[tip:perf/core] perf trace: Allow overriding global --max-stack per event

From: tip-bot for Arnaldo Carvalho de Melo
Date: Wed Jan 17 2018 - 11:36:51 EST


Commit-ID: bd3dda9ab0fbdb8a91a2e869d93a0c9692b8444f
Gitweb: https://git.kernel.org/tip/bd3dda9ab0fbdb8a91a2e869d93a0c9692b8444f
Author: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
AuthorDate: Mon, 15 Jan 2018 11:33:53 -0300
Committer: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
CommitDate: Wed, 17 Jan 2018 10:23:33 -0300

perf trace: Allow overriding global --max-stack per event

The per-event max-stack setting wasn't overriding the global --max-stack
setting:

# perf trace --no-syscalls --max-stack 4 -e probe_libc:inet_pton/call-graph=dwarf,max-stack=2/ ping -6 -c 1 ::1
PING ::1(::1) 56 data bytes
64 bytes from ::1: icmp_seq=1 ttl=64 time=0.072 ms

--- ::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.072/0.072/0.072/0.000 ms
0.000 probe_libc:inet_pton:(7feb7a998350))
__inet_pton (inlined)
gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so)
__GI_getaddrinfo (inlined)
[0xffffaa39b6108f3f] (/usr/bin/ping)
#

Fix it:

# perf trace --no-syscalls --max-stack 4 -e probe_libc:inet_pton/call-graph=dwarf,max-stack=2/ ping -6 -c 1 ::1
PING ::1(::1) 56 data bytes
64 bytes from ::1: icmp_seq=1 ttl=64 time=0.073 ms

--- ::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.073/0.073/0.073/0.000 ms
0.000 probe_libc:inet_pton:(7f1083221350))
__inet_pton (inlined)
gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so)
#

Cc: Adrian Hunter <adrian.hunter@xxxxxxxxx>
Cc: David Ahern <dsahern@xxxxxxxxx>
Cc: Hendrick Brueckner <brueckner@xxxxxxxxxxxxxxxxxx>
Cc: Jiri Olsa <jolsa@xxxxxxxxxx>
Cc: Namhyung Kim <namhyung@xxxxxxxxxx>
Cc: Thomas Richter <tmricht@xxxxxxxxxxxxxxxxxx>
Cc: Wang Nan <wangnan0@xxxxxxxxxx>
Link: https://lkml.kernel.org/n/tip-ic3g837xg8ob3kcpkspxwz0g@xxxxxxxxxxxxxx
Signed-off-by: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
---
tools/perf/builtin-trace.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index ee85c29..531d43b 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -1644,7 +1644,7 @@ static int trace__resolve_callchain(struct trace *trace, struct perf_evsel *evse
struct addr_location al;

if (machine__resolve(trace->host, &al, sample) < 0 ||
- thread__resolve_callchain(al.thread, cursor, evsel, sample, NULL, NULL, trace->max_stack))
+ thread__resolve_callchain(al.thread, cursor, evsel, sample, NULL, NULL, evsel->attr.sample_max_stack))
return -1;

return 0;
@@ -2423,6 +2423,18 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
trace->multiple_threads = thread_map__pid(evlist->threads, 0) == -1 ||
evlist->threads->nr > 1 ||
perf_evlist__first(evlist)->attr.inherit;
+
+ /*
+ * Now that we already used evsel->attr to ask the kernel to setup the
+ * events, lets reuse evsel->attr.sample_max_stack as the limit in
+ * trace__resolve_callchain(), allowing per-event max-stack settings
+ * to override an explicitely set --max-stack global setting.
+ */
+ evlist__for_each_entry(evlist, evsel) {
+ if ((evsel->attr.sample_type & PERF_SAMPLE_CALLCHAIN) &&
+ evsel->attr.sample_max_stack == 0)
+ evsel->attr.sample_max_stack = trace->max_stack;
+ }
again:
before = trace->nr_events;