| 00:07.41 | _dion_ | stefan_schmidt: is http://people.openezx.org/stefan/patches/patches-2.6.16-2.6.16.13-ezx7.tar.bz2 last available patch? |
| 00:07.57 | stefan_schmidt | _dion_: Yes |
| 00:08.21 | _dion_ | i should patch 2.6.16 or 2.6.16.13 vanilla kernel? |
| 00:14.19 | _dion_ | stefan_schmidt: also i have question about floating point operations.. As we know, we have no hardware float, so we must use software float.. For original 2.4 kernel all userspace software should be build with patched gcc. How about 2.6-ezx kernel? |
| 00:17.37 | stefan_schmidt | _dion_: the .13 patch is included in our patchset. Vanilla 2.6.16 is the right choice. |
| 00:17.53 | _dion_ | ok. |
| 00:17.54 | stefan_schmidt | _dion_: I use still ezx-crosstool for 2.6 kernel |
| 00:18.18 | _dion_ | ok.. so i will try to build kernel now |
| 00:18.22 | stefan_schmidt | _dion_: I've tried with a new 4.1 oe toolchain some time ago and this even worked fine. |
| 00:19.19 | _dion_ | stefan_schmidt: should binaries from debian/arm works with float without recompilation? |
| 00:19.27 | _dion_ | recompiling |
| 00:21.15 | stefan_schmidt | _dion_: They should, but I'm not sure. To many special things with crosscompile and softfloat... |
| 00:21.26 | _dion_ | ok |
| 00:22.53 | jaebird | yo! |
| 00:23.04 | jaebird | i enjoy lurking |
| 00:23.45 | jaebird | you guys are making incredible progress...my E2 likes that :) |
| 00:24.22 | jaebird | my hesitation to jumping in is time and that my E2 is how i communicate :( |
| 00:25.34 | jaebird | i'd buy another one...but i really want something with wifi next time |
| 00:26.24 | alphaone | Yeah, it's pretty exciting :-) |
| 00:26.52 | alphaone | _dion_: We managed to get an strace. i had to modify the source though |
| 00:27.18 | _dion_ | hmm. strange.. why strace works for me? |
| 00:27.51 | alphaone | I don't really know |
| 00:28.16 | alphaone | maybe subtle changes in userland / glibc |
| 00:28.43 | _dion_ | maybe |
| 00:28.48 | alphaone | stefan_schmidt is just going through the strace and commenting the reads/writes |
| 00:31.39 | stefan_schmidt | I will checkin somekind of mux<->tapisrv communication protocol later this night. |
| 00:32.49 | stefan_schmidt | _dion_: quilt push -a should be all you need |
| 00:33.07 | stefan_schmidt | _dion_: need to extract the patches dir in kernel-tree root |
| 00:33.19 | _dion_ | oops.. i extract them to separate dir |
| 00:33.57 | _dion_ | yeah..it works |
| 00:34.19 | stefan_schmidt | :) |
| 00:35.17 | _dion_ | stefan_schmidt: should i use defconfig-e680? |
| 00:35.29 | stefan_schmidt | _dion_: As you're going this way, it would be cute if you could document your steps in the wiki. :) |
| 00:35.42 | stefan_schmidt | yes defconfig is fine |
| 00:35.58 | _dion_ | ok. will do this but i already have installed crosstool |
| 00:36.21 | stefan_schmidt | _dion_: That's fine. crosstool is already described. |
| 00:36.46 | _dion_ | btw my gcc binary is arm-linux-gnu-gcc.. i don't know, why :) |
| 00:37.17 | _dion_ | hmm.. got error: arm-linux-gnu-ld: unrecognized option `-EL' |
| 00:37.42 | _dion_ | oops.. i forget this.. it works |
| 00:43.38 | _dion_ | got zImage |
| 00:46.00 | stefan_schmidt | nice |
| 00:46.09 | stefan_schmidt | Try it with boot_usb |
| 00:46.44 | _dion_ | hmm.. strange.. after using boot_usb got just white screen on phone |
| 00:48.26 | jaebird | hmm, parawireless is just down the road from me! |
| 00:49.03 | _dion_ | it works for third time |
| 00:49.54 | _dion_ | got kernel panic (Unable to mount root fs).. Sounds good |
| 00:50.52 | stefan_schmidt | _dion_: Yes, it tries to mount ext3 on your SD card. :) |
