IRC log of #maemo for Sunday, 2015-01-25

*** Haudegen has quit IRC00:14
*** Haudegen has joined #maemo00:16
*** Pali has quit IRC00:24
*** heroux has quit IRC00:25
*** heroux has joined #maemo00:42
*** githogori_ has quit IRC00:44
*** vakkov has joined #maemo00:51
*** Majmun has joined #maemo00:54
*** Majmun has left #maemo00:56
*** hurrian has joined #maemo00:58
*** githogori has joined #maemo01:04
*** githogori_ has joined #maemo01:05
*** babytoes has quit IRC01:07
*** githogori has quit IRC01:08
* Maxdamantus wonders what he's doing wrong trying to boot a 2.6.28.10-{cssu1,power53} kernel from uboot, without attachboot.01:10
Maxdamantustrying something like: ext2load mmc 1:1 0x82000000 /k/zImage-2.6.28.10-cssu1.u; bootm 0x8200000001:11
Maxdamantuswhich works for a 3.14 mainline.01:11
*** Luke-Jr has quit IRC01:16
*** Luke-Jr has joined #maemo01:17
*** r00t|home has joined #maemo01:17
MaxdamantusAhah, finally.01:28
* Maxdamantus wonders if it was the lack of setup_omap_atag.01:28
*** hurrian has quit IRC01:29
*** Guest13078 is now known as mario01:35
*** mario is now known as Guest1888901:36
*** ssvb has quit IRC01:36
*** Guest18889 has quit IRC01:36
*** Guest18889 has joined #maemo01:36
*** Guest18889 is now known as mario01:36
*** hurrian has joined #maemo01:44
*** wotan147 has quit IRC01:45
*** arcean has quit IRC01:49
*** vakkov has quit IRC01:58
*** geaaru has quit IRC02:00
*** florian has quit IRC02:13
*** vakkov has joined #maemo02:26
*** r00t|home has quit IRC02:34
*** r00t|home has joined #maemo02:46
*** erstazi has quit IRC02:50
*** erstazi has joined #maemo02:51
*** erstazi has quit IRC02:51
*** erstazi has joined #maemo02:51
*** auenf has quit IRC02:51
*** auenf has joined #maemo02:52
*** RiD has quit IRC02:57
*** FReaper-PC has joined #maemo03:03
*** futpib has quit IRC03:04
*** xes has quit IRC03:30
*** hubutm20 has joined #maemo03:30
*** discopig has quit IRC03:47
*** githogori_ has quit IRC03:47
*** auenf has quit IRC03:52
*** disco_stu_droid has joined #maemo04:07
*** nox- has quit IRC04:08
*** arossdotme1 has joined #maemo04:09
*** mattaust1n has joined #maemo04:09
*** peetah_ has joined #maemo04:09
*** dimw1t has joined #maemo04:10
*** ecloud_ has joined #maemo04:12
*** drathir87 has joined #maemo04:12
*** obsed has quit IRC04:16
*** Venusaur has quit IRC04:16
*** ruskie has quit IRC04:16
*** Kilroo has quit IRC04:16
*** peetah has quit IRC04:16
*** mattaustin has quit IRC04:16
*** aloril has quit IRC04:16
*** at1as has quit IRC04:16
*** Elleo has quit IRC04:16
*** arossdotme has quit IRC04:16
*** disco_stu has quit IRC04:16
*** ecloud has quit IRC04:16
*** sledges has quit IRC04:16
*** drathir has quit IRC04:16
*** drathir87 is now known as drathir04:16
*** ruskie has joined #maemo04:16
*** disco_stu_droid is now known as disco_stu04:17
MaxdamantusHm. Is it typical for fcam to pick up apparent predictable sensor noise?04:17
*** Elleo has joined #maemo04:17
*** Elleo has quit IRC04:17
*** Elleo has joined #maemo04:17
MaxdamantusIf I take a few pictures in low light, there's a static pattern of apparently full-colour pixels (#ff0000, #00ff00, #0000ff)04:18
ecc3ga lot of camera sensors are like that when the ISO is set very high and "shutter" speed set very slow.04:19
*** Kilroo has joined #maemo04:19
*** aloril has joined #maemo04:19
*** Venusaur has joined #maemo04:20
*** sledges has joined #maemo04:21
MaxdamantusHm. With an actual static pattern?04:22
Maxdamantusie, the same pixels are like that across different pictures.04:22
*** M4rtinK has quit IRC04:22
*** auenf has joined #maemo04:22
ecc3gtry it.  usually yes, and camera software will subtract it out04:24
MaxdamantusYeah, was just thinking, might be something that the software remembers once and cancels out.04:24
* Maxdamantus gets the same thing with 0.5s exposure, ISE 36004:24
ecc3gyes the default camera app does that, but the raw cam programs do not.04:25
ecc3git wouldnt be raw if it did :D04:25
MaxdamantusDo you know what the phenomenon is called?04:26
MaxdamantusSo I can Google it.04:26
ecc3gI never tested my n900 for crappy pixels.04:26
ecc3g"hot pixels" I think is the common name for these.04:26
ecc3gMy Nikon D5100 DSLR does not have that many hot pixels but even it will start getting these with very long shutter and high ISO04:27
*** qt_junkie is now known as 17SABLJSY04:27
*** qt_junkie has joined #maemo04:27
MaxdamantusYeah, I suspect this has a lot.04:28
ecc3gI noticed this phenomenon on my old olympus cameras... low light high iso I get these crappy hot pixels that are distracting...04:29
