* [GIT PULL 00/18] perf/core improvements and fixes
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
0 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, Alexander Shishkin,
Andi Kleen, Andy Lutomirski, Dave Hansen, David Ahern,
H . Peter Anvin, Jin Yao, Jiri Olsa, Joerg Roedel, Kan Liang,
Namhyung Kim, Naveen N . Rao, Peter Zijlstra, Ravi Bangoria,
Sandipan Das
Hi Ingo,
Please consider pulling,
- Arnaldo
Test results at the end of this message, as usual.
The following changes since commit 5aafae8d097e2161ee5c6a12ad532100f8885d2b:
Merge tag 'perf-core-for-mingo-4.18-20180516' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2018-05-16 17:56:43 +0200)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-4.18-20180519
for you to fetch changes up to 19422a9f2a3be7f3a046285ffae4cbb571aa853a:
perf tools: Fix kernel_start for PTI on x86 (2018-05-19 06:42:51 -0300)
----------------------------------------------------------------
perf/core improvements and fixes:
- Record min/max LBR cycles (>= skylake) and add 'perf annotate' TUI
hotkey to show it (c) (Jin Yao)
- Fix machine->kernel_start for PTI on x86 (Adrian Hunter)
- Make machine->env->arch always available, e.g. in 'perf top', not
just when reading that info from perf.data files (Adrian Hunter)
- Reduce the number of files read at 'perf' start, leaving information such as
cacheline size, tracefs mount point determination, max_stack, etc, to be
lazily read as tools needs then (Arnaldo Carvalho de Melo)
- Fixup BPF include and examples install messages (Arnaldo Carvalho de Melo)
- Fixup callchain addresses and symbol offsets in 'perf script', to help
correlating with objdump output (Sandipan Das)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
----------------------------------------------------------------
Adrian Hunter (2):
perf machine: Add machine__is() to identify machine arch
perf tools: Fix kernel_start for PTI on x86
Arnaldo Carvalho de Melo (12):
perf config: Call perf_config__init() lazily
tools lib api: The tracing_mnt variable doesn't need to be global
tools lib api: Unexport 'tracing_path' variable
tools lib api fs tracing_path: Introduce get/put_events_file() helpers
perf tools: Reuse the path to the tracepoint /events/ directory
perf parse-events: Use get/put_events_file()
tools lib api fs tracing_path: Introduce opendir() method
tools lib api fs tracing_path: Make tracing_events_path private
tools include compiler-gcc: Add __pure attribute helper
perf tools: Read the cache line size lazily
perf tools: No need to unconditionally read the max_stack sysctls
perf bpf: Fixup include and examples install messages
Jin Yao (2):
perf annotate: Record the min/max cycles
perf annotate: Create hotkey 'c' to show min/max cycles
Sandipan Das (2):
perf script: Show virtual addresses instead of offsets
perf script: Show symbol offsets by default
tools/include/linux/compiler-gcc.h | 3 +
tools/lib/api/fs/tracing_path.c | 40 +++++++++---
tools/lib/api/fs/tracing_path.h | 9 ++-
tools/perf/Makefile.perf | 2 +
tools/perf/builtin-script.c | 26 ++++----
tools/perf/builtin-top.c | 2 +-
tools/perf/builtin-trace.c | 2 +-
tools/perf/perf.c | 24 +------
tools/perf/tests/parse-events.c | 9 +--
.../tests/shell/record+probe_libc_inet_pton.sh | 12 ++--
tools/perf/ui/browsers/annotate.c | 8 +++
tools/perf/util/annotate.c | 51 ++++++++++++---
tools/perf/util/annotate.h | 11 +++-
tools/perf/util/config.c | 16 ++---
tools/perf/util/config.h | 1 -
tools/perf/util/env.c | 18 ++++++
tools/perf/util/env.h | 2 +
tools/perf/util/evsel.c | 2 +-
tools/perf/util/machine.c | 18 +++++-
tools/perf/util/machine.h | 2 +
tools/perf/util/parse-events.c | 73 +++++++++++++---------
tools/perf/util/probe-file.c | 3 +-
tools/perf/util/sort.c | 4 +-
tools/perf/util/sort.h | 4 +-
tools/perf/util/trace-event-info.c | 11 ++--
tools/perf/util/trace-event.c | 8 ++-
tools/perf/util/util.c | 34 +++++++++-
tools/perf/util/util.h | 4 +-
28 files changed, 279 insertions(+), 120 deletions(-)
Test results:
The first ones are container (docker) based builds of tools/perf with
and without libelf support. Where clang is available, it is also used
to build perf with/without libelf, and building with LIBCLANGLLVM=1
(built-in clang) with gcc and clang when clang and its devel libraries
are installed.
The objtool and samples/bpf/ builds are disabled now that I'm switching from
using the sources in a local volume to fetching them from a http server to
build it inside the container, to make it easier to build in a container cluster.
Those will come back later.
Several are cross builds, the ones with -x-ARCH and the android one, and those
may not have all the features built, due to lack of multi-arch devel packages,
available and being used so far on just a few, like
debian:experimental-x-{arm64,mipsel}.
The 'perf test' one will perform a variety of tests exercising
tools/perf/util/, tools/lib/{bpf,traceevent,etc}, as well as run perf commands
with a variety of command line event specifications to then intercept the
sys_perf_event syscall to check that the perf_event_attr fields are set up as
expected, among a variety of other unit tests.
Then there is the 'make -C tools/perf build-test' ones, that build tools/perf/
with a variety of feature sets, exercising the build with an incomplete set of
features as well as with a complete one. It is planned to have it run on each
of the containers mentioned above, using some container orchestration
infrastructure. Get in contact if interested in helping having this in place.
# dm
1 alpine:3.4 : Ok gcc (Alpine 5.3.0) 5.3.0
2 alpine:3.5 : Ok gcc (Alpine 6.2.1) 6.2.1 20160822
3 alpine:3.6 : Ok gcc (Alpine 6.3.0) 6.3.0
4 alpine:3.7 : Ok gcc (Alpine 6.4.0) 6.4.0
5 alpine:edge : Ok gcc (Alpine 6.4.0) 6.4.0
6 amazonlinux:1 : Ok gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28)
7 amazonlinux:2 : Ok gcc (GCC) 7.3.1 20180303 (Red Hat 7.3.1-5)
8 android-ndk:r12b-arm : Ok arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
9 android-ndk:r15c-arm : Ok arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
10 centos:5 : Ok gcc (GCC) 4.1.2 20080704 (Red Hat 4.1.2-55)
11 centos:6 : Ok gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-18)
12 centos:7 : Ok gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-16)
13 debian:7 : Ok gcc (Debian 4.7.2-5) 4.7.2
14 debian:8 : Ok gcc (Debian 4.9.2-10+deb8u1) 4.9.2
15 debian:9 : Ok gcc (Debian 6.3.0-18+deb9u1) 6.3.0 20170516
16 debian:experimental : Ok gcc (Debian 7.3.0-19) 7.3.0
17 debian:experimental-x-arm64 : Ok aarch64-linux-gnu-gcc (Debian 7.3.0-19) 7.3.0
18 debian:experimental-x-mips : Ok mips-linux-gnu-gcc (Debian 7.3.0-19) 7.3.0
19 debian:experimental-x-mips64 : Ok mips64-linux-gnuabi64-gcc (Debian 7.3.0-18) 7.3.0
20 debian:experimental-x-mipsel : Ok mipsel-linux-gnu-gcc (Debian 7.3.0-19) 7.3.0
21 fedora:20 : Ok gcc (GCC) 4.8.3 20140911 (Red Hat 4.8.3-7)
22 fedora:21 : Ok gcc (GCC) 4.9.2 20150212 (Red Hat 4.9.2-6)
23 fedora:22 : Ok gcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6)
24 fedora:23 : Ok gcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6)
25 fedora:24 : Ok gcc (GCC) 6.3.1 20161221 (Red Hat 6.3.1-1)
26 fedora:24-x-ARC-uClibc : Ok arc-linux-gcc (ARCompact ISA Linux uClibc toolchain 2017.09-rc2) 7.1.1 20170710
27 fedora:25 : Ok gcc (GCC) 6.4.1 20170727 (Red Hat 6.4.1-1)
28 fedora:26 : Ok gcc (GCC) 7.3.1 20180130 (Red Hat 7.3.1-2)
29 fedora:27 : Ok gcc (GCC) 7.3.1 20180303 (Red Hat 7.3.1-5)
30 fedora:28 : Ok gcc (GCC) 8.1.1 20180502 (Red Hat 8.1.1-1)
31 fedora:rawhide : Ok gcc (GCC) 8.0.1 20180324 (Red Hat 8.0.1-0.20)
32 gentoo-stage3-amd64:latest : Ok gcc (Gentoo 6.4.0-r1 p1.3) 6.4.0
33 mageia:5 : Ok gcc (GCC) 4.9.2
34 mageia:6 : Ok gcc (Mageia 5.5.0-1.mga6) 5.5.0
35 opensuse:42.1 : Ok gcc (SUSE Linux) 4.8.5
36 opensuse:42.2 : Ok gcc (SUSE Linux) 4.8.5
37 opensuse:42.3 : Ok gcc (SUSE Linux) 4.8.5
38 opensuse:tumbleweed : Ok gcc (SUSE Linux) 7.3.1 20180323 [gcc-7-branch revision 258812]
39 oraclelinux:6 : Ok gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-18.0.7)
40 oraclelinux:7 : Ok gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28.0.1)
41 ubuntu:12.04.5 : Ok gcc (Ubuntu/Linaro 4.6.3-1ubuntu5) 4.6.3
42 ubuntu:14.04.4 : Ok gcc (Ubuntu 4.8.4-2ubuntu1~14.04.3) 4.8.4
43 ubuntu:14.04.4-x-linaro-arm64 : Ok aarch64-linux-gnu-gcc (Linaro GCC 5.5-2017.10) 5.5.0
44 ubuntu:16.04 : Ok gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
45 ubuntu:16.04-x-arm : Ok arm-linux-gnueabihf-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
46 ubuntu:16.04-x-arm64 : Ok aarch64-linux-gnu-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
47 ubuntu:16.04-x-powerpc : Ok powerpc-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
48 ubuntu:16.04-x-powerpc64 : Ok powerpc64-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
49 ubuntu:16.04-x-powerpc64el : Ok powerpc64le-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
50 ubuntu:16.04-x-s390 : Ok s390x-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
51 ubuntu:16.10 : Ok gcc (Ubuntu 6.2.0-5ubuntu12) 6.2.0 20161005
52 ubuntu:17.04 : Ok gcc (Ubuntu 6.3.0-12ubuntu2) 6.3.0 20170406
53 ubuntu:17.10 : Ok gcc (Ubuntu 7.2.0-8ubuntu3.2) 7.2.0
54 ubuntu:18.04 : Ok gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
#
# uname -a
Linux seventh 4.17.0-rc5-00069-g3acf4e395260 #2 SMP Fri May 18 12:13:59 -03 2018 x86_64 x86_64 x86_64 GNU/Linux
# perf --version
perf version 4.17.rc5.g19422a9
# perf test
1: vmlinux symtab matches kallsyms : Ok
2: Detect openat syscall event : Ok
3: Detect openat syscall event on all cpus : Ok
4: Read samples using the mmap interface : Ok
5: Test data source output : Ok
6: Parse event definition strings : Ok
7: Simple expression parser : Ok
8: PERF_RECORD_* events & perf_sample fields : Ok
9: Parse perf pmu format : Ok
10: DSO data read : Ok
11: DSO data cache : Ok
12: DSO data reopen : Ok
13: Roundtrip evsel->name : Ok
14: Parse sched tracepoints fields : Ok
15: syscalls:sys_enter_openat event fields : Ok
16: Setup struct perf_event_attr : Ok
17: Match and link multiple hists : Ok
18: 'import perf' in python : Ok
19: Breakpoint overflow signal handler : Ok
20: Breakpoint overflow sampling : Ok
21: Breakpoint accounting : Ok
22: Number of exit events of a simple workload : Ok
23: Software clock events period values : Ok
24: Object code reading : Ok
25: Sample parsing : Ok
26: Use a dummy software event to keep tracking : Ok
27: Parse with no sample_id_all bit set : Ok
28: Filter hist entries : Ok
29: Lookup mmap thread : Ok
30: Share thread mg : Ok
31: Sort output of hist entries : Ok
32: Cumulate child hist entries : Ok
33: Track with sched_switch : Ok
34: Filter fds with revents mask in a fdarray : Ok
35: Add fd to a fdarray, making it autogrow : Ok
36: kmod_path__parse : Ok
37: Thread map : Ok
38: LLVM search and compile :
38.1: Basic BPF llvm compile : Ok
38.2: kbuild searching : Ok
38.3: Compile source for BPF prologue generation : Ok
38.4: Compile source for BPF relocation : Ok
39: Session topology : Ok
40: BPF filter :
40.1: Basic BPF filtering : Ok
40.2: BPF pinning : Ok
40.3: BPF prologue generation : Ok
40.4: BPF relocation checker : Ok
41: Synthesize thread map : Ok
42: Remove thread map : Ok
43: Synthesize cpu map : Ok
44: Synthesize stat config : Ok
45: Synthesize stat : Ok
46: Synthesize stat round : Ok
47: Synthesize attr update : Ok
48: Event times : Ok
49: Read backward ring buffer : Ok
50: Print cpu map : Ok
51: Probe SDT events : Ok
52: is_printable_array : Ok
53: Print bitmap : Ok
54: perf hooks : Ok
55: builtin clang support : Skip (not compiled in)
56: unit_number__scnprintf : Ok
57: mem2node : Ok
58: x86 rdpmc : Ok
59: Convert perf time to TSC : Ok
60: DWARF unwind : Ok
61: x86 instruction decoder - new instructions : Ok
62: Use vfs_getname probe to get syscall args filenames : Ok
63: Check open filename arg using perf trace + vfs_getname: Ok
64: probe libc's inet_pton & backtrace it with ping : Ok
65: Add vfs_getname probe to get syscall args filenames : Ok
#
$ make -C tools/perf build-test
make: Entering directory '/home/acme/git/perf/tools/perf'
- tarpkg: ./tests/perf-targz-src-pkg .
make_debug_O: make DEBUG=1
make_no_backtrace_O: make NO_BACKTRACE=1
make_with_babeltrace_O: make LIBBABELTRACE=1
make_no_libaudit_O: make NO_LIBAUDIT=1
make_no_libnuma_O: make NO_LIBNUMA=1
make_with_clangllvm_O: make LIBCLANGLLVM=1
make_static_O: make LDFLAGS=-static
make_no_libperl_O: make NO_LIBPERL=1
make_util_map_o_O: make util/map.o
make_no_libelf_O: make NO_LIBELF=1
make_no_slang_O: make NO_SLANG=1
make_install_prefix_slash_O: make install prefix=/tmp/krava/
make_no_libdw_dwarf_unwind_O: make NO_LIBDW_DWARF_UNWIND=1
make_help_O: make help
make_no_libbionic_O: make NO_LIBBIONIC=1
make_no_libunwind_O: make NO_LIBUNWIND=1
make_no_newt_O: make NO_NEWT=1
make_no_demangle_O: make NO_DEMANGLE=1
make_no_ui_O: make NO_NEWT=1 NO_SLANG=1 NO_GTK2=1
make_no_scripts_O: make NO_LIBPYTHON=1 NO_LIBPERL=1
make_perf_o_O: make perf.o
make_install_O: make install
make_util_pmu_bison_o_O: make util/pmu-bison.o
make_tags_O: make tags
make_install_prefix_O: make install prefix=/tmp/krava
make_minimal_O: make NO_LIBPERL=1 NO_LIBPYTHON=1 NO_NEWT=1 NO_GTK2=1 NO_DEMANGLE=1 NO_LIBELF=1 NO_LIBUNWIND=1 NO_BACKTRACE=1 NO_LIBNUMA=1 NO_LIBAUDIT=1 NO_LIBBIONIC=1 NO_LIBDW_DWARF_UNWIND=1 NO_AUXTRACE=1 NO_LIBBPF=1 NO_LIBCRYPTO=1 NO_SDT=1 NO_JVMTI=1
make_clean_all_O: make clean all
make_no_gtk2_O: make NO_GTK2=1
make_no_auxtrace_O: make NO_AUXTRACE=1
make_no_libpython_O: make NO_LIBPYTHON=1
make_no_libbpf_O: make NO_LIBBPF=1
make_install_bin_O: make install-bin
make_doc_O: make doc
make_pure_O: make
OK
make: Leaving directory '/home/acme/git/perf/tools/perf'
$
^ permalink raw reply [flat|nested] 22+ messages in thread* [PATCH 01/18] perf config: Call perf_config__init() lazily
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Taeung Song, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
We check what perf_config__init() does at each perf_config() call,
namely if the static perf_config instance was created, so instead of
bailing out in that case, try to allocate it, bailing if it fails.
Now to get the perf_config() call out of the start of perf's main()
function, doing it also lazily.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Taeung Song <treeze.taeung@gmail.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-4bo45k6ivsmbxpfpdte4orsg@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/perf.c | 1 -
tools/perf/util/config.c | 16 +++++++++-------
tools/perf/util/config.h | 1 -
3 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index 20a08cb32332..cd6ea55d4b0c 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -458,7 +458,6 @@ int main(int argc, const char **argv)
srandom(time(NULL));
- perf_config__init();
err = perf_config(perf_default_config, NULL);
if (err)
return err;
diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c
index 84eb9393c7db..5ac157056cdf 100644
--- a/tools/perf/util/config.c
+++ b/tools/perf/util/config.c
@@ -707,6 +707,14 @@ struct perf_config_set *perf_config_set__new(void)
return set;
}
+static int perf_config__init(void)
+{
+ if (config_set == NULL)
+ config_set = perf_config_set__new();
+
+ return config_set == NULL;
+}
+
int perf_config(config_fn_t fn, void *data)
{
int ret = 0;
@@ -714,7 +722,7 @@ int perf_config(config_fn_t fn, void *data)
struct perf_config_section *section;
struct perf_config_item *item;
- if (config_set == NULL)
+ if (config_set == NULL && perf_config__init())
return -1;
perf_config_set__for_each_entry(config_set, section, item) {
@@ -735,12 +743,6 @@ int perf_config(config_fn_t fn, void *data)
return ret;
}
-void perf_config__init(void)
-{
- if (config_set == NULL)
- config_set = perf_config_set__new();
-}
-
void perf_config__exit(void)
{
perf_config_set__delete(config_set);
diff --git a/tools/perf/util/config.h b/tools/perf/util/config.h
index baf82bf227ac..bd0a5897c76a 100644
--- a/tools/perf/util/config.h
+++ b/tools/perf/util/config.h
@@ -38,7 +38,6 @@ struct perf_config_set *perf_config_set__new(void);
void perf_config_set__delete(struct perf_config_set *set);
int perf_config_set__collect(struct perf_config_set *set, const char *file_name,
const char *var, const char *value);
-void perf_config__init(void);
void perf_config__exit(void);
void perf_config__refresh(void);
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 02/18] tools lib api: The tracing_mnt variable doesn't need to be global
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
(?)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Its only used in the file it is defined, so just make it static.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-p5x29u6mq2ml3mtnbg9844ad@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/lib/api/fs/tracing_path.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/tools/lib/api/fs/tracing_path.c b/tools/lib/api/fs/tracing_path.c
index 7b7fd0b18551..4f8ec7d476b8 100644
--- a/tools/lib/api/fs/tracing_path.c
+++ b/tools/lib/api/fs/tracing_path.c
@@ -13,8 +13,7 @@
#include "tracing_path.h"
-
-char tracing_mnt[PATH_MAX] = "/sys/kernel/debug";
+static char tracing_mnt[PATH_MAX] = "/sys/kernel/debug";
char tracing_path[PATH_MAX] = "/sys/kernel/debug/tracing";
char tracing_events_path[PATH_MAX] = "/sys/kernel/debug/tracing/events";
@@ -129,7 +128,7 @@ int tracing_path__strerror_open_tp(int err, char *buf, size_t size,
snprintf(buf, size,
"Error:\tNo permissions to read %s/%s\n"
"Hint:\tTry 'sudo mount -o remount,mode=755 %s'\n",
- tracing_events_path, filename, tracing_mnt);
+ tracing_events_path, filename, tracing_path_mount());
}
break;
default:
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH 03/18] tools lib api: Unexport 'tracing_path' variable
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (2 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
One should use tracing_path_mount() instead, so more things get done
lazily instead of at every 'perf' tool call startup.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-fci4yll35idd9yuslp67vqc2@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/lib/api/fs/tracing_path.c | 4 ++--
tools/lib/api/fs/tracing_path.h | 1 -
tools/perf/perf.c | 5 +----
tools/perf/util/probe-file.c | 3 +--
4 files changed, 4 insertions(+), 9 deletions(-)
diff --git a/tools/lib/api/fs/tracing_path.c b/tools/lib/api/fs/tracing_path.c
index 4f8ec7d476b8..6f5fe942eff4 100644
--- a/tools/lib/api/fs/tracing_path.c
+++ b/tools/lib/api/fs/tracing_path.c
@@ -14,7 +14,7 @@
#include "tracing_path.h"
static char tracing_mnt[PATH_MAX] = "/sys/kernel/debug";
-char tracing_path[PATH_MAX] = "/sys/kernel/debug/tracing";
+static char tracing_path[PATH_MAX] = "/sys/kernel/debug/tracing";
char tracing_events_path[PATH_MAX] = "/sys/kernel/debug/tracing/events";
@@ -75,7 +75,7 @@ char *get_tracing_file(const char *name)
{
char *file;
- if (asprintf(&file, "%s/%s", tracing_path, name) < 0)
+ if (asprintf(&file, "%s/%s", tracing_path_mount(), name) < 0)
return NULL;
return file;
diff --git a/tools/lib/api/fs/tracing_path.h b/tools/lib/api/fs/tracing_path.h
index 0066f06cc381..1b65decedfc0 100644
--- a/tools/lib/api/fs/tracing_path.h
+++ b/tools/lib/api/fs/tracing_path.h
@@ -4,7 +4,6 @@
#include <linux/types.h>
-extern char tracing_path[];
extern char tracing_events_path[];
void tracing_path_set(const char *mountpoint);
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index cd6ea55d4b0c..d5a0878de816 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -238,7 +238,7 @@ static int handle_options(const char ***argv, int *argc, int *envchanged)
(*argc)--;
} else if (strstarts(cmd, CMD_DEBUGFS_DIR)) {
tracing_path_set(cmd + strlen(CMD_DEBUGFS_DIR));
- fprintf(stderr, "dir: %s\n", tracing_path);
+ fprintf(stderr, "dir: %s\n", tracing_path_mount());
if (envchanged)
*envchanged = 1;
} else if (!strcmp(cmd, "--list-cmds")) {
@@ -463,9 +463,6 @@ int main(int argc, const char **argv)
return err;
set_buildid_dir(NULL);
- /* get debugfs/tracefs mount point from /proc/mounts */
- tracing_path_mount();
-
/*
* "perf-xxxx" is the same as "perf xxxx", but we obviously:
*
diff --git a/tools/perf/util/probe-file.c b/tools/perf/util/probe-file.c
index 4ae1123c6794..b76088fadf3d 100644
--- a/tools/perf/util/probe-file.c
+++ b/tools/perf/util/probe-file.c
@@ -84,8 +84,7 @@ int open_trace_file(const char *trace_file, bool readwrite)
char buf[PATH_MAX];
int ret;
- ret = e_snprintf(buf, PATH_MAX, "%s/%s",
- tracing_path, trace_file);
+ ret = e_snprintf(buf, PATH_MAX, "%s/%s", tracing_path_mount(), trace_file);
if (ret >= 0) {
pr_debug("Opening %s write=%d\n", buf, readwrite);
if (readwrite && !probe_event_dry_run)
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 04/18] tools lib api fs tracing_path: Introduce get/put_events_file() helpers
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (3 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
To make reading events files a tad more compact than with
get_tracing_files("events/foo").