| 00:51.15 | stefan_schmidt | _dion_: We can't use /dev/mtd2 atm. It's broken. |
| 00:51.32 | stefan_schmidt | _dion_: Is working. Moment |
| 00:51.47 | _dion_ | will try to boot again |
| 00:52.53 | stefan_schmidt | _dion_: Compile your kernel with this CONFIG_CMD: root=/dev/nfs nfsroot=192.168.1.10:/export/opie-image rootdelay=5 3 |
| 00:53.03 | stefan_schmidt | And add it to the wiki. ;) |
| 00:53.27 | _dion_ | ok.. but i can't boot again.. got just white screen |
| 00:53.58 | stefan_schmidt | white screen or garbage? |
| 00:54.11 | stefan_schmidt | We had this several time on our own. |
| 00:54.14 | _dion_ | white screen with some blue lines (vertical) |
| 00:54.22 | _dion_ | or cyan.. |
| 00:54.25 | _dion_ | not sure :) |
| 00:54.55 | stefan_schmidt | Guess there are still some uninitialized hardware problems |
| 00:55.34 | _dion_ | yeah.. or possible it's because we boots it from blue screen |
| 00:55.45 | _dion_ | i will try to flash it later |
| 01:01.38 | _dion_ | hmm. kernel is too large to flash |
| 01:01.58 | _dion_ | it's 1130652 bytes.. |
| 01:02.36 | _dion_ | and CG32 is 00020000-0011ffff |
| 01:03.20 | alphaone | _dion_: can you compile some drivers as module? |
| 01:03.35 | alphaone | That way you would still have the functionality |
| 01:03.55 | _dion_ | i think that yes, but i still have no rootfs.. |
| 01:08.40 | stefan_schmidt | _dion_: http://people.openezx.org/laforge/openezx-debian-sarge-root-20060512.README |
| 01:09.12 | _dion_ | looks like all stuff is already documented :) |
| 01:09.13 | stefan_schmidt | _dion_: Or oe OPIE root-fs. Your choice. http://people.openezx.org/mickeyl/README.html |
| 01:09.30 | _dion_ | i don't want gui for first time :) |
| 01:09.51 | stefan_schmidt | _dion_: Most of it. But it's still hard to find for people jump new into the project. |
| 01:10.31 | stefan_schmidt | _dion_: I use the OPIE image beacuse it's smaller. Most time without GUI. Just usbnet or serial console. |
| 01:10.32 | _dion_ | hmm.. only 2 of 10 boots was successfull |
| 01:11.18 | stefan_schmidt | hmm, not good |
| 01:11.18 | _dion_ | i prefer debian-way.. will try to move all to modules and flash kernel |
| 01:12.17 | stefan_schmidt | ok, that's fine |
| 01:13.45 | alphaone | good luck |
| 01:18.05 | _dion_ | hmm. tarball with all required debs is 40 mbytes for etch |
| 01:18.56 | *** join/#openezx nevarrie_ (n=jgrant@CPE-24-163-233-209.kc.res.rr.com) |
| 01:27.40 | _dion_ | almost noting to build as modules |
| 01:42.03 | *** join/#openezx nh_home (n=nh@aurora.msi.umn.edu) |
| 01:43.08 | _dion_ | something strange but finally it's booted and same kernel panic about rootfs |
| 01:44.24 | _dion_ | stefan_schmidt: the most interesting is that it can be booted without baterry.. just connect charger and wait :) |
| 01:45.22 | _dion_ | but each bot is sucessful |
| 01:45.22 | _dion_ | boot |
| 01:51.21 | stefan_schmidt | alphaone looks very tired. :( |
| 01:51.22 | _dion_ | tomorrow i will write something to wiki about compiling and flashing kernel |
| 01:51.34 | stefan_schmidt | _dion_: nice |
| 01:51.35 | _dion_ | :) |
| 01:52.12 | _dion_ | stefan_schmidt: is it any way to get serial console without miniusb port? |
| 01:52.31 | stefan_schmidt | _dion_: solder wires on the pcb |
| 01:53.08 | _dion_ | i want to use miniusb charger as power suply |
| 01:55.20 | stefan_schmidt | http://svn.openezx.org/trunk/doc/ipc/BP-tapisrv-AT-communication.txt |
| 01:55.35 | stefan_schmidt | _dion_: It's 3:00 am here. :) |
| 01:56.31 | _dion_ | :) so you have at least one hour :) |
| 01:57.06 | alphaone | _dion_: good night |
