| 00:06.11 | stefan_schmidt | florian: The one on A1200? |
| 00:07.39 | florian | stefan_schmidt: no, on one of our customers device :-) |
| 00:07.45 | florian | unluckily |
| 00:08.17 | stefan_schmidt | ah, ok |
| 00:08.25 | stefan_schmidt | ~lart customers |
| 00:08.29 | stefan_schmidt | ;) |
| 00:23.59 | *** part/#openezx agd5f (n=DeucherA@65.242.105.130) |
| 03:24.32 | *** join/#openezx dotmonkey (i=dotmonke@218.5.61.143) |
| 03:25.57 | *** join/#openezx blkhawk[zzz] (i=blkhawk@pD9E072FD.dip.t-dialin.net) |
| 03:29.14 | *** join/#openezx dotmonkey_ (n=dotmonke@218.5.61.143) |
| 03:37.02 | *** join/#openezx dotmonkey (n=dotmonke@218.5.61.143) |
| 07:13.26 | *** join/#openezx proberts (n=proberts@cpe-76-167-78-52.socal.res.rr.com) |
| 07:58.05 | *** join/#openezx dion (n=dion@xen.inhex.net) |
| 08:41.38 | Hopscotch | good morning |
| 08:47.44 | *** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1) |
| 09:44.10 | *** join/#openezx dotmonkey (n=dotmonke@59.61.59.250) |
| 09:45.33 | alphaone | Cool |
| 09:45.59 | alphaone | I got an E680 booting 2.6 that doesn't turn off |
| 09:48.21 | alphaone | Stefan will probably write a mail about our progress today. |
| 09:52.24 | dotmonkey | alphaone:Great!and that means you can force AP&BP into normal mode? |
| 09:53.07 | koen | alphaone: yay! |
| 09:54.33 | stefan_schmidt | alphaone: wow, I leave your home and you get it working. :) |
| 09:54.37 | stefan_schmidt | alphaone: congrats |
| 09:55.01 | koen | using opentapi? |
| 09:55.05 | stefan_schmidt | dotmonkey: No, we really flash the kernel into phone. Still not able to reset the BP |
| 09:55.10 | stefan_schmidt | koen: yes |
| 09:55.14 | koen | cool |
| 09:56.02 | koen | stock opentapi or patched? |
| 09:56.15 | dotmonkey | alphaone:If we flash the 2.6 kernel and not handshake with BP,BP will die? |
| 09:56.17 | stefan_schmidt | dotmonkey: I just answer one of your mails. No time for E2 at the moment, but I'll help you later. |
| 09:56.52 | stefan_schmidt | koen: Must be stocked. alphaone did you changed something yesterday? |
| 09:57.12 | koen | cool |
| 09:57.34 | dotmonkey | stefan_schmidt:I haven't recive the mail,but I check just a moment ago... |
| 09:58.17 | stefan_schmidt | dotmonkey: Handshake is fine now on A780/E680. Problem is that you need a working mux driver and send special AT command under 60 sec from phone start. |
| 09:59.01 | dotmonkey | I got it,Thanks! |
| 09:59.10 | koen | stefan_schmidt: seperate initscript to do that? |
| 09:59.16 | stefan_schmidt | dotmonkey: ah, sorry. It's in my local notebook spool. University have a broken mail setup for me. Should ariive around 12:00 (Next time I have working network) |
| 09:59.43 | stefan_schmidt | dotmonkey: Really, that means uni has fixed the mail setup, yay. :) |
| 10:00.27 | dotmonkey | Nothing,I may try other stuff first,I just get bit-bang ssp pcap work right now :) |
| 10:00.28 | stefan_schmidt | koen: Right now we do it in opentapi initscript. But we had now luck until 4:00am |
| 10:00.54 | stefan_schmidt | alphaone: Go to THW today or not? |
| 10:00.56 | koen | stefan_schmidt: I was thinking with root-on-sd the fsck could that too long |
| 10:01.21 | koen | s/that/take/ |
| 10:01.43 | stefan_schmidt | koen: Main problem is that the 2.6 kernel startup takes longer then 2.4. We need to optimize it a bit. |
| 10:02.06 | stefan_schmidt | I still don't know what alphaone changed to get it working right now. |
| 10:02.45 | koen | stefan_schmidt: have a look at http://free-electrons.com/articles/optimizations/ |
| 10:03.04 | koen | stefan_schmidt: we could even use an initramfs to send the at command |
| 10:04.18 | stefan_schmidt | koen: Thanks for the link. Take a look later. Initrd should work. Just were tto tired yesterday. |
| 10:04.32 | dotmonkey | Can we do it in the kernel itself? :) |
| 10:05.24 | stefan_schmidt | dotmonkey: Doing it in kernel is really ugly. Thought about it already. |
| 10:06.10 | stefan_schmidt | dotmonkey: With some more tests we will find out if it's fine in a early initscript or initrd is needed. Should be fine. |
| 10:07.10 | stefan_schmidt | cu all later. |
| 10:07.50 | dotmonkey | stefan_schmidt:and will you write a document about that so that I can try to boot E2? |
| 10:08.07 | stefan_schmidt | dotmonkey: Expect a mail today |
| 10:08.12 | dotmonkey | cu |
| 10:40.05 | dion | hmm interesting... |
| 10:50.29 | *** join/#openezx florian_kc (n=fuchs@217.146.132.69) |
| 11:31.14 | *** join/#openezx dotmonkey (n=root@59.61.59.250) |
| 11:40.41 | florian_kc | hi all |
| 11:40.59 | stefan_schmidt | hi florian_kc |
| 11:41.13 | dion | florian_kc: |
| 12:31.46 | *** join/#openezx [g2] (n=g2@nslu2-linux/g2) |
| 13:09.14 | nh_work | morning all, congrats! :D |
| 13:10.55 | florian_kc | hi nh_home |
| 13:10.59 | florian_kc | eh nh_work |
| 13:11.47 | nh_work | morning florian_kc. Sounds like lots of progress. :) |
| 13:12.44 | *** join/#openezx webmind (i=webmind@feather.perl6.nl) |
| 13:22.29 | *** join/#openezx dotmonkey_ (n=dotmonke@59.61.59.250) |
| 13:34.17 | *** join/#openezx dotmonkey (n=dotmonke@59.61.59.250) |
| 14:09.08 | dion | stefan_schmidt: are you here? |
| 14:09.45 | *** join/#openezx agd5f (n=DeucherA@65.242.105.130) |
| 14:10.01 | stefan_schmidt | dion: More or less. I'm at work right now. |
| 14:10.29 | dion | can you give me link to 2.4 kernel without splash screen? i forget it |
| 14:11.16 | stefan_schmidt | dion: zImage or patch? |
| 14:11.35 | dion | zImage |
| 14:13.01 | stefan_schmidt | dion: moment I upload one |
| 14:13.29 | dion | ok.. thanks |
| 14:15.39 | stefan_schmidt | dion: http://people.openezx.org/stefan/tmp/zImage-2.4.20-mit-bildschirmausgabe |
| 14:15.52 | dion | thanks |
| 14:16.27 | stefan_schmidt | the postfix means with text framebuffer |
| 14:17.05 | dotmonkey | stefan_schmidt:have got your mail :) |
| 14:17.55 | stefan_schmidt | dotmonkey: fine |
| 14:18.33 | dotmonkey | Now just waiting for you EZX8 :) |
| 14:18.43 | stefan_schmidt | heh |
| 14:19.08 | dotmonkey | have tried to write pcap driver,but couldn't read the registers. |
| 14:21.00 | stefan_schmidt | hi mickeyl |
| 14:21.34 | mickeyl | morning stefan_schmidt. good work! |
| 14:21.53 | stefan_schmidt | mickeyl: Thanks goes also to Shoragan and alphaone |
| 14:21.58 | koen | http://lists.openezx.org/pipermail/openezx-devel/2007-January/001005.html |
| 14:22.35 | stefan_schmidt | It's always 'three hackers on one couch' fun. :) |
| 14:22.51 | mickeyl | sounds like a party hacking |
| 14:22.58 | stefan_schmidt | Mostly |
| 14:25.01 | dion | hmm.. interesting.. i just flashed A780 BP stuff to E680i |
| 14:25.48 | *** join/#openezx florian_kc (n=fuchs@217.146.132.69) |
| 14:26.47 | dion | heh.. phone boots without backlight and then turns off after 1-2 mins |
| 14:36.17 | *** join/#openezx florian_ (n=fuchs@dns.kernelconcepts.de) |
| 14:36.30 | *** join/#openezx florian_ (n=fuchs@dns.kernelconcepts.de) |
| 14:39.49 | *** join/#openezx florian (n=fuchs@dns.kernelconcepts.de) |
| 17:12.01 | *** join/#openezx vmaster_ (i=vmaster@p549B4E06.dip.t-dialin.net) |
| 17:47.41 | *** join/#openezx florian (n=fuchs@dns.kernelconcepts.de) |
| 17:58.07 | *** join/#openezx florian (n=fuchs@dns.kernelconcepts.de) |
| 18:20.00 | *** join/#openezx koen (n=koen@dominion.kabel.utwente.nl) |
| 18:30.39 | alphaone | stefan_schmidt, koen: No, stock opentapi. But I'll try to improve opentapi tomorrow |
| 18:35.49 | alphaone | But it's not needed to keep the phone alive |
| 18:38.35 | alphaone | I just moved opentapi to the very beginning of the init process. |
| 18:40.09 | alphaone | Best thing would be to find a way to extend the period the BP lets you connect to it |
| 18:40.32 | *** join/#openezx proberts (n=wataku__@rrcs-64-183-51-174.west.biz.rr.com) |
