| 00:33.12 | *** join/#htc-linux Marex (n=vasum7am@auryn.karlin.mff.cuni.cz) |
| 00:53.27 | *** join/#htc-linux MethoS (n=lem@dyndsl-085-016-163-139.ewe-ip-backbone.de) |
| 01:01.05 | *** join/#htc-linux Marex (n=vasum7am@auryn.karlin.mff.cuni.cz) |
| 01:01.44 | Marex | dcordes_, Kevin2: hi, you around ? |
| 01:11.17 | *** join/#htc-linux Moku (n=John@f049100199.adsl.alicedsl.de) |
| 01:13.44 | Marex | mickeyl: hi, you around ? |
| 01:13.55 | Marex | ~botmail |
| 01:14.12 | Marex | ah ... |
| 01:17.15 | Marex | ~botmail for cr2: N560 loader... set PSPR=0xa1b2c3d4, reserve memory node at 0xa0000000 (see arch/arm/mm/mmu.c:640) as first four bytes are trashed by the bootloader, !! IMPORANT STEP - set DoCG4 into Deep Power-Down mode !!. If you now suspend, you should be able to wakeup from address 0xa0100000 (that's where bootloader jumps). The principle is simple, if DoCG4 is in DPD-mode, it returns invalid data if you read from registers (so the ID the |
| 01:21.30 | Marex | ~botmail for cr2: one more thing - the size of memory mode is fine with 1 page (0x1000). Here's a bonus question - how to switch DoCG4 into DPD mode ? ;-) |
| 01:21.58 | Marex | ~botmail for cr2: *memory node |
| 01:22.25 | Marex | argh damn ... botmail sucks :E |
| 01:30.01 | *** join/#htc-linux noodlesgc (n=noodles@97.72.68.195) |
| 02:03.13 | *** join/#htc-linux madCoder- (n=madcoder@c-71-225-238-170.hsd1.pa.comcast.net) |
| 02:39.16 | *** part/#htc-linux wasabi (n=wasabi@ubuntu/member/wasabi) |
| 02:41.05 | *** join/#htc-linux lama (i=lama@netbsd.pl) |
| 03:08.51 | *** join/#htc-linux mrmoku|easter (n=mrmoku@ppp-93-104-124-128.dynamic.mnet-online.de) |
| 03:55.50 | *** join/#htc-linux PoohbaLT (n=BlkPoohb@c-69-141-71-249.hsd1.nj.comcast.net) |
| 05:04.49 | *** join/#htc-linux swc|666 (n=swc@unaffiliated/swc666/x-4934821) |
| 05:32.46 | *** join/#htc-linux Magorium (i=Othello@gateway/tor/x-5f2a3ebb5b682872) |
| 05:54.12 | *** join/#htc-linux Sti_02391 (n=Where_is@140.154-65-87.adsl-dyn.isp.belgacom.be) |
| 05:57.11 | *** join/#htc-linux Othello (i=Othello@gateway/tor/x-096b156be6a41f69) |
| 06:25.53 | *** join/#htc-linux mugsie (n=graham@78.16.72.164) |
| 06:30.43 | *** join/#htc-linux dzo (n=dzo@121-98-128-127.bitstream.orcon.net.nz) |
| 06:36.52 | *** join/#htc-linux ellisway (n=ellis@80-46-67-47.static.dsl.as9105.com) |
| 06:38.59 | *** part/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey) |
| 06:41.48 | *** join/#htc-linux ellis (n=ellis@80-46-67-47.static.dsl.as9105.com) |
| 07:20.41 | *** join/#htc-linux mib_1cg1omfb (i=75c483fb@gateway/web/ajax/mibbit.com/x-3be4c85f3825f613) |
| 07:21.35 | mib_1cg1omfb | hello.. |
| 07:21.38 | *** join/#htc-linux exception13 (n=exceptio@testdrive.kgts.ru) |
| 07:23.41 | duncang | Morning. |
| 07:24.14 | *** join/#htc-linux Xime (n=xime@bankize.net) |
| 07:27.13 | mib_1cg1omfb | may i ask u something.? |
| 07:28.30 | duncang | You can try. I'm probably one of the least knowledgable people on this channel, though. |
| 07:29.37 | mib_1cg1omfb | k..u know to upgrade ubuntu 7.10 to latest..?? |
| 07:31.00 | tmzt | not really the channel for that |
| 07:31.10 | tmzt | what do you mean by latetst? |
| 07:31.39 | tmzt | unless you mean ubutnu on your phone then this the right channel |
| 07:32.47 | duncang | I use real Debian exactly because Ubuntu's upgrades are such a pain. In general, though: |
| 07:32.48 | duncang | * aptitude update |
| 07:32.48 | duncang | * aptitude upgrade |
| 07:32.48 | duncang | * sed -e 's|current-version-name|new-version-name|' /etc/apt/sources.list |
| 07:32.48 | duncang | * aptitude update |
| 07:32.48 | duncang | * aptitude dist-upgrade |
| 07:32.50 | duncang | Then keep doing "aptitude update" until there's no more to do. Then reboot. |
| 07:33.30 | tmzt | mib* left |
| 07:37.15 | duncang | Sorry, I meant keep doing "aptitude upgrade". Just refreshing the cache is pointless ;-) |
| 07:38.51 | duncang | notices the person who asked the question left before he answered it... Doh. |
| 08:17.08 | *** join/#htc-linux groton_ (n=groton@unaffiliated/groton) |
| 08:17.34 | groton_ | hello, anyone could tell me where to get a SSPL for 1.93 HTC Touch Diamond? |
| 08:18.12 | tmzt | #xda-devs |
| 08:18.43 | groton_ | ok, sorry for being OT :) |
| 08:19.32 | tmzt | it's just that your more likely to find an answer there |
| 08:19.53 | groton_ | yeah, thanks again :) |
| 08:19.53 | tmzt | this channel is about running linux on these devices |
| 08:20.10 | groton_ | Android right? This would be awesome :) |
| 08:20.35 | tmzt | android, angstrom, debian, gentoo, etc. |
| 08:20.46 | duncang | It is awesome - or will be when we have sound and can charge the battery... |
| 08:21.01 | tmzt | yeah |
| 08:21.09 | tmzt | what device dunc? |
| 08:21.27 | groton_ | i am connected thru hsdpa with my htc d.touch now, it is pretty fast as a modem (1.4Mbps) |
