Groups > linux.kernel > #1221665 > unrolled thread

[RFC 00/13] perf_env/CPU socket reorg/fixes

Started by	Arnaldo Carvalho de Melo <acme@kernel.org>
First post	2015-09-09 22:00 +0200
Last post	2015-09-14 03:30 +0200
Articles	20 on this page of 35 — 6 participants

Back to article view | Back to linux.kernel

  [RFC 00/13] perf_env/CPU socket reorg/fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
      RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method "Liang, Kan" <kan.liang@intel.com> - 2015-09-09 23:50 +0200
        Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-10 15:20 +0200
          RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method "Liang, Kan" <kan.liang@intel.com> - 2015-09-10 22:10 +0200
            RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method "Liang, Kan" <kan.liang@intel.com> - 2015-09-10 22:20 +0200
            Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-10 22:20 +0200
          Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method "Wangnan (F)" <wangnan0@huawei.com> - 2015-09-11 12:30 +0200
            Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-11 16:50 +0200
              Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-11 17:40 +0200
                Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method Namhyung Kim <namhyung@kernel.org> - 2015-09-11 18:20 +0200
                  Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-11 18:40 +0200
                [tip:perf/urgent] perf header:   Fixup reading of HEADER_NRCPUS feature tip-bot for Arnaldo Carvalho de Melo <tipbot@zytor.com> - 2015-09-14 11:20 +0200
      [tip:perf/core] perf env: Introduce read_cpu_topology_map()   method tip-bot for Arnaldo Carvalho de Melo <tipbot@zytor.com> - 2015-09-15 09:10 +0200
    [PATCH 07/13] perf hists browser: Fixup the "cpu" column width calculation Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
      Re: [PATCH 07/13] perf hists browser: Fixup the "cpu" column width  calculation "Wangnan (F)" <wangnan0@huawei.com> - 2015-09-11 13:00 +0200
    [PATCH 12/13] perf evsel: Remove forward declaration of 'struct perf_evlist' Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 02/13] perf env: Rename some leftovers from rename to perf_env Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 01/13] perf env: Move perf_env out of header.h and session.c into separate object Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 13/13] Revert "perf evlist: Add backpointer for perf_env to evlist" Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 08/13] perf machine: Add pointer to sample's environment Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 06/13] perf top: Cache the cpu topology info when "-s socket" is used Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 03/13] perf env: Adopt perf_header__set_cmdline Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 09/13] perf event: Use machine->env to find the cpu -> socket mapping Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
    [PATCH 10/13] perf report: Do not blindly use env->cpu[al.cpu].socket_id Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-09 22:00 +0200
      Re: [PATCH 10/13] perf report: Do not blindly use env->cpu[al.cpu].socket_id "Wangnan (F)" <wangnan0@huawei.com> - 2015-09-11 14:00 +0200
    Re: [RFC 00/13] perf_env/CPU socket reorg/fixes Jiri Olsa <jolsa@redhat.com> - 2015-09-10 11:20 +0200
      Re: [RFC 00/13] perf_env/CPU socket reorg/fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-10 16:20 +0200
    Re: [RFC 00/13] perf_env/CPU socket reorg/fixes "Wangnan (F)" <wangnan0@huawei.com> - 2015-09-11 14:30 +0200
      Re: [RFC 00/13] perf_env/CPU socket reorg/fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-11 15:10 +0200
        Re: [RFC 00/13] perf_env/CPU socket reorg/fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-11 15:30 +0200
          Re: [RFC 00/13] perf_env/CPU socket reorg/fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-11 15:40 +0200
            Re: [RFC 00/13] perf_env/CPU socket reorg/fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-09-11 15:40 +0200
              Re: [RFC 00/13] perf_env/CPU socket reorg/fixes "Wangnan (F)" <wangnan0@huawei.com> - 2015-09-14 03:40 +0200
        Re: [RFC 00/13] perf_env/CPU socket reorg/fixes "Wangnan (F)" <wangnan0@huawei.com> - 2015-09-14 03:30 +0200

Page 1 of 2 [1] 2 Next page →

#1221665 — [RFC 00/13] perf_env/CPU socket reorg/fixes

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-09 22:00 +0200
Subject	[RFC 00/13] perf_env/CPU socket reorg/fixes
Message-ID	<q6THb-4vo-3@gated-at.bofh.it>

Hi,

	Please take a look at these changes to fix the problems reported by
Wang Nan wrt accesses to the cpu_topology_map information.

	The fixes are present on these following two csets:

  perf event: Use machine->env to find the cpu -> socket mapping
  perf report: Do not blindly use env->cpu[al.cpu].socket_id

	The rest are fixes made while working on this, infrastructure to enable
the fixes, reverts for things that ended up not being necessary and some
cleanups.

	It is available at:

 git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git perf/env

	Please let me know if I can have your Acked-by, Tested-by or
Reviewed-by.

- Arnaldo

Arnaldo Carvalho de Melo (13):
  perf env: Move perf_env out of header.h and session.c into separate object
  perf env: Rename some leftovers from rename to perf_env
  perf env: Adopt perf_header__set_cmdline
  perf env: Introduce read_cpu_topology_map() method
  perf sort: Set flag stating if the "socket" key is being used
  perf top: Cache the cpu topology info when "-s socket" is used
  perf hists browser: Fixup the "cpu" column width calculation
  perf machine: Add pointer to sample's environment
  perf event: Use machine->env to find the cpu -> socket mapping
  perf report: Do not blindly use env->cpu[al.cpu].socket_id
  Revert "perf evsel: Add a backpointer to the evlist a evsel is in"
  perf evsel: Remove forward declaration of 'struct perf_evlist'
  Revert "perf evlist: Add backpointer for perf_env to evlist"
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1221667 — [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-09 22:00 +0200
Subject	[PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q6THc-4vo-11@gated-at.bofh.it>
In reply to	#1221665

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

Out of the code to write the cpu topology map in the perf.data file
header.

