/srv/irclogs.ubuntu.com/2009/12/17/#ubuntu-arm.txt

=== NCommander is now known as Guest48610
=== Guest48610 is now known as NCommander
=== NCommander is now known as Guest88322
=== Guest88322 is now known as NCommander
=== asac_ is now known as asac
persiaGrr..  Illegal instruction in mksquashfs in dove livefs build.  Is this a different bug than the one we saw for imx51 livefs last week?07:58
loolpersia: It might be that the dove kernels on the buildds have to be updated08:47
loolpersia: I think there was an errata for thumb 2 traps for dove which dave mentionned ina bug08:47
loolActually Eric did08:47
loolbug #49483108:48
ubot4Launchpad bug 494831 in linux-mvl-dove "Alignment trap/Unhandled fault errors on boot" [Critical,In progress] https://launchpad.net/bugs/49483108:48
persiaI definitely heard about that.  Do you think that would affect makesquashfs?08:48
loolpersia: If we rebuilt it with thumb2 it might08:48
loolBut I thought it had been rebuilt with -marm to workaround that08:48
persiaThat was my memory as well, and we ended up having working livefs builds for imx51.08:49
persiaDo you have hardware that could run a mksquashfs test to replicate the bug?08:49
loolI do but it's not online09:01
persiaOK.  Maybe someone else will volunteer then.09:02
loolChecking the build log, I don't see the mksquashfs version09:02
persiaIf it works in a local environment, then it's a config thing on the builder, and if it doesn't, then it's something we need to tweak, and I'm not sure which applies.09:02
loolI checked livecd-rootfs, and mksquashfs is run in the chroot09:04
loolYou could check on the porter box, but that's imx5109:04
loolpersia: BTW you might want to allow both for aufs and unionfs in your script09:05
lool(Ideally mount --union and fuse union as well; I think that's all casper supports)09:06
loolI have my own script which OOPSes the kernel  :-(09:06
persiaheh.  Guess we don't want to merge that.09:06
persiaYeah, I was looking at /proc to try to figure out which filesystems the running kernel supported, and do some case statement to use them in some preferred order.09:07
persiaBut I didn't figure it out in 10 minutes, and figured I'd just merge plars' stuff because it does make it simpler for the lucid work.09:07
loolpersia: /proc/filesystems only list the modprobe-d or builtin ones though; aufs and unionfs were often built as modules09:10
persialool: Hrm.  Which means I'd have to do some probing, which gets annoying (and the installer prober warns it may crash the machine)09:11
loolgrmpf09:18
persiaSometimes I'm tempted to try to set up a distributed network of proxies, each attached to a separate (specific) node, and forwarding me only unique entries.09:19
ogra_persia, did anyone ask lamont to update the squashfs-tools version on clementine ?09:27
persiaogra_: No idea.  Is that something that would be different between imx51 and dove?09:28
ogra_no, clementine is the livefs builder for both subarches09:28
persiaHrm.09:30
* persia looks at logs again09:30
ogra_the crontab entry is a bit silly09:30
ogra_seems imx51 is never tried if the other arches have a non zero exit code09:31
persiaAh, that explains why I didn't get any mail about imx5109:33
ogra_the crontab entry connects the two builds with &&09:33
ogra_should be changed to ;09:33
persiaWho can change that?09:34
ogra_me if i'm back from vacation, StevenK, lool, cjwatson etc etc09:37
ogra_wotn help the illegal instruction though09:37
* ogra_ wonders if the -marm actually applied to the current build thats used on clementine09:38
persiaDunno.09:40
persiaSo it's ubuntu-cdimage?09:40
persiaI'll bug someone from the team to look at it soon then.09:40
ogra_its not that important09:45
persiaWell, except that I promised to watch the daily builds this week.09:46
ogra_the illegal instruction will be identical no matter for which target you build09:46
persiaAnd if they aren't working, that needs to be investigated.09:46
