| 00:00.50 | Kevin2 | NetRipper: Ok. |
| 00:01.39 | cr2 | Kevin2: yes. i'm looking for the hermes git with the keyboard driver. i've rechecked the hermes cpld gpios, and would like to fix the serial setup |
| 00:02.31 | NetRipper | i'll make a proper diff later.. bedtime now |
| 00:02.33 | NetRipper | night all |
| 00:02.33 | NetRipper | :) |
| 00:02.47 | NetRipper | (current diff is missing the new font file) |
| 00:03.27 | snakefreak | kevin2 do you have time for me? |
| 00:05.58 | Kevin2 | snakefreak: In what way? |
| 00:06.54 | snakefreak | can you tell me how to boot an filesystem image with haret? |
| 00:07.19 | Kevin2 | Hrmm. Do you have an initrd ready? |
| 00:07.53 | snakefreak | no. |
| 00:08.57 | Kevin2 | Okay - that's a lot of work then. Take a look at openembedded.org or one of the other distros. First step is to get an initrd. Once you have that, then you can wire it into haret. |
| 00:10.26 | Kevin2 | cr2: Cool - so you have a renewed interest in hermes? Big thing before was getting SD working. |
| 00:10.57 | snakefreak | hm ok i know how to vreate one. but what is the integrents |
| 00:11.13 | snakefreak | -v +c |
| 00:11.48 | cr2 | Kevin2: i#d like to fix the easy things first. |
| 00:12.25 | cr2 | Kevin2: but i still have my notes for ATI, and a lot of ATI SD traces. in hardcopy :) |
| 00:13.45 | tmzt_ | how big can the initrd be on universal? |
| 00:13.49 | Kevin2 | snakefreak: Build an initrd with whatever you want in it. Haret wont care. |
| 00:14.50 | snakefreak | i do not understand. |
| 00:15.52 | snakefreak | http://wiki.neilandtheresa.co.uk/Titchy_Linux/Installing_Debian |
| 00:16.45 | snakefreak | i have all this things and i have maked a image from the rootfs tar file. |
| 00:17.30 | Kevin2 | snakefreak: Hrmm. I'm not sure what you are trying to do. Haret is just a bootloader. You need a kernel and filesystem for your device to boot into. Once you have that, you can boot via haret. Without that, there is not much you can do with haret. |
| 00:18.16 | snakefreak | yes thats clear. |
| 00:19.16 | snakefreak | i have a file system. and a kernel. but i dont know how the startup.txt should look |
| 00:20.54 | Kevin2 | See http://www.handhelds.org/moin/moin.cgi/HaRET_20Documentation -- read the section on booting linux. |
| 00:21.04 | snakefreak | set cmdline "root=/dev/mmcblk0p1 rw is clear mmc/sd card 0 partition 0=1 1=2 .... but what ist that for an image on fat partition |
| 00:22.18 | snakefreak | kevin2 i have read it. but it brings me not forward. |
| 00:22.29 | Kevin2 | Hrmm. It would be "root=/dev/xyz". I forget which one it typically is. |
| 00:24.19 | snakefreak | yes i know mmc sd card partition2 is mmcblk0p1 |
| 00:24.30 | snakefreak | and loop for an image. |
| 00:24.43 | Kevin2 | Yeah - that sounds right - so make sure "root=/dev/mmcblk0p1" is in your cmdline. |
| 00:25.03 | Kevin2 | Did you want to build a all-in-one exe to install and boot from? |
| 00:27.20 | snakefreak | no i want only boot an image via haret. |
| 00:27.37 | snakefreak | actually it looks: |
| 00:29.54 | snakefreak | kevin2 why root=mmcblk0p1 it should not a partition it should a file on a partition. |
| 00:30.20 | snakefreak | set cmdline "root=/dev/loop looproot=/dev/mmcblk0p1:/qtopia.linux.img rw console=tty0 rootdelay=5 fbcon=rotate:1" |
| 00:31.27 | Kevin2 | I'm confused. You have your filesystem in a file on another file system? Why not just install the filesystem on the sd card directly? |
| 00:33.56 | snakefreak | i havent a card reader which is linux compatible. and i will youse multible linux systems to find out the best for me. |
| 00:37.55 | tmzt_ | /debian/dists/testing/main/installer-armel/current/images/versatile/netboot/ |
| 00:38.02 | tmzt_ | ftp.us.debian.org |
| 00:39.35 | tmzt_ | put the Titchy root .tar.bz2 on your fat partition |
| 00:41.18 | snakefreak | i will try next day have to sleep. good night all and thanks for help! |
| 00:41.38 | tmzt_ | use the Titchy zImage, create a default.txt with "set KERNEL zImage" "set INITRD initrd.gz" |
| 00:42.10 | snakefreak | ok |
| 00:43.05 | *** part/#htc-linux wasup (n=Tech@208.47.103.106) |
| 00:43.12 | tmzt_ | "set CMDLINE init=/linuxrc", initramfs doesn't need root= does it? |
| 00:46.43 | tmzt_ | dcordes: do you know the url for the diamond thread? |
| 00:48.31 | dcordes | tmzt_: http://forum.xda-developers.com/showthread.php?t=402002 |
| 00:49.31 | cr2 | Kevin2: can you apply the latest mach-types and NetRipper's patch for msm7201A ? i#ve tested it and it works as expected |
| 00:50.10 | cr2 | Kevin2: the gpio bank5 is missing, but i'll check tomorrow where its registers are located. today it's too late ;) |
| 00:55.27 | cr2 | hehe. crashed raph100 with permissivemmutrace |
| 00:55.40 | cr2 | wiif |
| 00:55.45 | cr2 | wifi is unhappy |
| 00:57.20 | cr2 | nice. does not resume from sleep. softreset needed. |
| 01:00.11 | tmzt_ | someone said a g1 emulator is posted, might this have a real touchscreen coordinates? |
| 01:02.17 | dcordes | tmzt_: you can grab a .zip file from google which has the dalvik development files and the qemu based emulator source |
| 01:05.04 | *** join/#htc-linux surgex (i=surge@pool-98-118-166-131.bflony.fios.verizon.net) |
| 01:06.03 | cr2 | at@bric=0 |
| 01:06.39 | dcordes | is that the command used for bricking the device? |
| 01:06.43 | *** join/#htc-linux wasup (n=Tech@208.47.103.106) |
| 01:06.52 | cr2 | i guess so :) |
| 01:07.04 | wasup | can anyone tell me why it's failing to mount my sdcard |
| 01:07.19 | dcordes | wasup: it being..? |
| 01:07.19 | wasup | it shows the sdcard...it says its found |
| 01:07.33 | cr2 | 10.1.101.63 ? |
| 01:07.48 | wasup | being? |
| 01:07.54 | dcordes | wasup: titan? |
| 01:07.57 | wasup | yeah |
| 01:08.00 | wasup | titan |
| 01:08.11 | cr2 | dcordes: what is this address ? |
| 01:08.22 | dcordes | cr2: a gprs one? |
| 01:08.33 | wasup | i feel like a noob.... it was working before |
| 01:08.41 | wasup | i reformated my sd to fat32 |
| 01:08.51 | cr2 | port 7275 ? |
| 01:09.21 | cr2 | does not respond. |
| 01:09.43 | wasup | copied the new files 1.0 and now it says |
| 01:09.43 | wasup | mounting /dev/block/mmcb1k0p1 on /sdcard failed..... |
| 01:09.44 | tmzt_ | are they keeping the mmu on or turning it off on diamond? |
| 01:09.48 | wasup | invalid arguement |
| 01:09.57 | dcordes | wasup: linux won't mount a vfat rootfs. the default.txt has a rootfs= value that defines the path to the device with the root filesystem |
| 01:10.22 | dcordes | cr2: it does for me |
| 01:10.27 | wasup | oh...so what do i change it to? |
| 01:10.49 | dcordes | wasup: depends. what do you want to do? |
| 01:10.57 | wasup | just boot android |
| 01:11.01 | cr2 | dcordes: i have proxy |
| 01:11.03 | wasup | the new version |
| 01:11.21 | tmzt_ | cr2: are they keeping the mmu on or turning it off on diamond? |
| 01:11.52 | cr2 | don't know |
| 01:11.54 | dcordes | wasup: can you give me a url? |
| 01:12.10 | cr2 | no nmea at 0x1f. strange |
| 01:12.24 | wasup | a url of what? |
| 01:13.11 | dcordes | wasup: you want to boot this http://it029000.massey.ac.nz/vogue/android-new.zip ? |
| 01:13.19 | wasup | oh yeah... |
| 01:13.24 | tmzt_ | cr2: swetland said it's possible to access periphrials with mmu off, I had the log and dcordes posted it in that thread |
| 01:13.29 | wasup | sorry im confused |
| 01:13.41 | wasup | yeah thats the package i downloaded |
| 01:13.41 | cr2 | 1a210 |
| 01:13.44 | dcordes | looks like dzo has all the shiny things working :) sound, video playback etc |
| 01:14.26 | cr2 | dcordes: diamond/raphael will work too |
| 01:15.35 | cr2 | my values are different from diamond ?? |
| 01:15.37 | dcordes | well vogue is usable for a while now. I mean he has everything working the device has to offer. except for usb maybe |
| 01:15.53 | cr2 | $PSTIS |
| 01:16.23 | wasup | dcordes: has anyone else with a titan been experiencing problems??? |
| 01:16.42 | wasup | or am i that unluck SOB....:) |
| 01:16.47 | wasup | unlucky* |
| 01:17.05 | dcordes | wasup: maybe the kernel image in that zip file only has fat16 |
| 01:17.58 | wasup | hmm.... so how would i fix that? |
| 01:18.07 | wasup | reformat it? |
| 01:18.14 | dcordes | reformat the card and have on fat16 partition |
| 01:18.25 | dcordes | *one |
| 01:18.43 | wasup | can i do that via windoes? |
| 01:18.47 | wasup | windows* |
| 01:22.44 | cr2 | good night |
| 01:24.38 | dcordes | wasup: I don't know. but if it worked before the same way (with the previous zip file), it's likely that dzo forgot to include fat32 |
| 01:25.06 | tmzt_ | vfat is fat16 and fat32 |
| 01:25.16 | wasup | true... i am reformatting my card as just FAT |
| 01:25.21 | wasup | so maybe it will work... |
| 01:25.22 | wasup | idk |
| 01:25.40 | wasup | could it have to do with anything in the default.txt? |
| 01:25.52 | wasup | set MTYPE 1667 |
| 01:25.53 | wasup | set KERNEL zImage |
| 01:25.53 | wasup | set initrd initrd.gz |
| 01:25.53 | wasup | #set initrd initrd-android |
| 01:25.53 | wasup | #set initrd initrd-console |
| 01:25.53 | wasup | set cmdline "ppp.username=mobile@jamamobile ppp.password=telecom msmsdcc_fmax=48000000" |
| 01:25.55 | wasup | # console=ttyMSM0" |
| 01:25.57 | cr2 | hm. hitachi lcd panel on my raph100 |
| 01:25.57 | wasup | boot |
| 01:26.01 | tmzt_ | wait, CONFIG_FAT_FS is fat, and CONFIG_VFAT_FS is fat32, and fat16 with long file name support |
| 01:26.56 | cr2 | the VIC setup should be different |
| 01:27.08 | dcordes | wasup: ok sorry forget what I said about 16/32 |
| 01:27.23 | wasup | why me? seriously...%##@$ |
| 01:27.43 | dcordes | huh? :) |
| 01:27.46 | tmzt_ | what are the filenames in the zip? |
| 01:27.58 | wasup | data.gz |
| 01:28.05 | wasup | intrid.gz |
| 01:28.18 | wasup | initrid.gz* |
| 01:28.25 | wasup | system.gz |
| 01:28.33 | dcordes | data.gz and initrd.gz are gzipped ext3 loopfiles. initrd.gz is a gzipped cpio initial ramdisk |
| 01:28.38 | wasup | then the zImage |
| 01:28.54 | dcordes | data.gz and system.gz are gzipped ext3 loopfiles. initrd.gz is a gzipped cpio initial ramdisk |
| 01:28.56 | tmzt_ | are they in the root of the zip or in a directory? |
| 01:29.00 | dcordes | is tired |
| 01:29.17 | wasup | huh? |
| 01:29.27 | wasup | they are in the root of my sdcard |
| 01:30.01 | wasup | i had it working perfect before... |
| 01:30.06 | tmzt_ | they were not in files in the zip? |
| 01:30.09 | dcordes | tmzt_: it boots into the initrd which has a script that extracts the loopfiles |
| 01:30.16 | wasup | now... i'm worthless |
| 01:30.23 | tmzt_ | so it has to mount the first partition |
| 01:30.28 | dcordes | yea |
| 01:30.28 | wasup | sii |
| 01:31.03 | dcordes | mounting mmcblk0p1 is the first thing the script does |
| 01:31.04 | wasup | i am useless POS to the htclinux community now |
| 01:31.35 | cr2 | GPS: [NMEAThread - ReadData] No listeners currently. Stop reading NMEA from SMD |
| 01:32.27 | dcordes | -irq here- |
| 01:32.36 | wasup | does the MTYPE have anything to do with anything? |
| 01:33.10 | dcordes | wasup: yea it tells haret on what device it runs. it should not have changed since the last release from dzo |
| 01:33.37 | wasup | well i haven't booted the last 2 releases |
| 01:33.54 | wasup | do you know if it's different on the titan? |
| 01:33.59 | tmzt_ | even if it should be different for titan, if the MTYPE matches the kernel it should work |
| 01:34.20 | wasup | hmm..so what could be causing the failing mount? |
| 01:34.26 | wasup | should i try another SD card? |
| 01:34.30 | dcordes | yea |
| 01:34.40 | wasup | ok...just a sec...i think i have one in my car |
| 01:35.03 | tmzt_ | dcordes: does it extract those ext2's or just loop mount them? |
| 01:36.36 | wasup | ok im trying another one |
| 01:39.03 | dcordes | tmzt_: http://rafb.net/p/pqCZSm10.html |
| 01:39.12 | *** join/#htc-linux sikkle (n=sikkle@modemcable132.57-201-24.mc.videotron.ca) |
| 01:39.14 | tmzt_ | it forces vfat in /init |
| 01:39.47 | dcordes | yea mount -t vfat /dev/block/mmcblk0p1 /sdcard |
| 01:39.50 | wasup | no F***ing way.... it mounted |
| 01:39.55 | dcordes | great |
| 01:40.08 | wasup | its the exact same type of card |
| 01:40.11 | tmzt_ | wasup: did you get any text error messages before |
| 01:40.23 | wasup | nope |
| 01:41.28 | wasup | so any theories on why the exact same type card (same size,brand,everything) works and the other doesn't? |
| 01:42.31 | wasup | now for the special question.... how do i unlock the screen? |
| 02:14.55 | *** part/#htc-linux wasup (n=Tech@208.47.103.106) |
| 02:24.45 | *** join/#htc-linux Zoolooc_ (n=fredsiba@nrbg-4dbff7a0.pool.einsundeins.de) |
| 03:12.51 | *** join/#htc-linux MrYellow (i=radek@dynamic-78-8-97-119.ssp.dialog.net.pl) |
| 03:41.01 | *** join/#htc-linux webPragmatist (n=none@99-6-241-169.lightspeed.rcsntx.sbcglobal.net) |
| 03:51.35 | par | hi again San ;) |
| 04:00.16 | *** join/#htc-linux rob_w (n=bob@Mb67f.m.pppool.de) |
| 04:16.58 | *** join/#htc-linux Robwoerle (n=bob@Ma454.m.pppool.de) |
| 04:29.38 | *** join/#htc-linux MnemonicPunk (n=Atlantis@dslb-084-060-024-040.pools.arcor-ip.net) |
| 04:36.48 | MnemonicPunk | Good morning. Can anyone in here tell me how to boot from an image with a custom kernel? |
| 04:37.11 | tmzt_ | the "set KERNEL" line in startup.txt/default.txt |
| 04:38.53 | MnemonicPunk | Not the kernel is my problem but how to tell it to run the image instead of booting from a partition. =) |
| 04:39.28 | MnemonicPunk | (The disk image, not the kernel image.) |
| 04:39.51 | tmzt_ | what device, and what kernel were you using before? |
| 04:40.34 | MnemonicPunk | htc magician, current kernel (checked out and compiled yesterday) - no Linux before that |
| 04:41.16 | MnemonicPunk | It starts the boot process but can't mount the mmcblk0p2 so I hope to get it running with a disk image at least. |
| 04:41.19 | tmzt_ | ok, what kind of image are you trying to boot then |
| 04:41.56 | tmzt_ | the way it boots from an image is a script that is embedded in an initramfs that is in the kernel image |
| 04:42.11 | MnemonicPunk | openmoko from goxboxlives images but any other (GPE/Qtopia) would be fine for me as well |