| 18:41.46 | alphaone | Or maybe pretend that the mux devices were opened in the kernel (i.e. open via the DLCI channels) instead of doing it when the actual file is opened. |
| 18:42.10 | alphaone | That should buy us enough time to start opentapi afterwards, I think. |
| 18:42.14 | stefan_schmidt | alphaone: hmm, we still need to figure out how we can reset the BP. |
| 18:42.27 | koen | or via initrd/initramfs |
| 18:43.01 | stefan_schmidt | right |
| 18:44.20 | alphaone | No, it's a different approach |
| 18:44.43 | alphaone | IMHO initrd/initramfs is not the way to go, either |
| 18:44.43 | stefan_schmidt | alphaone: initrd or reset? |
| 18:45.19 | stefan_schmidt | alphaone: Then we need to speed up 2.6 boottime |
| 18:45.28 | alphaone | Since once the tapiapplication closes the mux files you can't reconnect to them |
| 18:45.35 | alphaone | Which happened in 2.4 |
| 18:45.48 | koen | let me move the opentapi initscript a bit up in OE |
| 18:46.12 | koen | stefan_schmidt: surely the kernel doesn't take 60 seconds to boot? |
| 18:46.15 | alphaone | koen: I moved it to S01 in /etc/rcS.d |
| 18:46.42 | stefan_schmidt | koen: Not 60s but way longer then 2.4 |
| 18:46.42 | alphaone | It's more like 15-20 seconds after the LED is on |
| 18:46.48 | stefan_schmidt | We should benchmark it someday |
| 18:46.53 | alphaone | That seems to be the critical time |
| 18:47.10 | alphaone | Maybe it isn't about boot time at all. |
| 18:47.31 | stefan_schmidt | You mean decompression? |
| 18:47.31 | alphaone | Maybe it is n seconds after driver initialization. |
| 18:47.37 | alphaone | No |
| 18:47.52 | alphaone | The failure to open the muxN |
| 18:48.18 | alphaone | Maybe that has to be done directly after driver initialization. |
| 18:48.22 | stefan_schmidt | ah, sorry. I was at BP kill phone atm |
| 18:48.30 | alphaone | That would make more sense. |
| 18:49.59 | alphaone | Maybe the BP expects the muxN to be opened after initialization otherwise a timeout occurs. |
| 18:51.34 | stefan_schmidt | hmm, good question |
| 18:52.02 | stefan_schmidt | We have time to play with this tomorrow |
| 18:54.58 | alphaone | Yup |
| 18:55.21 | alphaone | That and the problems we are still experiencing in the driver |
| 18:55.38 | stefan_schmidt | ~lart kfree bug |
| 18:56.24 | proberts | stefan_schmidt: I think I came across a BP reset call in one of the ezx libs. |
| 18:56.35 | alphaone | I'll dig in to LDD3 some more and see if I can find the bug in the mux driver that prevents opentapi from reading anything |
| 18:57.25 | stefan_schmidt | proberts: Indeed there is a reset function. Harald never got it working. Once we have finished more stuff. We'll take a look at it again. |
| 18:58.18 | alphaone | Has anyone tried using the reset function and re-registering the handshake interrupt function? |
| 19:01.19 | proberts | Tonight I can try running the bp reset with the mux logger running. |
| 19:09.29 | *** join/#openezx koen (n=koen@dominion.kabel.utwente.nl) |
| 19:15.59 | alphaone | proberts: Great. |
| 19:16.45 | alphaone | proberts: There's one interrupt in arch/arm/mach-pxa/ezx.c: bp_rdy_intr |
| 19:17.47 | proberts | Anyone have a suggestion for changing (or forcing) the uid and gid file attributes in a tar file? Looks like the app manager on the E2 (at least 45P) preserves these on extract, which can screw up write permissions. |
| 19:18.43 | alphaone | If the handshake passes once the function modifies itself so to handle something about usbipc (a NOP in 2.6) so it wont complete the handshake a second time unless you modify the function |
| 19:20.06 | proberts | I'll see what Moto's EZX dumps out in the mux logger. Hopefully that can help you guys... I haven't dug too deep into that stuff yet- at least not nearly as deep as you guys. |
| 19:20.53 | alphaone | proberts: From tar: --group=NAME force NAME as group for added files |
| 19:20.58 | alphaone | Same for --owner |