ogra_so squashfs-tools needs to be fixed first ... else you just waste buildd time09:46
ogra_if -marm doesnt work, i know that -O0 is supposed to09:47
persiaAnd it's the same group that can check versions on the builder, or do I need a buildd OSA?09:48
ogra_you need lamont09:50
persiaHeh.  OK.09:50
ogra_he pinned the karmic version until last sunday09:50
persiaThat's straightforward enough.09:50
ogra_and had a cronjob to release the pinning09:50
ogra_but probably it just unpinned and didnt upgrade to the version built with -marm09:51
persiaRight.  That needs to be checked, and maybe force-upgraded.09:51
persiaIf it *did* upgrade, it needs to be recompiled with -O009:51
ogra_so first check if the -marm built version is there ... if so, rebuild squashfs-tools with -O009:51
ogra_and if that doesnt work, we're out of ideas and probably need to pin the karmic version for longer09:52
persiaCool.  That's a set of instructions that can be executed :)09:52
persiaWell, except we'd need to do a manual downgrade, etc.09:52
ogra_until someone can investigate deeper09:52
persiaBut yeah.09:52
loolI've updated the crontab09:55
persialool: Cool.  Now I'll get twice as many emails :)09:56
ogra_http://launchpadlibrarian.net/36706216/buildlog_ubuntu-lucid-armel.squashfs_1:4.0-3ubuntu1_FULLYBUILT.txt.gz seems to have -marm09:56
ogra_but -O209:56
persiaogra_: Yep, but it's a question of which is installed.09:56
ogra_right, i just wanted to make sure it built09:56
ogra_being on vacation i dont check such stuff ;)09:56
persiaheh :)09:57
loolNote that it said illegal instructions, not sigbus10:12
lool(sigbus would have said: bus error  ...)10:12
persiaShould a -mthumb2 vs. -marm issue report a bus error?  I thought it was just illegal instructions.10:14
=== ogra_ is now known as ogra
loolpersia: My understanding is that the boards support thumb2 but mishandle unaligned accesses in thumb210:23
lool(They should handle these, but not without a fixed kernel)10:23
persiaThat was my understanding as well10:24
persiaBut the side discussion about squashfs-tools might provide a workaround for the short term (or did I misunderstand? )10:25
ograwe're supposed to get a kernel with mtd support at some point so we can replace the existing broken one10:25
ograor a mtd module for the existing one10:25
persiaMTD?  Aren't those the on-chip flash devices?10:25
ograright10:25
ograthats where the kernel lives in these boards10:25
persiaThat's surely a different issue.10:25
ograand we have no access to the devices10:25
persiaOh, you mean we don't have access to update the kernels?10:25
persiaAh.10:25
ograso we cant reflash with a new kernel atm10:25
ograa driver exists but not for this specifc kernel binary10:26
persiaRIght.10:26
persiaI understand.10:27
persiaSo the issue exists *only* on the buildds?  Should replication testing not happen on other hardware?10:27
persiaI'd think it'd be interesting to fix it elsewhere, and then look at the buildds, but maybe I am behind.10:27
ograit doesnt occur elsewhere10:27
ograand nobody of us has the buildd HW10:27
persiaAh.10:28
ograwell10:28
ogradavidm has10:28
persiaIf that's already been tested, I won't ask for testing :)10:28
ograand i replicated the original issue on his board remotely10:28
ograbut he shut off access to it afterwards10:28
persiaThis is the action item from the 20091208 meeting?10:28
ograyes10:28
persiaOK.  I understand then.10:29
ogradmart has a board as well, but i'm not sure it has the exact same kernel10:29
persiaWell, neither of them seem to be in-channel right now.10:29
ograhe brought up the two fixes (-marm and -O0)10:29
ograand davidm wont be until new year10:29
persiaSo is the squashfs-tools a useful workaround?10:30
ogra?10:30
persiaWon't we still encounter other issues running lucid code on those?10:30
ograwe shouldnt according to dmart10:30
ograeither of the above compile options should fix it10:30
persiaI don't quite understand why *only* mksquashfs is affected10:30
ograkernel module vs userspace10:31
