linux-kernel - [PATCH] [GIT PULL] tracing: optimize global_trace

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Tue, 15 Sep 2009 12:32:34 -0400
From:	Steven Rostedt <rostedt@...dmis.org>
To:	LKML <linux-kernel@...r.kernel.org>
Cc:	Ingo Molnar <mingo@...e.hu>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Frederic Weisbecker <fweisbec@...il.com>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: [PATCH] [GIT PULL] tracing: optimize global_trace_clock cachelines


Ingo,

Please pull the latest tip/tracing/core3 tree, which can be found at:

  git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace.git
tip/tracing/core3


Steven Rostedt (1):
      tracing: optimize global_trace_clock cachelines

----
 kernel/trace/trace_clock.c |   24 ++++++++++++++----------
 1 files changed, 14 insertions(+), 10 deletions(-)
---------------------------
commit 6ca6cca31ddc7cc8b1dc38b12d7593d2667defe8
Author: Steven Rostedt <srostedt@...hat.com>
Date:   Tue Sep 15 12:24:22 2009 -0400

    tracing: optimize global_trace_clock cachelines
    
    The prev_trace_clock_time is only read or written to when the
    trace_clock_lock is taken. For better perfomance, they
    should share the same cache line.
    
    Reported-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
    Signed-off-by: Steven Rostedt <rostedt@...dmis.org>

diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c
index b588fd8..20c5f92 100644
--- a/kernel/trace/trace_clock.c
+++ b/kernel/trace/trace_clock.c
@@ -66,10 +66,14 @@ u64 notrace trace_clock(void)
  * Used by plugins that need globally coherent timestamps.
  */
 
-static u64 prev_trace_clock_time;
-
-static raw_spinlock_t trace_clock_lock ____cacheline_aligned_in_smp =
-	(raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED;
+/* keep prev_time and lock in the same cacheline. */
+static struct {
+	u64 prev_time;
+	raw_spinlock_t lock;
+} trace_clock_struct ____cacheline_aligned_in_smp =
+	{
+		.lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED,
+	};
 
 u64 notrace trace_clock_global(void)
 {
@@ -88,19 +92,19 @@ u64 notrace trace_clock_global(void)
 	if (unlikely(in_nmi()))
 		goto out;
 
-	__raw_spin_lock(&trace_clock_lock);
+	__raw_spin_lock(&trace_clock_struct.lock);
 
 	/*
 	 * TODO: if this happens often then maybe we should reset
-	 * my_scd->clock to prev_trace_clock_time+1, to make sure
+	 * my_scd->clock to prev_time+1, to make sure
 	 * we start ticking with the local clock from now on?
 	 */
-	if ((s64)(now - prev_trace_clock_time) < 0)
-		now = prev_trace_clock_time + 1;
+	if ((s64)(now - trace_clock_struct.prev_time) < 0)
+		now = trace_clock_struct.prev_time + 1;
 
-	prev_trace_clock_time = now;
+	trace_clock_struct.prev_time = now;
 
-	__raw_spin_unlock(&trace_clock_lock);
+	__raw_spin_unlock(&trace_clock_struct.lock);
 
  out:
 	raw_local_irq_restore(flags);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/