IRC log of #maemo-ssu for Friday, 2012-03-16

*** NIN102 has quit IRC00:05
*** Pali has joined #maemo-ssu00:10
*** trbs has quit IRC00:23
*** javispedro has joined #maemo-ssu00:40
*** Pali has quit IRC00:42
*** LaoLang_cool has joined #maemo-ssu01:24
*** M4rtinK has joined #maemo-ssu01:41
*** LaoLang_cool has quit IRC01:59
*** LaoLang_cool has joined #maemo-ssu02:19
*** BCMM has quit IRC02:28
*** LaoLang_cool has quit IRC02:56
*** arcean has quit IRC03:39
*** M4rtinK has quit IRC03:51
*** LaoLang_cool has joined #maemo-ssu03:58
*** LaoLang_cool has quit IRC04:00
*** LaoLang_cool has joined #maemo-ssu04:01
*** macmaN has quit IRC04:03
*** M4rtinK has joined #maemo-ssu04:10
*** macmaN has joined #maemo-ssu04:11
*** M4rtinK has quit IRC04:17
*** javispedro has quit IRC04:28
*** psycho_oreos has joined #maemo-ssu04:42
*** psycho_oreos has quit IRC04:50
*** amiconn has quit IRC05:10
*** amiconn_ has joined #maemo-ssu05:10
*** amiconn_ is now known as amiconn05:11
*** xnt14 has quit IRC05:12
*** MohammadAG has quit IRC05:13
*** psycho_oreos has joined #maemo-ssu05:14
*** psycho_oreos has quit IRC05:16
*** psycho_oreos has joined #maemo-ssu05:17
*** LaoLang_cool has quit IRC05:24
*** wmarone__ is now known as wmarone05:47
*** wmarone_ has joined #maemo-ssu06:26
*** wmarone has quit IRC06:27
*** LaoLang_cool has joined #maemo-ssu07:21
*** LaoLang_cool_ has joined #maemo-ssu08:09
*** LaoLang_cool has quit IRC08:09
*** neal has quit IRC08:40
DocScrutinizerhah08:56
DocScrutinizerhonestly, useless effort. There's nothing you could do different when knowing what kernel user flashed08:59
DocScrutinizernot asking user if he wants kernel to be updated "from your current version XY?" BAD idea08:59
DocScrutinizerno matter what kernel flavour is currently on device, you *always* have to ask user what she wants to do about it. That's another reason why including kernel into CSSU is not going to happen when the more reasonable guys have to also give a vote09:02
DocScrutinizeras a mandatory question "dear user, do you want to install a)PK49 b)PK50 c)PK49 with uBoot d)PK50 with uBoot e)stock kernel latest f)your own flavour ? you currently seem to have installed: kernel-foobar-201147999_with_multiboothack".... errr. What's THAT worth?09:04
DocScrutinizerwhen extras-infra is borked, goddamn pester xfade to finally do sth about it. The fact that it's broken since months in repo is NO SOUND RATIONALE to get it into CSSU as a mandatory "optional on interactive question" package - and even less you can forcefeed PK to all CSSU users without "optional on interactive question"09:08
DocScrutinizerwhy the hack can't you see it as a great opportunity to fix PK to base on git rather than a convoluted patchset, fix some long pending issues, and roll it out under pkgname Kernel-Power-Improved09:10
DocScrutinizerbtw to determine content of mtd3, the only working method is to calculate a md5sum and compare that against what you expect to find. All other tags you look for may as well show up in a user custom kernel that is different to what you *think* it was judging by such tag09:27
DocScrutinizerthen you need to evaluate if whatever method you chose to find out about kernel also works when there are bad blocks in that section of NAND09:31
DocScrutinizerwhich is an extremely tedious thing to do on RE basis, as we don't really know how NOLO handles bad blocks in mtd309:32
DocScrutinizeranyway keep in mind that, other than mmc mtd has no transparent bad block management. It's done in upper layer whatever that is for particular partition, for rootfs it's done in filesystem09:34
DocScrutinizerafaik there's no such filesystem on mtd309:35
DocScrutinizerso there has to be some other means to cope with bad blocks in mtd3, probably buried deep in NOLO09:36
freemangordonDocScrutinizer, seems you missed something, the idea behind what Pali is coding re mtd3 is to ASK user "You have kernel xyz installed, do you want to upgrade it to CSSU one?".09:36
DocScrutinizerexactly that is nonsense09:37
DocScrutinizerreread my posts, I commented on exactly that09:37
