[PATCH 6/6][AArch64] Update prefetch tuning parameters for falkor and qdf24xx tunings.

[Maxim Kuvyrkov <maxim.kuvyrkov@linaro.org>]

[-- Attachment #1: Type: text/plain, Size: 183 bytes --]

This patch enables software prefetching at -O3 for Qualcomm's qdf24xx cores.

Bootstrapped and regtested on x86_64-linux-gnu and aarch64-linux-gnu.

--
Maxim Kuvyrkov
www.linaro.org


[-- Attachment #2: 0006-Update-prefetch-tuning-parameters-for-falkor-and-qdf.patch --]
[-- Type: application/octet-stream, Size: 1165 bytes --]

From f3dc0ae7203ec095e2f54894a0fbf13e807d3d46 Mon Sep 17 00:00:00 2001
From: Maxim Kuvyrkov <maxim.kuvyrkov@linaro.org>
Date: Thu, 15 Dec 2016 12:53:49 +0000
Subject: [PATCH 6/6] Update prefetch tuning parameters for falkor and qdf24xx
 tunings.

	* config/aarch64/aarch64.c (tune_params qdf24xx_tunings,)
	(tune_params falkor_tunings): Update prefetch settings, which, in turn,
	enable prefetching by default at -O3.

Change-Id: I2f0da54a8c262f6fbd0dcfde4584141bb09f1013
---
 gcc/config/aarch64/aarch64.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index e1b03ec..ccee947 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -774,7 +774,7 @@ static const struct tune_params qdf24xx_tunings =
   2,	/* min_div_recip_mul_sf.  */
   2,	/* min_div_recip_mul_df.  */
   0,	/* max_case_values.  */
-  AARCH64_PREFETCH_BENEFICIAL (0, -1, 64, -1),
+  AARCH64_PREFETCH_BENEFICIAL (4, 32, 64, 1024),
   tune_params::AUTOPREFETCHER_STRONG,	/* autoprefetcher_model.  */
   (AARCH64_EXTRA_TUNE_NONE)		/* tune_flags.  */
 };
-- 
2.7.4