Now if one needs the CPU topology map for the running machine, one needs
to call perf_env__read_cpu_topology_map(perf_env) and the info will be
stored in perf_env.cpu.

For now we're using a global perf_env variable, that will have its
contents freed after we run a builtin.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-j181xo95og1w5q145dwjm3dw@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/perf.c        |  2 ++
 tools/perf/util/env.c    | 28 ++++++++++++++++++++++++++++
 tools/perf/util/env.h    |  2 ++
 tools/perf/util/header.c | 26 ++++++++++----------------
 4 files changed, 42 insertions(+), 16 deletions(-)

diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index f2fc019b3671..1fded922bcc8 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -8,6 +8,7 @@
  */
 #include "builtin.h"
 
+#include "util/env.h"
 #include "util/exec_cmd.h"
 #include "util/cache.h"
 #include "util/quote.h"
@@ -369,6 +370,7 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
 
 	status = p->fn(argc, argv, prefix);
 	exit_browser(status);
+	perf_env__exit(&perf_env);
 
 	if (status)
 		return status & 0xff;
diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c
index ca1e33a2203e..6af4f7c36820 100644
--- a/tools/perf/util/env.c
+++ b/tools/perf/util/env.c
@@ -1,3 +1,4 @@
+#include "cpumap.h"
 #include "env.h"
 #include "util.h"
 
@@ -56,3 +57,30 @@ out_free:
 out_enomem:
 	return -ENOMEM;
 }
+
+int perf_env__read_cpu_topology_map(struct perf_env *env)
+{
+	int cpu, nr_cpus;
+
+	if (env->cpu != NULL)
+		return 0;
+
+	if (env->nr_cpus_avail == 0)
+		env->nr_cpus_avail = sysconf(_SC_NPROCESSORS_CONF);
+
+	nr_cpus = env->nr_cpus_avail;
+	if (nr_cpus == -1)
+		return -EINVAL;
+
+	env->cpu = calloc(nr_cpus, sizeof(env->cpu[0]));
+	if (env->cpu == NULL)
+		return -ENOMEM;
+
+	for (cpu = 0; cpu < nr_cpus; ++cpu) {
+		env->cpu[cpu].core_id	= cpu_map__get_core_id(cpu);
+		env->cpu[cpu].socket_id	= cpu_map__get_socket_id(cpu);
+	}
+
+	env->nr_cpus_avail = nr_cpus;
+	return 0;
+}
diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
index 70124d9a1624..c4e36323d91e 100644
--- a/tools/perf/util/env.h
+++ b/tools/perf/util/env.h
@@ -38,4 +38,6 @@ void perf_env__exit(struct perf_env *env);
 
 int perf_env__set_cmdline(struct perf_env *env, int argc, const char *argv[]);
 
+int perf_env__read_cpu_topology_map(struct perf_env *env);
+
 #endif /* __PERF_ENV_H */
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 151b8310ac70..d4c8aa2f4db7 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -415,8 +415,6 @@ struct cpu_topo {
 	u32 thread_sib;
 	char **core_siblings;
 	char **thread_siblings;
-	int *core_id;
-	int *phy_pkg_id;
 };
 
 static int build_cpu_topo(struct cpu_topo *tp, int cpu)
@@ -479,9 +477,6 @@ try_threads:
 	}
 	ret = 0;
 done:
-	tp->core_id[cpu] = cpu_map__get_core_id(cpu);
-	tp->phy_pkg_id[cpu] = cpu_map__get_socket_id(cpu);
-
 	if(fp)
 		fclose(fp);
 	free(buf);
@@ -509,7 +504,7 @@ static struct cpu_topo *build_cpu_topology(void)
 	struct cpu_topo *tp;
 	void *addr;
 	u32 nr, i;
-	size_t sz, sz_id;
+	size_t sz;
 	long ncpus;
 	int ret = -1;
 
@@ -520,9 +515,8 @@ static struct cpu_topo *build_cpu_topology(void)
 	nr = (u32)(ncpus & UINT_MAX);
 
 	sz = nr * sizeof(char *);
-	sz_id = nr * sizeof(int);
 
-	addr = calloc(1, sizeof(*tp) + 2 * sz + 2 * sz_id);
+	addr = calloc(1, sizeof(*tp) + 2 * sz);
 	if (!addr)
 		return NULL;
 
@@ -532,10 +526,6 @@ static struct cpu_topo *build_cpu_topology(void)
 	tp->core_siblings = addr;
 	addr += sz;
 	tp->thread_siblings = addr;
-	addr += sz;
-	tp->core_id = addr;
-	addr += sz_id;
-	tp->phy_pkg_id = addr;
 
 	for (i = 0; i < nr; i++) {
 		ret = build_cpu_topo(tp, i);
@@ -554,7 +544,7 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
 {
 	struct cpu_topo *tp;
 	u32 i;
-	int ret;
+	int ret, j;
 
 	tp = build_cpu_topology();
 	if (!tp)
@@ -579,11 +569,15 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
 			break;
 	}
 
