Revert r51017 that was reverting r51016.
In fact, it seems that cmlib assert error, whatever that eats randomly doesn't come from software issue inside ReactOS, but from a build issue.
Indeed, as you may see in the different tests I performed it shows that the same build produces the same failure, again and again. But, new build shows that there is no assert anylonger with that commit.
Don't ask me why or how it happens, I just don't get it.
But most of the time, it works, no one knows why ("as usual" (tm)). But it's cool.
Someone from BE team and/or buildslaves managers should take a serious look at this issue.
Just in case: if that commit re-hits cmlib assert, don't blame it immediatly, just commit something else, and see whether trunk is still broken.
But, as you can see in testman, we have now test results for r51016 ;-)
svn path=/trunk/; revision=51025
- Move the drag detection routine to kernel space, this will decrease the number of kernel calls that is noticeable with slower systems running an emulator.
svn path=/trunk/; revision=51021
- use well defined macro instead of weird pointer calculation
- reset thread specific data on a thread detach call
Fixes crash in heroes 3 and probably others
svn path=/trunk/; revision=51019
Revert r51016 that quite magically breaks Qemu test bot (Registry corruption?).
Everything was (and is still fine) on VirtualBox.
It will be put back online once the issue will have been fixed.
svn path=/trunk/; revision=51017
Implemented FstubTranslatorNull, FstubTranslateResource, FstubTranslateRequirement, xHalGetInterruptTranslator.
In other words: please welcome interrupt translator in ReactOS.
It's not used yet.
svn path=/trunk/; revision=51012
- EnableUserModePnpManager: StartServiceW can fail due to the service already running. Check last error for such a case. Should fix bug #5846.
svn path=/trunk/; revision=51006
- Fix calculation of dest rect in IntSetDIBits
- Rewrite NtGdiStretchDIBitsInternal. It now calls NtGdiSetDIBitsToDeviceInternal in case of SRCCOPY and no stretching and NtGdiCreateDIBitmapInternal + IntEngStretchBlt in the default case. This fixes RLE compressed dibs.
- The code isn't yet perfect, Windows behaviour is pretty complex, especially in regard to topdown/bottomup DIBs and the whole code needs to be rewritten. It will cause one more failed bitmap test, but another one that was formerly only passing with legacy 9x behaviour now behaves correctly.
See issue #5886 for more details.
svn path=/trunk/; revision=51005
Implemented FstubFixupEfiPartition():
Purpose is to let ReactOS handle realistic partitions length in case of an EFI/GPT protective MBR.
Indeed, when formatting a disk to GPT, you add a protective MBR that contains a fake partition entry, with type 0xEE having the following property: its size is set to maximum allocatable size in MBR.
In that case, if disk is small, you may go beyond disk size.
This is why a fix to partition length may be needed when re-reading partition entry.
This is the way ReactOS creates protective MBR.
You may find some more information here: http://msdn.microsoft.com/en-us/windows/hardware/gg463525
Fixed xHalIoReadPartitionTable():
- Allocate a buffer that has the proper size (instead of allocating a buffer which is too big). Especially since this leads to partial init only some lines after.
- Remove redundant & useless affectation.
- Check/fix 0xEE partition length in case of an EFI/GPT protective MBR (see upper point).
svn path=/trunk/; revision=51002
Fix IntCreateCompatibleBitmap, previously when the given DC was a memory DC with a dibsection selected, the function would only work if the dibsection was <= 8 bpp.
svn path=/trunk/; revision=50999
clip the destination rect against the bounds of the destination and source surface. Fixes a crash of gdi32_winetest bitmap that occurs as soon as IntCreateCompatibleBitmap is fixed. Other Eng* code needs to be checked as well.
svn path=/trunk/; revision=50998
- Print supported CPU features.
- Simplify detection of non working but claimed SYSENTER support for intel processors
- Workaround a virtualbox "feature" : claim that SYSENTER isn't supported despite the fact that it is.
Fix hang on registration of mshtml.dll in 2nd stage when using Virtualbox.
svn path=/trunk/; revision=50991
Use GreCreateBitmapEx instead of EngCreateBitmap in NtGdiSetDIBitsToDeviceInternal, so that the RLE hack works. Fix initialization of XLATEOBJ.
svn path=/trunk/; revision=50989
- Remove the old SetDIBBits, it severed us well.... Hold on to the win32k call.
- Tested, Area.exe, wine gdi32 bitmaps test, AbiWord 2.8.6, OOo 2.4.3, SM 2.0.11 and ReactOS applications.
- Aimp 2.61.583 (FULL, pinted okay), CoolPlayer 219, winamp 0.98d and winamp 2.95 (not FUll). The rest have drawing issue with DIB. See bug 5886.
svn path=/trunk/; revision=50987
- Rafal Harabien: Sync UniATA to 0.40a5. Changelog:
* Fixed bug with BSOD on newer SATA/AHCI Intel chips.
* Added support for different number of devices on different channls.
* Updated AHCI support code (not ready yet).
* All ReactOS specific changes have been left.
See issue #5976 for more details.
svn path=/trunk/; revision=50985
- Implement parameters checking in FsRtlIsNameInExpressionPrivate.
- Add two shortcuts for common wildcard invocations to make the function faster.
- Second (main part of the function) is still under review.
svn path=/trunk/; revision=50981
FlushFileBuffers:
- In spite of what MSDN pretends, it appears that FlushFileBuffers flushes console input buffer when it's called with a console handle. Then, do it here as well.
- Get rid of SetLastErrorByStatus here...
svn path=/trunk/; revision=50980
- RtlpInsertUnCommittedPages: Don't rely on linked list data in an UCR Descriptor after destroying it.
No functionality change because the linked list data was still "as expected", but not something we want to rely on.
- RtlpCreateUnCommittedRange: Fix a typo that caused the head of UCR Segment list of the Heap to be treated as an UCR Segment header.
Side effects of the typo: When this list wasn't empty, the (fake) UCR Segment it described was grown to contain more UCR descriptors.
Due to the data involved, this always happened to be the initial Heap Segment, which contains the Heap Header.
Writing the new UCR descriptors caused the Heap Header and trailing Heap Entries to be partially corrupted.
svn path=/trunk/; revision=50978
- Use one instead of two critical sections to synchronize service start and control actions.
- Fail, if an already running service is started again.
svn path=/trunk/; revision=50977
