Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

trap 10.04

6 views
Skip to first unread message

kbr_n...@guzzi.demon.nospam.nl

unread,
May 7, 2012, 4:19:14 PM5/7/12
to

10.04 just trapped when I opened a new tab (had 5 already open)



Part of the exeptq dump:

OS2/eCS Version: 2.45
# of Processors: 4
Physical Memory: 2047 mb
Virt Addr Limit: 1536 mb
Exceptq Version: 7.10 (Mar 1 2011)

______________________________________________________________________

Exception C0000005 - Access Violation
______________________________________________________________________

Process: F:\COMM\FIREFOX\FIREFOX.EXE
PID: A3 (163)
TID: 01 (1)
Priority: 200

Filename: F:\COMM\FIREFOX\XUL.DLL
Address: 005B:117F8570 (0001:01468570)
Cause: Attempted to read from 27B9B910
(unallocated memory)
OS2/eCS Version: 2.45
# of Processors: 4
Physical Memory: 2047 mb
Virt Addr Limit: 1536 mb
Exceptq Version: 7.10 (Mar 1 2011)

______________________________________________________________________

Exception C0000005 - Access Violation
______________________________________________________________________

Process: F:\COMM\FIREFOX\FIREFOX.EXE
PID: A3 (163)
TID: 01 (1)
Priority: 200

Filename: F:\COMM\FIREFOX\XUL.DLL
Address: 005B:117F8570 (0001:01468570)
Cause: Attempted to read from 27B9B910
(unallocated memory)


Registers
______________________________________________________________________

EAX : 23A9F000 EBX : 23A9F189 ECX : 00000000 EDX : 27B9B910
ESI : 00000083 EDI : 27B9B910
ESP : 0012F3A4 EBP : 00000015 EIP : 117F8570 EFLG : 00010202
CS : 005B CSLIM: FFFFFFFF SS : 0053 SSLIM: FFFFFFFF
DS : 0053 ES : 0053 FS : 150B GS : 0000

EAX : read/write memory allocated by XUL
EBX : read/write memory allocated by XUL
ECX : not a valid address
EDX : unallocated memory
ESI : not a valid address
EDI : unallocated memory





Cheers, Bjorn.
--
-----------------------------------------------------------
k...@guzzi.demon.nl
-----------------------------------------------------------

Dave Yeo

unread,
May 7, 2012, 9:00:06 PM5/7/12
to
kbr_n...@guzzi.demon.nospam.nl wrote:
> 10.04 just trapped when I opened a new tab (had 5 already open)

It's the bit just below what you posted that is of interest, the call
stack and the first few labels on the stack.
Dave

kbr_n...@guzzi.demon.nospam.nl

unread,
May 8, 2012, 12:19:33 AM5/8/12
to
In <lOidnbmb0Z297TXS...@mozilla.org>, on 05/07/12
Stack Info for Thread 01
______________________________________________________________________

Size Base ESP Max Top
00100000 00130000 -> 0012F3A4 -> 000FD000 -> 00030000

______________________________________________________________________

Call Stack
______________________________________________________________________

EBP Address Module Obj:Offset Nearest Public Symbol
-------- --------- -------- ------------- -----------------------
Invalid EBP 00000015

______________________________________________________________________

Labels on the Stack
______________________________________________________________________

