From 2713f6bb90765de81954275a30c62432d30e1d68 Mon Sep 17 00:00:00 2001
From: Wilco Dijkstra <wilco.dijkstra@arm.com>
Date: Thu, 14 Nov 2024 14:34:17 +0000
Subject: [PATCH] AArch64: Add FULLY_PIPELINED_FMA to tune baseline

Add FULLY_PIPELINED_FMA to tune baseline - this is a generic feature that is
already enabled for some cores, but benchmarking it shows it is faster on all
modern cores (SPECFP improves ~0.17% on Neoverse V1 and 0.04% on Neoverse N1).

gcc:
	* config/aarch64/aarch64-tuning-flags.def (AARCH64_EXTRA_TUNE_BASE):
	Add AARCH64_EXTRA_TUNE_FULLY_PIPELINED_FMA.
	* config/aarch64/tuning_models/ampere1b.h: Remove redundant
	AARCH64_EXTRA_TUNE_FULLY_PIPELINED_FMA.
	* config/aarch64/tuning_models/neoversev2.h: Likewise.
---
 gcc/config/aarch64/aarch64-tuning-flags.def   | 3 ++-
 gcc/config/aarch64/tuning_models/ampere1b.h   | 3 +--
 gcc/config/aarch64/tuning_models/neoversev2.h | 3 +--
 3 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/gcc/config/aarch64/aarch64-tuning-flags.def b/gcc/config/aarch64/aarch64-tuning-flags.def
index 7ebf390ef818..60967aac9037 100644
--- a/gcc/config/aarch64/aarch64-tuning-flags.def
+++ b/gcc/config/aarch64/aarch64-tuning-flags.def
@@ -49,6 +49,7 @@ AARCH64_EXTRA_TUNING_OPTION ("fully_pipelined_fma", FULLY_PIPELINED_FMA)
 AARCH64_EXTRA_TUNING_OPTION ("avoid_pred_rmw", AVOID_PRED_RMW)
 
 /* Baseline tuning settings suitable for all modern cores.  */
-#define AARCH64_EXTRA_TUNE_BASE (AARCH64_EXTRA_TUNE_CHEAP_SHIFT_EXTEND)
+#define AARCH64_EXTRA_TUNE_BASE (AARCH64_EXTRA_TUNE_CHEAP_SHIFT_EXTEND	\
+				 | AARCH64_EXTRA_TUNE_FULLY_PIPELINED_FMA)
 
 #undef AARCH64_EXTRA_TUNING_OPTION
diff --git a/gcc/config/aarch64/tuning_models/ampere1b.h b/gcc/config/aarch64/tuning_models/ampere1b.h
index c541623993d2..2ad6003d65c1 100644
--- a/gcc/config/aarch64/tuning_models/ampere1b.h
+++ b/gcc/config/aarch64/tuning_models/ampere1b.h
@@ -103,8 +103,7 @@ static const struct tune_params ampere1b_tunings =
   0,	/* max_case_values.  */
   tune_params::AUTOPREFETCHER_STRONG,	/* autoprefetcher_model.  */
   (AARCH64_EXTRA_TUNE_BASE
-   | AARCH64_EXTRA_TUNE_AVOID_CROSS_LOOP_FMA
-   | AARCH64_EXTRA_TUNE_FULLY_PIPELINED_FMA), /* tune_flags.  */
+   | AARCH64_EXTRA_TUNE_AVOID_CROSS_LOOP_FMA), /* tune_flags.  */
   &ampere1b_prefetch_tune,
   AARCH64_LDP_STP_POLICY_ALIGNED,   /* ldp_policy_model.  */
   AARCH64_LDP_STP_POLICY_ALIGNED    /* stp_policy_model.  */
diff --git a/gcc/config/aarch64/tuning_models/neoversev2.h b/gcc/config/aarch64/tuning_models/neoversev2.h
index 4fabe4df2de7..b000fb465709 100644
--- a/gcc/config/aarch64/tuning_models/neoversev2.h
+++ b/gcc/config/aarch64/tuning_models/neoversev2.h
@@ -220,8 +220,7 @@ static const struct tune_params neoversev2_tunings =
   (AARCH64_EXTRA_TUNE_BASE
    | AARCH64_EXTRA_TUNE_CSE_SVE_VL_CONSTANTS
    | AARCH64_EXTRA_TUNE_MATCHED_VECTOR_THROUGHPUT
-   | AARCH64_EXTRA_TUNE_AVOID_PRED_RMW
-   | AARCH64_EXTRA_TUNE_FULLY_PIPELINED_FMA),	/* tune_flags.  */
+   | AARCH64_EXTRA_TUNE_AVOID_PRED_RMW),	/* tune_flags.  */
   &generic_armv9a_prefetch_tune,
   AARCH64_LDP_STP_POLICY_ALWAYS,   /* ldp_policy_model.  */
   AARCH64_LDP_STP_POLICY_ALWAYS	   /* stp_policy_model.  */
-- 
GitLab