| 00:45.28 | Kevin2 | cr2: Are you around? |
| 00:55.31 | *** join/#htc-linux lamikr (n=chatzill@aragorn.kortex.jyu.fi) |
| 01:39.38 | *** join/#htc-linux ljp (n=lpotter@203.94.178.46) |
| 03:46.41 | *** join/#htc-linux rob__w (n=bob@p85.212.175.5.tisdip.tiscali.de) |
| 06:20.59 | *** join/#htc-linux rob_w (n=rob_w@p549B8F2F.dip0.t-ipconnect.de) |
| 08:34.06 | *** join/#htc-linux florian_kc (n=fuchs@dns.kernelconcepts.de) |
| 09:17.40 | *** join/#htc-linux psokolovsky (n=psokolov@236.usernat.ip.net.ua) |
| 10:06.02 | *** join/#htc-linux rob_w (n=rob_w@p549B8F2F.dip0.t-ipconnect.de) |
| 11:14.59 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 13:16.06 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 13:56.56 | rob__w | lamikr, u there , PLEASE !! |
| 14:03.10 | rob__w | lamikr, please check your sourceforge mails !! |
| 14:28.21 | lamikr | rob_w: Hi |
| 14:30.15 | rob__w | hey lamikr |
| 14:30.42 | rob__w | did you notice http://projects.linuxtogo.org/projects/gomunicator/ |
| 14:31.41 | rob__w | if you u like please create account there , otherwise i need to sync your changes every time to there |
| 14:33.37 | lamikr | rob_w: Well, yes I noticed that it is now also there. I have used the old repo as I have commit access there and I want to make essential changes as fast as possible because I have wanted to get my phone usable. |
| 14:34.53 | lamikr | rob_w: Sorry about that, I have now some sms phone number related fixes also pending that are not yet committed. |
| 14:34.59 | rob__w | <PROTECTED> |
| 14:35.47 | rob__w | i would be glad to have u there, ur the gomunicator guy these days |
| 14:36.15 | rob__w | but i can also sync the cvs and then you just need to create your account |
| 14:36.46 | rob__w | bring in your next changes, then i could do today evening the sync and get in the account |
| 14:37.14 | lamikr | I just read some email from the handhelds mailing list which made me a little bit nervous from the move to the linuxtogo, so I just skipped that thing right now. One thing was that there were none discussion from the move in public lists. |
| 14:38.11 | rob__w | hmm , yet xanadux should be the a hardware thing and gomunicator just lives in their cvs |
| 14:41.13 | rob__w | anyway its your decision, at least if you still use xanadux it will keep me busy integrating ur good stuff |
| 14:41.28 | lamikr | Well that's kind of true also. But I do not understand the move of most of the other projects from handhelds to another one... Both the oe, xanadux and handhelds have all served me fine. So sudden capture of multiple projects to one place and move of the libraries from gpl to lgpl is something I really want to think before I made my move. |
| 14:41.55 | rob__w | oh , what moved to lgpl ? |
| 14:45.50 | florian_kc | lamikr: hmm... what changed to lgpl? |
| 14:46.03 | lamikr | rob_w: I have not followed the details but I read some mails that gpe libraries are tried to relicence. I do not know has that already happened because I have been quite busy with kernel (and now with gomunicator). My plan was to make still couple of small changes to gomunicator and once I have got my phone to daily usable form take a break and then check what this linuxtogo is. (It might be... |
| 14:46.05 | lamikr | ...very nice thing, the thing just is that I have not had time to check it... :-) |
| 14:46.29 | lamikr | florian_kc: Hi |
| 14:46.35 | florian_kc | lamikr: hi :-) |
