| 00:00.11 | cr2 | gpiob 0x20 |
| 00:00.23 | cr2 | is tested after reset |
| 00:01.47 | lkcl | brb baby... |
| 00:01.55 | lkcl | ahh. |
| 00:02.59 | cr2 | no more refs. but this file is broken, don't forget. |
| 00:03.22 | cr2 | there is code to pull gpiob 0x20 on/off |
| 00:03.48 | cr2 | it's gpiob5 |
| 00:04.18 | cr2 | hmm. tested ? maybe it's irq ? |
| 00:04.54 | cr2 | pof: still here ? |
| 00:09.46 | *** join/#htc-linux willwork4foo (n=bjdemora@82-33-99-181.cable.ubr10.azte.blueyonder.co.uk) |
| 00:10.09 | willwork4foo | hey cr2: you know that list you gave of all the IPs that the xda-dev firewall blocks? |
| 00:11.07 | willwork4foo | I think someone has inadvertently blocked the Telewest gateway IP or something - because nobody AT ALL on Telewest can access xda-dev |
| 00:11.17 | willwork4foo | can get on FTP, but not on any websites |
| 00:12.40 | lkcl | babyfeeding back 30mins or so |
| 00:13.38 | willwork4foo | hi lkcl - are you on telewest in the UK too? (blueyonder) |
| 00:19.15 | *** join/#htc-linux vitb_home (n=vitb@cdma-42-209.msk.skylink.ru) |
| 00:40.09 | pof | willwork4foo: i posted the list... i post it on XDA-Dev mod's forum, u know which one is the Telewest gateway? |
| 00:40.42 | willwork4foo | pof: no idea |
| 00:40.49 | willwork4foo | wait a sec - lemme try a traceroute |
| 00:41.01 | *** join/#htc-linux mreimer (n=Matt@bilbo.vpop.net) |
| 00:41.16 | mreimer | any haret hackers here? |
| 00:41.24 | pof | ok |
| 00:41.58 | willwork4foo | pof: try 194.117.136.197 |
| 00:42.18 | willwork4foo | that's the last telewest IP I get before I start getting xs4all IP addresses |
| 00:43.27 | pof | it's not on the blocked list |
| 00:43.59 | pof | can you privmsg me the full traceroute... |
| 00:50.53 | psokolovsky | mreimer, Kevin2 is |
| 00:51.00 | psokolovsky | mreimer, what's up with haret? |
| 00:51.35 | markit | is qtek 9100 going to be supported? |
| 00:53.17 | lkcl | willwork4foo: telewest ... heard of that - can't remember where. telewest... |
| 00:54.32 | lkcl | telewest. nope - am with bt at the moment, soon-to-be-sky |
| 00:54.44 | lkcl | and if sky screw up the transfer i am xxxxed :) |
| 00:55.35 | willwork4foo | pof: one mo |
| 00:57.15 | mreimer | psokolovsky: I'm trying to get haret to work on a i.MX PDA, at least well enough to read some registers |
| 00:57.31 | mreimer | psokolovsky: but I have to run now. thanks for the answer |
| 00:57.33 | psokolovsky | mreimer, wow, cool! |
| 00:57.40 | mreimer | psokolovsky: olympus R1018 |
| 00:57.56 | psokolovsky | just tell what's that PDA, if you can! |
| 00:57.59 | psokolovsky | cool, thanks! |
| 00:58.03 | psokolovsky | see you later |
| 00:58.08 | mreimer | bye paul |
| 00:58.10 | *** part/#htc-linux mreimer (n=Matt@bilbo.vpop.net) |
| 01:09.39 | Kevin2 | Hi |
| 01:20.17 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 01:27.46 | lkcl | allo Kevin2 |
| 01:31.44 | Kevin2 | lkcl: Good luck. :-) |
| 01:31.47 | lkcl | hehe |
| 01:32.11 | lkcl | hmmm... says it's already running... |
| 01:32.56 | lkcl | oops i think i buggered it with an rmmod... |
| 01:34.03 | lkcl | hm i think i should try again, switching wireless off in wince, first |
| 01:34.26 | lkcl | oops it already was |
| 01:39.29 | lkcl | hmm.... not exactly successful, that.. |
| 01:40.34 | *** join/#htc-linux rejon (n=rejon@64-121-195-22.c3-0.sfpo-ubr4.sfrn-sfpo.ca.cable.rcn.com) |
| 01:50.36 | lkcl | ah HA! |
| 01:50.52 | lkcl | the chip was recognised - except i haven't the firmware (must try again) |
| 01:51.32 | lkcl | i powered up when i had the acx100 running from wince. |
| 01:52.28 | psokolovsky | lkcl, what is base address? |
| 01:52.48 | lkcl | same as universal - 0x08000000 |
| 01:53.01 | lkcl | lemme go find some firmware and try again. |
| 01:53.03 | psokolovsky | cool, thanks |
| 01:53.51 | lkcl | clearly i haven't quite got the gpio init sequence right. oops |
| 01:53.54 | psokolovsky | do it, do it! kick us old lazy h4000, which detects chip too, but whose mainatiners are too lazy to make it work in full ;-( |
| 01:55.13 | psokolovsky | s/us/ass/ ;-E |
| 01:56.26 | lkcl | oh dear! |
| 01:56.54 | lkcl | well i'm copying htcuniversal pretty much verbatim... |
| 01:57.16 | lkcl | btw, why is koen so impatient and so negative towards me? |
| 01:57.43 | psokolovsky | lkcl, what happened? ;-) |
| 01:57.45 | psokolovsky | ok |
| 02:00.34 | lkcl | mum fell asleep with baby, that's what |
| 02:00.49 | psokolovsky | that's good ;-) |
| 02:00.58 | psokolovsky | but what about Koen? |
| 02:01.47 | lkcl | take a look on openezx irc logs. |
| 02:02.19 | lkcl | yesterday's should be there by now if they roll over at GMT 0000 |
| 02:02.23 | psokolovsky | for what date? yesterday? |
| 02:02.24 | lkcl | it was really quite silly |
| 02:02.28 | lkcl | yeh, yesterday. |
| 02:02.34 | lkcl | ok baby needs feeding gotta go |
| 02:02.41 | psokolovsky | have good time! |
| 02:02.49 | lkcl | firmware load worked but eCPU failed |
| 02:02.54 | lkcl | gotta get later fw |
| 02:12.11 | Kevin2 | psokolovsky: When you get a chance, can you checkout latest haret and see if linux booting still runs okay for you? |
| 02:15.23 | psokolovsky | Kevin2, will do over weekend, ok |
| 02:26.05 | Kevin2 | Sounds good - thanks. |
| 02:41.02 | psokolovsky | <PROTECTED> |
| 02:41.11 | psokolovsky | Whoa-a-a! |
| 02:41.21 | psokolovsky | Am I dreaming here?! |
| 02:41.25 | psokolovsky | Kudos! |
| 02:45.24 | *** part/#htc-linux rejon (n=rejon@64-121-195-22.c3-0.sfpo-ubr4.sfrn-sfpo.ca.cable.rcn.com) |
| 02:47.26 | lkcl | well bugger me - the hw6915 firmware initialises the acx correctly. |
| 02:48.50 | lkcl | it bloody works. well. mostly. |
| 02:55.50 | psokolovsky | lkcl, connected to AP already? |
| 02:55.54 | lkcl | no not yet. |
| 02:56.01 | psokolovsky | ok |
| 02:56.01 | lkcl | i'm a bit far away from my AP |
| 02:56.20 | lkcl | but it didn't complain ok i got batteries gonna take a walk |
| 02:56.26 | lkcl | closer to my AP |
| 02:57.26 | lkcl | unfortunately i have wireless tools v 19 and v20 of course is the 2.6.18 kernel |
| 03:00.43 | lkcl | is link quality etc. all supposed to be listed? |
| 03:11.30 | lkcl | oops. setting registers fails... |
| 03:25.43 | lkcl | hmm, initialisation of acx works only if it was running when booting from wince... |
| 03:44.53 | lkcl | http://wiki.xda-developers.com/index.php?pagename=SableWireless |
| 03:44.57 | lkcl | clues anyone? |
| 03:49.43 | Kevin2 | lkcl: Do you have the poweron sequence for acx? |
| 03:49.48 | lkcl | i think so |
| 03:50.09 | Kevin2 | Okay - then I don't know why it would fail. |
| 03:50.28 | lkcl | well the old firmware (dlink) definitely doesn't work |
| 03:50.48 | lkcl | the device only is recognised when booting from wince if the acx is powered on |
| 03:51.01 | lkcl | so my guess is that i don't have the poweron sequence right |
| 03:51.25 | lkcl | and the firmware is being overwritten, live. |
| 03:51.26 | lkcl | oops :) |
| 03:51.48 | lkcl | but... no that don't make sense because i run the reset for 100ms and that's definitely correct |
| 04:01.53 | lkcl | nope - initialisation definitely doesn't work if the device is off in wince. |
| 04:02.03 | lkcl | more research needed, but sleep needed more :) |
| 04:02.06 | lkcl | night folks |
| 08:37.22 | *** join/#htc-linux psokolovsky_ (n=psokolov@239.usernat.ip.net.ua) |
| 08:39.32 | *** join/#htc-linux RoEn_PC (n=roen@p54A668B2.dip.t-dialin.net) |
