Back to Home / #uml / 2007 / 07 / Prev Day | Next Day
#uml IRC Logs for 2007-07-06

---Logopened Fri Jul 06 00:00:28 2007
00:28|-|Urgleflogue [~plamen@87-126-143-181.btc-net.bg] has quit [Remote host closed the connection]
01:04|-|_horst [~horst@a89-182-8-58.net-htp.de] has joined #uml
02:52|-|Netsplit charon.oftc.net <-> reticulum.oftc.net quits: Hunger
02:57|-|Hunger [Hunger.hu@Hunger.hu] has joined #uml
06:44|-|Ancalagon [~PtitKosmi@215.208-78-194.adsl-fix.skynet.be] has joined #uml
07:06|-|kokoko1 [~Slacker@203.148.65.8] has joined #uml
08:58|-|ram [~ram@pool-71-117-236-168.ptldor.fios.verizon.net] has joined #uml
09:16|-|Ancalagon [~PtitKosmi@215.208-78-194.adsl-fix.skynet.be] has quit [Quit: ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007050800]]
09:46|-|_horst [~horst@a89-182-8-58.net-htp.de] has quit [Remote host closed the connection]
09:47|-|_horst [~horst@a89-182-8-58.net-htp.de] has joined #uml
10:24|-|mgross [~mgross@pool-71-245-108-184.ptldor.fios.verizon.net] has joined #uml
10:26|-|hfb [~hfb@pool-72-87-254-188.lsanca.dsl-w.verizon.net] has joined #uml
10:27<mgross>I'm trying to boot a x86_64 uml kernel 2.6.22rc7 and passing a hand built root fs (ext3, with a copy of a 32bit NFS root image I built from a LFS CD last year) I'm getting a kernel pannic: EXT3-fs: mounted filesystem with ordered data mode. VFS: Mounted root (ext3 filesystem) readonly. Kernel panic - not syncing: No init found. Try passing init= option to kernel
10:28<mgross>any hints?
10:29|-|hfb [~hfb@pool-72-87-254-188.lsanca.dsl-w.verizon.net] has left #uml []
10:29<_horst>did you supply a root= option ?
10:30<mgross>mgross@mgross:~/_dev/UML/linux-2.6.22-rc7$ ./linux mem=513M ubda=/mnt/embedded/new_filesystem
10:30<_horst>try: ./linux ubda=<file> root=/dev/ubda
10:31<mgross>mgross@mgross:~/_dev/UML/linux-2.6.22-rc7$ ./linux mem=513M ubda=/mnt/embedded/new_filesystem root=/dev/ubda
10:31<mgross>no joy
10:32<mgross>perhaps it doesn't want to exececute a 32 bit init bin?
10:33<_horst>does the kernel have CONFIG_IA32_EMULATION ?
10:34<mgross>no
10:35<_horst>but it is required to run 32-bit apps on a 64bit kernel
10:36<mgross>I'm looking for it in the .config but its not there and I'm not finding anything under menuconfig either. Where would I find it?
10:37<_horst>"/Executable file formats/IA32 Emulation"
10:40<peterz>iirc uml doesn't support 32 on 64
10:41<mgross>ok, I can build a i386 kernel usign make ARCH=i386, but how can I build a i386 uml kernel? make ARCH=uml SUBARCH=i386?
10:43<mgross>oh, why doe the kernel make pause after: CC arch/um/sys-i386/user-offsets.s
10:43<mgross>? Waiting for a return key press?
10:44<peterz>ARCH=um
10:44<peterz>not =uml
10:44<peterz>but yeah, that should work
10:45<peterz>although the SUBARCH thing sometimes fails on me
10:45<mgross>yeah, I saw that after I typed it. it failes for me too : include/linux/sched.h:339: error: field ‘mmap_sem’ has incomplete type
10:46<mgross>I'll chroot into the LFS image and build the UML kernrl using those tools maybe that will work.
10:49<mgross>well that didn't work too well either.
10:51<mgross>where can I get a x86_64 uml root fs to use?
10:53<mgross>I just want to do some kernel development and uml will be the quickest way to go for fiddling witht he driver model /sysfs code.
11:07<mgross>What is the latest kernel,org kernel I should try building ARCH=um?
