From 1252d63cc2ecc398dd2d36b4a77a1af212cc6cb2 Mon Sep 17 00:00:00 2001
From: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Date: Tue, 18 Oct 2022 23:56:31 +0200
Subject: [PATCH] aco: Pre-split result of bvh64_intersect_ray_amd.

Avoids  later moves with extractions from the vector.

Reduces VALU operation in the raytrace loop by ~6%, increasing
the RT performance in Q2RTX on a 6800 XT by about ~1.3%.

Suggested by Georg.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19148>
---
 src/amd/compiler/aco_instruction_selection.cpp | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/amd/compiler/aco_instruction_selection.cpp b/src/amd/compiler/aco_instruction_selection.cpp
index e48f128..29fe8f3 100644
--- a/src/amd/compiler/aco_instruction_selection.cpp
+++ b/src/amd/compiler/aco_instruction_selection.cpp
@@ -6156,6 +6156,8 @@ visit_bvh64_intersect_ray_amd(isel_context* ctx, nir_intrinsic_instr* instr)
    mimg->dmask = 0xf;
    mimg->unrm = true;
    mimg->r128 = true;
+
+   emit_split_vector(ctx, dst, instr->dest.ssa.num_components);
 }
 
 static std::vector<Temp>
-- 
2.7.4