kern/167106: [eventtimers] eventtimer tsc smp_test unreliable
Jim Harris
jimharris at freebsd.org
Wed Jul 25 05:40:06 UTC 2012
The following reply was made to PR kern/167106; it has been noted by GNATS.
From: Jim Harris <jimharris at freebsd.org>
To: bug-followup at FreeBSD.org, claudius at ambtec.de
Cc:
Subject: Re: kern/167106: [eventtimers] eventtimer tsc smp_test unreliable
Date: Tue, 24 Jul 2012 22:30:31 -0700
I am almost certain kern/167106 is fixed by r238755, committed today.
kib@ has already approved this to be MFC'd to stable/9 for upcoming
9.1 release. MFC will happen on Friday (27th July) barring any
unforeseen issues with the patch.
Full commit message below.
Regards,
-Jim
Author: jimharris
Date: Tue Jul 24 22:10:11 2012
New Revision: 238755
URL: http://svn.freebsd.org/changeset/base/238755
Log:
Add rmb() to tsc_read_##x to enforce serialization of rdtsc captures.
Intel Architecture Manual specifies that rdtsc instruction is not serialized,
so without this change, TSC synchronization test would periodically fail,
resulting in use of HPET timecounter instead of TSC-low. This caused
severe performance degradation (40-50%) when running high IO/s
workloads due to
HPET MMIO reads and GEOM stat collection.
Tests on Xeon E5-2600 (Sandy Bridge) 8C systems were seeing TSC
synchronization
fail approximately 20% of the time.
Sponsored by: Intel
Reviewed by: kib
MFC after: 3 days
Modified:
head/sys/x86/x86/tsc.c
Modified: head/sys/x86/x86/tsc.c
==============================================================================
--- head/sys/x86/x86/tsc.c Tue Jul 24 20:15:41 2012 (r238754)
+++ head/sys/x86/x86/tsc.c Tue Jul 24 22:10:11 2012 (r238755)
@@ -328,6 +328,7 @@ init_TSC(void)
#ifdef SMP
+/* rmb is required here because rdtsc is not a serializing instruction. */
#define TSC_READ(x) \
static void \
tsc_read_##x(void *arg) \
@@ -335,6 +336,7 @@ tsc_read_##x(void *arg) \
uint32_t *tsc = arg; \
u_int cpu = PCPU_GET(cpuid); \
\
+ rmb(); \
tsc[cpu * 3 + x] = rdtsc32(); \
}
TSC_READ(0)
More information about the freebsd-bugs
mailing list