| 09:03.23 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 09:08.48 | *** join/#htc-linux willwork4foo (n=bjdemora@82-33-99-181.cable.ubr10.azte.blueyonder.co.uk) |
| 09:20.26 | *** join/#htc-linux WizMaui__ (n=WizMaui@62.112.90.179) |
| 09:20.26 | *** join/#htc-linux willwork4foo (n=bjdemora@82-33-99-181.cable.ubr10.azte.blueyonder.co.uk) [NETSPLIT VICTIM] |
| 09:20.26 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 09:20.26 | *** join/#htc-linux RoEn_PC (n=roen@p54A668B2.dip.t-dialin.net) [NETSPLIT VICTIM] |
| 09:20.26 | *** join/#htc-linux psokolovsky_ (n=psokolov@239.usernat.ip.net.ua) |
| 09:20.26 | *** join/#htc-linux TripleDES (n=sergio@84.77.161.251) [NETSPLIT VICTIM] |
| 09:20.26 | *** join/#htc-linux Kevin2 (n=Kevin@207-237-52-30.c3-0.avec-ubr12.nyr-avec.ny.cable.rcn.com) [NETSPLIT VICTIM] |
| 09:20.26 | *** join/#htc-linux scatterp (n=deminish@212.32.67.196) |
| 09:20.26 | *** join/#htc-linux ivor_ (n=ivor@kde/developer/ivor) [NETSPLIT VICTIM] |
| 09:20.26 | *** join/#htc-linux vitb (n=vitb@85.21.88.2) [NETSPLIT VICTIM] |
| 09:20.26 | *** join/#htc-linux Speedy2 (n=Javier_6@cpe-70-95-210-2.san.res.rr.com) |
| 09:20.26 | *** join/#htc-linux pierrox (n=pierrot@82.240.21.171) |
| 09:20.27 | *** join/#htc-linux cr2 (n=konversa@crpl22.physik.uni-wuppertal.de) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux baud123 (n=baud@ver78-1-82-240-29-152.fbx.proxad.net) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux hlbot (n=adm@iclem.net) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux pof (n=pof@62.57.1.173) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux anYc (i=mario@hadince17.hadiko.uni-karlsruhe.de) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux toi (n=pleemans@d5152D3B4.access.telenet.be) |
| 09:20.27 | *** join/#htc-linux parmaster (i=par@dipole.idlepattern.com) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux Kmarc (i=kari@markos.biz) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux TeringTuby (n=maarten@189-66-dsl.ipact.nl) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux gw280 (i=authdeni@81.91.110.54) [NETSPLIT VICTIM] |
| 09:20.27 | *** join/#htc-linux ljp (n=lpotter@203.94.178.46) |
| 09:21.30 | willwork4foo | morning y'all |
| 09:33.19 | willwork4foo | anyone awake in here? |
| 09:36.53 | *** join/#htc-linux ScottWaye (n=scott@213-152-36-33.dsl.eclipse.net.uk) |
| 09:37.01 | *** join/#htc-linux rejon (n=rejon@64-121-195-22.c3-0.sfpo-ubr4.sfrn-sfpo.ca.cable.rcn.com) |
| 09:38.06 | ScottWaye | WimMaui__: Are you the one hosting the hg kernel for the wizard? |
| 09:38.33 | WizMaui__ | jap ;) |
| 09:39.22 | ScottWaye | What is the state of that? I would like to build a kernel, but I think I must be missing some things. |
| 09:40.09 | WizMaui__ | hmm...the currently available kernel should be able to build |
| 09:40.17 | WizMaui__ | do you have the necessary buildroot stuff |
| 09:40.50 | ScottWaye | yes I got that. So the instructions at http://omap850.om.funpic.de/downloads/wizard/readme are not necessary? |
| 09:41.07 | WizMaui__ | the instructions are correct |
| 09:41.13 | WizMaui__ | though you don't need to patch the kernel |
| 09:41.20 | WizMaui__ | it's already patched for htc-wiz |
| 09:42.04 | ScottWaye | so I can just go straight to make mrproper with the hg kernel? |
| 09:43.51 | WizMaui__ | you can in fact skip make mrproper |
| 09:44.37 | ScottWaye | Ok, thanks I need to reboot to try. |
| 09:44.52 | *** part/#htc-linux ScottWaye (n=scott@213-152-36-33.dsl.eclipse.net.uk) |
| 09:49.37 | goxboxlive | morning |
| 10:04.35 | goxboxlive | cr2 are you around? |
| 10:08.14 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 10:32.42 | *** join/#htc-linux rejon (n=rejon@64-121-195-22.c3-0.sfpo-ubr4.sfrn-sfpo.ca.cable.rcn.com) |
| 10:51.54 | *** join/#htc-linux Scott_Waye (n=scott@213-152-36-33.dsl.eclipse.net.uk) |
| 11:12.14 | *** join/#htc-linux FossiFoo (n=Fossi@e176122008.adsl.alicedsl.de) |
| 11:13.39 | Scott_Waye | WizMaui: That builds fine, but I guess you knew that ;-) |
| 11:15.32 | Scott_Waye | I take it when it writes all the dots at the top of the screen, thats it uncompressing the cc.ext2? |
| 11:17.41 | *** join/#htc-linux florian (n=fuchs@84.245.185.102) |
| 11:18.25 | florian | good morning |
| 11:47.01 | RoEn_PC | moin |
| 13:15.51 | *** join/#htc-linux dullard (n=jim@adsl-static-1-30.uklinux.net) |
| 13:41.08 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 13:52.06 | *** join/#htc-linux lilliput (n=Touriste@82-47-214-98.stb.ubr11.brad.blueyonder.co.uk) |
| 14:02.48 | *** join/#htc-linux booba (n=booba@81.56.252.135) |
| 14:11.13 | *** join/#htc-linux FossiFoo (n=Fossi@e176107103.adsl.alicedsl.de) |
| 14:13.08 | *** part/#htc-linux booba (n=booba@81.56.252.135) |
| 14:43.23 | *** join/#htc-linux rjtd (n=rjtd@195-23-8-113.net.novis.pt) |
| 14:43.30 | rjtd | greets |
| 14:43.40 | rjtd | anyone working on magician? |
| 14:46.51 | cr2 | rjtd: pH5 |
| 14:47.14 | rjtd | do you know what is the problem with openembeded and magician? |
| 14:47.34 | cr2 | lkcl: ask somebody with windows to dump the files. |
| 14:47.45 | cr2 | rjtd: no idea. |
| 14:48.40 | rjtd | ok |
| 14:48.41 | rjtd | thanks |
| 14:49.56 | rjtd | I was amazed with how well it is working |
| 14:50.00 | rjtd | at least it boots now :) |
| 14:59.53 | Fossi | how fast are the sd card readers. any place to find that data? |
| 15:00.04 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 15:46.45 | goxboxlive | hi |
| 15:46.51 | goxboxlive | cr2 are you around? |
| 16:03.21 | *** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 16:06.40 | *** join/#htc-linux lilliput (n=Touriste@82-47-214-98.stb.ubr11.brad.blueyonder.co.uk) |
| 16:23.31 | *** join/#htc-linux TripleDES (n=sergio@84.77.161.251) |
| 16:24.00 | TripleDES | hi |
| 16:27.31 | cr2 | hi |
| 16:27.42 | cr2 | goxboxlive: i have read your e-mail |
| 16:30.56 | goxboxlive | cr2: What do you think, will it work? What do we (or rather you) have to do with the kernel. Is it just to make it as the apache? |
| 16:32.41 | cr2 | this keymap is only for the busybox environment. |
| 16:33.01 | cr2 | is there a "normal" loadkeys package in oe ? |
| 16:33.49 | cr2 | i think there was a readme there how to convert the ascii map into binary. |
| 16:34.25 | goxboxlive | ok, so it's only for busybox. But that doesnt mather or what? We are using busybox |
| 16:34.37 | goxboxlive | I can check with OE |
| 16:39.05 | cr2 | do you have an initramfs with fdisk,mkfs* ? i think i2c tools, irdadump and hcidump can be useful there too. |
| 16:39.15 | cr2 | and ts_* tools |
| 16:41.08 | goxboxlive | no i dont have that. The kernel is getting to big to boot. I had one, but that was with restricted functions. |
| 16:42.09 | cr2 | too big ? |
| 16:42.36 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 16:42.50 | cr2 | we should talk to Kevin2 about it. |
| 16:49.43 | goxboxlive | yes, haret wouldnt boot it |
| 17:03.09 | cr2 | well, since it's so painful, we can link zimage _and_ initrd with haret. |
| 17:04.40 | cr2 | then we need an oe "package" that will create such image. |
| 17:14.46 | *** join/#htc-linux pleemans (n=peter@d51A5E421.access.telenet.be) |
| 17:14.57 | cr2 | psokolovsky_: i'm not that happy with the corgi-bl backlight driver. it is not doing what i need. it has funny name. it is gplv2-only which i see as a silly political statement. |
| 17:16.10 | lkcl | allo darlins |
| 17:16.12 | psokolovsky_ | cr2, what you mean it doens't do? issues I saw you reported seems like machine impl. issues. It works perfectly on h4000, hx4700, ... |
| 17:16.15 | psokolovsky_ | Hi! |
| 17:16.37 | lkcl | cr2: Speedy offered - but doesn't know what to do |
| 17:20.14 | cr2 | i have documented the complete backlight operation on blueangel. imho it will be the same on himalaya. it requires the full pwm0 and pwm1 config which is completely omitted in the ipaq-asic3.h header. hmm. i need to patch that too. |
| 17:20.23 | lkcl | cr2: great! |
| 17:20.37 | *** join/#htc-linux pierrox (n=pierrot@ns1.1000wallpapers.com) |
| 17:21.13 | cr2 | i have added some definitions before, from code of cp15, but they are incomplete, and therefore a bit misleading. |
| 17:21.52 | cr2 | so we will have asic3 pwm0 , pwm1 and spi fully documented now. |
| 17:22.17 | psokolovsky_ | cr2: so, what are technical, not political/cosmetical issues with corgi-bl? |
| 17:23.14 | cr2 | i need to configure asic3 pwm0 and pwm1 to make backlight operational on blueangel. how can i do it with corgi-bl ? |
| 17:23.55 | cr2 | it will be easy to reflash blueangel, and drop all htc stuff, because it uses pxa flash. |
| 17:24.22 | cr2 | i will research the blueangel init further. |
| 17:25.06 | cr2 | we have a code for boardid 4,5,6. 4 is actually the same as 5. there are only the differences between sony and toppoly lcds. |
| 17:25.12 | psokolovsky_ | cr2, what you mean "configure"? do that in machine initialization. or if you mean "reconfigure on each brightness change", do that in set_intensity handler |
| 17:26.08 | cr2 | i need to keep the parent/child asic3 structure, and to have a proper suspend/resume functions. |
| 17:26.50 | cr2 | lkcl: is somebody capable of decoding the sable nbf ? |
| 17:28.06 | *** join/#htc-linux pierrox (n=pierrot@82.240.21.171) |
| 17:29.57 | *** join/#htc-linux pierrot1 (n=pierrot@ns1.1000wallpapers.com) |
| 17:36.50 | cr2 | lkcl: why do you have only 2 wifi power gpios ? |
| 18:15.23 | psokolovsky_ | cr2: btw, if you guys ever proceed with idea of having static irq base for asic3, I will make next step of questioning if asic3 parent/child pattern gives us more benefits or troubles. |
| 18:16.02 | psokolovsky_ | cr2, I mean, be careful to unlish the power of re-designing basic subsystems we use. |
| 18:16.34 | psokolovsky_ | cr2, but regarding BL, still, what's the issue? suspend/resume of BL is handled *completely* and *correctly* by corgi-bl |
| 18:16.50 | psokolovsky_ | again, h4000 & hx4700 work ok |
| 18:17.51 | psokolovsky_ | I'm not in scope to investigate what's wrong with your case (universal?), but pls don't make quick decisions to move away |
| 18:18.40 | psokolovsky_ | my concern is really following - if we'll flip-flop on trivial thingg like backlight, who and when bother with those asic3 clocks stuff and other real issues? |
| 18:18.44 | psokolovsky_ | just my IMHO |
| 18:24.56 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 18:25.31 | *** join/#htc-linux lilliput (n=Touriste@82-47-214-98.stb.ubr11.brad.blueyonder.co.uk) |
| 18:28.50 | *** join/#htc-linux BabelOued_ (n=Fabrice@82.238.28.28) |
| 18:32.24 | *** join/#htc-linux Ro_En_PC (n=roen@p54A67497.dip.t-dialin.net) |
| 18:43.38 | *** join/#htc-linux rejon (n=rejon@64-121-195-22.c3-0.sfpo-ubr4.sfrn-sfpo.ca.cable.rcn.com) |
| 18:43.55 | *** join/#htc-linux rejon (n=rejon@64-121-195-22.c3-0.sfpo-ubr4.sfrn-sfpo.ca.cable.rcn.com) |
| 18:47.47 | TripleDES | cr2: have you seen my privmsg ? |
| 18:53.43 | pierrox | lkcl: about the g500 gps : it is made by royaltek, and it uses a firmware that is updatable by a UART |
| 18:57.03 | cr2 | psokolovsky: i'm not going to change right now anyway, but there are other complex issues linked to it. |
| 18:57.25 | cr2 | TripleDES: yes. thanks, i could not reply. but the files are still broken ;) |
| 18:57.39 | psokolovsky_ | ok, let's have discussed some time later |
| 18:57.43 | cr2 | pierrox: it's a normal sirf3 ? |
| 18:58.01 | TripleDES | broken? |
| 18:58.17 | cr2 | TripleDES: yes. ida is unhappy. |
| 18:58.53 | cr2 | and there is no .data seg ? is it a testwm5 dump ? |
| 18:59.32 | pierrox | cr2: i don't know, i have few info, what do you mean by normal ? |
| 19:00.19 | cr2 | pierrox: realtek is not a gps chipset producer. it used the chipsets by uNav and by sirf, afaik. |
| 19:01.07 | pierrox | cr2: ok, so it should be sirf then |
| 19:01.42 | cr2 | pierrox: then it's easy for you. you can get sirf binary and nmea data. |
| 19:02.16 | cr2 | btw, why don't you extend haret to dump the samsung CPU subsystem data ? |
| 19:03.03 | pierrox | cr2: you mean the values of the various controler registers ? |
| 19:04.41 | TripleDES | cr2: yeah, they're all from testwm5 coz we tried with alpine and typhoon scripts for decoding the nbf file and then img tools and nk.exe never came out |
| 19:04.43 | cr2 | pierrox: yes. it was extremely helpful for the pxa devices. afaik the rx3715 data is all dumped by had ;) |
| 19:05.23 | cr2 | TripleDES: but this produces broken files. |
| 19:05.34 | TripleDES | cr2: I think I could give you the two dlls from non-testwm5 dump if u want |
| 19:05.45 | cr2 | yes. |
| 19:06.04 | TripleDES | but I don't know why the nk.exe couldn't be extracted |
| 19:06.06 | cr2 | testwm5 does not work for recent devices at all. |
| 19:06.21 | TripleDES | ok, 1min |
| 19:06.25 | pof | cr2: but nk.exe is only dumped with TestWM5, not possible to extract it from OS.nb using viewimgfs.exe and rdmsfhsl.pl |
| 19:06.44 | cr2 | pof: then it's a bug in these scripts, too. |
| 19:06.48 | Kevin2 | Hi |
| 19:06.54 | cr2 | Kevin2: hi |
| 19:07.07 | Kevin2 | pierrox: Can you try latest haret on the samsung, and let me know if it works? |
| 19:07.27 | cr2 | pierrox: http://wiki.xda-developers.com/index.php?pagename=HermesMemoryMap |
| 19:07.37 | pof | cr2 is it because newer roms don't have a linear rom -> vmem mapping ?? |
| 19:07.45 | pierrox | Kevin2: ok, not now because i will go but tomorrow i will give a try |
| 19:07.48 | *** join/#htc-linux NickBee (i=HydraIRC@221.134.52.117) |
| 19:08.01 | Kevin2 | The binary is at: svn://www.shadowmite.com/home/svn/ppc6700/haret -- use haret.exe |
| 19:08.10 | Kevin2 | Cool, thanks. |
| 19:08.40 | Kevin2 | cr2: Can you try haret on your universal and strongarm? |
| 19:08.52 | cr2 | pierrox: if you'll dump the data for all these subsystems, you'll get a lot of useful data, that saves a lot of dev. time and guessing. |
| 19:09.07 | Kevin2 | BTW, I added the ability to do dynamic linking to functions in dlls in the mainline haret. |
| 19:09.20 | cr2 | Kevin2: universal and BA. sa only tomorrow. |
| 19:09.27 | cr2 | Kevin2: great. |
| 19:09.45 | pierrox | cr2: most of the time i use the memory dump to see the values of registers, but that's not specially easy of course |
| 19:10.04 | NickBee | Hi Kevin, Can i help with my universal? |
| 19:10.05 | cr2 | pierrox: look at 'dump pxa27x*' commands. |
| 19:10.24 | pierrox | cr2: ok, i will have a look |
| 19:10.48 | pierrox | by the way have some of you have a problem with haret and function pointers with cegcc ? |
| 19:10.54 | Kevin2 | NickBee - sure. Download the latest haret from svn above, and just let me know if it works okay for you. (commands work, linux boots, etc.) |
