IRC log of #maemo-ssu for Tuesday, 2012-07-17

*** lizardo has quit IRC00:00
DocScrutinizer5116 days uptime and my twekr and profile(master?) profiles are defect :-/00:47
DocScrutinizer51actually I guess my systray menu is fecked00:48
DocScrutinizer51meh, reboot00:48
*** freemangordon has quit IRC00:55
*** Guest26773 has quit IRC00:59
*** _rd has quit IRC00:59
lufDocScrutinizer51: I can't be 16 days up without sleeping :D01:00
*** andre__ has quit IRC01:22
*** taziff1 has joined #maemo-ssu02:34
*** taziff has quit IRC02:34
*** Sicelo has quit IRC02:45
*** Sicelo has joined #maemo-ssu03:03
*** Sicelo has joined #maemo-ssu03:03
*** Estel_ has joined #maemo-ssu03:04
*** Estel_ has quit IRC03:04
*** Estel_ has joined #maemo-ssu03:04
*** povbot has joined #maemo-ssu03:15
*** ChanServ sets mode: +v povbot03:15
*** gri has quit IRC03:19
*** Estel_ is now known as ashdgf03:25
*** ashdgf is now known as Estel_03:25
*** gri has joined #maemo-ssu03:27
*** M4rtinK has quit IRC03:39
*** arcean_ has quit IRC04:15
*** Estel_ has quit IRC04:34
*** LaoLang_cool has joined #maemo-ssu04:35
*** luf has quit IRC04:35
*** LaoLang_cool has quit IRC04:37
*** LaoLang_cool has joined #maemo-ssu04:38
*** LaoLang_cool has quit IRC04:51
*** LaoLang_cool has joined #maemo-ssu04:59
*** LaoLang_cool has quit IRC04:59
*** LaoLang_cool has joined #maemo-ssu05:00
*** LaoLang_cool has quit IRC05:01
*** LaoLang_cool has joined #maemo-ssu05:03
*** amiconn has quit IRC05:08
*** amiconn_ has joined #maemo-ssu05:08
*** amiconn_ is now known as amiconn05:08
*** gri has quit IRC05:15
*** gri has joined #maemo-ssu05:17
*** LaoLang_cool has quit IRC06:27
*** mase76 has quit IRC06:35
*** mirandir has joined #maemo-ssu08:23
*** taziff1 has quit IRC08:38
*** taziff has joined #maemo-ssu08:40
*** amiconn has quit IRC09:05
*** amiconn has joined #maemo-ssu09:05
*** Pali has joined #maemo-ssu09:11
*** LaoLang_cool has joined #maemo-ssu10:17
*** Pali has quit IRC10:21
*** Pali has joined #maemo-ssu11:26
*** LaoLang_cool has quit IRC11:39
*** LaoLang_cool has joined #maemo-ssu11:42
*** LaoLang_cool has quit IRC12:05
*** arcean has joined #maemo-ssu12:09
*** freemangordon has joined #maemo-ssu12:11
*** DocScrutinizer06 is now known as DocScrutinizer0512:14
*** freemangordon has quit IRC12:29
*** M4rtinK has joined #maemo-ssu12:31
*** Pali has quit IRC12:59
*** amiconn has quit IRC13:24
*** amiconn has joined #maemo-ssu13:25
*** Pali has joined #maemo-ssu13:47
*** lizardo has joined #maemo-ssu13:52
*** Pali has quit IRC14:20
*** Pali has joined #maemo-ssu14:22
*** LaoLang_cool has joined #maemo-ssu16:14
*** LaoLang_cool has quit IRC16:40
*** andre__ has joined #maemo-ssu16:41
*** andre__ has joined #maemo-ssu16:41
*** DocScrutinizer05 has quit IRC16:52
*** DocScrutinizer06 has joined #maemo-ssu16:52
*** DocScrutinizer has quit IRC16:52
*** DocScrutinizer has joined #maemo-ssu16:52
*** mase76 has joined #maemo-ssu17:17
*** freemangordon has joined #maemo-ssu18:23
merlin1991freemangordon, Pali, arcean, chem|st, MohammadAG, DocScrutinizer, andre__ ping18:55
freemangordonmerlin1991: pong18:56
freemangordonmerlin1991: btw any progress with luf's libcurl?18:57
andre__merlin1991, pong?18:57
merlin1991andre__: we'll have a meeting today if possible :)18:57
andre__uh, when?18:57
merlin1991thoguht you might be interested aswell :)18:57
merlin1991anywhere between now and midnight when the key players are online18:57
andre__I'll probably lurk, quite busy, but still want to read the backlog, sure :)18:57
Palimerlin1991, pong18:58
freemangordon18:56 <freemangordon> merlin1991: btw any progress with luf's libcurl?19:00
merlin1991ah sry read it, head the answer mentally typed, it just didn't went to the keyboard :P19:00
merlin1991yes, I've applied the patches (with a bit of trickery) and will push them to gitoirous after I've done a test or two19:01
merlin1991bah typo day19:01
PaliI do not know when I will go away, so some my info:19:03
Palipatched osso-xterm working fine on my n900 - it can go to cssu-devel19:03
Palialso patched busybox can go to cssu-devel19:04
merlin1991Pali you can push anything you want to cssu-devel ;)19:04
merlin1991what's patched in that busybox?19:04
Paliportrait mode support19:04
Palibefore busybox show a lot of free lines19:04
Paliwhen you try to rotate19:04
Paliwith this patch forced rotation working19:05
merlin1991ah so you still have to force xterm to rotate?19:05
Paliand is blacklisted19:05
Palibecause of that busybox error19:05
Palibash working fine19:05
kerioPali: does it work correctly with busybox-power?19:06