| 01:57.44 | _dion_ | alphaone: thanks... |
| 01:57.49 | stefan_schmidt | _dion_: good night |
| 01:57.58 | _dion_ | stefan_schmidt: doc is good.. not so difficult to write tapisrv replacement |
| 01:58.26 | stefan_schmidt | _dion_: Yes, not difficult, but still some work. |
| 02:03.23 | nh_home | stefan: anything interesting been going on in the last couple days? |
| 02:03.55 | stefan_schmidt | nh_home: We straced tapisrv to capture communication with BP |
| 02:04.28 | nh_home | stefan: neat, did you figure out what's keeping the phone on? |
| 02:04.33 | stefan_schmidt | nh_home: alphaone is on the way the write a small replacement for it. We hope to keep the phone alive this way. |
| 02:04.46 | nh_home | stefan: that's great! |
| 02:04.54 | stefan_schmidt | nh_home: We found some interesting looking AT commands. We will see. |
| 02:06.51 | *** join/#openezx jonwil (n=jonwil@220-245-202-116-wa-pppoe.tpgi.com.au) |
| 02:06.58 | jonwil | hi |
| 02:10.33 | stefan_schmidt | hi jonwil |
| 03:39.21 | alphaone | I used my opentapi instead and the phone is alive for 7 mins already |
| 03:39.51 | stefan_schmidt | :) |
| 03:40.13 | alphaone | :-) |
| 03:40.14 | stefan_schmidt | 4:40 am is the right time for doing work. ;) |
| 03:40.31 | alphaone | And I guess I'm past feeling tired. |
| 03:41.03 | alphaone | I hope that makes up for me missing out yesterday :-) |
| 04:02.09 | *** join/#openezx chairuou_home (n=chairuou@58.187.147.233) |
| 05:17.45 | *** join/#openezx _dion_ (n=dionua@xen.inhex.net) |
| 05:28.58 | *** join/#openezx jonwil (n=jonwil@220-245-202-116-wa-pppoe.tpgi.com.au) |
| 05:34.41 | alphaone | good morning |
| 05:44.34 | jonwil | hi |
| 05:48.58 | stefan_schmidt | ok, one last test with opentapi and 2.6. Afterwards fallling asleep. It's 6:50 am here and we are _still_ awake. |
| 05:49.04 | stefan_schmidt | btw, hi jonwil |
| 05:53.48 | *** join/#openezx nevarrie_ (n=jgrant@CPE-24-163-233-209.kc.res.rr.com) |
| 05:55.17 | jonwil | have you made any progress? |
| 06:01.02 | alphaone | short resume before I fall asleep: |
| 06:02.29 | alphaone | We managed to strace our E680 and the string _dion_ found is the same here. |
| 06:03.24 | jonwil | ok, so what does this mean? |
| 06:03.35 | alphaone | We now have a program called opentapi that can be started instead of the tapisrv and it keeps the phone alive. |
| 06:03.44 | jonwil | ok, so it actually works with 2.6 kernel? |
| 06:04.02 | jonwil | is this opentapi available? |
| 06:04.06 | alphaone | No, it is unable to open /dev/muxN in 2.6 |
| 06:04.20 | jonwil | so thats a mux_cli issue with 2.6? |
| 06:04.29 | alphaone | http://svn.openezx.org/trunk/src/userspace/opentapi/ |
| 06:05.05 | alphaone | Yeah, we believe that it is usbipc that is still not handling things correctly. |
| 06:05.30 | jonwil | what is this socket opentapi opens? |
| 06:06.41 | jonwil | so basicly opentapi opens the dev/muxn devices and then reads everything they spew out |
| 06:06.48 | jonwil | and then writes something back? |
| 06:07.19 | alphaone | It is supposed to be a way to send and receive commands to/from the BP |
| 06:07.36 | alphaone | But as you can see it's not yet implemented :-) |
| 06:07.41 | jonwil | :) |
| 06:07.45 | alphaone | yes |
| 06:08.10 | alphaone | The commands it writes back are based on the observations from the strace |
| 06:08.15 | jonwil | so those 3 commands are the 3 commands that must be sent to the BP to keep it alive? |
| 06:08.29 | jonwil | the +CBC, +USBSTAT and +CBSM? |
| 06:09.06 | alphaone | Next thing to try is if it is enouch to just send one or two of them but these three definately keep the phone alive |
