Back to Home / #uml / 2007 / 02 / Prev Day | Next Day
#uml IRC Logs for 2007-02-05

---Logopened Mon Feb 05 00:00:23 2007
00:30|-|karrde_ [] has quit [Ping timeout: 480 seconds]
00:39|-|karrde_ [] has joined #uml
01:23|-|nessie_ changed nick to nessie
01:42|-|motp [~motp@] has joined #uml
02:02|-|kos_tom [] has joined #uml
04:17|-|motp [~motp@] has quit [Quit: Leaving]
05:13|-|karrde_ [] has quit [Ping timeout: 480 seconds]
05:22|-|karrde_ [] has joined #uml
05:25|-|Ws [~sare@] has joined #uml
05:38|-|Ws [~sare@] has quit [Read error: Connection reset by peer]
05:45|-|richardw [~richardw@] has joined #uml
06:34|-|infowolfe [] has joined #uml
06:37<infowolfe>anybody know when 2.6.20 support will be ready for testing?
06:37[~]peterz fails to parse that
06:38<infowolfe>2.6.20 skas patches, sorry ;-)
06:38<peterz>ah, see that works ;-)
06:38<peterz>dunno, srry
06:38<peterz>are there any skas3 for recent -rcs?
06:38<infowolfe>i didn't see any
06:39<infowolfe>i don't know where his git tree is kept though
06:39<infowolfe>it seems that 2.6.19-v9-pre9 applies cleanly to 2.6.20
06:40[~]infowolfe recompiles his kernel
06:48<kokoko1>infowolfe, please let me know if you can successful boot the host with 2.6.20 + skas3 and you UML run successfully
06:48<infowolfe>i don't know what to look for to verify that skas3 is actually active
06:48<kokoko1>so you booted into 2.6.20 with skas3 ?
06:48<kokoko1>and your UML started?
06:49<kokoko1>damn i just can't jump to it, coz all my machines are remote :S
06:49<infowolfe>Checking for the skas3 patch in the host:
06:49<infowolfe> - /proc/mm...not found
06:49<infowolfe> - PTRACE_FAULTINFO...not found
06:49<infowolfe> - PTRACE_LDT...not found
06:49<infowolfe>that's not a good sign
06:50<infowolfe>crap, i booted the wrong kernel
06:51<kokoko1>which kernel you are using on your UMLs ?
06:51<infowolfe>trying again, this time with 2.6.20-skas3-v9-pre9
06:51<infowolfe>vanilla 2.6.20
06:52<kokoko1>I am running
06:52<peterz>vanilla ice
06:52<infowolfe>looks to be booting properly at 2.6.20-skas3-v9-pre9
06:52<kokoko1>ls -l /proc/mm ?
06:53<infowolfe>got a panic
06:53<kokoko1>you used the same .conf with this kernel?
06:53<infowolfe>one's 2.6.20 vanilla the other has skas applied
06:54<kokoko1>well try it
06:54<infowolfe>try which?
06:54<kokoko1>by make oldconfig
06:54<kokoko1>the any config you have for 2.6.x with these new kernel
06:54<kokoko1>imean the last working kernel .config
06:54<infowolfe>the only thing different is the host has the skas3 applied and the other doesn't
06:55<kokoko1>its okay, just patch the vanilla copy the old config to new kernel source directory
06:55<kokoko1>and then 'make oldconfig
06:56<kokoko1>it will prompt you for any new things which has been added in 2.6.20
06:56<infowolfe>new machine
06:56<infowolfe>haven't booted it on anything other than 2.6.20 ;-)
06:56<infowolfe>and this is my first time using uml since kernel 2.5
07:01<infowolfe>same breakage
07:02<kokoko1>when until -bb release skas3 for .20 :P
07:02<infowolfe>are you running x86_64 or i386?
07:03<infowolfe>should work for you i think
07:04<kokoko1>I just can't take a risk with Host machine however I will soon shift all UMLs to .20
07:04<kokoko1>As i told you host machines remote to me thousand mills away :S
07:04<infowolfe>that oops is directly attributed to x86_64
07:04<infowolfe>look for err = 22 and follow the conversation
07:06<infowolfe>if i want to re-do my debian install, i can run maybe?
