Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core

Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:

New features:

 - Add 'L' hotkey to dynamicly set the percent threshold for histogram
   entries and callchains, i.e. dynamicly do what the --percent-limit
   command line option to 'top' and 'report' does. (Namhyung Kim)

Infrastructure changes:

 - Per hists field and sort lists, that will be used, for instance,
   in the c2c tool (Jiri Olsa)

Documentation changes:

 - Update documentation of --sort and --perf-limit options
   for 'perf report' (Namhyung Kim)

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

14 files changed, 425 insertions, 236 deletions

diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 8a301f6afb37..89cab84e92fd 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -117,6 +117,22 @@ OPTIONS
 	And default sort keys are changed to comm, dso_from, symbol_from, dso_to
 	and symbol_to, see '--branch-stack'.
 
+	If the --mem-mode option is used, the following sort keys are also available
+	(incompatible with --branch-stack):
+	symbol_daddr, dso_daddr, locked, tlb, mem, snoop, dcacheline.
+
+	- symbol_daddr: name of data symbol being executed on at the time of sample
+	- dso_daddr: name of library or module containing the data being executed
+	on at the time of the sample
+	- locked: whether the bus was locked at the time of the sample
+	- tlb: type of tlb access for the data at the time of the sample
+	- mem: type of memory access for the data at the time of the sample
+	- snoop: type of snoop (if any) for the data at the time of the sample
+	- dcacheline: the cacheline the data address is on at the time of the sample
+
+	And the default sort keys are changed to local_weight, mem, sym, dso,
+	symbol_daddr, dso_daddr, snoop, tlb, locked, see '--mem-mode'.
+
 	If the data file has tracepoint event(s), following (dynamic) sort keys
 	are also available:
 	trace, trace_fields, [<event>.]<field>[/raw]
@@ -151,22 +167,6 @@ OPTIONS
 	By default, every sort keys not specified in -F will be appended
 	automatically.
 
-	If --mem-mode option is used, following sort keys are also available
-	(incompatible with --branch-stack):
-	symbol_daddr, dso_daddr, locked, tlb, mem, snoop, dcacheline.
-
-	- symbol_daddr: name of data symbol being executed on at the time of sample
-	- dso_daddr: name of library or module containing the data being executed
-	on at the time of sample
-	- locked: whether the bus was locked at the time of sample
-	- tlb: type of tlb access for the data at the time of sample
-	- mem: type of memory access for the data at the time of sample
-	- snoop: type of snoop (if any) for the data at the time of sample
-	- dcacheline: the cacheline the data address is on at the time of sample
-
-	And default sort keys are changed to local_weight, mem, sym, dso,
-	symbol_daddr, dso_daddr, snoop, tlb, locked, see '--mem-mode'.
-
 -p::
 --parent=<regex>::
         A regex filter to identify parent. The parent is a caller of this
@@ -351,7 +351,10 @@ OPTIONS
 
 --percent-limit::
 	Do not show entries which have an overhead under that percent.
-	(Default: 0).
+	(Default: 0).  Note that this option also sets the percent limit (threshold)
+	of callchains.  However the default value of callchain threshold is
+	different than the default value of hist entries.  Please see the
+	--call-graph option for details.
 
 --percentage::
 	Determine how to display the overhead percentage of filtered entries.
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index cc5c1267c738..cfe366375c4b 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -245,7 +245,7 @@ static int __cmd_annotate(struct perf_annotate *ann)
 			hists__collapse_resort(hists, NULL);
 			/* Don't sort callchain */
 			perf_evsel__reset_sample_bit(pos, CALLCHAIN);
-			hists__output_resort(hists, NULL);
+			perf_evsel__output_resort(pos, NULL);
 
 			if (symbol_conf.event_group &&
 			    !perf_evsel__is_group_leader(pos))
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 72ed0b46d5a1..1eab50ac1ef6 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -507,7 +507,7 @@ static void report__output_resort(struct report *rep)
 	ui_progress__init(&prog, rep->nr_entries, "Sorting events for output...");
 
 	evlist__for_each(rep->session->evlist, pos)
-		hists__output_resort(evsel__hists(pos), &prog);
+		perf_evsel__output_resort(pos, &prog);
 
 	ui_progress__finish();
 }
@@ -912,15 +912,6 @@ repeat:
 		symbol_conf.cumulate_callchain = false;
 	}
 
-	if (setup_sorting(session->evlist) < 0) {
-		if (sort_order)
-			parse_options_usage(report_usage, options, "s", 1);
-		if (field_order)
-			parse_options_usage(sort_order ? NULL : report_usage,
-					    options, "F", 1);
-		goto error;
-	}
-
 	/* Force tty output for header output and per-thread stat. */
 	if (report.header || report.header_only || report.show_threads)
 		use_browser = 0;
@@ -930,6 +921,15 @@ repeat:
 	else
 		use_browser = 0;
 
+	if (setup_sorting(session->evlist) < 0) {
+		if (sort_order)
+			parse_options_usage(report_usage, options, "s", 1);
+		if (field_order)
+			parse_options_usage(sort_order ? NULL : report_usage,
+					    options, "F", 1);
+		goto error;
+	}
+
 	if (report.header || report.header_only) {
 		perf_session__fprintf_info(session, stdout,
 					   report.show_full_info);
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index bf01cbb0ef23..a75de3940b97 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -252,7 +252,8 @@ static void perf_top__print_sym_table(struct perf_top *top)
 	char bf[160];
 	int printed = 0;
 	const int win_width = top->winsize.ws_col - 1;
-	struct hists *hists = evsel__hists(top->sym_evsel);
+	struct perf_evsel *evsel = top->sym_evsel;
+	struct hists *hists = evsel__hists(evsel);
 
 	puts(CONSOLE_CLEAR);
 
@@ -288,7 +289,7 @@ static void perf_top__print_sym_table(struct perf_top *top)
 	}
 
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(evsel, NULL);
 
 	hists__output_recalc_col_len(hists, top->print_entries - printed);
 	putchar('\n');
@@ -540,6 +541,7 @@ static bool perf_top__handle_keypress(struct perf_top *top, int c)
 static void perf_top__sort_new_samples(void *arg)
 {
 	struct perf_top *t = arg;
+	struct perf_evsel *evsel = t->sym_evsel;
 	struct hists *hists;
 
 	perf_top__reset_sample_counters(t);
@@ -547,7 +549,7 @@ static void perf_top__sort_new_samples(void *arg)
 	if (t->evlist->selected != NULL)
 		t->sym_evsel = t->evlist->selected;
 
-	hists = evsel__hists(t->sym_evsel);
+	hists = evsel__hists(evsel);
 
 	if (t->evlist->enabled) {
 		if (t->zero) {
@@ -559,7 +561,7 @@ static void perf_top__sort_new_samples(void *arg)
 	}
 
