摘要:
提出了一種基于LGDFG(粗顆粒度數(shù)據(jù)流圖)的系統(tǒng)級存儲優(yōu)化方法,該方法為被優(yōu)化的程序建立LGDFG模型,把每個(gè)程序段作為該模型的一個(gè)節(jié)點(diǎn),分析前后節(jié)點(diǎn)之間的數(shù)據(jù)產(chǎn)生消耗關(guān)系,從而得到不同數(shù)據(jù)塊對應(yīng)的生存周期,對于生存周期相互間不重疊的數(shù)據(jù)段可以實(shí)現(xiàn)存儲共享.以音頻AC3解碼程序?yàn)槔?用LGDFG模型分析了程序中不同數(shù)據(jù)塊之間的產(chǎn)生消耗關(guān)系,從而實(shí)現(xiàn)不同數(shù)據(jù)塊間的存儲共享.試驗(yàn)結(jié)果表明使用該方法AC3解碼程序數(shù)據(jù)存儲容量減少了25.8%,有效地減少了數(shù)據(jù)存儲容量.%This paper proposes a system-level memory optimization approach,which uses LGDFG (Large Grain Data Flow Graph) model to share memory space of application programs.LGDFG model comprises nodes and edges,which denote sub-functions and producing-consuming relationship of variables respeetively.After checking the LGDFG,the live time of each global variable is de-termined.Variables which have non-overlapping live time can share the same memory space in different periods.Finally,AC3 de-coding program is employed as an example to describe the approach in detail.Implementation results show that the memory space has decreased 25.8% for AC3 decoding program.This approach has reduced the data memory size of applieation programs significantly.