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-do6xgtwpmfl8zjs1euxsd2du@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/lib/api/fs/tracing_path.c | 15 +++++++++++++++
tools/lib/api/fs/tracing_path.h | 5 +++++
tools/perf/util/trace-event-info.c | 11 +++++------
3 files changed, 25 insertions(+), 6 deletions(-)
diff --git a/tools/lib/api/fs/tracing_path.c b/tools/lib/api/fs/tracing_path.c
index 6f5fe942eff4..9cd282425929 100644
--- a/tools/lib/api/fs/tracing_path.c
+++ b/tools/lib/api/fs/tracing_path.c
@@ -86,6 +86,21 @@ void put_tracing_file(char *file)
free(file);
}
+char *get_events_file(const char *name)
+{
+ char *file;
+
+ if (asprintf(&file, "%s/events/%s", tracing_path_mount(), name) < 0)
+ return NULL;
+
+ return file;
+}
+
+void put_events_file(char *file)
+{
+ free(file);
+}
+
int tracing_path__strerror_open_tp(int err, char *buf, size_t size,
const char *sys, const char *name)
{
diff --git a/tools/lib/api/fs/tracing_path.h b/tools/lib/api/fs/tracing_path.h
index 1b65decedfc0..3b32fb439f12 100644
--- a/tools/lib/api/fs/tracing_path.h
+++ b/tools/lib/api/fs/tracing_path.h
@@ -12,5 +12,10 @@ const char *tracing_path_mount(void);
char *get_tracing_file(const char *name);
void put_tracing_file(char *file);
+char *get_events_file(const char *name);
+void put_events_file(char *file);
+
+#define zput_events_file(ptr) ({ free(*ptr); *ptr = NULL; })
+
int tracing_path__strerror_open_tp(int err, char *buf, size_t size, const char *sys, const char *name);
#endif /* __API_FS_TRACING_PATH_H */
diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
index d7f2113462fb..c85d0d1a65ed 100644
--- a/tools/perf/util/trace-event-info.c
+++ b/tools/perf/util/trace-event-info.c
@@ -103,11 +103,10 @@ static int record_file(const char *file, ssize_t hdr_sz)
static int record_header_files(void)
{
- char *path;
+ char *path = get_events_file("header_page");
struct stat st;
int err = -EIO;
- path = get_tracing_file("events/header_page");
if (!path) {
pr_debug("can't get tracing/events/header_page");
return -ENOMEM;
@@ -128,9 +127,9 @@ static int record_header_files(void)
goto out;
}
- put_tracing_file(path);
+ put_events_file(path);
- path = get_tracing_file("events/header_event");
+ path = get_events_file("header_event");
if (!path) {
pr_debug("can't get tracing/events/header_event");
err = -ENOMEM;
@@ -154,7 +153,7 @@ static int record_header_files(void)
err = 0;
out:
- put_tracing_file(path);
+ put_events_file(path);
return err;
}
@@ -243,7 +242,7 @@ static int record_ftrace_files(struct tracepoint_path *tps)
char *path;
int ret;
- path = get_tracing_file("events/ftrace");
+ path = get_events_file("ftrace");
if (!path) {
pr_debug("can't get tracing/events/ftrace");
return -ENOMEM;
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 05/18] perf tools: Reuse the path to the tracepoint /events/ directory
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (4 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
When using for_each_event() we needlessly rebuild the whole path to
the tracepoint directory, reuse the dir_path instead, saving some cycles
and reducing the size of the next patch.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-54bcs15n0cp6gwcgpc4hptyc@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/parse-events.c | 15 +++++++--------
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 2fc4ee8b86c1..f9d5bbd63484 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -156,13 +156,12 @@ struct event_symbol event_symbols_sw[PERF_COUNT_SW_MAX] = {
(strcmp(sys_dirent->d_name, ".")) && \
(strcmp(sys_dirent->d_name, "..")))
-static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
+static int tp_event_has_id(const char *dir_path, struct dirent *evt_dir)
{
char evt_path[MAXPATHLEN];
int fd;
- snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", tracing_events_path,
- sys_dir->d_name, evt_dir->d_name);
+ snprintf(evt_path, MAXPATHLEN, "%s/%s/id", dir_path, evt_dir->d_name);
fd = open(evt_path, O_RDONLY);
if (fd < 0)
return -EINVAL;
@@ -171,12 +170,12 @@ static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
return 0;
}
-#define for_each_event(sys_dirent, evt_dir, evt_dirent) \
+#define for_each_event(dir_path, evt_dir, evt_dirent) \
while ((evt_dirent = readdir(evt_dir)) != NULL) \
if (evt_dirent->d_type == DT_DIR && \
(strcmp(evt_dirent->d_name, ".")) && \
(strcmp(evt_dirent->d_name, "..")) && \
- (!tp_event_has_id(sys_dirent, evt_dirent)))
+ (!tp_event_has_id(dir_path, evt_dirent)))
#define MAX_EVENT_LENGTH 512
@@ -204,7 +203,7 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
if (!evt_dir)
continue;
- for_each_event(sys_dirent, evt_dir, evt_dirent) {
+ for_each_event(dir_path, evt_dir, evt_dirent) {
scnprintf(evt_path, MAXPATHLEN, "%s/%s/id", dir_path,
evt_dirent->d_name);
@@ -2119,7 +2118,7 @@ void print_tracepoint_events(const char *subsys_glob, const char *event_glob,
if (!evt_dir)
continue;
- for_each_event(sys_dirent, evt_dir, evt_dirent) {
+ for_each_event(dir_path, evt_dir, evt_dirent) {
if (event_glob != NULL &&
!strglobmatch(evt_dirent->d_name, event_glob))
continue;
@@ -2199,7 +2198,7 @@ int is_valid_tracepoint(const char *event_string)
if (!evt_dir)
continue;
- for_each_event(sys_dirent, evt_dir, evt_dirent) {
+ for_each_event(dir_path, evt_dir, evt_dirent) {
snprintf(evt_path, MAXPATHLEN, "%s:%s",
sys_dirent->d_name, evt_dirent->d_name);
if (!strcmp(evt_path, event_string)) {
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 06/18] perf parse-events: Use get/put_events_file()
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (5 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Instead of accessing the trace_events_path variable directly, that may
not have been properly initialized wrt detecting where tracefs is
mounted.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-id7hzn1ydgkxbumeve5wapqz@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/tests/parse-events.c | 7 +++---
tools/perf/util/parse-events.c | 50 +++++++++++++++++++++++++++--------------
tools/perf/util/trace-event.c | 8 +++++--
3 files changed, 43 insertions(+), 22 deletions(-)
diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c
index 6829dd416a99..6d57d7082637 100644
--- a/tools/perf/tests/parse-events.c
+++ b/tools/perf/tests/parse-events.c
@@ -1328,7 +1328,7 @@ static int count_tracepoints(void)
TEST_ASSERT_VAL("Can't open events dir", events_dir);
while ((events_ent = readdir(events_dir))) {
- char sys_path[PATH_MAX];
+ char *sys_path;
struct dirent *sys_ent;
DIR *sys_dir;
@@ -1339,8 +1339,8 @@ static int count_tracepoints(void)
|| !strcmp(events_ent->d_name, "header_page"))
continue;
- scnprintf(sys_path, PATH_MAX, "%s/%s",
- tracing_events_path, events_ent->d_name);
+ sys_path = get_events_file(events_ent->d_name);
+ TEST_ASSERT_VAL("Can't get sys path", sys_path);
sys_dir = opendir(sys_path);
TEST_ASSERT_VAL("Can't open sys dir", sys_dir);
@@ -1356,6 +1356,7 @@ static int count_tracepoints(void)
}
closedir(sys_dir);
+ put_events_file(sys_path);
}
closedir(events_dir);
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index f9d5bbd63484..24668300b327 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -189,19 +189,19 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
int fd;
u64 id;
char evt_path[MAXPATHLEN];
- char dir_path[MAXPATHLEN];
+ char *dir_path;
sys_dir = opendir(tracing_events_path);
if (!sys_dir)
return NULL;
for_each_subsystem(sys_dir, sys_dirent) {
-
- snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path,
- sys_dirent->d_name);
+ dir_path = get_events_file(sys_dirent->d_name);
+ if (!dir_path)
+ continue;
evt_dir = opendir(dir_path);
if (!evt_dir)
- continue;
+ goto next;
for_each_event(dir_path, evt_dir, evt_dirent) {
@@ -217,6 +217,7 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
close(fd);
id = atoll(id_buf);
if (id == config) {
+ put_events_file(dir_path);
closedir(evt_dir);
closedir(sys_dir);
path = zalloc(sizeof(*path));
@@ -241,6 +242,8 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
}
}
closedir(evt_dir);
+next:
+ put_events_file(dir_path);
}
closedir(sys_dir);
@@ -511,14 +514,19 @@ static int add_tracepoint_multi_event(struct list_head *list, int *idx,
struct parse_events_error *err,
struct list_head *head_config)
{
- char evt_path[MAXPATHLEN];
+ char *evt_path;
struct dirent *evt_ent;
DIR *evt_dir;
int ret = 0, found = 0;
- snprintf(evt_path, MAXPATHLEN, "%s/%s", tracing_events_path, sys_name);
+ evt_path = get_events_file(sys_name);
+ if (!evt_path) {
+ tracepoint_error(err, errno, sys_name, evt_name);
+ return -1;
+ }
evt_dir = opendir(evt_path);
if (!evt_dir) {
+ put_events_file(evt_path);
tracepoint_error(err, errno, sys_name, evt_name);
return -1;
}
@@ -544,6 +552,7 @@ static int add_tracepoint_multi_event(struct list_head *list, int *idx,
ret = -1;
}
+ put_events_file(evt_path);
closedir(evt_dir);
return ret;
}
@@ -2091,7 +2100,7 @@ void print_tracepoint_events(const char *subsys_glob, const char *event_glob,
DIR *sys_dir, *evt_dir;
