[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <tip-d8bced79af1db6734f66b42064cc773cada2ce99@git.kernel.org>
Date: Tue, 9 Feb 2016 08:08:12 -0800
From: tip-bot for Andy Lutomirski <tipbot@...or.com>
To: linux-tip-commits@...r.kernel.org
Cc: dvlasenk@...hat.com, mcgrof@...e.com, oleg@...hat.com,
aryabinin@...tuozzo.com, luto@...capital.net,
linux-kernel@...r.kernel.org, hpa@...or.com, tglx@...utronix.de,
dave.hansen@...ux.intel.com, mingo@...nel.org, bp@...en8.de,
peterz@...radead.org, toshi.kani@...com, akpm@...ux-foundation.org,
bp@...e.de, torvalds@...ux-foundation.org, luto@...nel.org,
brgerst@...il.com
Subject: [tip:x86/mm] x86/mm: If INVPCID is available,
use it to flush global mappings
Commit-ID: d8bced79af1db6734f66b42064cc773cada2ce99
Gitweb: http://git.kernel.org/tip/d8bced79af1db6734f66b42064cc773cada2ce99
Author: Andy Lutomirski <luto@...nel.org>
AuthorDate: Fri, 29 Jan 2016 11:42:59 -0800
Committer: Ingo Molnar <mingo@...nel.org>
CommitDate: Tue, 9 Feb 2016 13:36:11 +0100
x86/mm: If INVPCID is available, use it to flush global mappings
On my Skylake laptop, INVPCID function 2 (flush absolutely
everything) takes about 376ns, whereas saving flags, twiddling
CR4.PGE to flush global mappings, and restoring flags takes about
539ns.
Signed-off-by: Andy Lutomirski <luto@...nel.org>
Reviewed-by: Borislav Petkov <bp@...e.de>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Andrey Ryabinin <aryabinin@...tuozzo.com>
Cc: Andy Lutomirski <luto@...capital.net>
Cc: Borislav Petkov <bp@...en8.de>
Cc: Brian Gerst <brgerst@...il.com>
Cc: Dave Hansen <dave.hansen@...ux.intel.com>
Cc: Denys Vlasenko <dvlasenk@...hat.com>
Cc: H. Peter Anvin <hpa@...or.com>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Luis R. Rodriguez <mcgrof@...e.com>
Cc: Oleg Nesterov <oleg@...hat.com>
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Toshi Kani <toshi.kani@...com>
Cc: linux-mm@...ck.org
Link: http://lkml.kernel.org/r/ed0ef62581c0ea9c99b9bf6df726015e96d44743.1454096309.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@...nel.org>
---
arch/x86/include/asm/tlbflush.h | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 8b57683..fc9a2fd 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -152,6 +152,15 @@ static inline void __native_flush_tlb_global(void)
{
unsigned long flags;
+ if (static_cpu_has(X86_FEATURE_INVPCID)) {
+ /*
+ * Using INVPCID is considerably faster than a pair of writes
+ * to CR4 sandwiched inside an IRQ flag save/restore.
+ */
+ invpcid_flush_all();
+ return;
+ }
+
/*
* Read-modify-write to CR4 - protect it from preemption and
* from interrupts. (Use the raw variant because this code can
Powered by blists - more mailing lists