-	for (i = 0; i < tp->cpu_nr; i++) {
-		ret = do_write(fd, &tp->core_id[i], sizeof(int));
+	perf_env__read_cpu_topology_map(&perf_env);
+
+	for (j = 0; j < perf_env.nr_cpus_avail; j++) {
+		ret = do_write(fd, &perf_env.cpu[j].core_id,
+			       sizeof(perf_env.cpu[j].core_id));
 		if (ret < 0)
 			return ret;
-		ret = do_write(fd, &tp->phy_pkg_id[i], sizeof(int));
+		ret = do_write(fd, &perf_env.cpu[j].socket_id,
+			       sizeof(perf_env.cpu[j].socket_id));
 		if (ret < 0)
 			return ret;
 	}
-- 
2.1.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221749 — RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	"Liang, Kan" <kan.liang@intel.com>
Date	2015-09-09 23:50 +0200
Subject	RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q6VpD-703-7@gated-at.bofh.it>
In reply to	#1221667

> diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index
> 151b8310ac70..d4c8aa2f4db7 100644
> --- a/tools/perf/util/header.c
> +++ b/tools/perf/util/header.c
> @@ -415,8 +415,6 @@ struct cpu_topo {
>  	u32 thread_sib;
>  	char **core_siblings;
>  	char **thread_siblings;
> -	int *core_id;
> -	int *phy_pkg_id;
>  };
> 
>  static int build_cpu_topo(struct cpu_topo *tp, int cpu) @@ -479,9 +477,6
> @@ try_threads:
>  	}
>  	ret = 0;
>  done:
> -	tp->core_id[cpu] = cpu_map__get_core_id(cpu);
> -	tp->phy_pkg_id[cpu] = cpu_map__get_socket_id(cpu);
> -
>  	if(fp)
>  		fclose(fp);
>  	free(buf);
> @@ -509,7 +504,7 @@ static struct cpu_topo *build_cpu_topology(void)
>  	struct cpu_topo *tp;
>  	void *addr;
>  	u32 nr, i;
> -	size_t sz, sz_id;
> +	size_t sz;
>  	long ncpus;
>  	int ret = -1;
> 
> @@ -520,9 +515,8 @@ static struct cpu_topo *build_cpu_topology(void)
>  	nr = (u32)(ncpus & UINT_MAX);
> 
>  	sz = nr * sizeof(char *);
> -	sz_id = nr * sizeof(int);
> 
> -	addr = calloc(1, sizeof(*tp) + 2 * sz + 2 * sz_id);
> +	addr = calloc(1, sizeof(*tp) + 2 * sz);
>  	if (!addr)
>  		return NULL;
> 
> @@ -532,10 +526,6 @@ static struct cpu_topo *build_cpu_topology(void)
>  	tp->core_siblings = addr;
>  	addr += sz;
>  	tp->thread_siblings = addr;
> -	addr += sz;
> -	tp->core_id = addr;
> -	addr += sz_id;
> -	tp->phy_pkg_id = addr;
> 
>  	for (i = 0; i < nr; i++) {
>  		ret = build_cpu_topo(tp, i);
> @@ -554,7 +544,7 @@ static int write_cpu_topology(int fd, struct
> perf_header *h __maybe_unused,  {
>  	struct cpu_topo *tp;
>  	u32 i;
> -	int ret;
> +	int ret, j;
> 
>  	tp = build_cpu_topology();
>  	if (!tp)
> @@ -579,11 +569,15 @@ static int write_cpu_topology(int fd, struct
> perf_header *h __maybe_unused,
>  			break;
>  	}
> 
> -	for (i = 0; i < tp->cpu_nr; i++) {
> -		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> +	perf_env__read_cpu_topology_map(&perf_env);
> +

I think we need to handle error here.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222183 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-10 15:20 +0200
Subject	Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q79VE-2ta-17@gated-at.bofh.it>
In reply to	#1221749

Em Wed, Sep 09, 2015 at 09:41:18PM +0000, Liang, Kan escreveu:
> > -	for (i = 0; i < tp->cpu_nr; i++) {
> > -		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> > +	perf_env__read_cpu_topology_map(&perf_env);
> > +
> 
> I think we need to handle error here.

Ok, adding a test, updated patch below. While doing that, noticed that
the in the thread_sib case the "ret = do_something()" doesn't makes the
whole function fail, will fix later.

- Arnaldo

commit c98989f4a47a6fd3f3f8bd937b96faaced8b203e
Author: Arnaldo Carvalho de Melo <acme@redhat.com>
Date:   Wed Sep 9 10:37:01 2015 -0300

    perf env: Introduce read_cpu_topology_map() method
    
    Out of the code to write the cpu topology map in the perf.data file
    header.
    
    Now if one needs the CPU topology map for the running machine, one needs
    to call perf_env__read_cpu_topology_map(perf_env) and the info will be
    stored in perf_env.cpu.
    
    For now we're using a global perf_env variable, that will have its
    contents freed after we run a builtin.
    
    v2: Check perf_env__read_cpu_topology_map() return in
        write_cpu_topology() (Kan Liang)
    
    Cc: Adrian Hunter <adrian.hunter@intel.com>
    Cc: Borislav Petkov <bp@suse.de>
    Cc: David Ahern <dsahern@gmail.com>
    Cc: Frederic Weisbecker <fweisbec@gmail.com>
    Cc: Jiri Olsa <jolsa@redhat.com>
    Cc: Kan Liang <kan.liang@intel.com>
    Cc: Namhyung Kim <namhyung@kernel.org>
    Cc: Stephane Eranian <eranian@google.com>
    Cc: Wang Nan <wangnan0@huawei.com>
    Link: http://lkml.kernel.org/r/1441828225-667-5-git-send-email-acme@kernel.org
    Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index f2fc019b3671..1fded922bcc8 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -8,6 +8,7 @@
  */
 #include "builtin.h"
 