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(evsel, NULL);
 }
 
 static void *display_thread_tui(void *arg)
@@ -1243,6 +1245,13 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 	/* display thread wants entries to be collapsed in a different tree */
 	sort__need_collapse = 1;
 
+	if (top.use_stdio)
+		use_browser = 0;
+	else if (top.use_tui)
+		use_browser = 1;
+
+	setup_browser(false);
+
 	if (setup_sorting(top.evlist) < 0) {
 		if (sort_order)
 			parse_options_usage(top_usage, options, "s", 1);
@@ -1252,13 +1261,6 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 		goto out_delete_evlist;
 	}
 
-	if (top.use_stdio)
-		use_browser = 0;
-	else if (top.use_tui)
-		use_browser = 1;
-
-	setup_browser(false);
-
 	status = target__validate(target);
 	if (status) {
 		target__strerror(target, status, errbuf, BUFSIZ);
diff --git a/tools/perf/tests/hists_cumulate.c b/tools/perf/tests/hists_cumulate.c
index 5e6a86e50fb9..ecf136c385d5 100644
--- a/tools/perf/tests/hists_cumulate.c
+++ b/tools/perf/tests/hists_cumulate.c
@@ -191,7 +191,7 @@ static int do_test(struct hists *hists, struct result *expected, size_t nr_expec
 	 * function since TEST_ASSERT_VAL() returns in case of failure.
 	 */
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(hists_to_evsel(hists), NULL);
 
 	if (verbose > 2) {
 		pr_info("use callchain: %d, cumulate callchain: %d\n",
diff --git a/tools/perf/tests/hists_filter.c b/tools/perf/tests/hists_filter.c
index 351a42463444..34b945a55d4d 100644
--- a/tools/perf/tests/hists_filter.c
+++ b/tools/perf/tests/hists_filter.c
@@ -145,7 +145,7 @@ int test__hists_filter(int subtest __maybe_unused)
 		struct hists *hists = evsel__hists(evsel);
 
 		hists__collapse_resort(hists, NULL);
-		hists__output_resort(hists, NULL);
+		perf_evsel__output_resort(evsel, NULL);
 
 		if (verbose > 2) {
 			pr_info("Normal histogram\n");
diff --git a/tools/perf/tests/hists_output.c b/tools/perf/tests/hists_output.c
index b231265148d8..23cce67c7e48 100644
--- a/tools/perf/tests/hists_output.c
+++ b/tools/perf/tests/hists_output.c
@@ -156,7 +156,7 @@ static int test1(struct perf_evsel *evsel, struct machine *machine)
 		goto out;
 
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(evsel, NULL);
 
 	if (verbose > 2) {
 		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
@@ -256,7 +256,7 @@ static int test2(struct perf_evsel *evsel, struct machine *machine)
 		goto out;
 
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(evsel, NULL);
 
 	if (verbose > 2) {
 		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
@@ -310,7 +310,7 @@ static int test3(struct perf_evsel *evsel, struct machine *machine)
 		goto out;
 
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(evsel, NULL);
 
 	if (verbose > 2) {
 		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
@@ -388,7 +388,7 @@ static int test4(struct perf_evsel *evsel, struct machine *machine)
 		goto out;
 
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(evsel, NULL);
 
 	if (verbose > 2) {
 		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
@@ -491,7 +491,7 @@ static int test5(struct perf_evsel *evsel, struct machine *machine)
 		goto out;
 
 	hists__collapse_resort(hists, NULL);
-	hists__output_resort(hists, NULL);
+	perf_evsel__output_resort(evsel, NULL);
 
 	if (verbose > 2) {
 		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
index 61d578bf4ffd..a5a5390476ac 100644
--- a/tools/perf/ui/browsers/hists.c
+++ b/tools/perf/ui/browsers/hists.c
@@ -1095,7 +1095,7 @@ static int hist_browser__show_entry(struct hist_browser *browser,
 
 		hist_browser__gotorc(browser, row, 0);
 
-		perf_hpp__for_each_format(fmt) {
+		hists__for_each_format(browser->hists, fmt) {
 			if (perf_hpp__should_skip(fmt, entry->hists) ||
 			    column++ < browser->b.horiz_scroll)
 				continue;
@@ -1175,7 +1175,7 @@ static int hists_browser__scnprintf_headers(struct hist_browser *browser, char *
 			return ret;
 	}
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(browser->hists, fmt) {
 		if (perf_hpp__should_skip(fmt, hists)  || column++ < browser->b.horiz_scroll)
 			continue;
 
@@ -1441,7 +1441,7 @@ static int hist_browser__fprintf_entry(struct hist_browser *browser,
 	if (symbol_conf.use_callchain)
 		printed += fprintf(fp, "%c ", folded_sign);
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(browser->hists, fmt) {
 		if (perf_hpp__should_skip(fmt, he->hists))
 			continue;
 
@@ -2029,6 +2029,42 @@ static void hist_browser__update_nr_entries(struct hist_browser *hb)
 	hb->nr_non_filtered_entries = nr_entries;
 }
 
+static void hist_browser__update_percent_limit(struct hist_browser *hb,
+					       double percent)
+{
+	struct hist_entry *he;
+	struct rb_node *nd = rb_first(&hb->hists->entries);
+	u64 total = hists__total_period(hb->hists);
+	u64 min_callchain_hits = total * (percent / 100);
+
+	hb->min_pcnt = callchain_param.min_percent = percent;
+
+	if (!symbol_conf.use_callchain)
+		return;
+
+	while ((nd = hists__filter_entries(nd, hb->min_pcnt)) != NULL) {
+		he = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (callchain_param.mode == CHAIN_GRAPH_REL) {
+			total = he->stat.period;
+
+			if (symbol_conf.cumulate_callchain)
+				total = he->stat_acc->period;
+
+			min_callchain_hits = total * (percent / 100);
+		}
+
+		callchain_param.sort(&he->sorted_chain, he->callchain,
+				     min_callchain_hits, &callchain_param);
+
+		/* force to re-evaluate folding state of callchains */
+		he->init_have_children = false;
+		hist_entry__set_folding(he, false);
+
+		nd = rb_next(nd);
+	}
+}
+
 static int perf_evsel__hists_browse(struct perf_evsel *evsel, int nr_events,
 				    const char *helpline,
 				    bool left_exits,
@@ -2064,6 +2100,7 @@ static int perf_evsel__hists_browse(struct perf_evsel *evsel, int nr_events,
 	"E             Expand all callchains\n"				\
 	"F             Toggle percentage of filtered entries\n"		\
 	"H             Display column headers\n"			\
+	"L             Change percent limit\n"				\
 	"m             Display context menu\n"				\
 	"S             Zoom into current Processor Socket\n"		\
 
