Back to Home / #uml / 2007 / 10 / Prev Day | Next Day
#uml IRC Logs for 2007-10-04

---Logopened Thu Oct 04 00:00:23 2007
00:24|-|balbir [~balbir@122.167.79.239] has quit [Ping timeout: 480 seconds]
02:05|-|balbir [~balbir@59.145.136.1] has joined #uml
03:29|-|aroscha [~aroscha@chello213047053193.30.11.tuwien.teleweb.at] has joined #uml
03:34|-|SNy [7fc75ed2e5@bmx-chemnitz.de] has joined #uml
03:37|-|balbir [~balbir@59.145.136.1] has quit [Ping timeout: 480 seconds]
03:45|-|tyler29 [~tyler@ARennes-256-1-101-24.w90-32.abo.wanadoo.fr] has joined #uml
04:15|-|tyler29 [~tyler@ARennes-256-1-101-24.w90-32.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
05:03|-|aroscha [~aroscha@chello213047053193.30.11.tuwien.teleweb.at] has quit [Quit: aroscha]
05:41|-|kokoko1 [~Slacker@203.148.65.8] has quit [Ping timeout: 480 seconds]
05:43|-|besonen_mobile [~besonen_m@71-220-225-178.eugn.qwest.net] has joined #uml
05:45|-|besonen_mobile_ [~besonen_m@71-220-231-201.eugn.qwest.net] has quit [Ping timeout: 480 seconds]
06:09|-|kokoko1 [~Slacker@203.148.65.8] has joined #uml
06:26|-|Netsplit cation.oftc.net <-> panulirus.oftc.net quits: Hunger
06:39|-|tyler29 [~tyler@ARennes-256-1-24-59.w90-32.abo.wanadoo.fr] has joined #uml
06:53|-|Netsplit over, joins: Hunger
07:04|-|kokoko1 [~Slacker@203.148.65.8] has quit [Ping timeout: 480 seconds]
07:16|-|kokoko1 [~Slacker@203.148.65.19] has joined #uml
07:49|-|tyler29 [~tyler@ARennes-256-1-24-59.w90-32.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
08:00|-|krau [~cktakahas@200.184.118.132] has quit [Ping timeout: 480 seconds]
08:04|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has joined #uml
08:17|-|krau [~cktakahas@200.184.118.132] has joined #uml
09:26|-|mgross [~mgross@jffwprtest.jf.intel.com] has joined #uml
09:39|-|hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has joined #uml
10:46|-|ram [~ram@pool-71-117-247-220.ptldor.fios.verizon.net] has quit [Ping timeout: 480 seconds]
11:00|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has quit [Remote host closed the connection]
11:06|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has joined #uml
11:07|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has quit [Remote host closed the connection]
11:10|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has joined #uml
11:10|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has quit [Remote host closed the connection]
11:13|-|jdike [~jdike@pool-71-248-190-161.bstnma.fios.verizon.net] has joined #uml
11:13<jdike-#uml->>Hi guys
11:13|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has joined #uml
11:13<Magotari-#uml->>Hell, jdike.
11:13<Magotari-#uml->>Hello, I mean.
11:14<jdike-#uml->>hell?
11:14<jdike-#uml->>haha
11:14<Magotari-#uml->>:)
11:15|-|Baltam [~WIKIMOKI@tor-irc.dnsbl.oftc.net] has quit [Ping timeout: 480 seconds]
11:15|-|ram [~ram@bi01p1.co.us.ibm.com] has joined #uml
11:16<Magotari-#uml->>I am afraid I have some bad news. I can crash my UML quite reliably using crashme. Sometimes a kernel mode fault, sometimes something else, but it goes down fairly quickly.
11:16|-|Baltam [~WIKIMOKI@tor-irc.dnsbl.oftc.net] has joined #uml
11:16<jdike-#uml->>what crashme?
11:16<Magotari-#uml->>The commandline which does the trick is "crashme 2000.0 137 5"
11:16<Magotari-#uml->>Crashme is a program for OS testing.
11:16<jdike-#uml->>I tried the one in LTP, but that just hangs
11:16<Magotari-#uml->>Creates random program and runs them.
11:16<jdike-#uml->>I know
11:17<jdike-#uml->>where'd you get it?