kerioalso, what's this cssu-devel?19:06
kerioi like the sound of it19:06
merlin1991kerio: obviously not :D19:06
merlin1991the same set of patches has to go into bs-power :)19:07
kerioby the way, why would anyone want to use osso-xterm in portrait?19:07
kerioyou have no keyboard there!19:07
merlin1991kerio: people are strange ;)19:07
Palikerio, I do not have busybox-power, so I do not know19:08
Palikerio, but you can enable forced rotation in /usr/share/hildon-desktop/... and remove osso-xterm from blacklist19:09
kerioPali: is there a way to forcibly *disable* portrait mode everywhere?19:09
Palikerio, and then you can test if busybox power working with rotation19:09
merlin1991Pali: I fear it's going to break installations of busybox power, but I can look into the packaging and write a fix to the instal scripts in case19:09
merlin1991kerio: orientation lock? :D19:09
MohammadAGmerlin1991, pong19:12
keriomerlin1991: another icon in the status area, ew :c19:13
Palikerio, button in status area or gconf-tool via xterm19:13
merlin1991hm we just need chem|st, then we can start19:14
Palimerlin1991, can you create busybox and upstart repository on gitorious cssu?19:24
merlin1991dang where's chem|st19:25
merlin1991kerio: yes, mamemo uses upstart as init19:26
kerio...what's wrong with SysV init? :c19:26
keriohe's here!19:26
kerioeverybody is here!19:26
merlin1991MohammadAG, freemangordon still here?19:26
ShadowXkerio, upsart is mainly even orientated .... and to be honest sysV init sux :)19:28
keriono u sux19:28
kerioand what kind of event does boot time have?19:28
kerio"system booted, start *everything*"19:28
Palimerlin1991, code pushed19:29
merlin1991come on people don't pop in and out :D19:30
ShadowXkerio, no idea why maemo is using upstart init ..but you can think of upstart as sysV++ .... kind of19:30
MohammadAGmerlin1991, yes19:30
merlin1991freemangordon: you're last ;)19:31
freemangordonmerlin1991: I will be here in about 20-30 minutes19:32
keriobut... you're here right now!19:33
kerioyou just typed a message!19:33
kerioi saw it!19:34
DocScrutinizer51merlin1991: what's the meeting today now? gotta scheule my execution ;-)19:41
merlin1991DocScrutinizer51: somehwere around 19:00 (cet)19:42
DocScrutinizer51too early19:42
DocScrutinizer51dafaq, I'll simply leave office right now19:43
DocScrutinizer51a bit pissed anyway19:43
chem|stsounds like me 45mins ago19:44
MohammadAGmerlin1991, gnome-shell doesn't pop notifications19:47
MohammadAGso I only get highlights on my iPhone19:47
keriosee? Apple is great!19:48
* kerio flees19:48
merlin1991kerio: you traitor :D19:49
keriome? he (MohammadAG)'s the one with an iPhone19:50
freemangordonmerlin1991: Now I am here19:53
merlin1991whatabout the rest? chem|st, Pali, MohammadAG, DocScrutinizer51 ?19:54
DocScrutinizer51ok, I'm semi-available for next +-40min19:54
DocScrutinizer51then fully available19:54
merlin1991freemangordon: I belive you have something to say?19:55
freemangordonmerlin1991: sure :)19:56
freemangordonMohammadAG: what has happened with those thumb-compiled binaries in latest CSSU update?19:56
MohammadAGthey stayed there, every time I could update gnome-vfs you say there are other thumb binaries19:57
freemangordonAnd a bit more general question: MohammadAG, are you still a functioning maintainer of CSSU-T?19:57
MohammadAGI'm not sure19:57
freemangordonno, the last time we discussed that you said you're going to fix all of them19:58
MohammadAGI said I was going to upload gnome-vfs, then you said there were other binaries19:58
freemangordonnow, we have a problem here19:58
MohammadAGso that didn't happen19:58
DocScrutinizer51general notion seems: cssu-t maintenance needs some more manpower19:59
DocScrutinizer51y whatever mean19:59
freemangordonwell, before solving the thumb issue, I think we should solve the general problem20:00
freemangordonDocScrutinizer51: :nod:20:00
MohammadAGso, two maintainers or one is the question?20:00
DocScrutinizer51MohammadAG: I'd suggest you ask for, or accept offered support/co-maintainer20:01
MohammadAGI never rejected any offers20:01
MohammadAGat least I don't recall I have20:01
DocScrutinizer51i don't see a reason to deprive MohammadAG rom maintainer status20:01
merlin1991MohammadAG: I think 2 or 1 depends on how much of your time you still want to invest into the maemo cssu20:02
merlin1991also I agree with doc20:02
freemangordonnoone asks for mag's removal, AIUI20:02
DocScrutinizer51but I'd urgently suggest merlin1991 takes comaintainership of T20:02
freemangordonDocScrutinizer51: you'd better ask merlin if he has the time needed to do that20:02
