| --- | Log | opened Wed Jan 23 00:00:05 2008 |
| --- | Day | changed Wed Jan 23 2008 |
| 00:00 | -!- | balbir [~balbir@59.145.136.1] has joined #uml |
| 00:44 | -!- | aindilis [~aindilis@75.146.96.198] has joined #uml |
| 04:36 | -!- | kos_tom [~thomas@humanoidz.org] has joined #uml |
| 04:41 | -!- | scipio [~scipio@61.6.65.198] has quit [Server closed connection] |
| 04:42 | -!- | scipio [~scipio@61.6.65.198] has joined #uml |
| 07:51 | -!- | balbir [~balbir@59.145.136.1] has quit [Remote host closed the connection] |
| 08:34 | -!- | kokoko1 [~Slacker@203.148.64.252] has joined #uml |
| 08:34 | <kokoko1:#uml> | hi |
| 08:35 | -!- | dang [~dang@nemesis.fprintf.net] has quit [Quit: Leaving.] |
| 08:58 | -!- | balbir [~balbir@122.167.98.14] has joined #uml |
| 09:01 | -!- | dang [~dang@aa-redwall.ghs.com] has joined #uml |
| 09:12 | -!- | balbir [~balbir@122.167.98.14] has quit [Read error: Operation timed out] |
| 09:25 | -!- | balbir [~balbir@122.167.116.196] has joined #uml |
| 09:28 | -!- | marcel [~marcel@wc-50.r-195-35-150.atwork.nl] has joined #uml |
| 09:57 | -!- | jdike [~jdike@pool-72-93-84-53.bstnma.fios.verizon.net] has joined #uml |
| 09:57 | <jdike:#uml> | Hi guys |
| 09:59 | -!- | hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has joined #uml |
| 10:03 | <kokoko1:#uml> | hi jeff |
| 10:15 | <jdike:#uml> | back to figuring out why skas4 32-on-64 doesn't work |
| 10:23 | <jdike:#uml> | what oh what did I break? |
| 10:33 | <peterz:#uml> | mmap layouts? |
| 10:43 | <jdike:#uml> | no |
| 10:43 | <jdike:#uml> | it's switch_mm |
| 10:43 | <jdike:#uml> | when I switch into a new address space, IP and SP are crap |
| 10:45 | <jdike:#uml> | 64-on-64 is fine, which really reduces the amount of code that could be hiding the bug |
| 10:46 | <jdike:#uml> | hmm |
| 10:47 | <jdike:#uml> | pulling 64-bit IP from registers and putting it in AX: |
| 10:47 | <jdike:#uml> | 0xffffffff8020f4bc <ptrace_ip+30>: mov 0x80(%rdi),%rax |
| 10:47 | <jdike:#uml> | 32-bit: |
| 10:48 | <jdike:#uml> | 0xffffffff8020f4b4 <ptrace_ip+22>: lea 0x330(%rdi),%rax |
| 10:48 | <jdike:#uml> | 32-bit should be mov, with a smaller offset |
| 10:52 | <jdike:#uml> | 0x0000000000000d38 <ptrace_ip+22>: mov 0x30(%rdi),%eax |
| 10:52 | <jdike:#uml> | that's more like it |
| 10:52 | <jdike:#uml> | return ptrace_ip32(®s->u.regs32); |
| 10:53 | <jdike:#uml> | #define ptrace_ip32(regs) (unsigned long) (regs)[EIP] |
| 10:53 | <jdike:#uml> | The '&' seems to matter |
| 10:53 | <peterz:#uml> | :-) |
| 10:54 | <peterz:#uml> | which is why akpm hates macros |
| 10:54 | <jdike:#uml> | u.regs32 is an array |
| 10:55 | <jdike:#uml> | one more reboot |
| 10:57 | -!- | hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has quit [Quit: Leaving] |
