* Re: [uml-devel] [Pkg-uml-pkgs] Bug#495611: user-mode-linux: page_mapcount(page) went negative!
[not found] <20080818230509.23819.90290.reportbug@carrera.home.cobb.me.uk>
@ 2008-08-19 12:01 ` Mattia Dongili
2008-08-26 20:22 ` Jeff Dike
0 siblings, 1 reply; 5+ messages in thread
From: Mattia Dongili @ 2008-08-19 12:01 UTC (permalink / raw)
To: UML-devel; +Cc: 495611, Graham Cobb
Hello again Jeff,
one more bug report on 2.6.26.2. It looks like .25 is affected as well
are both suffering from it.
On Tue, Aug 19, 2008 at 12:05:09AM +0100, Graham Cobb wrote:
> Package: user-mode-linux
> Version: 2.6.24-1um-1
> Severity: normal
>
> I use Debian lenny guest environment in UML to automatically build software
> for the Nokia Maemo environment.
>
> The overnight all work fine using UML 2.6.24-1um-1. However, UML
> 2.6.25-1um-2 and UML 2.6.26-1um-2 both crash, early in each build, with
> the kernel panic shown below. I can easily reproduce this with my automated
> build system although I have not yet found another stress test (even doing a
> kernel build in the guest environment) which reproduces it.
>
> The crash details are:
>
> Eeek! page_mapcount(page) went negative! (-1)
> page pfn = 2cc
> page->flags = 400
> page->count = 1
> page->mapping = 00000000
> vma->vm_ops = 0x83accc8
> vma->vm_ops->fault = special_mapping_fault+0x0/0x60
> BUG: failure at mm/rmap.c:669/page_remove_rmap()!
> Kernel panic - not syncing: BUG!
>
> EIP: 0073:[<080a407a>] CPU: 0 Not tainted ESP: 007b:bf9ec8fc EFLAGS: 00000246
> Not tainted
> EAX: ffffffda EBX: 00008000 ECX: 001b6000 EDX: 00000005
> ESI: 00000812 EDI: 00000004 EBP: 00000000 DS: 007b ES: 007b
> 277ebd38: [<0809ec74>] notifier_call_chain+0x34/0x70
> 277ebd5c: [<08311f2a>] panic+0x71/0xff
> 277ebd78: [<080cb4e1>] page_remove_rmap+0x151/0x160
> 277ebd90: [<080c3f99>] unmap_vmas+0x2c9/0x600
> 277ebda4: [<08060fd3>] flush_tlb_page+0x113/0x1f0
> 277ebdf8: [<080c7895>] unmap_region+0xa5/0x150
> 277ebe2c: [<080c8a98>] do_munmap+0x1d8/0x290
> 277ebe58: [<080c9544>] mmap_region+0xd4/0x590
> 277ebe90: [<080c7230>] arch_get_unmapped_area+0x0/0x160
> 277ebeb8: [<080b4870>] generic_file_mmap+0x0/0x60
> 277ebec4: [<080c9c1a>] do_mmap_pgoff+0x21a/0x300
> 277ebf00: [<08060846>] sys_mmap2+0x76/0xe0
> 277ebf30: [<080627aa>] handle_syscall+0x8a/0xc0
> 277ebf4c: [<080607d0>] sys_mmap2+0x0/0xe0
> 277ebf78: [<080789ca>] userspace+0x48a/0x510
> 277ebf90: [<08075675>] os_set_thread_area+0x25/0x50
> 277ebfec: [<0805f72d>] fork_handler+0x5d/0x70
>
>
> -- System Information:
> Debian Release: lenny/sid
> APT prefers testing
> APT policy: (900, 'testing')
> Architecture: i386 (i686)
>
> Kernel: Linux 2.6.25-2-686 (SMP w/1 CPU core)
> Locale: LANG=en_IE@euro, LC_CTYPE=en_IE@euro (charmap=ISO-8859-15) (ignored: LC_ALL set to en_IE@euro)
> Shell: /bin/sh linked to /bin/bash
>
> Versions of packages user-mode-linux depends on:
> ii uml-utilities 20070815-1.1 User-mode Linux (utility programs)
>
> user-mode-linux recommends no packages.
>
> Versions of packages user-mode-linux suggests:
> ii konsole [x-terminal- 4:3.5.9.dfsg.1-2+b1 X terminal emulator for KDE
> pn linux-patch-skas <none> (no description available)
> pn rootstrap <none> (no description available)
> ii rxvt [x-terminal-emu 1:2.6.4-14 VT102 terminal emulator for the X
> pn slirp <none> (no description available)
> pn user-mode-linux-doc <none> (no description available)
> ii xterm [x-terminal-em 235-1 X terminal emulator
>
> -- no debconf information
>
>
>
> _______________________________________________
> Pkg-uml-pkgs mailing list
> Pkg-uml-pkgs@lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/pkg-uml-pkgs
>
--
mattia
:wq!
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [uml-devel] [Pkg-uml-pkgs] Bug#495611: user-mode-linux: page_mapcount(page) went negative!
