IRC Logs for #crux Monday, 2013-05-20

*** doomicide has quit IRC00:15
*** hawksbill has joined #crux01:46
*** tilman has quit IRC02:03
*** tilman has joined #crux02:05
*** __mavrick61 has quit IRC02:28
*** __mavrick61 has joined #crux02:30
*** hawksbill has quit IRC02:52
*** hawksbill has joined #crux03:00
*** v33 has joined #crux03:19
v33we had a guy in here named tuxhat, right?03:20
*** v33 has quit IRC03:46
*** mike_k has joined #crux06:04
*** Rotwang has joined #crux06:51
*** hawksbill has quit IRC06:52
*** jue has quit IRC07:06
*** jue has joined #crux07:17
*** Rotwang has quit IRC07:58
*** dkoby has joined #crux08:26
cruxbot[core.git/3.0]: attr: updated to 2.4.4709:22
cruxbot[core.git/3.0]: acl: updated to 2.2.5209:22
tilmananyone else having trouble building firefox 21?  "make[4]: write error" pretty early in the build process. just running make again seems to make it work10:39
frinnstno issues here with the betas or final10:45
tilmanpretty sure my hw is broken10:47
tilmanmachine completely freezes some minutes into the build10:47
nrxtxtilman: ram and swapspace full?11:06
frinnstwonderful day so far.. home with man-flu and the neighbour clogs his zink. sludge spewed out in my kitchen11:23
frinnstthe stench was.. ...11:24
nogagplzwhat the hell is man-flu11:30
*** doomicide has joined #crux11:42
teK_nrxtx, jaeger: using another stick solved the problem so it's either kernel upgrade or stickswitch. Thanks for the suggestions11:42
tilmanman-flu, lol11:43
tilmannrxtx: dont have swap. 16 gigs of ram, 12 used for a tmpfs in /tmp. that's where i'm building firefox11:44
tilmanjust ran 1 pass of memtest, no errors. will run it longer tonight i guess11:44
teK_building the latest glade3 yields many many 'write error' warnings, too  (then stalls)11:46
nogagplzah ok, that's pretty boss11:46
tilmanteK_: interesting. maybe i should confirm what make was trying to write...11:47
teK_getting back to dev-business :p11:47
frinnstno swap, teK_ ?11:49
teK_glade3 builds fine on this machine11:49
frinnstsame here11:49
teK_frinnst: the machine in question has 8GB RAM and glade3 is rather small11:50
teK_glade-3.8.3 _does_ fail on my desktop machine (^-----)11:50
teK_will hace to strace the root of thate11:50
frinnstand you run swap on that?11:50
teK_no because 8GB are more than enough11:50
teK_I don't swap on any of my personal machines11:50
frinnstsure, but would be interesting to test :)11:51
teK_like. ever.11:51
teK_will keep that in mind11:51
frinnstme neither.. well i have 10gig swap activated11:51
frinnstnever seen anything being swapped to it though11:51
teK_currently the machine is unavailable as in my son is oplaying FIFA 1311:51
Romsterfirefox built over distcc so it's not a high number of concurrent jobs issue.11:51
teK_yeah Romster is our grid guy11:52
Romsteri'm still hunting for more core 2 boxens to add to my collection11:54
frinnstget an i7 instead. you'll save $$ on the electric bill :)11:56
nogagplzyeah seriously, 2 would probably demolish your current 256 386 ray tracing cluster11:57
Romsteravtualylw hen i can afford it i'd liek to get a stack of mini mobos to use in a distcc cluster.12:00
Romsterdunno what i'd need 4gb ram 512MB flash card, gigabit lan, fast cpu i guess.12:01
joacimi'll crush your cluster with my PS2+linuxkit cluster12:05
nrxtxRomster: what about the hardkernel boards?12:08
tilmanwtf is a hardkernel?12:09
nrxtx the ODROID-X2 for example if you line up several of those :D12:10
tilmanyes yes, i said nvm ;p12:10
Romsterjoacim, lol12:11
Romstertesting clang on firefox over distcc nrxtx12:12
