The main motivation to port all that back is to get rid of Win32K-BSODs 0x50 like CORE-13907
that could be triggered by user-mode-applications.
Technical reason was CORE-14857 "RtlImageNtHeaderEx needs SEH" and issues with hooks+callbacks.
All other tickets mentioned in here were regressions that we experienced on the long
journey towards perfecting the fixes on master head, and those were not happening yet in the older builds,
so they are mentioned in here only to allow tracking where that code went into.
Thanks to Jim Tabor, Mark Jansen & Thomas Faber
By squashed backmerge of the following commits:
0.4.15-dev-3440-g a89844f740
0.4.15-dev-3430-g 9cff384c22
0.4.14-dev-1287-g 568b6d0558
0.4.14-dev-1255-g 177ae91bf6
0.4.14-dev-1241-g 915a5764a9
0.4.14-dev-1240-g d8add40e89
0.4.14-dev-1002-g feb7df42b4
JIRA-tickets: CORE-13907, CORE-16769, CORE-14857, CORE-17856, CORE-17857, partially CORE-13019.
We also had some more dupes of CORE-13907 in JIRA, but I linked all of them (transitively)
towards the initial report, and resolved them without setting a fixVer for them.
In more detail:
--------------
[USER32] Fix BSOD 0x50 in 'WineVDM + Castle Of Winds' CORE-17856 CORE-17857
CORE-17856 BSOD 0x50 when starting Castle of the Winds second time, with WineVDM
CORE-17857 BSOD 0x50 on shutdown after closing Castle of the Winds with WineVDM
0.4.15-dev-3430-g 9cff384c22
0.4.15-dev-3440-g a89844f740
--------------
[NTUSER] Fix Strings and Format to Hooks
Allocate heap instead of data segment to be used for callbacks on user side.
Move and correct initial hook call out setup. Use it in more than one hook call.
This fixes issues with strings out of alignment and use of kernel pointers.
See CORE-13907 (HXD-portable BSOD) and CORE-16769 (HXD-portable exception)
Small wow update.
cherry picked from commit 0.4.14-dev-1287-g 568b6d0558
--------------
[RTL] Introduce RtlpImageNtHeader,
which implements the required functionality.
ntdll and ntoskrnl now have a wrapper for this, with SEH.
This protects the function against malformed / bad images,
whilst still being able to use the code in freeldr et al.
Idea from Thomas.
CORE-14857
cherry picked from commit 0.4.14-dev-1255-g 177ae91bf6
--------------
[WIN32SS] Form Sanity to Hook Callbacks
Fix WH_CALLWNDPROC/RET data to user hook calls.
Helps with CORE-13907 "(HXD-portable BSOD)"
Fixes the logging of "(win32ss/user/ntuser/callback.c:748) err: Failure to make Callback! Status 0xc00000fd" within CORE-13019, but not the whole ticket.
cherry picked from commit 0.4.14-dev-1241-g 915a5764a9
--------------
[USER32] Fix null return.
See CORE-16769 "HxD 1.7.7.0 portable unhandled exception"
cherry picked from commit 0.4.14-dev-1240-g d8add40e89
--------------
[WIN32SS] Fix a typo in dbg print
cherry picked from commit 0.4.14-dev-1002-g feb7df42b4
[USER32] GetQueueStatus() should not return 0 when passing QS_ALLINPUT flag that includes QS_RAWINPUT (#4115)
GTK applications call GetQueueStatus(QS_ALLINPUT), where QS_ALLINPUT
specifies the QS_RAWINPUT flag as well, when these are compiled for
Windows XP+, and they expect the call to succeed on this platform.
On one side, ReactOS does not currently support this flag at all, but
since it claims to be XP/2003-compatible, applications may implicitly
expect the flag to be supported by GetQueueStatus() and the function
*NOT* failing when this flag is set.
(Later GTK apps don't care and just call GetQueueStatus(QS_ALLINPUT)
that includes QS_RAWINPUT, and therefore would fail as well on e.g.
Windows 2000...)
Otherwise, an observable effect is that some versions of libgdk-win32-2.0.0.dll
enter into an infinite loop when calling GetQueueStatus(QS_ALLINPUT),
since this call always failed on ReactOS.
On the other side, however, we should honour our winetests that handle
the presence of the QS_RAWINPUT flag and behave differently accordingly.
But since we do not support QS_RAWINPUT yet, we should keep their old
behaviour where QS_RAWINPUT is unused.
Thus, in order to accomodate both sides, we don't fail the GetQueueStatus()
call, but just set the ERROR_INVALID_FLAGS last error and continue it.
This fixes:
'All user32:TrackMouseEvent tests'
CORE-15686: Gimp 2.8.22 from rapps hangs after startup with 90%-100% CPU usage
and many dupes of that ticket like:
CORE-17551: Ardour hangs at the first start (GTK2 issue)
CORE-15151: newer "Inkscape 0.92.3" is unusable due to WIN32SS message queue lockup
CORE-14086: RawTherapee 5.0-r1-gtk2 fails to start (using 100% CPU)
CORE-11850: Geany 1.28 hangs with endless MsqSendMessage timed out
CORE-8475: Wireshark hangs after launch
It will also slightly appease, but not entirely fix:
CORE-8217: 3D Text ScreenSaver freezes system
Fix picked from 0.4.15-dev-3407-g 9c4397afdf
---------------------
[NTUSER] Do not remove message from the Msg Queue if it is not for us. (#4129)
CORE-8217 This part of the fix keeps the buttons working (Cancel/Ok/top-Right-X) even under high CPU-load
Patch from 'I_Kill_Bugs' contributor.