| 04:42.48 | tmzt_ | you built the kernel yourself, not with oe or anything? |
| 04:43.19 | MnemonicPunk | Sorry to sound stupid but what is OE? ^^ |
| 04:43.27 | tmzt_ | openembedded |
| 04:43.35 | MnemonicPunk | Ah. |
| 04:43.53 | MnemonicPunk | No I crosscompiled on my Arch64 desktop Linux. |
| 04:44.56 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 04:45.39 | MnemonicPunk | I've got no problem with compiling the kernel now that I know how to do that, problem is I don't have the slightest idea why it can't mount the second partition (ext2) |
| 04:46.24 | tmzt_ | building the kernel is pretty easy compared to alot of other parts |
| 04:46.47 | tmzt_ | in this case, you need to build a linuxrc script and an initramfs |
| 04:48.08 | MnemonicPunk | How would I do that? |
| 04:49.07 | tmzt_ | git.openembedded.net, tree, packages, initrdscripts, |
| 04:49.43 | tmzt_ | org.openembedded.dev.git, tree |
| 04:51.07 | MnemonicPunk | You lost me there. o.o |
| 04:52.13 | MnemonicPunk | Ah found initrdscripts =) |
| 04:54.50 | MnemonicPunk | Okay, be back in a moment, gotta boot Linux |
| 04:57.55 | *** join/#htc-linux MnemonicPunk (n=atlantis@dslb-084-060-024-040.pools.arcor-ip.net) |
| 04:58.01 | MnemonicPunk | And back. |
| 04:59.51 | tmzt_ | ok, just put all those files in a folder, basically copy and paste the blob link into a file with the filename |
| 05:00.09 | tmzt_ | no, raw link, just save it |
| 05:00.19 | MnemonicPunk | all files from the initrd directory? |
| 05:00.33 | MnemonicPunk | er... initfs? |
| 05:01.00 | MnemonicPunk | Hold on a moment, gotta reopen that site (or is there a simple command line command to do that? ^^) |
| 05:01.18 | tmzt_ | files |
| 05:03.47 | MnemonicPunk | I'm fine if I do wget <raw link>? |
| 05:04.11 | tmzt_ | I don't know what name you'll get, maybe wget -O init.sh ... |
| 05:08.37 | MnemonicPunk | This is what I have now: 10-initfs.sh 80-loopboot.sh 85-blockboot.sh 90-check-modules.sh 30-bootmenu.sh 80-nfsboot.sh 87-kexecboot.sh init.sh |
| 05:08.47 | tmzt_ | you really need initfs bootmenu loopboot blockboot and init.sh |
| 05:08.59 | tmzt_ | but rename init.sh to init |
| 05:09.08 | MnemonicPunk | Don't keep the numbers in the names? |
| 05:09.08 | tmzt_ | make sure they are chmod +x * |
| 05:09.14 | tmzt_ | yes, keep the numbers |
| 05:09.30 | tmzt_ | ok, that's fine |
| 05:09.30 | MnemonicPunk | Okay. =) |
| 05:10.33 | tmzt_ | checkmodules looks like it just crashes things when modules aren't in the rootfs |
| 05:10.50 | tmzt_ | your drivers are probably builtin, so you don't need that |
| 05:11.53 | MnemonicPunk | I compiled my kernel with "y" for all the ext2, ext3 and fat modules. Can't tell you much about the rest though, didn't touch that so it is in the "default" state I guess. |
| 05:12.51 | tmzt_ | right, I'm just saying leave out checkmodules |
| 05:13.14 | MnemonicPunk | Just delete the file? |
| 05:13.18 | tmzt_ | yes |
| 05:13.29 | MnemonicPunk | okay. |
| 05:13.44 | tmzt_ | alright, run find . and see if it lists all the files |
| 05:14.32 | MnemonicPunk | ./10-initfs.sh ./init ./85-blockboot.sh ./30-bootmenu.sh ./80-nfsboot.sh ./80-loopboot.sh ./87-kexecboot.sh |
| 05:14.45 | tmzt_ | ok |
| 05:15.24 | tmzt_ | cd .. |
| 05:15.42 | tmzt_ | what is the directory you put those in called? |
| 05:15.57 | MnemonicPunk | ~/initfs |
| 05:17.34 | tmzt_ | (cd initfs; find | cpio -o -H newc |gzip -c ) > initfs.gz |
| 05:17.46 | MnemonicPunk | Woah. xD |
| 05:18.28 | tmzt_ | zcat initfs.gz |cpio -t |
| 05:19.08 | MnemonicPunk | Copy the output here? |
| 05:19.21 | tmzt_ | just tell me if you got the same filenames |
| 05:20.17 | MnemonicPunk | Yes, same files in the .gz as in the dir. =) |
| 05:20.25 | tmzt_ | good |
| 05:20.43 | tmzt_ | copy that to your card and add "set INITRD initfs.gz" to startup.txt/default.txt |
| 05:21.28 | MnemonicPunk | Okay, let me try that. |
| 05:21.39 | MnemonicPunk | Thanks for your help so far, hope it works. =) |
| 05:22.03 | tmzt_ | add "set CMDLINE root=/dev/loop looproot=/dev/mmcblk0p1 rootdelay=5" |
| 05:22.04 | *** join/#htc-linux LunohoD_ (n=alex@e180068185.adsl.alicedsl.de) |
| 05:22.35 | tmzt_ | add "set CMDLINE root=/dev/loop looproot=/dev/mmcblk0p1:image.img rootdelay=5" |
| 05:25.00 | tmzt_ | most important is probably rootdelay, so you can use a partition again |
| 05:25.42 | MnemonicPunk | I guess I have to change the image.img to a valid image on Partition 1? |
| 05:25.48 | tmzt_ | yes |
| 05:28.42 | MnemonicPunk | Can I dd if=mysecondpartition of=myimage.img? |
| 05:29.33 | tmzt_ | you can, but if you have rootdelay, you can just use root=/dev/mmcblk0p2 |
| 05:30.51 | tmzt_ | if you make an image, resize2fs it |
| 05:34.31 | MnemonicPunk | Alright, now you got me confused. ^^ The image.img has to be an image of my root filesystem, right? |
| 05:35.36 | tmzt_ | what I am saying is a possible reason why you could not use a partition before is rootdelay is what makes the kernel wait for the mmc card to be detected |
| 05:36.27 | MnemonicPunk | So it tried to mount a device that wasn't detected at that point. |
| 05:36.32 | tmzt_ | yes |
| 05:36.49 | MnemonicPunk | Okay, that fits the errors I get. |
| 05:37.09 | tmzt_ | the kernel also has a table of devices internally that can be used a root devices, and not all possible block devices are in it |
| 05:37.37 | tmzt_ | this initramfs mounts the device using the device file and then pivots to it |
| 05:38.15 | MnemonicPunk | Hm. |
| 05:38.38 | tmzt_ | I think I missed something |
| 05:38.38 | MnemonicPunk | As you put it, it should be able to mount the partition itself now? Or am I misunderstanding you? |
| 05:38.47 | tmzt_ | yes |
| 05:38.55 | MnemonicPunk | Ah okay. =) |
| 05:38.59 | MnemonicPunk | What did you miss? |
| 05:42.31 | tmzt_ | <PROTECTED> |
| 05:44.18 | MnemonicPunk | Okay, created the directory. Now what to put in it? |
| 05:46.05 | tmzt_ | mknod dev/console c 5 0 |
| 05:46.13 | tmzt_ | mknod dev/console c 5 1 |
| 05:47.09 | *** join/#htc-linux goxboxlive1 (n=goxboxli@195.159.97.196) |
| 05:47.34 | tmzt_ | (cd initfs; find | cpio -o -H newc |gzip -c ) > initfs.gz |
| 05:49.33 | MnemonicPunk | Are you sure that is right even when crosscompiling? It fills the /dev dir with my PCs devices. o.o |
| 05:50.02 | MnemonicPunk | Oh my mistake |
| 05:50.10 | tmzt_ | a linux (unix) device is just a special file, that is a file that has a major and minor number |
| 05:50.21 | MnemonicPunk | I entered ls /dev and not ls ./dev |
| 05:50.29 | tmzt_ | ls -lh dev |
| 05:51.02 | tmzt_ | will show c---- on the perms, which is character device |
| 05:51.07 | tmzt_ | and the major/minor numbers |
| 05:52.32 | MnemonicPunk | Alright. |
| 05:52.43 | MnemonicPunk | Can I testdrive that one now? Or still something to do? |
| 05:53.01 | tmzt_ | yes, just copy that over |
| 05:53.11 | tmzt_ | yes, just copy initfs.gz over |
| 05:55.46 | *** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net) |
| 05:56.05 | MnemonicPunk | It fails to execute /init =( |
| 05:56.18 | tmzt_ | add init=/init to CMDLINE |
| 05:56.45 | tmzt_ | do you get the "found initramfs" ? |
| 05:57.18 | MnemonicPunk | Already reset. xD I'll tell you on next try, just a moment. |
| 06:01.38 | MnemonicPunk | I'm using "set CMDLINE root=/dev/mmcblk0p2 rootdelay=5 init=/init" but it kernel panics. |
| 06:01.49 | tmzt_ | what happens? |
| 06:02.11 | tmzt_ | you have quotes around the root../init part? |
| 06:02.22 | MnemonicPunk | Nope. |
| 06:02.29 | MnemonicPunk | *adds* |
| 06:03.43 | MnemonicPunk | quotes around the initfs.gz, too? |
| 06:04.32 | tmzt_ | no |
| 06:04.44 | MnemonicPunk | Alright. |
| 06:04.45 | tmzt_ | just if there's more than one word |
| 06:05.17 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbff7a0.pool.einsundeins.de) |
| 06:06.31 | MnemonicPunk | "mmcblk0:(4) failed to execute /init" |
| 06:07.35 | MnemonicPunk | But just before it tries to run /init, it says "freeing init memory: 88k", is that related? |
| 06:07.52 | tmzt_ | it means the kernel is ready to jump to userspace |
| 06:08.12 | tmzt_ | ls -lh initfs, see if you have x's on the files |
| 06:08.47 | MnemonicPunk | I have. o.o |
| 06:09.40 | MnemonicPunk | -rwxr-xr-x for all of them |
| 06:10.27 | tmzt_ | we need busybox /bin/sh, that's why using oe for this makes sense! |
| 06:10.53 | tmzt_ | what root fs are you using? |
| 06:11.02 | MnemonicPunk | I would use it, I just don't really know what it is. xD |
| 06:11.43 | tmzt_ | that was to myself, oe is for building whole distributions like openmoko and angstrom |
| 06:11.44 | MnemonicPunk | You mean my PCs fs? or the first partitions fs on the sd? |
| 06:11.52 | tmzt_ | second partition |
| 06:11.57 | MnemonicPunk | ext2 |
| 06:12.10 | tmzt_ | I mean openmoko, or whatever |
| 06:12.23 | MnemonicPunk | openmoko, yes. |
| 06:12.26 | tmzt_ | we need a shell |
| 06:12.30 | tmzt_ | ok |
| 06:12.40 | MnemonicPunk | But I don't really care which one I use as long as I get one running. xD |
| 06:12.48 | tmzt_ | can you mount that on your pc? |
| 06:13.41 | MnemonicPunk | Did that. |
| 06:13.49 | *** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 06:14.09 | tmzt_ | you'll have to ask dcordes how to setup oe, and make a busybox unless openmoko has a small enough one |
| 06:14.45 | tmzt_ | ok, ls -lh /sd/bin/sh and see what it links to |
| 06:15.19 | MnemonicPunk | It links to busybox. |
| 06:15.36 | tmzt_ | ls -lh /sd/.../busybox and see what size it is |
| 06:15.51 | tmzt_ | then ldd /sd/.../busybox and see what it needs |
| 06:16.34 | tmzt_ | if you don't have ldd, use your cross ldd |
| 06:17.42 | MnemonicPunk | which dir is the ...? |
| 06:17.44 | *** join/#htc-linux kiozen (n=oeichler@p54929D48.dip0.t-ipconnect.de) |
| 06:18.02 | tmzt_ | wherever busybox is, /sd/bin /sd/, /sd/sbin/, whatever |
| 06:21.03 | MnemonicPunk | ldd says "not a dynamic executable". (The one in my path so it's x64 and not arm I guess) |
| 06:23.05 | tmzt_ | it's static then, so where good |
| 06:23.13 | tmzt_ | copy busybox to initfs/bin/ |
| 06:23.17 | tmzt_ | cd initfs/bin |
| 06:23.23 | tmzt_ | ln -sf busybox sh |
| 06:23.30 | tmzt_ | cd .. |
| 06:23.31 | tmzt_ | cd .. |
| 06:23.46 | tmzt_ | (cd initfs; find | cpio -o -H newc |gzip -c ) > initfs.gz |
| 06:24.25 | *** join/#htc-linux pleemans (n=toi@212-123-1-140.iFiber.telenet-ops.be) |
| 06:25.26 | tmzt_ | do you have CONFIG_AEABI=y and CONFIG_OABI_COMPAT=y in your .config |
| 06:26.32 | MnemonicPunk | Yes, both set to y |
| 06:26.50 | tmzt_ | great, were still not done |
| 06:27.01 | tmzt_ | in initfs/bin: |
| 06:27.08 | tmzt_ | ln -sf busybox mount |
| 06:27.10 | tmzt_ | ln -sf busybox modprobe |
| 06:27.25 | tmzt_ | not the last one |
| 06:28.14 | MnemonicPunk | Okay, did that. |
| 06:29.20 | tmzt_ | ln -sf busybox stty |
| 06:29.23 | MnemonicPunk | Wait... "not the last one" = "Don't do ln -sf busybox modprobe" ot "not the last one" = "More coming up"? xD |
| 06:29.52 | tmzt_ | modprobe should be in sbin/ |
| 06:29.57 | tmzt_ | cd sbin |
| 06:30.10 | tmzt_ | ln -sf ../bin/busybox modprobe |
| 06:30.14 | tmzt_ | cd .. |
| 06:30.34 | tmzt_ | cd bin |
| 06:30.39 | tmzt_ | ln -sf busybox mknod |
| 06:30.40 | tmzt_ | cd .. |
| 06:30.43 | tmzt_ | cd .. |
| 06:30.49 | tmzt_ | (cd initfs; find | cpio -o -H newc |gzip -c ) > initfs.gz |
| 06:31.42 | MnemonicPunk | Now it's grown from 20 blocks to 1382 blocks. |
| 06:31.59 | tmzt_ | yeah, there's actually something in there now |
| 06:32.06 | MnemonicPunk | Heh. ^^ |
| 06:32.35 | tmzt_ | I wonder if it wouldn't just make more sense to add rootdelay to the kernel source |
| 06:34.03 | MnemonicPunk | Hm. |
| 06:34.11 | tmzt_ | I hope I told you to copy busybox to initfs/bin and you aren't doing all this on the sd card |
| 06:34.26 | MnemonicPunk | No, don't worry. =) |
| 06:34.37 | MnemonicPunk | DId that in ~/initfs |
| 06:34.47 | tmzt_ | ok, try booting it |
| 06:34.53 | MnemonicPunk | Just a sec. |
| 06:37.13 | MnemonicPunk | WM2003 boot times are slooow... o.o |
| 06:37.46 | tmzt_ | that runs in ram doesn't it? |
| 06:38.13 | MnemonicPunk | Yep. |
| 06:38.35 | MnemonicPunk | Eats almost half the devices RAM. -.- |
| 06:38.37 | tmzt_ | is there a wm5 for that one? |
| 06:38.50 | MnemonicPunk | Anyway, boot messages didn't change at all. =( |
| 06:38.55 | MnemonicPunk | No, sadly not. |
| 06:39.02 | tmzt_ | failed to execute /init ? |
| 06:39.10 | MnemonicPunk | Yes. |
| 06:40.24 | tmzt_ | not syncing: no init found? |
| 06:40.36 | MnemonicPunk | Yes. |
| 06:41.31 | tmzt_ | try init=/bin/sh |
| 06:45.30 | MnemonicPunk | Still fails to execute. Same messages but with /bin/sh this time. |
| 06:46.01 | tmzt_ | where did you get the openmoko fs? |
| 06:46.49 | MnemonicPunk | from goxboxlives images (http://www.linuxtogo.org/~htcpxa/htcmagician/images/OpenMoko/) |
| 06:50.52 | MnemonicPunk | I managed to boot it with those two files, problem is that the kernel in the Openmoko-Magician-Linux.exe doesn't have my display driver included (magician apparently comes with 2 different displays) and didn't give me any graphics. |
| 06:51.18 | MnemonicPunk | The rest appeared to work though, since it responded to touchscreen taps with sound effects. |
| 06:52.35 | tmzt_ | try rdinit= instead of init= |
| 06:52.50 | MnemonicPunk | Alright, just a moment. |
| 06:53.50 | MnemonicPunk | set CMDLINE "root=/dev/mmcblk0p2 rootdelay=5 rdinit=/init" |
| 06:54.08 | MnemonicPunk | Like this? |
| 06:54.17 | tmzt_ | yes |
| 06:54.46 | tmzt_ | the failed to execute is coming from the kernel, though |
| 06:57.25 | MnemonicPunk | Maybe it can't find or mount the initfs.gz? |
| 06:57.32 | MnemonicPunk | (Same boot messages, sadly.) |
| 06:57.58 | MnemonicPunk | One more thing. |
| 06:58.27 | MnemonicPunk | When I used the openmoko with the missing driver, I let install in image form, not on second partition. |
| 07:00.24 | MnemonicPunk | Do you think I should just try that with this one as well? |
| 07:11.36 | MnemonicPunk | Anyway, thanks for all that patience and help so far. =) |
