| 00:00.08 | cr2 | Speedy2: hx4700 is called HTC Roadster |
| 00:00.48 | psokolovsky | Speedy2, we boot with haret (I boot my hx4700 such too). that appears to be rather related |
| 00:10.04 | *** part/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 00:18.00 | goxboxlive | Ok, i tried suspendeing the universal, and the lcd backlight where turned off. So that is good. But, still the lcs floates after resume. It's just minor, but i guess there is still some bugz. |
| 00:21.51 | cr2 | goxboxlive: i can revert one more code change. will you test it ? |
| 00:23.05 | goxboxlive | yes |
| 00:24.42 | goxboxlive | r2: Do you mean revert to the ones we made in the .12? While you are in there chcking suspend/resume is it much work to make the power button work? So it will suspend the machine to? It's only reume it now. |
| 00:26.14 | cr2 | then it will be necessary to revert the gpio-keys too. |
| 00:26.40 | goxboxlive | i c |
| 00:27.03 | psokolovsky | goxboxlive, suspending is a matter of user space, not kernel |
| 00:27.27 | psokolovsky | goxboxlive, follow comon buttonmap proposal and all your issues with keys will be solved ;-) |
| 00:27.49 | cr2 | goxboxlive: done. |
| 00:27.54 | goxboxlive | While you are "reverting". Maybe we should take a look at the power module's plug'N'Pplay to. It also worked in the .12 |
| 00:28.17 | cr2 | :) too much for today . |
| 00:28.29 | cr2 | check the LCD. |
| 00:28.52 | goxboxlive | i will :-) |
| 00:29.15 | cr2 | psokolovsky: if i will split gpio-key (POWER), asic3-keys and the keyboard, should i get 3 input devices ? |
| 00:29.16 | goxboxlive | it's avaialable now |
| 00:29.29 | psokolovsky | cr2, yes, you should |
| 00:29.47 | cr2 | well, that will break a lot of stuff |
| 00:30.14 | cr2 | what is the right way to deal with 10 input devices ? |
| 00:30.47 | psokolovsky | cr2, userspace deals with them transaprently. I was unable to prove otherwise. |
| 00:31.27 | cr2 | i don't understand how this transparency will be achieved. |
| 00:32.32 | cr2 | i get /dev/input/event0,1,2,3,4,... in unknown sequence, how do i put all these events together ? |
| 00:33.33 | goxboxlive | cr2: The lcd screen is still the same, just a little floating crystals over the lethers in the shell. |
| 00:34.52 | cr2 | hmm. don't really know. the pxafb may switch the CKEN not at the same time as wince. then there is a missing .lccr4 problem. |
| 00:35.04 | goxboxlive | ok |
| 00:35.50 | cr2 | pxafb is not really tuned for pxa27x, the hw overlays are missing. |
| 00:36.18 | cr2 | ok, i need some sleep :) |
| 00:36.21 | cr2 | good night. |
| 00:36.26 | goxboxlive | g, night |
| 04:04.46 | psokolovsky | Goddamn! |
| 04:04.49 | psokolovsky | $E@#$! |
| 04:04.59 | psokolovsky | I've seen the light! |
| 04:05.17 | psokolovsky | meaning, I hear fschking sound! |
| 04:05.41 | psokolovsky | goxboxlive, you rules! |
| 04:06.39 | psokolovsky | and just another confirmation that no bug will survive! ;-E |
| 04:07.00 | psokolovsky | let's bisect now though... |
| 05:00.29 | zsircusr | uh |
| 05:00.33 | zsircusr | on which device ? |
| 05:02.50 | psokolovsky | hx4700, htcuniversal, etc |
| 05:03.17 | scatterp | so you can hear sound on the htcuniversal ? |
| 05:03.43 | psokolovsky | on hx4700 |
| 05:04.01 | scatterp | oh thats a good step forward for the hx4700 |
| 05:04.08 | scatterp | do you think it will effect the htc universal |
| 05:04.11 | scatterp | sound problems ? |
| 05:04.17 | psokolovsky | of course it will |
| 05:04.22 | psokolovsky | we have common problem |
| 05:04.25 | scatterp | oh thats great |
| 05:05.02 | scatterp | i have a htc universal here if you need me to test anything ... it seems the others are not around |
| 05:05.06 | scatterp | im a novice thru |
| 05:05.16 | scatterp | if not ill get some sleep and look forward to mp3s :) |
| 05:06.37 | psokolovsky | it's not that easy and yes, right now there's nothing to test yet |
| 05:06.59 | psokolovsky | but at least it was reproduced - both working and not working |
| 05:07.01 | scatterp | ok well thanks for the great news |
| 06:01.26 | Kevin2 | Hi |
| 06:05.23 | BabelOued | hi |
| 06:52.42 | *** join/#htc-linux rob_w (n=rob_w@p549B97E5.dip0.t-ipconnect.de) |
| 06:53.48 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 07:10.07 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 07:43.52 | goxboxlive | psokolovsky: Wake up |
| 07:44.13 | goxboxlive | YOu have been asleep for over three hours now |
| 07:44.38 | goxboxlive | Is the sound working on the .18 kernel or is it the .12 ? |
| 07:52.01 | *** join/#htc-linux TripleDES (n=sergio@84.77.155.74) |
| 08:00.52 | *** join/#htc-linux dullard (n=jim@62.245.36.30) |
| 08:06.04 | *** join/#htc-linux Fossi (n=cmewes@195.177.48.115) |
| 08:26.04 | rob_w | ~seen lpotter |
| 08:26.37 | apt | lpotter <n=ljp@c210-49-127-23.rochd1.qld.optusnet.com.au> was last seen on IRC in channel #opie, 26d 8h 2m 16s ago, saying: 'i believe it can'. |
| 08:34.11 | *** join/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com) |
| 08:56.18 | *** join/#htc-linux pH5 (n=ph5@p5485FD14.dip.t-dialin.net) |
| 09:03.43 | cr2 | pH5: hi |
| 09:05.01 | cr2 | static void snd_pxa2xx_i2sound_i2slink_on(void) |
| 09:05.25 | cr2 | <PROTECTED> |
| 09:05.29 | cr2 | <PROTECTED> |
| 09:05.33 | cr2 | <PROTECTED> |
| 09:05.37 | cr2 | <PROTECTED> |
| 09:05.41 | cr2 | <PROTECTED> |
| 09:05.45 | cr2 | <PROTECTED> |
| 09:05.46 | pH5 | hey cr2 |
| 09:05.49 | cr2 | <PROTECTED> |
| 09:05.53 | cr2 | <PROTECTED> |
| 09:05.53 | cr2 | <PROTECTED> |
| 09:05.53 | cr2 | <PROTECTED> |
| 09:05.54 | cr2 | <PROTECTED> |
| 09:05.54 | cr2 | this should go away ;) |
| 09:06.07 | cr2 | it is device-specific, and different on the universal and hx4700 ;-) |
| 09:06.43 | pH5 | are there any news on the sound dma issues? |
| 09:06.55 | cr2 | pH5: your PASIC3 CPLD driver cab be reused not only on blueangel, but also on the Alpine :) |
| 09:07.04 | cr2 | pH5: i think this is the bug. |
| 09:07.07 | rob_w | hey cr2 |
| 09:07.15 | cr2 | pxa_gpio_mode(GPIO28_BITCLK_IN_I2S_MD); |
| 09:07.20 | rob_w | i am working now a pxa270 unit but it feels dead slow |
| 09:07.34 | rob_w | can you tell me more about iwmmxt ? |
| 09:07.47 | cr2 | this clock is on different gpio on hx4700 and universal, and probably on apache too, i must have a look. |
| 09:08.12 | cr2 | rob_w: the battery monitor on blueangel is completely wrong code. |
| 09:08.45 | cr2 | imho, if it is slow then iwmmxt is probably not going to help you. |
| 09:11.25 | rob_w | cr2, the cvs code is wrong ... it was wrongly ported version from the .12 stuff |
| 09:11.42 | rob_w | they didnt took the latest code .. but even that one used it in the wrong order |
| 09:11.46 | cr2 | rob_w: i think it monitors the backup battery. |
| 09:12.24 | cr2 | them main battery is served by the OWM in a small ASIC. |
| 09:12.41 | rob_w | the thing it is only one showing as i remmeber .. but comparing the tsc2200 it is a correct thing .. |
| 09:12.43 | cr2 | why didn't they use asic3 is a mystery for me. |
| 09:12.55 | rob_w | what is iwmmxt ? |
| 09:13.00 | cr2 | we need to recheck the gpios once again. |
| 09:13.16 | cr2 | "multimedia" extensions. |
| 09:13.58 | rob_w | well i do have a companion video chip on that pxa270 named 2700G .. will this help for him ? |
| 09:14.37 | rob_w | it feels that gtk is too slow as it should be .. what can i do ? |
| 09:15.37 | pH5 | is there a simple command line util to decode sms pdu somewhere? |
| 09:21.30 | cr2 | rob_w: is this chip supported by linux in accelerated mode ? |
| 09:21.49 | cr2 | pH5: look at gnokii and libgsm |
| 09:22.16 | rob_w | they do have a special fb driver for it .. and i already can run 800x600 .. but i am unsure if this is all i can get |
| 09:24.32 | rob_w | cr2, how do u make sure that you build oe properly for the universal .. aas it also has the pxa270 ? |
| 09:27.44 | cr2 | yes. |
| 09:46.49 | goxboxlive | hi cr2: |
| 09:47.29 | goxboxlive | I readin the log that psokolovsku managed to get sound working on his hx4700. But i dont know if it was the .12 or .18 kernel |
| 10:00.51 | cr2 | goxboxlive: i think it was .12, but i don't know. |
| 10:02.04 | cr2 | hmm, it may be useful to write a linux tool that does "dump pxa27x*" |
| 10:02.44 | cr2 | with /dev/mem it should be very easy to port the code. |
| 10:02.53 | goxboxlive | ok |
| 10:18.17 | psokolovsky | Hi! |
| 10:18.25 | psokolovsky | So, quick update from me: |
| 10:18.49 | psokolovsky | I reproduced working sound on hx4700 with 2.6.12-hh3 |
| 10:19.14 | psokolovsky | Thanks to goxboxlive for *very good* hint |
| 10:19.31 | psokolovsky | And it doesn't work with 2.6.13-hh1 |
| 10:19.43 | psokolovsky | 2 weeks and kernel upgrade delimit those |
| 10:20.09 | goxboxlive | Is the changes in the drivers big between .12 and .13? |
| 10:21.00 | psokolovsky | So, if anyone wants to bisect mainline while applying our patches, please do, because I don't know how much I'll need to rest after this ;-) |
| 10:21.12 | psokolovsky | goxboxlive, as usual 5Mb patch.gz |
| 10:21.22 | goxboxlive | ok |
| 10:24.15 | Speedy2 | What was the hint? |
| 10:24.34 | cr2 | i think we should really port the "dump pxa27x*" functions to linux. |
| 10:30.55 | psokolovsky | oh, btw, I ported xanadux' driver to 2.6.18. same story. |
| 10:31.07 | psokolovsky | so, bisecting is the only way, imho |
| 10:31.48 | psokolovsky | or at least I'm tired of adhoc creeping around |
| 10:35.14 | cr2 | we should check the I2S setup, but dumping raw registers is a too painful way to do it. |
| 11:06.54 | goxboxlive | Ok, now i dont understand a thing |
| 11:07.13 | goxboxlive | Get ready for something ubeliveable |
| 11:07.30 | goxboxlive | cr2 / psokolvsky : Are you redy for something fun :-) |
| 11:08.04 | goxboxlive | SOUND WORKS WHEN I USEING THE HEADPHONES. |
| 11:08.19 | goxboxlive | and yes, with the .18 kernel. |