persiaBut I'll try to sort that, and we'll see what breaks next.10:31
persiaOh, it's an ABI thing between the kernel squashfs interface and mksquashfs?10:31
persiaSo it wouldn't affect other stuff?10:31
ograwell, it might if you have userspace things using kernel ABI ...10:32
persiaWould linux-libc-dev be affected?10:32
persiaThat's the only part of the toolchain that I think regularly uses that sort of stuff.10:33
ograhmm, i doubt it, we would have seen more fallout if it had10:33
persiaOK.  Then we have a workaround.10:34
persiaBe a few hours before we have access to look though :(10:34
ograyeah10:34
armin76rabeeh: do i get a present? :)13:23
persiaDoes anyone happen to know how Soyuz handles two source packages in Ubuntu that build a binary package with the same name?13:24
persia-ECHANNEL13:26
asacok chromium building now in qemu-arm-static chroot ;)14:48
ojnasac: how many hours? :)15:04
asacnot sure ... i hope less than native ;)15:09
asacbut maybe i am too optimistic ;)15:10
asacon native it takes 19h or so with more or less fast usb drive (with tests disabled)15:10
asacon babbage 315:10
asacor 16 ... not so sure atm ;)15:11
persiaSurely a big chunk of that is swapping, no?15:11
asacwell... whenever i looked most of CPU was really on action ;)15:11
asacbut swapping is probably a big thing for the linking15:11
ojnwould be interesting to see how long it'd take with distcc:ing over to a non-emulated cross compiler too. :)15:12
persiaI was thinking for some of the libraries as well.  I've been doing a build that includes webkit on a 512MB powerpc box and it keeps swapping out just to make more cache space available.  I'd expect the same for armel.15:12
asaci hope that qemu is in the middle of native cross and all native15:13
persiaDepends on the RAM of the host, really.15:13
ojnqemu system emulation was quite a bit slower than native for me. Hopefully static is quicker.15:13
persiaIf you have lots of RAM, qemu gets pretty fast.15:13
persiaIf you don't, qemu isn't.15:13
asaci had to take one ram module out because of brokenness recently ... so sad ;)15:13
asaci think the idea is to do that in the cloud ;)15:14
persiaojn: at what comparative rates?  For me, qemu seems to perform at about 40% of native speed, which is faster than some of my ARM hardware :)15:14
ojnpersia: I don't remember the details, let's see if I wrote them down15:14
asacit feels faster for me atm15:15
asacnot lightning speed, but i hope that will cut it by half15:15
ojnLooks like the only numbers I actually wrote down was building bash native for x86 vs for arm inside a scratchbox2 environment. I think the qemu system emulation was too slow to even consider, so I didn't keep track of performance.15:16
persiaYeah, compared to cross-compilation it's slow.15:17
persiaBut cross-compilation has it's own collection of issues :)15:17
ojnoh, definitely.15:17
ojndistcc:ing over to a cross compiler is the best of both worlds, really.15:17
persiaWell, sometimes.  I remember someone doing that for test building some stuff for jaunty, and they didn't get caught by a buildchain bug, which was confusing when doing the full native build later.15:18
zumbi_ojn: which script do you use?15:26
ojnzumbi_: for what?15:27
zumbi_distcc:ing over to a cross compiler15:27
loolicecream provides that IIRC15:27
ojnzumbi_: no script needed. Stick a couple of links to distcc named gcc in a directory, make sure it's at the front of your path, then just build. Set DISTCC_HOSTS correspondingly. Caveat: I have yet to try it for a debian package rebuild.15:27
ojnI used it all the time for powerpc kernel builds back in "the days".15:28
zumbi_i need to test this. Do you know if it is compatible with buildbot?15:29
zumbi_Well, i guess it won't be hard to integrate15:29
ojn^^^ see three lines up. :)15:29
ojni.e. no idea15:29
suihkulokkiqemu system emulation compiling could be improved quite a bit by using virtio disk15:30
zumbi_suihkulokki: yes, and it is great :-)15:39