| 19:10.55 | cr2 | pierrox: a lot of things i have written are portable, so you can simply use them, too. like 'powermon' 'pwr*' and others. |
| 19:11.02 | TripleDES | cr2: you've got msg |
| 19:11.21 | NickBee | Ok kevin, give me 5 minutes |
| 19:11.33 | cr2 | TripleDES: thanks. |
| 19:11.38 | Kevin2 | pierrox: No, I haven't had any problems with cegcc. gnu-haret uses the videolan gcc toolset, which I have run in to several minor problems. |
| 19:11.50 | TripleDES | cr2: IIRC those were extracted with rdmsfhsl.pl |
| 19:12.19 | cr2 | ok. |
| 19:14.01 | pierrox | things are moving fast on haret, cvs update gaves me a lot of changes, cool :-) |
| 19:14.53 | *** join/#htc-linux florian_ (n=fuchs@87.193.30.11) |
| 19:14.56 | cr2 | TripleDES: you should ask itsme about nk.exe |
| 19:15.38 | NickBee | kevin, i have to execute haret.exe on universal and see if linux boots, right? |
| 19:16.19 | Kevin2 | NickBee. That would be great, yes. |
| 19:16.35 | Kevin2 | NickBee - are you familiar with the telnet interface? If so, you could try other commands as well. Linux booting is the most important part, however. |
| 19:17.29 | NickBee | i am using winxp, can i do telnet to universal once linux boots through windows? |
| 19:17.33 | pierrox | Kevin2: does not work for me, but i used a modified haret for the g500 |
| 19:17.35 | NickBee | if you can tell me, i can try |
| 19:18.34 | Kevin2 | NickBee: You can telnet to haret while the machine still runs wince. |
| 19:18.35 | cr2 | pof: we need to fix testwm5. it's a _very_ valuable thing. |
| 19:18.49 | Kevin2 | pierrox: In what way does haret not work? |
| 19:18.58 | NickBee | let me give it a try |
| 19:19.20 | pierrox | Kevin2: i saw two wide blue line in the middle of the screen, then the screen has been shut off |
| 19:19.38 | pierrox | Kevin2: i saw a progress bar before |
| 19:20.19 | Kevin2 | pierrox: light blue and dark blue lines? |
| 19:20.24 | Kevin2 | Do you use an initrd? |
| 19:20.43 | cr2 | Kevin2: did you set the initrd base to +8M ? |
| 19:21.03 | pierrox | Kevin2: for the blue line probably : it has been very fast so i am not sure |
| 19:21.06 | Kevin2 | cr2: No, it is at +5M as was done by psokolovsky_. |
| 19:21.19 | cr2 | they need it on samsung for some reason. |
| 19:21.58 | NickBee | kevin, yes telnet is connected to universal |
| 19:22.11 | NickBee | what commands do we need to execute? |
| 19:22.40 | Kevin2 | See that "help" works. Check that "wi" is in the list. Does the greeting show "Universal found"? |
| 19:22.50 | pierrox | Kevin2: i use a initrd and previously i updated the initrd base to +8M manually directly in the source, i didn't try with the default 4M value since a long time |
| 19:23.54 | NickBee | kevin: Yes, when i started haret on uni, it said: Found machine "HTC Universal" |
| 19:24.14 | NickBee | Kevin: Yes, Help works |
| 19:24.21 | cr2 | Kevin2: one dumb question from me. is it difficult to add 'WI' to the linux kernel ? with /proc/wicmd and /proc/wiout ? |
| 19:24.46 | Kevin2 | pierrox: Did you compile haret yourself, or use the one from svn? |
| 19:25.04 | pierrox | Kevin2: i used the one from svn |
| 19:25.28 | Kevin2 | cr2: I guess you could, but wouldn't it be easier to just put a bunch of printk's in the kernel? |
| 19:25.44 | NickBee | kevin: no, i can't see WI in the list (from help), do you mean WIRQ by any chance? |
| 19:26.10 | cr2 | Kevin2: yes, but there are binary-only modules ;) |
| 19:26.47 | Kevin2 | NickBee: Yes, WIRQ. cool. Try running "wi 1" - it should return a bunch of data for about one second. |
| 19:27.03 | NickBee | ok |
| 19:27.33 | NickBee | yes, a huge list (mostly irq 7(os timer) |
| 19:28.00 | Kevin2 | pierrox:I pushed a new binary to svn - haret.exe-8MB - if it works, I'll modify haret to be able to do this on the fly. |
| 19:28.07 | cr2 | Kevin2: it works. |
| 19:29.24 | cr2 | Kevin2: the window is completey modal, there is not way to close it. |
| 19:29.25 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 19:29.33 | NickBee | shall i paste the output of wi 1 somewhere for you to see? |
| 19:29.50 | Kevin2 | cr2: Cool. Which is working? |
| 19:30.00 | Kevin2 | NickBee: No, that's okay. See if you can boot linux. |
| 19:30.05 | cr2 | 0.4.3 |
| 19:30.16 | NickBee | ok, trying it now |
| 19:30.21 | Kevin2 | cr2: Oh, 0.4.3 is a little old. Try haret.exe. |
| 19:30.30 | cr2 | ok. |
| 19:30.32 | pierrox | Kevin2: does not work neither, but it is due to something in the g500 i am sure |
| 19:30.50 | pierrox | i must go now, bye |
| 19:31.57 | NickBee | kevin: yes, linux boots |
| 19:32.40 | Kevin2 | NickBee: Cool, thanks. Do you use an initrd? |
| 19:33.12 | NickBee | will have to check :) |
| 19:33.45 | pof | cr2: sorry I was away... u know exactly what's wrong with TestWM5 ? |
| 19:34.19 | cr2 | pof: no. |
| 19:34.31 | cr2 | Kevin2: it detects Blueangel. |
| 19:35.00 | NickBee | the kernel commandline: root=/dev/mmcbk0p2 console=tty0 mem=64 fbon=rotate. so i guess, no initrd |
| 19:35.07 | cr2 | Kevin2: and i can close it. on the universal there is no exit button ;) |
| 19:35.39 | Kevin2 | NickBee: Does your default.txt have something like: SET INITRD=XXX |
| 19:35.49 | Kevin2 | Or do you have a file "initrd" in the same directory as haret? |
| 19:36.00 | *** part/#htc-linux rejon (n=rejon@64-121-195-22.c3-0.sfpo-ubr4.sfrn-sfpo.ca.cable.rcn.com) |
| 19:36.07 | NickBee | i am in linux, let me check |
| 19:36.46 | Kevin2 | cr2: That is odd. I don't know much about the gui stuff. It has an exit button on the apache - maybe it is the vga screen on universal. |
| 19:36.53 | pof | cr2: what u do to see if the dumped file is correct or not? open in IDA pro? |
| 19:37.43 | cr2 | pof: it is all zeros at the last 25% |
| 19:38.09 | cr2 | ida complains about the import section broken. |
| 19:38.26 | cr2 | and there is no second text/code section at all. |
| 19:38.38 | pof | ok, i'll try to have a look next week... or bother mamaich to see if he can explain how to fix at least |
| 19:39.23 | cr2 | pof: that will be great. |
| 19:39.45 | cr2 | Kevin2: it happens only in the landscape mode. |
| 19:40.37 | NickBee | kevin: No, the folder that contains haret.exe does not contains initrd |
| 19:40.47 | Kevin2 | cr2: Does booting linux work on BA? I'm also curious if initrd works. |
| 19:40.56 | Kevin2 | NickBee: Okay thanks. |
| 19:41.31 | NickBee | no problems |
| 19:41.57 | cr2 | Kevin2: i don't have an "installable" distribution. should this binary also run on sa11x0 ? |
| 19:42.22 | Kevin2 | cr2: Yes, I think it should run on SA. |
| 19:42.33 | Kevin2 | (I don't know any reason it wouldn't, but of course, I haven't tested it.) |
| 19:42.42 | cr2 | ok. |
| 19:46.43 | cr2 | hmm. you linked in the toolhelp.dll ? |
| 19:47.04 | Kevin2 | Yeah, is that a problem? (The original haret does that too.) |
| 19:47.14 | cr2 | it is not present on jornada820 ROM, so i had to copy it from evc3. |
| 19:47.45 | cr2 | it is needed only for 'ps' and such things, not for booting. |
| 19:48.27 | cr2 | it's not a general problem, but can be annoying for the people who don't know how to solve it. |
| 19:48.36 | Kevin2 | Someone changed haret to have a "-k" command line option that kills off welcome.exe. That is why they linked in toolhelp. |
| 19:49.00 | cr2 | ah. ok. |
| 19:49.02 | Kevin2 | I can easily yank it. The commands you have in gnu-haret are much more powerful, and we can easily do dynamic linking to the dll. |