| 08:21.40 | duncang | RAPH120, aka O2 Xda Serra. And let's not encourage Gentoo, eh? |
| 08:21.49 | tmzt | d.touch? |
| 08:21.52 | groton_ | I guess this feature is not supported by the linux kernel runnning on the HTCs device |
| 08:22.02 | groton_ | tmzt, Diamond Touch |
| 08:22.21 | tmzt | gentoo is used to build an arm stage3 |
| 08:22.34 | tmzt | not to build from source on the device |
| 08:22.54 | duncang | Not without battery charging working, at least ;-) |
| 08:23.41 | tmzt | raph120 is the same as raph100? |
| 08:24.07 | tmzt | sound could be made to work |
| 08:24.42 | duncang | We think so, at least approximately. The UTB builds work. I have sent and received SMS in Android; just silently... |
| 08:25.09 | duncang | My current userland is Android + bits of Debian armel on half an 8GB SD card. |
| 08:26.34 | tmzt | cool |
| 08:26.48 | tmzt | I wish cdma raph was that usable |
| 08:27.26 | tmzt | do you mean phone audio or music/sound effects? |
| 08:28.24 | duncang | There is no sound at all of any kind, in or out. |
| 08:28.45 | tmzt | ok |
| 08:28.55 | duncang | However, IP-over-ethernet-over-USB works fine. But ssh/scp stalls a lot because of lack of entroy. Sliding the keyboard seems to create some. Would be nice to get the accelerometers working and use them for entropy... |
| 08:29.05 | tmzt | but they are different and the way to fix them is different |
| 08:29.06 | duncang | Need battery charging first, though. |
| 08:29.30 | tmzt | yeah, that would be interesting |
| 08:30.30 | duncang | Am currently transplanting research info from xda-dev wiki to htc-linux.org. |
| 08:30.51 | duncang | Then need to figure out why git head freezes on boot on my device. |
| 08:31.56 | tmzt | freezes at what point? |
| 08:32.31 | duncang | Hang on; let me boot it. Before the SD driver comes up, certainly... |
| 08:32.56 | duncang | has to wait for WinCE to boot first. |
| 08:34.25 | duncang | Ah ha. Now need to look what kernels I have available... |
| 08:34.50 | tmzt | using rootdelay? |
| 08:35.07 | duncang | Have to with SD root fs. |
| 08:36.13 | duncang | Interesting; the vfat partition's not shown up with WinCE in usb mass storage mode... |
| 08:36.42 | duncang | plugcycles USB cable. |
| 08:37.03 | tmzt | try unpluging repluging sd card |
| 08:37.58 | duncang | That requires opening the back cover ;-). Got it now anyway. Many kernels to test. Let's start with git head. |
| 08:38.16 | tmzt | I know |
| 08:38.25 | tmzt | but not turning off phone |
| 08:38.28 | tmzt | ok |
| 08:38.34 | duncang | I should mention I built it with a conventional Debian cross-compile toolchain, not the CodeSourcery one recommended. |
| 08:38.48 | tmzt | that should be ok |
| 08:39.48 | duncang | My "known good" kernel is UTB's from 19 Feb. I haven't finished bisecting the changes since then. Or even checking that it is the same as the then git head... |
| 08:44.06 | duncang | OK, so we have HaRET 20090404-kern98000, which thinks RAPH120 is a generic MSM7xxxA |
| 08:44.34 | duncang | And I am about to do this: |
| 08:44.34 | duncang | set MTYPE 1910 |
| 08:44.34 | duncang | set KERNEL zImage |
| 08:44.34 | duncang | set RAMADDR 0x10000000 |
| 08:44.34 | duncang | set cmdline "root=/dev/mmcblk0p2 rootdelay=5 init=/init console=tty0 msmvkeyb_toggle=off msmts_calib=0x7e.0x63.0x35f.0x386" |
| 08:44.37 | duncang | Boot |
| 08:45.20 | duncang | It's hung at console handover. |
| 08:46.07 | duncang | Oh, no, sorry. It just paused. |
| 08:48.32 | duncang | thinks it's safe to say it's hung now. |
| 08:49.12 | tmzt | try a few times |
| 08:50.57 | duncang | Looks like it stuck at: |
| 08:50.57 | duncang | [1.405...] usb: reset |
| 08:50.57 | duncang | [1.411...] usb: portchange USB_SPEED_HIGH |
| 08:50.57 | duncang | [1.416...] clock-wince: set_mdsn_host_clock: 19, freq=1 |
| 08:51.03 | *** join/#htc-linux StarLite (n=nnscript@s55916ca6.adsl.wanadoo.nl) |
| 08:51.34 | duncang | And it's giving off enough RF and/or USB noise to bugger up my wireless keyboard. |
| 08:52.22 | duncang | Is haret-20090404-kern98000 OK to use, btw? What does the suffix mean? |
| 08:52.59 | duncang | waits for WinCE to boot again. |
| 08:54.44 | duncang | Console handover was instantaneous that time. |
| 08:54.51 | *** join/#htc-linux Untouchab1e (n=Untoucha@cm-84.209.89.86.getinternet.no) |
| 08:54.54 | Untouchab1e | Good morning all |
| 08:54.59 | Untouchab1e | Happy easter |
| 08:56.04 | duncang | Morning. tmzt is helping me watch my RAPH120 hang during boot. |
| 08:56.46 | Untouchab1e | ah, cool |
| 08:56.50 | Untouchab1e | How's it going? |
| 08:58.57 | duncang | This time we got to: |
| 08:58.57 | duncang | [1.52..] clock-wince: pc_clk_set_rate: id=19, 144000 |
| 08:58.57 | duncang | [1.53...] clock-wince: set mdns; 19, 144000; bitidx=7, offset=a4, ns=0 |