DocScrutinizer51if you'd accept that, merlin199120:03
DocScrutinizer51well, if there's too much workload from S for merlin1991 then I think I heard chem|st would be willing to take comaintainership for S20:04
freemangordonchem|st: ^^^ ?20:04
DocScrutinizer51so we'd have two maintainers for each flavour20:04
DocScrutinizer51rather 1.5 mainainers per distro20:05
*** taziff has quit IRC20:05
infobot1 is a number, silly20:05
freemangordonthanks bot20:05
*** taziff has joined #maemo-ssu20:05
infobot1 is a number, silly20:06
DocScrutinizer51thanks infobot20:06
infobotde nada, DocScrutinizer5120:06
freemangordonmerlin1991, chem|st: would you please give a more deffinitive answer, so we can move on20:07
DocScrutinizer51yep, me need an 'official' I accept20:08
merlin1991I agree20:08
chem|stI accept20:08
DocScrutinizer51we nee an official "i accept"20:08
DocScrutinizer51MohammadAG: fine with you?20:08
MohammadAGYes, fine by me20:09
MohammadAGso is that 2 more, or 1?20:09
freemangordon2 more AIUI20:09
merlin1991it's you and me for T and chemist and me for S20:09
DocScrutinizer51so you 3 guys will know whom too ping to make that clear20:09
*** taziff has quit IRC20:10
freemangordonoops, 1 more , infobot was right :D20:10
DocScrutinizer51we got one more mapower, two more roles20:10
DocScrutinizer51new role? merlin1991 maintains T20:11
freemangordon1 more person20:11
DocScrutinizer51new role? chem|st maintains S20:11
DocScrutinizer51remaining roles: moh for T and merlin1991 for S20:13
freemangordonso, we only need an "official" "I swear I am ok" from chem|st and merlin1991 and we're ready to go20:13
DocScrutinizer51we already had that20:13
freemangordonsorry, i missed it20:14
DocScrutinizer51so let's move on20:14
* chem|st puts his left hand on his n900 and his right on his n9 and starts calling souls from hell20:14
freemangordonyeah, that is what I was talking about20:15
freemangordonanyway, we have at least one thumb-compiled binary in latest T20:15
chem|stI missed 2 auctions and lost on 3... I'd like to get another n900...20:15
DocScrutinizer51fine, can we move on?20:16
freemangordonmy suggestion is to have a new update with all the stuff from latest re-compiled.20:16
DocScrutinizer51chem|st: noted20:16
DocScrutinizer51I'll pin you in case...20:17
DocScrutinizer51and getrid of damn operator widget dep hell20:17
chem|stDocScrutinizer51: business as usual20:18
DocScrutinizer51afk for 10 min20:18
DocScrutinizer51I'll read backscroll @home20:19
freemangordonarcean, Pali, merlin1991: is there something ready for T? since the last update?20:19
merlin1991there's a bunch of things from Pali to be merged20:19
freemangordonbut those are for devel afaik20:19
freemangordonhmm, what is fixed?20:20
freemangordonin ke-recv20:20
arceanhmm, Pali^20:20
Paliit is for devel too20:20
Paliautomounting ntfs20:20
Paliand auto usb network mode in pc suite mode (when user create file /etc/something...)20:21
freemangordonmerlin1991: is it a good idea to hold the update so libcurl to be included too?20:21
kerioautomounting ntfs? neat20:22
keriowith ntfs-3g?20:22
Palino, with kernel driver20:22
Palikerio, also fix for other parittions type20:22
Palithere is bug that sometimes hal does not send fs type20:22
freemangordonmerlin1991: after all we can't do much of a testing besides it works/not20:23
Paliso when ke-recv send empty fstype, mount script will ask blkid for fstype20:23
PaliI know what should go to testing20:23
freemangordonand it is?20:23
Paliupdated tvout package which move package from section user/*20:23
merlin1991Pali: did you write the call to blkid in a way that it does not use the cache? (there's a parameter for that)20:23
Palimerlin1991, I think no20:24
merlin1991the blkid cache contains all sorts of evil, so better change that to use that option20:24
Paliplease look at tvout status menu plugin and push new version (which only move package from section user/desktop to nonuser)20:24
Palimerlin1991, ok20:25
DocScrutinizer06k, back20:25
Paliblkid:  -c <file>   cache file (default: /etc/, /dev/null = none)20:25
DocScrutinizer06gave us days of laughter20:27
keriowhat's the blkid cache?20:27
DocScrutinizer06the idea of a crackhead20:28
DocScrutinizer06kerio: blkid caches the results of last system scan / last invocation in some private file and reuses them ininitely20:29
merlin1991kerio: blkid gives you detail information about your partitions, and it has a cache which stores all kinds of evil :D20:29
kerioi see20:29
freemangordonhmm, and what with cbs-widget?20:29