struct dirent *sys_dirent, *evt_dirent;
char evt_path[MAXPATHLEN];
- char dir_path[MAXPATHLEN];
+ char *dir_path;
char **evt_list = NULL;
unsigned int evt_i = 0, evt_num = 0;
bool evt_num_known = false;
@@ -2112,11 +2121,12 @@ void print_tracepoint_events(const char *subsys_glob, const char *event_glob,
!strglobmatch(sys_dirent->d_name, subsys_glob))
continue;
- snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path,
- sys_dirent->d_name);
+ dir_path = get_events_file(sys_dirent->d_name);
+ if (!dir_path)
+ continue;
evt_dir = opendir(dir_path);
if (!evt_dir)
- continue;
+ goto next;
for_each_event(dir_path, evt_dir, evt_dirent) {
if (event_glob != NULL &&
@@ -2132,11 +2142,15 @@ void print_tracepoint_events(const char *subsys_glob, const char *event_glob,
sys_dirent->d_name, evt_dirent->d_name);
evt_list[evt_i] = strdup(evt_path);
- if (evt_list[evt_i] == NULL)
+ if (evt_list[evt_i] == NULL) {
+ put_events_file(dir_path);
goto out_close_evt_dir;
+ }
evt_i++;
}
closedir(evt_dir);
+next:
+ put_events_file(dir_path);
}
closedir(sys_dir);
@@ -2184,19 +2198,19 @@ int is_valid_tracepoint(const char *event_string)
DIR *sys_dir, *evt_dir;
struct dirent *sys_dirent, *evt_dirent;
char evt_path[MAXPATHLEN];
- char dir_path[MAXPATHLEN];
+ char *dir_path;
sys_dir = opendir(tracing_events_path);
if (!sys_dir)
return 0;
for_each_subsystem(sys_dir, sys_dirent) {
-
- snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path,
- sys_dirent->d_name);
+ dir_path = get_events_file(sys_dirent->d_name);
+ if (!dir_path)
+ continue;
evt_dir = opendir(dir_path);
if (!evt_dir)
- continue;
+ goto next;
for_each_event(dir_path, evt_dir, evt_dirent) {
snprintf(evt_path, MAXPATHLEN, "%s:%s",
@@ -2208,6 +2222,8 @@ int is_valid_tracepoint(const char *event_string)
}
}
closedir(evt_dir);
+next:
+ put_events_file(dir_path);
}
closedir(sys_dir);
return 0;
diff --git a/tools/perf/util/trace-event.c b/tools/perf/util/trace-event.c
index 16a776371d03..1aa368603268 100644
--- a/tools/perf/util/trace-event.c
+++ b/tools/perf/util/trace-event.c
@@ -75,6 +75,7 @@ void trace_event__cleanup(struct trace_event *t)
static struct event_format*
tp_format(const char *sys, const char *name)
{
+ char *tp_dir = get_events_file(sys);
struct pevent *pevent = tevent.pevent;
struct event_format *event = NULL;
char path[PATH_MAX];
@@ -82,8 +83,11 @@ tp_format(const char *sys, const char *name)
char *data;
int err;
- scnprintf(path, PATH_MAX, "%s/%s/%s/format",
- tracing_events_path, sys, name);
+ if (!tp_dir)
+ return ERR_PTR(-errno);
+
+ scnprintf(path, PATH_MAX, "%s/%s/format", tp_dir, name);
+ put_events_file(tp_dir);
err = filename__read_str(path, &data, &size);
if (err)
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 07/18] tools lib api fs tracing_path: Introduce opendir() method
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (6 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
That takes care of using the right call to get the tracing_path
directory, the one that will end up calling tracing_path_set() to figure
out where tracefs is mounted.
One more step in doing just lazy reading of system structures to reduce
the number of operations done unconditionaly at 'perf' start.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-42zzi0f274909bg9mxzl81bu@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/lib/api/fs/tracing_path.c | 13 +++++++++++++
tools/lib/api/fs/tracing_path.h | 3 +++
tools/perf/tests/parse-events.c | 2 +-
tools/perf/util/parse-events.c | 8 ++++----
4 files changed, 21 insertions(+), 5 deletions(-)
diff --git a/tools/lib/api/fs/tracing_path.c b/tools/lib/api/fs/tracing_path.c
index 9cd282425929..9b451af0721c 100644
--- a/tools/lib/api/fs/tracing_path.c
+++ b/tools/lib/api/fs/tracing_path.c
@@ -101,6 +101,19 @@ void put_events_file(char *file)
free(file);
}
+DIR *tracing_events__opendir(void)
+{
+ DIR *dir = NULL;
+ char *path = get_tracing_file("events");
+
+ if (path) {
+ dir = opendir(path);
+ put_events_file(path);
+ }
+
+ return dir;
+}
+
int tracing_path__strerror_open_tp(int err, char *buf, size_t size,
const char *sys, const char *name)
{
diff --git a/tools/lib/api/fs/tracing_path.h b/tools/lib/api/fs/tracing_path.h
index 3b32fb439f12..904d085b2ae7 100644
--- a/tools/lib/api/fs/tracing_path.h
+++ b/tools/lib/api/fs/tracing_path.h
@@ -3,9 +3,12 @@
#define __API_FS_TRACING_PATH_H
#include <linux/types.h>
+#include <dirent.h>
extern char tracing_events_path[];
+DIR *tracing_events__opendir(void);
+
void tracing_path_set(const char *mountpoint);
const char *tracing_path_mount(void);
diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c
index 6d57d7082637..b9ebe15afb13 100644
--- a/tools/perf/tests/parse-events.c
+++ b/tools/perf/tests/parse-events.c
@@ -1323,7 +1323,7 @@ static int count_tracepoints(void)
DIR *events_dir;
int cnt = 0;
- events_dir = opendir(tracing_events_path);
+ events_dir = tracing_events__opendir();
TEST_ASSERT_VAL("Can't open events dir", events_dir);
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 24668300b327..15eec49e71a1 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -191,7 +191,7 @@ struct tracepoint_path *tracepoint_id_to_path(u64 config)
char evt_path[MAXPATHLEN];
char *dir_path;
- sys_dir = opendir(tracing_events_path);
+ sys_dir = tracing_events__opendir();
if (!sys_dir)
return NULL;
@@ -578,7 +578,7 @@ static int add_tracepoint_multi_sys(struct list_head *list, int *idx,
DIR *events_dir;
int ret = 0;
- events_dir = opendir(tracing_events_path);
+ events_dir = tracing_events__opendir();
if (!events_dir) {
tracepoint_error(err, errno, sys_name, evt_name);
return -1;
@@ -2106,7 +2106,7 @@ void print_tracepoint_events(const char *subsys_glob, const char *event_glob,
bool evt_num_known = false;
restart:
- sys_dir = opendir(tracing_events_path);
+ sys_dir = tracing_events__opendir();
if (!sys_dir)
return;
@@ -2200,7 +2200,7 @@ int is_valid_tracepoint(const char *event_string)
char evt_path[MAXPATHLEN];
char *dir_path;
- sys_dir = opendir(tracing_events_path);
+ sys_dir = tracing_events__opendir();
if (!sys_dir)
return 0;
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 08/18] tools lib api fs tracing_path: Make tracing_events_path private
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (7 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Not anymore accessed outside this library, keep it private.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-wg1m07flfrg1rm06jjzie8si@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/lib/api/fs/tracing_path.c | 3 +--
tools/lib/api/fs/tracing_path.h | 2 --
2 files changed, 1 insertion(+), 4 deletions(-)
diff --git a/tools/lib/api/fs/tracing_path.c b/tools/lib/api/fs/tracing_path.c
index 9b451af0721c..120037496f77 100644
--- a/tools/lib/api/fs/tracing_path.c
+++ b/tools/lib/api/fs/tracing_path.c
@@ -15,8 +15,7 @@
static char tracing_mnt[PATH_MAX] = "/sys/kernel/debug";
static char tracing_path[PATH_MAX] = "/sys/kernel/debug/tracing";
-char tracing_events_path[PATH_MAX] = "/sys/kernel/debug/tracing/events";
-
+static char tracing_events_path[PATH_MAX] = "/sys/kernel/debug/tracing/events";
static void __tracing_path_set(const char *tracing, const char *mountpoint)
{
diff --git a/tools/lib/api/fs/tracing_path.h b/tools/lib/api/fs/tracing_path.h
index 904d085b2ae7..a19136b086dc 100644
--- a/tools/lib/api/fs/tracing_path.h
+++ b/tools/lib/api/fs/tracing_path.h
@@ -5,8 +5,6 @@
#include <linux/types.h>
#include <dirent.h>
-extern char tracing_events_path[];
-
DIR *tracing_events__opendir(void);
void tracing_path_set(const char *mountpoint);
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 09/18] tools include compiler-gcc: Add __pure attribute helper
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (8 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Adopt it from the kernel sources, will be used soon.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-oubheiqj8edo5rzewt11cbn0@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/include/linux/compiler-gcc.h | 3 +++
1 file changed, 3 insertions(+)
diff --git a/tools/include/linux/compiler-gcc.h b/tools/include/linux/compiler-gcc.h
index a3a4427441bf..70fe61295733 100644
--- a/tools/include/linux/compiler-gcc.h
+++ b/tools/include/linux/compiler-gcc.h
@@ -21,6 +21,9 @@
/* &a[0] degrades to a pointer: a different type from an array */
#define __must_be_array(a) BUILD_BUG_ON_ZERO(__same_type((a), &(a)[0]))
+#ifndef __pure
+#define __pure __attribute__((pure))
+#endif
#define noinline __attribute__((noinline))
#ifndef __packed
#define __packed __attribute__((packed))
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 10/18] perf tools: Read the cache line size lazily
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (9 preceding siblings ...)
(?)
@ 2018-05-19 10:54 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:54 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
It is not read as commonly as 'page_size', so it makes sense to read it
lazily, caching its value when it is first read.
