| 02:45.25 | *** join/#htc-linux keizer1 (n=keizer@c-69-139-42-47.hsd1.fl.comcast.net) |
| 04:29.31 | *** join/#htc-linux sladen (i=paul@starsky.19inch.net) |
| 04:33.55 | *** join/#htc-linux sladen (i=paul@starsky.19inch.net) [NETSPLIT VICTIM] |
| 04:43.00 | *** join/#htc-linux sladen (i=paul@starsky.19inch.net) [NETSPLIT VICTIM] |
| 04:43.07 | *** join/#htc-linux sladen_ (i=paul@starsky.19inch.net) |
| 04:50.34 | *** join/#htc-linux sladen_ (i=paul@starsky.19inch.net) [NETSPLIT VICTIM] |
| 05:06.59 | *** join/#htc-linux the_sys0p (i=nroberts@117.121.80.203) |
| 05:09.48 | *** join/#htc-linux sladen_ (i=paul@starsky.19inch.net) [NETSPLIT VICTIM] |
| 05:15.50 | *** join/#htc-linux AC_Mast27 (i=s@gateway/tor/x-1ed1d67074abfe41) |
| 05:43.25 | *** join/#htc-linux sladen (i=paul@starsky.19inch.net) |
| 05:47.34 | *** join/#htc-linux goxboxlive (n=goxboxli@139.80-202-160.nextgentel.com) |
| 05:51.42 | goxboxlive | morning |
| 06:09.55 | *** join/#htc-linux the_sys0p (i=nroberts@117.75.146.114) |
| 06:11.22 | *** join/#htc-linux sladen (i=paul@starsky.19inch.net) |
| 06:23.23 | *** join/#htc-linux |dion| (n=dion@inhex.net) |
| 08:33.09 | *** join/#htc-linux rob_w (n=bob@X0984.x.pppool.de) |
| 09:16.45 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-103-152-149.karneval.cz) |
| 10:06.42 | *** join/#htc-linux pH5 (n=ph5@p5485C695.dip.t-dialin.net) |
| 10:15.11 | *** join/#htc-linux |dion| (n=dion@inhex.net) |
| 10:34.15 | *** join/#htc-linux NAiL (n=repvik@nslu2-linux/pdpc.active.NAiL) |
| 10:36.04 | NAiL | Is there no support for HTC P3300 (Artemis)? |
| 10:38.16 | pH5 | hey NAiL. not yet, BabelO_ is working on it. |
| 10:39.52 | NAiL | hiya pH5 :) |
| 10:40.10 | NAiL | Ok. I was looking at the wiki and couldn't see the artemis anywhere |
| 10:41.16 | NAiL | I just can't get stuff working with WM6, so I thought I'd try linux instead. |
| 10:44.19 | pH5 | heh. I wish. I have a P4350 (Herald) here, but the i2c driver doesn't work yet and with it the external i2c gpio expander that is used to power quite some of the hardware components. |
| 10:45.34 | pH5 | the omap850 framebuffer and udc works, so you can boot linux and log in via usbnet, but that's about it afaik. |
| 10:47.01 | NAiL | ah, ok |
| 10:51.05 | NAiL | Bit too early in other words |
| 10:51.33 | pH5 | NAiL: well, you're always to come and join in with the kernel hacking :) |
| 10:52.17 | NAiL | hehe, I've struggled enough with kernel hacking on all the NAS-devices I've added to OE ;) |
| 10:52.31 | pH5 | I bet :) |
| 10:53.16 | NAiL | generic question... OMAP850 has an SD hostcontroller. Is it possible to use SDHC? I've heard it's the software that differs. |
| 10:55.32 | pH5 | NAiL: yes, it even works in wince. |
| 10:56.05 | NAiL | well, not on my device it doesn't :-P |
| 10:56.42 | pH5 | really? on the herald with wm5 it works. |
| 10:57.04 | NAiL | My Artemis with wm6 it doesn't see the card at all. |
| 10:57.16 | pH5 | I also used my sdhc card on old magician and hx4700 devices that don't support it under windows. |
| 10:57.36 | NAiL | well, good to know that it's the software that needs to be fixed |
| 10:57.49 | NAiL | I've got a 4gb card lying around wanting to be used ;) |
| 11:04.35 | pH5 | the hardware is clearly capable of sdhc (the only thing it needs to support compared to sd are big block transfers) |
| 11:05.03 | NAiL | dunno, but if the herald has wm6 too, it should have drivers that support sdhc that I can use |
| 11:05.41 | pH5 | I have wm5, but there should be a wm6 update floating around. |
| 11:06.10 | pH5 | but I'm more looking forward to the linux update ;) |
| 11:06.15 | NAiL | hehe |
| 11:07.05 | *** join/#htc-linux Mineque (n=Mineque@195.60.65.50) |
| 11:07.10 | NAiL | I've never tried linux on a phone yet |
| 11:07.35 | NAiL | Is spamfiltering sms'es possible? Blacklisting phonenumbers? |
| 11:07.53 | NAiL | Those are my two "most wanted" features ;) |
| 11:09.49 | pH5 | NAiL: I don't know of any implementation yet, but that can be done easily in userspace. |
| 11:13.17 | NAiL | are there hooks I could use with python/shellscript/perl/<insert favourite scriptlanguage>? |
| 11:14.52 | NAiL | There's a website that keeps a list of numbers that I want to block. It'd be nice to pull it as a textfile once in a while and just blacklist the whole thing :) |
| 11:16.25 | *** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 11:17.05 | pH5 | NAiL: I think that's business of the phone/ringer application/service, you should bring up that idea with the openmoko/qtopia people or whatever software stack you intend to use. |
| 11:17.07 | pH5 | hey BabelO |
| 11:17.52 | NAiL | pH5: yeah |
| 11:19.01 | pH5 | if the phone supports the cpu <-> modem connection like neo1973 does you could even have it take the call, play a 'sorry you are blocked' message and hang up automatically. |
| 11:19.59 | NAiL | that would obviously rock the most |
| 11:20.43 | *** join/#htc-linux eway (n=ellis@149.254.192.192) |
| 11:20.47 | NAiL | even more fun would be to send telemarketers into an endless maze |
| 11:21.12 | pH5 | NAiL: but for that you'd want to forward them to some landline, because otherwise they can DoS you phone |
| 11:21.26 | NAiL | oh? |
| 11:21.27 | BabelO | hi, NAiL pH5 |
| 11:21.40 | NAiL | DoS how? Stay on the line? |
| 11:21.57 | pH5 | NAiL: yup, as long as you phone is in a call with a telemarketer, you won't be able to use it. |
| 11:22.29 | NAiL | heh, if there's another incoming call (call waiting obviously), the telemarketer is just ditched anyway ;) |
| 11:23.32 | pH5 | right, then the only argument I've got left is draining the battery. |
| 11:23.43 | NAiL | got a point there |
| 11:24.24 | NAiL | but I'd say it's worth it :-P |
| 11:25.02 | BabelO | NAiL: which wm rom do you haave ? |