| 08:58.58 | duncang | [1.54...] usb: portchange USB_SPEED_HIGH |
| 08:58.58 | duncang | [1.55...] usb: reset |
| 09:00.14 | duncang | Who build the kernels for the UTB downloads? Do they use git head from msm-2.6.27 or some other patchset? |
| 09:00.48 | Untouchab1e | duncang, the connect-utb.com packs are built from various people |
| 09:00.54 | Untouchab1e | not just one guy.. |
| 09:01.05 | Untouchab1e | I just put the work of many together, test it and release it |
| 09:01.33 | Untouchab1e | but I am pretty sure they use the git head from msm-2.6.27 (at least the later releases.. the earlier ones were 2.6.26) |
| 09:01.50 | duncang | OK. Who built the kernel for http://connect-utb.com/index.php?option=com_jdownloads&Itemid=58&task=view.download&cid=24 ? |
| 09:02.13 | duncang | That's my current known good. |
| 09:03.12 | duncang | I'm just going to try the kernel from the 20090316 "Raph 800" release too. |
| 09:03.39 | Untouchab1e | Hmm.. I think it was Majrep who did the 20090219 |
| 09:03.55 | Untouchab1e | the 20090316 will boot on the Raph100 if you edit the default.txt a bit.. |
| 09:04.06 | duncang | Is he on this channel? |
| 09:04.14 | Untouchab1e | He used to be online 24/7 |
| 09:04.23 | Untouchab1e | But holidays might have gotten the better of him right now |
| 09:04.26 | Untouchab1e | he is also from the US |
| 09:05.15 | Untouchab1e | Since the only thing that was significantly different on the latest "Raph 800" build was the color fix, I decided to throw in a raph800 specific default.txt instead of a Raph100 one as I normally do |
| 09:05.43 | Untouchab1e | But are you just trying to get Android booted? |
| 09:05.53 | mickeyl | duncang: HEAD is broken for Raph |
| 09:05.57 | mickeyl | fall back to HEAD^3 |
| 09:05.59 | mickeyl | or ^4 |
| 09:06.01 | mickeyl | last known good |
| 09:06.27 | mickeyl | a7bbc84bc783acc349b030a4621541eab84165e0 |
| 09:08.40 | duncang | mickeyl: Think I have build my own for almost every commit between current head and: committer Martijn Stolk Wed, 18 Feb 2009 21:56:57 +0000 (22:56 +0100) commit ed619deca2d062ba190f7e79aafcd8428559ccac tree 7f6b8e5f4bba172b0bee42070d20aa23b5267ab1 |
| 09:08.59 | duncang | goes to check the konsole window with the git logs in... |
| 09:10.08 | duncang | Yeah; lots of "git reset HEAD~1" in the history. Let me find the kernel you're recommending. |
| 09:10.56 | duncang | F*** me. The 20090316 kernel hadn't hung at console handover. Android is now booting... |
| 09:11.23 | mickeyl | "Update the MSM clock API" breaks it |
| 09:11.35 | mickeyl | head^4 |
| 09:13.07 | duncang | I need to be sure a kernel I built works before I start breaking things. Just fast-forwarding my old tree now. |
| 09:15.51 | Untouchab1e | duncang, it would be awesome if you reported a summary of your development progress over at htc-android.com. Lavender and me are trying to get some development documentation going there to make it easier for new developers coming in |
| 09:16.00 | Untouchab1e | anyways, I need to get going.. :) bbl |
| 09:16.42 | duncang | Untouchable: Are you co-ordinating with htc-linux.org? We've been transplanting dev docs from xda-dev's wiki there. |
| 09:16.59 | Untouchab1e | Yeah, or In the process of doing so |
| 09:16.59 | duncang | tuts. |
| 09:17.29 | Untouchab1e | Thanks for the tip |
| 09:17.39 | Untouchab1e | See you soon |
| 09:18.08 | duncang | Not sure who has the lock on the main page (Mickey?). I am adding to the RaphealResearch section. |
| 09:18.41 | mickeyl | no, not me |
| 09:19.25 | duncang | Are you the Mickey who is the other registered user? |
| 09:20.22 | mickeyl | yep |
| 09:21.55 | mickeyl | can you enable file uploads? |
| 09:22.16 | duncang | I am not Admin. |
| 09:22.24 | duncang | I don't know who is. |
| 09:22.24 | mickeyl | k, will pester dcordes then |
| 09:53.32 | *** join/#htc-linux hollo_ (n=hollo@3e6b7b2c.rev.stofanet.dk) |
| 09:55.51 | duncang | Victory! |
| 09:55.51 | duncang | bash-3.2# dmesg |
| 09:55.51 | duncang | <5>[ 0.000000] Linux version 2.6.27-00248-ga7bbc84-dirty (duncan@edna) (gcc version 4.3.3 (Debian 4.3.3-6) ) #8 PREEMPT Sat Apr 11 10:49:22 BST 2009 |
| 09:55.51 | duncang | <4>[ 0.000000] CPU: ARMv6-compatible processor [4117b362] revision 2 (ARMv6TEJ), cr=00c5387f |
| 09:55.51 | duncang | <4>[ 0.000000] Machine: HTC Raphael GSM phone (aka HTC Touch Pro) |
| 10:02.29 | mickeyl | which git rev? |
| 10:04.02 | duncang | mickeyl: a7bbc84bc783acc349b030a4621541eab84165e0 plus manual http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=commitdiff;h=12008ce2c5c1ab657c46fdc6f6129ccf6730760a (turn off Blackstone) |
| 10:04.35 | mickeyl | ah, known good then |
| 10:04.46 | duncang | Is anyone maintaining a Raph branch where things that don't boot on our devices don't get committed? |
