-143-17-173.hsd1.va.comcast.net)
23:21.49Fry-ku1why doesn't this silly debug work?? :(

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