| 06:09.29 | alphaone | The +CBC is Battery charge AFAIK and can probably be omitted |
| 06:09.57 | jonwil | yes AT+CBC gets battery charge |
| 06:09.58 | alphaone | The other two are not documented in any of the standards we looked through |
| 06:10.09 | jonwil | CBSM sounds like a GSM standard |
| 06:10.11 | jonwil | with the C |
| 06:10.35 | alphaone | There were a few starting with C we weren't able to find |
| 06:11.18 | jonwil | does anyone have a log of what the BP sends back on the /dev/mux lines? |
| 06:16.14 | alphaone | yes |
| 06:19.50 | alphaone | http://people.openezx.org/alphaone/logs/ |
| 06:22.59 | jonwil | the +CLK response seems to be from the RTC in the baseband |
| 06:23.18 | jonwil | +EBPV is the baseband processor version |
| 06:23.32 | jonwil | +CEME is the emergency numbers |
| 06:25.30 | jonwil | I think its the AT+USBSTAT command that is doing the work |
| 06:25.43 | jonwil | i.e the +USBSTAT=255 is what makes the BP realize the AP is awake |
| 06:25.48 | jonwil | or whatever |
| 06:26.54 | alphaone | That's my most likely candidate, too. |
| 06:27.47 | alphaone | However, stefan_schmidt has left 30 mins ago with the phone and I really need to get to bed now. |
| 06:28.13 | alphaone | So good night everyone |
| 06:28.34 | alphaone | Or good day or whatever it is :-) |
| 06:28.35 | [g2] | alphaone THX and sweet dreams! |
| 07:58.12 | Hopscotch | good morning |
| 08:26.23 | jonwil | hi |
| 08:41.01 | *** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1) |
| 09:05.08 | jonwil | is "base" the name of a mobile phone company? |
| 09:25.10 | *** join/#openezx kenws (n=ken@dslb-088-064-139-249.pools.arcor-ip.net) |
| 12:33.19 | *** join/#openezx _dion_ (n=dionua@xen.inhex.net) |
| 12:33.38 | _dion_ | hiall |
| 12:55.55 | jonwil | hi |
| 12:56.12 | jonwil | has anyone heard of a mobile phone carrier called BASE? |
| 12:56.48 | Hopscotch | yes, that's german e-plus |
| 12:56.49 | _dion_ | hmm. not /me |
| 13:04.18 | alphaone | jonwil: Yeah, that's the carrier I am at |
| 13:08.45 | jonwil | ok, that means that the AT+ESPN command asks the BP for the SP name |
| 13:09.28 | jonwil | I can see in the logs from the tapisrv that the phone does AT+ESPN? |
| 13:09.47 | jonwil | and gets back +ESPN: "BASE" OK |
| 13:09.53 | jonwil | except that "BASE" is in hex |
| 13:10.55 | jonwil | Hopefully someone with a 2.4 phone that boots and works can do AT+CLAC? on all AT command mux ports and see what the BP says it supports :) |
| 13:10.57 | jonwil | would be good to know that |
| 13:11.25 | jonwil | then we can use the GSM specs, google and whatever else to track down what those commands do |
| 13:11.30 | jonwil | :) |
| 13:17.19 | _dion_ | jonwil: i will do this later.. |
| 13:17.27 | jonwil | ok, sounds great |
| 13:17.30 | jonwil | post the results somewhere |
| 13:17.48 | jonwil | such as put them somewhere and post link to maillist |
| 13:26.53 | _dion_ | ok.. i will make serial console adapter tomorrow.. |
| 13:27.08 | _dion_ | or any idea how to do this without it? |
| 13:27.35 | alphaone | _dion_: yes. I did it without. |
| 13:27.49 | _dion_ | alphaone: how? |
| 13:29.12 | alphaone | We added a file /etc/rc.d/rc2.d/S70usbnet.sh to the rootfs with following content: |
| 13:29.22 | alphaone | #!/bin/sh |
| 13:29.22 | alphaone | echo MotNet > /proc/usbd-switch |
| 13:29.22 | alphaone | sleep 3 |
| 13:29.22 | alphaone | ifconfig usbl0 up 192.168.1.2 netmask 255.255.255.0 mtu 900 |
| 13:29.57 | alphaone | chmod a+x it and you should have usbnet after you boot the kernel. |
| 13:30.29 | alphaone | And /diska is an smb mount |
