KonstantinSeurer
/
mesa
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ac/llvm: implement 16-bit and 64-bit fpow correctly 

LLVM converts to 32 bits and back for llvm.pow, so we can't use it.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9395>
This commit is contained in:
Marek Olšák 2021-02-12 01:05:19 -05:00 committed by Marge Bot
parent 3475c79328
commit 46ce67a331
1 changed files with 9 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/amd/llvm/ac_nir_to_llvm.c
View File

@ -843,6 +843,15 @@ static void visit_alu(struct ac_nir_context *ctx, const nir_alu_instr *instr)
result = ac_build_frexp_mant(&ctx->ac, src[0], instr->dest.dest.ssa.bit_size);
break;
case nir_op_fpow:
if (instr->dest.dest.ssa.bit_size != 32) {
/* 16 and 64 bits */
result = emit_intrin_1f_param(&ctx->ac, "llvm.log2",
ac_to_float_type(&ctx->ac, def_type), src[0]);
result = LLVMBuildFMul(ctx->ac.builder, result, ac_to_float(&ctx->ac, src[1]), "");
result = emit_intrin_1f_param(&ctx->ac, "llvm.exp2",
ac_to_float_type(&ctx->ac, def_type), result);
break;
}
result = emit_intrin_2f_param(&ctx->ac, "llvm.pow", ac_to_float_type(&ctx->ac, def_type),
src[0], src[1]);
break;