IRC log for #openezx on 20080628

00:40.07*** join/#openezx sabrod (n=laptop@lns-bzn-40-82-251-139-252.adsl.proxad.net)
03:15.24*** join/#openezx sabrod (n=laptop@lns-bzn-40-82-251-139-252.adsl.proxad.net)
04:31.27CIA-5OpenEZX: wyrm * r2248 /branches/linux-2.6-arm/ (ezx-mci.patch ezx-pcap.patch):
04:31.28CIA-5OpenEZX: * change module description
04:31.28CIA-5OpenEZX: * change set_sw helper logic
04:31.28CIA-5OpenEZX: * implement set_vaux helper
04:31.28CIA-5OpenEZX: * move mmcsd functions to machine file
04:31.28CIA-5OpenEZX: * rename some defines
05:13.13CIA-5OpenEZX: wyrm * r2249 /branches/linux-2.6-arm/ezx-pcap.patch: convert to mfp
05:39.39*** join/#openezx gmzhuo (n=gmzhuo@221.237.75.24)
05:59.36*** join/#openezx drath_ (i=vmaster@p5B07D22E.dip.t-dialin.net)
06:45.30*** join/#openezx jonwil (n=jonwil@121.221.230.164)
07:09.01*** join/#openezx shoragan (n=shoragan@debian/developer/shoragan)
07:28.03*** join/#openezx dion (n=dion@inhex.net)
07:49.58*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
09:53.30*** join/#openezx blkhawk (i=blkhawk@g228067105.adsl.alicedsl.de)
10:37.20*** join/#openezx pH5 (n=ph5@p5485C15A.dip.t-dialin.net)
10:48.57*** join/#openezx rettichschnidi (n=reto@246-143.203-62.cust.bluewin.ch)
11:41.18*** join/#openezx shoragan (n=shoragan@debian/developer/shoragan)
11:41.18*** join/#openezx janneg (n=janne@tichy.grunau.be) [NETSPLIT VICTIM]
11:41.18*** join/#openezx stefan_schmidt (n=sschmidt@sicherheitsschwankung.de)
12:22.42*** join/#openezx gmzhuo (n=gmzhuo@222.210.144.27)
12:24.41gmzhuoI boot linux-2.6-arm,rootfs use mmcblkp2,sys pannic after free init memory:"Kernel panic - not syncing: Attempted to kill init
12:25.00gmzhuothe rootfs is ok for 2.6.21.4?
12:27.00gmzhuoanybody here?
12:34.39shoraganstefan_schmidt, http://cgi.ebay.de/Motorola-A1600-Neu-BT2-0-A2DP-3-2-MP-Touchscreen-GPS_W0QQitemZ180258920042QQihZ008QQcategoryZ69245QQrdZ1QQssPageNameZWD1VQQcmdZViewItemQQ_trksidZp1638Q2em118Q2el1247
13:13.55jonwilis the A1600 linux?
13:15.01jonwilI know the a1800 is not (most likely because its dual CDMA and GSM and moto havent written or cant write (well maybe cant write and be complient with both GPL and CDMA licensing) a CDMA stack for linux)
13:17.33stefan_schmidtjonwil: CDMA stack is on another ship. Same situation as GSM, so problem with the license for this. Moto states that A1800 is no linux. We will see if this is true.
13:17.55jonwilI see no reason for Motorola to lie
13:17.56stefan_schmidtshoragan: hmm, atm I don't have the money to buy just another phone. :)
13:18.41stefan_schmidtjonwil: I don't say they lie. Still there can be communication problems between engineering at management.
13:18.44jonwilAlthough it may be that the guy who said "a1800 is not linux" was using incorrect information
13:18.46stefan_schmidtAnyway, we will see
13:19.13jonwilI for one intend to focus any time I spend on this on my Z6
13:19.21jonwiland to a lesser extent the V8,E8 and U9
13:19.41jonwilas those 4 phones are what most people out for a moto linux phone will have available to pick from
13:19.53stefan_schmidtAnyone with such a phone tested dimtriys stuff yet?
13:20.02jonwilno, I havent had the time
13:20.36stefan_schmidtSo, we have the same situation as before.
13:21.16jonwilthe thing that annoys me the most is that despite all the promises, Motorola continues to violate the GPL left right and center and makes very few moves to correct it.
13:21.27jonwilFor example, no sign of the firmware for the E8 at all
13:21.37jonwilAnd no sign of the kernel for the E6 IIRC
13:22.49jonwilnot to mention the fact that there are a number of firmware revisions for any given phone and moto has not released matching source code for every revision that exists on a retail phone
13:23.52stefan_schmidtI know. That is the reason why I'm getting angry everytime Moto mentions hie "open source" involvement.
13:24.32stefan_schmidtPerhaps it is time for a "why opensource.motorola.com sucks" blogpost
13:24.49jonwilI think the issue for moto is that each source release (each phone and each firmware revision for that phone) is a totally seperate release
13:24.51stefan_schmidtJust to make my point once again
13:25.30stefan_schmidtWell, I don't care. It's motos problem to get the workflow right. They already know about this for 3 years.
13:25.59jonwilMore than anything, I think moto is trying to be very carefull NOT to give away any more information than they have to
13:26.05jonwilboth in terms of the hardware on the phones
13:26.11jonwiland in terms of how the software works
13:26.14jonwiland how the security works
13:26.16jonwiland etc
13:26.25jonwilbut yes they DO need to get better at it
13:26.55jonwilI for one want the source code to the R60_G_80.33.27R firmware on my Z6
13:27.11stefan_schmidtok, I'll setup an open letter for moto
13:27.32stefan_schmidtjonwil: If you have some hard facts I can integrate there please mail them to me.
13:27.38stefan_schmidtIt will take some days.
13:27.45jonwilwhat sort of facts?
13:27.56stefan_schmidtNeed to read all forum entries at the opensource side, etc..
13:28.06stefan_schmidtjonwil: What kernels, version are missing, etc
13:28.26jonwilThe only hard fact I have is that they havent published source code for that firmware revision
13:28.29stefan_schmidtI need to dig into this and that will cost time
13:28.33jonwilthe 27R revision
13:29.06stefan_schmidtSomething like this. But it in a mail if you like. Could same me some time
13:29.27jonwilI dont have any other information, thats the only usefull bit I have
13:29.54stefan_schmidtAnyway, I search for it myself
13:30.12jonwilI dont think that the GPL violations are deliberate (as in moto saying "we are going to violate GPL and we dont care"). Some companies are deliberatly violating GPL but I dont think moto is one of them anymore
13:30.18stefan_schmidtgoes to do his homework
13:30.45stefan_schmidtNo, but still they miss stuff.
13:31.14jonwilI think its a lack of time and resources to devote to these phones.
13:31.22stefan_schmidtBut discussion in a group of 2-3 people make no sense.
13:31.31stefan_schmidtI don't care if they lack time.
13:31.38stefan_schmidtThey had three years
13:31.48jonwilok, well you give them this letter
13:31.56jonwilThey need to compy with GPL for everything
13:32.38jonwilIt would be even better if they released more than they had to (e.g. userland bits for old EZX platform) but I doubt that could ever happen
13:33.06stefan_schmidtIt will not happen.
13:33.28stefan_schmidtMy point is that they should not say they are OSS friendly if they are not.
13:33.49stefan_schmidt<- away, doing some research for facts
13:36.08jonwilok
13:36.10jonwildo that
13:36.12jonwil:)
13:40.55*** join/#openezx Ikkebr (n=ikkibr@189.73.167.63)
14:03.20*** join/#openezx Ikkebr2 (n=ikkibr@189-72-2-36.smace700.dsl.brasiltelecom.net.br)
14:10.15*** join/#openezx johanbr (n=j@blk-103-247-181.eastlink.ca)
14:27.10*** join/#openezx pleemans (n=peter@d51A5E76A.access.telenet.be)
14:28.42*** join/#openezx Ikkebr (n=ikkibr@201-66-201-155.smace700.dsl.brasiltelecom.net.br)
14:44.48*** join/#openezx Ikkebr2 (n=ikkibr@201-66-203-103.smace700.dsl.brasiltelecom.net.br)
14:47.23WyrMgood morning.
