just replacing our win32k.sys with the one from Win2003SP2 68000 does not load, asserts in MM Assertion 'MI_IS_ROS_PFN(Pfn1) == TRUE' failed at /srv/buildbot/Trunk_x86_GCCLin_Debug/build/ntoskrnl/mm/freelist.c line 473 67900 does not load, asserts in MM Assertion 'MI_IS_ROS_PFN(Pfn1) == TRUE' failed at /srv/buildbot/Trunk_x86_GCCLin_Debug/build/ntoskrnl/mm/freelist.c line 473 67820 does not load, asserts in MM Assertion 'MI_IS_ROS_PFN(Pfn1) == TRUE' failed at /srv/buildbot/Trunk_x86_GCCLin_Debug/build/ntoskrnl/mm/freelist.c line 473 67805 does not load, asserts in MM Assertion 'MI_IS_ROS_PFN(Pfn1) == TRUE' failed at /srv/buildbot/Trunk_x86_GCCLin_Debug/build/ntoskrnl/mm/freelist.c line 473 67798 does not load, asserts in MM Assertion 'MI_IS_ROS_PFN(Pfn1) == TRUE' failed at /srv/buildbot/Trunk_x86_GCCLin_Debug/build/ntoskrnl/mm/freelist.c line 473 67794 does not load, asserts in MM (even more assertions beforehand! but also the well-known one Assertion 'MI_IS_ROS_PFN(Pfn1) == TRUE' failed at /srv/buildbot/Trunk_x86_GCCLin_Debug/build/ntoskrnl/mm/freelist.c line 473 67793 does not load, asserts in MM (even more assertions beforehand! but also the well-known one Assertion 'MI_IS_ROS_PFN(Pfn1) == TRUE' failed at /srv/buildbot/Trunk_x86_GCCLin_Debug/build/ntoskrnl/mm/freelist.c line 473 67792 loads, LATER dies which is expected: dll/ntdll/ldr/ldrinit.c:829) LDR: DLL_PROCESS_ATTACH for dll "user32.dll" (InitRoutine: 77A4B419) failed -> *** Fatal System Error: 0xc0000142 (0x0031FED4,0x0031FECC,0x00000000,0x00000000) 67790 loads, LATER dies which is expected: dll/ntdll/ldr/ldrinit.c:829) LDR: DLL_PROCESS_ATTACH for dll "user32.dll" (InitRoutine: 77A4B419) failed -> *** Fatal System Error: 0xc0000142 (0x0031FED4,0x0031FECC,0x00000000,0x00000000) 67749 loads, LATER dies which is expected GUILTY REVISION IS 67793! r67793 | tkreuzer | 2015-05-17 02:34:45 +0200 (So, 17 Mai 2015) | 2 lines [NTOSKRNL] Use the VAD tree to search for MEMORY_AREAS, except when inserting an ARM3 based MA. In this case we still nee d to use the old tables, since the related VAD is already inserted into the VAD tree. HOWEVER, KEEP IN MIND THAT A LOT OF STUFF HAS BEEN TOUCHED IN NTOSKRNL BY THE SURROUNDING REVS BY TKREUZER r67784-r67799!