On Oct 21, 2016, at 3:08 AM, Michal Hocko [off-list ref] wrote:

Interesting.
$ cat /debug/tracing/available_tracers 
function_graph preemptirqsoff preemptoff irqsoff function nop

Do I have to configure anything specially? And if I do why isn't it any
better to simply add a start tracepoint and make this available also to
older kernels?

Well, you just need to enable the function_graph tracer in the tracing directory:
echo function_graph > current_tracer,

set funcgraph-proc in trace_options to get process information:
echo funcgraph-proc > trace_options,

set funcgraph-abstime in trace_options to get timestamp,
echo funcgraph-abstime > trace_options

set all the functions we’d like to observe as filters. For e.g.
echo __alloc_pages_nodemask > set_ftrace_filter

and enable the tracepoints we would like to get information from.

I didn’t add a begin tracepoint for this as Steven Rostedt had suggested
using function graph instead of begin/end tracepoints (we already have
a tracepoint in __alloc_pages_nodemask - trace_mm_page_alloc to get 
some information about the allocation and we can just use function graph 
to see how long __alloc_pages_nodemask() takes).

Janani

-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>