| 11:08.33 | goxboxlive | I guess the updates of pxa had something to do with it. |
| 11:08.40 | goxboxlive | I'll test the mic now. |
| 11:08.52 | goxboxlive | But, the main speakers doesent work |
| 11:10.41 | goxboxlive | Even the Opie volume applet works now. |
| 11:12.29 | cr2 | hmm. the only difference is the power gpio. |
| 11:12.53 | cr2 | goxboxlive: can you record from the headphone mic ? |
| 11:13.29 | goxboxlive | I tried the mic, but it is buzzy, i can actually hera what i am saying. Sounds like a AM radio band when searching for a radioa channel. You know what i mean |
| 11:13.52 | cr2 | yes, that means it works. |
| 11:13.56 | cr2 | (read_cpuid(CPUID_ID) & 0x3f0) != 0x110 |
| 11:14.08 | cr2 | this can be a bit more verbose ;) |
| 11:14.14 | goxboxlive | ok |
| 11:14.34 | cr2 | then we can remove the gpio setup from htcuniversal_audio.c |
| 11:14.36 | goxboxlive | I want SO to make a call, Is it a fast way to fix the mic? |
| 11:14.46 | cr2 | fix ? |
| 11:14.59 | cr2 | the mic is controlled over i2c. |
| 11:15.01 | goxboxlive | repair, make it work |
| 11:15.30 | cr2 | the headphone is powered up, then the complete control goes over i2c. |
| 11:15.40 | goxboxlive | ok |
| 11:16.41 | goxboxlive | im so happy now. |
| 11:17.07 | *** part/#htc-linux pH5 (n=ph5@p5485FD14.dip.t-dialin.net) |
| 11:17.15 | *** join/#htc-linux pH5 (n=ph5@p5485FD14.dip.t-dialin.net) |
| 11:18.16 | Fossi | goxboxlive: just think about the importance of your achievement in relation to the universe |
| 11:18.27 | Fossi | will calm you down a bit ;) |
| 11:18.49 | Fossi | (still cool though) |
| 11:19.06 | goxboxlive | Fossi: Yes think about that. Lets bot stop with this sound shit. Lets make peace all over the world. |
| 11:19.24 | Fossi | and get high? |
| 11:19.42 | goxboxlive | yes by living |
| 11:19.51 | goxboxlive | halleluhja |
| 11:30.41 | goxboxlive | cr2: I tried the phone, but the sound does wtill come out of the lcd speaker. Not the HP even though it is connected. The mic didnt work, but that's probably because the HP where dead. |
| 11:33.29 | goxboxlive | I had a typo about the mic: I CAN'T hear what i am saying, But the form the input meter i can see that it is catching my voice, so it works even though it is buzzy. |
| 11:53.30 | goxboxlive | cr2 / psokolovsky: But there is another thing with it. It doesnt work right after boot. I have to reload the snd.htcuniversal_audio driver to get it working |
| 12:07.31 | *** join/#htc-linux jeanseb (n=jeanseb@88.164.32.155) |
| 12:11.22 | *** join/#htc-linux RoEn_PC (n=roen@p54A64906.dip.t-dialin.net) |
| 12:37.36 | cr2 | goxboxlive: phone needs to get some AT@AUDIOSET commands to reroute audio. this stuff is universal-specific. |
| 12:37.51 | goxboxlive | ok |
| 12:38.15 | cr2 | if you need to reload the modules, then they do setup something on the cpu. |
| 12:38.45 | cr2 | i2s gpio is one thing, but i have put the same code in htcuniversal_audio.c (seems to be redundant) |
| 12:38.59 | cr2 | we should check other CPU settings done by these modules |
| 12:39.47 | goxboxlive | ok |
| 12:46.38 | cr2 | goxboxlive: the purpose (and the parameters) of the at@talk command are unclear to me. |
| 12:47.24 | goxboxlive | What do you mean cr2? |
| 12:47.35 | cr2 | what are we going to do with an opie phone app ? |
| 12:47.59 | cr2 | when you hangup, the ATH is always sent with the at@talk command. |
| 12:48.26 | goxboxlive | well, i guess, there wouldnt be any. I have no clue how to make it. But we have Openmoko in january, and as far as i know it will be available in the oe.dev |
| 12:48.38 | cr2 | moko is gtk |
| 12:49.02 | goxboxlive | ok |
| 12:49.19 | cr2 | we may try to clone the documented qtopia interface |
| 12:49.21 | goxboxlive | , what about qomunicator |
| 12:49.42 | goxboxlive | Is there a lot that's left to get it working? |
| 12:49.43 | cr2 | and use the gomunicator/openmoko library. |
| 12:50.08 | goxboxlive | I tried to compile it a few days ago, but it failes. lkcl said that it wouldnt work anyway |
| 12:50.28 | cr2 | i will separate 3 problems: the gui, the api, and the lowlevel lib. |
| 12:50.43 | goxboxlive | ok |
| 12:50.48 | cr2 | for the gui maybe the kpnone/pi ? |
| 12:51.04 | cr2 | i don't know how advanced it is. |
| 12:51.11 | goxboxlive | But lpotter where talking about makeing binary for us, but i havent seen him for a while |
| 12:51.24 | cr2 | binary does not count. |
| 12:51.34 | goxboxlive | yes, or the at command app in console. Have you seen it? |
| 12:51.36 | cr2 | we need free software. |
| 12:52.06 | cr2 | if i'd be happy with the binary, i'd use wince instead ;) |
| 12:52.07 | goxboxlive | Well, if he would have made us one i would have been used it until we had a replacement. |
| 12:52.16 | goxboxlive | relsy |
| 12:52.20 | goxboxlive | realy |
| 12:52.42 | cr2 | he needs to put a lot of universal-specific stuff in the lib anyway. |
| 12:52.49 | goxboxlive | ok |
| 12:52.56 | cr2 | and this will be a huge waste of time, if the source code is not released. |
| 12:53.07 | cr2 | under gpl, i mean. |
| 12:53.15 | goxboxlive | ok |
| 12:54.06 | cr2 | i think somebody created a lowlevel lib from gomunicator. |
| 12:54.11 | goxboxlive | Haveyou seen the opie-conslo at application? |
| 12:54.19 | cr2 | no. |
| 12:54.47 | goxboxlive | ok, it's seom predifined at sommands in it |
| 12:55.26 | cr2 | i'd like to know the openmoko phone software architecture, but we are fed with the press releases now. |
| 12:55.47 | goxboxlive | I havce tried to use it to cennect as a data call, but with no luck. |
| 12:55.51 | cr2 | and lkcl can't read his greenphone cd, so we will see what they are doing. |
| 12:56.05 | cr2 | is it a shell script ? |
| 12:56.09 | goxboxlive | does lkcl has a greenphone? |
| 12:56.18 | cr2 | afaik yes. |
| 12:56.29 | cr2 | he has all the phones :) |
| 12:56.34 | goxboxlive | hmm ok :-) |
| 12:57.34 | goxboxlive | Why cant he read his cd? |
| 12:57.34 | cr2 | notebook dvd is flaky |
| 12:57.34 | goxboxlive | ok |
| 12:57.35 | cr2 | and the greenphone/openmoko are so open, that we don't even know what BT chipset they have. |
| 12:57.59 | goxboxlive | :-) |
| 12:58.12 | cr2 | i guess we will have the BT headset working on universal before ;) |
| 12:59.03 | goxboxlive | we will :-) |
| 12:59.18 | cr2 | i think it is possible to send raw binary commands with 'chat' from pppd. |
| 12:59.46 | cr2 | then we can send the BT speed change and other useful BT commands from the shell prompt. |
| 13:00.11 | cr2 | properly escaped. |
| 13:01.55 | cr2 | there is one problem, though. |
| 13:01.57 | goxboxlive | cr2: i treid the greenphone stuff |
| 13:02.39 | goxboxlive | to connect to gprs |
| 13:02.39 | cr2 | i don't know the purpose of the BT irq. |
| 13:02.39 | goxboxlive | some of it wrked, but i couldnt ping anything |
| 13:02.39 | goxboxlive | it made a ppp0 |
| 13:02.39 | cr2 | do you have the kernel from it ? |
| 13:03.03 | goxboxlive | no i just tried the script |
| 13:03.18 | cr2 | they don't release the source, but i'm used to it. with wince nk.exe ;-) |
| 13:03.31 | cr2 | script ? |
| 13:03.39 | goxboxlive | http://doc.trolltech.com/qtopia2.1/html/qtopia-gprs-networking.html |
| 13:04.26 | cr2 | heh. you need the multiplexor on the USB port. |
| 13:04.32 | goxboxlive | ok |
| 13:04.34 | cr2 | look at openezx wiki |
| 13:04.50 | cr2 | the famous mux_cli from a780. |
| 13:05.43 | cr2 | it can be lifted/(joined with a780 version) from openmoko |
| 13:05.49 | cr2 | when it will come out. |
| 13:06.26 | cr2 | and ATH0 is not enough on the universal as i have said. |
| 13:07.05 | cr2 | and pppd will run not on /dev/ttyS0 but on some mux pseudo-tty. |
| 13:07.47 | cr2 | on blueangel it is /dev/ttyS8 for DPRAM. |
| 13:07.48 | cr2 | but i have broke it in .16, need to check the code again. |
| 13:11.06 | cr2 | \ddd Collapse the octal digits (ddd) into a single ASCII character and send that character. |
| 13:11.15 | cr2 | from 'man chat' |
| 13:13.42 | goxboxlive | ok |
| 14:06.15 | pH5 | cr2: isn't the purpose of the bt irq waking up the pxa from sleep? |
| 14:23.59 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 14:27.04 | cr2 | pH5: no. it can be observed when the BTS script is uploaded to the BRF6150 chip. |
| 14:27.41 | cr2 | but it must have some universal purpose, because it is present on all other phones. |
| 14:32.04 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 14:47.53 | psokolovsky | Hi! |
| 14:47.59 | psokolovsky | Any news on sound? |
| 14:50.10 | cr2 | i hope that this code is correct (read_cpuid(CPUID_ID) & 0x3f0) != 0x110 |
| 14:50.17 | cr2 | but it can be more verbose |
| 14:52.12 | psokolovsky | goxboxlive, btw, if you still has that .12 around, can you please check how it does regarding suspend/resume |
| 14:54.19 | Kevin2 | psokolovsky: Did you see my email to hh-kernel? If we can get a dump of the chip registers when in works, I think we can figure out what setting is incorrect. |
| 14:56.09 | psokolovsky | Kevin2, Yes, I saw. No, I don't any longer believe in magic ways to solve it ;-) |
| 14:56.20 | goxboxlive | psokolovsky: Res the log. I have managed to get sound in .18 kernel today |
| 14:56.30 | goxboxlive | i mean read the log |
| 14:56.35 | psokolovsky | Kevin2, it maybe the case that registers are still ok, but code handling them gone mad ;-) |
| 14:56.40 | psokolovsky | Kevin2, bisect! |
| 14:56.53 | psokolovsky | goxboxlive, ok, will do, thanks! |
| 15:00.24 | cr2 | lkcl: i hope you have read that: |
| 15:00.27 | cr2 | <PROTECTED> |
| 15:00.27 | cr2 | <PROTECTED> |
| 15:00.52 | cr2 | mode is cool as hell -- |
| 15:00.52 | cr2 | exclusive to Global Locate. It will download ephemeris data over the |
| 15:00.52 | cr2 | internet. |
| 15:01.21 | cr2 | it's exactly what is needed on the "free software" phone. |
| 15:01.56 | cr2 | some moronic binary-only library that will connect to global locate over internet. |