| 11:25.25 | NAiL | I'm using HTC's latest WWE rom from their FTP |
| 11:25.46 | BabelO | like me |
| 11:26.36 | NAiL | why'd you ask? |
| 11:27.05 | BabelO | for sdhc, i read something at xanadux |
| 11:27.26 | NAiL | ah, you read backlog |
| 11:27.40 | BabelO | yes |
| 11:28.59 | NAiL | The two things I hate about this rom is that I can't get norwegian soft-keyboard on it, and that it doesn't support sdhc |
| 11:30.04 | BabelO | me i have FRE from htc website i just look at xanadux for something about sdhc driver |
| 11:32.48 | pH5 | BabelO: any news on the artemis linux front? |
| 11:37.11 | BabelO | pH5: no, but with new register list, maybe we can do something clean ****, like header files for this cpu ? |
| 11:39.06 | *** join/#htc-linux ellisway (n=ellis@homealone.good1.com) |
| 11:40.17 | BabelO | NAiL: http://www.pocketnavigation.de/article/view_609_1586__HTC-P3300/2.4.67.html |
| 11:42.03 | NAiL | My german isn't too good. I've read that page, but I haven't heard of anyone that actually has sdhc working on artemis |
| 11:42.27 | NAiL | htc support said it supported it, then they retracted the statement |
| 11:42.32 | NAiL | bbiaw |
| 11:42.54 | BabelO | NAiL: :) badddddddlink |
| 11:43.10 | BabelO | NAiL: http://forum.xda-developers.com/showthread.php?t=310456&page=4&highlight=sdhc |
| 11:43.35 | *** join/#htc-linux eway (n=ellis@149.254.192.195) |
| 11:47.28 | *** join/#htc-linux ellisway (n=ellis@homealone.good1.com) |
| 12:14.27 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-103-152-149.karneval.cz) |
| 12:51.02 | cr2 | hi |
| 12:51.15 | cr2 | NAiL: can you extract nk.exe for artemis ? |
| 12:51.46 | cr2 | BabelO: what should i do with libts ? |
| 12:52.33 | BabelO | hi cr2 |
| 12:52.35 | pH5 | hi cr2 |
| 12:52.48 | BabelO | now Qtopia need new libts-1.0 |
| 12:53.03 | cr2 | pH5: i'm booting BA from the raw SD now. |
| 12:53.09 | cr2 | BabelO: .ipk ? |
| 12:53.10 | BabelO | cr2: i make antother image in the afternoon |
| 12:53.15 | cr2 | BabelO: ok. |
| 12:53.27 | pH5 | cr2: very nice, with eol? |
| 12:53.27 | BabelO | i need to go now, bbl |
| 12:53.38 | cr2 | pH5: with linboot.exe |
| 12:54.08 | cr2 | root=/dev/mmcblk0 |
| 12:54.27 | pH5 | ah, ok |
| 12:55.38 | cr2 | pH5: so it should go the same way with mtd. |
| 13:00.51 | cr2 | pH5: linux people on xda-dev are cheap :) |
| 13:01.40 | cr2 | there are a lof of questions about running linux on omap850 phones, and nothing for athena. |
| 13:05.27 | pH5 | btw, it looks like on LAK there is some movement to get gpiolib up and running again, so as soon as this is in mainline we can start to push all our gpio extenders. |
| 13:13.26 | cr2 | pH5: what do you think about updating to .23 ? |
| 13:13.41 | cr2 | on hh.org |
| 13:15.13 | pH5 | cr2: I still haven't gotten pxamci to work. As long as I don't have either booting from mtd or mmc on .23, I won't invest any time into the hh.org update as it is my fallback. |
| 13:17.01 | cr2 | pH5: goxboxlive used the zaurus .23 kernel on universal. |
| 13:18.19 | cr2 | pH5: how can we manage a jump from 0x40000 on an mtd partition without loading some kind of SPL ? |
| 13:18.39 | NAiL | cr2: sure, why? |
| 13:18.54 | cr2 | i#d be happy to do some research and flash the BA too. |
| 13:18.59 | NAiL | BabelO: I tried those drivers. Didn't work for me :( |
| 13:19.11 | cr2 | NAiL: how do i extract xip from .nb ? |
| 13:19.30 | *** join/#htc-linux keizer1 (n=keizer@c-69-139-42-47.hsd1.fl.comcast.net) |
| 13:20.55 | NAiL | cr2: Well, I'm using the hypercore kitchen, which does all that stuff for me |
| 13:21.15 | cr2 | NAiL: ok, then extract it for me :) |
| 13:24.17 | NAiL | can I mail it to you? |
| 13:25.02 | cr2 | can you upload it somewhere ? |
| 13:25.52 | NAiL | yeah |
| 13:27.29 | NAiL | http://mirror.kynisk.com/nk.exe |
| 13:27.38 | *** join/#htc-linux Babel (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 13:30.54 | cr2 | NAiL: thanks. |
| 13:30.57 | NAiL | np |
| 13:50.56 | pH5 | cr2: at least on my devices, the htc bootloader jumps to 0x40000 without enabling ram, and linux needs ram enabled (not sure about xip though). |
| 13:51.20 | NAiL | BabelO: The sdhc cab makes my wireless disappear :( |
| 13:54.09 | cr2 | pH5: what about stack ? |
| 13:54.34 | cr2 | pH5: well, if you will keep everything in the registers, it will work. |
| 13:54.36 | pH5 | cr2: there is no stack in the cold-boot codepath |
| 13:54.43 | cr2 | ok. |
| 13:55.08 | cr2 | btw, there will be a similar problem on the universal with 128MB ram |
| 13:55.28 | cr2 | the most smart IPL is on athena. |
| 13:59.11 | BabelO | NAiL: yes wifi disappear as guy say, but should work... some of them report a working sdhc |
| 14:04.33 | BabelO | cr2: where pixmaps for roadmap should go ? |
| 14:07.25 | *** join/#htc-linux ivucica (n=ivucica@217.14.216.214) |
| 14:09.26 | NAiL | BabelO: "Not installed" :( |
| 14:11.21 | ivucica | hi! |
| 14:11.27 | cr2 | BabelO: pixmaps are portable (.png). so /usr/share/roadmap |
| 14:12.11 | ivucica | can someone provide some help with how to compile kernel for h3800 on i386? step by step? |
| 14:12.36 | ivucica | imagine that i have a pure debian machine with only tools needed for compiling kernel for i386 :) |
| 14:13.59 | ivucica | where do i take the kernel from? kernel.org, handhelds.org, linuxtogo.<something> or somewhere else? |
| 14:14.08 | ivucica | cr2: do you know perhaps? |
| 14:16.04 | cr2 | ivucica: http://www.handhelds.org/moin/moin.cgi/KernelDocumentation |
| 14:16.10 | ivucica | checking |
| 14:19.59 | ivucica | cr2: i'm following the steps. can you tell me which libc6 should i use in compiling kernel? |
| 14:20.07 | ivucica | there are several versions in debian arm |
| 14:20.30 | ivucica | i'm getting files "libc6-dev_2.6.1-6_arm.deb" ... :) |
