[SYCL][CUDA][libclc] Add bf16 builtins and optimize half builtins for fma, fmin, fmax and fmax#5724

Merged

bader merged 21 commits into

t4c1:libclc_bf16_tmp

Mar 14, 2022

Commits on Feb 18, 2022

Commits on Feb 21, 2022

Commits on Feb 24, 2022

Commits on Mar 2, 2022

[NVPTX] Expose float tys min, max, abs, neg as builtins

jchlanda
authored and
t4c1
committed
[NVPTX] Add more FMA intriniscs/builtins

jchlanda
authored and
t4c1
committed
[NVPTX] Add ex2.approx.f16/f16x2 support

npmiller
authored and
t4c1
committed
bugfixes
t4c1
committed

Commits on Mar 3, 2022