Back to Home / #uml / 2007 / 08 / Prev Day | Next Day
#uml IRC Logs for 2007-08-08

---Logopened Wed Aug 08 00:00:17 2007
00:50|-|Shel2476 [] has quit []
03:47|-|horst [] has joined #uml
04:11|-|horst [] has quit [Remote host closed the connection]
05:57|-|tyler [] has joined #uml
07:08|-|krau [~cktakahas@] has joined #uml
08:40|-|dang [] has quit [Quit: Leaving.]
08:59|-|dang [] has joined #uml
09:05|-|ram [] has joined #uml
09:24|-|jdike [] has joined #uml
09:25<jdike>Hi guys
09:31<caker>good morning
10:08|-|hfb [] has joined #uml
10:44|-|tyler [] has quit [Read error: Connection reset by peer]
10:45|-|tyler [] has joined #uml
10:53|-|tyler [] has quit [Ping timeout: 480 seconds]
10:55|-|baroni [] has joined #uml
11:00|-|tyler [] has joined #uml
11:53|-|krau [~cktakahas@] has quit [Read error: Connection reset by peer]
11:59|-|krau [~cktakahas@] has joined #uml
12:00|-|baroni [] has quit [Read error: Connection reset by peer]
12:00|-|baroni [] has joined #uml
12:38|-|tyler [] has quit [Remote host closed the connection]
13:58|-|baroni [] has quit [Read error: Connection reset by peer]
14:56|-|horst [] has joined #uml
14:57|-|baroni [] has joined #uml
15:12|-|jjkola [] has joined #uml
15:21<jjkola>I have an uml which segfaults when starting it and this happens when uml kernel is 2.6.23-rc2 (stock + exec patch + humfs patches)
15:21<jjkola>here is backtrace for it:
15:22<jjkola>the host is using kernel
15:23<jjkola>I have also a separate issue which is that if I use 2.6.23-rc2 with skas patches as host then whichever uml kernel I use they segfault even if they work with older kernels
15:24<jjkola>but from that situation I don't have backtrace at the moment
15:27<jjkola>I don't know if it has anything to do with this but I recently updated the host to a newer distro (Ubuntu 6.10 -> Ubuntu 7.04)
15:30<jdike>jjkola, that's fixed in current -mm and upcoming -stable
15:30<jjkola>ah, ok
15:32<jdike>just stop using a COW file if you just want to avoid the bug
15:32|-|baroni [] has quit [Ping timeout: 480 seconds]
15:43|-|baroni [] has joined #uml
15:48<jjkola>hmm, I can't seem to find the patch
15:49<jjkola>would you like to point out the place where I can get it?
15:51<jjkola>ah, it's that one
15:58<jjkola>hmm, I got "Reversed (or previously applied) patch detected!" when I tried to apply the patch
15:58<jjkola>so I think it is already in rc2
15:58<jdike>could be
15:58<jdike>but rc2 doesn't die like that
16:12<jjkola>why is sector_mask set to 0 on line 1069 in arch/um/drivers/ubd_kern.c?
16:12<jjkola>it seems like it will not be set to different value before it's used
16:12<jjkola>or did I read the code wrong?
16:13|-|ram [] has quit [Remote host closed the connection]
16:13<jjkola>ops, I read it wrong as the value isn't used but the size of it
16:13|-|ram [] has joined #uml
16:53<jjkola>hmm, shouldn't the comparator be >= instead of > on the line 1038 in arch/um/drivers/ubd_kern.c?
16:53<jjkola>because I think that the size could be excatly 65536
16:54<jjkola>as it's the case in the backtrace
17:00|-|dang [] has quit [Quit: Leaving.]
17:01<jjkola>what I mean is that the data length and the buffer size can be same
17:05|-|jvaughan [] has quit [Remote host closed the connection]
17:05|-|jvaughan [] has joined #uml
17:06|-|Netsplit <-> quits: silug, hfb, VS_ChanLog, da-x, nessie, kokoko1, balbir, caker
17:06|-|Netsplit over, joins: nessie
17:07|-|hfb [] has joined #uml
17:07|-|VS_ChanLog [] has joined #uml
17:07|-|caker [] has joined #uml
17:07|-|kokoko1 [~Slacker@] has joined #uml
17:07|-|silug [] has joined #uml
17:07|-|balbir [~balbir@] has joined #uml
17:07|-|da-x [] has joined #uml
17:09|-|kokoko1 [~Slacker@] has quit [Ping timeout: 480 seconds]
17:09|-|kokoko1 [~Slacker@] has joined #uml
17:19|-|silug_ [] has joined #uml
17:20|-|Netsplit <-> quits: silug, hfb, VS_ChanLog, da-x, balbir, caker
17:20|-|caker [] has joined #uml
17:22|-|jvaughan [] has quit [Remote host closed the connection]
17:24|-|VS_ChanLog [] has joined #uml
17:25|-|da-x [] has joined #uml
17:26|-|balbir [~balbir@] has joined #uml
17:27<jdike>65536 should be OK
17:27<jdike>just anything larger is a problem
17:27|-|hfb [~hfb@] has joined #uml
17:28<jjkola>(sizeof(req->sector_mask) * 8) << 9 gives value 16384
17:29<jjkola>at least in gdb
17:32<jjkola>so it would mean that anything bigger than 16384 is considered bad
17:35|-|VS_ChanLog [] has quit [Ping timeout: 480 seconds]
17:36|-|VS_ChanLog [] has joined #uml
17:45|-|Netsplit <-> quits: VS_ChanLog
17:47|-|Netsplit over, joins: VS_ChanLog
17:52<jjkola>maybe the problem lies in there that you can have two different max amounts of buffer lengths 16384 and 65536 depending if you are reading cow file or not
17:54<jjkola>I mean the patch which you pointed creates that situation and then there is that check which checks for the lower amount (16384) and panics if it happens to be the bigger one
17:55|-|horst [] has quit [Remote host closed the connection]
17:57|-|silug_ changed nick to silug
17:59<jjkola>I think that last theory is right one for the problem
18:01|-|VS_ChanLog [] has quit [Ping timeout: 480 seconds]
18:05|-|VS_ChanLog [] has joined #uml
18:14|-|VS_ChanLog [] has quit [Server closed connection]
19:06|-|hfb [~hfb@] has quit [Quit: Leaving]
19:11|-|ram [] has quit [Ping timeout: 480 seconds]
19:32|-|rbaroni [] has joined #uml
19:33|-|jdike [] has quit [Quit: Leaving]
19:51|-|rbaroni [] has quit [Quit: :wq]
20:16|-|jjkola [] has quit [Quit: *pop*]
20:25|-|VS_ChanLog [] has joined #uml
22:36|-|IRCFrEAK [] has joined #uml
22:59|-|VS_ChanLog [] has left #uml [Rotating Logs]
22:59|-|VS_ChanLog [] has joined #uml
23:27|-|IRCFrEAK [] has quit [Ping timeout: 480 seconds]
23:31|-|IRCFrEAK [] has joined #uml
---Logclosed Thu Aug 09 00:00:59 2007