07:08<kokoko1>Hmm yes
07:08<kokoko1>you install 64bit debian?
07:08<infowolfe>in the host, yes
07:08<infowolfe>rather in the guest
07:08<infowolfe>the host is gentoo
07:08<kokoko1>okay then install 32bit and try on host :S
07:26|-|richardw_ [] has joined #uml
07:27|-|richardw [~richardw@] has quit [Read error: Connection reset by peer]
08:09|-|da-x [] has quit [Server closed connection]
08:10|-|da-x [] has joined #uml
10:34|-|jdike [] has joined #uml
10:34<jdike>Hi guys
10:34<infowolfe>hi jdike
10:34<infowolfe>any progress on the x86_64 guest stuff?
10:36<jdike>Which stuff, specifically?
10:37<infowolfe>that stuff
10:37<infowolfe>saw some mention of something very similar on the ml
10:37<infowolfe>but never saw a resolution of any kind
10:38<jdike>that's a known problem
10:38<infowolfe>i know
10:38<infowolfe>i'm wondering if there's a solution short of skas0
10:38<jdike>with a known solution
10:39<infowolfe>i didn't see the solution anyplace, hence why i asked
10:39<jdike>the host's /proc/sys/vm/max_map_count is too small
10:40<infowolfe>what's a reasonable figure (i'm showing 65536 currently)
10:40<infowolfe>btw, the 2.6.19-skas3-v9-pre9 applies cleanly to 2.6.20
10:40<jdike>it depends on the RSS of the UML process, which depends on the UML physical memory size
10:41<infowolfe>i want to run 2 umls, with 1GB ram allocated to each
10:41<jdike>65536 gives a UML process a max 256M RSS
10:41<infowolfe>so i need to *4 the 65536 or *8 it?
10:41<jdike>4 times will do
10:42<jdike>that will cover a process using the whole iG
10:42<infowolfe>updated to 524288
10:43<infowolfe>and i still get the same panic/oops
10:44<jdike>ran UML after bumping it?
10:44<infowolfe>want another pastebin?
10:44<jdike>anything in the host's dmesg?
10:44<infowolfe>not since my bridges came up on boot
10:45<jdike>what's the UML process that's doing it?
10:46<jdike>whoops, it's different and I didn't notice
10:46<jdike>want to run with a debug patch?
10:47<jdike>I need more information from that
10:47<infowolfe>not a problem
10:48<jdike>now I'm remembering
10:48<jdike>this happened a long time ago and I never figured out what was happening
10:49<jdike>'s the boot log?
10:49<infowolfe>i don't have one defined
10:49<jdike>can you paste that?
10:49<infowolfe>./linux con=pts con0=fd:0,fd:1 eth0=tuntap,umlmail mem=512M root=/dev/root rootflags=/home/umlmail/umlroot rootfstype=hostfs ubd1=swap_fs
10:49<infowolfe>that's my bootline
10:50<jdike>the console output
10:50<jdike>from the boot
10:50<infowolfe>1 sec
10:50<infowolfe>i cleared that kernel out already, gotta recompile it (won't take long)
10:50<jdike>this is happening to init?
10:52<infowolfe>full paste
10:53<jdike> - /proc/mm...found
10:53<jdike> - PTRACE_FAULTINFO...not found
10:53<jdike> - PTRACE_LDT...found
10:53<jdike>it's not seeing PTRACE_FAULTINFO
10:54<jdike>which is why it's going into this seldom-used piece of code
10:54[~]infowolfe wonders why it wouldn't be
10:54|-|less [] has quit [Ping timeout: 480 seconds]
10:55<jdike>what skas patch are you using?
10:56<infowolfe>2.6.19-skas3-v9-pre9 applied to 2.6.20
10:57<jdike>when did this last work?
10:58<infowolfe>new install
10:58<infowolfe>so that'd be a 'never' because nothing farther back has been attempted
10:59<jdike>caker, there?
10:59<infowolfe>last time i ran uml was 2.5 series
11:00<infowolfe>but there are some fixups that i require that are in 2.6.20 and i really don't feel like mucking about with git, since 2.6.20 had some pretty major changes
11:00<jdike>2.6.20 host or guest?
