Parallelizing compile method, parallelizing compiler, parallelizing compile apparatus, and onboard apparatus

A parallelizing compiler causing a computer to perform:

• dividing a sequential program for an embedded system for an onboard apparatus into a plurality of macro tasks according to a predetermined method, the sequential program being executed by a single processor system;
• specifying, based on a control dependency between the macro tasks,
• (i) a starting end task of the macro tasks as a starting point, wherein the starting end task has a processing branched into different ones of the macro tasks, and
• (ii) a termination end task of the macro tasks, wherein the termination end task is either one of the macro tasks commonly executed among a plurality of series of processings, which are sequentially executed from the starting end task;
• fusing
• (i) the starting end task specified by the specifying,
• (ii) the termination end task executed from the starting end task in the processings, and
• (iii) a group of the macro tasks executed after execution of the starting end task and before execution of the termination end task into a new macro task;
• extracting a group of new macro tasks, which are parallelly executable by a plurality of processor units included in a multiprocessor system mounted to the onboard apparatus, from the new macro tasks fused in the fusing, based on a data dependency;
• performing a static scheduling assigning the new macro tasks to the processor units, so that the group of new macro tasks is parallelly executable by the processor units;
• and
• generating a parallelizing program executed by the multiprocessor system, based on a result of the static scheduling.

A parallelizing compiler apparatus comprising:

• a division device to divide a sequential program for an embedded system for an onboard apparatus into a plurality of macro tasks according to a predetermined method, the sequential program being executed by a single processor system;
• a specifying device to specify, based on a control dependency between the macro tasks,
• (i) a starting end task of the macro tasks as a starting point, wherein the starting end task has a processing branched into different ones of the macro tasks, and
• (ii) a termination end task of the macro tasks, wherein the termination end task is either one of the macro tasks commonly executed among a plurality of series of processings, which are sequentially executed from the starting end task;
• a fusion device to fuse
• (i) the starting end task specified by the specifying,
• (ii) the termination end task executed from the starting end task in the processings, and
• (iii) a group of the macro tasks executed after execution of the starting end task and before execution of the termination end task into a new macro task;
• an extraction device to extract a group of new macro tasks, which are parallelly executable by a plurality of processor units included in a multiprocessor system mounted to the onboard apparatus, from the new macro tasks fused in the fusing, based on a data dependency;
• a scheduling device to perform a static scheduling assigning the new macro tasks to the processor units, so that the group of new macro tasks is parallelly executable by the processor units;
• and
• a generation device to generate a parallelizing program executed by the multiprocessor system, based on a result of the static scheduling.

An onboard apparatus comprising a multiprocessor system operated by a parallelizing program generated by a parallelizing compile method,

• wherein the parallelizing compile method includes:
• dividing a sequential program for an embedded system for the onboard apparatus into a plurality of macro tasks according to a predetermined method, the sequential program being executed by a single processor system;
• specifying, based on a control dependency between the macro tasks,
• (i) a starting end task of the macro tasks as a starting point, wherein the starting end task has a processing branched into different ones of the macro tasks, and
• (ii) a termination end task of the macro tasks, wherein the termination end task is either one of the macro tasks commonly executed among a plurality of series of processings, which are sequentially executed from the starting end task;
• fusing
• (i) the starting end task specified by the specifying,
• (ii) the termination end task executed from the starting end task in the processings, and
• (iii) a group of the macro tasks executed after execution of the starting end task and before execution of the termination end task into a new macro task;
• extracting a group of new macro tasks, which are parallelly executable by a plurality of processor units included in the multiprocessor system mounted to the onboard apparatus, from the new macro tasks fused in the fusing, based on a data dependency;
• performing a static scheduling assigning the new macro tasks to the processor units, so that the group of new macro tasks is parallelly executable by the processor units;
• and
• generating the parallelizing program executed by the multiprocessor system, based on a result of the static scheduling.