ecc3gbut newer digital cameras subtract out these things...or at least they're a lot better...04:29
ecc3gI have two Canons which I don't notice (then again they don't raw), two olympus (both show hot pixels) and the Nikon DSLR (don't see too many)04:31
ecc3gThe N900 I'm sure I'll find hot pixels but don't care, if I cared that much I'm sure I'll be using the DSLR instead :D04:32
*** swartulv has quit IRC04:35
*** 17SABLJSY has quit IRC04:35
*** arossdotme1 has quit IRC04:35
*** r00t|home has quit IRC04:35
*** jrayhawk has quit IRC04:35
*** swartulv has joined #maemo04:36
Maxdamantushttp://i1.minus.com/ibdKXdpY89O3zI.jpg04:36
*** arossdotme has joined #maemo04:36
hurrianwhoa wtf happened to that sensor?04:37
ecc3gthat's normal with a raw and very low light/high iso04:38
hurrianhmm, might try that when I pull my N900 out of the closet one of these dats04:38
hurrians/dats/days/04:38
infobothurrian meant: hmm, might try that when I pull my N900 out of the closet one of these days04:38
ecc3gwell not "normal" but typical with cheap sensors04:38
*** povbot_ has joined #maemo04:43
-weber.freenode.net- [freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff: http://freenode.net/faq.shtml#gettinghelp04:43
*** luke-jr_ has joined #maemo04:43
*** HRH_H_Cr1b has joined #maemo04:44
*** kylanpaj_ has joined #maemo04:44
*** uhhimhere has quit IRC04:44
*** joga_ has joined #maemo04:45
*** lool- has joined #maemo04:45
*** clopez has quit IRC04:45
*** honestly_ has joined #maemo04:45
*** wnd_ has joined #maemo04:45
*** Raimu-X has joined #maemo04:45
ecc3gIt sort of is a dilemma now, one of my Olympus has a F1.8 lens, but has awful hot pixels.  My Nikon only stops down to F3.5 but its sensor is much cleaner...04:46
*** inz_ has joined #maemo04:47
*** clopez_ has joined #maemo04:47
*** Humpelst1lzchen has joined #maemo04:48
*** thedead1440 has joined #maemo04:48
*** CR0W- has joined #maemo04:48
*** Luke-Jr has quit IRC04:49
*** darkschneider has quit IRC04:49
*** skule has quit IRC04:49
*** CR0W has quit IRC04:49
*** HRH_H_Crab has quit IRC04:49
*** thedead1440_ has quit IRC04:49
*** Humpelstilzchen has quit IRC04:49
*** LjL has quit IRC04:49
*** inz has quit IRC04:49
*** lool has quit IRC04:49
*** ds3 has quit IRC04:49
*** honestly has quit IRC04:49
*** joga has quit IRC04:49
*** povbot has quit IRC04:49
*** teotwaki has quit IRC04:49
*** AndrewX192 has quit IRC04:49
*** Raimu has quit IRC04:49
*** wnd has quit IRC04:49
*** kylanpaj has quit IRC04:49
*** mickname has quit IRC04:49
*** LjL^ has joined #maemo04:49
*** inz_ is now known as inz04:49
*** LjL^ is now known as LjL04:49
ecc3gthough I would have to say that N900 photo is probably the worst I've ever seen :D04:50
*** teotwaki has joined #maemo04:52
*** AndrewX192 has joined #maemo04:55
*** wnd_ is now known as wnd04:55
*** skule has joined #maemo04:55
*** darkschneider has joined #maemo04:58
*** uhhimhere has joined #maemo04:59
*** mickname has joined #maemo05:02
*** uhhimhere has quit IRC05:02
*** obsed has joined #maemo05:05
*** Xjs|moonshine has quit IRC05:05
*** Xjs|moonshine has joined #maemo05:09
*** SpeedEvil has quit IRC05:13
*** SpeedEvil has joined #maemo05:15
*** luke-jr_ is now known as Luke-Jr05:31
*** nslu2-log_ has joined #maemo05:37
*** Natch_d has joined #maemo05:40
*** at1as has joined #maemo05:40
*** emma_ has joined #maemo05:41
*** valdyn_ has joined #maemo05:41
*** mhlavink_afk has joined #maemo05:43
*** mirda_ has joined #maemo05:44
*** githogori has joined #maemo05:45
*** protem has joined #maemo05:45
*** DrCode has quit IRC05:47
*** githogori_ has joined #maemo05:49
*** githogori has quit IRC05:49
*** darkschneider2 has joined #maemo05:50
*** mavhk has joined #maemo05:51
*** mavhc has quit IRC05:52
*** ruskie has quit IRC05:52
*** darkschneider has quit IRC05:52
*** dimw1t has quit IRC05:52
*** valdyn has quit IRC05:52
*** mirda has quit IRC05:52
*** N-Mi has quit IRC05:52
*** nslu2-log has quit IRC05:52
*** infobot has quit IRC05:52
*** emma has quit IRC05:52
*** valdyn_ is now known as valdyn05:52
*** mavhk is now known as mavhc05:52
*** mirda_ is now known as mirda05:52
*** Natch has quit IRC05:52
*** mhlavink has quit IRC05:52
*** nslu2-log_ is now known as nslu2-log05:52
*** N-Mi has joined #maemo05:53
*** emma_ is now known as emma05:55
*** ruskie has joined #maemo05:55
*** DrCode has joined #maemo06:00
*** lxp1 has joined #maemo06:01
*** CR0W- has quit IRC06:02
*** lxp has quit IRC06:04
*** CR0W has joined #maemo06:07
MaxdamantusswScaler: Compile-time maximum width is 2048 change VOF/VOFW and recompile06:07