| 19:49.31 | cr2 | agreed. |
| 19:50.52 | NickBee | anybody, can you point me to a link where i can read and start using wlan on universal |
| 19:51.13 | NickBee | at present i get an error: "Starting interface failed" |
| 19:51.17 | Kevin2 | cr2: Does that mean haret.exe runs on the SA? |
| 19:51.35 | NickBee | when i try to start wlan from "network" |
| 19:55.33 | cr2 | Kevin2: will test tomorrow. |
| 19:58.46 | cr2 | lkcl: are you here ? |
| 20:06.21 | goxboxlive | psokolovsky_ Why are you preventing building libopie2 from cvs? It checkout the 1.2.2 tag and not the head. |
| 20:06.49 | psokolovsky_ | i don't do anything like that |
| 20:06.53 | psokolovsky_ | be specific |
| 20:07.15 | *** join/#htc-linux rtertetertet (n=345rt@84.232.212.62) |
| 20:07.17 | psokolovsky_ | goxboxlive, also, do any keys work for you in opie? |
| 20:07.21 | goxboxlive | I can see that you have updated the libopie2_cvs.bb |
| 20:07.35 | psokolovsky_ | and? |
| 20:07.46 | goxboxlive | Yes, the keys works, but i have to do a hack in the qte-mt.inc file |
| 20:07.48 | goxboxlive | and |
| 20:08.31 | goxboxlive | when i start to build it it choose the 1.2.2 tag instead of the main (latest files) |
| 20:08.48 | rtertetertet | Hello all ... |
| 20:08.48 | rtertetertet | <PROTECTED> |
| 20:08.49 | rtertetertet | <PROTECTED> |
| 20:08.49 | goxboxlive | rest of opie is building from main, but not liboie2 |
| 20:08.49 | rtertetertet | <PROTECTED> |
| 20:08.51 | rtertetertet | <PROTECTED> |
| 20:08.56 | goxboxlive | rest of opie is building from main, but not liboie2 |
| 20:09.35 | psokolovsky_ | goxboxlive, opie is broken in OE.dev. don't expect any specific behavior unless it's fixed |
| 20:09.41 | psokolovsky_ | also, what's "main"? |
| 20:09.52 | goxboxlive | Is it HEAD tag then |
| 20:09.59 | psokolovsky_ | ah, HEAD |
| 20:10.03 | goxboxlive | :-) |
| 20:10.39 | lkcl | no am not here cr2 |
| 20:10.47 | psokolovsky_ | so well, building from HEAD is not going to be supported at all. HEAD is always a moving target, and OE is about building stable distros. |
| 20:11.03 | goxboxlive | not OE.dev |
| 20:11.13 | lkcl | allo rtertetertet |
| 20:11.16 | cr2 | lkcl: i've read that bt does not work in the greenphone, and that it's the 2.4.19 kernel ;) |
| 20:11.26 | lkcl | reverse-engineering, rtertetertet |
| 20:11.27 | rtertetertet | allo |
| 20:11.32 | lkcl | yep, it's 2.4.19. hooray. |
| 20:11.44 | psokolovsky_ | goxboxlive, so, if you need to build from HEAD, do that. have config for that and expect to follow any changes that happen in OPIE CVS or OE.dev, that's all I can say ;-) |
| 20:11.59 | cr2 | lkcl: and there is no source code ? |
| 20:12.03 | lkcl | rter: it's mostly guess-work, and lots of analysis. |
| 20:12.23 | lkcl | oh there should be lots of source code on the cd, which i haven't even opened yet been too busy with the hw6915 |
| 20:12.34 | cr2 | :) |
| 20:12.40 | cr2 | open() mount() |
| 20:12.45 | rtertetertet | are there any tools to help that quessing? |
| 20:12.48 | lkcl | brb. laptop. drink. wife |
| 20:12.56 | lkcl | yes there are - gnuharet |
| 20:12.59 | lkcl | TestWM5.exe |
| 20:13.08 | cr2 | broken. |
| 20:13.10 | lkcl | start reading wiki.xda-developers.com |
| 20:13.32 | rtertetertet | strted - no king of info on the htc oxygen though |
| 20:13.54 | Fossi | i guess it's mostly dumping memory, registers and signals, and reading assembly code :) |
| 20:13.59 | cr2 | why are they all so half-assed about bluetooth ?! |
| 20:14.01 | goxboxlive | psokolovsky_In the OE, there are two bb files ofr each application. In example opie-console_1.2.2.bb and opie-console_cvs.bb. The 1.2.2 file use the cvs tag 1.2.2, the cvs.bb file use HEAD tag. But now that is broken. I saw that you had done changes to the libopie2 so i thought that there was a reason for it, but i guess theres not then. |
| 20:14.02 | Fossi | at least from what i've seen here |
| 20:14.46 | psokolovsky_ | goxboxlive, _cvs - not true. Read about SRCDATE. |
| 20:15.38 | rtertetertet | are the manufacturers of htc devices going to chnge therir perspectives any soon and start to release complete arc docs about their devices ? i mean , isn't in their interest that a second complete OS(linux) might load their device |
| 20:15.44 | goxboxlive | ok, i am usning a future SRCDATE to get the latest changes, because the universal is not fully supported in the 1.2.2. |
| 20:15.46 | rtertetertet | ? |
| 20:16.06 | Speedy2 | rtertetertet: How much did you pay for your Orange C100? |
| 20:16.09 | psokolovsky_ | goxboxlive, as a maintainer of OPIE in OE.dev, I have reasons to do that. And the aim is known - support OPIE in Angstrom. To achieve it, it will nee dto go thru some changes and even breakage. |
| 20:16.33 | rtertetertet | 214.8 USD |
| 20:16.40 | rtertetertet | from orange romania |
| 20:16.46 | psokolovsky_ | goxboxlive, you can use *any* SRCDATE which you know will work - it depends on *you*, not me. |
| 20:16.47 | rtertetertet | here the spv is very cheap |
| 20:17.22 | Speedy2 | Wow, you're in Romania? |
| 20:17.22 | Fossi | i paid 70 euro for my prophet |
| 20:17.44 | alex_begin | pretty far , aaa? |
| 20:18.06 | Speedy2 | I have a friend there |
| 20:18.13 | goxboxlive | psokolovsky_ ok, there was only ment as a question, not critic |
| 20:18.17 | Speedy2 | He's in the peace corps |
| 20:18.49 | alex_begin | hey , this is a nice country ; too bad the leaders outhere aree n00b |
| 20:19.18 | Speedy2 | alex_begin: Where is Romania are you ? He's is "Foscani" |
| 20:19.43 | *** join/#htc-linux JTRipper (n=jtripper@p54B1EBF0.dip.t-dialin.net) |
| 20:19.54 | lkcl | allo speedy |
| 20:19.55 | alex_begin | im in the capital -> Bucharest |
| 20:19.56 | Speedy2 | focscani |
| 20:20.00 | Speedy2 | Hey lkcl! |
| 20:20.38 | lkcl | you still up for that thingy-running windows running stint? |
| 20:20.43 | lkcl | cr2 knows what's needed |
| 20:22.10 | Speedy2 | lkcl : Yes. Tell me what is needed, I will help |
| 20:22.29 | Speedy2 | As I mentioned, no WINCE or PocketPC so it must be an x86 Win program |
| 20:22.43 | lkcl | yep, it is - i just don't know what, but cr2 does |
| 20:22.44 | psokolovsky_ | goxboxlive, np, just make sure you read OE manually recenetly and know how to pinpoint versions of packages you need. I can't disallow that, vice-versa, I'm making that to be more flexible/standard than it was before. |
| 20:23.01 | Speedy2 | cr2: Ping. If you need me to run a windows program, let me know |
| 20:23.40 | lkcl | in the meantime i'm going to get openembedded up-and-running on my other laptop, cos this one's out of space |
| 20:24.17 | Speedy2 | lkcl: I'm really excited about Linux on these HTC devices. I'm going to wait on getting a PDA until Linux on HTC (OE, etc) is all kosher |
| 20:24.29 | ljp | cr2: bt does work on the gp. just not the version needed for qtopia to work properly. we are working on this. |
| 20:24.50 | lkcl | hey that already all works, dude. if you want just a pda, then the 2.6.12 kernel does the job |
| 20:24.55 | lkcl | thanks ljp |
| 20:25.00 | lkcl | btw what chipset is the bt? |
| 20:25.38 | goxboxlive | psokolovsky_ Ok, but what i tried to tell you was, wich ever SRCDATE i write for libopie2 it doesnt mather, it chooses 1.2.2 anyway. If i wanted to build form 1.2.2 i woulcnt need to use any SRCDATE. But since i want to do that shouldnt i have the posibility to do it? |
