[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130623215327.GA995@krava.redhat.com>
Date: Sun, 23 Jun 2013 23:53:27 +0200
From: Jiri Olsa <jolsa@...hat.com>
To: Greg Price <price@....edu>
Cc: Arnaldo Carvalho de Melo <acme@...stprotocols.net>,
linux-kernel@...r.kernel.org,
Peter Zijlstra <a.p.zijlstra@...llo.nl>,
Paul Mackerras <paulus@...ba.org>,
Ingo Molnar <mingo@...hat.com>, David Ahern <dsahern@...il.com>
Subject: Re: [PATCH] perf report: Add option to collapse undesired parts of
call graph
On Sat, Jun 22, 2013 at 11:17:20PM -0400, Greg Price wrote:
> For example, in an application with an expensive function
> implemented with deeply nested recursive calls, the default
> call-graph presentation is dominated by the different callchains
> within that function. By treating the function as a black box,
> we can collect the callchains leading into the function and
> compactly identify what to blame for expensive calls.
>
> For example, in this report the callers of garbage_collect() are
> scattered across the tree:
> $ perf report -d ruby 2>- | grep -m10 ^[^#]*[a-z]
> 22.03% ruby [.] gc_mark
> --- gc_mark
> |--59.40%-- mark_keyvalue
> | st_foreach
> | gc_mark_children
> | |--99.75%-- rb_gc_mark
> | | rb_vm_mark
> | | gc_mark_children
> | | gc_marks
> | | |--99.00%-- garbage_collect
>
> If we make garbage_collect() a black box, its callers are coalesced:
> $ perf report --blackbox garbage_collect -d ruby 2>- | grep -m10 ^[^#]*[a-z]
> 72.92% ruby [.] garbage_collect
> --- garbage_collect
> vm_xmalloc
> |--47.08%-- ruby_xmalloc
> | st_insert2
> | rb_hash_aset
> | |--98.45%-- features_index_add
> | | rb_provide_feature
> | | rb_require_safe
> | | vm_call_method
Seems useful, sort of oposite to parent option (-p)
few comments below
SNIP
> * Special case: if there's an argument left then assume that
> diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
> index 67bdb9f..abec83d 100644
> --- a/tools/perf/builtin-top.c
> +++ b/tools/perf/builtin-top.c
> @@ -775,8 +775,7 @@ static void perf_event__process_sample(struct perf_tool *tool,
> sample->callchain) {
> err = machine__resolve_callchain(machine, evsel,
> al.thread, sample,
> - &parent);
> -
> + &parent, NULL);
> if (err)
> return;
> }
Any reason why not add this for top?
> diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
> index b2ecad6..a14489c 100644
> --- a/tools/perf/util/machine.c
> +++ b/tools/perf/util/machine.c
> @@ -11,6 +11,10 @@
> #include <stdbool.h>
> #include "unwind.h"
>
> +regex_t blackbox_regex;
> +const char *blackbox_pattern;
> +int have_blackbox = 0;
util/sort.c mich be better place for this
It could also make sense to allow sorting on this
the same way as we do for '-s parent' and report only
'[other]' and 'blackbox' entries.
Also I dont like the 'blackbox' option name, it should
complement the parent option somehow.. but no idea ;-)
I tested this on separate example and numbers seem ok.
jirka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists