| 00:02.31 | BruteFrce | someone got android running on an elf? |
| 00:03.02 | NetRipper | cr2, if you start hacking on kernel: http://pastebin.com/f38b6d9a7 |
| 00:03.26 | NetRipper | *really off* |
| 00:09.37 | *** join/#htc-linux datachaos (n=datachao@189.27.180.67.adsl.gvt.net.br) |
| 00:11.05 | cr2 | NetRipper: the g1 code is much better than halibut |
| 00:11.31 | cr2 | the pwm MDDI regs are undocumented |
| 00:20.41 | cr2 | "B table" is "htc_acoustic" |
| 00:24.40 | cr2 | uh, wifi and compass calibration data is packed into ATAGs |
| 00:26.23 | dcordes_ | isn't that the stuff used to put the nand partitions? |
| 00:26.46 | dcordes_ | what wifi data do you mean? |
| 00:27.02 | *** join/#htc-linux surgex (i=surge@pool-98-118-165-224.bflony.fios.verizon.net) |
| 00:27.50 | miknix | hi all |
| 00:28.15 | dcordes_ | hello miknix |
| 00:29.10 | miknix | dcordes_, my wall fluorescent lamp just destroyed my IR remote control sensor on the laptop |
| 00:29.17 | miknix | I'm really pissed off |
| 00:30.03 | dcordes_ | man that sucks |
| 00:30.46 | cr2 | the compass driver is a port of windows code. a hack :) |
| 00:31.01 | cr2 | dcordes_: it's difficult to say |
| 00:33.44 | *** join/#htc-linux jbailey (n=jeffbail@nat/google/x-2b01d5829721d4bc) |
| 00:34.05 | cr2 | hmm. trout: Disable DMA for wifi interface (for now) |
| 00:35.30 | dcordes_ | sounds hacky as well :) |
| 00:36.27 | dcordes_ | cr2, might that be in relation with the kind of dma problem msm7200 msm_sdcc suffers from? |
| 00:36.44 | cr2 | i think so |
| 00:40.58 | cr2 | Kevin2: now you can compare your egpio driver with the 'genuine' htc driver :) |
| 00:41.25 | cr2 | dcordes_: the SRAM layout is interesting on A cpus |
| 00:42.01 | cr2 | the g1 framebuffer is at 0x03800000 |
| 00:42.32 | cr2 | #define MSM_RAM_CONSOLE_BASE MSM_EBI_BASE + 0x6d00000 |
| 00:42.54 | cr2 | this is something for 'pd' :) |
| 00:46.43 | cr2 | dcordes_: don't see the wifi driver ? |
| 00:48.35 | dcordes_ | cr2, not yet. I was busy with the beagle |
| 01:01.34 | dcordes_ | cr2, checked out |
| 01:05.10 | dcordes_ | cr2, pwr_sink is power collapse for the standby? |
| 01:07.59 | dcordes_ | I didn't know there was direct cooperation with htc developers |
| 01:08.08 | dcordes_ | -> h2w e.g. |
| 01:09.03 | dcordes_ | serial debug on uart3 |
| 01:10.38 | dcordes_ | cr2, where are wifi and camera driver? |
| 01:11.28 | dcordes_ | I was hoping a wifi base was defined somewhere so I could put it in the acx sdio code |
| 01:13.33 | dcordes_ | swetland, any hints on where to find the wifi driver? |
| 01:16.13 | swetland | wifi driver is not out yet |
| 01:16.22 | swetland | it'll be its own repository (it's frickin' enormous) |
| 01:16.51 | swetland | should be up tomorrow sometime |
| 01:20.03 | dcordes_ | ok great |
| 01:23.59 | dcordes_ | swetland, camera remains binary? |
| 01:27.05 | swetland | yeah, the camera is a userspace binary |
| 01:27.17 | swetland | we're working with qualcomm to move much of it into a kernel driver in the future |
| 01:29.07 | dcordes_ | with qualcomm? I thought the cameras were seperate devices |
| 01:30.28 | dcordes_ | swetland, http://www.aptina.com/products/image_sensors/mt9t013d00stc/ something like that |
| 01:31.26 | cr2 | swetland: 6 pwm register names are missing |
| 01:31.40 | swetland | there are several pieces. image sensor driver, autofocus stuff, video front-end / dsp bits, etc |
| 01:38.56 | dcordes_ | good night |
| 01:39.15 | *** join/#htc-linux Magorium (i=Othello@gateway/tor/x-136a5a5e407826c8) |
| 01:49.23 | BruteFrce | sweetland device? |
| 01:56.35 | SanMehat | swetland: :) |
| 01:58.18 | Kevin2 | cr2: There's an official htc egpio driver now? |
| 01:58.30 | *** join/#htc-linux Magorium (i=Othello@gateway/tor/x-f2b691806dcac9f2) |
| 02:00.24 | cr2 | Kevin2: look here http://git.android.com/?p=kernel.git;a=blob;f=arch/arm/mach-msm/board-trout-gpio.c;h=f2564483f1cae1d1aef8201d11c4f8b28f53430a;hb=f771d538d7b979442096252f8c4772303191b511 |
| 02:01.07 | cr2 | SanMehat: will the dma be fixed for wifi ? |
| 02:01.30 | SanMehat | cr2: yeah but after i come back from my honeymoon :) |
| 02:01.36 | cr2 | lol |
| 02:02.05 | SanMehat | well you asked... |
| 02:02.11 | cr2 | :) |
| 02:03.59 | Kevin2 | cr2: I think ph5 got the egpio driver into mainline. The one ph5 and I worked on looks better than the google one - ours supported per-board configs. |
| 02:04.25 | SanMehat | Kevin2: We welcome your patches :) |
| 02:04.27 | cr2 | Kevin2: yes, i agree. |
| 02:04.58 | Kevin2 | SanMehat: You work for google? |
| 02:05.04 | SanMehat | Kevin2: yup |
| 02:07.24 | Kevin2 | SanMehat: I'm pulling latest kernel git to check if the driver really was committed to mainline. |
| 02:08.16 | SanMehat | Kevin2: well if you have any improvements / cleanups / etc, we *do* welcome patches and whatnot. |
| 02:09.10 | cr2 | SanMehat: the htc-egpio driver is used on 5+ htc devices now. |
| 02:09.26 | Kevin2 | SanMehat: Yeah - it's in drivers/mfd/htc-egpio.c |
| 02:09.39 | SanMehat | cr2: cool |
| 02:09.47 | SanMehat | someone file a bug against me to look at that :) |
| 02:10.20 | Kevin2 | SanMehat: Okay - thanks. I haven't had much time to look at linux on mobile phones recently. The only suggestion I can give relative to the htc-egpio is to use the mainline one (or send a patch to improve it if it doesn't work). |
| 02:10.57 | cr2 | hm. and vogue and kaiser too |
| 02:11.48 | cr2 | Kevin2: will you update hermes to 2.6.27 eventually ? |
| 02:12.07 | SanMehat | Kevin2: ok i'll look into it |
| 02:12.13 | Kevin2 | cr2: I haven't had any time to look at this stuff recently. I don't know when I will be able to. |
| 02:12.19 | cr2 | ok |
| 02:12.55 | cr2 | heh. 4:12 |
| 02:12.58 | cr2 | good night ;) |
| 02:13.06 | Kevin2 | cr2: Good night. |
| 02:14.35 | tmzt | Kevin2: can you tell me what these traces mean?: |
| 02:14.47 | tmzt | 002.388 TRACES a5500010(1 2 5 6 10 19 22)=63 |
| 02:14.56 | tmzt | 002.397 TRACES a5500010(1 2 5 6 10 19 22)=480405 |
| 02:15.12 | tmzt | from tracing ssp1 on pxa27x with haretconsole |
| 02:17.17 | Kevin2 | tmzt: It means the value at a5500010 was read at 63 and then at 480405. |
| 02:17.37 | BruteFrce | hows it goin tmzt? |
| 02:18.06 | tmzt | Kevin2: sorry, not sure what "read at" number is |
| 02:18.11 | Kevin2 | tmzt: What are you trying to do? If you're trying to watch the content going to and from the ssp port, you don't want to use TRACES. |
| 02:18.16 | tmzt | Kevin2: also, can I adapt scanssp |
| 02:18.34 | tmzt | Kevin2: or where do I get the insn values in haret raw trace? |
| 02:18.41 | Kevin2 | The register at a5500010 had a value of 63. Then 9 ms later it had a value of 480405. |
| 02:18.44 | tmzt | Kevin2: ok, that's what I want to do |
| 02:19.15 | tmzt | Kevin2: oh, I wondered what the numbers in () where, bits? |
| 02:20.07 | Kevin2 | tmzt: Yes. However, the bits that have changed look really odd - are you sure you copied them correctly? |
| 02:20.23 | tmzt | hold on, I can paste the haretconsole/raw |
| 02:20.25 | Kevin2 | If you want to watch what is going to/from the ssp port, you really want to use mmutrace. |
| 02:20.57 | tmzt | I plugged/unplugged the pc and got output that differed each time |
| 02:21.22 | tmzt | I mean usb to pc |
| 02:22.30 | tmzt | http://pastebin.no/60885 |
| 02:22.35 | tmzt | haretconsole |
| 02:25.33 | Kevin2 | tmzt: Hrmm. Looks odd. Do you have the raw traffic? |
| 02:28.07 | tmzt | http://pastebin.no/60887 |
| 02:44.33 | tmzt | Kevin2: is you want to give me the mmutrace instructions, I can test with pc/charger |
| 02:44.41 | tmzt | Kevin2: if |
| 02:54.08 | Kevin2 | tmzt: Ahh - I missed a line on the initial paste. The bit defs make sense now. |
| 02:54.25 | Kevin2 | tmzt: Which register has the data in it? Is it a5500010? |
| 02:54.34 | tmzt | ssp1 read/write register |
| 02:54.38 | tmzt | from pxa specs |
| 02:54.52 | *** join/#htc-linux Sni1 (n=igor@195.208.14.74) |
| 02:54.52 | Kevin2 | I don't have the specs handy. Is it a5500010? |
| 02:55.04 | tmzt | that's virt I think |
| 02:55.42 | tmzt | 0x4100_0010 |
| 02:56.16 | Kevin2 | So, run "dump mmu 2 0x41000010" |
| 02:56.32 | tmzt | physical for mmu? |
| 02:56.41 | Kevin2 | Then run "addlist mmutrace 0xabc" for each address that was reported by the above command. |
| 02:56.51 | tmzt | oh, the mappings |
| 02:56.53 | tmzt | ok |
| 02:57.16 | Kevin2 | tmzt: Yes - dump mmu takes physical addresses. You then want to add all the virtual addresses to mmutrace. |
| 02:57.37 | tmzt | 49000000 85500000 a5500000 |
| 02:57.46 | tmzt | small 4k, 1mb, 1mb |
| 02:58.20 | Kevin2 | You probably don't have to worry about the 4k one. So you want "addlist mmutrace 0xa5500010" and "addlist mmutrace 0x85500010". |
| 02:58.22 | tmzt | the base address of xxxx0010 |
| 02:58.28 | tmzt | or |
| 02:58.30 | tmzt | ok |
| 02:59.01 | tmzt | wi? |
| 02:59.07 | Kevin2 | Yes - then run "wirq". |
| 03:04.49 | tmzt | Kevin2: ok, I found ldr/str block |
| 03:08.26 | *** part/#htc-linux Sni1 (n=igor@195.208.14.74) |
| 03:15.14 | *** join/#htc-linux ali12341 (n=al@robotfuzz.co.uk) |
| 03:23.04 | tmzt | Kevin2: how do I decode this/fix scanssp.py to do it? |
| 04:01.03 | tmzt | Kevin2: http://pastebin.no/60917 |
| 04:27.33 | Kevin2 | tmzt: scanssp.py looks like one of my old scripts - it's only for apache. |
| 04:29.10 | Kevin2 | I found the documentation for the dac chip that was connected to the ssp port and from that wrote a decode program. |
| 04:29.20 | tmzt | I can see re matches for insn, I just need to get the bytes |
| 04:29.43 | tmzt | but it looks like there a 4 bytes written at once, then read back? |
| 04:31.33 | Kevin2 | That's how ssp works - You have to write something in order to read something. |
| 04:31.36 | tmzt | 001306: 003f4455: mmutrace 8014c87c: e5823010(str) a5500010 01b0701c (00000000) |
| 04:31.39 | tmzt | 001306: 003f4480: mmutrace 8014c8a4: e5923010(ldr) a5500010 00480404 (00000000) |
| 04:31.43 | tmzt | 001306: 003f458d: mmutrace 8014c87c: e5823010(str) a5500010 82480405 (00000000) |
| 04:31.46 | tmzt | 001306: 003f45ad: mmutrace 8014c8a4: e5923010(ldr) a5500010 00480404 (00000000) |
| 04:31.49 | tmzt | 001306: 003f4658: mmutrace 8014c87c: e5823010(str) a5500010 07ef74d0 (00000000) |
| 04:31.52 | tmzt | 001306: 003f4678: mmutrace 8014c8a4: e5923010(ldr) a5500010 007bcf7f (00000000) |
| 04:31.55 | tmzt | 001306: 003f4714: mmutrace 8014c87c: e5823010(str) a5500010 847bcf7e (00000000) |
| 04:31.58 | tmzt | 001306: 003f4734: mmutrace 8014c8a4: e5923010(ldr) a5500010 007bcf7f (00000000) |
| 04:32.01 | tmzt | 001306: 003f47d9: mmutrace 8014c87c: e5823010(str) a5500010 a4000063 (00000000) |
| 04:32.04 | tmzt | 001306: 003f47f9: mmutrace 8014c8a4: e5923010(ldr) a5500010 00000060 (00000000) |
| 04:32.07 | tmzt | 001306: 003f4898: mmutrace 8014c87c: e5823010(str) a5500010 a8400000 (00000000) |
| 04:32.10 | tmzt | 001306: 003f48b8: mmutrace 8014c8a4: e5923010(ldr) a5500010 0048320b (00000000) |
| 04:32.13 | tmzt | :no |
| 04:33.23 | tmzt | Kevin2: sorry, wrong screen instance. I need spi though, not ssp |
| 04:33.42 | tmzt | the first two sets repeat with some time in between |
| 04:33.50 | Kevin2 | Same thing. Both are synchronous protocols - which means transmits happen at the same time as receives. |
| 04:34.06 | Kevin2 | So, in order to receive something, one has to write something. |
| 04:34.18 | Kevin2 | So, you should always see a str followed by an ldr. |
| 04:35.58 | tmzt | what are the fields in that dump? |
| 04:38.00 | Kevin2 | haretconsole should make that more apparent. The lines with "str" is a write to the register (eg, with a value of a8400000). The lines with "ldr" show the value read from the register (eg, 0048320b). |
| 04:38.50 | tmzt | ok, the number before the nuemonic?? is the arm instruction, what is before that? |
| 04:40.00 | tmzt | that's the virtual address trapped? |