@@ -2104,7 +2141,7 @@ static int perf_evsel__hists_browse(struct perf_evsel *evsel, int nr_events,
 	memset(options, 0, sizeof(options));
 	memset(actions, 0, sizeof(actions));
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(browser->hists, fmt) {
 		perf_hpp__reset_width(fmt, hists);
 		/*
 		 * This is done just once, and activates the horizontal scrolling
@@ -2219,6 +2256,24 @@ static int perf_evsel__hists_browse(struct perf_evsel *evsel, int nr_events,
 				top->zero = !top->zero;
 			}
 			continue;
+		case 'L':
+			if (ui_browser__input_window("Percent Limit",
+					"Please enter the value you want to hide entries under that percent.",
+					buf, "ENTER: OK, ESC: Cancel",
+					delay_secs * 2) == K_ENTER) {
+				char *end;
+				double new_percent = strtod(buf, &end);
+
+				if (new_percent < 0 || new_percent > 100) {
+					ui_browser__warning(&browser->b, delay_secs * 2,
+						"Invalid percent: %.2f", new_percent);
+					continue;
+				}
+
+				hist_browser__update_percent_limit(browser, new_percent);
+				hist_browser__reset(browser);
+			}
+			continue;
 		case K_F1:
 		case 'h':
 		case '?':
diff --git a/tools/perf/ui/gtk/hists.c b/tools/perf/ui/gtk/hists.c
index 0f8dcfdfb10f..32cc38a5b57f 100644
--- a/tools/perf/ui/gtk/hists.c
+++ b/tools/perf/ui/gtk/hists.c
@@ -306,7 +306,7 @@ static void perf_gtk__show_hists(GtkWidget *window, struct hists *hists,
 
 	nr_cols = 0;
 
-	perf_hpp__for_each_format(fmt)
+	hists__for_each_format(hists, fmt)
 		col_types[nr_cols++] = G_TYPE_STRING;
 
 	store = gtk_tree_store_newv(nr_cols, col_types);
@@ -317,7 +317,7 @@ static void perf_gtk__show_hists(GtkWidget *window, struct hists *hists,
 
 	col_idx = 0;
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(hists, fmt) {
 		if (perf_hpp__should_skip(fmt, hists))
 			continue;
 
@@ -367,7 +367,7 @@ static void perf_gtk__show_hists(GtkWidget *window, struct hists *hists,
 
 		col_idx = 0;
 
-		perf_hpp__for_each_format(fmt) {
+		hists__for_each_format(hists, fmt) {
 			if (perf_hpp__should_skip(fmt, h->hists))
 				continue;
 
diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c
index bf2a66e254ea..1ba4117d9c2d 100644
--- a/tools/perf/ui/hist.c
+++ b/tools/perf/ui/hist.c
@@ -371,7 +371,20 @@ static int64_t hpp__nop_cmp(struct perf_hpp_fmt *fmt __maybe_unused,
 	return 0;
 }
 
-#define HPP__COLOR_PRINT_FNS(_name, _fn)		\
+static bool perf_hpp__is_hpp_entry(struct perf_hpp_fmt *a)
+{
+	return a->header == hpp__header_fn;
+}
+
+static bool hpp__equal(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b)
+{
+	if (!perf_hpp__is_hpp_entry(a) || !perf_hpp__is_hpp_entry(b))
+		return false;
+
+	return a->idx == b->idx;
+}
+
+#define HPP__COLOR_PRINT_FNS(_name, _fn, _idx)		\
 	{						\
 		.name   = _name,			\
 		.header	= hpp__header_fn,		\
@@ -381,9 +394,11 @@ static int64_t hpp__nop_cmp(struct perf_hpp_fmt *fmt __maybe_unused,
 		.cmp	= hpp__nop_cmp,			\
 		.collapse = hpp__nop_cmp,		\
 		.sort	= hpp__sort_ ## _fn,		\
+		.idx	= PERF_HPP__ ## _idx,		\
+		.equal	= hpp__equal,			\
 	}
 
-#define HPP__COLOR_ACC_PRINT_FNS(_name, _fn)		\
+#define HPP__COLOR_ACC_PRINT_FNS(_name, _fn, _idx)	\
 	{						\
 		.name   = _name,			\
 		.header	= hpp__header_fn,		\
@@ -393,9 +408,11 @@ static int64_t hpp__nop_cmp(struct perf_hpp_fmt *fmt __maybe_unused,
 		.cmp	= hpp__nop_cmp,			\
 		.collapse = hpp__nop_cmp,		\
 		.sort	= hpp__sort_ ## _fn,		\
+		.idx	= PERF_HPP__ ## _idx,		\
+		.equal	= hpp__equal,			\
 	}
 
-#define HPP__PRINT_FNS(_name, _fn)			\
+#define HPP__PRINT_FNS(_name, _fn, _idx)		\
 	{						\
 		.name   = _name,			\
 		.header	= hpp__header_fn,		\
@@ -404,22 +421,25 @@ static int64_t hpp__nop_cmp(struct perf_hpp_fmt *fmt __maybe_unused,
 		.cmp	= hpp__nop_cmp,			\
 		.collapse = hpp__nop_cmp,		\
 		.sort	= hpp__sort_ ## _fn,		\
+		.idx	= PERF_HPP__ ## _idx,		\
+		.equal	= hpp__equal,			\
 	}
 
 struct perf_hpp_fmt perf_hpp__format[] = {
-	HPP__COLOR_PRINT_FNS("Overhead", overhead),
-	HPP__COLOR_PRINT_FNS("sys", overhead_sys),
-	HPP__COLOR_PRINT_FNS("usr", overhead_us),
-	HPP__COLOR_PRINT_FNS("guest sys", overhead_guest_sys),
-	HPP__COLOR_PRINT_FNS("guest usr", overhead_guest_us),
-	HPP__COLOR_ACC_PRINT_FNS("Children", overhead_acc),
-	HPP__PRINT_FNS("Samples", samples),
-	HPP__PRINT_FNS("Period", period)
+	HPP__COLOR_PRINT_FNS("Overhead", overhead, OVERHEAD),
+	HPP__COLOR_PRINT_FNS("sys", overhead_sys, OVERHEAD_SYS),
+	HPP__COLOR_PRINT_FNS("usr", overhead_us, OVERHEAD_US),
+	HPP__COLOR_PRINT_FNS("guest sys", overhead_guest_sys, OVERHEAD_GUEST_SYS),
+	HPP__COLOR_PRINT_FNS("guest usr", overhead_guest_us, OVERHEAD_GUEST_US),
+	HPP__COLOR_ACC_PRINT_FNS("Children", overhead_acc, OVERHEAD_ACC),
+	HPP__PRINT_FNS("Samples", samples, SAMPLES),
+	HPP__PRINT_FNS("Period", period, PERIOD)
 };
 
-LIST_HEAD(perf_hpp__list);
-LIST_HEAD(perf_hpp__sort_list);
-
+struct perf_hpp_list perf_hpp_list = {
+	.fields	= LIST_HEAD_INIT(perf_hpp_list.fields),
+	.sorts	= LIST_HEAD_INIT(perf_hpp_list.sorts),
+};
 