15:29.13*** join/#openezx flyhorse|linux (n=flyhorse@219.132.251.173)
16:00.22gmzhuonow i can boot 2.6.24/25/2.6-arm with machineID
16:00.22WyrMgmzhuo: hi.
16:00.36WyrMgmzhuo: what was wrong before?
16:01.09gmzhuoI excute boot_usb kernel_image,without the machineID
16:01.52gmzhuoI found some problems about can't use usb_boot on gen_blob
16:02.18WyrMwhat is wrong?
16:02.41gmzhuoI even can't send packet to the phone,
16:02.49WyrMboot_usb works with gen-blob usb mode just fine on a780, e680 and a1200.
16:03.41WyrMgmzhuo: are you able to compile gen-blob from the sources?
16:04.03gmzhuoyes
16:04.07WyrMgood!
16:04.31gmzhuoon gen-blob,ret = usb_bulk_write(hdl, phone.out_ep, buf, cur, 0);
16:04.32gmzhuo<PROTECTED>
16:04.32gmzhuo<PROTECTED>
16:04.32gmzhuo<PROTECTED>
16:04.32gmzhuo<PROTECTED>
16:04.46gmzhuoit fail in there
16:05.51WyrMgmzhuo: probably on e6 the original MBM will not leave the EOC chip routing to the pxa udc pins.
16:06.10gmzhuothen i swap the ep[0] to USB_OUT,ep[1] to USB_IN
16:06.32WyrMand it works?
16:06.48gmzhuousb_bulk_write works
16:07.07gmzhuobut ep2_recv cant' get any frame
16:07.23WyrMhum...
16:07.25gmzhuothis no intterupt
16:08.19WyrMlooks like eoc isnt configured properly.
16:08.26gmzhuobut when I plug out and plug in,there have intterupt\
16:08.33WyrMgen-blob relies on the eoc configuration left by mbm.
16:08.48gmzhuoand I can get descript of usb device of phone
16:09.04gmzhuoso I think eoc configuration is right
16:09.08WyrMgmzhuo: are you trying with a flashed gen-blob?
16:09.24gmzhuolet me try
16:09.35WyrMusb is not expected to work if you sent gen-blob itself via usb.
16:15.27gmzhuothe same
16:15.50gmzhuoand there is another problem
16:16.28WyrMgmzhuo: revert your changes to boot_usb before testing.
16:16.43gmzhuoI boot 2.6-arm,mmcblk0p2 as rootfs/ext3
16:17.26gmzhuothe root mout success,but show not sync....,kernel kill init
16:18.22gmzhuoI have test ther before gen-blob
16:18.24*** join/#openezx pH5 (n=ph5@p5485C15A.dip.t-dialin.net)
16:18.29WyrMi have no idea what may be causing the panic.
16:19.01*** join/#openezx jamesmck (n=james@gimel.esc.cam.ac.uk)
16:19.35gmzhuoI add the ezx-eoc of 2.6.25 to 2.6-arm,kernel panic too
16:19.47WyrMgmzhuo: please, dont do that.
16:20.12gmzhuowhy?:) I just test it!
16:20.24WyrMits very likely that the eoc code is causing the panic :)
16:20.56gmzhuoI think too
16:21.13gmzhuobut i want to use nfs rootfs
16:21.26WyrMthere is no relation on the mci code and eoc, i tested mci on a1200, so i think it is working on e6 too.
16:21.45gmzhuomci work on e6 too
16:22.12gmzhuorootfs has mounted,but panic
16:22.21WyrMgmzhuo: if you want more hardware support, use .25. the linux2.6-arm branch doesnt support usb yet.
16:22.49WyrMgmzhuo: i dont think that the mci code is panicing on e6. I think that it is the EOC changes you did yourself.
16:23.51gmzhuothere some problem too,My mmcblk02 is ext3,but if I select ext3 in kernel,the kernel will stop,no oops message
16:24.34WyrMgmzhuo: are you building ext3 support built-in?
16:24.51Zba_Phyhi, do you know where I can find the TrueFFS driver for 2.6 branch ?
16:24.57WyrMi am working with an ext2 fs, so i cant really help you there.