Maxdamantus:(06:07
* Maxdamantus is seeing if he can encode from FCam to ffv1 using ffmpeg.06:08
MaxdamantusMm, seems to be doing about 1 fps, basically max CPU, I think 850 MHz06:13
*** Gh0sty has quit IRC06:18
ecc3gn900s were never known to be the most speedy cpus...06:20
*** Gh0sty has joined #maemo06:20
MaxdamantusI'm surprised it actually manages 1 fps.06:22
Maxdamantusby encode from FCam, I mean 2592*1968 frames.06:22
Maxdamantus(ffv1 is a lossless video format)06:22
MaxdamantusSo essentially compressed 826 MiB of raw UYVY images to 44 MiB.06:24
Maxdamantusthough the images are mostly black in this case :\06:25
ecc3gcan the emmc even write 44MB/sec?06:25
MaxdamantusIt's not 44 MiB/sec06:25
MaxdamantusThat's 44 MiB for 108 frames.06:25
Maxdamantuswhich should be about 108 seconds.06:26
ecc3gthat makes more sense, a lot slower...06:27
MaxdamantusWell, the raw data will be about 16 MiB per frame06:27
ecc3gnormally when talking about video it's all bitrates06:27
Maxdamantuswhich the eMMC probably won't handle.06:27
MaxdamantusI'm still writing that raw data, 16 MiB/s06:27
Maxdamantusit's just compressed.06:27
hurrian16MiB/s is likely doable for the N900's eMMC06:28
ecc3gyes06:28
Maxdamantusbut through the MMC controller it has?06:28
hurrianI think I topped it out at around 24MiB/s? The OneNAND is way faster at 50-80MiB/s (depending on condition) though06:28
ecc3g44MB/sec is too high06:28
MaxdamantusAh, I can read it at 20 MiB/s, at least.06:29
MaxdamantusI guess writing through a filesystem is just slower.06:30
ecc3gwhat's the fastest anyone has gotten through the microsd slot?06:31
hurrianI got 11MiB/s sequential write and 16MiB/s sequential read06:31
Maxdamantusand actually, it's not 16 MiB/s raw, only 7.6 MiB/s06:31
hurrianif you get a good enough microSD it should be comparable to the eMMC, they're both connected to the same controller iirc06:32
ecc3gnot sure if either/both use parallel mode or not06:32
hurrianon a side note, >mfw my Note 4's eMMC is faster than my desktop SSD06:33
MaxdamantusYeah, I think the bottleneck is usually the controller, not the MMC.06:33
* Maxdamantus wants to try making an SD card ring some time.06:33
Maxdamantuscan't seem to find any documentation of anyone actually trying it though.06:34
hurrianMaxdamantus: >SPI chaining06:34
MaxdamantusThe only evidence I can see of it existinrg is in the SD specification.06:34
*** vakkov has quit IRC06:34
hurrianthat's gonna be... fun.06:34
MaxdamantusWould be pretty cool to be able to buy four 256 GB SD cards and put them in a mugen case.06:35
Maxdamantus1 TB phone.06:35
hurrianis SPI breadboardable though? last time I tried a SD hack for my router, there was too much noise on the wires. had to spin a board for it.06:36
Maxdamantusregarding the NAND being way faster than the MMC, how sensible would it be to use that for swap?06:54
* Maxdamantus is basically just using that as a modifiable initramfs.06:56
*** b1101 has joined #maemo07:02
*** FReaper-PC has quit IRC07:09
*** FReaper-PC has joined #maemo07:09
*** Natch_d has left #maemo07:10
*** Natch has joined #maemo07:10
MaxdamantusHeh, noticed it said the device was using more CPU than the charger is providing when I ran my thing.07:20
Maxdamantuser, more power*07:21
MaxdamantusMaybe it'd be a better idea to just lzop the raw stream.07:22
*** discopig has joined #maemo07:22
*** auenf has quit IRC07:30
*** auenfx4 has joined #maemo07:30
*** githogori_ has quit IRC07:31
MaxdamantusHm .. futex(0x3aac0, FUTEX_WAIT, 0, NULL^C <unfinished ...>07:46
Maxdamantuswonder why that happens while it's meant to be locked.07:46
Maxdamantus(in fcam)07:46
*** Kabouik has quit IRC07:52
*** Kabouik has joined #maemo07:53
*** blotter has joined #maemo08:01
*** ashneo76 has quit IRC08:03
*** ashneo76 has joined #maemo08:05
MaxdamantusMaybe I need to disable sleep.08:07
Maxdamantusnot actual sleep, but some sort of powersaving mode.08:08
*** bef0rd has quit IRC08:11
*** b1101 has quit IRC08:32
kerioMaxdamantus: nand is small08:36
*** b1101 has joined #maemo08:46
Maxdamantuskerio: it's still almost as large as the RAM.08:58
kerioyou want it to be a fuckton bigger08:59
kerioso it doesn't get fragmented08:59
MaxdamantusAlso, turns out my battery calibration was wrong.08:59
MaxdamantusI think.08:59
Maxdamantusseemed to jump from ~40% to 5%, then die.08:59
*** blotter has quit IRC09:10
*** blotter has joined #maemo09:26
Maxdamantus8/1486 mAh09:32
*** githogori has joined #maemo09:39
*** githogori has quit IRC09:39
*** joga_ is now known as joga09:40
*** joga has quit IRC09:40
*** joga has joined #maemo09:40
*** protem has quit IRC09:52
*** ds3 has joined #maemo10:04
*** _rd has joined #maemo10:12
Maxdamantushttps://drive.google.com/file/d/0B2Ob19L8oYDMcHZnMEI4MWdXb3c/view10:49
MaxdamantusNeed tape .. and to remember to change focus.10:49
Maxdamantusand probably different exposure or something.10:49
Maxdamantus(that video is cropped and downscaled)10:50
*** Pali has joined #maemo10:52
*** blotter has quit IRC10:59
*** sparetire_ has quit IRC11:02
*** heroux has quit IRC11:15
*** heroux has joined #maemo11:17
*** _rd has quit IRC11:37
*** Kabouik_ has joined #maemo11:41
*** Kabouik has quit IRC11:43
*** ssvb has joined #maemo11:43
*** Vajb has quit IRC11:47
*** infobot has joined #maemo11:56
infobotDocScrutinizer: infobot joined!11:56
*** uhhimhere has joined #maemo12:07
*** futpib has joined #maemo12:16
*** Haudegen has quit IRC12:18
*** Haudegen has joined #maemo12:21
*** hubutm20 has quit IRC12:30
*** hubutm20 has joined #maemo12:30
*** arcean has joined #maemo12:31
*** florian has joined #maemo12:41
Pali~seen FatPhil12:53
infobotfatphil <~pcarmody@87-119-188-75.tll.elisa.ee> was last seen on IRC in channel #maemo, 59d 19h 33m 24s ago, saying: 'I know I'll get in trouble for asking, but what kernel are you running?'.12:53
Pali~seen FatPhil_12:54
infobotfatphil_ <~pcarmody@87-119-183-129.tll.elisa.ee> was last seen on IRC in channel #maemo-ssu, 493d 14h 5m 45s ago, saying: 'yup, those are the US americans'.12:54
*** ssvb has quit IRC13:07
*** hubutm20 has quit IRC13:10
freemangordonPali: we'd better write an email13:11
*** hubutm20 has joined #maemo13:11
Paliok, will you write it?13:12
freemangordonI will, but not now13:12
Paliok13:13
*** Vajb has joined #maemo13:24
honestly_anybody using yappari?13:42
honestly_seems it doesn't workanymore13:42
freemangordonhonestly_: check the new thread on TMO13:43
honestly_TMO?13:43
freemangordon~tmo13:43
infobotmethinks tmo is http://en.wikipedia.org/wiki/TMO, or http://talk.maemo.org, or http://de.wikipedia.org/wiki/Terrestrial_Trunked_Radio#TMO. It's *not* T-MO (see ~T-MO) or trolls, morons, oxen.13:43
honestly_ah.13:44
honestly_what thread?13:45
honestly_oh nevermind13:45
honestly_found it13:45
honestly_so tl;dr there is a new version but it's not in the normal repo yet?13:47
MaxdamantusHm. So does anyone know why fcam (the library) would lag/freeze when the screen turns off?13:47
Maxdamantuswhen not connected to USB.13:48
Maxdamantusobviously something to do with powersaving .. but what exactly?13:48
bencohhonestly_: it's probably in -devel13:48
bencohMaxdamantus: what do you mean by "lag" ?13:49
honestly_bencoh: well, I don't want to enable -devel and pull in lots of other stuff13:52
Maxdamantusit might take a few shots, then the process using the library hangs on a futex.13:52
bencohhonestly_: personally I do apt-pinning13:52
Maxdamantusand this in the kernel messages: [15704.067108] sbl overflow 0040000013:52
bencohMaxdamantus: taking shots when screen is off using fcam ?13:53
Maxdamantusbencoh: yes.13:53
bencohhow do you actually take shots ?13:53
bencoh(fcam API, or fcamera app ?)13:53
Maxdamantusfcam is a library and some kernel modules.13:53
MaxdamantusAPI.13:53
bencohoh, okay :)13:53
honestly_bencoh: yeah I might try that if I'm not sleep-deprived at some point in the near future13:54
freemangordonPali: how to attach gdb to qemu?14:00
Paliappend -S -s to qemu command line14:00
freemangordonok14:00
Paliand in gdb run: target remote localhost:123414:00
freemangordonok14:00
Paliyou need gdb with armel target compiled14:00
freemangordonBTW I am trying to build qemu with traces enabled, for some reason I see nothing :(14:01
*** wotan147 has joined #maemo14:05
*** xes has joined #maemo14:11
*** darkschneider2 is now known as darkschneider14:14
*** ssvb has joined #maemo14:19
freemangordonPali: ping14:25
Palifreemangordon: pong14:25
freemangordonPali: in /proc/iomem, there is "02000000-0201ffff : omap2-onenand"14:26
freemangordonwhat is omap2-onenand?14:26
Palimemory range: start=0x02000000 size=0x01000000 is 16MB gpmc for smc91c96 ethernet14:28
Paliavailable only in qemu and development boards14:29
Palirange start=0x01000000 size=0x01000000 is 16MB for onenand14:29
freemangordonhmm, this is what I see on the device itself14:29
freemangordonNokia-N900:/dev# cat /proc/iomem14:29
freemangordon00000000-00000000 : omapflash.014:29
freemangordon01000000-0100000f : smc91x14:29