 #undef HPP__COLOR_PRINT_FNS
 #undef HPP__COLOR_ACC_PRINT_FNS
@@ -485,63 +505,60 @@ void perf_hpp__init(void)
 		hpp_dimension__add_output(PERF_HPP__PERIOD);
 }
 
-void perf_hpp__column_register(struct perf_hpp_fmt *format)
-{
-	list_add_tail(&format->list, &perf_hpp__list);
-}
-
-void perf_hpp__column_unregister(struct perf_hpp_fmt *format)
-{
-	list_del(&format->list);
-}
-
-void perf_hpp__register_sort_field(struct perf_hpp_fmt *format)
+void perf_hpp_list__column_register(struct perf_hpp_list *list,
+				    struct perf_hpp_fmt *format)
 {
-	list_add_tail(&format->sort_list, &perf_hpp__sort_list);
+	list_add_tail(&format->list, &list->fields);
 }
 
-void perf_hpp__column_enable(unsigned col)
+void perf_hpp_list__register_sort_field(struct perf_hpp_list *list,
+					struct perf_hpp_fmt *format)
 {
-	BUG_ON(col >= PERF_HPP__MAX_INDEX);
-	perf_hpp__column_register(&perf_hpp__format[col]);
+	list_add_tail(&format->sort_list, &list->sorts);
 }
 
-void perf_hpp__column_disable(unsigned col)
+void perf_hpp__column_unregister(struct perf_hpp_fmt *format)
 {
-	BUG_ON(col >= PERF_HPP__MAX_INDEX);
-	perf_hpp__column_unregister(&perf_hpp__format[col]);
+	list_del(&format->list);
 }
 
 void perf_hpp__cancel_cumulate(void)
 {
+	struct perf_hpp_fmt *fmt, *acc, *ovh, *tmp;
+
 	if (is_strict_order(field_order))
 		return;
 
-	perf_hpp__column_disable(PERF_HPP__OVERHEAD_ACC);
-	perf_hpp__format[PERF_HPP__OVERHEAD].name = "Overhead";
+	ovh = &perf_hpp__format[PERF_HPP__OVERHEAD];
+	acc = &perf_hpp__format[PERF_HPP__OVERHEAD_ACC];
+
+	perf_hpp_list__for_each_format_safe(&perf_hpp_list, fmt, tmp) {
+		if (acc->equal(acc, fmt)) {
+			perf_hpp__column_unregister(fmt);
+			continue;
+		}
+
+		if (ovh->equal(ovh, fmt))
+			fmt->name = "Overhead";
+	}
+}
+
+static bool fmt_equal(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b)
+{
+	return a->equal && a->equal(a, b);
 }
 
-void perf_hpp__setup_output_field(void)
+void perf_hpp__setup_output_field(struct perf_hpp_list *list)
 {
 	struct perf_hpp_fmt *fmt;
 
 	/* append sort keys to output field */
-	perf_hpp__for_each_sort_list(fmt) {
-		if (!list_empty(&fmt->list))
-			continue;
-
-		/*
-		 * sort entry fields are dynamically created,
-		 * so they can share a same sort key even though
-		 * the list is empty.
-		 */
-		if (perf_hpp__is_sort_entry(fmt)) {
-			struct perf_hpp_fmt *pos;
+	perf_hpp_list__for_each_sort_list(list, fmt) {
+		struct perf_hpp_fmt *pos;
 
-			perf_hpp__for_each_format(pos) {
-				if (perf_hpp__same_sort_entry(pos, fmt))
-					goto next;
-			}
+		perf_hpp_list__for_each_format(list, pos) {
+			if (fmt_equal(fmt, pos))
+				goto next;
 		}
 
 		perf_hpp__column_register(fmt);
@@ -550,27 +567,17 @@ next:
 	}
 }
 
-void perf_hpp__append_sort_keys(void)
+void perf_hpp__append_sort_keys(struct perf_hpp_list *list)
 {
 	struct perf_hpp_fmt *fmt;
 
 	/* append output fields to sort keys */
-	perf_hpp__for_each_format(fmt) {
-		if (!list_empty(&fmt->sort_list))
-			continue;
+	perf_hpp_list__for_each_format(list, fmt) {
+		struct perf_hpp_fmt *pos;
 
-		/*
-		 * sort entry fields are dynamically created,
-		 * so they can share a same sort key even though
-		 * the list is empty.
-		 */
-		if (perf_hpp__is_sort_entry(fmt)) {
-			struct perf_hpp_fmt *pos;
-
-			perf_hpp__for_each_sort_list(pos) {
-				if (perf_hpp__same_sort_entry(pos, fmt))
-					goto next;
-			}
+		perf_hpp_list__for_each_sort_list(list, pos) {
+			if (fmt_equal(fmt, pos))
+				goto next;
 		}
 
 		perf_hpp__register_sort_field(fmt);
@@ -579,20 +586,29 @@ next:
 	}
 }
 
-void perf_hpp__reset_output_field(void)
+
+static void fmt_free(struct perf_hpp_fmt *fmt)
+{
+	if (fmt->free)
+		fmt->free(fmt);
+}
+
+void perf_hpp__reset_output_field(struct perf_hpp_list *list)
 {
 	struct perf_hpp_fmt *fmt, *tmp;
 
 	/* reset output fields */
-	perf_hpp__for_each_format_safe(fmt, tmp) {
+	perf_hpp_list__for_each_format_safe(list, fmt, tmp) {
 		list_del_init(&fmt->list);
 		list_del_init(&fmt->sort_list);
+		fmt_free(fmt);
 	}
 
 	/* reset sort keys */
-	perf_hpp__for_each_sort_list_safe(fmt, tmp) {
+	perf_hpp_list__for_each_sort_list_safe(list, fmt, tmp) {
 		list_del_init(&fmt->list);
 		list_del_init(&fmt->sort_list);
+		fmt_free(fmt);
 	}
 }
 
@@ -606,7 +622,7 @@ unsigned int hists__sort_list_width(struct hists *hists)
 	bool first = true;
 	struct perf_hpp dummy_hpp;
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(hists, fmt) {
 		if (perf_hpp__should_skip(fmt, hists))
 			continue;
 
@@ -626,20 +642,12 @@ unsigned int hists__sort_list_width(struct hists *hists)
 
 void perf_hpp__reset_width(struct perf_hpp_fmt *fmt, struct hists *hists)
 {
-	int idx;
-
 	if (perf_hpp__is_sort_entry(fmt))
 		return perf_hpp__reset_sort_width(fmt, hists);
 
-	for (idx = 0; idx < PERF_HPP__MAX_INDEX; idx++) {
-		if (fmt == &perf_hpp__format[idx])
-			break;
-	}
-
-	if (idx == PERF_HPP__MAX_INDEX)
-		return;
+	BUG_ON(fmt->idx >= PERF_HPP__MAX_INDEX);
 
-	switch (idx) {
+	switch (fmt->idx) {
 	case PERF_HPP__OVERHEAD:
 	case PERF_HPP__OVERHEAD_SYS:
 	case PERF_HPP__OVERHEAD_US:
@@ -667,7 +675,7 @@ void perf_hpp__set_user_width(const char *width_list_str)
 	struct perf_hpp_fmt *fmt;
 	const char *ptr = width_list_str;
 
-	perf_hpp__for_each_format(fmt) {
+	perf_hpp_list__for_each_format(&perf_hpp_list, fmt) {
 		char *p;
 
 		int len = strtol(ptr, &p, 10);
diff --git a/tools/perf/ui/stdio/hist.c b/tools/perf/ui/stdio/hist.c
index 691e52ce7510..1a6e8f7f38c4 100644
--- a/tools/perf/ui/stdio/hist.c
+++ b/tools/perf/ui/stdio/hist.c
@@ -384,7 +384,7 @@ static int hist_entry__snprintf(struct hist_entry *he, struct perf_hpp *hpp)
 	if (symbol_conf.exclude_other && !he->parent)
 		return 0;
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(he->hists, fmt) {
 		if (perf_hpp__should_skip(fmt, he->hists))
 			continue;
 
@@ -453,7 +453,7 @@ size_t hists__fprintf(struct hists *hists, bool show_header, int max_rows,
 
 	init_rem_hits();
 
-	perf_hpp__for_each_format(fmt)
+	hists__for_each_format(hists, fmt)
 		perf_hpp__reset_width(fmt, hists);
 
 	if (symbol_conf.col_width_list_str)
@@ -464,7 +464,7 @@ size_t hists__fprintf(struct hists *hists, bool show_header, int max_rows,
 
 	fprintf(fp, "# ");
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(hists, fmt) {
 		if (perf_hpp__should_skip(fmt, hists))
 			continue;
 
@@ -488,7 +488,7 @@ size_t hists__fprintf(struct hists *hists, bool show_header, int max_rows,
 
 	fprintf(fp, "# ");
 
-	perf_hpp__for_each_format(fmt) {
+	hists__for_each_format(hists, fmt) {
 		unsigned int i;
 
 		if (perf_hpp__should_skip(fmt, hists))
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index 098310bc4489..12f2d794dc28 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -958,10 +958,11 @@ out:
 int64_t
 hist_entry__cmp(struct hist_entry *left, struct hist_entry *right)
 {
+	struct hists *hists = left->hists;
 	struct perf_hpp_fmt *fmt;
 	int64_t cmp = 0;
 
-	perf_hpp__for_each_sort_list(fmt) {
+	hists__for_each_sort_list(hists, fmt) {
 		cmp = fmt->cmp(fmt, left, right);
 		if (cmp)
 			break;
@@ -973,10 +974,11 @@ hist_entry__cmp(struct hist_entry *left, struct hist_entry *right)
 int64_t
 hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
 {
+	struct hists *hists = left->hists;
 	struct perf_hpp_fmt *fmt;
 	int64_t cmp = 0;
 
-	perf_hpp__for_each_sort_list(fmt) {
+	hists__for_each_sort_list(hists, fmt) {
 		cmp = fmt->collapse(fmt, left, right);
 		if (cmp)
 			break;
@@ -1117,10 +1119,11 @@ void hists__collapse_resort(struct hists *hists, struct ui_progress *prog)
 
 static int hist_entry__sort(struct hist_entry *a, struct hist_entry *b)
 {
+	struct hists *hists = a->hists;
 	struct perf_hpp_fmt *fmt;
 	int64_t cmp = 0;
 
-	perf_hpp__for_each_sort_list(fmt) {
+	hists__for_each_sort_list(hists, fmt) {
 		if (perf_hpp__should_skip(fmt, a->hists))
 			continue;
 
@@ -1197,19 +1200,13 @@ static void __hists__insert_output_entry(struct rb_root *entries,
 	rb_insert_color(&he->rb_node, entries);
 }
 
-void hists__output_resort(struct hists *hists, struct ui_progress *prog)
+static void output_resort(struct hists *hists, struct ui_progress *prog,
+			  bool use_callchain)
 {
 	struct rb_root *root;
 	struct rb_node *next;
 	struct hist_entry *n;
 	u64 min_callchain_hits;
-	struct perf_evsel *evsel = hists_to_evsel(hists);
-	bool use_callchain;
-
-	if (evsel && symbol_conf.use_callchain && !symbol_conf.show_ref_callgraph)
-		use_callchain = evsel->attr.sample_type & PERF_SAMPLE_CALLCHAIN;
-	else
-		use_callchain = symbol_conf.use_callchain;
 
 	min_callchain_hits = hists__total_period(hists) * (callchain_param.min_percent / 100);
 
@@ -1239,6 +1236,23 @@ void hists__output_resort(struct hists *hists, struct ui_progress *prog)
 	}
 }
 
+void perf_evsel__output_resort(struct perf_evsel *evsel, struct ui_progress *prog)
+{
+	bool use_callchain;
+
+	if (evsel && symbol_conf.use_callchain && !symbol_conf.show_ref_callgraph)
+		use_callchain = evsel->attr.sample_type & PERF_SAMPLE_CALLCHAIN;
+	else
+		use_callchain = symbol_conf.use_callchain;
+
+	output_resort(evsel__hists(evsel), prog, use_callchain);
+}
+
+void hists__output_resort(struct hists *hists, struct ui_progress *prog)
+{
+	output_resort(hists, prog, symbol_conf.use_callchain);
+}
+
 static void hists__remove_entry_filter(struct hists *hists, struct hist_entry *h,
 				       enum hist_filter filter)
 {
@@ -1567,7 +1581,7 @@ int perf_hist_config(const char *var, const char *value)
 	return 0;
 }
 
-int __hists__init(struct hists *hists)
+int __hists__init(struct hists *hists, struct perf_hpp_list *hpp_list)
 {
 	memset(hists, 0, sizeof(*hists));
 	hists->entries_in_array[0] = hists->entries_in_array[1] = RB_ROOT;
@@ -1576,6 +1590,7 @@ int __hists__init(struct hists *hists)
 	hists->entries = RB_ROOT;
 	pthread_mutex_init(&hists->lock, NULL);
 	hists->socket_filter = -1;
+	hists->hpp_list = hpp_list;
 	return 0;
 }
 
@@ -1612,7 +1627,7 @@ static int hists_evsel__init(struct perf_evsel *evsel)
 {
 	struct hists *hists = evsel__hists(evsel);
 