| 04:43.17 | tmzt | mmutrace 8014c87c: *(str) |
| 04:43.52 | *** join/#htc-linux goxboxlive (n=goxboxli@195.159.97.196) |
| 04:50.30 | Kevin2 | tmzt: Read through the info at: http://www.handhelds.org/moin/moin.cgi/HaRET_20Documentation |
| 04:51.16 | Kevin2 | The address after mmutrace is the address of the insn that caused the trap. |
| 04:51.37 | tmzt | I'm looking at the haretconsole |
| 04:51.55 | tmzt | it's starting to make sense, but it's doing spi before getting any gpio I can see |
| 04:57.40 | *** join/#htc-linux the_sys0p (n=the_sys0@cpe-75-85-249-111.bak.res.rr.com) |
| 05:14.05 | *** join/#htc-linux silven_ (n=zmc@adsl-64-237-232-217.prtc.net) |
| 05:58.52 | *** join/#htc-linux kiozen (n=oeichler@p54929FAB.dip0.t-ipconnect.de) |
| 06:42.49 | *** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz) |
| 06:53.28 | *** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz) |
| 06:56.16 | *** join/#htc-linux ginge (n=ginge@host81-149-155-58.in-addr.btopenworld.com) |
| 07:09.15 | *** join/#htc-linux BabelO (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 07:29.14 | *** join/#htc-linux petur (n=petur@kr-lun-155-153-233-83.3.cust.bredband2.com) |
| 07:30.59 | petur | So, people are holding theire final versions of the G1 |
| 07:31.10 | petur | but I still can't find a raw dump of the Android |
| 07:31.46 | NetRipper | haven't seen raw dumps yet, just filesystem dumps |
| 07:31.59 | NetRipper | i'm actually curious if we can get hold of the /proc/config.gz |
| 07:32.05 | NetRipper | or without .gz |
| 07:32.06 | NetRipper | ;) |
| 07:33.23 | petur | why wouldnt we? |
| 07:46.03 | *** join/#htc-linux LanceHaig (i=c2d8ed19@foresight/member/lhaig) |
| 07:48.20 | *** part/#htc-linux LanceHaig (i=c2d8ed19@foresight/member/lhaig) |
| 07:48.33 | *** join/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 07:49.26 | BabelO_ | morning |
| 07:51.01 | swetland | yawns |
| 07:53.20 | BabelO_ | hello swetland |
| 08:01.37 | BabelO_ | kiozen: do you plan to add elevation to garmin too ? :) |
| 08:02.31 | NetRipper | petur, haven't seen it in a filesystem dump yet |
| 08:04.23 | *** join/#htc-linux petur_ (n=petur@kr-lun-155-153-233-83.3.cust.bredband2.com) |
| 08:14.08 | *** join/#htc-linux lpotter (n=ljp@CPE-124-191-144-181.vic.bigpond.net.au) |
| 08:18.20 | kiozen | BabelO_: that is a bit tricky |
| 08:18.39 | kiozen | BabelO_: to do relief shading you need elevation lines, tooo |
| 08:32.05 | swetland | if you check out the android-msm-htc-2.6.25 branch |
| 08:32.09 | swetland | and make msm_defconfig |
| 08:32.29 | swetland | your .config should be identical to the configuration used in the production g1 kernel |
| 08:35.23 | BabelO_ | kiozen: ok, i try img's yesturday, works well :) |
| 08:38.11 | kiozen | BabelO_: on win32, too? |
| 08:38.41 | BabelO_ | kiozen: i can try right now |
| 08:38.53 | BabelO_ | wait few minutes i have nothing to do |
| 08:39.17 | kiozen | has too much to do :( |
| 08:41.00 | *** join/#htc-linux Marex (n=vasum7am@auryn.karlin.mff.cuni.cz) |
| 08:43.03 | BabelO_ | kiozen: you say to me to not include hard coded path, i 'm agree, but remove this one from CMakeFileList.txt :) C:/Programme/FWTools2.2.6 |
| 08:43.37 | kiozen | BabelO_: I didn't say :) it was andrew |
| 08:44.13 | kiozen | BabelO_: but you are right, we need a FWTool check to get the path |
| 08:44.15 | BabelO_ | kiozen: lol i know :) anyway the path i have added is not used, and cmake can't find it |
| 08:44.33 | BabelO_ | kiozen: you have to set it in your default environment |
| 08:45.00 | kiozen | BabelO_: better to do something like in the unix section |
| 08:48.25 | BabelO_ | kiozen: it seem your change to platform.h are ok, have to try with some img from yesturday |
| 09:06.52 | BabelO_ | kiozen: how can i do if i ve to img files ? |
| 09:06.57 | BabelO_ | sorry tdb file |
| 09:07.41 | kiozen | BabelO_: tdb register to the map collection like qmap files |
| 09:08.11 | BabelO_ | kiozen: ok io found a tdb |
| 09:08.16 | BabelO_ | it works under windows too :) |
| 09:08.34 | kiozen | BabelO_: how fast is zooming and moving |
| 09:08.37 | kiozen | ? |
| 09:09.33 | par | on the android raw dump, would be easy to use buzz lightyear's tool ... just set the sd card up with it and then plug it into the device.. it dumps the rom straight to the SD card |
| 09:10.10 | BabelO_ | kiozen: hmm, with 100km zoom, it is slow to move with 50km it is better, below is is acceptable ... zoom in /out is fast |
| 09:10.59 | kiozen | BabelO_: well that is osm, too much senseless stuff on low detail level |
| 09:11.15 | BabelO_ | kiozen: no it is not an OSM ima map :) |
| 09:11.49 | kiozen | BabelO_: ok :) |
| 09:12.13 | BabelO_ | kiozen: i ve to try OSM |
| 09:15.30 | *** join/#htc-linux dante_ (n=dante@host176-48-static.15-79-b.business.telecomitalia.it) |
| 09:24.02 | cr2 | par: ??? |
| 09:24.13 | cr2 | kiozen: how do i load .tbd in GT ? |
| 09:25.03 | kiozen | cr2: "load map" and set the file filter to *.tdb |
| 09:25.27 | kiozen | if you do not have the filter you forgot to enable garmin support in cmake |
| 09:25.33 | par | cr2: wasnt' someone looking for a raw dump from the device? |
| 09:35.45 | BabelO_ | cr2: i explain you yesturday :) |
| 09:36.07 | BabelO_ | kiozen: once i choose img for the tdb, how can i change img files iwant to load ? |
| 09:37.22 | kiozen | BabelO_: you picked the wrong basemap? |
| 09:37.50 | kiozen | BabelO_: in this case you have to edit ~/.config/QLandkarteGT/QLandkarteGT.conf |
| 09:41.23 | *** join/#htc-linux silven_ (n=zmc@adsl-64-237-230-239.prtc.net) |
| 09:41.45 | cr2 | kiozen: ok |
| 09:45.39 | *** join/#htc-linux BruteFrce (n=IrcNet@m7a0e36d0.tmodns.net) |
| 09:48.30 | BabelO_ | kiozen: no .conf under windows :( should be in regedit |
| 09:49.18 | kiozen | BabelO_: arrrg, right, QSettings is operating the registry on windows. Happy hacking :) |
| 09:54.26 | *** join/#htc-linux mooky (n=suzannet@unaffiliated/mooky) |
| 09:57.31 | BabelO_ | lol |
| 10:03.43 | *** join/#htc-linux LunohoD (n=alex@e180070085.adsl.alicedsl.de) |
| 10:04.08 | cr2 | kiozen: it took 20 sec on start, to dump the debug info to stdout :) |
| 10:07.03 | cr2 | hmm. and i don't see anything on the screnn |
| 10:12.19 | kiozen | cr2: indexing takes some time, try F1->F3 to center the map |
| 10:21.17 | cr2 | kiozen: still nothing |
| 10:21.30 | cr2 | kiozen: i'll try to enter the coordinates directly |
| 10:22.06 | cr2 | hm. don't know how |
| 10:24.34 | *** join/#htc-linux kaze_ (n=kaze@ABordeaux-152-1-42-72.w82-125.abo.wanadoo.fr) |
| 10:26.07 | cr2 | kiozen: is it possible to add a callback for clicking on the coordinates in the lower right corner ? |
| 10:28.24 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) |
| 10:31.06 | *** join/#htc-linux sni1 (n=igor@92.124.16.18) |
| 10:32.33 | *** part/#htc-linux sni1 (n=igor@92.124.16.18) |
| 10:34.02 | kiozen | cr2: hehe, you ever wondered what the search tool view is good for ? :) |
| 10:35.17 | kiozen | cr2: you try osm map right? |
| 10:38.06 | cr2 | yes |
| 10:38.22 | dcordes | kiozen, do you have debian packages with 3d plot? |
| 10:38.31 | dcordes | still didn't succeed with the build |
| 10:38.54 | cr2 | kiozen: which search tool ? |
| 10:39.07 | kiozen | dcordes: I don't do any packages, sorry, that would consum too much time |
| 10:39.13 | NetRipper | swetland, ok thanks |
| 10:39.53 | *** join/#htc-linux ImCoKeMaN (n=imcokema@pool-96-249-151-72.hrbgpa.fios.verizon.net) |
| 10:39.54 | kiozen | cr2: left hand tab widget. The is an input line with the nice and friendly word "Search" |
| 10:40.06 | dcordes | NetRipper, something must have went wrong with the last commit since I don't see any updates |
| 10:40.19 | kiozen | any google compliant coord format will do |
| 10:40.31 | NetRipper | dcordes, commit? |
| 10:40.44 | kiozen | a link to the internet is needed, of course |
| 10:40.49 | dcordes | NetRipper, yes, do you see any changes? |
| 10:40.54 | NetRipper | dcordes, where? |
| 10:41.08 | NetRipper | dcordes, android git? |
| 10:41.25 | dcordes | NetRipper, in the android-msm-htc-2.6.25 branch of google git, yes |
| 10:41.44 | NetRipper | i fetched it 4 hours ago |
| 10:42.35 | NetRipper | dcordes, last update on that head was 4 days ago |
| 10:43.07 | NetRipper | dcordes, what update were you expecting? |
| 10:43.18 | dcordes | NetRipper, see the log, 10:30 |
| 10:43.43 | NetRipper | dcordes, yes i believe he didn't just check it in, it was a reference to use that defconfig, |
| 10:44.11 | NetRipper | that defconfig was updated 4 days ago as well |
| 10:44.20 | NetRipper | so they probably updated the defconfig along with the trout |
| 10:44.24 | dcordes | well the statement can't be true because how would you get the production kernel with no wifi? |
| 10:44.43 | NetRipper | wifi is a module on G1 anyway, not built into kernel |
| 10:46.16 | *** join/#htc-linux Marnux (n=Marnux@83.101.56.57) |
| 10:46.29 | NetRipper | was using an older defconfig anyway, so made sense to me what he said :p |
| 10:48.37 | dcordes | nvm guess wifi code will hit git soon |
| 10:50.23 | dcordes | NetRipper, did you try your things on top of the trout code? |
| 10:50.45 | NetRipper | yes, need to make the same adjustment in irq.c before it continues booting, but haven't checked where it hangs now |
| 10:50.59 | NetRipper | i'm @ work now, so can't play until tonight |
| 10:54.38 | dcordes | so it didn't fix magivally? |
| 10:54.50 | dcordes | s/magivally/magically/ |
| 10:56.44 | NetRipper | dcordes, no it still hang while initializing the DEBUG_TIMER and UART interrupts |
| 10:58.23 | kiozen | cr2: found it? |
| 11:00.36 | kiozen | cute http://www.spiegel.de/fotostrecke/fotostrecke-36321.html |
| 11:01.38 | dcordes | kiozen, I've heard it runs a 2.4 kernel |
| 11:01.59 | dcordes | but it's tiny. on other pictures it looks like a useless zaurus clone |
| 11:02.22 | kiozen | just need a gps and a rugged version :) |
| 11:03.02 | dcordes | don't like lack of usb 2.0 |
| 11:03.18 | dcordes | I'd rather go for openpandora |
| 11:03.24 | kiozen | it's funny, pda's started like this and died because nobody wanted them like this. |
| 11:03.30 | dcordes | then you cna run GT on it :) |
| 11:05.20 | kiozen | dcordes: looks nice, too :) |
| 11:07.19 | dcordes | omap3* is ace. I have it on the beagle board and I use will get a pandora. |
| 11:08.08 | kiozen | dcordes: btw what keeps you from compiling GT? |
| 11:09.40 | dcordes | it doesn't like my gdal |
| 11:09.46 | dcordes | CCreateMapGeoTiff.cpp:(.text+0x90ba): undefined reference to `OGRSpatialReference::exportToProj4(char**) const' |
| 11:10.21 | kiozen | dcordes: you compiled GDAL? or binary? |
| 11:10.23 | dcordes | I get plenty of those. but it is after I made a wrong symlink, because it didn't find gdal at all |
| 11:10.31 | dcordes | no, installed debian package |
| 11:10.48 | dcordes | in debian testint/lenny |
| 11:10.56 | dcordes | sorry, can't type today |
| 11:10.59 | kiozen | compile it yourself, looks like the compiled it withput OGR support |
| 11:13.04 | dcordes | ok |
| 11:13.37 | kiozen | check after ./configure if ogr is enabled |
| 11:14.04 | dcordes | where to find the official latest source? |
| 11:14.30 | kiozen | http://www.gdal.org/ |
| 11:16.38 | dcordes | isn't it possible that I symlinked wrong? |
| 11:17.35 | kiozen | don't know I never play with that kind of stuff :) to much danger to f.... the system |
| 11:18.18 | kiozen | the error message simply says that the linker can't find that symbol in the lib |
| 11:20.46 | dcordes | <PROTECTED> |
| 11:21.15 | kiozen | nice :) |
| 11:28.29 | dcordes | kiozen, ok that solved it |
| 11:29.03 | kiozen | nothing competes a fine manual compilation of good software |
| 11:29.11 | dcordes | oh that was early. still the same error |
| 11:29.16 | kiozen | LOL |