freemangordon02000000-0201ffff : omap2-onenand14:29
Paliabove info is from n900 DT comments14:32
Palibut in u-boot is #define ONENAND_MAP             0x20000000      /* OneNand addr */14:32
Paliso maybe comments in n900 DT linux file are incorrect14:32
Palihm... in qemu 3.19 kernel I see: 04000000-0401ffff : omap2-onenand14:34
Palikernel can remap it??14:34
PaliI can normally mount ubifs in that kernel from rootfs nand14:35
Paliso nand is 100% working14:35
Palifreemangordon: if kernel say it is onenand it would be onenand...14:36
Palimaybe omap iommu can do some magic...14:37
freemangordonPali: well, ok. Any idea how to dump the bootrom?14:37
Paliyou can try via /dev/mem14:38
Paliopen it, mmap needed areay to userspace14:38
Paliand try to read it14:38
freemangordonsure, but how? "dd if=/dev/mem of=/home/user/MyDocs/flash bs=4096 count=1" gives me "badd address"14:38
freemangordonoh, mmap14:38
Palior try to use devmem214:39
Pali(or how is that program called)14:39
freemangordonhmm, iirc I have it somewhere14:39
Palihttps://github.com/XVilka/htools/blob/master/dump/dump_bootrom.c14:41
Palihttp://www.droid-developers.org/wiki/Application_Processor_Boot_ROM#Secure_part_of_Application_Processor_Boot_ROM14:46
Palihttp://www.droid-developers.org/wiki/Booting_chain14:47
freemangordonyeah, nice14:47
freemangordonanyway, I'll try to compile that module14:47
*** M4rtinK has joined #maemo14:49
Paliin that module are interesting data14:49
PaliROM CODE starts at 0x0001400014:49
freemangordonyeah14:49
Paliif (is_sram_locked) { ... if (omap_type() == OMAP2_DEVICE_TYPE_SEC) rom_boot_size = 0x700014:49
freemangordonNokia-N900:~# ./devmem2 0x0001400014:50
freemangordon/dev/mem opened.14:50
freemangordonMemory mapped at address 0x4001f000.14:50
freemangordonValue at address 0x14000 (0x4001f000): 0xEA00022E14:50
freemangordonseems I don;t need kernel module14:50
freemangordonok, module built, lets see :)14:52
*** uhhimhere has quit IRC14:53
*** vakkov has joined #maemo15:06
*** realitygaps has quit IRC15:15
*** realitygaps has joined #maemo15:18
*** realitygaps has quit IRC15:18
*** realitygaps has joined #maemo15:18
*** shentey has joined #maemo15:40
*** qt_junkie has quit IRC15:55
*** futpib has quit IRC16:08
freemangordonPali: bootrom dumped, seems like standart OMAP boot rom, version 14.5716:11
Palifreemangordon: ok16:11
Paliand what is at that address called by nolo??16:11
freemangordonnot part of it :D16:12
*** LauRoman|Alt has quit IRC16:12
Paliso it is secure part...16:12
freemangordonyeah, maybe16:12
Palican you access that address via devmem2?16:12
*** shentey has quit IRC16:13
*** auenfx4 has quit IRC16:15
freemangordonno, "bus error"16:15
*** auenf has joined #maemo16:15
Palifreemangordon: this is interesting: http://www.droid-developers.org/wiki/Mbmloader#How_to_know_in_advance_whether_a_given_mbmloader_can_work_on_the_Milestone16:22
PaliI bet mbmloader has same header as our xloader (header strucutre where is stored certificate and other data)16:23
freemangordonalready checked, the structure is different (we have "NOLO" and "RX-51" things)16:26
freemangordonoh, wait16:27
freemangordonthis is different16:27
*** at1as has quit IRC16:27
*** LauRoman has joined #maemo16:29
Palicheck also http://www.droid-developers.org/wiki/ISW16:30
freemangordonPali: maybe we should simply brute-force the private key and end this saga :D16:32
freemangordon1024 RSA shouldn't be impossible for distributed brute force16:32
Palican you extract public key?16:33
Palimaybe it could be weak :-)16:33
freemangordonthe public key is in xloader afaik16:33
Paliand it would be different for different hw revision...16:34
*** peetah_ is now known as peetah16:34
freemangordonI doubt16:35
Paliwe can try to check that sha1 checksum from ASIC id from different hw revisions...16:35
freemangordonas AFAIK this new xloader can be flashed on 2204 for example16:35
Paliwhy then there are different versions of xloader for different hw revisions?16:36
freemangordonNFC, honestly, maybe the HW have special needs16:36
freemangordonbut it makes no sense to have different keys16:36
Palisee: http://pastebin.com/WYPnsttB16:36
Pali2204 has same xloader as 210116:37
freemangordonyep16:37
Palibut 2301 different16:37
freemangordonbut...16:37
freemangordonint the very beginning of xloader is the HW revision16:37
Palineed to compare ASIC ID from more n900s...16:38
freemangordonPali: the easiest way - cat xloader | xxd -ps16:38
freemangordonthen diff/meld16:38
freemangordonHW rev is on offset 48416:39
