| 00:00.00 | BruteForce | wow this keypad sucks |
| 00:00.28 | tmzt | can't telnet in case the telnetd won't start even before init |
| 00:00.30 | BabelO | cr2_: what do you think about samsung addic ? it is wince 6.1 ... i ve to check it |
| 00:00.39 | BruteForce | cant telnet in because telnetd wont start even before init |
| 00:00.48 | tmzt | oh |
| 00:01.27 | BabelO | cr2_: http://fr.samsungmobile.com/pid47-step2-cell_idSGH-I900/specifications.html |
| 00:02.22 | tmzt | blackjack? |
| 00:02.25 | BruteForce | dont really have a way to troubleshoot that |
| 00:02.36 | tmzt | on android? |
| 00:02.46 | BruteForce | ya |
| 00:03.03 | tmzt | is it bringing up usb0? |
| 00:03.21 | BruteForce | i guess wecan patch the linwizard kernel |
| 00:03.33 | BabelO | cr2_: better here http://www.pdadb.net/index.php?m=specs&id=1328 |
| 00:03.34 | BruteForce | and use that i know telnet works on it |
| 00:03.54 | tmzt | what kernel are you using now? |
| 00:04.02 | BruteForce | ya usb is workig fine just telnetd |
| 00:04.26 | BruteForce | i am not sre what he is using, sonething that was easier to patch for android |
| 00:04.38 | tmzt | htc-omap? |
| 00:04.57 | BruteForce | even when we get it working no one is working on the gsm drivers for omap850 |
| 00:04.58 | ponicg | heh, the first time always takes forever:) |
| 00:05.00 | tmzt | if usbnet is working, how could it be kernel? |
| 00:05.04 | BruteForce | i dont know tmt |
| 00:05.26 | BruteForce | its not usbnet its telnetd |
| 00:06.39 | BruteForce | dont know how its kernel related but the same kernel with the linwizad userspace telnetd works fine |
| 00:06.56 | cr2_ | BabelO: Ecran principal : Résolution 240 x 400 pixels |
| 00:07.22 | BabelO | cr2_: yes that's the bad part, a bit lower |
| 00:07.27 | cr2_ | pxa312 is ok |
| 00:07.53 | BruteForce | but hey he fixed the framebuffer problem so thats good |
| 00:08.47 | BruteForce | i want to know how my device can beam and use irda with no visible ir port |
| 00:08.50 | cr2_ | BabelO: how are they doing umts ? |
| 00:09.43 | BabelO | cr2_: i don't know, it seem that it support GPRS + EDGE + 3G + 3G+ |
| 00:09.52 | tmzt | BruteForce: what? don't think you have irda |
| 00:09.52 | BabelO | cr2_: i think i have to run haret on it :) |
| 00:10.05 | cr2_ | BabelO: sure :) |
| 00:10.21 | BruteForce | tmzt phonescoop says i do lol |
| 00:11.51 | tmzt | you have rubber coating around almost the whole thing, right? |
| 00:12.40 | BruteForce | ya lol i cant find an irport |
| 00:13.53 | BruteForce | im going to find wmirc and see if its any better |
| 00:13.57 | cr2_ | good night |
| 00:14.51 | BabelO | good night |
| 00:15.02 | tmzt | have you tried zairc? |
| 00:15.58 | tmzt | zsirc |
| 00:16.22 | BruteForce | no never heard of i |
| 00:16.55 | tmzt | wicked in #hpcdev uses it on blackjack, I had that in my logs when we talked about it |
| 00:17.23 | BruteForce | doesnt seem to have a smartphone version just ppc |
| 00:19.21 | tmzt | ok, he was testing it, not uses it |
| 00:19.50 | tmzt | if have net right now anyway, so I can try it. |
| 00:20.12 | tmzt | and zaTelnet again |
| 00:20.39 | BruteForce | wow what the heck? i have some weird icon i have never seen in the top bar its red and orange kinda resembles a battery with some kinda bared meter next to it? |
| 00:26.54 | *** join/#htc-linux BruteSource (n=BruteSou@m7d0e36d0.tmodns.net) |
| 00:32.20 | *** join/#htc-linux BruteSource (n=BruteSou@m3d0e36d0.tmodns.net) |
| 00:32.33 | BruteSource | oh i forgot its not free |
| 00:33.46 | BruteSource | but its defenitly better even supports logging |
| 00:33.52 | *** join/#htc-linux fnord (n=fnord@24-151-90-116.static.nwtn.ct.charter.com) |
| 00:34.28 | BruteSource | <PROTECTED> |
| 00:35.07 | BruteSource | tmzt still here? |
| 00:35.41 | tmzt | yeah |
| 00:36.46 | BruteSource | hows your q kernel? |
| 00:41.01 | tmzt | we found out that it is very much like ezx (motorola linux) phones |
| 00:41.20 | tmzt | and can load a kernel over usb, but it doesn't start running |
| 00:41.29 | BruteSource | why |
| 00:42.43 | tmzt | might be signed, might be bad protocol/chunk size |
| 00:42.50 | tmzt | hope it's not signed |
| 00:42.59 | BruteSource | is motorolas regular os for razr linux? |
| 00:43.29 | tmzt | older than razr2v8 |
| 00:43.52 | tmzt | ming, really cool flip cover touchcreen phone 2+ years before iPhone/prada |
| 00:46.10 | BruteSource | what do the newer ones use? |
| 00:46.45 | tmzt | motomagx |
| 00:47.18 | tmzt | ezx also includes a few other touchscreen phones with t9 |
| 00:47.45 | tmzt | keypad |
| 00:47.49 | BruteSource | motomagx isnt nix? |
| 00:47.56 | *** join/#htc-linux OpnSrc1 (i=harisha1@gateway/tor/x-53e0e74436c51a64) |
| 00:48.06 | tmzt | yes, motomagx is linux but it's locked down |
| 00:48.12 | BruteSource | wow this client supports multiple servers as well |
| 00:48.17 | tmzt | razr2v9 uses something else |
| 00:48.21 | BruteSource | o i c |
| 00:54.04 | *** join/#htc-linux PoohbaLT (n=Poohba@c-98-235-52-97.hsd1.nj.comcast.net) |
| 01:03.22 | BruteSource | hi poo |
| 01:24.17 | szsoftware | cr2_: still around? |
| 01:24.51 | szsoftware | cr2_: ahh ok, already dreaming .. |
| 01:25.38 | szsoftware | tmzt: could you help me eventually? kernel related, maybe quick-fix |
| 01:25.56 | tmzt | I can try |
| 01:26.08 | szsoftware | pastebin follows |
| 01:29.08 | szsoftware | http://pastebin.com/m4f1e373b |
| 01:35.18 | tmzt | I sounds like someone commited something wrong |
| 01:38.13 | szsoftware | what do you mean? When I do make htckaiser_defconfig, it compiles fine, with make htckaiser_defconf and changing phone to polaris (what IMHO only changes board type ) then this error appears |
| 01:38.31 | tmzt | the arch_idle thing |
| 01:38.49 | szsoftware | dcordes and I adapting existing patches to kaiser respectivly to polaris |
| 01:38.53 | par | cr2_: with the original 2577 autorun the haret in the /linux partition should simply be named haret.exe , correct? |
| 01:39.31 | szsoftware | tmzt: any hint how to fix that? |
| 01:40.17 | tmzt | grep -r arch_idle arch/arm/mach-msm, look at examples make your board the same |
| 01:40.46 | tmzt | it might be .idle is not set in you mach definition at the end of your board |
| 01:40.56 | tmzt | can you paste the board file? |
| 01:41.06 | szsoftware | tmzt: I reviewed the patches and their success and I guess, I JUST FOUND THE SOLUTION |
| 01:52.31 | *** join/#htc-linux ALoGeNo (n=QUAKEIII@195.Red-81-32-138.dynamicIP.rima-tde.net) |
| 01:52.37 | szsoftware | tmzt: ok kernel built :-) |
| 01:52.47 | ALoGeNo | loall |
| 01:52.55 | szsoftware | lo ALoGeNo |
| 01:53.04 | ALoGeNo | :) |
| 01:53.16 | ALoGeNo | hi tmzt man machine |
| 01:53.45 | ALoGeNo | the man that never stop :) |
| 02:00.33 | druidu | what device are you people working on? |
| 02:00.36 | tmzt | ALoGeNo: hey |
| 02:00.50 | druidu | hello |
| 02:01.15 | tmzt | I think szsoftware is working on polaris |
| 02:05.14 | *** join/#htc-linux PoohbaLT (n=Poohba@c-98-235-52-97.hsd1.nj.comcast.net) |
| 02:08.20 | szsoftware | dcordes: android-msm-htc-2.6.25 boots on polaris |
| 02:08.35 | szsoftware | dcordes: touch screen is responding |
| 02:09.08 | szsoftware | dcordes: framebuffer needstobe reviewed, the geometry is wrong or so |
| 02:09.41 | szsoftware | dcordes: sd works so far as I can access sd-partitions |
| 02:10.00 | szsoftware | time to say good night! |
| 02:15.58 | *** join/#htc-linux heise2k (n=arrrrrrr@65-78-40-52.c3-0.upd-ubr6.trpr-upd.pa.cable.rcn.com) |
| 03:05.14 | *** join/#htc-linux ChanServ (ChanServ@services.) |
| 03:05.14 | *** join/#htc-linux TeringTuby (n=maarten@195-241-125-243.ip.telfort.nl) |
| 03:05.14 | *** join/#htc-linux kaze (n=kaze@pac33-1-82-235-251-34.fbx.proxad.net) |
| 03:05.14 | *** join/#htc-linux tmzt (n=tmzt@adsl-99-155-214-182.dsl.akrnoh.sbcglobal.net) |
| 03:05.14 | *** mode/#htc-linux [+o ChanServ] by irc.freenode.net |
| 03:06.30 | *** join/#htc-linux BruteSource (n=BruteSou@m7d0e36d0.tmodns.net) |
| 03:45.38 | *** join/#htc-linux surgex (n=surge@pool-98-118-165-224.bflony.fios.verizon.net) |
| 04:08.02 | *** join/#htc-linux tmzt_ (n=zsirc@7.sub-75-219-80.myvzw.com) |
| 04:08.19 | tmzt_ | #join #openezx |
| 04:11.15 | ALoGeNo | xd |
| 04:24.12 | *** join/#htc-linux datachaos (n=datachao@189.27.191.132.adsl.gvt.net.br) |
| 04:38.05 | *** join/#htc-linux AstainHellbring (n=AstainHe@unaffiliated/astainhellbring) |
| 04:51.07 | *** join/#htc-linux penguinz (n=opera@66-168-215-122.dhcp.athn.ga.charter.com) |
| 04:57.28 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 05:01.35 | parmaster | hi ALoGeNo |
| 05:01.52 | ALoGeNo | hey par :) |
| 05:01.59 | parmaster | did you get a newer version of android working on the exec yet? |
| 05:02.29 | parmaster | i'm running angstrom on ba boardid 4 |
| 05:02.53 | parmaster | it still is keeping the camera flash light lit constantly |
| 05:03.33 | parmaster | i need to try to get a fix for that.. its been a problem for several years now |
| 05:05.02 | parmaster | ALoGeNo: had luck with a newer version of android on the uni? |
| 05:05.24 | ALoGeNo | nopes :( |
| 05:05.40 | ALoGeNo | i test the new one for vogue in the uni and nothing :/ |
| 05:06.09 | parmaster | it supports ARM5 ? |
| 05:06.41 | ALoGeNo | the vogue image is used to kaiser |
| 05:09.11 | parmaster | and the code the image was built from supports ARM5 though right? |
| 05:15.11 | ALoGeNo | http://it029000.massey.ac.nz/vogue/android-24-10-08.zip |
| 05:15.21 | ALoGeNo | http://it029000.massey.ac.nz/vogue/ |
| 05:15.32 | tmzt | ALoGeNo: that's root fs for android? |
| 05:15.41 | ALoGeNo | yep |
| 05:15.46 | tmzt | nice |
| 05:16.07 | ALoGeNo | well you have there data.gz initrd and system.gz |
| 05:16.22 | tmzt | which is put on the card? which partition? |
| 05:16.54 | ALoGeNo | you need support for yastt partitions tipe |
| 05:17.03 | ALoGeNo | or reconvert the files i think |
| 05:17.07 | tmzt | yaffs2 |
| 05:17.12 | ALoGeNo | yes |
| 05:17.27 | tmzt | which on is yaffs2, the initrd? |
| 05:17.43 | ALoGeNo | i think that is the same crap that os images from sdk.. is a tedious work |
| 05:18.09 | ALoGeNo | i dunno that becose i not have vogue that not works for me |
| 05:19.32 | ALoGeNo | but i was try android.bin like initrd from kaise (i think that is the same buil but rebuilded..) and works.. (i can see the red ball but lather gona crazy.. i think becose the kaiser have a lot of memory and uni only 64) |
| 05:20.42 | ALoGeNo | (that isnt a new one initrd i think becose the new android that i saw in the android emu boots with a little android insted of red ball |
| 05:20.44 | tmzt | you can't be using the same kernel on both. is the one for uni fixed for android? |
| 05:21.05 | ALoGeNo | yeah of course allways the same 2.6.21hh20 |
| 05:21.30 | tmzt | still waiting then, I think I'm going to try the linwizard test first |
| 05:21.36 | ALoGeNo | o dont know how and where learn or get that uni patched kernel :/ |
| 05:21.47 | ALoGeNo | nobody is working in that i think |
| 05:22.04 | tmzt | you find the poky source? |
| 05:22.05 | ALoGeNo | I* |
| 05:22.23 | parmaster | yeah, so and no patched kernel for uni means no patched kernel for ba either |
| 05:23.19 | tmzt | parmaster: (wow) we are trying to get the pxa's to new 2.6.26 kernel |
| 05:23.25 | ALoGeNo | no, my problem is that is a lot of info over there, and i walk a little confuse with all of stuff of procesors, devices and sources :/ |
| 05:23.44 | tmzt | but there are parts missing that make it less complete than hh.org |
| 05:24.09 | ALoGeNo | you guys are trying to do that kernel works? |
| 05:24.20 | ALoGeNo | 2.6.26?? |
| 05:24.39 | tmzt | hh.org supports usb with pxa udc, the upstream doesn't yet in a standard way |
| 05:25.26 | ALoGeNo | i dont know why the fucking 2.6.23 from poky and autobuilded for uni not work in the uni.. holy crap.. |
| 05:26.00 | parmaster | tmzt: kool |
| 05:26.04 | ALoGeNo | what is the problem :/ |
| 05:27.47 | parmaster | tmzt: do you anticipate that pxa will be on 2.6.26 sometime next year? |
| 05:28.39 | tmzt | it should be straightforward to get it booting, if someone wants to start with magician.c |
| 05:29.59 | tmzt | that is what I am using on motoq, but I need to add pcap support from openezx for that device |
| 05:31.28 | tmzt | android requires modifying framebuffer (pxafb) a little |
| 05:32.29 | parmaster | it just means i really need to find how to shut that flash light off on ba boardid 4 |
| 05:33.05 | tmzt | what's wrong with boardid 0x4? |
| 05:33.18 | parmaster | there is a camera on that model |
| 05:33.40 | parmaster | and a built in superbright led which operates as a camera 'flashbulb' |
| 05:34.10 | parmaster | when current angstrom's running on it, it stays lit continuously and will burn out the light eventually |
| 05:34.48 | parmaster | its been a problem since 2005 |
| 05:35.01 | *** join/#htc-linux AstainHellbring (n=AstainHe@unaffiliated/astainhellbring) [NETSPLIT VICTIM] |
| 05:35.01 | *** join/#htc-linux Xanthin- (n=xanthin@salvia.nieva.com) [NETSPLIT VICTIM] |
| 05:35.01 | *** join/#htc-linux SanMehat (n=san@nat/google/x-260282151d9118de) |
| 05:35.01 | *** join/#htc-linux NetRipper (n=netrippe@netripper.nl) [NETSPLIT VICTIM] |
| 05:35.01 | *** join/#htc-linux W8TVI (n=me@71.238.236.234) |
| 05:35.28 | parmaster | also in coming into 2.6.21 there were some jittery ts problems |
| 05:35.47 | parmaster | it was fine before.. but had to be redone for that version |
| 05:36.43 | parmaster | in testing it, it seems like it could still use some work |
| 05:38.33 | parmaster | acx support is supposed to be working in 2.6.21 now, but i haven't got that far in testing out everything |
| 05:39.06 | parmaster | cr2 mentioned to test out the cdma modem also |
| 05:39.59 | parmaster | (ba boardid 4 has cdma instead of gsm) |
| 05:40.26 | parmaster | and the board with gsm have acx for wifi |
| 05:40.42 | parmaster | the cdma version doesn't have acx |
| 05:42.24 | parmaster | other than the camera flashbulb problem boardid 4 now appears to function the same as the other boards (yaaay) |
| 05:42.57 | BruteSource | what camera driver did you use? |
