On 09/27/2011 03:08 PM, Eric Anholt wrote:
> They're both implemented the same in GLSL IR (since round() has
> undefined behavior for N.5).
>
> Fixes glsl-1.30/compiler/built-in-functions/round*
> ---
> src/glsl/ir_constant_expression.cpp | 10 ++++++++++
> 1 files changed, 10 insertions(+), 0 deletions(-)
>
> diff --git a/src/glsl/ir_constant_expression.cpp
> b/src/glsl/ir_constant_expression.cpp
> index b3fe6cf..59a7e67 100644
> --- a/src/glsl/ir_constant_expression.cpp
> +++ b/src/glsl/ir_constant_expression.cpp
> @@ -196,6 +196,13 @@ ir_expression::constant_expression_value()
> }
> break;
>
> + case ir_unop_round_even:
> + assert(op[0]->type->base_type == GLSL_TYPE_FLOAT);
> + for (unsigned c = 0; c < op[0]->type->components(); c++) {
> + data.f[c] = rint(op[0]->value.f[c]);
It's not obvious to me that rint() rounds correctly.
>From the GLSL 1.30 spec:
"Returns a value equal to the nearest integer to x. A fractional part
of 0.5 will round toward the nearest even integer. (Both 3.5 and 4.5 for
x will return 4.0.)"
rint() does nearest integer, but probably not the mandated behavior for
0.5. At least, not unless we configure the rounding mode. Which I
don't think we have...
> + }
> + break;
> +
> case ir_unop_ceil:
> assert(op[0]->type->base_type == GLSL_TYPE_FLOAT);
> for (unsigned c = 0; c < op[0]->type->components(); c++) {
> @@ -1324,6 +1331,9 @@ ir_call::constant_expression_value()
> * op[1]->value.f[c];
> }
> }
> + } else if (strcmp(callee, "round") == 0 ||
> + strcmp(callee, "roundEven") == 0) {
> + expr = new(mem_ctx) ir_expression(ir_unop_round_even, op[0]);
> } else if (strcmp(callee, "sign") == 0) {
> expr = new(mem_ctx) ir_expression(ir_unop_sign, type, op[0], NULL);
> } else if (strcmp(callee, "sin") == 0) {
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
