| 00:00 | Squarc | Does htc-linux have a tut on how to build (with android) ? |
| 00:01 | scope009 | anyone? |
| 00:02 | scope009 | i'm getting a connection refused error when i try to access the repository. is the server down or is the problem on my end? |
| 00:02 | Captnoord | Squarc: its on the wiki |
| 00:03 | Captnoord | and there isn't a special way to build for android |
| 00:03 | Captnoord | the main config is android compatible |
| 00:03 | Squarc | ok |
| 00:05 | Squarc | not the htc-linux.org/wiki right? |
| 00:06 | Marajin | well I think I'm going to get some sleep in |
| 00:06 | Marajin | I'ma need the energy tomorrow :p |
| 00:06 | rafyvitto | if you were using the build from june, you were running open gl stuff software renderd, not hardware renderd |
| 00:07 | scope009 | so no one is having problems accessing git.linuxtogo.org? |
| 00:07 | Squarc | the webgit; no |
| 00:08 | scope009 | no, i can access the webgit |
| 00:08 | scope009 | i'm trying to checkout the kernel with git and it says "connection refused" |
| 00:08 | Squarc | ahh |
| 00:09 | Squarc | yes I have that too right now |
| 00:09 | scope009 | okay, looks like they are having server problems |
| 00:09 | Squarc | git.linuxtogo.org[0: 188.40.83.200]: errno=Connection refused |
| 00:09 | Squarc | uhu |
| 00:09 | scope009 | is there a mirror up somewhere? |
| 00:10 | Squarc | well.. I would think they allready have that with a loadbalancer or something |
| 00:11 | Squarc | why aren't there any nightly builds or so ? |
| 00:12 | scope009 | alright, looks like i'll just have to wait for the server to come back up |
| 00:13 | Squarc | uhu |
| 00:14 | Squarc | hmm, I think the server is just overloaded |
| 00:14 | Squarc | way overloaded |
| 00:14 | Squarc | over 66% packet loss |
| 00:15 | Squarc | and super high traffic times, though the server is in germany (I'm in Holland,, right next to it) |
| 00:19 | Squarc | hmm |
| 00:19 | Squarc | that htc-msm-2.6.27 branch doesnt include android right? |
| 00:20 | Squarc | though it is based on android-msm-2.6.27 that google provided...? how does that work? |
| 00:21 | Captnoord | its simple |
| 00:21 | Captnoord | android |
| 00:21 | Captnoord | is a linux kernel |
| 00:22 | Captnoord | with a java vm on top |
| 00:22 | Captnoord | the linux kernel is generic |
| 00:22 | Captnoord | and that is what we are doing here |
| 00:22 | Captnoord | android and the java vm |
| 00:22 | Captnoord | isn't something we need to develop |
| 00:22 | Squarc | is altering the linux kernel (layer) so that it works on htc devices |
| 00:22 | Squarc | I know |
| 00:22 | Squarc | and the android just runs on top of the linux kernel |
| 00:22 | Squarc | when the linux kernel works, android does |
| 00:23 | Captnoord | http://www.connect-utb.com/ |
| 00:23 | Captnoord | go there |
| 00:23 | Captnoord | and download the latest packet |
| 00:23 | Captnoord | for your phobe |
| 00:23 | Captnoord | phone |
| 00:23 | Squarc | is that the one with new sound included? |
| 00:23 | Captnoord | nope |
| 00:23 | Captnoord | not yet |
| 00:23 | Captnoord | that one you have to build yourself |
| 00:23 | Squarc | then I allready have that one :p |
| 00:23 | Squarc | yes |
| 00:23 | Captnoord | your running windows? |
| 00:24 | Captnoord | you can try building it trough cygwin |
| 00:24 | Captnoord | but |
| 00:24 | Captnoord | I recomment it building on linux |
| 00:24 | Captnoord | I use ubuntu |
| 00:24 | Squarc | thats why I was wondering if that branch includes the android packages or if its only the linux kernel (I'm guessing the last one, and then Im wondering how to add android to the img that I need to run on HaReT) |
| 00:24 | Squarc | I have Kubuntu for over a year.. and I wont load Vista anymore sorry |
| 00:24 | Squarc | so I'm going to use GCC ;) |
| 00:24 | Captnoord | hehe |
| 00:24 | Captnoord | nah |
| 00:24 | Captnoord | I need to hit the bed |
| 00:25 | Captnoord | so |
| 00:25 | Captnoord | talk to ya later |
| 00:25 | Squarc | ok :) |
| 00:25 | Captnoord | google me if you need to find me |
| 00:25 | Squarc | well I'll find out how that android stuff is to be added later :0 |
| 00:25 | Squarc | ok |
| 00:25 | Squarc | cya :) thx |
| 00:32 | MrPippy | i'm trying to boot the current connect-utb build on diamond, but it keeps having trouble with loop and mounting /system |
| 00:32 | MrPippy | i have cramfs built into my kernel |
| 00:58 | rafyvitto | does anyone have the devtools.apk? |
| 01:28 | mdrobnak | Hey everyone |
| 01:33 | rafyvitto | sup mate |
| 01:38 | mdrobnak | I see there was a lot of productive chat ealier, lol |
| 01:38 | mdrobnak | :-) |
| 01:39 | mdrobnak | I'm trying to look at my initrd to set up some swap, to see if the extra ram allievates the crashing problems. |
| 01:40 | rafyvitto | do you ahve swap enable rootfs? |
| 01:40 | rafyvitto | have* |
| 01:40 | mdrobnak | I'm working on that. |
| 01:40 | rafyvitto | great =D |
| 01:41 | rafyvitto | this 2-4mb ram isnt working xD |
| 01:41 | mdrobnak | Oh, this is nice, the initrd, just sets up the rootfs.img file |
| 01:41 | mdrobnak | which is good |
| 01:41 | mdrobnak | easier to modify |
| 01:41 | rafyvitto | oh nice |
| 01:42 | rafyvitto | hry did you tryed the new overclock tool from nue2chem? |
| 01:42 | rafyvitto | hey* |
| 01:42 | mdrobnak | Nope. |
| 01:42 | mdrobnak | I'm runing EnergyROM 3.0 WinMo 6.5, and it's fast enough for me, honestly. |
| 01:42 | rafyvitto | i tried it on my fuze, it freezes on AC and reboots on battery xD, just like nue2chem told me xD |
| 01:42 | mdrobnak | lol |
| 01:43 | rafyvitto | its working for most touch por users |
| 01:43 | mdrobnak | then why'd you try? |
| 01:43 | mdrobnak | ah |
| 01:43 | rafyvitto | =D |
| 01:43 | rafyvitto | how fuck uo it is to have two phones, which are the same, but when you go into there programing THERE COMPLETLY DIFRENT =D |
| 01:44 | mdrobnak | Well, it depends on if it's the CDMA or GSM varient...but I would think the RAPH100 and 110 are very close. |
| 01:44 | rafyvitto | enginiering etc |
| 01:44 | rafyvitto | yeh well att decided to have there own little touch pro cake |
| 01:44 | rafyvitto | -__- |
| 01:45 | rafyvitto | i should have bought a touch pro >< |
| 01:45 | rafyvitto | but.. coulndt way to much, compare to my fuze |
| 01:45 | rafyvitto | i paid 250$ for it on ebay |
| 01:46 | rafyvitto | a touch pro in similar conditions was like 350$ |
| 01:47 | mdrobnak | I bought the fuze new for $350 on two year agreement... |
| 01:47 | mdrobnak | so, 250 is good |
| 01:47 | mdrobnak | hehe |
| 01:47 | rafyvitto | mine's used btw lol |
| 01:47 | mdrobnak | I figured tht |
| 01:48 | rafyvitto | but its in almost mint condition |
| 01:48 | rafyvitto | few little nick here and there |
| 01:48 | rafyvitto | but hey, i love my fuze heheh |
| 01:49 | rafyvitto | such a nice looking device |
| 01:49 | rafyvitto | and the hardware for me is perfect |
| 01:49 | mdrobnak | It's nice, a little too dense though. |
| 01:49 | mdrobnak | And I wish the screen were bigger. |
| 01:49 | mdrobnak | But the keyboard is good. |
| 01:49 | rafyvitto | i love the onboard gpu |
| 01:50 | rafyvitto | i've got almost all the open gl ES games on my sd card |
| 01:50 | rafyvitto | q3ce,thp2 etc |
| 01:52 | mdrobnak | hmm |
| 01:52 | mdrobnak | Ok, found the issues with the modules it tries to load |
| 01:53 | rafyvitto | swap's not working? |
| 01:53 | mdrobnak | /bin/ramzswap on rootfs |
| 01:53 | mdrobnak | it tries to load lzo_compress / decompress / xvmalloc and ramzswap.ko |
| 01:53 | mdrobnak | but that's from the Vogue original 2.6.25 branch |
| 01:53 | rafyvitto | ic |
| 01:54 | mdrobnak | if you wish to do what I'm going to.. |
| 01:54 | mdrobnak | do this: |
| 01:54 | mdrobnak | mount -o loop rootfs.img somedir |
| 01:54 | mdrobnak | vi somedir/init |
| 01:54 | rafyvitto | im not on my linux box xD |
| 01:54 | mdrobnak | replace the line that says /bin/ramzswap and change it to createswap() |
| 01:55 | mdrobnak | Well, now it's in the logs :-) |
| 01:56 | rafyvitto | im lazy , when it comes to switching os's lol, should have use sunvm |
| 01:56 | rafyvitto | for ubuntu, thats what i got |
| 01:56 | mdrobnak | You mean VirtualBox? |
| 01:57 | rafyvitto | yyeh |
| 01:57 | rafyvitto | im using wubi, for ubuntu, but im just lazy dont feel like restaring the pc |
| 01:58 | mdrobnak | heh |
| 01:58 | rafyvitto | have you tryed cygwin? |
| 01:59 | mdrobnak | Yeah, always found it a pain to set up. |
| 02:00 | mdrobnak | I use Ubuntu at work, MacOS at home, and Linux and Windows in the datacenter. |