11:01<jdike>I want to ask caker what he's running
11:01<jdike>be he's ignoring me :-)
11:08<infowolfe>i'll stick around for a while :-)
11:10<jdike>infowolfe, apply and see what it says
11:12<infowolfe>host or guest?
11:13<infowolfe>arch/x86_64/kernel/asm-offsets.c:41: error: invalid use of undefined type 'struct x8664_pda'
11:13<infowolfe>fails to compile
11:13<jdike>you built UML before, right?
11:16<infowolfe>i'm tired, apologies
11:19<infowolfe>jdike, patch applied, doesn't seem to have changed any
11:20<jdike>there should be an extra line of output
11:20<infowolfe>[42949374.780000] mmap args - addr = 0x7fbfffe000, fd = 3, offset = 244000
11:20<infowolfe>that the one?
11:21<jdike>one of those number is allegedly bad
11:21<infowolfe>which one?
11:22<jdike>if I knew, that would be the bug
11:22<jdike>but it must be the offset
11:22<infowolfe>do you have access to a x86_64 box?
11:22<jdike>yes, but without skas
11:22<infowolfe>gimme your ssh pubkey?
11:23<jdike>and now that I'm thinking about it some more, skas3 on x86_64 has always been iffy
11:26<infowolfe>if there's something else you'd rather have me test i'm more than willing
12:22|-|less [] has joined #uml
13:06|-|albertito [] has joined #uml
14:07|-|nessie_ [] has joined #uml
14:08|-|ram [] has joined #uml
14:10|-|nessie [] has quit [Ping timeout: 480 seconds]
14:18|-|Netsplit <-> quits: tchan
14:23|-|tchan [] has joined #uml
14:38|-|ram [] has quit [Ping timeout: 480 seconds]
15:01|-|ram [] has joined #uml
15:26|-|rw__ [~richardw@] has joined #uml
15:33|-|richardw_ [] has quit [Ping timeout: 480 seconds]
15:37|-|kos_tom [] has quit [Quit: I like core dumps]
16:05|-|hfb [] has joined #uml
16:13[~]jdike is scared to look
16:14<jdike>suprisingly innuous
16:16<jdike>which reminds me
16:16<jdike>action is needed if I'm not to be in trouble with a certain someone
16:17<dgraves>jdike: who is that someone? :)
16:17<dgraves>besides, unlike some people, i don't post nsfw links here. :)
16:17<jdike>a certain female someone
16:17<dgraves>just not safe for waistline ones.
16:18<dgraves>can't you just code her a new uml feature?
16:19<dgraves>awww.. that would have been neat. ;)
16:19<dgraves>how about an edible arrangement?
16:19<jdike>it would be, yes
16:19<jdike>never tried one of those
16:19<dgraves>my friend got one for his other last year.
16:19<dgraves>he said it was really good.
16:19<dgraves>i keep meaning to get one for mine.
16:20[~]jdike spots his home town there
16:20<jdike>#3 on the new stores list
16:22<dgraves>i want to surf CA sometime.
16:22<dgraves>but that would require actually surfing again
16:24<dgraves>i'm saving for a new board.
16:24<dgraves>but i always seem to be working. :-P
16:24<dgraves>the whole paying the bills thing, i think.
16:24<dgraves>i like snowboarding too.
16:24<dgraves>what's your activies jdike?
16:26<dgraves>cross country?
16:28<jdike>more downhill of late
16:28<dgraves>i was reading a lot of people started telemarking downhill cause it was too easy to downhill downhill.
16:29<jdike>I don't see too much telemarking
16:30<dgraves>i've seen some.
16:30<dgraves>but its always the good ones on the double blacks.
16:30<dgraves>you know, the hills i look at? :)
16:35<dgraves>notice i didn't say "go down them".
16:35<dgraves>i'm horrible. :)
16:36<dgraves>so what's next on the uml train, now that kvd and smp is in?
16:37<jdike>uhh, they're not, and they're next on the UML train
16:37<jdike>along with some bug-smashing
16:37<jdike>with which I'm not having notable success
16:38<dgraves>hrm. i thought kvd had gone in 20 for you.
16:38<jdike>-bash-3.1# host
16:38<jdike>Segmentation fault
16:38<dgraves>i must have imagined things.
16:38<dgraves>yeah, that's no tmuch success, alright.
16:38<jdike>kvm did
16:38<dgraves>ah, right.
16:38<jdike>but the UML client is still fictional
16:39<dgraves>such is life, i guess.
16:42<jdike>well, I hope
16:42<jdike>I'd prefer my life to be non-fiction
16:42<jdike>if it were fiction, I'd have to fire the writer
16:44<dgraves>speaking of fiction, you watch the sB last night?
16:44<dgraves>what a shame.
16:44<dgraves>it might have been a good game, if not for the rain.
16:45<dgraves>started off well enough.
16:45<jdike>charlan had it on, but I was studying chinese in the kitchen
16:45<jdike>but listening enough to tell what was happening
16:45<jdike>especially the first play
16:46<jdike>thought it was going to be a long night for the Colts
16:46<dgraves>studying chinese???
16:46<dgraves>wow, that's wicked cool.
16:46<jdike>half my family is now chinese
16:46<dgraves>but yeah, i agree.
16:47<dgraves>they held on pretty good. but 3 & out for your offense all the time doesn't leave much rest for the defense.
16:47<dgraves>oh, didn't know. that's cool.
16:47<jdike>I've always had a sore spot for the Bears after the 86 superbowk
16:51<dgraves>anyway, audi!
17:10|-|albertito [] has quit [Quit: Reconnecting]
17:10|-|albertito [] has joined #uml
17:50|-|hfb [] has quit [Quit: Leaving]
17:56|-|rw__ [~richardw@] has quit [Remote host closed the connection]
18:25|-|ram [] has quit [Ping timeout: 480 seconds]
18:49|-|alb [] has joined #uml
18:50|-|Netsplit <-> quits: tchan, albertito
18:50|-|tchan [~tchan@] has joined #uml
19:04|-|jdike [] has quit [Quit: Leaving]
19:13<infowolfe>caker, ping?
19:17<infowolfe>you run x86_64 hosts and guests?
19:21<infowolfe>well, i gave jdike access to an x86_64 box he's welcome to use for testing
19:21<infowolfe>let's see what he comes up with :-p
19:33<caker>infowolfe == Antoine Martin?
19:33<caker>heh, ok
19:33<caker>ok just read scrollback
19:38|-|cmantito [] has quit [Quit: I aer quitted.]
20:02|-|VS_ChanLog [] has quit [Server closed connection]
20:04|-|VS_ChanLog [] has joined #uml
20:16|-|cmantito [] has joined #uml
20:34|-|ram [] has joined #uml
21:42|-|infowolfe [] has quit [Ping timeout: 480 seconds]
22:28|-|infowolfe [] has joined #uml
22:37|-|Nem^1 [] has joined #uml
22:38|-|kokoko1 [~Slacker@] has quit [Server closed connection]
22:39|-|kokoko1 [~Slacker@] has joined #uml
22:45|-|Nem^ [] has quit [Ping timeout: 480 seconds]
22:45|-|Nem^1 changed nick to Nem^
22:57|-|SNy [] has quit [Remote host closed the connection]
22:57|-|SNy [] has joined #uml
22:58|-|VS_ChanLog [] has left #uml [Rotating Logs]
22:58|-|VS_ChanLog [] has joined #uml
23:12|-|infowolfe_ [] has joined #uml
23:18|-|Netsplit <-> quits: infowolfe, VS_ChanLog, alb, Nem^, ram
23:24|-|VS_ChanLog [] has joined #uml
23:24|-|ram [] has joined #uml
23:25|-|Nem^ [] has joined #uml
23:25|-|albertito [] has joined #uml
23:32|-|SNy_ [] has joined #uml
23:35|-|tchan1 [] has joined #uml
23:38|-|Netsplit <-> quits: kokoko1, SNy, tchan, cmantito, infowolfe_
23:40|-|Netsplit over, joins: cmantito
23:49|-|Netsplit <-> quits: VS_ChanLog, tchan1, cmantito, SNy_, Nem^, albertito, ram
---Logclosed Tue Feb 06 00:00:04 2007