Jetpack/kernel/kernel-4.9/rt-patches/0043-x86-mm-cpa-avoid-wbinv...

45 lines
1.5 KiB
Diff

From cbc11b257df45f51ca07a46a436d19f6575506b7 Mon Sep 17 00:00:00 2001
From: John Ogness <john.ogness@linutronix.de>
Date: Mon, 30 Jan 2017 09:41:21 +0100
Subject: [PATCH 043/352] x86/mm/cpa: avoid wbinvd() for PREEMPT
Although wbinvd() is faster than flushing many individual pages, it
blocks the memory bus for "long" periods of time (>100us), thus
directly causing unusually large latencies on all CPUs, regardless
of any CPU isolation features that may be active.
For 1024 pages, flushing those pages individually can take up to
2200us, but the task remains fully preemptible during that time.
Cc: stable-rt@vger.kernel.org
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
arch/x86/mm/pageattr.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 1271bc9..83d9fde 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -215,7 +215,15 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
int in_flags, struct page **pages)
{
unsigned int i, level;
+#ifdef CONFIG_PREEMPT
+ /*
+ * Avoid wbinvd() because it causes latencies on all CPUs,
+ * regardless of any CPU isolation that may be in effect.
+ */
+ unsigned long do_wbinvd = 0;
+#else
unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
+#endif
BUG_ON(irqs_disabled());
--
2.7.4