d2d5858be1
scalar values (that ispc used to smear across the array/struct
elements). Now, initializers in variable declarations must be
{ }-delimited lists, with one element per struct member or array
element, respectively.
There were a few problems with the previous implementation of the
functionality to initialize from scalars. First, the expression
would be evaluated once per value initialized, so if it had side-effects,
the wrong thing would happen. Next, for large multidimensional arrays,
the generated code would be a long series of move instructions, rather
than loops (and this in turn made LLVM take a long time.)
While both of these problems are fixable, it's a non-trivial
amount of re-plumbing for a questionable feature anyway.
Fixes issue #50.
19 lines
498 B
Plaintext
19 lines
498 B
Plaintext
|
|
export uniform int width() { return programCount; }
|
|
|
|
|
|
|
|
struct Foo { float f; };
|
|
|
|
export void f_fu(uniform float RET[], uniform float aFOO[], uniform float b) {
|
|
float a = aFOO[programIndex];
|
|
uniform Foo foo[17];
|
|
for (uniform int i = 0; i < 17; ++i)
|
|
foo[i].f = a;
|
|
++foo[a].f;
|
|
uniform int i[16] = { 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16 };
|
|
RET[programIndex] = foo[i[programIndex]].f;
|
|
}
|
|
|
|
export void result(uniform float RET[]) { RET[programIndex] = 2+programIndex; }
|