[ros-dev] Regressions and bugs

[James Tabor]

I know about this one! It was worse before and I removed the routine
due to the kernel changes and added it back when I thought it was
fixed. ATM just type "cont" at the debug prompt. I'm not sure how it
is reentering even after removing the DCE from the window. Thread lock
should have catch it and it's very random. Only pops up when exiting
explorer. Mixing owned with classed DCE?

On Thu, Apr 15, 2010 at 8:14 AM, Gabriel ilardi
<gabrielilardi at hotmail.it> wrote:
> Last but not least there's a regression not filed yet to my knowledge:
>
> (subsystems/win32/win32k/ntuser/vis.c:75) ATM the Current Window or Parent
> is dead!
> (subsystems/win32/win32k/ntuser/vis.c:75) ATM the Current Window or Parent
> is dead!
> (subsystems/win32/win32k/ntuser/windc.c:738) Not POWNED or CLASSDC
> hwndCurrent -> 20168
> Assertion 'FALSE' failed at subsystems/win32/win32k/ntuser/windc.c line 739
> PuTTYEntered debugger on embedded INT3 at 0x0008:0x808e5e5e.
> kdb:> bt
> Eip:
> <ntoskrnl.exe:e5e5f (lib/rtl/i386/debug_asm.S:33 (DbgBreakPoint at 0))>
> Frames:
> <win32k.sys:7b79e (subsystems/win32/win32k/ntuser/windc.c:739
> (@DceFreeWindowDCE at 4))>
> <win32k.sys:7f69c (subsystems/win32/win32k/ntuser/window.c:488
> (co_UserFreeWindow))>
> <win32k.sys:7fab7 (subsystems/win32/win32k/ntuser/window.c:2747
> (@co_UserDestroyWindow at 4))>
> <win32k.sys:7faed (subsystems/win32/win32k/ntuser/window.c:2771
> (NtUserDestroyWindow at 4))>
> <ntoskrnl.exe:6fb7 (ntoskrnl/ke/i386/traphdlr.c:1564
> (@KiSystemServiceHandler at 8))>
> <ntoskrnl.exe:e041d (ntoskrnl/ke/i386/trap.s:127
> (KiInterruptTemplateDispatch))>