aco: always use rtne for fquantize2f16
authorGeorg Lehmann <dadschoorse@gmail.com>
Thu, 10 Aug 2023 18:55:04 +0000 (20:55 +0200)
committerMarge Bot <emma+marge@anholt.net>
Fri, 11 Aug 2023 12:37:23 +0000 (12:37 +0000)
commitc4f356faf490cb7d4ba63470e1240e92b8317742
treeb80fe7a15fc0b97af6426a1c803edcc17692d56a
parent144546f4345a4b822c531b867e1f3a1a1eb992e0
aco: always use rtne for fquantize2f16

The SPIR-V spec says:

If Value is positive with a magnitude too large to represent as a
16-bit floating-point value, the result is positive infinity.
If Value is negative with a magnitude too large to represent as a
16-bit floating-point value, the result is negative infinity.

This is only the case for rtne v_cvt_f16_f32

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24617>
src/amd/compiler/aco_instruction_selection.cpp