| 05:49.59 | parmaster | the last patch i see is for 2.6.9 |
| 05:49.59 | parmaster | http://ftp.arm.linux.org.uk/pub/armlinux/people/xscale/mainstone/02-25-2005/src/kernel/patch-2.6.9-intc1 |
| 05:50.04 | parmaster | i don't even think its included in the distribution |
| 05:50.09 | BruteSource | xscale? |
| 05:50.18 | parmaster | doesn't matter we just need it off |
| 05:50.41 | parmaster | yeah pxa's |
| 05:50.58 | tmzt | no one knows why it stays on? |
| 05:51.13 | parmaster | not that i've heard |
| 05:52.11 | parmaster | anyway i'm sure that patch just covers the 'camera' |
| 05:52.29 | parmaster | nothing to do with this blaring bright led flash light |
| 05:53.11 | parmaster | just need to disable it |
| 05:53.30 | parmaster | worry about the camera stuff when support comes in for it |
| 05:54.25 | *** join/#htc-linux AstainHellbring (n=AstainHe@unaffiliated/astainhellbring) |
| 05:55.21 | tmzt | are you booting from haret? |
| 05:55.29 | parmaster | yes |
| 05:55.53 | tmzt | obviously the led isn't on when you boot. how far into the kernel messages does it come on? |
| 05:56.34 | ALoGeNo | any version of the kernel that acx run proppely in the uni? |
| 05:57.05 | parmaster | ALoGeNo: yes, i'm certain ofit |
| 05:57.09 | ALoGeNo | or that is a problem bettwen arm and acx stuff? |
| 05:57.39 | parmaster | cr2 would know which one he has a uni and had it working i think |
| 05:57.49 | ALoGeNo | parmaster the BA have the crappy acx too? |
| 05:57.54 | parmaster | yes |
| 05:58.10 | ALoGeNo | damn :) |
| 05:58.11 | parmaster | (all boards except for boardid 4 ) |
| 05:58.25 | ALoGeNo | roger.. |
| 05:58.37 | tmzt | cmda ba is called blue angel too? |
| 05:58.45 | parmaster | harrier |
| 05:58.51 | ALoGeNo | (im learning a little more this days reading you guys.. |
| 05:58.55 | tmzt | ok, xv6600 |
| 05:59.02 | tmzt | ? |
| 05:59.12 | parmaster | yep |
| 05:59.29 | tmzt | when does it come on? |
| 05:59.32 | BruteSource | anybody make any sense of this strace output for android userspace init |
| 05:59.36 | BruteSource | http://pastebin.com/d62649620 |
| 05:59.46 | parmaster | i'll look right now |
| 06:01.09 | BruteSource | it just loops that |
| 06:01.55 | *** join/#htc-linux BruteSource (n=BruteSou@m7d0e36d0.tmodns.net) |
| 06:02.01 | BruteSource | ooops |
| 06:03.28 | parmaster | tmzt: soon as it said jumping to kernel |
| 06:04.02 | BruteSource | anyone lol? |
| 06:04.30 | tmzt | sounds like it is in the gpio config |
| 06:04.44 | parmaster | yep |
| 06:04.49 | parmaster | i agree |
| 06:05.01 | tmzt | where it sets up direction/alt function/etc. |
| 06:05.15 | BruteSource | are you talking to me? |
| 06:05.20 | tmzt | can you trace with haret? |
| 06:06.02 | tmzt | nledset comamnd |
| 06:06.47 | BruteSource | ? |
| 06:06.55 | BruteSource | me? lol |
| 06:10.18 | tmzt | BruteSource: there is no where near enough information there |
| 06:10.37 | tmzt | where did you put the strace? |
| 06:11.21 | BruteSource | tmzt thats all it gives |
| 06:11.30 | BruteSource | it just loops |
| 06:13.53 | BruteSource | its penguinz not me he hasnt replied yet |
| 06:13.58 | *** join/#htc-linux data|2 (n=datachao@189.27.191.132.adsl.gvt.net.br) |
| 06:17.47 | parmaster | tmzt: i better ask cr2 and maybe kevin2 about this, it could be that this problem has already been addressed |
| 06:18.22 | BruteSource | tmzt i dunno what you mean its strace for /init of android |
| 06:18.58 | tmzt | how was strace started? that is what I am asking |
| 06:19.04 | parmaster | i guess it would just be that the setting needs to be put in the defaults.txt file before loading on boardid4 |
| 06:19.23 | tmzt | different mtype? |
| 06:20.29 | tmzt | I don't mean nledset is a setting for default.txt (though you could use it there) but that you can use it to turn on off the led if it is done the standard way in ce |
| 06:20.37 | tmzt | and trace gpio writes when you do |
| 06:21.22 | tmzt | there must another gpio on the other boards that is the same as your led, which is set to 1 by the kernel |
| 06:21.42 | parmaster | tmzt: well i mean it needs to be used there and thats a place neither here nor there as far as the distribution is concerned. Its probably been worked out.. so i'm going to check. |
| 06:21.56 | BruteSource | we just ran it from a terminal before the init? |
| 06:22.21 | tmzt | so you started with init=/bin/sh and strace -ff /init ? |
| 06:22.42 | tmzt | distribution? |
| 06:22.51 | parmaster | yeah angsrtom |
| 06:22.56 | parmaster | angstrom |
| 06:23.23 | tmzt | this is really a kernel thing though, oe gets it's source from handhelds.org on those devies |
| 06:23.36 | tmzt | angstrom is a set of packages built by oe |
| 06:23.46 | parmaster | it just means another boardid check in the code like the one i was talking about earlier |
| 06:24.14 | BruteSource | lol tmzt you were right about those #android guys |
| 06:24.43 | parmaster | another piece missing from it |
| 06:24.55 | tmzt | not all, there are a lot of google people and a lot of others that really know what they are talking about |
| 06:24.59 | parmaster | the one before was related to the screen.. |
| 06:25.34 | BruteSource | there telling me its really not linux and its all locked down, like motorolas os |
| 06:25.53 | tmzt | I understand the boardid chack posted earlier, that can be done it just needs someone to know the gpio |
| 06:25.56 | parmaster | i know i spent about a minute on that channel |
| 06:26.09 | parmaster | and just saw what i'll call "root beggars" |
| 06:26.27 | parmaster | and so i laugh like hahahahahha bahahahaaah |
| 06:26.44 | parmaster | and switch windows |
| 06:26.55 | BruteSource | HE DID IT! |
| 06:27.12 | BruteSource | he has the userspace |
| 06:27.18 | BruteSource | he is in |
| 06:27.51 | BruteSource | tmzt omg |
| 06:27.57 | BruteSource | and the dpad works |
| 06:28.33 | tmzt | yeah, they are talking on linwizard about dpad being broken on wizard too apparently |
| 06:28.46 | parmaster | tmzt: the boardid check i was talking about earlier was fixed and committed into the code.. |
| 06:28.48 | tmzt | so it's a gizard problem at least for X |
| 06:29.12 | BruteSource | well our dpad works in linwizard just not for arrows |
| 06:29.16 | parmaster | so yeah i'm willing to do some legwork to find out what it is, but not before checking if someone's already worked it out., |
| 06:29.31 | BruteSource | bt he as working androis userspace on excalibur! |
| 06:29.42 | BruteSource | androis |
| 06:29.47 | BruteSource | android |
| 06:29.56 | tmzt | who? |
| 06:30.18 | BruteSource | penguinz! |
| 06:30.24 | BruteSource | he did it |
| 06:30.29 | tmzt | and swetland has been here for months, providing information on msm chips |
| 06:30.33 | BruteSource | he is a fucking genius |
| 06:30.57 | BruteSource | well he doesnt work for google no matter what he says |
| 06:31.20 | tmzt | par also knows San |
| 06:31.44 | tmzt | ok, here, go to git.android.org and check the signed-off-bys |
| 06:31.53 | tmzt | no |
| 06:31.59 | BruteSource | for what? |
| 06:31.59 | tmzt | git.android.com |
| 06:32.11 | BruteSource | omg android working on excalibur |
| 06:32.22 | BruteSource | only took him 3 days |
| 06:32.42 | tmzt | penguinz: you here? |
| 06:32.48 | BruteSource | god we need those radio drivers |
| 06:32.52 | BruteSource | so bad |
| 06:33.18 | tmzt | you need support for ts0707 in the ril first |
| 06:33.24 | BruteSource | he is trying to fix the key map currently cant get back to home screcen |
| 06:33.33 | BruteSource | works already |
| 06:33.38 | tmzt | KEY_HOME |
| 06:33.55 | BruteSource | you sure? |
| 06:34.12 | tmzt | no, don't know what android uses, sorry |
| 06:35.02 | BruteSource | lol |
| 06:35.33 | BruteSource | funny the android project leader himself doesnt know |
| 06:35.36 | BruteSource | lmfao |
| 06:35.55 | *** join/#htc-linux surgex (n=surge@pool-98-118-165-224.bflony.fios.verizon.net) |
| 06:42.29 | penguinz | yes, i'm here tmzt |
| 06:42.37 | penguinz | hard at work :) |
| 06:44.51 | tmzt | penguinz: BruteSource says you have android (userspace) on omap? what did you fix to make it work? |
| 06:44.59 | penguinz | yes, that is correct. |
| 06:45.00 | tmzt | penguinz: and what was that strace with the anon maps |
| 06:45.22 | penguinz | i needed futex support in the kernel |
| 06:45.49 | penguinz | You also need to make the omap frambuffer doublebuffer for display to work. |
| 06:45.59 | tmzt | oh, didn't know that was exposed to userspace |
| 06:46.01 | penguinz | In addition to applying the android specific patches. |
| 06:46.23 | penguinz | well, their gui uses directfb like qtopia or something |
| 06:46.49 | tmzt | I know about that. there are still ipc's in the kernel? |
| 06:48.09 | penguinz | I basically took Martin's work with the vogue and used what he had working from it. Disabled initialization of the radio and a few other changes. |
| 06:50.40 | BruteSource | maybe his radio drivers will magicly work |
| 06:54.24 | swetland | glances at scrollback and laughs |
| 06:56.28 | tmzt | I suggested he look at the signed-off-bys on git.android.com, but I guess maybe you could have registered that too |
| 06:57.44 | tmzt | oh yeah, love the link on the search page, too bad a lot of people use the one in the browser toolbar now |
| 06:57.54 | tmzt | I don't see the special ones anymore either |
| 06:58.11 | tmzt | home page |
| 06:58.35 | swetland | I was just amused to see myself being reference-checked in another channel I was on |
| 06:58.53 | swetland | and hey, for all he knows you are just another account I have logged in! |
| 06:58.53 | BruteSource | even the official google press release states "our implementation of this linux kernel is fast and reliable" |
| 06:59.25 | BruteSource | lol no I know tmzt |
| 06:59.38 | BruteSource | he knows android is linux as well ask him |
| 07:01.03 | swetland | this is an is-a/has-a distinction, is all I'm saying. android is a platform which has quite a few parts, one of which is a linux kernel. |
| 07:01.41 | swetland | I'm not sure how that translates to some grand conspiracy ^^ |
| 07:01.45 | BruteSource | which happens to be the heart of the entire os? |
| 07:02.49 | BruteSource | yes a kernel is just a working part in any linux system... |
| 07:03.33 | parmaster | lol |
| 07:03.37 | parmaster | is par btw. |
| 07:04.11 | BruteSource | just because its customized and doesnt use xserver and some weird vm doesnt mean its not linux anymore |
| 07:07.05 | *** join/#htc-linux datachaos (n=datachao@189.27.191.132.adsl.gvt.net.br) |
| 07:07.07 | BruteSource | this is a silly argument |
| 07:13.12 | penguinz | my clock keeps going back to 1970 |
| 07:13.19 | penguinz | whoever's idea that was.... =( |
| 07:13.19 | BruteSource | well now his desktop is frozen terminalis bitching some power managment problem |
| 07:13.27 | BruteSource | ya what he said |
| 07:15.13 | BruteSource | seems everyone went to sleep |
| 07:16.27 | tmzt | <PROTECTED> |
| 07:16.27 | tmzt | 17:57 < tmzt> but msm support was released by google last fall |
| 07:16.27 | tmzt | 17:57 < tmzt> and updated this whole year |
| 07:16.27 | tmzt | 18:06 < tmzt> wasup: no, but a lot of people in #android, do. apparently it's only an android device if it shipped with android |
| 07:16.30 | tmzt | 18:06 < tmzt> and android is not linux, it's java. |
| 07:16.33 | tmzt | 18:06 < tmzt> and posix does not exist |
| 07:16.35 | tmzt | 18:06 < tmzt> and things like firewalls and proxy servers have to be ported to android |
| 07:16.38 | tmzt | 18:07 < tmzt> so people can access the internet |
| 07:16.50 | tmzt | I should be more careful with my words |
| 07:17.36 | tmzt | so used to android referring to everything from parts of the kernel tree, to the git, to anything having to do with android |
| 07:18.01 | parmaster | yes, best to let sleeping dogs lie |
| 07:18.05 | parmaster | lol |
| 07:20.20 | swetland | I'm sure a bunch of folks out there have no idea, nor interest in what's under the hood |
| 07:20.39 | swetland | and obviously there are quite a few who do and are having fun with it |
| 07:20.43 | swetland | so, hey, everyone's happy |
| 07:23.47 | tmzt | yeah, which is why I stayed out of there. then someone here said they were doing kernel stuff so I looked into it |
| 07:24.46 | *** join/#htc-linux ulaas (n=ulaas@88.253.14.253) |
| 07:25.02 | tmzt | I am happy for there to be great applications, it's just that some things already exist for linux or in some cases are in the kernel that they don't need to be rewritten for android in java compiled to dvm code |
| 07:26.22 | BruteSource | what does sunos use for gui? |
| 07:26.30 | tmzt | now? |
| 07:26.45 | penguinz | BruteSource : X |
| 07:27.39 | BruteSource | oh i thought they had something else |
| 07:28.28 | penguinz | is there a rtc in any htc phones? |
| 07:28.46 | penguinz | and on a related note, where does linux get date / time info if not. |
| 07:29.04 | tmzt | there is supposed to be an rtc in pxa, but it is broken, it was just discussed on arm-kernel |
| 07:29.22 | tmzt | their are different registers to access the time in different ways |
| 07:29.50 | tmzt | on omap? don't know |
| 07:29.54 | penguinz | So basically, each time i load up with haret it's 1970 again. |
| 07:30.15 | penguinz | which apparently was not an issue until just now. |
| 07:30.20 | tmzt | I have time set in ce.net in my mp900c, boot linux with haret and it's 1970 (0 unix time) |
| 07:30.38 | tmzt | the driver is their but doesn't work |
| 07:31.12 | BruteSource | aw so it means it cant get the time |
| 07:31.28 | BruteSource | not that its getting the rong time from somewhere |
| 07:31.34 | tmzt | have you done hwclock --hwtosys ? |
| 07:32.08 | tmzt | hwclock -hwtosys |
| 07:32.43 | *** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz) |
| 07:32.46 | BruteSource | does a htc phone have something like a bios? |
| 07:32.57 | tmzt | it has a boot loader |
| 07:33.10 | tmzt | if you mean a setup program, no |
| 07:33.48 | BruteSource | something for cmos |
| 07:34.41 | tmzt | the rtc is the hardware battery backed (but using the main battery and a capacitor usually) which is what one of the functions of the pc cmos is |
