[tip:perf/core] perf/x86: Fix wrmsrl() debug wrapper

[tip:perf/core] perf/x86: Fix wrmsrl() debug wrapper

[tip-bot for Peter Zijlstra]

Commit-ID:  1c2ac3fde3e35279958e7b0408e2dcf866465301
Gitweb:     http://git.kernel.org/tip/1c2ac3fde3e35279958e7b0408e2dcf866465301
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Mon, 14 May 2012 15:25:34 +0200
Committer:  Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 6 Jun 2012 17:23:22 +0200

perf/x86: Fix wrmsrl() debug wrapper

Move the wrmslr() debug wrapper to the common header now that all the
include games are gone. Also clean it up a bit to avoid multiple
evaluation of the argument.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-l4gkfnivwv4yi5mqxjlovymx@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
 arch/x86/kernel/cpu/perf_event.c |   11 -----------
 arch/x86/kernel/cpu/perf_event.h |   12 ++++++++++++
 2 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index c4706cf..43c2017 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -35,17 +35,6 @@
 
 #include "perf_event.h"
 
-#if 0
-#undef wrmsrl
-#define wrmsrl(msr, val) 					\
-do {								\
-	trace_printk("wrmsrl(%lx, %lx)\n", (unsigned long)(msr),\
-			(unsigned long)(val));			\
-	native_write_msr((msr), (u32)((u64)(val)), 		\
-			(u32)((u64)(val) >> 32));		\
-} while (0)
-#endif
-
 struct x86_pmu x86_pmu __read_mostly;
 
 DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index 7241e2f..23b5710 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -14,6 +14,18 @@
 
 #include <linux/perf_event.h>
 
+#if 0
+#undef wrmsrl
+#define wrmsrl(msr, val) 						\
+do {									\
+	unsigned int _msr = (msr);					\
+	u64 _val = (val);						\
+	trace_printk("wrmsrl(%x, %Lx)\n", (unsigned int)(_msr),		\
+			(unsigned long long)(_val));			\
+	native_write_msr((_msr), (u32)(_val), (u32)(_val >> 32));	\
+} while (0)
+#endif
+
 /*
  *          |   NHM/WSM    |      SNB     |
  * register -------------------------------

Re: [tip:perf/core] perf/x86: Fix wrmsrl() debug wrapper

[H. Peter Anvin]

On 06/06/2012 09:15 AM, tip-bot for Peter Zijlstra wrote:
> 
> perf/x86: Fix wrmsrl() debug wrapper
> 
> Move the wrmslr() debug wrapper to the common header now that all the
> include games are gone. Also clean it up a bit to avoid multiple
> evaluation of the argument.
> 
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> Link: http://lkml.kernel.org/n/tip-l4gkfnivwv4yi5mqxjlovymx@git.kernel.org
> Signed-off-by: Ingo Molnar <mingo@kernel.org>

Could you please call this something like trace_wrmsrl() instead of
silently tacking something onto something else that already had meaning?

	-hpa

-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.

Re: [tip:perf/core] perf/x86: Fix wrmsrl() debug wrapper

[Peter Zijlstra]

On Wed, 2012-06-06 at 09:21 -0700, H. Peter Anvin wrote:
> On 06/06/2012 09:15 AM, tip-bot for Peter Zijlstra wrote:
> > 
> > perf/x86: Fix wrmsrl() debug wrapper
> > 
> > Move the wrmslr() debug wrapper to the common header now that all the
> > include games are gone. Also clean it up a bit to avoid multiple
> > evaluation of the argument.
> > 
> > Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> > Link: http://lkml.kernel.org/n/tip-l4gkfnivwv4yi5mqxjlovymx@git.kernel.org
> > Signed-off-by: Ingo Molnar <mingo@kernel.org>
> 
> Could you please call this something like trace_wrmsrl() instead of
> silently tacking something onto something else that already had meaning?

Its #if 0, its broken on CONFIG_PARAVIRT etc.. The whole purpose was to
hijack wrmsrl() for perf_event*.c so that I don't have to do
s/wrmsrl/trace_&/ on those files.

Its a crude debug aid, nothing more.