| 14:46.57 | florian_kc | lamikr: if you refer to this nonsens reenoo posted to the gpe mailinglist... |
| 14:47.11 | lamikr | I am really not sure has that happened but at least I read that there has been some plans. Do you know anything about that? |
| 14:47.30 | florian_kc | one of the basic ideas of gpe was to have the libraries lgpl and the applications gpl. |
| 14:47.42 | florian_kc | somehow we lost this in some libraries... |
| 14:48.16 | rob__w | florian_kc, but that only applies to libgpe or such ,, or will gtk itself move to lgpl ? |
| 14:48.32 | lamikr | florian_kc: Yes, I read that from the list. I do not neccessarily agree that lgpl:ing all libs is good idea. Sometimes most of the functionality is there and UI has just the cores. |
| 14:48.36 | florian_kc | rob__w: gtk _is_ lgpl |
| 14:49.02 | rob__w | really ?!? hang , that turns me view of reality around |
| 14:49.41 | florian_kc | lamikr: yes that's true, but when people started with gpe (that was before my time) they had in mind that maybe people want to start some closed crap ontop of it which should be possible at least. |
| 14:51.50 | lamikr | florian_kc: Yes, but the tendency to closed side in Linux seems to be growing all the time... I would like to keep some kind of balance there and avoid not putting everything. |
| 14:52.23 | lamikr | I for example agree that gomunicator will for example need to be splitted to multiple libs, but I would like to see the SMS and MMC encoding/decoding libs under gpl for example if I start to work with them. |
| 14:52.47 | rob__w | its TRUE , gtk includes carry lgpl headers ,, |
| 14:52.51 | cr2 | rob__w: gtk being lgpl is the only reason why "commercial" interests are attracted to it. |
| 14:53.43 | rob__w | wow , i was not aware of that for a long time now |
| 14:53.54 | florian_kc | right, thats an important thing for companies... without that fact everyone used qt :-/ |
| 14:54.22 | cr2 | i don't really like c++, but will chose qt over gtk for any gui development ;) |
| 14:54.59 | rob__w | well leaves the question if a the usage of bitbake injects gpl rights ,, any ideas ? |
| 14:56.14 | lamikr | Sorry, I need to go now. I will try to talk with you in this weekend! Propably tomorrow. |
| 14:56.20 | cr2 | florian_kc: i like xlib, but it's too complicated to get a decent looking UI with it. |
| 14:56.22 | rob__w | ok |
| 14:56.49 | florian_kc | cr2: indeed... mallum tried for a ling time and finally failed. |
| 14:57.11 | florian_kc | s/ling/long |
| 15:00.11 | cr2 | and the enormous complexity of gtk leads to gtk-c# and whatever crap microsoft/novell have invented today. |
| 15:01.35 | florian_kc | scary... |
| 15:05.05 | rob__w | cr2, well normally you dont need more then what gtk serves you .. the need for more ui thingi `s is a proof of making it bloated and just fancy |
| 15:10.47 | cr2 | i can give you an example: there is a gps/routeplanner program roadmap(zroadmap) it is multitoolkit, and we wanted to add the rotated utf8 labels to the streets. i patched the qt source in 5 minutes after reading the docs. i gave up reading gtk docs after 20 minutes, because it looked scary complicated. |
| 15:13.20 | cr2 | rob__w: the phone program gui is the most important factor, how people will make an opinion about linux. |
| 15:32.51 | *** join/#htc-linux jeanseb (n=jeanseb@88.164.32.155) |
| 15:33.38 | *** join/#htc-linux goxboxlive (n=jrs@9.80-202-160.nextgentel.com) |