| 20:26.32 | goxboxlive | But, it doesnt mather now, i went back a revision and it build from the HEAD tag now. |
| 20:26.51 | psokolovsky_ | goxboxlive, why don't you want to read manual? you've spent enough time with all this already, don't you want to master it? |
| 20:27.09 | goxboxlive | I actually have read it |
| 20:27.11 | cr2 | Speedy2: do you know about this http://wiki.xda-developers.com/index.php?pagename=RomTools |
| 20:27.27 | Speedy2 | lkcl: I meant PDA+phone capabilities. |
| 20:27.36 | psokolovsky_ | goxboxlive, what you nead is PREFERRED_VERSION_libopie2 = "1.2.2+cvs20061101" |
| 20:27.40 | lkcl | ahh |
| 20:27.42 | lkcl | yerrs |
| 20:27.49 | psokolovsky_ | SRCDATE_libopie2 = "20061101" |
| 20:27.52 | Speedy2 | cr2: I didn't know about that page. Is there an executable I need to run? |
| 20:27.56 | psokolovsky_ | choose date as you with, as I told |
| 20:28.33 | cr2 | Speedy2: it's not that easy. you need to install the Activstate perl |
| 20:29.07 | goxboxlive | I know this psokolovsky_ . But as i told you, it worked for a week ago, and stopped working today, so i thought that maybe there was something wrong with it. Just as a "mini |
| 20:29.19 | goxboxlive | " bug report u know. |
| 20:29.21 | Speedy2 | cr2: I'm wondering why this must be done on a Windows machine. |
| 20:30.03 | lkcl | jesus this is crazy. the monotone #oe archive is now a SEVENTY megabyte .sqlite database. |
| 20:30.14 | lkcl | it was stupid enough O(N*N) with only 30mb |
| 20:31.28 | lkcl | wow my line rate seems to have been whacked up to 8mb without me asking for it. cool! |
| 20:32.43 | lkcl | wow. 60mb downloaded so far already. |
| 20:33.40 | lkcl | cr2: which dll is that acx stuff in? |
| 20:34.07 | cr2 | Speedy2: that's the problem. there is a windows dll for decompressing. there is a linux libmspack in cabextract, but nobody has integrated it so far. |
| 20:34.18 | cr2 | lkcl: at least pcmcia.dll |
| 20:35.09 | lkcl | ok. |
| 20:35.42 | cr2 | tiacxwln.dll, but i don't think you'll get any useful from it. it's very obfuscated. |
| 20:35.53 | lkcl | o look there's a whole stack o shit on my server :) |
| 20:36.11 | cr2 | ? in ~cr2 ? |
| 20:36.15 | lkcl | yeh :) |
| 20:36.19 | cr2 | lol |
| 20:36.35 | cr2 | only 152MB |
| 20:37.06 | cr2 | ida is 52MB included |
| 20:39.18 | Speedy2 | cr2: Ok, but what does compiling those perl files have to do with the windows dll for decompressing? |
| 20:40.18 | cr2 | perl script needs this windows lib. |
| 20:41.27 | Speedy2 | Ok. What all do I need to do? I've used ActiveState's TclTk thing |
| 20:42.07 | alex_begin | does anybody know what chip HTC Tornado uses for the GSM? |
| 20:43.34 | lkcl | hang on alex i'll find a page for you to look at. in the meantime do a search on wiki.xda-developers for pages with the name 'Tornado'... |
| 20:44.04 | alex_begin | thnks a lot |
| 20:44.40 | lkcl | http://wiki.xda-developers.com/index.php?pagename=HTC_Tornado |
| 20:56.28 | lkcl | ok has anyone got an example config i can start from? or some suggestions? |
| 20:56.47 | lkcl | have we actually got e.g. himalaya or any other confs in openembedded.org? |
| 20:57.21 | lkcl | yep we have bluangel.. |
| 20:59.34 | cr2 | lkcl: universal. |
| 20:59.58 | lkcl | yep i saw that |
| 20:59.59 | cr2 | blueangel is for the .12 kernel |
| 21:00.06 | cr2 | it is all very confusing. |
| 21:00.57 | cr2 | i wish we had 1 image for all our phones. |
| 21:03.03 | lkcl | ack. |
| 21:03.19 | lkcl | btw what's with the /etc/bluetooth/TIInit_3.2.26.bts script? |
| 21:03.26 | cr2 | lkcl: i have looked the the himalaya binaries. |
| 21:03.49 | cr2 | lkcl: it's actually a questionable thing, if it is really needed. |
| 21:03.55 | lkcl | ah yes i saw you mentioned that, cr2 - something to do with ASIC3? |
| 21:04.03 | cr2 | no, BRF6150 |
| 21:04.08 | lkcl | oh right. |
| 21:04.10 | cr2 | himalaya ? |
| 21:04.49 | lkcl | brf6150 - TI single-chip bluetooth |
| 21:05.06 | cr2 | lkcl: we need a way to find out which ROM and FW has the BRF6150, to avoid loading/sending wrong commands. |
| 21:05.19 | lkcl | oops |
| 21:05.36 | cr2 | the "switch to 921600" is very simple, and can be done by 'chat' ;) |
| 21:06.20 | cr2 | without any binary blob like TIInit_3.2.26.bts which i suspect is intended to another ROM/FW pair . |
| 21:06.41 | lkcl | oh right. oh - it's a binary blob?? _greaaat_ |
| 21:06.57 | cr2 | this whole TI BTS stuff can be done by chat. |
| 21:06.57 | lkcl | oh well. too much info |
| 21:07.07 | cr2 | lkcl: i know how to decode it |
| 21:07.12 | lkcl | cool |
| 21:07.25 | cr2 | rripped out the code from hciattach :) |
| 21:07.34 | lkcl | cr2!! :) |
| 21:07.57 | lkcl | can you give me a hand getting acx initialised, cr2? |
| 21:08.18 | lkcl | i should really just get you one of these hw6915s, really, shouldn't i |
| 21:08.21 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 21:08.34 | lkcl | allo wiz |
| 21:08.50 | cr2 | i can send you the ba in exchange :) |
| 21:08.59 | lkcl | it's ok i got one already, in holland |
| 21:09.05 | lkcl | and i'm going back in two weeks |
| 21:09.17 | cr2 | i've told you 2 gpios i saw in pcmcia.dll, did you consider them ? |
| 21:09.21 | cr2 | ok. |
| 21:09.44 | lkcl | yes i got those - and a couple more. i need to look at it, myself, to double-check |
| 21:10.08 | cr2 | we can ask pof to try to extract it in another way. |
| 21:10.12 | lkcl | cos it's not working. well - if the wireless was already running (from wince) then i get to the point of being able to initialise the device |
| 21:10.30 | cr2 | then the reset does not work. |
| 21:10.35 | cr2 | which is strange. |
| 21:10.38 | Speedy2 | cr2: I am happy to help with the WIndows side of things. What all do I need to do? Which .pl files do I need to compile? Where do I get the Windows DLL? |
| 21:10.49 | lkcl | yep |
| 21:10.58 | cr2 | Speedy2: i'll look. |
| 21:11.03 | lkcl | which is why i wanted to have a look at it myself. |
| 21:12.27 | cr2 | Speedy2: http://wiki.xda-developers.com/index.php?pagename=RomTools%2Frdmsflsh.pl |
| 21:13.03 | cr2 | for LZX decompression, ?CompressUtils needs access to a live wm2005 device via activesync. |
| 21:13.08 | cr2 | this is wrong. |
| 21:13.34 | cr2 | but pof knows how to to it properly. |
| 21:22.34 | *** join/#htc-linux RoEn_PC (n=roen@p54A67497.dip.t-dialin.net) |
| 21:28.22 | lkcl | cr2, can you save an .idb for me on the pcmcia.dll, or point me in the right direction for setting up the pxa regs etc? |
| 21:30.10 | cr2 | hmm. not only pxa, asic3. i run 5.0 demo locally, because it's much faster. wait. |
| 21:30.51 | lkcl | k |
| 21:35.35 | *** join/#htc-linux willwork4foo (n=bjdemora@82-33-99-181.cable.ubr10.azte.blueyonder.co.uk) |
| 21:36.03 | cr2 | sent you a cut from the window. |
| 21:36.22 | cr2 | PXA GPIO is 0xAEE00000 |
| 21:36.40 | willwork4foo | hi cr2 - do you know if there is another IRC channel with people from xda-developers that's focused on windows mobile instead of linux? |
| 21:36.51 | cr2 | no. |
| 21:37.18 | willwork4foo | hmm. |
| 21:38.20 | willwork4foo | I've been trying without much success to get hold of the patch for the HTC Magician's bluetooth stack to get a bit more functionality out of it - because frankly, the microsoft bluetooth stack is s**t. |
| 21:38.26 | Speedy2 | cr2: /msg me when you're ready for me to handle the Windows side of things. |
| 21:38.41 | lkcl | ta dude |