| 14:20.46 | ivucica | and "libc6_2.6.1-6_arm.deb" ofc |
| 14:33.33 | *** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be) |
| 14:48.40 | Kevin2 | Hi |
| 14:49.00 | BabelO | hi Kevin2 |
| 14:51.39 | Kevin2 | cr2: Are you going to port the asic3 sd driver to hermes? |
| 14:52.08 | Kevin2 | I looked, and I can't tell which one is better to use - the asic3 or tmio driver.. |
| 14:52.25 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 14:57.30 | ivucica | oi Kevin2 |
| 14:58.23 | ivucica | Kevin2: when building the h3800 kernel, can i use these versions of libc6 from debian for arm? |
| 14:58.24 | ivucica | libc6_2.6.1-6_arm.deb libc6-dev_2.6.1-6_arm.deb |
| 14:59.29 | Kevin2 | ivucica: I've never used debian, but you shouldn't need libc at all to build a kernel. |
| 14:59.54 | ivucica | kk -- i'm following some tutorial which mentions it |
| 15:00.19 | ivucica | Kevin2: this -- http://www.handhelds.org/moin/moin.cgi/DevelopmentToolchains |
| 15:00.28 | ivucica | the first link tells i need to install libc |
| 15:00.36 | ivucica | oops, sorry, third |
| 15:01.57 | Kevin2 | You'll need the libc packages to cross compile any userspace programs. But not for the kernel. |
| 15:02.48 | ivucica | Kevin2: ok |
| 15:04.43 | ivucica | Kevin2: i asked cr same thing, should i use the handhelds.org kernel tree? or is the active development done elsewhere? |
| 15:04.44 | cr2 | Kevin2: asic3 is better, we only need to strip the SD_HWCTRL stuff. |
| 15:04.56 | ivucica | i saw that GPE forked/moved/something to linuxtogo.org |
| 15:05.16 | cr2 | ivucica: just try it, you ask about a very old device. |
| 15:05.29 | ivucica | cr2: i will ;) |
| 15:06.24 | ivucica | well, i certainly hope that i can return something to the community, you guys have been tolerating me for over a week now ;) |
| 15:06.25 | cr2 | Kevin2: btw, i'm successfully using the linload.exe, because it does not force me to include initrd. |
| 15:06.48 | Kevin2 | cr2: You don't need initrd with bootbundle - just specify an initrd of /dev/null. |
| 15:06.49 | cr2 | which i don't actually need. |
| 15:07.06 | cr2 | Kevin2: hm. ok. |
| 15:07.20 | Kevin2 | (If you look through the makefile, linload just turns an empty initrd into /dev/null for you.) |
| 15:07.31 | cr2 | Kevin2: and i also can't trace ATI on blueangel like BabelO |
| 15:07.39 | ivucica | Kevin2: i guess that developers of bootbundle should make that the default behaviour, hmm? |
| 15:07.39 | ivucica | :) |
| 15:07.53 | cr2 | Kevin2: it checks if i have 4 arguments on the command line, so i could not drop it. |
| 15:08.23 | cr2 | <PROTECTED> |
| 15:08.24 | Kevin2 | Yeah, Paul wrote that script and it looks like it has some bugs. |
| 15:09.18 | ivucica | oh ye, i noticed that development around h3800 has stopped, even images are not really produced -- do you guys know who in Angstrom/Familiar/elsewhere should i contact regarding testing the images? |
| 15:09.22 | cr2 | anyway, i can boot off the raw SD disk now. |
| 15:09.29 | ivucica | i could make an image and test it every week or so |
| 15:09.42 | cr2 | ivucica: you should do it yourself. imho. |
| 15:10.02 | ivucica | yep, i will... it's just that i'd like to help other people like me |
| 15:10.18 | Kevin2 | ivucica: I too am confused on where to go for distros.. |
| 15:10.19 | ivucica | i'd like to give something back -- like testing the images, and uploading them somewhere |
| 15:10.34 | ivucica | perhaps someone should start another one? ;) |
| 15:10.47 | Kevin2 | cr2: booting from SD on which machine? |
| 15:10.56 | cr2 | Kevin2: on BA |
| 15:11.11 | Kevin2 | BA uses ati sd? |
| 15:11.35 | Kevin2 | cr2: What commands did you use to trace ati on BA? |
| 15:11.46 | ivucica | i'm really thinking about, once i figure this out, starting a project to make installing linux to pocketpc a "wizard-like" process |
| 15:11.49 | cr2 | Kevin2: it means that if somebody wants to check linux, he does not need to do any partitioning. only rawwrite.exe |
| 15:12.11 | cr2 | Kevin2: BA has ati sd, but it is not used. it uses asic3. |
| 15:12.20 | ivucica | it'd probably use a portion of debian/arm (just like familiar)... |
| 15:12.37 | cr2 | hmm. the commands you told me to use on hermes. |
| 15:12.59 | Kevin2 | The virtual addresses may be different on BA. Do you have a "dump mmu" handy? |
| 15:13.03 | BabelO | cr2: this one is tested and it boot on mine, just wait 5 min after you boot first time ;) http://linuxtogo.org/~htcpxa/htcblueangel/O.P.I.E-II/ |
| 15:14.38 | cr2 | Kevin2: i'll create one now, but i doubt something new will come out: http://wiki.xda-developers.com/index.php?pagename=BlueangelVirtualMemoryMapping |
| 15:15.03 | ivucica | cr2, Kevin2: are the extra PPC modules from the "handhelds" kernels built into mainstream kernel? if not, why? |
| 15:17.57 | BabelO | cr2: i also add flite .. but you have to install it if you want |
| 15:24.12 | *** part/#htc-linux rmoravcik (n=rmoravci@ip-89-103-152-149.karneval.cz) |
| 15:25.15 | cr2 | BabelO: ok, i'll check haret now, and then will try to boot your image. |
| 15:36.08 | cr2 | Kevin2: ab000000 | 08000000 | 1MB section | D=0 AP=1M |
| 15:37.09 | Kevin2 | cr2: There are usually several virtual mappings - you'll need to watch them all to be sure. |
| 15:37.31 | cr2 | addlist mmutrace 0xab000000 1024*1024 |
| 15:40.16 | *** join/#htc-linux rob_w (n=bob@X160a.x.pppool.de) |
| 15:40.44 | cr2 | set permissivemmutrace 1 |
| 15:41.56 | Kevin2 | You don't need permissivemmutrace if you're tracing L1 section mappings. |
| 15:42.21 | cr2 | there is one more mapping at 0x4e000000 |
| 15:43.50 | cr2 | HaRET(8)# show mmutrace |
| 15:43.50 | cr2 | 000: 0xab000000 1048576 rw 00000000 none |
| 15:43.50 | cr2 | 001: 0x4e000000 1048576 rw 00000000 none |