| 15:53.38 | rob__w | cr2, u are also right with that |
| 16:24.43 | goxboxlive | rob_w are you still here? |
| 16:25.11 | goxboxlive | or anyone else |
| 16:26.14 | goxboxlive | whats the diffrents of fat 32 (LBA) and fat 32? It's a choice i have in fdisk to format a disk |
| 16:27.04 | cr2 | which disk ? |
| 16:27.47 | goxboxlive | it's a usb disk i'm gonna use in my car |
| 16:28.01 | cr2 | in windows ? |
| 16:28.04 | goxboxlive | i found it: LBA = Logical Block Addressing |
| 16:28.21 | goxboxlive | I guess i dont need that to my car radio :-) |
| 16:28.46 | cr2 | it is some weird legacy bios related setting. |
| 16:29.07 | cr2 | LBA is better. |
| 16:31.15 | goxboxlive | ok |
| 16:31.23 | cr2 | florian_kc: i'm looking what the first stage bootloader is doing. 1. resume 2. load the second stage bootloader from the doc g3 "block device". the 2. does not look very complicated. |
| 16:31.54 | goxboxlive | Now i am gonna have a 80GB 2,5" harddisk connected to my car radio. That's a lot of music :-). |
| 16:33.08 | cr2 | resume sets some gpio & memctl values (which our drivers need to verify!) and jumps to 0xa0040000. |
| 16:33.31 | cr2 | goxboxlive: you must be a pirate :D |
| 16:34.16 | goxboxlive | it's me and captain Hook. No, i am not a pirate. |
| 16:35.43 | florian_kc | cr2: that sounds quite simple, indeed |
| 16:36.37 | rob__w | ho ho captian hook |
| 16:39.51 | cr2 | florian_kc: there are some control registers at 0x1000+, and some stuff is written there. do you have a link to the g3 docs ? |
| 16:42.42 | cr2 | nice code: ldr r6, =0x8a0f and then strb r6, [r1,#0x3e] repeated five times, where r1=0x1000 ;) |
| 16:42.46 | florian_kc | cr2: not offhand, but it was easy to find on their website |
| 16:43.00 | cr2 | the links from hh.org wiki are dead. |
| 16:54.19 | cr2 | sample code in the manual looks good. with some asic register names :) |
| 16:57.38 | cr2 | 0x200 is the doc chipid. ok. |
| 16:58.53 | *** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde) |
| 16:59.19 | *** join/#htc-linux pH5 (n=ph5@p5485F8BA.dip.t-dialin.net) |
| 17:05.09 | cr2 | hehe. not all registers are documented. htc uses more of them. |
| 17:05.19 | pH5 | cr2: what exactly should OWM_INT signal? I can write to the DS2762 now, but I lose bytes when reading. |
| 17:05.54 | cr2 | wait some time ? |
| 17:09.02 | pH5 | probably. I haven't figured out that part in the wince code yet. but I know GPIO107 starts to trigger after sending commands |
| 17:10.05 | cr2 | maybe you need to read only when it comes ? |
| 17:10.16 | *** join/#htc-linux Jayrom (n=jay@146.66.209.88.dslw0.libello.mc) |
| 17:11.02 | cr2 | pH5: do you have doc g3 ? |
| 17:11.12 | cr2 | probably not. |
| 17:12.43 | Jayrom | hello |
| 17:12.59 | Jayrom | anyone here working on the wizard-linux? |
| 17:15.37 | WizMaui | -> #linwizard ;) |
| 17:28.54 | pH5 | cr2: I don't. pxa272 has some kind of intel strataflash |
| 17:30.09 | pH5 | I guess I should figure out how the interrupt is triggered, first. |
| 17:35.27 | psokolovsky | Hi guys! |
| 17:35.46 | psokolovsky | Someone can tell me how can I check if I have ds2762 in my battery or no? ;-) |
| 17:39.00 | pH5 | hi psokolovsky. disassemble your wince's battery.dll |
| 17:40.02 | psokolovsky | pH5: Hi Philipp! Well, that will probably wait then. too cumbersome ;-). at least I know exact dll name. |