freemangordonI reread them several times, maybe it is my English, but AIUI you want exactly the same thing09:39
freemangordonUser to be able to opt-out09:39
DocScrutinizerof course you can host as many powerkernel pkgs as you like, in a way analog to what we got for orientation-lock applet in S now09:39
DocScrutinizerI.E. as a loosely attached optional pkg that incidentally lives on CSSU09:40
freemangordonaah, you don't like the name change or what?09:40
DocScrutinizerI don't like the idea to taint CSSU with a nagging of user about installing any of a ever growing selection of pkgs (here for kernel) that are not even depending on CSSU, not to discuss how CSSU depends on those pkgs which it *obviously* doesn't09:41
DocScrutinizerCSSU and kernel are obviously orthogonal09:42
freemangordonDocScrutinizer, "community kernel" is aiming 2 things: 1 - the place where reported kernel bugs are fixed. 2 - imrovements(in kernel) which has happend since last Nokia omap kernel was released09:43
freemangordonIt just does not make sense CSSU to depend on an external package, i.e. kernel-power in extras-devel09:44
DocScrutinizerexactly, and it doesn't and never will do09:45
freemangordonwell, I think that maybe you still misunderstand the idea09:46
DocScrutinizernot at all09:46
DocScrutinizerit seems you misunderstand the fact that PK has done for kernel since ages what CSSU is now doing for *rootfs*/system09:47
DocScrutinizerPK *is* the CSSU for kernel, and it's orthogonal to rootfs09:47
DocScrutinizerthus has to stay a indeoendent part not attached to / included in CSSU09:48
freemangordonthe "community kernel" will be a part of CSSU, on CSSU upgrade if the user has flashed other than Nokia omap kernel there will be a question "do you want your kernel zapped". Otherwise community kernel will live in CSSU repos. And we will take KP50 as base.09:48
DocScrutinizerand there are a plethora of possible configs regarding kernel, which CSSU impossibly can handle. Plus most users of CSSU are not interested in PK and even despise it getting shipped with CSSU09:49
DocScrutinizeruhuh, says who09:50
freemangordonIf by KP you mean OC, that is not mandatory for at all09:50
DocScrutinizerBS I'm not an idiot09:50
freemangordonWell, I am not too. Seems still I am short on coffee, lets continue the discussion on the matter when there are more CSSU devs here09:52
freemangordonDocScrutinizer, if memory serves me well, it was you to say "don't fork KP, put it in CSSU". But still, lets discuss community-kernel next time there are more people active here that just me and you09:54
freemangordonbb09:54
*** LaoLang_cool_ has quit IRC10:05
*** andre__ has joined #maemo-ssu10:18
*** andre__ has joined #maemo-ssu10:18
DocScrutinizerwell, please don't start quoting me on fake staements now. I definitely can't recall ever having been in a mood so I might even consider saying such thing10:23
*** dafox has joined #maemo-ssu10:26
DocScrutinizerI might have suggested (and did again some lines above) to *host* PK on CSSU repo, just like we do with orientation-lock applet pkg now. This isn't exactly nice as this repo is meant for things that can work and thus get installed under CSSU *only*, but it doesn't hurt THAT much if we get one or two pkgs there as well that could also work under stock PR1.3.110:27
DocScrutinizerthough clearly our "manifest" says "CSSU is *not* about anything that can get deployed via normal repo" - PK obviously can and has been since ages10:28
*** Pali has joined #maemo-ssu10:34
*** freemangordon_ has joined #maemo-ssu10:35
DocScrutinizeranyway even hosting PK on CSSU repo would basically mean users need to install CSSU to be able to get it (unless they do in a way not supposed to be "the starndard way", like adding CSSU manually to catalogs of their stocl PR1.3.1, or using wget or whatever)10:37
*** neal has joined #maemo-ssu10:49
*** ivgalvez has joined #maemo-ssu10:53
*** Pali has quit IRC10:55
ivgalvezWow, latest discussions around introducing KP in CSSU are useless!!11:01
ivgalvezIt's up to the developers that are actually doing the job11:01
