초록 ▼

A scheduling algorithm is provided for selecting the placement of instructions with internal slack into a schedule of instructions within a loop. The algorithm achieves this by pinning nodes with internal slack to corresponding nodes on the critical path of the code that have similar properties in

A scheduling algorithm is provided for selecting the placement of instructions with internal slack into a schedule of instructions within a loop. The algorithm achieves this by pinning nodes with internal slack to corresponding nodes on the critical path of the code that have similar properties in terms of the data dependency graph, such as earliest time and latest time. The effect is that nodes with internal slack are more often optimally placed in the schedule, reducing the need for rotating registers or register copy instructions. The benefit of the present invention can primarily be seen when performing instruction scheduling or software pipelining on loop code, but can also apply to other forms of instruction scheduling when greater control of placement of nodes with internal slack is desired.

대표청구항 ▼

What is claimed is: 1. A method in a data processing system for scheduling a set of instructions, the method comprising: building a data dependency graph for the set of instructions, wherein a subset of the instructions form a critical path in the data dependency graph; identifying an internal slac

What is claimed is: 1. A method in a data processing system for scheduling a set of instructions, the method comprising: building a data dependency graph for the set of instructions, wherein a subset of the instructions form a critical path in the data dependency graph; identifying an internal slack node in the data dependency graph, wherein the internal slack node has internal slack; pinning the internal slack node to a critical path node in the critical path, wherein the critical path node has at least one property that is similar to the internal slack node, wherein the at least one property includes an average of an earliest time and a latest time; and scheduling the set of instructions such that the internal slack node is scheduled close to the critical path node. 2. The method of claim 1, wherein the set of instructions is a loop. 3. The method of claim 2, wherein scheduling the set of instructions includes performing software pipelining on the loop. 4. The method of claim 1, further comprising: analyzing the data dependency graph to determine properties of nodes in the graph. 5. The method of claim 4, wherein the properties include at least one of height, depth, earliest time, latest time, and slack. 6. The method of claim 1, wherein identifying an internal slack node includes: marking nodes that have slack whose slack is external to the critical path; and identifying at least one unmarked node with slack.