Paliyes, there is hw rev16:41
freemangordonthat changes the hash16:41
freemangordonBTW it seems we really have the same ROMBL as droid16:42
Palidroid has also hw revision at offset 484?16:43
Palianyway hw revision is also in xloader-qemu.bin16:43
freemangordonnot xloader, but ROMBL16:43
freemangordonROM bootloader that is, the one that is at address 0x1400016:43
Paliok16:44
*** incrys has joined #maemo16:45
freemangordonPali: http://www.droid-developers.org/wiki/File:Omap_3430_bootrom.idb.gz16:45
freemangordoneven the revision at the end is the same16:45
*** incrys has left #maemo16:46
freemangordonnot that it helps much, as those unknown foncs are not in rombl16:46
freemangordon*funcs16:47
*** Smily has quit IRC16:49
*** Smily has joined #maemo16:51
*** sq-one has joined #maemo16:55
*** _jester_ has joined #maemo17:00
_jester_Hello World17:00
_jester_quick question, nokia n900, flashed, cssu thumb installed. clock/alarms are not working, I get an error libgq-gconf.so.0 is missing17:02
freemangordonwhere is that error?17:03
_jester_oh and apt-get install libgq-gconf0 returns, no installation candidate17:03
_jester_from the gui, it basically never starts, to debug i tried running 'worldclock' from the console17:03
freemangordonhmm, weird17:04
freemangordonii  libgq-gconf0                                        0.2-5+0m5                                  a Qt wrapper for GConf17:05
_jester_oh sorry, apt-get actually returns "libgq-gconf0 is already the newest version17:05
_jester_actually i think with the new updates, a few old apps are kinda getting broken17:05
freemangordonhow did you install cssu-thumb?17:05
_jester_flashed the n900, clicked on the cssu-thumb link on the maemo.org site17:06
_jester_installation went smoothly17:06
_jester_oh and enabled the extra testing repo17:06
freemangordoncould you try "ldd /usr/bin/worldclock | grep conf" from the terminal17:07
_jester_yup17:08
_jester_give me a minute17:08
freemangordonalso, try to start worldclock from terminal17:08
_jester_i did, thats how i got the error message about the libgq-gconf.so.0 missing17:09
freemangordondid you install anything else but cssu-thumb? like bash or whatnot?17:09
*** RiD has joined #maemo17:10
_jester_i installed a lot of apps17:10
freemangordon:)17:10
_jester_you think one of the apps messed up the library ?17:10
freemangordonfor some reason a library that is in /usr/lib/ can;t be found17:11
freemangordonI guess so17:11
_jester_cause an "apt-get install --reinstall libgq-gconf0" doesnt work17:12
*** shentey has joined #maemo17:12
_jester_it says its not possible and the package cannot be downloaded, so im assuming its part of the core, as it should be17:12
freemangordoncould you try "LD_LIBRARY_PATH=/usr/lib worldclock"?17:12
freemangordon_jester_: enable your okia repos17:12
freemangordon*nokia17:12
freemangordonin HAM17:13
_jester_ah, youre right, i have those disabled17:13
freemangordon*DO NOT* disable those17:13
_jester_yes sir !17:13
freemangordon:)17:13
*** shentey has quit IRC17:14
_jester_and after enabling them, i should try and reinstall the library ?17:14
freemangordonfirst try the LD_LIBRARY_PATH trick17:15
_jester_nopes, i still get the error for the missing libraries17:16
freemangordontry to apt-get install --reinstall17:17
freemangordonthough that shouldn;t be needed for a library in /usr/lib17:17
freemangordonunless your flash is broken17:18
*** shentey has joined #maemo17:18
_jester_nopes, just reinstalled it and everything works now17:18
freemangordonweird17:18
_jester_freemangordon, youre a superstar dude17:18
freemangordonoh, come on17:18
freemangordon:)17:18
freemangordon_jester_: anything in dmesg log?17:19
freemangordonanithing weird that is17:19
freemangordon*anything17:19
_jester_well you essentially saved me a reflash17:19
_jester_cause this was annoying the hell out of me17:19
freemangordonyeah, I can imagine17:20
_jester_nopes, just the usual slider open, wlan0: authenticated messages17:20
freemangordonhmm, weird17:21
freemangordonanyway, good that it works now17:21
_jester_yea man, thanks a lot ... this is much appreciated17:22
*** shentey has quit IRC17:43
freemangordonvakkov: actually I am not REing NOLO, but xloader. Or rather trying to find what xloader does to enable the AES HW module17:44
*** blotter has joined #maemo17:44
*** _jester_ has quit IRC17:48
*** shentey has joined #maemo17:50
vakkovfreemangordon: i see; if someone knows where in NOLO is the hardcoded value (size) of the ram, please do ping me.17:51