| 11:29.53 | dcordes | maybe I should remove that symlink and rerun ccmake to configure |
| 11:30.00 | kiozen | try to find all libgdal in the usual paths |
| 11:30.12 | kiozen | and remove the bad ones |
| 11:30.23 | kiozen | bad == older |
| 11:30.33 | dcordes | /usr/lib/libgdal.so: symbolic link to `/usr/lib/ogdi/libgdal.so' |
| 11:30.54 | dcordes | /usr/lib/libgdal1.5.0.so.1: symbolic link to `libgdal1.5.0.so.1.12.2' |
| 11:31.43 | kiozen | the first one is strange |
| 11:36.01 | dcordes | kiozen, how must I invoke the cmake configuration again? |
| 11:36.06 | *** join/#htc-linux NiZoX (n=none@2a01:e35:8a13:a2b0:21c:c0ff:fe25:ff68) |
| 11:37.08 | kiozen | http://www.qlandkarte.org/index.php?option=com_content&view=article&id=3&Itemid=4 |
| 11:40.23 | dcordes | kiozen, thanks. by default, GDAL_INCLUDE var points to /usr/include but there's no gdal here? |
| 11:45.32 | dcordes | kiozen, where must it point? |
| 11:46.21 | kiozen | where did "make install" copy them too? |
| 11:46.50 | kiozen | search for gdal.h |
| 11:48.43 | dcordes | <PROTECTED> |
| 11:48.49 | dcordes | next problem |
| 11:49.43 | kiozen | is qt4.4 installed on the system? with all development stuff? |
| 11:50.04 | dcordes | well I installed all of the qt4,4 and X dev stuff which I found |
| 11:50.12 | dcordes | poor apt :( |
| 11:51.43 | kiozen | hm, that's frustrating. |
| 11:51.57 | kiozen | what about opensuse :) |
| 11:52.13 | dcordes | noooo |
| 11:53.08 | kiozen | but that stuff works at least |
| 11:56.46 | dcordes | kiozen, it must be me, not debian. |
| 11:59.03 | kiozen | lol, that reminds me on my last kubuntu test run. I gave up, because I could not install a working dev system and did not want to compile everything from scratch |
| 12:01.21 | dcordes | I gave up on ubuntu. now using debian on both of my x86 systems |
| 12:04.02 | dcordes | kiozen, do I want garmin ? |
| 12:04.47 | kiozen | dcordes: hard to tell. nice hw lousy sw |
| 12:05.02 | kiozen | what do you want to do with it? |
| 12:05.31 | dcordes | idk, that's why I ask. maybe I want to import map data or so at some point |
| 12:05.54 | dcordes | aarrgh CMakeFiles/qlandkartegt.dir/CTrack3DWidget.o: In function `CTrack3DWidget::mouseDoubleClickEvent(QMouseEvent*)': |
| 12:08.27 | kiozen | dcordes: with a garmin you can just use vector maps. This is quite ok for city areas and biking. |
| 12:08.51 | kiozen | dcordes: if you do hikes in the mountains vector maps are of no real use. |
| 12:09.23 | dcordes | ok then I will hike mountains for now ;) you know what that 3dWidget error is? I guess I need some extra headers for the 3d plot ? |
| 12:10.20 | kiozen | dcordes: can't tell, the one line is not enought information |
| 12:11.18 | dcordes | kiozen, can you have a look http://rafb.net/p/VmFC0p13.html |
| 12:11.55 | cr2 | kiozen: hehe. all the widgets below GPS device were invisible |
| 12:13.28 | kiozen | dcordes: missing OpenGL stuff |
| 12:14.38 | kiozen | cr2: right, I never manged to do a propper initial setup of a splitter |
| 12:15.08 | dcordes | kiozen, do you know which suse package provides that? |
| 12:16.52 | kiozen | dcordes: should be the usual opengl stuff, think in my system this is provided by nvidia |
| 12:23.09 | dcordes | kiozen, the kernel module nvidia?? |
| 12:25.11 | kiozen | dcordes: no, if i remember right the driver installation will install some opengl stuff too |
| 12:25.39 | kiozen | elsa mesa (sw opengl) is the way to go :( |
| 12:26.25 | dcordes | well I have the open source radeon driver installed here. so this machine will be capable of hardware rendering |
| 12:27.40 | kiozen | that should go with some opengl stuff too, can't tell, on suse it just works |
| 12:37.45 | cr2 | kiozen: the geometry/layout management is strange |
| 12:38.43 | kiozen | cr2: why? |
| 12:39.59 | cr2 | the gps device combobox takes 1/3 of the vertical space, and the text in the upper part is cut at top and booton |
| 12:40.59 | cr2 | NetRipper: i think i'll create the DreamGPIO wiki page |
| 12:41.36 | cr2 | NetRipper: the clock api needs a replacement too |
| 12:41.49 | kiozen | cr2: the left hand GUI is 3 splitters, you have to move the two splitter handles to the correct position for your screen GT will remember the location next time |
| 12:42.03 | kiozen | cr2: qt does not provide better splitter setup |
| 12:43.16 | cr2 | ok |
| 12:44.12 | *** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix) |
| 12:46.39 | dcordes | kiozen, can't figure what to install in order to sort the undefined gluUnProject problem |
| 12:48.15 | kiozen | dcordes: I have to admit that I am no opengl expert either, maybe BabelO_ can tell |
| 12:48.35 | dcordes | BabelO, are you around? |
| 12:50.14 | BabelO_ | dcordes: yes |
| 12:50.55 | dcordes | do you know what's missing here? http://rafb.net/p/VmFC0p13.html |
| 12:57.22 | ginge | gl/glut libs |
| 12:58.07 | ginge | libgl-dev in poobuntu not sure on the rest of the distros |
| 12:59.16 | kiozen | likes the word "poobuntu" :P |
| 13:03.56 | *** join/#htc-linux timebomb (n=timebomb@e176108158.adsl.alicedsl.de) |
| 13:12.49 | *** join/#htc-linux datachaos (n=datachao@189.6.250.25) |
| 13:20.19 | *** join/#htc-linux LanceHaig (i=c2d8ed19@foresight/member/lhaig) |
| 13:27.52 | BabelO_ | dcordes: yes ginge is right, you are missing Glut |
| 13:30.25 | kaze_ | any kaiser kernel compiled with latest android git patches? |
| 14:27.38 | *** join/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 14:28.50 | *** join/#htc-linux goxboxlive_ (n=goxboxli@139.84-48-212.nextgentel.com) |
| 14:38.10 | *** join/#htc-linux Dinde (i=kayser@81-65-176-209.rev.numericable.fr) |
| 14:59.55 | *** join/#htc-linux chile (n=chatzill@AToulouse-158-1-40-214.w90-50.abo.wanadoo.fr) |
| 15:13.54 | *** join/#htc-linux Magorium (i=Othello@gateway/tor/x-0af9f128518be856) |
| 15:15.58 | *** join/#htc-linux raven_coda (n=raven_co@64.50.35.156) |
| 15:16.46 | *** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde) |
| 15:17.11 | chiledog | greetings all |
| 15:17.57 | jbailey | Vaguely off-topic for here, but we just did the code drop for Android. Details in #android. |
| 15:19.01 | chiledog | thank you jbailey |
| 15:22.50 | chiledog | could anyone possibly tell me what the gizard release on http://tinderbox.dev.gentoo.org/embedded/linwizard/ is for ? |
| 15:23.09 | chiledog | it seems to work great for me on htc elfin |
| 15:23.55 | chiledog | what i wanted to ask was how can i find out what kernel was used ? |
| 15:25.42 | chiledog | i tried compiling from the linwizard git repo but don't get the same results |
| 15:26.03 | swetland | http://android.kernel.org/?p=platform/system/wlan/ti.git;a=summary |
| 15:26.11 | swetland | probably of much interest to the folks here |
| 15:26.56 | tmzt | chiledog: the kernel you used has the support for elf/elfin, the initrd should work on any arm but at least omap850 |
| 15:27.08 | tmzt | chiledog: you can ask in #linwizard about that |
| 15:27.19 | chiledog | thanks tmzt |
| 15:27.30 | tmzt | chiledog: if you want to compile your own kernel (zImage) you need to see what was changed |
| 15:27.44 | tmzt | chiledog: at least the omapfb settings, maybe keys |
| 15:28.48 | chiledog | is there a way to take the zImage and decompile it ? |
| 15:30.02 | chiledog | and yes omapfb settings (at least in the default.txt) had been changed |
| 15:30.26 | tmzt | swetland: what is the git:// for that? |
| 15:31.04 | swetland | git://android.kernel.org/platform/system/wlan/ti.git |
| 15:31.14 | tmzt | thanks |
| 15:31.19 | swetland | enjoy! |
| 15:38.56 | cr2 | swetland: .ko in the repository ? |
| 15:39.09 | swetland | sources are there |
| 15:39.38 | swetland | we don't ship binary modules since they're petty specific to the kernel binary |
| 15:39.43 | *** join/#htc-linux chiledog (n=chatzill@AToulouse-158-1-40-214.w90-50.abo.wanadoo.fr) |
| 15:40.18 | cr2 | swetland: i'm surprised that it's there |
| 15:40.58 | swetland | what, there is one? |
| 15:41.18 | cr2 | yes |
| 15:41.33 | cr2 | [platform/system/wlan/ti.git] / sta_dk_4_0_4_32 / fw / |
| 15:42.38 | swetland | argh. we'll fix that. |
| 15:42.47 | cr2 | ok |
| 15:44.11 | cr2 | hehe. md5sum does not match g1 |
| 15:45.03 | BabelO_ | who do this code ? i mean the wifi ? |
| 15:45.23 | *** join/#htc-linux Bendy_ (n=sniveri@bb-81-175-215-228.dsl.phnet.fi) |
| 15:46.06 | cr2 | Copyright © Texas Instruments Incorporated (Oct 2005) |
| 15:46.18 | cr2 | Copyright © Google Inc (Feb 2008) |
| 15:46.59 | cr2 | BabelO: wifi on arte ? |
| 15:47.23 | BabelO_ | cr2: yes :) should be the same -> date 29-Dec-2004 |
| 15:48.37 | BabelO_ | cr2: poor, busaccess is done via shm |
| 15:50.09 | cr2 | BabelO: better than no driver ;) |
| 15:50.23 | BabelO_ | cr2: yes agreee |
| 15:50.40 | BabelO_ | cr2: look at whalBus_Config in busaccess/shmbus.c |
| 15:50.53 | NetRipper | cr2, you mean clock api needs to be replaced with vogue's? |
| 15:50.58 | cr2 | the main msm tree also needs a lot of patching here and there. |
| 15:51.19 | cr2 | NetRipper: yes, something like that. |
| 15:51.19 | BabelO_ | (SDIO_Stop/SDIO_Start) :) |
| 15:51.59 | *** join/#htc-linux kiozen (n=oeichler@rgnb-5d87c658.pool.einsundeins.de) |
| 15:52.02 | cr2 | NetRipper: the proc_comm() should be overridden |
| 15:53.09 | cr2 | for clock control |
| 15:53.11 | NetRipper | ok |
| 15:53.20 | cr2 | vogue is a bit hacky |
| 15:53.25 | BabelO_ | cr2: ok i found the right file tnetw_sdio.c :) |
| 15:53.38 | BabelO_ | include mmc_omap_api.h |
| 15:53.45 | cr2 | because it patched the drivers directly, instead of overriding the clock api itself |
| 15:53.51 | BabelO_ | with a ifdef MSM7000 just before |
| 15:54.15 | cr2 | NetRipper: we may do #ifdef ANDROID |
| 15:54.32 | cr2 | and #else for direct clock control |
| 15:55.57 | *** join/#htc-linux imfloflo (i=58b44979@gateway/web/ajax/mibbit.com/x-a3fb2867f4fecff1) |
| 15:55.57 | *** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net) |
| 15:56.19 | imfloflo | drcode have you see this http://source.android.com/ |
| 15:56.22 | NetRipper | i can transfer functions from vogue to htc, but i have no idea what i'm actually doing |
| 15:56.30 | *** join/#htc-linux exco (n=exco@e181069079.adsl.alicedsl.de) |
| 15:57.41 | *** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
| 15:57.54 | NetRipper | i mean like i can't check whether code makes any sence |
| 15:57.57 | NetRipper | sense |
| 15:59.49 | imfloflo | hey guys have you make progress on porting android ? since the source are avalaible now |
| 15:59.50 | cr2 | NetRipper: the clk_api is simply commented out |
| 16:00.14 | cr2 | imfloflo: android itself ? |
| 16:00.17 | druidu | hey |
| 16:00.23 | druidu | check out android git |
| 16:00.25 | druidu | new stuff :) |
| 16:00.41 | cr2 | druidu: you have diamond ? |
| 16:00.45 | druidu | yes |
| 16:00.50 | druidu | they added a new branch |
| 16:00.55 | cr2 | druidu: with 128M ram ? |
| 16:01.02 | druidu | android-msm-htc |
| 16:01.06 | druidu | 192 I think |
| 16:01.14 | cr2 | 128+64 ? |
| 16:01.27 | BabelO_ | cr2: not for you or me :) -> "To build the Android source under Linux, you will need Ubuntu" |
| 16:01.38 | druidu | yes, 192 |
| 16:02.02 | cr2 | druidu: it's 64MB sram, which you don't use anyway. yet. |
| 16:02.09 | cr2 | and 128 SDRAM |
| 16:02.41 | imfloflo | http://source.android.com/ |
| 16:02.58 | druidu | doesn't say |
| 16:03.08 | druidu | it's like: RAM: 128 MB FLASH: 256 |
| 16:03.12 | imfloflo | ./download/ |
| 16:03.27 | druidu | wohooo |
| 16:03.27 | druidu | :) |
| 16:03.29 | ali12341 | 2GB... jeez |
| 16:03.39 | druidu | tonight we're gonna have android |
| 16:03.46 | druidu | on both diamond and raphael |
| 16:03.49 | druidu | I can see it! :) |
| 16:03.57 | cr2 | imfloflo: we are doing the kernel side, because it's the first thing to do. |