| 07:12.10 | tmzt_ | yeah, but we aren't getting that far |
| 07:12.30 | tmzt_ | the initfs.gz is loaded into ram by haret, just like the kernel |
| 07:12.44 | MnemonicPunk | Hm. |
| 07:12.54 | tmzt_ | if you get the found initramfs image, and then free init memory, it found the initfs |
| 07:12.57 | MnemonicPunk | So Haret would error when it can't find it? |
| 07:13.15 | tmzt_ | no, it just boots without it, but you don't get anywhere |
| 07:14.29 | tmzt_ | we can try one more thing, copy the /lib/ld-linux.so.? to initfs/lib/ |
| 07:14.47 | tmzt_ | if you can, run ldd from your cross toolchain on busybox |
| 07:15.38 | MnemonicPunk | Where is the .so file? o.o |
| 07:15.53 | tmzt_ | under the openmoko fs, lib/ |
| 07:16.03 | tmzt_ | second partition |
| 07:17.08 | tmzt_ | go ahead and copy glibc, uclibc, zlib/libz and whatever else under lib to initfs/lib/ |
| 07:17.15 | tmzt_ | not modules |
| 07:17.29 | MnemonicPunk | I just replaced that by an image of the partition. >_> Let me try that (it worked before, maybe it does now), if that doesn't I'll restore the 2 partition layout. |
| 07:17.38 | tmzt_ | either way |
| 07:18.03 | tmzt_ | mkdir mnt; mount -o loop,ro image.img mnt |
| 07:22.08 | MnemonicPunk | What the... |
| 07:22.27 | MnemonicPunk | I seriously killed something there. |
| 07:22.29 | MnemonicPunk | o.o |
| 07:22.55 | MnemonicPunk | And why can't my arch find any /dev/loop? I used that just yesterday. |
| 07:23.33 | tmzt_ | your arch? |
| 07:23.53 | MnemonicPunk | Arch Linux. ^^ |
| 07:24.08 | MnemonicPunk | Ubuntu does it, I know that. But that is so slow. >_> |
| 07:24.35 | tmzt_ | /dev/loop just tells that initramfs to use loop file |
| 07:24.46 | tmzt_ | modprobe loop |
| 07:24.53 | MnemonicPunk | Ah right. |
| 07:25.08 | MnemonicPunk | Did that yesterday too, just forgot I did. xD |
| 07:31.36 | *** join/#htc-linux LunohoD (n=alex@e180068185.adsl.alicedsl.de) |
| 07:31.36 | *** join/#htc-linux Kevin2 (n=Kevin2@207-237-52-122.c3-0.avec-ubr12.nyr-avec.ny.cable.rcn.com) |
| 07:31.36 | *** join/#htc-linux ptl (n=patola@unaffiliated/ptl) |
| 07:32.20 | tmzt_ | I need to go, check that busybox with the cross ldd /opt/arm/*/bin/arm-none-eabi-ldd (or whatever) and see if its static |
| 07:32.47 | MnemonicPunk | Okay, I'll try. =) Thanks again. |
| 07:32.50 | tmzt_ | is /bin/sh is as symlink to busybox, it must have a sh applet |
| 07:33.23 | tmzt_ | if it is dynamic, make sure all the libs it needs are on the initfs/lib |
| 07:33.40 | tmzt_ | run ldconfig initfs/lib |
| 07:33.51 | tmzt_ | rebuild and try booting |
| 07:34.29 | tmzt_ | ask dcordes later (a few hours) about OE and building a new busybox, or getting the lcd patches into handhelds kernel in OE |
| 07:34.37 | tmzt_ | so you can build an openmoko image |
| 07:34.37 | MnemonicPunk | As soon as I mount the SD it appears empty. When I unmount it, I see the files on it. wtf? |
| 07:35.04 | tmzt_ | you extracted to your rootfs, so its empty, usually |
| 07:35.46 | tmzt_ | you mean the second partition |
| 07:35.53 | tmzt_ | ? |
| 07:36.04 | MnemonicPunk | Nope. The first. |
| 07:36.26 | MnemonicPunk | I think that was just my mistake though ^^ |
| 07:37.39 | tmzt_ | I wish this would work, but I've tried building these kernels before and switching to an sd card, and have gotten nowhere |
| 07:38.09 | MnemonicPunk | So it's normal for SD cards to cause problems? |
| 07:39.02 | tmzt_ | it's a bunch of things, sd being part of it |
| 07:40.13 | tmzt_ | if you look at debian or ubuntu, which use initramfs now, its similar problems, they work until you change things |
| 07:41.16 | MnemonicPunk | Never looked at those parts of Ubuntu. ^^ I do enjoy using Linux but I'm happy when I don't have to deal with getting the kernel to work. *g* |
| 07:41.32 | tmzt_ | since you said this same image worked before, it can't be the cpu not buing compatible |
| 07:41.55 | MnemonicPunk | Yes, it totally worked before (except the display of course) |
| 07:42.07 | tmzt_ | you got the openmoko .tar.bz from that url you posted? |
| 07:42.12 | MnemonicPunk | Yes. |
| 07:42.13 | *** join/#htc-linux kiozen (n=oeichler@p54929D48.dip0.t-ipconnect.de) |
| 07:42.31 | tmzt_ | can you run file initfs/bin/busybox ? |
| 07:42.40 | MnemonicPunk | If you put it on an SD on only one partition, it'll install to an image. |
| 07:43.35 | MnemonicPunk | Er... o.o |
| 07:43.48 | tmzt_ | #file initfs/bin/busybox |
| 07:44.00 | MnemonicPunk | How do I run that? Can't get to any shell or anything on the device. =( |
| 07:44.06 | tmzt_ | on the pc |
| 07:44.14 | *** join/#htc-linux nebi (n=nebi@c-498fe253.02-145-7570701.cust.bredbandsbolaget.se) |
| 07:44.46 | MnemonicPunk | bash: ./busybox: cannot execute binary file |
| 07:46.51 | tmzt_ | file is the command, it should be in arch |
| 07:46.51 | *** join/#htc-linux ltxda00 (n=ltxda@c-98-201-10-242.hsd1.tx.comcast.net) |
| 07:46.51 | tmzt_ | this initrdscripts is confusing, those .sh need to be in initfs/initrd.d/ |
| 07:47.13 | tmzt_ | cd ~; file initfs/bin/busybox |
| 07:47.59 | MnemonicPunk | busybox: ELF 32-bit LSB executable, ARM, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.14, stripped |
| 07:48.20 | tmzt_ | where is your cross compiler installed? |
| 07:49.09 | MnemonicPunk | In ~/arm-2007q1 |
| 07:49.49 | tmzt_ | ~/arm-2007q1/bin/arm-none-eabi-ldd initfs/bin/busybox |
| 07:50.35 | MnemonicPunk | it doesn't find any file named arm-none-eabi-ldd |
| 07:50.52 | MnemonicPunk | And that's true, the file doesn't exist. The prefix is correct, though. |
| 07:52.06 | MnemonicPunk | I used the cross compiler linked on this wikipage (http://www.handhelds.org/moin/moin.cgi/UniversalKernel). Maybe that one is outdated or something? |
| 07:53.11 | tmzt_ | I don't have it either, should be in binutils |
| 07:53.34 | tmzt_ | strings initfs/bin/busybox |grep so |
| 07:54.54 | MnemonicPunk | There's a package that might contain it: community/cross-arm-elf-binutils 2.17-1 |
| 07:55.03 | tmzt_ | ok |
| 07:55.15 | MnemonicPunk | Should I try installing that or does that clash with my arm-2007q1? |
| 07:55.35 | tmzt_ | it shouldn't, that's specific to codesourcery |
| 07:56.06 | tmzt_ | it will probably be in /usr/bin/arm-*-gcc or whatever |
| 07:57.00 | MnemonicPunk | what the... o.o |
| 07:57.39 | MnemonicPunk | When trying to install that package, it reports "not found" for most mirrors, then downloads it from one and says the package was corrupt. o.o |
| 07:58.01 | tmzt_ | do you have to update something? |
| 07:58.14 | MnemonicPunk | Maybe. |
| 07:58.25 | MnemonicPunk | Let me pacman -Syu and then try again. |
| 08:00.06 | MnemonicPunk | Okay, there were some updates, some related to autoconf and libtool. |
| 08:02.48 | MnemonicPunk | libdrm o.O Is that what it seems to be? |
| 08:03.13 | tmzt_ | related to video, direct rendering manager |
| 08:03.21 | MnemonicPunk | Ah lol. |
| 08:03.35 | MnemonicPunk | Thought DRM in the M$ meaning. xD |
| 08:05.07 | tmzt_ | you basically need to copy uclibc from openmoko image /lib to initfs/lib |
| 08:05.53 | MnemonicPunk | Alright, I'll try. |
| 08:07.38 | *** join/#htc-linux miknix (n=miknix@81.193.83.113) |
| 08:10.14 | MnemonicPunk | Can't find it in the image (tried /bin, there's lots of libs but no uclibc) |
| 08:10.32 | tmzt_ | what is in /lib, glibc? |
| 08:10.46 | MnemonicPunk | Er... lib I mean |
| 08:10.48 | tmzt_ | also, strings initfs/bin/busybox |grep so |
| 08:11.19 | *** join/#htc-linux rwhitby (n=rwhitby@nslu2-linux/rwhitby) |
| 08:11.24 | MnemonicPunk | I can see a libc-2.5.so |
| 08:13.31 | tmzt_ | ok, copy that and libc.so.6 and ld-linux.so.3 and ld-2.5.so to initfs/lib |
| 08:13.47 | tmzt_ | those numbers might be different |
| 08:17.06 | MnemonicPunk | All found. =) |
| 08:18.05 | tmzt_ | (cd initfs; find | cpio -o -H newc |gzip -c ) > initfs.gz |
| 08:18.28 | tmzt_ | try booting that, rdinit=/init is fine |
| 08:26.24 | MnemonicPunk | Hm. |
| 08:26.46 | MnemonicPunk | For some reason it still fails to run. =( |
| 08:32.53 | MnemonicPunk | (The errors are the same as before. ) |
| 08:34.45 | tmzt_ | ask dcordes about OE later, I will be back in a few hours |
| 08:34.55 | MnemonicPunk | Okay. ^^ |
| 08:35.24 | MnemonicPunk | I'll tell ya if I can get it to work somehow, although I doubt I can do it on my own. ^^ |
| 08:35.38 | MnemonicPunk | Thanks for the help. =) |
| 08:51.14 | *** join/#htc-linux miknix (n=miknix@81.193.83.113) |
| 09:16.59 | *** join/#htc-linux lpotter (n=ljp@CPE-124-191-144-181.vic.bigpond.net.au) |
| 09:47.37 | *** join/#htc-linux pleemans (n=toi@212-123-1-140.iFiber.telenet-ops.be) |
| 10:12.23 | *** join/#htc-linux LunohoD_ (n=alex@e180075167.adsl.alicedsl.de) |
| 10:43.04 | *** join/#htc-linux cousin_luigi (n=luigi@unaffiliated/cousinluigi/x-395723) |
| 10:54.45 | cousin_luigi | hello |
| 10:56.20 | cousin_luigi | I'm not sure this is the right place to ask, but is it possible to sync a P3450 using stock fw with linux? |
| 11:02.49 | MnemonicPunk | Okay, I've got a new problem... my initfs is mounted correctly but it doesn't have the correct path. Problem is: It tells me that it can't find basic commands like echo or mkdir |
| 11:03.30 | MnemonicPunk | Any ideas what causes this? |
| 11:15.23 | dcordes | cousin_luigi: there is "synce" |
| 11:16.46 | cousin_luigi | heard about that |
| 11:17.53 | dcordes | I never had much success with it personally though |
| 11:18.36 | MnemonicPunk | Never worked for me, sadly. =( |
| 11:19.09 | MnemonicPunk | dcordes: Do you have any ideas for my problem? |
| 11:19.14 | dcordes | MnemonicPunk: sounds like something is wrong with your initramfs |
| 11:19.46 | MnemonicPunk | Yeh, thought so. But no idea what. What could cause init not to find basic shell commands? |
| 11:20.33 | dcordes | what's the error you get? |
| 11:20.52 | MnemonicPunk | for example "echo: command not found " |
| 11:25.47 | MnemonicPunk | Or "mkdir: command not found" =( |
| 11:26.10 | dcordes | what's that initrd? did you build it yourself? |
| 11:27.09 | MnemonicPunk | More or less. |
| 11:27.26 | MnemonicPunk | tmzt_ guided me through it up to this point. |
| 11:27.40 | dcordes | I will look in the log.. |
| 11:28.07 | MnemonicPunk | Okay. =) |
| 11:28.36 | MnemonicPunk | One thing about the logs: I made a mistake mid-way and copied the libs to a file /lib instead to a directory. |
| 11:28.43 | MnemonicPunk | I fixed that myself now, though. |
| 11:31.55 | dcordes | hm I never built an initrd manually if you don't count the several angstrom and android ones which are only compressed rootfilesystems |
| 11:32.28 | dcordes | so you are doing this because the kernel doesn't like to boot from the SD directly? |
| 11:32.43 | MnemonicPunk | Yep. |
| 11:32.54 | MnemonicPunk | I'm not even sure why it won't do that. |
| 11:33.10 | dcordes | maybe I would stay there first |
| 11:33.30 | dcordes | do you actually know it works? |
| 11:33.40 | MnemonicPunk | It does with the HH18 kernel. |
| 11:33.54 | MnemonicPunk | I can't use that one though because it doesn't recognize my display. |
| 11:34.12 | dcordes | k building the kernel manually is better anyway |
| 11:34.16 | MnemonicPunk | Plays sound and reacts to touchscreen input but only gives me a white screen. |
| 11:34.28 | dcordes | so does your kernel .config have ext2 support? |
| 11:35.02 | MnemonicPunk | I checked it in "make xconfig" and the vars in .config are set to y. |
| 11:35.20 | MnemonicPunk | Any other stuff I have to make sure for ext2 support? |
| 11:36.06 | dcordes | I usually use menuconfig because I find that easier to use. but I only tick the ext2 bock. maybe you built it as a module? |
| 11:36.13 | dcordes | s/bock/box/ |
| 11:37.27 | MnemonicPunk | How do I decide whether it will be built as module or not? |
| 11:39.09 | dcordes | in menu config it says M for module and * for built in |
| 11:39.51 | dcordes | click the checkbox again in xconfig and it will switch modes I think (space in menuconfig) |
| 11:40.40 | MnemonicPunk | Alright. let me doublecheck that. |
| 11:41.29 | dcordes | you can disable module support in the main kconfig menu |
| 11:41.39 | dcordes | then everything is built in |
| 11:42.23 | MnemonicPunk | dcordes: You were ger, right? |
| 11:42.34 | dcordes | ja |
| 11:42.41 | MnemonicPunk | Gut zu wissen. =) |
| 11:45.22 | MnemonicPunk | Also wenn ich in menuconfig den "Loadable Module Support" ganz abschalte compiliert der das alles mit ein? |
| 11:45.51 | dcordes | I think so |
| 11:46.02 | dcordes | just check manually for the ext2 |
| 11:46.27 | MnemonicPunk | That's enabled by default if I deploy magician_defconfig first. =) |
| 11:51.00 | dcordes | ok so that can't be the problem. on the other end, are you sure the filesystem you have is really ext2? not ext3 or something else? |
| 11:52.03 | MnemonicPunk | I checked both ext2 and ext3 this time and am compiling right now. |
| 11:52.13 | MnemonicPunk | But I tried both and they didn't work as of yet. |
| 11:57.27 | dcordes | ok. do you specify the right CMDLINE? can you paste it? |
| 11:58.36 | MnemonicPunk | I'll try the new kernel for the first time now. What I meant was I tried that yesterday and it didn't work. |
| 11:59.18 | *** join/#htc-linux MRFriend (i=prettacs@82-35-152-46.cable.ubr05.enfi.blueyonder.co.uk) |
| 11:59.27 | MnemonicPunk | But just to be safe, I want to use "root=/dev/mmcblk0p2" as root partition. |
| 12:00.32 | MnemonicPunk | That is not mistyped or wrong in another way I hope? o.o |
| 12:01.09 | dcordes | and the second partition of the sd card has an ext2 filesystem? |
| 12:01.49 | MnemonicPunk | Yes it has. (Just doublechecked) |
| 12:03.26 | dcordes | ok you can try to mount the card in your initramfs? |
| 12:03.46 | dcordes | the actual block you specify in root= |
| 12:03.59 | *** join/#htc-linux sikkle (n=sikkle@MTLXPQAK-1176053010.sdsl.bell.ca) |
| 12:04.11 | MnemonicPunk | Let me first try to simply boot it w/o initramfs and look for errors. |
| 12:06.58 | MnemonicPunk | My whole CMDLINE is 'set CMDLINE "root=/dev/mmcblk0p2 rootdelay=5 init=/init"', any mistakes in there so far? |