| 15:02.19 | cr2 | http://lists.openmoko.org/pipermail/community/2006-November/000249.html |
| 15:03.32 | cr2 | as if the people can't predict the satellite orbits since Kepler, and need some stupid patented algorithms. |
| 15:05.49 | psokolovsky | goxboxlive, well, I tested headphones on hx4700 (already had such a magic in my dejavues) - no avail |
| 15:07.19 | goxboxlive | pskolovsky: But did you reload your shn-hx4700_audio module'? |
| 15:07.22 | goxboxlive | snd |
| 15:08.07 | goxboxlive | It doesnt work after boot, but when i am reloading the module the sound work in the headphone. |
| 15:08.39 | psokolovsky | well, I'm tired of this crap. it's completely mindless, and putting mind into it is useles, imho. so I won't be tracking registers and interpreting their values (but someone really should do that!), but will instead grow up to making bisects on the kernel code, plus with applying our patches on top. |
| 15:08.51 | psokolovsky | it may take time, of course. But it's universal, and formal, approach |
| 15:09.09 | psokolovsky | goxboxlive, I did many funky things ;-) |
| 15:09.19 | goxboxlive | :-) i c |
| 15:09.29 | psokolovsky | goxboxlive, anything else I should know before trying to reproduce that? |
| 15:10.31 | goxboxlive | No, not anything else that the snd-machine file must be a module and not compiled in to the kernel image. |
| 15:13.25 | psokolovsky | goxboxlive, will try if I won't get dizzy and sick by booting my hx for the purpose of trying another ak idea ;-) |
| 15:14.13 | psokolovsky | goxboxlive, but thanks of course, I'm very glad we all keep together and communicate on things - I would never get idea to try really old kernel myself... |
| 15:14.34 | psokolovsky | goxboxlive, btw, if you'll be able to test sound over suspend on it, it would be nice (no hurry) |
| 15:15.58 | goxboxlive | I have it running now, i'll try. |
| 15:20.33 | goxboxlive | pskolovsky: yes the audio is resuming |
| 15:21.34 | psokolovsky | goxboxlive, resuming?! |
| 15:21.51 | goxboxlive | yes, it starts again after resume |
| 15:21.57 | psokolovsky | wtf, pxa27x *itself* cannot resume normally in .12! |
| 15:22.03 | *** join/#htc-linux bd2_ (n=ufo@89.113.84.151) |
| 15:22.04 | psokolovsky | hopy crap |
| 15:22.30 | psokolovsky | bd2_: |
| 15:22.30 | psokolovsky | <goxboxlive> I have it running now, i'll try. |
| 15:22.30 | psokolovsky | <goxboxlive> pskolovsky: yes the audio is resuming |
| 15:22.30 | psokolovsky | <psokolovsky> goxboxlive, resuming?! |
| 15:22.30 | psokolovsky | <goxboxlive> yes, it starts again after resume |
| 15:22.31 | psokolovsky | <psokolovsky> wtf, pxa27x *itself* cannot resume normally in .12! |
| 15:23.08 | bd2_ | :-) |
| 15:23.23 | psokolovsky | bd2_, so, some brave soul needs to do bisect on 2.6.13 kernel process. any ideas? |
| 15:23.50 | goxboxlive | Well, it works here. I did cat "mp3 file" > /dev/dsp" form the shell and there where only buzz in the headphones (maybe lack uf codec). I did apm --suspend and after i wake it up the buzz came back in to the headphones again |
| 15:24.45 | cr2 | goxboxlive: do cat file.wav > /dev/dsp |
| 15:24.55 | bd2_ | hi all. psokolovsky, but you know.. that's a little success actually. I'm in similar situation to, (glibc-2.3.x-cvs->2.4 bisectioning.. that's hard to find now) |
| 15:25.01 | cr2 | there are plenty of them in opie |
| 15:25.13 | goxboxlive | cr2: i dont have any working wav files, but i will try with mp321 |
| 15:25.22 | cr2 | ? |
| 15:25.51 | psokolovsky | bd2_, but with kernel, we have git ;-) |
| 15:26.30 | cr2 | goxboxlive: lol, openmoko does not have BT. |
| 15:27.16 | bd2_ | psokolovsky, yes.. but bisecting anything is boooring :-) |
| 15:28.27 | psokolovsky | bd2_, exactly ;-). but with it we at least find the colution, whereas what I've been doing last 3 days was complete chase-in-the-dark |
| 15:31.24 | goxboxlive | The easiest way for me to test suspend/resume is with opie. Is there a easy way to fix the opie suspend? |
| 15:31.48 | psokolovsky | bd2_, so you lazy boy, didn't start with your local git tree idea, and won't tell me how to do that? ;-) |
| 15:34.55 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 15:35.44 | bd2_ | psokolovsky, nope.. I didn't start my git tree :-( though.. I've git'ed -hh tree (now linux/kernel26 directory handled by two RCS :-) |
| 15:36.24 | psokolovsky | bd2_, by git-import-cvs? (or how is it called?) |
| 15:36.49 | bd2_ | nope. git add . git commit -a |
| 15:37.45 | bd2_ | I'm using git just as advanced "working copy" |
| 15:38.02 | psokolovsky | ;-( |
| 15:38.42 | psokolovsky | but we need git, now it's no game. we should be ready and armed for problems like this. |
| 15:39.08 | bd2_ | heh.. yes, yes.. I'm completely with you here |
| 15:39.59 | goxboxlive | I have to try later with the wav file. The ones in opie is just some beeps for a second. I am home with a sick child so i have to take a little care of him to. I'll be back ina cuople of hors. |
| 16:10.07 | Kevin2 | Hrmm. I added a printk to snd_ak4641_hwsync_read_all -- I can confirm that the i2c bus is working and that the ak chip is properly reset. |
| 16:15.23 | Kevin2 | Sound is working on Apache. |
| 16:15.46 | psokolovsky | Kevin2, ??? |
| 16:16.37 | Kevin2 | I followed goxboxlive's instructions - modprobe snd_htcapache_audio ; rmmod snd_htcapache_audio ; rmmod snd_ak4641 ; rmmod snd_pxa2xx_i2sound ; modprobe snd_htcapache_audio |
| 16:16.46 | Kevin2 | It works for me on both headphones and speaker. |
| 16:16.50 | psokolovsky | lol |
| 16:16.57 | psokolovsky | and goddamn |
| 16:17.27 | psokolovsky | something like that was suggested by bd2_ too |
| 16:20.03 | psokolovsky | Kevin2, do you have i2c-pxa bultin? |
| 16:22.07 | psokolovsky | Kevin2, booting from card or nfs? |
| 16:23.20 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 16:23.43 | Kevin2 | psokolovsky: No, I used a module. |
| 16:23.57 | Kevin2 | I boot from sd card. (I've never tried nfs boot.) |
| 16:24.02 | psokolovsky | ok |
| 16:24.10 | psokolovsky | will be reproducing now |
| 16:27.19 | Kevin2 | Record doesn't work - I get: |
| 16:27.32 | cr2 | so. is it a module load sequence error ? |
| 16:27.37 | Kevin2 | Nov 24 11:24:47 h2200 user.debug kernel: ALSA sound/core/oss/pcm_oss.c:940: HW_PARAMS failed: -22 |
| 16:27.53 | psokolovsky | hw initialization erroe apparently |
| 16:28.03 | psokolovsky | unload puts hw in sane state |
| 16:28.07 | psokolovsky | just guess, of course |
| 16:28.26 | psokolovsky | or maybe indeed order?!! |
| 16:28.30 | psokolovsky | fsck! |
| 16:28.32 | psokolovsky | ;-)) |
| 16:28.35 | cr2 | what is there hardware-related in unload ? |
| 16:29.13 | cr2 | we have some module dependence bugs ;) the kernel makes oops if i rmmod htcuniversal_acx, for example. |
| 16:29.21 | psokolovsky | uninitialization? |
| 16:29.40 | psokolovsky | lots, not some, actually ;-) |
| 16:31.01 | psokolovsky | Kevin2, using CVS HEAD, right? |
| 16:31.04 | cr2 | then the init process is incomplete, if it requires a reset. |
| 16:32.37 | psokolovsky | apparently |
| 16:32.41 | Kevin2 | cvs head as of last night - plus my apache changes. |
| 16:33.11 | psokolovsky | I have forward-ported drivers from 2.6.12 xanadux here rigth now, and that tricked didn't help ;-( |
| 16:33.19 | psokolovsky | switching to pristine HEAD now |
| 16:37.41 | goxboxlive | Aanyone know how i can get rid of the ACX debug message. I have set "0" in the acx_config, but there is still some debug messages. |
| 16:38.03 | goxboxlive | Kevin2. Is your sounddriver resumeing after suspend? |
| 16:39.36 | psokolovsky | Kevin2, can you please drop me a .config you use? |
| 16:39.42 | Kevin2 | I've never tried suspend. Apache isn't ready for it yet. |
| 16:41.01 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 16:41.18 | psokolovsky | i wonder, what's the state of suspend for htcsable ;-) |
| 16:41.19 | Kevin2 | psokolovsky: Sent. |
| 16:41.26 | psokolovsky | Kevin2, thanks! |
| 16:49.51 | goxboxlive | Ok, i tried out a wav file now, and it is resumeing after wakup (suspending). But i have problems witt dmesg becasue the acx is filling the dmesg with crap output. |
| 16:51.35 | Kevin2 | goxboxlive - you might have better luck with "logread -f" |
| 16:51.43 | psokolovsky | Kevin2, here, following that procedure just lockups the boxy hard |
| 16:53.51 | goxboxlive | ov 23 17:54:17 htcuniversal daemon.notice apmd[1038]: Suspending now |
| 16:54.58 | cr2 | Kevin2: you have the same suspend as the universal. |
| 16:55.01 | goxboxlive | Nov 23 17:54:48 htcuniversal user.debug kernel: HTC Universal Audio HTC Universal Audio: suspend |
| 16:55.01 | goxboxlive | Nov 23 17:54:48 htcuniversal user.debug kernel: i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 16:55.01 | goxboxlive | Nov 23 17:54:48 htcuniversal user.notice kernel: snd_htcuniversal_audio_set_headphone_power: 0 |
| 16:55.01 | goxboxlive | Nov 23 17:54:48 htcuniversal user.debug kernel: i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 16:55.04 | goxboxlive | Nov 23 17:54:48 htcuniversal user.notice kernel: snd_htcuniversal_audio_set_codec_power: 0 |
| 16:55.06 | goxboxlive | Nov 23 17:54:48 htcuniversal user.debug kernel: ak4641-i2c 0-0012: suspend |
| 16:55.08 | goxboxlive | Nov 23 17:54:48 htcuniversal user.debug kernel: eeprom 0-0051: suspend |
| 16:55.10 | goxboxlive | Nov 23 17:54:48 htcuniversal user.debug kernel: acx-mem acx-mem.0: suspend |
| 16:56.24 | goxboxlive | This rely doesnt help me, because after resume the ssh connection doesnt resume. |
| 16:56.45 | cr2 | goxboxlive: is this the "tx:" message that produces spam ? |
| 16:56.50 | psokolovsky | yes "modprobe snd-hx4700_audio ; rmmod snd-hx4700_audio ; rmmod snd_ak4641 ; rmmod snd_pxa2xx_i2sound ; modprobe snd-hx4700_audio" hangs hx4700 here |
| 16:57.23 | bd2_ | nothing hangs for me |