| 07:35.17 | tmzt | battery backed clock |
| 07:35.29 | penguinz | my busybox is not compiled with hwclock |
| 07:35.45 | tmzt | using gizard? |
| 07:36.32 | penguinz | not currently. |
| 08:15.00 | parmaster | tmzt: what kind of work do you do? |
| 08:21.10 | *** join/#htc-linux TeringTu1y (n=maarten@195-241-125-243.ip.telfort.nl) |
| 08:33.10 | *** join/#htc-linux ALoGeNo (n=QUAKEIII@118.Red-83-49-136.dynamicIP.rima-tde.net) |
| 08:40.46 | BruteSource | well android on omap850 is coming along nicely at least |
| 08:42.24 | *** join/#htc-linux dante_ (n=dante@79.15.48.176) |
| 08:48.17 | parmaster | did you get a copy of the omap850 technical reference manual? |
| 08:50.16 | swetland | here's a fun tidbit for free |
| 08:50.36 | swetland | the reference ril configuration has some mention of sooner |
| 08:50.47 | swetland | sooner was our original omap850 based development platform |
| 08:51.22 | swetland | not sure how much overlap there is with other 850 based devices, but might not be entirely useless |
| 08:51.43 | BruteSource | ya we looked, thats where the original vogue android got up and running |
| 08:51.56 | BruteSource | vox |
| 08:52.39 | swetland | did ti actually release the 850trm? |
| 08:52.52 | parmaster | i don't believe so |
| 08:52.59 | swetland | I was amazed to see them finally publishing the 3430trm publicly. they had been making noises about it since 95 |
| 08:53.21 | parmaster | although some shady black bag type characters had been handing it out |
| 08:53.23 | parmaster | lol |
| 08:54.57 | *** join/#htc-linux goxboxlive_ (n=goxboxli@139.84-48-212.nextgentel.com) |
| 08:55.05 | parmaster | actually they could have released it the same s the 3430trm i'm not sure |
| 08:57.15 | swetland | I don't recall it having much specific to the modem stuff, so I'm not sure what would be sensitive |
| 08:57.54 | *** join/#htc-linux nebi (n=nebi@c-498fe253.02-145-7570701.cust.bredbandsbolaget.se) |
| 08:59.36 | parmaster | well i can show you what was already in circulation i guess |
| 09:00.19 | swetland | don't need it |
| 09:01.02 | parmaster | right, ;) |
| 09:02.22 | BruteSource | god i loce trolling ##linux, i never troll anywhere but those guys are so egotistical |
| 09:07.12 | *** join/#htc-linux lpotter (n=ljp@CPE-124-191-144-181.vic.bigpond.net.au) |
| 09:09.13 | *** join/#htc-linux eufrat (n=eufrat@dsl-trebrasgw1-fe5dfa00-103.dhcp.inet.fi) |
| 09:13.49 | ALoGeNo | hey we can load yaffs support like a module in kernel? |
| 09:14.20 | ALoGeNo | or the yatffs is a propyetary driver in android? |
| 09:15.30 | BruteSource | just use ext2 |
| 09:15.37 | BruteSource | thats what we did saved alota hassle |
| 09:19.15 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 09:22.00 | parmaster | yeah with busybox and everything its an industrial build system for sure |
| 09:22.28 | parmaster | at least its not vxworks |
| 09:23.49 | ALoGeNo | hum... |
| 09:25.51 | *** join/#htc-linux kiozen (n=oeichler@rgnb-5d87dc0c.pool.einsundeins.de) |
| 09:26.13 | parmaster | ALoGeNo: are you working on bringing pxa's up to 2.6.26? |
| 09:26.43 | ALoGeNo | that will be okey but im not developer :( |
| 09:26.59 | parmaster | me either |
| 09:27.10 | ALoGeNo | maybe can do the work of betatester or whatever but not coding at all :/ |
| 09:27.16 | parmaster | also, i saw how long it took to move up kernels last time |
| 09:27.38 | swetland | for whatever it's worth, we're moving to 2.6.27 soon |
| 09:27.50 | swetland | (android and android-msm branches) |
| 09:28.19 | ALoGeNo | swetland but something to BA and uni? |
| 09:28.31 | ALoGeNo | (and good morning) |
| 09:28.54 | parmaster | heh, yeah i don't think they care much about pxa |
| 09:29.15 | parmaster | they are advancing.. |
| 09:29.31 | swetland | half asleep and didn't notice it was htc-linux, where you may not particularly care about where android is going ^^ |
| 09:30.02 | parmaster | i have no idea about android |
| 09:30.12 | parmaster | don't know anything about it really |
| 09:30.37 | parmaster | just another qtextended or gpe or openmoko to me i guess |
| 09:33.21 | BruteSource | i do care lol but you dont know lol |
| 09:37.06 | parmaster | yeah, i'm not a developer too ALoGeNo.. but even fixing a small problem is stil fixing a problem... |
| 09:37.34 | *** join/#htc-linux BabelO (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 09:37.34 | parmaster | so i think i'll try for fixing the boardid4 problem(s) for right now |
| 09:37.54 | parmaster | hey BabelO |
| 09:38.06 | BabelO | hi |
| 09:39.13 | parmaster | do you remember back when you were working a lot on ba stuff.. a bunch of testers were complaining about a bright led camera flash-light on it |
| 09:39.30 | parmaster | on the boardid 4 version i mean |
| 09:40.39 | parmaster | the one with the camera and flashlightbul led |
| 09:41.13 | parmaster | i saw a lot of the guys trying to fix the problem |
| 09:41.46 | parmaster | was wondering if anyone ever found the cause, so that we can put the proper boardid check in the code for it |
| 09:43.16 | *** join/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 09:44.29 | *** join/#htc-linux Othello (i=Othello@gateway/tor/x-6fb748c772156d47) |
| 09:47.04 | *** join/#htc-linux pete_ (n=pete@91.104.31.172) |
| 09:47.31 | pete_ | yo |
| 09:49.04 | BabelO | parmaster: no,never heard that, but you can look at htc aic driver to fix that i think |
| 09:49.22 | *** join/#htc-linux Othello (i=Othello@gateway/tor/x-65c911ea5d4d7768) |
| 09:50.26 | BabelO | swetland: i still can t compile android on intel x8_64 :( |
| 09:50.29 | BabelO | swetland: so you say we have to look at android ril code |
| 09:53.04 | BabelO | swetland : latest try with htc artemis on 2.6.26rc9 was ok i ve to move to 2.6.27 then if i want android patch on omap850 branch |
| 10:00.26 | pete_ | guys I'm interested in Linux- Am running Ubuntu on all my home computers (including this one), and I have an HTC Diamond that's I'm looking forward to using Linux on... |
| 10:00.34 | ALoGeNo | parmaster you was tested some poky kernel in BA? |
| 10:00.49 | ALoGeNo | the 2.6.23 one? |
| 10:00.54 | parmaster | ALoGeNo: nope |
| 10:01.02 | pete_ | My question is, where is a good place to start if I want to learn about coding? Is there a starter's resource |
| 10:01.23 | BruteSource | tldp |
| 10:01.31 | BruteSource | best resource there is |
| 10:01.39 | BruteSource | tldp.net |
| 10:01.40 | ALoGeNo | pete_ the best way i think that is learn assembler and later C or C++ |
| 10:01.43 | BruteSource | or org |
| 10:01.44 | pete_ | I've never programmed before so I don't have any experience |
| 10:02.32 | BruteSource | start with an interpreted language so you can understand syntax than work up to oo languages |
| 10:02.46 | pete_ | Just so I know, what is the difference between C and C++ |
| 10:02.49 | BruteSource | thats how you learn in a cs course |
| 10:02.52 | parmaster | ALoGeNo: gimme a link for their project |
| 10:03.30 | ALoGeNo | give me one minute.. |
| 10:04.28 | ALoGeNo | http://www.pokylinux.org/getit/ |
| 10:04.41 | pete_ | http://tldp.net/HOWTO/HOWTO-INDEX/programming.html I don't even know where to start |
| 10:05.46 | ALoGeNo | parmaster http://pokylinux.org/releases/pinky-3.1/ |
| 10:06.02 | ALoGeNo | there isnt ba autobuild :/ |
| 10:06.15 | pete_ | brute what is a Cs course? |
| 10:06.16 | ALoGeNo | relase* |
| 10:07.16 | ALoGeNo | if you use the rootfs of the uni sure that work parmaster |
| 10:07.53 | parmaster | ugh, its a qemu |
| 10:07.55 | ALoGeNo | and maybe you can use the ba config qith the universal branch to compile the kernel no? |
| 10:08.29 | parmaster | could give it a shot i guess |
| 10:08.43 | ALoGeNo | you say that all that is for qemu? the kernel too? |
| 10:08.44 | parmaster | i need to rebuild my entire environment |
| 10:09.00 | parmaster | bitbake all that crap |
| 10:09.26 | ALoGeNo | what you mean with bitbake? |
| 10:10.53 | parmaster | nah, i see the universal image in there |
| 10:11.48 | ALoGeNo | yeah you have the image and kernel but the kernel dont boot :/ |
| 10:12.34 | ALoGeNo | nothing, it hangs in the tiny letter at the begginig |
| 10:13.30 | parmaster | hahaha |
| 10:13.42 | parmaster | you never got it to work ALoGeNo |
| 10:13.44 | parmaster | ? |
| 10:14.11 | ALoGeNo | only rootfs with the hh.org kernel |
| 10:15.05 | ALoGeNo | i will try to compile my own kernel with poky source.. |
| 10:15.49 | ALoGeNo | i think that is in the same way that for the hh.org kernel that im using |
| 10:16.14 | parmaster | aha |
| 10:17.00 | ALoGeNo | im bored using allways the same 21hh20 kernel and his fucked acx X) |
| 10:17.09 | parmaster | hahaahh |
| 10:17.19 | parmaster | did you bring up the wifi through the acx? |
| 10:17.31 | ALoGeNo | (and thanks that it work mid fine ;)) ) |
| 10:18.22 | ALoGeNo | yeah it works for me.., but when i look the opie and another distros runing.. it charge only the htcuniversal_acx and not the acx one |
| 10:18.32 | parmaster | and don't you worry.. it will become even more fucked all over again every time we advance a kernel its like to start from scratch |
| 10:19.08 | ALoGeNo | later i must do morprobe acx and ifup the wlan that the fucked show like wlan1 and not wlan0 (wtf??) |
| 10:19.24 | parmaster | hahaha |
| 10:19.46 | parmaster | might be in the device setup during boot |
| 10:20.15 | ALoGeNo | heh i konw man.., but i so nervious.. and i wnat all quickly you know.. hehe.. |
| 10:20.42 | ALoGeNo | yeah that is not the problem.. |
| 10:20.58 | parmaster | maybe wifi0 is device destroyed somehow lol |
| 10:21.27 | parmaster | so it auto makes it wifi1 |
| 10:22.42 | parmaster | hey, ya know another fucked up thing i noticed is that in the older versions of the distro the phone worked fine |
| 10:22.59 | ALoGeNo | in debian i have all configurated and running problelly only the crap of that is wlan1 (i dont know why..) but all work.., wifi conect and all airdump run like a charm but some time (many times.. maybe..) the driver is gone crazy and put a lot of tash in the screen and hang the system or only cut the connection... bad thing.. |
| 10:23.00 | parmaster | i could make calls and talk never had to modify anything |
| 10:24.00 | ALoGeNo | yeah what are you using? |
| 10:24.01 | parmaster | but i notice in 21hh20 it refernces some openmokodevice for the phone portion and fails on that device when i try to place calls |
| 10:24.03 | ALoGeNo | opie? |
| 10:24.08 | ALoGeNo | titchy? |
| 10:24.13 | parmaster | no the regular angstrom distro |
| 10:24.32 | ALoGeNo | oh gpe x11? |
| 10:24.34 | parmaster | its gpe |
| 10:24.35 | parmaster | yes |
| 10:24.39 | ALoGeNo | nice.. |
| 10:25.19 | parmaster | well, its cool most of the stuff looks good |
| 10:25.23 | ALoGeNo | im with debian titchy mised by shimomura |
| 10:25.28 | parmaster | but the device support is all fucked up |
| 10:25.38 | parmaster | acx doesn't set up automatically |
| 10:25.51 | ALoGeNo | no? |
| 10:25.57 | ALoGeNo | in the uni yes |
| 10:26.09 | parmaster | or at all.. theres a procedure somewhere for how you manually configure it i saw |
| 10:26.17 | ALoGeNo | list the modules.. |
| 10:26.25 | ALoGeNo | you have both? |
| 10:26.47 | parmaster | i can't ifconfig a device that isn't there :) |
| 10:27.05 | ALoGeNo | well in the x11 i cant remember but in opie it runs, but i must charge the acx by hand |
| 10:27.28 | parmaster | yeah so i need to look at it more.. do the manual process for enabling the driver and put it in a boot script once i have it hammered |
| 10:28.04 | ALoGeNo | heh |
| 10:28.24 | parmaster | those aren't real problems.. |
| 10:28.36 | *** join/#htc-linux Xanthin- (n=xanthin@salvia.nieva.com) [NETSPLIT VICTIM] |
| 10:28.36 | *** join/#htc-linux SanMehat (n=san@nat/google/x-260282151d9118de) |
| 10:28.36 | *** join/#htc-linux NetRipper (n=netrippe@netripper.nl) [NETSPLIT VICTIM] |
| 10:28.36 | *** join/#htc-linux W8TVI (n=me@71.238.236.234) |
| 10:28.40 | ALoGeNo | yep.. |
| 10:28.46 | parmaster | the led differences between the boards burning out a camera flashbulb on the harrier.. thats a big deal |
| 10:28.58 | ALoGeNo | the unique important diference about your ba and my uni is the screen resolution no par? |
| 10:29.12 | parmaster | what 240x320? |
| 10:29.16 | ALoGeNo | yep |
| 10:29.28 | ALoGeNo | 640x480 here :) |
| 10:29.48 | ALoGeNo | that is the unique? |
| 10:29.58 | parmaster | i don't remember |
| 10:30.04 | ALoGeNo | hum... |
| 10:30.06 | parmaster | let me look at the qvga specs |
| 10:31.18 | parmaster | 320 x 240 pixels with 65536 colours |
| 10:32.09 | parmaster | but you have only 64mb ram right |
| 10:32.20 | parmaster | ba has 128 |
| 10:33.30 | ALoGeNo | yep im saw that 128 hehe nice memory :) |
| 10:34.36 | parmaster | yeah the little sucker was a workhorse for years when it first came out.. outpowering at least two models the came after it even |
| 10:36.40 | parmaster | i won't throw it away no matter how many new phones i get |
| 10:36.50 | parmaster | even the cryptophone is based on it. |
| 10:37.10 | ALoGeNo | lol |
| 10:37.52 | ALoGeNo | cryptophone sounds like something strange gadget of spys movie.. |
| 10:37.53 | ALoGeNo | XD |
| 10:38.01 | parmaster | a cryptophone is just a ba with a specially cooked winmo rom |
| 10:38.13 | parmaster | well it is actually |
| 10:38.14 | ALoGeNo | oh.. nice.. |
| 10:38.27 | parmaster | european security services use it |
| 10:38.49 | ALoGeNo | maybe i can use it you buy hemp to my camell |
| 10:38.51 | ALoGeNo | XD |
| 10:39.01 | parmaster | ahahhahah |
| 10:39.30 | parmaster | yes its good for drug running, arms dealing.. all the fun stuff you might do under gsm. |
| 10:39.32 | ALoGeNo | heh :) |
| 10:39.41 | parmaster | LOL@ |
| 10:39.56 | ALoGeNo | x,D |
| 10:40.51 | ALoGeNo | i need the acx driver or wifi running quickly becose im thinking in buy some eeepc... |
| 10:41.04 | ALoGeNo | and if wifi works.. i dont need eeepc.. |
| 10:41.14 | parmaster | well then bring it up |
| 10:41.48 | parmaster | you said it was running tho right? |
