perf scripts python: exported-sql-viewer.py: Add call tree

Add a new report to display a call tree. The Call Tree report is very similar to the Context-Sensitive Call Graph, but the data is not aggregated. Also the 'Count' column, which would be always 1, is replaced by the 'Call Time'. Committer testing: $ cat simple-retpoline.c /* https://lkml.kernel.org/r/20190109091835.5570-6-adrian.hunter@intel.com $ gcc -ggdb3 -Wall -Wextra -O2 -o simple-retpoline simple-retpoline.c $ objdump -d simple-retpoline */ __attribute__((noinline)) int bar(void) { return -1; } int foo(void) { return bar() + 1; } __attribute__((indirect_branch("thunk"))) int main() { int (*volatile fn)(void) = foo; fn(); return fn(); } $ $ perf record -o simple-retpoline.perf.data -e intel_pt/cyc/u ./simple-retpoline $ perf script -i simple-retpoline.perf.data --itrace=be -s ~acme/libexec/perf-core/scripts/python/export-to-sqlite.py simple-retpoline.db branches calls $ python ~acme/libexec/perf-core/scripts/python/exported-sql-viewer.py simple-retpoline.db And in the GUI select: "Reports" "Call Tree" Call Path | Object | Call Time (ns) | Time (ns) | Time (%) | Branch Count | Brach Count (%) | > simple-retpolin > PID:TID > _start ld-2.28.so 2193855505777 156267 100.0 10602 100.0 unknown unknown 2193855506010 2276 1.5 1 0.0 > _dl_start ld-2.28.so 2193855508286 137047 87.7 10088 95.2 > _dl_init ld-2.28.so 2193855645444 9142 5.9 326 3.1 > _start simple-retpoline 2193855654587 7457 4.8 182 1.7 > __libc_start_main <SNIP> <SNIP> > main simple-retpoline 2193855657493 32 0.5 12 6.7 > foo simple-retpoline 2193855657493 14 43.8 5 41.7 <SNIP>

