摘要:
A method, a translator, and a computer-readable medium for translating compiled programming code from a first code state to a second code state are disclosed. The programming code in the first state has a plurality of basic blocks, where each basic block has a set of instructions. At least one basic block ends in a dynamic branch, the dynamic branch being a transfer to one of a set of destinations based on a calculation of a destination address. The plurality of basic blocks in the first state of the programming code are identified, as are links between the identified basic blocks. A control flow graph (CFG) of the programming code is then constructed based on the identified basic blocks and identified links, where the CFG is in a preliminary form. At least one basic block ending in a dynamic branch is identified, and all identified basic blocks that lead to the dynamic branch are explored, based on the CFG, as far back as is necessary to fully determine a set of destination addresses for the dynamic branch. The set of destination addresses defines the set of destinations from the dynamic branch. Such set of destinations is examined to identify a branch table, and the CFG is updated to reflect the set of destinations and the identified branch table. The programming code is then translated from the first state to the second state based at least in part on the updated CFG.
摘要:
A method, a translator, and a computer-readable medium for translating compiled programming code from a first code state to a second code state are disclosed. The programming code in the first state has a plurality of basic blocks, where each basic block has a set of instructions. At least one basic block ends in a dynamic branch, the dynamic branch being a transfer to one of a set of destinations based on a calculation of a destination address. The plurality of basic blocks in the first state of the programming code are identified, as are links between the identified basic blocks. A control flow graph (CFG) of the programming code is then constructed based on the identified basic blocks and identified links, where the CFG is in a preliminary form. At least one basic block ending in a dynamic branch is identified, and all identified basic blocks that lead to the dynamic branch are explored, based on the CFG, as far back as is necessary to fully determine a set of destination addresses for the dynamic branch. The set of destination addresses defines the set of destinations from the dynamic branch. Such set of destinations is examined to identify a branch table, and the CFG is updated to reflect the set of destinations and the identified branch table. The programming code is then translated from the first state to the second state based at least in part on the updated CFG.
摘要:
A method for communicating active messages among nodes of a parallel processing computer system is disclosed. The active messages are defined by .mu.threads, and the method comprises the steps of: (a) generating a .mu.thread comprising an instruction pointer, frame pointer, and Local Parameters pointer from a first node to a second node; and (b) performing a procedure on a data structure in accordance with the .mu.thread. The instruction pointer points to an application specific procedure in system memory, and the frame pointer points to an application specific data structure in system memory. The Local Parameters pointer points to one or more words of additional data or parameters stored in memory mapped device registers or system memory.