BBBsmoke / angleproject

Automatically exported from code.google.com/p/angleproject
Other
0 stars 0 forks source link

Fix GetShaderPrecisionFormat on DirectX 11 backend #919

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago
What steps will reproduce the problem?
1. Run dEQP tests dEQP-GLES3.functional.shaders.operator.unary_operator.minus.*

What is the expected output? What do you see instead?
Expected: All pass
Actual: lowp and mediump unsigned integer tests fail.

The reason for the failure is that the tests obtain the expected integer 
wraparound point from GetShaderPrecisionFormat, which returns wrong values. 
DirectX 9 didn't yet mandate native 32-bit integers, so only a limited integer 
range was available (usually simulated by floats). ANGLE assumes this kind of 
limited integer range. Instead, range[0] = 31 and range[1] = 30 should be 
returned on the DirectX 11 backend, which is guaranteed to have native 32-bit 
integers.

Original issue reported on code.google.com by oetu...@nvidia.com on 16 Feb 2015 at 2:07

GoogleCodeExporter commented 9 years ago
This was fixed a while ago.

Original comment by oetu...@nvidia.com on 27 Feb 2015 at 2:11