| 02:00 | rafyvitto | oh ic |
| 02:00 | rafyvitto | on my laptop i have win7,winvista and ubuuntu |
| 02:01 | mdrobnak | On my work laptop I have Debian 5, Windows XP, and MacOS 10.4, and at one point OpenSolaris 2008.05 lol |
| 02:02 | mdrobnak | Ok, trying my updated rootfs. Lets see what happens. |
| 02:04 | rafyvitto | nice lets see that beast of ram RAWRR! |
| 02:04 | mdrobnak | Yay! 64MB of swap now. |
| 02:04 | rafyvitto | OMFG |
| 02:04 | rafyvitto | NICE |
| 02:04 | mdrobnak | Wow and it's responsive now lol |
| 02:04 | mdrobnak | lol ok manybe not |
| 02:04 | rafyvitto | i want to try it xD |
| 02:05 | rafyvitto | do the apps crash? |
| 02:05 | rafyvitto | as often? |
| 02:06 | mdrobnak | Hmm. |
| 02:06 | mdrobnak | Not really using swap. |
| 02:06 | rafyvitto | great! |
| 02:06 | mdrobnak | lol don't hit the camera app by accident |
| 02:07 | mdrobnak | It's better I think |
| 02:08 | mdrobnak | Gmail loads wtihout issue, haven't seen the connection go down yet. |
| 02:08 | rafyvitto | is the home proccess crashing? |
| 02:09 | mdrobnak | It gave me a force quit or wait once. |
| 02:09 | mdrobnak | But it's a little better since. |
| 02:09 | mdrobnak | I killed the clock widget. |
| 02:10 | mdrobnak | Ok, now I will try the phone killer. |
| 02:10 | mdrobnak | Shazam |
| 02:10 | mdrobnak | lol |
| 02:10 | rafyvitto | lol |
| 02:10 | mdrobnak | Shazam would hard freeze the phone, every time. |
| 02:10 | mdrobnak | I've got logcat running |
| 02:11 | mdrobnak | Crah |
| 02:11 | mdrobnak | *Crash |
| 02:11 | mdrobnak | lol |
| 02:12 | mdrobnak | W/AudioRecord( 117): obtainBuffer timed out (is the CPU pegged?) user=00000000, server=00000000 |
| 02:12 | mdrobnak | That's the last thing I see |
| 02:13 | mdrobnak | I guess audio in a call vs audio recording isn't quite the same :-) |
| 02:13 | mdrobnak | Anyone know how to get the ramconsole info after a crash? |
| 02:20 | rafyvitto | away for a bit watching lost =D |
| 02:30 | druidu | mdrobnak: you configured at address 0xe0000 ? |
| 02:30 | druidu | if so, do this in haret: pwf log.txt 0xe0000 0x20000 |
| 02:31 | druidu | first 13 bytes are crap (console header), rest is the console log |
| 02:35 | mdrobnak | druidu: Whatever the default is. |
| 02:35 | mdrobnak | Will that work with WinMo 6.5? |
| 02:38 | mdrobnak | And how can I make the thing go to swap more easily? |
| 02:38 | mdrobnak | Change the /proc/sys/vm/swappiness ? It was set to 20 |
| 02:38 | mdrobnak | I tried 40 |
| 02:38 | mdrobnak | didn't help much |
| 02:47 | mdrobnak | damn. |
| 02:47 | mdrobnak | RIL crashed once |
| 02:47 | mdrobnak | I wish I could tell what was going on |
| 03:00 | mdrobnak | Am I going to have to get the whole android src tree just to rebuild the ril? |
| 03:06 | dzo | hi, no just the libs should be enough. |
| 03:06 | dzo | but i haven't tried it. |
| 03:06 | dzo | mdrobnak: you know logcat -b radio will tell you what it's doing. |
| 03:07 | mdrobnak | doh |
| 03:07 | mdrobnak | no. |
| 03:07 | mdrobnak | lol |
| 03:07 | mdrobnak | Oh, that's nice. |
| 03:07 | mdrobnak | lol |
| 03:09 | mdrobnak | I/RILJ ( 186): 'rild' socket closed |
| 03:09 | mdrobnak | I/RILJ ( 186): java.io.IOException: Connection reset by peer |
| 03:09 | mdrobnak | I/RILJ ( 186): at android.net.LocalSocketImpl.readba_native(Native Method) |
| 03:09 | mdrobnak | I/RILJ ( 186): at android.net.LocalSocketImpl.access$400(LocalSocketImpl.java:29) |
| 03:09 | mdrobnak | I/RILJ ( 186): at android.net.LocalSocketImpl$SocketInputStream.read(LocalSocketImpl.java:92) |
| 03:09 | mdrobnak | I/RILJ ( 186): at com.android.internal.telephony.gsm.RIL.readRilMessage(RIL.java:397) |
| 03:09 | mdrobnak | I/RILJ ( 186): at com.android.internal.telephony.gsm.RIL.access$200(RIL.java:186) |
| 03:09 | mdrobnak | I/RILJ ( 186): at com.android.internal.telephony.gsm.RIL$RILReceiver.run(RIL.java:499) |
| 03:09 | mdrobnak | I/RILJ ( 186): at java.lang.Thread.run(Thread.java:1058) |
| 03:09 | mdrobnak | I/RILJ ( 186): Disconnected from 'rild' socket |
| 03:09 | mdrobnak | Why is it asking fr my IMEI from the modem??? |
| 03:10 | druidu | mdrobnak: if you have latest git kernel and diamond, it's ok, otherwise most likely it won't survive reboot |
| 03:10 | dzo | that's the java code crashing after tthe lib dies. |
| 03:10 | mdrobnak | Ah. |
| 03:10 | mdrobnak | There's so much stuff going on, it's hard to follow everything.. |
| 03:11 | dzo | it does ask for the imei but i don't think thats a problem. |
| 03:11 | dzo | it's probably in the network registration code somewhere. |
| 03:11 | mdrobnak | looks that way |
| 03:12 | dzo | probably the format from the modem is different, what does it do for COPS |
| 03:13 | druidu | dzo: what mddi client does diamond have? toshiba? |
| 03:13 | mdrobnak | D/AT ( 734): AT> AT+CGACT? |
| 03:13 | mdrobnak | I/GSM ( 186): PDP connection has dropped (active=false case). Reconnecting |
| 03:13 | mdrobnak | D/GSM ( 186): [DataConnectionTracker] Clean up connection due to null |
| 03:13 | mdrobnak | D/RILJ ( 186): [0098]> DEACTIVATE_DEFAULT_PDP 1 |
| 03:14 | dzo | yes |
| 03:14 | mdrobnak | I see often. |
| 03:14 | druidu | mddi/mdp/fb stuff is almost completely rewritten in 2.6.29 / aurora branch |
| 03:14 | dzo | none of the panel init works though. |
| 03:14 | druidu | much more complex |
| 03:15 | druidu | the vsync gpio, do we know it? |
| 03:15 | druidu | 97? |
| 03:15 | dzo | yes, vsync works fine. |
| 03:15 | mdrobnak | dzo: What's the best way to log this stuff...Here's what I see...No service, then signal, then PPP, then no service...over and over. Doesn't seem to be coming back this time. |
| 03:16 | dzo | logcat -b radio |
| 03:16 | mdrobnak | I got that. :-) |
| 03:16 | mdrobnak | It flies by |
| 03:16 | dzo | + logcat for the crash dumps |
| 03:16 | mdrobnak | ok. |
| 03:16 | dzo | logcat -b radio > logfile |
| 03:18 | mdrobnak | W/pppd ( 1069): LCP: timeout sending Config-Requests |
| 03:18 | dzo | druidu: my changes to the clock make the display much better, before it was taking 33ms to update, now it's < 16 and so happens in 1 vsync time. |
| 03:18 | mdrobnak | Gonna try going to airplane mode and back. |
| 03:19 | dzo | mdrobnak: i'd be interested to see where it crashes because it never crashes on kaiser. |
| 03:19 | mdrobnak | dzo: Well at first I thought it was because of the RAM. But I have 64 MB of swap, with no low memory killer...and it's still crashing. |
| 03:19 | druidu | git still not up huh? |
| 03:20 | dzo | swap is not really a good idea, will kill your sd card. |
| 03:20 | mdrobnak | D/RILJ ( 186): [0141]< REQUEST_GET_NEIGHBORING_CELL_IDS error: com.android.internal.telephony.gsm.CommandException: REQUEST_NOT_SUPPORTED |
| 03:20 | mdrobnak | Well, only will do it until the RAM banks are fixed. :_) |
| 03:20 | mdrobnak | :-) |
| 03:21 | dzo | why doesn't bank 2 work on raph? |
| 03:22 | dzo | druidu: git is up but the permissions are wrong, can't push. |
| 03:22 | mdrobnak | No idea. If I boot with anything more then 76 M, it doesnt boot to android |
| 03:22 | mdrobnak | Wow that crashed fast. |
| 03:28 | mdrobnak | dzo: Will the stack trace help? |
| 03:30 | dzo | perhaps but it's probably easier to add some more debugging to the code. |
| 03:32 | mdrobnak | D/RILJ ( 186): [UNSL]< UNSOL_RESPONSE_NETWORK_STATE_CHANGED |
| 03:32 | mdrobnak | D/RILJ ( 186): [0182]> OPERATOR |
| 03:32 | mdrobnak | That's interesting, no? It's probably a switch from GSM to UMTS |
| 03:33 | mdrobnak | I'm gonna capture once more, one last time. I don't see anything which screams to me here it is, then I'm gonna pastebin both logs. I'm running -c on both before I start to clear it. |
| 03:37 | mdrobnak | D/AT ( 2133): AT> AT+COPS? |
| 03:37 | mdrobnak | D/AT ( 2133): AT< +CME ERROR: 30 |
| 03:41 | mdrobnak | Hmm. Radio log too large, lol |
| 03:41 | mdrobnak | I'm annoyed that it doesn't have timestamps on anything. |
| 03:42 | mdrobnak | I'm gonna shut up now, you can have it add time. |
| 03:42 | mdrobnak | Ok, one last time, with time stamps. |
| 03:44 | mdrobnak | Ok, got it, with time stamps. :-) |
| 03:45 | tmzt | hey, can anyone help with something? |
| 03:46 | mdrobnak | tmzt: Depends. Trying to debug RIL over here...seems it's getting unsolicited network info and crashing |
| 03:47 | dzo | I'm going for lunch, back later. |
| 03:48 | mdrobnak | dzo: http://pastebin.ca/1559196 Radio Log |
| 03:49 | mdrobnak | dzo: http://pastebin.ca/1559199 Full Log |
| 03:49 | mdrobnak | Ttyl. I hope that helps to give a clue. |
| 03:49 | tmzt | people.openezx.org/tmzt/green7.S |
