Accuracy of some operations was recently improved in the R600 backend, at the cost of slower code. This is required for compute shaders, but not for graphics shaders. Add unsafe-fp-math hint to make LLVM generate faster but possibly less accurate code.
Piglit didn't indicate any regressions.
0e9cded radeon/llvm: enable unsafe math for graphics shaders
src/gallium/drivers/radeon/radeon_llvm_emit.c | 5 +++++
1 file changed, 5 insertions(+)
Upstream: cgit.freedesktop.org