nrxtxRomster: nice, but as discussed yesterday the performance of firefox might be lower than a firefox compiled with gcc12:13
Romstereh that hardkernel wouldn't even match what i'm after.12:14
Romsteryeah i know compile times != run time12:14
Romsteri usually prefer -fomgoptimise over speed of compiling.12:14
Romsterhmm plus they are arm.12:15
Romsterthough i could crosscompile.12:15
tilmanjust froze again building on the ssd :/12:16
nogagplzsign that your cells have been burnt out12:17
tilmancould this be the cpu overheating? or does that result in a reboot usually?12:17
tilmannogagplz: riiiiight12:17
nogagplzpfft. you'd eat that word if it was actually true...12:18
Romstercpu over heating usually either freeze or just runs supper slow.12:18
Romsteror in some boards it'll power off.12:19
Romsterclang failed with an error in the firefox code.12:27
teK_frinnst: why no tmpfs anymore?12:30
*** joacim has quit IRC12:38
*** joacim has joined #crux12:38
Romsterfirefox built in 64 minutes again just fine tilman, you must have a serious problem. it might be power supply related. if it's not ram/hdd/sdd13:09
jaegerthat narrows it down :)13:09
Romsterthat's on a sysup just done like before building firefox.13:10
Romsterman this gas lift chair is annoying it keeps going down on me.13:12
Romsterworst invention ever13:12
Romsterat least the threaded screw chairs never went down13:13
fireglowDon't anger it or it'll shot its steel bolt up your bottom13:13
Romstershoot. shot is in the past and already done.13:14
fireglowoh right13:16
Romsteryou got me worried now -_-13:17
*** horrorSt1uck has joined #crux13:17
*** horrorStruck has quit IRC13:20
frinnstRomster: real23m12.526s13:22
frinnstfirefox build :)13:22
cruxbot[opt.git/3.0]: p5-html-parser: 3.70 -> 3.7113:27
cruxbot[opt.git/3.0]: gstreamer1: 1.0.6 -> 1.0.713:27
cruxbot[opt.git/3.0]: gst-plugins-base1: 1.0.6 -> 1.0.713:27
cruxbot[opt.git/3.0]: distribute: 0.6.36 -> 0.6.4013:27
Romsteri must be doing something wrong unless frinnst has a mainframe13:29
Romster2.0Ghz dual core, 2.0Ghz dual core, 2.6Ghz dual core, 3.4Ghz quad core13:30
Romsteris what i'm running in distcc, the other quad core needs work done to it.13:30
jaegerI assume frinnst is running on his i7 which cuts out the overhead of network transfers and distcc entirely, plus it's several generations newer than any of those hosts, I bet13:31
Romsterwell my desktop is a phenon II13:32
Romsterthe rest are older core 2 intells after the pentiums came out but before the i313:32
Romsteri probably should do a test run on my quadcore without distcc and see what my times are too.13:33
Romsteri haven't had time to see if i can optimize it any better.13:34
Romsterit's all over gigabit lan too.13:34
frinnstRomster: while a "quadcore" might sound modern and quick, sandy-bridge is insanely fast13:37
*** deus_ex has joined #crux13:37
frinnsti never imagined it would be that much difference from my core2quad q660013:37
frinnsti was wrong :)13:37
Romsteris that your supposed 24 core system that's using two 6 core cpus with HT13:38
frinnstno, this is my desktop i7 2600k - 4 cores + ht13:38
frinnstthe cost is not *that* high. i strongly recommend it :)13:39
Romsteri might have to leave AMD if they don't come out witha  decent new CPU soon. the newer ones withthe APU crap are less performance than this phenom II 96513:40
nogagplzyeah romster the 2600k really kicked tubs in crux13:41
nogagplzwas using one myself in it and boy did it run rings around my old q660013:41
frinnstyeah, and its not even very new.. i've had mine for close to two years13:41
nogagplztruly a magical piece of kit13:42
RomsterLGA1155 or the newer 2011 socket?13:44
