New windowing code

New windowing code

[Øyvind Harboe]

Where can I find the ChangeLog for Xwin?

Is there a mailing list where I can read what changes that were made and
which files where changed?

I'm running a build that I compiled from 
:pserver:anoncvs@freedesktop.org:/cvs/xorg based upon Harolds
instructions. Is this the repository where you checked in the changes?


(The reason I ask is that I've got a different(nicer) icon in the
taskbar than the one I get if I run the build I get from CygWin
setup.exe installer)

Øyvind

Re: New windowing code

[Alexander Gottwald]

On Sat, 21 Feb 2004, Øyvind Harboe wrote:

> Where can I find the ChangeLog for Xwin?
> 
> Is there a mailing list where I can read what changes that were made and
> which files where changed?

http://pdx.freedesktop.org/cgi-bin/mailman/listinfo/xorg-commit

> I'm running a build that I compiled from 
> :pserver:anoncvs@freedesktop.org:/cvs/xorg based upon Harolds
> instructions. Is this the repository where you checked in the changes?

Yes. it is.

> (The reason I ask is that I've got a different(nicer) icon in the
> taskbar than the one I get if I run the build I get from CygWin
> setup.exe installer)

Harold has another tree where he build the release from. This tree slightly 
differs (only minimal) from the cvs tree. The icon is one of the small 
differences.

bye
	ago
-- 
 Alexander.Gottwald@s1999.tu-chemnitz.de 
 http://www.gotti.org           ICQ: 126018723
 Chemnitzer Linux-Tag 2004 - 6. und 7. März 2004
 http://www.tu-chemnitz.de/linux/tag

Re: New windowing code

[Harold L Hunt II]

Alexander Gottwald wrote:

> On Wed, 25 Feb 2004, Harold L Hunt II wrote:
> 
> 
>>>I changed the condition when IncludeMakefile is defined as sinclude (the
>>>former condition seemed broken) and define HasMakefileSafeInclude in cygwin.cf
> 
> 
> The second was already in CVS. So I only changed the condiation. 
> 
> 
>>Interesting.  Do you have any idea as to why your commit message didn't 
>>go through?
> 
> 
> I've seen the commit message. But i was hidden between same hundred commit 
> messages from kaleb *g*.

That explains it... now I remember deleting about 200 messages because 
they overflowed my account limit :)

Harold

Re: New windowing code

[Alexander Gottwald]

On Wed, 25 Feb 2004, Harold L Hunt II wrote:

> > I changed the condition when IncludeMakefile is defined as sinclude (the
> > former condition seemed broken) and define HasMakefileSafeInclude in cygwin.cf

The second was already in CVS. So I only changed the condiation. 

> 
> Interesting.  Do you have any idea as to why your commit message didn't 
> go through?

I've seen the commit message. But i was hidden between same hundred commit 
messages from kaleb *g*.

http://pdx.freedesktop.org/pipermail/xorg-commit/2004-February/000377.html

Index: Imake.rules
===================================================================
RCS file: /cvs/xorg/xc/config/cf/Imake.rules,v
retrieving revision 1.1.4.3
retrieving revision 1.1.4.3.2.1
diff -u -d -r1.1.4.3 -r1.1.4.3.2.1
--- a/Imake.rules	30 Nov 2003 16:46:49 -0000	1.1.4.3
+++ b/Imake.rules	23 Feb 2004 21:39:04 -0000	1.1.4.3.2.1
@@ -1865,7 +1865,7 @@
  * The double-@ is to ensure no leading spaces on the line.
  */
 #ifndef IncludeMakefile
-#if HasClearmake
+#if HasClearmake || HasMakefileSafeInclude
 #define IncludeMakefile(file) @@sinclude file
 #else
 #if HasBsdMake

ClearMake seems to be a commercial product and has (IMO) nothing to do 
with sinclude. 

bye
	ago
-- 
 Alexander.Gottwald@s1999.tu-chemnitz.de 
 http://www.gotti.org           ICQ: 126018723
 Chemnitzer Linux-Tag 2004 - 6. und 7. März 2004
 http://www.tu-chemnitz.de/linux/tag

Re: New windowing code

[Harold L Hunt II]

Alexander,

I have messages from before and after your commit, and the mailing list 
archive seems to have caught your commit too:

http://pdx.freedesktop.org/pipermail/xorg-commit/2004-February/000376.html
http://pdx.freedesktop.org/pipermail/xorg-commit/2004-February/000377.html

Did you happen to get a copy of those sent to you?  I wonder if my copy 
is sitting in a queue somewhere for the next week...

