On 06/22/2010 05:08 AM, Zhang, Yanmin wrote:
Something that is worrying is that we don't expose group information.
perf will multiplex the events for us, but there will be a loss in accuracy.
#ifdef CONFIG_HAVE_HW_BREAKPOINT
#includeasm/hw_breakpoint.h
#endif
@@ -753,6 +752,20
On Tue, 2010-06-22 at 12:12 +0300, Avi Kivity wrote:
On 06/22/2010 05:08 AM, Zhang, Yanmin wrote:
Something that is worrying is that we don't expose group information.
perf will multiplex the events for us, but there will be a loss in
accuracy.
#ifdef
On 06/22/2010 12:25 PM, Zhang, Yanmin wrote:
I'm talking about the guest/host interface. So you have one vmexit and
many host perf calls.
I understood what you were speaking. I mean, perf generic codes operate
perf_event
one by one. At low layer, we just know one perf_event before
The 2nd patch is to change the definition of perf_event to facilitate
perf attr copy when a hypercall happens.
Signed-off-by: Zhang Yanmin yanmin_zh...@linux.intel.com
---
--- linux-2.6_tip0620/include/linux/perf_event.h2010-06-21
15:19:52.821999849 +0800
+++
On 06/21/2010 12:31 PM, Zhang, Yanmin wrote:
The 2nd patch is to change the definition of perf_event to facilitate
perf attr copy when a hypercall happens.
Signed-off-by: Zhang Yanminyanmin_zh...@linux.intel.com
---
--- linux-2.6_tip0620/include/linux/perf_event.h2010-06-21
On Mon, 2010-06-21 at 15:00 +0300, Avi Kivity wrote:
On 06/21/2010 12:31 PM, Zhang, Yanmin wrote:
The 2nd patch is to change the definition of perf_event to facilitate
perf attr copy when a hypercall happens.
Signed-off-by: Zhang Yanminyanmin_zh...@linux.intel.com
---
---