| 03:49 | tmzt | based on htcfb but it's not working |
| 03:51 | mdrobnak | lol data's up long enough for Babbler to work. |
| 03:52 | mdrobnak | tmzt: Sorry, that's way out of my expertise. |
| 03:52 | tmzt | yeah, I've been working on it all morning |
| 03:52 | tmzt | and it doesn't even vibrate |
| 04:03 | dzo | tmzt: is mmu off when you run this? |
| 04:03 | tmzt | it's started from haret, so I think so |
| 04:03 | tmzt | I'm trying to verfify that |
| 04:03 | tmzt | since haret used to leave it with an identity mapping |
| 04:04 | dzo | if so it won't work because the device memory needs a special mapping. |
| 04:04 | tmzt | ah, ok |
| 04:04 | tmzt | so I want to use the virtual address? |
| 04:04 | dzo | you need to set up the peripheral port remap register. |
| 04:05 | dzo | google it and you'll find the arm docs. |
| 04:06 | tmzt | this? http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.faqs/ka4167.html |
| 04:06 | dzo | i tried it once (on non A devices) and never got it to work but it may be ok on A devices. |
| 04:06 | dzo | yes |
| 04:07 | mdrobnak | dzo: requestOperator must not return error when radio is on This seems to be a problem |
| 04:07 | dzo | yes i looked at your logs, there's a lot happening! |
| 04:07 | mdrobnak | LOL |
| 04:08 | mdrobnak | Weird thing is, once it decides to calm down, it'll hold a connection for a while without a problem. |
| 04:09 | tmzt | dzo: well ok, how do I find the address? I guess in linux |
| 04:10 | dzo | looks like in requestoperator, that long at command replies with OK and that stuffs it up. |
| 04:10 | dzo | tmzt: which address? |
| 04:10 | tmzt | the register itself |
| 04:11 | dzo | you just need to set the pprr to cover any physical addresses where devices are, in your case the mdp regs etc. |
| 04:12 | dzo | also the csr otherwise your proc comm is never called. |
| 04:12 | mdrobnak | dzo: Well, looks like this one isn't as important: CME ERROR: 30 -- which means -- No network service |
| 04:13 | dzo | in fact just tell it to cover all the io regions. |
| 04:13 | mdrobnak | http://www.activexperts.com/xmstoolkit/sms/gsmerrorcodes/ |
| 04:14 | dzo | mdrobnak: try sending the AT+COPS=3,0;..... command to /dev/smd0 and see what it does. |
| 04:15 | dzo | bye for now. |
| 04:15 | mdrobnak | ok |
| 04:15 | mdrobnak | later |
| 04:15 | mdrobnak | Thanks for the help |
| 04:17 | tmzt | http://forum.xda-developers.com/member.php?s=40047e5d2ba0ce39d326bd47b5327f8c&u=1267334 |
| 04:18 | tmzt | druidu: you here? |
| 04:18 | tmzt | druidu: I'm trying to port your htc_fb_con to asm (booted from haret) |
| 04:18 | tmzt | people.openezx.org/tmzt/green7.S |
| 04:18 | tmzt | dzo says I need the periphral remap to work but I can't find where that would be done |
| 04:19 | tmzt | NetRipper: or you? :) |
| 04:20 | tmzt | I know this is old, but it's actually AMSS/not the kernel in case that was never cleared up |
| 04:21 | tmzt | Open Kernel Labs (OK Labs) [...] announced that the company's OKL4 embedded hypervisor is employed on the Qualcomm chipset inside the market's first-ever commercial Android-powered handset. |
| 04:22 | mdrobnak | tmzt: So what's the problem with the htc_fb_con as-is? |
| 04:22 | tmzt | nothing, I'm just trying to build a test program for a bootloader |
| 04:22 | tmzt | and see if I can get the graphics to update from asm |
| 04:23 | tmzt | although it has dma2_cfg including *AHB which is not used (and is 0) |
| 04:23 | tmzt | but it's confusing because MDDI is atually used instead |
| 04:25 | tmzt | we're the only ones! http://www.google.com/search?q=msm+%22peripheral+remap%22&ie=utf-8&oe=utf-8&aq=t&rls=com.ubuntu:en-US:official&client=firefox-a |
| 04:26 | tmzt | 20:54.16NetRippersirox, not sure, but i'd think so... haret simply jumps to the memory address of the kernel... so if you make your own code and set the binary in haret as kernel, it'll probably jump to it |
| 04:26 | tmzt | 20:54.48NetRippersirox, note that the MMU will be off, you will need to remap the gpios you want to use using the peripheral remap register |
| 04:26 | tmzt | 20:54.55 NetRipper before you can use gpios |
| 04:26 | tmzt | if you are after the flashlight gpio you can make it available when MMU is off using a haret command: SET CP(15, 0, 15, 2, 4) 0xa920000b |
| 04:27 | tmzt | (NetRipper) |
| 04:27 | tmzt | so is this a mask? |
| 04:27 | tmzt | http://www.heyrick.co.uk/assembler/coprocmnd.html |
| 04:36 | mdrobnak | tmzt: lol @ google result |
| 04:37 | mdrobnak | I was messing with the rild stuff |
| 04:38 | druidu | tmzt: here |
| 04:38 | druidu | ok |
| 04:38 | druidu | so you only need one tiny thing to do |
| 04:39 | druidu | to make haret update the screen while writing stuff |
| 04:39 | druidu | htc_fb_console_update() |
| 04:39 | druidu | and actually, just the dma part |
| 04:40 | druidu | you have 8 writel commands, that's all |
| 04:40 | dzo | tmzt: if you set that CP reg to cover the whole i/o region it should work. |
| 04:40 | druidu | just remember you have to pass a physical address here: writel(fbram, MSM_MDP_BASE + MDP_FULL_BYPASS_WORD34); |
| 04:41 | dzo | druidu: he's trying to do it with mmu off so needs to set up the i/o memory first. |
| 04:41 | druidu | constant for that file are in drivers/video/msm/mdp_hw.h |
| 04:41 | tmzt | any idea how to do that? |
| 04:41 | tmzt | I can only find that I can, not how |
| 04:41 | druidu | check arm11 datasheet |
| 04:41 | tmzt | 0xa4 I think |
| 04:42 | druidu | you have it? |
| 04:42 | druidu | DDI0211J_arm1136_r1p5_trm.pdf |
| 04:42 | dzo | try set cp in haret before booting. |
| 04:42 | tmzt | I see it |
| 04:42 | tmzt | ok, downloaded |
| 04:42 | tmzt | thanks |
| 04:42 | druidu | you just need one identity mapping |
| 04:42 | druidu | for the region you're writing to |
| 04:43 | druidu | I had the code for that, once, when me & NetRipper were trying to blindly boot the kernel |
| 04:43 | tmzt | dzo: is there a reason to do it that way? (before booting) |
| 04:43 | tmzt | druidu: did you get proc_comm to work? |
| 04:43 | druidu | set cp? |
| 04:44 | tmzt | yeah |
| 04:44 | tmzt | oh, set cp |
| 04:44 | tmzt | a command |
| 04:44 | dzo | just to test, otherwise write to the pprr in your code. |
| 04:44 | tmzt | I get it |
| 04:44 | tmzt | ok, I'll look at trm |
| 04:44 | dzo | with mrc |
| 04:44 | druidu | hmmm, proc_comm works I guess... what do you mean exactly? |
| 04:44 | tmzt | just vibrate |
| 04:44 | tmzt | I'm sure the code is running, it's just not doing anything right now |
| 04:44 | tmzt | because of the memory mapping |
| 04:45 | druidu | btw |
| 04:45 | druidu | you can also check out the arm bootup code |
| 04:45 | dzo | tmzt: i don't think you can use an identity page, mmu has to be off to boot correctly. |
| 04:45 | tmzt | yeah, so pprr |
| 04:46 | AstainHellbring | hows it going? |
| 04:46 | tmzt | I misunderstood your 'set cp' thing, I thought you meant I needed to do it in haret |
| 04:46 | druidu | arch/arm/kernel/head.S |
| 04:46 | tmzt | yeah, is the pprr stuff there? I didn't see it |
| 04:47 | tmzt | or is haret currently setting that up for us, since I don't see any msm specific code in early kernel |
| 04:47 | tmzt | I guess I just need the address we use if that's the case |
| 04:48 | dzo | no, pprr isn't used at all, you have to set it up using a coprocessor instruction. |
| 04:49 | tmzt | dzo: any reason to use svc mode? |
| 04:49 | druidu | Peripheral Port Memory Remap Register |
| 04:49 | tmzt | svc32 |
| 04:49 | druidu | what exactly do you want to access anyway? |
| 04:49 | tmzt | fb and proc_comm (for vibrate) |
| 04:49 | tmzt | just trying to see how to write a bootloader and need a console for feedback |
| 04:49 | tmzt | so I'm converting yours |
| 04:50 | dzo | tmzt: you will be in svc mode already i think. |
| 04:50 | tmzt | ok |
| 04:50 | druidu | hey |
| 04:50 | tmzt | no need to set that then |
| 04:50 | druidu | you should also check out |
| 04:51 | druidu | http://android.git.kernel.org/?p=platform/bootable/bootloader/legacy.git;a=summary |
| 04:51 | tmzt | yeah |
| 04:51 | druidu | they have a few libs useful for writing bootloaders |
| 04:52 | tmzt | but it's c and I haven't figured out how to set up my stack yet |
| 04:52 | tmzt | this is the first step |
| 04:52 | druidu | including basic mddi/mdp support |
| 04:52 | tmzt | right |
| 04:52 | tmzt | I saw that |
| 04:52 | druidu | mddi_console |
| 04:54 | druidu | so, how do you get your bootloader to load with no hardware initialized? or you just load the bootloader using haret then bootloader should load linux? :) |