ivgalvezDocScrutinizer: how can you argue with freemangordon in that way?11:02
ivgalvezIt seems very unpolite and nonsense11:02
*** dafox has quit IRC11:03
ivgalvezAnd there is a clear reason to have a community kernel, if it's finally possible to move to introduce errata for thumb2 in kernel, building the CSSU with thumb2 would requiere that kernel, wouldn't it?11:04
*** dafox has joined #maemo-ssu11:05
ivgalveznot to mention the possibility to move to newer gcc or glibc11:05
*** freemangordon_ has quit IRC11:28
*** Sicelo has quit IRC11:29
*** Sicelo has joined #maemo-ssu11:30
*** M4rtinK has joined #maemo-ssu11:32
andre__PK = packagekit?11:35
ivgalvezHe meant Kernel Power11:37
andre__ahaha. okay.11:37
andre__so much for abbreviations.11:38
*** neal has quit IRC11:58
*** BCMM has joined #maemo-ssu12:00
*** andre__ has quit IRC12:05
*** freemangordon_ has joined #maemo-ssu12:28
*** andre__ has joined #maemo-ssu12:30
*** andre__ has joined #maemo-ssu12:30
*** MohammadAG has joined #maemo-ssu12:32
*** xnt14 has joined #maemo-ssu12:33
*** DocScrutinizer has quit IRC12:34
*** DocScrutinizer has joined #maemo-ssu12:35
*** lizardo has joined #maemo-ssu12:40
*** M4rtinK has quit IRC12:56
*** ivgalvez has quit IRC13:17
*** freemangordon_ has left #maemo-ssu13:20
*** FireFly has quit IRC14:11
*** FireFly has joined #maemo-ssu14:11
*** neal has joined #maemo-ssu14:51
*** LinuxCode has joined #maemo-ssu15:17
Lava_Croftmeh15:26
Lava_Crofthttps://fbcdn-sphotos-a.akamaihd.net/hphotos-ak-ash4/422005_379468958743709_100000418249745_1336895_1361373448_n.jpg15:26
Lava_Croftshitty official nokia case dies so fast15:26
merlin1991Lava_Croft: WTF?15:26
merlin1991why is there this leather abdomination on top of the n900? :D15:27
Lava_Croftto protect it15:27
Lava_Croftalso makes the device way less 'smooth' and 'slippery'15:27
Lava_Croftso it dont drop from breastpockets, or from tables that easily15:27
merlin1991ruins all the fun :D15:28
Lava_Crofttoo bad this 20e case is still rather cheap15:28
Lava_Croftthere's probably about 50+ times a day that the n900 gets removed and put back in clothing pockets, seems the case dont like it15:28
Lava_Croftpretty bad construction with leather, carton and some rubbery black shit on the sides to close off the layers of carton and leather15:28
Lava_Crofti need something like this, only made from a proper material15:29
freemangordonmerlin1991, you said MohammadAG is going to build new CSSU update, but it was on Wednesday. Any news?15:29
Lava_Croftyes, i thought about just ductaping the entire device15:29
merlin1991yesterday around 21 Cet he was still on it15:30
merlin1991since then no reply15:30
freemangordonstill on it as in? because most of the packages are built and are on your server. What exactly MohammadAG is doing re new update?15:31
merlin1991there was a slight issue with the camera-ui l10n packages15:32
freemangordonwhat issue?15:32
freemangordonis it in logs?15:32
freemangordonso I won;t waste your time15:32
freemangordon*won't15:33
merlin1991nah it isn't but it's going to be resolved by mag anyway :D15:33
*** LinuxCode has quit IRC15:33
freemangordonwhat was the issue, as I have never had an issue building tohse, last build was on 12th of march15:33
merlin1991it's not a build issue but a maintaining issue15:34
freemangordonaaah, ok15:34
merlin1991has todo with the way the mr0 depends on the other packages15:34
merlin1991and how we want to depend on that from our mp15:34
freemangordonok15:34
Raimu"My n900 loves to wear leather."15:35
Raimu"Nothing but"15:35
Lava_Croftleather can be taken loosely here15:36
Lava_Croftits leather, but its more carton15:36
Raimufreemangordon: Is the KP50 that's trying to be repoed now the same build as the latest one in your dox.abv.bg, BTW?15:37
freemangordonRaimu, yes, absolutely the same15:39
freemangordonmerlin1991, I think to put a link to the "devel repo URL" on TMO, I know I was agains it, but... What do you think?15:40
merlin1991I suggest you wait untill the T release15:41
freemangordonwell, ok :)15:41
freemangordonRaimu, any problemswith KP50?15:41