| 17:40.33 | psokolovsky | pH5, and how does code look like? Does it access OWM contoller registers directly? |
| 17:41.17 | psokolovsky | pH5, I try to disassamble wifi driver on my h4000 to find out base address of acx100. needless to say, i found nothing in ~1hr I peered into it. |
| 17:41.32 | pH5 | psokolovsky: I guess. you have asic3, right? I don't. My OWM controller is on a chip that I don't know much about yet. |
| 17:42.40 | psokolovsky | pH5, yes, it's asic3. Once I tried to port owm asic3 driver from hx4700. But I got only timeouts. That's why I wonder how to check if there at all or no (ds2762) |
| 17:43.07 | psokolovsky | for example, h1910 lacks it, and h4000 somewhat close to it in architecture |
| 17:44.04 | pH5 | you could try to watch OWM register traffic with haret |
| 17:44.15 | pH5 | or try to find out what else happens when pulling the plug |
| 17:44.26 | psokolovsky | Oh, yeah, nice idea! |
| 17:45.29 | psokolovsky | exactly what I was looking for. my mind is thinking about other things last times ;-) |
| 17:49.48 | pH5 | I understood that there is a ds2762 in my battery only months after first discovering the traffic while checking out the PXA_CSx_PHYS addresses with haret :) |
| 17:52.22 | *** join/#htc-linux florian (n=fuchs@p5083B092.dip0.t-ipconnect.de) |
| 17:55.07 | pH5 | hey florian |
| 18:18.20 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 18:20.01 | *** join/#htc-linux pleemans (n=peter@d51A5E421.access.telenet.be) |
| 18:28.17 | *** join/#htc-linux psokolovsky_ (n=psokolov@ip.85.202.124.203.dyn.sub-9.broadband.voliacable.com) |
| 18:44.52 | cr2 | psokolovsky_: back here ? |
| 18:45.22 | psokolovsky_ | cr2: Hi! |
| 18:45.47 | cr2 | hi. you have asked a lot of questions, and a lot of them have very simple answers :) |
| 18:46.19 | psokolovsky_ | cr2: "use haret"? |
| 18:46.39 | cr2 | you need to add your model to haret.cpp first. |
| 18:46.50 | cr2 | h4000 is too generic. |
| 18:48.02 | psokolovsky_ | cr2: ok, I will be looking into that |
| 18:48.12 | cr2 | btw, acx base address can be found in the tiacwlan.dll, but there are more simple ways. |
| 18:48.32 | cr2 | your cpu is pxa27x ? pxa26x ? |
| 18:48.41 | psokolovsky_ | cr2: I wonder, if you happen to be, or know, owner og gnu-haret repo @ xanadux.sf.net? |
| 18:48.48 | psokolovsky_ | pxa255 ;-) |
| 18:49.05 | psokolovsky_ | cr2: tell me simple ways, please! |
| 18:49.20 | cr2 | hmm. irq tracing will be more tricky hten. |
| 18:49.23 | *** join/#htc-linux florian (n=fuchs@p5083B092.dip0.t-ipconnect.de) |
| 18:49.45 | cr2 | first. are you using pxa flash for the bootloader ? or mdoc g3 ? |
| 18:50.08 | psokolovsky_ | cr2, its some kind of mdoc |
| 18:50.24 | cr2 | florian: m-systems documentation is bad. but it seems we will write a better one. |
| 18:50.51 | cr2 | psokolovsky_: is it pxa255 without flash ? |
| 18:51.10 | psokolovsky_ | cr2: yes, afaik, all pxa25x w/o flash |
| 18:51.15 | cr2 | ok. |
| 18:51.42 | cr2 | can you dump and decode the wince rom ? |
| 18:52.16 | psokolovsky_ | cr2: dump, sure. as for decoding, I did kind of that too. |
| 18:52.32 | psokolovsky_ | i have dump on my hands |
| 18:52.41 | cr2 | oh. there is an easier way. dump RAM starting from 0xa0400000 |
| 18:52.45 | florian | cr2: heh |
| 18:52.54 | cr2 | look at the first page. |
| 18:54.00 | cr2 | or if you have the nk.exe, run arm-wince-pe-objdump -d nk.exe and post the starting code to pastebin. |
| 18:54.01 | psokolovsky_ | cr2: keep telling, because I can't do that right now. or is feedback from me needed? |
| 18:54.17 | psokolovsky_ | cr2: let me check |
| 18:54.40 | cr2 | ok. you will be able to see the wince static page table. you can also run "dump mmu" in haret. |
| 18:55.01 | cr2 | there are many ways to achieve good results :) |
| 18:56.18 | cr2 | florian: the documented only some registers, but we will be able to write a simple (readonly?) mtd block driver. |
| 18:56.53 | psokolovsky_ | cr2: ok, and should that make obvious what acx100 base adress is? look for a page surrounded by nothing? |
| 18:58.10 | cr2 | you will be able to write something like that: http://wiki.xda-developers.com/index.php?pagename=UniversalMemoryMap |
| 18:58.55 | florian | cr2: wasn't that quite trivial to use... like nor or so? |
| 18:59.27 | cr2 | since we know where is your asic3 (and there is haret support for it) usually you have only 2 choices for the wifi address space. |
| 19:01.21 | psokolovsky_ | cr2, btw, gnu-haret doesn't support asic3 in h4000 and few other models (at least, some older version didn't) |
| 19:02.01 | cr2 | what is h4000 ? it is trivial to add your model to haret. |
| 19:03.00 | cr2 | there are some minor adjustments needed, because you have pxa255. |
| 19:03.10 | psokolovsky_ | cr2: yep, and it should be done. |
| 19:04.04 | psokolovsky_ | cr2: anyway, are you owner of gnu-haret repo? there was some movement recently to revive original haret, and have single common tool. Would you be interested? |
| 19:04.24 | cr2 | then it will be userfriendly. |
| 19:04.38 | cr2 | i'm not the owner, but i have write access. |
| 19:05.03 | cr2 | it is not going to be very easy. |
| 19:06.05 | psokolovsky_ | yep, not easy |
| 19:06.07 | cr2 | there is a lot of ugly and pxa27x-specific code there. |
| 19:06.48 | psokolovsky_ | cr2: exactly the problem I have with gnu-haret ;-) so, backporting stuff to orig. haret would allow to clean up it at the same time... |
| 19:07.20 | cr2 | it uses AllocPhysMem for booting the kernel, which is not available in the older wince versions. |
| 19:08.23 | cr2 | i think good parts from gnu-haret should be simply taken and put into mainline. there are many useful and portable things there that will run on any wince. even non-arm. |
| 19:11.02 | cr2 | there is a name clash of h4000 and h4700, if we will add 'iPaq h4' to haret.cpp |
| 19:11.48 | cr2 | the asic3 addresses are ok. http://wiki.xda-developers.com/index.php?pagename=ASIC3 |
| 19:14.51 | cr2 | pH5: the IPL is changing the CCCR on resume. and the GPIO settings, but it is another story. |
| 19:15.40 | psokolovsky_ | cr2: yep, it hsould be hx4700. mach-types is wrong there ;-) |
| 19:16.27 | pH5 | cr2: what is IPL? |
| 19:17.07 | psokolovsky_ | cr2: remind me how can I reach handheld by activesync usb? I suddenly getting feeling I can't ;-) |
| 19:17.58 | cr2 | psokolovsky_: telnet to haret ? |
| 19:18.09 | psokolovsky_ | cr2: ip address? ;-) |
| 19:18.15 | cr2 | pH5: wince5 has 2 stage bootloader. this is the first stage. |
| 19:18.48 | cr2 | psokolovsky_: hmm. route.exe ? |
| 19:19.10 | cr2 | i don't use windows :) it is some standard address. |
| 19:20.00 | psokolovsky_ | cr2: that's why I ask. last time I get strange holes in memory. ;-) Did activesync ever up'ed an interface for usb connection? I have deja vu feeling of "no" |
| 19:20.28 | cr2 | it should show you the icon in the tray. |