| 15:47.47 | cr2 | Kevin2: ? |
| 15:47.49 | cr2 | Warning! Tracing non-section mapping (4e000000) not well supported |
| 15:47.50 | cr2 | If you really want to do this, run 'set permissivemmutrace 1' and retry |
| 15:48.01 | Kevin2 | Can you post the "dump mmu" output? |
| 15:49.52 | cr2 | 4e000000 | a7a0b000 | Coarse | D=0M |
| 15:49.52 | cr2 | <PROTECTED> |
| 15:50.18 | cr2 | ok, that's not that i want |
| 15:51.02 | cr2 | they are all mapped as single pages. weird. |
| 15:51.47 | Kevin2 | Heres is the same way. So, if the whole 1MB is mapped as 4K pages, then you can use 'set permissivemmutrace 1' |
| 15:51.54 | Kevin2 | s/Heres/Hermes/ |
| 15:52.46 | cr2 | ok. |
| 15:52.53 | cr2 | but it does not show anything. |
| 15:53.02 | cr2 | i'll try it on athena. |
| 15:53.07 | cr2 | ba has pxa263 |
| 15:53.15 | Kevin2 | cr2: Can you pastebin the entire "dump mmu"? |
| 15:53.44 | cr2 | yes, but is it not too big ? |
| 15:53.59 | Kevin2 | pastebin should handle it. |
| 15:54.38 | cr2 | ok, i'll try. |
| 15:55.09 | cr2 | and may not exceed 150,000 bytes. |
| 15:55.26 | cr2 | wc -c typescript |
| 15:55.32 | cr2 | 288189 typescript |
| 15:56.03 | cr2 | ok, 2 halves. |
| 15:56.05 | Kevin2 | Hrmm. Okay - I've thought I've seen dump posted before. |
| 15:57.52 | *** join/#htc-linux comicinker (n=simeon@R0b32.r.pppool.de) |
| 15:59.45 | cr2 | http://pastebin.ca/751610 |
| 15:59.50 | cr2 | the first part |
| 16:00.28 | cr2 | http://pastebin.ca/751613 |
| 16:00.31 | cr2 | the second part |
| 16:00.58 | comicinker | hi. how was it, the bootloader is not available for htc-magician? it is implossible to soft-reset such a device? |
| 16:04.35 | Kevin2 | cr2: What is it you are trying to trace? |
| 16:05.22 | *** join/#htc-linux goxboxlive (n=goxboxli@139.80-202-160.nextgentel.com) |
| 16:05.37 | cr2 | Kevin2: the accesses to the ATI registers. |
| 16:07.05 | Kevin2 | Not ATI SD (0x0e000000) or ati gpio (0x0c000000)? |
| 16:07.39 | cr2 | Kevin2: these are ASIC3 SD and GPIO. |
| 16:08.14 | cr2 | hima/ba have both ATI and ASIC3. |
| 16:11.23 | goxboxlive | hi |
| 16:11.38 | Kevin2 | Okay - try "addlist mmutrace 0x4e000000 1024*1024" and repeat for 0x4e100000, 4e200000, 4e300000, 8b000000, 8b100000, 8b200000, 8b300000, 8b400000, 8b500000, 8b600000, 8b700000, 8b800000, 8b900000. 8ba00000. 8bb00000. 8bc00000, 8bd00000, 8be00000, 8bf00000, ab000000 ... abf00000 |
| 16:12.15 | cr2 | ok. |
| 16:12.38 | pH5 | comicinker: soft-reset is a winceism, just press the reset-button and it will reboot without clearing ram completely. I don't understand the first part of the question. if you ask about a linux bootloader that can load linux from flash, then the answer is not yet. |
| 16:14.28 | comicinker | pH5: in my case, nothing happens when I push the soft-reset- button under linux |
| 16:15.38 | pH5 | comicinker: in that case, yes, there is no 'soft-reset' under linux. you have to press cam+power+reset to do the 'hard-reset'. |
| 16:20.13 | *** join/#htc-linux AC_Mast27 (i=s@gateway/tor/x-b31f2b974b474387) |
| 16:20.22 | cr2 | Kevin2: yes, now it works. |
| 16:20.26 | cr2 | 029.665 07080de0: e4851004 (??) # a=4e080d78 v=0019001a |
| 16:20.26 | cr2 | 029.665 07080e38: e5812000 (??) # a=4e010220 v=0000035f |
| 16:20.26 | cr2 | 029.665 07089c20: e5912000 (??) # a=4e01021c v=0000035f |
| 16:20.26 | cr2 | 029.665 07089c34: e5912000 (??) # a=4e010220 v=0000035f |
| 16:20.26 | cr2 | 029.665 07089c58: e5912000 (??) # a=4e010140 v=00000110 |
| 16:20.27 | cr2 | 029.665 070ae330: e1d010b0 (??) # a=4e1001b0 v=000012b6 |
| 16:20.47 | cr2 | ?? means objdump not found ? |
| 16:20.51 | Kevin2 | Yes |
| 16:20.52 | *** join/#htc-linux Keizer (n=keizer@c-69-139-42-47.hsd1.fl.comcast.net) |
| 16:21.10 | cr2 | ok. i don't have it on my notebook. |
| 16:21.26 | cr2 | still weird. |
| 16:21.34 | Kevin2 | It's in http://handhelds.org/~koconnor/haret/haretconsole-0.5.0.tar.gz |
| 16:22.01 | cr2 | i have used CVS |
| 16:22.24 | cr2 | hm. i'm not running from the notebook. |
| 16:22.26 | Kevin2 | Yeah - just pointing you to a precompiled objdump if you want it. |
| 16:22.39 | comicinker | pH5: with a hard-reset, do I clear my phonebook entries etc? |
| 16:22.43 | cr2 | should i just add the path ? |
| 16:22.43 | Kevin2 | Oh, you can also use the cegcc one. |
| 16:23.11 | Kevin2 | Put it in the same directory as the python scripts. |
| 16:23.24 | cr2 | as 'objdump' ? |
| 16:23.42 | Kevin2 | arm-linux-objdump |
| 16:24.17 | cr2 | Kevin2: it's nice anyway. i've captured all the ATI camera datastreadm. |
| 16:24.50 | pH5 | comicinker: in linux? no. in wince? yes. they are already gone as soon as you replace wince with linux in RAM. the only thing you can do is to have wince store them in flash, but installed applications will still be gone. |
| 16:25.57 | comicinker | pH5: how can I get the newest version of a distro compiled for the ARM architecture? |
| 16:30.03 | pH5 | comicinker: I'm sorry, that question is a bit too generic for me to be able to come up with a useful answer. |
| 16:32.48 | comicinker | no problem. I will just wait until somebody generates a newer image of openmoko for the htc magician device |
| 16:37.41 | *** join/#htc-linux Mineque (n=Mineque@195.60.65.50) |
| 16:39.28 | pH5 | ah, openmoko. the latest images I built are at http://linuxtogo.org/~ph5/magician/, the stuff in untested/ might or might not work. |
| 16:48.20 | cr2 | BabelO: i've sorted the ATI capture data on BA. the datastream goes through the ringbuffer. the registers are documented, but i don't know if there is some code using them. |
| 16:54.08 | *** part/#htc-linux comicinker (n=simeon@R0b32.r.pppool.de) |
| 16:54.35 | cr2 | Kevin2: 000.001(0004937) IRQS ICIP=00000080: HW uart=1 |
| 16:54.46 | cr2 | pxa27x does not have hwuart. |
| 16:56.14 | Kevin2 | cr2: Isn't BA a pxa26x? |
| 16:56.58 | Kevin2 | The defs for the registers are in haretconsole/regs_pxa.py |
| 17:01.07 | cr2 | Kevin2: i'm checking athena now. |
| 17:01.08 | cr2 | ok. |
| 17:07.33 | Kevin2 | cr2: Latest cvs haret you can do: addlist mmutrace 0x4e000000 4*1024*1024 ; addlist mmutrace 0x8b000000 16*1024*1024 ; addlist mmutrace 0xab000000 16*1024*1024 |
| 17:11.51 | cr2 | ok, i'll try. |
| 17:12.36 | cr2 | the irq handler is accessed at 0x4e* on ba, but 0x4c* on athena. and there are much more ATI mappings on athena. |
| 17:13.56 | cr2 | bbl. |
| 17:14.52 | cr2 | Kevin2: will you write the atiwxxxx-sdio driver ? |
| 17:17.33 | Kevin2 | cr2: Not in short term. |
| 17:50.04 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-103-152-149.karneval.cz) |
| 17:50.18 | ivucica | does anyone here use Debian? |
| 17:50.33 | ivucica | and, what's the easiest way to install the cross-compiling tools? |
| 17:50.42 | ivucica | installing GCC for ARM looks like pain in the *something*. |
| 17:50.49 | ivucica | i'm torturing myself with that for quite some time now |
| 17:54.35 | rob_w | ivucica: use openembedded.org |
| 17:54.45 | ivucica | rob_w: mkay |
| 17:55.12 | ivucica | rob_w: do they prepare a Debian package? ;) |
| 17:55.27 | rob_w | ivucica: no , but a nice GettingStarted |
| 17:55.32 | ivucica | :) |
| 17:56.03 | rob_w | hang on |
| 17:56.24 | rob_w | http://www.openembedded.org/wiki/GettingStarted |
| 17:56.32 | ivucica | awww jeez |
| 17:56.34 | ivucica | ye, i found it |
| 17:56.42 | ivucica | looks gross |
| 17:56.57 | ivucica | all i wanted was a plain-and-simple mingw-like compiler :( |
| 17:57.04 | rob_w | if you go through this and get to use the bitbake environment you will end up have mighty power of many cross platforms |
| 17:57.20 | ivucica | i guess this has to wait some other day ... |
| 17:57.22 | rob_w | ivucica: so ok then hang on one more time |
| 17:57.29 | ivucica | ok ;) |
| 17:57.47 | ivucica | rob_w: all i want is recompile the kernel in order to add support for MMC into it |
| 17:57.58 | ivucica | or support for usbnet |
| 17:58.07 | ivucica | the images i found dont seem to have it built in |
| 17:58.34 | ivucica | the thing works -- if i flash it in ;) |
| 17:58.46 | ivucica | however, i want to boot off of an ext2 partition on my MMC card ... |
| 17:58.53 | ivucica | i just want a quick fix :) |
| 17:59.09 | ivucica | well, "quick" -- i mean a simple way of recompiling the kernel |
| 17:59.16 | rob_w | hmm usbnet is one major thing on those devices .. they will always have that , i hope |
| 17:59.34 | ivucica | i'm not sure this one does |
| 17:59.43 | ivucica | i'm trying to boot using haret on h3800 |
| 17:59.57 | ivucica | it boots alright, but doesnt correctly introduce itself to the host |
| 18:00.08 | ivucica | i'm getting "error -71", whatever that means |
| 18:00.20 | rob_w | maybe that helps http://www.angstrom-distribution.org/unstable/toolchain/ |
| 18:00.40 | ivucica | i got it with some other devices in the past which were broken -- this one isnt (usb works in wince) |
| 18:00.42 | ivucica | ok |
| 18:01.11 | rob_w | ivucica: your host is not allowed to load a module called ipaq_something !! try to prevent that |
| 18:01.27 | ivucica | i rmmod it |
| 18:01.32 | ivucica | but ... hmm |
| 18:01.36 | ivucica | i'll try blacklisting it |
| 18:01.48 | ivucica | however, i dont think that's an issue |
| 18:02.13 | ivucica | see, i added rootdelay of 5 sec, then quickly rmmodded ipaq, unplugged and re-plugged |
| 18:02.16 | ivucica | it didnt work |
| 18:02.40 | ivucica | i really think that usbnet isn't there, but that's odd since NFS and friends ARE |
| 18:02.58 | rob_w | hmm move the ipaq.ko away - depmod again then the module is not available for your hosts hotplug |
| 18:03.10 | ivucica | (i suppose i'm not taking the uclibc version?) |
| 18:03.16 | ivucica | (of the toolchain) |
| 18:03.34 | rob_w | ivucica: if you are unsure then choose not the uclibc |
| 18:03.38 | ivucica | rob_w: i'll try blacklisting it in modprobe |
| 18:03.40 | ivucica | ok |
| 18:04.06 | ivucica | rob_w: further, should i use handhelds.org kernel or do the angstrom have their own? |
| 18:04.32 | rob_w | for a h3800 ? |
| 18:04.47 | pH5 | ivucica: one possibility to get a cross-compiler might be to apt-get install emdebian-tools and then use emsetup and emchain to set up a cross-compilation toolchain. |
| 18:04.58 | ivucica | rob_w: yes, h3800 |
| 18:04.59 | pH5 | but I don't have experience with that either, I use openembedded, too. |
| 18:05.11 | ivucica | pH5: hmmm, i'll try that |
| 18:05.31 | ivucica | i'll try ANYTHING |
| 18:05.53 | rob_w | http://www.angstrom-distribution.org/unstable/images/h3800/ |
| 18:06.01 | ivucica | my "deadline" is monday -- there's the "three of us" having a total of 11 PDAs >: ) |
| 18:06.14 | ivucica | out of that, 3 are PocketPC, rest are PalmOS |
| 18:06.28 | ivucica | now i'm trying to be the "first one with PocketPC with Linux" >: ) |
| 18:06.31 | rob_w | what is that deadline for ? who did the most linux pda `s ?? |
| 18:06.41 | ivucica | rob_w: those are images i tested :D |
| 18:06.53 | ivucica | rob_w: on monday we're having this class called "Operating Systems" on our faculty |
| 18:07.00 | rob_w | those will have prober usbnet i think |
| 18:07.08 | ivucica | and well, we're trying to disgoust our professor with all the PDAs on the table |
| 18:07.20 | ivucica | in addition to that ... |
| 18:07.22 | rob_w | so there will be only one with linux ?? |
| 18:07.35 | ivucica | most of the things on that subject are regarding unices ... |
| 18:07.44 | ivucica | it's stuff like semaphores, signals, etc. |
| 18:07.54 | rob_w | sounds dry |
| 18:07.58 | ivucica | indeed... |