frinnsti have 115513:44
nogagplzthe 2011 stuff is damn expensive13:44
Romsteryeah really i was like oh geez13:45
jaegerI'm on the same one as frinnst, it's really nice13:45
frinnstnot sure what the intel future for their sockets look like13:45
RomsterSPECIAL  Intel Core i7 3770 (3.40Ghz / 8MB / LGA1155 / Quad Core)  $31513:46
Romsterthought they were more than that... price must of came down alot.13:46
RomsterSPECIAL  Intel Core i7 3770K (3.50Ghz / 8MB / LGA1155 / Quad Core / Overclocking Enabled)  for $35513:47
frinnstiirc the price was the same when i bought it13:47
frinnstmaybe a bit more13:47
frinnstthe only change between sandy and ivy bridge are the onboard gpu's iirc13:48
nogagplzI have the 3770k now romster and onthe same hsf as yours it can get to 4.3GHz pretty damn easily13:48
nogagplzmaybe even higher but I don't havethe patience for the fiddling13:48
Romsteryeah that heatsink is a massive whooper. i like it.13:49
Romsteri'd seriously want to go with triple or quad channel ram on my next build but that's gonna be some time away.13:56
jaegerI'll probably get an i5 instead of i7 next time I upgrade. I do most of my building on a VM server now and my games/emulators don't need 8 logical CPUs14:01
Romsteri really only got them core 2's for moose setup but i decided to throw them in distcc for the heck of it too.14:02
nogagplzdoesn't you care about your epenis size though jaeger :P14:04
jaegeronly when it's cost-effective :P14:06
jaegerbesides, I've got more than enough virtual e-penis in VMware :D14:06
nogagplzok you win :P14:07
Romsternever thought i'd see the day nogagplz said that.14:14
Romsters/said that/would say that/14:14
nogagplzit's one thing to have an answer to everything, but it's another entirely to know when you're outmatched14:15
*** dkoby has quit IRC14:35
cruxbot[contrib.git/3.0]: gst-plugins-good1: 1.0.6 -> 1.0.714:37
cruxbot[contrib.git/3.0]: gst-plugins-bad1: 1.0.6 -> 1.0.714:37
cruxbot[contrib.git/3.0]: gst-plugins-ugly1: 1.0.6 -> 1.0.714:37
cruxbot[contrib.git/3.0]: deco-archive: 1.5.1 -> 1.614:37
cruxbot[contrib.git/3.0]: deco: 1.6.2 -> 1.6.314:37
cruxbot[contrib.git/3.0]: dmidecode: 2.11 -> 2.1214:37
cruxbot[contrib.git/3.0]: ffmpeg-compat: 0.10.6 -> 0.10.714:37
cruxbot[contrib.git/3.0]: ffmpeg: 1.2 -> 1.2.114:37
cruxbot[contrib.git/3.0]: fftw: 3.3.2 -> 3.3.314:37
cruxbot[contrib.git/3.0]: libbluray: 0.2.2 -> 0.2.314:37
Romsterlike my beard out matches you :D14:38
nogagplzyou have to sleep some time, and I know how to use a razor ;P14:43
Romsterapparently i never sleep according to those that ask if i ever sleep.14:44
Romsteri've done enough for one night i'm heading to bed.14:44
jaegerspeaking of v-e-penis I wonder if firefox would benefit from 11, 16, or 24 cores and tons of RAM.... might have to test :)15:09
tilmanjaeger: doubt it. there's no hope for firefox15:13
jaegerI'll try it, need to work the new VMware cluster a bit anyway :)15:15
horrorStrucksoon this browser will be an OS if they... wait WHAT15:17
jaegerSoon we'll need to switch the "kitchen sink" jokes from emacs to <insert browser name here>15:21
frinnsttilman: check out the chromium tarball :)15:22
frinnst 178M jan  8 04:08 /usr/ports/src/chromium-24.0.1312.52.tar.bz215:22
tilmancan i have web1.0 back?15:23
jaegertilman: here you go:
frinnstdear god, its glorious!15:24
jaegerIsn't it? :D15:25
*** acrux has quit IRC15:27
*** acrux has joined #crux15:35
jaegerheh, just noticed I typed 11 cores earlier, meant 1215:36
jaegercurse you, fingers!15:36
*** kris2 has joined #crux15:37
*** timcowchip has joined #crux16:21
timcowchip16 updates I'm afraid to do sysup16:23