merlin1991creates nice phantom / invalid partitions when you don't call it with the proper commandline20:29
DocScrutinizer06freemangordon: kill it for now?20:30
keriocbs-widget needs to go, but this is not my 2c, this is my 0.0000000002c20:30
DocScrutinizer06or move it to optional?20:30
DocScrutinizer06or rename so it replaces original instead of conflicting with it?20:30
freemangordonif we kill it now, it will never make it again (if you ask me), so decide wisely20:30
merlin1991it actually does not conflict with the original20:31
merlin1991it only conflicts with the other "custom" one20:31
DocScrutinizer06I'd opt for moving it to optional, but aiui merlin1991hasn't finished that yet ;-)20:31
merlin1991basically what we have todo is to move the patched lib into its own package and then have a user decision on the actual widget package20:31
DocScrutinizer06i aint not a packager20:32
DocScrutinizer06so nfc20:32
merlin1991so we have libcurl20:32
merlin1991recompiled osso-gnome-vfs20:32
DocScrutinizer06merlin1991: could you take a note on your last post pelase20:32
* DocScrutinizer06 throws off shoes and jacket20:33
freemangordonmerlin1991: not only  osso-gnome-vfs20:34
freemangordonmerlin1991: one need to check all of the binaries from the latest update for being thumb-compiled20:35
merlin1991yes, but we know osso-gnome-vfs for sure, so that has to be done asap20:36
freemangordondo we have a list of updated packages?20:36
DocScrutinizer06is there a reliable tool meanwhile to spot thumb binaries20:36
freemangordonhmm, lemme try to find the bug20:36
merlin1991freemangordon: is accurate20:38
DocScrutinizer06hell, there must be some sort of fingerprint to spot thumb binaries from arm binaries, no?20:38
*** arcean_ has joined #maemo-ssu20:38
*** DocScrutinizer06 is now known as DocScrutinizer0520:40
*** arcean has quit IRC20:41
DocScrutinizer05so would you mind me shooting myself now?20:41
freemangordonthere was some method described on bugs.maemo.org20:41
DocScrutinizer05great, i'd just use a .45 on the forehead20:42
merlin1991libcurl has builddepends of openssh-server20:43
merlin1991hm seems I'll have to bounce it back to luf again20:46
povbotBug 10333: Thumb feature is not enable in the PR1.2 ROM20:49
DocScrutinizer05wtf? bug: NOT enable...?20:50
freemangordonYou can find components using thumb functions with something like this: readelf -s $file | grep -v "00000001" | grep FUNC |grep ":[0-9a-f]*[13579bdf] "20:50
freemangordonaccording to comment 420:50
DocScrutinizer05I see20:50
DocScrutinizer05makes sense20:51
freemangordonDocScrutinizer05: what do you think, it is above my scripting abilities20:51
freemangordonwell, I am lazy :)20:51
DocScrutinizer05it searches in readelf output for functions starting at impair addr20:51
freemangordonyeah, makes sense20:52
DocScrutinizer05though I wonder if thumb functions actiually START at impair addr or just the jumps to those thumb functions are to impair addr, while the actual target is one off20:52
freemangordonyeah, the functions start at pair address20:53
DocScrutinizer05umm, prolly not20:53
DocScrutinizer05probably thumbified functions actually start their code at impair addr20:53
DocScrutinizer05since any jump to a pair addr in thumb code would switch back to ARM, no?20:54
DocScrutinizer05and actually packing could make impair function start addr for thumb, that's the whole rationale of BX aiui20:54
freemangordonno, LSB is the one that is in charge20:55
freemangordonbut it is ignored20:55
freemangordonwhen the address is calculated20:55
DocScrutinizer05so what now? is it ignored or not?20:55
freemangordonit is20:55
DocScrutinizer05then no function label will ever start at an impair addr, and obove search is BS20:56
DocScrutinizer05wait, maybe that's the function jump / call, not the function label?20:56
DocScrutinizer05:0089ef81   is a label, no?20:58
DocScrutinizer05== grep ":[0-9a-f]*[13579bdf] "20:58
DocScrutinizer05freemangordon: could you do me a favour and readelf -s yourACMEthumb_binary|pastebin20:59
freemangordonbut i don't see readelf on n900, lemme check in VM20:59
DocScrutinizer05get it with binutils21:00
merlin1991yep need binutils for it21:00
freemangordonok, will do it in SB21:00
DocScrutinizer05or you run the cmd on your favourite PC21:00
DocScrutinizer05no need for SB21:01
freemangordonno other favourite PC with linux :)21:01
DocScrutinizer05readelf should work platform agnostic I'd think21:01
DocScrutinizer05IroN900:~# readelf -s /bin/devlocktool |pastebinit21:05
DocScrutinizer05THIS API HAS BEEN DISABLED. Please use Pastebin's new API.