Adrian Hunter committed 7 years ago

Commit ae8b8871 parent 254c0d81 file · +186 −9Browse files

Showing the changes to tools/perf/scripts/python (1 of 1 file). Show all

Unified · Split

tools/perf/scripts/python/exported-sql-viewer.py+186−9

@@ -688,6 +688,150 @@ class CallGraphModel(CallGraphModelBase): · 687 unmodified lines
688	688	ids.insert(0, query.value(1))
689	689	return ids
690	690
	691	+# Call tree data model level 2+ item base
	692	+
	693	+class CallTreeLevelTwoPlusItemBase(CallGraphLevelItemBase):
	694	+
	695	+ def __init__(self, glb, row, comm_id, thread_id, calls_id, time, branch_count, parent_item):
	696	+ super(CallTreeLevelTwoPlusItemBase, self).__init__(glb, row, parent_item)
	697	+ self.comm_id = comm_id
	698	+ self.thread_id = thread_id
	699	+ self.calls_id = calls_id
	700	+ self.branch_count = branch_count
	701	+ self.time = time
	702	+
	703	+ def Select(self):
	704	+ self.query_done = True;
	705	+ if self.calls_id == 0:
	706	+ comm_thread = " AND comm_id = " + str(self.comm_id) + " AND thread_id = " + str(self.thread_id)
	707	+ else:
	708	+ comm_thread = ""
	709	+ query = QSqlQuery(self.glb.db)
	710	+ QueryExec(query, "SELECT calls.id, name, short_name, call_time, return_time - call_time, branch_count"
	711	+ " FROM calls"
	712	+ " INNER JOIN call_paths ON calls.call_path_id = call_paths.id"
	713	+ " INNER JOIN symbols ON call_paths.symbol_id = symbols.id"
	714	+ " INNER JOIN dsos ON symbols.dso_id = dsos.id"
	715	+ " WHERE calls.parent_id = " + str(self.calls_id) + comm_thread +
	716	+ " ORDER BY call_time, calls.id")
	717	+ while query.next():
	718	+ child_item = CallTreeLevelThreeItem(self.glb, self.child_count, self.comm_id, self.thread_id, query.value(0), query.value(1), query.value(2), query.value(3), int(query.value(4)), int(query.value(5)), self)
	719	+ self.child_items.append(child_item)
	720	+ self.child_count += 1
	721	+
	722	+# Call tree data model level three item
	723	+
	724	+class CallTreeLevelThreeItem(CallTreeLevelTwoPlusItemBase):
	725	+
	726	+ def __init__(self, glb, row, comm_id, thread_id, calls_id, name, dso, count, time, branch_count, parent_item):
	727	+ super(CallTreeLevelThreeItem, self).__init__(glb, row, comm_id, thread_id, calls_id, time, branch_count, parent_item)
	728	+ dso = dsoname(dso)
	729	+ self.data = [ name, dso, str(count), str(time), PercentToOneDP(time, parent_item.time), str(branch_count), PercentToOneDP(branch_count, parent_item.branch_count) ]
	730	+ self.dbid = calls_id
	731	+
	732	+# Call tree data model level two item
	733	+
	734	+class CallTreeLevelTwoItem(CallTreeLevelTwoPlusItemBase):
	735	+
	736	+ def __init__(self, glb, row, comm_id, thread_id, pid, tid, parent_item):
	737	+ super(CallTreeLevelTwoItem, self).__init__(glb, row, comm_id, thread_id, 0, 0, 0, parent_item)
	738	+ self.data = [str(pid) + ":" + str(tid), "", "", "", "", "", ""]
	739	+ self.dbid = thread_id
	740	+
	741	+ def Select(self):
	742	+ super(CallTreeLevelTwoItem, self).Select()
	743	+ for child_item in self.child_items:
	744	+ self.time += child_item.time
	745	+ self.branch_count += child_item.branch_count
	746	+ for child_item in self.child_items:
	747	+ child_item.data[4] = PercentToOneDP(child_item.time, self.time)
	748	+ child_item.data[6] = PercentToOneDP(child_item.branch_count, self.branch_count)
	749	+
	750	+# Call tree data model level one item
	751	+
	752	+class CallTreeLevelOneItem(CallGraphLevelItemBase):
	753	+
	754	+ def __init__(self, glb, row, comm_id, comm, parent_item):
	755	+ super(CallTreeLevelOneItem, self).__init__(glb, row, parent_item)
	756	+ self.data = [comm, "", "", "", "", "", ""]
	757	+ self.dbid = comm_id
	758	+
	759	+ def Select(self):
	760	+ self.query_done = True;
	761	+ query = QSqlQuery(self.glb.db)
	762	+ QueryExec(query, "SELECT thread_id, pid, tid"
	763	+ " FROM comm_threads"
	764	+ " INNER JOIN threads ON thread_id = threads.id"
	765	+ " WHERE comm_id = " + str(self.dbid))
	766	+ while query.next():
	767	+ child_item = CallTreeLevelTwoItem(self.glb, self.child_count, self.dbid, query.value(0), query.value(1), query.value(2), self)
	768	+ self.child_items.append(child_item)
