[patch 25/24] x86-64: Mark rdtsc as sync only for netburst, not for core2
From: Chris Wright
Date: Fri Dec 15 2006 - 13:19:54 EST
2.6.18-stable review patch. If anyone has any objections, please let us know.
------------------
From: Arjan van de Ven <arjan@xxxxxxxxxxxxxxx>
On the Core2 cpus, the rdtsc instruction is not serializing (as defined
in the architecture reference since rdtsc exists) and due to the deep
speculation of these cores, it's possible that you can observe time go
backwards between cores due to this speculation. Since the kernel
already deals with this with the SYNC_RDTSC flag, the solution is
simple, only assume that the instruction is serializing on family 15...
The price one pays for this is a slightly slower gettimeofday (by a
dozen or two cycles), but that increase is quite small to pay for a
really-going-forward tsc counter.
Signed-off-by: Arjan van de Ven <arjan@xxxxxxxxxxxxxxx>
Signed-off-by: Andi Kleen <ak@xxxxxxx>
[chrisw: backported to 2.6.18]
Signed-off-by: Chris Wright <chrisw@xxxxxxxxxxxx>
---
Commit: f3d73707a1e84f0687a05144b70b660441e999c7
Author: Arjan van de Ven <arjan@xxxxxxxxxxxxxxx>
Date: Thu Dec 7 02:14:12 2006 +0100
arch/x86_64/kernel/setup.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
--- linux-2.6.18.5.orig/arch/x86_64/kernel/setup.c
+++ linux-2.6.18.5/arch/x86_64/kernel/setup.c
@@ -1010,7 +1010,10 @@ static void __cpuinit init_intel(struct
if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
(c->x86 == 0x6 && c->x86_model >= 0x0e))
set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
- set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
+ if (c->x86 == 15)
+ set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
+ else
+ clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
c->x86_max_cores = intel_num_cpu_cores(c);
srat_detect_node();
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/