kerioDocScrutinizer05: hahahahaha21:06
DocScrutinizer05FSCK U!21:06
DocScrutinizer05anyway, you even may paste it to a /query to me, freemangordon21:07
merlin1991DocScrutinizer05: as I told you swap the url for the bloody pastebin ;)21:07
freemangordonDocScrutinizer05: lemme try it here first, but what I see sofar is that addresses start at /4 offset :(21:08
DocScrutinizer05     5: 0000894c   924 FUNC    GLOBAL DEFAULT  UND abort@GLIBC_2.4 (2)21:08
DocScrutinizer05     6: 00011910     4 OBJECT  GLOBAL DEFAULT   23 stderr@GLIBC_2.4 (2)21:08
DocScrutinizer05     7: 00008958   552 FUNC    GLOBAL DEFAULT  UND change_devlock_code21:08
DocScrutinizer05seems like *my* /bin/devlocktool has all pair addr in there21:08
freemangordonhmm, "715: 0000b8b9   764 FUNC    GLOBAL DEFAULT   10 main"21:10
*** NIN101 has joined #maemo-ssu21:11
freemangordonbut the last grep does not work for some reason21:11
freemangordonseems grep in SB is broken somehow21:12
merlin1991works on device21:13
freemangordonmerlin1991: grep?21:13
merlin1991the whole grep thingy from the bug report21:14
merlin1991readelf -s $file | grep -v "00000001" | grep FUNC |grep ":[0-9a-f]*[13579bdf] "21:14
DocScrutinizer05715: .* FUNC21:14
DocScrutinizer05dafaq, there's a space between ":" and first digit of addr21:15
DocScrutinizer05look at >>  5: 0000894c   924 FUNC ...21:16
freemangordon readelf -s hildon-home.launch | grep -v "00000001" | grep FUNC |grep ":[ 0-9a-f]*[13579bdf]"21:17
freemangordondoes the job21:17
merlin1991that ones is a bad regex21:17
freemangordonwell, it should be one liner21:18
freemangordonanyway, function offsets are +1, i.e. odd21:19
freemangordonmerlin1991: ^^^21:20
freemangordonwe need space in regex21:20
freemangordoni.e. [ 0-9a-f]21:20
merlin1991i think the regex is cleaner if the space goes between : and the selection for numbers21:21
freemangordonreadelf -s hildon-home.launch | grep -v "00000001" | grep FUNC |grep ": [0-9a-f]*[13579bdf]"21:22
merlin1991thus: readelf -s $FILE | grep -v "00000001" | grep FUNC |grep ": [0-9a-f]*[13579bdf] "21:22
* merlin1991 boots nonthumb n900 to check files21:22
DocScrutinizer05readelf -s /bin/devlocktool |grep '^ *[0-9]*: [0-9abcdef]*[13579bdf] *[0-9]* FUNC'21:22
merlin1991hm best way to find all binaries on the fs?21:23
merlin1991also, does anybody have an idea which one of my n900s has -T on it? :D21:23
freemangordonfind -t=x?21:23
merlin1991when they are all powered down it's hard to know xD21:23
DocScrutinizer05find $(echo $PATH|sed "s/:/ /")21:24
DocScrutinizer05or sth like that21:25
merlin1991hm can only find a type for file there's not one for binary21:26
DocScrutinizer05find $(echo $PATH|sed "s/:/ /g")21:26
freemangordonyeah :(21:26
DocScrutinizer05IroN900:~# find $(echo $PATH|sed "s/:/ /g")|wc -l21:27
DocScrutinizer05find: "/usr/sbin/gnu": No such file or directory21:28
merlin1991hah found it a proper way21:28
freemangordonunfortunately most if the .so files does not have x bit set21:28
freemangordonmerlin1991: ?21:28
DocScrutinizer05yep, .so was what occured to me as well this very moment21:28
merlin1991readelf -h21:29
merlin1991should be considerable cheap and does return 1 if not a binary21:29
merlin1991DocScrutinizer05: that should only print the elf header of the file (if htere is one)21:29
DocScrutinizer05merlin1991: you don't want to check ALL files of your device?21:29
merlin1991well i'M bored?21:29
freemangordonhe wants ;)21:29
freemangordonmerlin1991: you could attach USB cable, so mydocs gets unmounted21:30
merlin1991or just rund umount ;)21:30
infobotmerlin1991 meant: or just run umount ;)21:30
freemangordonand you will have enough juice :P21:30
DocScrutinizer05merlin1991: you should check if your method works for .so21:30
merlin1991DocScrutinizer05: it does21:31
DocScrutinizer05then don't forget to |tee21:31
freemangordonmerlin1991: well, you could unmount and attach charger, but plugging usb cable is easier21:31
merlin1991hm if I have echo "something | tool#1 && tool#2  can I somehow have the piped stuff in tool#2 aswell?21:31
infobotmerlin1991 meant: hm if I have echo something | tool#1 && tool#2  can I somehow have the piped stuff in tool#2 aswell?21:32
DocScrutinizer05err wut?21:32
DocScrutinizer05tee --help? maybe?21:32
merlin1991I need the input left from the pipe in both tools right from the pipe21:33
DocScrutinizer05named pipe21:33
merlin1991I'll need an example :D21:34
DocScrutinizer05mkfifo /tmp/foobar21:34
DocScrutinizer05echo xyz | tee /tmp/foobar|tool1 &;  cat /temp/foobar|tool221:35
DocScrutinizer05or the other way round21:36
DocScrutinizer05mkfifo foobar21:37
DocScrutinizer05cat foobar|tool2&21:37
DocScrutinizer05echo xyz | tee foobar|tool121:37
DocScrutinizer05I forgot if you start with reading or with writing on a fifo21:38
DocScrutinizer05probably with writing as otherwise reading process wouldn't know which EOF is the right one21:39
