/srv/irclogs.ubuntu.com/2009/09/22/#ubuntu-arm.txt

neonfreonwg 202:23
gaspa has anyone ever dealt with mmap and virt_to_phys address translation on arm? ogra?10:17
ograwhats your prob ?10:18
gaspabriefly this: http://linux.omap.com/pipermail/davinci-linux-open-source/2009-September/016102.html10:19
gaspawell, not so brief... :P10:19
gaspaogra: I think virt_to_phys behave differently in i386 and arm, but I'm not finding what.10:20
ograi know the minimal addresses are different on arm vs intel ... but that shouldnt stop you from using it10:22
gaspaI think so, but I got only zeroes.10:22
gaspadid you read the link above?10:23
loolgaspa: Hmm isn't the kmalloc area already remapped?11:26
loolgaspa: Anyway you want to bring this up on some kernel chan; we're not so much into kernel dev here11:26
ali1234gaspa: that sounds exactly like what the android pmem driver does11:28
gaspalool: I tried also calling pfn_range with kmalloc_area directly...(if that's what you meant... otherwise I didn't catch what you said ;) )11:30
gaspaali1234: I don't know at all android, have you any pointer to this pmem driver?11:30
ali1234"Android/PMEM: The PMEM (physical memory) driver is used to provide contiguous physical memory regions to userspace libraries that interact with the digital signal processor (DSP) and other hardware that cannot cope with scatter-gather."11:32
ali1234source: http://android.git.kernel.org/?p=kernel/common.git;a=blob;f=drivers/misc/pmem.c;h=44c0ae8515bac186a7a0a3060204b29ef7b21130;hb=f92ea0328d3a0bb5386d52fd68ae4e10ca1ebce611:33
gaspaali1234: seen, but I'm not facing with device memories, it's just a test to learn using mmap so i'm currently using normal RAM11:34
gaspaanyway, i'll take a look. ;)11:34
ali1234ok, well, the memory address by pmem is like any other, it just happens to be shared with devices so outside what linux considers general purpose ram11:35
ali1234ultimately it just remaps it and makes it into a device file in userspace11:36
ali1234it's a bit of a hack to make obfuscated userspace drivers work11:36
ali1234there's the /dev/mem driver too which does the same thing but on the whole of the address space11:37
ali1234that one is probably simpler/nicer11:38
gaspayes, in fact my code went from that (indirectly)11:38
gaspaali1234: BTW: thanks!11:42
gaspapmem seems interesting, but for other purposes of mine ;)11:42
ograeggonlea, i'm just working on rootstock12:06
looleggonlea: Hey12:06
ograeggonlea, it needs some massive changes to work with the new upstart ... init=/bin/bash doesnt work anymore12:06
eggonleahi12:06
ograwell, it does work, but you cant start any services anymore, i'm working since this morning to find a proper way around that12:07
eggonleado you mean the rootfs is created but the /etc/init.d/udev start should be removed12:07
ograthe bugfix for teh networking stuff is already in my local branch but doesnt make sense to push unless the rest is fixed12:07
ogranew upstart prevents /etc/init.d/xxxx scripts from beiong executed directly12:08
ograyou need to use the service utility for that, which only works if upstart is running12:08
eggonleaso... will upstart work with "chroot"?12:09
ograbut since rootstock uses something similar to init=/bin/bash upstart isnt running and the service utility doesnt find the upstart soocket to send its messages to12:09
ograno, and it doesnt have to12:09
ograit needs to work with a VM12:09
ograand the rootstock installer needs to be spawned by upstart12:10
ograso i'm trying to convert everything to it atm12:10
eggonleaI see. so rootstock/qemu should be ok after fixing the init scripts between first_run and second_run12:10
ograwell, the second run actually has to become an upstart job12:11
ograand thats what i work on12:11
eggonleaplease let me know if I could contribute on this.12:11
ograwell, do you know much about upstart jobs ?12:11
* ogra is in the middle of rewriting the installer12:12
eggonleaknow some high level configuration but no deep understanding into the code12:12
* ogra fires up another test, cross your fingers ;)12:14
eggonleaplease12:15
eggonleamy pleasure12:15
ograi switched the first stage completely to qemu-arm-static btw ... it takes less than 10min now to get to the VM12:15
ograinstead of 30+12:15
eggonleait never costs so much time before.12:15
eggonleaI think ~10 minutes for ubuntu-minimum before12:16
ograyou mean it doesnt sit eternally in "I: Configuring core packages..." after it switched to VM on your side ?12:16
ograit surely does for everyone else12:17
ograbut if the whole thing took 10min for you before it will only take 2 now ;)12:17
eggonlealet me see if I could find the log back.12:17