11:17<Magotari-#uml->>Go to debian pages, the one from there.
11:17<Magotari-#uml->>They have patches for it.
11:17<Magotari-#uml->>You can get the source and the patches.
11:17<Magotari-#uml->>Want the link?
11:17<Magotari-#uml->>http://packages.debian.org/stable/devel/crashme
11:18<jdike-#uml->>yup, got it
11:19<Magotari-#uml->>You need to run the crashme line that I provided a few times, but usually two are enough. So far just one run did not do any harm, but sooner or later if you keep repeating it...
11:19<Magotari-#uml->>"watch crashme 2000.0 137 5" does the trick under a minute.
11:22<jdike-#uml->>doesn't kill UML
11:23<Magotari-#uml->>Works for me really reliably.
11:23<Magotari-#uml->>Maybe my toolchain...
11:23<Magotari-#uml->>Did you run it many times?
11:24<Magotari-#uml->>Once ain't enough, never.
11:24|-|mjf [~mjf@r5bb59.net.upc.cz] has joined #uml
11:26<jdike-#uml->>been running it for the last 5 minutes at least
11:27<jdike-#uml->>plus 10-20 times before that by hand
11:27<jdike-#uml->>I also have the example from the man page running
11:27<Magotari-#uml->>http://rafb.net/p/w1rjwF28.html
11:27<Magotari-#uml->>This is what I got this time.
11:28<Magotari-#uml->>Sometimes I get something else...
11:28<Magotari-#uml->>I often ran crashme for hours before something happened.
11:28<Magotari-#uml->>It was nondeterministic in crashing uml. Could not predict when or how it will go down.
11:28<Magotari-#uml->>That one line ran many times was the best I found.
11:28<Magotari-#uml->>Ah yes.
11:29<Magotari-#uml->>Higher verbose levels did the trick faster too.
11:29|-|tyler29 [~tyler@ARennes-256-1-55-192.w90-32.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
11:30<Magotari-#uml->>I guess in the end this is not majorly important, but...
11:31<jdike-#uml->>crashes need fixing
11:31<jdike-#uml->>I'll leave these running and see what happens
11:32<Magotari-#uml->>Now I got this, but no crash...
11:32<Magotari-#uml->>do_syscall_stub : ret = 6, offset = -1350979588, data = 086c0ffc
11:32<Magotari-#uml->>do_syscall_stub: syscall 0 failed, return value = 0x6, expected return value = 0x0 syscall parameters: 0x0 0x0 0x0 0x0 0x0 0x0
11:32<Magotari-#uml->>Failed to flush page for address 0xaf79a000
11:32<jdike-#uml->>OK, I was just about to ask for that
11:33<jdike-#uml->>nothing but zeros
11:33<Magotari-#uml->>do_syscall_stub : ret = 6, offset = -1354526724, data = 0836effc
11:33<Magotari-#uml->>do_syscall_stub: syscall 0 failed, return value = 0x6, expected return value = 0x3d81e675 syscall parameters: 0x5d2d7448 0xa4158bc3 0x89080f16 0xf16a015 0x9bc8a108 0xc085080e
11:33<Magotari-#uml->>Failed to flush page for address 0xaf438000
11:33<Magotari-#uml->>do_syscall_stub : ret = 6, offset = -1354502148, data = 08188ffc
11:33<Magotari-#uml->>This was shortly before it died.
11:34<Magotari-#uml->>Love the offset.
11:34<jdike-#uml->>I love the system call
11:36<Magotari-#uml->>When I don't open it in gdb it dies more spectacularly, usually. Prints a *lot* of numbers, in four columns. Then it dies.
11:41<Magotari-#uml->>Ah, yes. And I reproduced it with a rc3 2/2 kernel, so this is not a -mm thing.
11:43<jdike-#uml->>���Lrocess 1� Barfed
11:43|-|balbir [~balbir@122.167.79.239] has joined #uml
11:43<jdike-#uml->>it managed to corrupt its own strings, it seems...
11:46|-|tyler29 [~tyler@ARennes-256-1-15-250.w90-31.abo.wanadoo.fr] has joined #uml
12:01<jdike-#uml->>30 minutes and no problems here
12:04<Magotari-#uml->>Must be my bad toolchain then. It must be my fault somehow.