merlin1991hm coming to think of it, it's actually not what I want :D21:40
DocScrutinizer05A process can open a FIFO in non-blocking mode.  In this case, opening for read only will succeed even if no-one has opened on the write side yet, opening       for write only will fail with ENXIO (no such device or address) unless the other end has already been opened.21:40
merlin1991I need to input the paths from find into readelf -h $FILE && readelf -s $FILE ....21:40
merlin1991I think I'll better user xargs + a scripts21:41
DocScrutinizer05or a temprary file with all your filenames fromn find21:41
merlin1991am I remembering right, that $0 is name of the running script?21:42
DocScrutinizer05unless it's "bash"21:42
merlin1991what happens then?21:42
DocScrutinizer05bash -c myscript will have bash in $0 iirc21:43
DocScrutinizer05generally $0 is the first lexem on cmdline21:43
freemangordonreadelf -A $FILE21:44
freemangordonTag_THUMB_ISA_use: Thumb-221:44
freemangordonDocScrutinizer05: , merlin1991 ^^^21:44
merlin1991hah that soudns easier :)21:44
DocScrutinizer05and more sane21:44
freemangordonwell, that is binutils 2.2221:45
freemangordonmerlin1991: try it on n90021:45
merlin1991currently adding binutils to n90021:45
freemangordonfrom extras-devel?21:45
merlin1991from sdk repo21:46
freemangordonworks too21:47
* DocScrutinizer05 just wonders why he never seen this Tag_THUMB_ISA_use on the allegedly thumbified modest binary he hammered with all flavours of readelf and whatnot else21:47
DocScrutinizer05I spend several hours on that until I gave up and concluded there's no proper way to find out of thumbification directly and verbatim21:48
freemangordonDocScrutinizer05: NFC, but lemme check osso-gnome-vfs21:48
merlin1991freemangordon: it says thumb1 on my t device21:49
merlin1991thumb2 on my thumb device21:49
freemangordonwhich one?21:49
merlin1991but Tag_THUMB_ISA_use: Thumb-1 is on all binaries I have21:50
freemangordondamn :(21:50
DocScrutinizer05well, I'm not completely braindamaged then21:50
freemangordonso, back to scripting21:50
merlin1991err this is getting interesting21:51
freemangordonmerlin1991: which one?21:51
merlin1991osso-gnome-vfs on my -testing device does not have the stuff that the grep results in anything21:51
merlin1991only on my -thumb device the grep has a result21:51
freemangordonmerlin1991: could you pastebin the result from readelf -s on your -testing device?21:52
freemangordonand ls -al of .so21:53
freemangordondamn, maybe it is interworked :(21:54
freemangordonmerlin1991: ^^^21:54
merlin1991-rw-r--r--    1 root     root        339112 Apr 10  2012 /usr/lib/
freemangordon"Apr 10  2012" ?21:55
freemangordonmerlin1991: are you sure it is not on -stable branch?21:56
merlin1991ii  mp-fremantle-community 21.2011.38-1Tmaemo3.1  Community SSU package, includes latest fixes from Gitorious21:56
merlin1991it's not on T421:56
* merlin1991 headdesks21:56
freemangordonversion (testing): 21.2011.38-1Tmaemo4; (stable): 21.2011.38-1Smaemo3 / - Set by merlin1991 at 02:59 on Вторник 12 Юни 201221:57
* merlin1991 runs the update21:57
merlin1991I feel silly now :P21:57
DocScrutinizer05I feel ghosty now21:58
merlin1991why ghosty?21:58
* DocScrutinizer05 ghosts himself21:58
merlin1991hm wrong ham on the device21:59
merlin1991ah yeah I remember21:59
merlin1991I didn't update to T4 in order to be able to easily test palis patches21:59
merlin1991find / -type f -print0 | xargs -0 ./thumbtest.sh22:07
merlin1991with readelf -h $1 && readelf -s $1 | grep -v "00000001" | grep FUNC |grep ": [0-9a-f]*[13579bdf] " > /dev/null && echo $122:07
merlin1991in the script22:07
merlin1991that should work22:07
kerioor just -exec22:08
merlin1991exec not suitable here, since I need to execute a complex statement22:08
kerioi mean22:08
kerio-exec sh "{}" ";"22:09
merlin1991that's almost the same xD22:09
merlin1991but iirc busybox find has no -exec22:09
keriothat's why you have /usr/bin/gnu early in your PATH22:10
keriolike a doc22:10
merlin1991okay osso-gnome-vfs is now thumb22:11
kerioand that's bad22:12
DocScrutinizer05except messybox find doesn't support -exec22:12
DocScrutinizer05aah you already said that22:13
freemangordonmerlin1991: what is the result of readelf -A ?22:13
freemangordonand for the others? thumb1?22:14
freemangordonso, after all we have a more human readable thumb2 criteria22:15
freemangordonmerlin1991: could you run that on
freemangordoni think it is in /opt/22:17
merlin1991don't have it22:18
merlin1991gotta go for dinner, bbl22:18
freemangordonmerlin1991: "/opt/maemo/usr/lib/sharing/plugins"22:19
freemangordonDocScrutinizer05: ^^^ ?22:20
freemangordoncould you run readelf -A against  /opt/maemo/usr/lib/sharing/plugins/
DocScrutinizer51that line is"mt c+p able22:23
DocScrutinizer51well could try deleting 'against'22:24
freemangordon"readelf -A  /opt/maemo/usr/lib/sharing/plugins/"22:24
DocScrutinizer51readelf: Error: ' /opt/maemo/usr/lib/sharing/plugins/': No such file22:30
DocScrutinizer51and that's the way I like it ;P22:31
merlin1991hm xargs doesn't like me today22:32
merlin1991or find22:34
merlin1991or busybox22:36
DocScrutinizer51fsck messybox22:38
merlin1991if I put an echo $1 into my script22:38
merlin1991and do a find . | xargs ./myscript22:38
kerioDocScrutinizer05: i don't think fsck can fix it22:38
infobotmessy... err busybox is meant for lean scripting. Regarding all the missing options and immanent limitations (see su) it's not really the interactive shell of choice. A lot of people hate busybox because a lot of system integrators don't understand the difference between busybox and a decent user interactive shell plus unix utils22:38
merlin1991I only get the first line from find back22:38
DocScrutinizer51merlin1991: use tmpfiles22:40
merlin1991hah xargs -0n 122:40
DocScrutinizer51dang, no 10k lines prolly22:40
merlin1991silly though that I have todo that22:40
merlin1991okay it's running22:42
merlin1991and It's already getting odd22:42
merlin1991but it's finally running properly22:43
merlin1991find /usr /lib /bin /home -type f -print0 | xargs -0n 1 ./ 2>/dev/null | tee thumbs.txt22:43
*** arcean has joined #maemo-ssu22:46
merlin1991hm now that I think of it I should have pushed the current file to stderr so I see where it is :/22:46
*** arcean__ has joined #maemo-ssu22:47
*** arcean has quit IRC22:47
merlin1991uh oh, a few in here are really bad22:47
DocScrutinizer51tee &222:47
*** arcean has joined #maemo-ssu22:48
merlin1991it's ham and gnomevfs22:50
freemangordonwhy7 is that?22:50
merlin1991but it feels baaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaad22:50
keriooh god why22:50
* kerio installs kp51 RIGHT FUCKING NOW22:51
kerioPali: PUSH IT GOOD22:51
*** arcean__ has quit IRC22:51
merlin1991kerio: I don't see the reason for that :D22:51
Palikerio, there are problems on autobuilder22:51
keriomerlin1991: thumb errata workaround?22:51
Paliit cannot compile kp5122:51
PaliI will look at it later22:52
Palinow compiling xbmc for rpi :-)22:52
* merlin1991 wonders if ham is a false positive or really thumb22:52
Pali(I cannot compile all at one time)22:52
merlin1991Pali: you got an rpi?22:52
Palimerlin1991, yes22:52
freemangordonmerlin1991: 20.2010.36-2maemo14 (11.04.2011) was the last HAM update according to changelog22:52
merlin1991I finally managed to order one 2 days ago, thus yay anotehr 11 weeks22:53
PaliI ordered in april and got it last week22:53
*** luf has joined #maemo-ssu22:53
merlin1991dang luf22:53
merlin1991hopping in and out22:53
freemangordonmerlin1991: besides HAM we have /usr/libexec/apt-worker.real too, i don't believe both are false positives22:55
merlin1991hm lemme recompile ham on my sb and see if it still detects it then22:56
freemangordon"/usr/bin/maemo-confirm-text.launch" is part of HAM package too22:56
merlin1991yeah I know everything not osso gnomvfs is ham22:56
*** amiconn has quit IRC22:57
*** amiconn_ has joined #maemo-ssu22:57
*** amiconn_ is now known as amiconn22:57
*** NIN101 has quit IRC22:58
*** lizardo has quit IRC22:58
*** trumee has quit IRC22:58
*** trumee_ has joined #maemo-ssu22:59
*** trumee_ is now known as trumee22:59
*** _rd has joined #maemo-ssu23:00
*** rd_ has joined #maemo-ssu23:00
Palimerlin1991, can I push all my cssu patches packages to cssu-devel?23:00
*** rd_ is now known as Guest9250223:00
kerioPali: why not testing? :D23:01
merlin1991Pali: you can push whatever you want to to cssu-devel :)23:01
merlin1991it's meant to be living on the edge23:01
Palithen I will push also problematic HAM23:01
Palithen I will see what will do patched HAM if will be installed from *trusted* repository23:01
merlin1991btw Pali the patch to add cssu repos to trusted, was that part of the enabler or ham?23:03
merlin1991or both?23:03
Paliof enabler23:03
*** NIN101 has joined #maemo-ssu23:04
merlin1991freemangordon: ham seems to be positive23:06
merlin1991my compile does not get detected23:06
kerioshit, HAM is important23:06
keriowhy does it... work, though?23:06
*** arcean_ has quit IRC23:07
merlin1991because the problem with thumb is not that it always fails, but only from time to time23:07
merlin1991kerio: ie. modest in cssu has been thumb for quite a long time23:08
merlin1991it worked, quite well but had unpredictable crashes once upon a time23:09
keriomerlin1991: yeah, and we see how that went23:09
kerioquite well=slow as shit?23:09
merlin1991well it doesn't crash anymore but is still slow as shit :D23:09
merlin1991so there is no difference there :P23:10
kerioso... fixing HAM is important, i feel23:10
keriowill the update need HAM?23:10
keriothat could be a problem23:10
merlin1991update can be done by apt-get aswell23:10
merlin1991but kerio there have been quite a few cssu updates since we introduced the broken ham, so you'll be able to apply this ham update with ham aswell :)23:11
merlin1991actually 19 updates since it came :P23:12