| 13:32.11 | ao2 | _dion_, if you can find mini-usb connector you can also work around that. For example I'll use a female-female usb adapter connecting it to the usual cable. |
| 13:32.54 | ao2 | or simply build the serial adpter with a female usb port |
| 13:34.20 | alphaone | _dion_: We didn't manage to get a shell on the serial console. The output was always broken somehow. |
| 13:36.03 | alphaone | And then we managed to get usbnet working and it was much nicer since you can copy files from / to the samba share. But if you want to try to get serial console working under 2.4 go ahead. |
| 13:36.21 | jonwil | so is it possible to send AT commands to the BP on 2.4? |
| 13:36.50 | alphaone | jonwil: Yes, that's what opentapi does. |
| 13:37.11 | jonwil | in that case sending AT+CLAC? and listening to the output should be easy enough :) |
| 13:39.28 | alphaone | jonwil: Yeah, but we forgot yesterday and now stefan_schmidt has his phone back :-) |
| 13:39.39 | jonwil | true |
| 13:39.53 | jonwil | all we need then is someone else with a working 2.4 kernel |
| 13:39.54 | jonwil | to try it |
| 13:39.56 | alphaone | But if _dion_ wants to try it I could guide him through the process |
| 13:40.27 | ao2 | hoping that at least one mux device (mux4?) does understand +CLAC |
| 13:40.31 | jonwil | would be good if _dion_ would try it :) |
| 13:40.31 | jonwil | or anyone else :) |
| 13:40.44 | _dion_ | i will try |
| 13:40.47 | jonwil | then we can start building up on wiki infos about the AT commands BP accepts |
| 13:41.41 | alphaone | I would try sending it to mux0 and mux4 |
| 13:41.59 | _dion_ | alphaone: i will try both methods.. this night i tried to flash 2.6 kernel to phone.. |
| 13:42.41 | alphaone | _dion_: okay. Just tell me if you need any help. |
| 13:43.00 | jonwil | any of mux0,1,2,3,4 might understand it |
| 13:43.06 | jonwil | maybe even 5,6,7 too |
| 13:44.06 | _dion_ | alphaone: ok. thanks.. i buy MAX3232 and USB cable like orignal data cable by motorola.. |
| 13:44.18 | alphaone | BTW: For USBNet you have to have a kernel that doesn't switch to UART :-) |
| 13:44.25 | alphaone | okay |
| 13:45.23 | jonwil | so does this mean we can run test or is more hardware required? |
| 13:46.06 | alphaone | _dion_ should have everything required for USBNet |
| 13:46.15 | jonwil | good :) |
| 13:46.43 | _dion_ | jonwil: is it ok to just kill tapisrv and try to chat with /dev/muxX? |
| 13:46.52 | jonwil | I dont know |
| 13:46.53 | jonwil | no clue |
| 13:47.04 | jonwil | alphaone, any idea? |
| 13:47.21 | jonwil | should be able to kill tapisrv and run a modified copy of opentapi |
| 13:47.34 | jonwil | i.e. modify opentapi to send AT+CLAC as well as all the other stuff that is sent |
| 13:47.36 | _dion_ | jonwil: is opentapi in SVN? |
| 13:47.44 | jonwil | yes it is in openezx SVN |
| 13:48.02 | jonwil | http://svn.openezx.org/trunk/src/userspace/opentapi/' |
| 13:48.04 | jonwil | http://svn.openezx.org/trunk/src/userspace/opentapi/ |
| 13:48.09 | _dion_ | ok.. will do svn up now |
| 13:49.12 | *** join/#openezx lkcl (n=lkcl@5ac267a5.bb.sky.com) |
| 13:49.27 | alphaone | _dion_: No, it doesn't work.. |
| 13:49.49 | _dion_ | alphaone: what don't work? |
| 13:50.01 | alphaone | The mux devices don't respond the way they use to |
| 13:50.59 | alphaone | Also you have to start opentapi (or tapisrv for that matter) quite early in the boot process. Otherwise opening them fails |
| 13:51.04 | alphaone | Pretty strange |
| 13:51.24 | alphaone | I changed the tapisrv init script to start opentapi instead |
| 13:51.50 | _dion_ | alphaone: possbile because tapisrv read all banners |