16:25.40gmzhuoI think EOC will cause 2.6-arm panic,but first there no EOC code,it panic too after mounted rootfs and free init memory
16:25.57WyrMgmzhuo: with what error?
16:26.35WyrMgmzhuo: probably, it is trying to mount "3e00" instead of mmcblk0p2, can you confirm this to me?
16:27.24WyrMIf this is the case, it is already known, and easy to solve. The kernel is reading the ATAGS set-up by MBM for the 2.4 kernel.
16:28.12WyrMin this case, just let your phone rest without battery for 2 minutes, and boot it to MBM red screen without booting the 2.4 kernel.
16:28.36WyrMor change the params addr on ezx.c
16:28.49gmzhuoit is do mout mmcbllk02
16:29.34WyrMthen something is wrong with your ext3 support or your cmdline.
16:30.24gmzhuoI'm recompiling the code without eoc
16:30.54WyrMgmzhuo: please, do so. I cant debug your problem if I dont have access to the code you are using :)
16:31.59gmzhuoCONFIG_CMDLINE="console=tty1 noinitrd root=/dev/mmcblk0p2 rw rootfstype=ext3 rootdelay=1 ip=192.168.16.2:192.168.16.1:192.168.16.1:255.255.255.0 debug mem=32M@0xA0000000 mem=16M@0xAC000000"
16:32.38gmzhuoMY LINUX IS TOO SLOW :(
16:32.47WyrMgmzhuo: on the linux2.6-arm branch there is no need to use the mem= argument on cmdline.
16:34.08gmzhuoCan you give me you config file?
16:34.56WyrMsure! 'netcat wyrm.is.dreaming.org 57483
16:35.07WyrM> .config'
16:36.12WyrMdone?
16:37.31gmzhuofail
16:39.24WyrMlet me upload it to somewhere, just a min.
16:39.30gmzhuoOk
16:40.47WyrMhttp://people.openezx.org/wyrm/defconfig-2.6-arm-20080628
16:41.05gmzhuoVFS: Mounted root (ext3 filesystem)
16:41.10gmzhuo\
16:41.24gmzhuoFreeing init memory :108k
16:41.35gmzhuothen kernel panic
16:41.58WyrMpanic why?
16:42.25gmzhuonot syncing;attempte to kill init
16:42.50WyrMcan you read where is PC?
16:42.59WyrMand have you removed your EOC changes?
16:43.23gmzhuoI can't no more message
16:44.11WyrMsend me a photo of your screen please.
16:46.58gmzhuoIt's not enougth clear to read on the photo:(
16:47.37gmzhuohow about this ?error: 'pxa25x_device_udc' undeclared here (not in a function)
16:48.51WyrMgmzhuo: quilt pop -a; git-pull -f; (cd patches && svn up); quilt push -a
16:49.19WyrMyour tree is not up-to-date
16:50.49gmzhuoquilt pop -a ???
16:51.12WyrMand note that we are using the pxa branch of linux2.6-arm. see with 'git-branch' if you are working on the same branch.
16:51.42WyrMyes, pop all quilt patches, update your git tree, update your exz patchset, quilt push all patches.
16:52.12gmzhuothere no this command on my system
16:52.44WyrMgmzhuo: do you exec patch -p1 < file.patch for each ezx patch?
16:52.44*** join/#openezx FosforoEee (n=Fosforo@200.222.210.2)
16:53.13WyrMhow do you apply our patchset to your kernel tree?
16:53.33gmzhuono ,I'm not family with git,So I copy it out
16:53.47gmzhuoyes,I use patch
16:54.03gmzhuoBut in another copy
16:54.11WyrMdont do that.
16:54.15gmzhuoso I need update git tree ?
16:54.18WyrMuse git and quilt.
16:54.42WyrMyes, please. Install git and quilt, and work with that.
16:54.51gmzhuohow do it update git tree to the newst ?
16:55.19WyrMgmzhuo: have you git-cloned it? or donwloaded from somewhere else?
16:55.45gmzhuoI use git-cloned
16:56.12WyrMthen, see if you are working on the pxa branch with 'git-branch'
16:56.43gmzhuoI am not clear
16:57.50WyrMlets start from scratch.