| 19:21.02 | cr2 | but with 'route' you should see if you have the 192.168.* local interface |
| 19:21.23 | psokolovsky_ | cr2: oh yes, it does. but no if in ipconfig. what you guys use to access haret ? |
| 19:21.37 | cr2 | to save such headaches, i use wifi :) |
| 19:22.10 | cr2 | i heard it is possible to use bluetooth, but probably only on wince5. |
| 19:22.25 | psokolovsky_ | cr2: me too! But this time I decided to figure out if it can be done by AS. lol, no |
| 19:22.27 | cr2 | the irda will not work afaik. |
| 19:22.50 | psokolovsky_ | cr2: ok, np, again, just checking my memory - maybe I missing something. will use wifi ;-) |
| 19:23.05 | cr2 | irda is PITA anyway. |
| 19:27.04 | psokolovsky_ | cr2: Ok, I have gnu-haret's dump mmu before me. anything I can look for in this long dump? |
| 19:28.34 | pH5 | cr2: does it restore CCCR from some memory location? I had a problem with the bootloader setting CCCR to some bogus value on resume. |
| 19:28.44 | cr2 | 0x08000000 and 0x0a000000 , where they are mapped. |
| 19:29.08 | cr2 | pH5: LDR r4, =0x2000190 |
| 19:29.37 | cr2 | ldr r2, CCCR |
| 19:29.58 | cr2 | str r4, [r2] |
| 19:30.18 | cr2 | mov r0, #2 |
| 19:30.32 | cr2 | mcr p14, 0, r0,c6,c0 |
| 19:30.36 | cr2 | that's it. |
| 19:31.38 | psokolovsky_ | cr2: well, I clearly see asic3 mapped: |
| 19:31.39 | psokolovsky_ | <PROTECTED> |
| 19:31.39 | psokolovsky_ | <PROTECTED> |
| 19:31.40 | psokolovsky_ | <PROTECTED> |
| 19:31.59 | psokolovsky_ | but neither 0x08000000 nor 0x0a000000 |
| 19:32.00 | cr2 | asic3 gpio. |
| 19:32.15 | cr2 | 0x14000000 ? |
| 19:32.49 | cr2 | because asic3 sdio is at 0x10000000 |
| 19:32.52 | psokolovsky_ | <PROTECTED> |
| 19:32.52 | psokolovsky_ | <PROTECTED> |
| 19:32.52 | psokolovsky_ | <PROTECTED> |
| 19:32.52 | psokolovsky_ | <PROTECTED> |
| 19:32.52 | psokolovsky_ | <PROTECTED> |
| 19:32.52 | psokolovsky_ | <PROTECTED> |
| 19:32.54 | psokolovsky_ | yep |
| 19:33.08 | psokolovsky_ | so, 14-16 are candidates? |
| 19:33.24 | cr2 | no. htc uses them for the debugging card. |
| 19:33.33 | cr2 | 15 and 16 i mean :) |
| 19:33.38 | cr2 | so it should be 0x14 |
| 19:34.25 | psokolovsky_ | cr2: well, cool analisys! Many thanks! I'll try to plug that soon. |
| 19:34.46 | cr2 | you may find the 0xbde00000 somewhere in the tiacxwlan.dll but it is very obfuscated code there. |
| 19:35.35 | psokolovsky_ | cr2: do you say that wince uses static mapping, known at compile-time? |
| 19:36.14 | cr2 | yes. 'pd 0xa040000 0x4000' and you will see it in the hexdump. |
| 19:36.26 | cr2 | pd 0xa0400000 0x4000 |
| 19:36.36 | cr2 | damn. |
| 19:36.44 | cr2 | pd 0xa0040000 0x4000 |
| 19:36.49 | pH5 | :-) |
| 19:37.09 | psokolovsky_ | cr2: ok, thanks |
| 19:38.20 | cr2 | the table is made of triples: virtual,physical,size. it can be relatively easily seen. usually it is followed by the pxa gpio setup pairs. |
| 19:43.14 | psokolovsky_ | cr2: just fyi, pd 0xa0040000 0x4000 is all zeroes for me. but I'm on wm2003 anyway. |
| 19:44.11 | cr2 | all ? |
| 19:46.26 | cr2 | if you have nk.exe, you can try objdump. |
| 19:58.43 | cr2 | florian: http://wiki.xda-developers.com/index.php?pagename=HTC_DiskOnChip |
| 19:59.50 | *** join/#htc-linux pH5_ (n=ph5@p5485DF7A.dip.t-dialin.net) |
| 20:01.08 | florian | cr2: very good! |
| 20:03.09 | cr2 | the undocumented regs are used by the bootloader load code on the universal. |
