Intel 815 Manuel D’Utilisation
Intel® 815 Chipset: Graphics Controller PRM, Rev 1.0
R
142
10.3.
GC Register Programming
All of the GC registers (except for the PCI Configuration registers) are memory mapped. The base
address of this 512 KB memory block is programmed in the MMADR PCI Configuration register. Note
that 2D control registers (VGA and Extended VGA registers) are also located at their standard I/O
locations. For more information on the registers, refer to the System Address Map chapter.
address of this 512 KB memory block is programmed in the MMADR PCI Configuration register. Note
that 2D control registers (VGA and Extended VGA registers) are also located at their standard I/O
locations. For more information on the registers, refer to the System Address Map chapter.
10.4.
GC Instruction Streams
This section describes how instruction streams can be used to control GC operation and perform GC
operations.
operations.
10.4.1. Instruction
Use
GC instructions are used to control drawing engines and various GC functional units:
• 3D Instructions. 3D instructions are used to program the 3D pipeline state and perform 3D
rendering operations, including "StretchBlt" operations.
• Motion Compensation. A special "GFXBLOCK" instruction is used to perform Motion
Compensation operations via the Mapping Engine.
• 2D Instructions (BLT). These instructions are used to perform BLT operations.
• Instruction Parser (IP) Instructions. The IP instructions can be used to control and synchronize
• Instruction Parser (IP) Instructions. The IP instructions can be used to control and synchronize
the instruction stream as well as perform various GC auxiliary functions (e.g., define graphics buffer
attributes, perform display/overlay flips, etc.)
attributes, perform display/overlay flips, etc.)
10.4.2. Instruction Transport Overview
Instructions are not written directly to the GC; instead, they are placed in memory by software and later
read (via DMA) by the GC's Instruction Parser unit. The primary mechanism used to transport
instructions is through the use of two ring buffers (RB): Low Priority RB and Interrupt RB. A secondary
mechanism for instruction transport is through the use of batch buffers. The IP uses a set of rules to
determine the order in which instructions are executed. Following sections in this chapter provide
descriptions of the ring buffers, batch buffers, and IP rules.
read (via DMA) by the GC's Instruction Parser unit. The primary mechanism used to transport
instructions is through the use of two ring buffers (RB): Low Priority RB and Interrupt RB. A secondary
mechanism for instruction transport is through the use of batch buffers. The IP uses a set of rules to
determine the order in which instructions are executed. Following sections in this chapter provide
descriptions of the ring buffers, batch buffers, and IP rules.