| 04:55 | cr2 | dzo: you are overwriting the spl by your own resume vector ? |
| 04:57 | dzo | yes, that's needed for power collapse. |
| 04:57 | cr2 | dzo: hehe. 512k more of free ram :) |
| 04:57 | dzo | doesn't work yet though. it works perfectly once then doesn't wake properly after that. |
| 04:58 | cr2 | ok |
| 04:58 | mdrobnak | dzo: Still working on that test.. |
| 04:59 | mdrobnak | dzo: Also, would the radio log from a G1 be useful at all? |
| 04:59 | dzo | which test? and probably not. |
| 04:59 | mdrobnak | the operator related commands |
| 05:00 | cr2 | i need SPI_WRITE16 for my lcd panel. working on that now. |
| 05:01 | cr2 | mdrobnak: you can log all AT commands in wince, or, dump AT fifos, or, run strings on ril.dll |
| 05:02 | mdrobnak | cr2: good idea. |
| 05:02 | mdrobnak | cr2: We're looking at why some are failing on the RAPH using the Vogue RIL |
| 05:03 | cr2 | mdrobnak: check the init string after ATZ |
| 05:03 | mdrobnak | cr2: In WINCE or AND? |
| 05:03 | cr2 | both |
| 05:03 | cr2 | if they match, and what is different |
| 05:03 | mdrobnak | They do it one command at a time here. it's long. |
| 05:04 | mdrobnak | I have pastebins earlier |
| 05:04 | cr2 | the init is one command |
| 05:04 | mdrobnak | http://pastebin.ca/1559196 Radio Log |
| 05:04 | mdrobnak | It's just ATZV1 |
| 05:05 | cr2 | i can check my dumps |
| 05:07 | cr2 | ATZE0S0=0Q0V0X3&C1&D1 |
| 05:07 | cr2 | AT+GTKC=2;+CMEE=1;+CRC=1;+CR=1;+CREG=1;+FCLASS=0;+CMGF=0;+CSCS="HEX";+CLIP=1;+COLP=0;+CCWA= |
| 05:07 | cr2 | 1;+CMUT=0;+CLIR=0;+CNMI=1,2,2,2,0;+CGREG=1;+CUSD=1;+CPPP=1 |
| 05:07 | cr2 | this is the init string |
| 05:07 | mdrobnak | That's for which model? |
| 05:07 | cr2 | raph100 |
| 05:07 | cr2 | and then comes a lot of weird stuff |
| 05:07 | mdrobnak | Ok, here's the issue I think. |
| 05:08 | cr2 | like AT+HTCmaskW1=1048575,162161 |
| 05:08 | mdrobnak | With the way it's set up now.. |
| 05:08 | mdrobnak | # echo -e "AT+COPS=3,0\r\n" > /dev/smd0 |
| 05:08 | mdrobnak | # echo -e "AT+COPS?\r\n" >/dev/smd0 |
| 05:08 | cr2 | AT+HTCNV=1,12,6 |
| 05:08 | mdrobnak | OK |
| 05:08 | mdrobnak | OK |
| 05:08 | mdrobnak | which isn't the right answers |
| 05:08 | mdrobnak | lol |
| 05:08 | cr2 | AT@BRIC=0 |
| 05:09 | cr2 | AT+BANDSET=0 |
| 05:09 | cr2 | AT@HTCDORMANCYSET=3 |
| 05:09 | cr2 | AT@HTCPDPFD=0 |
| 05:09 | cr2 | AT+COPS=3,0;+COPS? |
| 05:09 | mdrobnak | I'll look, but a lot of that is similar. |
| 05:09 | cr2 | AT+CPBS? |
| 05:09 | cr2 | AT+CSCA? |
| 05:10 | cr2 | AT+COPS=3,2;+COPS? |
| 05:10 | mdrobnak | Sometimes I get a +CME ERROR:30 |
| 05:10 | mdrobnak | Which means network not available |
| 05:11 | cr2 | why not |
| 05:11 | mdrobnak | Not sure. |
| 05:12 | cr2 | with the same init, the performance should be the same like in wince |
| 05:14 | cr2 | ok, done with the log. |
| 05:14 | cr2 | checking g1 code |
| 05:15 | mdrobnak | Ok, I put it into stupid mode for a min by accident.. |
| 05:15 | mdrobnak | Trying the +COPS again |
| 05:16 | mdrobnak | Ok, that worked that time. |
| 05:16 | cr2 | 131 #define SPI_WRITE(reg, val) \ |
| 05:16 | cr2 | 132 { SSITX, 0x00010000 | (((reg) & 0xff) << 8) | ((val) & 0xff) }, \ |
| 05:17 | cr2 | ok, a bit more involved for 16bit here. |
| 05:18 | cr2 | SSICTL=0x170 |
| 05:19 | cr2 | SSITX=0x80000|0x10 |
| 05:19 | cr2 | SSITX= 0x00010000 | ((reg) & 0xffff) |
| 05:20 | cr2 | SSICTL=0x172 |
| 05:20 | cr2 | SSICTL=0x170 |
| 05:20 | cr2 | SSITX=0x80000|0x12 |
| 05:20 | cr2 | SSITX= 0x00010000 | ((val) & 0xffff) |
| 05:20 | cr2 | SSICTL=0x172 |
| 05:21 | cr2 | "a bit more" ;) |
| 05:25 | tmzt | for lcd init? |
| 05:25 | tmzt | dzo: did you fix the length of the lcm init table? you had the right length but ARRAY_SIZE was of the wrong array |
| 05:26 | dzo | tmzt: that may explain why it doesn't work, thanks. |
| 05:27 | tmzt | I think I missed you before |
| 05:27 | tmzt | why is it lcm though? you don't know the manufacture? |
| 05:28 | cr2 | tmzt: for hitachi lcd |
| 05:28 | tmzt | also, I assume the registers are the same as the others, should we clean that up to eliminate the magic value feel? |
| 05:28 | cr2 | tmzt: for sharp (druidu) it's different |
| 05:28 | cr2 | ? |
| 05:28 | tmzt | I mean add the register names |
| 05:28 | tmzt | instead of just the offsets |
| 05:28 | cr2 | it's the panel setup |
| 05:28 | tmzt | I know |
| 05:29 | tmzt | lcm is lcd control module |
| 05:29 | tmzt | or similar |
| 05:29 | cr2 | SSICTL and SSITX are already there |
| 05:29 | tmzt | no, the registers on the client chip |
| 05:29 | tmzt | look at the other tables |
| 05:29 | cr2 | only toshiba MDDI client registers are "known" |
| 05:29 | tmzt | ah |
| 05:30 | cr2 | and these are different registers on the panel itself |
| 05:30 | tmzt | well there's certainly some overlap, like the last entry is START I think |
| 05:30 | tmzt | ok, I see |
| 05:30 | cr2 | where ? |
| 05:30 | tmzt | we can worry about that later |
| 05:30 | tmzt | let me look |
| 05:31 | cr2 | omg. toppoly init even reads from the panel |
| 05:31 | tmzt | #define START (LCD_CONTROL_BLOCK_BASE|0x08) |
| 05:31 | cr2 | but it was the same already on n560 |
| 05:31 | tmzt | yeah, and reading the gpio block for panel id |
| 05:31 | tmzt | I can assume that might break somehow |
| 05:31 | cr2 | tmzt: these are toshiba MDDI registers |
| 05:31 | cr2 | have nothing to do with SPI panel init. |
| 05:31 | tmzt | lcm init |
| 05:31 | tmzt | {0x00110008,0x00000001}, |
| 05:32 | tmzt | 2nd from bottom in board-htcraphael-panel.c |
| 05:32 | tmzt | any reason not to say that's START? |
| 05:32 | cr2 | this table is not correct anyway |
| 05:32 | tmzt | not sure, was just trying to catch up on the patches while I was offline this morning |
| 05:33 | cr2 | it will be replaced soon. also with readable names |
| 05:33 | tmzt | ok |
| 05:33 | tmzt | also, if this is diam stuff should it be in diam file? |
| 05:33 | tmzt | I've also been looking at spliting the cdma fixup, the current one doesn't work on my phone |
| 05:33 | tmzt | which I think needs us to use the mtype I registered |
| 05:33 | cr2 | is there board-diam-panel.c ? |
| 05:33 | tmzt | no |
| 05:34 | tmzt | until there's a proper lcdc class in linux of course |
| 05:34 | cr2 | lcdc is a different beast on msm |
| 05:34 | tmzt | yes, but the non-core part of the client chips are similar to other devices |
| 05:34 | tmzt | so it's like a remote lcdc |
| 05:35 | cr2 | cdma (and blac100) has epson MDDI chip, the init will be completely different anyway. |
| 05:35 | tmzt | we need to remove AHB from the dma configs for mddi, it's either AHB or MDDI |
| 05:35 | tmzt | not both |
| 05:35 | tmzt | yeah, but I don't have pm yet anyway |
| 05:35 | tmzt | we will need it |
| 05:35 | cr2 | which AHB dma config ? |
| 05:35 | tmzt | and so while the bootloader |
| 05:35 | tmzt | in druidu console, but I assume in msm_fb also |
| 05:36 | cr2 | i'd like to get the codeaurora msmfb |
| 05:36 | tmzt | yeah |
| 05:36 | tmzt | hah |
| 05:36 | tmzt | he's back |
| 05:36 | cr2 | g1 msmfb seems to be crippled |
| 05:36 | tmzt | I want to look at using drm and it's mode setting |
| 05:36 | cr2 | does not support tvout and dma3 |
| 05:37 | druidu | shit, my external seagate drive fell of the table, on the carpet... wasn't even a big fall... now it's making beeping sounds and no sign of it on usb :-s |
| 05:37 | |Jason8|Raph| | pwned. |
| 05:37 | tmzt | I think the way google did the page flipping might be unnessacary |
| 05:37 | cr2 | hehe |
| 05:37 | tmzt | I thought it was a requirement of the hardware but I don't think so anymore |
| 05:37 | tmzt | and X doesn't need/support it |
| 05:37 | cr2 | ok |
| 05:38 | tmzt | it works for them, but they use 3d anyway |
| 05:38 | cr2 | druidu: are there raph/diam with toppoly panel ? |
| 05:38 | tmzt | this will make our transition to 3d eaiser, either by wrapping/trapping syscalls from the htc driver (and mapping them to drm instead of pmem) |
| 05:38 | tmzt | or by getting the basic 2d support for the 3d gpu to work |
| 05:38 | cr2 | ok |
| 05:39 | cr2 | can we get rid of pmem ? |
| 05:39 | tmzt | the PMEM stuff is qualcomm, not google/android |
| 05:39 | tmzt | there's strings about it that will be printed to amss |
| 05:39 | tmzt | console |
| 05:39 | tmzt | but we can't read from there yet |
| 05:39 | cr2 | yeah, there is even a pmem rpc |
| 05:39 | druidu | dunno |
| 05:39 | druidu | strange thing though |
| 05:39 | tmzt | that would be very helpful for debugging rpc if we could do it |
| 05:39 | druidu | I bootez 2.6.29 codeaurora kernel |
| 05:39 | tmzt | it prints errors for bad calls |