11:07<mgross>because now I'm getting : arch/um/kernel/signal.c:51: error: ‘HOST_RAX’ undeclared (first use in this function)
11:14|-|jdike [~jdike@pool-71-174-247-82.bstnma.fios.verizon.net] has joined #uml
11:14<jdike>Hi guys
11:14[~]jdike back from CA
11:15<mgross>I"m going to try :http://uml.nagafix.co.uk/Debian-3.1/Debian-3.1-AMD64-root_fs.bz2
11:21<mgross>why the pause in the early build of the make ARCH=um ? It stops every time waiting for a return key press.
11:27<jdike>there's no pause there
11:28<mgross>It pauses for me :( host=ubuntu6.10 x86_64.
11:29<mgross>oh well.
11:29<mgross>Say I just tryed the above debiean image and I get a new crash.
11:30<mgross>./linux ubda=/mnt/embedded/Debian-3.1-AMD64-root_fs
11:30<mgross> give me: EXT3-fs: mounted filesystem with ordered data mode.
11:30<mgross>VFS: Mounted root (ext3 filesystem) readonly.
11:30<mgross>Stub registers -
11:30<mgross> 0 - 7f7fcd2af8
11:30<mgross> 1 - 7f7fcd2790
11:30<mgross> bla bla bla : Kernel panic - not syncing: handle_trap - failed to wait at end of syscall, errno = 0, status = 2943
11:30<mgross>any thought?
11:33<jdike>paste the boot log to pastebin or something
11:33<jdike>what version of UML?
11:35<mgross>2.6.22-rc7 from kernel.org. is there an additional patch I need?
11:35<jdike>nope
11:35<jdike>should be OK
11:35<mgross>whate is pastebin?
11:36<jdike>pastebin.com
11:40<mgross>its being slow :(
11:40<jdike>rafb.net is another one
11:42<mgross> http://rafb.net/p/EkBvNH78.html
11:57<jdike>rc7 is fine on my x86_64 box
12:00<jdike>what's the host?
12:09<mgross>ubuntu 6.10
12:10<mgross>Linux mgross 2.6.15-28-amd64-generic #1 SMP PREEMPT Thu May 10 09:46:40 UTC 2007 x86_64 GNU/Linux
12:14<kokoko1>Hi all
12:27|-|kos_tom [~thomas@humanoidz.org] has joined #uml
12:27<jdike>mgross, is it possible to upgrade the host to something more recent?
12:31<mgross>Can I get by with just updating the host kernel to 2.6.22-rc7?
12:32<jdike>yup
12:32<mgross>ok, any config switches I should make sure to use?
12:32<jdike>although 2.6.21 should be fine if you don't want to experiment with a -rc
12:32<jdike>nope
12:32<jdike>just use your standard config
12:33<jdike>BTW, by "host", I meant host kernel
12:33<jdike>not the whole distro
12:34<mgross>ah. updating my whole distro wan't going to happen.
12:35<mgross>I'll let you know how it goes in about 30 min.
13:29|-|mgross [~mgross@pool-71-245-108-184.ptldor.fios.verizon.net] has quit [Quit: Leaving]
14:17|-|mgross [~mgross@pool-71-245-108-184.ptldor.fios.verizon.net] has joined #uml
14:18<mgross>booting 2.6.22rc7 on the host fixed my problems. (although my host sdb, and sdc drivers are not mounting now....)
14:19<mgross>Thanks for the help. What was special of 2.6.22rc7 vrs 2.6.15-ubuntu6.10 that made things work? Its kindof fishy for a user modeprocess to depend so much on the kernel it running on....
14:21<jdike>host bugs
14:21<jdike>you can't count on processes to run correctly in the presence of bugs in the kernel
14:22<mgross>That make sence. I was just wondering what types of bugs you exposed. If its a big list then don't bother answering. I was just wondering.
14:30<jdike>ptrace
14:30<jdike>dark corners of ptrace and associated stuff in entry.S
14:50<mgross>Thanks for the help. Now if I can get all three of my HD's to show up on my host running 2.6.22rc7 I'll be happy.
