cvs commit: src/lib/msun/src s_exp2.c s_exp2f.c src/lib/msun/ld128 s_exp2l.c src/lib/msun/ld80 s_exp2l.c

[Bruce Evans]

bde         2008-02-13 10:44:45 UTC

  FreeBSD src repository

  Modified files:
    lib/msun/src         s_exp2f.c s_exp2.c 
    lib/msun/ld128       s_exp2l.c 
    lib/msun/ld80        s_exp2l.c 
  Log:
  Fix exp2*(x) on signaling NaNs by returning x+x as usual.
  
  This has the side effect of confusing gcc-4.2.1's optimizer into more
  often doing the right thing.  When it does the wrong thing here, it
  seems to be mainly making too many copies of x with dependency chains.
  This effect is tiny on amd64, but in some cases on i386 it is enormous.
  E.g., on i386 (A64) with -O1, the current version of exp2() should
  take about 50 cycles, but took 83 cycles before this change and 66
  cycles after this change.  exp2f() with -O1 only speeded up from 51
  to 47 cycles.  (exp2f() should take about 40 cycles, on an Athlon in
  either i386 or amd64 mode, and now takes 42 on amd64).  exp2l() with
  -O1 slowed down from 155 cycles to 123 for some args; this is unimportant
  since the i386 exp2l() is a fake; the wrong thing for it seems to
  involve branch misprediction.
  
  Revision  Changes    Path
  1.3       +1 -1      src/lib/msun/ld128/s_exp2l.c
  1.3       +1 -1      src/lib/msun/ld80/s_exp2l.c
  1.6       +1 -1      src/lib/msun/src/s_exp2.c
  1.8       +1 -1      src/lib/msun/src/s_exp2f.c