(svn r26211) -Add: specialised non-animated SS2 blitter (MJP)

With 32bpp base set about 30% faster than 32bpp-optimized, or about 10% for 8bpp base sets in the Draw function. Respectively about 5 and 1% of total run time

1 files changed, 1 insertions, 1 deletions

diff --git a/src/video/win32_v.cpp b/src/video/win32_v.cpp
index 186ac1025..02cae68e2 100644
--- a/src/video/win32_v.cpp
+++ b/src/video/win32_v.cpp
@@ -1077,7 +1077,7 @@ static bool AllocateDibSection(int w, int h, bool force)
 	bi->bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
 
 	bi->bmiHeader.biWidth = _wnd.width = w;
-	bi->bmiHeader.biHeight = -(_wnd.height = h);
+	bi->bmiHeader.biHeight = -(_wnd.height = h+1); // Allocate extra room to prevent out-of-bounds when SSE reads a 16B block at the end of the buffer.
 
 	bi->bmiHeader.biPlanes = 1;
 	bi->bmiHeader.biBitCount = BlitterFactory::GetCurrentBlitter()->GetScreenDepth();