[tip:locking/core] locking/qspinlock/x86: Increase _Q_PENDING_LOOPS upper bound
From: tip-bot for Will Deacon
Date: Fri Apr 27 2018 - 05:39:06 EST
Commit-ID: b247be3fe89b6aba928bf80f4453d1c4ba8d2063
Gitweb: https://git.kernel.org/tip/b247be3fe89b6aba928bf80f4453d1c4ba8d2063
Author: Will Deacon <will.deacon@xxxxxxx>
AuthorDate: Thu, 26 Apr 2018 11:34:18 +0100
Committer: Ingo Molnar <mingo@xxxxxxxxxx>
CommitDate: Fri, 27 Apr 2018 09:48:47 +0200
locking/qspinlock/x86: Increase _Q_PENDING_LOOPS upper bound
On x86, atomic_cond_read_relaxed will busy-wait with a cpu_relax() loop,
so it is desirable to increase the number of times we spin on the qspinlock
lockword when it is found to be transitioning from pending to locked.
According to Waiman Long:
| Ideally, the spinning times should be at least a few times the typical
| cacheline load time from memory which I think can be down to 100ns or
| so for each cacheline load with the newest systems or up to several
| hundreds ns for older systems.
which in his benchmarking corresponded to 512 iterations.
Suggested-by: Waiman Long <longman@xxxxxxxxxx>
Signed-off-by: Will Deacon <will.deacon@xxxxxxx>
Acked-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Acked-by: Waiman Long <longman@xxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: boqun.feng@xxxxxxxxx
Cc: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
Cc: paulmck@xxxxxxxxxxxxxxxxxx
Link: http://lkml.kernel.org/r/1524738868-31318-5-git-send-email-will.deacon@xxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
---
arch/x86/include/asm/qspinlock.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/x86/include/asm/qspinlock.h b/arch/x86/include/asm/qspinlock.h
index 90b0b0ed8161..da1370ad206d 100644
--- a/arch/x86/include/asm/qspinlock.h
+++ b/arch/x86/include/asm/qspinlock.h
@@ -7,6 +7,8 @@
#include <asm-generic/qspinlock_types.h>
#include <asm/paravirt.h>
+#define _Q_PENDING_LOOPS (1 << 9)
+
#define queued_spin_unlock queued_spin_unlock
/**
* queued_spin_unlock - release a queued spinlock