-	__hists__init(hists);
+	__hists__init(hists, &perf_hpp_list);
 	return 0;
 }
 
@@ -1631,3 +1646,9 @@ int hists__init(void)
 
 	return err;
 }
+
+void perf_hpp_list__init(struct perf_hpp_list *list)
+{
+	INIT_LIST_HEAD(&list->fields);
+	INIT_LIST_HEAD(&list->sorts);
+}
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index d4ec4822a103..1c7544a8fe1a 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -75,6 +75,7 @@ struct hists {
 	u64			event_stream;
 	u16			col_len[HISTC_NR_COLS];
 	int			socket_filter;
+	struct perf_hpp_list	*hpp_list;
 };
 
 struct hist_entry_iter;
@@ -128,6 +129,7 @@ int hist_entry__sort_snprintf(struct hist_entry *he, char *bf, size_t size,
 			      struct hists *hists);
 void hist_entry__delete(struct hist_entry *he);
 
+void perf_evsel__output_resort(struct perf_evsel *evsel, struct ui_progress *prog);
 void hists__output_resort(struct hists *hists, struct ui_progress *prog);
 void hists__collapse_resort(struct hists *hists, struct ui_progress *prog);
 
@@ -185,7 +187,7 @@ static inline struct hists *evsel__hists(struct perf_evsel *evsel)
 }
 
 int hists__init(void);
-int __hists__init(struct hists *hists);
+int __hists__init(struct hists *hists, struct perf_hpp_list *hpp_list);
 
 struct rb_root *hists__get_rotate_entries_in(struct hists *hists);
 bool hists__collapse_insert_entry(struct hists *hists __maybe_unused,
@@ -214,28 +216,56 @@ struct perf_hpp_fmt {
 			    struct hist_entry *a, struct hist_entry *b);
 	int64_t (*sort)(struct perf_hpp_fmt *fmt,
 			struct hist_entry *a, struct hist_entry *b);
+	bool (*equal)(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b);
+	void (*free)(struct perf_hpp_fmt *fmt);
 
 	struct list_head list;
 	struct list_head sort_list;
 	bool elide;
 	int len;
 	int user_len;
+	int idx;
 };
 
-extern struct list_head perf_hpp__list;
-extern struct list_head perf_hpp__sort_list;
+struct perf_hpp_list {
+	struct list_head fields;
+	struct list_head sorts;
+};
+
+extern struct perf_hpp_list perf_hpp_list;
+
+void perf_hpp_list__column_register(struct perf_hpp_list *list,
+				    struct perf_hpp_fmt *format);
+void perf_hpp_list__register_sort_field(struct perf_hpp_list *list,
+					struct perf_hpp_fmt *format);
+
+static inline void perf_hpp__column_register(struct perf_hpp_fmt *format)
+{
+	perf_hpp_list__column_register(&perf_hpp_list, format);
+}
 
-#define perf_hpp__for_each_format(format) \
-	list_for_each_entry(format, &perf_hpp__list, list)
+static inline void perf_hpp__register_sort_field(struct perf_hpp_fmt *format)
+{
+	perf_hpp_list__register_sort_field(&perf_hpp_list, format);
+}
+
+#define perf_hpp_list__for_each_format(_list, format) \
+	list_for_each_entry(format, &(_list)->fields, list)
 
-#define perf_hpp__for_each_format_safe(format, tmp)	\
-	list_for_each_entry_safe(format, tmp, &perf_hpp__list, list)
+#define perf_hpp_list__for_each_format_safe(_list, format, tmp)	\
+	list_for_each_entry_safe(format, tmp, &(_list)->fields, list)
 
-#define perf_hpp__for_each_sort_list(format) \
-	list_for_each_entry(format, &perf_hpp__sort_list, sort_list)
+#define perf_hpp_list__for_each_sort_list(_list, format) \
+	list_for_each_entry(format, &(_list)->sorts, sort_list)
 
-#define perf_hpp__for_each_sort_list_safe(format, tmp)	\
-	list_for_each_entry_safe(format, tmp, &perf_hpp__sort_list, sort_list)
+#define perf_hpp_list__for_each_sort_list_safe(_list, format, tmp)	\
+	list_for_each_entry_safe(format, tmp, &(_list)->sorts, sort_list)
+
+#define hists__for_each_format(hists, format) \
+	perf_hpp_list__for_each_format((hists)->hpp_list, fmt)
+
+#define hists__for_each_sort_list(hists, format) \
+	perf_hpp_list__for_each_sort_list((hists)->hpp_list, fmt)
 
 extern struct perf_hpp_fmt perf_hpp__format[];
 
@@ -254,19 +284,14 @@ enum {
 };
 
 void perf_hpp__init(void);
-void perf_hpp__column_register(struct perf_hpp_fmt *format);
 void perf_hpp__column_unregister(struct perf_hpp_fmt *format);
-void perf_hpp__column_enable(unsigned col);
-void perf_hpp__column_disable(unsigned col);
 void perf_hpp__cancel_cumulate(void);
+void perf_hpp__setup_output_field(struct perf_hpp_list *list);
+void perf_hpp__reset_output_field(struct perf_hpp_list *list);
+void perf_hpp__append_sort_keys(struct perf_hpp_list *list);
 
-void perf_hpp__register_sort_field(struct perf_hpp_fmt *format);
-void perf_hpp__setup_output_field(void);
-void perf_hpp__reset_output_field(void);
-void perf_hpp__append_sort_keys(void);
 
 bool perf_hpp__is_sort_entry(struct perf_hpp_fmt *format);
-bool perf_hpp__same_sort_entry(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b);
 bool perf_hpp__is_dynamic_entry(struct perf_hpp_fmt *format);
 bool perf_hpp__defined_dynamic_entry(struct perf_hpp_fmt *fmt, struct hists *hists);
 
@@ -381,4 +406,6 @@ int parse_filter_percentage(const struct option *opt __maybe_unused,
 			    const char *arg, int unset __maybe_unused);
 int perf_hist_config(const char *var, const char *value);
 
+void perf_hpp_list__init(struct perf_hpp_list *list);
+
 #endif	/* __PERF_HIST_H */
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
index 898e4b0724bf..de620f7f40f4 100644
--- a/tools/perf/util/sort.c
+++ b/tools/perf/util/sort.c
@@ -1441,20 +1441,6 @@ struct hpp_sort_entry {
 	struct sort_entry *se;
 };
 
-bool perf_hpp__same_sort_entry(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b)
-{
-	struct hpp_sort_entry *hse_a;
-	struct hpp_sort_entry *hse_b;
-
-	if (!perf_hpp__is_sort_entry(a) || !perf_hpp__is_sort_entry(b))
-		return false;
-
-	hse_a = container_of(a, struct hpp_sort_entry, hpp);
-	hse_b = container_of(b, struct hpp_sort_entry, hpp);
-
-	return hse_a->se == hse_b->se;
-}
-
 void perf_hpp__reset_sort_width(struct perf_hpp_fmt *fmt, struct hists *hists)
 {
 	struct hpp_sort_entry *hse;
@@ -1540,6 +1526,33 @@ static int64_t __sort__hpp_sort(struct perf_hpp_fmt *fmt,
 	return sort_fn(a, b);
 }
 