Harold

Re: New windowing code

[Harold L Hunt II]

Alexander Gottwald wrote:

> On Wed, 25 Feb 2004, Harold L Hunt II wrote:
> 
> 
>>Alexander,
>>
>>I didn't see a commit message for this, nor is there a log of it in CVS:
> 
> 
> It is in the config/cf directory. The Makefile template uses the macro
> IncludeMakefile(DependFileName) to include the dependencies but this 
> macros was defined as /**/ for a long time. So no dependency information
> was used in the makefile.
> 
> I changed the condition when IncludeMakefile is defined as sinclude (the
> former condition seemed broken) and define HasMakefileSafeInclude in cygwin.cf

Interesting.  Do you have any idea as to why your commit message didn't 
go through?

Harold

Re: New windowing code

[Alexander Gottwald]

On Wed, 25 Feb 2004, Harold L Hunt II wrote:

> Alexander,
> 
> I didn't see a commit message for this, nor is there a log of it in CVS:

It is in the config/cf directory. The Makefile template uses the macro
IncludeMakefile(DependFileName) to include the dependencies but this 
macros was defined as /**/ for a long time. So no dependency information
was used in the makefile.

I changed the condition when IncludeMakefile is defined as sinclude (the
former condition seemed broken) and define HasMakefileSafeInclude in cygwin.cf

bye
	ago
-- 
 Alexander.Gottwald@s1999.tu-chemnitz.de 
 http://www.gotti.org           ICQ: 126018723
 Chemnitzer Linux-Tag 2004 - 6. und 7. März 2004
 http://www.tu-chemnitz.de/linux/tag

Re: New windowing code

[Harold L Hunt II]

Alexander Gottwald wrote:
> Alexander Gottwald wrote:
> 
> 
>>On Mon, 23 Feb 2004, Takuma Murakami wrote:
>>
>>
>>>Looks 'sinclude ...' is missing.
>>
>>Hm, I thought I had checked that in. I'll check tonight.
> 
> 
> I forgot th check it in. I've done it now. make Makefile should create
> a Makefile which honours the dependencies created by make depend.

Alexander,

I didn't see a commit message for this, nor is there a log of it in CVS:

http://pdx.freedesktop.org/cgi-bin/viewcvs.cgi/xc/programs/Xserver\
/hw/xwin/?cvsroot=xorg&only_with_tag=CYGWIN&sortby=date#dirlist

Harold

Re: New windowing code

[Alexander Gottwald]

Alexander Gottwald wrote:

> On Mon, 23 Feb 2004, Takuma Murakami wrote:
>
> > Looks 'sinclude ...' is missing.
>
> Hm, I thought I had checked that in. I'll check tonight.

I forgot th check it in. I've done it now. make Makefile should create
a Makefile which honours the dependencies created by make depend.

bye
	ago


NP: Blutengel - Our Time
-- 
 Alexander.Gottwald@informatik.tu-chemnitz.de
 http://www.gotti.org           ICQ: 126018723

Re: New windowing code

[Alexander Gottwald]

On Mon, 23 Feb 2004, Takuma Murakami wrote:

> Looks 'sinclude ...' is missing.

Hm, I thought I had checked that in. I'll check tonight.

bye
	ago
-- 
 Alexander.Gottwald@s1999.tu-chemnitz.de 
 http://www.gotti.org           ICQ: 126018723
 Chemnitzer Linux-Tag 2004 - 6. und 7. März 2004
 http://www.tu-chemnitz.de/linux/tag

Re: New windowing code

[Takuma Murakami]

Alexander,

> Hm. It seems the dependencies are broken for you.
> 
> Can you check this is in hw/xwin/Makefile?
> 
> # ----------------------------------------------------------------------
> # dependencies generated by makedepend
> 
> sinclude Makefile.dep
> 
> if this is present then just do "make depend" and a simple make should
> compile even files where only includes were changed.

I got the session:

~/x-devel/build/debug/programs/Xserver/hw/xwin>tail -4 Makefile

# ----------------------------------------------------------------------
# dependencies generated by makedepend

~/x-devel/build/debug/programs/Xserver/hw/xwin>

Looks 'sinclude ...' is missing.  In fact, I don't understand
well about the build process.  For the first time I did
cd ~/x-devel/build/debug && makeg World
and since then, I only do
cd ~/x-devel/build/debug/programs/Xserver && makeg XWin.exe
for every rebuild.  Could you tell me what shall be done for
appropriate rebuilds?