teK_do one today, the next tomorrow etc. O_o16:25
timcowchipthere will be a lot of revdep16:27
timcowchipsomebody once said the arch devs updated things too much16:28
timcowchipthey got nothing on you guys16:28
timcowchipexcept that systemd fiasco16:29
timcowchipI should say they got nothing on Romster16:30
teK_doing 9 updates at once16:30
teK_<---- bad ass16:31
jaegeroh snap, we got a badass here16:31
tilmanno issues yet ._o16:34
tilman(except that it's slow ;p)16:34
*** Kaishi has joined #crux16:38
teK_that's badass too16:40
tilmanbuild succeeded btw16:40
tilmanwtf is this16:40
teK_shitty makefiles? :o16:40
tilmanteK_: the firefox build used to _hang_ the system. reproduced it at least twice16:41
tilmannot due to a makefile :)16:41
*** Rotwang has joined #crux16:42
timcowchipdoes libass really need enca?16:45
timcowchipjaeger: I'm using your port of libass and it lists enca as a dep16:46
timcowchipdoomicide has a libass port with only fribidi as a dep16:48
*** noname3 has joined #crux16:50
jaegerIt may not, it did at some point in the past16:58
tilmanseems to be the cpu overheating16:58
tilmanaccording to k10temp, the cpu gets at least 5-10 degrees hotter when all cores are running on full throttle16:59
timcowchipok thanks jaeger17:06
*** guzzano has joined #crux17:55
*** Kaishi has quit IRC18:08
timcowchiphandbrake doesn't break18:14
timcowchipbacklite doesn't break18:23
*** lasso has joined #crux18:39
timcowchipgstreamer1 footprint needs updating18:59
timcowchipcausing it to fail when doing prt-get sysup18:59
timcowchipwhich, in turn , cuases gst-plugins-base1 to need rebuilding19:00
timcowchipwhich, in turn, cause gst-plugins-ugly1 and gst-plugins-bad1 to need rebuilding19:01
timcowchipand anything that depends on any of them to need rebuilding19:02
timcowchipI've been maually updating footprint files with almost every prt-get sysup19:04
timcowchipand of course ports -u reverses that footprint update19:05
timcowchipI've noticed a lot of commits are just Pkgfile and .md5sum no .footprint19:06
timcowchipso who knows how old the .footprint file in the repo is19:08
timcowchipif I do prt-get -uf will it download and use whatever .footprint file is in the repo?19:09
timcowchipif so ,might it be better to do prt-get if19:10
timcowchip"prt-get -uf sypup" and "prt-get -if sysup" that is19:10
timcowchip"prt-get -uf sysup"19:11
guzzanoSomeone you can share your .config for virtualbox? thanks19:16
poulecacahi ther19:29
poulecacaI have the found a old disk with crux 2.7.1 for x86_64 and I'm wondering if I will have any problem to update it with 3.0 iso19:31
poulecacaMy guess is that I will end up with useless files like /usr/port/opt-x86_6419:32
poulecacabut besides that this should be ok19:32
tilmanpoulecaca: the old crux for x86_64 was not a multilib system iirc19:38
tilmani'm not sure how easy it would be to upgrade it to crux 3.019:39
jaegerEasier than x86 at least19:39
poulecacahm I missed that point. I'll think about it but certainly end up erasing all / and /usr partitions19:42
jaegeryou can always erase /usr/ports and replace the files in /etc/ports so that part's pretty easy. upgrading from a pure x86_64 to multilib was pretty easy, too. replace the toolchain with the stuff from the CD and add a kernel option19:52
jaegerfirefox 21 on 12 cores with -j12: real    17m51.568s19:55
jaegeron to the 16-core test!19:58
*** Kaishi has joined #crux20:01
poulecacathx jaeger20:01
jaegernp - there's even an outdated wiki article on the process here:
jaegerthe git instructions aren't necessary since ports are multilib now but the process is basically the same20:02
