cvs commit: src/sys/vfs/hammer hammer.h hammer_blockmap.c hammer_io.c hammer_ondisk.c hammer_vnops.c

[Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxxxxx>]

dillon      2008/02/10 10:58:23 PST

DragonFly src repository

  Modified files:
    sys/vfs/hammer       hammer.h hammer_blockmap.c hammer_io.c 
                         hammer_ondisk.c hammer_vnops.c 
  Log:
  HAMMER 28A/many: Translation and write performance optimizations
  
  * Cache a buffer's virtual (zoneX->zone2) translation to avoid unnecessary
    re-translations when a hot-cached buffer is available.
  
  * Use the sequential heuristic provided by the kernel to detect sequential
    writes and call bawrite() instead bdwrite() in such cases.  This
    prevents the buffer cache from hitting slow fallback cases and
    completely solves the write performance issue.
  
    Note that the undo fifo is not yet in place and performance will drop
    slightly once it is, but what you see now is basically what you're gonna
    get in the final.
  
  * Performance is now on par with UFS.
  
  Revision  Changes    Path
  1.37      +2 -1      src/sys/vfs/hammer/hammer.h
  1.2       +0 -8      src/sys/vfs/hammer/hammer_blockmap.c
  1.21      +1 -1      src/sys/vfs/hammer/hammer_io.c
  1.30      +18 -7     src/sys/vfs/hammer/hammer_ondisk.c
  1.31      +10 -0     src/sys/vfs/hammer/hammer_vnops.c


http://www.dragonflybsd.org/cvsweb/src/sys/vfs/hammer/hammer.h.diff?r1=1.36&r2=1.37&f=u
http://www.dragonflybsd.org/cvsweb/src/sys/vfs/hammer/hammer_blockmap.c.diff?r1=1.1&r2=1.2&f=u
http://www.dragonflybsd.org/cvsweb/src/sys/vfs/hammer/hammer_io.c.diff?r1=1.20&r2=1.21&f=u
http://www.dragonflybsd.org/cvsweb/src/sys/vfs/hammer/hammer_ondisk.c.diff?r1=1.29&r2=1.30&f=u
http://www.dragonflybsd.org/cvsweb/src/sys/vfs/hammer/hammer_vnops.c.diff?r1=1.30&r2=1.31&f=u