ESP Address Module Obj:Offset Nearest Public Symbol
-------- --------- -------- ------------- -----------------------
0012F47C 10C759F0 XUL 0001:008E59F0
nsViewManager::GetRootWidget (in nsViewManager.o)
0012F490 117F8A1C XUL 0001:01468A1C between
__moz_cairo_os2_surface_paint_window + 12C and
__moz_cairo_os2_surface_set_size - 614 (both in cairo-os2-surface.o)
0012F4A0 115DC8C5 XUL 0001:0124C8C5 between NS_TableDrivenQI +
55 and nsMemory::GetGlobalMemoryService - 2B (in nsISupportsImpl.o and
nsMemory.o)
0012F4D8 10C75560 XUL 0001:008E5560 between
nsIView::COMTypeInfo::kIID + 10 and
nsIInterfaceRequestor::COMTypeInfo::kIID - 20 (both in nsView.o)
0012F4DC 10C75500 XUL 0001:008E5500
ViewWrapper::COMTypeInfo::kIID (in nsView.o)
0012F4F0 10C734B8 XUL 0001:008E34B8 between
nsIView::DestroyWidget + 78 and nsView::GetOffsetTo - 68 (both in
nsView.o)
0012F520 114D2F8E XUL 0001:01142F8E between
nsWindow::DispatchEvent + 3E and nsWindow::ReparentNativeWidget - 12
(both in nsWindow.o)
0012F540 114D71D8 XUL 0001:011471D8 between nsWindow::OnPaint +
608 and nsWindow::Show - 188 (both in nsWindow.o)
0012F600 1FDCE6BA PMMERGE 0004:000FE6BA between Ring0GetNextSysMsg
+ FA and Ring0ClearWakeBit - A
0012F618 1FDCE183 PMMERGE 0004:000FE183 between ReadNextSysMsg + 2F
and PMREQUESTMUTEXSEM - 6D
0012F630 1FDD1906 PMMERGE 0004:00101906 between GetNextSysMsg + 4E
and GetWC2 - 206
0012F694 0FCD1400 KERNEL32 0001:00011400
0012F710 114DAD5E XUL 0001:0114AD5E between
nsWindow::ProcessMessage + 48E and fnwpNSWindow - 1C2 (both in
nsWindow.o)
0012F728 1FDFA7BB PMMERGE 0004:0012A7BB between GetHookToCall + 1E3
and AccelFEnableTranslate - 21
0012F740 114EB3C7 XUL 0001:0115B3C7 between
nsBaseWidget::QueryInterface + 27 and
gfxUnknownSurface::~gfxUnknownSurface - 189 (both in nsBaseWidget.o)
0012F748 114EC200 XUL 0001:0115C200 between
nsISupports::{vtable} + 20 and nsISupports::COMTypeInfo::kIID - 20 (both
in nsBaseWidget.o)
0012F74C 10395F10 XUL 0001:00005F10
nsISupports::COMTypeInfo::kIID (in nsRDFResource.o)
0012F760 115D9367 XUL 0001:01249367 between
nsCOMPtr_base::assign_from_qi + 27 and
nsCOMPtr_base::assign_from_gs_contractid_with_error - 29 (both in
nsCOMPtr.o)
0012F768 10395F10 XUL 0001:00005F10
nsISupports::COMTypeInfo::kIID (in nsRDFResource.o)
0012F790 114DAFBE XUL 0001:0114AFBE between fnwpNSWindow + 9E
and nsWindow::{vtable} - C2 (both in nsWindow.o)
0012F7E0 1FDD59F9 PMMERGE 0004:001059F9 between WIN32DISPATCHMSG +
F5 and CheckLoseBoost - 1B
0012F7F4 114DAF20 XUL 0001:0114AF20 fnwpNSWindow (in
nsWindow.o)

















>Dave

Dave Yeo

unread,
May 8, 2012, 1:18:40 AM5/8/12
to dev-po...@lists.mozilla.org
On 05/07/12 09:19 pm, kbr_n...@guzzi.demon.nospam.nl wrote:
>> >kbr_n...@guzzi.demon.nospam.nl wrote:
>>> >> 10.04 just trapped when I opened a new tab (had 5 already open)

Was Flash active in one of the other tabs?
Dave

kbr_n...@guzzi.demon.nospam.nl

unread,
May 8, 2012, 1:26:33 PM5/8/12
to
In <mailman.26157.13364543...@lists.mozilla.org>, on
05/07/12
Not active, but I had played some flash content.