zumbi_suihkulokki: imagine adding distcc to that :)15:41
* zumbi_ has to go..15:41
plarsasac: have you done any benchmarking to see how much faster it is to build things under qemu-arm-static environment?  Does it only benefit you for large packages?16:18
plarsor rather, things that take a long time to build16:18
asacplars: atm, i dont know nothing - first time i use it ;). will get better stats when this is done16:19
plarsasac: what is your host system? how much ram, what cpu, etc?16:22
asaconly 2G ...one module is dead16:22
asacCPU too old to remember ;)16:23
asacso my time will be least possible win16:23
asac&& GYP_GENERATORS=make GYP_DEFINES="target_arch=arm disable_nacl=1 v816:24
asacgood16:24
plarsasac: heh, I had meant to try it out a while back but wasn't able to at the time.  I have an old p4, 3GHz box that I revived recently, may install it on there to mess with16:24
asacplars: its really easy. install qemu-arm-static package ...16:24
asacrun build-arm-chroot ... like debootstrap16:24
asacand then you are done16:24
asacyou can chroot into it and just use it16:24
plarsasac: yeah, I have a link to the wiki page that ogra did for it, looks simple.  I just didn't have a machine to do it on a while back16:24
asackk16:24
asacfta: not sure whats going on, but atm its stilil python dominting my CPU: 14014 asac      20   0  201m  82m 2816 R  100  4.1   3:21.12 python16:29
asacah ... all fine16:29
asacnow make is there ;)16:29
ftaasac, did you take the branch? rev389?16:29
asaci picked latest gyp from daily16:30
asacppa16:30
ftafor chromium16:30
fta?16:30
asaclatest daily version as well from latest bzr16:30
asacyes16:30
asacits16:30
ftaok16:31
asac17:05 < asac> starting over16:31
asac17:06 < asac> 389.16:31
asacso yeah16:31
ftagreat16:31
fta389 packaging revisions \o/ but still nothing in the archives /o\16:33
asacmight be allusion16:33
asacillusion16:33
asacbut it seems to be indeed faster compiling stuff16:33
asacis there a feature that would allow to dump the full compile line just for errors?16:34
asacrather than making everything verbose?16:34
armin76grep16:37
armin76*g*16:37
armin76asac: got tired building on native? :D16:38
plarsasac: getting a lot of qemu: Unsupported syscall: 335 from things when I try to apt-get install packages16:39
asacplars: i get that too ... but seems not to hurt16:40
ftaasac, nope :(16:41
ojnasac: did you hit any snags setting up the qemu-static environment?16:41
asacojn: no. just install qemu-arm-static ... then run build-arm-chroot16:42
plarsojn: I just set one up a bit ago for lucid, very simple16:42
asacand then oyou have a working chroot16:42
asacthat is arm16:42
ojncool16:42
plarsojn: https://wiki.ubuntu.com/ARM/BuildEABIChroot16:42
ojnAh, nice16:42
ojnplars: thanks16:42
plarsojn: getting a nice stream of spam regarding the unsupported syscalls16:42
ojnplars: are you on i386 or x86_64?16:44
plarsojn: i38616:44
ojnhm, there is a 335 syscall on i38616:44
plarswow, that's enough spam to cause my gnome-terminal session to eat ~33% cpu!16:45
ojnwait, is it from qemu or in dmesg on the host?16:45
plarsojn: qemu16:46
asacits from qemu16:46
asacdrops that syscall i guess16:46
asacnot sure what 335 is16:46
ojnI don't see a 335 syscall in current mainline sources for arm. I wonder what's using it.16:46
ojnoh, nevermind. bad grep.16:47
ojn#define __NR_pselect6                   (__NR_SYSCALL_BASE+335)16:47
ojnneeds to be added to qemu16:48
plarsah yes16:48
plarsI think pselect is missing on arm actually16:48
plarsor it used to be at least16:48
ojnwell, it seems to be plumbed up now16:48
plarsojn: not in our kernels yet16:49
plarsCALL(sys_ni_syscall)            /* eventually pselect6 */16:49
ojnplars: but in glibc, it seems. :)16:49
plarsapt-get failed16:50
plarslooks like postfix and bsd-mailx failed to finish installing, don't care thouth16:51
asacplars: what did you install?16:54
asaci only bootstrapped -> worked16:54
plarsasac: pbuilder16:54
plarsbunch of deps16:54