| 16:57.34 | bd2_ | but I'll try exactly that sequence |
| 16:57.34 | goxboxlive | cr2: Yes it is the tx message that spams me. |
| 16:57.36 | Kevin2 | Well, I didn't run all those commands in rapid succession. But otherwise it worked for me. |
| 16:58.20 | cr2 | Kevin2: why don't you add the suspend ? |
| 16:59.03 | Kevin2 | cr2: I just haven't gotten the time yet. I looked through universal code yesterday. Do I need both _power.c and _pm.c? |
| 16:59.27 | goxboxlive | cr2: look at this shit, is there a way i can get rid of it. I just want it to be like g_ether (g_ether is not resuming after suspend): http://pastebin.ca/256780 |
| 17:00.28 | cr2 | printk( KERN_NOTICE "tx: got desc %u, %u remain\n", |
| 17:00.28 | cr2 | this one ? |
| 17:00.29 | cr2 | Kevin2: you don't need power. |
| 17:00.29 | cr2 | only the _pm |
| 17:00.40 | bd2_ | psokolovsky, no.. no hang |
| 17:01.05 | cr2 | goxboxlive: ok send KERN_NOTICE to /dev/null. |
| 17:01.12 | bd2_ | goxboxlive, how do you suspending it, btw? |
| 17:01.20 | cr2 | hmm. it will kill many useful messages. |
| 17:01.35 | goxboxlive | bd2_ apm --suspend |
| 17:02.57 | bd2_ | I see.. |
| 17:03.40 | *** part/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com) |
| 17:09.16 | cr2 | <PROTECTED> |
| 17:09.16 | cr2 | <PROTECTED> |
| 17:09.16 | cr2 | <PROTECTED> |
| 17:09.16 | cr2 | <PROTECTED> |
| 17:09.27 | cr2 | this is done on suspend |
| 17:12.26 | psokolovsky | bd2_, did you try HEAD with my latest commit? maybe reset is done twice on purpose, lol? |
| 17:12.51 | cr2 | Kevin2: do we zero all requests in haret ? |
| 17:12.53 | goxboxlive | cr2: Why isent suspend work from Opie. It doesnt work form the "icon" and it is not working form the opie-console when i am calling it with "apm --suspend" |
| 17:13.19 | cr2 | goxboxlive: the power button does not report to /dev/input/event0 |
| 17:13.58 | Kevin2 | cr2: Zero what requests in haret? |
| 17:14.10 | cr2 | i have removed it from htcuniversal_kbd.c, but i'll put it back. unless we will have a sane solution for handling multiple input devices. |
| 17:14.32 | goxboxlive | cr2: What have that to do with apm in opie? |
| 17:14.39 | cr2 | #define DRCMR2 __REG(0x40000108) /* Request to Channel Map Register for I2S receive Request */ |
| 17:14.39 | cr2 | #define DRCMR3 __REG(0x4000010c) /* Request to Channel Map Register for I2S transmit Request */ |
| 17:15.06 | cr2 | Kevin2: these are DRCMRTXSADR = 0; DRCMRRXSADR = 0; |
| 17:15.31 | *** join/#htc-linux psokolovsky_ (n=psokolov@237.usernat.ip.net.ua) |
| 17:15.48 | goxboxlive | hmm i am also trying to build "btsco" but it failes. Doesnt find my bluetooth.h file. |
| 17:16.11 | cr2 | goxboxlive: you have OE locally ? then add a printk in odevice_htc.cpp |
| 17:16.24 | cr2 | where the suspend it handled. |
| 17:16.57 | goxboxlive | ok, i'll take a look at it. |
| 17:17.06 | goxboxlive | i know where that file is |
| 17:17.26 | Kevin2 | cr2: gnu-haret does. New mainline haret does not. |
| 17:17.36 | Kevin2 | But I checked - linux kernel is zeroing on its own. |
| 17:18.30 | cr2 | Kevin2: ok. |
| 17:20.24 | psokolovsky_ | Kevin2, I have few patches for haret to do strict dma reset, but it doesn't make any difference |
| 17:21.48 | Kevin2 | psokolovsky_: I think mainline code is more correct than gnu-haret code. |
| 17:22.08 | bd2_ | psokolovsky, yes, I'm runned cvs up recently |
| 17:22.51 | psokolovsky_ | Kevin2, then compare pxa25x code & pxa27x ;-). but I have patches, I'll post them. but later, as it is just cleanup ;-( |
| 17:22.56 | psokolovsky_ | bd2_, thanks |
| 17:26.25 | Kevin2 | psokolovsky_: Right, I think pxa25x code is less correct - I didn't change it because I can't test. |
| 17:26.54 | psokolovsky_ | less correct? 8-) |
| 17:27.14 | goxboxlive | cr2: What is this: |
| 17:27.16 | goxboxlive | <PROTECTED> |
| 17:27.16 | goxboxlive | <PROTECTED> |
| 17:27.16 | goxboxlive | <PROTECTED> |
| 17:27.16 | goxboxlive | <PROTECTED> |
| 17:28.17 | psokolovsky_ | goxboxlive, that's opie braindeadness |
| 17:28.48 | goxboxlive | :-) |
| 17:29.48 | psokolovsky_ | goxboxlive, did you consider my proposal re: having common buttonmap? |
| 17:32.30 | goxboxlive | psokolovsky: That is not up to me. But i guess e have to do it because we are not able to have the keyboard resuming now aftre suspend. What do you mean cr2? |
| 17:33.41 | psokolovsky_ | goxboxlive, well, that thing mostly related to having OPIE/GPE working out of the box, not to kernel probs |
| 17:34.10 | bd2_ | seems like x rule |
| 17:34.11 | Kevin2 | psokolovsky_: The existing code writes to the DINT register when the docs state the register is read-only. It writes to DDADRx registers when DMA could be in progress and docs say to not do that. I prefer a simpler approach - turn off all interrupts and be done with it. IMO, these paranoia writes to hardware chips cause more problems then they solve (witness the usb client problems). |
| 17:34.21 | psokolovsky_ | goxboxlive, so, you can try youself to use _KEY_POWER as the keycode for power button, and that should work OOB if you build libopie2_cvs |
| 17:34.57 | Kevin2 | s/turn off all interrupts/turn off all DMA/ |
| 17:35.29 | psokolovsky_ | Kevin2, well, you know more about that. I'd guess, request stops on channels, wait for compeltion, reset all regs to predefined state is more clean still... |
| 17:37.17 | psokolovsky_ | Kevin2, so, can you commit snd_htcapache_audio.c? maybe it does some magic other devices don't? |
| 17:37.39 | Kevin2 | psokolovsky_: I have to see if my haret access gives kernel access to. :-) |
| 17:38.14 | psokolovsky_ | Kevin2, ahh. sorry, I forget you don't have commit... nope, access control is ok on hh.org ;-) |
| 17:38.17 | goxboxlive | psokolovsky_ I would like to do it, but i dont know how to do it, and need help from cr2 to do the changes. If it is bether, then i think we shall go for it. |
| 17:38.31 | psokolovsky_ | well, feel free to submit it to k-d then... |
| 17:39.20 | psokolovsky_ | goxboxlive, ok. you could you h4000_keys as sample for includes to use, etc. but well, that needs more consideration indeed |
| 17:39.58 | Kevin2 | Yeah, I get permission denied. |
| 17:46.29 | bd2_ | who is interested in sound, check out k-d. ;-) |
| 17:57.33 | scatterp | when would there be a new version uploaded with sound support on the htc universal ? |
| 17:57.56 | Kevin2 | bd2_: Interesting. I was able to get sound on apache by unloading/reloading snd modules. However, both times I had 8805 in 0x40400000. (Oddly, 0x4040000c differs, so maybe it is still an i2s issue.) |
| 18:03.05 | *** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 18:05.59 | BabelOued | hi |
| 18:07.57 | cr2 | works 0x40400060 | 0000000d 00000000 00000000 00000000 |
| 18:08.25 | cr2 | is it SACR0 ? the I2S speed is set ? |
| 18:09.23 | cr2 | hmm. no |
| 18:09.54 | cr2 | SADIV. |
| 18:11.09 | cr2 | <PROTECTED> |
| 18:11.13 | cr2 | <PROTECTED> |
| 18:11.17 | cr2 | <PROTECTED> |
| 18:11.21 | cr2 | <PROTECTED> |
| 18:11.25 | cr2 | <PROTECTED> |
| 18:11.25 | cr2 | <PROTECTED> |
| 18:11.25 | cr2 | <PROTECTED> |
| 18:11.26 | cr2 | <PROTECTED> |
| 18:13.57 | Kevin2 | cr2: Were you able to get sound working by writing to 0x40400000? |
| 18:14.18 | cr2 | it's the post to k-d |
| 18:14.49 | cr2 | can somebody put a printk for card.i2slink.sadiv ? |
| 18:15.07 | bd2_ | easily |
| 18:15.10 | cr2 | well, and sacr0 to this function. |
| 18:16.29 | psokolovsky_ | cr2, I had lotsa printk's here. wanna patch? ;-) |
| 18:16.30 | cr2 | the differences are in SACR0 and SADIV, and the sound does not work when they are improperly initialized. |
| 18:17.02 | cr2 | "sounds" logical ;-) |
| 18:17.22 | cr2 | psokolovsky_: i don't do real testing :) |
| 18:18.00 | psokolovsky_ | for me: |
| 18:18.00 | psokolovsky_ | 0x40400000 | 00008805 00000000 00000000 00000005 |
| 18:18.01 | psokolovsky_ | 0x40400010 | 00000000 00000078 00000000 00000000 |
| 18:18.05 | psokolovsky_ | non-working |
| 18:18.49 | cr2 | print +0x00 and +0x60 |
| 18:18.56 | Kevin2 | psokolovsky_: That is also what I see - when it works I see: |
| 18:19.01 | cr2 | SADIV should be 0xd |
| 18:19.08 | Kevin2 | 0x40400000 | 00008805 00000000 00000000 0000007f |
| 18:19.09 | cr2 | dump pxa27xsound |
| 18:19.12 | psokolovsky_ | SADIV, yeah, 0x0d |
| 18:19.14 | cr2 | in haret |
| 18:19.41 | cr2 | otherwise you have a weird clock speed. |
| 18:19.48 | psokolovsky_ | an 0x4040000c is status reg |
| 18:19.53 | psokolovsky_ | so, no differences ;-) |
| 18:20.30 | cr2 | do you have the same SACR0 and SADIV like in wince ? |
| 18:21.16 | psokolovsky_ | cr2, if me, I dunno about wince ;-) |
| 18:21.34 | cr2 | i'll look in the channel logs. |
| 18:22.17 | cr2 | <PROTECTED> |
| 18:22.42 | *** join/#htc-linux RoEn (n=roen@p54A64906.dip.t-dialin.net) |
| 18:22.42 | cr2 | <PROTECTED> |
| 18:22.48 | bd2_ | _SADIV: 0xD |
| 18:22.48 | bd2_ | SACR0: 0x8 |
| 18:22.48 | bd2_ | SADIV: 0xC |
| 18:22.49 | cr2 | it's on magician. |
| 18:22.53 | psokolovsky_ | cr2, I even tried to install i2s interrupt handler. it is never called ;-). nor on hx4700(dead) nor on h4000 (work) |
| 18:23.31 | psokolovsky_ | I bet we follow false cause here ;-) |
| 18:23.31 | bd2_ | http://paste.lisp.org/display/30623 here is code |
| 18:24.23 | bd2_ | sound not working as now |
| 18:24.40 | psokolovsky_ | bd2_, : |
| 18:24.41 | psokolovsky_ | ALSA sound/arm/pxa2xx-i2sound.c:585: Set SADIV to 0xc |
| 18:24.41 | psokolovsky_ | ALSA sound/arm/pxa2xx-i2sound.c:436: Started DMA on channel 10 |
| 18:24.41 | psokolovsky_ | ALSA sound/core/pcm_lib.c:1714: playback write error (DMA or IRQ trouble?) |
| 18:24.41 | psokolovsky_ | ALSA sound/arm/pxa2xx-i2sound.c:445: Stopping DMA on channel 10 |
| 18:24.41 | psokolovsky_ | ALSA sound/arm/pxa2xx-i2sound.c:305: DMA IRQ for channel 10, DCSR=0x20000008 |
| 18:24.45 | psokolovsky_ | ALSA sound/arm/pxa2xx-i2sound.c:307: I2S SASR0 register=0x00000005 |