| 10:04.56 | mickeyl | I'm afraid not |
| 10:05.19 | mickeyl | i guess the consensus is that the devices are so similar that this would be contraproductive |
| 10:05.25 | mickeyl | i wonder whether this is true though |
| 10:05.38 | mickeyl | since I'm not a kernel guy, I refrain from commenting though |
| 10:06.14 | mickeyl | apparantly, no kernel developer has access to more than just one or two of those devices, that's a major problem |
| 10:06.32 | duncang | Yes, known good source, but the first time I've successfully booted something I built. Hence proof my toolchain is OK. |
| 10:07.39 | mickeyl | ah, you have a custom toolchain? |
| 10:07.42 | mickeyl | using OE |
| 10:07.53 | duncang | Since Untouchable is sure it's the clock_wince API changes that break the Raphael, it should be possible to (a) merge the other changes, and (b) figure out why that change won't boot. |
| 10:09.11 | duncang | Toolchain is standard armel-cross from Debian mostly-Squeeze (gcc version 4.3.3 (Debian 4.3.3-6)) rather than CodeSourcery's. |
| 10:10.55 | duncang | (ie it is newer and not a random binary from the internet) |
| 10:14.55 | tmzt | I think cr2 said that was the cause |
| 10:15.15 | tmzt | or somebody, it's a regression though |
| 10:15.37 | tmzt | mickeyl: what do you recommend for tracking these kind of things? |
| 10:15.55 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 10:18.11 | duncang | I'm going to pull HEAD again and then revert just the clock-wince changes from the commit that broke my world. |
| 10:18.16 | tmzt | maybe CIA? |
| 10:19.07 | mickeyl | well, as these kind of things are not testable without the actual hardware, i'd recommend adding a 'testing' branch where folks commit new patches -- plus having a mentoring crew, one mentor for every machine. only if all mentors have successfully tested the change, it would be pushed into master |
| 10:19.24 | mickeyl | i do think though that one branch per machine slows development down too much |
| 10:19.32 | mickeyl | so this mentoring system would be something in between |
| 10:19.56 | mickeyl | CIAbot shoudl be present anyways, yeah |
| 10:20.01 | mickeyl | it's wonderful |
| 10:21.58 | tmzt | I don't think this project/channel has had much infrastructure |
| 10:22.14 | tmzt | everything was hh.org at one point, cvs |
| 10:22.27 | tmzt | now with ltg and git that makes sense |
| 10:23.36 | mickeyl | yep |
| 10:24.34 | mickeyl | i do consider HTClinux as being one of the most promising anti-vendor-ports, hence it's vital for the future of mobile linux. and as such it should be upleveled to gain more visibility and "professionalism" |
| 10:25.41 | tmzt | a lot of that is thanks to google/android |
| 10:26.03 | tmzt | which is working with a vendor (htc) to develop a phone |
| 10:26.19 | tmzt | but can't assist directly or something |
| 10:26.55 | mickeyl | i know |
| 10:27.09 | mickeyl | still in essence what we are doing here is not much different from an anti-vendor-port |
| 10:27.16 | mickeyl | only that we get a kernel-kickstart |
| 10:27.23 | tmzt | wasn't sure of the anti-vendor thing |
| 10:27.30 | tmzt | ok |
| 10:28.04 | mickeyl | i love using this term to highlight that without direct support from the vendor, what we are doing is more or less against the vendor's interest, hence the -- maybe offensive -- term anti-vendor-port :) |
| 10:28.23 | tmzt | selling their phones? |
| 10:28.45 | mickeyl | reverse engineering their protection |
| 10:28.45 | tmzt | I like stated or implied interest |
| 10:29.00 | mickeyl | "freeing" the device |
| 10:29.11 | mickeyl | if they wanted they would have released info, so IMO it's anti. |
| 10:29.19 | mickeyl | but that's details |
| 10:29.21 | mickeyl | policits |
| 10:29.28 | tmzt | not in the sense of lawyer-client "in the client's best interest" |
| 10:29.42 | mickeyl | hehe |
| 10:29.53 | duncang | It's in the vendor's interest to sell more widgets. Hence in their interest to have more options for OEMs and users wrt OS. |
| 10:30.15 | duncang | Since HTC make the G1, they must already have upset M$ as much as they can ;-) |
| 10:30.53 | tmzt | Microsoft get some amount of money in license |
| 10:30.55 | mickeyl | well, lets talk again, when you need some info for an asic or something |
| 10:30.58 | *** join/#htc-linux tsdogs (n=tsdogs@net203-187-146.mclink.it) |
| 10:30.59 | tmzt | for software |
| 10:31.00 | mickeyl | and then lets see how open you find HTC |
| 10:31.02 | mickeyl | *shrug* |
| 10:31.32 | tmzt | I'm not saying I find them open |
| 10:31.32 | mickeyl | we have been trying to get asic info from MOTO for years |
| 10:31.42 | mickeyl | but they just lough their asses off upon us |
| 10:32.05 | mickeyl | even if the IP used for the chip is ages old |
| 10:32.11 | mickeyl | and no longer relevant for competitors |
