qrintfã®æœ€é©åŒ– - 開発日記
H2Oã®å†…部ã§ã¯ï¼Œæ•°å€¤ï¼Œæ–‡å—列ã®ãƒ•ã‚©ãƒ¼ãƒžãƒƒã‚¿ã¨ã—ã¦sprintfãŒç”¨ã„られã¦ã„ã¾ã™ï¼Ž 我々ã¯ï¼Œæœ€è¿‘ã¾ã§sprintf専用ソースコード変æ›å™¨, qrintfã®ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã‚’è¡Œã£ã¦ã„ã¾ã—ãŸï¼Ž 本稿ã§ã¯qrintfã®æ¦‚è¦ã¨ä»Šå›žé©å¿œã—ãŸå·¥å¤«ï¼Œãã—ã¦ä»Šå¾Œã®èª²é¡Œã«ã¤ã„ã¦ã¾ã¨ã‚ã¦ãŠãã¾ã™ï¼Ž 1. qrintfã¨ã¯ï¼Ÿ qrintfã¨ã¯Cコンパイラã®ï¼‘ã¤ã§ã‚ã‚‹gccã®ãƒ—リプãƒã‚»ãƒƒã‚µã®ãƒ©ãƒƒãƒ‘ーã§ã‚り，snprintfを高速化ã™ã‚‹ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰å¤‰æ›å™¨ã§ã™ï¼Ž 本稿執ç†æ™‚点ã§ã®qrintfã®å¤‰æ›å¯¾è±¡ã¯sprintfã¨snprintfã§ã™ï¼Žä»¥ä¸‹æœ¬æ–‡ä¸ã§ã¯ç‰¹ã«ã“ã¨ã‚ã‚Šã®ãªã„é™ã‚Šsprintfã¨è¨˜è¿°ã—ãŸå ´åˆsprintf, snprintfã®ä¸¡æ–¹ã‚’指ã—ã¾ã™ï¼Ž qrintfã¯ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã«å‡ºç¾ã™ã‚‹sprintfã§åˆ©ç”¨ã™ã‚‹ãƒ•ã‚©ãƒ¼ãƒžãƒƒãƒˆæ–‡å—列を解æžã—，型ã”ã¨ã«ç”¨æ„ã•ã‚ŒãŸé–¢æ•°å‘¼å‡ºã«ã‚³ãƒ¼ãƒ‰ã‚’書ãæ›ãˆsprintfã®é«˜é€ŸåŒ–ã‚’è¡Œã„ã¾ã™ï¼Ž
Cã®æ¼”ç®—åã¨ãƒ“ット演算関連ã®æœ€é©åŒ– - mirichiã®æ—¥è¨˜
mrubyã®æœ€æ–°ç‰ˆã«å®Ÿè¡Œé€Ÿåº¦æ”¹å–„ã®ä¿®æ£ãŒå…¥ã£ãŸã€‚arena_aiã‚’å¿…è¦ãªã¨ã“ã‚ã®ã¿ã«å…¥ã‚Œã‚‹ã€while最é©åŒ–ã€ç„¡é§„ãªOP_LOADNILã®å‰Šé™¤ã€ã 。 ã“ã‚Œã«ã‚ˆã‚Štimesã®å‘½ä»¤åˆ—ã¯ã“ã®ã‚ˆã†ã«ãªã£ãŸã€‚ irep 150 nregs=6 nlocals=4 pools=0 syms=3 000 OP_ENTER 0:0:0:0:0:0:1 001 OP_LOADI R2 0 002 OP_JMP 012 003 OP_MOVE R4 R1 004 OP_MOVE R5 R2 005 OP_LOADNIL R6 006 OP_SEND R4 :call 1 007 OP_MOVE R4 R2 008 OP_LOADI R5 1 009 OP_LOADNIL R6 010 OP_ADD R4 :+ 1 011 OP_MOVE R2 R4 012 OP_MOVE R4 R2 013 OP_LOADS
LinkTimeOptimization - GCC Wiki
Status The branch has been merged to trunk and is now closed. Final merge announcement: http://gcc.gnu.org/ml/gcc/2009-10/msg00060.html Background Link Time Optimization (LTO) gives GCC the capability of dumping its internal representation (GIMPLE) to disk, so that all the different compilation units that make up a single executable can be optimized as a single module. This expands the scope of in
1
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
処ç†ã‚’実行ä¸ã§ã™
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
Improving LLVM-Generated Code Size for X86 Processors
{{#tags}}- {{label}}
{{/tags}}