| 00:47.54 | *** join/#openezx gmzhuo (n=Administ@58.59.133.221) |
| 00:55.26 | gmzhuo | Hi WyrM |
| 03:15.13 | *** join/#openezx LoneStar (n=xinzhen@218.80.193.59) |
| 03:57.30 | *** join/#openezx gmzhuo (n=Administ@58.59.133.221) |
| 06:46.05 | *** join/#openezx gmzhuo (n=Administ@58.59.133.221) |
| 07:18.40 | *** join/#openezx muromec (n=muromec@212-178-23-184.broadband.tenet.odessa.ua) |
| 07:29.46 | *** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 07:55.57 | *** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1) |
| 08:40.16 | *** join/#openezx l403 (n=l@85.132.159.239) |
| 09:12.34 | *** join/#openezx Sleep_Walker (n=Sleep@nat/novell/x-fugfbvrtbdunltzg) |
| 09:45.11 | *** join/#openezx florian_kc (n=fuchs@port-217-146-132-69.static.qsc.de) |
| 09:53.14 | *** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 10:05.43 | stefan_n_ | ao2 , i was not able to get the camera frames faster :( 15 fps seems to be the maximum |
| 10:06.24 | ao2 | that is at 640x480? |
| 10:07.10 | stefan_n_ | if i manually set the divisor that gets calculated to 0 and use a master clock of 52 i get 30 hz but the images are bad |
| 10:07.19 | stefan_n_ | yes that is 640 by 480 |
| 10:07.37 | stefan_n_ | also the same with 640 by 512 |
| 10:08.26 | ao2 | stefan_n_, you could try asking to linux-media or linux-arm-kernel, there are people certainly more helpful than me |
| 10:08.26 | stefan_n_ | ok thanks |
| 10:09.39 | ao2 | in particular Guennadi Liakhovetski and Robert Jarzmik worked on pxa_camera and mt9m111 |
| 10:10.26 | ao2 | they both are on linux-media, but it would be better to put them in CC |
| 10:13.18 | *** join/#openezx marex (n=marex@thor.hackndev.com) |
| 10:24.18 | *** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 10:48.37 | ao2 | muromec, no more Oops, but still not passing handshake: http://pastebin.com/d5a6185cc |
| 10:48.46 | ao2 | problems opening /dev/mux7 |
| 10:49.06 | ao2 | TS07.10: MUX: Can't connected channel 7! |
| 10:49.42 | muromec | hm |
| 10:49.52 | muromec | ao2, maby i missed numbers |
| 10:50.05 | ao2 | the test program is always the same, it quits on the firs open error |
| 10:50.49 | muromec | ao2, which muxes did you open in old driver? |
| 10:50.56 | ao2 | muromec, you mean missed number in the line list for gen1? |
| 10:51.02 | muromec | yep |
| 10:51.04 | ao2 | muromec, from 0 to 7 IIRC |
| 10:51.24 | ao2 | ezxd does so |
| 10:52.33 | *** part/#openezx stefan_n_ (n=swelker@neuro16.informatik.uni-freiburg.de) |
| 10:54.57 | muromec | strange |
| 10:55.05 | ao2 | does my dumb test program work on gen2? |
| 10:57.36 | muromec | checking now |
| 10:59.09 | muromec | yep, working |
| 10:59.22 | muromec | opened muxes up to 10 |
| 11:02.38 | muromec | hm |
| 11:02.53 | muromec | in old driver dlci 7 was mux6 |
| 11:03.06 | muromec | mux6 is first mux to control gprs |
| 11:03.33 | muromec | ao2, can you open mux8 now? |
| 11:04.03 | ao2 | muromec, I'll try to skip 7 and open 8 |
| 11:04.19 | muromec | i think it should not open too |
| 11:04.43 | muromec | ao2, did bp paniced after this attempt? |
| 11:05.29 | ao2 | muromec, the phone powered off, but I didn't see any panic message |
| 11:06.02 | muromec | huh |
| 11:08.09 | muromec | ao2, bp panic message is "bp lowered wdi line" |
| 11:08.25 | ao2 | I didn't see it |
| 11:08.47 | muromec | if you dont see it, then you passed |
| 11:11.05 | ao2 | but the phone shut off, let me take more attention and see if the message is there |
| 11:12.03 | muromec | its strange, as we just log bp panics |
| 11:12.38 | muromec | dont know, maby there is hardware watchdog on gen1 |
