1. merge si_set_ls_return_value_for_tcs into si_llvm_ls_build_end because they
do the same job to return value
2. stop return value when mono mode with different thread count, in which case
TCS use LS input as its input instead of LS output
3. use si_insert_input_ret_float
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24204>
Qiang Yu
401a40a5f4