asacthen i installed builddeps16:54
asacand stuff16:54
asachmm16:54
asacpbuilder inside of chroot? ;)16:54
asaci would rather make pbuilder use that16:54
plarsasac: heh, thought it might be interesting to try :)16:54
asactry the other approach16:54
asactweak pbuilder config to use build-emu-static for constructing the chroot16:55
asac(if thats possible)16:55
plarsasac: ah, that would probably be more useful, yet16:55
asaci dont use pbuilder, but feels doable16:55
ftaasac, what the plan to test the perf of chromium & ff? which tools/methodology?16:58
asacfta: those tasks ar assigned to dave16:59
asacthey seem to have something internally for that16:59
ftainternally?17:02
asacfta: dave is from arm18:03
asacso: out/Release/obj.target/webcore/third_party/WebKit/WebCore/dom/Document.o18:04
asacfeels slower than native ;)18:04
plarsasac: I have a test build running on both, building something sizable enough, but not too insane18:12
plarsasac: I have them both building bash, I started on the babbage much earlier and it already finished: real30m21.368s18:13
plarsasac: I wouldn't be too surprised if it was a bit slower, but might be interesting to see if there is good-enough benefit from running them in parallel with distcc18:13
plarsasac: also, this is a pretty slow machine - yes, much faster than any arm proc I have sitting on my desk, but still18:14
ojnplars: hah, i just started the same exact test ;)18:14
plarsojn: no kidding, with bash even?18:15
ojnyep18:15
plarsha18:15
plarsyou'd think we used to work together or something :)18:15
ojn:)18:15
plarsojn: will be interesting to see how your numbers compare to mine, I suspect my time from the babbage is not great (running off a 16G usb stick at the moment)18:16
ojngrmbl, -j<largenumber> doesn't seem to work with dpkg-buildpackage though.18:16
ojnI had 34m real time here, running off of internal flash on the pegatron18:16
plarsojn: I was just about to ask asac if he knew how to make debuild call make with -j18:16
ojnactually, 34m54s.18:16
persiaojn: DEB_BUILD_OPTIONS=parallel=${n}18:16
plarsojn: ok, so not too much different from mine anyway18:16
plarspersia: thanks!18:17
persia(you need a debian/rules that supports that)18:17
ojndpkg-buildpackage has a -j switch, but it'll fail due to some of the jobs trying to get into the directory before it's created. :)18:17
persiaIt's not well tested :)18:17
ojnI'll do an apples-to-apples and do single-thread first.18:17
* ojn watches 3.5 cores being idle on his $600 nehalem box. :)18:17
persiaGrueMaster: I had a try at an inital sort for mobile-lucid-arm-suspend-resume-testplan : http://paste.ubuntu.com/343650/ What do you think?18:19
GrueMasterLooks reasonable.  I don't know enough about eSATA, but it might be more of a middle option.  Isin't it like an external USB drive (separate power from the system, but able to take sleep/wake commands)?18:21
persiaNo, it draws power from the system.18:21
persiaBut it's not nearly as common as USB.18:22
persiaI suppose there are also eSATA enclosures with separate power, but they aren't so interesting :)18:22
persiaMy basic thought was as follows:18:22
GrueMasterAre you sure?  Every motherboard I have in recent years has come with a chassis eSATA connector, which is just a sata extension cable.18:22
persiaBase is the stuff that we can't function without18:22
GrueMasterright.18:23
persiaStandard is the stuff that we expect to find in most hardware18:23
persiaand Extra is the rest of the stuff.18:23
persiaI expect that we'll find Extra stuff in most hardware, but I don't expect to find all of it in anything.18:23
GrueMastereSATA should be middle then.  Most nettop systems come with eSATA now.18:23
persiaWhereas I could easily imagine something with all of Standard.18:23
persiaOK.  Handhelds don't.18:24
persia(too much power draw)18:24
GrueMasterTrue.  Nor do laptops or netbooks.18:24
persiaWell, some laptops do18:24
persiaBut those are the big ones :)18:24
GrueMasterYou mean the next gen luggables?18:25