| 18:08.18 | ivucica | ...they even use weird Croatian translations for stuff like "pointer" ... |
| 18:08.21 | ivucica | (i'm from Croatia) |
| 18:08.22 | pH5 | ivucica: emchain doesn't seem to work here out of the box :( |
| 18:08.37 | rob_w | faculty `S still try to teach "the lowest knowledge possible " but nothing usefull for life at all |
| 18:08.46 | rob_w | pH5: hehe |
| 18:08.49 | ivucica | kazaljka ==> pointer, but the one on the clock |
| 18:08.54 | ivucica | pH5: confirmed |
| 18:09.18 | rob_w | well sure we do the same in germany .. zeiger == pointer |
| 18:09.31 | ivucica | rob_w: however, this sounds odd and weird |
| 18:09.32 | rob_w | but that is a quite good translation in that sense |
| 18:09.43 | ivucica | it should be "pokazivac" |
| 18:09.49 | ivucica | which is roughly "person who is pointing" |
| 18:09.54 | ivucica | and not the "thing on the clock" |
| 18:09.56 | rob_w | heahehaeaahe |
| 18:10.12 | ivucica | and that's not all ... |
| 18:10.14 | Kevin2 | ivucica: If you're just trying to build a cross compiler, I've used buildroot with success - http://buildroot.uclibc.org/ |
| 18:10.16 | ivucica | "dretva" instead of "nit" (thread) |
| 18:10.24 | rob_w | like the american indian langugeas ." Those who ... " "the one which .. " |
| 18:10.35 | ivucica | Kevin2: i'm trying to build the kernel -- and i'm trying to get a working cross-compiler :) |
| 18:11.01 | ivucica | rob_w: well it isn't like that ... |
| 18:11.09 | ivucica | this is not just "person" .... |
| 18:11.20 | ivucica | it'd be "that which points" |
| 18:11.26 | rob_w | ivucica: so i ve you are basis are the angstrom images then go for openembedded .. you might not succeed to setup all up and get to know it till monday but on the long run you will climb heaven |
| 18:11.28 | ivucica | same as "pointer"... |
| 18:11.57 | ivucica | hmm |
| 18:12.07 | ivucica | as i said at the moment i'm looking for a quick fix ;) |
| 18:12.11 | ivucica | i'm limited in time and disk space |
| 18:12.15 | rob_w | in the end you can show your faculty "linux on ANY device " or "the os which runs them all" |
| 18:12.41 | ivucica | damned NTFS partition takes up 40 out of 60GB, and i've filled the ext3 partitions :) |
| 18:12.44 | rob_w | ivucica: then check the existing images i really bet it must have propoer usbnet |
| 18:12.55 | ivucica | i'll see |
| 18:13.26 | ivucica | on the other hand, getting MMC to work will be a cause for ecstasy since then i can mount the 512mb MMC and boot from that ;) |
| 18:13.27 | rob_w | it should end up in something like 192.168.0.202 |
| 18:13.51 | ivucica | rob_w, i tried it during the period of time when it was "flashed in" |
| 18:14.05 | ivucica | i first flashed linux -- but that's not really what i want to do |
| 18:14.23 | ivucica | especially since getting GCC installed on that limited "disk" space will be a problem |
| 18:14.29 | rob_w | then you will need the kernel & initrd to load the mmc and pivot_root to it |
| 18:14.53 | rob_w | yet you will reset windows ce hard after each boot .. i think |
| 18:15.06 | rob_w | bbl |
| 18:16.20 | ivucica | rob_w: yes, i will :) |
| 18:16.29 | ivucica | i mean, i will hardreset it |
| 18:17.02 | ivucica | it will hardreset it |
| 18:17.09 | ivucica | but that is no problem for me |
| 18:18.54 | ivucica | ipaq 3870 has a nice backup utility ^^ i've learned to use it quite well during the last week |
| 18:19.04 | ivucica | (and i learned to hate the initial calibration screen) |
| 18:23.04 | *** join/#htc-linux pH5_ (n=ph5@p5485D7F2.dip.t-dialin.net) |
| 18:26.38 | cr2 | back |
| 18:27.05 | cr2 | Kevin2: ok, so i need to try to write the atiwxxx-mmc driver. |
| 18:29.39 | Kevin2 | You'd have my thanks. :-) |
| 18:29.50 | cr2 | :) |
| 18:36.55 | BabelO | cr2: so you have traced ati on BA :) |
| 18:42.02 | ivucica | i just HAVE to say this |
| 18:42.04 | ivucica | you guys ROCK :) |
| 18:43.24 | cr2 | BabelO: yes. |
| 18:43.54 | BabelO | cr2: good good good, so i can get back my adcm driver for i2c :) |
| 18:44.02 | cr2 | BabelO: so you only need to know how to operate the ringbuffer. |
| 18:44.27 | cr2 | BabelO: i see the i2s ATI regs too. but on w2284 :) |
| 18:44.36 | cr2 | no sound on w3220 |
| 18:45.05 | BabelO | cr2: did you rtry latest image ? |
| 18:45.33 | cr2 | not yet. i was so happy that tracing works ;) |
| 18:45.47 | BabelO | :) |
| 18:48.37 | ivucica | oh boy oh boy here we go |
| 18:48.38 | ivucica | :) |
| 18:54.53 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-103-152-149.karneval.cz) |
| 19:19.29 | *** join/#htc-linux |dion| (n=dion@inhex.net) |
| 19:24.33 | cr2 | heh. ldrb from dpram, but strh. |
| 19:25.13 | ivucica | cr2: interesting |
| 19:25.27 | cr2 | ldrh from tx buffer. |
| 19:25.46 | ivucica | LDRH == load register, half word? |
| 19:26.10 | ivucica | if it is, then "woah! i remember some of that ARM ASM we were taught last year" |
| 19:28.30 | *** join/#htc-linux goxboxlive (n=goxboxli@139.80-202-160.nextgentel.com) |
| 19:30.52 | cr2 | yes. |
| 19:31.14 | cr2 | half register. |
| 19:31.34 | cr2 | ldrh on alpha is a 32bit load. |
| 19:32.10 | cr2 | nice. dpram control seems to be relatively simple. |
| 19:32.34 | ivucica | cr2: alpha?! where did you dig that processor out from? :D |
| 19:33.22 | cr2 | nice 64bit cpu without any fluff, like on amd64 ;) |
| 19:35.24 | ivucica | hehe |
| 19:35.38 | ivucica | found in a ppc or elsewhere? |
| 19:36.05 | cr2 | have a ppc too |
| 19:36.52 | ivucica | er, did that mean "have a powerpc too" or "found in pocketpc too"? |
| 19:38.47 | ivucica | daamn, only 17mb free space left on root |
| 19:44.47 | cr2 | have a powerpc too |
| 19:46.53 | ivucica | cr2: :) |
| 19:49.55 | cr2 | hmm. still can't find the usb host. |