| 11:13.01 | ao2 | I don't remember if once passed handshake the mux lines can be closed safely. Closing then can still make the phone shut down? |
| 11:13.51 | muromec | yep, you can close safely after handshake |
| 11:14.12 | ao2 | yes, there is some hardware watchdog, that's why we had ezxd in rcS IIRC |
| 11:14.25 | muromec | damm |
| 11:14.58 | ao2 | but I don't know/remember all the details about BP |
| 11:15.15 | ao2 | let me check the wiki |
| 11:15.15 | muromec | me too |
| 11:22.45 | ao2 | mux8 can't be opened either as you guessed, and still poweroff and no panic message. I'll try to run "mux" earlier retrying if it fails witn ENODEV on ttyIPC0, ok? |
| 11:24.00 | *** join/#openezx gDD|mobilE (n=gDD@unaffiliated/gdd) |
| 11:24.14 | muromec | ENODEV is ok |
| 11:29.42 | muromec | as 7 and 8 are gprs controlling lines, they cant require some actions on bp side to initialize |
| 11:29.57 | muromec | like some AT command |
| 11:30.03 | ao2 | muromec, what is the name of the mux driver in 2.4? |
| 11:30.14 | muromec | lol |
| 11:30.22 | muromec | its not in kernel tree |
| 11:30.28 | muromec | its named mux_cli |
| 11:30.42 | ao2 | ah, that's why I can't find it in http://git.openezx.org/motorola-2.4.git |
| 11:30.43 | ao2 | <PROTECTED> |
| 11:30.50 | muromec | well, maby its in gen1 kernel tree |
| 11:30.51 | ao2 | thanks |
| 11:31.39 | ao2 | when you developed the new one, did you based on the latest one for A1200? Did you checked the difference with previous versions? |
| 11:32.32 | ao2 | I am getting it from opensource.motorola.com now |
| 11:33.18 | ao2 | ah, moto released a910i sources as a delta against a910 |
| 11:52.08 | *** join/#openezx marex (n=marex@thor.hackndev.com) |
| 11:56.29 | muromec | ao2, current driver is based on old :) |
| 11:57.01 | muromec | dont know, what can be wrong except line numbers |
| 12:08.51 | *** join/#openezx gmzhuo (n=Administ@58.59.133.221) |
| 12:23.59 | ao2 | muromec, http://people.openezx.org/ao2/tmp/mux_cli_from_a780_to_33R.diff and http://people.openezx.org/ao2/tmp/mux_cli_from_33R_to_a910i.diff |
| 12:25.13 | ao2 | but I also think it is about dlci lines, will try with gen2 ones |
| 12:27.13 | muromec | ao2, try to open them all |
| 12:27.21 | ao2 | ok |
| 12:27.34 | ao2 | fom 1 to ? |
| 12:28.28 | ao2 | 16? |
| 12:30.04 | muromec | yep, 1 - 6 |
| 12:30.07 | muromec | *16 |
| 12:34.10 | ao2 | Scheduling while atomic... |
| 12:36.54 | ao2 | muromec, http://pastebin.com/d4f1f73e0 |
| 12:37.22 | ao2 | the fact that also mux5 fails could be due to the watchdog, I'll put mux in rcS again |
| 12:49.11 | *** join/#openezx sabrod (n=sabrod@mic92-8-82-234-143-184.fbx.proxad.net) |
| 13:01.35 | *** join/#openezx sabrod (n=sabrod@mic92-8-82-234-143-184.fbx.proxad.net) |
| 13:12.12 | *** join/#openezx sabrod1 (n=sabrod@mic92-8-82-234-143-184.fbx.proxad.net) |
| 13:37.22 | ao2 | muromec, If i run "mux" in rcS I can indeed open also mux7. Not mux8 and upper, tho. Here's the messages: http://pastebin.com/d369c72ef Still no messages about bp status, and the phone still powers off... |
| 13:47.37 | *** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 16:25.37 | *** join/#openezx marex (n=marex@thor.hackndev.com) |
| 16:55.48 | *** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 17:00.56 | muromec | ao2, interesting |
| 17:01.25 | muromec | so dlci7 can be opened, it just timing error |
| 17:02.07 | muromec | ao2, from where did you launched last time? |
| 17:02.24 | muromec | and how did you opened ipc |
| 17:22.44 | ao2 | muromec, this mux.c http://pastebin.com/dc488c56 from rcS |
| 17:33.33 | muromec | ao2, and when it stopps? |
| 17:34.36 | ao2 | it can't open mux lines >= 8 ,and after a while the phone powers off |
| 17:34.45 | muromec | hmm |
| 17:35.42 | ao2 | do you think I should give a look to ezx-bp or to the mux driver itself (harder for me)? |
| 17:36.02 | muromec | i dont think so |
| 17:38.48 | muromec | both drivers are versy simple, just encoding packets and transmitting data |
| 17:38.51 | muromec | *very |
| 17:39.10 | muromec | dont understand where is the problem |
| 17:40.13 | WyrM | Hi |
| 17:40.17 | muromec | WyrM, |
| 17:40.21 | WyrM | Try to delay the handshake :) |
| 17:40.21 | ao2 | hi WyrM |
| 17:41.25 | muromec | WyrM, this can brake usb :( |
| 17:41.25 | WyrM | With the handshake earlier, and mux.c running later may be too late for the watchdog. |
| 17:41.26 | WyrM | muromec: not really. |
| 17:41.35 | muromec | hm |
| 17:41.46 | WyrM | can't you leave the phone on gen-blob screen for any amount of time? |
| 17:42.08 | WyrM | as long as you don't try to use usb before the handshake, this shouldn't be a problem. |
| 17:42.16 | muromec | the trick is that usb initialised on bp side after handshake |
| 17:42.30 | WyrM | just sync the timing between the handshake and the usb access, and it should work., |
| 17:42.33 | muromec | yep, ohci initialisation should be delayd too |
| 17:42.45 | muromec | dont know, how to do this |
| 17:43.12 | WyrM | I don't think that we need to do anything. |
| 17:43.22 | muromec | wrong |
| 17:43.28 | WyrM | as bp will only be seen after the handshake. |
| 17:44.10 | muromec | keep in mind, that bp is stupid |
| 17:44.19 | WyrM | yeah, I know :) |
| 17:44.26 | muromec | it looks connected from ap side, when its really not |
| 17:44.38 | WyrM | but as far as I remember, before the handshake you can't see a thing on the usb., |
| 17:45.09 | ao2 | well, I see "Motorola PCS" with lsusb |
| 17:45.22 | muromec | yep |
| 17:45.23 | WyrM | ao2: with ezx-bp.c disabled? |
| 17:45.34 | muromec | if not, you`l not be able to open muxes :) |
| 17:45.45 | ao2 | WyrM, no, enabled |
| 17:46.19 | WyrM | a quick hack would be to remove the call to handshake() from ezx-bp.c's probe |
| 17:46.47 | muromec | and call it from ipc open() ? |
| 17:46.55 | WyrM | and call it on whatever code is called on /dev/ttyipc open() |
| 17:46.58 | WyrM | yes. |
| 17:47.15 | ao2 | ok, hanging on gen-blob too much prevents even /dev/ttyIPC0 to be opened, as I expected |
| 17:47.27 | ao2 | ok, let's try this |
| 17:47.35 | muromec | WyrM, btw, i tried to merge mux_usb.c and ezx-bp.c and turn ipc into usb-serial subdriver |
| 17:47.53 | WyrM | muromec: yes, this is a good thing :) |
| 17:47.56 | muromec | that was working, but after upstream update it stopped :( |
| 17:47.57 | WyrM | I think. :) |
| 17:48.17 | WyrM | well, the timing is relative to step 2 of handshake, IIRC |
| 17:48.45 | WyrM | so it may be needed to adjust the delays between usb detection and actual usb probe. |
| 17:48.59 | WyrM | iirc there are some fields on ohci_info that can be abused to do this. |
| 17:49.26 | WyrM | they are not really meant for this, but I have used them in the past, to _not_ use old_scheme_first |
| 17:49.52 | muromec | heh |
| 17:50.06 | WyrM | Also, gmzhuo did a lot of tests compiling ezx-bp.c and the mux drivers as modules. |
| 17:50.06 | muromec | old_scheme_first _removes_ delays |
| 17:50.19 | WyrM | and this also worked to solve the timing issues |
| 17:50.27 | muromec | huh |
| 17:50.51 | WyrM | muromec: well, in fact, BP can be probed by _both_ schemes. |
| 17:50.52 | muromec | well, compile ohci-hcd and mux as modules can solve this :) |
| 17:51.01 | WyrM | I mean. |
| 17:51.09 | muromec | hmm |
| 17:51.10 | WyrM | probed _first_ by both schemes. |
| 17:51.30 | WyrM | not after the driver gives up and try the other scheme. |
| 17:51.47 | muromec | never saw this |
| 17:52.12 | muromec | iirc i always had -62 on new scheme, than fallback |
| 17:52.26 | WyrM | the trick is to get the correct delay relative to enumeration start. |
| 17:52.37 | muromec | yep |
| 17:53.38 | WyrM | I did some tests with this. |
| 17:54.01 | WyrM | and I remember getting a780 to pass usb probe on the first try |
| 17:54.18 | WyrM | with a big delay on some ohci_info field. |
| 17:54.32 | muromec | heh. as you have both gen1 and gen2, try do fix this propertly :) |
| 17:54.38 | muromec | *to |
| 17:54.41 | WyrM | muromec: heh |
| 17:55.02 | WyrM | I told you that your work for a1200 would break gen1. :) |
| 17:55.02 | WyrM | This really was expected. |
| 17:55.17 | muromec | yep, i remember |
| 17:55.22 | WyrM | but you work is great anyway, that it was worth to allow the gen1 breakage. |
| 17:56.29 | muromec | heh, thanks |
| 17:56.37 | WyrM | As for fixing it myself, im very low on time to do this now. |
| 17:56.55 | WyrM | but I can send you a gen1 phone if you have the time to do it. |
| 17:57.18 | muromec | it would be difficult for me :( |
| 17:59.07 | WyrM | Its your call. :) |
| 17:59.28 | WyrM | If you want to work on gen1 too, just send me an email and I send you a device |
| 17:59.41 | WyrM | or if you can get one locally, I can send the $ too. |
| 18:00.35 | muromec | WyrM, we have crazy customs laws |
| 18:01.07 | muromec | i`l pay 20% of device price for this |
| 18:01.43 | WyrM | just 20%? |
| 18:01.49 | WyrM | I pay 88% :) |
| 18:01.53 | muromec | o_O |
| 18:01.55 | WyrM | 60% for imports |
| 18:02.11 | WyrM | plus 18% of state tax, caleed ICMS |
| 18:02.17 | muromec | wow |
| 18:02.24 | WyrM | And these are over the product price + the shipping cost. |
| 18:02.27 | WyrM | LOL |
| 18:02.41 | muromec | this world is crazy |
| 18:03.03 | WyrM | I bought a nice USB oscilloscope |
| 18:03.09 | WyrM | Costs US$249,00 |
| 18:03.21 | WyrM | shipping was $60 |
| 18:03.22 | tmzt | maybe silven can get one in pr? how odes that work? |
| 18:03.38 | WyrM | Taxes were R$ 630,00 |
| 18:04.09 | muromec | WyrM, awesome :) |
| 18:04.19 | WyrM | R$ 630 = US$350,00 |
| 18:04.32 | WyrM | so, total taxes were more than the product cost |
| 18:28.40 | *** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 18:41.52 | *** join/#openezx florian (n=fuchs@f049034046.adsl.alicedsl.de) |
| 19:37.56 | *** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 20:12.05 | *** join/#openezx tombhadAC (n=tombhad@pD95DC931.dip.t-dialin.net) |
| 20:39.51 | *** join/#openezx tombhadAC (n=tombhad@pD95DD199.dip.t-dialin.net) |
| 20:53.40 | *** join/#openezx Simon- (i=simon@proxima.lp0.eu) |
| 21:29.47 | *** join/#openezx tombhadAC (n=tombhad@pD95DF52E.dip.t-dialin.net) |
| 21:45.41 | Simon- | I've got an usb0 device but I'm not receiving any packets on it :/ |
| 21:46.38 | WyrM | send the kernel to gen-blob(black screen) instead of the original bootloader (blue or red screen) |
| 21:47.24 | Simon- | oh, sorry, this is just the stock firmware |
| 21:47.41 | Simon- | so far I can only telnet to it from windows |
| 21:47.41 | WyrM | hum. |
| 21:47.55 | WyrM | then your ip configuration is wrong. |
| 21:48.11 | Simon- | I have the right IP |
| 21:48.24 | Simon- | it's not responding to DHCP requests either, which windows appears to be able to do |