Fix picked from 0.4.15-dev-3499-g 7d1b50394b
---------------------
[3DTEXT] Fix 3dtext.scr using near 100% CPU (#4125) CORE-17866, CORE-8217
Fix picked from 0.4.15-dev-3443-g 5c9fdcb1de
Fixes drawing of caption bar buttons (min/max/close in Mozilla Firefox titlebar when its menubar is disabled).
See CORE-8808.
The fix is a partial port of 0.4.14-dev-1314-g acf135d16d
I left out all changes to the logging and took only the essence of this patch.
LOAD_LIBRARY_AS_IMAGE_RESOURCE and LOAD_LIBRARY_SEARCH_SYSTEM32 are Vista+
This fixes some blurry icons in systray when 2k3sp2 kernel32.dll is used in ros CORE-12004
That part in user32 was a regression of SVN r71609 == git a44dfe6c76
While we are at it we do fix some other modules as well, that used those NT6+flags.
The fix was picked from 0.4.15-dev-788-g a04831677e
This prevents the tracking menu from erroneously closing itself right after it was opened in some cases.
Fixes CORE-17338 which got unhidden by 0.4.15-dev-1126-g 58b0558f94
And fixes CORE-15760 which got unhidden by SVN r74972 == git 19dd22d422
Since both symptoms look very similar but the unhiding revisions did differ,
that could mean we do have some kind of race condition here.
I guess this fix is more like a workaround.
2 Fixes picked:
0.4.15-dev-3313-g 6417b2323d [NTUSER][USER32] Workaround tracking menu on CORE-17338 CORE-15760 (#4048)
to prevent the menu from auto-closing
and
0.4.15-dev-3297-g 037c744eb1 [EXPLORER] Send WM_POPUPSYSTEMMENU asynchronously CORE-16353 (#4047)
to make the taskbar button turn into pressed state when the menu opens
This commit deals with the following tickets:
CORE-14994 "Spotify 1.0.29.92 has no icons on the buttons"
which regressed by 0.4.10-dev-466-g 35f62fc5ba
CORE-15166 "Media Player Classic HomeCinema MPC HC 1.7.13 'elapsed time' in status bar uses incredibly tiny font"
which regressed by 0.4.10-dev-466-g 35f62fc5ba
CORE-15303 "Firefox 28 does not longer show the checkmarks in the menubar"
which regressed by 0.4.10-dev-466-g 35f62fc5ba
CORE-16125 "Marlett, the 'top-right-closing-x' drawing regressed"
which regressed in master by 0.4.13-dev-110-g 19fc9415bf (in older releases by different commits)
---------------
The fix that manages to fix them all is a backport of the code-base that evolved up to 0.4.12-release-24-g4572c75. This was the source of porting.
*Before* the fix I found the following state for 0.4.11-release-39-gdeae917 by testing:
CORE-14994 ok
CORE-15166 ok
CORE-15303 ok
CORE-16125 affected
*Before* the fix I found the following state for 0.4.10-release-48-g72a4ecb by testing:
CORE-14994 affected
CORE-15166 affected
CORE-15303 affected
CORE-16125 ok
So the before-state was kind of exclusive-OR.
*After* the fix all of the mentioned tickets are fixed for releases/0.4.10 and releases/0.4.11
Ftr:
0.4.12-release-24-g4572c75 had all 4 tickets fixed, because it did receive the needed patches beforehand.
0.4.9-release-56-g9be0765 had all 4 tickets fixed, because it was never hit yet by the guilty commits in master branch. It does not heavily need those fixes yet therefore.
And last but not least, the backport to releases/0.4.10 will also inherently contain the fix for
CORE-15331 "[WIN32SS][FONT] Fix NEWTEXTMETRIC.ntmCellHeight value" (which was already fixed in initial releases/0.4.11)
In sum this backport contains the following commits (at least):
0.4.13-dev-483-g 641a0ea99b (#1670) CORE-14994 == 0.4.12-RC-38-g d5c85cd
0.4.13-dev-110-g 19fc9415bf (#1535) CORE-15303 == 0.4.12-RC-16-g e1ac9dc
0.4.12-RC-43-g 6f651b6ec5 CORE-16125 (please note, that this is an interim solution and NOT yet the final master-fix of 0.4.13-dev-699-g b8383b4 which would have fixed that as well)
0.4.11-dev-914-g 88ddb37acb (#1093) CORE-15166
0.4.11-dev-656-g 7f679a116b (#1010) CORE-15331
Porting any of those fixes on its own is dangerous, only their combination leads to tolerable results.
A squashed port of
0.4.13-dev-702-g 1e91a1690d
0.4.13-dev-809-g 65e4efd475
0.4.14-dev-186-g b3033b81a4
which in sum brings us to the nice state of 0.4.13-release-0-g4ee3778a4ed2d80c96e7a0f7a1e52e78d3a6568e
where I used the same trio successfully already.
CORE-13211 "RIGHT-expansion-arrow in submenus being drawn too large"
did affect many apps, e.g: UltraISO and Abiword
It was fixed by 0.4.13-dev-702-g 1e91a1690d (#1752)
but on its own it did introduce 2 regressions:
The first was that the
"upwards scrolling arrow in language selection submenu of UltraISO was now drawn much too large"
(discussion and pics also in CORE-13211)
That was fixed by the addendum 0.4.13-dev-809-g 65e4efd475 (#1753)
But we had still another regression from the first commit to solve:
CORE-16470 "The checkmarks in Abiwords menubar became black boxes"
That last regression was fixed by 0.4.14-dev-186-g b3033b81a4 (#2009)
CORE-16721 "Scrolling by pressing the scrollbar arrows does not draw the arrows as pressed"
Fixed by a patch of JIRA-user 'I_kill_bugs'. Thank you very much!
The regression was introduced by 0.4.11-dev-586-g 929a2c6
(which itself aimed to fix CORE-13986,
but not CORE-13918 as erroneously stated in commit comment)
cherry-picked from 0.4.14-dev-1167-g 8bb9434
and CORE-15557 "disabled scrollbars being clickable"
gets fixed as well due to squash of 0.4.14-dev-243-g 19fbdda
Fix CORE-5177 "TaskMgr and ProcExp positioning when maximized."
which regressed by SVN r47264 == git
0ac861114e
fix cherry picked from commit 0.4.12-dev-954-g
1f619b2781
-----------
but since above commit did introduce a new regression CORE-15893
symptom "Most of maximized windows have extra row of pixels around"
I had to port back those 2 follow-up-master-commits as well:
0.4.12-dev-956-g
197c4f875d3332dbc4656a6360e3c7e9def2ed5
and 0.4.13-dev-397-g
3cc50cedaf
Side-note: This results in the same final state as 0.4.12rls0, because
I ported 0.4.13-dev-397-g3cc50ce into 0.4.12-RC-29-gab7f77a earlier.
This fixes Civ II trying to create scrollbars based on 'native' scrollbars.
Fixes 2 application crashes CORE-15576 & CORE-13924
cherry picked from commit 0.4.12-dev-417-g
e022d8d7cb
- Implement user32!SwitchToThisWindow by NtUserCallTwoParam TWOPARAM_ROUTINE_SWITCHTOTHISWINDOW.
- Improve user32!CloseWindow with using SetActiveWindow and ShowWindow (synchronized).
CORE-15165
In Windows, user32!CloseWindow minimizes the window. It doesn't close the window actually. This is not a joke. This bad function naming is responsible to MS.
This fix is needed for the Delphi GUI programs.
- Don't enumerate the windows whose owner window exists.
- Use PostMessage SC_RESTORE instead of ShowWindowAsync SW_RESTORE.
CORE-15165
[LPK][USER32] Fix reading layout checks.
- Both functions don't care about DT_RTLREADING flag.
- Added a different method to check reading order.
[USER32] Remove gcc-build breaker semicolon.
This function is a condensed version of LpkPSMTextOut() (from lpk.dll),
but not bug-for-bug equivalent.
The main difference is that the underscore is drawn without taking into
account any kind of script processing.
Added callback function that redirects calls to GreExtTextOutW that didn't went through lpk BiDi processing, calls that are from the kernel.
Completely solves JIRA issue CORE-6910.
Observed via API monitoring.
This corresponds to the case where Win32k creates a window station whose name is based on the logon session identifier for the calling process.
Add also a note about the fact that we need to use a per-session-based WindowStation directory name, as done already in Win32k.
CORE-11933 and PR #621.
When we unselect text after the WM_CBLOSTTEXTFOCUS message,
make sure we also forget we have been focused at all;
otherwise the edit may become focused again, but with an
empty text selection.
CORE-10266
Additional changes:
- Adapt the message-box dialog procedure to correctly take into account
the presence or absence of a CANCEL button and take appropriate actions.
- Cache the strings retrieved by MB_GetString() (add FIXME notices too).
- Remove usage of ButtonLen and ibtn-pointer arrays, to make the code
more generic and less dependent on the (maximal) number of buttons
in the message boxes.
- Move around the code that computes the positions and sizes of the
window controls.
Also, improve NtUserSetWindowStationUser() capture order, make psid optional as it should (and avoid a user-mode triggered BSOD), and initialize luidUser only when everything succeeded.
The keyboard shortcuts Win+D and Win+M are also enabled.
- Implement IShellDispatch4::ToggleDesktop().
- Implement some commands in CTrayWindow.
- Add "sdk/include/reactos/traycmd.h" for tray commands.
- Fix task window switching.
- Improve the user32!SwitchToThisWindow() function and use it.
CORE-14318, CORE-13157
See also: CORE-14806 and CORE-8723