| 10:32.13 | tmzt | moto has evoke qa4, trying to make magx a mainstream platform |
| 10:32.30 | tmzt | but the kernel is very odd |
| 10:32.39 | duncang | Motorola are founder members of LiMo, so have their penguin eggs in that basket. |
| 10:33.59 | tmzt | I see it as OM has enabled the software part and android has enabled the hardware part |
| 10:34.02 | tmzt | currently |
| 10:34.06 | mickeyl | hmm |
| 10:34.09 | duncang | needs entroy to use scp... |
| 10:34.12 | mickeyl | i completely disagree wrt hardware |
| 10:34.24 | mickeyl | i don't see android bringing us more free devices |
| 10:34.44 | tmzt | I'm using android as google said to use that to refer to the organization |
| 10:35.09 | tmzt | well, I think qualcomm soc's is the future, maybe omap3 |
| 10:35.21 | tmzt | and later tegra and x86 |
| 10:36.00 | tmzt | but that we won't be going back to pxa or devices with fully open (sufficeintly) documentation |
| 10:36.06 | duncang | would also like kexec boot under Android... |
| 10:36.24 | tmzt | msm is not the ideal situation |
| 10:38.02 | *** join/#htc-linux MethoS (n=lem@dyndsl-085-016-163-139.ewe-ip-backbone.de) |
| 10:39.04 | *** join/#htc-linux hollo_ (n=hollo@3e6b7b2c.rev.stofanet.dk) |
| 10:39.35 | duncang | Booting Android with zImage_34cb6db2f11843fd561d034e5f0066f11e80bedb_minus_clock_wince... |
| 10:40.19 | tmzt | hey sorry, it's not that the android os has made hardware free, it's that the kernel source has given us the opportunity to use the OM software on COTS hardware |
| 10:40.44 | tmzt | which we otherwise would not have |
| 10:41.27 | mickeyl | sure sure |
| 10:41.32 | mickeyl | i agree with that point |
| 10:41.41 | mickeyl | (that's what i was referring to with kickstart) |
| 10:41.53 | mickeyl | still it's unclear whether android will damage the movement of free software more than it helps |
| 10:42.07 | mickeyl | and it's also still unclear whether we will manage to complete the kernel towards standard interfaces |
| 10:42.17 | mickeyl | and the blobs |
| 10:42.50 | mickeyl | but again, I'm grateful for all we have so far, it's just that we have been so close for years now |
| 10:42.55 | mickeyl | but we're still not there :) |
| 10:43.09 | mickeyl | no matter whether htclinux, openezx, or what not |
| 10:43.11 | tmzt | swetland has made a great effort to keep binary components out of the kernel memory space on arm11 core |
| 10:43.17 | mickeyl | swetland? |
| 10:43.29 | tmzt | Brian Swetland at Google |
| 10:43.33 | mickeyl | ah |
| 10:45.06 | tmzt | there are still components that mmap physical memory from userspace |
| 10:45.18 | tmzt | but kernel parts are gpl source |
| 10:45.33 | tmzt | and those are not required to boot/use the kernel |
| 10:45.51 | tmzt | not even with android which is apache 2 licensed |
| 10:46.02 | tmzt | Android Open Source Project |
| 10:47.40 | tmzt | the other binary components are the microkernel that runs on arm9 core |
| 10:48.02 | tmzt | known as Anvanced Mobile Subscriber Software or AMSS |
| 10:48.28 | tmzt | this is essentially the equivalent of calypso firmware |
| 10:48.52 | mickeyl | right |
| 10:48.57 | tmzt | but it also serves some vital hardware related functions on the g1 |
| 10:49.10 | tmzt | such as vregs, gpios and clocks |
| 10:49.13 | duncang | Such as? |
| 10:49.16 | duncang | Ah. |
| 10:49.36 | mickeyl | and the only communication is some undocumented command langauge? |
| 10:49.44 | mickeyl | differing from device to device? |
| 10:49.52 | tmzt | this was done in part to protect disclosure of details of qualcomm hardware details |
| 10:50.12 | duncang | Grrr... Idiocy. |
| 10:50.21 | tmzt | the g1 version is somewhat documented in kernel source |
| 10:50.30 | mickeyl | hmm |
| 10:50.36 | tmzt | the ce version is not documented though |
| 10:50.37 | mickeyl | that doesn't make me hope for everythign != g1 |
| 10:50.59 | mickeyl | i have seen so many uncomplete ports since 2000 |
| 10:50.59 | tmzt | so tracing of the read/writes was required |
| 10:51.06 | ali12341 | all the omap devices use the same radio core |
| 10:51.14 | tmzt | hey ali |
| 10:51.15 | ali12341 | and i mean exactly the same |
| 10:51.18 | ali12341 | hi :) |
| 10:51.26 | mickeyl | bbl, need to run some errands |
| 10:54.58 | *** join/#htc-linux DJW|Home (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) |
| 11:06.21 | *** join/#htc-linux captnoord (i=4da104ed@gateway/web/ajax/mibbit.com/x-d2b2578c9adaf84c) |
| 11:07.36 | *** join/#htc-linux Untouchab1e (n=Untoucha@cm-84.209.89.86.getinternet.no) |
| 11:14.31 | *** join/#htc-linux DJW|Home (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) |
| 11:15.30 | duncang | has to do some DIY. Back later. |
| 11:15.45 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) |
