- IntCreateWindow: Fill out MaximumLength field of the window name. Fixes potential buffer overflow in at least NtUserDefSetText.
svn path=/trunk/; revision=50135
- remove useless memory reallocation : the buffer size must be dword aligned, we have no way to guarantee the buffer location will be.
- pass BITMAPINFO pointer size to NtGdiGetDIBitsInternal
[WIN32K]
- Improve "not enough memory" check when creating a bitmap
- use correct function to set the last error.
svn path=/trunk/; revision=50133
- NtGdiGetDIBitsInternal: Copy the requested amount of scanlines into the return buffer, not the whole bitmap. Should fix bug #5766.
svn path=/trunk/; revision=50113
Fix rounding error when calculating display size
Loosely based on patch by neoman.
Fixes gdi32_winetest:mapping
See issue #5790 for more details.
svn path=/trunk/; revision=50112
The installer now completes given enough ram in NEWCC=1, and writes an
understandable FS. There's at least one more problem preventing the
resulting installation from booting however.
Fix a bug in rmap where we were checking the next (unfortunately named
current) rmap for a segment membership.
Fix a mistake in cache trim.
#ifdef detection of DirectMapping in section in NEWCC mode in page out.
svn path=/trunk/; revision=50110
- Fix a typo in region size calculation of special memory areas, which unbreaks OllyDbg & co. Thanks Michael Martin for the hint.
See issue #5692 for more details.
svn path=/trunk/; revision=50109
- Fix incompatibility with newer KiEnterTrap implementation. Should fix spontaneous issues I possibly introduced with my previous commit, and finally fixes that VirtualBox issue.
svn path=/trunk/; revision=50107
- Always initialize TrapFrame segment selectors in KiEnterTrap, because further C code relies on at least TrapFrame->SegFs being correct. Running Arwinss on VirtualBox exposed this bug.
I wouldn't find solution for this bug without Timo's great help!
svn path=/trunk/; revision=50095
Fix for a buffer overflow and then a buffer overrun (if ever it fixes something)
The way filenames are handled for FAT entries should be REALLY simplified. This would prevent such errors.
Thus, there are more magic values in fastfat driver than everywhere else in ReactOS which makes proper fixing hard (impossible?).
Finally, the code for that fix is crappy, but I don't care, it fits the rest of the fastfat driver code.
*pissed off*
Fixes CID #2502
svn path=/trunk/; revision=50090
ScsiPortConvertPhysicalAddressToUlong takes an ULONG_PTR as input, fix function accordingly and do the "calculation" directly instead of calling RtlConvertUlongToLargeInteger
Spotted by Samuel.
svn path=/trunk/; revision=50089
Actually, code was correct, but ugly (who said "normal, that's a hack).
So, when that code is called from SetupLDR, KeyHandle is required to be to 1. This what code does. Thing we do in the if condition when it appears we are called from SetupLDR.
To avoid any further question, or warning, adding more parenthesis to show that we know what we do.
So, to sum up, there's no comparaison for KeyHandle
svn path=/trunk/; revision=50083
- DisplayDevicePropertyText: Use full DWORD value, not just one byte of it (CID 1803).
[WINMM]
- MCI_Close: Use UINT instead of UINT16 for the device id (CID 1804).
svn path=/trunk/; revision=50080
- ConvertStringSidToSidW: Stop writing 1 subauthority too much. Fixes a DWORD sized buffer overflow. Should fix bug #5764.
- ConvertStringSidToSidW: Don't leak an allocated SID in case of failure.
[SETUPAPI]
- SetupDiClassNameFromGuidExW: Rewrite to prevent a buffer overflow and pass additional winetests. Should fix bug #5474.
- SetupDiClassNameFromGuidExA: Return the required buffer size in failure cases.
svn path=/trunk/; revision=50066
