Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1276535

Re: [PATCH 34/37] perf hists browser: Support flat callchains

Show all headers | View raw

Em Tue, Nov 24, 2015 at 02:27:08PM +0900, Namhyung Kim escreveu:
> On Mon, Nov 23, 2015 at 04:16:48PM +0100, Frederic Weisbecker wrote:
> > On Thu, Nov 19, 2015 at 02:53:20PM -0300, Arnaldo Carvalho de Melo wrote:
> > > From: Namhyung Kim <namhyung@kernel.org>
> > [...]
> > 
> > > +int callchain_node__make_parent_list(struct callchain_node *node)
> > > +{
> > > +	struct callchain_node *parent = node->parent;
> > > +	struct callchain_list *chain, *new;
> > > +	LIST_HEAD(head);
> > > +
> > > +	while (parent) {
> > > +		list_for_each_entry_reverse(chain, &parent->val, list) {
> > > +			new = malloc(sizeof(*new));
> > > +			if (new == NULL)
> > > +				goto out;
> > > +			*new = *chain;
> > > +			new->has_children = false;
> > > +			list_add_tail(&new->list, &head);
> > > +		}
> > > +		parent = parent->parent;
> > > +	}
> > > +
> > > +	list_for_each_entry_safe_reverse(chain, new, &head, list)
> > > +		list_move_tail(&chain->list, &node->parent_val);
> > > +
> > > +	if (!list_empty(&node->parent_val)) {
> > > +		chain = list_first_entry(&node->parent_val, struct callchain_list, list);
> > > +		chain->has_children = rb_prev(&node->rb_node) || rb_next(&node->rb_node);
> > > +
> > > +		chain = list_first_entry(&node->val, struct callchain_list, list);
> > > +		chain->has_children = false;
> > 
> > I'm a bit puzzled with this, can't we rewind through the parents on printing or adding
> > to the flat rbtree instead of having this parent_val field?
> 
> Yes, this code is to simplify things on parent nodes.  Maybe we could
> go up to parents and print the callchain list there as you said.
> 
> However, problem I think is how to handle 'has_children' information
> on parents.  That info controls folding status of each callchain.  As
> the info is in the struct callchain_list and flat or folded callchain
> mode require the info should be in the top-most entry, I cannot share
> entries in parent nodes.
> 
> Thus I simply copied callchain lists in parents to leaf nodes.  Yes,
> it will consume some memory but can simplify the code.

I haven't done any measuring, but I'm noticing that 'perf top -g' is
showing more warnings about not being able to process events fast enough
and so ends up losing events, I tried with --max-stack 16 and it helped,
this is just a heads up.

Perhaps my workstation workloads are gettning deeper callchains over
time, but perhaps this is the cost of processing callchains that is
increasing, I need to stop and try to quantify this.

We really need to look at reducing the overhead of processing
callchains.

- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[GIT PULL 00/37] perf/core improvements and fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 31/37] perf callchain: Add count fields to struct callchain_node Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
    Re: [PATCH 31/37] perf callchain: Add count fields to struct  callchain_node Frederic Weisbecker <fweisbec@gmail.com> - 2015-11-23 15:40 +0100
      Re: [PATCH 31/37] perf callchain: Add count fields to struct  callchain_node Namhyung Kim <namhyung@kernel.org> - 2015-11-24 06:20 +0100
  [PATCH 27/37] perf tools: Fix machine__create_kernel_maps to put kernel dso refcount Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 06/37] perf bpf: Add BPF_PROLOGUE config options for further patches Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 17/37] perf test: Print result for each LLVM subtest Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 30/37] perf callchain: Abstract callchain print function Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 10/37] perf bpf: Allow BPF program config probing options Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 35/37] perf hists browser: Support folded callchains Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 37/37] perf ui/gtk: Support folded callchains Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 36/37] perf ui/gtk: Support flat callchains Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 34/37] perf hists browser: Support flat callchains Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
    Re: [PATCH 34/37] perf hists browser: Support flat callchains Frederic Weisbecker <fweisbec@gmail.com> - 2015-11-23 16:20 +0100
      Re: [PATCH 34/37] perf hists browser: Support flat callchains Namhyung Kim <namhyung@kernel.org> - 2015-11-24 06:30 +0100
        Re: [PATCH 34/37] perf hists browser: Support flat callchains Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com> - 2015-11-24 15:50 +0100
          Re: [PATCH 34/37] perf hists browser: Support flat callchains Namhyung Kim <namhyung@kernel.org> - 2015-11-25 02:30 +0100
            Re: [PATCH 34/37] perf hists browser: Support flat callchains Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com> - 2015-11-25 02:40 +0100
              Re: [PATCH 34/37] perf hists browser: Support flat callchains Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com> - 2015-11-25 03:20 +0100
                Re: [PATCH 34/37] perf hists browser: Support flat callchains Namhyung Kim <namhyung@kernel.org> - 2015-11-25 22:10 +0100
  [PATCH 03/37] tools: Adopt memdup() from tools/perf, moving it to tools/lib/string.c Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 02/37] tools: Fix selftests_install Makefile rule Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 33/37] perf hists browser: Factor out hist_browser__show_callchain_list() Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 32/37] perf report: Add callchain value option Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:00 +0100
  [PATCH 08/37] perf bpf: Allow BPF program attach to uprobe events Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 21/37] perf machine: Fix machine__findnew_module_map to put registered map Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 14/37] perf test: Fix 'perf test BPF' when it fails to find a suitable vmlinux Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 13/37] perf test: Test the BPF prologue adding infrastructure Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 16/37] perf tests: Pass the subtest index to each test routine Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 18/37] perf test: Print result for each BPF subtest Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 11/37] perf bpf: Add prologue for BPF programs for fetching arguments Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 19/37] perf test: Mute test cases error messages if verbose == 0 Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 22/37] perf machine: Fix machine__destroy_kernel_maps to drop vmlinux_maps references Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 12/37] perf bpf: Generate prologue for BPF programs Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 20/37] perf probe: Fix to free temporal Dwarf_Frame Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 25/37] perf tools: Fix to put new map after inserting to map_groups in dso__load_sym Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 28/37] perf machine: Fix machine__findnew_module_map to put dso Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 26/37] perf tools: Fix __dsos__addnew to put dso after adding it to the list Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 04/37] tools: Clone the kernel's strtobool function Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 23/37] perf machine: Fix to destroy kernel maps when machine exits Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  [PATCH 05/37] bpf tools: Load a program with different instances using preprocessor Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-11-19 19:10 +0100
  RE: [GIT PULL 00/37] perf/core improvements and fixes 平松雅巳 / HIRAMATU，MASAMI   <masami.hiramatsu.pt@hitachi.com> - 2015-11-20 11:10 +0100
    Re: [GIT PULL 00/37] perf/core improvements and fixes 'Arnaldo Carvalho de Melo' <acme@kernel.org> - 2015-11-20 13:10 +0100
      RE: [GIT PULL 00/37] perf/core improvements and fixes 平松雅巳 / HIRAMATU，MASAMI   <masami.hiramatsu.pt@hitachi.com> - 2015-11-20 18:00 +0100
  Re: [GIT PULL 00/37] perf/core improvements and fixes Ingo Molnar <mingo@kernel.org> - 2015-11-23 09:20 +0100

csiph-web