Back to Home / #uml / 2007 / 09 / Prev Day | Next Day
#uml IRC Logs for 2007-09-12

---Logopened Wed Sep 12 00:00:47 2007
01:23|-|aroscha [] has joined #uml
01:42|-|balbir [~balbir@] has quit [Ping timeout: 480 seconds]
01:59|-|flinkdeldinky [] has quit [Ping timeout: 480 seconds]
03:03|-|aroscha [] has quit [Quit: aroscha]
03:12|-|krau [] has joined #uml
03:38|-|dsoul_ [] has joined #uml
03:44|-|dsoul [] has quit [Ping timeout: 480 seconds]
03:50|-|Baltam54 [] has quit [Remote host closed the connection]
03:56|-|Baltam54 [] has joined #uml
05:01|-|aroscha [] has joined #uml
05:07|-|tyler [] has joined #uml
05:26|-|tyler [] has quit [Read error: Connection reset by peer]
05:48|-|tyler [] has joined #uml
06:05|-|Baltam54 [] has quit [Quit: using sirc version 2.211]
06:06|-|Blissex [] has joined #uml
06:34|-|Baltam [debian-tor@] has joined #uml
06:37|-|aroscha [] has quit [Quit: aroscha]
06:49|-|aroscha [] has joined #uml
06:50|-|aroscha [] has quit []
06:52|-|tyler [] has quit [Ping timeout: 480 seconds]
07:04|-|tyler [] has joined #uml
07:31|-|flinkdeldinky [] has joined #uml
08:02|-|tyler [] has quit [Ping timeout: 480 seconds]
08:12|-|tyler [] has joined #uml
08:20|-|dang [~dang@] has joined #uml
08:37|-|jdike [] has joined #uml
08:37<jdike>Hi guys
08:38<dang>Morning, Jeff
08:44|-|balbir [~balbir@] has joined #uml
08:46|-|the_hydra [~the_hydra@] has joined #uml
08:46<the_hydra>jdike: hi
08:53|-|krau [] has quit [Remote host closed the connection]
09:00|-|BlaisorBlade [] has joined #uml
09:01|-|krau [] has joined #uml
09:23|-|Urgleflogue [~plamen@] has quit [Remote host closed the connection]
09:23|-|krau [] has quit [Read error: Connection reset by peer]
09:28|-|balbir [~balbir@] has quit [Ping timeout: 480 seconds]
09:28|-|Blissex [] has quit [Remote host closed the connection]
09:34|-|Urgleflogue [~plamen@] has joined #uml
09:34|-|tyler [] has quit [Quit: ++]
09:45|-|tyler29 [] has joined #uml
09:47|-|tyler29 [] has quit []
09:47|-|tyler [] has joined #uml
09:48|-|hfb [] has joined #uml
09:59|-|ram [] has joined #uml
10:26|-|aroscha [] has joined #uml
10:27|-|aroscha [] has quit []
10:27|-|aroscha [] has joined #uml
11:24|-|tyler [] has quit [Read error: Connection reset by peer]
11:30|-|tyler [] has joined #uml
12:00|-|tyler [] has quit [Ping timeout: 480 seconds]
12:17|-|tyler [] has joined #uml
12:39|-|dsoul_ [] has quit [Ping timeout: 480 seconds]
12:40|-|dsoul [] has joined #uml
13:04|-|dex [] has joined #uml
13:05|-|BlaisorBlade [] has quit [Ping timeout: 480 seconds]
13:09|-|dsoul [] has quit [Ping timeout: 480 seconds]
13:09|-|dsoul [] has joined #uml
13:11|-|netdrake [] has joined #uml
13:22|-|the_hydra [~the_hydra@] has left #uml []
13:27|-|netdrake [] has quit [Remote host closed the connection]
13:28|-|netdrake [] has joined #uml
13:30|-|netdrake [] has quit []
13:30|-|dsoul [] has quit [Ping timeout: 480 seconds]
13:31|-|dsoul [] has joined #uml
13:32|-|netdrake [] has joined #uml
14:00[~]jdike gets tickless ready for -mm
14:00<netdrake>can anybody help me with build external kernel module for uml, i have some troubles with include headers (using 2.6.21 kernel)plase
14:01<dang>netdrake: I always build in a chroot of the uml filesystem, using it's kernel headers, or else inside the UML itself.
14:01<dang>That way you don't have to worry about accidentally getting the system headers involved.
14:02<jdike>The usual method should work for UML
14:02<jdike>it's no different from any other kernel in this regard, AFAIK
14:02|-|tyler [] has quit [Ping timeout: 480 seconds]
14:03<netdrake>dang: hm nice idea, i will try it .. thanks :)
14:06<netdrake>jdike: usual method ? did you mean CFLAGS ... make script ... or .. something else ?
14:06<jdike>not something I usually do
14:07<jdike>but I recall a little Makefile which does a make -C /path/to/kernel/tree or something along those lines
14:08<netdrake>ah ok :)
14:09|-|Urgleflogue [~plamen@] has quit [Ping timeout: 480 seconds]
14:13[~]dgraves builds like that all the time, outside uml.
14:13<dgraves>its way faster.
14:14<dgraves>netdrake: how are you attempting to build it?
14:14<dgraves>jdike: did you fix the speed up? :)
14:14<jdike>you mean the accounting stuff from yesterday?
14:14[~]dgraves thinks his keyboard ish drinked.