| 18:24.55 | bd2_ | _SADIV: 0xC |
| 18:24.55 | bd2_ | SACR0: 0x8805 |
| 18:24.55 | bd2_ | SADIV: 0xC |
| 18:25.10 | bd2_ | ^^^ that is after echo mem > /sys/power/state and sound now working. |
| 18:25.35 | cr2 | 0xc ? |
| 18:26.00 | psokolovsky_ | 0c is still valid for sadiv |
| 18:26.28 | psokolovsky_ | 0x0000_000C 12.235 MHz 3.058 MHz 47.794K (closest std = 48 kHz) |
| 18:26.41 | bd2_ | _SADIV: 0xC |
| 18:26.41 | bd2_ | SACR0: 0x8 |
| 18:26.41 | bd2_ | SADIV: 0xC |
| 18:26.58 | bd2_ | ^^^ that is after echo standby > /sys/power/state and sound not working |
| 18:27.48 | bd2_ | look at my first _SADIV (0xD) it's printed from variable.. and SADIV (0xC) printed from register. they're differ |
| 18:28.01 | cr2 | 000 1101 = Selects an I2S_BITCLK of 2.836 MHz |
| 18:28.07 | cr2 | this is for 0xd |
| 18:28.09 | psokolovsky_ | 48Khz vs 44.1Khz |
| 18:28.53 | psokolovsky_ | there's something else to it, guys ;-) |
| 18:29.05 | Kevin2 | Okay, I can get apache to work by playing with I2S writes. |
| 18:29.23 | psokolovsky_ | bd2_, and about 0x77xx in SACR0 - stupid thing just doesn't reinit it after resume %) |
| 18:29.32 | psokolovsky_ | Kevin2, how?! |
| 18:30.28 | Kevin2 | Sequence was this: modprobe snd-htcapache-audio ; x -w 0x40400000 0x880d ; x -w 0x40400060 0x0d ; x -w 0x40400000 0x8805 |
| 18:30.55 | cr2 | psokolovsky_: if snd_pxa2xx_i2sound_i2slink_on(void) is called it should be reinitialized. |
| 18:30.55 | psokolovsky_ | Kevin2, YOU RULEZZZ!!! |
| 18:30.56 | *** join/#htc-linux anderson (n=dirk@p54B3F67B.dip.t-dialin.net) |
| 18:30.57 | psokolovsky_ | ;-D |
| 18:31.18 | psokolovsky_ | here noise here ;-D |
| 18:31.20 | anderson | hallo |
| 18:31.23 | anderson | :_) |
| 18:31.26 | anderson | :-) |
| 18:31.28 | Kevin2 | :-) |
| 18:31.42 | anderson | spricht hier jemand deutsch ?? |
| 18:33.05 | goxboxlive | anderson: The most of the people hanging here are germans, but the channel language is english. |
| 18:33.06 | psokolovsky_ | Kevin2, so you reset it? But driver does the same, fsck! I thought it lacks delay! |
| 18:33.15 | anderson | ok |
| 18:34.00 | bd2_ | as I said.. value from variable and register differs.. most probably delay needed |
| 18:34.51 | psokolovsky_ | bd2_, sorry, I missed that you meant that. well, but how that could happen? |
| 18:35.15 | bd2_ | psokolovsky_, don't know.. maybe we should consult with intel's errata :-) I recalling much i2s issues |
| 18:37.30 | psokolovsky_ | "1. If SACR0[ENB] is toggled in the middle of a normal operation, the SACR0[RST] bit must also be set and cleared to |
| 18:37.30 | psokolovsky_ | reset all I2S controller registers." |
| 18:38.46 | cr2 | <PROTECTED> |
| 18:38.46 | cr2 | <PROTECTED> |
| 18:38.46 | cr2 | <PROTECTED> |
| 18:38.47 | psokolovsky_ | and what's "normal operation"? |
| 18:38.56 | cr2 | is this code correct ? |
| 18:39.20 | scatterp | when would there be a new version uploaded with sound support on the htc universal ? |
| 18:39.34 | cr2 | scatterp: when it will work ;) |
| 18:39.56 | Kevin2 | cr2: Does i2s writes work on universal? |
| 18:39.57 | scatterp | oh i thought psokolovsky fixed it or something ? |
| 18:40.37 | cr2 | Kevin2: i don't have a testing environment. |
| 18:41.00 | scatterp | cr2 want to telnet my phone ? |
| 18:42.30 | cr2 | anderson: wir haben uns geeinigt daß hier nur englisch gesprochen wird. dann können die anderen auch was davon verstehen :) |
| 18:42.43 | *** join/#htc-linux ljp (n=lpotter@203.94.178.46) |
| 18:43.13 | cr2 | scatterp: it does not make much sense. we need to change the kernel and the modules. |
| 18:43.53 | Kevin2 | I'll be back later. |
| 18:44.14 | *** join/#htc-linux ljp (n=lpotter@203.94.178.46) |
| 18:48.36 | goxboxlive | sctterp: Do , me a favour, would you? |
| 18:49.43 | goxboxlive | s/sctterp:/scatterp:/ |
| 18:57.41 | scatterp | ok |
| 18:58.10 | goxboxlive | scatterp: Do you have opie running? |
| 18:58.22 | scatterp | yes but its kinda buggy |
| 18:58.27 | goxboxlive | how come? |
| 18:58.44 | scatterp | remember i said about the cursor comeing through the display |
| 18:58.48 | scatterp | from the console |
| 18:58.54 | scatterp | but its useable |
| 18:59.17 | goxboxlive | ohh, i have uploaded new images since then |
| 18:59.21 | goxboxlive | do you have a mp3 file on your universal? |
| 18:59.24 | scatterp | ok ill grab them now or later depending on what you need |
| 18:59.28 | scatterp | yes in windows i do |
| 18:59.35 | scatterp | i can mount it too |
| 18:59.43 | goxboxlive | good |
| 19:00.04 | goxboxlive | Do you have a logout button in the opie-menu? |
| 19:00.14 | scatterp | yes i think so |
| 19:00.19 | goxboxlive | ok, hit it |
| 19:00.19 | scatterp | shall i boot opie / |
| 19:00.26 | scatterp | ok booting it |
| 19:00.30 | goxboxlive | good |
| 19:00.38 | goxboxlive | remember taht we need a mp3 file |
| 19:00.53 | scatterp | yeap |
| 19:01.03 | scatterp | its booting |
| 19:01.04 | goxboxlive | We shall just test to see if the sound is also working on your older image. |
| 19:01.11 | goxboxlive | good |
| 19:01.24 | scatterp | opie/splash screen .... |
| 19:02.04 | goxboxlive | ok, when opie is running hit the logout button in the opie-menu |
| 19:02.08 | scatterp | opies up |
| 19:02.24 | goxboxlive | are you back in the shell? |
| 19:02.48 | scatterp | sorry i was mistaken no logout button just shutdown |
| 19:03.01 | scatterp | i can ctrl alt backspace |
| 19:03.02 | goxboxlive | hmm, ok |
| 19:03.03 | scatterp | should i ? |
| 19:03.06 | goxboxlive | no |
| 19:03.09 | scatterp | ok |
| 19:03.25 | goxboxlive | hit the shutown icon, and hit terminate opie |
| 19:03.30 | scatterp | ok |
| 19:03.36 | goxboxlive | lets hope it works |
| 19:03.41 | scatterp | dont |
| 19:03.43 | scatterp | im in console |
| 19:03.47 | goxboxlive | good |
| 19:03.49 | scatterp | login and mount ? |
| 19:03.53 | goxboxlive | lodin |
| 19:03.57 | goxboxlive | login as root |
| 19:04.02 | goxboxlive | no passwd |
| 19:04.18 | goxboxlive | ok |
| 19:04.24 | goxboxlive | still in shell right? |
| 19:04.37 | scatterp | yes |
| 19:04.42 | scatterp | mp3s mounted |
| 19:05.03 | goxboxlive | type "/etc/init.d/opie start" without the quotes |
| 19:05.04 | scatterp | /media/mmcblk0p3 |
| 19:05.08 | scatterp | ok |
| 19:05.57 | scatterp | done |
| 19:06.06 | goxboxlive | opie is running right? |
| 19:06.35 | scatterp | yes |
| 19:06.41 | goxboxlive | open opie-console |
| 19:06.54 | scatterp | done |
| 19:07.19 | goxboxlive | is the mp3 file partition mounted? |
| 19:07.28 | scatterp | i believe its at |
| 19:07.30 | scatterp | /media/mmcblk0p3 |
| 19:07.32 | scatterp | let me check |
| 19:07.38 | goxboxlive | check with "df" |
| 19:07.58 | goxboxlive | or "cat /proc/mounts" |
| 19:08.02 | scatterp | i see it in both |
| 19:08.05 | scatterp | at |
| 19:08.07 | goxboxlive | good |
| 19:08.07 | scatterp | /media/mmcblk0p3 |
| 19:08.32 | goxboxlive | type: "modprobe -r snd-htcuniversal_audio" |
| 19:09.12 | scatterp | done |
| 19:09.12 | goxboxlive | and then "modprobe snd-htcuniversal_audio" |
| 19:09.20 | goxboxlive | non of them with quotes |
| 19:09.54 | scatterp | ok got an error |
| 19:09.55 | scatterp | on the second time |
| 19:09.55 | goxboxlive | when reloading the module? |
| 19:09.58 | scatterp | modprobe snd-htcuniversal_audio |
| 19:10.14 | goxboxlive | failed to insert module? |
| 19:10.31 | scatterp | yes |
| 19:10.48 | goxboxlive | o, then sound is not working with your image. |
| 19:11.03 | goxboxlive | I'll make a new one when we have sound working. |
| 19:11.32 | scatterp | FATAL: Error inserting snd_htcuniversal_audio (/lib/modules/2.6.17-hh1/kernel/sound/arm/snd-htcuniversal_audio.ko) : Operation not permited |
| 19:11.35 | scatterp | that was the error |
| 19:11.42 | scatterp | i can reinstall with the other version |
| 19:11.46 | scatterp | if you give me the url |
| 19:12.01 | scatterp | i mean like a newer version of opie |
| 19:15.28 | goxboxlive | hold on |
| 19:15.35 | scatterp | k |
| 19:16.05 | goxboxlive | http://gnulinux.biz/files//Universal/tmp/Angstrom/ both zimage anf opie |
| 19:16.24 | scatterp | ok installing |
| 19:21.19 | scatterp | downloaded copying to phone |
| 19:28.02 | scatterp | ok booting linux to untar everything to the partion |
| 19:29.57 | goxboxlive | ok, i am trying to make a new image now with the latest kernel drivers, but there is only sound in the headphones. |
| 19:30.19 | scatterp | well thats a big step forward |
| 19:33.14 | *** join/#htc-linux pleemans (n=peter@d51A5E421.access.telenet.be) |
| 19:43.16 | scatterp | untaring... |
| 19:43.30 | goxboxlive | gtood |
| 19:43.55 | scatterp | phone seems a little slugish today |
| 19:43.56 | scatterp | hmm |
| 20:01.30 | *** join/#htc-linux psokolovsky__ (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com) |
| 20:02.53 | scatterp | booting windows |
| 20:03.47 | goxboxlive | ok |
| 20:04.23 | scatterp | got to fix default.txt |
| 20:05.25 | goxboxlive | yes, because of the swap partition |
| 20:05.55 | goxboxlive | psokolovsky_ What did you do? hh.org cvs is down for anoncvs again. |
| 20:05.59 | goxboxlive | :-) |
| 20:06.19 | psokolovsky__ | goxboxlive, you did that! I don't use anoncvs ;-) |
| 20:07.08 | goxboxlive | actually, i build a image for about 1/2 an hour ago and the cvs where updated, but after your update it dont work. |
| 20:08.23 | scatterp | booting linux |
| 20:08.38 | scatterp | starting opie |
| 20:09.06 | psokolovsky__ | goxboxlive, lol. tried traceroure? |
| 20:09.21 | psokolovsky__ | 14 p5-0.core01.bos01.atlas.cogentco.com (66.28.4.117) 149.29 ms 152.161 ms 151.433 ms |
| 20:09.21 | psokolovsky__ | 15 g8.ba21.b002250-1.bos01.atlas.cogentco.com (66.250.14.210) 170.922 ms 148.248 ms 147.026 ms |
| 20:09.21 | psokolovsky__ | 16 MIT.demarc.cogentco.com (38.112.2.214) 149.805 ms 163.805 ms 149.492 ms |
