Description

The GR32_LowLevel unit contains performance-critical low-level primitives used extensively across Graphics32. It provides optimized memory fills and copies, fast division, fast arithmetic shifts, range clamping and wrapping functions, and accelerated fast-rounding routines (utilizing CPU instruction set extensions such as SSE2 and SSE4.1 where available).

Key capabilities provided by this unit include:

• Optimized Memory Operations: 32-bit and 16-bit array filling and copying routines (FillLongword, FillWord, MoveLongword, MoveWord).
• Stack Memory Allocation: Stack-based dynamic memory allocation primitives (StackAlloc, StackFree).
• Fast Value Manipulation: Swapping operations (Swap, Swap32, Swap16, Swap64), conditional swapping (TestSwap), and range clipping check (TestClip).
• Boundary Clamping, Wrapping & Reflection: Flexible coordinate mapping functions including Clamp, Constrain, SwapConstrain, Wrap, WrapMem, WrapPow2, Mirror, Reflect, and ReflectPow2.
• Optimal Wrap Function Resolvers: GetOptimalWrap, GetOptimalReflect, GetWrapProc, and GetWrapProcEx for dynamically selecting optimized wrapping procedure delegates.
• Fast Integer Division: Accelerated division by 255 and 127 (Div255, FastDiv255, Div255Round, Div127).
• Fast Rounding Functions: High-performance replacement for RTL rounding and truncation routines (FastFloor, FastCeil, FastTrunc, FastRound).
• Arithmetic Right Shifts: Sign-preserving bitwise right shifts (SAR_3 through SAR_16).
• Color Processing: Color format conversion between VCL/WinAPI TColor and 32-bit TColor32 via ColorSwap.

---