Takuma Murakami

Re: New windowing code

[Alexander Gottwald]

Øyvind Harboe wrote:

> - Clipboard causes winapps to deadlock

I've noticed some of the deadlocks too. but reproducing them is very hard
since they only happen after long and intensive usage.

bye
    ago
-- 
 Alexander.Gottwald@informatik.tu-chemnitz.de
 http://www.gotti.org           ICQ: 126018723

Re: New windowing code

[Alexander Gottwald]

Takuma Murakami wrote:

> I confirmed the problem after 'make clean', with
> Harold's correction.

Hm. It seems the dependencies are broken for you.

Can you check this is in hw/xwin/Makefile?

# ----------------------------------------------------------------------
# dependencies generated by makedepend

sinclude Makefile.dep

if this is present then just do "make depend" and a simple make should
compile even files where only includes were changed.


bye
    ago
NP: Assemblage 23 - Naked
-- 
 Alexander.Gottwald@informatik.tu-chemnitz.de
 http://www.gotti.org           ICQ: 126018723

New windowing code

[Øyvind Harboe]

>- I saw the crash below in the my text mode GDB session when maximizing
>Evolution. Don't know what to make of it. 

This crash happened only once and I was not able to reproduce it. 

Not surprising really.

All the remaining bugs in cygwin-xfree are at this 
point hard to reproduce and relatively rare, because all the others have been 
fixed! 

:-)

Here is my list of areas where I'm trying to corner those rare and hard
to reproduce bugs:

- Multimonitor rendering SEGFAULTs. Harold has probably fixed a couple here.
- Clipboard causes winapps to deadlock



Øyvind

Re: New windowing code

[Massimiliano Hofer]

On Sunday February 22 2004 3:37 pm, Takuma Murakami wrote:

> I want to leave it because I don't have the software and
> the debug without symptoms is pretty hard.  A question I have
> is: was the problem observed with older versions or is newly
> introduced in release-45?

I understand the difficulty. The problem seems to be new. In the coming days 
I'll do more testing and see if I can pin down a more accurate and 
reproducible scenario.

-- 
Bye,
   Massimiliano Hofer
        Nucleus

Re: New windowing code

[Takuma Murakami]

Massimiliano,

> The new code seems to fix the problems for me. I had easily reproducible 
> problems and the new version never exhibited them during my tests.

That's good.  Thank you for reporting.

> I had just one minor glitch in non maximized windows with shadowed popup 
> menues, where, for some reason, a couple of lines is drawn on the main window 
> title bar. This happened only with CrossOver applications and may be due to 
> the way commands are converted from GDI to X, producing unusal usage 
> patterns. If you need snapshots or debug info, please let me know.

I want to leave it because I don't have the software and
the debug without symptoms is pretty hard.  A question I have
is: was the problem observed with older versions or is newly
introduced in release-45?

Takuma Murakami

Re: New windowing code

[Massimiliano Hofer]

On Saturday February 21 2004 2:34 pm, Takuma Murakami wrote:

> This should fix the following problems about windowing.
> 1) Maximize->Minimize->Maximize loses the window contents.
> 2) After 1, window size cannot be correctly restored.
> 3) Left and top borders does not resize but move the window.
> 4) Moved window occasionally fails to update its contents.
>
> A note for test: if you compile it from CVS, you need to do
> 'touch winallpriv.c' after 'cvs update' or XWin.exe crashes
> on exiting.
>
> Any comments, questions, test/bug reports will be
> greatly appreciated.

The new code seems to fix the problems for me. I had easily reproducible 
problems and the new version never exhibited them during my tests.

I had just one minor glitch in non maximized windows with shadowed popup 
menues, where, for some reason, a couple of lines is drawn on the main window 
title bar. This happened only with CrossOver applications and may be due to 
the way commands are converted from GDI to X, producing unusal usage 
patterns. If you need snapshots or debug info, please let me know.

Great work. Thank you!

-- 
Bye,
   Massimiliano Hofer
        Nucleus

Re: New windowing code

[Takuma Murakami]

> > - To get things to compile, I had to make the attached smallchange.txt
> 
> I think that change is not necessary.  I changed the type of
> HandleCustomWM_COMMAND from int to Bool in my previous
> commit on few days ago.  Is the definition of the function in
> winprefs.c typed as Bool?

I confirmed the problem after 'make clean', with
Harold's correction.