| 10:41.54 | ALoGeNo | but if i dont have something... im bored... but im poor to to walk buying things.. heh.. and you have the crisis now.. |
| 10:41.55 | ALoGeNo | XDDD |
| 10:41.59 | *** join/#htc-linux szsoftware (n=sven@e178204134.adsl.alicedsl.de) |
| 10:42.16 | ALoGeNo | and we have the crisis* i mean |
| 10:42.16 | ALoGeNo | xd |
| 10:42.45 | *** join/#htc-linux szsoftware (n=sven@e178204134.adsl.alicedsl.de) |
| 10:42.49 | ALoGeNo | yep yep, its running.., but for times.. |
| 10:43.01 | parmaster | ohh right the crshing problem |
| 10:43.06 | szsoftware | morning |
| 10:43.06 | parmaster | everyone told about that |
| 10:43.12 | ALoGeNo | now yes,, not now.,.. not now.. now yes.. and so.. |
| 10:43.20 | parmaster | one guy went.. dude i ran ping and it crashed acx lol |
| 10:43.29 | ALoGeNo | morning szsoftware |
| 10:43.41 | ALoGeNo | yes! |
| 10:43.48 | ALoGeNo | thats it..! |
| 10:44.21 | ALoGeNo | it works like a charm.. but you transfer data and at the few seconds.. craped.. |
| 10:44.22 | ALoGeNo | Xd |
| 10:44.41 | ALoGeNo | ifdown, ifup,... and pray.. |
| 10:44.43 | ALoGeNo | XD |
| 10:45.20 | ALoGeNo | parmaster you dont have this problems? |
| 10:51.52 | parmaster | one reason i didn't run it |
| 10:52.03 | parmaster | i thought maybe it had improved in this build |
| 10:52.16 | parmaster | but that doesn't sound very promising :) |
| 10:54.08 | ALoGeNo | :)) |
| 10:57.10 | parmaster | hmmm |
| 10:57.30 | parmaster | i may have a more lower level problem than i thought: |
| 10:57.32 | parmaster | CONFIG_MACH_BLUEANGEL_LEDS=y |
| 10:59.07 | *** join/#htc-linux ezak (n=kaze@ABordeaux-152-1-8-229.w82-125.abo.wanadoo.fr) |
| 11:01.03 | *** join/#htc-linux BabelO (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 11:01.07 | ALoGeNo | xd |
| 11:03.21 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) |
| 11:07.17 | *** join/#htc-linux fnord (n=fnord@24-151-90-116.static.nwtn.ct.charter.com) |
| 11:21.14 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) |
| 11:23.15 | dcordes | hi |
| 11:23.22 | szsoftware | yo |
| 11:25.13 | swetland | 'morning |
| 11:26.11 | szsoftware | morning swetland :-) |
| 11:27.37 | szsoftware | dcordes: update, msm_fb.c works now on polaris, left vogue-changes untouched, took the msm_fb.c from your tarball into android-msm-htc-2.6.25 |
| 11:27.46 | dcordes | morning there |
| 11:27.54 | dcordes | ay caramba |
| 11:28.18 | dcordes | anybody reviewed that kaiser wifi code? |
| 11:28.52 | dcordes | szsoftware, that's great what's up with the touchscreen? |
| 11:29.13 | szsoftware | dcordes: it's working :-) |
| 11:29.13 | ALoGeNo | parmaster http://nemustech.blogspot.com/ nice info.. ;) |
| 11:29.36 | szsoftware | dcordes: but without onscreen kb :-( |
| 11:29.59 | dcordes | ok you don't need it anyway if you have wifi |
| 11:30.20 | swetland | alogeno: looks like from M5 era, possibly earlier. init now wants the ashmem driver and no longer does anything in /tmp. neat tidbits though |
| 11:30.27 | szsoftware | dcordes: yeah, but it would be a nice feature (to mplayer on the console, haha) |
| 11:31.02 | ALoGeNo | ohhh thanks for the info swetland :) |
| 11:31.08 | szsoftware | s/to/to start/ |
| 11:34.11 | dcordes | we just need 400$ so somebody will help with the kaiser wifi |
| 11:34.30 | dcordes | I know that guy |
| 11:34.40 | dcordes | he will do it but he needs the $ |
| 11:34.53 | szsoftware | dcordes: ok, I know one who does it for 300 bucks |
| 11:35.34 | dcordes | szsoftware, did you use the wifi patch as a base? |
| 11:35.46 | dcordes | did you test phone? |
| 11:35.59 | dcordes | we need to fix the GPS channel |
| 11:36.04 | szsoftware | dcordes: I did'nt touch wifi, untill I have prepared the polaris diff |
| 11:36.48 | dcordes | so you used the first trout patch I put? |
| 11:43.00 | szsoftware | dcordes: I patched the clean android-msm-htc-2.6.25 with your patch git-diff-android-msm-htc-2.6.25_kaiser_20082210.diff |
| 11:44.10 | szsoftware | dcordes: then fiddled polaris stuff in. Will send you the patch now. It is polaris-only, after your patch, devided into adding files and editing files |
| 11:56.25 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) |
| 11:56.53 | dcordes | damn wifi. need 30m lan cable. |
| 11:57.23 | dcordes | szsoftware, ok |
| 11:57.37 | dcordes | I will merge it in and create a new diff including everything |
| 11:58.19 | dcordes | SanMehat, got my mail? |
| 11:58.20 | parmaster | ALoGeNo: is that link the source of the Benno patch? |
| 11:58.54 | ALoGeNo | i dunno but read that swetland said |
| 11:59.15 | ALoGeNo | [13:30] <swetland> alogeno: looks like from M5 era, possibly earlier. init now wants the ashmem driver and no longer does anything in /tmp. neat tidbits though |
| 11:59.23 | parmaster | yep |
| 11:59.44 | szsoftware | dcordes: explain, why you need 30m lan cable? |
| 12:00.09 | dcordes | so I can get rid of all wifi at home |
| 12:02.17 | szsoftware | dcordes: :-) you could hang deeper the lamps, plug a switch instead of the lamp, put led-beamers on it, powered over ethernet, but at least leave one port for your eth device |
| 12:02.35 | szsoftware | dcordes: and do it in every room |
| 12:02.57 | dcordes | lol wtf |
| 12:04.32 | szsoftware | dcordes: I try booting android-1.0-r1 from rootsystem. With the new kernel, after freeing init memory, Warning: unable to open an initial console |
| 12:04.38 | szsoftware | dcordes: any hint? |
| 12:06.44 | dcordes | no |
| 12:19.47 | parmaster | ALoGeNo i'm pretty sure that was the Benno patch used in that image you have running on the uni |
| 12:23.21 | BabelO | kiozen: ohhh, text is there :) |
| 12:30.16 | szsoftware | dcordes: do you know how to debug fb? how can I blit stuff over it? For instance, from the loaded (buffered) fb keyboard image |
| 12:30.50 | dcordes | idk can't think horrible hangover bbiab |
| 12:30.52 | szsoftware | any other? Who has experience with framebuffer? |
| 12:31.03 | *** join/#htc-linux goxboxlive (n=goxboxli@139.84-48-212.nextgentel.com) |
| 12:31.08 | szsoftware | dcordes: :-) ok |
| 12:33.04 | dcordes | spent much money slowing down synaptic activity through use of highly neurotoxic substance |
| 12:34.53 | BabelO | hi goxboxlive |
| 12:35.32 | szsoftware | dcordes: yeah that's fine. Sometimes you need a total reset and preparing a tabula rasa, :-) |
| 12:35.48 | *** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix) |
| 12:35.51 | dcordes | BabelO, hi can you suggest a good gps/chipset with usb port? |
| 12:36.07 | szsoftware | dcordes: so you got pm working? |
| 12:36.15 | szsoftware | dcordes: regarding your brain, haha |
| 12:36.44 | szsoftware | dcordes: angstrom rootfs boots fine with new kernel |
| 12:37.02 | szsoftware | dcordes: I need that damn onscreen keyboard ^^ |
| 12:37.22 | BabelO | dcordes: not more idea than kiozen, i have only one gps at home |
| 12:37.56 | dcordes | yes, hardreset. no you want wifi. I sent the patch with my full approach along with an outline of how I did what to SanMehat and dr. barry carter |
| 12:40.08 | szsoftware | dcordes: yes I want wifi too. Of course. |
| 12:40.24 | szsoftware | dcordes: ok we can talk about less complicated, if you want |
| 12:40.39 | *** join/#htc-linux Dubbeld00 (n=opera@82-168-116-233.ip.telfort.nl) |
| 12:40.45 | *** part/#htc-linux Dubbeld00 (n=opera@82-168-116-233.ip.telfort.nl) |
| 12:40.46 | szsoftware | dcordes: what do you think about reorganising kaiser/polaris stuff |
| 12:40.51 | *** join/#htc-linux Dubbeld00 (n=opera@82-168-116-233.ip.telfort.nl) |
| 12:41.37 | szsoftware | dcordes: Could we resolve some redundancies in the code? |
| 12:41.46 | dcordes | BabelO, do you share kiozen's concern about the sirf III being problematic? |
| 12:42.17 | dcordes | szsoftware, what's wrong with it? for the framebuffer keyboard, I would contact dzo again and ask him to provide further help |
| 12:42.54 | szsoftware | For instance, when you put android-msm-htc+kaiser+polaris on linuxtogo, should'nt we provide only one board-file with ifdefs? |
| 12:43.06 | szsoftware | dcordes: since the boards are very similar |
| 12:43.22 | BabelO | dcordes: yes |
| 12:43.35 | BabelO | dcordes: i m not expert on GPS chipset |
| 12:45.34 | szsoftware | dcordes: do you know what linux/gpio_event.h does? its included in board-htcpolaris.c, but not in board-htckaiser.c |
| 12:46.03 | dcordes | szsoftware, I don't know we could as well keep the two seperate board files. |
| 12:46.32 | szsoftware | dcordes: what is MSM_ONRPCROUTER? it's actually deselectet, but depends on kaiser-smd, not on polaris-smd |
| 12:46.58 | dcordes | but on the other hand you have a point there. arch-arm/ will get huge at the point merging with the vogue branch |
| 12:47.02 | szsoftware | dcordes: but if there are changes in the future, you have to double code |
| 12:47.39 | szsoftware | dcordes: ok merging with vogue brach is planned? |
| 12:47.53 | *** part/#htc-linux Dubbeld00 (n=opera@82-168-116-233.ip.telfort.nl) |
| 12:47.55 | dcordes | it might as well be a good idea to create subfolders for the machines like in the hh tree |
| 12:48.22 | dcordes | without something like that it will get horribly unorganized |
| 12:48.32 | szsoftware | dcordes: should'nt we optimize make menuconfig selection? You can select several boards at the same time, same is with phone type |
| 12:48.38 | dcordes | or we need a more consistent naming structure |
| 12:48.55 | dcordes | board-kaiser-keypad.c e.g. is a mess |
| 12:48.58 | swetland | boardfile-per-machinetype seems preferable |
| 12:49.24 | dcordes | what do you think about subfolders, swetland? |
| 12:49.34 | swetland | maintaining one uberboardfile with a ratsnest of ifdefs would not be fun |
| 12:50.01 | swetland | dcordes: might be advisable for variants on smd, etc |
| 12:50.13 | swetland | mach-msm is starting to get a bit crowded, it is true |
| 12:50.14 | dcordes | since you outsourced the trout specific bits like mmc there are plenty of files only for that board |
| 12:50.19 | szsoftware | dcordes: swetland, if there are so much similarities, why not one board-file with if defs and symlinking to boardfile-<phone/machine>? |
| 12:50.59 | swetland | well, the cleanup we're working on in the patches that went into 2.6.28rc1 tries to reduce duplication in board files |
| 12:51.05 | swetland | moving common stuff into devices.c, etc |
| 12:51.24 | szsoftware | swetland: that sound good |
| 12:51.49 | swetland | I do think we'll need to think about the proliferation of board-$machine-$thing.c files |
| 12:51.53 | szsoftware | swetland: so the next release of android kernel will be 2.6.28? |
| 12:51.59 | swetland | I kinda like having them split up a bit, but it does get overwhelming |
| 12:52.04 | swetland | sz: 2.6.27 |
| 12:52.24 | swetland | we'll hopefully have android-2.6.27 / msm-2.6.27 out soonish and those will become the active working branches |
| 12:52.45 | swetland | .28 was just where we got the next batch of stuff into mainline |
| 12:52.48 | swetland | it's slow going |
| 12:53.11 | swetland | http://git.android.com/?p=kernel.git;a=shortlog;h=for-rmk <- got merged the other day |
| 12:53.32 | *** join/#htc-linux nebi (n=nebi@c-498fe253.02-145-7570701.cust.bredbandsbolaget.se) |
| 12:53.48 | szsoftware | swetland: what do you think about MSM7K HS USB? is that very MSM7x00A specific? |
| 12:54.12 | swetland | 7200 doesn't have a hsusb controller |
| 12:54.18 | swetland | 720[01]A does |
| 12:55.11 | NetRipper | swetland, did you guys ever use cdc_ether or usb serial when developing? |
| 12:55.14 | szsoftware | swetland: uhh, ok. Do you know how is usb in 7200 realized |
| 12:55.18 | szsoftware | swetland: ? |
| 12:55.21 | swetland | the family naming scheme is pretty annoying, which is one reason we shifted from ARCH_MSM7X01A to ARCH_MSM |
| 12:55.36 | swetland | sz: fullspeed controller. we never got specs because of gpl issues |
| 12:55.55 | dcordes | swetland, for the htc msm devices it is kept similar. we have kaiser-smd.c vogue-smd.c kaiser-keypad etc etc |
| 12:56.14 | swetland | dcordes: that sounds reasonable to me |
| 12:57.22 | dcordes | swetland, could you do us a big favor and review my approach adding ti1251 support for kaiser/polaris? I have problems with the "card" not being detected by the driver |
| 12:57.22 | szsoftware | dcordes: swetland what are the chances getting usb work on 7200? Is the controller in the SoC? What is needed to get it running? |
| 12:57.44 | dcordes | szsoftware, dzo looked into it. he says it is nontrivial. |
| 12:57.45 | swetland | dcordes: I honestly have no clue about the wifi/sd stuff |
| 12:58.03 | dcordes | swetland, ok I hope SanMehat can help us out, I emailed him about it. thanks anyway. |
| 12:58.05 | swetland | San is the guy to ask, but he's getting married next week, so he may be a bit hard to track down |
| 12:58.29 | dcordes | yes I know, still he kindly offered to have a look. very nice. |
| 13:01.03 | dcordes | cr2_, did you see my answer to your hint about the sdio wifi sd controller? I think it has clock because the msm_sdcc output in the kaiser linux dmesg shows a clock on both controllers. |
| 13:01.05 | szsoftware | dcordes: swetland: ok, what about bluetooth? Is that also a nontrivial thing? What are the diferences there between 7x00A and 7200? |
| 13:02.02 | dcordes | szsoftware, I think for bluetooth and usb the 7200 vs 7500 differences are not big. barry looked at the bluetooth. it is connected to a uart. |
| 13:02.41 | swetland | sz: dunno how bt works on other htc devices (but suspect it would be similar), but on dream, it's a ti brf6300 on uart1 |
| 13:03.38 | swetland | basic bt is probably not too tricky, but routing is going to depend on talking to the baseband, to tell it to send voice over the sco link, etc |
| 13:04.16 | dcordes | http://wiki.xda-developers.com/index.php?pagename=KaiserBT TI BRF6300 |
| 13:06.17 | dcordes | swetland, ok first things first we will look into establishing the basic bluetooth communication first |
| 13:07.35 | dcordes | cr2_, regarding sdcc clocks: http://rafb.net/p/qtJydJ77.html |
| 13:07.43 | dcordes | mmc0: MMC clock 100000 -> 32000000 Hz, PCLK 4294967274 Hz |
| 13:07.51 | dcordes | mmc1: MMC clock 100000 -> 32000000 Hz, PCLK 66000000 Hz |
| 13:09.11 | dcordes | something wrong with it? mmc1 holds the sdio wifi. 66000000 looks rounded |
| 13:15.51 | dcordes | and way too low? |
| 13:16.28 | dcordes | I don't want to play with clocks because I fear it could fry parts of kaiser hardware |
| 13:43.34 | *** join/#htc-linux tuxydo (n=joop@ip5455c588.speed.planet.nl) |
| 13:51.46 | *** join/#htc-linux f0rbid` (n=f0rbid@host210-17-dynamic.17-79-r.retail.telecomitalia.it) |
| 13:53.42 | *** part/#htc-linux f0rbid` (n=f0rbid@host210-17-dynamic.17-79-r.retail.telecomitalia.it) |
| 14:09.58 | *** join/#htc-linux tuxydo (n=joop@ip5455c588.speed.planet.nl) |
| 14:16.16 | *** join/#htc-linux heise2k (n=arrrrrrr@65-78-40-52.c3-0.upd-ubr6.trpr-upd.pa.cable.rcn.com) |
| 14:16.42 | szsoftware | dcordes: can you tell me something about proc_comm? what is it? We don't use, dzo does? |
| 14:17.11 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 14:18.24 | szsoftware | dcordes: nevermind, also please let me understand what proc_comm is in general |
| 14:18.38 | szsoftware | s/also/only/ |
| 14:22.22 | dcordes | szsoftware, interfacing with shared memory, it provides us with access to several arm9 functions |
| 14:25.09 | *** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net) |
| 14:27.30 | dcordes | what's the difference between clck and pclck ? |
| 14:28.28 | dcordes | <PROTECTED> |
| 14:28.28 | dcordes | <PROTECTED> |
| 14:28.39 | dcordes | cr2_, how can I find out the correct values? |
| 14:30.49 | dcordes | cr2_, btw, wifi poweron is not as easy as leaving it in the power on state from wince |
| 14:31.26 | dcordes | SanMehat, says the wlan driver will not initialize correctly with the firmware already running on the sdio wlan |
| 14:47.27 | szsoftware | dcordes: thanks. Could you explain something about cpld vs egpio? We have a choice to use this or the other? We use egpio, dzo cpld? COULD we use cpld too? Why the difference? |
| 14:50.52 | *** join/#htc-linux miknix_ghost (n=miknix@81.193.80.227) |
| 14:51.40 | *** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix) |
| 14:56.30 | szsoftware | ok, I've got to go, getting my son back from his grandma :-). We will see us in the night .. |
| 15:17.14 | *** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde) |
| 15:19.17 | *** join/#htc-linux data|2 (n=datachao@189.27.191.132.adsl.gvt.net.br) |
| 15:24.34 | *** join/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes) |
| 15:28.43 | *** join/#htc-linux joop (n=joop@ip5455c588.speed.planet.nl) |
| 15:31.56 | *** join/#htc-linux fnord (n=fnord@24-151-90-116.static.nwtn.ct.charter.com) [NETSPLIT VICTIM] |
| 15:31.56 | *** join/#htc-linux penguinz (n=opera@66-168-215-122.dhcp.athn.ga.charter.com) [NETSPLIT VICTIM] |
| 15:32.14 | *** join/#htc-linux dante_ (n=dante@79.15.48.176) |
| 15:36.28 | *** join/#htc-linux Alex[sp3dev] (n=alex_dfr@86.110.163.25) |
| 15:42.28 | *** join/#htc-linux FRITZ (n=FRITZ@189-19-168-237.dsl.telesp.net.br) |
| 15:59.36 | *** part/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 16:09.59 | cr2_ | dcordes: i'll point you to the problematic code. |
| 16:10.15 | cr2_ | druidu: are you here ? |
| 16:14.13 | *** join/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 16:14.31 | dcordes | cr2_, clock.c ? |
| 16:15.35 | cr2_ | 0xfbe00000 0xb2d00000 0xb2e00000 0xa8600000 1 MSM_CLK_CTL, DEX |
| 16:15.35 | cr2_ | <PROTECTED> |
| 16:17.23 | cr2_ | but it's not all |
| 16:17.52 | cr2_ | the wiki is not complete, need to cross-check it with vogue code. |
| 16:18.39 | dcordes | atm the sdc1 (wifi) clock_params are a copy of the sdc2 |
| 16:19.03 | dcordes | <PROTECTED> |
| 16:19.03 | dcordes | <PROTECTED> |
| 16:19.03 | dcordes | <PROTECTED> |
| 16:19.03 | dcordes | <PROTECTED> |
| 16:19.03 | dcordes | <PROTECTED> |
| 16:19.04 | dcordes | <PROTECTED> |
| 16:19.06 | dcordes | <PROTECTED> |
| 16:19.08 | dcordes | <PROTECTED> |
| 16:19.10 | dcordes | <PROTECTED> |
| 16:19.12 | dcordes | <PROTECTED> |
| 16:19.21 | SanMehat | i can't believe i'm watching I love lucy |
| 16:20.26 | dcordes | SanMehat, enjoy |
| 16:20.44 | *** join/#htc-linux LunohoD_ (n=alex@e180078103.adsl.alicedsl.de) |
| 16:21.27 | dcordes | cr2_, 0xff9e0b78 is a memory address and 576000 a "pclck"? |
| 16:21.38 | Alex[sp3dev] | hey, guys, is it possible to read a register of an i2c device without registering i2c client and occupying the bus? i mean, i have some i2c driver assigned with my i2c device, and want to just read registers from another one? |
| 16:22.16 | SanMehat | dcordes: how is it going? |
| 16:22.55 | dcordes | SanMehat, thanks, all good, hagover is survived. you? |
| 16:23.09 | dcordes | s/hagover/hangover/ |
| 16:23.49 | SanMehat | :) |
| 16:23.49 | cr2_ | Alex[sp3dev]: the i2c-tools ? |
| 16:23.55 | SanMehat | just getting up |
| 16:24.10 | *** join/#htc-linux tuxydo (n=joop@ip5455c588.speed.planet.nl) |
| 16:24.21 | cr2_ | dcordes: this table looks very strange to me. |
| 16:24.43 | cr2_ | dcordes: but i mean somethign different, i'll look into the vogue wiki |
| 16:25.34 | Alex[sp3dev] | cr2 no, i mean, i need to do it from my driver in kernel, not from shell |
| 16:25.36 | dcordes | cr2_, I don't know what clock parameters are. are they the same on both controllers? |
| 16:26.01 | dcordes | Alex[sp3dev], maybe you can look up in the i2c-tools code how it operates and then adapt it in your driver code? |
| 16:26.48 | cr2_ | Alex[sp3dev]: on pxa ? |
| 16:27.01 | cr2_ | dcordes: he want to do it in the kernel |
| 16:27.04 | Alex[sp3dev] | cr2_ yes,on pxa270 |
| 16:29.13 | cr2_ | Alex[sp3dev]: you can dump the control register |
| 16:31.13 | Alex[sp3dev] | will that be ok if i have multiple devices on i2c? maybe it will be better to move the needed functions from old i2c driver to mine. |
| 16:32.24 | cr2_ | yes, this is the purpose of i2c_id |
| 16:32.51 | cr2_ | and the major difference to spi. |
| 16:38.01 | cr2_ | dcordes: ill try to update the kaiser wiki page |
| 16:39.16 | *** join/#htc-linux EA2 (n=chipper@cpe-024-074-138-191.carolina.res.rr.com) |
| 16:40.25 | dcordes | cr2_, nice |
| 16:42.50 | cr2_ | dcordes: the a0 and a4 clock regs are SD0 ? |
| 16:43.06 | dcordes | out of clue |
| 16:43.59 | cr2_ | where is the vogue git link ? |
| 16:45.02 | *** join/#htc-linux FRITZ (n=FRITZ@189-19-168-237.dsl.telesp.net.br) [NETSPLIT VICTIM] |
| 16:45.02 | *** join/#htc-linux penguinz (n=opera@66-168-215-122.dhcp.athn.ga.charter.com) [NETSPLIT VICTIM] |
| 16:49.31 | dcordes | cr2_, http://www.linuxtogo.org/~lgorris/kaiser-bootkit/README.txt I keep links here |
| 16:49.55 | dcordes | also use it for my own reference http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=shortlog;h=refs/heads/htc-vogue |
| 16:50.09 | cr2_ | dcordes: grep for +int set_sdcc_host_clock(unsigned freq) { |
| 16:50.39 | druidu | back |
| 16:51.03 | cr2_ | dcordes: the file where you ripped the main clocks out |
| 16:51.06 | cr2_ | + writel(sd_clock_parameters[n].a0val,MSM_CLK_CTL_BASE + 0xa0); |
| 16:51.06 | cr2_ | + writel(sd_clock_parameters[n].a4val,MSM_CLK_CTL_BASE + 0xa4); |
| 16:51.07 | cr2_ | + writel(readl(MSM_CLK_CTL_BASE)|(1<<7),MSM_CLK_CTL_BASE); |
| 16:51.18 | cr2_ | dcordes: this is for SD0 |
| 16:51.52 | cr2_ | dcordes: for SD1 you need +0xa8, 0xac and 1<<8 (0x100) |
| 16:52.02 | druidu | I think nothing is working because of smd interface, need to patch that for diamond/raphael |
| 16:54.24 | cr2_ | dcordes: i've edited the wiki, check CLK_CTL |
| 16:55.01 | cr2_ | druidu: there is a lot of other broken things too. |
| 16:55.47 | cr2_ | druidu: i propose to nominate dcordes as the repo maintainer, and send him the pathces :) |
| 16:56.34 | cr2_ | dcordes: can you try to 'pd' these clock regs one by one ? |
| 16:57.06 | druidu | cr2_: can't we get commit acces on main android git? |
| 16:57.09 | druidu | it's open source now |
| 16:57.19 | dcordes | cr2_, yes but I must buy foodstuffs first |
| 16:57.26 | druidu | it would be cool if vogue would be merged to that too |
| 16:57.47 | dcordes | druidu, I doubt it. but if you want to maintain the diamond code on your own you can register for a linuxtogo account. |
| 16:58.07 | druidu | dcordes: how can I access your git or what do you use? |
| 16:58.08 | cr2_ | it's already merged, but we need to fix many things. |
| 16:58.26 | druidu | I don't have time right now to mantain my own... |
| 16:58.31 | cr2_ | there is no way in hell the vogue code will ever be merged into mainline ;) |
| 16:58.54 | druidu | I know it's a terrible hack right now, but we could make it better/cleaner |
| 16:58.58 | dcordes | it will be time consuming to merge kaiser/polaris with vogue and the product with google |
| 16:59.38 | cr2_ | druidu: merging vogue will create a real mess |
| 16:59.45 | druidu | anyway, we could try to make it compatible with actual msm-htc branc from google and submit some patches once we get it working |
| 16:59.56 | druidu | no merging with vogue |
| 17:00.07 | dcordes | druidu, ok then hand me the codestuffs! |
| 17:00.18 | cr2_ | druidu: it's much better to have some plan, and merge the clean code, and reimplementation. |
| 17:00.20 | druidu | for example, for smd... we can modifiy existing driver so that it includes a pluggable board-part (like a header) |
| 17:00.49 | dcordes | I will then put a new branch in linuxtogo, based on the trout branch, that has raph/diamond and the kaiser/polaris code |
| 17:01.07 | druidu | so they'll surely accept it like that because it will be more general, all the general stuff remains in smd*.c/h and we'll have board-xxx-smd.h with board-specific stuff |
| 17:01.09 | cr2_ | dcordes: imho, they overlap |
| 17:01.43 | druidu | no, let's branch of kernel/msm.git android-msm-htc-2.6.25 branch |
| 17:01.43 | dcordes | druidu, we have kaiser-smd.c and vogue-smd.c so we could add a raphael-smd.c |
| 17:01.53 | dcordes | cr2_, what exactly does overlap? |
| 17:01.59 | cr2_ | druidu: hehe, we need to fix the gpio setup, and clock control first. before even thinking about smd. |
| 17:02.54 | druidu | dcordes: it's not a good idea imho, a lot of code is shared for the two, and vogue-smd.c for example lacks a lot of the functionality smd.c provides |
| 17:03.14 | druidu | and about clock controls, I kind of think it's not working because of smd |
| 17:03.21 | druidu | at least that's why the RTC is not working |
| 17:04.02 | cr2_ | well, rtc is the last thing one really needs. |
| 17:04.20 | *** join/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes) |
| 17:04.21 | cr2_ | i'd like to have working SD and wifi first. |
| 17:04.32 | *** join/#htc-linux N3m3515 (n=nembnc@p5B09DF40.dip.t-dialin.net) |
| 17:04.54 | cr2_ | so we can have a rootfs and a comm channel. |
| 17:04.57 | druidu | a busybox would be helpful before that |
| 17:05.15 | druidu | we need to fix proc_comm, a lot of stuff is based on that and just bypassing all that will get us nowhere |
| 17:05.52 | dcordes_ | cr2_, druidu, does raphael/diamond bluetooth work ootb with the trout code? |
| 17:06.08 | cr2_ | gpio setup and clocks are independent of proc_comm. |
| 17:06.22 | cr2_ | it's only android g1 code that uses it. |
| 17:06.30 | cr2_ | because of thier amss |
| 17:06.41 | druidu | no thet're not, arch/arm/mach-msm/clock.c is full of msm_proc_comm calls |
| 17:06.48 | cr2_ | we will ifdef that code anyway. |
| 17:07.18 | druidu | well, it's the *core*... enabling clocks is done via proc_comm as I see it |
| 17:07.19 | cr2_ | druidu: it's for the main cpu clock. we may stay with 528MHz first, and do not touch it. |
| 17:07.36 | druidu | pc_clk_enable(id) => msm_proc_comm(PCOM_CLKCTL_RPC_ENABLE, &id, 0); |
| 17:08.00 | cr2_ | yes, it will not work on "our" amss |
| 17:08.18 | cr2_ | because the clocks are controlled directly |
| 17:08.46 | cr2_ | and all the PCOM_GPIO weird stuff will fail too. |
| 17:08.47 | druidu | ah, ok, then I don't know very much about that... do we have some kind of diagrams of how they're connected? all controlled via gpios? |
| 17:10.23 | cr2_ | check thr wiki for the 0xa8600000 block |
| 17:10.27 | cr2_ | also for kaiser |
| 17:11.53 | cr2_ | dcordes: this code should be replaced by a "switch" |
| 17:11.56 | cr2_ | +static int pc_clk_enable(unsigned id) |
| 17:11.56 | cr2_ | +{ |
| 17:11.57 | cr2_ | + if(id==SDC1_CLK) |
| 17:11.57 | cr2_ | + writel(readl(MSM_CLK_CTL_BASE)|(1<<7),MSM_CLK_CTL_BASE); |
| 17:11.57 | cr2_ | + return 0; |
| 17:12.01 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) |
| 17:12.21 | cr2_ | dcordes: the same for the clock divisors |
| 17:12.41 | cr2_ | there are much more devices than SDC1 |
| 17:14.20 | dcordes_ | ok bb in 20 |
| 17:15.16 | *** join/#htc-linux surgex (i=surge@pool-98-118-165-224.bflony.fios.verizon.net) |
| 17:16.38 | druidu | maybe an array if we have many of them |
| 17:18.55 | cr2_ | i think a switch is more readable |
| 17:19.22 | cr2_ | actually we talk about the i2c, sd,wifi and cam clocks right now. |
| 17:20.26 | cr2_ | there are more other clocks, but since we don't have the docs we can only guess |
| 17:20.40 | cr2_ | and these other clocks are not controlled by wince |
| 17:21.37 | cr2_ | druidu: i still can't undertstand how to get the right oemsbl offset/entry |
| 17:21.56 | cr2_ | druidu: the pic parts are ok, but some memory offsets are completely broekn. |
| 17:22.08 | szsoftware | dzo, good morning! Are you around already? |
| 17:26.26 | *** join/#htc-linux tuxydo (n=joop@ip5455c588.speed.planet.nl) |
| 17:28.43 | szsoftware | cr2_: would you review vogue-ts.c and give a hint/short explanation where we can abstract cpld/egpio stuff? |
| 17:29.45 | cr2_ | szsoftware: look at the n560 code in hh.org CVS |
| 17:30.47 | NetRipper | dcordes_, no, the trout code does not work ootb |
| 17:33.47 | szsoftware | cr2_: please give me a hint. Where to look: http://handhelds.org/cgi-bin/cvsweb.cgi/#dirlist |
| 17:36.40 | cr2_ | szsoftware: we have many devices using htc-egpio |
| 17:36.47 | cr2_ | looking for a link |
