Chinaunix首页 | 论坛 | 博客
  • 博客访问: 3132352
  • 博文数量: 117
  • 博客积分: 10003
  • 博客等级: 上将
  • 技术积分: 5405
  • 用 户 组: 普通用户
  • 注册时间: 2007-01-23 09:34
文章分类

全部博文(117)

文章存档

2011年(1)

2010年(10)

2009年(69)

2008年(37)

分类: LINUX

2008-03-13 09:18:07

  M_MXFAST is the maximum request size used for "fastbins", special bins
  that hold returned chunks without consolidating their spaces. This
  enables future requests for chunks of the same size to be handled
  very quickly, but can increase fragmentation, and thus increase the
  overall memory footprint of a program.
  This malloc manages fastbins very conservatively yet still
  efficiently, so fragmentation is rarely a problem for values less
  than or equal to the default.  The maximum supported value of MXFAST
  is 80. You wouldn't want it any higher than this anyway.  Fastbins
  are designed especially for use with many small structs, objects or
  strings -- the default handles structs/objects/arrays with sizes up
  to 8 4byte fields, or small strings representing words, tokens,
  etc. Using fastbins for larger objects normally worsens
  fragmentation without improving speed.
  M_MXFAST is set in REQUEST size units. It is internally used in
  chunksize units, which adds padding and alignment.  You can reduce
  M_MXFAST to 0 to disable all use of fastbins.  This causes the malloc
  algorithm to be a closer approximation of fifo-best-fit in all cases,
  not just for larger requests, but will generally cause it to be
  slower.
阅读(3297) | 评论(1) | 转发(0) |
0

上一篇:Linux的帧缓冲设备

下一篇:O2的代价

给主人留下些什么吧!~~

chinaunix网友2008-06-05 18:23:06

明天给我讲讲啊