+bool perf_hpp__is_sort_entry(struct perf_hpp_fmt *format)
+{
+	return format->header == __sort__hpp_header;
+}
+
+static bool __sort__hpp_equal(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b)
+{
+	struct hpp_sort_entry *hse_a;
+	struct hpp_sort_entry *hse_b;
+
+	if (!perf_hpp__is_sort_entry(a) || !perf_hpp__is_sort_entry(b))
+		return false;
+
+	hse_a = container_of(a, struct hpp_sort_entry, hpp);
+	hse_b = container_of(b, struct hpp_sort_entry, hpp);
+
+	return hse_a->se == hse_b->se;
+}
+
+static void hse_free(struct perf_hpp_fmt *fmt)
+{
+	struct hpp_sort_entry *hse;
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+	free(hse);
+}
+
 static struct hpp_sort_entry *
 __sort_dimension__alloc_hpp(struct sort_dimension *sd)
 {
@@ -1561,6 +1574,8 @@ __sort_dimension__alloc_hpp(struct sort_dimension *sd)
 	hse->hpp.cmp = __sort__hpp_cmp;
 	hse->hpp.collapse = __sort__hpp_collapse;
 	hse->hpp.sort = __sort__hpp_sort;
+	hse->hpp.equal = __sort__hpp_equal;
+	hse->hpp.free = hse_free;
 
 	INIT_LIST_HEAD(&hse->hpp.list);
 	INIT_LIST_HEAD(&hse->hpp.sort_list);
@@ -1571,9 +1586,23 @@ __sort_dimension__alloc_hpp(struct sort_dimension *sd)
 	return hse;
 }
 
-bool perf_hpp__is_sort_entry(struct perf_hpp_fmt *format)
+static void hpp_free(struct perf_hpp_fmt *fmt)
 {
-	return format->header == __sort__hpp_header;
+	free(fmt);
+}
+
+static struct perf_hpp_fmt *__hpp_dimension__alloc_hpp(struct hpp_dimension *hd)
+{
+	struct perf_hpp_fmt *fmt;
+
+	fmt = memdup(hd->fmt, sizeof(*fmt));
+	if (fmt) {
+		INIT_LIST_HEAD(&fmt->list);
+		INIT_LIST_HEAD(&fmt->sort_list);
+		fmt->free = hpp_free;
+	}
+
+	return fmt;
 }
 
 static int __sort_dimension__add_hpp_sort(struct sort_dimension *sd)
@@ -1587,14 +1616,15 @@ static int __sort_dimension__add_hpp_sort(struct sort_dimension *sd)
 	return 0;
 }
 
-static int __sort_dimension__add_hpp_output(struct sort_dimension *sd)
+static int __sort_dimension__add_hpp_output(struct perf_hpp_list *list,
+					    struct sort_dimension *sd)
 {
 	struct hpp_sort_entry *hse = __sort_dimension__alloc_hpp(sd);
 
 	if (hse == NULL)
 		return -1;
 
-	perf_hpp__column_register(&hse->hpp);
+	perf_hpp_list__column_register(list, &hse->hpp);
 	return 0;
 }
 
@@ -1804,6 +1834,14 @@ bool perf_hpp__is_dynamic_entry(struct perf_hpp_fmt *fmt)
 	return fmt->cmp == __sort__hde_cmp;
 }
 
+static void hde_free(struct perf_hpp_fmt *fmt)
+{
+	struct hpp_dynamic_entry *hde;
+
+	hde = container_of(fmt, struct hpp_dynamic_entry, hpp);
+	free(hde);
+}
+
 static struct hpp_dynamic_entry *
 __alloc_dynamic_entry(struct perf_evsel *evsel, struct format_field *field)
 {
@@ -1828,6 +1866,7 @@ __alloc_dynamic_entry(struct perf_evsel *evsel, struct format_field *field)
 	hde->hpp.cmp = __sort__hde_cmp;
 	hde->hpp.collapse = __sort__hde_cmp;
 	hde->hpp.sort = __sort__hde_cmp;
+	hde->hpp.free = hde_free;
 
 	INIT_LIST_HEAD(&hde->hpp.list);
 	INIT_LIST_HEAD(&hde->hpp.sort_list);
@@ -2065,40 +2104,54 @@ static int __sort_dimension__add(struct sort_dimension *sd)
 
 static int __hpp_dimension__add(struct hpp_dimension *hd)
 {
-	if (!hd->taken) {
-		hd->taken = 1;
+	struct perf_hpp_fmt *fmt;
 
-		perf_hpp__register_sort_field(hd->fmt);
-	}
+	if (hd->taken)
+		return 0;
+
+	fmt = __hpp_dimension__alloc_hpp(hd);
+	if (!fmt)
+		return -1;
+
+	hd->taken = 1;
+	perf_hpp__register_sort_field(fmt);
 	return 0;
 }
 
-static int __sort_dimension__add_output(struct sort_dimension *sd)
+static int __sort_dimension__add_output(struct perf_hpp_list *list,
+					struct sort_dimension *sd)
 {
 	if (sd->taken)
 		return 0;
 
-	if (__sort_dimension__add_hpp_output(sd) < 0)
+	if (__sort_dimension__add_hpp_output(list, sd) < 0)
 		return -1;
 
 	sd->taken = 1;
 	return 0;
 }
 
-static int __hpp_dimension__add_output(struct hpp_dimension *hd)
+static int __hpp_dimension__add_output(struct perf_hpp_list *list,
+				       struct hpp_dimension *hd)
 {
-	if (!hd->taken) {
-		hd->taken = 1;
+	struct perf_hpp_fmt *fmt;
 
-		perf_hpp__column_register(hd->fmt);
-	}
+	if (hd->taken)
+		return 0;
+
+	fmt = __hpp_dimension__alloc_hpp(hd);
+	if (!fmt)
+		return -1;
+
+	hd->taken = 1;
+	perf_hpp_list__column_register(list, fmt);
 	return 0;
 }
 
 int hpp_dimension__add_output(unsigned col)
 {
 	BUG_ON(col >= PERF_HPP__MAX_INDEX);
-	return __hpp_dimension__add_output(&hpp_sort_dimensions[col]);
+	return __hpp_dimension__add_output(&perf_hpp_list, &hpp_sort_dimensions[col]);
 }
 
 static int sort_dimension__add(const char *tok,
@@ -2191,6 +2244,26 @@ static int sort_dimension__add(const char *tok,
 	return -ESRCH;
 }
 