Takuma Murakami

Re: New windowing code

[Takuma Murakami]

Øyvind,

Thank you for testing and reporting.

> - To get things to compile, I had to make the attached smallchange.txt

I think that change is not necessary.  I changed the type of
HandleCustomWM_COMMAND from int to Bool in my previous
commit on few days ago.  Is the definition of the function in
winprefs.c typed as Bool?

> - I saw the crash below in the my text mode GDB session when maximizing
> Evolution. Don't know what to make of it. 

I cannot reproduce this crash but it can be caused by bad usage
of ConfigureWindow, because I coded it without any document.
Indeed, I really want a document about the function.  Anyway,
could you try the release-45 server distributed via Cygwin
setup and confirm the crash is still there?

> - I was able to resize Evolution window by dragging the upper left
> corner of the window(which is listed as fixed in the posting by Takuma
> Murakami)

Thanks for the positive report.

> - I was not able to reproduce the duplicate windows + wrong offset for
> mouse cursor problem after maximizing Evolution on my second monitor. To
> say that the problem is definitely gone is a bit premature, since the
> problem was intermittant.

Such kind of things are supposed to be fixed.

Takuma Murakami

Re: New windowing code

[Harold L Hunt II]

Great work Takuma!

Harold

Takuma Murakami wrote:

> I committed a couple of changes on windowing in multiwindow
> mode to the xorg CVS tree.  Here I mean move, resize,
> minimize, maximize, and restore operations as 'windowing'.
> 
> This should fix the following problems about windowing.
> 1) Maximize->Minimize->Maximize loses the window contents.
> 2) After 1, window size cannot be correctly restored.
> 3) Left and top borders does not resize but move the window.
> 4) Moved window occasionally fails to update its contents.
> 
> A note for test: if you compile it from CVS, you need to do
> 'touch winallpriv.c' after 'cvs update' or XWin.exe crashes
> on exiting.
> 
> Any comments, questions, test/bug reports will be
> greatly appreciated.
> 
> Takuma Murakami
> 
> 

Re: New windowing code

[Harold L Hunt II]

Øyvind Harboe wrote:
> I did a cvs update and took the new XWin.exe for a spin.
> 
> - To get things to compile, I had to make the attached smallchange.txt

Good catch.  Instead of changing the return type from Bool to int I just 
included Xdefs.h in winprefs.h.

> - I saw the crash below in the my text mode GDB session when maximizing
> Evolution. Don't know what to make of it. 

Don't know either.

> - I was able to resize Evolution window by dragging the upper left
> corner of the window(which is listed as fixed in the posting by Takuma
> Murakami)

Good.

> - I was not able to reproduce the duplicate windows + wrong offset for
> mouse cursor problem after maximizing Evolution on my second monitor. To
> say that the problem is definitely gone is a bit premature, since the
> problem was intermittant.

Interesting.


Harold

New windowing code

[Øyvind Harboe]

[-- Attachment #1: Type: text/plain, Size: 7137 bytes --]

I did a cvs update and took the new XWin.exe for a spin.

- To get things to compile, I had to make the attached smallchange.txt

- I saw the crash below in the my text mode GDB session when maximizing
Evolution. Don't know what to make of it. 

- I was able to resize Evolution window by dragging the upper left
corner of the window(which is listed as fixed in the posting by Takuma
Murakami)

- I was not able to reproduce the duplicate windows + wrong offset for
mouse cursor problem after maximizing Evolution on my second monitor. To
say that the problem is definitely gone is a bit premature, since the
problem was intermittant.




startxwin.bat - Starting on Windows NT/2000/XP/2003
GNU gdb 2003-09-20-cvs (cygwin-special)
Copyright 2003 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for
details.
This GDB was configured as "i686-pc-cygwin"...
(gdb) run
Starting program: /usr/X11R6/bin/XWin.exe -multiwindow -clipboard
-multiplemonit
ors
--Type <return> to continue, or q <return> to quit--

Program received signal SIGSEGV, Segmentation fault.
0x004d3228 in miComputeClips (pParent=0x103a1800, pScreen=0x100f8e80,
    universe=0x22e6a4, kind=VTMove, exposed=0x22e684) at mivaltree.c:259
259         dx = pParent->drawable.x -
pParent->valdata->before.oldAbsCorner.x;
(gdb) thread apply bt all
(gdb) thread apply all bt