| 20:11.01 | *** join/#htc-linux TimRiker (n=timr@rikers.org) |
| 20:14.49 | cr2 | Kevin2: if i have mapped a big area for tracing, is it possible to mask some spamming addresses from it ? |
| 20:20.35 | Kevin2 | cr2: Sure. |
| 20:21.26 | Kevin2 | You need to make two mappings - one before the troublesome address, and then one after the troublesome address. (Thus not "mapping" the spamming register.) |
| 20:22.04 | cr2 | ok. |
| 20:22.08 | cr2 | 006724: mmutrace 090d4398: e18020b3(strh) a9200900 00000705 (00000000) |
| 20:22.08 | cr2 | 006957: mmutrace 090d4398: e18020b3(strh) a9200900 00000403 (00000000) |
| 20:22.08 | cr2 | 007379: mmutrace 090d4398: e18020b3(strh) a9200900 00000507 (00000000) |
| 20:22.08 | cr2 | 008871: mmutrace 090d4398: e18020b3(strh) a9200900 00000434 (00000000) |
| 20:22.08 | cr2 | 010302: mmutrace 090d4398: e18020b3(strh) a9200900 00000434 (00000000) |
| 20:22.28 | cr2 | CMD5, CMD3, CMD7, CMD52 |
| 20:22.38 | cr2 | nice. |
| 20:23.58 | Kevin2 | You can also have a mmutrace on a register and then suppress duplicate reads/writes - this involves setting up a "mask" and "compare value". This really only works if the mmutrace size is 4 bytes or less, however. |
| 20:30.13 | cr2 | ok, but there are 2 such registers. |
| 20:30.48 | Kevin2 | 2 registers that are "spamming"? Do you want to suppress them entirely or just suppress duplicates? |
| 20:32.07 | cr2 | suppress duplicates |
| 20:34.25 | Kevin2 | So: addlist mmutrace <start> <len_to_reg1> ; addlist mmutrace <reg1addr> 4 'rw' 0 last ; addlist mmutrace <reg1addr + 4> <len_to_reg2> ; addlist mmutrace <reg2addr> 4 'rw' 0 last ; addlist mmutrace <reg2addr + 4> <len_to_end> |
| 20:36.35 | Kevin2 | You can also make use of the fact that the first mmutrace takes priority so - addlist <reg1addr> 4 'rw' 0 last ; addlist <reg2addr> 4 'rw' 0 last ; addlist <start> <total_len> -- should also work. |
| 20:36.54 | Kevin2 | (make sure you do a "clear mmutrace" first though.) |
| 20:38.10 | cr2 | great. i need to create a wiki page for the tracing recipies :) |
| 20:39.26 | Kevin2 | I tried to start one at http://www.handhelds.org/moin/moin.cgi/HaRET_20Documentation but never got to filling it it. :-( |
| 20:41.42 | rob_w | are you guys hacking the ati video these days ? |
| 20:42.35 | cr2 | rob_w: ati-sdio |
| 20:42.50 | rob_w | oh i see |
| 20:42.55 | cr2 | but the ati multimedia asic is also there. |
| 20:43.02 | cr2 | video and i2s audio. |
| 20:43.03 | ivucica | i guess i'll go now ;) |
| 20:43.14 | cr2 | i've a register list here |
| 20:43.30 | cr2 | http://wiki.xda-developers.com/index.php?pagename=ATIWxxxx |
| 20:43.32 | ivucica | hopefully i'll manage to get linux compiled tomorrow |
| 20:43.33 | ivucica | bye! |
| 20:43.39 | rob_w | cr2: i wonder when one could really use the 3d capability`s of such devices and render a real nice gui |
| 20:43.39 | *** part/#htc-linux ivucica (n=ivucica@217.14.216.214) |
| 20:44.39 | cr2 | rob_w: w32x0 and w228x do not have 3d. |
| 20:45.01 | rob_w | i though the one on the ba had some of this |
| 20:45.17 | cr2 | even the 2d accel is not great in linux, because there are zero docs. |
| 20:45.22 | cr2 | ba has w32x0 |
| 20:46.33 | cr2 | do you know if one can get an imageon SDK for wince somewhere ? |
| 20:46.41 | rob_w | there is this new touch-only htc device- whats is codename ? |
| 20:46.59 | BabelO | rob_w: htc elf |
| 20:47.06 | BabelO | ? |
| 20:47.39 | cr2 | rob_w: it's omap850. pita. |
| 20:47.47 | BabelO | :) |
| 20:48.13 | cr2 | BabelO: you can trace the CAM at 0x4e000000 |
| 20:48.33 | rob_w | BabelO: right |
| 20:48.43 | cr2 | rob_w: do you still have a ba ? |
| 20:48.45 | rob_w | but i like the design |
| 20:48.55 | rob_w | cr2: unfortunately not |
| 20:49.07 | cr2 | rob_w: other htc phone? |
| 20:49.19 | rob_w | maybe a elf soon enough ;-) |
| 20:50.21 | BabelO | rob_w: this custom works on all omap based phone, i have exactly same on my artemis, and it exists same on some other even ba can have wm6 with 3d cube |
| 20:51.00 | rob_w | BabelO: how can the ba do that , you mean ? |
| 20:51.53 | BabelO | rob_w: it is just a wm6 custom from htc ;) |
| 20:52.06 | rob_w | what ? |
| 20:52.10 | rob_w | custom ? |
| 20:53.30 | rob_w | 385,00 EUR for a elf .. hmm |
| 20:53.35 | BabelO | rob_w: yes it is a cab file :) you can look at xanadux forum for each device it exist a wm6 with that cab file |
| 20:54.01 | rob_w | so you mean a 3d test app which shows a spinning cube ? |
| 20:54.32 | rob_w | can the latest haret load a kernel out of wm6 ? |
| 20:54.44 | BabelO | yes to switch between contacts, app, and multimedia |
| 20:54.58 | BabelO | rob_w: i can start linux from wm6 with haret on artemis |
| 20:55.09 | rob_w | sounds good |
| 20:55.21 | BabelO | it is just a wince5 light ;) |
| 20:56.11 | *** join/#htc-linux comicinker (n=simeon@R0b32.r.pppool.de) |
| 20:56.50 | rob_w | 438,80 € for a elf without contract |
| 20:57.07 | rob_w | BabelO: WHAT .. wm6 is meant as a wm5 light ?? what ms crap is that again |
| 20:57.08 | comicinker | on which distros does the phone of the htc work? |
| 20:57.59 | rob_w | oh guys , i had a chance to see a ms xp embedded presentation .. i hardly hit the floor laughing to death .. you cant believe this |
| 20:58.02 | cr2 | rob_w: too much for an omap850 phone ? |
| 20:58.29 | cr2 | rob_w: it seems more like a qualcomm-based phone. |
| 20:58.35 | rob_w | hmm cr2 i dont know .. when i saw it first i had this strange feeling in my stomach |
| 20:58.58 | rob_w | cr2: what would u say ? |
| 21:00.24 | cr2 | http://wiki.xda-developers.com/index.php?pagename=HTC_Elf |
| 21:01.17 | rob_w | no linux mentioned yet .. so its time !! |
| 21:01.28 | rob_w | rob_w will come back to the world of HTC ! |
| 21:02.14 | rob_w | if i find the elf below 400 without contract ,, i will buy :-) |