| 21:46.02 | cr2 | lkcl: they use a more optimizing compiler on wince5, so some things are more hidden by offsets instead of direct references. |
| 21:46.08 | lkcl | joy. |
| 21:46.19 | lkcl | FRAK! |
| 21:50.31 | *** join/#htc-linux lamikr|lap (n=chatzill@dyn244-218.adsl.spy.dnainternet.fi) |
| 21:51.07 | lkcl | right. that does it. i'm doing a wi on asic3 gpio. |
| 21:52.25 | cr2 | i see one more place where it says 'bt coexistance' |
| 21:52.50 | cr2 | there are 3 such 'asic3gpio-like accesses' |
| 21:53.06 | cr2 | don't know ABCD, though. |
| 21:53.33 | lkcl | ok. that's ok i managed to get those from the asic3gpio trace. |
| 21:53.54 | lkcl | are there any sleep/delays in between the first and second gpio-access? |
| 21:54.03 | cr2 | where ? |
| 21:54.25 | lkcl | in pcmcia.dll |
| 21:54.45 | cr2 | for reset it is 1,sleep 0x64, 0, sleep 0x32, tst other gpio. |
| 21:54.46 | lkcl | yes because remember bt and wifi are both on 2.4ghz - they get in each other's way esp cos there are two separate chips (!) |
| 21:54.59 | lkcl | 1 then zero?? |
| 21:55.04 | lkcl | argh :) |
| 21:55.10 | lkcl | that'll be it, then |
| 21:55.20 | lkcl | it's similar to the hx4700 |
| 21:55.21 | lkcl | argh |
| 21:55.55 | lkcl | ok that test is an irq. |
| 21:56.07 | lkcl | the test on the other gpio is a loop, right? |
| 21:56.11 | lkcl | or not? |
| 21:56.23 | cr2 | wait a bit. |
| 21:56.26 | lkcl | k |
| 21:56.44 | lkcl | well at least it will help if i get the reset signals the right way round wouldn't it *sigh* |
| 21:57.07 | cr2 | 0x8000 at gplr3 ? |
| 21:57.39 | cr2 | http://wiki.xda-developers.com/index.php?pagename=SableGPIO |
| 21:57.53 | cr2 | not very readable, copy the bitmasks from universalgpio |
| 21:57.59 | lkcl | 1sec less have a look |
| 21:58.27 | cr2 | hmm. gpio111 |
| 21:58.52 | cr2 | input. unlikely. |
| 21:58.55 | lkcl | huh. why didn't gpio111 appear in the wasic3gpio?? |
| 21:58.58 | lkcl | oh well |
| 21:59.24 | cr2 | it's the #124 offset. usually it's clear gplr3 |
| 21:59.54 | lkcl | ah. |
| 22:00.06 | cr2 | hmm. since it's a memory dump you can peek at these addresses directly :) |
| 22:00.15 | cr2 | to know the base. |
| 22:00.24 | cr2 | for the offset at runtime. |
| 22:00.42 | lkcl | cr2? |
| 22:00.56 | cr2 | &7fff |
| 22:01.03 | lkcl | rats. wlan%d: ACX_VENDOR_ID: 0 after 501 retries, is this fatal? |
| 22:01.03 | lkcl | acx: eCPU is already running. reset_dev() FAILED |
| 22:01.09 | lkcl | cr2? |
| 22:01.48 | cr2 | &fff7 |
| 22:02.13 | lkcl | &7fff -> ~(1<<15) |
| 22:02.22 | cr2 | &feff |
| 22:02.44 | cr2 | at three different locations. let's look at sableasic3 |
| 22:02.48 | lkcl | &fff7 -> ~(1<<3) |
| 22:03.04 | cr2 | 8000, 80, 100 |
| 22:03.27 | lkcl | &feff -> ~(1<<8) |
| 22:03.32 | lkcl | yep - those are the three. |
| 22:03.57 | cr2 | 8000 is acx_power_1 |
| 22:04.36 | lkcl | B15 - yep. |
| 22:04.37 | cr2 | 100 is acx_power_3, but isn't it the reset line ? |
| 22:04.53 | lkcl | errr.... |
| 22:05.01 | lkcl | 1mo... |
| 22:05.14 | lkcl | yep wrong name lemme correct it |
| 22:05.36 | cr2 | 80 gpiod7 ? |
| 22:06.09 | cr2 | these 3 are "disable BT coexistance" |
| 22:06.38 | lkcl | 80? you mean 8 not 80 which is 3 so it's D3 |
| 22:06.45 | lkcl | yes? |
| 22:07.17 | lkcl | &fff7 -> ~(1<<3) |
| 22:07.23 | cr2 | they are referenced 38, 40, 3c |
| 22:07.24 | lkcl | not D7 |
| 22:07.41 | cr2 | ah. |
| 22:08.03 | lkcl | so, let me check, is the sequence you're seeing: |
| 22:08.21 | lkcl | B15 1 |
| 22:08.23 | lkcl | D3 1 |
| 22:08.24 | cr2 | gpiod3 - acx_power_2 |
| 22:08.42 | lkcl | C8 0 |
| 22:08.45 | lkcl | mdelay(100) |
| 22:08.47 | lkcl | C8 1 |
| 22:08.51 | lkcl | mdelay(50) |
| 22:08.57 | lkcl | oops. |
| 22:09.01 | lkcl | cancel that |
| 22:09.02 | lkcl | B15 1 |
| 22:09.04 | lkcl | D3 1 |
| 22:09.07 | lkcl | C8 1 |
| 22:09.10 | lkcl | mdelay(100 |
| 22:09.14 | lkcl | C8 0 |
| 22:09.16 | lkcl | mdelay(50) |
| 22:09.17 | lkcl | C8 1 |
| 22:09.18 | lkcl | ? |
| 22:09.32 | cr2 | yes, 38,40,3c -> B D C |
| 22:09.38 | lkcl | ok. |
| 22:09.40 | lkcl | let's try that. |
| 22:13.53 | lkcl | arse. nope. |
| 22:14.15 | lkcl | why does it work if the wifi is running whilst wince is up?? |
| 22:15.39 | cr2 | you are missing one more power pin. |
| 22:15.52 | lkcl | oops :) |
| 22:16.04 | cr2 | pcmcia may reference not all of them. |
| 22:16.28 | cr2 | the driver name is HPWIFIDVR1 |
| 22:16.35 | lkcl | i'm tempted to open up the device and trace back the power pins to see if one goes to the pxa. |
| 22:16.58 | cr2 | the best way to check it is to modify haret, and send the powerdown/powerup event only to that driver. |
| 22:17.17 | cr2 | do you have haret source at hand ? |
| 22:17.25 | lkcl | yep. |
| 22:17.29 | lkcl | ah ha! yes of course |
| 22:17.49 | lkcl | yes i do |
| 22:17.55 | lkcl | well. gnu-haret, anyway. |
| 22:18.08 | cr2 | LPWSTR dev=L"{98C5250D-C29A-4985-AE5F-AFE5367E5006}\\TIACXWLN1 |
| 22:18.13 | cr2 | in power.cpp |
| 22:18.26 | cr2 | look how it is called. |
| 22:18.39 | cr2 | and replace the TIACXWLN1 |
| 22:18.51 | cr2 | by HPWIFIDVR1 |
| 22:19.52 | lkcl | ack |
| 22:20.07 | cr2 | right now it's BKL: |
| 22:20.10 | lkcl | yep. |
| 22:20.25 | lkcl | brb gotta go say hello to baby+wife |
| 22:20.37 | cr2 | then you'll know what wince does. |
| 22:20.38 | cr2 | ok. |
| 22:30.53 | *** join/#htc-linux RoEn_PC (n=roen@p54A67497.dip.t-dialin.net) |
| 22:35.30 | lkcl | nuts - of course: i need to run from scripts because this is wifi :) |
| 22:36.42 | lkcl | ok how am i gonna do this.... |
| 22:37.30 | lkcl | i need a script to turn power on and then off |
| 22:40.56 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 22:41.50 | lkcl | allo goboxlive |
| 22:43.06 | goxboxlive | hello |
| 22:51.06 | lkcl | 111! |
| 22:51.08 | lkcl | gpio 111 |
| 22:57.22 | lkcl | bollocks - 111 is an input. |
| 22:59.12 | lkcl | yep - 111 goes 0 when off and 1 when on |
| 23:04.10 | *** join/#htc-linux RoEn_PC (n=roen@p54A67497.dip.t-dialin.net) |
| 23:10.27 | lkcl | ok let's delay things find out what the status of 111 is |
| 23:13.51 | goxboxlive | what is the gpio 111 doing lkcl? |
| 23:14.19 | lkcl | changing? :) |
| 23:14.53 | lkcl | argh i must have left wifi on when booting into linux |
| 23:17.27 | cr2 | lkcl: so #124 was a real gpio offset ? |
| 23:17.47 | lkcl | ah i was confused about the #124 dude |
| 23:18.25 | cr2 | the code just picked some address from the kernel space, and added #124 |
| 23:18.35 | cr2 | usually it means gpcr3. |
| 23:18.38 | lkcl | uhh?? |
| 23:18.41 | lkcl | ermmm... |
| 23:18.58 | cr2 | and it's the same code block where they mess with asic3 gpios. |
| 23:19.00 | lkcl | oh that's better it tailed |
| 23:19.22 | cr2 | it's all done by macros, so you can see the similar binary code structure. |
| 23:20.06 | goxboxlive | cr2: Have you had any progress with the universal today cr2? |
| 23:20.34 | cr2 | goxboxlive: i still don't have a good idea how to understand the audio routing. |
| 23:20.47 | goxboxlive | Wjhat about the power? |
| 23:21.23 | cr2 | i have identified all gpio changes, but they are linked to some ioctl("WAV1:) , where the numbers don't tell me much. |
| 23:21.52 | cr2 | i see on suspicious difference to hx4700. |
| 23:22.07 | cr2 | we don't have "CODEC RESET" gpio. |
| 23:22.52 | cr2 | maybe something is wrong there, the codec is always powered. otherwise why it repsnds to i2cdetect ? |