14:50|-|mgross [~mgross@pool-71-245-108-184.ptldor.fios.verizon.net] has quit [Quit: Leaving]
14:55<kokoko1>jdike, what's up?
14:55<jdike>getting back to work after a long weekend
14:56<kokoko1>ah right resuming working on smp ?
14:58<kokoko1>[PATCH] UML - Improve host PTRACE_SYSEMU check , 2.6.21.5
14:59<jdike>that fixes UML on FC6
14:59<kokoko1>thanks God we don't have these hangups
14:59<kokoko1>yes we are running UMLs on fc6
15:02<jdike>well, the utrace in FC6 breaks PTRACE_SYSEMU
15:04<kokoko1>jdike, where is UML now compare to host? getting better ?
15:04<kokoko1>I thinks xen is 75%?
15:04<jdike>you mean performance?
15:04<kokoko1>yes
15:05<jdike>here with a kernel build, UML is 70-75% native
15:05<jdike>depending somewhat on the host kernel
15:05<kokoko1>cool that's where xen is right?
15:05<jdike>pretty much, from what I hear
15:06<kokoko1>jdike, i thinks you should be looking for some lobbiest in RH camp :)
15:06<kokoko1>they are only propagating xen, which in turn attracts clients like my boss
15:07<jdike>true, but they're not going to be propagating Xen for long
15:07<kokoko1>recently got new host with all my struggle for uml he asked me to go for xen.
15:07<kokoko1>yes, that's what i heard from you and some article , rh is weighing kvm
15:07<jdike>yup
15:08<kokoko1>jdike, why not Intel supports uml, imean in term of propagating it?
15:08<kokoko1>i know they support uml :)
15:09<jdike>what can Intel do, really
15:09<kokoko1>by supporting you
15:09<jdike>there is no Intel distro
15:09<kokoko1>UML was the hostest virtuallization around when our boss started business according to him.
15:10<kokoko1>but no longer hotest.
15:11<kokoko1>heh still our critical VMs on uml :)
15:12<kokoko1>one uml was giving problem 'crashing' after building latest kernel for it, its cool know.
15:12<jdike>how'd you fix it?
15:16<kokoko1>hmm duno latest kernel fixed it i thinks
15:17<kokoko1>ateast if it crashes again we will get the core
15:20<kokoko1>heh, recently migrated our authoritive DNS from domaindirect, and the guy (xen fan) did it on uml
15:20<kokoko1>look like he was not aware what virtual running that particular VM :)
15:27|-|kokoko1 [~Slacker@203.148.65.8] has quit [Remote host closed the connection]
16:16|-|ram [~ram@pool-71-117-236-168.ptldor.fios.verizon.net] has quit [Ping timeout: 480 seconds]
16:24|-|kos_tom [~thomas@humanoidz.org] has quit [Quit: I like core dumps]
16:35|-|ram [~ram@bi01p1.co.us.ibm.com] has joined #uml
17:37|-|_horst [~horst@a89-182-8-58.net-htp.de] has quit [Remote host closed the connection]
18:20|-|jdike [~jdike@pool-71-174-247-82.bstnma.fios.verizon.net] has quit [Quit: Leaving]
19:45|-|ram [~ram@bi01p1.co.us.ibm.com] has quit [Ping timeout: 480 seconds]
20:54|-|tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has quit [Read error: Connection reset by peer]
21:04|-|ram [~ram@pool-71-117-236-168.ptldor.fios.verizon.net] has joined #uml
21:07|-|Nem^ [~Nem@dslb-084-056-246-174.pools.arcor-ip.net] has quit [Ping timeout: 480 seconds]
21:17|-|Nem^ [~Nem@dslb-084-056-246-250.pools.arcor-ip.net] has joined #uml
21:23|-|ram [~ram@pool-71-117-236-168.ptldor.fios.verizon.net] has quit [Read error: Connection reset by peer]
21:23|-|tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has joined #uml
21:23|-|ram [~ram@pool-71-117-236-168.ptldor.fios.verizon.net] has joined #uml
22:59|-|VS_ChanLog [~stats@ns.theshore.net] has left #uml [Rotating Logs]
22:59|-|VS_ChanLog [~stats@ns.theshore.net] has joined #uml
---Logclosed Sat Jul 07 00:00:04 2007