	769	+ self.child_count += 1
	770	+
	771	+# Call tree data model root item
	772	+
	773	+class CallTreeRootItem(CallGraphLevelItemBase):
	774	+
	775	+ def __init__(self, glb):
	776	+ super(CallTreeRootItem, self).__init__(glb, 0, None)
	777	+ self.dbid = 0
	778	+ self.query_done = True;
	779	+ query = QSqlQuery(glb.db)
	780	+ QueryExec(query, "SELECT id, comm FROM comms")
	781	+ while query.next():
	782	+ if not query.value(0):
	783	+ continue
	784	+ child_item = CallTreeLevelOneItem(glb, self.child_count, query.value(0), query.value(1), self)
	785	+ self.child_items.append(child_item)
	786	+ self.child_count += 1
	787	+
	788	+# Call Tree data model
	789	+
	790	+class CallTreeModel(CallGraphModelBase):
	791	+
	792	+ def __init__(self, glb, parent=None):
	793	+ super(CallTreeModel, self).__init__(glb, parent)
	794	+
	795	+ def GetRoot(self):
	796	+ return CallTreeRootItem(self.glb)
	797	+
	798	+ def columnCount(self, parent=None):
	799	+ return 7
	800	+
	801	+ def columnHeader(self, column):
	802	+ headers = ["Call Path", "Object", "Call Time", "Time (ns) ", "Time (%) ", "Branch Count ", "Branch Count (%) "]
	803	+ return headers[column]
	804	+
	805	+ def columnAlignment(self, column):
	806	+ alignment = [ Qt.AlignLeft, Qt.AlignLeft, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight ]
	807	+ return alignment[column]
	808	+
	809	+ def DoFindSelect(self, query, match):
	810	+ QueryExec(query, "SELECT calls.id, comm_id, thread_id"
	811	+ " FROM calls"
	812	+ " INNER JOIN call_paths ON calls.call_path_id = call_paths.id"
	813	+ " INNER JOIN symbols ON call_paths.symbol_id = symbols.id"
	814	+ " WHERE symbols.name" + match +
	815	+ " ORDER BY comm_id, thread_id, call_time, calls.id")
	816	+
	817	+ def FindPath(self, query):
	818	+ # Turn the query result into a list of ids that the tree view can walk
	819	+ # to open the tree at the right place.
	820	+ ids = []
	821	+ parent_id = query.value(0)
	822	+ while parent_id:
	823	+ ids.insert(0, parent_id)
	824	+ q2 = QSqlQuery(self.glb.db)
	825	+ QueryExec(q2, "SELECT parent_id"
	826	+ " FROM calls"
	827	+ " WHERE id = " + str(parent_id))
	828	+ if not q2.next():
	829	+ break
	830	+ parent_id = q2.value(0)
	831	+ ids.insert(0, query.value(2))
	832	+ ids.insert(0, query.value(1))
	833	+ return ids
	834	+
691	835	# Vertical widget layout
692	836
693	837	class VBox():
@@ -772,6 +916,29 @@ class CallGraphWindow(TreeWindowBase): · 78 unmodified lines
772	916
773	917	AddSubWindow(glb.mainwindow.mdi_area, self, "Context-Sensitive Call Graph")
774	918
	919	+# Call tree window
	920	+
	921	+class CallTreeWindow(TreeWindowBase):
	922	+
	923	+ def __init__(self, glb, parent=None):
	924	+ super(CallTreeWindow, self).__init__(parent)
	925	+
	926	+ self.model = LookupCreateModel("Call Tree", lambda x=glb: CallTreeModel(x))
	927	+
	928	+ self.view = QTreeView()
	929	+ self.view.setModel(self.model)
	930	+
	931	+ for c, w in ((0, 230), (1, 100), (2, 100), (3, 70), (4, 70), (5, 100)):
	932	+ self.view.setColumnWidth(c, w)
	933	+
	934	+ self.find_bar = FindBar(self, self)
	935	+
	936	+ self.vbox = VBox(self.view, self.find_bar.Widget())
	937	+
	938	+ self.setWidget(self.vbox.Widget())
	939	+
	940	+ AddSubWindow(glb.mainwindow.mdi_area, self, "Call Tree")
	941	+
775	942	# Child data item finder
776	943
777	944	class ChildDataItemFinder():
@@ -1890,10 +2057,10 @@ def GetEventList(db): · 1112 unmodified lines
1890	2057
1891	2058	# Is a table selectable
1892	2059
1893		−def IsSelectable(db, table):
	2060	+def IsSelectable(db, table, sql = ""):
1894	2061	query = QSqlQuery(db)
1895	2062	try:
1896		− QueryExec(query, "SELECT * FROM " + table + " LIMIT 1")
	2063	+ QueryExec(query, "SELECT * FROM " + table + " " + sql + " LIMIT 1")
1897	2064	except:
1898	2065	return False
1899	2066	return True
@@ -2302,9 +2469,10 @@ p.c2 { · 402 unmodified lines
2302	2469	</style>
2303	2470	<p class=c1><a href=#reports>1. Reports</a></p>
2304	2471	<p class=c2><a href=#callgraph>1.1 Context-Sensitive Call Graph</a></p>
2305		−<p class=c2><a href=#allbranches>1.2 All branches</a></p>
2306		−<p class=c2><a href=#selectedbranches>1.3 Selected branches</a></p>