eggonleacool12:19
eggonleaI cannot get the log back.12:19
eggonleaI'll record them next time12:19
ograafter it switched to VM it did a lot of debootstrap processing where it got very slow ... and just sat there12:20
ograi had a lot of user complaints about that12:20
ograi moved all of debootstrap out of the VM and it operates a lot faster now12:20
eggonleaI have a question: could we leave the second part to run on the real board intead of qemu?12:20
ograi a later version, not for karmic12:21
eggonleawhy? any tech obstacle?12:21
ogracould you file a whishlist bug about that ?12:22
ograwe're past feature freeze12:22
eggonleagot it.12:22
ograi wont add new features to it now12:22
ograonly fix up existing stuff and add speedups12:22
eggonleaSure, I'll do that.12:23
eggonleaso, when do you expect to have a workable one?12:23
ograi hope later today12:23
eggonleawhat a good news!12:24
ograupstart doesnt seem to be happy to run in a VM though ... i need to find out why first12:24
eggonleathanks and wait... no more question to interrupt your work, then.12:25
eggonleaagain, if you have anything need a man to test or debug, please let me know.12:25
ograwill do, thanks a lot for the offer :)12:25
eggonleaor, you could drop your engineering version to me.12:26
eggonleacreate a branch on LP, send it on ubuntu-arm maillist or to aawlbt@gmail.com directly.12:27
ograhttp://paste.ubuntu.com/275799/12:29
ograi'm in the middle of reworking it, it wont work, but you can take a look12:29
eggonleagot it.12:30
ogranote that the new version needs qemu-arm-static installed12:33
* ogra goes to find some lunch while another test runs12:34
* eggonlea go home for supper12:47
gaspaali1234, lool: solved... aerhm. I was mmap'ing with PROT_NONE ;)14:21
ograI: ARM rootfs created as /var/build/rootstock-fixing/armel-rootfs-200909221611.tgz15:24
ograI: Cleaning up...15:24
ogra.....15:24
ograI: A logfile was saved as /var/build/rootstock-fixing/rootstock-200909221611.log15:24
ograI: done ...15:24
ograWOHOO !15:24
loologra: WTF is this -Nqemu-arm-static stuff??15:31
ogra-N ?15:32
ograno idea, i surely didnt add any -N15:32
ograi guess kirkland added that the statci branch doesnt exist in the other builds, its a second pass with a copy of the sourcetree15:36
loolIt's ridiculous15:37
ograremove it if it doesnt break the build15:37
ograyou cant build -static in the same pass as nonstatic15:37
loologra: The rules are absolutely horrid16:10
loolI've never seen so many policy violations and programming errors in the same rules16:10
loolStarting with binary: binary-static binary-indep binary-arch16:10
loolGeez clean depends on config.status16:17
MartynWhere can I find the launchpad files that generate the karmic-desktop-armel+dove files?16:22
Martynfound 'em16:26
armin76Martyn: :D16:35
Martynarmin76 : generating an image for pbx-a9 is going to be fun16:36
Martyn"fun" in the "I think I need a rusty 10 foot stake shoved up ..."16:36
Martynway16:36
ogralool, i refuse any responsibility for clean ! :) but you can blame me for everything with -static16:51
ograsigh, i really wonder how livecd-rootfs still works with the upstart changes16:52
ogracups is definately unhappy about Failed to connect to socket /com/ubuntu/upstart: Connection refused16:53
loolWhat does it changes for livecd-rootfs?16:57
loolHmm probably we should disable start/stop/service there16:57
ograi'm comparing the code atm16:58
ograthere is not a single change from Keybuk16:58
ograi wonder why it works16:58
ograoh, it diverts usr/sbin/invoke-rc.d17:01
loolYes17:03
loolYou dont?17:03
ogranope17:03
ograi run a VM17:03
ograup to now i didnt divert anything ... there was never a need to17:04
ograi dont see it touching start-stop-daemon though17:05
ogra(i know it did once in the past)17:05
loolI thought you were using a chroot not a vm17:05
ograoriginal rootstock: chroot for first debootstrap stage, VM for rest17:06
ogranew rootstock: chroot for complete debootstrap stage, VM for rest17:06
ografuture rootstock (if the binfmt issues are fixed): chroot for complete build17:07
armin76Martyn: gimme ssh access to the babbage board plz :P17:13
Martynarmin76: Sure, give me a couple hours though17:23
Martynarmin76: I'm in the middle of doing a new Karmic buildroot on the babbage and on the PBX-a917:24
Martynarmin76: Do you have my IM?17:24
Martynarmin76: Babbage v1 is okay?17:25
armin76Martyn: armv7 is okay :)17:26
armin76how much space?17:26
ogra5x5" :P17:27
armin76in fact i wasn't expceting you would say yes17:27
* armin76 is just happy with ssh+being able to chroot17:27
* armin76 blames ogra for martyn running away17:38
ograsure sure, blame me !17:39