| 12:11.13 | MnemonicPunk | Alright, my current error is: VFS: Cannot open root device "mmcblk0p2" or unknown-block(254,2) |
| 12:12.49 | cr2 | MnemonicPunk: you have a magician ? |
| 12:13.02 | MnemonicPunk | Yes. |
| 12:14.04 | *** part/#htc-linux cousin_luigi (n=luigi@unaffiliated/cousinluigi/x-395723) |
| 12:15.22 | MnemonicPunk | Why, is there something special to take care of for the magician? |
| 12:17.18 | dcordes | MnemonicPunk: raise the rootdelay. try 10 or 15 seconds |
| 12:17.54 | MnemonicPunk | Okay. |
| 12:18.30 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 12:21.15 | MnemonicPunk | No change. =( |
| 12:21.37 | MnemonicPunk | I'll try recompiling the kernel with all SD drivers I can find. xD |
| 12:23.44 | dcordes | good point, we didn't even check if you build the SD.. but defconfig should include it Iguess |
| 12:24.19 | MnemonicPunk | I think it does but there are several device drivers for that and some are ticked by default, some are not. |
| 12:24.25 | cr2 | MnemonicPunk: pxamci should be enough |
| 12:24.54 | cr2 | MnemonicPunk: does the dmesg say something about the partition table on your card ? |
| 12:25.18 | MnemonicPunk | No, it apparently doesn't even find the location. |
| 12:25.26 | MnemonicPunk | That's why I think drivers. |
| 12:28.58 | *** join/#htc-linux infernix (i=nix@unaffiliated/infernix) |
| 12:39.45 | cr2 | dcordes: the button panel on raph100 is touch-sensitive too (not only the LCD). i've traced such an irq yesterday. |
| 12:55.02 | dcordes | pendown irq? which #? |
| 12:58.06 | cr2 | dcordes: not for the main ts |
| 12:59.17 | cr2 | didn't see any pendown irq. at least not in the first gpio banks |
| 13:13.32 | *** join/#htc-linux infernixx (i=nix@unaffiliated/infernix) |
| 13:16.11 | *** join/#htc-linux Leonhardt (n=Misio@194.193.58.253) |
| 13:17.03 | MnemonicPunk | I need to talk to goxboxlive I guess. |
| 13:17.36 | dcordes | MnemonicPunk: youc an also try a kernel you know works to sort it's nod your card |
| 13:17.52 | MnemonicPunk | That's why I have to talk to him. ^^ |
| 13:18.19 | MnemonicPunk | His hh18 image does work except for the screendriver (which was fixed later). |
| 13:41.41 | *** join/#htc-linux ltxda (n=ltxda@c-98-201-10-242.hsd1.tx.comcast.net) |
| 13:45.42 | MnemonicPunk | Could my card be not working because one of the partitions is >1 GB? |
| 13:46.04 | MnemonicPunk | (Just saw that some cards don't work, especially >1 GB) |
| 13:46.40 | *** join/#htc-linux ltxda (n=ltxda@c-98-201-10-242.hsd1.tx.comcast.net) |
| 13:54.27 | *** join/#htc-linux SmallR2003 (n=SmallR20@79-77-79-186.dynamic.dsl.as9105.com) |
| 13:55.33 | SmallR2003 | ok, you guys will be interested in the G1/Dream enough to talk to me :p |
| 13:55.52 | dcordes | you have one? |
| 13:56.24 | marajin | No! ve shall znub the Dream, ve do not belief in ze dream! |
| 13:56.32 | marajin | :p |
| 13:56.35 | SmallR2003 | i haven't |
| 13:56.36 | marajin | yo dcordes |
| 13:56.39 | SmallR2003 | and i was thinking about it |
| 13:56.53 | dcordes | marajin: hi there |
| 13:56.55 | SmallR2003 | at first it sounded like a great idea |
| 13:56.57 | marajin | it looks vaguely ok but we don't know huge amounts about it yet beyond well, hardware specs and stuff |
| 13:57.19 | SmallR2003 | i mean, looks functional, the os is basically linux, should be nice to play with and even work with |
| 13:57.27 | SmallR2003 | then i started thinking about the google aspect of all this |
| 13:57.28 | marajin | assuming it isn't locked down yeah |
| 13:57.43 | SmallR2003 | well i've read that it wont be locked down |
| 13:57.57 | dcordes | the question is, what is locked down? |
| 13:58.16 | SmallR2003 | google are starting to worry me, they have all my emails, they track my searching and my response to advertisements... |
| 13:58.28 | SmallR2003 | do i want them knowing which apps i have, what my phone number is, and where i am? |
| 13:58.54 | dcordes | and when you look for a pizza place :( |
| 13:59.11 | marajin | dude you think they don't already know ? :p |
| 13:59.14 | marajin | you think MS don't? |
| 13:59.34 | SmallR2003 | ms are likely to have a lot of data on me, agreed |
| 13:59.51 | SmallR2003 | but i think google might have a little trouble coordinating things |
| 14:00.15 | SmallR2003 | ms have *raw* access to my phone, google just send me emails :p |
| 14:00.23 | SmallR2003 | anyway, so, i was thinking: |
| 14:00.33 | SmallR2003 | a) do any of you have concerns about this? |
| 14:01.19 | SmallR2003 | b) do you feel that porting some of their apps to another mobile linux distro, and the said mobile linux distro to the phone would be beneficial in the long run |
| 14:01.25 | kiozen | there is no such thing like a free lunch. You give them information, they give service to you. If you don't like it don't do it |
| 14:01.40 | SmallR2003 | c) do you feel that the privacy concern is smaller than the gain of linux? |
| 14:02.05 | kiozen | privacy is a strong reason not to do it |
| 14:02.07 | SmallR2003 | kiozen, my issue is that you sign in with your google account, it syncs *everything* to that account |
| 14:02.33 | kiozen | I don't have a google account :) |
| 14:02.49 | SmallR2003 | this would cause problems for using the phone, methinks |
| 14:02.50 | kiozen | guess why? |
| 14:03.18 | SmallR2003 | right, the trade deal :p |
| 14:04.02 | kiozen | but if people don't care to turn their inside out on TV why should they bother on google |
| 14:04.30 | SmallR2003 | hmmm |
| 14:04.56 | SmallR2003 | so would you say the gain of linux is better than the loss of privacy? |
| 14:05.12 | SmallR2003 | to be fair to google, you can never lose a contact or appointment again |
| 14:05.34 | kiozen | sure tell your neighbour, they won't forget too |
| 14:06.09 | kiozen | I don't think that technical benefit and privacy are linked in this case |
| 14:06.50 | kiozen | you can analyse their code, but you don't have to use one of their devices |
| 14:07.33 | SmallR2003 | i'd totally say that there's a massive benefit from google throwing this code at *us* |
| 14:07.55 | SmallR2003 | especially as i expect there will be home brews of it soon enough |
| 14:08.40 | kiozen | yes, that is what's happening, however the nice parts like DRM and GSM control are not public |
| 14:08.52 | kiozen | for obvious reasons :) |
| 14:09.46 | dcordes | kiozen: on the qualcomm it's more than just gsm. the arm9 has not only the radio but also GPS, camera, ... |
| 14:10.11 | SmallR2003 | humph :p |
| 14:10.16 | dcordes | SmallR2003: that's something one should also bear in mind when it comes to privacy and personal security |
| 14:10.30 | SmallR2003 | right |
| 14:10.48 | dcordes | and on the android phones, guess who has their custom firmware on the arm9.... |
| 14:10.49 | SmallR2003 | i know google maps gets my position within a few streets without gps |
| 14:11.21 | SmallR2003 | so gps would be just the next step |
| 14:11.27 | kiozen | and usually you agree that it belongs all to google :P |
| 14:12.14 | dcordes | who knows which contractors got their hands on G1 AMSS .. |
| 14:13.06 | SmallR2003 | but then again, all this stuff is accessible to ms |
| 14:13.11 | SmallR2003 | who's the greater evil? |
| 14:13.21 | kiozen | but most users won't care, this always reminds me at "Minority Report" |
| 14:14.55 | *** join/#htc-linux johndrinkwater (n=beta@nextra.plus.com) |
| 14:15.43 | kiozen | SmallR2003: both, just tell your kids: "Do not give away privat data lighthearted" |
| 14:16.09 | kiozen | it's a bit like "Do not talk with the stranger on your way to school" |
| 14:16.32 | SmallR2003 | hehe |
| 14:16.53 | kiozen | if you give away your private live to anyone (MS, Google,...) you make a fault. |
| 14:17.05 | *** join/#htc-linux TeringTu1y (n=maarten@195-241-125-243.ip.telfort.nl) |
| 14:20.03 | SmallR2003 | hmmm |
| 14:20.11 | SmallR2003 | i guess i'm just paranoid |
| 14:20.14 | SmallR2003 | btw, hi johndrinkwater |
| 14:20.28 | johndrinkwater | lo |
| 14:20.34 | SmallR2003 | the worst thing i do is avoid writing tax forms till the last minute |
| 14:22.25 | kiozen | SmallR2003: let data specialist decide in future if you do bad or good things, they will tell you <evil grin> |
| 14:22.41 | kiozen | ok, have to collect my son, bbl |
| 14:25.00 | SmallR2003 | so, johndrinkwater, you came in just after the main conversation tailed off :p |
| 14:25.06 | SmallR2003 | johndrinkwater, what are your exact concerns |
| 14:25.29 | dcordes | SmallR2003: I had a similar discussion with a friend who is using the fic freerunner phone with openmoko software. we were deliberating about the privacy benefits of the free software (and hardware) on the moko phone. |
| 14:26.48 | dcordes | SmallR2003: if it comes to the question, who knows who did I contact from where, at least in germany, the cellular provider will know either way |
| 14:26.59 | dcordes | no matter what software you run on the phone |
| 14:27.46 | dcordes | the legislator decided to make the operators store that data for a certain period of time (for law enforcement etc) |
| 14:28.30 | dcordes | from that position one could say free software on the phone is pointless, so one could install android or whatever without getting any privacy drawbacks |
| 14:28.59 | SmallR2003 | hmmm |
| 14:29.11 | SmallR2003 | so, calling is not an issue either way |
| 14:29.23 | SmallR2003 | but what about calendar stuff and anything else google is sending back to base? |
| 14:29.41 | SmallR2003 | i hear their new browser tells them more about your hair style than bugs |
| 14:30.17 | dcordes | that's the point. on the one hand the government makes people save your calling information anyway, I bet also in your country, but what runs on the phone, you can decide on your own. |
| 14:30.50 | dcordes | thereby who gets those informartion like calendar, web usage etc etc |
| 14:31.12 | dcordes | positioning |
| 14:31.56 | *** join/#htc-linux ltxda (n=ltxda@c-98-201-10-242.hsd1.tx.comcast.net) |
| 14:32.51 | dcordes | nobody can contest the platform is huge eyecandy and highly reactive and usable, also the nice services everybody likes to use in their everydaylife. but what about the things beyond that? interesting point you have there.. |
| 14:33.20 | SmallR2003 | see, i was looking at it and thinking 'wow, this looks amazing, i want one' |
| 14:33.47 | SmallR2003 | then i read the bit about syncing data and thought 'hmmm... good idea... but what if i don't want to sync it?' |
| 14:33.51 | SmallR2003 | i mean, it's a great idea really |
| 14:33.58 | SmallR2003 | syncing with a popular central service |
| 14:34.29 | dcordes | hm what kiozen said, you won't get the lunch for free :) |
| 14:35.53 | *** join/#htc-linux EA2 (n=chipper@cpe-024-074-138-191.carolina.res.rr.com) |
| 14:38.37 | dcordes | bbl |
| 14:51.29 | SmallR2003 | johndrinkwater, did you work out how many possible combinations the screen lock thing has? |
| 14:52.31 | *** join/#htc-linux timebomb_ (n=timebomb@p5B10C255.dip.t-dialin.net) |
| 14:56.03 | *** join/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 15:10.33 | *** join/#htc-linux chab7 (n=kvirc@212.92.4.114) |
| 15:10.37 | *** join/#htc-linux kiozen (n=oeichler@rgnb-5d87c6db.pool.einsundeins.de) |
| 15:15.38 | Raketemensch_ | FINE |
| 15:15.41 | Raketemensch_ | drinks water |
| 15:16.32 | MnemonicPunk | goxboxlive, talk to me. >.< |
| 15:17.21 | MnemonicPunk | I need someone to tell me how to mount my SD. >.< |
| 15:18.53 | MnemonicPunk | And by "mount" I mean "make my Magician recognize it so I finally have a root partition to boot from"! xD |
| 15:21.28 | MnemonicPunk | Or can someone explain to me how to transform an ext2 partition into an image I can boot from? |
| 15:22.51 | johndrinkwater | SmallR2003, it wasnât so much the amount of combinations, just that if an easy way was found to bypass it, all my google account (cc, contacts, mail) is linked to the device. wanted to know how to kill access (and phone data) remotely |
| 15:23.12 | SmallR2003 | /nod |
| 15:23.18 | SmallR2003 | that's a good point |
| 15:23.31 | SmallR2003 | they should allow locking of the phone from the google account |
| 15:24.26 | johndrinkwater | -locking +wiping |
| 15:24.42 | MnemonicPunk | Nuking! |
| 15:36.29 | marajin | SmallR2003: It's a simple mathmatic sum to calculate maximum variations |
| 15:36.30 | *** join/#htc-linux GPFerror (n=gpferror@cpe-76-187-41-132.tx.res.rr.com) |
| 15:40.33 | *** join/#htc-linux chab7 (n=kvirc@212.92.4.114) |
| 15:47.14 | cr2 | SmallR2003: unless they will force you to activate G1 at the buytime, you can always format the NAND |
| 15:48.01 | MnemonicPunk | Honestly guys, do you have ANY ideas why the SD can't be found? |
| 15:48.14 | cr2 | SmallR2003: i agree with dcordes, the microkernel is a much more dangerous piece of software than the A11 stuff |
| 15:49.56 | cr2 | kiozen: ready for mmutrace ? |
| 15:50.23 | kiozen | cr2: later :) |
| 15:50.27 | cr2 | MnemonicPunk: it's some problem with pxamci init. looks strange though |
| 15:50.30 | cr2 | kiozen: ok. |
| 15:50.59 | dcordes | MnemonicPunk: did you try mounting in your initrd? |
| 15:51.03 | kiozen | cr2: http://pastebin.no/38166 |
| 15:51.16 | MnemonicPunk | cr2: Honestly? Or do you just want me to shut up? xD |
| 15:51.32 | kiozen | cr2: it do recording of extended information per trackpoint |
| 15:51.40 | cr2 | dcordes: the message about card detection should come first, and the partition table message |
| 15:51.53 | kiozen | cr2: any value you miss there? |
| 15:52.05 | cr2 | kiozen: float ? |
| 15:52.11 | kiozen | arm is float |
| 15:52.14 | MnemonicPunk | If you tell me how to do that I can give you any logs you need. xD |
| 15:52.44 | cr2 | kiozen: maybe u_int32_t is better ? |
| 15:53.01 | kiozen | cr2: and I don't think that any of these values challenge a float |
| 15:53.36 | kiozen | hm velocity is no integer |
| 15:54.05 | cr2 | MnemonicPunk: what comes before the VFS: Cannot open root device "mmcblk0p2" or unknown-block(254,2) |
| 15:54.30 | kiozen | cr2: I would like to know if we do a sirf binary if there is more information we would like to record |