| 11:03 | <jdike:#uml> | here goes nothing |
| 11:03 | <jdike:#uml> | cool |
| 11:20 | <dang:#uml> | jdike: Working? |
| 11:20 | <jdike:#uml> | yup |
| 11:20 | <dang:#uml> | Sweet. |
| 11:21 | <jdike:#uml> | one more round of testing on the patches I intend to release to make sure there's no last-minute breakage |
| 11:23 | <jdike:#uml> | 64-bit host looks good |
| 11:26 | <jdike:#uml> | 64-bit guest |
| 11:28 | <jdike:#uml> | 32-bit host |
| 11:30 | -!- | hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has joined #uml |
| 11:33 | <jdike:#uml> | 32-bit guest |
| 11:34 | <jdike:#uml> | 32-on-64 |
| 11:35 | <dang:#uml> | How about 64-on-32? :P |
| 11:36 | <jdike:#uml> | where'd I put my gun? |
| 12:12 | <jdike:#uml> | Off they go |
| 12:12 | <jdike:#uml> | 10 little patches trooping off into the distance |
| 12:44 | * | dgraves:#uml watches them hit a land mine. |
| 12:46 | <kokoko1:#uml> | lol |
| 12:56 | <jdike:#uml> | this time uml-user gets them |
| 12:57 | <jdike:#uml> | I had uml-user aliased to uml-devel last time and wondered why they never showed up on uml-user |
| 12:58 | <dgraves:#uml> | ::L:: |
| 12:58 | <dgraves:#uml> | that'd be funny. :) |
| 12:59 | <jdike:#uml> | haha |
| 13:00 | <jdike:#uml> | wait till I alias LMLK to you |
| 13:00 | <jdike:#uml> | LKML |
| 13:00 | <jdike:#uml> | see how funny it is then |
| 13:25 | <dgraves:#uml> | :( |
| 13:25 | <dgraves:#uml> | be like my first kernel submit. |
| 13:25 | <dgraves:#uml> | 348 spam in the next day. |
| 14:15 | -!- | remus [~remus@76.231.178.131] has joined #uml |
| 14:40 | -!- | ram [~ram@pool-71-117-243-240.ptldor.fios.verizon.net] has joined #uml |
| 14:42 | -!- | ram_ [~ram@pool-71-117-243-240.ptldor.fios.verizon.net] has quit [Read error: Connection reset by peer] |
| 14:55 | -!- | dfgf [~ghjgh@213.37.216.213.dyn.user.ono.com] has joined #uml |
| 14:55 | <dfgf:#uml> | hi |
| 14:56 | <dfgf:#uml> | is possible run uml (debian and kernel 32 bits) on a machine 64 bits? :S |
| 14:58 | <jdike:#uml> | yes |
| 14:58 | <dfgf:#uml> | i can not :S |
| 14:58 | <dfgf:#uml> | # ./vmlinux_debian_etch_32b ubd0=/mnt/debian_etch_32bits_uml |
| 14:58 | <dfgf:#uml> | -bash: ./vmlinux_debian_etch_32b: No existe el fichero o el directorio |
| 14:59 | <dfgf:#uml> | kernel is 32b |
| 14:59 | <dfgf:#uml> | local server is 64bits |
| 14:59 | <dfgf:#uml> | i have compile the kernel in a 32 bits machine, because en this machine (64 bits) i can not compile :S |
| 14:59 | <jdike:#uml> | that seems to say vmlinux_debian_etch_32b doesn't exist |
| 15:00 | <dfgf:#uml> | # make linux ARCH=um SUBARCH=i386 |
| 15:00 | <dfgf:#uml> | i get errors |
| 15:00 | <dfgf:#uml> | no |
| 15:00 | <dfgf:#uml> | # /./vmlinux_debian_etch_32b ubd0=/mnt/debian_etch_32bits_uml |
| 15:00 | <dfgf:#uml> | -bash: /./vmlinux_debian_etch_32b: cannot execute binary file |
| 15:00 | <dfgf:#uml> | i'm sorry |
| 15:01 | <jdike:#uml> | ldd it |
| 15:01 | <dfgf:#uml> | # ldd /./vmlinux_debian_etch_32b |
| 15:01 | <dfgf:#uml> | /usr/bin/ldd: line 161: /lib/ld-linux.so.2: cannot execute binary file |
| 15:01 | <dfgf:#uml> | /usr/bin/ldd: line 171: /lib/ld-linux.so.2: cannot execute binary file |
| 15:01 | <dfgf:#uml> | ldd: /lib/ld-linux.so.2 exited with unknown exit code (126) |
| 15:01 | <dfgf:#uml> | uhms |
| 15:01 | <dfgf:#uml> | i have /lib32/ |
| 15:02 | <jdike:#uml> | by default, UML will need some 32-bit shared libraries |
| 15:02 | <dfgf:#uml> | yes i have 32bits librerys installed |
| 15:02 | <jdike:#uml> | you might enable CONFIG_STATIC_LINK and see if that works any better |
| 15:02 | <dfgf:#uml> | may be i need any export for can use 32bits libreries? |
| 15:03 | <jdike:#uml> | ldd is supposed to tell you that |
| 15:03 | <jdike:#uml> | but it seems confused |
| 15:03 | <dfgf:#uml> | aha... i'm trying with CONFIG_STATIC_LINK=y |
| 15:04 | <jdike:#uml> | does the host have 32-bit compatibility configured? |
| 15:06 | <dfgf:#uml> | # Multiarch support |
| 15:06 | <dfgf:#uml> | /lib/x86_64-linux-gnu |
| 15:06 | <dfgf:#uml> | /usr/lib/x86_64-linux-gnu |
| 15:06 | <dfgf:#uml> | yes |
| 15:06 | <dfgf:#uml> | mm |
| 15:06 | <dfgf:#uml> | wait |
| 15:06 | <dfgf:#uml> | /lib32 |
| 15:06 | <dfgf:#uml> | in /etc/ld.so.conf.d/x86_64-linux-gnu.conf |
| 15:07 | <jdike:#uml> | the host kernel needs CONFIG_IA32_EMULATION |
| 15:07 | <dfgf:#uml> | uhms! |
| 15:07 | <dfgf:#uml> | wait please |
| 15:07 | <jdike:#uml> | if that's not there, what's in the filesystem doesn't matter at all |
| 15:08 | <dfgf:#uml> | # CONFIG_IA32_EMULATION is not set |
| 15:08 | <dfgf:#uml> | yes i have this problem |
| 15:08 | <dfgf:#uml> | i'm recompiling this kernel |
| 15:08 | <jdike:#uml> | didn't you just say you can't compile? |
| 15:09 | <dfgf:#uml> | i can not compile um kernel in SUBARCH=i386 |
| 15:10 | <jdike:#uml> | but you can build a 64-bit UML |
| 15:10 | <jdike:#uml> | if you can build the host kernel |
| 15:11 | <dfgf:#uml> | yes but i need 32bits for to use aplications NO opensource |
| 15:11 | <dfgf:#uml> | :\ |
| 15:11 | <dfgf:#uml> | glftpd |
| 15:11 | <jdike:#uml> | OK |
| 15:11 | <dfgf:#uml> | uml 64 bits work very good |
| 15:11 | <dgraves:#uml> | dfgf, did you do a clean before you rebuilt 32 bit? |
| 15:11 | <jdike:#uml> | what kernel version is the host running? |
| 15:12 | <fo0bar:#uml> | jdike: heh, gmail merged your emails to both uml-devel and uml-user, but was a little random about it. from my perspective, you sent patches 9, 7 and 6 to -user, and the rest to -devel :) |
