Store sine & cosine tables as doubles instead of floats

Since they're used as doubles in the calculations, might as well store them
that way instead of converting when read from the table at runtime.

Costs about 4k more storage in the binary for the doubled tables.

Also gets rid of 900 clang warnings, of the form:

Clock.c:1708:11: warning: implicit conversion loses floating-point precision:
      'double' to 'float' [-Wconversion]
0.000000, 0.001745, 0.003490, 0.005235, 0.006981, 0.008726, 0.010471, 0.012217,

(one for each table entry)

Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>

0 files changed, 0 insertions, 0 deletions