| 19:21.03 | alphaone | okay, thanks |
| 19:21.05 | proberts | alphaone: I can only get that to work if my system has a user with the same name. |
| 19:21.23 | proberts | (and group) |
| 19:21.44 | alphaone | proberts: And if you add --numeric-owner can you give numbers instead? |
| 19:22.35 | proberts | Same problem... it fails if the uid/gid don't already exist on the system. |
| 19:22.51 | alphaone | hmm |
| 19:23.51 | proberts | I can write something that post-processes the tar, but I would think there would be a way to have tar either ignore uid/gid or set it to any possible number. |
| 19:23.52 | alphaone | proberts: Creating an archive fails? |
| 19:24.06 | proberts | Yeah - (no user "ezx") |
| 19:24.22 | proberts | same with number... - no user 2000 |
| 19:24.56 | alphaone | Hmm, it just succeeded here |
| 19:25.04 | alphaone | tar -cvf test.tar --numeric-owner --owner=892 foo bar |
| 19:25.15 | proberts | what version of tar do you have? |
| 19:26.02 | alphaone | tar (GNU tar) 1.16 |
| 19:27.17 | proberts | Hmmm... I don't know what version I have at home where I tried it. Here I have 1.15 here at work and it works. |
| 19:27.33 | proberts | thanks |
| 19:27.47 | alphaone | np |
| 19:28.26 | proberts | Is user ezx uid=2000 on all the phones? |
| 19:29.08 | proberts | same with group.... I'll update the wiki |
| 19:29.46 | alphaone | I don't know |
| 19:32.10 | proberts | I wonder why motorola decided to do that. Permissions aren't preserved... just uid and gid |
| 19:49.53 | *** join/#openezx florian_ (n=fuchs@87.193.28.13) |
| 19:52.26 | florian_ | re |
| 19:55.11 | stefan_schmidt | wb florian |
| 19:55.52 | florian | hi stefan_schmidt |
| 19:56.51 | *** join/#openezx dion (n=dion@xen.inhex.net) |
| 20:01.56 | *** join/#openezx kenws (n=ken@dslb-088-064-141-255.pools.arcor-ip.net) |
| 20:54.32 | *** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1) |
| 21:02.02 | alphaone | proberts: on the image I have ezx is uid 2000 and gid 233 |
| 21:02.56 | alphaone | Sorry for AFK, I had to fix the network here. Damn arp spoofer |
| 21:17.57 | stefan_schmidt | Time for bed. Good night all. |
| 21:18.40 | alphaone | good night |
| 21:18.53 | alphaone | stefan_schmidt: When will you be here tomorrow? |
| 21:19.23 | stefan_schmidt | alphaone: We have communications systems course together at 9:45 am |
| 21:20.19 | alphaone | yeah but what time will you be here? :-) |
| 21:20.45 | alphaone | In case I miss that lecture or something similar unexpected happens |
| 21:20.51 | stefan_schmidt | heh |
| 21:21.01 | stefan_schmidt | 11:15 at your flat. |
| 21:21.04 | alphaone | :-) |
| 21:21.30 | alphaone | 11:15: mobile communications :-) |
| 21:22.03 | stefan_schmidt | heh |
| 21:22.05 | stefan_schmidt | night |
| 21:45.06 | *** join/#openezx koen (n=koen@dominion.kabel.utwente.nl) |
| 21:56.32 | proberts | alphaone: I wonder if there's a way to strip that out of the tar file all together? |
| 21:59.47 | alphaone | proberts: Strip what? |
| 21:59.56 | proberts | uid and gid from the tar file |
| 22:01.03 | alphaone | seems to work only the other way around |
| 22:01.26 | alphaone | if you untar you can tell the owner to be yourself |
| 22:02.16 | proberts | Yeah - I'm still trying to figure out why Mot would decide to do that... since it didn't work that way in v33P on the E2. I wonder if it's some lame attempt at security. |
| 22:09.31 | alphaone | hmm |
| 22:10.18 | alphaone | What are you trying to achieve? |
| 22:12.19 | proberts | I've got it working now. The problem I had is my installed mpkg app couldn't write to it's own install directory when installed on the phone (for some reason under 45P the mpkg fails to install to the mem card- don't know why yet). The problem turned out to be permission based- all the files had my user uid/gid. |
| 22:17.46 | alphaone | ah |
| 23:34.33 | *** part/#openezx agd5f (n=DeucherA@65.242.105.130) |