Dave Yeo

unread,
May 8, 2012, 10:08:32 PM5/8/12
to
kbr_n...@guzzi.demon.nospam.nl wrote:
> In<mailman.26157.13364543...@lists.mozilla.org>, on
> 05/07/12
> at 10:18 PM, Dave Yeo<dave....@gmail.com> said:
>
>> On 05/07/12 09:19 pm, kbr_n...@guzzi.demon.nospam.nl wrote:>>
>>> kbr_n...@guzzi.demon.nospam.nl wrote:
>>>>>>> 10.04 just trapped when I opened a new tab (had 5 already open)
>
>> Was Flash active in one of the other tabs?
>
> Not active, but I had played some flash content.
>

I notice that kernel32 is in the labels on stack list, whether it had
anything to do with the crash I don't know. It appears that the Cairo
surface was corrupted somehow though I'm far from expert at diagnosing
these but there is quite a few interactions between the system and
Firefox, throw in Flash and the chances of something being out of
bounds, corrupted or whatever probably goes up.
I haven't seen anyone else post anything similar so hopefully it's a one
time thing.
Dave

Dave Yeo

unread,
May 8, 2012, 10:14:08 PM5/8/12
to
kbr_n...@guzzi.demon.nospam.nl wrote:
> 0012F600 1FDCE6BA PMMERGE 0004:000FE6BA between Ring0GetNextSysMsg
> + FA and Ring0ClearWakeBit - A

And thanks for taking the time to install the sym files.
Dave

kbr_n...@guzzi.demon.nospam.nl

unread,
May 9, 2012, 5:25:50 PM5/9/12
to
In <l9-dnaHsj8puTzTS...@mozilla.org>, on 05/08/12
You mean the firefox sym files? Or the kernel ones? For the latter thanks
go to Steven, who has an in depth description on his website:-)

Thank you for taking the time to compile and upload the Mozilla stuff!

Dave Yeo

unread,
May 9, 2012, 9:24:48 PM5/9/12
to
kbr_n...@guzzi.demon.nospam.nl wrote:
>> And thanks for taking the time to install the sym files.
> You mean the firefox sym files? Or the kernel ones? For the latter thanks
> go to Steven, who has an in depth description on his website:-)

The OS/2 system ones.
Dave

Steven Levine

unread,
May 9, 2012, 9:54:16 PM5/9/12
to
On Wed, 9 May 2012 21:25:50 UTC, kbr_n...@guzzi.demon.nospam.nl
wrote:

Hi Bjorn,

> >kbr_n...@guzzi.demon.nospam.nl wrote:
> >> 0012F600 1FDCE6BA PMMERGE 0004:000FE6BA between Ring0GetNextSysMsg
> >> + FA and Ring0ClearWakeBit - A

If you would like me to take a look at the .trp file, send it to me
unsullied. :-)

Steven


--
---------------------------------------------------------------------
Steven Levine <ste...@earthlink.bogus.net>
eCS/Warp/DIY etc. www.scoug.com www.ecomstation.com
---------------------------------------------------------------------

Dave Yeo

unread,
May 10, 2012, 12:15:29 AM5/10/12
to
Steven Levine wrote:
> On Wed, 9 May 2012 21:25:50 UTC, kbr_n...@guzzi.demon.nospam.nl
> wrote:
>
> Hi Bjorn,
>
>>> kbr_n...@guzzi.demon.nospam.nl wrote:
>>>> 0012F600 1FDCE6BA PMMERGE 0004:000FE6BA between Ring0GetNextSysMsg
>>>> + FA and Ring0ClearWakeBit - A
>
> If you would like me to take a look at the .trp file, send it to me
> unsullied. :-)
>

Patches uploaded as gecko_10_04_patches.zip
Dave

Steven Levine

unread,
May 10, 2012, 3:21:32 AM5/10/12
to
On Thu, 10 May 2012 04:15:29 UTC, Dave Yeo <dave....@gmail.com>
wrote:

Hi Dave,

> Patches uploaded as gecko_10_04_patches.zip

Thanks. BTW, where do the sdks patches live. If don't seem to have a
directory/.hg directory. I ignored this before because I was not
planning to build anything.