| 17:43.02 | cr2_ | http://handhelds.org/cgi-bin/cvsweb.cgi/linux/kernel26/arch/arm/mach-pxa/looxc550/looxc550.c?rev=1.21&content-type=text/x-cvsweb-markup |
| 17:43.16 | cr2_ | grep for 'gpio_set_value' |
| 17:44.01 | dcordes_ | I will do the wifi clock later |
| 17:44.19 | cr2_ | dcordes_: think about i2c clock too |
| 17:44.29 | cr2_ | dcordes_: it's also needed here . |
| 17:44.37 | cr2_ | dcordes_: ok kaiser too. |
| 17:44.43 | cr2_ | s/ok /on / |
| 17:47.22 | szsoftware | cr2_: thanks for that link/hint |
| 17:51.44 | *** join/#htc-linux Marnux (n=Marnux@83.101.56.249) |
| 17:55.00 | *** part/#htc-linux penguinz (n=opera@66-168-215-122.dhcp.athn.ga.charter.com) |
| 18:04.06 | ALoGeNo | lo |
| 18:11.29 | *** join/#htc-linux BruteSource (n=BruteSou@m2b0e36d0.tmodns.net) |
| 18:15.22 | BabelO | cr2_: better is to use the egpio driver i use for artemis at htc-omap in git.linuxtogo.org ... |
| 18:24.22 | druidu | cr2_: the kitchen for raphael doesn't seem to work very |
| 18:24.27 | szsoftware | dcordes_: around? who did the polaris adaption of vogue-ts.c? |
| 18:24.56 | druidu | good |
| 18:26.32 | *** part/#htc-linux Jones (i=jjones@butters.mozor.net) |
| 18:33.58 | *** join/#htc-linux dante_ (n=dante@net-93-148-251-28.t2.dsl.vodafone.it) |
| 19:07.45 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 19:18.09 | *** join/#htc-linux datachaos (n=datachao@189.27.191.132.adsl.gvt.net.br) |
| 19:21.36 | cr2_ | druidu: if you want to have some fun, you can write a touchscreen driver |
| 19:25.24 | druidu | I'll do that as soon as I can get a busybox booting or something :P |
| 19:25.46 | druidu | I mean, can we test anything until we finish the basic low-level stuff? |
| 19:26.18 | NetRipper | i have an initrd that'll give you a busybox shell |
| 19:26.19 | cr2_ | yes. it's a kernel driver |
| 19:26.22 | NetRipper | if you want |
| 19:26.32 | cr2_ | it will spit the messages on the screen. |
| 19:26.44 | NetRipper | im doing something else right now, but i'll be back in 30 |
| 19:26.46 | cr2_ | when you touch the screen you get an irq |
| 19:26.49 | druidu | did it work for you NetRipper ? |
| 19:26.59 | cr2_ | and integration begins |
| 19:27.08 | NetRipper | druidu, yes i got a shell, though i couldn't do anything but type the 1, 4 and ~ characters |
| 19:27.14 | NetRipper | (via hardware buttons) |
| 19:27.16 | druidu | I could get to the freeing init memory, it hanged right after hcrtctosys |
| 19:27.31 | druidu | send me please |
| 19:27.40 | cr2_ | when you release the stylus, you get another irq, and only need to read the x/y regs. imho that's all |
| 19:28.08 | druidu | and maybe a `git-diff` to see what else you disabled there |
| 19:28.10 | NetRipper | druidu, when im finished what im doin now i'll remove my debug crap and give it |
| 19:28.13 | cr2_ | a spinlock, and 2 irq handlers |
| 19:28.16 | druidu | ok, thanks |
| 19:28.39 | druidu | for the moment I disabled all the msm drivers and trying to get them up one by one, fully functional :) |
| 19:29.01 | druidu | and well, proc_comm is the first <sigh> |
| 19:29.45 | druidu | cr2_ : theoretically, shouldn't the vogue proc_comm code work? the memory map you put together seems to confirm the addresses |
| 19:30.07 | cr2_ | it will |
| 19:34.35 | druidu | where are the docs we have on the touchscreen? you said in a post that the protocol is understood |
| 19:35.11 | cr2_ | hmm. |
| 19:35.23 | cr2_ | mostly in the wiki |
| 19:35.43 | cr2_ | and from the trace. |
| 19:36.16 | druidu | the there's nothing about the touchscreen in raphael research pagte |
| 19:36.17 | druidu | *page |
| 19:36.24 | druidu | except "Built-in MSM7201A touchscreen controller. |
| 19:37.02 | druidu | btw, where's google's touchscreen driver? |
| 19:38.11 | cr2_ | look at memorymap page |
| 19:38.40 | druidu | by default msm config enables synaptics i2c rmi |
| 19:38.51 | druidu | which should be the driver g1 is using for touchscreen |
| 19:39.02 | druidu | msm_defconfig: CONFIG_TOUCHSCREEN_SYNAPTICS_I2C_RMI=y |
| 19:39.21 | cr2_ | INT_TCHSCRN1 is pen down |
| 19:39.38 | cr2_ | INT_TCHSCRN2 is pen up |
| 19:39.57 | cr2_ | http://wiki.xda-developers.com/index.php?pagename=Raphael_IRQ |
| 19:40.17 | druidu | and how do we know the position? |
| 19:40.39 | cr2_ | <PROTECTED> |
| 19:41.03 | druidu | can we verify in haret? |
| 19:41.15 | cr2_ | <PROTECTED> |
| 19:41.16 | cr2_ | <PROTECTED> |
| 19:41.32 | cr2_ | yes. add this area to mmutrace |
| 19:41.38 | NetRipper | oh god |
| 19:41.52 | NetRipper | i just extracted an initrd to my own linux root filesystem |
| 19:41.55 | cr2_ | i had nice traces on raphael |
| 19:42.02 | druidu | fuck?! |
| 19:42.07 | druidu | did you erase everything? |
| 19:42.28 | NetRipper | i sure hope it didnt overwrite anything major |
| 19:42.36 | druidu | it should interfere too much, you android uses /sys and /system which don't mean much in linux |
| 19:42.44 | NetRipper | well... it wasn't an android initrd |
| 19:42.53 | cr2_ | hmm. |
| 19:42.56 | druidu | oh... |
| 19:42.56 | NetRipper | anyway |
| 19:43.00 | NetRipper | i'll continue as long as i dont crash |
| 19:43.14 | cr2_ | l works ? |
| 19:43.23 | cr2_ | s/l /ls / |
| 19:43.26 | NetRipper | yes |
| 19:43.51 | cr2_ | <PROTECTED> |
| 19:44.56 | druidu | initrds don't have many stuff in /etc, hopefully |
| 19:45.07 | cr2_ | druidu: only TS init needs better docs. and the TSSC_STATUS bits. |
| 19:45.32 | druidu | that means TS is not i2c right? |
| 19:45.40 | NetRipper | heh it was a small initrd luckily, though it overwrite some important libraries |
| 19:46.09 | *** join/#htc-linux tuxydo (n=joop@ip5455c588.speed.planet.nl) |
| 19:46.14 | druidu | what distro? |
| 19:46.21 | NetRipper | ubuntu |
| 19:46.45 | NetRipper | my gcc library and libcrypt got overwritten |
| 19:47.03 | NetRipper | i'll fix it later |
| 19:47.19 | NetRipper | --no-absolute-filenames is a must-know switch for working with cpio from now on |
| 19:47.27 | cr2_ | druidu: forget about g1,vogue,kaiser et al. |
| 19:47.46 | cr2_ | NetRipper: it uses rpm ? |
| 19:47.56 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 19:48.22 | druidu | apt-get --reinstall libc6 libxcrypt1 |
| 19:48.40 | cr2_ | druidu: it'll be a completely standalone driver. and it seems that most dirty work is done in hardware. |
| 19:48.55 | NetRipper | druidu, thx, first lets get that initrd for raph done :p |
| 19:49.20 | druidu | we just need /busybox statically linked :) |
| 19:49.37 | NetRipper | i got a working busybox |
| 19:49.39 | druidu | I compiled one that might work, but couldn't test it... gotta see why I don't get to freeing init memory |
| 19:50.01 | NetRipper | eh, you dont get to freeing init? |
| 19:50.15 | druidu | it stops exactly after hcrtctosys |
| 19:50.18 | druidu | though that is not the problem |
| 19:51.02 | cr2_ | dcordes_: added CAM reset gpio |
| 19:51.03 | druidu | gonna run some more tests to see |
| 19:51.17 | druidu | did we get usb to work? |
| 19:51.53 | druidu | serial usb connection would be much better, it would be easy to write a terminal and hook it with a busybox or something |
| 19:54.24 | NetRipper | druidu, i once had usb working up to the point where my linux recognized the device, but i can't get it back to working.. i'll work on that once i have a proper initrd first |
| 19:55.01 | NetRipper | as it's an initrd that is prepared for cdc_ether, though it probably doesnt work for raph |
| 19:55.05 | NetRipper | /diam |
| 19:56.42 | cr2_ | NetRipper: somebody managed to switch to g1 bootloader, but he says it's also not recognized by linux |
| 19:57.23 | NetRipper | cr2_, eh, like, flashing a new spl? |
| 19:58.53 | cr2_ | lol |
| 19:58.59 | cr2_ | dumping first |
| 20:00.46 | druidu | if we flash a rom and brick the device, can we repair it? :) |
| 20:01.03 | NetRipper | not if you flash a faulty spl |
| 20:01.05 | druidu | I'm thinking that at some point we're gonna package a nice ROM for the android |
| 20:01.11 | cr2_ | if it's the wince area, then yes |
| 20:01.22 | druidu | do we need another spl or can we use a wince spl? |
| 20:01.29 | cr2_ | we can use wince |
| 20:01.35 | NetRipper | we hope |
| 20:02.01 | druidu | hehe, so a broken spl means brick for good? |
| 20:02.16 | cr2_ | i think wince spl is probably irrelevant |
| 20:02.23 | cr2_ | aka appsbl |
| 20:02.34 | cr2_ | the real thing is the oemsbl |
| 20:02.43 | cr2_ | because it can deal with the flash too |
| 20:03.23 | cr2_ | appsbl is a wince "console" app |
| 20:03.42 | cr2_ | so it looks like on the older phones |
| 20:04.13 | cr2_ | druidu: so we will need the offset and entry for oemsbl :) |
| 20:04.41 | druidu | where, think it like this: the hardware needs to know where to boot from, so it's probably something very static |
| 20:05.39 | druidu | it's pretty easy to look at some arm asm and see if it's junk or not |
| 20:09.51 | NetRipper | linux just uses its current address as base, it can boot from anywhere in the memory afaik |
| 20:10.15 | druidu | well yeah, but there is still the bootsector on the hdd |
| 20:10.20 | NetRipper | yes |
| 20:10.20 | druidu | which is initialized |
| 20:10.34 | NetRipper | haret is doing the bootsector's job\ |
| 20:10.38 | druidu | we just have the nand flash, so our "bootsector" must be somewhere in there |
| 20:11.16 | druidu | no, I'm talking about the spl... cr2 is trying to figure out entry point in oemsbl to disassemble the code |
| 20:11.23 | NetRipper | ah ok |
| 20:17.36 | NetRipper | btw, the t-mobile G1 logo is no good, it's too wide, fucks up the screen |
| 20:17.54 | cr2_ | lol |
| 20:18.05 | NetRipper | maybe i can try rotating the screen |
| 20:18.11 | NetRipper | i'll do that later |
| 20:20.22 | NetRipper | druidu, http://www.netripper.nl/raphael/initrd-netripper-busybox-usb.cpio.gz |
| 20:21.04 | NetRipper | and do yourself a favor, remove 8x16 font from kernel... you get much more info on screen using 8x8 font |
| 20:21.08 | NetRipper | :) |
| 20:23.32 | cr2_ | NetRipper: screenshot ? |
| 20:24.03 | NetRipper | not much to see though |
| 20:24.09 | cr2_ | with such dpi, you can even use a smaller font |
| 20:28.47 | *** join/#htc-linux ssapphiree (n=sapphire@ppp91-77-158-123.pppoe.mtu-net.ru) |
| 20:29.12 | NetRipper | well yea but i dont want to pain my eyes :P |
| 20:29.27 | NetRipper | rofl |
| 20:29.30 | NetRipper | terminal blanked |
| 20:31.51 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfc46d.pool.einsundeins.de) |
| 20:34.00 | NetRipper | cr2_, http://www.netripper.nl/raphael/pics/20081025_screenshot1.jpg |
| 20:37.29 | cr2_ | druidu: the 0x800 header must go away |
| 20:37.47 | cr2_ | druidu: i've checked with direct sram dump |
| 20:38.05 | NetRipper | we're missing some stuff in the g1 kernel git |
| 20:38.17 | cr2_ | NetRipper: which nand chip is it ? |
| 20:38.40 | cr2_ | NetRipper: hehe. the 6 nand partitions :) |
| 20:38.57 | NetRipper | cr2_, http://www.netripper.nl/raphael/g1_config_unrecognized_defines.txt |
| 20:38.59 | NetRipper | i meant |
| 20:39.02 | cr2_ | smc91x should go away too |
| 20:39.17 | NetRipper | ANDROID_GADGET |
| 20:39.21 | NetRipper | would be very nice to have |
| 20:39.23 | cr2_ | NetRipper: create the board-rapahel |
| 20:39.39 | cr2_ | we will remove irrelevant things, and add correct gpios |
| 20:39.57 | NetRipper | ok, should we setup a git somewhere for our stuff? |
| 20:40.03 | cr2_ | then we will bash android guys to remove the g1-specific code from the device drivers. |
| 20:40.17 | cr2_ | if you'll manage it :) |
| 20:40.32 | NetRipper | i cant host it myself |
| 20:40.36 | NetRipper | it's too big |
| 20:40.40 | cr2_ | 5501bcec |
| 20:40.45 | cr2_ | linuxtogo |
| 20:40.58 | cr2_ | where the kaiser and vogue are hosted |
| 20:41.09 | NetRipper | how should we call the head? htc-raphael or htc-diamond or..? |
| 20:41.11 | cr2_ | or even htc-linux.org |
| 20:41.22 | cr2_ | 7201a ? |
| 20:41.35 | cr2_ | no, that includes g1 ;) |
| 20:41.38 | NetRipper | hehe yea |
| 20:41.41 | NetRipper | it's too generic |
| 20:41.47 | cr2_ | why not |
| 20:42.07 | cr2_ | it shows that g1 code is not really designed for multiple boards |
| 20:42.39 | cr2_ | omg |
| 20:42.44 | cr2_ | ps/2 mouse ? |
| 20:43.17 | cr2_ | halibut_keypad should be rapahel_keypad. with corrected gpios :) |
| 20:43.20 | NetRipper | well, i'm not really going for making a kernel that runs everything or is perfectly clean.. i just want it working for raph/diamond ;) |
| 20:43.27 | NetRipper | yes |
| 20:43.48 | cr2_ | compass can go away, it'll only spam on the i2c bus |
| 20:44.10 | NetRipper | problem with removing some drivers is that sometimes it wont compile anymore |
| 20:44.19 | cr2_ | ok, mmc needs the clock |
| 20:44.27 | cr2_ | for wifi and sd |
| 20:44.52 | cr2_ | rtc fails because of proc_comm probably |
| 20:46.22 | cr2_ | dcordes_: documented 2 more clocks |
| 20:47.19 | cr2_ | maker ec device bc |
| 20:47.53 | cr2_ | <PROTECTED> |
| 20:48.10 | NetRipper | what the hell is a ram buffer console |
| 20:48.14 | cr2_ | we can hardcode the table |
| 20:48.27 | cr2_ | it's some android buffer |
| 20:48.34 | NetRipper | we dont need it do we |
| 20:48.44 | cr2_ | like wince "dmesg" buffer |
| 20:48.54 | NetRipper | ah |
| 20:49.11 | cr2_ | you can add a 'cat' of this area to your initrd startup |
| 20:49.26 | NetRipper | i dont know where it puts it |
| 20:49.34 | cr2_ | check the g1 code |
| 20:49.38 | NetRipper | i will |
| 20:49.48 | cr2_ | that's why i'm for the DreamMemoryMap wiki page :) |
| 20:50.47 | cr2_ | NetRipper: if you are the tough hacker, you can enable the SD clock in the board init :) |
| 20:50.57 | NetRipper | why tough? |
| 20:50.58 | NetRipper | :p |
| 20:51.05 | cr2_ | don't know if it'll survive switching :) |
| 20:51.23 | NetRipper | lol worst case i brick it? |
| 20:51.24 | NetRipper | :p |
| 20:51.28 | NetRipper | (or damage) |
| 20:51.29 | cr2_ | because you need to know the addresses and values |
| 20:51.41 | cr2_ | you can write the vogue values :) |
| 20:51.45 | cr2_ | wait, i'll check |
| 20:52.04 | cr2_ | btw, the i2c clock reg location is different between 7200 and 7200a |
| 20:53.21 | cr2_ | hm. it seems they uses some kind of clock api. |
| 20:53.34 | cr2_ | for the newer devices |