| 15:54.49 | cr2 | kiozen: magnetic variation comes from a model, unless you really have a compass |
| 15:55.01 | MnemonicPunk | Can hardly read it on the screen. xD Gimme a sec. |
| 15:55.17 | cr2 | kiozen: sirf binary provides XYZ as int32_t |
| 15:55.55 | kiozen | cr2: right in most receivers you won't get a sane value for magnetic, never the less I need the placeholder here |
| 15:56.12 | cr2 | kiozen: grmn *dop is a bit magic for me. |
| 15:56.12 | kiozen | cr2: right what unit is XYZ? |
| 15:56.21 | cr2 | meters |
| 15:56.34 | kiozen | cr2: ok I add XYZ |
| 15:57.08 | kiozen | cr2: they convert the unitless dop int something with meter |
| 15:57.15 | cr2 | 3D cartesian coordinates from the WGS84 earth center and usual lat/long convention for the axes |
| 15:58.05 | cr2 | kiozen: i thought it was a geometric DOP (documented even in gpsd sources), which you can calculate from the ephemeris |
| 15:58.12 | cr2 | kiozen: but it isn't |
| 15:58.24 | kiozen | ok |
| 15:58.34 | kiozen | cr2: anything else to record? |
| 15:58.43 | cr2 | kiozen: i mostly have 2m since i use 76cx with an external antenna on the car roof |
| 15:59.28 | kiozen | so you can remote control your car with OSM data :P |
| 16:00.23 | cr2 | xyz, and v_x, v_y, v_z are probabably the basic measured data. |
| 16:00.50 | kiozen | cr2: v_x ? velocity? |
| 16:01.03 | cr2 | yes |
| 16:01.22 | cr2 | x,y,z,dx/dt,dy/dt,dz/dt |
| 16:01.24 | MnemonicPunk | "in te(?)pio _get: shift=1, bit=04, pos=5, addrbase=0xc480000(6)mmcblk0: mmc0: b368 1997312KiB (ro)" |
| 16:01.31 | MnemonicPunk | That's what I could read. xD |
| 16:01.48 | MnemonicPunk | The (?) means I'm not sure about the two letters before it. |
| 16:01.50 | cr2 | MnemonicPunk: ok, so the driver works, and the card is detected |
| 16:02.49 | cr2 | MnemonicPunk: either your partition table is broken, or card has badblocks, or your root filesystem is not ext2, or you forgot to add the ext2 fs driver |
| 16:03.08 | kiozen | cr2: is that a float or int32? |
| 16:03.26 | cr2 | kiozen: int32_t for sirf binary |
| 16:03.30 | kiozen | ok |
| 16:03.55 | cr2 | kiozen: sirf2 must have the carrier phase too, but it's compless stuff. |
| 16:03.58 | MnemonicPunk | The card can be read and written to on my desktop (Arch64 and Ubuntu 8.10) and I doublechecked the driver. |
| 16:04.10 | MnemonicPunk | How can I check the partition tables? |
| 16:04.11 | kiozen | cr2: hm, let's make it a float for portability sake... |
| 16:04.12 | cr2 | kiozen: then you'd save the per SV/PRN data |
| 16:04.36 | kiozen | you want that per point? |
| 16:04.54 | cr2 | not really |
| 16:05.05 | cr2 | at the current state of things |
| 16:05.22 | cr2 | rinex per point will be nice of course :) |
| 16:05.29 | cr2 | for post-processing |
| 16:06.29 | cr2 | once i#ve taked to a guy at a conference, who told me how to measure water waves strength by putting the gps receiver at the beach :) |
| 16:06.35 | cr2 | so even that is possible. |
| 16:06.43 | MnemonicPunk | o.o |
| 16:06.45 | kiozen | cr2: what does rinex specify per point? |
| 16:07.00 | kiozen | cr2: any link? |
| 16:07.39 | cr2 | http://artico.lma.fi.upm.es/numerico/miembros/antonio/async/ |
| 16:08.40 | cr2 | http://artico.lma.fi.upm.es/numerico/miembros/antonio/async/node3.html |
| 16:11.01 | cr2 | i think gpstk should be able to deal with rinex. |
| 16:12.08 | cr2 | http://www.gpstk.org/bin/view/Documentation/WebHome |
| 16:12.19 | kiozen | cr2: think the rinex format stuff has to go to GT, but for M I have to make sure to record all needed data in the ext section |
| 16:14.12 | cr2 | put a void * in the ext structure, and version as the first member |
| 16:14.45 | cr2 | then you can change the versions without such pita like with the .gcp |
| 16:15.50 | cr2 | rinex writer should be an M "plugin" |
| 16:16.30 | kiozen | cr2: M stores the data a bit different to be more flexible and nice to resources. |
| 16:16.41 | cr2 | surveying, or something like that. just create the gui window for it, we can fill it later. |
| 16:16.52 | kiozen | once the data has been transfered to GT we can do everything |
| 16:16.53 | cr2 | kiozen: ok, it was just an idea |
| 16:17.09 | kiozen | but I do someting similar |
| 16:17.27 | kiozen | the header tells me what data sections I have |
| 16:17.28 | cr2 | ok |
| 16:17.44 | kiozen | but once I released a data section it's locked. |
| 16:17.50 | cr2 | like the elf format |
| 16:17.59 | kiozen | yes, somehow |
| 16:18.14 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfc055.pool.einsundeins.de) |
| 16:18.25 | kiozen | and I don't wnat to open a new section for each single var |
| 16:18.38 | MnemonicPunk | cr2: Just found this in the wiki: "hh0 1. MMC internals changed a bit, adding "rootdelay=N" param on kernel command line is now required to boot from SD/MMC. " Is that still needed? o.o |
| 16:20.35 | kiozen | cr2: think I do the satellite info in a separate section |
| 16:25.40 | tmzt_ | MnemonicPunk: do you still have the initrd in the haret config? |
| 16:26.03 | MnemonicPunk | Nope, took it out for the moment. |
| 16:26.13 | MnemonicPunk | But I kept it and it is able to start for now. |
| 16:26.22 | tmzt_ | that would explain the VFS error |
| 16:26.39 | tmzt_ | do you still get the cannot execute message, or whatever it was? |
| 16:26.45 | MnemonicPunk | Nope. |
| 16:26.48 | MnemonicPunk | I found my mistake. |
| 16:27.18 | MnemonicPunk | When I copied the libs, I accidentally copied them to the file initfs/lib instead of a directory. |
| 16:27.37 | MnemonicPunk | That part works now. |
| 16:27.37 | tmzt_ | yep, got to mkdir initfs/lib first |
| 16:27.57 | tmzt_ | add the rootdelay=15 |
| 16:27.58 | MnemonicPunk | But it won't recognize basic commands like echo, mkdir and so on. |
| 16:28.03 | MnemonicPunk | Any idea what is missing? |
| 16:28.19 | tmzt_ | cd bin; ln -sf busybox echo |
| 16:28.22 | tmzt_ | cd bin; ln -sf busybox mkdir |
| 16:28.40 | tmzt_ | I thought they were built in |
| 16:28.50 | tmzt_ | at least echo |
| 16:29.05 | MnemonicPunk | Me too, never would have thought of that. o.o |
| 16:29.29 | tmzt_ | if someone knows how to extract the initramfs from a zImage, you could use one that works |
| 16:29.51 | MnemonicPunk | Aaaand. Er. |
| 16:29.54 | MnemonicPunk | One more was there. |
| 16:29.56 | MnemonicPunk | *thinks* |
| 16:30.07 | MnemonicPunk | sleep |
| 16:30.14 | MnemonicPunk | Can I just add that the same way? |
| 16:31.45 | tmzt_ | yes |
| 16:32.08 | dcordes | on the kaiser I can use any filesystem that is smaller than 20MB as initrd with haret. only have to cpio pack it and gzip it |
| 16:32.15 | tmzt_ | and grep |
| 16:32.38 | MnemonicPunk | What about mount? |
| 16:33.00 | tmzt_ | dcordes: were looking for the tiny initramfs that implements rootdelay, looproot, etc. |
| 16:33.08 | tmzt_ | in the handhelds kernels |
| 16:33.23 | tmzt_ | I found it in the oe git, and we are trying to build it |
| 16:33.29 | tmzt_ | yes, mount |
| 16:33.33 | MnemonicPunk | Ah, mount is already there. Probably one of those you told me earlier. |
| 16:33.58 | tmzt_ | I thought so |
| 16:39.07 | MnemonicPunk | One step closer to booting! |
| 16:40.05 | MnemonicPunk | It sitll doesn't want to boot from p2. |
| 16:40.23 | MnemonicPunk | But at least now it gets somewhere and complains that tty isn't available. |
| 16:40.23 | MnemonicPunk | xD |
| 16:48.12 | *** join/#htc-linux MrYellow_ (i=radek@dynamic-78-8-96-97.ssp.dialog.net.pl) |
| 16:48.36 | MnemonicPunk | tmzt_: It lacks "expr", too, can I link that to busybox as well? |
| 16:49.28 | *** join/#htc-linux diogene31_ (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
| 16:55.58 | MnemonicPunk | Alright, I managed to get it running far enough to make it create device nodes. |
| 16:56.06 | MnemonicPunk | How do I continue from here, though? >.< |
| 16:57.51 | MnemonicPunk | It creates the nodes mtdblock0 - mtdblock6 and mmcblk0p1 and mmcblk0p2 |
| 17:04.46 | dcordes | MnemonicPunk: try to mount mmcblk0p2 |
| 17:05.08 | MnemonicPunk | How? |
| 17:05.17 | MnemonicPunk | I can't input anything. o.o |
| 17:09.39 | BabelO_ | kiozen: look at your email :) |
| 17:11.31 | *** join/#htc-linux ALoGeNo (n=QUAKEIII@207.Red-83-41-48.dynamicIP.rima-tde.net) |
| 17:17.50 | tmzt_ | do you get "Unable to mount root device" ? |
| 17:18.23 | MnemonicPunk | It says something about job control being disabled. |
| 17:18.36 | tmzt_ | that's a shell, what's your init/rdinit |
| 17:18.45 | tmzt_ | can you type then? |
| 17:18.48 | MnemonicPunk | And then outputs the "No valid root device was specified. Please add root=/dev/something to the kernel command-line and try again." |
| 17:19.16 | MnemonicPunk | How can I type? o.O The Magician doesn't have any keyboard. |
| 17:20.17 | tmzt_ | oh |
| 17:20.20 | MnemonicPunk | Is there some kind of on-screen keyboard for this state? =( |
| 17:20.45 | MnemonicPunk | If there isn't, I have to make it to the graphical interface w/o any input. |
| 17:21.05 | tmzt_ | did you move those .sh to initrd.d ? |
| 17:21.09 | MnemonicPunk | And yes, it shows a little # as the last line, I guess its a shell. |
| 17:21.17 | MnemonicPunk | Er. |
| 17:21.24 | MnemonicPunk | No should I? Q_Q |
| 17:21.29 | *** join/#htc-linux rob_w (n=bob@M822d.m.pppool.de) |
| 17:21.56 | MnemonicPunk | initfs/initrd.d? |
| 17:22.47 | *** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 17:22.51 | tmzt_ | yes |
| 17:22.55 | tmzt_ | mkdir initfs/initrd.d |
| 17:22.58 | tmzt_ | cd initfs |
| 17:23.01 | MnemonicPunk | Already done. =) |
| 17:23.03 | tmzt_ | mv *.sh initrd.d |
| 17:23.04 | tmzt_ | ok |
| 17:23.45 | MnemonicPunk | Init itself stays in the initfs root? |
| 17:23.50 | tmzt_ | yes |
| 17:24.08 | tmzt_ | did you symlink expr ? |
| 17:24.13 | MnemonicPunk | Yep. |
| 17:24.15 | MnemonicPunk | o.o |
| 17:24.23 | tmzt_ | try it then |
| 17:24.24 | MnemonicPunk | Hope that was right. xD |
| 17:24.43 | tmzt_ | what are rdinit and init ? |
| 17:25.14 | MnemonicPunk | init=/init |
| 17:25.24 | tmzt_ | get rid of init= |
| 17:25.25 | MnemonicPunk | rdinit is not set in default.txt atm. Do I need that? |
| 17:25.28 | MnemonicPunk | Okay. |
| 17:25.37 | tmzt_ | no, you shouldn't need either |
| 17:26.45 | MnemonicPunk | By the way, I noticed there are different commands to use at the end of default.txt |
| 17:26.54 | MnemonicPunk | I am using bootlinux, is that the right one? |
| 17:26.58 | tmzt_ | yes |
| 17:27.38 | tmzt_ | that last error message was from your /init script |
| 17:27.49 | tmzt_ | because it couldn't find its modules |
| 17:28.36 | MnemonicPunk | IT MOUNTS |
| 17:28.41 | MnemonicPunk | Cowabunga! |
| 17:28.41 | MnemonicPunk | xD |
| 17:29.07 | MnemonicPunk | Now locks up at /sbin/kexec but that should be the final missing executable. o.o |
| 17:29.26 | tmzt_ | mv the kexec .sh to the root, you don't need it |
| 17:29.37 | tmzt_ | it's for starting a new kernel |
| 17:30.28 | MnemonicPunk | cd initfs/initrd.d; mv **-kexec.sh ..? |
| 17:30.35 | tmzt_ | yes |
| 17:30.47 | tmzt_ | kexecboot.sh |
| 17:36.13 | MnemonicPunk | Alright, now it's missing switch_root. |
| 17:36.33 | MnemonicPunk | But that's already a few steps forward again. =) |
| 17:40.27 | tmzt_ | might be in busybox] |
| 17:43.09 | *** join/#htc-linux NetRipper (n=netrippe@netripper.nl) |
| 17:44.46 | MnemonicPunk | Alright, let's see how far I get. ^^ |
| 17:45.04 | MnemonicPunk | Woa. |
| 17:45.09 | MnemonicPunk | A splashscreen. |
| 17:45.12 | MnemonicPunk | Teh w00t! |
| 17:45.19 | NetRipper | woot |
| 17:45.24 | NetRipper | on which device? |
| 17:45.36 | MnemonicPunk | HTC Magician, Samsung Display. xD |
| 17:45.43 | NetRipper | sweet, good job |
| 17:45.43 | NetRipper | ;) |
| 17:45.58 | MnemonicPunk | Where do I contribute the kernel and initramfs? xD |
| 17:46.27 | tmzt_ | it worked, switch_root is in that busybox? |
| 17:46.32 | MnemonicPunk | Yes. |
| 17:47.08 | tmzt_ | so magician is pxa, pxamci, and pxafb |
| 17:47.23 | MnemonicPunk | The third doesn't ring a bell. |
| 17:47.36 | MnemonicPunk | But the first two shop up numerous times during boot. |
| 17:47.37 | NetRipper | MnemonicPunk, the original author, if there is one |
| 17:47.38 | NetRipper | :) |
| 17:48.06 | MnemonicPunk | NetRipper: tmzt then, he guided me through the process of making the initramfs =) |
| 17:49.10 | MnemonicPunk | tmzt_: Alright, I'll make sure to repay you if I get the chance some time. =) Thanks a lot. |
| 17:49.23 | NetRipper | :) |
| 17:49.40 | NetRipper | and you modified the kernel right? you can diff that and send it to one of the kernel devs |
| 17:50.06 | MnemonicPunk | I didn't modify anything. |
| 17:50.09 | dcordes | MnemonicPunk: you can always donate 300$ |
| 17:50.15 | NetRipper | oh |
| 17:50.17 | NetRipper | thought you did |
| 17:50.22 | NetRipper | heard something about MAX_PATH |
| 17:50.29 | MnemonicPunk | Yeh okay, that. |
| 17:50.33 | NetRipper | ;) |
| 17:50.42 | NetRipper | small change, but still |
| 17:50.57 | MnemonicPunk | But that was a dirty hack. I don't even know if I put it in the right spot. xD |
| 17:51.33 | MnemonicPunk | Also, that was probably a problem with my environment, not with the source. Some people with certain compiler environments under Ubuntu 8.10 get the same error in the exact same spot. |
| 17:51.47 | MnemonicPunk | (Even for x86 compilation, not ARM) |
| 17:52.24 | MnemonicPunk | You know how you never have the time to try your cool gadgets? |
| 17:52.46 | MnemonicPunk | Well, just as it has booted, my g/f wants me to pay attention to her... xD |
| 17:53.05 | MnemonicPunk | See you in a few hours. =) |
| 17:53.41 | NetRipper | haha ok |
| 17:53.42 | NetRipper | have fun |
| 17:55.38 | dcordes | NetRipper: anything exciting about raphael bootup? |
| 18:07.25 | *** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net) |
| 18:24.07 | BabelO | kiozen: no need of additionnal libray :) |
| 18:24.32 | kiozen | BabelO: oh, cool you did all by yourself? |
| 18:24.44 | BabelO | yes |
| 18:25.14 | BabelO | kiozen: that's why it take me from monday :) i forgot some basic thing, so need to learn again |
