[llvm] [X86] Remove SlowDivide tuning from GRTTuning (PR #84676)

[Phoebe Wang via llvm-commits]

https://github.com/phoebewang created https://github.com/llvm/llvm-project/pull/84676

The DIV32/64 throughput was improved since Goldmont in the Atom architecture. The Alder Lake-E shows similar number too. So we shouldn't add such tunings to Gracemont and later products.

Checked from Agner Fog's table and uops.info.

>From 083ea9d3760dd3c67f2d70c0e521736ad2b4bf54 Mon Sep 17 00:00:00 2001
From: Phoebe Wang <phoebe.wang at intel.com>
Date: Sun, 10 Mar 2024 21:22:29 +0800
Subject: [PATCH] [X86] Remove SlowDivide tuning from GRTTuning

The DIV32/64 throughput was improved since Goldmont in the Atom
architecture. The Alder Lake-E shows similar number too. So we shouldn't
add such tunings to Gracemont and later architectures.
---
 llvm/lib/Target/X86/X86.td | 2 --
 1 file changed, 2 deletions(-)

diff --git a/llvm/lib/Target/X86/X86.td b/llvm/lib/Target/X86/X86.td
index a2a65ce75d6b9a..8367f938c0ddfa 100644
--- a/llvm/lib/Target/X86/X86.td
+++ b/llvm/lib/Target/X86/X86.td
@@ -1237,8 +1237,6 @@ def ProcessorFeatures {
   // Gracemont
   list<SubtargetFeature> GRTTuning = [TuningMacroFusion,
                                       TuningSlow3OpsLEA,
-                                      TuningSlowDivide32,
-                                      TuningSlowDivide64,
                                       TuningFastScalarFSQRT,
                                       TuningFastVectorFSQRT,
                                       TuningFast15ByteNOP,