From: Kristian S. <Kri...@ik...> - 2003-10-19 15:47:14
|
Hi, I tried many different kernels and they all result in the same behaviour. During the boot UML hangs. The last lines messages: VFS: Mounted root (ext2 filesystem) readonly. Mounted devfs on /dev I noticed 3 processes linux [initd] (or similar). So I attached gdb to each one and extracted stack trace (traces at the bottom). I even tried running the UML under gdb from the start. I was able to trace the problem to execve("/sbin/init",...) in init(). Then I stepped the code for a while and noticed that it ended up in a while(1)-loop inside userspace(). I guess that is still ok? I also noticed that on every iteration of that loop, there was a page fault (just like in stack trace #1 below). Later, I tried running with mode=tt and UML booted ok. My host system is running 2.4.22 kernel with skas-patch for 2.4.21 (http://kernels.usermodelinux.org/host/skas-2.4.21.patch.bz2) on Debian unstable. I can't recall if there were any major problems during the patching. Probably not as I wouldn't have compiled the kernel. CONFIG_PROC_MM is enabled. The UML kernel is 2.4.19 vanilla + uml-patch-2.4.19-51. Any suggestions, how to get skas mode working? If you need more information, I'd be happy to help. Just tell me what information you need. [stack trace #1] #0 0xa00a5cdf in os_write_file (fd=12, buf=<incomplete type>, count=28) at file.c:143 #1 0xa00a1616 in map (fd=12, virt=1073745920, phys=4100096, len=4096, r=1, w=0, x=1) at mem_user.c:36 #2 0xa00a2758 in fix_range (mm=0xa03fe080, start_addr=1073745920, end_addr=1073750016, force=0) at tlb.c:46 #3 0xa00a297a in flush_tlb_range_skas (mm=0xa03fe080, start=1073745920, end=1073750016) at tlb.c:130 #4 0xa009c1a0 in flush_tlb_range (mm=0xa03fe080, start=1073745920, end=1073750016) at tlb.c:36 #5 0xa009c104 in flush_tlb_page (vma=0xa03fc1e0, address=1073746640) at tlb.c:15 #6 0xa0033185 in handle_mm_fault (mm=0xa03fe080, vma=0xa03fc1e0, address=1073746640, write_access=0) at memory.c:914 #7 0xa009c3e2 in handle_page_fault (address=1073746640, ip=0, is_write=0, is_user=1, code_out=<incomplete type>) at trap_kern.c:58 #8 0xa009c539 in segv (address=1073746640, ip=0, is_write=0, is_user=1, sc=0x0) at trap_kern.c:101 #9 0xa00a17ed in handle_segv (pid=394) at process.c:43 #10 0xa00a1af1 in userspace (regs=0xa09fc278) at process.c:142 #11 0xa00a21db in new_thread_handler (sig=10) at process_kern.c:69 #12 <signal handler called> #13 0xa0127421 in kill () [stack trace #2] #0 0xa013e7bd in read () at /home/krisu/virtual/linux-2.4.19/arch/um/include/../kernel/skas/include/uaccess.h:113 #1 0x0000003c in ?? () #2 0xa00d6e81 in io_thread (arg=<incomplete type>) at ubd_user.c:566 #3 0xa013f9a7 in clone () at /home/krisu/virtual/linux-2.4.19/arch/um/include/../kernel/skas/include/uaccess.h:113 [stack trace #3] #0 0xa013eeb6 in poll () at /home/krisu/virtual/linux-2.4.19/arch/um/include/../kernel/skas/include/uaccess.h:113 #1 0xa00958a6 in write_sigio_thread (unused=<incomplete type>) at sigio_user.c:194 #2 0xa013f9a7 in clone () at /home/krisu/virtual/linux-2.4.19/arch/um/include/../kernel/skas/include/uaccess.h:113 BR, -- Kristian Slavov |
From: Jeff D. <jd...@ad...> - 2003-10-19 19:26:04
|
Kri...@ik... said: > My host system is running 2.4.22 kernel with skas-patch for 2.4.21 > (http://kernels.usermodelinux.org/host/skas-2.4.21.patch.bz2) Don't use that patch on 2.4.22. Use my 2.4.19 skas3 patch instead. Jeff |
From: Goetz B. <bo...@bl...> - 2003-10-21 02:06:12
|
On Sun, Oct 19 '03 at 14:51, Jeff Dike wrote: > Kri...@ik... said: > > My host system is running 2.4.22 kernel with skas-patch for 2.4.21 > > (http://kernels.usermodelinux.org/host/skas-2.4.21.patch.bz2) > > Don't use that patch on 2.4.22. Why not? it does not seam to cause any problems on my box. > Use my 2.4.19 skas3 patch instead. Which? The host-skas3.patch from Nov 11 2002 or the host-skas3a-RH8.patch? Just wondering. -- /"\ Goetz Bock at blacknet dot de -- secure mobile Linux everNETting \ / (c) 2003 as GNU FDL 1.1 X [ 1. Use descriptive subjects - 2. Edit a reply for brevity - ] / \ [ 3. Reply to the list - 4. Read the archive *before* you post ] |
From: Jeff D. <jd...@ad...> - 2003-10-21 05:15:53
|
bo...@bl... said: > Why not? it does not seam to cause any problems on my box. This is confusing. I've heard from two people for whom that patch totally doesn't work. For one of them, my skas3 patch works, but for the other, it doesn't. So, I don't know what's going on right now. Jeff |
From: sean w. <sw...@de...> - 2003-10-21 06:49:24
|
for what it's worth, for me the 2.4.21 patch applied to 2.4.22 causes uml to hang at /sbin/init. i applied the 2.4.19 skas to 2.4.22 and it no longer hangs at init, and things seem to be running fine. it even forced me to fix a bug in my kernel module. -sean ----- Original Message ----- From: "Jeff Dike" <jd...@ad...> To: "Goetz Bock" <bo...@bl...> Cc: <use...@li...> Sent: Monday, October 20, 2003 7:52 PM Subject: Re: [uml-user] UML hangs after /sbin/init in skas mode > bo...@bl... said: > > Why not? it does not seam to cause any problems on my box. > > This is confusing. I've heard from two people for whom that patch totally > doesn't work. For one of them, my skas3 patch works, but for the other, > it doesn't. So, I don't know what's going on right now. > > Jeff > > > > ------------------------------------------------------- > This SF.net email is sponsored by OSDN developer relations > Here's your chance to show off your extensive product knowledge > We want to know what you know. Tell us and you have a chance to win $100 > http://www.zoomerang.com/survey.zgi?HRPT1X3RYQNC5V4MLNSV3E54 > _______________________________________________ > User-mode-linux-user mailing list > Use...@li... > https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user > > |
From: Kristian S. <Kri...@ik...> - 2003-10-21 19:39:06
|
Jeff Dike wrote: >Kri...@ik... said: > > >>My host system is running 2.4.22 kernel with skas-patch for 2.4.21 >>(http://kernels.usermodelinux.org/host/skas-2.4.21.patch.bz2) >> >> > >Don't use that patch on 2.4.22. Use my 2.4.19 skas3 patch instead. > > Jeff > > Fresh 2.4.22 + the host-skas3.patch (from the homepage) works. Thanks. -- Kristian Slavov |