| 20:09.21 | psokolovsky__ | 17 * * * |
| 20:09.28 | scatterp | im in opie |
| 20:09.52 | scatterp | also the battery indicater indicates chargeing in opie and i have an orange light am i chargeing ? |
| 20:09.55 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 20:10.08 | scatterp | and what should i do |
| 20:10.36 | goxboxlive | pskolovsky: i know |
| 20:10.55 | goxboxlive | scatterp: no i dont thinkl so |
| 20:11.11 | goxboxlive | do not use the the power modules, it's not safe |
| 20:11.50 | goxboxlive | scatterp: do as we did before ( not rebbot opie, just the snd-htcuniversal_audio stuff) |
| 20:12.13 | scatterp | umm not sure what you mean i didnt "use power modules" i just booted and see a power indicater in opie |
| 20:12.25 | goxboxlive | scatterp: open opie-console |
| 20:12.27 | scatterp | doing the audio stuff |
| 20:12.34 | goxboxlive | also do |
| 20:12.46 | goxboxlive | modprobe -r htcuniversal_power |
| 20:12.54 | scatterp | first ? |
| 20:12.58 | goxboxlive | doesnt mather |
| 20:13.21 | scatterp | ok i did that |
| 20:13.25 | scatterp | still have a orange light |
| 20:13.34 | goxboxlive | thats not a problem |
| 20:13.39 | scatterp | but opie power bar stoped moveing |
| 20:13.41 | goxboxlive | have you done the sound stuff? |
| 20:13.45 | scatterp | ill do the other commands nw |
| 20:13.51 | goxboxlive | forget the power now, it's ok |
| 20:14.15 | cr2 | goxboxlive: what is *#31# ? |
| 20:14.53 | goxboxlive | cr2: i dont know. |
| 20:14.57 | scatterp | ok done this time with out errors |
| 20:15.02 | goxboxlive | ok |
| 20:15.18 | goxboxlive | scattrep: connect a headphone to you universal |
| 20:15.30 | scatterp | ok |
| 20:15.49 | goxboxlive | And the open opie-mediaplayer |
| 20:17.19 | goxboxlive | file->add a file-> browse to your mp3 file-> hit "OK" -> hit the white "play" icon in the right top (triangle) and see what's happend. |
| 20:17.29 | goxboxlive | Does it play the mp3 file? |
| 20:18.01 | scatterp | when i clicked file |
| 20:18.05 | scatterp | i heard |
| 20:18.08 | scatterp | some clicks |
| 20:18.11 | scatterp | proceeding |
| 20:18.30 | psokolovsky__ | So I still cannot get sound work directly after boot, but now reinsert trick finally works for me |
| 20:19.05 | goxboxlive | psokolovsky_ Will the update affect the universal? Should i manually download the file and try it? |
| 20:19.15 | scatterp | no |
| 20:19.23 | scatterp | but i do have something |
| 20:19.35 | psokolovsky__ | goxboxlive, nope, never hurry. let me fix it fully first (or try at least ;-) ) |
| 20:19.51 | scatterp | i hear like that noise a sea shell makes when you put it to your ear |
| 20:20.02 | goxboxlive | psokolovsky_ i love living on the edge, but i can wait. |
| 20:20.15 | psokolovsky__ | ok ;-) |
| 20:20.48 | psokolovsky__ | I'll let you know when it give up to my pressure ;-) |
| 20:20.50 | goxboxlive | scatterp: Ok, then the half working sound was after yesterdays changes in CVS |
| 20:21.15 | scatterp | so now its broke again ? |
| 20:21.25 | goxboxlive | scatterp: Do you want a image with working headphones? |
| 20:21.28 | scatterp | i tryed changeing volume didnt effect anything |
| 20:21.48 | scatterp | yes goxboxlive that would be great |
| 20:22.06 | scatterp | it should for safty maybe have the power command auto on bootup |
| 20:22.39 | goxboxlive | scatterp: I have removed it in this image |
| 20:23.44 | goxboxlive | scatterp: i am uploading it now, but just be aware of that you have to reload the sound module after boot to get it working. And as i said, it is only headphones that works. |
| 20:23.45 | scatterp | cool yeah that would be good |
| 20:24.00 | scatterp | thats great |
| 20:24.08 | scatterp | cant i add those commands to a file |
| 20:24.12 | goxboxlive | scatterp: ANd use the provided zImage, the one you downloaded os now obselete |
| 20:24.17 | scatterp | something like /etc/rc |
| 20:24.28 | scatterp | that does commands on bootup |
| 20:25.03 | scatterp | goxboxlive you understand that i can generate sound with this by open closeing the file right ? |
| 20:25.04 | goxboxlive | scatterp: Yes you can make a script to do it, but i guess in a couple of days the "gurus" has made it work from boot, so there is no need for that. |
| 20:25.19 | goxboxlive | ?? |
| 20:25.35 | scatterp | you know when you turn on an mp3 player |
| 20:25.36 | goxboxlive | You mean buzz, not real sound? |
| 20:25.38 | scatterp | you hear a noise |
| 20:25.43 | scatterp | not real sound |
| 20:25.50 | scatterp | i wouldnt describe it as a buzz sound |
| 20:25.51 | goxboxlive | yes like "zzzzhhhhhhhhhhhhhhhhhssssssssssssssss" |
| 20:25.58 | scatterp | more a "on" sound |
| 20:26.09 | goxboxlive | yes, i know that |
| 20:26.12 | scatterp | ok cool |
| 20:27.06 | goxboxlive | scatterp: They are now uploaded. |
| 20:27.30 | scatterp | http://gnulinux.biz/files//Universal/tmp/Angstrom/ |
| 20:27.33 | scatterp | are they there? |
| 20:28.22 | goxboxlive | Yes as allways |
| 20:28.28 | scatterp | great |
| 20:28.39 | scatterp | well i hope that test was some help to you guys |
| 20:28.46 | scatterp | if you need me to test anything else let me know |
| 20:30.50 | goxboxlive | scatterp: I was just curioues to see if the sound where fixed last night or before. |
| 20:31.06 | scatterp | yeah |
| 20:31.22 | scatterp | so its just mic and power to fix now reallt |
| 20:31.23 | scatterp | so its just mic and power to fix now really |
| 20:33.29 | scatterp | brb going to have a smoke |
| 20:45.13 | goxboxlive | scatterp: The mic is partially working, but it records buzzy. Did the power-module load? |
| 20:46.06 | goxboxlive | psokolovsky_ he cvs is up again now. |
| 20:46.42 | psokolovsky__ | ok, nice |
| 20:47.45 | scatterp | goxboxlive i didn install the new opie yet |
| 20:47.55 | scatterp | i have low battery and i need to make a call |
| 20:48.12 | scatterp | i have ordered some spare enhanced batterys and an external charger |
| 20:49.46 | psokolovsky__ | ok, so far I seem to be reliably remove and resinsert modules and sound works |
| 20:51.44 | scatterp | http://www.boxwave.com/products/highcapacitybattery/electraspan-high-capacity-extended-battery-htc-universal_724.htm |
| 20:51.48 | goxboxlive | psokolovsky__ i tried the new pxa-i2sound my self. It didnt affect the universal |
| 20:52.30 | scatterp | goxboxlive ill let you know when ive updated and retested might be a little while |
| 20:52.39 | goxboxlive | scatterp: That was a big one. BTW do you know any free SIP providers? |
| 20:53.17 | scatterp | hmm i remember reading some stuff i will look in to it for you and get back to you with some info |
| 20:53.25 | goxboxlive | good |
| 20:53.36 | goxboxlive | need to make kphone/pi work. |
| 20:55.10 | scatterp | when i say good stuff i mean free calls to non sip numbers |
| 20:55.22 | scatterp | give me a while ill update you shortly |
| 20:55.51 | goxboxlive | cool |
| 20:59.20 | *** join/#htc-linux WizMaui_ (n=WizMaui@62.112.90.179) |
| 21:05.31 | goxboxlive | cr2: Are you around? |
| 21:10.16 | *** join/#htc-linux RoEn_PC (n=roen@p54A64906.dip.t-dialin.net) |
| 21:23.03 | cr2 | goxboxlive: yes. |
| 21:28.17 | goxboxlive | cr2: Where did you find : *#31# And did you find out what it means? |
| 21:32.27 | BabelOued | *#31# is masqurade call |
| 21:34.02 | goxboxlive | ok |
| 21:34.49 | psokolovsky__ | well, I still can't get sound after boot... ;-( |
| 21:35.18 | goxboxlive | me neither |
| 21:36.24 | goxboxlive | psokolovsy__ With the .12 kernel we where not able to put the snd-htcuniversal in ti the kernel Image |
| 21:36.47 | goxboxlive | We had to use it as a module, otherwise it wouldnt work. |
| 21:37.05 | goxboxlive | But we didnt need to load@reload it to let it work. |
| 21:37.44 | goxboxlive | I have also tried to boot the .18 kernel with the snd-htcuniversal not loaded at boot, and rather load it manually |
| 21:37.51 | goxboxlive | but that didnt help either. |
| 21:37.52 | psokolovsky__ | yeah, .12 works for me after boot too |
| 21:38.06 | cr2 | BabelOued: *#76# and *#77# ? |
| 21:38.09 | psokolovsky__ | and I don't see anything magic in it... |
| 21:38.10 | goxboxlive | But not as a part of the Image |
| 21:39.51 | psokolovsky__ | goxboxlive, for me, works as part |
| 21:40.02 | goxboxlive | realy |
| 21:40.27 | psokolovsky__ | CONFIG_SND_HX4700=y |
| 21:40.31 | psokolovsky__ | that's it |
| 21:40.35 | goxboxlive | Ok, we could not have it in the zImage, it failed |
| 21:40.37 | goxboxlive | right |
| 21:41.29 | goxboxlive | psokolovsky__ Is everything working in the .12 kernel for your hx4700? |
| 21:41.58 | psokolovsky__ | goxboxlive, nothing except sound ;-P |
| 21:42.28 | goxboxlive | :-) |
| 21:42.45 | psokolovsky__ | I have no idea what patches you had so it suspended for you |
| 21:43.01 | psokolovsky__ | decent pxa27x suspend/resume patches went into mainline only on .13 |
| 21:43.11 | psokolovsky__ | sor me, it just hangs on suspend |
| 21:43.15 | psokolovsky__ | for me |
| 21:43.24 | goxboxlive | ok |
| 21:50.56 | *** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 21:57.05 | BabelOued | cr2: i not have this command |
| 22:00.36 | cr2 | BabelOued: universal has both |
| 22:04.49 | lkcl | allo darlins |
| 22:05.50 | cr2 | lkcl: what are you going to buy today ? |
| 22:06.05 | lkcl | micro-sd cards for the hermes |
| 22:06.15 | cr2 | it has micro ? |
| 22:07.14 | cr2 | read that, item 4) : http://lists.openmoko.org/pipermail/community/2006-November/000249.html |
| 22:07.58 | lkcl | yeh |
| 22:14.15 | BabelOued | cr2: any idea about battery or WIFI on BA ? |
| 22:15.08 | cr2 | BabelOued: the battery needs an OWM driver, it seems pH5 has already written it, so it's better to talk to him. |
| 22:15.42 | cr2 | can't tell you anything useful about wifi. do we know the full powerup/powerdown gpio sequence ? |
