On Tue, Oct 23, 2012 at 03:13:52PM +0200, Peter Zijlstra wrote:
> > @@ -562,6 +565,7 @@ enum perf_event_type {
> > * { u64 stream_id;} && PERF_SAMPLE_STREAM_ID
> > * { u32 cpu, res; } && PERF_SAMPLE_CPU
> > * { u64
On Thu, 2012-10-18 at 16:19 -0700, Andi Kleen wrote:
> @@ -601,6 +602,7 @@ static inline void perf_sample_data_init(struct
> perf_sample_data *data,
> data->regs_user.abi = PERF_SAMPLE_REGS_ABI_NONE;
> data->regs_user.regs = NULL;
> data->stack_user_size = 0;
> +
On Thu, 2012-10-18 at 16:19 -0700, Andi Kleen wrote:
@@ -601,6 +602,7 @@ static inline void perf_sample_data_init(struct
perf_sample_data *data,
data-regs_user.abi = PERF_SAMPLE_REGS_ABI_NONE;
data-regs_user.regs = NULL;
data-stack_user_size = 0;
+ data-weight
On Tue, Oct 23, 2012 at 03:13:52PM +0200, Peter Zijlstra wrote:
@@ -562,6 +565,7 @@ enum perf_event_type {
* { u64 stream_id;} PERF_SAMPLE_STREAM_ID
* { u32 cpu, res; } PERF_SAMPLE_CPU
* { u64
From: Andi Kleen
For some events it's useful to weight sample with a hardware
provided number. This expresses how expensive the action the
sample represent was. This allows the profiler to scale
the samples to be more informative to the programmer.
There is already the period which is used
From: Andi Kleen a...@linux.intel.com
For some events it's useful to weight sample with a hardware
provided number. This expresses how expensive the action the
sample represent was. This allows the profiler to scale
the samples to be more informative to the programmer.
There is already the
6 matches
Mail list logo