BeeEngine/glslang-zig
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Out-of-range floats should overflow/underflow to infinity/0.0

Browse source 
glslang representing literal constants with double precision, so 1.0e40 and 1.0e-50 are normal values.

        Shader1：
        precision highp float;
        out vec4 my_FragColor;
        void main()
        {
        // Out-of-range floats should overflow to infinity
        // GLSL ES 3.00.6 section 4.1.4 Floats:
        // "If the value of the floating point number is too large (small) to be stored as a single precision value, it is converted to positive (negative) infinity"
        float correct = isinf(1.0e40) ? 1.0 : 0.0;
        my_FragColor = vec4(0.0, correct, 0.0, 1.0);
        }
        The expected ouput result of this test is vec4(0.0, 1.0, 0.0, 1.0),
        but it's vec4(0.0,0.0,0.0,1.0).Because the return value of isInf is
        false.

        precision highp float;
        out vec4 my_FragColor;
        void main()
        {
        // GLSL ES 3.00.6 section 4.1.4 Floats:
        // "A value with a magnitude too small to be represented as a mantissa and exponent is converted to zero."
        // 1.0e-50 is small enough that it can't even be stored as subnormal.
        float correct = (1.0e-50 == 0.0) ? 1.0 : 0.0;
        my_FragColor = vec4(0.0, correct, 0.0, 1.0);
        }
        The expected ouput result of this test is vec4(0.0, 1.0, 0.0, 1.0),
        but it's vec4(0.0,0.0,0.0,1.0).

        For f32 and f16 type, when the literal constant out of range of the f32
        and f16 number, the value should overflow or underflow to inf or zero.

            glcts test item
        KHR-GLES3.number_parsing.float_out_of_range_as_infinity
This commit is contained in:
“jimihe” 2023-11-17 17:02:10 +08:00 • committed by arcady-lunarg
parent 845e5d5a28
commit 854db99ff1
4 changed files with 54 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

28
Test/baseResults/overflow_underflow_toinf_0.out Normal file
View file

@ -0,0 +1,28 @@
overflow_underflow_toinf_0.frag
Shader version: 320
0:? Sequence
0:4 Function Definition: main( ( global void)
0:4 Function Parameters:
0:9 Sequence
0:9 Sequence
0:9 move second child to first child ( temp highp float)
0:9 'correct' ( temp highp float)
0:9 Constant:
0:9 1.000000
0:10 Sequence
0:10 move second child to first child ( temp highp float)
0:10 'correct1' ( temp highp float)
0:10 Constant:
0:10 1.000000
0:11 move second child to first child ( temp highp 4-component vector of float)
0:11 'my_FragColor' ( out highp 4-component vector of float)
0:11 Construct vec4 ( temp highp 4-component vector of float)
0:11 Constant:
0:11 0.000000
0:11 'correct' ( temp highp float)
0:11 'correct1' ( temp highp float)
0:11 Constant:
0:11 1.000000
0:? Linker Objects
0:? 'my_FragColor' ( out highp 4-component vector of float)

12
Test/overflow_underflow_toinf_0.frag Normal file
View file

@ -0,0 +1,12 @@
#version 320 es
precision highp float;
out vec4 my_FragColor;
void main()
{
// GLSL ES 3.00.6 section 4.1.4 Floats:
// "A value with a magnitude too small to be represented as a mantissa and exponent is converted to zero."
// 1.0e-50 is small enough that it can't even be stored as subnormal.
float correct = (1.0e-50 == 0.0) ? 1.0 : 0.0;
float correct1 = isinf(1.0e40) ? 1.0 : 0.0;
my_FragColor = vec4(0.0, correct, correct1, 1.0);
}

2
Test/runtests
View file

@ -343,6 +343,8 @@ run --enhanced-msgs -V --target-env vulkan1.2 --amb --aml enhanced.7.vert enhanc
diff -b $BASEDIR/enhanced.7.link.out $TARGETDIR/enhanced.7.link.out || HASERROR=1
run --enhanced-msgs -V --target-env vulkan1.2 --amb --aml spv.textureError.frag > $TARGETDIR/spv.textureError.frag.out
diff -b $BASEDIR/spv.textureError.frag.out $TARGETDIR/spv.textureError.frag.out || HASERROR=1
run -i overflow_underflow_toinf_0.frag > $TARGETDIR/overflow_underflow_toinf_0.out
diff -b $BASEDIR/overflow_underflow_toinf_0.out $TARGETDIR/overflow_underflow_toinf_0.out || HASERROR=1
#
# Final checking

12
glslang/MachineIndependent/Intermediate.cpp
View file

@ -2590,6 +2590,18 @@ TIntermConstantUnion* TIntermediate::addConstantUnion(double d, TBasicType baseT
{
assert(baseType == EbtFloat || baseType == EbtDouble || baseType == EbtFloat16);
if (isEsProfile() && (baseType == EbtFloat || baseType == EbtFloat16)) {
int exponent = 0;
frexp(d, &exponent);
int minExp = baseType == EbtFloat ? -126 : -14;
int maxExp = baseType == EbtFloat ? 127 : 15;
if (exponent > maxExp) { //overflow, d = inf
d = std::numeric_limits<double>::infinity();
} else if (exponent < minExp) { //underflow, d = 0.0;
d = 0.0;
}
}
TConstUnionArray unionArray(1);
unionArray[0].setDConst(d);