+#include "util/env.h"
 #include "util/exec_cmd.h"
 #include "util/cache.h"
 #include "util/quote.h"
@@ -369,6 +370,7 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
 
 	status = p->fn(argc, argv, prefix);
 	exit_browser(status);
+	perf_env__exit(&perf_env);
 
 	if (status)
 		return status & 0xff;
diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c
index ca1e33a2203e..6af4f7c36820 100644
--- a/tools/perf/util/env.c
+++ b/tools/perf/util/env.c
@@ -1,3 +1,4 @@
+#include "cpumap.h"
 #include "env.h"
 #include "util.h"
 
@@ -56,3 +57,30 @@ out_free:
 out_enomem:
 	return -ENOMEM;
 }
+
+int perf_env__read_cpu_topology_map(struct perf_env *env)
+{
+	int cpu, nr_cpus;
+
+	if (env->cpu != NULL)
+		return 0;
+
+	if (env->nr_cpus_avail == 0)
+		env->nr_cpus_avail = sysconf(_SC_NPROCESSORS_CONF);
+
+	nr_cpus = env->nr_cpus_avail;
+	if (nr_cpus == -1)
+		return -EINVAL;
+
+	env->cpu = calloc(nr_cpus, sizeof(env->cpu[0]));
+	if (env->cpu == NULL)
+		return -ENOMEM;
+
+	for (cpu = 0; cpu < nr_cpus; ++cpu) {
+		env->cpu[cpu].core_id	= cpu_map__get_core_id(cpu);
+		env->cpu[cpu].socket_id	= cpu_map__get_socket_id(cpu);
+	}
+
+	env->nr_cpus_avail = nr_cpus;
+	return 0;
+}
diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
index 70124d9a1624..c4e36323d91e 100644
--- a/tools/perf/util/env.h
+++ b/tools/perf/util/env.h
@@ -38,4 +38,6 @@ void perf_env__exit(struct perf_env *env);
 
 int perf_env__set_cmdline(struct perf_env *env, int argc, const char *argv[]);
 
+int perf_env__read_cpu_topology_map(struct perf_env *env);
+
 #endif /* __PERF_ENV_H */
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 151b8310ac70..d6437465f70f 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -415,8 +415,6 @@ struct cpu_topo {
 	u32 thread_sib;
 	char **core_siblings;
 	char **thread_siblings;
-	int *core_id;
-	int *phy_pkg_id;
 };
 
 static int build_cpu_topo(struct cpu_topo *tp, int cpu)
@@ -479,9 +477,6 @@ try_threads:
 	}
 	ret = 0;
 done:
-	tp->core_id[cpu] = cpu_map__get_core_id(cpu);
-	tp->phy_pkg_id[cpu] = cpu_map__get_socket_id(cpu);
-
 	if(fp)
 		fclose(fp);
 	free(buf);
@@ -509,7 +504,7 @@ static struct cpu_topo *build_cpu_topology(void)
 	struct cpu_topo *tp;
 	void *addr;
 	u32 nr, i;
-	size_t sz, sz_id;
+	size_t sz;
 	long ncpus;
 	int ret = -1;
 
@@ -520,9 +515,8 @@ static struct cpu_topo *build_cpu_topology(void)
 	nr = (u32)(ncpus & UINT_MAX);
 
 	sz = nr * sizeof(char *);
-	sz_id = nr * sizeof(int);
 
-	addr = calloc(1, sizeof(*tp) + 2 * sz + 2 * sz_id);
+	addr = calloc(1, sizeof(*tp) + 2 * sz);
 	if (!addr)
 		return NULL;
 
@@ -532,10 +526,6 @@ static struct cpu_topo *build_cpu_topology(void)
 	tp->core_siblings = addr;
 	addr += sz;
 	tp->thread_siblings = addr;
-	addr += sz;
-	tp->core_id = addr;
-	addr += sz_id;
-	tp->phy_pkg_id = addr;
 
 	for (i = 0; i < nr; i++) {
 		ret = build_cpu_topo(tp, i);
@@ -554,7 +544,7 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
 {
 	struct cpu_topo *tp;
 	u32 i;
-	int ret;
+	int ret, j;
 
 	tp = build_cpu_topology();
 	if (!tp)
@@ -579,11 +569,17 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
 			break;
 	}
 
-	for (i = 0; i < tp->cpu_nr; i++) {
-		ret = do_write(fd, &tp->core_id[i], sizeof(int));
+	ret = perf_env__read_cpu_topology_map(&perf_env);
+	if (ret < 0)
+		goto done;
+
+	for (j = 0; j < perf_env.nr_cpus_avail; j++) {
+		ret = do_write(fd, &perf_env.cpu[j].core_id,
+			       sizeof(perf_env.cpu[j].core_id));
 		if (ret < 0)
 			return ret;
-		ret = do_write(fd, &tp->phy_pkg_id[i], sizeof(int));
+		ret = do_write(fd, &perf_env.cpu[j].socket_id,
+			       sizeof(perf_env.cpu[j].socket_id));
 		if (ret < 0)
 			return ret;
 	}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222379 — RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	"Liang, Kan" <kan.liang@intel.com>
Date	2015-09-10 22:10 +0200
Subject	RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7gkp-3tg-7@gated-at.bofh.it>
In reply to	#1222183


> 
> Em Wed, Sep 09, 2015 at 09:41:18PM +0000, Liang, Kan escreveu:
> > > -	for (i = 0; i < tp->cpu_nr; i++) {
> > > -		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> > > +	perf_env__read_cpu_topology_map(&perf_env);
> > > +
> >
> > I think we need to handle error here.
> 
> Ok, adding a test, updated patch below. While doing that, noticed that the
> in the thread_sib case the "ret = do_something()" doesn't makes the
> whole function fail, will fix later.
> 

Thanks.
Except patch 11 & 13 (for reverting), the rest of patches are good to me.