| 20:54.39 | *** join/#htc-linux data|2 (n=datachao@189.27.191.132.adsl.gvt.net.br) |
| 20:57.54 | cr2_ | NetRipper: if you removed these drivers, remove the bluetooth too. its setup is completely wrong |
| 20:58.09 | cr2_ | then you can create linload.exe :) |
| 20:58.37 | cr2_ | druidu: which nand chip do you have ? |
| 20:59.00 | NetRipper | im starting with a clean .config again give me a moment to get console again |
| 20:59.17 | NetRipper | (in an attempt to get usb working again) |
| 21:00.01 | NetRipper | oh yea |
| 21:00.16 | NetRipper | Windows needs to install driver software for you Halibut |
| 21:00.16 | NetRipper | :D |
| 21:01.57 | cr2_ | lol |
| 21:05.30 | NetRipper | didnt have to disable PPP driver to boot btw |
| 21:05.56 | NetRipper | only using the IRQ patch now |
| 21:07.51 | szsoftware | cr2_: NetRipper: druidu: any experience with php or other language using classes as a certain term (IMHO java too) |
| 21:08.10 | NetRipper | szsoftware, term? |
| 21:08.58 | NetRipper | szsoftware, i have experience with php/java/c# and some more |
| 21:09.19 | szsoftware | NetRipper: ok I need understanding of c language in comparism to other |
| 21:09.36 | cr2_ | NetRipper: found the enable bits for more clk divisors :) |
| 21:09.42 | szsoftware | NetRipper: can I say a c struct is that, what in php a class could be? |
| 21:09.55 | NetRipper | szsoftware, no |
| 21:10.34 | szsoftware | NetRipper: f**ck, but it looks so similar in defining and usage |
| 21:10.43 | NetRipper | szsoftware, well you could compare it to a class with only public properties |
| 21:10.52 | NetRipper | szsoftware, but memory layout is completely different |
| 21:11.04 | cr2_ | hmm. it seems we have a problem here too |
| 21:11.09 | ali1234 | there's no reason why a struct cant have public members too, in the form of function pointers |
| 21:11.38 | NetRipper | a struct is a collection of data, it has no intelligence |
| 21:11.41 | szsoftware | NetRipper: ok,I just need to imagine new stuff |
| 21:12.20 | szsoftware | NetRipper: but in php I can use a class, simulating a c struct? |
| 21:12.51 | szsoftware | NetRipper: at least handling data |
| 21:13.07 | NetRipper | szsoftware, you can use it for the same purpose |
| 21:13.34 | szsoftware | NetRipper: ok thanks, I will dive into relating books again |
| 21:13.40 | NetRipper | szsoftware, defining a class with only public members will resemble a struct |
| 21:14.10 | NetRipper | szsoftware, what's your goal here? |
| 21:14.29 | NetRipper | cr2_, what problem? |
| 21:14.45 | szsoftware | NetRipper: sorry I abuse this channel. My goal? |
| 21:15.11 | NetRipper | szsoftware, i mean, why are you trying to make a 'struct' in php? |
| 21:16.17 | szsoftware | NetRipper: no, I just want have more understanding of c, and therefore I need comparism to what I know more than c. |
| 21:17.24 | szsoftware | NetRipper: I was wondering about defining the struct, and their usage. So I thought to derivation, which I mean is used regarding classes |
| 21:18.39 | NetRipper | szsoftware, ah, i'm not all that experience with c and cpp either.. just from some basic hacking and a little bit of past experience |
| 21:18.57 | ali1234 | szsoftware: in a nutshell, structs are classes without the namespace, inheritance and public/private separation rules, and without the syntactic shortcuts for defining function pointers |
| 21:20.54 | parmaster | htcblueangel and himalaya: code cleanup. better board_id=4 handling. |
| 21:21.02 | druidu | back, a little wireless problem |
| 21:21.23 | druidu | nand chip? not sure, where can I look? |
| 21:21.35 | parmaster | oleg's update from a year and half ago |
| 21:21.57 | szsoftware | ali1234: very much thanks, I might be wrong in using correct terms, but what You said, I had in mind :-) |
| 21:21.58 | parmaster | still camera flash led stays on tho. |
| 21:22.16 | druidu | NetRipper : so you use halibut board? |
| 21:22.33 | szsoftware | so let's go again to msm hacking! Yeah |
| 21:22.50 | szsoftware | s/msm/htc/ |
| 21:23.04 | ali1234 | druidu: nand chip type will probably be G3 or G4... you might have seen those terms in relation to unlocking utils, if you have researched those |
| 21:23.44 | NetRipper | druidu, yes, but with MACH_TROUT enabled |
| 21:23.53 | ginge | No, the I2C spikes when it is disabled. The internal reconnection of te GPIO is causing some oddities. |
| 21:24.03 | ginge | erm. Hi guys, wrong channel |
| 21:24.52 | NetRipper | cr2_, "MT9T013 Camera Driver implemented by HTC" you reckon that's the same cam as on raph? |
| 21:25.22 | druidu | camera is like the top of the cake, we don't have the cake yet! |
| 21:25.44 | NetRipper | im just wondering if it'd be the right driver ;) |
| 21:25.47 | dcordes_ | yo |
| 21:25.58 | cr2_ | NetRipper: yes |
| 21:26.17 | NetRipper | dcordes_, you mentioned setting up git on linuxtogo, how? :) |
| 21:26.43 | szsoftware | dcordes_: yo man, how it's going? |
| 21:26.49 | dcordes_ | szsoftware, "who did the polaris adaption of vogue-ts.c?" is there any? |
| 21:27.10 | cr2_ | i've opened big can of worms now. let me concentrate :) |
| 21:27.34 | druidu | btw, the equivalent of a struct in php would be either an associative array or an object (class extended from stdObject), the object using kind of the same notations for accessing structs |
| 21:28.08 | druidu | but in both java and php they are irrelevant -- they have automatic memory management and don't let you touch pointers, in C the pointer is the core of the stuff :) |
| 21:28.08 | szsoftware | dcordes_: someone put polaris ifdefs in and commented stuff from dzo out, and did things related to cpld vs htc-egpio, no? |
| 21:28.52 | ali1234 | druidu: szsoftware: right, the key to understanding C is understanding pointers |
| 21:29.13 | druidu | and especially the key to understanding kernel / lowlevel code |
| 21:29.23 | druidu | I can't imagine kernel code with no pointers :) |
| 21:29.59 | szsoftware | dcordes_: I have a good "pointers in a nutshell on the example of vogue-ts.c" written by druidu |
| 21:30.17 | druidu | :D |
| 21:30.23 | szsoftware | dcordes_: it's on a PM window in my irssi, not logged. |
| 21:30.36 | szsoftware | dcordes_: how to save the window's content? |
| 21:31.09 | druidu | gpm? |
| 21:31.13 | druidu | copy-paste |
| 21:31.18 | druidu | http://www.netripper.nl/raphael/pics/20081025_screenshot1.jpg |
| 21:31.39 | dcordes_ | szsoftware, premy (patrick remy) |
| 21:31.50 | dcordes_ | he made all of the polaris specific code |
| 21:32.28 | szsoftware | druidu: is'nt it a mess because of not scrolling the window, when marking text? |
| 21:32.38 | szsoftware | dcordes_: is irssi ncurses? |
| 21:32.46 | szsoftware | dcordes_: using? |
| 21:33.10 | szsoftware | dcordes_: premy used not to be here, or? |
| 21:33.17 | dcordes_ | I use xchat now |
| 21:33.35 | dcordes_ | he is busy at the moment |
| 21:34.06 | dcordes_ | I tried to contact him whene I tried to enable the on screen keyboard for polaris in the 2.6.25 branch |
| 21:35.02 | *** join/#htc-linux tuxydo (n=joop@ip5455c588.speed.planet.nl) |
| 21:35.46 | druidu | http://octav.oricum.ro/pointers.txt |
| 21:37.31 | szsoftware | ahh, very kind. Just did it with c&p, haha |
| 21:38.06 | szsoftware | druidu: now my dumbness is in the web :-) |
| 21:38.23 | druidu | :) |
| 21:38.38 | druidu | NetRipper : you could disable all networking btw, there's no need for it now |
| 21:38.43 | druidu | in the kernel config |
| 21:40.28 | NetRipper | not if i want usb network to work |
| 21:48.07 | *** join/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 21:48.48 | parmaster | heh, i looked at the pics... good job so far ;) |
| 21:48.56 | *** part/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 22:00.20 | *** join/#htc-linux datachaos (n=datachao@189.27.191.132.adsl.gvt.net.br) |
| 22:02.19 | cr2_ | ok, i have a better idea now. |
| 22:02.31 | cr2_ | NetRipper: can you run haret ? |
| 22:02.37 | NetRipper | i can |
| 22:02.41 | NetRipper | magically |
| 22:03.02 | cr2_ | :) |
| 22:03.39 | cr2_ | can you dump 8 registers with wifi on/off, and sd on/off ? |
| 22:03.58 | NetRipper | sure, give me the command |
| 22:04.15 | NetRipper | sd? |
| 22:04.16 | dcordes_ | ginge, did you see the wifi code? |
| 22:04.19 | NetRipper | sd card in/out you mean? |
| 22:04.29 | cr2_ | yes |
| 22:05.03 | cr2_ | pd of the 00 a0 a4 a8 ac b0 b4 b8 bc clock registers. |
| 22:05.46 | cr2_ | write it into the default.txt |
| 22:05.53 | cr2_ | and then just run the script. |
| 22:06.14 | cr2_ | pd 0xa8600000 4 |
| 22:06.17 | cr2_ | and so on |
| 22:06.32 | cr2_ | +0xa0 , 0xa4 ... |
| 22:07.04 | cr2_ | i wish haret could print the bits directly. |
| 22:07.45 | cr2_ | <PROTECTED> |
| 22:08.01 | cr2_ | hehe |
| 22:08.03 | cr2_ | %B |
| 22:08.34 | NetRipper | hehe |
| 22:10.08 | toer | :) |
| 22:10.42 | cr2_ | i think a0,a4 are sd and a8,ac are wifi |
| 22:11.49 | cr2_ | 48,24,19,12MHz and 144kHz |
| 22:11.52 | ginge | cr2_: same layout as kaiser then |
| 22:11.58 | ginge | dcordes_: not yet |
| 22:13.24 | cr2_ | ginge: you have running wifi on kaiser ? |
| 22:13.39 | cr2_ | ginge: i've edited the kaiser clock wiki just some time ago |
| 22:13.53 | ginge | cr2_: not tried yet, but should be easy enough assuming we know the clocks |
| 22:13.55 | ginge | oh |
| 22:14.00 | ginge | link? |
| 22:14.07 | cr2_ | ginge: A cpu has 0xb0,0xb4,0xb8 and 0xbc for the other SD slots |
| 22:14.25 | cr2_ | ginge: i need to validate the +0x00 clock bits. |
| 22:14.44 | cr2_ | and the divisor calculation in the vogue code is a bit strange. |
| 22:14.59 | ginge | cr2_: how so? |
| 22:15.09 | cr2_ | hehe. mmc runs at 12MHz. why not 19... |
| 22:15.19 | NetRipper | cr2_, www.netripper.nl/raphael/ and then the regs_*.txt files |
| 22:15.23 | cr2_ | ginge: it's some hardcoded table i think |
| 22:15.27 | cr2_ | NetRipper: thanks. |
| 22:15.43 | ginge | cr2_: yeah myself and dzo wrote them |
| 22:15.55 | ginge | we ripped the clocks from the spl and what ce does |
| 22:16.12 | cr2_ | ginge: still looks strange. |
| 22:16.57 | cr2_ | ginge: i see the function params in the spl |
| 22:17.51 | ginge | we know the wifi clocks dont work properly, just a clone of the sd clocks |
| 22:18.29 | cr2_ | slot is the first parameter. |
| 22:18.37 | cr2_ | then there are 3 others |
| 22:19.09 | cr2_ | i need to convert the mask into 3par combination |
| 22:19.26 | cr2_ | then we can talk about the math. |
| 22:20.37 | NetRipper | awww |
| 22:20.52 | NetRipper | even with fbcon rotated, the t-mobile g1 logo is still written the normal way |
| 22:21.38 | cr2_ | i guess we don't need the t-mob logo |
| 22:21.55 | dcordes_ | NetRipper, FB_MSM_LOGO |
| 22:21.57 | dcordes_ | ? |
| 22:22.15 | NetRipper | dcordes_, yes that and /logo.rle |
| 22:22.26 | dcordes_ | did you try and edit it? |
| 22:22.30 | NetRipper | it draws, but it seems like the width is wrong |
| 22:22.44 | NetRipper | it continues drawing on next pixel-line |
| 22:22.46 | cr2_ | ok, it starts with 144kHz |
| 22:23.33 | cr2_ | ginge: but the wifi clock is written into dmesg ? |
| 22:24.06 | NetRipper | it's funny to see it boot on high speed camera |
| 22:24.06 | dcordes_ | cr2_, yes |
| 22:24.14 | NetRipper | graphics corruption all the way |
| 22:24.23 | ginge | true, but the dividers and addresses are seemingly different |
| 22:24.45 | BruteSource | you have a high speed camera? |
| 22:24.47 | dcordes_ | cr2_, here's the dmesg http://rafb.net/p/0BuJ8V28.html |
| 22:24.54 | NetRipper | well, 60 fps |
| 22:25.07 | NetRipper | on my digicam |
| 22:25.28 | cr2_ | ginge: i don't see any such table in my spl |
| 22:25.29 | NetRipper | only 320x240 though, but still ;) |
| 22:25.59 | BruteSource | is that near the fps of the high speed cameras they use on mythbusters? |
| 22:26.03 | ginge | spl was used to determine the layout iirc the actual table was generated from the sd dll |
| 22:26.20 | NetRipper | BruteSource, dont think so, that's more like 1000 fps |
| 22:26.30 | BruteSource | damn |
| 22:26.33 | cr2_ | dcordes_: strange values. but i mean the wince dmesg |
| 22:27.02 | NetRipper | BruteSource, even worse: "High speed cameras can film up to 250,000 frame/s by running the film over..." |
| 22:27.05 | NetRipper | ;) |
| 22:27.07 | NetRipper | on wikipedia |
| 22:28.10 | cr2_ | ginge: the code does not depend on the slot. it's probably just some macro |
| 22:28.37 | dcordes_ | when I tried to load the wlan module 3 times in a sequence, the -22 error appears |
| 22:29.21 | ginge | if you use the same clock settings for the wifi and sd card slot, it locks on init |
| 22:29.24 | dcordes_ | do I need CONFIG_MMC_EMBEDDED_SDIO ? |
| 22:29.41 | ginge | the 20mhz clock is different for both devices, and the only wifi related clock that I think is right |
| 22:29.55 | ginge | dcordes_: your dmeg uses 32mhz which is too fast and probably wrong clock |
| 22:30.01 | cr2_ | ginge: the sd begins at 144kHz |
| 22:30.15 | druidu | ok, made some changes to my console driver, now will automagically take the best compiled font (should have width <= 8) and use that :D |
| 22:30.21 | cr2_ | as it should. it's the minimal clock |
| 22:30.32 | cr2_ | and 48MHz is the maximum |
| 22:30.36 | ginge | cr2_: ok. we were not sure on that |
| 22:30.58 | cr2_ | i didn't measure it, but that's what the spl code says |
| 22:31.31 | cr2_ | ginge: i'll check the spl |
| 22:31.36 | cr2_ | for wifi |
| 22:32.22 | cr2_ | wifi does 1,4,2 |
| 22:32.36 | cr2_ | sorry |
| 22:32.38 | cr2_ | 1,2,4 |
| 22:32.47 | BruteSource | our android userspace is working on excalibur, just once in while hw clock resets to 1970 and freezes |
| 22:33.10 | cr2_ | yes, "2" is 144kHz too |
| 22:33.35 | cr2_ | then it sends cmd0 |
| 22:33.56 | cr2_ | hm. i always thought cmd5 must be the first sdio command. |
| 22:34.13 | ginge | at least in mmc spec, yeah |
| 22:34.28 | cr2_ | then cmd5, then cmd3 |
| 22:34.32 | cr2_ | all at 144kHz |
| 22:34.53 | cr2_ | cmd7 and then cmd0x34 |
| 22:34.58 | cr2_ | 52 i think |
| 22:35.01 | cr2_ | io_read |
| 22:35.20 | cr2_ | and only then |
| 22:35.25 | cr2_ | it changes the clock to |
