linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Namhyung Kim <namhyung@kernel.org>
To: Jiri Olsa <jolsa@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Paul Mackerras <paulus@samba.org>, Ingo Molnar <mingo@kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Stephane Eranian <eranian@google.com>
Subject: Re: [PATCH 07/14] perf ui/hist: Add support for event group view
Date: Thu, 20 Dec 2012 11:59:01 +0900	[thread overview]
Message-ID: <20121220025901.GB3679@danjae> (raw)
In-Reply-To: <20121218154710.GN1283@krava.brq.redhat.com>

On Tue, Dec 18, 2012 at 04:47:10PM +0100, Jiri Olsa wrote:
> On Mon, Dec 17, 2012 at 03:38:58PM +0900, Namhyung Kim wrote:
> > From: Namhyung Kim <namhyung.kim@lge.com>
> > 
> > Show group members' overhead also when showing the leader's if event
> > group is enabled.  Use macro for defining hpp functions which looks
> > almost identical.
> 
> SNIP
> 
> > -
> > -static int hpp__color_overhead_us(struct perf_hpp_fmt *fmt __maybe_unused,
> > -				  struct perf_hpp *hpp, struct hist_entry *he)
> > +static int __hpp__percent_fmt(struct perf_hpp *hpp, struct hist_entry *he,
> > +			      u64 (*get_field)(struct hist_entry *),
> > +			      const char *fmt, hpp_snprint_fn print_fn)
> 
> it's hard to tell from diff, but the function looks like this (group part):

Exactly.

> 
>         if (symbol_conf.event_group) {
>                 int prev_idx, idx_delta, i;
>                 struct perf_evsel *evsel = hists_to_evsel(hists);
>                 struct hist_entry *pair;
>                 int nr_members = evsel->nr_members;
> 
>                 if (nr_members <= 1)
>                         return ret;
> 
>                 prev_idx = perf_evsel__group_idx(evsel);
> 
>                 list_for_each_entry(pair, &he->pairs.head, pairs.node) {
>                         u64 period = get_field(pair);
>                         u64 total = pair->hists->stats.total_period;
> 
>                         if (!total)
>                                 continue;
> 
>                         evsel = hists_to_evsel(pair->hists);
>                         idx_delta = perf_evsel__group_idx(evsel) - prev_idx - 1;
> 
>                         for (i = 0; i < idx_delta; i++) {
>                                 ret += print_fn(hpp->buf + ret, hpp->size - ret,
>                                                 fmt, 0.0);
>                         }
> 
>                         ret += print_fn(hpp->buf + ret, hpp->size - ret,
>                                         fmt, 100.0 * period / total);
> 
>                         prev_idx += 1 + idx_delta;
>                 }
> 
>                 idx_delta = nr_members - prev_idx - 1;
> 
>                 for (i = 0; i < idx_delta; i++) {
>                         ret += print_fn(hpp->buf + ret, hpp->size - ret,
>                                         fmt, 0.0);
>                 }
>         }
> 
> The temporary array (initial send) was more clear to me.. but if
> this is the preffered way I wont object, since this is another place
> that would benefit from hist_entries array linking.. when (and if)
> there's ever patch for that ;-)
> 
> I think this code now suffers from mischosen data layer

Yes, I admit that using array can make things simple.  But I also don't
like a temporary array due to its runtime cost.  Linking hist entries by
array can be a solution since it can be reused for output resorting and
(multiple) hpp column(s).  In addition, it might have a smaller memory
footprint than the list method for the leader-only sample feature since
it'll make every hist entries to be linked IMHO.

Thanks,
Namhyung

  reply	other threads:[~2012-12-20  2:59 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-12-17  6:38 [PATCH 00/14] perf report: Add support for event group view (v7) Namhyung Kim
2012-12-17  6:38 ` [PATCH 01/14] perf tools: Keep group information Namhyung Kim
2012-12-18 15:28   ` Jiri Olsa
2012-12-17  6:38 ` [PATCH 02/14] perf test: Add group test conditions Namhyung Kim
2012-12-18 15:28   ` Jiri Olsa
2012-12-17  6:38 ` [PATCH 03/14] perf header: Ensure read/write finished successfully Namhyung Kim
2012-12-18 15:30   ` Jiri Olsa
2013-01-25 11:06   ` [tip:perf/core] perf header: Ensure read/ write " tip-bot for Namhyung Kim
2012-12-17  6:38 ` [PATCH 04/14] perf header: Add HEADER_GROUP_DESC feature Namhyung Kim
2012-12-17  6:38 ` [PATCH 05/14] perf report: Make another loop for linking group hists Namhyung Kim
2012-12-18 15:32   ` Jiri Olsa
2012-12-17  6:38 ` [PATCH 06/14] perf hists: Resort hist entries using group members for output Namhyung Kim
2012-12-18 15:39   ` Jiri Olsa
2012-12-17  6:38 ` [PATCH 07/14] perf ui/hist: Add support for event group view Namhyung Kim
2012-12-18 15:47   ` Jiri Olsa
2012-12-20  2:59     ` Namhyung Kim [this message]
2012-12-18 19:30   ` Arnaldo Carvalho de Melo
2012-12-20  2:34     ` Namhyung Kim
2012-12-17  6:38 ` [PATCH 08/14] perf hist browser: " Namhyung Kim
2012-12-17  6:39 ` [PATCH 09/14] perf gtk/browser: " Namhyung Kim
2012-12-17  6:39 ` [PATCH 10/14] perf gtk/browser: Trim column header string when event group enabled Namhyung Kim
2012-12-17  6:39 ` [PATCH 11/14] perf report: Bypass non-leader events when event group is enabled Namhyung Kim
2012-12-17  6:39 ` [PATCH 12/14] perf report: Show group description " Namhyung Kim
2012-12-17  6:39 ` [PATCH 13/14] perf report: Add --group option Namhyung Kim
2012-12-17  6:39 ` [PATCH 14/14] perf report: Add report.group config option Namhyung Kim

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20121220025901.GB3679@danjae \
    --to=namhyung@kernel.org \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@ghostprotocols.net \
    --cc=eranian@google.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=paulus@samba.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).