GrueMaster:P18:25
persiaWell, I've seen people carrying eSATA enabled laptops to conferences, but yeah :)18:26
GrueMasterI have a friend with one of those beasts.  17" wide screen, dual core nvidia graphics, Core2Quad cpu 4G mem, 1T drive, blue-ray, etc.18:26
persiaYeah.18:26
persiaSo Extra?18:26
asacplars: any results yet?18:26
asac(comparison)18:26
persiaGrueMaster: So anyway, having sorted the list, I have no idea what needs to happen to solve the "Decide and document testcase priorities; sort into three classes; base, standard, extra: TODO" workitem.18:27
persiaDoes it belong on some specific wiki page?  Is more organisation needed?18:27
* persia suspects "Create suspend resume testplan page that will document how to run suspend resume test" should have been done first, but was at a bit of a loss18:28
GrueMasterThat I don't know.  I guess we sort the whiteboard with what you have and document each test case on wiki.18:28
persiaI can sort the whilteboard.18:28
persiaDo you know the Testing wiki hierarchy?18:29
persiaI just don't know where it belongs18:29
persia(other than somewhere under https://wiki.ubuntu.com/Testing )18:29
GrueMasterNot really.  I can figure it out though.18:29
persiaCool.  I'll sort the blueprint.  Let me know if you need me to populate a wiki page once you have a URL18:30
GrueMasterWasn't someone generating a wiki map?18:30
persiaheh.  But I'm only doing that for some namespaces, and /Testing/ isn't one of them18:30
plarsasac: not more than what I gave you earlier18:33
plarsso clearly, going to be MUCH slower on my system to use the qemu chroot18:33
plarspersia, GrueMaster: it was my understanding that they were trying to transition all testcases from w.u.c/Testing to testcases.qa.ubuntu.com18:36
GrueMasterplars: I may need your help then to publish there.  I can do the writeup, but I may need you to post it if I don't have the correct privileges.18:37
plarsGrueMaster: I don't think I had to get any special privileges for posting on that wiki, you just need to be logged in18:38
GrueMasterok.  I'll check when I'm ready to post.18:38
persiatestcases.qa.ubuntu.com might require an ubuntu-qa login though (the separate, parallel, single-sign-on exists for historical reasons)18:39
plarsah yes, I vaguely remember signing up for that ID18:41
plarsin any case GrueMaster, if you have problems posting to it, let us know, or ask about it somewhere like #ubuntu-testing.  You need to have access to that wiki if you don't18:42
GrueMasterok, will do.18:42
plarsI suspect you can just register for the account though, and it will let you post18:42
persiaSHould do.18:43
persiamobile-lucid-arm-suspend-resume-testplan whiteboard updated with sorting.18:46
plarsah, my qemu-arm-static build of bash is nearing the end finally18:57
ojnplars: Mine took _exactly_ as long on qemu as on native hardware without a -j option19:00
plarsojn: that's encouraging19:00
plarsmine will be longer by a fair margin for certain, but on pretty slow hardware19:00
plarsojn: did you kick it off with distcc also?19:01
ojnno, not at this time19:01
ojnI might give that a go next; I'm waiting on fedex though so once I get that I will be 100% busy with new board work. :)19:02
jetienneq. i would like to have an hw running arm and ubuntu. ideally a cortex a8 or a9. where i can find such hw ?19:02
ojnjetienne: I don't think you can buy a9 hardware anywhere today. Buying a freescale babbage eval board is the only supported platform to date, but some are using ubuntu on a beagleboard (much much cheaper)19:05
ojnsomeone from canonical should correct me if I'm wrong though. :)19:05
persiaThere's also some retail hardware that works, but again, it runs into the kernel issue.19:05
plarsthe ubuntu arm images will not support beagleboard, so you'd need your own kernel, but could run ubuntu packages on top of that.  Certainly I've heard of people doing it19:06
jetiennei look for a reference, something i can buy. this is mostly to test the perf of arm on linux19:06