| 18:25.15 | kiozen | BabelO: users will definetly thank you |
| 18:25.41 | kiozen | BabelO: yes, yes I am like a gold fish,too, every round a new experience :) |
| 18:25.49 | BabelO | kiozen: i hope result are good on most maps |
| 18:26.02 | kiozen | BabelO: did you try mine? |
| 18:26.27 | BabelO | i ve to try it now |
| 18:26.36 | BabelO | but should work too |
| 18:26.50 | kiozen | worst test case :) |
| 18:27.08 | BabelO | kiozen: i think i have bigger maps from you here :) |
| 18:30.19 | BabelO | kiozen: but we need a zoom factor set to 1 for better quality, it works for other zoom up to x3 but not efficieznt |
| 18:30.30 | kiozen | BabelO: whst about the caves? :P |
| 18:30.42 | BabelO | kiozen: wait :) |
| 18:31.31 | marajin | yo |
| 18:31.58 | kiozen | BabelO: well caves are important, but they can wait until April now. season is over |
| 18:32.30 | BabelO | kiozen: lol, but i have done the algorithm for wood :) "O_" and "O" symbols |
| 18:32.31 | BabelO | lol |
| 18:32.51 | kiozen | lol who cares for trees |
| 18:33.40 | BabelO | kiozen: season for fungus is not over |
| 18:33.58 | kiozen | BabelO: right, but I am really bad for that |
| 18:34.03 | kiozen | BabelO: too blind |
| 18:34.38 | BabelO | lol |
| 18:35.37 | kiozen | BabelO: I thing I can trip over them and won't see them |
| 18:36.32 | dcordes | cr2: can you help me with kaiser wifi? |
| 18:41.32 | *** join/#htc-linux ali1234 (n=al@62.24.214.38) |
| 18:44.03 | *** join/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes) |
| 18:51.00 | dcordes_ | BabelO: do you have experience with sdio wifi? |
| 18:52.52 | BabelO | dcordes_: no really, only pcmcia |
| 18:53.03 | BabelO | dcordes_: you have a driver ? |
| 18:54.34 | marajin | we do technically I think |
| 18:55.27 | dcordes_ | BabelO: yes. acx100 and kaiser sd controller works. |
| 18:58.45 | *** join/#htc-linux apt (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 18:58.45 | *** topic/#htc-linux is HTC Linux Channel: Find logs at http://apt.rikers.org/%23htc-linux/ | please check http://handhelds.org/moin/moin.cgi/HTC_2dPhones | http://wiki.xda-developers.com/index.php?pagename=Xanadux | <cr2> let's define a common setup. |
| 18:58.46 | dcordes_ | and only operational for sd card |
| 18:58.52 | BabelO | dcordes_: ah ok |
| 18:59.09 | BabelO | dcordes_: in google git, there is serial sdio driver example for GPS |
| 18:59.15 | BabelO | it works like pcmcia |
| 18:59.17 | dcordes_ | BabelO: the acx100 driver _should_ work for the kaiser sdio wifi (TNETW1251) |
| 18:59.24 | BabelO | it allocate meme region by asking sdio |
| 18:59.36 | BabelO | i do such for pcmcia driver |
| 19:00.16 | dcordes_ | ah I think I know which example you mean. I have seen that in the kaiser kconfig |
| 19:00.44 | dcordes_ | [ ] SDIO UART/GPS class support ? |
| 19:01.44 | BabelO | yes |
| 19:02.15 | BabelO | dcordes_: acx mem driver from hh kernel need an irq and a mem region to work |
| 19:02.54 | dcordes_ | BabelO: I talked with SanMehat about that wifi irq |
| 19:03.04 | dcordes_ | marajin: remember? |
| 19:03.13 | dcordes_ | I wonder when that was. |
| 19:04.58 | marajin | yeah |
| 19:05.48 | dcordes_ | <PROTECTED> |
| 19:06.03 | BabelO | kiozen: what about symbols /_\_ |
| 19:06.22 | BabelO | dcordes_: drivers/net/wireless/acx/ |
| 19:07.19 | dcordes_ | BabelO: oh, a lot of files in there.- |
| 19:08.18 | BabelO | kiozen: email |
| 19:08.38 | BabelO | dcordes_: yes, you can copy this directory, what is your current android kernel version ? |
| 19:08.55 | dcordes_ | BabelO: 2.6.25. but there is only 2.6.24 in the git |
| 19:10.00 | BabelO | dcordes_: ok, because newer kernel use a mac80211 stack so you need to change a little acx driver to be compatible |
| 19:10.10 | BabelO | not sure about 2.6.24 |
| 19:11.42 | dcordes_ | BabelO: kaiser/polaris/other-msm7200 2.6.25 tree has both, mac80211 and the old one |
| 19:11.56 | dcordes_ | <PROTECTED> |
| 19:11.59 | dcordes_ | <PROTECTED> |
| 19:13.38 | dcordes_ | BabelO: but I know acx100 also works with mac80211. will it be easier to use the hh acx? |
| 19:14.16 | BabelO | dcordes_: acx100 work with mac80211 but not the acx_mem version |
| 19:14.23 | dcordes_ | I see |
| 19:14.29 | BabelO | and you need this acx_mem version which work with old stack |
| 19:14.37 | BabelO | unless someone port it |
| 19:14.42 | dcordes_ | ok so that's no problem |
| 19:14.59 | BabelO | but not sure, i send my code to pH5, because i fail get it working with mac80211 |
| 19:15.05 | dcordes_ | when I copied the folder to the tree, where must I start to look at? |
| 19:15.34 | dcordes_ | I see machine specific .c files in it |
| 19:15.53 | dcordes_ | do I need a htckaiser_acx.c? |
| 19:16.31 | *** join/#htc-linux infernixx (i=gerben@unaffiliated/infernix) |
| 19:16.50 | dcordes_ | hi infernixx |
| 19:17.04 | SanMehat | hola |
| 19:17.15 | *** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net) |
| 19:17.22 | marajin | 'lo SanMehat |
| 19:17.29 | dcordes_ | hi |
| 19:17.34 | SanMehat | marajin: 'lo |
| 19:18.03 | infernixx | hi:) |
| 19:18.07 | imfloflo | hi dcordes i'm trying your v8 |
| 19:18.13 | dcordes_ | great |
| 19:18.30 | imfloflo | no keyboard appear when push up camera button |
| 19:18.39 | dcordes_ | grrrmmhjhmrm |
| 19:19.03 | dcordes_ | maybe something is missing in the keyboard driver |
| 19:19.27 | dcordes_ | did anything else change since test-lol zImage? |
| 19:19.44 | imfloflo | but it boots fine |
| 19:20.08 | kiozen | BabelO: hey, that's good! |
| 19:21.44 | BabelO | kiozen: yes symbols should be easy to find like this |
| 19:22.01 | kiozen | BabelO: caves? :) |
| 19:22.03 | dcordes_ | imfloflo: nothing changed about the display? |
| 19:22.06 | imfloflo | nothing |
| 19:22.17 | imfloflo | display is very good |
| 19:22.31 | dcordes_ | imfloflo: didn't you say before something strange happened when you pushed cam button? |
| 19:22.33 | BabelO | dcordes_: machine specifi is for power on / off acx, contain only gpio and irq definition + mem region definition |
| 19:22.37 | BabelO | empty file :) |
| 19:22.50 | imfloflo | nothings happens no keyboard |
| 19:23.04 | dcordes_ | no split screen or anything? |
| 19:23.14 | imfloflo | no works fine |
| 19:23.25 | dcordes_ | and before, did something happen when you pushed cam in the older test-*? |
| 19:24.29 | imfloflo | in the old version with split screen when i pushed cam button the screen change display but don't saw keyoard |
| 19:25.28 | dcordes_ | so the only builds where something happened where the ones which had the split screen? |
| 19:25.46 | imfloflo | yes ^^ |
| 19:29.43 | dcordes_ | hmm the function which triggers the keyboard is included. ohterwise you wouldn't have seen something happen before |
| 19:31.02 | imfloflo | ok |
| 19:33.41 | dcordes_ | imfloflo: I have no idea what I can try then. I'll make a diff and ask somebody else |
| 19:34.10 | imfloflo | ok thanks |
| 19:38.23 | kiozen | marajin: around, need a native speaker |
| 19:38.57 | dcordes_ | how does one remove file from git index again? |
| 19:39.05 | marajin | kiozen: 'sup? |
| 19:39.31 | kiozen | how would you call dy/dx of a road |
| 19:40.20 | dcordes_ | git-rm |
| 19:40.29 | marajin | er.. you mean on the horizontal or vertical? |
| 19:41.45 | marajin | kiozen: what's your native language? |
| 19:41.51 | kiozen | the derivation of a road in a normal human words |
| 19:41.53 | kiozen | german |
| 19:42.03 | marajin | say it in german, I can probably translate |
| 19:42.11 | kiozen | steigung |
| 19:43.11 | marajin | You probably mean slope? |
| 19:43.22 | marajin | gradient |
| 19:43.32 | kiozen | slope bingo, I won :) |
| 19:44.22 | kiozen | we just had a little discussion about it, thanks |
| 19:44.39 | marajin | no worries |
| 19:44.55 | marajin | why, what were the others saying? |
| 19:45.07 | kiozen | incline |
| 19:45.24 | marajin | well it would be near identical in meaning |
| 19:45.31 | marajin | slope is probably just a little more generic english |
| 19:45.57 | *** join/#htc-linux wdslbr (n=feld@dslb-088-069-202-082.pools.arcor-ip.net) |
| 19:45.58 | marajin | though incline would only be the positive ascent of a road between two points in altitute |
| 19:46.22 | kiozen | lol, marajin: :)) you made a woman happy |
| 19:46.39 | marajin | I have that effect ;) |
| 19:46.40 | marajin | :p |
| 19:47.01 | kiozen | lol |
| 19:47.11 | marajin | see there's incline and decline as an opposing word pair |
| 19:47.55 | marajin | As a general guideline I'd say a road slopes, a hill inclines |
| 19:48.05 | kiozen | right, but if you want to talk about the gradient/derivation of a recorded gps track I think slope is the way to go |
| 19:49.04 | dcordes_ | kiozen: do you use kalman filtering already to polish the tracks? |
| 19:49.30 | kiozen | no, sofar just a recursive lowpass filter |
| 19:51.25 | marajin | kiozen: If you're writing an english language set of strings for your nav software, I don't mind looking over the stuff as a review when you've done it |
| 19:52.01 | kiozen | marajin: you just want to amuse yourself :P |
| 19:52.28 | BabelO | kiozen: that's be easy for marajin with qt linguist :) |
| 19:52.39 | marajin | Not really, you obviously have a good enough grasp of english that you'd shame the average native speaker |
| 19:52.52 | marajin | I just know non-native speakers sometimes have trouble with slightly awkward phrasing |
| 19:52.59 | marajin | things are 'technically' correct but they sound unnatural |
| 19:53.24 | kiozen | lol, had loads of fun in british pubs in my youth |
| 19:53.31 | marajin | Ya know, using words like 'excursion' |
| 19:53.43 | marajin | sure, it's a valid english word but anyone saying it was probably being sarcastic :P |
| 19:54.03 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfc055.pool.einsundeins.de) |
| 19:54.20 | BabelO | marajin: did you solve your frozen virtualbox ? |
| 19:54.25 | kiozen | oh, I would hav thought of a nice sunday in the woods :) |
| 19:54.37 | marajin | BabelO: what frozen virtualbox? |
| 19:54.51 | marajin | BabelO: You mean my 'slow' virtualbox? |
| 19:55.16 | BabelO | marajin: yes :) does virtualbox use lot of cpu ? |
| 19:55.26 | marajin | BabelO: Yeah I just changed the cpu affinity for the process. It was shifting onto hyperthreaded 'cores' |
| 19:55.41 | BabelO | marajin: ok :) |
| 19:55.59 | BabelO | Virtualbox run really fine |
| 19:56.04 | marajin | locked it to the main cores and it's been great :) |
| 19:56.11 | marajin | WinXP has probably been running 3-4 days |
| 19:56.39 | marajin | System Up Time: 3 Days, 20 Hours, 34 Minutes, 14 Seconds |
| 19:56.45 | marajin | that's XP Pro SP3 in a VirtualBox |
| 19:56.45 | BabelO | :) |
| 19:57.03 | BabelO | marajin: i have sp3 on eeepc ... |
| 19:57.12 | marajin | heh |
| 19:57.17 | marajin | ironically my virtualbox may actually be faster |
| 19:57.40 | marajin | just RDPs to the virtualbox from his UMPC |
| 19:57.54 | marajin | adds an ABC to the sentence just for good measure :p |
| 19:58.46 | marajin | BabelO: So, you played with host networking yet? Once I gave it a kicking it actually works very well. No network problems at all in the last week |
| 19:59.21 | BabelO | marajin: yes i have it running at office with linux guest with latest too |
| 19:59.26 | marajin | and of course my Virtualmachines can grab real DHCP addresses and become part of the real LAN |
| 19:59.50 | BabelO | it seem that unbuntu run well in seamless mode but not opensuse, still looking why |
| 19:59.59 | marajin | heh =) |
| 20:00.02 | BabelO | marajin: yes my linux guest is a zabbix server |
| 20:00.09 | marajin | well I don't even use the virtualbox gui anymore |
| 20:00.21 | marajin | VBoxHeadless =) |
| 20:01.05 | marajin | I could kiss them for including that feature |
| 20:01.07 | BabelO | marajin: i like to have access to linux application over my windows desktop at office, so can do some funny thing more easily |
| 20:01.24 | BabelO | yes headless is good for server |
| 20:01.50 | marajin | well otherwise I must NX -> Server -> KDE -> Run VirtualBox |
| 20:01.55 | marajin | tis very silly |
| 20:02.35 | marajin | Also, just using the native RDP services of the windows guests is nice |
| 20:02.38 | BabelO | i did not look at vbox plugin for a while now |
| 20:02.41 | marajin | they're nowhere near as broken as the vbox one |
| 20:03.03 | BabelO | yes agree, but did not try with latest |
| 20:03.24 | marajin | works fine in 2.0.2 with WinXP for me as long as host networking is used |
| 20:03.48 | marajin | kinda funny you can capture vbox output to a video actually |
| 20:03.57 | BabelO | yeah |
| 20:04.18 | BabelO | oh oh oh |
| 20:04.27 | BabelO | two new wifi networks ;) |
| 20:05.29 | marajin | oh? |
| 20:05.42 | BabelO | yes, both with WEP key |
| 20:08.57 | marajin | oooo |
| 20:09.47 | *** join/#htc-linux romario (n=romario@ip-89-102-143-87.karneval.cz) |
| 20:11.45 | BabelO | that's the problem to make difference between B and 8 when you type a long long WPA key ... after many try, you simply switch to WEP and use that |
| 20:16.27 | dcordes_ | BabelO: is it possible to let a certain config in a Kconfig file depend on one machine OR another machine? like depends on MACH_HTCKAISER or MACH_HTCPOLARIS |
| 20:17.14 | BabelO | dcordes_: what do you want to do ? |
| 20:17.41 | dcordes_ | BabelO: add kaiser to drivers/net/wireless/acx/Kconfig |
| 20:18.11 | BabelO | dcordes_: yes you can, you have many example |
| 20:18.11 | dcordes_ | BabelO: but also MACH_HTCPOLARIS uses the tnetw sdio wifi |
| 20:18.57 | dcordes_ | depends on MACH_HTCKAISER && MACH_HTCPOLARIS ? |
| 20:19.06 | dcordes_ | I thought that will require both machines.. |
| 20:19.15 | *** join/#htc-linux gtt (n=ytuygy@87.252.227.52) |
| 20:21.35 | BabelO | dcordes_: || |
| 20:22.00 | dcordes_ | ok |
| 20:23.38 | dcordes_ | the acx board files all have a line saying (TI TNETW1100B), while kaiser has TNETW1251 |
| 20:24.32 | marajin | SanMehat said that he knew it though? the 1251 |
| 20:24.48 | SanMehat | marajin: hi |
| 20:24.50 | BabelO | kiozen: how do you know distance for each pixel from the zoom factor ? |