| 20:03.36 | cr2 | if you will find better docs... at least we know how they can be used. |
| 21:13.44 | goxboxlive | florian are you there? |
| 21:13.57 | florian | goxboxlive: yep |
| 21:14.12 | goxboxlive | i have problems building binutils-cross |
| 21:14.24 | goxboxlive | do you know why? |
| 21:20.11 | florian | goxboxlive: not really... whar kind of problems? |
| 21:20.55 | goxboxlive | it just wouldnt build. I just failes. I'll lookat the local.conf file |
| 21:21.09 | cr2 | goxboxlive: i have rebuilt opie today. didn't have any problems. |
| 21:22.29 | *** join/#htc-linux psokolovsky (n=psokolov@ip.85.202.124.203.dyn.sub-9.broadband.voliacable.com) |
| 21:23.12 | goxboxlive | my problem is that it wouldnt buil binutils-cross. Your where build a long time ago i guess. Check your version. is it 2.15 or .16 or .17? |
| 21:34.38 | goxboxlive | cr2: is your Opie image based on 2.6.17-hh1 or MAIN? If you are using the stock handhelds-pxa-2.6_cvs.bb file you use -hh1. Have you managed to change it to use the MAIN source? |
| 22:00.00 | goxboxlive | and now glibc failes |
| 22:00.08 | goxboxlive | I hate this starting all over again |
| 22:52.54 | *** join/#htc-linux psokolovsky_ (n=psokolov@ip.85.202.124.203.dyn.sub-9.broadband.voliacable.com) |
| 22:54.32 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 22:58.08 | goxboxlive | psokolovsky_ : are you familiar with OE bitbake? |
| 23:00.22 | goxboxlive | take a look here. I am failing building glibc: http_//pastebin.ca/186446 |
| 23:00.44 | goxboxlive | florian are you awake? |
| 23:03.18 | florian | yep |
| 23:03.37 | goxboxlive | any idea whats wrong here? pastebin.ca/186446 |
| 23:06.17 | florian | uh uglly |
| 23:06.38 | florian | is this latest stuff from .dev? |
| 23:09.51 | goxboxlive | yes |
| 23:14.37 | psokolovsky_ | goxboxlive, weird. I just built flibc from familiar-unstable. |
| 23:15.03 | goxboxlive | hmm, |
| 23:20.39 | goxboxlive | could it be that i miss gcc4 on my desktop pc? I have onsly gcc-3.4 |
| 23:20.43 | psokolovsky_ | goxboxlive, that would be hardly the cause, as that seems to be from arm cross-gcc |
| 23:20.43 | psokolovsky_ | goxboxlive, what's you local.conf still? |
| 23:20.46 | goxboxlive | hold on |
| 23:20.46 | goxboxlive | http://pastebin.ca/186456 |
| 23:20.49 | psokolovsky_ | goxboxlive, did that build for you before or is this first build? |
| 23:20.49 | goxboxlive | this is the first one. It's a fresh .dev source |
| 23:21.06 | psokolovsky_ | goxboxlive, you don't suspect ccache? |
| 23:21.30 | psokolovsky_ | goxboxlive, otherwise, my build cmdline is much longer than yours |
| 23:21.44 | goxboxlive | like? |
| 23:22.07 | psokolovsky_ | arm-linux-gcc -march=armv5te -mtune=xscale ../sysdeps/wordsize-32/divdi3.c -c -std=gnu99 -O2 -Wall -Winline -Wstrict-prototype |
| 23:22.08 | psokolovsky_ | s -Wwrite-strings -fexpensive-optimizations -frename-registers -isystem/home/pfalcon/linux-ppc/build-oe-familiar/tmp/staging/a |
| 23:22.08 | psokolovsky_ | rm-linux/include -fomit-frame-pointer -fPIC -I../include -I. -I/home/pfalcon/linux-ppc/build-oe-familiar/tmp/work/armv5te- |
| 23:22.14 | psokolovsky_ | linux/glibc-2.3.5+cvs20050627-r12/build-arm-linux/csu -I.. -I../libio -I/home/pfalcon/linux-ppc/build-oe-familiar/tmp/work/ar |
| 23:22.17 | psokolovsky_ | mv5te-linux/glibc-2.3.5+cvs20050627-r12/build-arm-linux -I../ports/sysdeps/arm/elf -I../ports/sysdeps/unix/sysv/linux/arm/linu |