| 13:52.37 | alphaone | yeah |
| 13:57.57 | _dion_ | i replaced tapisrv with opentapi.. will reflash phone now |
| 13:58.06 | jonwil | did you modify opentapi |
| 13:58.10 | jonwil | to send +CLAC? |
| 13:58.53 | _dion_ | no |
| 13:58.55 | _dion_ | should i do this? |
| 13:59.04 | jonwil | I think you need to |
| 13:59.18 | jonwil | unless opentapi provides a way to feed AT commnds directly to /dev/muxn or something |
| 14:08.05 | _dion_ | i will read source now.. |
| 14:08.08 | jonwil | ok |
| 14:09.18 | alphaone | No, it does not yet |
| 14:10.00 | alphaone | You can increment MAX_STATES and add the line in the response var |
| 14:11.44 | jonwil | that will send it to only one mux though, right? |
| 14:13.20 | _dion_ | no.. looks like it will send it to all mux-es |
| 14:14.20 | alphaone | No, only to one.. |
| 14:15.20 | alphaone | Only to mux4 |
| 14:17.49 | _dion_ | so looks like now it just open all mux devices and try to write three commands to each mux. am i right? |
| 14:20.10 | jonwil | no, code only sends to one mux |
| 14:26.35 | _dion_ | so looks like i should just add "AT+CLAC" to response[] and replace state < 3 to state < 4 |
| 14:27.20 | _dion_ | "AT+CLAC?\r\n" |
| 14:29.50 | alphaone | _dion_: svn up |
| 14:30.09 | alphaone | I just noticed that I didn't check in the last version |
| 14:34.03 | alphaone | But basically that's it |
| 14:34.51 | alphaone | With the new version you only have to change MAX_STATES |
| 14:36.53 | _dion_ | yes. changed.. just little question.. Pin check is disabled on my SIM. So should i enable it first? |
| 14:37.13 | alphaone | _dion_: No, all the better |
| 14:37.27 | alphaone | Just don't uncomment the AT+CPIN line |
| 14:37.50 | _dion_ | ok. and don't change MAX_STATES.. so no changes at all? |
| 14:40.56 | _dion_ | alphaone: will flash phone now |
| 14:40.59 | alphaone | MAX_STATES has to be 4 |
| 14:41.13 | alphaone | if you have 4 lines in response |
| 14:41.22 | alphaone | (Which I assume you do) |
| 14:41.35 | _dion_ | ok |
| 14:41.49 | alphaone | But this only sends ATCLAC? to mux4 |
| 14:42.33 | jonwil | which is not what we want to do |
| 14:42.50 | alphaone | jonwil: I'm working on it... |
| 14:42.55 | jonwil | good |
| 14:42.57 | jonwil | :) |
| 14:45.37 | alphaone | :-) |
| 14:45.38 | alphaone | Hehe |
| 14:45.42 | _dion_ | ok.. I'm flashing new CG33 |
| 14:56.21 | jonwil | does it work? |
| 14:56.52 | _dion_ | just finish flashing.. |
| 14:56.57 | jonwil | ok |
| 14:59.42 | _dion_ | jonwil: http://pastebin.ca/281135 |
| 14:59.45 | alphaone | _dion_: svn up |
| 15:00.53 | jonwil | that didnt even write AT+CLAC at all |
| 15:00.53 | alphaone | It should now send AT+CLAC? to all muxN |
| 15:00.53 | jonwil | to anywhere |
| 15:00.53 | alphaone | Hmm |
| 15:01.16 | alphaone | _dion_: But I'm not able to test it... |
| 15:01.58 | _dion_ | alphaone: so i should flash new opentapi? |
| 15:02.16 | _dion_ | from last SVN |
| 15:02.56 | alphaone | I'll look over it once more to make sure it does what it's supposed to do |
| 15:04.12 | _dion_ | hmm. phone wasn't turned off after 2 mins |
| 15:08.38 | alphaone | cool |
| 15:09.11 | _dion_ | i'm going away and will back after 20-30 mins. should i do anything before? |
| 15:09.39 | jonwil | as soon as the 2.6 USB issues are resolved we should have working 2.6 too |
| 15:10.44 | alphaone | jonwil: That's the idea to have a minimal tapi implementation to test 2.6 against |
| 15:10.49 | alphaone | _dion_: |
| 15:11.07 | _dion_ | i'm still here.. |
| 15:11.37 | _dion_ | but should go :) |
| 15:11.51 | alphaone | you could update svn and start flashing |