| 20:25.14 | BabelO | dcordes_: that for 802.11B |
| 20:25.31 | BabelO | dcordes_: do you know the acx100 memory region ? |
| 20:26.27 | BabelO | cr2: sleeping ? |
| 20:26.29 | dcordes_ | BabelO: http://wiki.xda-developers.com/index.php?pagename=KaiserMemoryMap |
| 20:26.49 | kiozen | BabelO: you need the map scal in [px/m], too |
| 20:26.49 | *** join/#htc-linux Type (n=typevr@87.252.227.52) |
| 20:27.01 | dcordes_ | BabelO: it has a SDIO1 and SDIO2 range |
| 20:27.11 | BabelO | kiozen: yes that is what i mean |
| 20:27.28 | dcordes_ | marajin: SDIO0 and SDIO1 sorry |
| 20:27.29 | BabelO | kiozen: just to filter the grid position from noise |
| 20:27.47 | kiozen | IMap::convertM2Pt(..) might be what you want |
| 20:28.23 | kiozen | or you have to extend the IMap API to read the scale factor and zoom factor |
| 20:28.48 | dcordes_ | BabelO: #define WLAN_BASEPXA_CS2_PHYS <- that must be changed to the correct sdio base? |
| 20:29.32 | dcordes_ | must I include msm_sdcc.h? |
| 20:29.35 | BabelO | dcordes_: yes, but sdio is like pcmcia, it use tuple like and should return you the acx base adress |
| 20:32.09 | dcordes_ | BabelO: I'm using the universal acx file as a base. asic3 is what provides access to the uni gpios? |
| 20:32.23 | BabelO | kiozen: look here , why i need to make filter :) http://linuxbox2.free.fr/maps/ |
| 20:32.42 | BabelO | dcordes_: yes |
| 20:33.03 | dcordes_ | BabelO: so I must use msm7x00 gpio code instead |
| 20:33.24 | BabelO | dcordes_: it is an extend to gpios with some other feature like ds1wm bus, mmc , pwm |
| 20:33.36 | BabelO | dcordes_: yes |
| 20:34.13 | dcordes_ | egpio is also on kaiser since you added it |
| 20:34.36 | BabelO | kiozen: in the blue you can see the noise, i can filter it because i know the direction of the point, but need another way to do it quickly too :) |
| 20:34.56 | BabelO | dcordes_: right :) do you swtich to the new egpio driver ? |
| 20:35.04 | dcordes_ | BabelO: no. don't know how to do it. |
| 20:35.19 | dcordes_ | I tried it once but ended up not knowing what I did |
| 20:36.26 | dcordes_ | <PROTECTED> |
| 20:36.43 | dcordes_ | don't know how the infernixx got in there ^^ |
| 20:37.31 | kiozen | BabelO: you can do correlation, but that is not fast :) |
| 20:37.47 | infernixx | no idea:) |
| 20:38.08 | dcordes_ | infernixx: do you think you can help with the wifi? |
| 20:38.59 | BabelO | kiozen: better to look around the probable position once i found first line |
| 20:39.13 | BabelO | kiozen: it is zoom x2 there |
| 20:39.26 | kiozen | right, but for that you need the gridspacing, too |
| 20:40.10 | BabelO | kiozen: not reallly just need to find redundant point |
| 20:40.28 | infernixx | dcordes_: what I can do is log on to my kaiser with ssh over public IP; if that's any help I can assist |
| 20:40.38 | infernixx | but i'm just to damn busy to do anything more right now :| |
| 20:41.03 | cr2 | BabelO: not yet. but i've lost the track |
| 20:41.57 | BabelO | cr2: do you see a specific mem region for acx on kaiser or raphael |
| 20:42.13 | dcordes_ | infernixx: ok nvm |
| 20:45.22 | gtt | ïî ðóññêè ãîâîðèò êòî-íèáóäü? |
| 20:47.24 | dcordes_ | BabelO: do you know where PXA_CS2_PHYS is defined in the hh tree? |
| 20:48.09 | BabelO | dcordes_: it is a PXA base adress some like 0x2c0000000 |
| 20:49.07 | BabelO | dcordes_: should be defined somewhere is /include/asm-arm/arch-pxa/pxa.h |
| 20:49.55 | dcordes_ | BabelO: ./include/asm-arm/arch-pxa/pxa-regs.h:#define PXA_CS2_PHYS0x08000000 |
| 20:50.33 | BabelO | gtt: try with ~x |
| 20:50.51 | BabelO | ~x en ru gtt : saying hello ? |
| 20:51.32 | dcordes_ | :) |
| 20:52.20 | BabelO | i hope babelfish did not make mistake while translation :) |
| 20:52.22 | dcordes_ | BabelO: in the UniversalMemoryMap 0x08000000 is labled "acx100". |
| 20:52.30 | dcordes_ | not sure what it is on kaiser |
| 20:52.47 | dcordes_ | BabelO: can you look in http://wiki.xda-developers.com/index.php?pagename=KaiserMemoryMap ? |
| 20:54.18 | *** part/#htc-linux Type (n=typevr@87.252.227.52) |
| 20:55.58 | BabelO | dcordes_: don't know, as i say before, acx got is adress from sdio bus i think. i have see something on a sdio driver... was just thinking that if you are lucky acx expose its memory region somewhere else like on Universal |
| 20:57.03 | dcordes_ | BabelO: I think I know what I must use. |
| 20:57.12 | dcordes_ | MSM_SDC2_PHYS 0xA0400000 |
| 20:57.55 | dcordes_ | because MSM_SDC1 is sd |
| 21:00.40 | *** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde) |
| 21:01.16 | dcordes_ | BabelO: do I need DMA? msm_sdcc.c also uses DMA |
| 21:01.54 | BabelO | dcordes_: don't think it works with only sdc1_phys |
| 21:02.08 | BabelO | structure is not the same |
| 21:02.14 | cr2 | BabelO: it may be 0x9* or 0x80000000 |
| 21:03.02 | cr2 | dcordes_: i think sdio itself is used only for some init sequence (like pcmcia on ba) |
| 21:03.07 | *** part/#htc-linux gtt (n=ytuygy@87.252.227.52) |
| 21:03.41 | cr2 | dcordes_: the data goes over mmaped registers, and the dma is not needed. |
| 21:04.03 | cr2 | dcordes_: it works on universal,hx470x,ba,hw6915,etc. |
| 21:04.13 | dcordes_ | cr2: do we know which mmaped registers? |
| 21:04.22 | cr2 | dcordes_: no. |
| 21:04.37 | cr2 | dcordes_: on ba we could not trace the wifi too |
| 21:04.46 | cr2 | but on universal it was visible |
| 21:04.53 | BabelO | cr2: sdio linux driver should return 0x8xxxx or 0x9xxxxx as meme region like it does on ba with pcmcia |
| 21:05.29 | cr2 | on universal i've seen the memory access in the SPL |
| 21:05.52 | cr2 | BabelO: return ? |
| 21:06.50 | BabelO | cr2: yes, on ba when pcmcia driver ask for windows region available for plugged pcmcia card it give 0x2c0000000 |
| 21:07.04 | dcordes_ | cr2: maybe the answer is in the msm_sdcc code? |
| 21:08.17 | dcordes_ | I mean if the same works for the SD card, why should we not see the regs/whatever for the WIFI on kaiser? |
| 21:09.27 | dcordes_ | http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=blob;f=drivers/mmc/host/msm_sdcc.c;hb=refs/heads/htc-msm |
| 21:11.39 | cr2 | BabelO: i think it's unlikely with sdio, but who knows. |
| 21:11.57 | dcordes_ | how can I try that, with the return? |
| 21:12.38 | cr2 | dcordes_: the sdio functions/register layout is fixed. there should be some tool to read the tuples |
| 21:12.49 | cr2 | like for the pcmcia |
| 21:13.08 | cr2 | cis_dump or something like that |
| 21:13.24 | dcordes_ | BabelO: do you know which tool I need? |
| 21:15.52 | BabelO | dcordes_: you can take sdio_uart as sample code |
| 21:16.59 | *** join/#htc-linux nebi (n=nebi@c-498fe253.02-145-7570701.cust.bredbandsbolaget.se) |
| 21:17.51 | dcordes_ | drivers/mmc/card/sdio_uart.c what is the interesting part there? |
| 21:17.59 | BabelO | dcordes_, cr2 http://sourceforge.net/project/showfiles.php?group_id=186068&package_id=216821&release_id=476499 |
| 21:18.17 | BabelO | there is an ar6000 sdio wlan patch |
| 21:20.17 | dcordes_ | BabelO: ok got the ar6000 patch |
| 21:20.19 | cr2 | dcordes_: uart is a completely different class |
| 21:20.38 | cr2 | dcordes_: but the routines for reading tuples should be there too |
| 21:20.55 | dcordes_ | what's a tuple? |
| 21:21.08 | dcordes_ | +/* |
| 21:21.09 | dcordes_ | + * This reads the card's CONFIG tuple to find its |
| 21:21.09 | dcordes_ | + * configuration registers. |
| 21:21.09 | dcordes_ | + */ |
| 21:21.21 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 21:21.37 | snake_schule | grml my screen has died... |
| 21:21.58 | BabelO | dcordes_: look for ar6000_cs.c |
| 21:22.15 | dcordes_ | what I pasted i in that file |
| 21:22.19 | dcordes_ | is |
| 21:26.46 | BabelO | dcordes_: you need someone like dzo or ginge to explore the hif_internal.h and the whole patch :) with the acx_cs pcmcia driver and meme driver they should be able to do a mix :) |
| 21:28.15 | dcordes_ | no way to find the regs using haret? |
| 21:29.18 | BabelO | lol |
| 21:29.21 | BabelO | + help |
| 21:29.21 | BabelO | +good luck. |
| 21:29.34 | BabelO | that is the atheros driver description :) |
| 21:30.05 | BabelO | dcordes_: it seem that the wlan sdio base function is sdio_memory |
| 21:30.28 | BabelO | so like cr2 say you can try to set 0x9xxx or 0x8xxx and boot with wifi on from wince |
| 21:30.57 | BabelO | i have to try sdio driver on artemis :) |
| 21:31.13 | snakefreak | when i install network interface rndis/ethernet gadget in win xp my network conection brakes when i connect my uni to the pc anybody has a solution? |
| 21:32.04 | dcordes_ | cr2: BabelO why 0x8xxx or 0x9xxx ? |
| 21:32.33 | BabelO | snakefreak: it is just windows :) one netwrok interface at a time |
| 21:33.11 | snakefreak | hm grml |
| 21:34.24 | cr2 | dcordes_: because the CPLD is at 0x98000000 |
| 21:36.37 | dcordes_ | cr2: so I define the CPLD base fas WLAN_BASE? |
| 21:36.42 | *** join/#htc-linux infernixx (i=gerben@unaffiliated/infernix) |
| 21:39.42 | cr2 | no |
| 21:39.55 | cr2 | it should be some similar area |
| 21:40.58 | dcordes_ | 0x94000000 ? I wasn't able to trace that area |
| 21:42.16 | *** join/#htc-linux Type (n=typevr@87.252.227.52) |
| 21:43.32 | dcordes_ | I could try several values but first I must edit the rest of the acx board file. what about wifi gpios? I assume power gpios are not needed when wifi is on in wince? |
| 21:44.54 | cr2 | scoi think theyare known |
| 21:44.55 | cr2 | ? |
| 21:46.41 | dcordes_ | cr2: on the cpld, but only two: wifi_pwr1 (c3) wifi_pwr2 (c2) |
| 21:47.07 | dcordes_ | s/c2/d1/ |
| 21:47.28 | dcordes_ | while universal has 3 |
| 21:48.14 | cr2 | it's a different chipset... |
| 21:48.21 | cr2 | what about msm gpio ? |
| 21:48.49 | BabelO | cr2: what is the replacement for linux/config.h in newer kernel, i always forgot |
| 21:48.58 | BabelO | i look for it ... |
| 21:49.59 | *** join/#htc-linux infernixxx (i=gerben@91.141.202.136) |
| 21:50.07 | dcordes_ | cr2: ok so kaiser needs only 2 power gpios. not sure what you mean what about msm gpio? |
| 21:51.06 | cr2 | BabelO: don't know |
| 21:51.11 | BabelO | cr2: ok touch include/linux/config.h is ok for me |
| 21:51.12 | BabelO | lol |
| 21:51.32 | cr2 | dcordes_: the third gpio may be an msm gpio, not cpld |
| 21:51.56 | cr2 | NetRipper: which LCD panel do you have ? |
| 21:52.23 | NetRipper | dcordes_, no news on my side |
| 21:52.35 | NetRipper | cr2, don't know about type, isn't it on raphael wiki specs? |
| 21:53.11 | cr2 | NetRipper: mine is hitachi |
| 21:53.16 | dcordes_ | cr2: oh right, there are 6 entries labled SD2 related (wifi) and then there is the on/off IRQ gpio on msm gpio |
| 21:53.29 | NetRipper | cr2, how do i check? |
| 21:53.40 | NetRipper | and aren't all raphaels the same? |
| 21:53.41 | NetRipper | :) |
| 21:53.44 | cr2 | i've picked it looking at the 'dmesg' |
| 21:53.48 | cr2 | don't know |
| 21:53.52 | NetRipper | er, you booted raphael? |
| 21:54.00 | cr2 | there may be a toppoly devices |
| 21:54.02 | dcordes_ | there's the wince dmesg |
| 21:54.05 | cr2 | noit yet |
| 21:54.05 | NetRipper | oh |
| 21:54.10 | cr2 | wince dmesg |
| 21:54.12 | NetRipper | didn't know about wince dmesg |
| 21:54.22 | cr2 | pd 0x17200000 0x100000 |
| 21:54.30 | cr2 | or |
| 21:54.33 | cr2 | better |
| 21:54.45 | cr2 | pwf wince-dmesg 0x17200000 0x100000 |
| 21:55.51 | NetRipper | lets see |
| 21:56.42 | dcordes_ | cr2: I guess "29 0x1d 0,3 13 I wifi IRQ (off,on)" that and the two power gpios on the cpld must be critical for the init/exit of kaiser wifi? |
| 21:56.42 | NetRipper | er, i should've dumped that to file lol |
| 21:58.29 | NetRipper | that last cmd does not return any data |
| 21:58.34 | NetRipper | the pwf |
| 21:58.36 | NetRipper | the pd does |
| 21:58.54 | cr2 | NetRipper: pwf means 'physical write to file' |
| 21:59.00 | NetRipper | oh |
| 21:59.01 | NetRipper | lol |
| 21:59.08 | NetRipper | ok got it |
| 21:59.09 | cr2 | wince-dmesg |
| 21:59.12 | NetRipper | i used redir for that normally |
| 21:59.16 | cr2 | ok |
| 21:59.29 | cr2 | it'll be a normal ascii file |
| 21:59.43 | NetRipper | only see some [DISP] +HitachiPowerOn and power off lines |
| 21:59.44 | cr2 | maybe with a little binary blob at the beginning |
| 21:59.56 | cr2 | that's good. |
| 21:59.58 | NetRipper | and a lot of other stuff obviously |
| 22:00.02 | cr2 | no toppoly message ? |
| 22:00.25 | cr2 | i'm looking for the bank5 addresses |
| 22:00.33 | cr2 | to fix the gpio traceing |
| 22:00.38 | NetRipper | toppoly? |
| 22:00.55 | NetRipper | the string toppoly does not exist |
| 22:01.24 | cr2 | ok |
| 22:01.34 | cr2 | so you have a hitachi panel too |
| 22:01.35 | cr2 | good |
| 22:01.57 | NetRipper | i'd be concerned if raphaels have different lcd panels |
| 22:02.23 | NetRipper | funny how you see my every touch on the lcd panel |
| 22:02.28 | NetRipper | coordinate-wise |
| 22:03.02 | cr2 | NetRipper: have you seen that the lower keypad is touchsensitive too ? |
| 22:03.21 | NetRipper | yes |
| 22:03.30 | NetRipper | i gave you a link to the htc debug program |
| 22:03.32 | NetRipper | that shows it |
| 22:03.33 | cr2 | it pops the respective gpio irq, i've added it to the wiki yesterday |
| 22:03.55 | *** join/#htc-linux wasup (n=Tech@208.47.103.106) |
| 22:04.24 | dcordes_ | cr2: we already have those kaiser cpld gpios defined in board.h |
| 22:04.33 | dcordes_ | e.g. #define EGPIO_3_1_WIFI_PWR2 KAISER_EGPIO(3, 1) |
| 22:08.20 | NetRipper | arf anyone knows how i make a proper diff _excluding_ any CVS directories? |
| 22:08.21 | NetRipper | :) |
| 22:08.42 | dcordes_ | in git I use git-diff |
| 22:08.45 | NetRipper | oh |
| 22:08.47 | NetRipper | yes |
| 22:08.50 | NetRipper | cvs diff it is |
| 22:08.54 | cr2 | diff -x ? |
| 22:09.30 | cr2 | --exclude |
| 22:09.54 | NetRipper | oh yea |
| 22:10.06 | NetRipper | i looked through the --help a number of times |
| 22:10.09 | cr2 | -u is also nicer than < > |
| 22:10.11 | NetRipper | guess i was sleepy |
| 22:10.17 | NetRipper | yes i prefer unified |
| 22:12.52 | dcordes_ | cr2: I only know how to set egpios when the machine boots.. but do I need them to try out wifi_base registers? |
