| *** bzo has quit IRC | 00:11 | |
| *** co_ajak_cewe_ngi has joined #xdandroid | 01:41 | |
| *** co_ajak_cewe_ngi has left #xdandroid | 01:41 | |
| *** Xawen has quit IRC | 03:13 | |
| *** Xawen has joined #xdandroid | 03:14 | |
| *** hardwalker has quit IRC | 05:06 | |
| *** arrrghhh has quit IRC | 05:34 | |
| *** arrrghhh has joined #xdandroid | 05:36 | |
| *** ke- has joined #xdandroid | 06:13 | |
| *** drakaz has joined #xdandroid | 08:14 | |
| *** helicopter88 has joined #xdandroid | 09:06 | |
| *** fishhead2567 has joined #xdandroid | 09:11 | |
| *** NeoMatrixJR_away is now known as NeoMatrixJR | 10:25 | |
| *** helicopter88 is now known as helicAWAY | 11:41 | |
| *** xdandroid has joined #xdandroid | 12:29 | |
| *** xdandroid` has joined #xdandroid | 12:33 | |
| *** xdandroid` is now known as xdandroid | 12:33 | |
| *** ke- has quit IRC | 12:40 | |
| *** xdandroid has joined #xdandroid | 23:09 | |
| hyc | http://pastebin.com/XFR0d9m7 | 23:09 |
|---|---|---|
| hyc | see the bottom of the paste | 23:09 |
| hyc | the command spits out a bunch of stuff on stdout | 23:09 |
| hyc | but also sends it to the kernel | 23:09 |
| d3tul3 | oh crap it just came up | 23:10 |
| d3tul3 | but i have no idea what i did | 23:10 |
| hyc | heh heh | 23:10 |
| hyc | let's see your dmesg now | 23:10 |
| d3tul3 | yeah looks the same as yours | 23:10 |
| d3tul3 | i see wl0.1 | 23:10 |
| hyc | yep | 23:11 |
| hyc | so this was using the original firmware right? | 23:11 |
| hyc | 4.217.74.0 | 23:11 |
| d3tul3 | er not sure i was just trying the apsta firmware from your link | 23:12 |
| d3tul3 | but that didn't load properly | 23:12 |
| hyc | ok | 23:12 |
| hyc | weird | 23:12 |
| arrrghhh | hyc, my evil rootfs was the reason for the firmware copy :meep: | 23:12 |
| hyc | arrrghhh: you and your stupid patches to unconditionally copy files around..... :P | 23:13 |
| arrrghhh | lol | 23:13 |
| arrrghhh | i was asking about how to do that w/o copying every time | 23:13 |
| d3tul3 | how did you get the tether notification to go away | 23:13 |
| arrrghhh | couldn't think of a reasonable way to do it | 23:13 |
| hyc | tether notification to go away? | 23:13 |
| hyc | up in the status bar? | 23:13 |
| d3tul3 | yeah | 23:14 |
| hyc | just ifconfig down / rmmod | 23:14 |
| d3tul3 | that worked | 23:14 |
| hyc | yeah, no problems | 23:14 |
| d3tul3 | ok now to reproduce | 23:15 |
| hyc | heh heh | 23:15 |
| hyc | and find out what ioctls they're using | 23:15 |
| hyc | oh, we have strace, good | 23:16 |
| d3tul3 | is there any source for this ultra binary? | 23:18 |
| hyc | no idea | 23:18 |
| hyc | ask this guy http://forum.xda-developers.com/showthread.php?t=697441 | 23:19 |
| d3tul3 | https://gist.github.com/824341 | 23:22 |
| hyc | excellent | 23:22 |
| d3tul3 | back to not working http://pastebin.com/DngN51Et | 23:27 |
| hyc | non-BDC packet received | 23:28 |
| hyc | I saw a lot of messages like that on the old firmware | 23:28 |
| hyc | could try it again | 23:29 |
| hyc | or try rmmod and try the new firmware again | 23:30 |
| d3tul3 | well it worked | 23:32 |
| d3tul3 | afer | 23:32 |
| hyc | ? | 23:32 |
| d3tul3 | after stuffing the _apsta firmware in BCM4325_apsta.bin | 23:32 |
| hyc | ok | 23:32 |
| hyc | so I guess we really need the new firmware | 23:32 |
| d3tul3 | at least the new apsta firmware | 23:33 |
| hyc | might as well have a matching set. assuming the new one doesn't break your regular wifi | 23:34 |
| *** rpierce99 has quit IRC | 23:36 | |
| d3tul3 | so those ioctls on 235 and 241 all we need? | 23:36 |
| hyc | probably | 23:36 |
| hyc | mabye also 163 | 23:38 |
| hyc | i.e, I wonder if 163 is what tells the android listener that a tetherable interface appeared. | 23:39 |
| hyc | I don't know... | 23:39 |
| hyc | or maybe it's just watching for wl0.1 to be created, that's also likely | 23:39 |
| d3tul3 | so what do you think patch softap in userland? | 23:40 |
| hyc | mmm. seems kinda gross | 23:41 |
| d3tul3 | or whatever android controller that is currently crappign out | 23:41 |
| hyc | was thinking it might be nicer to patch the bcm driver to handle the ioctls android is using | 23:42 |
| hyc | dunno | 23:42 |
| hyc | strange | 23:45 |
| d3tul3 | i mean that SIOCSIWPRIV ioctl is clearly in place | 23:45 |
| hyc | SoftapController.cpp seems to already have code to generate this ascii command | 23:46 |
| *** BulitPruf has joined #xdandroid | 23:47 | |
| hyc | maybe the problem is just a sequencing thing | 23:50 |
| hyc | looks like the userland tries to start the interface before configuring it | 23:50 |
| hyc | and ultra_bcm configures it before starting it | 23:50 |
| hyc | or mebbe driver start is a no-op | 23:52 |
| hyc | could try just commenting out the SoftapControler::startDriver() stuff and see if skipping it is enough to make it work | 23:54 |
| d3tul3 | i think it's pooping out somewhere in setsoftap | 23:54 |
| hyc | nope, it's not getting that far | 23:54 |
| hyc | the caller does softap stop / softap fwreload / softap start | 23:55 |
| hyc | it fails at softap start | 23:55 |
| hyc | the next thing after that is softap set | 23:55 |
| hyc | and then finally softap startap | 23:55 |
| hyc | so it never gets to softap set | 23:55 |
| d3tul3 | i see softap start returning | 23:56 |
| hyc | I see it returning 1 | 23:56 |
| hyc | when it should return 0 | 23:56 |
| d3tul3 | sure but that wouldn't trigger an exception right | 23:56 |
| hyc | line 9 of my paste | 23:56 |
| hyc | dunno | 23:57 |
| hyc | I guess it's possible, we need some more debugging to know for sure | 23:57 |
| hyc | but there's never any log from Softap set | 23:58 |
| hyc | so I believe it never got there | 23:58 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!