jaegerupdate ports, pkgutils, glibc, gcc, binutils, mpc, mpfr, gmp and install glibc-3220:02
poulecacaawesome thx20:02
jaegernp, hope it helps20:03
poulecacaJust so I know. Why is the reason crux 3.0 is made multilib and not pure 64bit ?20:04
jaegerBecause maintaining one version is far easier than 2 and the extra stuff required for multilib isn't very large20:09
teK_guzzano: VBox creates ~/.config/.. ? not for me.20:12
poulecacahm if this isn't large that makes sense :)20:12
jaegerteK_: perhaps he meant a kernel config20:16
teK_sounds legit20:17
teK_although I don't think that there are that many vbox-regarindg options to enable besides module loading support20:20
jaegeryeah, nothing really specific. just need the right controller and network drivers like anything else20:20
*** lasso has quit IRC20:30
joacimi love that album!20:32
joacimi've got it on cassette20:33
teK_of course20:33
teK_frinnst: will you reformat your btrfs after the layout changes in 3.10(?)20:39
*** Rotwang has quit IRC20:58
*** doomicide has quit IRC21:36
frinnstnot sure21:51
frinnstthe metadata thingy?21:52
teK_did not read about it more deeply.. thought you could guide me ;)21:55
frinnstme neither21:56
frinnstbut apparantly you can switch between the new and old21:56
frinnstand its a simple tune-switch21:57
frinnstjue has a snapshot of btrfsprogs packaged. just change the $version to 2013051022:01
frinnst-x enable skinny metadata extent refs22:01
teK_really? I read that you'd have to mkfs the thingy22:05
teK_hm, gotta catchup tomorrow :)22:05
frinnst[  153.957276] btrfs: has skinny extents22:09
frinnstonly on my /~22:10
frinnstseems the fs needs to be unmounted for the convertion, problematic on /22:10
teK_hm must have mismemorised it then22:13
teK_or skimmed to quickly over the paragraphs :22:14
frinnstnice, still starting firefox22:15
frinnsthas been running for 5 mins22:16
frinnst  613 be/4 root     1336.54 K/s    0.00 B/s  0.00 % 99.99 % [btrfs-cache-1]22:16
frinnstah, there we go22:17
teK_uh, sweet22:18
teK_"EU wants to cheapen power"22:19
teK_after the banana thing (regulate the shape etc.) I've got a new favourite EU law: you are not allowed to put olive oil on the table on a plate. No it has to come from a bottle.22:20
teK_THANK GOD. Remember when I told you about Zakytnhos? The people there did that mischievous all the time!1122:21
teK_I was really worried about that ;>22:21
teK_and now I'm off to bed.22:21
teK_get better soon (again)22:21
cruxbot[contrib.git/3.0]: webkit: 2.0.1 -> 2.0.222:50
*** timcowchip has joined #crux23:16
timcowchipah webkit23:16
timcowchipnot me23:16
timcowchipno webkit, no firefox, no problem, that's my motto23:17
prologicMy desktop at home has finally died.23:32
prologicNeed to buy a replacement or some sort.23:32
prologicQ: If I want to do software raid - do I need to setup an initrf?23:33
timcowchipput livarp on it
timcowchipat the bottom of the page is a howto for building your onw debian-based OS23:43
prologicI want CRUX on it :)23:44
timcowchipwhen is 3.1 released?23:45
timcowchipmight be worth waiting23:46
prologicwe have slow release cyclces23:47
prologicno point waiting :)23:48
prologicnormally it's just an ISO update23:48
prologicwith kernel and toolchain updates (and packages)23:48
prologicfrom what I've read you need an inited for software raid23:48
prologicso I answered my own question :)23:48
prologicwhich makes sense23:48
prologiclilo/grub will only see the raw disks23:48
timcowchipok then, thanks for schooling me :)23:50
prologicno problems :)23:52

Generated by 2.11.0 by Marius Gedminas - find it at!