ogra.oO(was martyn -static ?)17:39
loologra: Why did you name the package qemu-arm-static instead of qemu-static?18:04
ograbecause its arm onyl18:05
ogra*only18:05
ograand needs its own postinst18:05
loolBut we could easily have multiple binaries in there18:05
ograwe shouldnt18:05
loolWhy not?18:05
loolqemu-system-arm is in qemu-kvm-extras18:05
loolAmong others18:06
ograi personally dont want ppc or mips if i need just an armel chroot18:06
ograyeah, i dont like that either18:06
loolYou have qemu-system-mips64el and qemu-system-ppc64 and a bunch others with qemu-kvm-extras18:06
ograi'd prefer a binary for every arch18:06
ograbut qemu system doesnt need the binfmt stuff18:06
loolThat would be huge18:06
ogranor does it need the sysctl settings18:06
lool15 binary packages18:06
ografine18:07
loolThat's huge18:07
ograi wouldnt mind 15 binary packages if i can then install only for one arch i care for18:07
loolThe qemu-arm-static approach isn't consistent with qemu-kvm-extras18:07
ograi created it before qemu-kvm was in the archive18:07
loolbah it was the same before the qemu-kvm package18:08
ograi just tried to merge the -static stuff into the existing package on kirklands request18:08
ograin any case there exists no -static flavour thats integrated for karmic18:08
loolI'm saying it's inconsistent to have one package per target for the static flavour and one package for all targets in the shared flavour18:08
ograapart from arm18:09
loolSo?18:09
loolIt's still inconsistent18:09
ograso the naming is fine for the moment18:09
loologra: What would it change for you if it was named qemu-static?18:09
loolOr qemu-kvm-static18:09
ograbeyond i highly object having to have ppc sysctl and binfmt mangling installed just because i want arm18:09
loolYou wont since it only has arm18:10
ograthey might even conflict18:10
ograppc might be added in L18:10
ograi know at least one person trying to work out support for mips and ppc in debian over the next months18:10
loolYeah so your argument that there exists no -static flavour thats integrated for karmic apart from arm is moot  :-)18:11
ograno18:11
ograthe package adds binfmt handlers and sysctl settings18:11
ograits different from the non static version18:11
loolYes18:11
ograthats why static binaries should have one package per arch18:12
ograand the proper conflict/breaks settings if needed18:12
loolDoes debian/binfmt-arm correspond to ARM binaries?18:12
ograyes18:12
loolDoes it match e.g. ppc binaries18:13
ograno18:13
loolSo I dont see the issue you mention18:13
ograthe magic is different18:13
ograsysctl is the issue18:13
ograbinfmt is just ugly18:13
loolIsn't there a common sysctl we can find for other arches?18:14
ograi dont want to have a ton binfmt handlers installed if i only need one arch ... as i said before18:14
loolI dont want to have 15 packages cluttering Packages.gz18:14
ograif it comes to sysctl there might even be conflicting settings18:14
ograyou wont convince me :) if you feel its needed to change the package, do it ...18:14
ograi'll live with it ...18:15
loolOk18:15
ograbut you wont convince me :)18:15
ograjust tell me how it's called in the end so i can change the arm docs accordingly18:15
loolI personally prefer keeping things consistent between static and shared18:15
loolIf it creates issue we can look into them and see if we need to split18:15
ograits a different purpose18:15
loolI think it's premature optimisation to name the package -arm and doesn't match the shared package; I personally find it misleading that qemu-static-arm changes system wide settings in a way not specific to ARM stuff18:16
loolSuch as the sysctl knob you mention18:16
* ogra wonders about "udevd[392]: specified group 'fuse' unknown" in the recent rootstock test18:16
ograright, kees and i tried a better approach18:17
ograbut the support doesnt work yet18:17
ograso i had to resort to sysctl18:17
ograthere are plans for a per binary solution in the security team18:17
ograjust not there ydet18:17
ogra(see changelog, i went back and forth through several solutions)18:18
ogra(guided by kees)18:18
ograhmm, why dont i see the fuse errors in livefs buildlogs ?18:20
ograthere is nothing in livecd-rootfs that would touch /etc/group18:21
ograUnpacking linux-headers-2.6.31-101-imx51 (from .../linux-headers-2.6.31-101-imx51_2.6.31-101.8_armel.deb) ...18:24
ograhrm18:24
ograwhy is that pulled in by ubuntu-desktop18:24
* ogra sees "apt-get -y --purge remove $HEADERPACKAGES </dev/null || true" in livecd-rootfs18:26
ograoh my18:26
ograso many hacks18:26
ograsigh, rootstock is so time consuming18:58

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