| 16:04.15 | druidu | The source is approximentely 2.1GB in size. You will need 6GB free to complete the build. |
| 16:04.22 | druidu | wow, they do have a lot of stuff there |
| 16:04.26 | HardDisk_WP | lol |
| 16:04.28 | cr2 | peanuts :) |
| 16:04.31 | imfloflo | but drcodes say he wait release |
| 16:04.51 | cr2 | imfloflo: it's for vogue/kaiser |
| 16:05.09 | cr2 | imfloflo: on diamond/raphael we need to boot the kernel first |
| 16:05.18 | exco | who'd have thought? ... Android is now Open Source |
| 16:05.23 | cr2 | because it's a different cpu |
| 16:06.15 | cr2 | exco: port wine to wince. then wince is dead. |
| 16:06.24 | imfloflo | on polaris |
| 16:06.37 | druidu | git clone git://android.kernel.org/kernel/msm.git |
| 16:06.39 | druidu | :D |
| 16:06.44 | cr2 | imfloflo: polaris is kaiser. |
| 16:07.02 | imfloflo | no :p |
| 16:07.05 | imfloflo | no keyboard |
| 16:07.11 | cr2 | the kernel side |
| 16:07.19 | cr2 | so ask dcordes :) |
| 16:07.34 | imfloflo | yes |
| 16:07.40 | cr2 | ok |
| 16:07.57 | exco | cr2: is that going to happen? |
| 16:08.17 | druidu | very nice of them to release sources before the stores open |
| 16:09.43 | *** part/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 16:09.49 | BabelO_ | android seem to be well packaged :) syncing repository now |
| 16:12.07 | *** join/#htc-linux sroecker (n=sroecker@e180071202.adsl.alicedsl.de) |
| 16:15.16 | *** join/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 16:17.05 | BabelO_ | swetland: where do we report bug ? :) cannot fetch libpcap |
| 16:17.22 | swetland | we're seeing it too |
| 16:17.34 | BabelO_ | bad md5sum :) |
| 16:17.39 | swetland | looks like a replication issue (something stepped on a tarball). the open source team is working on it |
| 16:18.04 | chiledog | i think they must be syncing the mirrors or something and thats why |
| 16:19.01 | BabelO_ | swetland: any idea of the approx size needed to build android ? |
| 16:19.43 | swetland | you'll need ~2GB to check out the entire thing and 2GB to build it |
| 16:20.37 | *** join/#htc-linux LunohoD_ (n=alex@e180076221.adsl.alicedsl.de) |
| 16:21.24 | *** join/#htc-linux FRITZ (n=FRITZ@189-19-168-237.dsl.telesp.net.br) |
| 16:21.29 | BabelO_ | swetland: ok, thanks, maybe a good idea to add that to the Get sources howto :) |
| 16:21.33 | BabelO_ | bbl |
| 16:21.55 | cr2 | bootloader source is there too ? |
| 16:22.47 | *** part/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) |
| 16:22.58 | swetland | "The source is approximentely[sic] 2.1GB in size. You will need 6GB free to complete the build" |
| 16:23.03 | swetland | http://sites.google.com/a/android.com/opensource/download |
| 16:23.34 | swetland | cr2: g1 bootloader is htc proprietary. there is some reference bootloader source we use for surf |
| 16:24.03 | cr2 | swetland: ok. |
| 16:24.23 | cr2 | swetland: is there some special driver for brf6300 ? |
| 16:24.31 | swetland | not particularly |
| 16:24.50 | swetland | everything is in the kernel, afaik |
| 16:24.58 | SanMehat | morning |
| 16:25.13 | cr2 | and who loads the firmware patches/commands ? |
| 16:25.33 | cr2 | hi SanMehat. a good day for all of us :) |
| 16:26.27 | SanMehat | cr2: yes :) |
| 16:26.34 | HardDisk_WP | BabelO, you remember me yesterday with the A702? |
| 16:26.43 | HardDisk_WP | where can I get toolchain for it? |
| 16:29.06 | cr2 | swetland: a minibootloader is still a very helpful thing. it can be loaded as htc diagnostics code, i guess. |
| 16:30.30 | swetland | I'm actually hoping to get a more complete generic android bootloader put together in the near future |
| 16:30.43 | swetland | http://android.kernel.org/?p=platform/bootloader/legacy.git;a=summary |
| 16:30.50 | swetland | has useful tidbits for bootloader stuff |
| 16:31.14 | swetland | in particular an implementation of our fastboot protocol which is what we use to download/flash/boot system images |
| 16:31.45 | cr2 | yes, i'm looking at it now. |
| 16:32.51 | *** join/#htc-linux webPragmatist (n=none@99-6-241-169.lightspeed.rcsntx.sbcglobal.net) |
| 16:33.34 | cr2 | swetland: you are able to redirect the uart3 on g1 to the htc usb connector pins ? |
| 16:33.58 | swetland | cr2: that's the default state |
| 16:34.38 | swetland | I'm trying to arrange for the Y debug cables we use (usb on one side, usb w/ usb-serial on the other) to be available for sale somewhere. still talking to people about that |
| 16:50.31 | *** join/#htc-linux schlash (n=har@adsl-76-212-172-60.dsl.sndg02.sbcglobal.net) |
| 16:53.19 | schlash | hey, im looking or some advice on getting linux up and running on a htc touch dia., anyone here that has some experience with that one? |
| 16:58.02 | *** join/#htc-linux sroecker_ (n=sroecker@e180071202.adsl.alicedsl.de) |
| 17:01.13 | druidu | swetland: is there any usb-serial-console or something, for htc? |
| 17:02.20 | druidu | until we get the lcd driver up |
| 17:02.56 | swetland | we always use the actual serial port |
| 17:03.07 | swetland | for development on dream/g1 |
| 17:06.40 | *** join/#htc-linux Kalemas (n=kalemas@mytlt.ru) |
| 17:06.52 | cr2 | swetland: the default state for g1 probably. i'm not quite sure that htc does it on other phones too. |
| 17:07.38 | druidu | well, is the serial port in end user devices? |
| 17:07.49 | cr2 | i've seen the pinout somewhere... |
| 17:07.50 | swetland | on g1, yes |
| 17:07.57 | swetland | other htc devices, dunno |
| 17:08.06 | *** join/#htc-linux GPFerror (n=gpferror@cpe-76-187-41-132.tx.res.rr.com) |
| 17:08.07 | druidu | schematics would be nice, yes :) |
| 17:08.28 | cr2 | druidu: it's easy to check. |
| 17:09.02 | druidu | although I don't wanna open my diamond, I'm surely gonna wreck it without proper tools |
| 17:09.09 | cr2 | druidu: unless one needs to set some gpios to enable it |
| 17:10.00 | cr2 | the diamond motherboard pictures were posted already. |
| 17:10.34 | cr2 | hm. i wish htc did it on all msm devices. |
| 17:11.00 | druidu | where where? |
| 17:11.08 | NetRipper | cr2, would it help if i checked if we can init any of the UARTs? i'm currently just skipping init'ing all uarts |
| 17:11.14 | cr2 | druidu: the pics ? |
| 17:11.17 | druidu | yes |
| 17:11.30 | BabelO | HardDisk_WP: codesourcery.com |
| 17:11.35 | NetRipper | cr2, interrupts of the uarts i mean |
| 17:12.06 | cr2 | NetRipper: you can write to a serial port with an oscilloscope connected to the htc usb port |
| 17:12.32 | cr2 | swetland: which pins are you using for the uart ? |
| 17:14.09 | druidu | oscilloscopes don't generate signals, they analyze them :P |
| 17:14.39 | NetRipper | he probs means using the oscilloscope to see if it receives output |
| 17:14.39 | NetRipper | ;) |
| 17:14.49 | swetland | I honestly don't know off the top of my head. Would need to poke around. We have Y-cables that break out the serial port and really haven't had to worry about the details |
| 17:15.11 | druidu | use e multimeter to trace the pins :P |
| 17:15.20 | druidu | I suppose there's no active part inside the cable |
| 17:15.22 | cr2 | http://forum.xda-developers.com/showpost.php?p=2610305&postcount=19 |
| 17:15.40 | cr2 | the extra pins are used for video on raphael |
| 17:15.51 | druidu | thanks cr2 |
| 17:16.15 | cr2 | and for multiplexing the usb+- pins one certainly needs some gpio... |
| 17:17.58 | HardDisk_WP | BabelO, target arch arm-linux or arm-uclinux? |
| 17:18.48 | BabelO | arm-linux |
| 17:20.10 | SanMehat | druidu: you'll need an RS232 level shifter |
| 17:20.41 | HardDisk_WP | once blew up one of these |
| 17:20.47 | HardDisk_WP | it was covered in hot glue |
| 17:21.00 | HardDisk_WP | I noticed that it was broken as the hot glue melted under the heat |
| 17:21.14 | HardDisk_WP | yet I still wonder what was broken, another chip perfectly worked with same wiring |
| 17:21.20 | druidu | SanMehat : level shifter as in they have different voltage levels for 1/0? |
| 17:22.01 | SanMehat | druidu: the device puts out TTL level signals.. you need to use an RS232 level shifter to get the voltages up to RS232 levels |
| 17:22.16 | SanMehat | druidu: you will likely need to power the level shifter off external power. |
| 17:22.25 | cr2 | druidu: http://forum.xda-developers.com/showthread.php?t=434535 |
| 17:22.29 | exco | ~seen lastik_ |
| 17:22.30 | apt | lastik_ <n=linuxrul@80.92.96.29> was last seen on IRC in channel #oe, 14d 22h 47m 13s ago, saying: 'Are anybody compiled libsvg ? It has an do_qa_staging fault'. |
| 17:22.36 | swetland | cr2: on g1, both usb and uart3 are present on different pins on the 11 pin connector, simultaneously |
| 17:23.29 | cr2 | swetland: ok, there are 6 extra pins. but g1 does not have video out ? |
| 17:23.41 | swetland | there is no video out on g1 |
| 17:24.22 | swetland | stereo audio out, mic in, uart |
| 17:24.26 | cr2 | xda-dev picture shows 6=2audio+video+3ground |
| 17:24.42 | cr2 | hmm. |
| 17:25.03 | cr2 | then it's on 3ground :) |
| 17:26.46 | cr2 | 1 mic, video out |
| 17:27.11 | cr2 | 2,6 audio out |
| 17:27.27 | cr2 | 5 audio ground |
| 17:27.53 | cr2 | so only 3 and 4 are left |
| 17:32.26 | SanMehat | button. |
| 17:33.39 | cr2 | ok, need to check it with haret then. |
| 17:33.54 | cr2 | SanMehat: but where are the txd and rxd then ? |
| 17:35.13 | SanMehat | i believe button re-uses one of the pins, not sure which. |
| 17:35.24 | *** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix) |
| 17:36.33 | *** join/#htc-linux goxboxlive (n=goxboxli@139.84-48-212.nextgentel.com) |
| 17:36.35 | druidu | maybe there is a special pin that activates secondary pin functions |
| 17:36.39 | druidu | just guessing |
| 17:37.57 | cr2 | druidu: gpio |
| 17:38.22 | druidu | SanMehat: good ol' max232 :) |
| 17:38.28 | druidu | or that |
| 17:38.44 | SanMehat | druidu: yeah.. the max232 will never die :) |
| 17:39.16 | druidu | ~seen druidu |
| 17:39.18 | apt | druidu is currently on #htc-linux (5d 5h 28m 44s). Has said a total of 423 messages. Is idling for 2s, last said: '~seen druidu'. |
| 17:39.27 | druidu | haha |
| 17:39.51 | cr2 | druidu: they have H2W clock and data. why not rxd and txd . |
| 17:40.22 | cr2 | ok. |
| 17:40.39 | cr2 | you can just bitbang rs232 there :) |
| 17:40.59 | SanMehat | cr2: you are evil |
| 17:40.59 | SanMehat | haha |
| 17:41.10 | cr2 | LOL |
| 17:41.28 | druidu | so it would be that simple to acces serial port, no soldering and disassembling? |
| 17:41.54 | cr2 | +ttl shifter |
| 17:42.11 | cr2 | you may talk spi or i2c. who cares. |
| 17:42.40 | druidu | I have some around, had a school project with an AVR |
| 17:42.50 | druidu | so you know which GPIO? |
| 17:43.15 | druidu | I don't like unmapped GPIOs |
| 17:43.23 | cr2 | on kaiser they certainly do something like spi bitbang for h2w |
| 17:44.28 | cr2 | i need to look closed on raphael, but all h2w related gpios are documented |
| 17:44.35 | cr2 | gpio1 is called 'h2w power'. maybe it's just h2w enable. |
| 17:44.48 | druidu | what exactly is h2w? |
| 17:45.00 | cr2 | htc 2wire |
| 17:45.09 | cr2 | according to the g1 source. |
| 17:45.10 | druidu | aah, like the 1wire in the kernel code |
| 17:45.52 | cr2 | it's like qualcomm |
| 17:46.11 | cr2 | sbi and ssbi patents ;) |
| 17:46.25 | cr2 | for 1wire and spi. |
| 17:46.30 | cr2 | typical NIH |
| 17:47.17 | druidu | I've got some AVR microcontrollers, I could write some interface code to connect to serial |
| 17:47.49 | druidu | right now, I could do with smoke signals if I had to, it would be better than my one-shot-lantern for sure |
| 17:48.04 | druidu | I think I restarted my htc diamond 1000+ in the last days |
| 17:48.17 | cr2 | spi feels old and dull. h2w sounds like "htc innovation" |
| 17:48.26 | cr2 | lol |
| 17:48.41 | druidu | atmel avr supports spi directly :P |
| 17:48.44 | druidu | it would be easier |
| 17:49.10 | druidu | where do we connect it physically anyway? (spi/h2w) |