| 21:16.02 | rob_w | 376.83 € .. and sold ! |
| 21:16.09 | *** part/#htc-linux rmoravcik (n=rmoravci@ip-89-103-152-149.karneval.cz) |
| 21:39.17 | comicinker | wahoo! i put the last testing zImage on my magician, and now I cannot do anything with it. can somebody help me? |
| 21:41.30 | comicinker | may it help until I wait the internal battery is used off? |
| 21:43.45 | comicinker | wow sh... thought I couldn't crash my device playing like that... |
| 22:00.38 | pH5_ | comicinker: resetting with power+camera buttons held always works for me. |
| 22:01.01 | comicinker | pH5: well, that worked for me too till now |
| 22:01.26 | comicinker | oh |
| 22:01.35 | comicinker | the batterycase was shifted |
| 22:01.50 | comicinker | I'm sorry |
| 22:01.52 | pH5 | heh, that's the same as holding down reset :) |
| 22:04.58 | comicinker | in which status is the device after opening the battery case and then closing it again? |
| 22:11.30 | comicinker | I mean- it is not off, is it? I thought you cannot switch off the device? but is not running as well, so in which status is the device? |
| 22:11.52 | pH5 | comicinker: the same as after pressing the reset button. it starts executing from address 0x0 (in flash) and the bootloader either jumps into the wince kernel at 0x40000 (in flash) or starts the bootloader interface if power+cam are pressed after about 500ms |
| 22:13.53 | comicinker | is it possible to write the adress 0x40000? or is it possible to mount this area after starting linux? |
| 22:15.18 | pH5 | comicinker: or, wait. the battery switch probably asserts the battery fault line, so most probably it is in deep sleep until you press reset again. |
| 22:15.37 | pH5 | comicinker: yes, it is possible to write the address 0x40000 |
| 22:16.01 | comicinker | so is it possible to place the linux kernel there for booting? |
| 22:16.24 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-103-152-149.karneval.cz) |
| 22:16.54 | pH5 | either from linux by writing to mtdblock1 after loading the mtd drivers, or via the htc-bootloader (an early but working usb flasher tool is at http://linuxtogo.org/~ph5/htc-tools/) |
| 22:17.11 | pH5 | but we can't put the kernel at 0x40000 directly because it needs ram enabled. |
| 22:17.56 | pH5 | there needs to be at least some glue code that enables ram and then jumps into the kernel, which could be stored at 0x80000 then |
| 22:22.57 | cr2 | pH5: there was a hermes flasher on xda-dev. does your code have something in common ? |
| 22:23.33 | comicinker | I think I understand. but why needs the ram at 0x40000 to be write enabled? I thought it is writeable? plus: is this the area where the the o2-software was stored, too? |
| 22:24.29 | cr2 | comicinker: it's NOR flash there. |
| 22:24.40 | pH5 | cr2: no idea, but I'm using the hermes bootloader password generation code by 'TheBlasphemer' |
| 22:24.51 | cr2 | pH5: ok. |
| 22:25.21 | cr2 | unsigned long |
| 22:25.27 | cr2 | is not professional :) |
| 22:26.50 | cr2 | pH5: but then you can flash the radio too ? |
| 22:28.09 | pH5 | cr2: I didn't try that yet. |
| 22:28.26 | cr2 | pH5: how did sdg solve that problem on hx4700 ? |
| 22:28.42 | cr2 | or is the sdram enabled there ? |
| 22:30.02 | pH5 | cr2: sdg have a complete second bootloader including flashing facilities and whatnot after the htc one. the sdram is enabled by bootldr. |
| 22:30.26 | cr2 | yes, but it must be flashed at 0x40000 too. |
| 22:30.35 | pH5 | exactly |
| 22:30.46 | cr2 | ok. |
| 22:31.04 | cr2 | how big is it ? |
| 22:31.13 | cr2 | it's the old cpq bootldr ? |
| 22:32.08 | pH5 | ahh |
| 22:32.10 | pH5 | < 256k |
| 22:32.32 | cr2 | so 512k is reserved for the bootloaders. |
| 22:32.50 | cr2 | btw, looking at your code. i want hima and ba there too :) |
| 22:33.10 | pH5 | cr2: feel free to test and send patches :) |
| 22:33.16 | cr2 | and 'long' should be replaced by int32. |
| 22:33.30 | cr2 | otherwise it will blow everything on my old notebook ;) |
| 22:33.34 | cr2 | ok. |
| 22:34.37 | pH5 | basically only the machine id strings have to be added, and I need to think about how to properly handle the partition layout for different devices (people could have flashed stuff like bigstorage roms etc.) |
| 22:34.59 | cr2 | i came to a conclusion that people using 'long' usually don't have a clue what they are doing, and what pain they are inflicting later. |
| 22:35.14 | pH5 | heh, the setup.h header actually comes from the kernel |
| 22:35.15 | cr2 | you can dump the rom first, and look for the magic strings. |
| 22:35.50 | pH5 | I should rewrite the passgen.c part anyway, but you know how it is once 'it works' :) |
| 22:38.48 | cr2 | pH5: http://forum.xda-developers.com/showthread.php?t=296436 |
| 22:46.47 | pH5 | cr2: wow, I wish I had known about that earlier |
| 22:47.13 | pH5 | good night |
| 22:55.36 | goxboxlive | cr2: i dont understand. The power button on A780 isent hooked up to either event0 .......event2 |
| 22:55.43 | goxboxlive | Where the heck is it? |
| 22:57.04 | cr2 | maybe it has an own irq handler ? |
| 22:57.13 | cr2 | in the kernel. |
| 22:57.19 | goxboxlive | ohh |
| 22:58.01 | goxboxlive | I am trying to make keys working in Qtopia with it. |
| 22:58.28 | goxboxlive | I have it allmost complete, but some issues wuth the numeric ones |
| 23:09.18 | cr2 | it's also necessary to pick out the pxa_camera driver. |
| 23:09.27 | cr2 | does the cam work for you on a780 ? |
| 23:13.00 | goxboxlive | cr2: what should i look for? |
| 23:13.05 | goxboxlive | /dev/video? |
| 23:13.20 | goxboxlive | i dont have that |
| 23:15.42 | goxboxlive | i have a ovcamchip.h in the include source |
| 23:16.42 | *** join/#htc-linux AC_Mast27 (i=s@gateway/tor/x-e2479e830fcff97a) |