diff options
| author | Kenneth Graunke <[email protected]> | 2014-01-05 22:57:01 -0800 |
|---|---|---|
| committer | Kenneth Graunke <[email protected]> | 2014-01-07 12:54:57 -0800 |
| commit | 847bc36a38d42967ad6bf0492fe90a4892d9d799 (patch) | |
| tree | da704ff271f8302bd2d25333d9b1118ffd8cf37b | |
| parent | 5e3fd6a9dbd62ca5cd9965282fac01a34d23733e (diff) | |
glsl: Optimize pow(2, x) --> exp2(x).
On Haswell, POW takes 24 cycles, while EXP2 only takes 14. Plus, using
POW requires putting 2.0 in a register, while EXP2 doesn't.
I believe that EXP2 will be faster than POW on basically all GPUs, so
it makes sense to optimize it.
Looking at the savage2 subset of shader-db:
total instructions in shared programs: 113225 -> 113179 (-0.04%)
instructions in affected programs: 2139 -> 2093 (-2.15%)
instances of 'math pow': 795 -> 749 (-6.14%)
instances of 'math exp': 389 -> 435 (11.8%)
Signed-off-by: Kenneth Graunke <[email protected]>
Reviewed-by: Matt Turner <[email protected]>
| -rw-r--r-- | src/glsl/opt_algebraic.cpp | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/src/glsl/opt_algebraic.cpp b/src/glsl/opt_algebraic.cpp index 5e885f78648..332f0b77b87 100644 --- a/src/glsl/opt_algebraic.cpp +++ b/src/glsl/opt_algebraic.cpp @@ -88,6 +88,12 @@ is_vec_one(ir_constant *ir) } static inline bool +is_vec_two(ir_constant *ir) +{ + return (ir == NULL) ? false : ir->is_value(2.0, 2); +} + +static inline bool is_vec_negative_one(ir_constant *ir) { return (ir == NULL) ? false : ir->is_negative_one(); @@ -420,6 +426,11 @@ ir_algebraic_visitor::handle_expression(ir_expression *ir) /* 1^x == 1 */ if (is_vec_one(op_const[0])) return op_const[0]; + + /* pow(2,x) == exp2(x) */ + if (is_vec_two(op_const[0])) + return expr(ir_unop_exp2, ir->operands[1]); + break; case ir_unop_rcp: |