| 17:49.48 | cr2 | there are no docs |
| 17:49.55 | cr2 | but it can be measured. |
| 17:51.20 | druidu | acoustic_mmap |
| 17:51.28 | druidu | what's this acoustic thing? |
| 17:51.37 | cr2 | where ? |
| 17:52.12 | cr2 | time to run sed http://wiki.xda-developers.com/index.php?pagename=RaphaelMDDI on http://wiki.xda-developers.com/index.php?pagename=RaphaelLCD |
| 17:52.16 | druidu | Headset is connect via h2w?!? |
| 17:52.31 | druidu | all the docs are in kernel source |
| 17:52.35 | druidu | #defines |
| 17:52.50 | cr2 | yes, to identify the headset probably. |
| 17:53.08 | druidu | maybe for the button |
| 17:53.11 | druidu | 2 buttons |
| 17:53.20 | cr2 | SanMehat: 6 pwm registers stay unnamed http://wiki.xda-developers.com/index.php?pagename=RaphaelMDDI |
| 17:53.58 | druidu | yes, I'm right |
| 17:54.31 | cr2 | extmic, cable_in |
| 17:54.37 | druidu | http://android.kernel.org/?p=kernel/msm.git;a=blob;f=arch/arm/mach-msm/board-trout-h2w.c;h=c6dfd726adb6decf5b5e1a9d18606b2154b8f229;hb=android-msm-htc-2.6.25 |
| 17:54.38 | cr2 | gpio 17 and 18 |
| 17:55.01 | cr2 | gpio 45 cable_in2 |
| 17:55.35 | *** join/#htc-linux goxboxlive (n=goxboxli@139.84-48-212.nextgentel.com) |
| 17:55.41 | cr2 | gpio46 and 31. data and clock probably |
| 17:56.22 | cr2 | and kaiser the clock and data were obvious |
| 17:56.23 | par | whats the processor? |
| 17:56.30 | cr2 | did i document them ... |
| 17:56.36 | cr2 | par: ? |
| 17:56.52 | par | which microprocessor |
| 17:57.16 | cr2 | par: what are you talking about ? |
| 17:57.41 | par | well you said gpio46 and 31 |
| 17:57.46 | cr2 | ok, cable_in2 is the button |
| 17:58.08 | cr2 | msm7201A on raphael |
| 17:58.14 | cr2 | these are just gpios |
| 17:58.27 | par | okay so its omap850? |
| 17:58.40 | cr2 | msm7201A |
| 17:58.46 | druidu | cr2: who compiled all that stuff for raphael and how? :) |
| 17:59.06 | par | hmm i don't have ref for msm7201A |
| 17:59.06 | cr2 | druidu: ->history |
| 17:59.11 | par | seems close though |
| 18:01.09 | par | the datasheet floating around fo msm7201A at all |
| 18:01.12 | par | ? |
| 18:01.41 | cr2 | hehe. let me know if you'll find one :) |
| 18:01.49 | par | sure ;) |
| 18:05.24 | kiozen | cr2: you manged to see something on the map? |
| 18:05.44 | cr2 | kiozen: no. |
| 18:05.57 | kiozen | cr2: coodinate? |
| 18:06.21 | cr2 | 51 7 |
| 18:06.41 | cr2 | kiozen: i use the eu.tgz |
| 18:07.16 | cr2 | kiozen: .tdb contains the tile grid ? |
| 18:08.12 | kiozen | cr2: tdb lists all subfiles and their area coverage |
| 18:08.20 | cr2 | ok. |
| 18:08.35 | kiozen | there is plenty to see at N51 E7 |
| 18:08.40 | cr2 | maybe you'd draw the grid when no other features are found ? |
| 18:08.45 | druidu | cr2 ? |
| 18:08.55 | druidu | ah, nevermind :) |
| 18:08.56 | *** join/#htc-linux lastik_ (n=eraser@80.92.96.29) |
| 18:09.14 | cr2 | kiozen: trying again |
| 18:09.42 | cr2 | druidu: i suspect that 99% of the rapahel gpios are the same on diamond, but you'd be careful. |
| 18:10.10 | *** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz) |
| 18:10.16 | druidu | yeah, but different keys + keyboard + no flashlight |
| 18:10.20 | druidu | lots of little differences |
| 18:10.30 | kiozen | cr2: that's how it should look like http://www.qlandkarte.org/shot4.png |
| 18:10.32 | druidu | any idea if LCDs have W > H ? |
| 18:10.40 | druidu | how does the controller see them? |
| 18:10.46 | druidu | 640 x 480 or 480 x 640? |
| 18:11.05 | cr2 | kiozen: trying it again. |
| 18:11.18 | cr2 | druidu: imvho 480x640 |
| 18:11.53 | NetRipper | yes, 480x640 |
| 18:14.45 | cr2 | kiozen: N51 E7 does not show anything |
| 18:14.57 | kiozen | funny |
| 18:15.09 | kiozen | if you zoom out? |
| 18:15.23 | cr2 | kiozen: yes |
| 18:15.34 | kiozen | nothing at all |
| 18:15.41 | cr2 | kiozen: but the lower right corner shows different numbers |
| 18:16.14 | cr2 | N16° 39... E045° 53... |
| 18:16.48 | cr2 | kiozen: hm http request failed. for search. |
| 18:17.44 | druidu | gotta go, later |
| 18:17.45 | kiozen | nothing at N16° 39 E045° 53 |
| 18:17.50 | druidu | let's get this android baby movting :) |
| 18:17.55 | cr2 | kiozen: no. |
| 18:18.19 | kiozen | yes nothing there :) |
| 18:18.28 | HardDisk_WP | BabelO, btw, I have made the dump with wmtest |
| 18:18.49 | cr2 | kiozen: but if i zoom |
| 18:20.16 | cr2 | kiozen: btw N16. what does it have with the EU ? |
| 18:20.33 | *** join/#htc-linux Othello (i=Othello@gateway/tor/x-bb6ea4c6dc8d4c72) |
| 18:20.38 | kiozen | it's not on the map |
| 18:20.41 | cr2 | and E45 is certainly too far in the east. |
| 18:21.01 | cr2 | but center the map puts these values there. |
| 18:21.13 | schlash | the boot param that you can set in the startup script for haret .. "set RAMADDR=" is that hardware specific or just the address where id like to put the kernel? |
| 18:21.44 | schlash | or is it the virtual mem mapping to the start of the ram memory? |
| 18:22.04 | cr2 | kiozen: ok, i moved to 51N 7E manually, but there is still nothing there. |
| 18:22.27 | cr2 | schlash: hardware specific |
| 18:23.00 | cr2 | schlash: in 99.9% of cases it's the SDRAM start physical address |
| 18:23.03 | schlash | okay, it seems to be default hex 1000000 , for 7500, anyone know if it is the same for msm7200A |
| 18:23.13 | kiozen | I can't reproduce that, but I fail to see island |
| 18:23.18 | schlash | cr2: but it must be the virtual mem address and not phys? |
| 18:23.18 | cr2 | schlash: yes |
| 18:23.24 | schlash | okay thanks |
| 18:23.34 | cr2 | schlash: http://wiki.xda-developers.com/index.php?pagename=RaphaelMemoryMap |
| 18:23.50 | kiozen | I can understand that island wants to disappear, taking a vacation vom history, but not within my app |
| 18:24.00 | cr2 | kiozen: island is not EU |
| 18:24.28 | *** part/#htc-linux Kalemas (n=kalemas@mytlt.ru) |
| 18:24.31 | cr2 | ok. strictly speaking. |
| 18:24.39 | kiozen | shh don't tell those guy on OSM |
| 18:24.49 | cr2 | like Norway or Switzerland |
| 18:25.40 | kiozen | cr2: ah, maybe I have the bug, wait... |
| 18:28.47 | *** join/#htc-linux goxboxlive (n=goxboxli@139.84-48-212.nextgentel.com) |
| 18:28.57 | kiozen | cr2: svn up, fixed a very nasty bug |
| 18:29.08 | cr2 | ok |
| 18:29.58 | schlash | Is there any good way to actually debug the events when haret has loaded the kernel to 100%? cause all i know is the haretlog.txt but it doesnt print anything more ofcourse after the last "Go go go" message... and my problem is that haret gets to 100% stays there for 15 seconds than the device reboots |
| 18:30.32 | cr2 | kiozen: i need to run make clean to rebuild GT. something is wrong with the makefiles |
| 18:30.49 | kiozen | huh? |
| 18:31.13 | cr2 | kiozen: i've done svn up, and then make clean |
| 18:31.21 | cr2 | otherwise it does not see the changes |
| 18:31.36 | BabelO | kiozen: look at england, and you see the bug, depend of the zoom level |
| 18:31.47 | cr2 | schlash: it's very difficult because all the hardware is disabled |
| 18:31.52 | kiozen | BabelO: think I fixed it |
| 18:32.27 | cr2 | schlash: we use flashlight led on raphael, because it's on a cpu gpio. everythign else will be really difficult. |
| 18:32.27 | schlash | cr2: yepp, i know, :( i was just hoping someone else have had the problem and could give me a small hint :) |
| 18:32.38 | schlash | ahh i see |
| 18:32.40 | schlash | good idea |
| 18:32.43 | cr2 | schlash: on which device ? |
| 18:32.44 | BabelO | kiozen: ok, i m fighting with android source.... not compatible with python 2.5, i do it by hand :( |
| 18:33.24 | kiozen | BabelO: 2.5 is out for ages |
| 18:33.28 | schlash | cr2: sonyericsson Xperia |
| 18:33.42 | cr2 | schlash: ok. |
| 18:33.51 | cr2 | schlash: it's a 7201A phone ? |
| 18:33.54 | BabelO | kiozen: i know but google use python 2.4 and unbuntu as reference :) |
| 18:34.02 | cr2 | schlash: i know it's from htc. |
| 18:34.04 | schlash | 7200a |
| 18:34.16 | cr2 | 0 or 1 does not mate |
| 18:34.24 | cr2 | s/mate/matter/ |
| 18:34.32 | cr2 | damn notebook keyboard ;) |
| 18:34.36 | schlash | hehe |
| 18:34.55 | cr2 | schlash: do you have a mtype foe it ? |
| 18:35.07 | kiozen | lol OSM draws the LHC in Cern as private alley |
| 18:35.15 | schlash | i added my own |
| 18:35.24 | cr2 | otherwise you'd use halibut. |
| 18:35.28 | cr2 | ok. |
| 18:35.36 | cr2 | kiozen: lol |
| 18:35.46 | schlash | but i just extended the msm7200 that was in the haret code i compiled |
| 18:36.45 | cr2 | schlash: 0.5.2 supports A |
| 18:36.52 | schlash | hmm |
| 18:36.57 | schlash | ok |
| 18:37.29 | schlash | so there is someone who got linux up and running on a msm7200a |
| 18:38.25 | cr2 | schlash: haret != linux |
| 18:38.56 | cr2 | schlash: g1 obviously runs on msm7201a |
| 18:39.21 | schlash | true, haret is not linux, but im guessing its what it is used for |
| 18:39.24 | schlash | so |
| 18:40.05 | cr2 | you need a lot of device drivers to really boot linux. |
| 18:41.22 | schlash | yes |
| 18:45.32 | *** part/#htc-linux exco (n=exco@e181069079.adsl.alicedsl.de) |
| 18:45.40 | *** join/#htc-linux exco (n=exco@e181069079.adsl.alicedsl.de) |
| 18:48.42 | *** join/#htc-linux goxboxlive (n=goxboxli@139.84-48-212.nextgentel.com) |
| 19:12.13 | *** join/#htc-linux dr-dre (n=dr-dre@f048230144.adsl.alicedsl.de) |
| 19:17.37 | *** join/#htc-linux nebi (n=nebi@c-498fe253.02-145-7570701.cust.bredbandsbolaget.se) |
| 19:20.07 | tmzt | cr2: any possibility of using bt uart debugging? something in android kernel |
| 19:20.16 | tmzt | cr2: is it on arm11 uart? |
| 19:21.31 | *** join/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 19:24.09 | dcordes | oh good news |
| 19:25.31 | imfloflo | hey drcodes |
| 19:25.35 | imfloflo | how are u ? |
| 19:25.43 | imfloflo | so you see the big news |
| 19:26.22 | *** join/#htc-linux goxboxlive (n=goxboxli@139.84-48-212.nextgentel.com) |
| 19:34.37 | *** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net) |
| 19:35.34 | dcordes | imfloflo, yep |
| 19:36.00 | dcordes | imfloflo, you could easily make a small initrd now |
| 19:37.13 | *** join/#htc-linux Tinyboom (n=nahh@72.84-49-71.nextgentel.com) |
| 19:37.55 | imfloflo | ok i'm ok to test it :p |
| 19:51.06 | *** join/#htc-linux infernixx (n=nix@unaffiliated/infernix) |
| 19:51.20 | *** part/#htc-linux infernixx (n=nix@unaffiliated/infernix) |
| 19:54.25 | *** join/#htc-linux timebomb (n=timebomb@e176108158.adsl.alicedsl.de) |
| 19:54.45 | kaze | big news? |
| 19:55.00 | swetland | I need to get BUILD_TINY_ANDROID working again in the open source release |
| 19:55.06 | swetland | it's crazy handy for bringup debugging |
| 19:57.10 | SanMehat | swetland: I'll help you if you need |
| 19:57.27 | swetland | it's mostly build system muckery |
| 19:57.39 | swetland | and those guys have been slammed getting the open source infrastructure in place |
| 19:58.15 | diogene31 | Good evening everyone. |
| 19:59.12 | diogene31 | cr2: Do you know if G1 MSM7201A datasheet will be released one day ? |
| 20:00.24 | cr2 | diogene31: i doubt it |
| 20:00.52 | cr2 | tmzt: bt uart on pxa ? |
| 20:00.58 | diogene31 | cr2: Marvell / Intel did it before. Don't you think Qualcomm will go the same way ? |
| 20:01.24 | cr2 | diogene31: no. do you have ti gsm datasheets ? |
| 20:02.00 | diogene31 | cr2: No, never had this hardware. Why ? |
| 20:02.08 | cr2 | intel did it, because DEC did it, and both were always pretty open about the hw docs |
| 20:02.47 | cr2 | diogene31: because there are only leaked ti gsm datasheets available. nothing is published. |
| 20:03.10 | cr2 | diogene31: i don't know where to find the omap730/750 datasheet for example |
| 20:03.31 | dcordes | cr2, can you help me with the tnetw driver? I would like to try it on the kaiser. |
| 20:03.34 | cr2 | does marvell provide the pxa3xx datasheets ? |
