Optimization was disabled in this function to work around a bug in MSVC, use
a different solution that does not come with such a big performance regression.
When we switched to MSVC2019 and C++17 we seemingly
managed to trigger a code-gen bug with MSVC in the
AVX code-path.
This change works around the issue by (hopefully
temporary) disabling the optimizer for the fast_exp2f4
function, given it is only used in a single pass
of the denoiser and nowhere else, this is luckily
not as bad as it could have been.
Once the compiler is fixed or a different fix is
available we'll have to revisit this.
Details and link to the repro posted to MS is
available in T78047
BF-admins agree to remove header information that isn't useful,
to reduce noise.
- BEGIN/END license blocks
Developers should add non license comments as separate comment blocks.
No need for separator text.
- Contributors
This is often invalid, outdated or misleading
especially when splitting files.
It's more useful to git-blame to find out who has developed the code.
See P901 for script to perform these edits.
- In fresnel_dielectric, the differentials calculation sometimes divided by zero.
- When the normal map was (0.5, 0.5, 0.5), the code would try to normalize a zero vector. Now, it just uses the regular normal as a fallback.
- The approximate error function used in Beckmann sampling sometimes overflowed to inf while calculating r^16. The final value is 1 - 1/r^16, however,
so now it just returns 1 if the computation would overflow otherwise.
Our own implementation is in fact the same performance as in fast_math from
OpenShadingLanguage, but implementation from fast_math is using explicit madd
function, which increases chance of compiler deciding to use intrinsics.
