Mercurial > hg > truffle
diff src/share/vm/opto/matcher.hpp @ 4047:d8cb48376797
7097546: Optimize use of CMOVE instructions
Summary: Avoid CMove in a loop if possible. May generate CMove if it could be moved outside a loop.
Reviewed-by: never
author | kvn |
---|---|
date | Wed, 26 Oct 2011 06:08:56 -0700 |
parents | 95134e034042 |
children | 6729bbc1fcd6 |
line wrap: on
line diff
--- a/src/share/vm/opto/matcher.hpp Tue Oct 25 12:51:13 2011 -0700 +++ b/src/share/vm/opto/matcher.hpp Wed Oct 26 06:08:56 2011 -0700 @@ -360,6 +360,12 @@ // Anything this size or smaller may get converted to discrete scalar stores. static const int init_array_short_size; + // Some hardware needs 2 CMOV's for longs. + static const int long_cmove_cost(); + + // Some hardware have expensive CMOV for float and double. + static const int float_cmove_cost(); + // Should the Matcher clone shifts on addressing modes, expecting them to // be subsumed into complex addressing expressions or compute them into // registers? True for Intel but false for most RISCs