Dave Yeo

unread,
May 10, 2012, 10:33:19 AM5/10/12
to
Steven Levine wrote:
> On Thu, 10 May 2012 04:15:29 UTC, Dave Yeo<dave....@gmail.com>
> wrote:
>
> Hi Dave,
>
>> Patches uploaded as gecko_10_04_patches.zip
>
> Thanks. BTW, where do the sdks patches live. If don't seem to have a
> directory/.hg directory. I ignored this before because I was not
> planning to build anything.
>

Thought of that after uploading them :) comm-central/ldap/sdks.
If not building then it's fine to ignore them, otherwise running
comm-central/client.py checkout will pull it in.
Dave
http://hg.mozilla.org/releases/mozilla-esr10
http://hg.mozilla.org/releases/comm-esr10

Steven Levine

unread,
May 11, 2012, 11:30:10 PM5/11/12
to
On Thu, 10 May 2012 04:15:29 UTC, Dave Yeo <dave....@gmail.com>
wrote:

Hi Dave,

443112-readme-mc-v3.diff may have gotten stale.

I get rejects when it is applied against a checkout of
FIREFOX_10_0_4esr_RELEASE.

Dave Yeo

unread,
May 12, 2012, 12:03:58 AM5/12/12
to
Steven Levine wrote:
> On Thu, 10 May 2012 04:15:29 UTC, Dave Yeo<dave....@gmail.com>
> wrote:
>
> Hi Dave,
>
> 443112-readme-mc-v3.diff may have gotten stale.
>
> I get rejects when it is applied against a checkout of
> FIREFOX_10_0_4esr_RELEASE.
>

That's weird, it applies here,
[G:\cc-release-10\mozilla]hg qpush
applying 443112-readme-mc-v3.diff
patching file browser/installer/package-manifest.in
patching file widget/src/os2/Makefile.in
patching file widget/src/os2/README.OS2
patching file widget/src/os2/README.firefox
patching file widget/src/os2/README.seamonkey
patching file widget/src/os2/README.thunderbird
adding widget/src/os2/README.OS2
browser/installer/package-manifest.in
widget/src/os2/Makefile.in
widget/src/os2/README.OS2
now at: 443112-readme-mc-v3.diff

Did you do hg update -r FIREFOX_10_0_4esr_RELEASE before applying the
patches? If not then try popping them and update to 10_04. I haven't
pulled since TB 10_04 (two commits after FF 10_04 with no changes, just
tags) to stay at the same level as the releases.
Dave

Dave Yeo

unread,
May 12, 2012, 12:10:13 AM5/12/12
to
Dave Yeo wrote:
> Did you do hg update -r FIREFOX_10_0_4esr_RELEASE before applying the
> patches? If not then try popping them and update to 10_04. I haven't
> pulled since TB 10_04 (two commits after FF 10_04 with no changes, just
> tags) to stay at the same level as the releases.

Rereading your message, I see that you checked out the correct version.
How are you applying the patches? What is the error? Mercurial will
convert line endings (screwing up the REXX files) if you let it.
Dave

Dave Yeo

unread,
May 12, 2012, 12:11:44 AM5/12/12
to
Dave Yeo wrote:
> Mercurial will convert line endings (screwing up the REXX files) if you
> let it.

So will ZIP. The patches should generally have Unix line endings.
Dave

Steven Levine

unread,
May 12, 2012, 12:38:11 AM5/12/12
to
On Sat, 12 May 2012 04:03:58 UTC, Dave Yeo <dave....@gmail.com>
wrote:

Hi Dave,

> That's weird, it applies here,

OK.

> Did you do hg update -r FIREFOX_10_0_4esr_RELEASE before applying the
> patches?

Yes. It was pretty clear that this was required, athough there really
should be a cookbook doc for this.

I've also get dirstate corruption, so will probably just revert and
experiment with fixing the repository as a learning experience and
retry the qpush.

>I haven't
> pulled since TB 10_04 (two commits after FF 10_04 with no changes, just
> tags) to stay at the same level as the releases.

Which reminds me. What's the hg command to figure out what you have
checked out.

Steven Levine

unread,
May 12, 2012, 12:39:57 AM5/12/12
to
On Sat, 12 May 2012 04:10:13 UTC, Dave Yeo <dave....@gmail.com>
wrote:

> > pulled since TB 10_04 (two commits after FF 10_04 with no changes, just
> > tags) to stay at the same level as the releases.
>
> Rereading your message, I see that you checked out the correct version.
> How are you applying the patches? What is the error?

Just the typical rejects.

>Mercurial will
> convert line endings (screwing up the REXX files) if you let it.

Doesn't the .hgrc avoid this?

Steven Levine

unread,
May 12, 2012, 12:43:10 AM5/12/12
to
On Sat, 12 May 2012 04:11:44 UTC, Dave Yeo <dave....@gmail.com>
Your .zip is OK except for manifestp_dos_dis_ipc.diff.

FWIW, the wrapper scripts I generally use to apply patches correct the
line endings for just this reason.

Dave Yeo

unread,
May 12, 2012, 1:10:43 AM5/12/12
to dev-po...@lists.mozilla.org
On 05/11/12 09:39 pm, Steven Levine wrote:
> On Sat, 12 May 2012 04:10:13 UTC, Dave Yeo<dave....@gmail.com>
> wrote:
>
>>> pulled since TB 10_04 (two commits after FF 10_04 with no changes, just
>>> tags) to stay at the same level as the releases.
>>
>> Rereading your message, I see that you checked out the correct version.
>> How are you applying the patches? What is the error?
>
> Just the typical rejects.
>
>> Mercurial will
>> convert line endings (screwing up the REXX files) if you let it.
>
> Doesn't the .hgrc avoid this?

It should if the right settings are there and the copy in the repository
has the right line endings.
Dave


Dave Yeo

unread,
May 12, 2012, 1:26:36 AM5/12/12
to dev-po...@lists.mozilla.org
On 05/11/12 09:38 pm, Steven Levine wrote:
> On Sat, 12 May 2012 04:03:58 UTC, Dave Yeo<dave....@gmail.com>
> wrote:
>
> Hi Dave,
>
>> That's weird, it applies here,
>
> OK.
>
>> Did you do hg update -r FIREFOX_10_0_4esr_RELEASE before applying the
>> patches?
>
> Yes. It was pretty clear that this was required, athough there really
> should be a cookbook doc for this.

Yes, I've been thinking about that since I uploaded the latest patches.
To be honest I'm terrible at writing readmes.
Now I've become distracted trying to find the commit that broke 11.

>
> I've also get dirstate corruption, so will probably just revert and
> experiment with fixing the repository as a learning experience and
> retry the qpush.

Yes, the best way is to use mercurial queues, copy the patches to
.hg\patches and do hg qpush -a or for safety, hg qpush -a 2>&1 | tee
qpush.log.
https://developer.mozilla.org/en/Mercurial_Queues

>
>> I haven't
>> pulled since TB 10_04 (two commits after FF 10_04 with no changes, just
>> tags) to stay at the same level as the releases.
>
> Which reminds me. What's the hg command to figure out what you have
> checked out.
>

I'm no expert so I do hg serve -a 192.168.0.2 (adjust for your IP or use
localhost) in the repository and type 192.168.0.2:8000 into the browser.
Then you can look at everything. hg help serve | less for all parametres.
Dave
0 new messages