| 05:40 | druidu | and made the mddi code to work, and it detected the toshiba_vga client |
| 05:40 | druidu | *and* it detected the panel as toshiba! |
| 05:40 | tmzt | also, there seems to be a dedicated oncrpc router client on the arm9 |
| 05:40 | druidu | not sharp |
| 05:40 | cr2 | druidu: panel, or the controller ? |
| 05:40 | druidu | panel |
| 05:40 | tmzt | and it acts as a kind of proxy from arm9 programs to arm11 ones |
| 05:40 | druidu | lemme show you |
| 05:40 | tmzt | which is missing on non-A |
| 05:40 | tmzt | (old non-A) |
| 05:41 | cr2 | tmzt: yes. there is a lot R2R traffic there. |
| 05:41 | tmzt | yeah |
| 05:41 | druidu | https://www.codeaurora.org/gitweb/quic/la/?p=kernel/msm.git;a=blob;f=drivers/video/msm/mddi_toshiba_vga.c;h=f7e8482c62dfa99229c5c53590357801b6f4205e;hb=refs/heads/android-msm-2.6.29b |
| 05:41 | tmzt | do you think we can read the arm9 console? |
| 05:41 | cr2 | tmzt: it's the most of rpc data transmitted |
| 05:41 | tmzt | maybe push it to the dmesg? |
| 05:41 | druidu | line 121: id = mddi_get_client_id(); |
| 05:41 | tmzt | (with a prefix, like [amss]) |
| 05:42 | druidu | this correctly returns 0xd263, as we have it |
| 05:42 | tmzt | druidu: I think those id's are just gpio states |
| 05:42 | druidu | then in mddi_toshiba_panel_detect(), line 92 |
| 05:42 | druidu | lcd_gpio = toshiba_lcd_gpio_read(); |
| 05:42 | druidu | yeah, gpio state was 0xc0 |
| 05:42 | tmzt | they are really only for differentiating between two boards |
| 05:42 | druidu | which is toshiba panel they say |
| 05:43 | cr2 | druidu: these are different panels, and all this is depends on the gpio bits set by the manufacturer |
| 05:43 | tmzt | if we split the lcm's out and created a table for each board |
| 05:43 | druidu | well, for my panel... it was 0xc0 |
| 05:43 | tmzt | like .id = 0xxxx, .name = mddi-client-toshiba or mddi-lcd-toshiba or something |
| 05:43 | cr2 | pmdh -> toshiba mddi bridge -> lcm |
| 05:43 | tmzt | yeah |
| 05:44 | cr2 | pmdh -> epson mddi bridge -> lcm |
| 05:44 | tmzt | but pmdh is just a port |
| 05:44 | tmzt | it seems it's the packet dest that matters as well as the port in use |
| 05:44 | tmzt | but I'm not sure |
| 05:44 | druidu | anyway, I had low diskspace on my linux partition so I created a 30 GB file on my external drive, formated it with ext3, mounted it as a partition, and used it to compile android platform, work on codeaurora kernel, etc |
| 05:44 | cr2 | yes, each of these 3 have their control registers |
| 05:44 | druidu | now it's all fucked up, drive's toast :( |
| 05:44 | cr2 | :( |
| 05:45 | druidu | I only had some kits, all the android platform and the code aurora kernel on that one + hundreds of movies & kits... oh well |
| 05:45 | mdrobnak | that sucks |
| 05:45 | druidu | hope at least they'll replace it 'cause it's still under warranty |
| 05:45 | mdrobnak | anyway, I've stared at too many AT commands tonight. |
| 05:45 | cr2 | druidu: so if you have a toshiba mddi bridge, you need to query the gpio bits. the LCM init is then device specific |
| 05:45 | mdrobnak | Time for sleep. |
| 05:45 | mdrobnak | See yas. |
| 05:46 | druidu | LCM? |
| 05:46 | cr2 | LCD |
| 05:46 | druidu | well, gpio bits return 0xc0 |
| 05:46 | druidu | https://www.codeaurora.org/gitweb/quic/la/?p=kernel/msm.git;a=blob;f=drivers/video/msm/mddi_toshiba_vga.c;h=f7e8482c62dfa99229c5c53590357801b6f4205e;hb=refs/heads/android-msm-2.6.29b#l92 |
| 05:47 | cr2 | val = read_client_reg(GPIODATA) & 0x2C0; |
| 05:48 | druidu | exactly |
| 05:48 | druidu | that's what toshiba_lcd_gpio_read does |
| 05:48 | cr2 | you should check the raphaelLCD wiki for GPIODATA -> panel conversion |
| 05:48 | druidu | and my value was 0xc0 |
| 05:48 | druidu | not 0x80 how I'd expected (if my panel was sharp) |
| 05:48 | cr2 | the code above is useless for us |
| 05:49 | cr2 | because HTC used the gpios in a different way |
| 05:49 | druidu | all the mdp/mddi/fb stuff is rewritten in that kernel |
| 05:49 | druidu | ah, I see |
| 05:49 | cr2 | yes, in a mich better way |
| 05:50 | druidu | if only they'd document it :) |
| 05:50 | cr2 | read GPIODATA(150000) |
| 05:50 | cr2 | id=0 |
| 05:50 | cr2 | if bit 0x10 is not set -> id+=1 |
| 05:50 | cr2 | if bit 0x04 is not set -> id+=2 |
| 05:50 | cr2 | =0 -> panel id 5 (hitachi) |
| 05:50 | cr2 | =1 -> panel id 0xa (sharp) |
| 05:50 | cr2 | =2 -> panel id 0x11 (toppoly2) |
| 05:50 | cr2 | =3 -> panel id 2 (toppoly) |
| 05:50 | cr2 | this is for raph100, but i think it's the same for diam100 (to be checked) |
| 05:51 | druidu | what is 150000/ |
| 05:51 | druidu | ? |
| 05:51 | cr2 | GPIODATA |
| 05:51 | druidu | ah |
| 05:51 | druidu | well |
| 05:51 | druidu | I had C0 |
| 05:51 | druidu | id=0? |
| 05:51 | cr2 | masked after &0x2c0 |
| 05:52 | cr2 | print the unmasked GPIODATA value |
| 05:52 | cr2 | val = read_client_reg(GPIODATA) & 0x2C0; |
| 05:53 | druidu | ah, could try... but my kernel's on the toasted drive |
| 05:53 | cr2 | and here it's val = read_client_reg(GPIODATA) |
| 05:53 | cr2 | ok |
| 05:53 | druidu | managed to copy the stuff I was working on from cache |
| 05:53 | druidu | anyway, 2.6.29 aurora version is really much more flexible than android version |
| 05:53 | tmzt | from cache? |
| 05:53 | druidu | fs cache I guess |
| 05:54 | cr2 | druidu: yes. tvout,tvout,tvout :) |
| 05:54 | tmzt | which devices have ntsc/pal encoders? |
| 05:54 | druidu | usb device was long gone, I still had the file mounted as a partition... did a cp * /somedir/ and got some files |
| 05:54 | cr2 | raph100, raph800 |
| 05:54 | tmzt | oh yeah, that should work |
| 05:54 | druidu | yeah, and that only works because they've rewritten all the mdp/mddi stuff |
| 05:54 | tmzt | but you have to have a lot of free ram |
| 05:54 | druidu | it's 2-3 times more complex |
| 05:55 | tmzt | raph500? |
| 05:55 | cr2 | tmzt: don't have one |
| 05:55 | druidu | diamond hasn't got tvout |
| 05:55 | cr2 | and blac100 ? |
| 05:55 | cr2 | also not afair |
| 05:55 | tmzt | I think HD's do |
| 05:55 | tmzt | one of the features |
| 05:56 | cr2 | not sure about it, though |
| 05:57 | cr2 | the encoder is on 7x0x* |
| 05:57 | cr2 | and the DAC on pmic. i think |
| 05:57 | tmzt | not a chrontel chip? |
| 05:57 | cr2 | no |
| 05:57 | druidu | as for me... I think it's a sign, better get busy studying for an exam :D |
| 05:57 | cr2 | hehe |
| 05:58 | cr2 | druidu: the only real problem is BT now |
| 05:59 | cr2 | everythng else looks doable |
| 05:59 | druidu | I could live without BT for a while :) |
| 05:59 | cr2 | but don't know how to debug this one ;) |
| 05:59 | druidu | so what would the pb with BT be |
| 05:59 | druidu | ? |
| 05:59 | druidu | clocks? |
| 06:00 | cr2 | we need to check the clock first |
| 06:00 | tmzt | cr2: so arm9 console? that would likely help with clocks as well |
| 06:00 | cr2 | and then find out the reason for irq flood |
| 06:00 | cr2 | tmzt: what is arm9 console ? |
| 06:00 | tmzt | logging from the modem |
| 06:00 | druidu | of to bed or something, good luck with stuff |
| 06:00 | druidu | ;) |
| 06:00 | tmzt | it can log to SD or efs but I think it has a buffer in ram, maybe where we can read it |
| 06:01 | cr2 | druidu: good night |
| 06:01 | cr2 | i need to sleep 2 more hours too :) |
| 06:01 | cr2 | tmzt: in which ram ? |
| 06:01 | tmzt | probably protected ram |
| 06:01 | cr2 | sram or sdram ? |
| 06:01 | cr2 | find out the location |
| 06:01 | tmzt | not sure |
| 06:02 | cr2 | we may unprotect amss, and just 'pd' |
| 06:10 | cr2 | last thing for now: |
| 06:10 | cr2 | dzo: you are not supposed to run mddi at 245 |
| 06:10 | cr2 | this is from codeaurora |
| 06:10 | cr2 | 145 pinfo.clk_rate = 122880000; |
| 06:10 | cr2 | 146 pinfo.clk_min = 120000000; |
| 06:10 | cr2 | 147 pinfo.clk_max = 200000000; |
| 06:10 | tmzt | what was the mddi speed you pasted yesterday? |
| 06:10 | cr2 | good night |
| 06:10 | tmzt | the one wince yses? |
| 06:11 | cr2 | there are many. |
| 06:11 | cr2 | i think the raph100 value was 192MHz |
| 06:11 | cr2 | and g1 uses 122880000 |
| 06:11 | cr2 | ok |
| 07:41 | no2chem | sigh |
| 07:41 | no2chem | [12:19:09] <cr2> you probably need v=7 for 768, like for 528 |
| 07:42 | no2chem | PLEVEL0 - PLEVEL6 are all the same |
| 07:42 | no2chem | they are set in VDD_APC_PLEVEL0-VDD_APC_PLEVEL7, CLK_CTL_BASE 0x298 |
| 07:42 | no2chem | +* |
| 07:43 | no2chem | only PLEVEL7 is different |
| 07:43 | no2chem | only thing is supposedly "under a9 control" |
| 07:43 | no2chem | i can write to the register but |
