As an additional feature on top of the "allow non-existing functions to be exported" "feature", because rbuild generates and links STDCALL function names without the proper decoration (vs. enforcing decoration at linking, but only removing it at export-time), this allows the definition (as an export) of a STDCALL function that is completely different from the actual function itself.
For example, the 5-parameter Foo function is normally Foo@20, while the 3-parameter Foo function woudl be Foo@12. Linking one against the other would fail (say, 2 parameters were added to Foo in a newer version). However, under RBUILD, both of these would appear as "Foo", and the linker/compiler would happilly connect the caller of Foo@3 (that has pushed 3 parameters) to the receiving side of Foo@5 (that is about to pop 5 parameters).
Even -if- decorations WERE to be applied, Foo@12 would STILL succeed, because of the first feature, which would enable the export of Foo@12 even though no such function exist.
In a further, bizare, twist of fate, the behavior of this RBUILD "feature", when the target function is not found, is to link the exported DLL TO ITSELF.
Therefore, one can see how, previously to this patch, kernel32.dll would import a dozen functions from itself (all the non-existing functions).
To really seal the deal, the behavior of exported functions used by kernel32, but that are actually forwarded to another DLL deserves a special mention.
GetLastError, for example, merely forwards to RtlGetLastWin32Error, so it is normal behavior to use a #define in the C code so that all internal calls to the function are routed correctly.
This did not happen, so instead, kernel32 tried importing/linking/exporting GetLastError, but this symbol is not found in the binary, because it is only a forwarder.
This caused kernel32 to import from itself (the behavior when an exported symbol is not found). When importing from itself, the loader would now find the _forwarded_ for GetLastError, and correctly link with ntdll.
What should be a one-liner of assembly (inline TEB access) thus became a triple-call indirection (GetLastError@0->StubLastError@0->__impGetLastError@0->__impRtlGetLastWin32Error->RtlGetLastWin32Error.
While analyzing these issues, we also realized a strange macro SetLastErrorByStatus that manually tried to perform what there already exists a function for: RtlSetLastNtStatusFromWin32Error.
And, in an exciting coda, we also realized that our Server 2003 Kernel32 exports more than a dozen Windows 95 APIs, through an auto-stub generation mechanism within winebuild, that gets linked as an object behind the scenes.
[KERNEL32]: Removed all Win95 exports, cleaned up exports.
[KERNEL32]: Fixed up set/get error macros by making them inline and/or calling the correct ntdll function.
[KERNEL32]: Removed bizare calls to Wine-internal/specific APIs from our core Win32 DLL.
[KERNEL32]: Wrote stubs for all functions which should be exported, and set the correct number of parameters for them.
[KERNEL32]: Kernel32 is smaller, loads faster, does not export Windows 95 functions, does not export non-existing functions, and does not import from itself anymore.
Note: This is one of the many failings of RBUILD the CMAKE system has helped us discover. I believe these issues are serious enough to warrant an immediate sync with trunk, but rest assured, there are many more completely broken, infinitely-regressing things that we discovered while switching to CMAKE.
svn path=/trunk/; revision=48475
- Remove "implementations" of __lc_collate_cp, __lc_handle and __lc_codepage. These are not functions, but varuables, implemented as MSVCRT_*
- add ___lc_codepage_func and ___lc_handle_func taken from wine
- Fix msvcrt spec file
- Fixes compilation with gcc 4.5.1
svn path=/trunk/; revision=48473
- MAXIMUM_VOLUME_LABEL_LENGTH is in bytes, not in characters. Fix struct definition and access beyond the buffer. Spotted by Carlo Bramini.
- Trim trailing spaces from the volume label name, instead of stopping at the first encountered space. Fix by Carlo Bramini.
See issue #5505 for more details.
svn path=/trunk/; revision=48468
When creating a shortcut icon and we have 32 bpp, use GdiAlphaBlend to create the final bitmap. This is neccessary, since we want to support alpha channels and those are destroyed when using SRCPAINT or any similar ROPs on the alpha bitmap (not a bug, Windows works like that, too). We could use MaskBlt, if it would work correctly, but on reactos it destroys the alpha channel as well (bug!), it's also most likely slower then the current solution.
Fixes broken overlay icons.
See issue #5455 for more details.
svn path=/trunk/; revision=48463
- Fixed FsRtlIsNameInExpression to make it properly handle * in expressions
- Fix formatting
- Patch by Pierre Schweitzer
- Fixes everything
See issue #5541 for more details.
svn path=/trunk/; revision=48461
- Move generated precompiled headers to binary folders.
- Enable auto image base for ntdll.
- Update the location of generated PCH files for rtl, ntdll and crt.
svn path=/branches/cmake-bringup/; revision=48459
[CMAKE]: Set global linker flags we need for all libraries.
[CMAKE]: Set specific NTDLL linker flag.
Thanks to Amine for some of his help.
svn path=/branches/cmake-bringup/; revision=48457
- Sync lcformat.c to Wine-1.3. This fixes (a number of potential) problems introduced by new nls data files being synced recently and led to crashes on locales having different months genitives. Original issue confirmed by Sergey Gusev <evilslon@mail.ru> and Igor Paliychuk.
svn path=/trunk/; revision=48433
Implement a proper version of _assert. It nows shows a message box with the details and let's the user choose to either abort, debug or ignore the assertion.
See bug 5486, bug 5507, bug 5511
svn path=/trunk/; revision=48432