| 22:16.24 | cr2 | do you have an usb connection to ba to test switching wifi on/off ? |
| 22:18.10 | BabelOued | cr2: yes i have both |
| 22:18.24 | BabelOued | #define GPIOA_WLAN_PWR1 (1 << 3) |
| 22:18.37 | BabelOued | #define GPIOA_WLAN_PWR2 (1 << 5) |
| 22:18.54 | BabelOued | #define GPIOB_WLAN_PWR3 (1 << 13) |
| 22:19.26 | BabelOued | i've found this one |
| 22:19.33 | cr2 | i'll create a new haret version, so you'd be able to do 'tiacxon 0' 'tiacxon 1' |
| 22:19.52 | cr2 | and trace it with wasic3gpio. |
| 22:20.56 | BabelOued | ok |
| 22:23.11 | cr2 | we also need to revise the battery gpios. |
| 22:23.39 | cr2 | I would like my Jasjar to act as a WiFi Access Point for other devices, it would appear this is not possible with WM5; however is/would this possible with Linux ? |
| 22:23.39 | cr2 | Reply With Quote |
| 22:23.56 | cr2 | on xda-dev |
| 22:25.11 | lkcl | ok - i'm running the camera, and A0 gets initialised to 1 when the camera is turned on. |
| 22:28.30 | cr2 | BabelOued: gnuharet-200611242327.exe |
| 22:28.48 | cr2 | the command is 'tiacxctl' |
| 22:28.56 | BabelOued | same place as other ? |
| 22:28.58 | cr2 | 0 and 4 |
| 22:28.59 | cr2 | yes. |
| 22:30.00 | cr2 | lkcl: what is your cam i2c_id ? |
| 22:30.09 | lkcl | oo - i can find out in a mo.... |
| 22:30.17 | cr2 | great. |
| 22:30.22 | goxboxlive | cr2: I have build btsco. Will it work, or is it just waste of time to try it out? |
| 22:30.39 | lkcl | i'll do a scan under linux in oh uh... i have to find its gpio init first and leave them enabled in the htcsable init |
| 22:30.57 | Kevin2 | lkcl: I've been able to get sound working by touching the i2s registers. Paul has also gotten it to work on hx4700. Do you want to try on Sable? |
| 22:31.39 | cr2 | goxboxlive: can you connect to the bt headset ? |
| 22:32.00 | Kevin2 | (13:30:28) Kevin2: Sequence was this: modprobe snd-htcapache-audio ; x -w 0x40400000 0x880d ; x -w 0x40400060 0x0d ; x -w 0x40400000 0x8805 |
| 22:32.55 | goxboxlive | I havent tried yet. The only thing i have done is to see if the uni finds it, and it did. Havent made any rfcomm for it yet. I wanted to build btsco first ( it has failed the last half year) |
| 22:32.58 | lkcl | yaay! |
| 22:33.03 | lkcl | yespleasekevin |
| 22:33.17 | cr2 | lkcl: with 'set trace' on wince |
| 22:34.42 | cr2 | set trace 0xae301688 |
| 22:34.50 | lkcl | cr2: oh ok |
| 22:35.11 | cr2 | you'l see not only the id, but all the data. |
| 22:35.13 | Kevin2 | lkcl: sequence is above - you'll need to grab x.c from svn - svn://www.shadowmite.com/home/svn/ppc6700/tools |
| 22:35.17 | goxboxlive | kevin2: Should i try with the same adresses on the Universal? |
| 22:35.22 | lkcl | Kevin2: ack |
| 22:35.39 | Kevin2 | goxboxlive: It would help. |
| 22:36.04 | lkcl | ooooooo |
| 22:36.07 | lkcl | weeeeird |
| 22:36.37 | lkcl | ok i have a file 'x'... |
| 22:37.16 | cr2 | Kevin2: the 0x88 is the fifo setting, but 0xd -> 0x5 ? the lower bit is i2s enable, 0x4 is output, what is 0xd ? |
| 22:37.47 | cr2 | i.e. 0x8 ? |
| 22:38.17 | lkcl | brb baby.. |
| 22:39.19 | Kevin2 | It is reset+output+enable. We later write output+enable. |
| 22:40.02 | Kevin2 | psokolovsky doesn't think it has anything to do with the sequence -- it might just be the reads/writes themselves. I just know it worked for me on Apache. |
| 22:40.23 | goxboxlive | Kevin2: What do i do with the x.c to make it run? how do i compile it ? |
| 22:40.47 | psokolovsky__ | Kevin2, I now have "module load-unload" trick work for hx4700 too |
| 22:41.00 | psokolovsky__ | but your writes don't work in kernel driver |
| 22:41.17 | psokolovsky__ | more specifically, i2s regs are readonly somehow... |
| 22:41.41 | scatterp | goxboxlive fyi im installing now |
| 22:42.15 | *** join/#htc-linux g3gg0_ (n=g3gg0@ppp-82-135-69-143.dynamic.mnet-online.de) |
| 22:43.10 | goxboxlive | lkcl: Do you own a greenphone? |
| 22:43.17 | lkcl | yep. |
| 22:43.23 | goxboxlive | how did you et it? |
| 22:43.33 | lkcl | i got looootts of phones. |
| 22:43.35 | Kevin2 | goxboxlive: The first few lines of the file have the compile line. |
| 22:43.41 | Kevin2 | arm-linux-gcc -march=armv5te -static -Wall -O x.c -o x |
| 22:43.44 | lkcl | paid for it? |
| 22:44.03 | lkcl | fell off back of lorry. |
| 22:44.05 | goxboxlive | lkcl: orderd form where, trolltech? |
| 22:44.15 | lkcl | yes gobox |
| 22:44.21 | goxboxlive | i c |
| 22:44.36 | scatterp | talking about "fell off the back of a lorry" |
| 22:44.47 | scatterp | is it ever going to be possible to make the htc universal |
| 22:44.50 | scatterp | give free calls |
| 22:44.53 | scatterp | via an exploit |
| 22:44.58 | scatterp | i mean is it likely |
| 22:45.05 | cr2 | Kevin2: but what was the old driver doing ? |
| 22:45.32 | Kevin2 | cr2: I don't know - I never looked. |
| 22:45.32 | cr2 | scatterp: i think the capabilities of the gsm lib on greenphone are limited |
| 22:45.54 | scatterp | hmm |
| 22:45.55 | cr2 | but the docs mention the video calls. |
| 22:46.30 | cr2 | lkcl: will you also do with the greenphone what you have done with the sable ? :) |
| 22:47.22 | lkcl | yes i will |
| 22:47.42 | lkcl | i want a sable kernel before i go on to userspace tho |
| 22:47.49 | cr2 | :) then tell me the BT chipset used. |
| 22:47.51 | cr2 | ok. |
| 22:47.59 | cr2 | me too. |
| 22:48.03 | lkcl | also dont want to be 1st person to do 2.6 kernel :) |
| 22:48.45 | cr2 | is it samsung cpu ? |
| 22:49.03 | cr2 | so many phones, i forget it. |
| 22:49.46 | lkcl | ok. psoko on handhelds.org tried load -xa-i2s then suspend then resume THEN load audio |
| 22:49.51 | lkcl | from wince |
| 22:49.56 | lkcl | aand sound wroked |
| 22:50.23 | cr2 | no, it's pxa270. |
| 22:51.15 | cr2 | lkcl: bluetooth, sound, touchscreen. then you'll get your 2.6 |
| 22:53.48 | cr2 | have not found it in the arm machine registry. |
| 22:58.47 | goxboxlive | Kevin2 / cr2 : When i did the x stuff, i didnt need to reload the sound modules. But the main speakers are still not workimng, only sound from headphones |
| 23:00.36 | cr2 | goxboxlive: if you have the sound working with the headphones, you don't need this hack. |
| 23:00.44 | cr2 | your i2s is ok. |
| 23:01.50 | cr2 | dump the asic3 gpioA status register instead, or look in the dmesd output. are A1 and A6 gpios on ? |
| 23:02.01 | goxboxlive | cr2: But, to get it work with headphones, i have to remove the sounddriver and then load it again. Then i have sound in my headphones. So i guess we need it |
| 23:02.10 | cr2 | ok. |
| 23:02.28 | goxboxlive | It doesnt work after a bootup. |
| 23:02.40 | cr2 | but the speakers is only the i2c command and gpio issue. |
| 23:04.05 | goxboxlive | I have also tried to boot without the snd-htcuniversal_audio and loaded it when the machine has started opie, but i have to unload it and then load it again to get sound. |
| 23:05.38 | cr2 | ok. |
| 23:06.08 | cr2 | use the Kevin's x tool. |
| 23:06.34 | cr2 | on 0x10000018 (16bit reg) |
| 23:06.41 | Kevin2 | goxboxlive - the gpio pin for speaker is probably wrong. |
| 23:06.44 | psokolovsky | ok, it appears the gore is due to wrong codec activation sequence |
| 23:07.10 | cr2 | Kevin2: it works in .12 and i have traced it in wince, and seen in the wavedev.dll |
| 23:07.30 | Kevin2 | cr2: cut and paste error in .18 then? |
| 23:07.46 | cr2 | error ? |
| 23:08.01 | goxboxlive | no no poutput |
| 23:08.19 | cr2 | goxboxlive: dump the asic3 gpio area |
| 23:08.34 | goxboxlive | cr2: I dont know how to do that, you tell me then |
| 23:08.49 | cr2 | Kevin2: what is the 'x' syntax ? |
| 23:09.37 | Kevin2 | cr2: run "x" with no args, and it will tell you. |
| 23:10.22 | cr2 | i don't have a working device. |
| 23:10.29 | cr2 | x 0x10000000 0x200 |
| 23:10.33 | goxboxlive | root@htcuniversal:~$ x |
| 23:10.33 | goxboxlive | Usage: |
| 23:10.33 | goxboxlive | <PROTECTED> |
| 23:10.33 | goxboxlive | OR x [options] -w <address> <value> |
| 23:10.33 | goxboxlive | Options: |
| 23:10.34 | goxboxlive | <PROTECTED> |
| 23:10.36 | goxboxlive | <PROTECTED> |
| 23:10.38 | goxboxlive | <PROTECTED> |
| 23:10.40 | goxboxlive | <PROTECTED> |
| 23:11.07 | cr2 | goxboxlive: pastebin "x 0x10000000 0x200" |
| 23:11.28 | goxboxlive | cr2:. root@htcuniversal:~$ x 0x10000000 0x200 http://pastebin.ca/257081 |
| 23:12.02 | Kevin2 | cr2: If you want to access 16-bit regs, then use '-2' option. |
| 23:13.22 | cr2 | ok. 0xf9a5 -> 1111.1001.1010.0101, 1 - not set, 6 - not set => no speakers. |
| 23:13.44 | Kevin2 | psokolovsky: The xanadux code writes zero to SACR0 after reset, and then writes zero to SACR1 at the end of init. Did you try this in .18 code? |
| 23:14.10 | psokolovsky | Kevin2, sure ;-) |
| 23:14.26 | goxboxlive | cr2: here is the 16-bit reg http://pastebin.ca/257083 |
| 23:14.27 | psokolovsky | Kevin2, I even tried to write directly to CKEN and lotsa other tricks |
| 23:14.36 | psokolovsky | finally, something worked ;-) |
| 23:15.08 | psokolovsky | Kevin2, do you think you'll be able to test it in 1-2 next hours? |
| 23:15.31 | cr2 | goxboxlive: are you using the spekaers now ? |
| 23:15.49 | goxboxlive | cr2: no |
| 23:15.59 | goxboxlive | should i? |
| 23:16.08 | psokolovsky | bd2_, glibc-2.5, right? |
| 23:16.23 | bd2_ | psokolovsky, -2.4 is fine. |
| 23:16.23 | cr2 | goxboxlive: switch to the speakers, and do the dump. |
| 23:16.29 | goxboxlive | cr2: The headphones where unplugged |
| 23:16.48 | goxboxlive | But i had no musuic playing. should i do that? |
| 23:16.52 | cr2 | goxboxlive: play a long sound to the speakers and do the dump |
| 23:16.56 | cr2 | yes :) |
| 23:16.57 | goxboxlive | ok |
| 23:18.32 | goxboxlive | cr2: no headphones, and playing a song: root@htcuniversal:~$ x -2 0x10000000 0x200 gives me http://pastebin.ca/257090 |
| 23:19.39 | cr2 | f9ad -> 1111.1001.1010.1101 - both unset. the driver does not powerup the speakers. |
| 23:19.46 | cr2 | dmesg ? |
