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

Groups > linux.kernel > #1403992

[PATCH 06/11] perf callchain: Stop validating callchains by the max_stack sysctl

Show all headers | View raw

From: Arnaldo Carvalho de Melo <acme@redhat.com>

As thread__resolve_callchain_sample can be used for handling perf.data
files, that could've been recorded with a large max_stack sysctl setting
than what the system used for analysis has set.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: He Kuang <hekuang@huawei.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: Zefan Li <lizefan@huawei.com>
Link: http://lkml.kernel.org/n/tip-2995bt2g5yq2m05vga4kip6m@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/machine.c | 25 +++++--------------------
 1 file changed, 5 insertions(+), 20 deletions(-)

diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index bdc33ce40bdc..205d27017361 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1771,11 +1771,6 @@ static int resolve_lbr_callchain_sample(struct thread *thread,
 		 */
 		int mix_chain_nr = i + 1 + lbr_nr + 1;
 
-		if (mix_chain_nr > (int)sysctl_perf_event_max_stack + PERF_MAX_BRANCH_DEPTH) {
-			pr_warning("corrupted callchain. skipping...\n");
-			return 0;
-		}
-
 		for (j = 0; j < mix_chain_nr; j++) {
 			if (callchain_param.order == ORDER_CALLEE) {
 				if (j < i + 1)
@@ -1815,7 +1810,7 @@ static int thread__resolve_callchain_sample(struct thread *thread,
 	struct ip_callchain *chain = sample->callchain;
 	int chain_nr = chain->nr;
 	u8 cpumode = PERF_RECORD_MISC_USER;
-	int i, j, err, nr_entries, nr_contexts;
+	int i, j, err, nr_entries;
 	int skip_idx = -1;
 	int first_call = 0;
 
@@ -1830,8 +1825,7 @@ static int thread__resolve_callchain_sample(struct thread *thread,
 	 * Based on DWARF debug information, some architectures skip
 	 * a callchain entry saved by the kernel.
 	 */
-	if (chain_nr < sysctl_perf_event_max_stack)
-		skip_idx = arch_skip_callchain_idx(thread, chain);
+	skip_idx = arch_skip_callchain_idx(thread, chain);
 
 	/*
 	 * Add branches to call stack for easier browsing. This gives
@@ -1891,7 +1885,7 @@ static int thread__resolve_callchain_sample(struct thread *thread,
 	}
 
 check_calls:
-	for (i = first_call, nr_entries = 0, nr_contexts = 0;
+	for (i = first_call, nr_entries = 0;
 	     i < chain_nr && nr_entries < max_stack; i++) {
 		u64 ip;
 
@@ -1906,13 +1900,8 @@ check_calls:
 #endif
 		ip = chain->ips[j];
 
-		if (ip >= PERF_CONTEXT_MAX) {
-			if (++nr_contexts > sysctl_perf_event_max_contexts_per_stack)
-				goto out_corrupted_callchain;
-		} else {
-			if (++nr_entries > sysctl_perf_event_max_stack)
-				goto out_corrupted_callchain;
-		}
+		if (ip < PERF_CONTEXT_MAX)
+                       ++nr_entries;
 
 		err = add_callchain_ip(thread, cursor, parent, root_al, &cpumode, ip);
 
@@ -1921,10 +1910,6 @@ check_calls:
 	}
 
 	return 0;
-
-out_corrupted_callchain:
-	pr_warning("corrupted callchain. skipping...\n");
-	return 0;
 }
 
 static int unwind_entry(struct unwind_entry *entry, void *arg)
-- 
2.5.5

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

Thread

[GIT PULL 00/11] perf/core improvements and fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 04/11] perf top: Use machine->kptr_restrict_warned Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 02/11] perf machine: Do not bail out if not managing to read ref reloc symbol Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 10/11] perf trace: Only auto set call-graph to "dwarf" when syscalls are being traced Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 03/11] perf trace: Warn when trying to resolve kernel addresses with kptr_restrict=1 Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 09/11] perf annotate: Sort list of recognised instructions Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 08/11] perf annotate: Fix identification of ARM blt and bls instructions Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 06/11] perf callchain: Stop validating callchains by the max_stack sysctl Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 01/11] perf tools: Find vdso supporting cross-platform analysis Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 07/11] perf tools: Fix usage of max_stack sysctl Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 11/11] perf tools: Set buildid dir under symfs when --symfs is provided Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  [PATCH 05/11] perf trace: Fix exit_group() formatting Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 00:30 +0200
  Re: [GIT PULL 00/11] perf/core improvements and fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-05-20 17:10 +0200
    Re: [GIT PULL 00/11] perf/core improvements and fixes Ingo Molnar <mingo@kernel.org> - 2016-05-20 19:40 +0200

csiph-web