| 22:13.04 | dcordes_ | can't I just let wince setup the wifi to try the memory addresses? |
| 22:14.13 | cr2 | dcordes_: btw, can we set the msm gpios directly ? |
| 22:14.32 | cr2 | dcordes_: afair the android kernel uses some obfuscated proc_comm for that. |
| 22:14.48 | cr2 | dcordes_: i mean if we have the code for that |
| 22:15.09 | cr2 | because the htc code is full of it. i'm just looking for the bank5 regs |
| 22:15.48 | BabelO | good night |
| 22:15.54 | NetRipper | what do you think bank5 is used for? |
| 22:16.05 | NetRipper | sensor? |
| 22:16.45 | *** part/#htc-linux Type (n=typevr@87.252.227.52) |
| 22:16.48 | cr2 | http://wiki.xda-developers.com/index.php?pagename=Raphael_GPIO |
| 22:17.02 | dcordes_ | night BabelO |
| 22:17.28 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 22:18.47 | wasup | hmm... where has ali1234 and ginge been for the past 2 days? |
| 22:19.42 | dcordes_ | they might be hiding from making a titan keyboard driver ;) |
| 22:19.56 | dcordes_ | wasup: did you succeed creating your own kernel builds? |
| 22:20.13 | dcordes_ | it's easy now that dzo has the htc-vogue branch in linuxtogo git |
| 22:25.17 | wasup | nope not yet |
| 22:25.23 | wasup | hiding? |
| 22:26.13 | wasup | yeah? |
| 22:26.40 | wasup | so I should maybe be expecting a working keyboard sometime soon? |
| 22:27.07 | wasup | what is the button to open the on screen keyboard in DZO's build for the vogue? |
| 22:27.14 | dcordes_ | cam |
| 22:27.24 | wasup | hmmm...that didnt work on mine |
| 22:27.25 | dcordes_ | did you ask ginge for the titan keyboard diff? |
| 22:27.36 | wasup | no |
| 22:27.49 | wasup | i was talking to ali1234 mostly |
| 22:27.55 | dcordes_ | if you tell me where I can find it I can have a look |
| 22:28.33 | *** join/#htc-linux snakefreak (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 22:28.53 | wasup | find the zImage? |
| 22:29.12 | wasup | the keyboard diff? |
| 22:33.31 | *** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net) |
| 22:34.58 | dcordes_ | drivers/net/wireless/acx/common.c:2853: error: âstruct sk_buffâ has no member named âmacâ |
| 22:35.19 | cr2 | NetRipper: picked some more interesting things about 7201A |
| 22:36.13 | NetRipper | what did you find? |
| 22:36.47 | cr2 | editing wiki now |
| 22:38.08 | cr2 | hmm. it's bank5 related |
| 22:38.27 | NetRipper | oh sweet |
| 22:38.29 | dcordes_ | ..I guess the above is because the acx I use from hh is obsolete |
| 22:38.38 | *** join/#htc-linux MnemonicPunk (n=atlantis@dslb-084-060-024-040.pools.arcor-ip.net) |
| 22:39.25 | MnemonicPunk | Okay, now that I have a working kernel, how do I get another distro on my phone? xD |
| 22:39.38 | dcordes_ | MnemonicPunk: replace rootfs |
| 22:40.06 | dcordes_ | MnemonicPunk: especially easy when you have no module support and thus don't need to deal with modules |
| 22:40.08 | MnemonicPunk | But it is important which device the rootfs is for? |
| 22:40.11 | MnemonicPunk | Ah. |
| 22:40.27 | MnemonicPunk | So I can take rootfs' from other devices? |
| 22:40.43 | dcordes_ | MnemonicPunk: except for the modules, it's the arch the binaries are built for |
| 22:40.55 | dcordes_ | obviously you can't use x86 binaries on magician |
| 22:41.05 | MnemonicPunk | lol |
| 22:41.15 | dcordes_ | yes you can, assuming they are armv5 or lower |
| 22:41.24 | MnemonicPunk | Woa. |
| 22:41.34 | MnemonicPunk | Now this is getting interesting. xD |
| 22:43.01 | dcordes_ | I would try openmoko-fso |
| 22:43.06 | NetRipper | you tried android? |
| 22:46.12 | MnemonicPunk | What's the difference between normal openmoko and fso? |
| 22:47.15 | dcordes_ | the old openmoko gui is complete gtk based. fso frontend is e17 based and has a complete new design of the whole radio stuff |
| 22:47.29 | dcordes_ | written in python |
| 22:47.51 | dcordes_ | I'm aiming to make that radio stuff (frameworkd) work with msm7* modems |
| 22:47.54 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 22:48.10 | dcordes_ | it's very modular and allows different types of radios |
| 22:48.29 | dcordes_ | there are already radio types for the moko modem, ezx phone modems, .. |
| 22:49.55 | cr2 | NetRipper: can you have a look at the 7201A clock patch by swetland in android git ? |
| 22:50.47 | *** join/#htc-linux dzo (n=dzo@121.98.128.127) |
| 22:51.41 | MnemonicPunk | I'd love to try out something with xfce or gpe. o.o |
| 22:51.43 | dcordes_ | dzo: hi |
| 22:51.46 | NetRipper | cr2, i've just updated haret.exe for readable 8x8 font on www.netripper.nl/raphael/, and it also add sleep()s to make sure framebuffer prints are displayed |
| 22:52.09 | dzo | dcordes_: hi |
| 22:52.48 | dcordes_ | dzo: can I ask you a question regarding vogue on screen keyboard? |
| 22:52.55 | dzo | sure |
| 22:53.10 | NetRipper | Kevin2, i have placed my latest diff for haret on www.netripper.nl/raphael/, could you review and merge with cvs? primarily please look at my #define VGA in fbwrite.h, which is not elegant... either Makefile should be changed to make a haret-vga.exe or another more elegant solution perhaps :) |
| 22:53.58 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 22:54.41 | MnemonicPunk | Pass it as a config file command maybe? Would fit in with the rest of the parameters. o.o |
| 22:54.54 | dcordes_ | dzo: polaris is very similar to kaiser as you might now, so we use it in the kaiser tree. but it uses the vogue panel and has no keyboard, so we use the vogue-ts and the on screen keyboard you use on vogue. in 2.6.24 (the htc-msm git code) it worked just fine. now I put polaris in 2.6.25 and everything works well, except for the on screen keyboard not being displayed |
| 22:55.42 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 22:55.43 | dcordes_ | dzo: it is possible to type things, it even gives the vibrate response they keyboard is just not showing. I guess I messed up something in msm_fb.c . can you have a look? |
| 22:56.05 | dzo | dcordes_: the fb driver changed a lot since 2.6.24, I'll have a look. |
| 22:57.01 | dcordes_ | dzo: http://pastebin.ca/1211003 it is based on the fb code we use in 2.6.25 |
| 22:58.02 | cr2 | NetRipper: what can be on 7201A, what is not on 7201 ?? |
| 22:58.02 | NetRipper | cr2, last commit on android git is 2 months ago |
| 22:58.08 | dcordes_ | NetRipper: look at the bottom in the gitweb, there are different branches |
| 22:58.18 | NetRipper | dcordes_, thx |
| 22:58.23 | NetRipper | blushes |
| 23:00.30 | cr2 | http://git.android.com/?p=kernel.git;a=shortlog;h=for-rmk |
| 23:00.52 | dzo | that code was for my old kbd, it won't work with 2.6.25 since the mdp handler runs in interrupt context. Have a look at the fb.c in the vogue tree, lines 194..226 |
| 23:00.58 | NetRipper | looking |
| 23:00.58 | cr2 | this one |
| 23:01.01 | cr2 | http://git.android.com/?p=kernel.git;a=commitdiff;h=3b55b35ea0edc29e41ac22ac0f391a5444899684 |
| 23:01.12 | cr2 | d0,d4,d8 and dc clocks |
| 23:01.16 | cr2 | what's that ? |
| 23:01.36 | NetRipper | you're asking me? |
| 23:01.54 | cr2 | yes. |
| 23:02.05 | cr2 | some are documented on the kaisermemorymap page |
| 23:02.29 | cr2 | we need to correlate them with this arch/arm/mach-msm/clock-7x01a.c file :) |
| 23:03.51 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 23:06.15 | dcordes_ | dzo: ok. http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=blob;f=drivers/video/msm/msm_fb.c;hb=refs/heads/htc-vogue ? |
| 23:06.46 | NetRipper | interesting stuff on that branch |
| 23:08.55 | MnemonicPunk | By the way: Where exactly is the ExtraStorage folder located on the HTC Magician? |
| 23:09.04 | dzo | dcordes_: yes |
| 23:09.09 | MnemonicPunk | I mean physically. Is that flash or ram? |
| 23:10.40 | *** join/#htc-linux snake_schule (n=s@dyndsl-091-096-099-184.ewe-ip-backbone.de) |
| 23:11.25 | dcordes_ | dzo: hm the code I use doesn't even have the function msmfb_start_dma. I'm not sure where to put changes from that area you said |
| 23:11.32 | *** part/#htc-linux wasup (n=Tech@208.47.103.106) |
| 23:13.39 | cr2 | dzo: are you using PCOM_GPIO_CFG to access gpios ? |
| 23:14.17 | dzo | dcordes_: you can't be using the latest fb.c then. |
| 23:16.49 | dcordes_ | ok I will close down the wifi thing for today and try making the latest fb.c work on kaiser/polaris. |
| 23:17.00 | dzo | cr2: no i don't think that works on our version of amss. The gpios have a config register that puts them into different modes, that could be what it's doing. |
| 23:17.00 | dcordes_ | vogue has no wifi, does it? |
| 23:17.14 | dzo | no, titan does though. |
| 23:17.33 | dcordes_ | ok so the kaiser sdio wifi driver might be interesting for titan as well |
| 23:17.48 | dzo | yes for sure, does it work? |
| 23:18.37 | dcordes_ | no I've been messing with it the last couple of hours |
| 23:18.59 | cr2 | NetRipper: it was HSUSB at 0xa08* |
| 23:19.18 | dzo | have you tried to get bluetooth working, vogue has that. |
| 23:19.23 | NetRipper | looks to see if he can make any sense |
| 23:19.37 | dcordes_ | I'm stuck with this "drivers/net/wireless/acx/common.c:2853: error: âstruct sk_buffâ has no member named âmacâ |
| 23:20.05 | dcordes_ | dzo: we did try bluetooth on kaiser, but w/o success |
| 23:20.54 | NetRipper | cr2, ah |
| 23:20.59 | dzo | OK, I'm still fighting with power collapse atm, I can get it to suspend and resume once but it always crashes the second time. |
| 23:21.25 | NetRipper | memory map you meant |
| 23:21.58 | cr2 | dcordes_: uartDMA looks very complex... (for BT) |
| 23:22.53 | dcordes_ | dzo: sounds strange. maybe barry has an idea about that. he has applied some powermanagement code for the kaiser |
| 23:23.00 | cr2 | #define MSM_AD5_PHYS 0xAC000000 |
| 23:23.04 | cr2 | what's that ? |
| 23:23.12 | dcordes_ | dzo: we can turn off the display, and I know he messed with clocks |
| 23:24.23 | dcordes_ | main cpu clocks |
| 23:25.18 | dzo | thats the simple stuff, power collapse powers down the CPU completely but it's hideously complex and has to do different things depending on how it's woken up. Also is almost impossible to trace. |
| 23:33.06 | dcordes_ | dzo: does the change from the msm_fb.c to your current msm_fb.c require any changing beyond that file? |
| 23:34.56 | NetRipper | anyone been able to compile the for-rmk branch kernel? |
| 23:35.00 | *** join/#htc-linux MnemonicPunk (n=atlantis@dslb-084-060-028-090.pools.arcor-ip.net) |
| 23:35.09 | NetRipper | for msm |
| 23:35.23 | dcordes_ | NetRipper: I have it locally, I can give it a try |
| 23:35.37 | dzo | there's nothing new in that branch, it's just repackaged for mainline i think. |
| 23:35.37 | cr2 | NetRipper: try it. for halibut |
| 23:35.48 | NetRipper | binary of latest commit dcordes_ ? |
| 23:36.08 | NetRipper | dcordes_, can you put the zImage online? |
| 23:36.23 | MnemonicPunk | Hm. My touchscreen isn't reacting as it should. =( |
| 23:36.34 | NetRipper | my cross compile toolchain is epic fail |
| 23:36.47 | dcordes_ | use OE toolchain then |
| 23:36.55 | NetRipper | OE? |
| 23:37.16 | dzo | dcordes_: it probably needs mdp.c too do a git diff to the current commit and you will see what needs changing. |
| 23:37.22 | dcordes_ | it's always good to have OE if you want some entertaining build problems anyway (OE is highly moving target) |
| 23:38.10 | NetRipper | dcordes_, can you just send me the zImage for now? |
| 23:38.10 | NetRipper | :) |
| 23:39.10 | NetRipper | if you have it anyway, i mean |
| 23:39.15 | NetRipper | you dont need to compile for me :p |
| 23:40.55 | imfloflo | with the 1.0 i must unzip data.gz and system.gz on the root of the SD ? and i can put the other files in sdcard/android/*???? |
| 23:41.29 | MnemonicPunk | Are you trying android right now? o.o |
| 23:41.34 | imfloflo | yes |
| 23:41.46 | imfloflo | i succes lauching it on my polaris |
| 23:41.53 | dzo | imfloflo: yes, don't uncompress data.gz and system.gz, first boot will do that. |
| 23:42.14 | imfloflo | ok i will retry without uncompress it |
| 23:42.27 | imfloflo | in the root of the SD ? |
| 23:42.32 | cr2 | hehe, it's interesting how they check 7201A vs. 7200A :) |
| 23:42.36 | dzo | yes |
| 23:45.22 | dcordes_ | dzo: how can I compare two local branches? |
| 23:46.54 | MnemonicPunk | How can I try Android on my phone? o.o Is there an image or something? |
| 23:47.13 | imfloflo | go lauching SDK 1 on SD on polaris with zimage of dcordes |
| 23:47.33 | MnemonicPunk | << Magician only =( |
| 23:47.53 | *** join/#htc-linux infernixxx (n=nix@ip164-156-173-82.adsl2.static.versatel.nl) |
| 23:48.04 | NetRipper | also read myhtcphone.com has android 1.0 cab, though it wont work on your phone, you can use its system and data |
| 23:48.09 | imfloflo | very long without uncompressing files on pc before |
| 23:48.49 | NetRipper | oh |
| 23:49.01 | imfloflo | style in haret screen |
| 23:49.56 | NetRipper | MnemonicPunk, on http://it029000.massey.ac.nz/vogue/ there's a 'new sdk version (1.0)' link, you can probably extract android from that |
| 23:50.35 | MnemonicPunk | I often get errors with nodes when I uncompress tar.gz files of root images and I guess I'm just lacking a command line option here. Anyone know? o.o |
| 23:50.45 | dcordes_ | good night |
| 23:50.49 | NetRipper | night dcordes_ |
| 23:50.50 | imfloflo | too long |
| 23:50.54 | imfloflo | god night |
| 23:50.57 | MnemonicPunk | Okay, I'll give it a try tomorrow. =) |
| 23:51.21 | imfloflo | i prefere uncompressing on my pc its faster |
| 23:52.27 | MnemonicPunk | I meant on the pc. o.o It fails to create nodes, though. Don't know why. |
| 23:52.45 | cr2 | NetRipper: we need the g1 kernel, not the sdk. |
| 23:53.02 | NetRipper | cr2, obviously, but that was for MnemonicPunk, he already has a kernel for his magician |
| 23:53.20 | MnemonicPunk | =) |
| 23:53.49 | NetRipper | cr2 , who's that swetland guy? htc developer? |
| 23:56.19 | imfloflo | for me is success to uncompressing and after transfer on the SD it works |
| 23:56.41 | imfloflo | 7minute remaning to transfert to the sd :( too long |
| 23:58.02 | cr2 | NetRipper: android :) |