merlin1991because then I'll update the repo to that and import the stuff that would go on top15:41
freemangordonmerlin1991, ok15:41
Raimufreemangordon, ever run into trouble with fcam-drivers -using applications with modern KPs? I mean, sometimes they just crash out randomly and blurt out weird things into dmesg.15:41
merlin1991+ it would be a real repo then :)15:42
merlin1991freemangordon: I've got a repo with incoming dir set up already15:43
Raimufreemangordon: I can't think of anything problematic and KP50-related. Fcam-driver apps like hdrsomething and lowlight have always been kind of randomly unusable on my n900, so I doubt it relates to KP. Some dspbridge warnings show up on dmesg on the occasion.15:43
merlin1991just need to symlink the pool and dists dir in the webserver and it will be avaiable15:43
RaimuBut that could be due to many things like the Harmattan-nicked 720p hack.15:44
RaimuAIUI15:45
freemangordonRaimu, dspbridge "deprecated ioctl" messages are absolutely normal if that is what you mean15:45
freemangordonmerlin1991, great, will wait "cssu-devel" repo to be announced15:48
Raimufreemangordon: I think that's what I remembered.15:48
freemangordonRaimu, just ignore them, it is a real warning from dspbridge driver when some application issues "CLOSE" ioctl. it is deprecated, but all of the applications in Maemo are build using older DSP API.15:50
RaimuOkay, hdrcamera blew up again and started spitting out sbl overflow errors...15:51
Lava_Crofthdr thingy never has crashed here in nearly 2y of usage15:51
Lava_Croftand been on CSSU since quite a while15:52
Lava_Croftand kp15:52
RaimuAh, yeah. Figures.15:52
Lava_Crofti use it quite a lot too:\15:52
RaimuI remember fcam-drivers apps being unstable from the start.15:52
Lava_Croftlowlight does crash sometimes15:52
freemangordonNow the only thing that "CLOSE" ioctl handler in dspbridge driver does is to issue the warning in wuestion end to return :)15:52
freemangordon*question15:52
RaimuThese days, though, hdrcamera crashes make it so I can't use any fcamd apps before a reboot. It buggers them or something.15:53
Raimufreemangordon: Ha, OK.15:53
Raimufreemangordon, anyway, no problems attributable to KP50 on my n900.15:54
Lava_Croftoh, i know this is probably wait off topic16:01
Lava_Crofthttp://www.ted.com/talks/rob_reid_the_8_billion_ipod.html16:01
Lava_Croftbut enjoy Copyright Math16:01
*** Pali has joined #maemo-ssu16:01
*** LinuxCode has joined #maemo-ssu16:17
amiconnRaimu: Are you overclocking and/or undervolting your N900? If so, that's most probably the root cause of those crashes16:27
amiconnThe fact that the cpu runs stable at a certain clock and voltage doesn't mean the dsp will. And if the dsp crashes, apps using the dsp, like cameras, video players etc will experience all sorts of weird crashes16:29
* amiconn learned that from his own N90016:29
freemangordon...and that is why KP50 has additional /sy/power entry to tweak DSP voltage when SR is enabled ;)16:33
freemangordon“/sys/power”16:34
Raimuamiconn: Yeah, I've looked into it but it seems they do the same crash dance on stock values. Video players work fine on the "dsp" profile.16:35
DocScrutinizer51and this is also why it's generally strongly recommended to not mess with voltages at all, as TI and Nokia incvested quite a chunk of work to find optimum settings that guarantee errorfree operation while conserving as much power as possible16:35
RaimuYup, aware of this.16:38
Lava_Croftbut the amount of time they spent on it is probably nothing compared to the time that users of the device spend on that subject16:39
DocScrutinizer51haha16:40
DocScrutinizer51you have no idea, honestly16:40
RaimuAlthough the engineers' target was to get a compromise that will be guaranteed to work on every piece of hardware they put out.16:40
RaimuReading the forums users easily tend to veer for the "works for me" approach.16:41
RaimuWhich is, I guess, fine and OK if they're willing to take the blame when it goes awry.16:42
RaimuI'm probably talking out of my ass again, so silence I shall resume. :P16:43
DocScrutinizer51well, as long as you don't think either of the two applies to you, you're better off not even bothering about 'optimizing' such stuff: 1) your usage pattern or requirements are so etremely different to average as evaluated by Nokia that you need to change system to accomodate. 2) you're willing to throw more manpower and lab equipment on a particular detail than those who do this for their makking-money, and we're talking bout nnn16:44
DocScrutinizer51millions here16:44
DocScrutinizer51neither Nokia nor TI optimize for a "works under all circumstances for 9999/10000 of devices. They *always* optimize for maximum customer satisfaction16:46
DocScrutinizer51basically not even that, they optimize for bargain, but that's mostly the same16:47
DocScrutinizer51IOW regarding voltage: TI&Nokia optimized for a sweetspot between power conservation and stability already. Unless you're willing like whatever accept 10 timesw more random crashes for 4% increased standby time, it'snot worth looking into it again16:52
freemangordonDocScrutinizer51, if that was the case (i.e. Ti and Nokia spending all the night in the lab, trying to squeeze every bit of stability against power), could you tell me why the hell SR is not wotking out of the box on N900?16:54
DocScrutinizer51because there been a ticket against SR that suggested random crashes go thru the roof when activated16:55
freemangordonAnd calibration values look like some marketing guy put them in efuse16:55
DocScrutinizer51investigation on that takes time16:55
DocScrutinizer51muuuch time16:55
freemangordonticket? on Nokia bugtracker?16:56
DocScrutinizer51eventually somebody finds the full story incl root cause, and when you're lucky they even find a SW 'fix'16:57
DocScrutinizer51ticken on internal tracker, yes. Ask andre__ - I bet hecan look up such ticket16:57
DocScrutinizer51I dunno if he may disclose details16:58
freemangordonWell, Nokia can do absolutely nothing about the fact that the whole batch of omaps they’ve received from TI had fcked efuse SR calibration.16:58
DocScrutinizer51:nod: quite possible16:59
RaimuDocScrutinizer51, thanks for the clarif.16:59
DocScrutinizer51YW16:59
DocScrutinizer51bbl, daywork17:00
RaimuTo be honest I'm often quite guilty on fiddling on things that needn't fiddling.17:02
*** infobot has joined #maemo-ssu17:02
*** ChanServ sets mode: +v infobot17:02
*** LinuxCode has quit IRC17:18
*** Sicelo has quit IRC17:22
*** Sicelo has joined #maemo-ssu17:23
*** Sicelo has joined #maemo-ssu17:23
*** Sicelo has quit IRC17:26
amiconnDocScrutinizer51: There's an important difference between manufacturer and user optimisation though.17:28
amiconnThe manufacturer needs to optimise in a way that it works for *all* devices. The user can optimise for a single device17:29
*** Sicelo has joined #maemo-ssu17:39
*** Sicelo has quit IRC17:48
Paliping merlin1991, freemangordon17:49
merlin1991ping Pali17:49
Palimerlin1991, I updated my program kernel-version17:49
Palinow it use malloc() + read()17:49
Paliinstead mmap17:50
Palimtd kernel partition is 2MB, so it can be aloocated and readed by one read() call17:50
merlin1991what does it report in case of installed uboot?17:52
PaliVersion string not found17:52
Paliit ignore attached kernels to the end of some image17:52
Palicorrectly, program try to find that kernel gz archive in first 2^16 bytes17:53
Paliand attached kernel to u-boot binary is at the end17:53
merlin1991sure that's ok, but can we detect IF uboot is installed?17:54
*** Sicelo has joined #maemo-ssu17:54
PaliThis program detect if zImage kernel is installed17:54
Paliso if not then some other bootloader is installed...17:55
Palido we need to specify name of other image (e.g. u-boot)?17:55
amiconnTo me that doesn't sound logical. If you only check the first 2^16 bytes, you only need to read those, not the whole 2^2117:55
merlin1991My idea of the cssu kernel flasher is a check if it is the right stock kernel or older cssu version --> flash cssu kernel, if not ask17:55
merlin1991but if possible I'd like the ask part to be able to use an existing uboot in mtd3 and flash with that17:56
merlin1991something along the lines of: we detected you have uboot installed, do you want to flash the cssu kernel or flash the cssu kernel with your uboot image or not flash at all17:57
*** andre__ has quit IRC17:57
Palimerlin1991, flashing u-boot with kernel is not simple...17:57
merlin1991it's damn easy17:57
Paliyou need to build compined u-boot image17:57
Paliand that is different as fiasco...17:57
Paliand we have more u-boot version17:58
merlin1991you can go from fiasco to u-boot image17:58
merlin1991and you can read the existing uboot and attach a new kernel17:58
Paliso we cannot prepaire 2 versions (one zImage and one uboot+zimage)17:58
Paliand for flashing: you cannot flash zImage17:58
Paliin mtd3 is some NOLO header (maybe with checksum)17:59
merlin1991ffs17:59
Paliso you need to flash fiasco image17:59
merlin1991I thought we have more in mtd317:59
merlin1991damn17:59
merlin1991ah well then it's a do you want to or not? question :D17:59
*** Sicelo has quit IRC17:59
Paliand generating fiasco image is possible ony with SDK repo (with fiasco-gen package)17:59
*** Sicelo has joined #maemo-ssu18:00
Paliso dumping u-boot somehow from mtd3 attach to that new zImage, creating new fiasco image and flashing is hard and possible only in scratchbox with SDK repo18:01
Palionly we can do is ask question: do you want to flash new kernel image?18:02
merlin1991yeah18:02
merlin1991well better than flashing without asking18:02
Paliand for cssu kernel I suggest NOT to change version string18:02
Paliwe have more kernel packages in extras-devel repo (also joikuspot) which has modules agains default kernel18:03
Paliand if we does not change kernel interface (only bugfix patches) there will be no compatibility problem18:03
Palimerlin1991, I suggest to create $PACKAGE which dpkg-divert /sbin/fiasco-image-update18:05
Paliand use some new wrapper for that fiasco-* binary which will ask if you want to flash kernel if new version is not same (or similar) as old18:05
Palithis will also ask for any kernel updates (not only cssu), so users will see if some package will want to flash kernel18:06
merlin1991can we do it somehow to only have it with cssu updates?18:06
Paliwe can include that $PACKAGE as predepends for cssu kernel18:06
Paliand package can go to cssu repo18:07
*** lardman has joined #maemo-ssu18:09
*** ekze has quit IRC18:11
*** ekze has joined #maemo-ssu18:11
*** Sicelo has quit IRC18:11
*** Sicelo has joined #maemo-ssu18:14
freemangordonmerlin1991, the tool to create .po files from .mo file (l10n) is called msgunfmt, please remember that :D.18:33
merlin1991noting it down :)18:34
Palistored in IRC log :-)18:34
merlin1991is it a simple tool as in can anybody do it?18:34
freemangordongoing to RE osso-pdfviewerv l10n18:34
merlin1991because then I can just do it, and leave you the more interesting tasks ;)18:34
Palifreemangordon, if you have time, can you also RE tvout l10n?18:34
freemangordonyou have it in every distribution, gettext tool18:34
Paliwe have some new untranslated strings in cssu tvout control panel18:35
freemangordonat least it is I have it in scratchbox18:35
freemangordonand in ubuntu too18:35
freemangordonpretty damn standatr tool18:35
Palimerlin1991, did you check if new version of kernel-version.c program get correct version of yours kernels?18:36
freemangordonmerlin1991, http://www.gnu.org/savannah-checkouts/gnu/gettext/manual/html_node/msgunfmt-Invocation.html18:36
merlin1991freemangordon: that looks quite straightforward18:37
merlin1991Pali: no I didn't find the time yet18:37
merlin1991Pali: did you update the armel build?18:37
Paliyes18:38
freemangordonmerlin1991, if you can do l10n instead of me, I will really appreciate that, I am running out of time now and tomorrow will travel to countryside. On the other hand I almost made a promise  on TMO to do it for the next update18:38
merlin1991I can do it18:39
freemangordonI know you can :p18:39
freemangordonThe question is will you :)18:39
merlin1991okay I WILL do it :D18:39
freemangordon:D18:39
freemangordongreat18:39