jetienneplars: ok so no beagleboard. i need it to be easy to install19:06
plarsalso, if you really just want low cost, the $99 sheevaplugs run something based on Jaunty iirc19:07
jetiennecost is no issue19:07
persiajetienne: What sort of platform do you want?19:07
jetiennebut already installed is ultra good19:07
* persia is happy with the Sharp Netwalker, but it does need a custom kernel to run Ubuntu other than 9.0419:07
ojnGenesi Efika MX comes preinstalled with jaunty, but it's not officially supported.19:07
jetiennepersia: im part of a startup doing softward and hardward. the hw will be outsourced. currently im just evaluating the need19:07
plarsdev board from freescale will run jaunty, or also let you run lucid images that are already available19:08
persiajetienne: OK.  Do you want a dev board or a handheld?19:08
ojnjetienne: sounds like you should talk to the SoC vendors about getting some eval boards. Marvell Dove and Freescale Bababge officially support ubuntu, none of the others do.19:08
ojn(but with your own kernel most of them can run it)19:08
jetiennedev board is ok, handheld too19:08
jetienneojn: SoC = ?19:09
persiaFor a dev board, FSL sells babbage, which has kernel support in Ubuntu.19:09
ojnjetienne: Oh boy. SoC == system on a chip19:09
jetiennehow much is it ?19:09
persiaFor a handheld, Sharp sells the Netwalker, which is the same CPU, but needs a different kernel because of how ARM kernels work today.19:09
persiaI don't know of anything else available retail (but someone could correct me)19:09
ojnpersia: Ha, OMAP kernels don't have that issue. :)19:10
ojnthey can be multi-platform.19:10
persiaojn: Yes they do, just less so.  Try running the same kernel on OMAP1 and OMAP319:10
jetiennepersia: how much did you pay for the board19:10
ojnpersia: sure, but within a family they work. Freescale doesn't even support it that far, do they?19:10
persiajetienne: I paid 44800 yen for a Netwalker.  I don't have a babbage: you'd have to ask someone else19:11
* persia looks for a website19:11
GrueMasterThere is also the Beagle Board (http://beagleboard.org).  While not directly supported by Ubuntu (kernel), there is a help page for getting Ubuntu installed.19:11
GrueMasterIt uses the OMAP4 chip (I believe).  Still ARMv7.19:12
jetiennepersia: so around 400euro ? wow ultra cheap19:13
persiajetienne: If you get conics to import it, they'll charge a bit of a premium, but yeah.19:13
persiajetienne: Sorry.  I remember representatives from freescale saying the Babbage was available retail at UDS, but I can't figure out how to order it from their site.  Maybe you'll have better luck.19:16
plarshahah19:16
plarsreal107m58.740s19:16
plarsojn, asac ^19:16
ojnGrueMaster: Beagleboard has OMAP 3530, Cortex A819:16
jetienneGrueMaster: http://beagleboard.org/hardware19:16
ojnplars: Wow. I guess my machine is quite a bit faster19:16
ojn:)19:17
GrueMasterright.19:17
plarsojn: yeah, I told you this was an old box19:17
jetiennepersia: ok thanks19:17
ojn:)19:17
jetienneq. i read a lot of doc on ubuntu-arm, they talks about "lucid"... what is this ?19:18
armin76:D19:19
persiahttps://wiki.ubuntu.com/ is a good place to start for that sort of question.19:19
persiaIt's the code name for what will become the Ubuntu 10.04 release in April.19:19
persiahttps://wiki.ubuntu.com/LucidLynx has limited information along those lines19:20
jetiennethanks19:20
jetienneall those boxes got very limited ram (i have seen 128mbyte, or 512mbyte), what is the motivation behind that ?19:22
jetienneto save battery life ? or more to save hw cost19:22
ojnjetienne: most of the embeddded ARM chips use package-on-package memory (LPDDR), and while there might be options to get 1GB that way, it's usually very expensive.19:24
ojnSome chips can use regular DDR2 memory, and then you can go to larger sizes. I think babbage can, Marvell Dove should be able to as well.19:24
jetienneojn: thanks19:26
