ST10R272LT1 datasheet(16/77 Pages) STMICROELECTRONICS

16 / 77 page

16/77

ST10R272L - MULTIPLY-ACCUMULATE UNIT (MAC)

5.2

MAC Operation

Instruction pipelining

All MAC instructions use the 4-stage pipeline. During each stage the following tasks are

performed:

•

FETCH: All new instructions are double-word instructions.

•

DECODE: If required, operand addresses are calculated and the resulting operands are

fetched. IDX and GPR pointers are post-modified if necessary.

•

EXECUTE: Performs the MAC operation. At the end of the cycle, the Accumulator and the

MAC condition flags are updated if required. Modified GPR pointers are written-back

during this stage, if required.

•

WRITEBACK: Operand write-back in the case of parallel data move.

Note

At least one instruction which does not use the MAC must be inserted between two

instructions that read from a MAC register. This is because the Accumulator and the

status of the MAC are modified during the Execute stage. The CoSTORE instruction

has been added to allow access to the MAC registers immediately after a MAC

operation.

Address generation

MAC instructions can use some standard ST10 addressing modes such as GPR direct or

#data4 for immediate shift value.

New addressing modes have been added to supply the MAC with two new operands per

instruction cycle. These allow indirect addressing with address pointer post-modification.

Double indirect addressing requires two pointers. Any GPR can be used for one pointer, the

other pointer is provided by one of two specific SFRs IDX0 and IDX1. Two pairs of offset

registers QR0/QR1 and QX0/QX1 are associated with each pointer (GPR or IDXi). The GPR

pointer allows access to the entire memory space, but IDXi are limited to the internal Dual-

Port RAM, except for the CoMOV instruction.

ST10R272LT1 bảng dữ liệu(PDF) 16 Page - STMicroelectronics