14:15<jdike>somewhere in there is a complete, coherent, one-word answer
14:15<jdike>damned if I can find it though
14:16<dgraves>yeah, the accounting stuff.
14:16<dgraves>where you were magically making things take less time than they really did.
14:16<dgraves>so how are you putting tickless in?
14:16<jdike>I'm not convinced that the generic code gets that right
14:16<dgraves>but it works! ;0
14:16<jdike>does it?
14:17<jdike>Have you looked at a tickless i386 kernel?
14:17[~]dgraves actually doesn't know.
14:17<dgraves>what's tickless get you?
14:18<netdrake>dgraves: i trying use uml fs ...
14:18<jdike>looking at a single UML, somewhat better performance
14:18<jdike>it doesn't have to take 100 interrupts a second
14:18<netdrake>... because i need debug my module
14:19<jdike>looking at a lot of UMLs on a host, the host doesn't have to deliver 100 interrupts/sec to a bunch of idle UMLs
14:20|-|tyler [] has joined #uml
14:25<jdike>tickless gets me another 1-2% on a kernel build
14:27<dang>dgraves: That's why I usually build in chroot...
14:27<dang>Rather than in uml...
14:27<dang>I have a whole complex dev environment involving different OS versions in different chroot/uml combinations.
14:33<dgraves>netdrake: what's your build line?
14:34<dgraves>dang: not sure how that fixes it, but its the same idea, i suppose...
14:34<jdike>dang, I don't suppose you have a sparc cross-build environment in there somewhere?
14:35<dang>dgraves: I'm building in the same environment as the UML (modulo uname output), but at full speed.
14:35<dang>jdike: No, sorry. Only x86 and x86_64...
14:35<dgraves>dang: for a kernel, so am i. :) and since we've got 26 developers using the root image, its not like a chroot would easily work.
14:36|-|aroscha [] has quit [Quit: aroscha]
14:41<dang>dgraves: Well, I'm the only one making root images here... Others just use them.
14:42<dgraves>dang: yeah, our developers all use the same root image through the magic of cow. no sense setting up complex chroot stuff when the kernel build process (and module build) already handles different kernel versions.
14:43<dang>dgraves: You assume I'm working with sane kernel modules. :)
14:44<dang>You also assume the chroot environment is even remotely similar to the host box...
14:45<dang>Some of the environments inside the chroot/UML are *old*.
14:45<dang>I'm sure I could do it; it's just simpler to use a chroot and be sure I have the correct environment...
14:47<dgraves>dang: long as it works. :)
14:47<dang>Yep, it works. Quite well, I might add...
14:48<dang>Of course, I'm a bit of a chroot junky... :)
14:55|-|tyler [] has quit [Read error: Connection reset by peer]
14:57|-|tyler [] has joined #uml
14:58<jdike>11 patches ready to go, so 11 builds and boots to make sure
15:00<dang>Run all 11 at the same time. :)
15:16<jdike>one of them broke delay calibration
15:28|-|aroscha [] has joined #uml
15:51|-|tyler [] has quit [Ping timeout: 480 seconds]
16:11|-|tyler [] has joined #uml
16:14|-|tyler [] has quit [Read error: Connection reset by peer]
16:18<jdike>11 patches off
16:24|-|flinkdeldinky [] has quit [Ping timeout: 480 seconds]
16:25|-|tyler [] has joined #uml
16:27<dang>Good stuff.
16:31|-|dang [~dang@] has quit [Quit: Leaving.]
16:51|-|aroscha [] has quit [Quit: aroscha]
17:04|-|alb [] has quit [Ping timeout: 480 seconds]
17:18|-|Baltam [debian-tor@] has quit [Remote host closed the connection]
17:26|-|tyler [] has quit [Ping timeout: 480 seconds]
17:33[~]jdike off too
17:33|-|jdike [] has quit [Quit: Leaving]
17:49|-|Baltam [] has joined #uml
17:55|-|dex [] has quit [Quit: Bye bye ...]
18:09|-|mgross [] has joined #uml
18:15|-|hfb [] has quit [Quit: Leaving]
18:26|-|mgross [] has quit [Quit: Leaving]
21:29|-|force [] has joined #uml
21:35|-|Baltam [] has quit [Remote host closed the connection]
21:35<force>i am having a little trouble with uml networking
21:36<force>my topology is server->switch->router(nat)->internet
21:36|-|Baltam [] has joined #uml
21:37<force>i can't ping anything it just drops out
21:37<force>the tap0 interface IP address is set to netmask
21:37<force>within the uml the eth0 address is set to netmask
21:38<force>and i have uml_switch running
21:38<force>i tried to add a route both within the host and the uml
21:38<force>but i'm unsure exactly what it should be set to
21:39<force>ideally i want to run a couple uml kernels each with their own private IP address
21:40<force>none of the documentation i have read has anything for this topology
21:40<force>has anyone here got a solution for this?
22:07|-|silug [~steve@] has quit [Ping timeout: 480 seconds]
22:23|-|force [] has quit [Quit: leaving]
22:59|-|VS_ChanLog [] has left #uml [Rotating Logs]
22:59|-|VS_ChanLog [] has joined #uml
23:13|-|albertito [] has joined #uml
23:19|-|Baltam [] has quit [Remote host closed the connection]
23:21|-|Baltam [] has joined #uml
23:59|-|silug [~steve@] has joined #uml
---Logclosed Thu Sep 13 00:00:23 2007