| 15:12 | <jdike:#uml> | hehe |
| 15:12 | <jdike:#uml> | UML - confusing Google since 1999 |
| 15:12 | <dfgf:#uml> | mmmm |
| 15:13 | <dfgf:#uml> | yes i do make mrproper |
| 15:13 | <dfgf:#uml> | 2.6.19.2+grsec |
| 15:13 | <dfgf:#uml> | in uml 2.6.23.8 |
| 15:14 | <jdike:#uml> | there are host bugs which prevent 32-bit UML from running on 64-bit |
| 15:14 | <jdike:#uml> | until recently that is |
| 15:18 | <dfgf:#uml> | yes, i see |
| 15:18 | <dfgf:#uml> | i'm recompiling ... |
| 15:22 | <dfgf:#uml> | # /./vmlinux_debian_etch_32b_2 help |
| 15:22 | <dfgf:#uml> | Core dump limits : |
| 15:22 | <dfgf:#uml> | soft - 0 |
| 15:22 | <dfgf:#uml> | hard - NONE |
| 15:22 | <dfgf:#uml> | Checking that ptrace can change system call numbers...check_ptrace: PTRACE_OLDSETOPTIONS failed: Invalid argument |
| 15:22 | <dfgf:#uml> | :S |
| 15:23 | <dfgf:#uml> | # ldd /./vmlinux_debian_etch_32b |
| 15:23 | <dfgf:#uml> | linux-gate.so.1 => (0xffffe000) |
| 15:23 | <dfgf:#uml> | libutil.so.1 => /lib32/libutil.so.1 (0xf7f9d000) |
| 15:23 | <dfgf:#uml> | libc.so.6 => /lib32/libc.so.6 (0xf7e72000) |
| 15:23 | <dfgf:#uml> | /lib/ld-linux.so.2 (0xf7fad000) |
| 15:25 | <jdike:#uml> | that's one of the host bugs |
| 15:26 | <dfgf:#uml> | uhms |
| 15:26 | <dfgf:#uml> | how to i can fix this problem? :S |
| 15:26 | <dfgf:#uml> | do you know? |
| 15:26 | <jdike:#uml> | put a modern kernel on the box |
| 15:26 | <dgraves:#uml> | if jdike doesn't know, we're all in trouble. |
| 15:27 | <jdike:#uml> | or you can patch your kernel with the fixes |
| 15:27 | <dfgf:#uml> | uhms |
| 15:28 | <dfgf:#uml> | upgrade the kernel is a problem... it is a dedicated server and i dont like to make a lot of changes in the kernel :S |
| 15:28 | <dfgf:#uml> | i dont have iLO connection :S |
| 15:28 | <jdike:#uml> | it needs fixing one way or another |
| 15:29 | <dfgf:#uml> | what's the patchs? |
| 15:29 | <dfgf:#uml> | sorry |
| 15:29 | <dfgf:#uml> | where can i find the patchs? |
| 15:31 | <jdike:#uml> | hold on, I'll get some changelog ids |
| 15:32 | <dfgf:#uml> | Also, 2.6.19.5 is perfectly good (includes the attached patch). |
| 15:32 | <dfgf:#uml> | uhms |
| 15:32 | <dfgf:#uml> | i have 2.6.19.2 for grsec |
| 15:33 | <dfgf:#uml> | UML kernel binaries for 2.6.13 ... 2.6.18 kernels |
| 15:33 | <dfgf:#uml> | gRRRRRr |
| 15:34 | -!- | tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has quit [Ping timeout: 480 seconds] |
| 15:37 | <jdike:#uml> | can't find the PTRACE_OLDSETOPTIONS one |
| 15:37 | <jdike:#uml> | strange |
| 15:37 | <dfgf:#uml> | :S |
| 15:39 | <jdike:#uml> | crap |
| 15:39 | <jdike:#uml> | looking in the wrong place |
| 15:40 | <jdike:#uml> | 47c51dff4af25341c575e5089dbb92d61ac69cdf- fixes PTRACE_OLDSETOPTIONS |
| 15:40 | <dfgf:#uml> | uhms |