| 20:03.46 | diogene31 | cr2: So the bet for us, open communauty, will be waiting for a phone under a pxa3xx, under Android or assimilated ... |
| 20:04.08 | diogene31 | cr2: Not yet, I know Eric Miao at LAK has it. Maybe one day ... |
| 20:04.13 | dcordes | diogene31, what about omap3? |
| 20:04.19 | cr2 | dcordes: i need to checkout it first. we need to find the place for power,irq ... |
| 20:04.35 | dcordes | cr2, ok. it seems totally confusing to me. don't understand anything |
| 20:05.06 | dcordes | interesting that the g1 filesystem has a different firmware |
| 20:05.22 | cr2 | dcordes: it was known from the very beginning |
| 20:06.09 | diogene31 | dcordes: I don't really know anything about omap chips ... |
| 20:06.11 | cr2 | dcordes: if the xda people will dump the appsbl (aka spl), it may be the same as on other phones. |
| 20:06.59 | cr2 | dcordes: but it's interesting to look at it anyway. |
| 20:07.25 | BabelO | why are you speaking about omap ? :) |
| 20:07.27 | diogene31 | swetland: I assume you're in contact with Qualcomm very often :) Would you believe they'll release the datasheet one day ? |
| 20:07.39 | cr2 | you can't expect that they will publish it, because it contains keys, you can flash too many things and so on. |
| 20:07.55 | swetland | my experience with qualcomm leads me to believe the likelyhood of them publicly releasing databooks is pretty much near 0 |
| 20:08.40 | cr2 | diogene31: why do you need the datssheet ? |
| 20:08.40 | swetland | but hey, four years ago TI told me about how they were trying to get legal to scrub their databooks for public release, and these days you can download OMAP3 databooks with no clickthrough or anything |
| 20:08.45 | swetland | so who knows |
| 20:08.48 | dcordes | cr2, sorry, I don't mean amss but wifi firmware. you mentioned that earlier. I think I'm familiar with the arm9 situation |
| 20:08.57 | imfloflo | drcodes have you plan to relaesa new version for kaiser and polaris |
| 20:09.08 | dcordes | and do hope viperbjk & co will bring more light to it |
| 20:09.29 | diogene31 | swetland: Thanks, that's straight :) That will surely help me to choose my next kernel efforts. |
| 20:09.38 | dcordes | imfloflo, if somebody makes a small rootfs I can replace the old initrd with it, yes |
| 20:09.40 | cr2 | diogene31: it's only important if you want to mess with the umts DSP. but they will burn you for doing it anyway ;-) |
| 20:10.37 | diogene31 | cr2: You know what I do : make basic hardware work, develop drivers, and so on ... UMTS is not in my scope, as it is too expensive in France. |
| 20:11.18 | cr2 | dcordes: the wifi firmware is arm7tdmi assembler +data. since it was patched on hx4700, there are people who digged inside. |
| 20:11.24 | BabelO | is agree with diogene31 has he got a 3G+ usb dongle for free |
| 20:14.07 | diogene31 | cr2: The datasheets are important for people improving and extending existing drivers. I went through many pxa drivers, rewrote fully 2 of them, and I'm still improving others. That's what I like to do, make hardware sing :) |
| 20:14.27 | cr2 | diogene31: you may look at the raphaelmemorymap page, and look for "free" areas |
| 20:14.36 | diogene31 | cr2: I can't play with hardware without datasheets without a tremendous efforts ... |
| 20:14.46 | cr2 | diogene31: i'd be happy if the kernel will work ;) |
| 20:15.18 | cr2 | diogene31: ok, then pxa is your only choice. |
| 20:16.06 | cr2 | or you may hope that somebody posts a pdf. like for the ti calypso. |
| 20:17.02 | cr2 | on himalaya/blueangel you have all datasheets available |
| 20:17.17 | cr2 | for everything. |
| 20:17.27 | BabelO | cr2: adcm2650 :) |
| 20:17.35 | BabelO | cr2: w3220 |
| 20:17.47 | cr2 | BabelO: the driver is available. |
| 20:17.52 | *** part/#htc-linux silven (n=zmc@adsl-64-237-230-239.prtc.net) |
| 20:17.54 | cr2 | for 3220 too |
| 20:18.02 | cr2 | to a big extent. |
| 20:18.03 | BabelO | cr2: but for imageon we miss the capture part |
| 20:18.17 | diogene31 | cr2: The Qualcomm is too much under public attention. Changes of a .pdf leaked are pretty much 0. |
| 20:18.36 | cr2 | BabelO: yes, but there are more interesting problems to solve |
| 20:19.29 | BabelO | cr2: yes keyboard repeat |
| 20:19.54 | *** join/#htc-linux wasup (n=Tech@208.47.103.106) |
| 20:19.56 | cr2 | diogene31: localize the hardware control part of amss. |
| 20:20.07 | BabelO | cr2: i stop trying android, their repo system works like OE, and actually too much error |
| 20:20.26 | cr2 | BabelO: somebody should port asic3_mmc to 2.5.27+ first |
| 20:20.50 | BabelO | yes i know |
| 20:20.56 | dcordes | BabelO, I won't try for now. one buildsystem is enough |
| 20:21.06 | kaze | Usually the Android uses YAFFS as rootfs and uses the mtd device as storage media, the bad blocks in the mtd device seldom cause YAFFS file system to work abnormally << is it the problem faced on kaiser? |
| 20:21.09 | cr2 | BabelO: i'll try android only after i'll boot the kernel on raphael:) |
| 20:21.11 | diogene31 | cr2: amss ? Don't know what that is ... |
| 20:22.01 | dcordes | kaze, no |
| 20:22.01 | BabelO | dcordes: i sync the repo, but got some error due to my python2.5 ( don't want to insall old 2.4)so i remove from manifest.xml the problem... and now i m at end with a traceback error .. |
| 20:22.04 | cr2 | diogene31: are you running 2.6.27 on mio ? |
| 20:22.20 | dcordes | BabelO, lol. no thanks :) |
| 20:22.30 | diogene31 | cr2: Of course, I'm always up to date with kernel git. I'm even in mainline now :) |
| 20:22.49 | cr2 | diogene31: how do you control the serial port power ? |
| 20:23.21 | diogene31 | cr2: Through gpio sysfs. |
| 20:23.39 | diogene31 | cr2: That controls GSM (FFUART), Bluetooth(BTUART) and GPS (STUART). |
| 20:23.54 | cr2 | diogene31: well, let's try to push other phones then. there are only minor problems with them. (non asic3_mmc) |
| 20:24.25 | cr2 | diogene31: ok. it does not really depend on the cpu. |
| 20:25.03 | diogene31 | cr2: No, that's the beauty of it. It's the same way of doing whether you're on PXA, OMAP or MSM. |
| 20:25.26 | cr2 | ok. |
| 20:26.17 | cr2 | hmm. maybe i'd try 2.6.27 on athena :) |
| 20:26.31 | BabelO | diogene31: if you want to play, i can send you my htc blueangel :) |
| 20:26.48 | cr2 | BabelO: asic_mmc :( |
| 20:27.01 | BabelO | cr2: for diogene31:) |
| 20:27.26 | cr2 | BabelO: but to work on it, you need a working asic3_mmc driver |
| 20:27.40 | diogene31 | BabelO: Thank you, but I consider that one a bit ... old :) And a colleague at work has it, and I don't like it very much. |
| 20:27.46 | dcordes | I also have ba to borrow. board id 6 |
| 20:28.25 | dcordes | lol ba is certainly not the most fashionable device |
| 20:28.32 | cr2 | diogene31: then you should buy a 7201A device. |
| 20:28.47 | cr2 | dcordes: himalaya is the most solid device :) |
| 20:29.10 | diogene31 | cr2: Like the G1 ? |
| 20:29.11 | cr2 | but having a wallaby will be fun too. |
| 20:29.15 | BabelO | diogene31: oh yes i remeber the other person who have one |
| 20:29.27 | cr2 | diogene31: G1 is not for hacking. |
| 20:29.57 | diogene31 | cr2: Why not ? |
| 20:30.19 | cr2 | diogene31: if you like iphone, then blackstone/HD. otherwise raphael. |
| 20:30.44 | cr2 | because it has different amss from wince. |
| 20:30.48 | dcordes | diogene31, because the working kernel sources are present |
| 20:31.21 | cr2 | diogene31: so your work does not benefit running linux on wince phones. |
| 20:32.23 | cr2 | where you have some tricky problems like managing the vreg_* |
| 20:33.02 | cr2 | gpio control is 100% known now, the clock control 50%, and vreg is the most closed part. |
| 20:33.18 | *** part/#htc-linux wasup (n=Tech@208.47.103.106) |
| 20:33.26 | cr2 | with htc using some really bizarre ways to access it. |
| 20:33.51 | cr2 | like sending at commands on some hidden subchannel |
| 20:33.52 | dcordes | diogene31, you can always get a ksier as well :D |
| 20:34.23 | cr2 | dcordes: and get a lot of troubles with msm_fb :) |
| 20:34.56 | diogene31 | cr2: dcordes I'm really thinking about the kaiser. It's only 99 euros in France ATM ... |
| 20:35.29 | dcordes | diogene31, yea got cheap. but be aware. the sd driver is hairy |
| 20:36.21 | cr2 | diogene31: we know a lot of things about it, but msm_fb will be a pita. |
| 20:36.24 | dcordes | cr2, seems there will be more msm7201a than diamond raphael and blackstone |
| 20:36.58 | dcordes | cr2, which problem about msm_fb on the kaiser do you mean? making it work with something else than android? |
| 20:37.07 | dcordes | because that's the only problem I see |
| 20:37.25 | cr2 | dcordes: maybe, but these 3 cover most feastures you will really need. |
| 20:37.40 | cr2 | the speed of msm_fb there. |
| 20:38.08 | cr2 | there are no docs, and therefore no hope. |
| 20:38.50 | dcordes | what's the problem with the speed? |
| 20:39.39 | dcordes | you can do optically complex things in the android |
| 20:40.33 | dcordes | wtf is wireless man? |
| 20:40.57 | dcordes | http://pdadb.net/index.php?m=specs&id=1528 "Wireless:MAN: 802.16e-2005 " |
| 20:43.16 | dcordes | mobile wimax, on 2,3Ghz |
| 20:44.02 | imfloflo | drcodes how can we make rootfs ? |
| 20:45.30 | ali1234 | BabelO: is artemis bluetooth on one of the omap uarts? |
| 20:45.47 | dcordes | imfloflo, I have no clue. maybe there is documentation at android.com |
| 20:46.21 | BabelO | ali1234: yes, like gps too |
| 20:46.29 | dcordes | diogene31, if you need no keyboard, polaris is the same as kaiser without one. slightly different display |
| 20:46.45 | dcordes | and ts |
| 20:46.55 | ali1234 | BabelO: on linwizard we dont seem to have uarts working |
| 20:47.25 | ali1234 | does it need a special driver or are they just standard? |
| 20:48.34 | imfloflo | ok so we can't use existing files we must start bulding new files to have this version working on polaris |
| 20:49.00 | BabelO | ali1234: look at my artemis code bt.c, you need to activate clock and some gpio |
| 20:49.12 | BabelO | ali1234: clock let you detect the uart else nothing is seen |
| 20:49.15 | dcordes | imfloflo, no it is still the same storage problem. you can also use the g1 or older sdk builds. |
| 20:49.31 | dcordes | imfloflo, but it's probably nicer to compile it yourself |
| 20:49.52 | ali1234 | BabelO: ah i see, i thought the gpio was just turning on the BT chip |
| 20:50.08 | dcordes | imfloflo, as swetland said earlier they want to maek a "recipe" (or whatever they call their metadata) for a small image |
| 20:50.15 | imfloflo | compile what dont understand |
| 20:50.33 | imfloflo | how can i help in this? |
| 20:50.42 | *** join/#htc-linux SanMehat (n=san@nat/google/x-0c1076d7fd9da5cd) |
| 20:51.08 | BabelO | ali1234: yes gpio just turn on the bt chip, clock is just to activate the omap_serial.c detection |
| 20:51.09 | dcordes | imfloflo, ask SanMehat or swetland how you can get a root filesystem which is max ~20mb in size |
| 20:51.56 | imfloflo | SanMehat how you can get a root filesystem which is max ~20mb in size |
| 20:52.42 | ali1234 | BabelO: thanks, gonna try to trace the gpio on wizard, and that should be all i need |
| 20:53.37 | BabelO | ali1234: what about gsm ? miknix look at it ? |
| 20:53.40 | SanMehat | imfloflo: very carefully? |
| 20:54.42 | miknix | BabelO, Didn't work much on it since I had to re-implement it using usart |
| 20:54.43 | imfloflo | just want to have lastest version of android working well on Polaris |
| 20:54.55 | imfloflo | ad drcordes says i should ask you this |
| 20:54.58 | BabelO | miknix: ok |
| 20:57.18 | dcordes | SanMehat, what do you mean, carefully? |
| 21:02.17 | kaze | ok the size limit is the size of the initrd haret can load? |
| 21:02.46 | kaze | its not a storagze problem but a initrd size for haret? |
| 21:03.33 | imfloflo | so why don't modify haret if its haret limit ? |