| 11:24.50 | Untouchab1e | So we are moving it all over to htc-linux.org then |
| 11:25.09 | Untouchab1e | who would be person of contact there? |
| 11:25.42 | Untouchab1e | Maybe linking it up to the forums at htc-android.com? |
| 11:31.48 | elf | i think i may have accidentally just found the docs on the 802.11 chip for the raph800 by the backend manufacturer for TI. |
| 11:32.24 | elf | or at least some of em. |
| 11:33.01 | Marajin | morning |
| 11:33.04 | Marajin | well lunchtime |
| 11:33.05 | elf | mornin. |
| 11:33.07 | Marajin | hey elf |
| 11:33.09 | Untouchab1e | hi |
| 11:33.15 | elf | 'lo. |
| 11:33.40 | Marajin | elf you been haxing into something? :p getting docs for some of those chips is nigh impossible |
| 11:33.43 | Untouchab1e | docs say anything interesting? |
| 11:34.10 | elf | marajin: nothing that stupid, no. |
| 11:34.18 | elf | may have being operative. |
| 11:34.28 | elf | and its not full specs, just some diagrams and control lines. |
| 11:34.50 | elf | im still trying to read it and seeing if its the same chip. |
| 11:34.53 | Marajin | Every little helps |
| 11:35.09 | elf | yeah, if it is the same chip it will help a bit. |
| 11:36.01 | *** join/#htc-linux Pure4Real (n=pure4rea@93-45-193-152.ip103.fastwebnet.it) |
| 11:36.03 | elf | yeah, it is :) |
| 11:36.08 | Untouchab1e | Anyone of you have anything to do with the htc-linux.org wiki? |
| 11:36.12 | Untouchab1e | or htc-linux.org in general? |
| 11:36.19 | elf | on at least some of em. |
| 11:38.10 | Marajin | Untouchab1e: dcordes does I /think/ but don't quote me on it |
| 11:38.35 | Untouchab1e | ok, I will contact him :) |
| 11:38.42 | Untouchab1e | what timezone is he in again? |
| 11:38.54 | Marajin | well if he doesn't he'd probably know who does at least |
| 11:38.58 | Marajin | erm, GMT+! |
| 11:39.03 | Marajin | *GMT+1 |
| 11:39.27 | Untouchab1e | ah ok |
| 11:39.30 | Untouchab1e | so not US |
| 11:40.31 | Marajin | nope |
| 11:40.33 | Marajin | german AFAIK |
| 11:40.40 | Untouchab1e | Ah ok |
| 11:40.54 | Marajin | 'Not that timezones are that important to geeks, none of us seem to actually live on our own timezone |
| 11:41.09 | elf | how many non-htc devices use the msm7xx chipsets? |
| 11:41.11 | Untouchab1e | hahah |
| 11:41.15 | Untouchab1e | true, true |
| 11:41.18 | elf | (or the qsc chipsets) |
| 11:44.05 | Untouchab1e | Well, cool that you got the docs |
| 11:44.17 | Untouchab1e | is it a hard copy, or digital? (pdf or whatever?) |
| 11:52.42 | elf | pdf |
| 11:53.13 | elf | its not necessarily going to be useful, but it looks like it may be useful for something. |
| 12:42.33 | *** join/#htc-linux cr2 (n=cr2@ip-90-186-90-30.web.vodafone.de) |
| 12:47.34 | *** join/#htc-linux Marajin_ (n=marajin@87-194-102-189.bethere.co.uk) |
| 12:48.35 | *** join/#htc-linux fnord__ (n=fnord@24-151-90-116.static.nwtn.ct.charter.com) |
| 13:04.25 | *** join/#htc-linux _RZK333 (n=rzk@daemonet.ru) |
| 13:16.50 | *** join/#htc-linux timebomb (n=tb@d016182.adsl.hansenet.de) |
| 13:32.44 | *** join/#htc-linux ellisway (n=ellis@80-46-67-47.static.dsl.as9105.com) |
| 13:34.43 | *** join/#htc-linux MethoS- (n=lem@dyndsl-085-016-164-100.ewe-ip-backbone.de) |
| 13:41.34 | *** join/#htc-linux Marex (n=vasum7am@auryn.karlin.mff.cuni.cz) |
| 13:41.37 | Marex | cr2: hi |
| 13:41.54 | Marex | cr2: did you got my botmail ? |
| 13:43.11 | *** join/#htc-linux fnord__ (n=fnord@24-151-90-116.static.nwtn.ct.charter.com) |
| 13:57.17 | *** join/#htc-linux zycho (n=zycho@a89-182-23-78.net-htp.de) |
| 14:17.42 | Untouchab1e | hi all |
| 14:20.19 | Marex | hello |
| 14:21.48 | Untouchab1e | Its great that we are now putting some development information in one place, and I still believe though that a forum would be a good suplement to a wiki. Lavender and I thought that the htc-android.com forums could be a nice arena to comunicate on. |
| 14:21.56 | Untouchab1e | so maybe we could connect them as a forum and wiki? |
| 15:00.47 | ellisway | hi all |
| 15:06.52 | Marajin_ | yo |
| 15:08.01 | ellisway | hi Marajin |
| 15:12.32 | mickey|bbl | Untouchab1e: sounds good, as long as you don't give the false impression that htclinux is just about porting android. |
| 15:13.23 | Untouchab1e | yeah, I know |
| 15:16.31 | *** join/#htc-linux cr2 (n=cr2@ip-90-186-90-30.web.vodafone.de) |
| 15:23.46 | Marajin | lots of people getting uptight that most of the conversation in here seems to be about android yeah |
| 15:27.48 | ellisway | htc-linux i should have rreally got htc-nix |
| 15:52.29 | *** join/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey) |
| 15:53.38 | mickey|bbl | not uptight |
| 15:53.38 | mickey|bbl | sad |
| 16:00.18 | *** join/#htc-linux dcom (n=adsas@p54A84E34.dip.t-dialin.net) |
| 16:33.44 | ali1234 | every time i hear "porting android" it makes me cringe. what are you porting it to? win32? bsd? darwin? |
| 16:41.00 | *** join/#htc-linux groton (n=groton@unaffiliated/groton) |
| 16:50.50 | ali1234 | hi groton :) |
| 16:51.21 | groton | hi ali1234 |
| 16:51.29 | groton | guess i met you on c64? |
| 16:51.48 | ali1234 | ##amiga |
| 16:51.55 | ali1234 | are you porting android to amigaos? |
| 16:52.42 | groton | oh no |
| 16:53.20 | groton | i bought an htc diamond touch and perhaps i could upload on it a linux kernel |
| 16:54.41 | ali1234 | i think there is a kernel for that somewhere. it's msm though, and i only know about omap. |
| 16:55.43 | groton | msm? omap? :) |
| 16:56.24 | ali1234 | cpu make |
| 16:56.34 | *** join/#htc-linux iratsu (n=iratsu@modemcable018.234-59-74.mc.videotron.ca) |
| 16:57.15 | groton | ok |
| 16:57.32 | duncang | returns from the DIY vortex... |
| 16:57.57 | duncang | Untouchable: you still here? |
| 16:58.00 | ali1234 | groton: the first thing to do is find out all the different names of your phone, of which there are probably many. http://wiki.xda-developers.com/ can help with that (it has lots of photos) |
| 16:58.24 | groton | ok, thx |
| 17:02.25 | Untouchab1e | Im here, barely :) |
| 17:02.33 | groton | it is called Diamond :) Pretty much as its brand name :) |
| 17:06.15 | duncang | In case you missed it, I did as you suggested (approximately). msm-2.6.27 HEAD, except clock-wince.c from HEAD^4 built with normal Debian cross-gcc boots nicely on my raph120. |
| 17:06.38 | duncang | Need to debug the clock-wince changes.. |
| 17:07.52 | duncang | Also, some hours ago you were talking about forums. I /really/ think we need to use a mailling list with properly searchable archives. |
| 17:09.17 | groton | web like forums are really a mess to deal with, instead mailing lists are much easier to read (could delete useless posts, mark what is more important for you, and have a quick look to tens of post threads together) |
| 17:09.24 | duncang | I suggested yesterday we invade http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/mobile-linux-discuss which is quiet and has many of the msm-* committers. |
| 17:11.23 | duncang | I haven't seen the state of the Android forums, but xda-dev's are awash with clueless newbies^W^Wenthusiastic potential users. |
| 17:12.07 | duncang | We need to get the vital bits of the kernel off the ground (battery, sound). |
| 17:21.33 | mickeyl | yep. as soon as we can show folks a viable lternative to WM -- at least something that makes calls with sounds and sends/receives sms -- we're in the game |
| 17:22.00 | duncang | Send and receive SMS is working for me on raph120. |
| 17:22.22 | duncang | And I can place and receive calls - just the lack of audio that's a problem ;-) |
| 17:22.34 | mickeyl | audio, power management, suspend/resume |
| 17:22.59 | mickeyl | oh, and gprs is absent |
| 17:23.07 | mickeyl | at least on some models |
| 17:23.09 | duncang | I'd settle for just charging the battery at the moment. |
| 17:23.22 | mickeyl | yes, charging would be a good start |
| 17:23.44 | duncang | is currently transplanting MSM_VREG page to htc-linux.org |
| 17:26.26 | *** join/#htc-linux ClashTheBunny (i=47b8a74a@gateway/web/ajax/mibbit.com/x-7e1dab03274fd99e) |
| 17:57.59 | *** join/#htc-linux timebomb (n=tb@C6660.c.ppp-pool.de) |
| 18:14.42 | *** join/#htc-linux kiozen (n=oeichler@rgnb-5d87450e.pool.einsundeins.de) |
| 18:15.01 | cr2 | duncang: the clock-wince changes need to be extended, there are some bugs there |
| 18:18.20 | duncang | cr2: No kidding, but it's nice to know everything else it git head works on my device. |
| 18:18.46 | duncang | Sadly I have to go now. May be able to do more on Monday. |
| 18:50.27 | *** join/#htc-linux Untouchab1e (n=Untoucha@cm-84.209.89.86.getinternet.no) |
| 18:54.04 | *** join/#htc-linux drw-r--r-- (n=kvirc@78-105-175-81.zone3.bethere.co.uk) |
| 19:03.58 | *** join/#htc-linux groton (n=groton@unaffiliated/groton) |
| 19:29.18 | *** join/#htc-linux noodlesgc (n=noodles@97.72.68.195) |
| 19:45.58 | *** join/#htc-linux chab7 (n=kvirc@catv91EC01FC.pool.t-online.hu) |
| 19:48.08 | *** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
| 19:52.54 | *** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
| 20:09.15 | *** join/#htc-linux |chab7| (n=kvirc@catv91EC01E4.pool.t-online.hu) |
| 20:10.25 | *** part/#htc-linux exception13 (n=exceptio@testdrive.kgts.ru) |
| 20:12.15 | *** join/#htc-linux _chab7_ (n=kvirc@catv91EC01E6.pool.t-online.hu) |
| 20:13.20 | *** join/#htc-linux _chab7_3 (n=kvirc@catv91EC01FC.pool.t-online.hu) |
| 20:42.08 | Marex | cr2: wake up ;) |
| 20:42.48 | *** join/#htc-linux Untouchab1e (n=Untoucha@cm-84.209.89.86.getinternet.no) |
| 20:45.42 | *** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 21:36.11 | *** join/#htc-linux Marex (n=vasum7am@u-pl15.ms.mff.cuni.cz) |