Thread 13 (thread 2848.0x960):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c534 in ntdll!ZwWaitForSingleObject () from ntdll.dll
#2  0x77e7a62d in WaitForSingleObjectEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#3  0x00000468 in ?? ()

Thread 12 (thread 2848.0xf7c):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c534 in ntdll!ZwWaitForSingleObject () from ntdll.dll
#2  0x71a51f97 in DSA_DestroyCallback ()
   from /cygdrive/c/WINDOWS/System32/MSWSOCK.DLL
#3  0x0421fdfc in ?? ()
#4  0x0421fe90 in ?? ()
#5  0x10184bfc in ?? ()
#6  0x0421fe80 in ?? ()
#7  0x6efeb39c in ?? ()
#8  0x01c3f8c5 in ?? ()
#9  0xffffffff in ?? ()
#10 0x7fffffff in ?? ()
#11 0x002523d8 in ?? ()

Thread 11 (thread 2848.0xb18):
--Type <return> to continue, or q <return> to quit--
#0  0x7ffe0304 in ?? ()
#1  0x77f5c524 in ntdll!ZwWaitForMultipleObjects () from ntdll.dll
#2  0x77e75ee0 in WaitForMultipleObjectsEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#3  0x00000003 in ?? ()
#4  0x0401e978 in ?? ()
#5  0x00000001 in ?? ()

Thread 10 (thread 2848.0xe5c):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c534 in ntdll!ZwWaitForSingleObject () from ntdll.dll
#2  0x71a51f97 in DSA_DestroyCallback ()
   from /cygdrive/c/WINDOWS/System32/MSWSOCK.DLL
#3  0x03e1fdfc in ?? ()
#4  0x03e1fe90 in ?? ()
#5  0x10191c94 in ?? ()
#6  0x03e1fe80 in ?? ()
#7  0x6efeb39c in ?? ()
#8  0x01c3f8c5 in ?? ()
#9  0xffffffff in ?? ()
#10 0x7fffffff in ?? ()
#11 0x00252208 in ?? ()

Thread 9 (thread 2848.0x698):
--Type <return> to continue, or q <return> to quit--
#0  0x7ffe0304 in ?? ()
#1  0x77f5c534 in ntdll!ZwWaitForSingleObject () from ntdll.dll
#2  0x77e7a62d in WaitForSingleObjectEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#3  0x000005a4 in ?? ()

Thread 8 (thread 2848.0x218):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c534 in ntdll!ZwWaitForSingleObject () from ntdll.dll
#2  0x77e7a62d in WaitForSingleObjectEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#3  0x000005b8 in ?? ()

Thread 7 (thread 2848.0x42c):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c024 in ntdll!ZwRemoveIoCompletion () from ntdll.dll
#2  0x71a51a1b in DSA_DestroyCallback ()
   from /cygdrive/c/WINDOWS/System32/MSWSOCK.DLL
#3  0x0381ffac in ?? ()
#4  0x0381ffb0 in ?? ()
#5  0x71a5b740 in MSWSOCK!AcceptEx ()
   from /cygdrive/c/WINDOWS/System32/MSWSOCK.DLL
#6  0x00240608 in ?? ()
#7  0x77f9e28f in ntdll!RtlpNtEnumerateSubKey () from ntdll.dll
--Type <return> to continue, or q <return> to quit--
#8  0x00000003 in ?? ()
#9  0x00000004 in ?? ()
#10 0x0000ffff in ?? ()
#11 0x000000b8 in ?? ()

Thread 6 (thread 2848.0xce0):
#0  0x7ffe0304 in ?? ()
#1  0x77d43a09 in SetBitmapDimensionEx ()
   from /cygdrive/c/WINDOWS/system32/user32.dll
#2  0x77d443b5 in USER32!GetMessageA ()
   from /cygdrive/c/WINDOWS/system32/user32.dll
#3  0x61091f03 in wait4 () from /usr/bin/cygwin1.dll
#4  0x61002dae in forkpty () from /usr/bin/cygwin1.dll
#5  0x77e7d33b in RegisterWaitForInputIdle ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#6  0x610cfb50 in cygwin1!__ctype_ptr ()
#7  0x019e4200 in ?? ()
#8  0x610cfb50 in cygwin1!__ctype_ptr ()
#9  0x7ffd9000 in ?? ()
#10 0x0361ffc0 in ?? ()
#11 0x00000007 in ?? ()
#12 0xffffffff in ?? ()
#13 0x77e94809 in SetThreadExecutionState ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
--Type <return> to continue, or q <return> to quit--