| 07:43 | no2chem | i dont think the setting ever ends up in pmic |
| 07:44 | no2chem | probably needs something to be sent to VDD_APC_SSBI_ADDR, |
| 07:44 | no2chem | (CLK_CTL_BASE + 0x2B8) |
| 07:44 | no2chem | but |
| 07:44 | no2chem | i have no idea. |
| 07:53 | rafyvitto | hey nue, nows the overclock tool going =D |
| 07:54 | rafyvitto | hows* |
| 07:58 | rafyvitto | you there? |
| 08:02 | NetRipper | no2chem, so how is it possible you can change to 384 and then set it back to 528 (which makes it level 7 again)? or is that because of some automagic in the arm9? |
| 08:56 | MrPippy | trying to boot the connect-utb android on diam500, i'm getting the white screen and 'Unhandled fault: external abort on linefetch (0x806) at 0x00000000' in dmesg |
| 08:57 | MrPippy | android is just in a reset loop because SurfaceFlinger, power, batteryinfo, usagestats, and telephony.registry all die |
| 09:24 | Captnoord | MrPippy: try a diff package |
| 09:31 | MrPippy | i'm gonna give tac's a try, but i think its a kernel regression (at least for diam500), my own build that works with a month-old kernel just causes a reset now |
| 09:47 | MrPippy | yeah tac's and connect-utb are the exact same thing |
| 10:34 | tmzt | mrmoku|away: that's smd on cdma still broken |
| 10:36 | tmzt | mrp: |
| 10:36 | tmzt | sorry mrmoku |
| 10:36 | tmzt | mrpippy: |
| 10:41 | cr2 | hi |
| 10:41 | cr2 | on the train again |
| 11:38 | tmzt | mrc p15, 0, (rd), c15, c2, 4 |
| 11:38 | tmzt | mcr p15, 0, (rd), c15, c2, 4 |
| 11:38 | tmzt | rd |
| 11:38 | tmzt | wr |
| 11:38 | tmzt | but what about size? |
| 11:39 | tmzt | all of memory would be 1mb or b01011 |
| 11:39 | tmzt | all of the devices |
| 11:39 | tmzt | at least |
| 11:39 | tmzt | 4mb b01101 8mb b01110 32mb b10000 |
| 11:40 | tmzt | 31:12 base address |
| 11:40 | tmzt | 11:5 zero |
| 11:40 | tmzt | 4:0 size |
| 12:13 | Captnoord | untouchable's android packages... are really nice.... sound works.. only the interface is a bit slow.... |
| 12:54 | god4saken | anyone knows where to find the htcunid package?? |
| 13:26 | DJ_Steve | morning |
| 13:51 | DJ_Steve | mornin all |
| 13:52 | DJ_Steve | any haret experts in here |
| 13:52 | Captnoord | depending on what you call expert |
| 13:53 | DJ_Steve | ppl who can understad the output of haretconsole more specificly |
| 13:53 | Captnoord | I see |
| 15:21 | tuxhero | hi tmzt |
| 15:42 | Reactor16 | Hi All |
| 16:05 | gregnapola | Hi all... |
| 16:05 | gregnapola | does anyone have a git of earlier android builds (0.9) for Vogue ? |
| 16:21 | ninstaah | Hi, I am trying to boot Android for the first time - but it says "msm_i2c: Warning bus was busy - what did i do wrong? |
| 16:21 | ninstaah | On my Touch Diamond. |
| 16:43 | TaC | hi all |
| 16:44 | TaC | anyone can sign me an apk? |
| 16:46 | DJ_Steve | evening all |
| 16:46 | TaC | hi |
| 16:46 | DJ_Steve | hows all |
| 16:48 | TaC | i can't sign in this pc, so i need you to sign me one apk |
| 18:10 | Marajin | yo Captnoord |
| 18:10 | Captnoord | yo |
| 18:10 | Marajin | Captnoord: the staff training was pretty successful, they seemed so genuinely interested that they were learning very quickly |
| 18:11 | Captnoord | thank god |
| 18:11 | Captnoord | your saved |
| 18:11 | Captnoord | :P |
| 18:11 | Marajin | probably have forgotten it all by tomorrow of course |
| 18:11 | Marajin | but the basics are in ther |
| 18:11 | Marajin | e |
| 18:11 | Captnoord | nice nice nice |
| 18:11 | Captnoord | very cool it actualy works the way you want... |
| 18:11 | Marajin | yeah :) |
| 18:11 | Marajin | they love the new interactive boards |
| 18:12 | slawson | What field of work? |
| 18:17 | god4saken | anyone here has used htcunid? |
| 18:30 | Marajin | slawson: hmm, me? I own/run an independent school |
| 18:30 | Gnutoo | hi,I'll look how's the kernel for the dream is going on linuxtogo |
| 18:30 | Marajin | slawson: Only took over on the 1st june though, and over summer have upgraded the tech throughout the school amongst other things.. |
| 18:30 | slawson | Ah |
| 18:30 | slawson | Excellent :D |
| 18:31 | slawson | Happy it's going great for youi |
| 18:31 | Marajin | slawson: gone from athlon 1100s with 256mb of RAM to Phenom X2s @ 2.8ghz with 2 gigs of ram :) |
| 18:31 | slawson | :) |
| 18:31 | Marajin | windows/windows server 2000 -> Win7/Server 2008 |
| 18:31 | Marajin | 6 brand new active digitizer whiteboards |
| 18:31 | Marajin | lots of upgrades |
| 18:32 | Marajin | so I've had to train staff to use outlook/exchange/outlook web access/sharepoint + intranet stuff/windows 7 and the new whiteboard stuff |
| 18:32 | Marajin | 10 years of tech upgrades.. in 5 hours :p |
| 18:33 | Gnutoo | mmm where's the commit for alsa in msm? |
| 18:33 | DJ_Steve | nice upgrades @ marajin |
| 18:34 | Marajin | DJ_Steve: yeah the staff were all frightened for a bit but then they all got that 'woo! toys!' geek-like fascination, so they did really well learning |
| 18:34 | DJ_Steve | lol i bet |
| 18:34 | DJ_Steve | anyone understand haretconsole in here? |
| 18:35 | Gnutoo | I'll go |
| 18:35 | DJ_Steve | ? |
| 18:38 | Marajin | DJ_Steve: you'ev frightened 'em off by asking a question :p |
| 18:38 | DJ_Steve | LOL |
| 18:39 | miknix | DJ_Steve, what about haretconsole? |
| 18:39 | DJ_Steve | i need to work out from the console logs what gpios something uses :) |
| 18:41 | druidu | weee, edge works on android on my diam100 |
| 18:42 | DJ_Steve | lol got android w/wifi running on the polaris over here, |
| 18:43 | bertramt | Has anybody had luck with the CDMA radio in the Raph800? |
| 18:44 | Marajin | DJ_Steve: polaris? I weep for you :P |
| 18:44 | DJ_Steve | lol ive got a dream aswell :) |
| 18:44 | Marajin | DJ_Steve: I have a kaiser, which as I recall is more or less polaris without a keyboard :p |
| 18:44 | Marajin | er *with a keyboard |
| 18:44 | Marajin | and it's nasty slow :p |
| 18:44 | Marajin | then again, my kovsky (xperia x1) isn't that fast either =/ |
| 18:45 | DJ_Steve | was gonna say lol. i used to have kaiser, its fast as heck on android tho, just need to get gpios for camera and bluetooth found |
| 19:25 | rafyvitto | hello guys |
| 19:25 | druidu | hey |
| 19:26 | druidu | I was thinking... is there any way to set the resume address after a soft reset? |
| 19:27 | druidu | 'cause if the ram is still there but the hardware is initialized, it would be a nice way to test a bootloader |
| 19:27 | rafyvitto | your asking me ? xD |
| 19:33 | druidu | asking the room :) |
| 19:38 | rafyvitto | though so xD |
| 19:43 | Squarc | Hey :) |
| 19:43 | rafyvitto | hiya ;) |
| 19:44 | Squarc | git.linuxtogo.org is still down ? |
| 19:44 | Captnoord | seems so |
| 19:44 | Captnoord | btw |
| 19:44 | Captnoord | did you put the stuff into a directory Squarc? |
| 19:44 | Captnoord | or did you put the stuff into the root |
| 19:44 | Squarc | ..? how do you mean ? |
| 19:44 | Captnoord | ? |
| 19:44 | Captnoord | the package |
| 19:44 | mickeyl | git works fine here |
| 19:45 | mickeyl | did you remove your dns entry in /etc/hosts again? |
| 19:45 | Captnoord | git.linuxtogo.org[0: 188.40.83.200]: errno=No error |
| 19:45 | Captnoord | fatal: unable to connect a socket (No error) |
| 19:45 | Squarc | mickeyl: "git clone git://git.linuxtogo.org/home/groups/mobile-linux/kernel.git branches htc-msm-2.6.27" does work for you ? |
| 19:45 | Squarc | I have the same |
| 19:45 | rafyvitto | yep gits working |
| 19:45 | mickeyl | hmm, i'm using non-anonymous |
| 19:45 | mickeyl | let me try anonymous |
| 19:45 | Captnoord | I removed my host entry because I tought the server upgrade was complete |
| 19:46 | Squarc | Captnoord: I still dont get what you mean.. which "stuff" ? |
| 19:46 | Squarc | Captnoord: the build package that I tested on my phone first? |
| 19:46 | mickeyl | hmm, that git clone does't work for me either |
| 19:46 | Squarc | hmm |
| 19:46 | mickeyl | guess they didn't restart the anonymous git server |
| 19:46 | mickeyl | non-anonymous works though |
| 19:46 | mickeyl | (in case you have an account @ ltg) |
| 19:46 | Squarc | Is such an account free / open for everyone? |
| 19:46 | Captnoord | slawson: yup |
| 19:47 | Captnoord | as I tryed the new one and needed to do stuff at the root of my SD card |
| 19:47 | Squarc | ok |
| 19:47 | Captnoord | with stuff I mean |
| 19:47 | Captnoord | the android package... |
| 19:47 | Squarc | ok |
| 19:47 | Squarc | well |
| 19:48 | Squarc | I have the boot files located in the internal ROM on my phone... in /linux |
| 19:48 | Squarc | HaReT.exe is also there |
| 19:48 | Captnoord | hehe |
| 19:48 | Squarc | and I have the large packages (images) located in /tmp on the internal memory (4 Gig, non swappable) |
| 19:48 | Captnoord | and if you move them to the root directory |
| 19:48 | Captnoord | I tryed that to |
| 19:48 | Squarc | wont make a difference... |
| 19:48 | Squarc | also works |
| 19:48 | Captnoord | it doesn't seem to work correct |
| 19:48 | Squarc | :o ? |
| 19:49 | Squarc | as long as you change the config it will, right? |
| 19:49 | Captnoord | I changed the config correctly |
| 19:50 | Captnoord | but I have to admit that I didn't spend a lot of time on it |
| 21:33 | mdrobnak | Hey everyone. Half here. Half sick. :-/ |
| 22:07 | nanook | hello all! i was looking for android on the htc-diamond and found xda... i want to try it but latest images are from june, are there somewhere more recent images? |
| 22:09 | cr2 | nanook: there is a long thread on xda-dev about it |
| 22:09 | nanook | cr2: ok, thanks |
| 22:10 | nanook | cr2: i think you are talking about this one: http://wiki.xda-developers.com/index.php?pagename=RaphaelLinux |
| 22:11 | cr2 | nanook: no |
| 22:11 | nanook | cr2: at the end in 'progress' i can read that microphone is working but images are older, thats why i was asking... i'll have to wait i don't see myself compiling that succcesfully :_( |
| 22:11 | cr2 | http://forum.xda-developers.com/showthread.php?t=402002 |
| 22:12 | nanook | cr2: ok |
| 22:12 | nanook | cr2: thx again |
| 22:21 | cr2 | NetRipper: does the ltg git commit work now ? |
| 22:24 | NetRipper | cr2, eh? what commit? |
| 22:24 | NetRipper | i havent done anything with the git |
| 22:25 | Untouchab1e | hi all |
| 22:28 | cr2 | NetRipper: it does not work in the last days. dzo could not commit his latest changes |
| 22:29 | NetRipper | aha |
| 22:29 | NetRipper | out of disk space again? :p |
| 22:29 | cr2 | no idea |
| 22:29 | cr2 | dns problems too |
| 22:30 | cr2 | i have worked through the LCD init |
| 22:30 | Untouchab1e | TaC just released a new package 10 minutes ago.. Anyone know whats new? |
| 22:30 | cr2 | and hope that it will work, so we can try to boot from haret with LCD switched off |
| 22:31 | NetRipper | ah to test if it turns it on? |
| 22:31 | NetRipper | Untouchab1e, TaC? |
| 22:31 | NetRipper | Untouchab1e, for touch pro? |
| 22:31 | NetRipper | /dev/md2 720544880 338221028 346010532 50% / |
| 22:32 | cr2 | Untouchab1e: what about gps ? |
| 22:32 | NetRipper | enough disk space |
| 22:32 | NetRipper | :P |
| 22:32 | Untouchab1e | cr2, what you mean? |
| 22:32 | cr2 | Untouchab1e: the androed has some standalone closed source gps lib |
| 22:33 | Untouchab1e | Now you lost me.. are you talking about the TaC package now, or? |
| 22:33 | cr2 | no, in general |
| 22:34 | cr2 | i have no idea about android capabilities |
| 22:34 | Untouchab1e | ah, I see |
| 22:34 | cr2 | 3d worked as is , afaik |
| 22:34 | Untouchab1e | Btw, anyone know why Android isnt running on the Blackstone yet? |
| 22:35 | Untouchab1e | its pretty much identical to the Raphael in terms of hardware |
| 22:35 | Untouchab1e | only exception being the screen |
| 22:35 | cr2 | nobody has such device here,. |
| 22:35 | pdbogen | I'm interested in running Linux (any sort) on my Raphael (RAPH110). What should I look at? |
| 22:35 | cr2 | it needs the epson mddi init being finished |
| 22:35 | Untouchab1e | pdbogen, you mean what or where? |
| 22:36 | pdbogen | Untouchab1e: I consider the two questions synonmous, but I supposed the latter is more correct. |
| 22:36 | Untouchab1e | pdbogen, if you just want to run Android, check out connect-utb.com |
| 22:36 | Untouchab1e | and go to the download section |
| 22:37 | pdbogen | Thanks; looking at it now. |
| 22:40 | mdrobnak | cr2: What kind of patches from dzo is git missing? |
| 22:48 | NetRipper | screen -x |
| 22:48 | NetRipper | ^ignore |
| 22:50 | thedicemaster | i don't suppose anyone here is edvard under a different name? |
| 22:51 | Untouchab1e | thedicemaster, here |
| 22:51 | Untouchab1e | ;) |
| 22:52 | mdrobnak | NetRipper: Not a shell. ;-) |
| 22:52 | thedicemaster | found 4 more sensors on the diamond today. |
| 22:53 | Untouchab1e | what kind of sensors |
| 22:53 | Untouchab1e | ?* |
| 22:53 | thedicemaster | apparently the dial, hangup, home, and back button each have a touch-sensitive sensor in them. |
| 22:53 | thedicemaster | never knew because they aren't(visibly) used in windows mobile. |
| 22:54 | mdrobnak | On the RAPH, it's 1 big capacitive touch pad + buttons |
| 22:54 | Untouchab1e | yup |
| 22:54 | Untouchab1e | same with the diamond |
| 22:54 | Untouchab1e | the whole touch pad is capacitiive |
| 22:55 | Untouchab1e | capacitive* |
| 22:55 | thedicemaster | interesting. |
| 22:56 | mdrobnak | Hey, does anyone know how to manually communicate with the RIL daemon in Android? I would like to do some tests, and Android tries to re-connect too quickly after a crash... |
| 22:56 | thedicemaster | up untill now i thought the only capacitive touch part on the diamond was the ring around the ok button. |
| 22:56 | pdbogen | When it says to uncompress the package in the SD card root, does it mean I should have a directory called 'raph', or everything inside raph should be in the root? |
| 22:56 | Untouchab1e | pdbogen, sorry for that unclearity, but everything inside the raph folder should be in root |
| 22:57 | pdbogen | Thanks much. (In fairness, it may not have been unclear; I admit to a bit of laziness in asking before digging around in docs.) |
| 22:58 | thedicemaster | i was hoping to try out some more features on android, but it seems DIAM200 uses slightly different ways to handle data connections as well. |
| 22:58 | Untouchab1e | thedicemaster, to be honest, I dont know anything about the DIAM200. But you say it has a different button layout? |
| 22:58 | thedicemaster | as soon as i arrive at the step of turning on/off airplane mode it goes crazy |
| 22:58 | thedicemaster | yes it does. |
| 22:58 | Untouchab1e | Never heard of that before.. hah |
| 22:59 | Untouchab1e | NetRipper, or cr2, did you know that? |
| 22:59 | thedicemaster | it has dial and hangup on the top, and home/back on the bottom. |
| 22:59 | thedicemaster | for a picture search for HTC victor. |
| 22:59 | Untouchab1e | http://pdadb.net/img/htc_victor.jpg |
| 23:00 | Untouchab1e | shows home\back on the top? |
| 23:00 | thedicemaster | ... wtf? |
| 23:00 | Untouchab1e | hehe, I just googled HTC Victor and that is what came up :) |
| 23:00 | thedicemaster | i wish i could send you a pic of mine, but my camera is my phone. |
| 23:01 | Untouchab1e | hehe |
| 23:01 | Untouchab1e | no problems ;) |
| 23:01 | Untouchab1e | just facinated by how many different models of the Diamond and Touch Pro there are |
| 23:01 | Untouchab1e | I can understand the need for one GSM and one CDMA, but all these other variations.. I dont see the point |
| 23:02 | thedicemaster | http://www.techdigest.tv/mda-compact-iv-t-mobile-uk-july.jpg |
| 23:02 | thedicemaster | well, DIAM100 and DIAM200 i can understand |
| 23:02 | Untouchab1e | how so? |
| 23:04 | mdrobnak | Wow that's kinda funky. |
| 23:04 | thedicemaster | 100 is the actual HTC diamond, and 200 is an operator branded version, which doesn't have the characteristic diamond design to keep diamond an HTC-only thing. |
| 23:04 | high-rez | Gents - to build my own kernel with all the working audio - should I still follow the BuildTheRaphaelKernel wiki link on xda-developers ? |
| 23:04 | NetRipper | didn't even know a DIAM200 existed |
| 23:04 | thedicemaster | http://www.cellfanatic.com/wp-content/uploads/2008/07/htc_victor.jpg |
| 23:04 | Untouchab1e | NetRipper, neither did I untill recently |
| 23:05 | thedicemaster | around here it's mostly known as MDA compact IV(T-mobile is the only one selling it, and that's the name they gave to it) |
| 23:05 | Untouchab1e | I hate the fact that Carriers decide they want different looking phones just for the sake of being different |
| 23:05 | Untouchab1e | usually, you end up with a uglier phone compared to the original |
| 23:06 | NetRipper | i also have the home button at top, and connect/disconnect at bottom (so other way around) |
| 23:06 | thedicemaster | i think this was decided by HTC. |
| 23:06 | NetRipper | but i think this difference was already known |
| 23:06 | NetRipper | doesn't the fuze have it like that as well? |
| 23:06 | high-rez | Yeah those carrier guys suck. ;) |
| 23:06 | high-rez | ducks |
| 23:06 | Untouchab1e | NetRipper, home\back at the top is the "normal" layout |