| 21:36.49 | Marex | cr2: ping |
| 21:36.58 | Marex | he has to be sleeping like a log dammit |
| 21:53.40 | Marajin | Marex: He always does! |
| 21:54.42 | Marex | no he doesnt |
| 22:00.56 | Marex | Marajin: I'll tinker with him a little bit more, maybe I can wake him up ;) |
| 22:02.25 | cr2 | hey Marex |
| 22:03.17 | cr2 | Marex: it's a good idea to go into deep sleep, but imho you can't really do it |
| 22:07.31 | *** join/#htc-linux infernix (i=nix@unaffiliated/infernix) |
| 22:25.45 | *** join/#htc-linux osnr (n=osnr@ool-4351ae8c.dyn.optonline.net) |
| 22:25.59 | tmzt | Unt: have you seen zinx cupcake build in #android? |
| 22:35.07 | Marex | cr2: hi |
| 22:35.27 | Marex | cr2: what do you mean ? you can suspend the DoCG4 with "some magic sequence" |
| 22:35.34 | Marex | cr2: that's also the key to resume |
| 22:35.57 | Marex | (or you can rip away the DoCG4 chip while the device is asleep, it'd have the same effect ;) ) |
| 22:36.16 | Marex | the bootloader checks the DoCG4 device ID register |
| 22:36.21 | Marex | if it gets valid value, it resumes |
| 22:37.46 | cr2 | i think deep sleep is a chip pin |
| 22:38.17 | cr2 | we don't know if it's connected, or where it is connected |
| 22:38.37 | Marex | that's just one of the options |
| 22:38.49 | cr2 | if you will rip away g4, you will know :) |
| 22:39.33 | Marex | cr2: according to the manual, you can enter DPD without the pin being connected |
| 22:39.38 | cr2 | i think that reading out the spl from mdoc is an easier task |
| 22:39.40 | *** join/#htc-linux osnr_ (n=osnr@ool-4351ae8c.dyn.optonline.net) |
| 22:40.06 | cr2 | Marex: ok, then try it :) |
| 22:40.14 | Marex | see mdocg4_spec.pdf Chapter 6.3 |
| 22:40.42 | *** join/#htc-linux dzo (n=dzo@121-98-128-127.bitstream.orcon.net.nz) |
| 22:40.59 | Marex | cr2: it jumps into the DoCG4 only in case the chips isn't in DPD mode ... so there is only one option for wakeup - put the chip into DPD |
| 22:41.19 | Marex | *chips |
| 22:41.45 | Marex | also ... if the DPD pin is connected, it has to be connected to some EGPIO as PXA GPIO registers are set before reading the chips |
| 22:42.49 | Marex | I assume WM puts g4 into DPD mode and on wakeup it powers it up again |
| 22:43.09 | Marex | if the device was powercycled though, the g4 is powered-up and it jumps into SPL |
| 22:43.19 | cr2 | what is the DPD sequence ? |
| 22:43.33 | Marex | cr2: that's what I dont know |
| 22:43.40 | Marex | I thought you might know that |
| 22:43.43 | cr2 | this sounds good |
| 22:44.02 | cr2 | no, i have only read the docs |
| 22:44.14 | cr2 | hmm. where are all these docs... |
| 22:44.26 | cr2 | i want to upload all of them to htc-linux.org |
| 22:44.35 | Marex | cr2: well ... in qemu (without DoCG4), it shows very similar pattern to other bootloaders I saw |
| 22:44.53 | Marex | if I "emulate" DoCG4 also, it jumps into SPL then |
| 22:45.12 | Marex | so I assume if the G4 is in DPD, the wakeup process would be very simple |
| 22:45.38 | Marex | now the only thing needed is to figure out how to put the sleeping beauty DoCG4 into DPD-mode |
| 22:46.22 | Marex | cr2: btw. do you use jabber? |
| 22:46.30 | cr2 | yes |
| 22:46.37 | Marex | cr2: can you PM me your jid ? |
| 22:46.37 | cr2 | n |
| 22:46.41 | Marex | why? |
| 22:46.58 | cr2 | no time for weird fancy stuff |
| 22:47.27 | Marex | cr2: you gave up hacking on N560 ? |
| 22:47.54 | cr2 | well, i don't have the device, and it a difficult task. |
| 22:48.30 | cr2 | but i can help with many things while i still remember something about it. |
| 22:49.03 | Marex | cr2: that's why I asked you for your JID ... since you dont respond here much :-B |
| 22:49.29 | Marex | well I need the sequence to enter DPD-mode, that's all ;-) |
| 22:56.14 | Marex | darn :E what the hell is "Proper sequence must be writted into DiskOnChip" :E |
| 23:13.25 | *** join/#htc-linux WarptwistDK (n=chatzill@0x4dd49295.adsl.cybercity.dk) |
| 23:13.48 | WarptwistDK | Can i get linux on my HTC Touch ELF? |
| 23:18.42 | *** join/#htc-linux dzo (n=dzo@121-98-128-127.bitstream.orcon.net.nz) |
| 23:25.36 | tmzt | WarptwistDK: somewhat, that's omap though |
| 23:25.54 | tmzt | not the same as cdma touch (vogue) which is msm7200 |
| 23:27.09 | WarptwistDK | How do i do it then? any guides or something? |
| 23:31.55 | tmzt | try the image from excalibur forums on forum.xda-developers.com |
| 23:32.23 | tmzt | I don't know if there is a cupcake for omap |
| 23:32.36 | tmzt | you can search for wizard and android |
| 23:32.54 | tmzt | you also need the kernel (zImage) from linwizard |
| 23:44.07 | *** join/#htc-linux MethoS (n=lem@dyndsl-085-016-164-100.ewe-ip-backbone.de) |
| 23:46.42 | *** join/#htc-linux dzo (n=dzo@121-98-128-127.bitstream.orcon.net.nz) |
| 23:55.12 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) |