| 15:12.00 | alphaone | I think this should work now |
| 15:12.07 | _dion_ | ok.. sounds good |
| 15:12.25 | alphaone | :-) |
| 15:12.51 | _dion_ | so just flash without any changes? |
| 15:17.02 | alphaone | yes |
| 15:19.13 | *** join/#openezx leoncamel (n=leoncame@219.142.137.2) |
| 15:37.23 | *** join/#openezx [g2] (n=g2@nslu2-linux/g2) |
| 15:38.43 | jonwil | post results for AT+CLAC to all muxn somewhere' |
| 15:38.50 | jonwil | cya later |
| 15:46.47 | _dion_ | ok.. i'm back.. and phone is flased |
| 15:54.02 | _dion_ | alphaone: http://pastebin.ca/281177 |
| 16:00.56 | alphaone | Hmm |
| 16:01.11 | alphaone | So it doesn't implement it? |
| 16:02.28 | *** join/#openezx _dion_ (n=dionua@xen.inhex.net) |
| 16:23.28 | *** part/#openezx _dion_ (n=dionua@xen.inhex.net) |
| 16:29.35 | *** join/#openezx lkcl_ (n=lkcl@5ac5d83e.bb.sky.com) |
| 16:34.43 | *** join/#openezx _dion_ (n=dionua@xen.inhex.net) |
| 16:52.28 | *** part/#openezx _dion_ (n=dionua@xen.inhex.net) |
| 16:54.43 | *** join/#openezx dion (n=dionua@xen.inhex.net) |
| 18:04.36 | *** join/#openezx nh_home (n=nh@aurora.msi.umn.edu) |
| 18:24.35 | *** join/#openezx vmaster_ (i=vmaster@p549B574A.dip.t-dialin.net) |
| 18:48.23 | ao2 | does anyone know if kernel accepts root device specified using its major and minor number? |
| 18:54.09 | dion | alphaone: do you get my last log? |
| 18:55.22 | alphaone | _dion_ alphaone: http://pastebin.ca/281177 |
| 18:55.22 | alphaone | alphaone Hmm |
| 18:55.22 | alphaone | alphaone So it doesn't implement it? |
| 18:55.22 | alphaone | * _dion_ is now known as dion |
| 18:55.26 | alphaone | That one? |
| 18:55.57 | dion | yes |
| 19:01.06 | alphaone | yeah. Looks like it doesn't accept that command :-( |
| 19:02.01 | dion | but phone don't shutdown after 2 mins |
| 19:14.07 | alphaone | So the init sequence works. But AT+CLAC doesn't |
| 19:14.08 | ao2 | ok, passing root device using major and minor can be done, it is just root=Mm where M and m are major and minor number in hex |
| 19:14.21 | alphaone | You should see ring indication if someone calls you |
| 19:16.02 | dion | in log? |
| 19:17.37 | alphaone | yes |
| 19:17.46 | dion | hmm.. yeah, i see it |
| 19:18.19 | alphaone | opentapi writes everything it reads from the /dev/muxN to the logfile |
| 19:18.41 | dion | Read from /dev/mux0: |
| 19:18.42 | dion | +CLIN: 0 |
| 19:18.42 | dion | RING: 1 |
| 19:18.42 | dion | +CLIP: "+3<PHONE NUMBER>",145 |
| 19:19.21 | alphaone | yup |
| 19:21.23 | dion | i will try to use opentapi with 2.6 kernel later.. (After creating rootfs) |
| 19:22.44 | alphaone | dion: We tried it just one time without luck. But we were really tired at that time :-) |
| 19:24.37 | dion | we should hope that it will work |
| 19:32.36 | ao2 | I was able to mount a rootfs from extranal flash also with 2.4 kernel, opie starts up but do not get input (the whole rootfs is built for 2.6+udev so missing device files can be the cause :)) |
| 19:33.56 | ao2 | more details later on the ml |
| 19:34.17 | alphaone | cool |
| 19:35.43 | ao2 | now if only we could take back the BP from reflash mode to normal mode... |
| 19:36.19 | dion | ao2: you can just flash kernel :) it takes not so long time |
| 19:37.53 | ao2 | but ir requires some braveness :) |
| 19:37.58 | ao2 | s/ir/it/ |
| 19:38.40 | ao2 | s/braveness/bravery/ |
| 19:39.44 | ao2 | mmh, purl does not remember substitutions... |
| 19:42.39 | ao2 | now I think that you can replicate current phone fs on external flash, and reflesh a kernel that boots from it, so the development can have a further speed up |
