summaryrefslogtreecommitdiff
path: root/arch/arm/mach-imx/system.c
diff options
context:
space:
mode:
authorAndrey Smirnov <andrew.smirnov@gmail.com>2016-06-18 18:09:30 -0700
committerShawn Guo <shawnguo@kernel.org>2016-06-21 20:40:38 +0800
commit1d9e94779956b765d316924cc39af2165ad631fb (patch)
tree494f9ff55568aec4eb5260139a6ef23bf1570d03 /arch/arm/mach-imx/system.c
parentb829037136766abf34456e2406bab67d689b3538 (diff)
ARM: i.MX: system.c: Tweak prefetch settings for performance
Update Prefetch Control Register settings to match that of Freescale's Linux tree. As the commit e3addf1b773964eac7f797e8538c69481be4279c states (author Nitin Garg): "... set Prefetch offset to 15, since it improves memcpy performance by 35%. Don't enable Incr double Linefill enable since it adversely affects memcpy performance by about 32MB/s and reads by 90MB/s. Tested with 4K to 16MB sized src and dst aligned buffer..." Those results are also corroborated by our own testing. Acked-by: Arnd Bergmann <arnd@arndb.de> Tested-by: Chris Healy <cphealy@gmail.com> Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Diffstat (limited to 'arch/arm/mach-imx/system.c')
-rw-r--r--arch/arm/mach-imx/system.c8
1 files changed, 6 insertions, 2 deletions
diff --git a/arch/arm/mach-imx/system.c b/arch/arm/mach-imx/system.c
index 76d7ebe883d7..bf7ab779d5bf 100644
--- a/arch/arm/mach-imx/system.c
+++ b/arch/arm/mach-imx/system.c
@@ -111,8 +111,12 @@ void __init imx_init_l2cache(void)
val = readl_relaxed(l2x0_base + L310_PREFETCH_CTRL);
val |= L310_PREFETCH_CTRL_DBL_LINEFILL |
L310_PREFETCH_CTRL_INSTR_PREFETCH |
- L310_PREFETCH_CTRL_DATA_PREFETCH |
- L310_PREFETCH_CTRL_DBL_LINEFILL_INCR;
+ L310_PREFETCH_CTRL_DATA_PREFETCH;
+
+ /* Set perfetch offset to improve performance */
+ val &= ~L310_PREFETCH_CTRL_OFFSET_MASK;
+ val |= 15;
+
writel_relaxed(val, l2x0_base + L310_PREFETCH_CTRL);
}