| 22:35.28 | cr2_ | 1,4,4 |
| 22:35.59 | cr2_ | which is 12MHz |
| 22:36.59 | cr2_ | so wifi in the spl runs at 12MHz |
| 22:37.11 | dcordes_ | how can I set that? |
| 22:37.12 | cr2_ | sdio wifi |
| 22:37.39 | cr2_ | the current code is a big hack :) |
| 22:37.45 | cr2_ | i need to soft out the math. |
| 22:38.05 | cr2_ | s/soft/sort/ |
| 22:39.36 | cr2_ | going through sdhc dll is a pita, because it ioremaps the registers, and puts it into a big struct |
| 22:39.50 | ginge | yeah really |
| 22:39.59 | ginge | hence the hackiness ;) |
| 22:41.50 | cr2_ | 0x80 and 0x100 |
| 22:42.05 | cr2_ | ok, so the enable bits are preserved too. |
| 22:42.13 | cr2_ | compared no non-A |
| 22:43.04 | *** join/#htc-linux dah1979 (n=pocketir@94.163.25.125) |
| 22:44.18 | BruteSource | i wish i knew how to write the radio drivers |
| 22:50.53 | szsoftware | druidu: I make success --> http://pw1.netcom.com/~tjensen/ptr/pointers.htm: Chapter 4 and a lot of coffee :-) |
| 22:51.07 | *** join/#htc-linux LanceHaig (i=4e696aee@foresight/member/lhaig) |
| 22:51.13 | druidu | hehe, that's good :) |
| 22:51.24 | druidu | although chapter 5 has structures |
| 22:52.24 | druidu | speaking of radio, do we anything on the radio/rds chip in diamond/raphael? |
| 22:52.36 | druidu | fm/rds that is |
| 22:55.29 | cr2_ | no |
| 22:56.01 | cr2_ | it sends some .bts commands to BT |
| 22:56.16 | dcordes_ | szsoftware, nice website, thanks. |
| 22:56.19 | cr2_ | which may enable i2c |
| 22:57.05 | LanceHaig | dcordes_: does the android we use have to be activated on the google site? |
| 22:57.06 | cr2_ | and connect the pcm out somewhere. the fm quality is good |
| 22:57.18 | cr2_ | LanceHaig: lol |
| 22:57.26 | LanceHaig | the reason I ask is that I was at a LInux Show on a stand |
| 22:57.58 | LanceHaig | and these guys from openssh said that after I showed them my HTC with android |
| 22:58.07 | cr2_ | LanceHaig: don't miss the g1 and andrid source code |
| 22:58.13 | LanceHaig | I told the I did not know if taht was the caes |
| 22:59.38 | dcordes_ | LanceHaig, did they know we hacked it manually? it sounds like they assumed it's a production software |
| 23:00.07 | LanceHaig | Well I did tell them you guys were real geniuses |
| 23:00.17 | LanceHaig | and that you had proted it to the HTC |
| 23:00.25 | LanceHaig | and that are close to bootstrapping it |
| 23:01.09 | dcordes_ | cool. what's linux show? |
| 23:01.46 | LanceHaig | It was LInux LIve here in London |
| 23:02.06 | cr2_ | ginge: what do you think about cleaning the clock api ? |
| 23:02.14 | LanceHaig | I am part of the bongo-project |
| 23:02.17 | LanceHaig | and we had a stand |
| 23:02.25 | dcordes_ | I wanna go to one of those linux conferences one day :) |
| 23:03.01 | BruteSource | i wanna goto an e3 |
| 23:03.44 | LanceHaig | Well once you guys release your version I think they will REALLY want you to attend |
| 23:03.44 | cr2_ | dcordes_: the javascript killed my kwin |
| 23:04.31 | dcordes_ | cr2_, which javascript? |
| 23:04.40 | BruteSource | version for which device lol |
| 23:05.26 | LanceHaig | Well Mine of course :-) |
| 23:05.48 | dcordes_ | LanceHaig, if I go there and show people linux on an msm7* device, they will assume I come from android marketing |
| 23:05.49 | BruteSource | which is? |
| 23:06.10 | LanceHaig | TyTAN_II |
| 23:06.13 | BruteSource | lol then bring an omap device |
| 23:06.56 | dcordes_ | szsoftware, around? I just booted your angstrom build rootfs. fbi doesn't seem to work for me. |
| 23:07.03 | LanceHaig | dcordes_: I suppose but I am sure you guys are clever enough to change the branding to make it different :-) |
| 23:07.11 | dcordes_ | szsoftware, it is stuck saying it loads image |
| 23:08.01 | dcordes_ | that's a nice idea though. maybe it would get even more people on board |
| 23:08.09 | dcordes_ | LanceHaig, did you show it to some skilled kernel hackers :) ? |
| 23:08.31 | LanceHaig | No sorry there were no kernel guys there at all |
| 23:08.42 | LanceHaig | which was sad really |
| 23:09.04 | LanceHaig | just a few debian / ubuntu/centos/openssh/openbsd guys |
| 23:09.14 | LanceHaig | and som "openSource Businesses" |
| 23:09.19 | dcordes_ | oh cool I'm watching southpark on my kaiser :) |
| 23:09.21 | LanceHaig | if you get my meaning |
| 23:09.26 | dcordes_ | on 2.6.25 :) |
| 23:09.35 | LanceHaig | fantastic |
| 23:09.59 | dcordes_ | sadly it's like 0,5 fps |
| 23:10.15 | dcordes_ | oh now it's faster |
| 23:10.30 | parmaster | dcordes_: what media viewer? |
| 23:10.49 | dcordes_ | parmaster, javaplayer 3.4 :D |
| 23:11.03 | dcordes_ | parmaster, no just kidding. mplayer. |
| 23:11.09 | parmaster | haha |
| 23:11.23 | dcordes_ | it runs perfectly smooth nice. |
| 23:11.31 | parmaster | how far along is the kaiser now? |
| 23:12.00 | dcordes_ | I think the lags result from sd read |
| 23:12.52 | dcordes_ | parmaster, we lack wifi, sound, usb, bluetooth, precis touchscreen, more stable sd driver. |
| 23:14.54 | cr2_ | lol |
| 23:15.08 | druidu | why can't I find 'Freeing init memory' in kernel source? |
| 23:15.17 | druidu | it's the exact message I see on the screen... |
| 23:15.28 | cr2_ | javaplayer... is it the standard android media player ? |
| 23:15.51 | dcordes_ | cr2_, don't know but that's what I meant ^^ |
| 23:15.56 | cr2_ | LOL |
| 23:16.32 | druidu | nevermind, it's Freeing[D |
| 23:16.40 | druidu | "Freeing %s memory" damn |
| 23:17.04 | NetRipper | you still not getting the Freeing.. message druidu ? |
| 23:17.17 | cr2_ | when i'll make fm work on raph, i'll contact gez. |
| 23:17.33 | dcordes_ | how can I split a file in half with the stuff available in busybox? |
| 23:17.42 | cr2_ | dcordes_: dd ? |
| 23:18.16 | dcordes_ | good |
| 23:19.12 | szsoftware | dcordes_: thank druidu, but you're possibly interested in this one too: http://www.fh-schmalkalden.de/schmalkaldenmedia/Downloads/elektrotechnik/l/edv/ck.pdf |
| 23:20.08 | dcordes_ | looks good |
| 23:20.23 | *** join/#htc-linux BabelO (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 23:21.14 | dcordes_ | szsoftware, fbi doesn't work for me |
| 23:22.28 | druidu | NetRipper : btw, why isn't your keyboard working? you have like, full keyboard :) |
| 23:22.35 | ginge | cr2_: not entirely sure how we should standardise these clocks with an api. What are your thoughts? |
| 23:23.26 | NetRipper | druidu, driver isn't working |
| 23:23.28 | druidu | why not mimic clock.c from msm android? |
| 23:23.43 | ginge | druidu: we do, although in a hacky way |
| 23:23.54 | szsoftware | dcordes_: you have all needed libs/packages installed? I' remember a strange bug, the screen had to be updated once through another fb event. For instance, I had to press camera button, which shows the onscreen kb, and magically, the picture was shown |
| 23:24.22 | dcordes_ | szsoftware, that's an interesting point, it might be the answer to the X problem. |
| 23:24.25 | NetRipper | druidu, when you use that latest raph_*.exe on netripper.nl/raphael you don't get freeing init either? |
| 23:24.40 | dcordes_ | szsoftware, ufnortunately I don't have a button that would redraw anythin |
| 23:25.28 | szsoftware | dcordes_: so you need the onscreen kb working. As I said, YOU WONT ONSCREEN KEYBOARD! |
| 23:25.30 | druidu | NetRipper: I get it with my latest compiled kernel (don't remember if yours worked, if you give me a link I'll try it) |
| 23:25.31 | szsoftware | dcordes_: :-) |
| 23:25.32 | cr2_ | NetRipper: the gpio mess needs to be cleaned up |
| 23:25.47 | druidu | problem is, it hangs after freeing init memory, nothing shows up |
| 23:25.54 | NetRipper | druidu, ah, that's just a bad initrd |
| 23:25.57 | dcordes_ | szsoftware, ok you won |
| 23:25.57 | druidu | maybe because I'm still running on my console driver... |
| 23:26.00 | druidu | I used yours :P |
| 23:26.07 | druidu | with init=/init passed via cmdline |
| 23:26.17 | druidu | fuck |
| 23:26.20 | NetRipper | you dont have to pass it |
| 23:26.23 | druidu | my console is __init |
| 23:26.25 | druidu | LOL |
| 23:26.41 | szsoftware | dcordes_: but I'm close to get it working, starting now with Chapter 5 of pointers :-) |
| 23:26.54 | druidu | actually I do, default is /sbin/init /etc/init /bin/init /bin/sh |
| 23:27.43 | druidu | actually... only htc_fb_console_init is marked with __init, so it should work |
| 23:27.45 | NetRipper | druidu, i dont need a init=/init to make my initrd work |
| 23:28.10 | dcordes_ | because /init is automatically exectued |
| 23:28.19 | NetRipper | exactly |
| 23:28.27 | szsoftware | dcordes_: btw, I compile mplayer within angstrom. Is that in my delivered rootfs? |
| 23:28.41 | *** join/#htc-linux BruteSource (n=BruteSou@m2a0e36d0.tmodns.net) |
| 23:28.51 | dcordes_ | szsoftware, yes I just watched a movie with it |
| 23:29.30 | szsoftware | dcordes_: Ah southpark. Cool |
| 23:30.54 | druidu | code says something else: http://android.git.kernel.org/?p=kernel/msm.git;a=blob;f=init/main.c;h=99ce94930b09302f7af89f75c07d7daa6a7ef668;hb=android-msm-htc-2.6.25#l789 |
| 23:30.55 | ginge | dcordes_: re: your boot log: "# this error is known from SD card dma issue:" its just a backtrace and unrelated to dma. the wifi doesn't use dma, just pio |
| 23:31.05 | druidu | anyway, I don't get panic either |
| 23:31.27 | cr2_ | ginge: enumerate by id ? |
| 23:31.42 | dcordes_ | ginge, yes stupid me. it just looked the same way ^^ I originally removed that |
| 23:32.16 | cr2_ | ginge: i think that the g1 kernel has all names, at least for A cpu. but we know where are "our" clocks located , and how to activate them |
| 23:32.16 | ginge | dcordes_: that pclk looks a little odd |
| 23:33.12 | cr2_ | ginge: i'll try to document the spl speed setting, and the math behind it. but not now ;) |
| 23:33.33 | ginge | cr2_: ok, would be handy. Thanks. |
| 23:33.57 | druidu | ok, it fails on run_init_process("/init"); |
| 23:34.08 | druidu | gonna try with no init= in cmd line, but I see no way it is going to work... |
| 23:34.11 | ginge | where in the wiki is the clock docs? |
| 23:34.31 | cr2_ | druidu: old abi enabled ? i doubt that rootfs.cgz is eabi |
| 23:34.39 | druidu | nope |
| 23:34.42 | druidu | not enabled :) |
| 23:34.45 | druidu | thanks |
| 23:34.46 | NetRipper | oh |
| 23:34.50 | NetRipper | im using eabi |
| 23:34.51 | druidu | (actually, NetRipper should know) |
| 23:34.52 | cr2_ | ginge: the 0xa8600000 block |
| 23:35.08 | cr2_ | ginge: look for raphael and kaiser memorymap |
| 23:35.11 | druidu | NetRipper : you sure it can boot with no init= ? |
| 23:35.13 | BruteSource | wow looks like i was drunk trolling last night |
| 23:35.20 | NetRipper | druidu, i am booting with no init= |
| 23:35.30 | ginge | oh in the memory map. i was looking for another page :-S |
| 23:35.55 | druidu | what root= and what console= ? |
| 23:36.01 | cr2_ | NetRipper: you have compiled the ramdisk yourself ? |
| 23:36.03 | NetRipper | only mem="64M" |
| 23:36.05 | NetRipper | nothing else |
| 23:36.44 | NetRipper | cr2_, i have compiled the busybox, but i dont know if that was with or without eabi, works for me though |
| 23:36.57 | druidu | you have old abi enabled? |
| 23:37.07 | NetRipper | no i'm using EABI, which is the default in the kernel |
| 23:37.17 | NetRipper | CONFIG_AEABI=y |
| 23:37.19 | dcordes_ | NetRipper, you can pick any angstrom image and use it as initrd. they all have busybox |
| 23:37.20 | druidu | yeah, but for busybox |
| 23:37.25 | NetRipper | # CONFIG_OABI_COMPAT is not set |
| 23:37.33 | druidu | it has another option, CONFIG_OABI_COMPAT |
| 23:37.35 | druidu | ok |
| 23:37.35 | druidu | Å= |
| 23:37.36 | druidu | :) |
| 23:38.41 | druidu | I get kernel panic with no init= |
| 23:38.48 | ginge | cr2_: I see it now, did you just document what we already know/have, or is there something I missed? |
| 23:38.52 | druidu | did you set a default cmd line in the kernel config? |
| 23:39.05 | NetRipper | CONFIG_CMDLINE="mem=76M console=ttyMSM2,115200n8" |
| 23:39.15 | cr2_ | ginge: it's everything i know. |
| 23:39.17 | NetRipper | but it gets overwritten by CMDLINE in haret |
| 23:39.34 | cr2_ | NetRipper: btw, where is BT ? |
| 23:39.40 | NetRipper | cr2_, bluetooth? gone |
| 23:39.43 | ginge | okay. at least the code we have in place now confirms that. |
| 23:39.51 | ginge | despite it being horrid |
| 23:39.55 | druidu | there is just no way it can work... please check init/main.c lines 800-804 |
| 23:40.27 | druidu | unless you have ramdisk_execute_command set to something... there is only execute_command which is the init= thing |
| 23:40.47 | cr2_ | NetRipper: uart1,2 or 3 on diam/raph ? |
| 23:40.56 | NetRipper | cr2_, disable |
| 23:40.57 | NetRipper | d |
| 23:41.04 | NetRipper | cr2_, but the interrupts initialize |
| 23:41.19 | cr2_ | NetRipper: physically |
| 23:41.21 | NetRipper | druidu, oh, sec |
| 23:41.27 | cr2_ | not in your kernel |
| 23:41.35 | druidu | ah, ops, looks like lines 843-849 set ramdisk_execute_command |
| 23:41.40 | cr2_ | your kernel needs many changes to enable it anyway :) |
| 23:42.18 | NetRipper | druidu ;) |
| 23:42.27 | druidu | but mine ends up NULL |
| 23:42.54 | cr2_ | NetRipper: looks like uart1, but i need to check |
| 23:43.04 | NetRipper | cr2_, what ar eyou talking about lol |
| 23:43.07 | druidu | I disabled all filesystems, but I guess cpio needs none right? |
| 23:43.42 | NetRipper | druidu, default should be alright, although i enabled ext2/ext3 |
| 23:43.53 | NetRipper | with extended attributes |
| 23:44.13 | NetRipper | i think you may need it, as when creating the cpio, it takes the attributes too |
| 23:47.02 | cr2_ | htc is amazing. block code duplication is everywhere ;) |
| 23:47.34 | *** join/#htc-linux fnord (n=fnord@24-151-90-116.static.nwtn.ct.charter.com) |
| 23:49.40 | druidu | what drivers are you talking about? |
| 23:54.07 | cr2_ | ginge: time to create the MSM_CLK wiki page |
| 23:55.43 | druidu | kernel panic - attempted to kill init |
| 23:56.00 | BruteSource | does hsdpa support simmotanious voice and data? |
| 23:59.07 | dcordes_ | BruteSource, I can make calls and have internet enabled simultaneously |