16:58.06WyrMon another directory, clone linux2.6-arm again.
16:58.27WyrMthen: git-checkout pxa
16:58.58WyrMdownload our patchset, put it on the 'patches' directory inside the linux2.6-arm dir, and 'quilt push -a'
16:59.21WyrMinstall quilt if you dont have it installed, of course.
17:01.15gmzhuoI excute git-pull -f in the git-tree,have get the pxa27x.c and pxa25x.c,now I copy it out
17:03.12WyrMgmzhuo: your tree is probably at an inconsistent state. You have changed files by executing 'patch -p1 <' and will have to revert these changes.
17:04.47gmzhuothat path din't excut patch
17:06.11gmzhuoIt's Ok
17:07.05gmzhuoMaybe is my mem=32M@... CMDLINE make it crash
17:09.17*** join/#openezx rettichschnidi (n=reto@246-143.203-62.cust.bluewin.ch)
17:12.00*** join/#openezx Gottox (n=Gottox@s01.de)
17:14.00WyrMgmzhuo: no it cant be. the mem= lines are just ignored currently.
17:24.32gmzhuoI need to go to bed!bye
17:31.16*** join/#openezx johanbr (n=j@JBrannlund2.MathStat.Dal.Ca)
18:19.41*** join/#openezx rettichschnidi (n=reto@246-143.203-62.cust.bluewin.ch)
18:27.29*** join/#openezx rettichschnidi (n=reto@246-143.203-62.cust.bluewin.ch)
19:37.16*** join/#openezx flo_lap (n=fuchs@f049184217.adsl.alicedsl.de)
20:35.39CIA-5OpenEZX: wyrm * r2250 /branches/linux-2.6-arm/pxa27x-udc-no-up2ocr-mangling.patch: this patch fixes pxa27x-udc.c for OpenEZX
20:55.28*** join/#openezx sabrod (n=laptop@lns-bzn-54-82-251-74-134.adsl.proxad.net)
21:21.56WyrMmainline udc driver still doesnt work :(
21:23.11WyrMthe host recognizes the device, my usb0 goes up, and the error counter on ifconfig keeps counting. Any clues?
21:43.58WyrMph5: is the mainline gadget driver working (with ethernet gadget) for magician?
21:46.36pH5WyrM: yes, although I was worried about the missing up2ocr handling, too.
21:46.56pH5I used to reset UP2OCR in udc_command every time, but it seems to work even without that.
21:48.05WyrMi removed it from udc_enable, and i am setting it before calling set_udc_info().
21:48.51WyrMthis way i can see the interface getting configured. my host detects it and all, but i get only errors on /sbin/ifconfig :(
21:50.08pH5is the interface up on the device?
21:52.00WyrMit seems so, i can see "usb0: full speed config #1: 100ma, Ethernet Gadget, using CDC Ethernet Subset on the screen"
21:52.24WyrMs/set on the screen"/set" on the screen/
21:52.40WyrMafter a bunch of "USB reset"
21:54.41pH5I'm not sure if that is sufficient. I need to have the ip= cmdline parameter or have ifconfig usb0 $IPADDR up in the rootfs to force the interface up.
21:55.09pH5If I don't do that, the host recognizes an ethernet gadget but only times out in the netdev watchdog.
21:55.20WyrMahh, you mean the net iface? yes, i am using ip= on cmdline.
21:56.59pH5ah ok, then I'm out of ideas. up2ocr and ip address on the interface were the only issues I had with that driver so far.
22:30.05WyrMpH5: do you see ep_warn(ep, "usb_ep %s already enabled, doing nothing\n" warnings on magician?
22:36.15pH5WyrM: yes, and I complained to robert about it. he said it is harmless.
22:38.44*** part/#openezx blkhawk (i=blkhawk@g228067105.adsl.alicedsl.de)
23:28.37*** join/#openezx Ikkebr (n=ikkibr@201-66-201-169.smace700.dsl.brasiltelecom.net.br)
23:34.21*** join/#openezx gmzhuo (n=gmzhuo@222.210.144.27)
23:34.40*** join/#openezx johanbr (n=j@blk-103-247-181.eastlink.ca)

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