merlin1991Pali I can check tomorrow morning, gotta leave now :/18:42
Paliok18:42
Palilet me know if it working fine with wifi kp version...18:43
freemangordonPali, why the hell kernel-power 50 .deb is missing in autobuilder?18:43
freemangordonbut all other packages are there18:43
Palifucking extras and autobuilder18:43
freemangordonman, thats insane18:43
freemangordonPali, any sign of X-Fade?18:46
PaliI tried ping him today again18:46
freemangordonI’ll do that now18:46
*** M4rtinK has joined #maemo-ssu19:03
*** NIN101 has joined #maemo-ssu19:05
*** M4rtinK has quit IRC19:19
merlin1991damn you should have said that 3 hours ago19:26
merlin1991I had him fix the -testing repo19:26
merlin1991(there was a bogus entry in Sources.gz)19:26
*** arcean has joined #maemo-ssu19:34
*** M4rtinK has joined #maemo-ssu19:44
*** dafox has quit IRC20:05
*** BCMM has quit IRC20:10
*** M4rtinK has quit IRC20:26
*** M4rtinK has joined #maemo-ssu20:30
*** arcean_ has joined #maemo-ssu20:39
*** M4rtinK has quit IRC20:42
*** arcean has quit IRC20:42
*** arcean_ is now known as arcean20:43
*** Sicelo- has joined #maemo-ssu21:23
DocScrutinizerok, seems you all decided to go the mad hackers way and by all means force kernel into CSSU21:28
Sicelo-latest cssu comes with kernel?21:29
DocScrutinizerI'm afraid it will21:29
DocScrutinizerno really, I'm fine with that, if anybody could explain to me how that fits into CSSU manifest concept21:30
*** LinuxCode has joined #maemo-ssu21:30
DocScrutinizerCSSU has no dependencies on kernel. KErnel has no dependencies on any CSSU patch. And PK got published via extras repo since ages. 3 times no check, no argument why we want that stuff in CSSU21:31
DocScrutinizerbut obviously it's again en vogue to forcefeed another unrelated stuff to users of CSSU21:32
NIN101forcing kernels? sucks.21:34
DocScrutinizeryeah, and when we get some security patch to kernel, will we force a NIL update of CSSU to those that opted out of CSSU-kernel?21:40
DocScrutinizeror will that kernel patch only ship with next stable?21:40
DocScrutinizerand what we gonna do about the 57 flavours of uBoot-kernels?21:41
PaliDocScrutinizer, see log21:42
DocScrutinizerI had a cursory look at log and threw up21:42
PaliI'm writing program which ask user which has non nokia kernel flashed if he want to flash new version21:42
Palinow it write current version flashed in mtd321:43
DocScrutinizeruhuh, and those who got Nokia kernel? those have no choice or what?21:43
Paliwe should not change kernel string21:43
DocScrutinizeryou should not touch kernel at all21:44
Paliand we include bugfixes, so reason why to flash it21:44
DocScrutinizerI come over anybody touching my kernel without me asking for it, with fire21:44
NIN101yep.21:45
DocScrutinizerNO MATTER which kernel I decided to use that very minute21:45
DocScrutinizermy daily phone actually *is* running stock kernel, and I damn well know why21:47
DocScrutinizeryou can't even know if the system which is running CSSU been actually booted from kernel in mtd321:49
DocScrutinizercould have been booted from a kernel via vlasher&ramload, could be booted from mmc via kexec, with kernel God-knows-where21:50
DocScrutinizers/vlasher/flasher/21:50
infobotDocScrutinizer meant: could have been booted from a kernel via flasher&ramload, could be booted from mmc via kexec, with kernel God-knows-where21:50
DocScrutinizermaybe some cool haxor patches his NOLO to make it boot from mtd421:51
Palibut old version of nokia kernel in mtd3 can be updated to one which will have fixed security bugs21:52
DocScrutinizerhonestly keep your fingers from mandatory kernel updates unless you want a shitsorm hitting your inbox21:52
DocScrutinizerwho verified that "bugfixed" version, who audited it, who signed it off?21:53
DocScrutinizerwho tested it still works with e.g. uBoot, if only for mere limitaions of available free space in mtd3?21:54
DocScrutinizersorry, but in my book kernel is clearly where CSSU leaves own terrain and goes hamuk abroad - IOW there's a clear red line which CSSU mustn't cross21:57
DocScrutinizerI'm all happy with a CSSU-kernel as a associated but independent project21:58
DocScrutinizerbut CSSU as we know it is about userland, not kernel domain22:00
DocScrutinizereven flasher has an option to not flash kernel, or only flash kernel22:01
DocScrutinizerkernel is not userland, and userland is not kernel. full stop.22:02
DocScrutinizerusers don't appreciate kernel updates, unless they do it all under their own control, at the time they like, to the kernel they carefully picked or even built themselves22:04
DocScrutinizerand let's look at it from the diametrically other side: are you maintaining two locations for powerkernel then, or will CSSU refer tpo stuff outside CSSU repo, or will all the OC fanbois be forced to switch to CSSU then? If the latter, have fun with bugtrac! :-/22:06
LinuxCodeI agree with the kernel update statement22:07
LinuxCodestatement22:07
* LinuxCode wouldnt mind a userland tool, that informs me of an update though22:08
DocScrutinizersure, nothing against such a tool22:08
LinuxCodemaybe informing of the security issues/other bug fixes22:08
DocScrutinizer:nod:22:09
DocScrutinizerif it wasn't such unbearable crap regarding usability, that tool already had a name though: HAM22:10
RaimuDocScrutinizer: Your argument on keeping KP outta mainline CSSU is very, very convincing.22:11
DocScrutinizerand if Pali could get a hold of x-fade, to fix the lockup in extras repo, then nobody was *that* eager to get powerkernel into CSSU22:12
*** NIN101 has quit IRC22:28
DocScrutinizer( [2012-03-16 16:29:21] <amiconn> The manufacturer needs to optimise in a way that it works for *all* devices. The user can optimise for a single device ) sure, but the problem with user "optimizing" is your device won't differ that much from average due to immutable properties (there are no golden devices that are 3 times as fast as the worst case) - what manuf optimizes for is operation in a wide range of "environmental" conditions,22:35
DocScrutinizerlike extreme heat, cold, low battery, fresh battery, different GSM network peculiarities in different countries etc pp. The problem with user trying to compete against that is mainly that the user regularly only has limited knowledge about the fringe cases where the optimization actually needs to grip22:35
DocScrutinizervery usually that user "optimization" is done on a WFM basis, without *any* proper evaluation of even yield on the primary optimization goal (is it *really* faster / more power economic / whatever?) In 99.9% of cases this "optimization" is based on perceived 1337ness of doing such things on your own, plus a placebo effect that frequently comes in when testing stuff like standby times or performance without proper tools22:38
*** NIN101 has joined #maemo-ssu22:42
DocScrutinizera simple analogy: you might want to 'optimize' an electric motor. You reduce voltage so it consumes less power. All fine, it gets less warm, less wear, consumes less power. But what if the motor is regulated to keep a certain rotations per minute? you still can reduce voltage, but that will cause rotation speed to go down, and the regulation kicks in making the motor draw more electric current to compensate. Result: the consumed energy22:47
DocScrutinizeris probably higher due to motor not operation at optimum conditions, plus the higher current drawn will cause increased wear on the collector & brushes22:47
*** trbs has joined #maemo-ssu22:48
DocScrutinizerIOW: it's hard to do a proper optimization when you have no complete knowledge of how the system works. On a system complex as a SoC only the chip manufacturer has sufficient knowledge22:49
*** lizardo has quit IRC23:01
*** Sicelo has quit IRC23:09
*** Sicelo has joined #maemo-ssu23:10
*** Sicelo has joined #maemo-ssu23:10
*** dafox has joined #maemo-ssu23:10
*** bsdmaniak has joined #maemo-ssu23:13
*** Sicelo has quit IRC23:16
*** Sicelo- is now known as Sicelo23:17
*** dafox has quit IRC23:31
*** bsdmaniak has quit IRC23:36
*** nox- has joined #maemo-ssu23:47
*** nox- has joined #maemo-ssu23:47

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