| 21:03.38 | kaze | wow my questions are so stupid, nobody ever care to answer |
| 21:04.02 | kaze | i feel love =( |
| 21:05.47 | cr2 | kaze: it's mainly a decompression problem |
| 21:08.11 | kaze | thanks for answering |
| 21:10.04 | druidu | I THINK I JUST MADE THE SCREEN UPDATE! |
| 21:10.06 | druidu | WOHOHOOHOHOOO |
| 21:10.13 | kaze | :O |
| 21:10.17 | kaze | on diamond? |
| 21:10.18 | druidu | lemme try text |
| 21:10.27 | *** join/#htc-linux BruteFrce (n=IrcNet@m530e36d0.tmodns.net) |
| 21:10.31 | druidu | yes, right now it's white, and not black as usual :) |
| 21:10.36 | kaze | :O |
| 21:10.49 | druidu | of course, with screen already initialized by wince |
| 21:11.01 | BruteFrce | god this arrow thing is driving me mad |
| 21:11.50 | druidu | where are the compiled in fonts in the linux kernel? (file location) |
| 21:12.04 | druidu | nevermind :P |
| 21:12.20 | BruteFrce | i would guess var |
| 21:13.39 | dcordes | BruteFrce, it's likely he means the sauce kernel. |
| 21:14.06 | BruteFrce | sauce kernel? |
| 21:14.38 | kaze | excuse me again sirs, but did sanMehat elaborated a bit about building a small rootfs " carefully"? |
| 21:14.47 | diogene31 | dcordes: I like the keyboard very much. Raphael is too expensive for me. |
| 21:15.00 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 21:15.55 | druidu | sauce -> source? :) |
| 21:15.56 | BruteFrce | are the msm procs the fastest for ppc phones roght now? |
| 21:16.05 | druidu | found them anyway, drivers/video/console/ |
| 21:16.10 | BruteFrce | oh he means kernel source |
| 21:16.15 | kaze | diogene31 afaik there is no french provider selling the raphael |
| 21:16.21 | *** join/#htc-linux datachaos (n=datachao@201.22.213.114.static.gvt.net.br) |
| 21:16.37 | BruteFrce | eww why would you buy from cell company? |
| 21:16.50 | imfloflo | diogene31 31 like toulouse ? |
| 21:17.24 | diogene31 | imfloflo: Yes, Toulouse is where I live in. |
| 21:17.32 | kaze | i live in bordeaux |
| 21:17.35 | kaze | if you care |
| 21:17.38 | kaze | 33 |
| 21:17.41 | imfloflo | me too :) another htc user |
| 21:17.50 | kaze | :) |
| 21:18.06 | diogene31 | kaze: I think Bouygues Telecom sells it, under "Touch Pro" name. |
| 21:18.14 | kaze | oh |
| 21:18.18 | imfloflo | bordeau is too far away :p |
| 21:18.28 | kaze | not that far |
| 21:18.29 | imfloflo | yes bouygues sele rapahel |
| 21:18.34 | kaze | 200Km at most |
| 21:18.41 | kaze | damn i switched over to sfr |
| 21:18.47 | kaze | and got a diamond |
| 21:18.55 | kaze | but since i got a diamond i bought a kaiser |
| 21:19.04 | kaze | cause the keyboard is great |
| 21:19.22 | *** join/#htc-linux nebi_ (n=nebi@c-498fe253.02-145-7570701.cust.bredbandsbolaget.se) |
| 21:19.26 | BruteFrce | tmzt is making progress on a kaiser kernel ya know |
| 21:19.28 | kaze | maybe i could have gotten a raphale by renewing to bt |
| 21:19.38 | kaze | woowoo |
| 21:19.38 | imfloflo | sell diamond and kaiser and bought rapahel |
| 21:19.41 | dcordes | BruteFrce, where? |
| 21:20.03 | kaze | kaiser kernel, /me wants too |
| 21:20.23 | BruteFrce | where? like where does he live? |
| 21:20.27 | dcordes | BruteFrce, seems you are mixing things up |
| 21:20.30 | kaze | i think where to dl |
| 21:20.34 | diogene31 | kaze: There are a lot of people from Bordeaux lately :) |
| 21:20.47 | kaze | diogene31 in this chan? |
| 21:20.49 | dcordes | BruteFrce, dzo works with the vogue kernel |
| 21:20.50 | BruteFrce | last i heard he was having troublewith usb |
| 21:21.00 | BruteFrce | dzo? |
| 21:21.22 | dcordes | sorry, you said tmzt. I mixed ^^ |
| 21:21.24 | kaze | bruteFrce i think dcordes is the one doing the kernel |
| 21:21.49 | dcordes | most recent things for the kaiser kernel are done by ginge |
| 21:21.50 | kaze | i may be wrong |
| 21:21.51 | BruteFrce | oh well i know tmzt has been working with it |
| 21:21.57 | kaze | oh |
| 21:22.14 | kaze | btw dcordes the kernel on your site is kinda old? |
| 21:22.22 | BruteFrce | or was talking about the usb on his the other day at least |
| 21:22.46 | kaze | or does it have all latest patches? |
| 21:23.03 | dcordes | kaze, yes because nothing groundbreaking changed |
| 21:23.14 | BruteFrce | my kernel now has usb networking, full sd, x11, machbox wm, keyboard support for matchbox |
| 21:23.55 | druidu | what device? |
| 21:23.58 | BruteFrce | just cant get the damn keyboard arrows to work on any key |
| 21:24.06 | BruteFrce | s260 |
| 21:24.12 | BruteFrce | excalibur |
| 21:24.43 | dcordes | BruteFrce, wow that was rapid. I remember the other week you said the display doesn't init :) |
| 21:25.01 | BruteFrce | penguinz has been making great progress his new keymap is awesome |
| 21:25.16 | BruteFrce | ya have come far |
| 21:25.17 | dcordes | nice. do you guys use the hh kernel? |
| 21:25.32 | tmzt | linwizard right now, but I built htc-omap too |
| 21:25.36 | tmzt | before |
| 21:25.51 | BruteFrce | if ali1234 would hurry with his tty driver we would have pppd |
| 21:25.56 | dcordes | what kernel version are they at? |
| 21:26.08 | BruteFrce | the htc-omap never worked though |
| 21:26.19 | tmzt | could be wrong |
| 21:26.22 | tmzt | I |
| 21:26.47 | ali1234 | BruteFrce: i'm trying to get bluetooth working because it looks easy... |
| 21:26.57 | tmzt | no working on kaiser |
| 21:26.59 | tmzt | not |
| 21:27.10 | BruteFrce | oh my mistake |
| 21:27.57 | BruteFrce | ali1234 oh that sucks... i mean obviously you can work on whatever you want, but i have no use for bluetooth especially with no adio |
| 21:28.13 | BruteFrce | audio |
| 21:28.56 | BruteFrce | tty seems a little more important lol, but its your gig man |
| 21:29.18 | marajin | hmm I need to scrounge up some more DDR2-FB ECC .. |
| 21:29.19 | BruteFrce | wish i had the knowlage |
| 21:29.56 | dcordes | marajin, I have 2 laptop ddr2 modules if you like |
| 21:30.49 | marajin | dcordes: nah man I mean DDR2 Fully Buffered ECC-Registered. |
| 21:30.53 | marajin | dcordes: It's server RAM |
| 21:31.33 | ali1234 | BruteFrce: the trouble with gsm is the driver basically needs rewriting from scratch, and the guy who wrote the original driver is not around. and even then it will probably only work for phone calls. |
| 21:31.43 | BruteFrce | its great having x that you cant use grrrr |
| 21:32.12 | dcordes | BruteFrce, lol on the kaiser it's everything but X |
| 21:32.26 | tmzt | BruteFrce: does d-pad work now? |
| 21:32.38 | tmzt | you could make a fake mouse with that |
| 21:32.43 | BruteFrce | well we need somone who can get it working for sms mms calls and gprs im sure its possible |
| 21:32.44 | tmzt | that puts events into uinput |
| 21:32.46 | ali1234 | or get a bluetooth mouse |
| 21:33.01 | tmzt | a presenter with trackball would be nice |
| 21:33.09 | BruteFrce | tmzt dpad has been working fine its the arrow syms that dont work |
| 21:33.23 | marajin | ali1234: which device are you working on the bluetooth for? |
| 21:33.33 | tmzt | silven is getting omap upstream, we can start again from there |
| 21:33.36 | tmzt | and it should work |
| 21:33.47 | BruteFrce | dont need a fake mouse matchbox can be contolled with only arrow keys |
| 21:34.01 | ali1234 | marajin: wizard. the thing about it is that the artemis driver should basically work unmodified, except possibly for one gpio needing to be changed. |
| 21:34.05 | tmzt | <PROTECTED> |
| 21:34.32 | BruteFrce | artemis kernel wont work on my device |
| 21:34.32 | tmzt | and bt means bt pan if you can make it work on your pc there |
| 21:34.40 | BruteFrce | first thing i tried |
| 21:35.06 | tmzt | your device is practically the same as wizard/artemis/elf(in)/etc. even n95 nokia |
| 21:35.13 | marajin | ali1234: yes but it's never that simple :p |
| 21:35.14 | ali1234 | BruteFrce: artemis is approximately as different to wizard as it is to excalibur |
| 21:35.17 | tmzt | except for omapfb and keyboard |
| 21:35.34 | BruteFrce | well it didnt work even with lcd patch |
| 21:35.55 | tmzt | you change mtype too? |
| 21:36.19 | BruteFrce | n95 thatsmy dream device |
| 21:36.34 | tmzt | needs a symbian bootloader/haret |
| 21:36.56 | BruteFrce | it keept busting trying to init |
| 21:37.05 | BruteFrce | and having ramdisk failure |
| 21:37.28 | tmzt | you still using squashfs? |
| 21:37.59 | BruteFrce | for linwizard ya compresed ext2 was to big |
| 21:38.28 | druidu | cr2 NetRipper : looks like I can do DMA and update the screen, problem is that I need to give him a ioremapped addr I think |
| 21:38.42 | NetRipper | druidu, cool |
| 21:39.18 | *** join/#htc-linux mooky (n=suzannet@unaffiliated/mooky) |
| 21:39.54 | BruteFrce | ali could you try building the atemis kernel with the excaliburpatches for shits and giggles? |
| 21:40.02 | BruteFrce | ali1234 |
| 21:40.22 | BruteFrce | can i use the same userspace tmzt? |
| 21:40.31 | NetRipper | druidu, so you actually updated the screen? like blackened it or something? |
| 21:40.39 | ali1234 | BruteFrce: i dont have enough disk space for another git tree |
| 21:40.59 | NetRipper | brb |
| 21:41.02 | ali1234 | BruteFrce: userspaces are pretty much interchangable between any arm device |
| 21:41.03 | BruteFrce | lol sure okay, thx anyays |
| 21:41.32 | ali1234 | BruteFrce: for example i booted the vanilla openmoko userspace on top of linwizard kernel |
| 21:42.06 | BruteFrce | no disk space in anage where you can geta 1.5 tb hdd for 189 usd |
| 21:42.29 | BruteFrce | what is that like 95 quid |
| 21:42.59 | druidu | nope |
| 21:43.01 | druidu | it's white :) |
| 21:43.06 | druidu | full white |
| 21:43.09 | *** join/#htc-linux infernixx (n=nix@unaffiliated/infernix) |
| 21:43.16 | tmzt | I've never seen a driver written in COBOL before |
| 21:43.20 | druidu | problem is I need to ioremap some memory and write in that |
| 21:43.28 | druidu | hahaha, lol |
| 21:44.22 | druidu | it's like writing a driver in java |
| 21:44.22 | druidu | who would do that? |
| 21:44.34 | BruteFrce | i hate to give up after all this progress but i dunno howto approach this arrow problem, its frustrating s shit |
| 21:44.58 | tmzt | what problem? |
| 21:45.10 | BruteFrce | the arrows |
| 21:45.35 | BruteFrce | they dont work no matter what we map them too dpad or qwerty keys |
| 21:45.57 | BruteFrce | and we need them to navigate matchbox |
| 21:46.38 | BruteFrce | also figured out the ts calibration was what was messing up startx |
| 21:46.51 | BruteFrce | penguinz did |
| 21:47.58 | par | did he fix it? |
| 21:48.29 | BruteFrce | ya we can use startx now he also gave x keyboard support |
| 21:48.35 | par | oh cool |
| 21:48.42 | BruteFrce | but none of it matters with no arrows |
| 21:48.46 | par | how are things with the dpad |
| 21:48.59 | BruteFrce | good works fine |
| 21:49.10 | BruteFrce | just cant map it to arrows |
| 21:49.24 | BruteFrce | cant map anything to them |
| 21:49.32 | par | ok, so yuou know what the dpad is then already |
| 21:49.44 | par | what you mean arrows? |
| 21:49.58 | BruteFrce | ya we can map like lettters to the dpad and it works fine |
| 21:50.08 | BruteFrce | i mean keyboard arrows |
| 21:50.12 | par | ohhh |
| 21:50.34 | BruteFrce | key_up key_down key_left key_right |
| 21:50.48 | par | this is on which phone.. diamond? |
| 21:51.10 | BruteFrce | there useless, showkey shows the right keycode when there mapped, they just dont function |
| 21:51.12 | BruteFrce | excalibur |
| 21:51.17 | par | ah ok |
| 21:51.47 | BruteFrce | its all we need for full control of matchbox |
| 21:52.01 | BruteFrce | such a weird problem |
| 21:53.02 | par | yeah, i don't know what to say |
| 21:53.11 | par | what are the ascii code for those keys? |
| 21:53.41 | BruteFrce | dont know |
| 21:53.44 | *** join/#htc-linux echelon` (n=justin@76.211.5.164) |
| 21:53.51 | echelon` | Tim |
| 21:53.53 | echelon` | Tim Meade |
| 21:53.56 | echelon` | is he here? |
| 21:54.13 | BruteFrce | lol whats his nick? |