| 23:22.20 | psokolovsky_ | xthreads -I../ports/sysdeps/unix/sysv/linux/arm -I../ports/sysdeps/unix/sysv/linux -I../ports/sysdeps/unix/sysv -I../ports/sys |
| 23:22.23 | psokolovsky_ | deps/unix/arm -I../ports/sysdeps/unix -I../ports/sysdeps/arm/fpu -I../ports/sysdeps/arm/linuxthreads -I../ports/sysdeps/arm -I |
| 23:22.26 | psokolovsky_ | ../sysdeps/wordsize-32 -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754/dbl-64 -I../linuxthreads/sysdeps/unix/sysv/linux -I../ |
| 23:22.29 | psokolovsky_ | linuxthreads/sysdeps/pthread -I../sysdeps/pthread -I../linuxthreads/sysdeps/unix/sysv -I../linuxthreads/sysdeps/unix -I../sysd |
| 23:22.32 | psokolovsky_ | eps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet -I../sysdeps/unix/ |
| 23:22.35 | psokolovsky_ | sysv -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic -nostdinc -isyste |
| 23:22.38 | psokolovsky_ | m /home/pfalcon/linux-ppc/build-oe-familiar/tmp/cross/lib/gcc/arm-linux/3.4.4/include -isystem /home/pfalcon/linux-ppc/build-o |
| 23:22.43 | psokolovsky_ | e-familiar/tmp/cross/arm-linux/include -D_LIBC_REENTRANT -include ../include/libc-symbols.h -DPIC -DSHARED -DHAVE_INITFIN |
| 23:22.44 | goxboxlive | i c |
| 23:22.46 | psokolovsky_ | I -o /home/pfalcon/linux-ppc/build-oe-familiar/tmp/work/armv5te-linux/glibc-2.3.5+cvs20050627-r12/build-arm-linux/csu/divdi3.o |
| 23:22.49 | psokolovsky_ | s -MD -MP -MF /home/pfalcon/linux-ppc/build-oe-familiar/tmp/work/armv5te-linux/glibc-2.3.5+cvs20050627-r12/build-arm-linux/csu |
| 23:22.52 | psokolovsky_ | sorry for wrapping %) |
| 23:22.54 | psokolovsky_ | finally, I actually using binutils-cross-2.15.94.0.2.2 |
| 23:23.11 | psokolovsky_ | (just couldn't find binutils-cross-2.15.94.0.1 in local mirror ;-) |
| 23:24.14 | goxboxlive | ok, i am using 2.17 |
| 23:24.14 | psokolovsky_ | goxboxlive, sorry, but I don't have idea what can be wrong %) |
| 23:24.22 | psokolovsky_ | 2.17???? |
| 23:24.32 | goxboxlive | me neither, have to ask the gyus at #OE |
| 23:24.32 | psokolovsky_ | screw that, use recommended |
| 23:24.44 | goxboxlive | 2.15*** didnt build. |
| 23:25.07 | goxboxlive | maybe i should try to use 2..16 |
| 23:25.19 | psokolovsky_ | you cannot expect >1 year old glibc to be buildable with 2-month old binutils ;-) |
| 23:25.38 | psokolovsky_ | 2.15.94.0.2.2 worked from me. from nearest gentoo mirror. |
| 23:25.51 | psokolovsky_ | had to patch patches a bit, though |
| 23:26.23 | psokolovsky_ | 2.16 may be a better choice, if you really can't get 2.15.94.0.1 |
| 23:26.48 | goxboxlive | no, it failes. hrw at #oe told me to use 2.16 |
| 23:27.09 | goxboxlive | wich bitbake version are you using? |
| 23:27.18 | psokolovsky_ | 1.6.1 from SVN |
| 23:27.25 | goxboxlive | ok me to |
| 23:27.32 | goxboxlive | is it stable? |
| 23:27.37 | goxboxlive | like 1.4.2 |
| 23:27.53 | psokolovsky_ | yep, pretty nice. lacks my patch for SVN fetcher from trunk though ;-) |
| 23:28.02 | psokolovsky_ | 1.4.2 is *old* ;-) |
| 23:28.17 | goxboxlive | :-) i know, but it was stable as a rock :-) |
| 23:28.46 | psokolovsky_ | 1.6 seems to do all as expected for me ;-) |
| 23:29.58 | goxboxlive | :-) |
| 23:47.21 | goxboxlive | psokolovsky_ it seem like glibc is building now. Thx for the tip about binutils . I downgraded it to 2.16 |
| 23:48.34 | psokolovsky_ | goxboxlive, nice. just in case, pls keep in mind, that not all combinations work together ;-) |
| 23:48.50 | goxboxlive | i never learn :-) |