jetiennei think for my testing, i could use a modern phone, or archos hw... can they run ubuntu ?19:27
ojnjetienne: if you have hardware platform selection to do, I would recommend finding someone to assist you with it that knows about these things already. There are many details you want to consider when selecting a hardware base for your product.19:27
persiaDefinitely!19:28
jetienneojn: we will outsource it to a company we know in cambridge19:28
persiaGrabbing some random thing to examine the software stack is a completely different activity than selecting the basis of a product.19:28
jetienneojn: currently im just doing some prospective research19:29
ojnjetienne: sounds like you should have them assist you in the selection too, if you don't have any of that knowledge in-house.19:29
ojnjetienne: ok. :)19:29
jetiennefor example i need to know if it is feasible to do video transcoding on a modern arm, or do i need a intel cpu, or a special chip to encode mp419:29
ojnon arm you most definitely need to use the hardware offload assist to do it in a reasonable speed19:30
ojnubuntu doesn't provide that on any platform, as far as I know19:30
persiaDepending on the solution, that might be a separate chip, or it might be onboard (based on the block diagrams I've seen)19:30
jetienneojn: yep, the ubuntu on arm is only there to know if i can use arm chip to encode19:30
persiaojn: That's mostly a lack of available software that meets the licensing requirements.19:30
ojnjetienne: I can answer that right now. :-)19:31
persiaWe'd be *very* happy to provide that.19:31
jetiennethe personn doing x264, a well known software video encoder, is telling me it is possible to encode 320x240 in real time on a cortex a819:31
ojnpersia: sure, I'm just saying it's not included. The codec licensing paperwork is a nightmare on most platforms.19:31
persiaIndeed.19:31
ojnjetienne: ah, that low resolution might be possible. Sorry, I was thinking higher bandwidths.19:32
persiaAnd so far, nobody has submitted code to our nice public collaborative code review site for package inclusion :)19:32
jetiennei dont need real time. but i need to be sure this is actually possible19:32
jetienneojn: thats ok19:32
jetiennepersia: you are in canonical19:32
jetienne?19:32
jetiennein our case, we will use specific IP lawyers to get codec license, and i do agree this is a nightmare19:33
jetiennehttp://arstechnica.com/gadgets/news/2009/10/nokias-n900-is-an-important-step-forward-for-mobile-linux.ars <- ok i will get the company to get this toy :)19:34
jetiennethanks a lot for your helps guy19:34
persiajetienne: No promises that Ubuntu can be run on there.  I know several people tried with N810s and had various issues.19:34
jetiennei know the author behind strigi, he is paid to code on one of those nokia19:36
jetiennei will bug him to get running what i wnat :)19:36
ojnplars: halting bulding experiments, got new hardware toys. :)19:45
plarsojn: merry Christmas :)19:46
plarsojn: I also tested hello, it took about 4x as long on my qemu-arm-static environment as it took on my babbage :(19:46
plars1m5.991s vs 4m19.461s19:47
* plars needs some new hardware19:47
ojnplars: I was monitoring these for a while, snagged one when it showed up for $609. http://outlet.us.dell.com/ARBOnlineSales/topics/global.aspx/arb/online/en/InventorySearch?c=us&cs=22&l=en&s=dfh19:49
ojnuh, that link didn't work. XPS 435 mini tower.19:50
plarsojn: wow, not bad, cheapest at the moment is $73919:51
ojnyeah. they go up and down a bit depending on memory configs, etc.19:52
ojnmine was a bit low on memory, but I picked up some at fry's to fix that.19:52
ojnHm, is there a bootargs option to make init provide debug info on what it is spawning?23:14
persiaFor which release?23:15
ojnkarmic (upstart)23:15
ojnOh, there seems to be a --debug.23:15
persiaKill that, and add quiet (see /usr/share/doc/upstart/README.Debian.gx for more)23:16
=== jldugger is now known as pwnguin

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!