| 23:06 | thedicemaster | not only the physical location is swapped, internally they are swapped as well. |
| 23:06 | Untouchab1e | Just look what T-Mobile has done with the Touch Pro 2: http://i.zdnet.com/blogs/thumb_450_t-mobile-touch-pro-2.jpg |
| 23:07 | Untouchab1e | Compared to the original Touch Pro 2, it looks like a bucket of puke |
| 23:07 | thedicemaster | which is why my dial button brings up the menu. |
| 23:07 | Untouchab1e | pardon my french.. |
| 23:07 | thedicemaster | in android that is. |
| 23:07 | Bally3 | can anyone tell me whch kernel supports the Samsung SC32442 |
| 23:08 | Bally3 | please? |
| 23:08 | tmzt | linux? |
| 23:08 | high-rez | Hi tmzt. |
| 23:08 | Bally3 | yup |
| 23:08 | Bally3 | cr2 told me and I've forgotton now :S |
| 23:09 | thedicemaster | i actually prefer the victor look over the diamond look, i don't really like the weird battery cover and sharp corners. |
| 23:09 | Bally3 | this is the first chance I've had to work on it |
| 23:10 | tmzt | well, I haven't really gotten anywhere with pprr, turns out the physical addresses I have for the periphs are not in the same region of memory |
| 23:12 | thedicemaster | really annoys me how gprs won't work for me. |
| 23:12 | Untouchab1e | thedicemaster, I havent gotten it to work either.. you tried following the guide I linked to you, right? |
| 23:13 | tmzt | high-rez: hi, how have you been? |
| 23:13 | thedicemaster | well, i don't know if it actually works for the DIAM200 |
| 23:13 | dwaradzyn | how do i make standalone .exe with kernel+initramfs+default.txt using haret? any doc about it? |
| 23:13 | thedicemaster | when i get to the part where i turn airplane mode on android does a soft-reset. |
| 23:13 | high-rez | tmzt: Good. I see you guys have made some amazing progress. Congrats on that. :) |
| 23:14 | tmzt | make linboot? |
| 23:14 | dwaradzyn | yes |
| 23:14 | tmzt | high-rez: that's dzo/cr2/etc. |
| 23:14 | tmzt | mostly |
| 23:14 | high-rez | Well congrats all around. That's friggin cool/amazing/all of the above. |
| 23:15 | Squarc1 | tmzt: I have build the latest version that was on git and I've added android packages, I'm just wondering, is Internet working allrady because this Android needs to setup Google account... |
| 23:15 | tmzt | on gsm? |
| 23:15 | thedicemaster | internet works if you can figure ir out. |
| 23:15 | tmzt | you can set something in init.rc I think to disabled that |
| 23:15 | tmzt | ask in #android-root |
| 23:16 | Squarc1 | ok :) |
| 23:16 | Squarc1 | thx |
| 23:16 | thedicemaster | but you can cancel setting up that account usually. |
| 23:17 | thedicemaster | i got a question too now. i read that with the diamond the power button rotates the screen, but is that on all screens or only certain supported apps? |
| 23:18 | rafyvitto | guys, whats new on dzo diamond zimage?? |
| 23:18 | rafyvitto | he released one today |
| 23:21 | tmzt | wouldn't it be in gitweb? |
| 23:21 | rafyvitto | is not >< |
| 23:22 | high-rez | hmmf |
| 23:22 | rafyvitto | http://cs-alb-pc3.massey.ac.nz/vogue/files/?C=M;O=D |
| 23:22 | rafyvitto | its there |
| 23:22 | rafyvitto | bit there no change in the git |
| 23:22 | rafyvitto | but** |
| 23:25 | cr2 | which haret command disables the lcd ? |
| 23:25 | cr2 | setlcd 4 or setlcd 1 ? |
| 23:27 | NetRipper | 4 |
| 23:27 | NetRipper | 1=on, 2=standby, 3=suspend, 4=off |
| 23:27 | cr2 | can somebody try it on diam100 ? |
| 23:27 | cr2 | just put |
| 23:27 | cr2 | setlcd 4 |
| 23:27 | NetRipper | this is from haret docs |
| 23:27 | NetRipper | :) |
| 23:27 | cr2 | into default.txt |
| 23:28 | tmzt | NetRipper: have time to help with pprr? I'm very confused |
| 23:28 | cr2 | before booting linux |
| 23:28 | NetRipper | tmzt, not really, i gtg to bed now.. besides that... what is pprr? |
| 23:28 | NetRipper | :) |
| 23:28 | tmzt | peripheral port remap register |
| 23:28 | high-rez | Man its been a long time since I tried to get this working. |
| 23:29 | tmzt | same as the problem we had way back with booting with mmu off |
| 23:29 | NetRipper | why are you working on that again? |
| 23:29 | tmzt | I'm trying to write a simple assembler program for drawing to the screen and doing proc comm |
| 23:29 | tmzt | as a first step to writing a bootloader |
| 23:30 | tmzt | people.openezx.org/tmzt/green7.S |
| 23:30 | NetRipper | you can look into the source of haret |
| 23:30 | tmzt | yeah, that's the next step |
| 23:30 | tmzt | I just hope I can find the source, the cvs seems to be broken for me |
| 23:30 | NetRipper | im shit at assembler :p |
| 23:31 | NetRipper | is it? |
| 23:31 | tmzt | yeah, or I have the password wrong |
| 23:31 | NetRipper | its anonymous |
| 23:31 | tmzt | anyway |
| 23:31 | tmzt | yeah |
| 23:31 | tmzt | this is a mappping thing, I can get the assembler fixed now |
| 23:31 | tmzt | I just need the addresses for it to work |
| 23:32 | NetRipper | but in haret we didn't have to do anything with the pprr |
| 23:32 | high-rez | Are there any issues with harest and wimno 6.5 ? |
| 23:32 | NetRipper | i believe |
| 23:32 | high-rez | s/harest/haret |
| 23:33 | NetRipper | high-rez, not to my knowledge |
| 23:33 | dwaradzyn | cr2: i'm still trying to test sound on blac100. yesterday i tried with older angstrom initrd with mplayer, but "no sound device" for alsa, oss and sdl audio output. i built boot.exe with zimage+initramfs+default.txt but it still complains about "you need correct root="... what should i put into default.txt? |
| 23:34 | high-rez | netripper then I'm doing something very wrong ;) |
| 23:34 | kri5 | i could test it too but i still can't clone git repo :/ |
| 23:34 | tmzt | did you clear the host entry? |
| 23:34 | cr2 | dwaradzyn: i have |
| 23:34 | kri5 | tmzt: ? |
| 23:34 | cr2 | set cmdline "msmts_calib=0x6D.0x5D.0x340.0x375 msmvkeyb_toggle=hide lpj=2637824 board_htcraphael.disable_wifi=1 init=/init console=tty0 delay=2" |
| 23:35 | cr2 | ramboot |
| 23:35 | NetRipper | ramboot is only for linload |
| 23:35 | dwaradzyn | well, i have set cmdline "init=/init console=tty0 msmts_calib=0x395.0x3c9.0x79.0x77 msmvkeyb_toggle=on" |
| 23:35 | NetRipper | use 'boot' if you have haret and other files seperately |
| 23:36 | cr2 | dwaradzyn: ramboot ? |
| 23:36 | cr2 | make linload KERNEL=../kernel/arch/arm/boot/zImage INITRD=initrd-angstrom-2009-07-06 SCRIPT=script.txt |
| 23:36 | dwaradzyn | ah, i have boot instead of ramboot |
| 23:36 | cr2 | NetRipper: why does android need mem=0x59M ? |
| 23:37 | cr2 | NetRipper: they all boot with mem=76M |
| 23:37 | NetRipper | 59M? think it works with 76 as well |
| 23:37 | cr2 | and then android crashes on out-or-memory |
| 23:37 | NetRipper | cr2, android won't boot when the second memory bank is enabled |
| 23:37 | cr2 | 0x59, not 59 |
| 23:38 | cr2 | mem=95M |
| 23:38 | NetRipper | cr2, and i dont know the reason for it |
| 23:38 | NetRipper | my memory tests seem to work on the second memory bank (but couldnt test it all) |
| 23:38 | cr2 | with mem=96M it does nkit work too |
| 23:38 | mdrobnak | Mine only works with 76 |
| 23:38 | NetRipper | no i meant mem=76M |
| 23:38 | NetRipper | should be that |
| 23:38 | mdrobnak | Which really hurts things |
| 23:38 | NetRipper | any higher and android wont boot |
| 23:39 | cr2 | ok |
| 23:39 | NetRipper | (or, instead of mem=76M, set #if 0 for the 2nd memory bank in board-raphael.c) |
| 23:39 | NetRipper | :) |
| 23:40 | tmzt | the banks need fixed |
| 23:40 | tmzt | it's not the total memory that's the issue |
| 23:40 | mdrobnak | Do we know why it doesn't work? |
| 23:40 | NetRipper | tmzt, the annoying thing is, the 2nd memory bank works fine with plain linux |
| 23:40 | NetRipper | just not in android |
| 23:40 | tmzt | because of pmem? |
| 23:40 | NetRipper | the memory map should take care of that |
| 23:40 | NetRipper | and the second memory bank doesn't have reserved memory addreses afaik |
| 23:41 | NetRipper | reserved = protected |
| 23:41 | NetRipper | cr2, did you need me to try and commit something? |
| 23:44 | cr2 | NetRipper: not today |
| 23:45 | NetRipper | oki |
| 23:46 | cr2 | 0x1deadca7 |
| 23:47 | NetRipper | 0x1nmydr34ms |
| 23:47 | NetRipper | good night :) |
| 23:48 | cr2 | good night |
| 23:54 | dwaradzyn | playwa2 says: Could not open MSM SND driver: no such file or directory |
| 23:56 | dwaradzyn | the output and dmesg is at http://pastebin.com/m11ef4f35 |
| 23:58 | high-rez | Hmm. My kernel seems to be stuck in a big dmesg loop... "msm_gpio_enter_sleep gpio [xxxxxx]" .... "msmfb_start_dma 105.xxx ms after vsync request" over and over and over again |