| 21:54.26 | echelon` | no idea |
| 21:54.29 | echelon` | he's from Ohio too |
| 21:54.31 | echelon` | like me |
| 21:54.49 | BruteFrce | why would he be here? |
| 21:54.53 | par | cr2, are you here? were you serious about cdma modem on ba board id 4? |
| 21:54.55 | *** join/#htc-linux mickeyl (i=mickey@80.81.242.146) |
| 21:54.58 | echelon` | he told me he comes here |
| 21:55.29 | BruteFrce | well people dont usually disclose there real names here so i duno |
| 21:55.57 | BruteFrce | full retail g1 is harder to find than a left handed circle jerk |
| 21:56.40 | par | cr2, if apache was hard.. i would think that ba boardid 4 was already done? i dunno i haven't had a chance to look at the current code base to see where the ba is at. If you're serious i can test anyway.. even with no service i can still try to get the responses from the modem |
| 21:57.11 | cr2 | par: i#ve heard from somebody that cdma ba (harrier) work |
| 21:57.21 | par | ohhh ok then. |
| 21:57.33 | cr2 | par: do you have 'cu' ? |
| 21:57.43 | par | the call unix command? |
| 21:57.55 | cr2 | if at command will respond, then you'll manage it |
| 21:57.57 | cr2 | yes |
| 21:57.58 | BruteFrce | so does any device so far have working gprs/edge? |
| 21:58.01 | cr2 | or minicom |
| 21:58.06 | par | i'll have to build an image |
| 21:58.24 | par | for my ba boardid 4 |
| 21:58.34 | cr2 | take an existng one |
| 21:58.41 | ali1234 | BruteFrce: kaiser does |
| 21:58.55 | par | one of the angstrom autobuilds you mean? |
| 21:58.58 | cr2 | par: and pick the cu from angstrom repository |
| 21:58.59 | BruteFrce | but it has no x? |
| 21:59.04 | par | okay |
| 21:59.07 | cr2 | par: if it is missing |
| 21:59.17 | cr2 | par: linuxtogo~htcpxa |
| 21:59.31 | par | i'll have to do it later though.. need to sleep now.. working graveyard shift 2night ;) |
| 22:00.00 | BruteFrce | man that sucks net and no x, is it able to compile? |
| 22:00.07 | tmzt | reading source on git |
| 22:00.22 | par | is anyone actively trying to make kernel patches so that Android will run properly on the ba |
| 22:00.23 | echelon` | tmzt: don't ignore me |
| 22:00.28 | echelon` | it's your buddy from home! |
| 22:00.29 | echelon` | :P |
| 22:00.53 | tmzt | hey, I figured that out |
| 22:01.00 | echelon` | hahaha |
| 22:01.07 | echelon` | learn to be happy |
| 22:01.25 | druidu | swetland, you around? |
| 22:01.30 | par | i mean a sperate set of patches to be able to run android on the ba.. |
| 22:01.57 | druidu | was wondering if you can tell me some more about the FB_ADDR, what exactly it represents, is it board definied? |
| 22:02.30 | par | most of that stuff is already widely known for models like ba and universal |
| 22:03.07 | par | shouldn't be hard for someone to force 240x320 in hw with the fb |
| 22:03.47 | druidu | well, let's say I transfer stuff from memory to LCD via DMA |
| 22:03.49 | cr2 | druidu: no |
| 22:03.53 | druidu | i'm not sure where it gets it from |
| 22:04.07 | druidu | I get a white screen, so dma si succesful but it doesn't get data from main memory |
| 22:04.13 | dcordes | BruteFrce, yea it sucks a lot. I wish somebody who has the kaiser and knows more about those things as me could give it a go |
| 22:04.18 | cr2 | druidu: we will move it to SRAM eventually |
| 22:04.18 | druidu | (i have it garbage from kernel start) |
| 22:04.36 | druidu | **eventually** |
| 22:04.44 | druidu | right now I'd be happy with anything :) |
| 22:04.46 | dcordes | BruteFrce, tmzt helped me a lot debugging it and trying but no success |
| 22:04.48 | cr2 | when it work with the wince config |
| 22:04.59 | druidu | just wanna get the lcd to display text so I can debug properly |
| 22:05.04 | druidu | good ol' printk :) |
| 22:05.04 | cr2 | druidu: g1 runs from SRAM |
| 22:05.18 | cr2 | yes, it's clear |
| 22:05.21 | druidu | well, it has a fixed addr |
| 22:05.24 | druidu | does ioremap |
| 22:05.43 | druidu | and the startup logo at least is written in ioremapped region |
| 22:05.53 | cr2 | where do you want to ioremap ? |
| 22:06.02 | druidu | but ioremap implies it must corespond to an actual excternal peripherial memory? |
| 22:06.11 | druidu | anywhere really... |
| 22:06.14 | cr2 | NetRipper: how far do you come before crashing ? |
| 22:06.20 | druidu | but you need ioremap to comunicate with peripherals |
| 22:06.24 | NetRipper | cr2, don't know |
| 22:06.30 | druidu | end of start_kernel? |
| 22:06.32 | NetRipper | cr2, unlike what i thought, i didnt hav time |
| 22:06.41 | druidu | mine still crashes @ kmem_start_cache in start_kernel |
| 22:06.44 | NetRipper | druidu, it should be getting further |
| 22:06.48 | cr2 | druidu: ioremap just creates some ptable entries for your ram. |
| 22:06.52 | NetRipper | druidu, it does? hrm :x |
| 22:06.59 | cr2 | then you start it too early. |
| 22:07.08 | druidu | dunno, strange |
| 22:07.20 | druidu | cr2 : I don't, it's just the natural order of start_kernel |
| 22:07.56 | cr2 | NetRipper: i think we should start with the g1 kernel, remove noncritical drivers from config, fix gpios, and comment out the proc_comm() for the clocks. |
| 22:08.02 | cr2 | NetRipper: then it may boot. |
| 22:09.05 | NetRipper | druidu, if you do that ioremap before kmem_cache_init() it may not work.. it initialites stuff that has tod o with the page tables after that |
| 22:09.54 | NetRipper | druidu, however you can try using virtual addresses that are already predefined (as what i am doing with GPIO addresses to turn on flashlight) |
| 22:09.59 | NetRipper | i dont ioremap() |
| 22:10.47 | cr2 | hm. fbram may be statically remaped |
| 22:11.03 | cr2 | i think it is, but i'm not 100% sure |
| 22:11.43 | tmzt | cr2: any hope for uart then? |
| 22:11.50 | NetRipper | there are probably defines for fbram |
| 22:12.03 | druidu | it fails in kmem_cache_init -> kmem_cache_create -> kmem_cache_alloc -> cache_alloc_refill -> cache_grow -> cache_init_objs |
| 22:12.11 | cr2 | tmzt: where ? |
| 22:12.16 | druidu | or something like that, I'm tired of debugging with no screen :D |
| 22:12.36 | tmzt | htc extusb |
| 22:12.39 | cr2 | druidu: which kernel tree are you using ? |
| 22:12.41 | druidu | NetRipper: but GPIOs are ioremapped! |
| 22:12.46 | cr2 | tmzt: too much pita |
| 22:12.50 | druidu | otherwise it wouldn't work |
| 22:12.56 | druidu | (ioremapped by the kernel I think) |
| 22:13.06 | druidu | android-msm-htc-2.6.25 |
| 22:13.14 | cr2 | tmzt: if you will really drive the gpios directly. as h2w does. |
| 22:13.20 | tmzt | I had some old usb phone cables with pl2303 at 3.3v, will probably make a sio cable |
| 22:13.26 | tmzt | but then I don't have htc yet |
| 22:13.34 | NetRipper | cr2, by g1 kernel you mean that android-msm-htc-2.6.25 i presume |
| 22:13.35 | tmzt | but I can build an emu cable for q |
| 22:13.55 | cr2 | tmzt: it's possible to send some morse-type messages, but you need then to decode them off the extusb pins. |
| 22:14.16 | cr2 | and it is necessary to find out which pins are actially connected to which gpio. |
| 22:14.34 | cr2 | io pins, not the cable detect et al. |
| 22:14.46 | cr2 | the last are already known, and are just inut pins. |
| 22:15.30 | *** part/#htc-linux Kalemas (i=ejabberd@jabbus.org) |
| 22:15.35 | tmzt | so it's not 1wire or something, I thought that's what h2w was |
| 22:15.41 | *** join/#htc-linux Kalemas (i=ejabberd@217.147.29.117) |
| 22:16.12 | druidu | gonna try to add the fbmem to msm_map_common_io list |
| 22:16.27 | NetRipper | :) |
| 22:16.37 | NetRipper | and im gonna try to sleep |
| 22:16.39 | NetRipper | good night |
| 22:17.20 | tmzt | cr2: I have some ssp dumps from haret on pxa27x, do you know how to get the bytes read/written from there? |
| 22:19.30 | BabelO | cr2: google do the fix very fast :) |
| 22:22.36 | druidu | I think the remap worked, let's see... |
| 22:23.12 | druidu | cr2: hehe, we could also flash the screen like that or something and hook it to a webcam and decode it like that |
| 22:23.20 | druidu | yes, back to smoke signals :) |
| 22:24.09 | NetRipper | lol i thought of doing that with flashlight |
| 22:24.22 | NetRipper | like turn it off and on after each function to see how far it gets |
| 22:24.34 | NetRipper | but i figured my eyes aren't fast enough to count |
| 22:24.34 | NetRipper | :P |
| 22:25.04 | NetRipper | anyway, i was off already ;) |
| 22:26.40 | druidu | you could make a debug function with for( ) to wait |
| 22:27.16 | dcordes | druidu, http://headfuzz.co.uk/?q=kaiserledhack |
| 22:28.26 | druidu | cool, I actually thought of doing that |
| 22:28.51 | druidu | but my "flashlight" is one-shot at the moment :) |
| 22:29.26 | *** join/#htc-linux szsoftware (n=sven@e178210022.adsl.alicedsl.de) |
| 22:32.08 | druidu | I managed (I think) to remap the original physical wince fb, but still white screen |
| 22:36.17 | druidu | :q |
| 22:42.50 | druidu | YESSSSSSSS |
| 22:42.52 | druidu | it worked |
| 22:43.22 | druidu | I have some stuff on the screen |
| 22:43.24 | druidu | like noise |
| 22:43.25 | druidu | :) |
| 22:45.19 | dcordes | ~hail druidu |
| 22:45.20 | apt | ACTION bows down to druidu and chants, "I'M NOT WORTHY!!" |
| 22:45.47 | druidu | NetRipper was write, gotta ioremap it and send the lcd the physical address |
| 22:46.11 | druidu | which seems logical, although in the msm_fb code they send the ioremapped one (in which they write directly) |
| 22:46.20 | *** join/#htc-linux romak (n=romario@89-185-235-7.static.masterinter.net) |
| 22:47.20 | druidu | hahaha, I got the haret boot! |
| 22:47.33 | druidu | I mean, I got the haret boot messages which were written to FB |
| 22:48.41 | druidu | problem is I don't know how to write to that memory location |
| 22:48.42 | cr2 | druidu: the 'dots' go to /dev/null |
| 22:49.00 | druidu | dots? |
| 22:53.15 | dcordes | good night |
| 23:02.21 | ali1234 | BabelO: i did some tracing and it looks like wizard even uses same gpios for BT |
| 23:02.55 | BabelO | ali1234: good :) |
| 23:03.09 | BabelO | ali1234: you can try my code then |
| 23:03.58 | ali1234 | BabelO: the "normal" gpios are the same but i'm not sure about the egpio which looks like the most important one |
| 23:07.02 | ali1234 | BabelO: will i be able to tell if it is working by watching for the blue LED? |
| 23:07.33 | ali1234 | or does that require firmware download? |
| 23:07.38 | BabelO | ali1234: hmm for me blue led is independant |
| 23:07.57 | BabelO | ali1234: no firmware needed for normal use |
| 23:08.08 | *** part/#htc-linux szsoftware (n=sven@e178210022.adsl.alicedsl.de) |
| 23:08.11 | BabelO | blue led is activated by others gpio, not bt related |
| 23:14.01 | *** join/#htc-linux druidu (n=root@213.164.231.5) |
| 23:14.39 | cr2 | ali1234: i have not seen the bt cpld gpios on wiza |
| 23:20.33 | ali1234 | well this is what result i got tracing with haret: http://pastebin.com/m3da580d7 |
| 23:21.38 | BabelO | ali1234, cr2 i2c cpld activation is the blue LED i think |
| 23:23.39 | cr2 | BabelO: maybe the led, but i have not seen i2c-cpld in bt spl startup |
| 23:24.23 | ali1234 | does haret ignore egpio/cpld? i didnt see any trace output when the blue led flashed |
| 23:28.51 | BabelO | good night |
| 23:30.06 | *** join/#htc-linux BruteFrce (n=IrcNet@m430e36d0.tmodns.net) |
| 23:32.13 | *** join/#htc-linux infernixxx (i=gerben@89.205.241.44) |
| 23:41.54 | druidu | ok, gimme 1 hour and I'll make a console driver :D |
| 23:42.12 | druidu | NetRipper : do you have ram virtual console from android-msm tree? |
| 23:44.58 | *** join/#htc-linux BruteFrce (n=IrcNet@m430e36d0.tmodns.net) |
| 23:46.09 | BruteFrce | ever feed your dog an onion, its hillarious |
| 23:50.09 | *** join/#htc-linux OpnSrc1 (i=harisha1@gateway/tor/x-6e91c7a3b8279ed1) |
| 23:53.58 | *** join/#htc-linux surgex (i=surge@pool-98-118-165-224.bflony.fios.verizon.net) |