*** shentey has quit IRC17:58
freemangordonvakkov: why do you think the RAM size is hardcoded?18:00
*** zammy has joined #maemo18:01
zammythere is an issue with yappari18:02
zammyhttp://talk.maemo.org/showthread.php?s=f1b8b1b6f0b130fd878d5c7dbaa9232e&p=1458065#post145806518:02
zammyalso a irc channel #yappari18:02
vakkovas far as i understood Pali i t is hardocded in NOLO and nolo passes it to the kernel via an ATAG. the mainline kernel has it hardcoded though and ignores the atag. but older ones need the atag. freemangordon:18:03
Palifreemangordon: ram size is in DT18:04
Palialso in board rx51 file18:04
bencohzammy: see the other yappari thread18:04
bencoh(the new one)18:04
Paliand also in uboot rx51 include config18:04
zammylink me thanks18:04
Paliand also nolo pass it in atags18:04
bencohzammy: http://talk.maemo.org/showthread.php?t=9446518:04
vakkovisn't it possible to just make u-boot change the atag, Pali18:05
Paliyes, it is possible18:05
Paliuboot does not reuse memory atag from nolo18:05
Palibut it read it from uboot rx51 include file18:05
*** shentey has joined #maemo18:05
vakkovbut the old kernels do use the atag? so it needs to be changed somehow for them?18:06
*** shentey has quit IRC18:08
Palivakkov: yes18:11
freemangordonvakkov: you have n900 with 512 MB of RAM?18:11
*** shentey has joined #maemo18:12
Palivakkov: you can change uboot config18:12
Paliand you do not need to modify atags18:12
Paliuboot generate them18:12
vakkovfreemangordon: no, they are all stock for now. the plan is to upgrade to 1 gig using the n9 ram+nand chip18:12
vakkovthe nand in n9 and n900 is pin-compatible18:13
*** sleepee has joined #maemo18:13
zammybencoh what about repository from now to beyond18:25
zammyI use the wetransfer link to install yappari 2.0.2 and it' all fine but... repository and future updates?18:26
bencohzammy: I'm not ceene, I dont know :)18:26
zammyalright18:26
bencohhe probably didnt have time to push it to -devel18:26
bencoh"Cepiperez has uploaded changes to the code and a binary here, but latest .deb version is already available for download and is already on the autobuilder, so upgrading should be easily done via HAM in a few hours."18:27
freemangordonif I get it right, it is still not ready to be released in the repos18:27
zammyok18:27
* freemangordon does not use it, so doesn't really know18:28
* bencoh doesnt either :)18:28
bencohbut it's not in maemo.org/packages18:28
bencohfreemangordon: btw, any idea regarding my question on microb (chan next door) ?18:28
zammyfuck we are all whatsapp/yappari drugged18:28
freemangordonI meant I don't use yappari/whatsapp at all :)18:29
zammyok anyway thanks to all18:32
*** _rd has joined #maemo18:32
freemangordonbencoh: could you repeat your question? was it about kernels 3.x?18:33
bencoh23:35 < bencoh> bleh, now cc1plus segfaults ... http://pastebin.notk.org/pastebin.php?show=m4336cf71 :/18:33
bencoh23:36 < bencoh> freemangordon: which version/tree did you build ?18:33
bencoh:-)18:33
bencoh(I got past the js "unable to dump" error)18:33
freemangordonthe one on cssu gitorious18:33
bencohgit://gitorious.org/community-ssu/microb-engine.git ?18:34
freemangordonyep18:34
*** zammy has left #maemo18:34
bencohhmm ... then I dont understand18:34
bencohyou didnt get the cc1plus segfault ?18:34
freemangordonif you build in thumb target, you need to use the thumb-testing branch18:34
bencohI build for vanilla18:34
freemangordonthen master branch it is18:35
freemangordonno, never seen gcc segfaulting18:35
bencohactually for cssu-t, so I reverted the libpng/zlib patches18:35
freemangordonreverted?18:35
*** netkat has quit IRC18:35
bencohon a local branch :)18:35
freemangordonlibpng and zlib are in cssu-t18:35
bencoherm, in cssu-s, mybad18:35
freemangordonoh18:35
freemangordonanyway, gcc should not segfault18:36
freemangordonwhat linux is that?18:36
freemangordonVM?18:36
bencohyeah18:36
freemangordongive it plenty of RAM, like 4GB+swap18:36
bencohqemu-kvm on a i686/core2duo18:37
bencohI'll try with some swap18:37
bencohand ... yeah, more ram as well18:37
freemangordonyou'll wait a week, mind you(if it hits the swap)18:37
bencohhaha18:37
bencohtrue18:37
*** APTX has left #maemo18:37
*** netkat has joined #maemo18:38
freemangordonyou'd better use SDK vmware image18:38
freemangordonor setup SB in your host linux18:38
bencohhmmm, no I dont really want to try SB on a grsec host :)18:38
bencohI admit trying to build old mozilla stuff isn't sane behavior, but I'm not *that* crazy :)18:39