| 23:21.07 | goxboxlive | cr2: dmesg -> http://pastebin.ca/257093 |
| 23:21.07 | psokolovsky | cr2, Kevin2: so, gentlemen, guranteed, I've caught it. |
| 23:21.17 | Kevin2 | psokolovsky: What did you want me to test? |
| 23:21.17 | psokolovsky | welcome back your .12 stuff ;-) |
| 23:21.31 | psokolovsky | Kevin2, sound working as it did in .12 ;-) |
| 23:21.49 | psokolovsky | Kevin2, just would like to get confirm before tagging -hh1 |
| 23:21.56 | psokolovsky | but that's in ~1hr |
| 23:22.10 | cr2 | psokolovsky: great! |
| 23:22.15 | Kevin2 | I'd need to backport all the apache stuff in order to test xandux .12, and I'm not up for that. |
| 23:22.32 | cr2 | Kevin2: ? |
| 23:23.18 | cr2 | goxboxlive: <4>Alignment trap: opieplayer2 (1309) PC=0x40a859bc Instr=0xe5917038 Address=0x001196b5 FSR 0x013 |
| 23:23.23 | goxboxlive | psokolovsky: I can try it on the universal so you can get it confirmed if you like. |
| 23:23.29 | Kevin2 | I'm confused. What do you want me to test? |
| 23:23.34 | cr2 | can you play with a progran without bugs ? |
| 23:23.41 | psokolovsky | Kevin2, on .18 ;-) I mean, get best of your .12 on contemporary thing ;-) |
| 23:23.57 | psokolovsky | and stop telling me that .12 was better ;-) |
| 23:23.58 | goxboxlive | yes i can do cat |
| 23:24.05 | psokolovsky | goxboxlive, wonderful! |
| 23:24.25 | psokolovsky | Kevin2, I commit some fix to CSV HEAD soon |
| 23:24.27 | psokolovsky | CVS |
| 23:24.28 | cr2 | Kevin2: we (on the universal) still could not reach the state of .12 ;-) |
| 23:24.51 | cr2 | but thanks to psokolovsky it is real! :D |
| 23:25.46 | cr2 | goxboxlive: i don't see the speaker_on message because of this buggy alignment messages. |
| 23:27.29 | Kevin2 | psokolovsky: Do you think the refcounting was causing the problem? |
| 23:27.47 | psokolovsky | Kevin2, nope, but we do have leakage |
| 23:28.29 | psokolovsky | So, the issue is: |
| 23:28.31 | goxboxlive | cr2: here is the output when doing cat 'file' > /dev/dsp |
| 23:28.33 | goxboxlive | <7>i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 23:28.34 | goxboxlive | <5>snd_htcuniversal_audio_set_headphone_power: 1 |
| 23:28.34 | goxboxlive | <7>i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 23:28.47 | psokolovsky | here's 2.6.12-hh3: |
| 23:28.48 | psokolovsky | static int |
| 23:28.48 | psokolovsky | snd_pxa_i2sound_hx4700_activate(void) |
| 23:28.48 | psokolovsky | { |
| 23:28.48 | psokolovsky | <PROTECTED> |
| 23:28.48 | psokolovsky | <PROTECTED> |
| 23:28.50 | psokolovsky | <PROTECTED> |
| 23:28.56 | psokolovsky | <PROTECTED> |
| 23:28.58 | psokolovsky | and look at 2.6.18 ;-) |
| 23:29.11 | psokolovsky | and I'm not sure I'll be able to fix it nicely now |
| 23:29.20 | psokolovsky | because there's crappy control flow |
| 23:29.28 | goxboxlive | cr2: And that is without headphone attached |
| 23:29.53 | psokolovsky | it's ak4641 driver powers up the chip, whereas it really needed for i2s to work properly |
| 23:30.17 | goxboxlive | cr2: and this is with the headphone attached: |
| 23:30.19 | goxboxlive | <7>i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 23:30.20 | goxboxlive | <5>snd_htcuniversal_audio_set_headphone_power: 1 |
| 23:30.20 | goxboxlive | <7>i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 23:30.45 | cr2 | goxboxlive: set_headphone_power: 1, but why are the gpios low ? |
| 23:30.48 | goxboxlive | cr2: For me it seems like we are setting the power on the headphones no mather what |
| 23:31.18 | cr2 | goxboxlive: not we. it was the same on the .12 |
| 23:31.46 | goxboxlive | But it is the headphone and not the main speakers which is turned on. |
| 23:32.46 | cr2 | hmm. i'm blind. |
| 23:32.48 | cr2 | ;) |
| 23:32.51 | goxboxlive | :-) |
| 23:33.00 | cr2 | i'll look at the code again. |
| 23:33.53 | scatterp- | goxboxlive i got that new version installed im loading it up now |
| 23:34.36 | goxboxlive | scatterp- ok, there is no more to test now. |
| 23:35.02 | scatterp- | k |
| 23:36.10 | cr2 | goxboxlive: but it's the ak4641 code that calls the function, we are providing the function pointers... |
| 23:36.34 | goxboxlive | ok |
| 23:36.41 | goxboxlive | then what ? |
| 23:37.59 | cr2 | do you see other driver messages in dmesg ? |
| 23:38.47 | cr2 | and disable the i2c debugging. |
| 23:39.40 | goxboxlive | cr2: i dont think so, but i could disable the i2c debugging |
| 23:40.09 | cr2 | there should be a codec power, codec reset and so on messages. |
| 23:40.12 | goxboxlive | i have to compile a new kernel thow, but then i can add the stuff psokolovsky wantet me to test |
| 23:40.21 | BabelOued | cr2: i made this log by script, first on and after off http://linuxbox2.free.fr/blueangel/WIFI.txt |
| 23:41.09 | scatterp- | goxboxlive i installed the new version as you said and also i restarted the sound module and run opieplayer but it was the same as the last test no sound |
| 23:41.20 | cr2 | BabelOued: you should mask the pwm0 and 1 |
| 23:41.35 | cr2 | set iasic3gpioa(11) 1 |
| 23:41.45 | cr2 | set iasic3gpioa(12) 1 |
| 23:41.52 | goxboxlive | scatterp- that sounds strange. |
| 23:41.54 | cr2 | otherwise they spam the logs |
| 23:42.04 | psokolovsky | well, pxa2xx-i2sound code is a bit messy ;-\ |
| 23:42.21 | goxboxlive | cr2: there are only headphone_power and headphone detect |
| 23:42.33 | scatterp- | i dont need to use the original htc universal headset right ? im just useing sony earphones |
| 23:42.45 | scatterp- | (working new high quality ones) |
| 23:42.52 | cr2 | goxboxlive: hmm. is the headset detect active high or low ? |
| 23:43.06 | cr2 | but it should not have changed from .12 ??? |
| 23:43.20 | cr2 | BabelOued: set igpio(85) 1 |
| 23:43.31 | cr2 | and set igpio(88) 1 |
| 23:43.46 | goxboxlive | cr2: Only high when plugged in, and low when plugged out |
| 23:43.48 | cr2 | these are the memory gpios. |
| 23:44.16 | cr2 | goxboxlive: the gpio ? |
| 23:44.46 | goxboxlive | cr2: actually, it doesnt detect the unplug part before i start to play a song |
| 23:44.52 | psokolovsky | well, now I'm not sure if I commit that code, even if it will work |
| 23:44.54 | lkcl | cr2: http://hands.com/~lkcl/hp6915/wi.i2c.cam.on.txt |
| 23:45.03 | psokolovsky | would ned first to discuss it with Giorgio |
| 23:45.31 | goxboxlive | Actually, it never detect that i am unplugging my headphones. |
| 23:45.42 | cr2 | goxboxlive: ok, this is another story. is this message coreect : snd_htcuniversal_audio_set_headphone_detect: |
| 23:46.03 | goxboxlive | <5>snd_htcuniversal_audio_set_headphone_detect: 1 |
| 23:46.12 | cr2 | is it true ? |
| 23:46.17 | goxboxlive | yes |
| 23:46.22 | cr2 | lkcl: thanks |
| 23:46.25 | lkcl | k |
| 23:46.55 | cr2 | goxboxlive: ok. the irq 'sees' only the plugin event :) i need to think about it. |
| 23:47.05 | goxboxlive | It doesnt detect that i am unplugging the headphones. So it set's the power to the headphones even thow i have just unplugged it |
| 23:47.32 | BabelOued | cr2: ok done |
| 23:48.10 | goxboxlive | cr2: And with the .12 kernel the headphones and the main speaker where playing at same time, there where no detection there either. |
| 23:48.44 | BabelOued | cr2: play again script with delay 30s for each state |
| 23:49.07 | cr2 | lkcl: what is the resolution of the cam ? the id is 0x30 , ov9650 ? |
| 23:49.22 | cr2 | BabelOued: i'll have a look. |
| 23:49.32 | lkcl | well i did see a thing in the corner saying 1280x960 which is somewhat insane |
| 23:49.52 | lkcl | no - 1280x1024. that's insane. |
| 23:49.56 | cr2 | goxboxlive: yes, the driver does not watch for the 'plugout' event. it's tricky to have REFE on asic3. |
| 23:50.34 | cr2 | lkcl: look at the specs of ov9650 |
| 23:50.48 | lkcl | yeh it's the hp photosmart program lying |
| 23:50.55 | goxboxlive | psokolovsky: Why, as long as it is working i think you should, it is just to revert it if there is going to be some issues with it. If you dont, could you mail me the driver si i can test it, i have no idea how to aply that pasted stuff. |
| 23:51.06 | goxboxlive | cr2: ok |
| 23:51.48 | lkcl | dang. Array Size-SXGA(1300 x 1028) |
| 23:53.05 | cr2 | lkcl: so it's the same on the universal. |
| 23:53.41 | cr2 | lkcl: did you use i2cdetect with A0=1 or A0=0 ? |
| 23:53.57 | lkcl | not yet |
| 23:54.09 | cr2 | i still don't understand where is your RTC. |
| 23:54.21 | cr2 | not for the cam, it's irrelevant now. |
| 23:55.14 | scatterp- | goxboxlive http://www.freeworlddialup.com/ is the sip provider i was thinking about they offer many good services like free real phone numbers direct to your phone and that kind of thing there also a linux friendly group the real good info about them is on there forums but leave it with me and ill do the diging around and get you a list of good links |
| 23:55.39 | cr2 | lkcl: on power i2c ? you had the cken dump once... |
| 23:55.41 | scatterp- | also free calls to usa some times and other places |
| 23:56.23 | lkcl | hehe |
| 23:56.24 | goxboxlive | scatterp- Ok thanks, you can also try it out with the kphone/pi aplied in the image yuo just downloaded. |
| 23:56.46 | scatterp- | i got no sound so i cant really test much |
| 23:57.05 | psokolovsky | goxboxlive, but you do patches, right? |
| 23:57.21 | goxboxlive | yes |
| 23:57.33 | psokolovsky | ok |
| 23:57.34 | cr2 | lkcl: pwi2c off |
| 23:57.51 | scatterp- | ill see what i can do and get back to you gonna take a rest for a bit exhausted |
| 23:58.03 | cr2 | what is the purpose of having an RTC which you can't access ?? |
| 23:58.14 | lkcl | cr2: that a haret command? |
| 23:58.40 | cr2 | lkcl: no. dump pxa27xcken output in the wiki |
| 23:58.54 | lkcl | oh ok. |
| 23:59.11 | lkcl | http://wiki.xda-developers.com/index.php?pagename=Ipaq6915_Research |
| 23:59.40 | cr2 | we need to convince the OVcamchip maintainer to include the ov9650 and ov6650 into the code. |
| 23:59.50 | cr2 | yes, this wiki page |