| 19:43.12 | ao2 | I'll try that too |
| 19:44.35 | ao2 | well, if there are hardcoded fs locations in userland that the transition can be less smooth, but feasible anyway |
| 19:44.51 | ao2 | s/that/then/ |
| 20:37.51 | *** join/#openezx LetoAms (n=paul@tla.xelerance.com) |
| 22:43.23 | *** join/#openezx dion (n=dion@xen.inhex.net) |
| 22:47.20 | *** join/#openezx dion (n=dion@xen.inhex.net) |
| 23:27.19 | alphaone | anyone here? |
| 23:27.40 | alphaone | s/?/ with a phone?/ |
| 23:29.29 | dion | alphaone: /me is here |
| 23:30.13 | alphaone | dion: Anything new yet? |
| 23:30.41 | dion | no.. just back.. will deboostrap etch now |
| 23:30.46 | alphaone | ah |
| 23:32.22 | dion | have you any news? |
| 23:34.09 | alphaone | Another thing we should do is check which program connects to tapisrv and get a filedescriptor -> program matching |
| 23:34.43 | alphaone | That way we might be able to guess what some of the AT commands mean |
| 23:34.49 | alphaone | No, nothing new here |
| 23:35.14 | nh_home | alphaone: anything interesting in the last couple of days? |
| 23:35.32 | dion | my first strace log was was with process list |
| 23:35.59 | alphaone | dion: Where can I get it? |
| 23:36.15 | dion | hmm. give me 5 mins |
| 23:36.34 | alphaone | nh_home: We have a tapisrv replacement that can keep the phone alive on 2.4 |
| 23:36.42 | alphaone | dion: k, thanks |
| 23:36.57 | nh_home | alphaone: do we know what tapisrv is doing that keeps the phone alive? |
| 23:37.15 | alphaone | nh_home: http://people.openezx.org/alphaone/logs/opentapi.log |
| 23:37.27 | nh_home | thanks |
| 23:37.35 | alphaone | It's probably AT+USBSTAT=255,1 |
| 23:37.44 | alphaone | That command |
| 23:38.30 | nh_home | why do you think they did that? |
| 23:38.40 | alphaone | Though I just send everything that tapisrv sends which is also AT+CBC and AT+CBSM? |
| 23:39.12 | alphaone | nh_home: I don't have a clue. |
| 23:39.22 | dion | alphaone: http://inhex.net/dion/ps.gz http://inhex.net/dion/tapisrv_log.gz |
| 23:39.43 | nh_home | alphaone: motorola confuses me. |
| 23:39.43 | alphaone | tapisrv or opentapi have to be started pretty early to even be able to open the muxN devices |
| 23:44.03 | *** join/#openezx purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 23:44.16 | dion | jonwil: hi |
| 23:44.50 | alphaone | hi |
| 23:45.11 | alphaone | dion: I don't.. |
| 23:45.45 | dion | wiki page is empty :(. /me will wait for stefan_schmidt |
| 23:46.47 | dion | hmm. my irc client should save history |
| 23:46.59 | alphaone | accept(12, {sa_family=AF_UNIX, path="^UAÐý A^Tþÿ¿(ë^AAxA "}, [2]) = 13 |
| 23:47.14 | alphaone | Can I somehow get the pid of the originating program? |
| 23:47.36 | alphaone | I guess not |
| 23:48.44 | dion | hmm.. we should compile netstat and try again |
| 23:49.17 | dion | i will do this later because i should charge my batterry |
| 23:54.24 | jonwil | were you able to get opentapi to send AT+CLAC to the BP? |
| 23:55.02 | alphaone | cool |
| 23:55.34 | alphaone | jonwil: Yes |
| 23:55.37 | alphaone | _dion_ ok.. i'm back.. and phone is flased |
| 23:55.37 | alphaone | _dion_ alphaone: http://pastebin.ca/281177 |
| 23:55.37 | alphaone | alphaone Hmm |
| 23:56.25 | jonwil | it looks like all muxes are returning CME ERROR 4 for AT+CLAC |
| 23:57.33 | jonwil | what is CME ERROR 4? |
| 23:58.40 | alphaone | operation not supportet |
| 23:58.48 | alphaone | s/et/ed/ |
| 23:59.22 | jonwil | so it would sound like none of the BP muxes support AT+CLAC |
| 23:59.57 | jonwil | which actually makes sense since the only thing that should be talking to the BP is the AP |