bencohlet's try and give it moar ram18:40
bencohqemu: at most 2047 MB RAM can be simulated18:40
bencohlol18:40
freemangordon:)18:44
freemangordonyou'd better get VMWare SDK image and install LPAE kernel18:45
*** at1as has joined #maemo18:46
bencohoh btw, did you have to export ARM_FORCE_PLATFORM=v7l ?18:47
*** shentey has quit IRC18:48
bencohused mem including buffers/caches never goes higher than 400M (for a total of 2047M) when building it, and it still fails :/18:49
bencohI doubt it's a ram issue18:50
freemangordonnever seen that export before18:50
freemangordonyou try to cross-compile?18:50
bencohdpkg-buildpackage from sb18:50
freemangordonsomething is broken in your SB install18:50
freemangordonjust grab SDK image18:51
bencoh~sb18:51
infobotextra, extra, read all about it, scratchbox is a cross-compiling system that uses binfmt_misc, rpc calls, and an nfs mount to make a cross-build appear to be 100% native, and is found at http://www.scratchbox.org/, hosted by maemo now. Also at http://maemo.merlin1991.at/files/SB18:51
freemangordonyep, that's it18:51
*** shentey has joined #maemo18:52
freemangordonmerlin's one18:52
bencohfreemangordon: at which point did you have to add a symlink for 3.2->2.6 ?18:53
freemangordonafter it complained for missing build script18:53
freemangordonbut this is not needed in SDK image18:53
bencohthe thing is ... it didnt complain (yet) here18:54
*** shentey has quit IRC18:54
freemangordonwhich kernel?18:54
bencoh3.218:54
freemangordondunno18:54
freemangordongive it some time :)18:54
bencohhaha18:54
freemangordonit will, most probably18:54
bencohyeah I guess you're right about that18:54
*** shentey has joined #maemo18:57
ceenehi19:04
ceenei just realized that the autobuilder for yappari had failed19:04
*** sleepee has quit IRC19:04
ceenei'm just uploading it again19:04
ceenesorry :/19:04
*** LauRoman|Alt has joined #maemo19:06
*** sledges_ has joined #maemo19:08
bencohfreemangordon: invoking cc1plus manually works ....19:09
*** sq-one_ has joined #maemo19:09
*** kylanpaj has joined #maemo19:09
freemangordonbencoh: shouldn't be like that19:10
*** LjL^ has joined #maemo19:10
bencohyeah I guess so ...19:11
* freemangordon gtg19:12
freemangordonbbl, night guys19:12
bencohcya, and thx :)19:12
*** LjL has quit IRC19:13
*** LjL^ is now known as LjL19:13
*** sq-one has quit IRC19:15
*** LauRoman has quit IRC19:15
*** heroux has quit IRC19:15
*** N-Mi has quit IRC19:15
*** obsed has quit IRC19:15
*** kylanpaj_ has quit IRC19:15
*** sledges has quit IRC19:15
*** heroux has joined #maemo19:16
*** N-Mi has joined #maemo19:22
*** blotter has quit IRC19:29
*** at1as has quit IRC19:33
*** _rd has quit IRC19:50
*** _rd has joined #maemo20:04
*** spoofy has joined #maemo20:09
*** fmunozs has joined #maemo20:23
*** sparetire_ has joined #maemo20:30
*** M4rtinK has quit IRC20:41
*** at1as has joined #maemo20:56
*** wotan147 has quit IRC21:02
*** protem has joined #maemo21:03
*** futpib has joined #maemo21:03
*** wotan147 has joined #maemo21:05
*** maybeWTF has joined #maemo21:07
*** maybeHere has quit IRC21:10
*** sq-one_ has quit IRC21:14
*** Haudegen has quit IRC21:22
*** shentey has quit IRC21:22
*** Haudegen has joined #maemo21:25
*** shentey has joined #maemo21:26
*** RedM has joined #maemo21:27
*** RedW has quit IRC21:28
*** M4rtinK has joined #maemo21:30
*** RedW has joined #maemo21:30
*** RedM has quit IRC21:31
*** nox- has joined #maemo21:32
*** blotter has joined #maemo21:34
*** M4rtinK has quit IRC21:53
*** _rd has quit IRC21:54
*** fmunozs has quit IRC22:04
*** M4rtinK has joined #maemo22:06
*** fmunozs has joined #maemo22:19
*** obsed has joined #maemo22:20
*** shentey_ has joined #maemo22:22
*** silviof has joined #maemo22:24
*** jon_y has quit IRC22:25
*** M4rtinK has quit IRC22:27
*** shentey_ has quit IRC22:31
*** shentey_ has joined #maemo22:31
*** vakkov has quit IRC22:32
*** shentey_ has quit IRC22:37
*** shentey_ has joined #maemo22:38
*** arcean has quit IRC22:43
*** XDS2010_ has quit IRC22:45
*** arcean has joined #maemo22:45
*** silviof has quit IRC22:52
*** blotter has quit IRC22:54
*** hubutm20 has quit IRC23:00
*** hubutm20 has joined #maemo23:01
*** vakkov has joined #maemo23:05
*** silviof has joined #maemo23:14
*** shentey has quit IRC23:17
*** shentey has joined #maemo23:19
*** at1as has quit IRC23:26
*** Vajb has quit IRC23:29
*** Vajb has joined #maemo23:30
*** M4rtinK has joined #maemo23:33
*** shentey has quit IRC23:44
*** jon_y has joined #maemo23:58

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!