| 23:22.59 | goxboxlive | ok |
| 23:23.08 | goxboxlive | cr2: have you seen this? http://www.xs4all.nl/~itsme/projects/xda/ril.html |
| 23:23.08 | cr2 | you can't see the cams i2c addresses. they are powered off. |
| 23:23.13 | goxboxlive | ok |
| 23:23.39 | cr2 | hx4700 people got a gpio list from hp/cpq, it was much easier for them. |
| 23:23.51 | cr2 | yes. |
| 23:23.59 | *** join/#htc-linux RoEn_PC (n=roen@p54A67497.dip.t-dialin.net) |
| 23:24.16 | cr2 | but look at this RIL_SetAudioDevices |
| 23:24.23 | cr2 | DeviceIoControl("WAV1:", 0x001d000c, {0, 0x65, 0, ?1, 0}, 0x14}, 0x14, sp+0x18, 0x14, sp+0x10, 0) |
| 23:24.32 | cr2 | does it tell you something ? |
| 23:24.40 | goxboxlive | :-) |
| 23:25.07 | cr2 | me too. |
| 23:25.30 | cr2 | maybe we should document the terminology in wiki. |
| 23:26.10 | cr2 | i'm trying to look at other phones, the design logic and the terms used on all htc phones must be the same. |
| 23:26.27 | goxboxlive | coiuld sp be speaker and 18 an 10 be adresses |
| 23:26.35 | cr2 | so we can narrow the search. central limit theorem :) |
| 23:27.04 | cr2 | sp+ are some parameters put on the stack. |
| 23:27.23 | cr2 | for example. |
| 23:27.32 | cr2 | what is "handsfree" ? |
| 23:27.48 | cr2 | where does the sound come and where it goes. |
| 23:28.17 | cr2 | then we can watch gpios, disassemble something relevant, trace memory accesses, log AT commands. |
| 23:28.44 | cr2 | there is AT@AUDIOSET command, |
| 23:29.04 | cr2 | i think it has 4 variable parameters: 0,2,3,7 |
| 23:29.46 | cr2 | i think once i knew what they mean. |
| 23:29.46 | lkcl | cr2: they did?? |
| 23:29.46 | lkcl | cr2 duh i should ask for the gpio list for the hw 6915 |
| 23:29.58 | cr2 | ok. |
| 23:30.54 | goxboxlive | cr2: does this help? It os for the h2200: |
| 23:30.56 | cr2 | goxboxlive: there is a testril.cpp program in itsutils, we can try to trace it too, but first we need to port the code to haret. |
| 23:30.56 | goxboxlive | [HKEY_LOCAL_MACHINE\Drivers\Active\06] |
| 23:30.56 | goxboxlive | "Hnd"=dword:00030790 |
| 23:30.56 | goxboxlive | "Name"="WAV1:" |
| 23:30.56 | goxboxlive | "Key"="Drivers\\BuiltIn\\WaveDev" |
| 23:30.57 | goxboxlive | "InterfaceType"=dword:ffffffff |
| 23:31.03 | lkcl | the off routine is never called when htcsable_acx is rmmod'd |
| 23:31.44 | cr2 | lkcl: .probe .init .suspend .resume .remove ? |
| 23:32.37 | lkcl | acx_hardware_data_acx_data - stop_hw function isn't called. |
| 23:33.33 | cr2 | module_exit( htcsable_wlan_exit ); |
| 23:33.46 | cr2 | and it does not call any gpios ? |
| 23:36.05 | cr2 | there is something buggy in this stuff anyway. if i call rmmod htcuniversal_acx, then the kernel will crash. |
| 23:36.44 | cr2 | goxboxlive: can you formulate what is handsfree ? |
| 23:38.11 | goxboxlive | what do you mean? Handsfree as in BT handsfree, or speaker on loud? |
| 23:39.05 | cr2 | ok, let's forget about BT for now. |
| 23:39.38 | cr2 | so handsfree means send the phone sound to the stereo speakers ? |
| 23:40.01 | goxboxlive | yes, loud so you dont need to have it in to your ear |
| 23:40.30 | cr2 | mic is still the same. |
| 23:40.42 | goxboxlive | Nad there is the handsfree headset that comes with the universal wich you plug in to the 3,5 jack |
| 23:40.42 | *** join/#htc-linux TripleDES (n=sergio@84.77.161.251) |
| 23:40.56 | TripleDES | hi |
| 23:41.02 | cr2 | that's a different story. |
| 23:41.07 | goxboxlive | I dont think the mic is the same, i guess it activates a mic boost as in alsa |
| 23:41.25 | cr2 | TripleDES: hi. can you do us a favour ? |
| 23:41.29 | pof | hi TripleDES :) |
| 23:41.43 | goxboxlive | But it is the same mic AFAIK. |
| 23:41.54 | TripleDES | yes sure :) |
| 23:42.01 | TripleDES | what is it? |
| 23:42.10 | cr2 | goxboxlive: wait. the headphone is a different story. |
| 23:42.50 | cr2 | TripleDES: lkcl has a htc sable, and his pcmcia.dll is completely messed up by testwm5. is it possible to extract it by other means ? |
| 23:44.21 | goxboxlive | cr2: I know, but as i said, i guess when i use the universal as hanhdsfree, the sound comes out of the stereo speakers but i guess the mic is the same, only diffrents is that it is more senistive i guess. |
| 23:44.21 | cr2 | goxboxlive: i have inserted the 3.5 jack. the mic and the speakers (?) is switched over. what is handsfree now ? |
| 23:44.47 | pof | cr2 / lkcl : do u have any shipped rom upgrade for the Sable? |
| 23:44.55 | cr2 | which speakers, btw. big, or those on LCD ? |
| 23:44.59 | cr2 | pof: yes. |
| 23:44.59 | goxboxlive | cr2: the headset |
| 23:45.05 | goxboxlive | big |
| 23:45.08 | goxboxlive | stereo |
| 23:45.20 | TripleDES | cr2: what about itsutils? |
| 23:45.29 | cr2 | goxboxlive: it's called 'hp' as in headphone, in the driver. |
| 23:45.39 | goxboxlive | ok |
| 23:45.43 | cr2 | TripleDES: i don't have windows. |
| 23:45.54 | cr2 | and did not port all the code over to haret. |
| 23:46.43 | TripleDES | I only was able to extract files from windows... |
| 23:46.48 | pof | cr2: is the shipped rom in NBH or NBF format? |
| 23:47.08 | goxboxlive | cr2: When you plug in a 3.5 jack, i guess the units speakers and mic is muted. |
| 23:47.12 | cr2 | goxboxlive: we need to create such table, where is the mic/speaker and where the sound goes under which conditions. |
| 23:47.29 | cr2 | then i can try to correlate it with the traced hardware events. |
| 23:47.45 | cr2 | pof: nk.nbf |
| 23:48.03 | cr2 | goxboxlive: by a mixer ? |
| 23:48.29 | cr2 | that will mean and i2c command, which can be traced by 'wi' on the i2c port. |
| 23:48.41 | TripleDES | pof maybe typhoondecode.pl could do it? |
| 23:49.06 | cr2 | TripleDES: the alpinenbfdecode did not do it. |
| 23:49.12 | pof | use typhoonnbfdecode.pl with '-r' param to convert it to RAW format, then he can extract the files either with viewimgfs.exe or rdmsflsh.pl, but both need windows |
| 23:49.19 | goxboxlive | cr2: i dont know if it is a mixer or anything else, but i know that here is noe sound in the speakers when the 3.5 jack is activated. |
| 23:51.15 | cr2 | goxboxlive: we have gpios ( all tracked), AT commands (tracked) and the i2c commands (not tracked). maybe we should try to track and document them. the ak4641 datasheet describes all of them, so it's easy to follow. |
| 23:51.41 | goxboxlive | yes if that will help i think we should |
| 23:52.13 | pof | cr2: can you edit HTC_GSM wiki page at xda-dev? it's giving me an error after I press the "preview" button |
| 23:53.28 | cr2 | pof: i'll try |
| 23:54.29 | cr2 | pof: i can't preview, but i can edit and save. i've already seen it before. |
| 23:54.55 | pof | thanks, i'll edit it and put hermes stuff then :) |
| 23:55.07 | TripleDES | pof: do you if dumping with 'd2s' command will give me all wince files or just a single file? |
| 23:55.15 | TripleDES | ops...know |
| 23:55.17 | TripleDES | :) |
| 23:55.42 | pof | d2s dumps to SD in raw format (not fat32) |
| 23:56.12 | pof | but you can read it with 'dd if=/dev/sda of=/dev/sd-dump.bin' after |
| 23:56.47 | TripleDES | and then extract files? |
| 23:56.52 | pof | probably this is the best way, yes... |
| 23:57.03 | TripleDES | ok I'll try it tonight |
| 23:57.04 | cr2 | TripleDES: there is also a header+crc |
| 23:57.24 | TripleDES | hmm |
| 23:59.16 | cr2 | http://h18007.www1.hp.com/support/files/HandheldiPAQ/us/download/23839.html |
| 23:59.39 | cr2 | hmm. not _that_ one ;) |