by Jiri Olsa

[permalink] [raw]

Subject: Re: [PATCH 2/5] perf unwind: Do not look at globals

On Tue, Jan 16, 2018 at 05:05:20PM -0300, Arnaldo Carvalho de Melo wrote:

SNIP

> > >
> > > +/*
> > > + * Are there any events usind DWARF callchains?
> > > + *
> > > + * I.e.
> > > + *
> > > + * -e cycles/call-graph=dwarf/
> > > + */
> > > +bool dwarf_callchain_users;
> >
> > hum, I don't follow.. this bool seems to mirror the usage of
> > 'param->record_mode = CALLCHAIN_DWARF', whats the difference?
> >
> > also, the patch title says 'Do not look at globals', while inside you
>
> The first version didn't look at globals, the second one doesn't look at
> an _specific_ global variable, the global config for --call-graph, which
> is a global variable, callchain_param, which _we_ can't touch at
> apply_config_terms(), since that is about _just_ that event, not all of
> them.
>
> > add new global dwarf_callchain_users and work with it.. what do I miss?
> >
> > I'll check tomorrow with clean head ;-)
>
> Look closely at apply_config_terms() it passes a _local_ variable to
>
> perf_evsel__config_callchain(evsel, opts, &param);
>
> It will not affect any globals that tools/perf/util/unwind-libunwind-local.c
> could possibly use... and that is the problem. :-)

ah ok, that one is local.. just to set up the attrs

then adding that new variable is ok, though it could be
aded inside struct callchain_param, to keep callchain
config stuff together

thanks,
jirka

2018-01-17 16:37:02

by tip-bot for Vasyl Gomonovych

[permalink] [raw]

Subject: [tip:perf/core] perf unwind: Do not look just at the global callchain_param.record_mode

Commit-ID: eabad8c6856f185f876b54c426c2cc69fe0f0a7d
Gitweb: https://git.kernel.org/tip/eabad8c6856f185f876b54c426c2cc69fe0f0a7d
Author: Arnaldo Carvalho de Melo <[email protected]>
AuthorDate: Mon, 15 Jan 2018 16:48:46 -0300
Committer: Arnaldo Carvalho de Melo <[email protected]>
CommitDate: Wed, 17 Jan 2018 10:23:32 -0300

perf unwind: Do not look just at the global callchain_param.record_mode

When setting up DWARF callchains on specific events, without using
'record' or 'trace' --call-graph, but instead doing it like:

perf trace -e cycles/call-graph=dwarf/

The unwind__prepare_access() call in thread__insert_map() when we
process PERF_RECORD_MMAP(2) metadata events were not being performed,
precluding us from using per-event DWARF callchains, handling them just
when we asked for all events to be DWARF, using "--call-graph dwarf".

We do it in the PERF_RECORD_MMAP because we have to look at one of the
executable maps to figure out the executable type (64-bit, 32-bit) of
the DSO laid out in that mmap. Also to look at the architecture where
the perf.data file was recorded.

All this probably should be deferred to when we process a sample for
some thread that has callchains, so that we do this processing only for
the threads with samples, not for all of them.

For now, fix using DWARF on specific events.

Before:

# perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1
PING ::1(::1) 56 data bytes
64 bytes from ::1: icmp_seq=1 ttl=64 time=0.048 ms

--- ::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.048/0.048/0.048/0.000 ms
0.000 probe_libc:inet_pton:(7fe9597bb350))
Problem processing probe_libc:inet_pton callchain, skipping...
#

After:

# perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1
PING ::1(::1) 56 data bytes
64 bytes from ::1: icmp_seq=1 ttl=64 time=0.060 ms

--- ::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.060/0.060/0.060/0.000 ms
0.000 probe_libc:inet_pton:(7fd4aa930350))
__inet_pton (inlined)
gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so)
__GI_getaddrinfo (inlined)
[0xffffaa804e51af3f] (/usr/bin/ping)
__libc_start_main (/usr/lib64/libc-2.26.so)
[0xffffaa804e51b379] (/usr/bin/ping)
#
# perf trace --call-graph=dwarf --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1
PING ::1(::1) 56 data bytes
64 bytes from ::1: icmp_seq=1 ttl=64 time=0.057 ms

--- ::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.057/0.057/0.057/0.000 ms
0.000 probe_libc:inet_pton:(7f9363b9e350))
__inet_pton (inlined)
gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so)
__GI_getaddrinfo (inlined)
[0xffffa9e8a14e0f3f] (/usr/bin/ping)
__libc_start_main (/usr/lib64/libc-2.26.so)
[0xffffa9e8a14e1379] (/usr/bin/ping)
#
# perf trace --call-graph=fp --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1
PING ::1(::1) 56 data bytes
64 bytes from ::1: icmp_seq=1 ttl=64 time=0.077 ms

--- ::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms
0.000 probe_libc:inet_pton:(7f4947e1c350))
__inet_pton (inlined)
gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so)
__GI_getaddrinfo (inlined)
[0xffffaa716d88ef3f] (/usr/bin/ping)
__libc_start_main (/usr/lib64/libc-2.26.so)
[0xffffaa716d88f379] (/usr/bin/ping)
#
# perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=fp/ ping -6 -c 1 ::1
PING ::1(::1) 56 data bytes
64 bytes from ::1: icmp_seq=1 ttl=64 time=0.078 ms

--- ::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.078/0.078/0.078/0.000 ms
0.000 probe_libc:inet_pton:(7fa157696350))
__GI___inet_pton (/usr/lib64/libc-2.26.so)
getaddrinfo (/usr/lib64/libc-2.26.so)
[0xffffa9ba39c74f40] (/usr/bin/ping)
#

