perf callchain: Enable dwarf_callchain_users on arm64

Enable dwarf_callchain_users on arm64 which will be needed to do a
DWARF unwind in order to get the caller of the leaf frame.

Reviewed-by: James Clark <[e-mail]>
Signed-off-by: Alexandre Truong <[e-mail]>
Acked-by: Jiri Olsa <[e-mail]>
Cc: Alexander Shishkin <[e-mail]>
Cc: John Garry <[e-mail]>
Cc: Leo Yan <[e-mail]>
Cc: Mark Rutland <[e-mail]>
Cc: Mathieu Poirier <[e-mail]>
Cc: Namhyung Kim <[e-mail]>
Cc: Will Deacon <[e-mail]>
Cc: [e-mail]
Link: https://lore.kernel.org/r/[e-mail]
Signed-off-by: German Gomez <[e-mail]>
Signed-off-by: Arnaldo Carvalho de Melo <[e-mail]>

tools/perf/builtin-report.c

···
 		}
 	}
 
-	callchain_param_setup(sample_type);
+	callchain_param_setup(sample_type, perf_env__arch(&rep->session->header.env));
 
 	if (rep->stitch_lbr && (callchain_param.record_mode != CALLCHAIN_LBR)) {
 		ui__warning("Can't find LBR callchain. Switch off --stitch-lbr.\n"
···
 	 * on events sample_type.
 	 */
 	sample_type = evlist__combined_sample_type(*pevlist);
-	callchain_param_setup(sample_type);
+	callchain_param_setup(sample_type, perf_env__arch((*pevlist)->env));
 	return 0;
 }
 

tools/perf/builtin-script.c

···
 	 * on events sample_type.
 	 */
 	sample_type = evlist__combined_sample_type(evlist);
-	callchain_param_setup(sample_type);
+	callchain_param_setup(sample_type, perf_env__arch((*pevlist)->env));
 
 	/* Enable fields for callchain entries */
 	if (symbol_conf.use_callchain &&
···
 	struct perf_session *session = script->session;
 	u64 sample_type = evlist__combined_sample_type(session->evlist);
 
-	callchain_param_setup(sample_type);
+	callchain_param_setup(sample_type, perf_env__arch(session->machines.host.env));
 
 	if (script->stitch_lbr && (callchain_param.record_mode != CALLCHAIN_LBR)) {
 		pr_warning("Can't find LBR callchain. Switch off --stitch-lbr.\n"

tools/perf/util/callchain.c

···
 		map__zput(node->ms.map);
 }
 
-void callchain_param_setup(u64 sample_type)
+void callchain_param_setup(u64 sample_type, const char *arch)
 {
 	if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain) {
 		if ((sample_type & PERF_SAMPLE_REGS_USER) &&
···
 		else
 			callchain_param.record_mode = CALLCHAIN_FP;
 	}
+
+	/*
+	 * It's necessary to use libunwind to reliably determine the caller of
+	 * a leaf function on aarch64, as otherwise we cannot know whether to
+	 * start from the LR or FP.
+	 *
+	 * Always starting from the LR can result in duplicate or entirely
+	 * erroneous entries. Always skipping the LR and starting from the FP
+	 * can result in missing entries.
+	 */
+	if (callchain_param.record_mode == CALLCHAIN_FP && !strcmp(arch, "arm64"))
+		dwarf_callchain_users = true;
 }
 
 static bool chain_match(struct callchain_list *base_chain,

tools/perf/util/callchain.h

···
 			    u64 *branch_count, u64 *predicted_count,
 			    u64 *abort_count, u64 *cycles_count);
 
-void callchain_param_setup(u64 sample_type);
+void callchain_param_setup(u64 sample_type, const char *arch);
 
 bool callchain_cnode_matched(struct callchain_node *base_cnode,
 			     struct callchain_node *pair_cnode);