Less files open unconditionally at startup.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-35xhrq91u94uc1djtclek1ie@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/perf.c | 11 -----------
tools/perf/util/sort.c | 4 ++--
tools/perf/util/sort.h | 4 ++--
tools/perf/util/util.c | 21 ++++++++++++++++++++-
tools/perf/util/util.h | 2 +-
5 files changed, 25 insertions(+), 17 deletions(-)
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index d5a0878de816..cefd8f74630c 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -421,16 +421,6 @@ void pthread__unblock_sigwinch(void)
pthread_sigmask(SIG_UNBLOCK, &set, NULL);
}
-#ifdef _SC_LEVEL1_DCACHE_LINESIZE
-#define cache_line_size(cacheline_sizep) *cacheline_sizep = sysconf(_SC_LEVEL1_DCACHE_LINESIZE)
-#else
-static void cache_line_size(int *cacheline_sizep)
-{
- if (sysfs__read_int("devices/system/cpu/cpu0/cache/index0/coherency_line_size", cacheline_sizep))
- pr_debug("cannot determine cache line size");
-}
-#endif
-
int main(int argc, const char **argv)
{
int err;
@@ -444,7 +434,6 @@ int main(int argc, const char **argv)
/* The page_size is placed in util object. */
page_size = sysconf(_SC_PAGE_SIZE);
- cache_line_size(&cacheline_size);
if (sysctl__read_int("kernel/perf_event_max_stack", &value) == 0)
sysctl_perf_event_max_stack = value;
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
index e65903a695a6..4058ade352a5 100644
--- a/tools/perf/util/sort.c
+++ b/tools/perf/util/sort.c
@@ -2582,7 +2582,7 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok,
if (sort__mode != SORT_MODE__MEMORY)
return -EINVAL;
- if (sd->entry == &sort_mem_dcacheline && cacheline_size == 0)
+ if (sd->entry == &sort_mem_dcacheline && cacheline_size() == 0)
return -EINVAL;
if (sd->entry == &sort_mem_daddr_sym)
@@ -2628,7 +2628,7 @@ static int setup_sort_list(struct perf_hpp_list *list, char *str,
if (*tok) {
ret = sort_dimension__add(list, tok, evlist, level);
if (ret == -EINVAL) {
- if (!cacheline_size && !strncasecmp(tok, "dcacheline", strlen(tok)))
+ if (!cacheline_size() && !strncasecmp(tok, "dcacheline", strlen(tok)))
pr_err("The \"dcacheline\" --sort key needs to know the cacheline size and it couldn't be determined on this system");
else
pr_err("Invalid --sort key: `%s'", tok);
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index 035b62e2c60b..9e6896293bbd 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -186,13 +186,13 @@ static inline float hist_entry__get_percent_limit(struct hist_entry *he)
static inline u64 cl_address(u64 address)
{
/* return the cacheline of the address */
- return (address & ~(cacheline_size - 1));
+ return (address & ~(cacheline_size() - 1));
}
static inline u64 cl_offset(u64 address)
{
/* return the cacheline of the address */
- return (address & (cacheline_size - 1));
+ return (address & (cacheline_size() - 1));
}
enum sort_mode {
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index 1019bbc5dbd8..99ab52165680 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -38,7 +38,26 @@ void perf_set_multithreaded(void)
}
unsigned int page_size;
-int cacheline_size;
+
+#ifdef _SC_LEVEL1_DCACHE_LINESIZE
+#define cache_line_size(cacheline_sizep) *cacheline_sizep = sysconf(_SC_LEVEL1_DCACHE_LINESIZE)
+#else
+static void cache_line_size(int *cacheline_sizep)
+{
+ if (sysfs__read_int("devices/system/cpu/cpu0/cache/index0/coherency_line_size", cacheline_sizep))
+ pr_debug("cannot determine cache line size");
+}
+#endif
+
+int cacheline_size(void)
+{
+ static int size;
+
+ if (!size)
+ cache_line_size(&size);
+
+ return size;
+}
int sysctl_perf_event_max_stack = PERF_MAX_STACK_DEPTH;
int sysctl_perf_event_max_contexts_per_stack = PERF_MAX_CONTEXTS_PER_STACK;
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index c9626c206208..74d21dfe0d29 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -43,7 +43,7 @@ size_t hex_width(u64 v);
int hex2u64(const char *ptr, u64 *val);
extern unsigned int page_size;
-extern int cacheline_size;
+int __pure cacheline_size(void);
int fetch_kernel_version(unsigned int *puint,
char *str, size_t str_sz);
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 11/18] perf tools: No need to unconditionally read the max_stack sysctls
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (10 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Let tools that need to have those variables with the sysctl current
values use a function that will read them.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-1ljj3oeo5kpt2n1icfd9vowe@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/builtin-top.c | 2 +-
tools/perf/builtin-trace.c | 2 +-
tools/perf/perf.c | 7 -------
tools/perf/util/evsel.c | 2 +-
tools/perf/util/util.c | 13 +++++++++++++
tools/perf/util/util.h | 2 ++
6 files changed, 18 insertions(+), 10 deletions(-)
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 3c061c57afb6..7a349fcd3864 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -1264,7 +1264,7 @@ int cmd_top(int argc, const char **argv)
.proc_map_timeout = 500,
.overwrite = 1,
},
- .max_stack = sysctl_perf_event_max_stack,
+ .max_stack = sysctl__max_stack(),
.sym_pcnt_filter = 5,
.nr_threads_synthesize = UINT_MAX,
};
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index c7effcfc40ed..560aed7da36a 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -3162,7 +3162,7 @@ int cmd_trace(int argc, const char **argv)
mmap_pages_user_set = false;
if (trace.max_stack == UINT_MAX) {
- trace.max_stack = input_name ? PERF_MAX_STACK_DEPTH : sysctl_perf_event_max_stack;
+ trace.max_stack = input_name ? PERF_MAX_STACK_DEPTH : sysctl__max_stack();
max_stack_user_set = false;
}
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index cefd8f74630c..51c81509a315 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -426,7 +426,6 @@ int main(int argc, const char **argv)
int err;
const char *cmd;
char sbuf[STRERR_BUFSIZE];
- int value;
/* libsubcmd init */
exec_cmd_init("perf", PREFIX, PERF_EXEC_PATH, EXEC_PATH_ENVIRONMENT);
@@ -435,12 +434,6 @@ int main(int argc, const char **argv)
/* The page_size is placed in util object. */
page_size = sysconf(_SC_PAGE_SIZE);
- if (sysctl__read_int("kernel/perf_event_max_stack", &value) == 0)
- sysctl_perf_event_max_stack = value;
-
- if (sysctl__read_int("kernel/perf_event_max_contexts_per_stack", &value) == 0)
- sysctl_perf_event_max_contexts_per_stack = value;
-
cmd = extract_argv0_path(argv[0]);
if (!cmd)
cmd = "perf-help";
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 4cd2cf93f726..150db5ed7400 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -2862,7 +2862,7 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
return scnprintf(msg, size,
"Not enough memory to setup event with callchain.\n"
"Hint: Try tweaking /proc/sys/kernel/perf_event_max_stack\n"
- "Hint: Current value: %d", sysctl_perf_event_max_stack);
+ "Hint: Current value: %d", sysctl__max_stack());
break;
case ENODEV:
if (target->cpu_list)
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index 99ab52165680..eac5b858a371 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -62,6 +62,19 @@ int cacheline_size(void)
int sysctl_perf_event_max_stack = PERF_MAX_STACK_DEPTH;
int sysctl_perf_event_max_contexts_per_stack = PERF_MAX_CONTEXTS_PER_STACK;
+int sysctl__max_stack(void)
+{
+ int value;
+
+ if (sysctl__read_int("kernel/perf_event_max_stack", &value) == 0)
+ sysctl_perf_event_max_stack = value;
+
+ if (sysctl__read_int("kernel/perf_event_max_contexts_per_stack", &value) == 0)
+ sysctl_perf_event_max_contexts_per_stack = value;
+
+ return sysctl_perf_event_max_stack;
+}
+
bool test_attr__enabled;
bool perf_host = true;
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 74d21dfe0d29..dc58254a2b69 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -45,6 +45,8 @@ int hex2u64(const char *ptr, u64 *val);
extern unsigned int page_size;
int __pure cacheline_size(void);
+int sysctl__max_stack(void);
+
int fetch_kernel_version(unsigned int *puint,
char *str, size_t str_sz);
#define KVER_VERSION(x) (((x) >> 16) & 0xff)
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 12/18] perf script: Show virtual addresses instead of offsets
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (11 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users, Sandipan Das,
Jiri Olsa, Naveen N . Rao, Ravi Bangoria,
Arnaldo Carvalho de Melo
From: Sandipan Das <sandipan@linux.vnet.ibm.com>
When perf data is recorded with the call-graph option enabled, the
callchain shown by perf script shows the binary offsets of the symbols
as the ip. This is incorrect for kernel symbols as the ip values are
always off by a fixed offset depending on the architecture. If the
offsets from the start of the symbols are printed, they are also
incorrect for both kernel and userspace symbols.
Without the call-graph option, the callchain shows the virtual addresses
of the symbols rather than their binary offsets. The offsets printed in
this case are also correct.
This fixes the inconsistency in perf script's output.
This can be verified on a powerpc64le system running Fedora 27 as
follows:
# cat /proc/kallsyms | grep sys_write
...
c0000000004025a0 T sys_write
c0000000004025a0 T __se_sys_write
...
# perf probe -a sys_write
Before applying this patch:
# perf record -e probe:sys_write -g ~/test
# perf script -F ip,sym,symoff
4125b0 sys_write+0x8000000000008010
1b9e0 system_call+0x8000000000008058
118234 __GI___libc_write+0xffff0000f52c0024
92c74 _IO_file_write@@GLIBC_2.17+0xffff0000f52c0044
5afbfd8a [unknown]
91a60 new_do_write+0xffff0000f52c0090
94638 _IO_do_write@@GLIBC_2.17+0xffff0000f52c0038
94bbc _IO_file_overflow@@GLIBC_2.17+0xffff0000f52c014c
95a24 __overflow+0xffff0000f52c0064
84548 _IO_puts+0xffff0000f52c0218
440 main+0xffffffffe0000020
236a0 generic_start_main.isra.0+0xffff0000f52c0140
23898 __libc_start_main+0xffff0000f52c00b8
0 [unknown]
...
# perf record -e probe:sys_write ~/test
# perf script -F ip,sym,symoff
c0000000004025b0 sys_write+0x10
...
After applying this patch:
# perf record -e probe:sys_write -g ~/test
# perf script -F ip,sym,symoff
c0000000004025b0 sys_write+0x10
c00000000000b9e0 system_call+0x58
7fffb70d8234 __GI___libc_write+0x24
7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44
5afc1818 [unknown]
7fffb7051a60 new_do_write+0x90
7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38
7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c
7fffb7055a24 __overflow+0x64
7fffb7044548 _IO_puts+0x218
10000440 main+0x20
7fffb6fe36a0 generic_start_main.isra.0+0x140
7fffb6fe3898 __libc_start_main+0xb8
0 [unknown]
...