Acked-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: David Ahern <[email protected]>
Cc: Hendrick Brueckner <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Thomas Richter <[email protected]>
Cc: Wang Nan <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
---
tools/perf/builtin-c2c.c | 5 +++--
tools/perf/builtin-report.c | 5 +++--
tools/perf/builtin-script.c | 5 +++--
tools/perf/tests/dwarf-unwind.c | 1 +
tools/perf/util/callchain.c | 10 ++++++++++
tools/perf/util/callchain.h | 2 ++
tools/perf/util/unwind-libunwind-local.c | 9 +++------
7 files changed, 25 insertions(+), 12 deletions(-)

diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c
index c0debc3..c0815a3 100644
--- a/tools/perf/builtin-c2c.c
+++ b/tools/perf/builtin-c2c.c
@@ -2390,9 +2390,10 @@ static int setup_callchain(struct perf_evlist *evlist)
enum perf_call_graph_mode mode = CALLCHAIN_NONE;

if ((sample_type & PERF_SAMPLE_REGS_USER) &&
- (sample_type & PERF_SAMPLE_STACK_USER))
+ (sample_type & PERF_SAMPLE_STACK_USER)) {
mode = CALLCHAIN_DWARF;
- else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
+ dwarf_callchain_users = true;
+ } else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
mode = CALLCHAIN_LBR;
else if (sample_type & PERF_SAMPLE_CALLCHAIN)
mode = CALLCHAIN_FP;
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index dd4df9a..6593779 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -338,9 +338,10 @@ static int report__setup_sample_type(struct report *rep)

if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain) {
if ((sample_type & PERF_SAMPLE_REGS_USER) &&
- (sample_type & PERF_SAMPLE_STACK_USER))
+ (sample_type & PERF_SAMPLE_STACK_USER)) {
callchain_param.record_mode = CALLCHAIN_DWARF;
- else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
+ dwarf_callchain_users = true;
+ } else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
callchain_param.record_mode = CALLCHAIN_LBR;
else
callchain_param.record_mode = CALLCHAIN_FP;
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index c1cce47..08bc818 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -2919,9 +2919,10 @@ static void script__setup_sample_type(struct perf_script *script)

if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain) {
if ((sample_type & PERF_SAMPLE_REGS_USER) &&
- (sample_type & PERF_SAMPLE_STACK_USER))
+ (sample_type & PERF_SAMPLE_STACK_USER)) {
callchain_param.record_mode = CALLCHAIN_DWARF;
- else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
+ dwarf_callchain_users = true;
+ } else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
callchain_param.record_mode = CALLCHAIN_LBR;
else
callchain_param.record_mode = CALLCHAIN_FP;
diff --git a/tools/perf/tests/dwarf-unwind.c b/tools/perf/tests/dwarf-unwind.c
index ac40e05..26041896 100644
--- a/tools/perf/tests/dwarf-unwind.c
+++ b/tools/perf/tests/dwarf-unwind.c
@@ -173,6 +173,7 @@ int test__dwarf_unwind(struct test *test __maybe_unused, int subtest __maybe_unu
}

callchain_param.record_mode = CALLCHAIN_DWARF;
+ dwarf_callchain_users = true;

if (init_live_machine(machine)) {
pr_err("Could not init machine\n");
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index 082505d0..32ef7bd 100644
--- a/tools/perf/util/callchain.c
+++ b/tools/perf/util/callchain.c
@@ -37,6 +37,15 @@ struct callchain_param callchain_param = {
CALLCHAIN_PARAM_DEFAULT
};

+/*
+ * Are there any events usind DWARF callchains?
+ *
+ * I.e.
+ *
+ * -e cycles/call-graph=dwarf/
+ */
+bool dwarf_callchain_users;
+
struct callchain_param callchain_param_default = {
CALLCHAIN_PARAM_DEFAULT
};
@@ -265,6 +274,7 @@ int parse_callchain_record(const char *arg, struct callchain_param *param)
ret = 0;
param->record_mode = CALLCHAIN_DWARF;
param->dump_size = default_stack_dump_size;
+ dwarf_callchain_users = true;

tok = strtok_r(NULL, ",", &saveptr);
if (tok) {
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
index b79ef24..154560b 100644
--- a/tools/perf/util/callchain.h
+++ b/tools/perf/util/callchain.h
@@ -89,6 +89,8 @@ enum chain_value {
CCVAL_COUNT,
};

+extern bool dwarf_callchain_users;
+
struct callchain_param {
bool enabled;
enum perf_call_graph_mode record_mode;
diff --git a/tools/perf/util/unwind-libunwind-local.c b/tools/perf/util/unwind-libunwind-local.c
index 7a42f70..af87304 100644
--- a/tools/perf/util/unwind-libunwind-local.c
+++ b/tools/perf/util/unwind-libunwind-local.c
@@ -631,9 +631,8 @@ static unw_accessors_t accessors = {

static int _unwind__prepare_access(struct thread *thread)
{
- if (callchain_param.record_mode != CALLCHAIN_DWARF)
+ if (!dwarf_callchain_users)
return 0;
-
thread->addr_space = unw_create_addr_space(&accessors, 0);
if (!thread->addr_space) {
pr_err("unwind: Can't create unwind address space.\n");
@@ -646,17 +645,15 @@ static int _unwind__prepare_access(struct thread *thread)

static void _unwind__flush_access(struct thread *thread)
{
- if (callchain_param.record_mode != CALLCHAIN_DWARF)
+ if (!dwarf_callchain_users)
return;
-
unw_flush_cache(thread->addr_space, 0, 0);
}

static void _unwind__finish_access(struct thread *thread)
{
- if (callchain_param.record_mode != CALLCHAIN_DWARF)
+ if (!dwarf_callchain_users)
return;
-
unw_destroy_addr_space(thread->addr_space);
}