+static int setup_sort_list(char *str, struct perf_evlist *evlist)
+{
+	char *tmp, *tok;
+	int ret = 0;
+
+	for (tok = strtok_r(str, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		ret = sort_dimension__add(tok, evlist);
+		if (ret == -EINVAL) {
+			error("Invalid --sort key: `%s'", tok);
+			break;
+		} else if (ret == -ESRCH) {
+			error("Unknown --sort key: `%s'", tok);
+			break;
+		}
+	}
+
+	return ret;
+}
+
 static const char *get_default_sort_order(struct perf_evlist *evlist)
 {
 	const char *default_sort_orders[] = {
@@ -2285,7 +2358,7 @@ static char *setup_overhead(char *keys)
 
 static int __setup_sorting(struct perf_evlist *evlist)
 {
-	char *tmp, *tok, *str;
+	char *str;
 	const char *sort_keys;
 	int ret = 0;
 
@@ -2323,17 +2396,7 @@ static int __setup_sorting(struct perf_evlist *evlist)
 		}
 	}
 
-	for (tok = strtok_r(str, ", ", &tmp);
-			tok; tok = strtok_r(NULL, ", ", &tmp)) {
-		ret = sort_dimension__add(tok, evlist);
-		if (ret == -EINVAL) {
-			error("Invalid --sort key: `%s'", tok);
-			break;
-		} else if (ret == -ESRCH) {
-			error("Unknown --sort key: `%s'", tok);
-			break;
-		}
-	}
+	ret = setup_sort_list(str, evlist);
 
 	free(str);
 	return ret;
@@ -2344,7 +2407,7 @@ void perf_hpp__set_elide(int idx, bool elide)
 	struct perf_hpp_fmt *fmt;
 	struct hpp_sort_entry *hse;
 
-	perf_hpp__for_each_format(fmt) {
+	perf_hpp_list__for_each_format(&perf_hpp_list, fmt) {
 		if (!perf_hpp__is_sort_entry(fmt))
 			continue;
 
@@ -2404,7 +2467,7 @@ void sort__setup_elide(FILE *output)
 	struct perf_hpp_fmt *fmt;
 	struct hpp_sort_entry *hse;
 
-	perf_hpp__for_each_format(fmt) {
+	perf_hpp_list__for_each_format(&perf_hpp_list, fmt) {
 		if (!perf_hpp__is_sort_entry(fmt))
 			continue;
 
@@ -2416,7 +2479,7 @@ void sort__setup_elide(FILE *output)
 	 * It makes no sense to elide all of sort entries.
 	 * Just revert them to show up again.
 	 */
-	perf_hpp__for_each_format(fmt) {
+	perf_hpp_list__for_each_format(&perf_hpp_list, fmt) {
 		if (!perf_hpp__is_sort_entry(fmt))
 			continue;
 
@@ -2424,7 +2487,7 @@ void sort__setup_elide(FILE *output)
 			return;
 	}
 
-	perf_hpp__for_each_format(fmt) {
+	perf_hpp_list__for_each_format(&perf_hpp_list, fmt) {
 		if (!perf_hpp__is_sort_entry(fmt))
 			continue;
 
@@ -2432,7 +2495,7 @@ void sort__setup_elide(FILE *output)
 	}
 }
 
-static int output_field_add(char *tok)
+static int output_field_add(struct perf_hpp_list *list, char *tok)
 {
 	unsigned int i;
 
@@ -2442,7 +2505,7 @@ static int output_field_add(char *tok)
 		if (strncasecmp(tok, sd->name, strlen(tok)))
 			continue;
 
-		return __sort_dimension__add_output(sd);
+		return __sort_dimension__add_output(list, sd);
 	}
 
 	for (i = 0; i < ARRAY_SIZE(hpp_sort_dimensions); i++) {
@@ -2451,7 +2514,7 @@ static int output_field_add(char *tok)
 		if (strncasecmp(tok, hd->name, strlen(tok)))
 			continue;
 
-		return __hpp_dimension__add_output(hd);
+		return __hpp_dimension__add_output(list, hd);
 	}
 
 	for (i = 0; i < ARRAY_SIZE(bstack_sort_dimensions); i++) {
@@ -2460,7 +2523,7 @@ static int output_field_add(char *tok)
 		if (strncasecmp(tok, sd->name, strlen(tok)))
 			continue;
 
-		return __sort_dimension__add_output(sd);
+		return __sort_dimension__add_output(list, sd);
 	}
 
 	for (i = 0; i < ARRAY_SIZE(memory_sort_dimensions); i++) {
@@ -2469,12 +2532,32 @@ static int output_field_add(char *tok)
 		if (strncasecmp(tok, sd->name, strlen(tok)))
 			continue;
 
-		return __sort_dimension__add_output(sd);
+		return __sort_dimension__add_output(list, sd);
 	}
 
 	return -ESRCH;
 }
 
+static int setup_output_list(struct perf_hpp_list *list, char *str)
+{
+	char *tmp, *tok;
+	int ret = 0;
+
+	for (tok = strtok_r(str, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		ret = output_field_add(list, tok);
+		if (ret == -EINVAL) {
+			error("Invalid --fields key: `%s'", tok);
+			break;
+		} else if (ret == -ESRCH) {
+			error("Unknown --fields key: `%s'", tok);
+			break;
+		}
+	}
+
+	return ret;
+}
+
 static void reset_dimensions(void)
 {
 	unsigned int i;
@@ -2499,7 +2582,7 @@ bool is_strict_order(const char *order)
 
 static int __setup_output_field(void)
 {
-	char *tmp, *tok, *str, *strp;
+	char *str, *strp;
 	int ret = -EINVAL;
 
 	if (field_order == NULL)
@@ -2519,17 +2602,7 @@ static int __setup_output_field(void)
 		goto out;
 	}
 
-	for (tok = strtok_r(strp, ", ", &tmp);
-			tok; tok = strtok_r(NULL, ", ", &tmp)) {
-		ret = output_field_add(tok);
-		if (ret == -EINVAL) {
-			error("Invalid --fields key: `%s'", tok);
-			break;
-		} else if (ret == -ESRCH) {
-			error("Unknown --fields key: `%s'", tok);
-			break;
-		}
-	}
+	ret = setup_output_list(&perf_hpp_list, strp);
 
 out:
 	free(str);
@@ -2563,9 +2636,9 @@ int setup_sorting(struct perf_evlist *evlist)
 		return err;
 
 	/* copy sort keys to output fields */
-	perf_hpp__setup_output_field();
+	perf_hpp__setup_output_field(&perf_hpp_list);
 	/* and then copy output fields to sort keys */
-	perf_hpp__append_sort_keys();
+	perf_hpp__append_sort_keys(&perf_hpp_list);
 
 	return 0;
 }
@@ -2581,5 +2654,5 @@ void reset_output_field(void)
 	sort_order = NULL;
 
 	reset_dimensions();
-	perf_hpp__reset_output_field();
+	perf_hpp__reset_output_field(&perf_hpp_list);
 }