2307		−<p class=c2><a href=#topcallsbyelapsedtime>1.4 Top calls by elapsed time</a></p>
	2472	+<p class=c2><a href=#calltree>1.2 Call Tree</a></p>
	2473	+<p class=c2><a href=#allbranches>1.3 All branches</a></p>
	2474	+<p class=c2><a href=#selectedbranches>1.4 Selected branches</a></p>
	2475	+<p class=c2><a href=#topcallsbyelapsedtime>1.5 Top calls by elapsed time</a></p>
2308	2476	<p class=c1><a href=#tables>2. Tables</a></p>
2309	2477	<h1 id=reports>1. Reports</h1>
2310	2478	<h2 id=callgraph>1.1 Context-Sensitive Call Graph</h2>
@@ -2340,7 +2508,10 @@ v- ls · 29 unmodified lines
2340	2508	<h3>Find</h3>
2341	2509	Ctrl-F displays a Find bar which finds function names by either an exact match or a pattern match.
2342	2510	The pattern matching symbols are ? for any character and * for zero or more characters.
2343		−<h2 id=allbranches>1.2 All branches</h2>
	2511	+<h2 id=calltree>1.2 Call Tree</h2>
	2512	+The Call Tree report is very similar to the Context-Sensitive Call Graph, but the data is not aggregated.
	2513	+Also the 'Count' column, which would be always 1, is replaced by the 'Call Time'.
	2514	+<h2 id=allbranches>1.3 All branches</h2>
2344	2515	The All branches report displays all branches in chronological order.
2345	2516	Not all data is fetched immediately. More records can be fetched using the Fetch bar provided.
2346	2517	<h3>Disassembly</h3>
@@ -2366,10 +2537,10 @@ sudo ldconfig · 19 unmodified lines
2366	2537	Ctrl-F displays a Find bar which finds substrings by either an exact match or a regular expression match.
2367	2538	Refer to Python documentation for the regular expression syntax.
2368	2539	All columns are searched, but only currently fetched rows are searched.
2369		−<h2 id=selectedbranches>1.3 Selected branches</h2>
	2540	+<h2 id=selectedbranches>1.4 Selected branches</h2>
2370	2541	This is the same as the <a href=#allbranches>All branches</a> report but with the data reduced
2371	2542	by various selection criteria. A dialog box displays available criteria which are AND'ed together.
2372		−<h3>1.3.1 Time ranges</h3>
	2543	+<h3>1.4.1 Time ranges</h3>
2373	2544	The time ranges hint text shows the total time range. Relative time ranges can also be entered in
2374	2545	ms, us or ns. Also, negative values are relative to the end of trace. Examples:
2375	2546	<pre>
@@ -2380,7 +2551,7 @@ ms, us or ns. Also, negative values are relative to the end of trace. Examples: · 4 unmodified lines
2380	2551	-10ms- The last 10ms
2381	2552	</pre>
2382	2553	N.B. Due to the granularity of timestamps, there could be no branches in any given time range.
2383		−<h2 id=topcallsbyelapsedtime>1.4 Top calls by elapsed time</h2>
	2554	+<h2 id=topcallsbyelapsedtime>1.5 Top calls by elapsed time</h2>
2384	2555	The Top calls by elapsed time report displays calls in descending order of time elapsed between when the function was called and when it returned.
2385	2556	The data is reduced by various selection criteria. A dialog box displays available criteria which are AND'ed together.
2386	2557	If not all data is fetched, a Fetch bar is provided. Ctrl-F displays a Find bar.
@@ -2516,6 +2687,9 @@ class MainWindow(QMainWindow): · 129 unmodified lines
2516	2687	if IsSelectable(glb.db, "calls"):
2517	2688	reports_menu.addAction(CreateAction("Context-Sensitive Call &Graph", "Create a new window containing a context-sensitive call graph", self.NewCallGraph, self))
2518	2689
	2690	+ if IsSelectable(glb.db, "calls", "WHERE parent_id >= 0"):
	2691	+ reports_menu.addAction(CreateAction("Call &Tree", "Create a new window containing a call tree", self.NewCallTree, self))
	2692	+
2519	2693	self.EventMenu(GetEventList(glb.db), reports_menu)
2520	2694
2521	2695	if IsSelectable(glb.db, "calls"):
@@ -2576,6 +2750,9 @@ class MainWindow(QMainWindow): · 54 unmodified lines
2576	2750	def NewCallGraph(self):
2577	2751	CallGraphWindow(self.glb, self)
2578	2752
	2753	+ def NewCallTree(self):
	2754	+ CallTreeWindow(self.glb, self)
	2755	+
2579	2756	def NewTopCalls(self):
2580	2757	dialog = TopCallsDialog(self.glb, self)
2581	2758	ret = dialog.exec_()