2008-08-19 12:01 ` [uml-devel] [Pkg-uml-pkgs] Bug#495611: user-mode-linux: page_mapcount(page) went negative! Mattia Dongili
@ 2008-08-26 20:22 ` Jeff Dike
[not found] ` <200808271605.25584.graham@cobb.uk.net>
2008-09-02 23:01 ` Graham Cobb
0 siblings, 2 replies; 5+ messages in thread
From: Jeff Dike @ 2008-08-26 20:22 UTC (permalink / raw)
To: Mattia Dongili; +Cc: Graham Cobb, 495611, UML-devel
On Tue, Aug 19, 2008 at 09:01:55PM +0900, Mattia Dongili wrote:
> one more bug report on 2.6.26.2. It looks like .25 is affected as well
> are both suffering from it.
Sigh, I was hoping that this wasn't seen on anything later than 2.6.24.
Any chance it can be bisected? Since this is an overnight test, it
would take a week or two, probably.
Jeff
--
Work email - jdike at linux dot intel dot com
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [uml-devel] [Pkg-uml-pkgs] Bug#495611: user-mode-linux: page_mapcount(page) went negative!
[not found] ` <200808271605.25584.graham@cobb.uk.net>
@ 2008-08-27 17:13 ` Jeff Dike
0 siblings, 0 replies; 5+ messages in thread
From: Jeff Dike @ 2008-08-27 17:13 UTC (permalink / raw)
To: Graham Cobb; +Cc: Mattia Dongili, uml-devel
On Wed, Aug 27, 2008 at 04:05:24PM +0100, Graham Cobb wrote:
> However, how would I go about bisecting through the large number of patches
> which are the difference between these? Is there somewhere I can get interim
> kernel versions from?
Set up a git tree, and git-bisect will chose versions for you, telling
you which commit introduced the bug.
See http://lkml.org/lkml/2005/7/30/106
Jeff
--
Work email - jdike at linux dot intel dot com
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [uml-devel] [Pkg-uml-pkgs] Bug#495611: user-mode-linux: page_mapcount(page) went negative!
2008-08-26 20:22 ` Jeff Dike
[not found] ` <200808271605.25584.graham@cobb.uk.net>
@ 2008-09-02 23:01 ` Graham Cobb
2008-09-03 17:21 ` Jeff Dike
1 sibling, 1 reply; 5+ messages in thread
From: Graham Cobb @ 2008-09-02 23:01 UTC (permalink / raw)
To: Jeff Dike; +Cc: Mattia Dongili, 495611, UML-devel
The bisection is complete. It took longer than I expected as many of the
versions git-bisect wanted to test would not build UML without tracking down
some additional patches.
The bug seems to have been introduced with git commit
3963333fe6767f15141ab2dc3b933721c636c212 (uml: cover stubs with a VMA).
Note the bisection didn't quite complete: the previous commit (git commit
42a2b54ce8c7b9d4f418995a7950e7e2e15e52ce (uml: clean up TASK_SIZE usage))
also causes a panic but it is different from the one reported in this bug and
occurs immediately init is started so I presume it is not the same problem.
The version before these two commits works.
With hindsight I suppose I could have guessed that would be the commit: the
crash seemed to be when unmapping a VMA. If my reading of the page flags is
correct the problem is that the page has the PG_reserved flag set, so what is
it doing in the VMA which is being unmapped?
Let me know if you want me to run some more tests, for example if you want to
add some printk's in the code to understand more about what is going on.
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [uml-devel] [Pkg-uml-pkgs] Bug#495611: user-mode-linux: page_mapcount(page) went negative!
2008-09-02 23:01 ` Graham Cobb
@ 2008-09-03 17:21 ` Jeff Dike
0 siblings, 0 replies; 5+ messages in thread
From: Jeff Dike @ 2008-09-03 17:21 UTC (permalink / raw)
To: Graham Cobb; +Cc: Mattia Dongili, 495611, UML-devel
On Wed, Sep 03, 2008 at 12:01:19AM +0100, Graham Cobb wrote:
> The bisection is complete. It took longer than I expected as many of the
> versions git-bisect wanted to test would not build UML without tracking down
> some additional patches.
>
> The bug seems to have been introduced with git commit
> 3963333fe6767f15141ab2dc3b933721c636c212 (uml: cover stubs with a VMA).
Excellent, the backtrace now makes a bit of sense. It looks like the
VMAs at the top of the address space are being overmapped by a call to
mmap. The question is why does it look like there's a big enough hole
there, when it ends up unmapping the stubs in order to make room for
the mmap.
> With hindsight I suppose I could have guessed that would be the commit: the
> crash seemed to be when unmapping a VMA. If my reading of the page flags is
> correct the problem is that the page has the PG_reserved flag set, so what is
> it doing in the VMA which is being unmapped?
A page of kernel code is mapped into the process. This page (and all
other pages containing kernel text) is marked reserved during boot.
> Let me know if you want me to run some more tests, for example if you want to
> add some printk's in the code to understand more about what is going on.
I'm going to need some more information. I'll get back to you with a
patch...
Jeff
--
Work email - jdike at linux dot intel dot com
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2008-09-03 17:22 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20080818230509.23819.90290.reportbug@carrera.home.cobb.me.uk>
2008-08-19 12:01 ` [uml-devel] [Pkg-uml-pkgs] Bug#495611: user-mode-linux: page_mapcount(page) went negative! Mattia Dongili
2008-08-26 20:22 ` Jeff Dike
[not found] ` <200808271605.25584.graham@cobb.uk.net>
2008-08-27 17:13 ` Jeff Dike
2008-09-02 23:01 ` Graham Cobb
2008-09-03 17:21 ` Jeff Dike
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox