Marvellworld Trade Ltd.

Marvell International Limited

Liping Gao

Haitao Huang

Xinyu Qi

Ningsheng Jian

Pengfei Li

Lei Jin

Cheng-Hsueh Andrew Hsieh

XingXing Pan

Yuheng Zhang

Aspects of the disclosure provide a method for linking input files during a linking process. The method includes receiving an input section that is to be mapped to a memory segment by a linker circuit, determining whether an out-of-memory (OOM) event occurs when an available memory space of the memory segment is unable to accommodate the input section, estimating a memory expansion size that would be required for the memory segment to be able to accommodate the input section when an OOM event occurs, and creating by the linker circuit a map file that includes the estimated memory expansion size of the memory segment.

Memory expansion in a linking process

Aspects of the disclosure provide a method for identifying an induction variable in a loop during a compiling process. The method includes searching for a phi-function that includes a first operand and a second operand and defines a candidate basic induction variable (BIV), searching for an add/sub instruction that has a first register and a second register wherein the first register is the second operand of the phi-function, or the value in the first register is subsequently stored to the second operand of the phi-function through one or more move instructions, and determining the candidate BIV is a BIV when the second register of the add/sub instruction is the candidate BIV or stores a value that is passed from the candidate BIV through one or more move instructions.

Induction variable identification

System and methods are provided for register allocation. An original code block and a target code block associated with a branch of an execution loop are determined. An original allocation of a plurality of physical registers to one or more original variables associated with the original code block is detected. A target allocation of the plurality of physical registers to one or more target variables associated with the target code block is determined. One or more temporary registers are selected from the plurality of physical registers based at least in part on the original allocation and the target allocation. The original allocation is changed to the target allocation using the selected temporary registers. Specifically, one or more instructions are generated to change the original allocation to the target allocation using the selected temporary registers. The instructions are executed using one or more processors.

Systems and methods for register allocation

Aspects of the disclosure provide a method to support performance analysis. The method includes compiling bytecodes to generate native codes corresponding to the bytecodes in an electronic device, generating a file to include the bytecodes and the corresponding native codes in the file, collecting symbol information to map symbols in the bytecodes with offsets of corresponding native codes, and including the symbol information in the file to enable profiling.

Method and apparatus for supporting performance analysis

Aspects of the disclosure provide a method for code compilation. The method includes receiving instructions of a loop code for compiling, allocating one or more registers to variables before compiling the instructions into a loop body for the loop code, and compiling the instructions into the loop body based on the allocated registers.

Method and system for compiler optimization

Aspects of the disclosure provide a method for instruction scheduling. The method includes receiving a sequence of instructions, identifying redundant flag-register based dependency of the instructions, and re-ordering the instructions without being restricted by the redundant flag-register based dependency.

Method and system for instruction scheduling

System and methods are provided for loop process suspension. One or more loop instructions associated with a loop process are loaded in a code cache. One or more branch instructions associated with a branch of the loop process in the code cache are determined. A suspension event is detected. The branch instructions are replaced with one or more jump instructions in the code cache upon the detection of the suspension event. If the jump instructions are executed in the code cache, the branch instructions in the code cache are restored, and the loop process is suspended. One or more suspension instructions associated with the suspension event are executed in an interpreter.

Systems and methods for loop suspension in java programming

In one embodiment, the present invention includes a method for emitting a live range statement into a program for a memory variable to be cached during run-time that has at least one simulation state variable if the memory variable is dynamically mapped, and defining the simulation state variable at a first execution path of the program. In such manner, the program may be optimized using the live range statement and the simulation state variable. Also, a debugger may use the simulation state variables in obtaining and displaying the memory variable from a cache.

Growing community of inventors

Shanghai, China

Liping Gao