# perf record -e probe:sys_write ~/test
# perf script -F ip,sym,symoff
c0000000004025b0 sys_write+0x10
...
Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Link: http://lkml.kernel.org/r/20180517063326.6319-1-sandipan@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/machine.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 72a351613d85..7c777cb32806 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1764,7 +1764,7 @@ static int add_callchain_ip(struct thread *thread,
}
srcline = callchain_srcline(al.map, al.sym, al.addr);
- return callchain_cursor_append(cursor, al.addr, al.map, al.sym,
+ return callchain_cursor_append(cursor, ip, al.map, al.sym,
branch, flags, nr_loop_iter,
iter_cycles, branch_from, srcline);
}
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 13/18] perf script: Show symbol offsets by default
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (12 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users, Sandipan Das,
Jiri Olsa, Naveen N . Rao, Ravi Bangoria,
Arnaldo Carvalho de Melo
From: Sandipan Das <sandipan@linux.vnet.ibm.com>
Since the ip shown for a symbol is now always a virtual address, it
becomes difficult to correlate this with objdump output and determine
the exact instruction address. So, we always show the offset from the
start of the symbol.
This can be verified on a powerpc64le system running Fedora 27 as
follows:
# perf probe -a sys_write
# perf record -e probe:sys_write -g ~/test
Before applying this patch:
# perf script
test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0)
c0000000004025b0 sys_write (/lib/modules/4.17.0-rc4+/build/vmlinux)
c00000000000b9e0 system_call (/lib/modules/4.17.0-rc4+/build/vmlinux)
7fffb70d8234 __GI___libc_write (/usr/lib64/libc-2.26.so)
7fffb7052c74 _IO_file_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so)
5afc1818 [unknown] ([unknown])
7fffb7051a60 new_do_write (/usr/lib64/libc-2.26.so)
7fffb7054638 _IO_do_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so)
7fffb7054bbc _IO_file_overflow@@GLIBC_2.17 (/usr/lib64/libc-2.26.so)
7fffb7055a24 __overflow (/usr/lib64/libc-2.26.so)
7fffb7044548 _IO_puts (/usr/lib64/libc-2.26.so)
10000440 main (/home/sandipan/test)
7fffb6fe36a0 generic_start_main.isra.0 (/usr/lib64/libc-2.26.so)
7fffb6fe3898 __libc_start_main (/usr/lib64/libc-2.26.so)
0 [unknown] ([unknown])
...
After applying this patch:
# perf script
test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0)
c0000000004025b0 sys_write+0x10 (/lib/modules/4.17.0-rc4+/build/vmlinux)
c00000000000b9e0 system_call+0x58 (/lib/modules/4.17.0-rc4+/build/vmlinux)
7fffb70d8234 __GI___libc_write+0x24 (/usr/lib64/libc-2.26.so)
7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44 (/usr/lib64/libc-2.26.so)
5afc1818 [unknown] ([unknown])
7fffb7051a60 new_do_write+0x90 (/usr/lib64/libc-2.26.so)
7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38 (/usr/lib64/libc-2.26.so)
7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c (/usr/lib64/libc-2.26.so)
7fffb7055a24 __overflow+0x64 (/usr/lib64/libc-2.26.so)
7fffb7044548 _IO_puts+0x218 (/usr/lib64/libc-2.26.so)
10000440 main+0x20 (/home/sandipan/test)
7fffb6fe36a0 generic_start_main.isra.0+0x140 (/usr/lib64/libc-2.26.so)
7fffb6fe3898 __libc_start_main+0xb8 (/usr/lib64/libc-2.26.so)
0 [unknown] ([unknown])
...
Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Link: http://lkml.kernel.org/r/20180517063326.6319-2-sandipan@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/builtin-script.c | 26 ++++++++++++----------
.../tests/shell/record+probe_libc_inet_pton.sh | 12 +++++-----
2 files changed, 20 insertions(+), 18 deletions(-)
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index fa2c7a288750..cefc8813e91e 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -153,8 +153,8 @@ static struct {
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
- PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
- PERF_OUTPUT_PERIOD,
+ PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
+ PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD,
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
},
@@ -165,8 +165,9 @@ static struct {
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
- PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
- PERF_OUTPUT_PERIOD | PERF_OUTPUT_BPF_OUTPUT,
+ PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
+ PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD |
+ PERF_OUTPUT_BPF_OUTPUT,
.invalid_fields = PERF_OUTPUT_TRACE,
},
@@ -185,10 +186,10 @@ static struct {
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
- PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
- PERF_OUTPUT_PERIOD | PERF_OUTPUT_ADDR |
- PERF_OUTPUT_DATA_SRC | PERF_OUTPUT_WEIGHT |
- PERF_OUTPUT_PHYS_ADDR,
+ PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
+ PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD |
+ PERF_OUTPUT_ADDR | PERF_OUTPUT_DATA_SRC |
+ PERF_OUTPUT_WEIGHT | PERF_OUTPUT_PHYS_ADDR,
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
},
@@ -199,8 +200,8 @@ static struct {
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
- PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
- PERF_OUTPUT_PERIOD,
+ PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
+ PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD,
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
},
@@ -211,8 +212,8 @@ static struct {
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
- PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
- PERF_OUTPUT_SYNTH,
+ PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
+ PERF_OUTPUT_DSO | PERF_OUTPUT_SYNTH,
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
},
@@ -544,6 +545,7 @@ static int perf_session__check_output_opt(struct perf_session *session)
if (attr->sample_type & PERF_SAMPLE_CALLCHAIN) {
output[j].fields |= PERF_OUTPUT_IP;
output[j].fields |= PERF_OUTPUT_SYM;
+ output[j].fields |= PERF_OUTPUT_SYMOFFSET;
output[j].fields |= PERF_OUTPUT_DSO;
set_print_ip_opts(attr);
goto out;
diff --git a/tools/perf/tests/shell/record+probe_libc_inet_pton.sh b/tools/perf/tests/shell/record+probe_libc_inet_pton.sh
index ee86473643be..650b208f700f 100755
--- a/tools/perf/tests/shell/record+probe_libc_inet_pton.sh
+++ b/tools/perf/tests/shell/record+probe_libc_inet_pton.sh
@@ -16,18 +16,18 @@ nm -g $libc 2>/dev/null | fgrep -q inet_pton || exit 254
trace_libc_inet_pton_backtrace() {
idx=0
expected[0]="ping[][0-9 \.:]+probe_libc:inet_pton: \([[:xdigit:]]+\)"
- expected[1]=".*inet_pton[[:space:]]\($libc|inlined\)$"
+ expected[1]=".*inet_pton\+0x[[:xdigit:]]+[[:space:]]\($libc|inlined\)$"
case "$(uname -m)" in
s390x)
eventattr='call-graph=dwarf,max-stack=4'
- expected[2]="gaih_inet.*[[:space:]]\($libc|inlined\)$"
- expected[3]="(__GI_)?getaddrinfo[[:space:]]\($libc|inlined\)$"
- expected[4]="main[[:space:]]\(.*/bin/ping.*\)$"
+ expected[2]="gaih_inet.*\+0x[[:xdigit:]]+[[:space:]]\($libc|inlined\)$"
+ expected[3]="(__GI_)?getaddrinfo\+0x[[:xdigit:]]+[[:space:]]\($libc|inlined\)$"
+ expected[4]="main\+0x[[:xdigit:]]+[[:space:]]\(.*/bin/ping.*\)$"
;;
*)
eventattr='max-stack=3'
- expected[2]="getaddrinfo[[:space:]]\($libc\)$"
- expected[3]=".*\(.*/bin/ping.*\)$"
+ expected[2]="getaddrinfo\+0x[[:xdigit:]]+[[:space:]]\($libc\)$"
+ expected[3]=".*\+0x[[:xdigit:]]+[[:space:]]\(.*/bin/ping.*\)$"
;;
esac
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 14/18] perf annotate: Record the min/max cycles
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (13 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users, Jin Yao,
Alexander Shishkin, Andi Kleen, Jiri Olsa, Kan Liang,
Peter Zijlstra, Arnaldo Carvalho de Melo
From: Jin Yao <yao.jin@linux.intel.com>
Currently perf has a feature to account cycles for LBRs
For example, on skylake:
perf record -b ...
perf report or perf annotate
And then browsing the annotate browser gives average cycle counts for
program blocks.
For some analysis it would be useful if we could know not only the
average cycles but also the min and max cycles.
This patch records the min and max cycles.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1526569118-14217-2-git-send-email-yao.jin@linux.intel.com
[ Switch from max/min to min/max ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/annotate.c | 14 +++++++++++++-
tools/perf/util/annotate.h | 4 ++++
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 5d74a30fe00f..4fcfefea3bc2 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -760,6 +760,15 @@ static int __symbol__account_cycles(struct annotation *notes,
ch[offset].num_aggr++;
ch[offset].cycles_aggr += cycles;
+ if (cycles > ch[offset].cycles_max)
+ ch[offset].cycles_max = cycles;
+
+ if (ch[offset].cycles_min) {
+ if (cycles && cycles < ch[offset].cycles_min)
+ ch[offset].cycles_min = cycles;
+ } else
+ ch[offset].cycles_min = cycles;
+
if (!have_start && ch[offset].have_start)
return 0;
if (ch[offset].num) {
@@ -953,8 +962,11 @@ void annotation__compute_ipc(struct annotation *notes, size_t size)
if (ch->have_start)
annotation__count_and_fill(notes, ch->start, offset, ch);
al = notes->offsets[offset];
- if (al && ch->num_aggr)
+ if (al && ch->num_aggr) {
al->cycles = ch->cycles_aggr / ch->num_aggr;
+ al->cycles_max = ch->cycles_max;
+ al->cycles_min = ch->cycles_min;
+ }
notes->have_cycles = true;
}
}
diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
index f28a9e43421d..d50363d56f73 100644
--- a/tools/perf/util/annotate.h
+++ b/tools/perf/util/annotate.h
@@ -105,6 +105,8 @@ struct annotation_line {
int jump_sources;
float ipc;
u64 cycles;
+ u64 cycles_max;
+ u64 cycles_min;
size_t privsize;
char *path;
u32 idx;
@@ -186,6 +188,8 @@ struct cyc_hist {
u64 start;
u64 cycles;
u64 cycles_aggr;
+ u64 cycles_max;
+ u64 cycles_min;
u32 num;
u32 num_aggr;
u8 have_start;
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 15/18] perf annotate: Create hotkey 'c' to show min/max cycles
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (14 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users, Jin Yao,
Alexander Shishkin, Andi Kleen, Jiri Olsa, Kan Liang,
Peter Zijlstra, Arnaldo Carvalho de Melo
From: Jin Yao <yao.jin@linux.intel.com>
In the 'perf annotate' view, a new hotkey 'c' is created for showing the
min/max cycles.
For example, when press 'c', the annotate view is:
Percent│ IPC Cycle(min/max)
│
│
│ Disassembly of section .text:
│
│ 000000000003aab0 <random@@GLIBC_2.2.5>:
8.22 │3.92 sub $0x18,%rsp
│3.92 mov $0x1,%esi
│3.92 xor %eax,%eax
│3.92 cmpl $0x0,argp_program_version_hook@@G
│3.92 1(2/1) ↓ je 20
│ lock cmpxchg %esi,__abort_msg@@GLIBC_P
│ ↓ jne 29
│ ↓ jmp 43
│1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+
8.93 │1.10 1(5/1) ↓ je 43
When press 'c' again, the annotate view is switched back:
Percent│ IPC Cycle
│
│
│ Disassembly of section .text:
│
│ 000000000003aab0 <random@@GLIBC_2.2.5>:
8.22 │3.92 sub $0x18,%rsp
│3.92 mov $0x1,%esi
│3.92 xor %eax,%eax
│3.92 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x
│3.92 1 ↓ je 20
│ lock cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0
│ ↓ jne 29
│ ↓ jmp 43
│1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0
8.93 │1.10 1 ↓ je 43
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1526569118-14217-3-git-send-email-yao.jin@linux.intel.com
[ Rename all maxmin to minmax ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/ui/browsers/annotate.c | 8 ++++++++
tools/perf/util/annotate.c | 37 +++++++++++++++++++++++++++++++------
tools/perf/util/annotate.h | 7 ++++++-
3 files changed, 45 insertions(+), 7 deletions(-)
diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
index 3781d74088a7..8be40fa903aa 100644
--- a/tools/perf/ui/browsers/annotate.c
+++ b/tools/perf/ui/browsers/annotate.c
@@ -695,6 +695,7 @@ static int annotate_browser__run(struct annotate_browser *browser,
"O Bump offset level (jump targets -> +call -> all -> cycle thru)\n"
"s Toggle source code view\n"
"t Circulate percent, total period, samples view\n"
+ "c Show min/max cycle\n"
"/ Search string\n"
"k Toggle line numbers\n"
"P Print to [symbol_name].annotation file.\n"
@@ -791,6 +792,13 @@ static int annotate_browser__run(struct annotate_browser *browser,
notes->options->show_total_period = true;
annotation__update_column_widths(notes);
continue;
+ case 'c':
+ if (notes->options->show_minmax_cycle)
+ notes->options->show_minmax_cycle = false;
+ else
+ notes->options->show_minmax_cycle = true;
+ annotation__update_column_widths(notes);
+ continue;
case K_LEFT:
case K_ESC:
case 'q':
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 4fcfefea3bc2..6612c7f90af4 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -2498,13 +2498,38 @@ static void __annotation_line__write(struct annotation_line *al, struct annotati
else
obj__printf(obj, "%*s ", ANNOTATION__IPC_WIDTH - 1, "IPC");
- if (al->cycles)
- obj__printf(obj, "%*" PRIu64 " ",
+ if (!notes->options->show_minmax_cycle) {
+ if (al->cycles)
+ obj__printf(obj, "%*" PRIu64 " ",
ANNOTATION__CYCLES_WIDTH - 1, al->cycles);
- else if (!show_title)
- obj__printf(obj, "%*s", ANNOTATION__CYCLES_WIDTH, " ");
- else
- obj__printf(obj, "%*s ", ANNOTATION__CYCLES_WIDTH - 1, "Cycle");
+ else if (!show_title)
+ obj__printf(obj, "%*s",
+ ANNOTATION__CYCLES_WIDTH, " ");
+ else
+ obj__printf(obj, "%*s ",
+ ANNOTATION__CYCLES_WIDTH - 1,
+ "Cycle");
+ } else {
+ if (al->cycles) {
+ char str[32];
+
+ scnprintf(str, sizeof(str),
+ "%" PRIu64 "(%" PRIu64 "/%" PRIu64 ")",
+ al->cycles, al->cycles_min,
+ al->cycles_max);
+
+ obj__printf(obj, "%*s ",
+ ANNOTATION__MINMAX_CYCLES_WIDTH - 1,
+ str);
+ } else if (!show_title)
+ obj__printf(obj, "%*s",
+ ANNOTATION__MINMAX_CYCLES_WIDTH,
+ " ");
+ else
+ obj__printf(obj, "%*s ",
+ ANNOTATION__MINMAX_CYCLES_WIDTH - 1,
+ "Cycle(min/max)");
+ }
}
obj__printf(obj, " ");
diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
index d50363d56f73..5080b6dd98b8 100644
--- a/tools/perf/util/annotate.h
+++ b/tools/perf/util/annotate.h
@@ -61,6 +61,7 @@ bool ins__is_fused(struct arch *arch, const char *ins1, const char *ins2);
#define ANNOTATION__IPC_WIDTH 6
#define ANNOTATION__CYCLES_WIDTH 6
+#define ANNOTATION__MINMAX_CYCLES_WIDTH 19
struct annotation_options {
bool hide_src_code,
@@ -69,7 +70,8 @@ struct annotation_options {
show_linenr,
show_nr_jumps,
show_nr_samples,
- show_total_period;
+ show_total_period,
+ show_minmax_cycle;
u8 offset_level;
};
@@ -243,6 +245,9 @@ struct annotation {
static inline int annotation__cycles_width(struct annotation *notes)
{
+ if (notes->have_cycles && notes->options->show_minmax_cycle)
+ return ANNOTATION__IPC_WIDTH + ANNOTATION__MINMAX_CYCLES_WIDTH;
+
return notes->have_cycles ? ANNOTATION__IPC_WIDTH + ANNOTATION__CYCLES_WIDTH : 0;
}
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 16/18] perf bpf: Fixup include and examples install messages
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (15 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users,
Arnaldo Carvalho de Melo, Adrian Hunter, David Ahern, Jiri Olsa,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Before:
INSTALL lib
install include/bpf/*.h '/home/acme/lib/include/perf/bpf'
INSTALL lib
install examples/bpf/*.c '/home/acme/lib/examples/perf/bpf'
After:
INSTALL lib
INSTALL include/bpf
INSTALL lib
INSTALL examples/bpf
Reported-by: Ingo Molnar <mingo@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: dd8e4ead6e98 ("perf bpf: Add bpf.h to be used in eBPF proggies")
Fixes: 8f12a2ff00e5 ("perf bpf: Add 'examples' directories")
Link: https://lkml.kernel.org/n/tip-icljqe87e8pak8mu6mkki9d4@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/Makefile.perf | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index c63a3971d719..ecc9fc952655 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -770,9 +770,11 @@ endif
ifndef NO_LIBBPF
$(call QUIET_INSTALL, lib) \
$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perf_include_instdir_SQ)/bpf'
+ $(call QUIET_INSTALL, include/bpf) \
$(INSTALL) include/bpf/*.h '$(DESTDIR_SQ)$(perf_include_instdir_SQ)/bpf'
$(call QUIET_INSTALL, lib) \
$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perf_examples_instdir_SQ)/bpf'
+ $(call QUIET_INSTALL, examples/bpf) \
$(INSTALL) examples/bpf/*.c '$(DESTDIR_SQ)$(perf_examples_instdir_SQ)/bpf'
endif
$(call QUIET_INSTALL, perf-archive) \
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 17/18] perf machine: Add machine__is() to identify machine arch
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (16 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users, Adrian Hunter,
Alexander Shishkin, Andi Kleen, Andy Lutomirski, Dave Hansen,
H . Peter Anvin, Joerg Roedel, Peter Zijlstra, Thomas Gleixner,
x86, Arnaldo Carvalho de Melo
From: Adrian Hunter <adrian.hunter@intel.com>
Add a function to identify the machine architecture.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Tested-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: x86@kernel.org
Link: http://lkml.kernel.org/r/1526548928-20790-6-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/env.c | 18 ++++++++++++++++++
tools/perf/util/env.h | 2 ++
tools/perf/util/machine.c | 9 +++++++++
tools/perf/util/machine.h | 2 ++
4 files changed, 31 insertions(+)
diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c
index 4c842762e3f2..319fb0a0d05e 100644
--- a/tools/perf/util/env.c
+++ b/tools/perf/util/env.c
@@ -93,6 +93,24 @@ int perf_env__read_cpu_topology_map(struct perf_env *env)
return 0;
}
+static int perf_env__read_arch(struct perf_env *env)
+{
+ struct utsname uts;
+
+ if (env->arch)
+ return 0;
+
+ if (!uname(&uts))
+ env->arch = strdup(uts.machine);
+
+ return env->arch ? 0 : -ENOMEM;
+}
+
+const char *perf_env__raw_arch(struct perf_env *env)
+{
+ return env && !perf_env__read_arch(env) ? env->arch : "unknown";
+}
+
void cpu_cache_level__free(struct cpu_cache_level *cache)
{
free(cache->type);
diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
index c4ef2e523367..62e193948608 100644
--- a/tools/perf/util/env.h
+++ b/tools/perf/util/env.h
@@ -76,4 +76,6 @@ int perf_env__read_cpu_topology_map(struct perf_env *env);
void cpu_cache_level__free(struct cpu_cache_level *cache);
const char *perf_env__arch(struct perf_env *env);
+const char *perf_env__raw_arch(struct perf_env *env);
+
#endif /* __PERF_ENV_H */
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 7c777cb32806..107bae7676b1 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -2296,6 +2296,15 @@ int machine__set_current_tid(struct machine *machine, int cpu, pid_t pid,
return 0;
}
+/*
+ * Compares the raw arch string. N.B. see instead perf_env__arch() if a
+ * normalized arch is needed.
+ */
+bool machine__is(struct machine *machine, const char *arch)
+{
+ return machine && !strcmp(perf_env__raw_arch(machine->env), arch);
+}
+
int machine__get_kernel_start(struct machine *machine)
{
struct map *map = machine__kernel_map(machine);
diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
index 388fb4741c54..b31d33b5aa2a 100644
--- a/tools/perf/util/machine.h
+++ b/tools/perf/util/machine.h
@@ -188,6 +188,8 @@ static inline bool machine__is_host(struct machine *machine)
return machine ? machine->pid == HOST_KERNEL_ID : false;
}
+bool machine__is(struct machine *machine, const char *arch);
+
struct thread *__machine__findnew_thread(struct machine *machine, pid_t pid, pid_t tid);
struct thread *machine__findnew_thread(struct machine *machine, pid_t pid, pid_t tid);
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* [PATCH 18/18] perf tools: Fix kernel_start for PTI on x86
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
` (17 preceding siblings ...)
(?)
@ 2018-05-19 10:55 ` Arnaldo Carvalho de Melo
-1 siblings, 0 replies; 22+ messages in thread
From: Arnaldo Carvalho de Melo @ 2018-05-19 10:55 UTC (permalink / raw)
To: Ingo Molnar
Cc: Clark Williams, linux-kernel, linux-perf-users, Adrian Hunter,
Alexander Shishkin, Andi Kleen, Andy Lutomirski, Dave Hansen,
H . Peter Anvin, Joerg Roedel, Peter Zijlstra, Thomas Gleixner,
x86, Arnaldo Carvalho de Melo
From: Adrian Hunter <adrian.hunter@intel.com>
Opickn x86_64, PTI entry trampolines are less than the start of kernel text,
but still above 2^63. So leave kernel_start = 1ULL << 63 for x86_64.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Tested-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: x86@kernel.org
Link: http://lkml.kernel.org/r/1526548928-20790-7-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/machine.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 107bae7676b1..e011a7160380 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -2321,7 +2321,12 @@ int machine__get_kernel_start(struct machine *machine)
machine->kernel_start = 1ULL << 63;
if (map) {
err = map__load(map);
- if (!err)
+ /*
+ * On x86_64, PTI entry trampolines are less than the
+ * start of kernel text, but still above 2^63. So leave
+ * kernel_start = 1ULL << 63 for x86_64.
+ */
+ if (!err && !machine__is(machine, "x86_64"))
machine->kernel_start = map->start;
}
return err;
--
2.14.3
^ permalink raw reply related [flat|nested] 22+ messages in thread* Re: [GIT PULL 00/18] perf/core improvements and fixes
2018-05-19 10:54 ` Arnaldo Carvalho de Melo
@ 2018-05-19 11:33 ` Ingo Molnar
-1 siblings, 0 replies; 22+ messages in thread
From: Ingo Molnar @ 2018-05-19 11:33 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: Clark Williams, linux-kernel, linux-perf-users, Adrian Hunter,
Alexander Shishkin, Andi Kleen, Andy Lutomirski, Dave Hansen,
David Ahern, H . Peter Anvin, Jin Yao, Jiri Olsa, Joerg Roedel,
Kan Liang, Namhyung Kim, Naveen N . Rao, Peter Zijlstra,
Ravi Bangoria, Sandipan Das, Taeung Song, Thomas Gleixner,
Wang Nan, x86, Arnaldo Carvalho de Melo
* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> Hi Ingo,
>
> Please consider pulling,
>
> - Arnaldo
>
> Test results at the end of this message, as usual.
>
> The following changes since commit 5aafae8d097e2161ee5c6a12ad532100f8885d2b:
>
> Merge tag 'perf-core-for-mingo-4.18-20180516' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2018-05-16 17:56:43 +0200)
>
> are available in the Git repository at:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-4.18-20180519
>
> for you to fetch changes up to 19422a9f2a3be7f3a046285ffae4cbb571aa853a:
>
> perf tools: Fix kernel_start for PTI on x86 (2018-05-19 06:42:51 -0300)
>
> ----------------------------------------------------------------
> perf/core improvements and fixes:
>
> - Record min/max LBR cycles (>= skylake) and add 'perf annotate' TUI
> hotkey to show it (c) (Jin Yao)
>
> - Fix machine->kernel_start for PTI on x86 (Adrian Hunter)
>
> - Make machine->env->arch always available, e.g. in 'perf top', not
> just when reading that info from perf.data files (Adrian Hunter)
>
> - Reduce the number of files read at 'perf' start, leaving information such as
> cacheline size, tracefs mount point determination, max_stack, etc, to be
> lazily read as tools needs then (Arnaldo Carvalho de Melo)
>
> - Fixup BPF include and examples install messages (Arnaldo Carvalho de Melo)
>
> - Fixup callchain addresses and symbol offsets in 'perf script', to help
> correlating with objdump output (Sandipan Das)
>
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
>
> ----------------------------------------------------------------
> Adrian Hunter (2):
> perf machine: Add machine__is() to identify machine arch
> perf tools: Fix kernel_start for PTI on x86
>
> Arnaldo Carvalho de Melo (12):
> perf config: Call perf_config__init() lazily
> tools lib api: The tracing_mnt variable doesn't need to be global
> tools lib api: Unexport 'tracing_path' variable
> tools lib api fs tracing_path: Introduce get/put_events_file() helpers
> perf tools: Reuse the path to the tracepoint /events/ directory
> perf parse-events: Use get/put_events_file()
> tools lib api fs tracing_path: Introduce opendir() method
> tools lib api fs tracing_path: Make tracing_events_path private
> tools include compiler-gcc: Add __pure attribute helper
> perf tools: Read the cache line size lazily
> perf tools: No need to unconditionally read the max_stack sysctls
> perf bpf: Fixup include and examples install messages
>
> Jin Yao (2):
> perf annotate: Record the min/max cycles
> perf annotate: Create hotkey 'c' to show min/max cycles
>
> Sandipan Das (2):
> perf script: Show virtual addresses instead of offsets
> perf script: Show symbol offsets by default
>
> tools/include/linux/compiler-gcc.h | 3 +
> tools/lib/api/fs/tracing_path.c | 40 +++++++++---
> tools/lib/api/fs/tracing_path.h | 9 ++-
> tools/perf/Makefile.perf | 2 +
> tools/perf/builtin-script.c | 26 ++++----
> tools/perf/builtin-top.c | 2 +-
> tools/perf/builtin-trace.c | 2 +-
> tools/perf/perf.c | 24 +------
> tools/perf/tests/parse-events.c | 9 +--
> .../tests/shell/record+probe_libc_inet_pton.sh | 12 ++--
> tools/perf/ui/browsers/annotate.c | 8 +++
> tools/perf/util/annotate.c | 51 ++++++++++++---
> tools/perf/util/annotate.h | 11 +++-
> tools/perf/util/config.c | 16 ++---
> tools/perf/util/config.h | 1 -
> tools/perf/util/env.c | 18 ++++++
> tools/perf/util/env.h | 2 +
> tools/perf/util/evsel.c | 2 +-
> tools/perf/util/machine.c | 18 +++++-
> tools/perf/util/machine.h | 2 +
> tools/perf/util/parse-events.c | 73 +++++++++++++---------
> tools/perf/util/probe-file.c | 3 +-
> tools/perf/util/sort.c | 4 +-
> tools/perf/util/sort.h | 4 +-
> tools/perf/util/trace-event-info.c | 11 ++--
> tools/perf/util/trace-event.c | 8 ++-
> tools/perf/util/util.c | 34 +++++++++-
> tools/perf/util/util.h | 4 +-
> 28 files changed, 279 insertions(+), 120 deletions(-)
Pulled, thanks a lot Arnaldo!
Ingo
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [GIT PULL 00/18] perf/core improvements and fixes
@ 2018-05-19 11:33 ` Ingo Molnar
0 siblings, 0 replies; 22+ messages in thread
From: Ingo Molnar @ 2018-05-19 11:33 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: Clark Williams, linux-kernel, linux-perf-users, Adrian Hunter,
Alexander Shishkin, Andi Kleen, Andy Lutomirski, Dave Hansen,
David Ahern, H . Peter Anvin, Jin Yao, Jiri Olsa, Joerg Roedel,
Kan Liang, Namhyung Kim, Naveen N . Rao, Peter Zijlstra,
Ravi Bangoria, Sandipan Das, Taeung Song
* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> Hi Ingo,
>
> Please consider pulling,
>
> - Arnaldo
>
> Test results at the end of this message, as usual.
>
> The following changes since commit 5aafae8d097e2161ee5c6a12ad532100f8885d2b:
>
> Merge tag 'perf-core-for-mingo-4.18-20180516' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2018-05-16 17:56:43 +0200)
>
> are available in the Git repository at:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-4.18-20180519
>
> for you to fetch changes up to 19422a9f2a3be7f3a046285ffae4cbb571aa853a:
>
> perf tools: Fix kernel_start for PTI on x86 (2018-05-19 06:42:51 -0300)
>
> ----------------------------------------------------------------
> perf/core improvements and fixes:
>
> - Record min/max LBR cycles (>= skylake) and add 'perf annotate' TUI
> hotkey to show it (c) (Jin Yao)
>
> - Fix machine->kernel_start for PTI on x86 (Adrian Hunter)
>
> - Make machine->env->arch always available, e.g. in 'perf top', not
> just when reading that info from perf.data files (Adrian Hunter)
>
> - Reduce the number of files read at 'perf' start, leaving information such as
> cacheline size, tracefs mount point determination, max_stack, etc, to be
> lazily read as tools needs then (Arnaldo Carvalho de Melo)
>
> - Fixup BPF include and examples install messages (Arnaldo Carvalho de Melo)
>
> - Fixup callchain addresses and symbol offsets in 'perf script', to help
> correlating with objdump output (Sandipan Das)
>
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
>
> ----------------------------------------------------------------
> Adrian Hunter (2):
> perf machine: Add machine__is() to identify machine arch
> perf tools: Fix kernel_start for PTI on x86
>
> Arnaldo Carvalho de Melo (12):
> perf config: Call perf_config__init() lazily
> tools lib api: The tracing_mnt variable doesn't need to be global
> tools lib api: Unexport 'tracing_path' variable
> tools lib api fs tracing_path: Introduce get/put_events_file() helpers
> perf tools: Reuse the path to the tracepoint /events/ directory
> perf parse-events: Use get/put_events_file()
> tools lib api fs tracing_path: Introduce opendir() method
> tools lib api fs tracing_path: Make tracing_events_path private
> tools include compiler-gcc: Add __pure attribute helper
> perf tools: Read the cache line size lazily
> perf tools: No need to unconditionally read the max_stack sysctls
> perf bpf: Fixup include and examples install messages
>
> Jin Yao (2):
> perf annotate: Record the min/max cycles
> perf annotate: Create hotkey 'c' to show min/max cycles
>
> Sandipan Das (2):
> perf script: Show virtual addresses instead of offsets
> perf script: Show symbol offsets by default
>
> tools/include/linux/compiler-gcc.h | 3 +
> tools/lib/api/fs/tracing_path.c | 40 +++++++++---
> tools/lib/api/fs/tracing_path.h | 9 ++-
> tools/perf/Makefile.perf | 2 +
> tools/perf/builtin-script.c | 26 ++++----
> tools/perf/builtin-top.c | 2 +-
> tools/perf/builtin-trace.c | 2 +-
> tools/perf/perf.c | 24 +------
> tools/perf/tests/parse-events.c | 9 +--
> .../tests/shell/record+probe_libc_inet_pton.sh | 12 ++--
> tools/perf/ui/browsers/annotate.c | 8 +++
> tools/perf/util/annotate.c | 51 ++++++++++++---
> tools/perf/util/annotate.h | 11 +++-
> tools/perf/util/config.c | 16 ++---
> tools/perf/util/config.h | 1 -
> tools/perf/util/env.c | 18 ++++++
> tools/perf/util/env.h | 2 +
> tools/perf/util/evsel.c | 2 +-
> tools/perf/util/machine.c | 18 +++++-
> tools/perf/util/machine.h | 2 +
> tools/perf/util/parse-events.c | 73 +++++++++++++---------
> tools/perf/util/probe-file.c | 3 +-
> tools/perf/util/sort.c | 4 +-
> tools/perf/util/sort.h | 4 +-
> tools/perf/util/trace-event-info.c | 11 ++--
> tools/perf/util/trace-event.c | 8 ++-
> tools/perf/util/util.c | 34 +++++++++-
> tools/perf/util/util.h | 4 +-
> 28 files changed, 279 insertions(+), 120 deletions(-)
Pulled, thanks a lot Arnaldo!
Ingo
^ permalink raw reply [flat|nested] 22+ messages in thread