MIPS16e®

• A compressed instruction set that allows for reduced system costs through smaller code memory size requirements
• Applications compiled with MIPS16e are 30% smaller on average when compared to MIPS32®
• Compatibility with MIPS32 and MIPS64® instruction sets
• Ability to switch between MIPS16e mode and MIPS 32-bit mode

• MIPS16e instructions are 16-bits in size
• 8 of the 32 general purpose registers are accessible in MIPS16e mode; these include the registers commonly used for argument passing and return values
• 8-bit, 16-bit and 32-bit data-types are supported when used with MIPS32; when used with MIPS64, 64-bit data-types are added
• Mode switching between MIPS16e and MIPS32/64 allows for privileged architecture/floating-point/ASE support; software control of this mode switching is accomplished through a special jump instruction
• The EXTEND instruction allows the 16-bit instructions to efficiently create 16-bit immediate values
• SAVE/RESTORE instructions increase the code density for stack frame handling
• PC-relative addressing increases the code density for creating load/store addresses
• Some instructions use implied register specifiers to increase code density - the stack pointer or program counter for load/stores; a GPR designated for compare results for branches
• Support exists for application-specific macro instruction sequences

v2.60 (1356 KB)

v2.60 (1759 KB)