12:04<jdike-#uml->>I doubt it
12:05<jdike-#uml->>just I'm not reproducing the bug somehow
12:05[~]jdike #uml tries on x86_64#uml-> tries on x86_64
12:07<Magotari-#uml->>I will check in my config for any unusual things which might be causing this.
12:08<Magotari-#uml->>I have an idea.
12:08<Magotari-#uml->>I can give you my kernel and/or image to test this.
12:09<jdike-#uml->>yeah, I might want them
12:09<jdike-#uml->>at least the kernel
12:09<jdike-#uml->>let me see if I can make anything break here
12:09<Magotari-#uml->>Sure. Let me power up my http server.
12:15<Magotari-#uml->>http://89.76.71.182:8008/linux.bz2
12:15<Magotari-#uml->>http://89.76.71.182:8008/linux-failsafe.bz2
12:15<Magotari-#uml->>I can compile one for a different memsplit too.
12:15<Magotari-#uml->>This is my home computer, so the upload rates are small.
12:16<jdike-#uml->>grabbing the kernel
12:16<jdike-#uml->>~3 min
12:17<Magotari-#uml->>BlackBox expe # md5sum linux
12:17<Magotari-#uml->>7ae745bffc07496d37d496d87ae2a7e2 linux
12:17<Magotari-#uml->>BlackBox expe # md5sum linux-failsafe
12:17<Magotari-#uml->>a467c8f9b0ca9a0df984a8eb8ff7a98b linux-failsafe
12:17<Magotari-#uml->>Those are unpacked files.
12:17<jdike-#uml->>OK
12:19<jdike-#uml->>which one crashes?
12:19<Magotari-#uml->>Both.
12:19|-|tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has quit [Ping timeout: 480 seconds]
12:19<Magotari-#uml->>failsafe is rc3, the plain one is rc8-mm2 + patches from you.
12:19<jdike-#uml->>OK
12:19<Magotari-#uml->>For memsplit and printf.
12:20|-|mjf [~mjf@r5bb59.net.upc.cz] has quit [Quit: leaving]
12:22<Magotari-#uml->>I am turning the server off for now, if you want the files or the image I will put it up again.
12:28<jdike-#uml->>I got linux
12:28<jdike-#uml->>I'll play with that
12:29<jdike-#uml->>the hour-long run just finished
12:29<jdike-#uml->>hm
12:29<jdike-#uml->>that terminal froze
12:30<Magotari-#uml->>Yeeeees?
12:30<Magotari-#uml->>That is what usually happens.
12:30<Magotari-#uml->>Your main console is probably printing numbers now.
12:31<jdike-#uml->>no
12:31<Magotari-#uml->>Ah. Ok. Sorry.
12:31<jdike-#uml->>watch is happily watching crashme
12:32<Magotari-#uml->>:/
12:39|-|flatronf700B [~flatronf7@ns1.clipsalportal.com] has quit [Ping timeout: 480 seconds]
12:39|-|aroscha [~aroscha@chello213047053193.30.11.tuwien.teleweb.at] has joined #uml
12:40<kokoko1-#uml->>Hello
12:44<jdike-#uml->>ooh
12:44<jdike-#uml->>it died on x86_64
12:45<jdike-#uml->><3>Failed to get registers from stub, errno = 3
12:45<jdike-#uml->><0>Kernel panic - not syncing: wait_stub_done : failed to wait for SIGUSR1/SIGTRAP, pid = 5309, n = 5309, errno = 0, status = 0xb
12:45<jdike-#uml->>har
12:45<Magotari-#uml->>Not my bug though.
12:45<jdike-#uml->>not the same, but it's something
12:45<Magotari-#uml->>Yeah. It is something.
12:46<Magotari-#uml->>I am preemptively preparing the image to be shipped too.
12:47|-|balbir [~balbir@122.167.79.239] has quit [Read error: Operation timed out]
13:05|-|tyler29 [~tyler@ARennes-256-1-15-250.w90-31.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
13:19|-|tyler29 [~tyler@ARennes-256-1-63-88.w90-32.abo.wanadoo.fr] has joined #uml
13:42|-|tyler29 [~tyler@ARennes-256-1-63-88.w90-32.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
13:51|-|kos_tom [~thomas@humanoidz.org] has joined #uml
13:52|-|tyler29 [~tyler@ARennes-256-1-5-25.w90-31.abo.wanadoo.fr] has joined #uml
14:15|-|kokoko1 [~Slacker@203.148.65.19] has quit [Remote host closed the connection]
15:06<Magotari-#uml->>My dmesg is very full of crashed UML's.
15:06<Magotari-#uml->>Damn apostrophe.
15:06<Magotari-#uml->>Habit. Ah well.
15:10|-|dang [~dang@aa-redwall.nexthop.com] has joined #uml
15:33<Magotari-#uml->>jdike: Do you have any idea why crashme would be nondeterministic? It kills me that it not predictable, despite being predictable by design.
15:33<Magotari-#uml->>jdike: I have a new bug for you.
15:33<Magotari-#uml->>setup_physmem - mapping -753664 bytes of memory at 0x0x8800000 failed - errno = -12
15:33<Magotari-#uml->>I think this bit in my kernel command line broke things: mem=7M
15:34<Magotari-#uml->>8M boots fine.
15:40<jdike-#uml->>I think if you don't give it enough memory, it's OK to panic
15:40<Magotari-#uml->>Erm, no.
15:40<jdike-#uml->>the negative mmap?
15:40<Magotari-#uml->>When I boot 8M free returns 3M free.
15:40<Magotari-#uml->>There is room.
15:40<jdike-#uml->>OK
16:01|-|tyler29 [~tyler@ARennes-256-1-5-25.w90-31.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
16:09<Magotari-#uml->>I am playing with that memory thing a bit. Inserting printfs and so on.
16:11|-|tyler29 [~tyler@ARennes-256-1-20-245.w90-32.abo.wanadoo.fr] has joined #uml
16:15|-|tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has joined #uml
16:22<Magotari-#uml->>You know... This thing is early in the boot process...
16:22<Magotari-#uml->>Not much happening, not too much code to trace...
16:22|-|dang [~dang@aa-redwall.nexthop.com] has quit [Quit: Leaving.]
16:22<Magotari-#uml->>Mind if I take a stab at fixing it myself? It will take me a longer while, but I don't see many people using UML for embedded system testing right now...
16:24<jdike-#uml->>of course not
16:25<Magotari-#uml->>Great.
16:25<Magotari-#uml->>Thanks.
16:25|-|tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has quit [Remote host closed the connection]
16:25|-|tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has joined #uml
16:40|-|tyler29 [~tyler@ARennes-256-1-20-245.w90-32.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
16:53|-|tyler29 [~tyler@ARennes-256-1-24-119.w90-32.abo.wanadoo.fr] has joined #uml
16:54|-|hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has quit [Quit: Leaving]
17:00|-|kos_tom [~thomas@humanoidz.org] has quit [Quit: I like core dumps]
17:28|-|tyler29 [~tyler@ARennes-256-1-24-119.w90-32.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
17:32|-|mgross [~mgross@jffwprtest.jf.intel.com] has quit [Ping timeout: 480 seconds]
17:40|-|jdike [~jdike@pool-71-248-190-161.bstnma.fios.verizon.net] has quit [Quit: Leaving]
17:48|-|_Hunger [Hunger.hu@Hunger.hu] has joined #uml
17:49|-|Netsplit cation.oftc.net <-> panulirus.oftc.net quits: Hunger
19:31|-|aroscha [~aroscha@chello213047053193.30.11.tuwien.teleweb.at] has quit [Quit: aroscha]
19:35|-|ram [~ram@bi01p1.co.us.ibm.com] has quit [Ping timeout: 480 seconds]
20:55|-|balbir [~balbir@122.167.82.87] has joined #uml
22:58|-|VS_ChanLog [~stats@ns.theshore.net] has left #uml [Rotating Logs]
22:58|-|VS_ChanLog [~stats@ns.theshore.net] has joined #uml
23:44|-|Netsplit resistance.oftc.net <-> oxygen.oftc.net quits: Baltam, besonen_mobile, silug_
23:47|-|Netsplit over, joins: Baltam, besonen_mobile, silug_
---Logclosed Fri Oct 05 00:00:44 2007