kerio19 updates of what kind?23:12
*** lizardo has joined #maemo-ssu23:12
merlin1991cssu updates in general23:12
keriowell who gives a shit23:13
merlin1991broken ham came with 20.2010.36-2maemo14 (11.04.2011) and has been there ever since23:13
kerioi always apply those with fapman anyway :323:13
keriowhich still has a fucking awesome name23:13
merlin1991hm I need to find another usb charger so I can charge my stable n900 aswell to see if there is any thumb stuff there (though I doubt that)23:16
merlin1991why is it that my n900s always run out of battery at the same time?23:18
merlin1991probably related to the fact that I usually charge them simultaniously :P23:18
*** Pali has quit IRC23:19
freemangordonmerlin1991: you have compiled all of the stuff in -stable, ain't?23:20
merlin1991doing it just in case, I'm pretty sure I know why mags stuff was thumb and mine wasnt23:21
merlin1991when you install sb with the python installer it creates the target.ENVIRONMENT file23:22
merlin1991which in turn sets DEB_BUILD_OPTIONS to soemthing like mamemo-launcher thumb ....23:22
merlin1991and again some debian/rules files check for that "thumb" option23:22
*** luf has joined #maemo-ssu23:25
merlin1991hey luf23:25
freemangordonluf: ping23:25
keriohmm... who can you guys blame for this?23:26
merlin1991nokia is at fault :D23:26
merlin1991they provide that fsckd sb installer23:26
merlin1991when the manufacturer provided sdk already works against you, what can you do? :D23:26
merlin1991luf: you'll have to work on the libcurl patches a bit23:27
luffreemangordon: pong :D23:27
lufmerlin1991: What's the problem?23:27
merlin1991I tried to actually build it today, and it does not build because of unmet dependencys23:27
lufwhat dependencies?23:28
freemangordonluf: hi23:28
MohammadAG<merlin1991> when you install sb with the python installer it creates the target.ENVIRONMENT file23:28
MohammadAGfyi, I use the shell scripts :p23:28
merlin1991luf: libgcrypt11-dev libidn11-dev openssh-server stunnel423:28
freemangordonMohammadAG: why then you SB is screwed up23:28
MohammadAGfreemangordon, cause the sh scripts do that too I guess :P23:29
merlin1991I don't have a single .environment file :D23:29
lufmerlin1991: what about apt-get install? :)23:29
lufii  libgcrypt11-de 1.5.0~svn1440- LGPL Crypto library - development files23:29
merlin1991luf: those are in extras23:29
merlin1991cssu must live from nokia ssu / sdk repo23:29
merlin1991thus either we can work around them or we need to include the libs into cssu23:30
freemangordonluf: why do we need that?23:30
lufIt's only build dependencies ...23:30
lufIt's from debian build package.23:30
keriothe more i read about maemo, the more it feels like a minefield23:30
keriofor everything23:30
freemangordonmerlin1991: could you build libcurl3 without those?23:30
merlin1991freemangordon: I have nfc23:31
lufI'll try to remove this build dependencies and let you know ...23:31
freemangordonmerlin1991: hmm, why don't you just edit debian/control?23:31
*** NIN101 has quit IRC23:31
keriobtw, KP51 would've fixed this with no extra binaries23:32
lufBTW are someone interested in ?23:32
merlin1991freemangordon: because I want to know what happens if the lib is missing23:32
merlin1991just removing the dependency and hoping for the best is bound to backfire23:32
freemangordonmerlin1991: ok23:33
merlin1991luf: that's very interesting23:34
merlin1991I'm actually sitting on the code from the wiki page (obexd 0.37)23:34
MohammadAGmerlin1991, use -d23:34
MohammadAGand see if it fails23:34
freemangordonluf: why is that targeted to CSSU. Just asking for the reason, not saying it does not qualify.23:34
merlin1991but if your 0.46 works without sideeffect it's perfect23:35
merlin1991freemangordon: bluetooth stack is a system lib23:35
merlin1991thus cssu23:35
freemangordonaah, ok23:35
lufI also tried to upgrade bluez but it has some problems I can't go through yet.23:35
freemangordonluf: repo anywhere?23:35
luf0.46 has the the same buggy behaviour. But I hope it's fixed with eds patch.23:36
luffreemangordon: I have no public www/ftp server :(23:36
*** Guest92502 has quit IRC23:36
*** _rd has quit IRC23:36
freemangordonluf: gitorious/github ;)23:36
merlin1991MohammadAG: instafail :D23:36
luffreemangordon: ok, I'll try it ...23:37
luffreemangordon: but it's not very good for .deb files ;)23:38
merlin1991luf: see pm23:38
freemangordonluf: I mean for the source code23:39
*** Vlad_on_the_road has joined #maemo-ssu23:39
luffreemangordon: I'm working with patches so I can easiely upgrade to new debian updates ...23:40
freemangordonluf: sure, but noone can contribute23:40
luffreemangordon: noone contribute longer than year or too ;)23:41
luffreemangordon: I understand.23:41
*** mirandir has quit IRC23:46
*** Guest92502 has joined #maemo-ssu23:58
*** _rd has joined #maemo-ssu23:58
*** lizardo has quit IRC23:59

Generated by 2.15.1 by Marius Gedminas - find it at!