| 15:40 | <dfgf:#uml> | what is the kernel version for this patch? |
| 15:42 | <dfgf:#uml> | patch-2.6.21-rc3-git3.log |
| 15:42 | <dfgf:#uml> | :S |
| 15:42 | <dfgf:#uml> | yes i can see |
| 15:43 | -!- | hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has quit [Quit: Leaving] |
| 15:43 | <dfgf:#uml> | i have 2 ways.... kernel upgrade or patch, but patch is for others older kernels |
| 15:43 | <dfgf:#uml> | i must upgrade the kernel |
| 15:43 | <dfgf:#uml> | :\ |
| 15:43 | -!- | tchan [~tchan@c-24-13-84-219.hsd1.il.comcast.net] has joined #uml |
| 15:43 | <dfgf:#uml> | jdike, THANKS YOU!!! ;D |
| 15:44 | <jdike:#uml> | 2.6.23 should have all the fixes in it |
| 15:45 | <dfgf:#uml> | okok, grsec support 2.6.23.9 (beta) |
| 15:45 | <dfgf:#uml> | perfect! |
| 15:47 | -!- | gueb [~gueb@84.77.137.110] has joined #uml |
| 15:47 | <gueb:#uml> | hi |
| 15:58 | * | jdike:#uml growls at ptrace |
| 15:59 | <jdike:#uml> | seems to be singlestepping when it shouldn't |
| 16:04 | <jdike:#uml> | but not in skas0 mode |
| 16:05 | <jdike:#uml> | as Spock would say, "fascinating" |
| 17:00 | -!- | newbie [~jeff@ita4fw1.itasoftware.com] has joined #uml |
| 17:11 | <dgraves:#uml> | live long and prosper! |
| 17:18 | -!- | krau [~cktakahas@200.184.118.132] has quit [Quit: Varei!!!] |
| 17:20 | -!- | dang [~dang@aa-redwall.ghs.com] has quit [Quit: Leaving.] |
| 17:44 | <jdike:#uml> | who asked you? |
| 17:44 | <jdike:#uml> | this bug seems to be living long and prospering |
| 17:47 | <jdike:#uml> | although |
| 17:48 | * | jdike:#uml has an idea |
| 17:59 | -!- | dfgf [~ghjgh@213.37.216.213.dyn.user.ono.com] has quit [Remote host closed the connection] |
| 18:03 | <dgraves:#uml> | :) |
| 19:05 | -!- | silug [~steve@38.104.162.2] has quit [Ping timeout: 480 seconds] |
| 19:08 | -!- | kokoko1 [~Slacker@203.148.64.252] has quit [Read error: Operation timed out] |
| 20:25 | -!- | silug [~steve@38.104.162.2] has joined #uml |
| 20:29 | -!- | dang [~dang@nemesis.fprintf.net] has joined #uml |
| 20:46 | -!- | gueb [~gueb@84.77.137.110] has quit [Quit: leaving] |
| 20:58 | -!- | jdike [~jdike@pool-72-93-84-53.bstnma.fios.verizon.net] has quit [Quit: Leaving] |
| 21:03 | -!- | diatoid [~diatoid@12.44.178.253] has left #uml [Leaving] |
| 21:21 | -!- | balbir [~balbir@122.167.116.196] has quit [Ping timeout: 480 seconds] |
| 21:28 | -!- | balbir [~balbir@122.167.90.166] has joined #uml |
| 23:13 | -!- | remus [~remus@76.231.178.131] has quit [Remote host closed the connection] |
| 23:59 | -!- | VS_ChanLog [~stats@ns.theshore.net] has left #uml [Rotating Logs] |
| 23:59 | -!- | VS_ChanLog [~stats@ns.theshore.net] has joined #uml |
| --- | Log | closed Thu Jan 24 00:00:14 2008 |
![[FSF Associate Member]](/assets/fsf.png){kind=small}