Thread 5 (thread 2848.0xf28):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c524 in ntdll!ZwWaitForMultipleObjects () from ntdll.dll
#2  0x77e75ee0 in WaitForMultipleObjectsEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#3  0x00000001 in ?? ()
#4  0x0341fe88 in ?? ()
#5  0x00000001 in ?? ()

Thread 4 (thread 2848.0xd84):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c524 in ntdll!ZwWaitForMultipleObjects () from ntdll.dll
#2  0x77e75ee0 in WaitForMultipleObjectsEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#3  0x00000002 in ?? ()
#4  0x0321e058 in ?? ()
#5  0x00000001 in ?? ()

Thread 3 (thread 2848.0xa38):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c534 in ntdll!ZwWaitForSingleObject () from ntdll.dll
#2  0x77e7a62d in WaitForSingleObjectEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
--Type <return> to continue, or q <return> to quit--
#3  0x000005b0 in ?? ()

Thread 2 (thread 2848.0xde0):
#0  0x7ffe0304 in ?? ()
#1  0x77f5c524 in ntdll!ZwWaitForMultipleObjects () from ntdll.dll
#2  0x77e75ee0 in WaitForMultipleObjectsEx ()
   from /cygdrive/c/WINDOWS/system32/KERNEL32.DLL
#3  0x00000003 in ?? ()
#4  0x0191fe58 in ?? ()
#5  0x00000001 in ?? ()

Thread 1 (thread 2848.0x820):
#0  0x004d3228 in miComputeClips (pParent=0x103a1800,
pScreen=0x100f8e80,
    universe=0x22e6a4, kind=VTMove, exposed=0x22e684) at mivaltree.c:259
#1  0x004d4004 in miValidateTree (pParent=0x10110258, pChild=0x103a1800,
    kind=VTMove) at mivaltree.c:717
#2  0x004d1764 in miMoveWindow (pWin=0x1035c8c0, x=1658, y=56,
pNextSib=0x0,
    kind=VTMove) at miwindow.c:552
#3  0x00465436 in winMoveWindowMultiWindow (pWin=0x1035c8c0, x=1658,
y=56,
    pSib=0x0, kind=VTMove) at winmultiwindowwindow.c:870
#4  0x00406e70 in ConfigureWindow (pWin=0x1035c8c0, mask=15,
vlist=0x22e874,
    client=0x10365228) at window.c:2439
#5  0x00465713 in winAdjustXWindow (pWin=0x1035c8c0, hwnd=0x370286)
    at winmultiwindowwindow.c:977
--Type <return> to continue, or q <return> to quit--
#6  0x0045e847 in winTopLevelWindowProc (hwnd=0x370286, message=3,
wParam=0,
    lParam=3671674) at winmultiwindowwndproc.c:804
#7  0x77d43a50 in SetBitmapDimensionEx ()
   from /cygdrive/c/WINDOWS/system32/user32.dll
#8  0x00370286 in ?? ()
#9  0x00000003 in ?? ()
(gdb)




[-- Attachment #2: smallchange.txt --]
[-- Type: text/x-patch, Size: 481 bytes --]

Index: winprefs.h
===================================================================
RCS file: /cvs/xorg/xc/programs/Xserver/hw/xwin/winprefs.h,v
retrieving revision 1.1.6.2
diff -u -r1.1.6.2 winprefs.h
--- winprefs.h	17 Feb 2004 06:26:17 -0000	1.1.6.2
+++ winprefs.h	21 Feb 2004 22:08:34 -0000
@@ -132,7 +132,7 @@
 HandleCustomWM_INITMENU(unsigned long hwndIn,
 			unsigned long hmenuIn);
 
-Bool
+int
 HandleCustomWM_COMMAND (unsigned long hwndIn,
 			int           command);
 

New windowing code

[Takuma Murakami]

I committed a couple of changes on windowing in multiwindow
mode to the xorg CVS tree.  Here I mean move, resize,
minimize, maximize, and restore operations as 'windowing'.

This should fix the following problems about windowing.
1) Maximize->Minimize->Maximize loses the window contents.
2) After 1, window size cannot be correctly restored.
3) Left and top borders does not resize but move the window.
4) Moved window occasionally fails to update its contents.

A note for test: if you compile it from CVS, you need to do
'touch winallpriv.c' after 'cvs update' or XWin.exe crashes
on exiting.

Any comments, questions, test/bug reports will be
greatly appreciated.

Takuma Murakami