Thanks,
Kan

> - Arnaldo
> 
> commit c98989f4a47a6fd3f3f8bd937b96faaced8b203e
> Author: Arnaldo Carvalho de Melo <acme@redhat.com>
> Date:   Wed Sep 9 10:37:01 2015 -0300
> 
>     perf env: Introduce read_cpu_topology_map() method
> 
>     Out of the code to write the cpu topology map in the perf.data file
>     header.
> 
>     Now if one needs the CPU topology map for the running machine, one
> needs
>     to call perf_env__read_cpu_topology_map(perf_env) and the info will
> be
>     stored in perf_env.cpu.
> 
>     For now we're using a global perf_env variable, that will have its
>     contents freed after we run a builtin.
> 
>     v2: Check perf_env__read_cpu_topology_map() return in
>         write_cpu_topology() (Kan Liang)
> 
>     Cc: Adrian Hunter <adrian.hunter@intel.com>
>     Cc: Borislav Petkov <bp@suse.de>
>     Cc: David Ahern <dsahern@gmail.com>
>     Cc: Frederic Weisbecker <fweisbec@gmail.com>
>     Cc: Jiri Olsa <jolsa@redhat.com>
>     Cc: Kan Liang <kan.liang@intel.com>
>     Cc: Namhyung Kim <namhyung@kernel.org>
>     Cc: Stephane Eranian <eranian@google.com>
>     Cc: Wang Nan <wangnan0@huawei.com>
>     Link: http://lkml.kernel.org/r/1441828225-667-5-git-send-email-
> acme@kernel.org
>     Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> diff --git a/tools/perf/perf.c b/tools/perf/perf.c index
> f2fc019b3671..1fded922bcc8 100644
> --- a/tools/perf/perf.c
> +++ b/tools/perf/perf.c
> @@ -8,6 +8,7 @@
>   */
>  #include "builtin.h"
> 
> +#include "util/env.h"
>  #include "util/exec_cmd.h"
>  #include "util/cache.h"
>  #include "util/quote.h"
> @@ -369,6 +370,7 @@ static int run_builtin(struct cmd_struct *p, int argc,
> const char **argv)
> 
>  	status = p->fn(argc, argv, prefix);
>  	exit_browser(status);
> +	perf_env__exit(&perf_env);
> 
>  	if (status)
>  		return status & 0xff;
> diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c index
> ca1e33a2203e..6af4f7c36820 100644
> --- a/tools/perf/util/env.c
> +++ b/tools/perf/util/env.c
> @@ -1,3 +1,4 @@
> +#include "cpumap.h"
>  #include "env.h"
>  #include "util.h"
> 
> @@ -56,3 +57,30 @@ out_free:
>  out_enomem:
>  	return -ENOMEM;
>  }
> +
> +int perf_env__read_cpu_topology_map(struct perf_env *env) {
> +	int cpu, nr_cpus;
> +
> +	if (env->cpu != NULL)
> +		return 0;
> +
> +	if (env->nr_cpus_avail == 0)
> +		env->nr_cpus_avail = sysconf(_SC_NPROCESSORS_CONF);
> +
> +	nr_cpus = env->nr_cpus_avail;
> +	if (nr_cpus == -1)
> +		return -EINVAL;
> +
> +	env->cpu = calloc(nr_cpus, sizeof(env->cpu[0]));
> +	if (env->cpu == NULL)
> +		return -ENOMEM;
> +
> +	for (cpu = 0; cpu < nr_cpus; ++cpu) {
> +		env->cpu[cpu].core_id	= cpu_map__get_core_id(cpu);
> +		env->cpu[cpu].socket_id	=
> cpu_map__get_socket_id(cpu);
> +	}
> +
> +	env->nr_cpus_avail = nr_cpus;
> +	return 0;
> +}
> diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h index
> 70124d9a1624..c4e36323d91e 100644
> --- a/tools/perf/util/env.h
> +++ b/tools/perf/util/env.h
> @@ -38,4 +38,6 @@ void perf_env__exit(struct perf_env *env);
> 
>  int perf_env__set_cmdline(struct perf_env *env, int argc, const char
> *argv[]);
> 
> +int perf_env__read_cpu_topology_map(struct perf_env *env);
> +
>  #endif /* __PERF_ENV_H */
> diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index
> 151b8310ac70..d6437465f70f 100644
> --- a/tools/perf/util/header.c
> +++ b/tools/perf/util/header.c
> @@ -415,8 +415,6 @@ struct cpu_topo {
>  	u32 thread_sib;
>  	char **core_siblings;
>  	char **thread_siblings;
> -	int *core_id;
> -	int *phy_pkg_id;
>  };
> 
>  static int build_cpu_topo(struct cpu_topo *tp, int cpu) @@ -479,9 +477,6
> @@ try_threads:
>  	}
>  	ret = 0;
>  done:
> -	tp->core_id[cpu] = cpu_map__get_core_id(cpu);
> -	tp->phy_pkg_id[cpu] = cpu_map__get_socket_id(cpu);
> -
>  	if(fp)
>  		fclose(fp);
>  	free(buf);
> @@ -509,7 +504,7 @@ static struct cpu_topo *build_cpu_topology(void)
>  	struct cpu_topo *tp;
>  	void *addr;
>  	u32 nr, i;
> -	size_t sz, sz_id;
> +	size_t sz;
>  	long ncpus;
>  	int ret = -1;
> 
> @@ -520,9 +515,8 @@ static struct cpu_topo *build_cpu_topology(void)
>  	nr = (u32)(ncpus & UINT_MAX);
> 
>  	sz = nr * sizeof(char *);
> -	sz_id = nr * sizeof(int);
> 
> -	addr = calloc(1, sizeof(*tp) + 2 * sz + 2 * sz_id);
> +	addr = calloc(1, sizeof(*tp) + 2 * sz);
>  	if (!addr)
>  		return NULL;
> 
> @@ -532,10 +526,6 @@ static struct cpu_topo *build_cpu_topology(void)
>  	tp->core_siblings = addr;
>  	addr += sz;
>  	tp->thread_siblings = addr;
> -	addr += sz;
> -	tp->core_id = addr;
> -	addr += sz_id;
> -	tp->phy_pkg_id = addr;
> 
>  	for (i = 0; i < nr; i++) {
>  		ret = build_cpu_topo(tp, i);
> @@ -554,7 +544,7 @@ static int write_cpu_topology(int fd, struct
> perf_header *h __maybe_unused,  {
>  	struct cpu_topo *tp;
>  	u32 i;
> -	int ret;
> +	int ret, j;
> 
>  	tp = build_cpu_topology();
>  	if (!tp)
> @@ -579,11 +569,17 @@ static int write_cpu_topology(int fd, struct
> perf_header *h __maybe_unused,
>  			break;
>  	}
> 
> -	for (i = 0; i < tp->cpu_nr; i++) {
> -		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> +	ret = perf_env__read_cpu_topology_map(&perf_env);
> +	if (ret < 0)
> +		goto done;
> +
> +	for (j = 0; j < perf_env.nr_cpus_avail; j++) {
> +		ret = do_write(fd, &perf_env.cpu[j].core_id,
> +			       sizeof(perf_env.cpu[j].core_id));
>  		if (ret < 0)
>  			return ret;
> -		ret = do_write(fd, &tp->phy_pkg_id[i], sizeof(int));
> +		ret = do_write(fd, &perf_env.cpu[j].socket_id,
> +			       sizeof(perf_env.cpu[j].socket_id));
>  		if (ret < 0)
>  			return ret;
>  	}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222380 — RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	"Liang, Kan" <kan.liang@intel.com>
Date	2015-09-10 22:20 +0200
Subject	RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7gu6-3Ey-1@gated-at.bofh.it>
In reply to	#1222379


> Em Thu, Sep 10, 2015 at 08:00:54PM +0000, Liang, Kan escreveu:
> > > Em Wed, Sep 09, 2015 at 09:41:18PM +0000, Liang, Kan escreveu:
> > > > > -	for (i = 0; i < tp->cpu_nr; i++) {
> > > > > -		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> > > > > +	perf_env__read_cpu_topology_map(&perf_env);
> > > > > +
> 
> > > > I think we need to handle error here.
> 
> > > Ok, adding a test, updated patch below. While doing that, noticed
> > > that the in the thread_sib case the "ret = do_something()" doesn't
> > > makes the whole function fail, will fix later.
> 
> > Except patch 11 & 13 (for reverting), the rest of patches are good to me.
> 
> Ok, so, to take it more formally, can I turn this "good to me" as justification
> for adding "Acked-by: Kan Liang", to those patches?

Yes.

> 
> Please take a look at the other patches in the perf/env branch, I did it while
> trying to figure it out if the reverts should be kept.
>
OK

Thanks,
Kan
 
> - Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222381 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-10 22:20 +0200
Subject	Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7gu6-3Ey-3@gated-at.bofh.it>
In reply to	#1222379

Em Thu, Sep 10, 2015 at 08:00:54PM +0000, Liang, Kan escreveu:
> > Em Wed, Sep 09, 2015 at 09:41:18PM +0000, Liang, Kan escreveu:
> > > > -	for (i = 0; i < tp->cpu_nr; i++) {
> > > > -		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> > > > +	perf_env__read_cpu_topology_map(&perf_env);
> > > > +

> > > I think we need to handle error here.

> > Ok, adding a test, updated patch below. While doing that, noticed that the
> > in the thread_sib case the "ret = do_something()" doesn't makes the
> > whole function fail, will fix later.

> Except patch 11 & 13 (for reverting), the rest of patches are good to me.

Ok, so, to take it more formally, can I turn this "good to me" as
justification for adding "Acked-by: Kan Liang", to those patches?

Please take a look at the other patches in the perf/env branch, I did it
while trying to figure it out if the reverts should be kept.

- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222699 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	"Wangnan (F)" <wangnan0@huawei.com>
Date	2015-09-11 12:30 +0200
Subject	Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7tKG-7d0-29@gated-at.bofh.it>
In reply to	#1222183


On 2015/9/10 21:12, Arnaldo Carvalho de Melo wrote:

[SNIP]

> +
> +int perf_env__read_cpu_topology_map(struct perf_env *env)
> +{
> +	int cpu, nr_cpus;
> +
> +	if (env->cpu != NULL)
> +		return 0;
> +
> +	if (env->nr_cpus_avail == 0)
> +		env->nr_cpus_avail = sysconf(_SC_NPROCESSORS_CONF);
> +
> +	nr_cpus = env->nr_cpus_avail;
> +	if (nr_cpus == -1)
> +		return -EINVAL;
> +
> +	env->cpu = calloc(nr_cpus, sizeof(env->cpu[0]));
> +	if (env->cpu == NULL)
> +		return -ENOMEM;
> +
> +	for (cpu = 0; cpu < nr_cpus; ++cpu) {
> +		env->cpu[cpu].core_id	= cpu_map__get_core_id(cpu);
> +		env->cpu[cpu].socket_id	= cpu_map__get_socket_id(cpu);
> +	}

Shouldn't we check the failure of these two functions?

At this point perf_env__read_cpu_topology_map and build_cpu_topology are 
doing
similar things. build_cpu_topology() reads 
/sys/xxxxx/cpu%d/topology/{core,thread}_siblings_list,
perf_env__read_cpu_topology_map() reads /sys/xxxxx/cpu%d/topology/core_id,
but build_cpu_topology() returns error if any read failed, but
perf_env__read_cpu_topology_map() fills core_id and socket_id with -1 if 
read fail.

I tried to offline a core between build_cpu_topology() and 
perf_env__read_cpu_topology_map(),
and perf report say:

  # perf report -v --header-only -I
  build id event received for [kernel.kallsyms]: (...)
  core_id number is too big.You may need to upgrade the perf tool. <-- 
*see this warning*
  # ========
  # captured on: Sun Feb 15 15:01:05 2009
  # hostname : localhost
  ...
  # sibling cores   : ...
  # sibling threads : 7
  # Core ID and Socket ID information is not available
  # pmu mappings: not available
  # ========

Thank you.

> +
> +	env->nr_cpus_avail = nr_cpus;
> +	return 0;
> +}
> diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
> index 70124d9a1624..c4e36323d91e 100644
> --- a/tools/perf/util/env.h
> +++ b/tools/perf/util/env.h
> @@ -38,4 +38,6 @@ void perf_env__exit(struct perf_env *env);
>   
>   int perf_env__set_cmdline(struct perf_env *env, int argc, const char *argv[]);
>   
> +int perf_env__read_cpu_topology_map(struct perf_env *env);
> +
>   #endif /* __PERF_ENV_H */
> diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
> index 151b8310ac70..d6437465f70f 100644
> --- a/tools/perf/util/header.c
> +++ b/tools/perf/util/header.c
> @@ -415,8 +415,6 @@ struct cpu_topo {
>   	u32 thread_sib;
>   	char **core_siblings;
>   	char **thread_siblings;
> -	int *core_id;
> -	int *phy_pkg_id;
>   };
>   
>   static int build_cpu_topo(struct cpu_topo *tp, int cpu)
> @@ -479,9 +477,6 @@ try_threads:
>   	}
>   	ret = 0;
>   done:
> -	tp->core_id[cpu] = cpu_map__get_core_id(cpu);
> -	tp->phy_pkg_id[cpu] = cpu_map__get_socket_id(cpu);
> -
>   	if(fp)
>   		fclose(fp);
>   	free(buf);
> @@ -509,7 +504,7 @@ static struct cpu_topo *build_cpu_topology(void)
>   	struct cpu_topo *tp;
>   	void *addr;
>   	u32 nr, i;
> -	size_t sz, sz_id;
> +	size_t sz;
>   	long ncpus;
>   	int ret = -1;
>   
> @@ -520,9 +515,8 @@ static struct cpu_topo *build_cpu_topology(void)
>   	nr = (u32)(ncpus & UINT_MAX);
>   
>   	sz = nr * sizeof(char *);
> -	sz_id = nr * sizeof(int);
>   
> -	addr = calloc(1, sizeof(*tp) + 2 * sz + 2 * sz_id);
> +	addr = calloc(1, sizeof(*tp) + 2 * sz);
>   	if (!addr)
>   		return NULL;
>   
> @@ -532,10 +526,6 @@ static struct cpu_topo *build_cpu_topology(void)
>   	tp->core_siblings = addr;
>   	addr += sz;
>   	tp->thread_siblings = addr;
> -	addr += sz;
> -	tp->core_id = addr;
> -	addr += sz_id;
> -	tp->phy_pkg_id = addr;
>   
>   	for (i = 0; i < nr; i++) {
>   		ret = build_cpu_topo(tp, i);
> @@ -554,7 +544,7 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
>   {
>   	struct cpu_topo *tp;
>   	u32 i;
> -	int ret;
> +	int ret, j;
>   
>   	tp = build_cpu_topology();
>   	if (!tp)
> @@ -579,11 +569,17 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
>   			break;
>   	}
>   
> -	for (i = 0; i < tp->cpu_nr; i++) {
> -		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> +	ret = perf_env__read_cpu_topology_map(&perf_env);
> +	if (ret < 0)
> +		goto done;
> +
> +	for (j = 0; j < perf_env.nr_cpus_avail; j++) {
> +		ret = do_write(fd, &perf_env.cpu[j].core_id,
> +			       sizeof(perf_env.cpu[j].core_id));
>   		if (ret < 0)
>   			return ret;
> -		ret = do_write(fd, &tp->phy_pkg_id[i], sizeof(int));
> +		ret = do_write(fd, &perf_env.cpu[j].socket_id,
> +			       sizeof(perf_env.cpu[j].socket_id));
>   		if (ret < 0)
>   			return ret;
>   	}


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222891 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-11 16:50 +0200
Subject	Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7xOi-4C9-19@gated-at.bofh.it>
In reply to	#1222699

Em Fri, Sep 11, 2015 at 06:20:03PM +0800, Wangnan (F) escreveu:
> 
> 
> On 2015/9/10 21:12, Arnaldo Carvalho de Melo wrote:
> 
> [SNIP]
> 
> >+
> >+int perf_env__read_cpu_topology_map(struct perf_env *env)
> >+{
> >+	int cpu, nr_cpus;
> >+
> >+	if (env->cpu != NULL)
> >+		return 0;
> >+
> >+	if (env->nr_cpus_avail == 0)
> >+		env->nr_cpus_avail = sysconf(_SC_NPROCESSORS_CONF);
> >+
> >+	nr_cpus = env->nr_cpus_avail;
> >+	if (nr_cpus == -1)
> >+		return -EINVAL;
> >+
> >+	env->cpu = calloc(nr_cpus, sizeof(env->cpu[0]));
> >+	if (env->cpu == NULL)
> >+		return -ENOMEM;
> >+
> >+	for (cpu = 0; cpu < nr_cpus; ++cpu) {
> >+		env->cpu[cpu].core_id	= cpu_map__get_core_id(cpu);
> >+		env->cpu[cpu].socket_id	= cpu_map__get_socket_id(cpu);
> >+	}
> 
> Shouldn't we check the failure of these two functions?

Humm, the original code, in build_cpu_topology() was not checking
that, i.e. if you tried it with or without my patches the result will be
the same, no?
 
> At this point perf_env__read_cpu_topology_map and build_cpu_topology are
> doing
> similar things. build_cpu_topology() reads
> /sys/xxxxx/cpu%d/topology/{core,thread}_siblings_list,
> perf_env__read_cpu_topology_map() reads /sys/xxxxx/cpu%d/topology/core_id,
> but build_cpu_topology() returns error if any read failed, but
> perf_env__read_cpu_topology_map() fills core_id and socket_id with -1 if
> read fail.

> I tried to offline a core between build_cpu_topology() and
> perf_env__read_cpu_topology_map(),
> and perf report say:
> 
>  # perf report -v --header-only -I
>  build id event received for [kernel.kallsyms]: (...)
>  core_id number is too big.You may need to upgrade the perf tool. <-- *see
> this warning*
>  # ========
>  # captured on: Sun Feb 15 15:01:05 2009
>  # hostname : localhost
>  ...
>  # sibling cores   : ...
>  # sibling threads : 7
>  # Core ID and Socket ID information is not available
>  # pmu mappings: not available
>  # ========

So this is a problem before and after my patches, i.e. If I go on and
do, with what we have in acme/perf/core, i.e. none of the changes I'm
playing with in perf/env:

  $ git remote update acme
  Fetching acme
  $ git checkout -b tmp acme/perf/core
  Branch tmp set up to track remote branch perf/core from acme.
  Switched to a new branch 'tmp'
  $ git log --oneline | head -5
  7e150fb33a91 perf tests: Introduce iterator function for tests
  1765d9b26f84 perf test: Add entry for hists socket filter
  207bb55e9193 perf hists browser: Zoom in/out for processor socket
  9a2843a5f421 perf report: Introduce --socket-filter option
  99851c76436a perf tools: Introduce new sort type "socket" for the processor socket
  $ rm -rf /tmp/build/perf ; mkdir -p /tmp/build/perf ; make DEBUG=1 -C tools/perf O=/tmp/build/perf install-bin
  # echo 0 > /sys/devices/system/cpu/cpu2/online
  $ cat /sys/devices/system/cpu/cpu2/topology/core_id
  cat: /sys/devices/system/cpu/cpu2/topology/core_id: No such file or directory
  $ ls -la /sys/devices/system/cpu/cpu2/topology/
  ls: cannot access /sys/devices/system/cpu/cpu2/topology/: No such file or directory
  $ perf record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.012 MB perf.data (7 samples) ]
  [acme@felicio linux]$ perf report --header-only -I 
  # ========
  # captured on: Fri Sep 11 11:34:18 2015
  # hostname : felicio.ghostprotocols.net
  # os release : 4.2.0
  # perf version : 4.2.g7e150f
  #  arch : x86_64
  # nrcpus online : 4
  # nrcpus avail : 3
  # cpudesc : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
  <SNIP>
  # node0 meminfo  : total = 8085412 kB, free = 5317596 kB
  # node0 cpu list : 0-1,3
  <SNIP>
  $ 

We can see multiple bugs here, right? online/avail is swapped, and when
online != avail we simply do not record the cpu topology info at all!

If we get that CPU back online:

  # echo 1 > /sys/devices/system/cpu/cpu2/online

Then all works:

  $ perf record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.012 MB perf.data (7 samples) ]
  $ perf report --header-only -I
  # ========
  # captured on: Fri Sep 11 11:37:31 2015
  # hostname : felicio.ghostprotocols.net
  # os release : 4.2.0
  # perf version : 4.2.g7e150f
  # arch : x86_64
  # nrcpus online : 4
  # nrcpus avail : 4
  # cpudesc : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
  <SNIP>
  # sibling cores   : 0-3
  # sibling threads : 0
  # sibling threads : 1
  # sibling threads : 2
  # sibling threads : 3
  # CPU 0: Core ID 0, Socket ID 0
  # CPU 1: Core ID 1, Socket ID 0
  # CPU 2: Core ID 2, Socket ID 0
  # CPU 3: Core ID 3, Socket ID 0
  # node0 meminfo  : total = 8085412 kB, free = 5316992 kB
  # node0 cpu list : 0-3

So, again, this was not introduced by this patchkit, but it is good that you did
these offline tests, so we can fix it!

- Arnaldo
 
 
> >+
> >+	env->nr_cpus_avail = nr_cpus;
> >+	return 0;
> >+}
> >diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
> >index 70124d9a1624..c4e36323d91e 100644
> >--- a/tools/perf/util/env.h
> >+++ b/tools/perf/util/env.h
> >@@ -38,4 +38,6 @@ void perf_env__exit(struct perf_env *env);
> >  int perf_env__set_cmdline(struct perf_env *env, int argc, const char *argv[]);
> >+int perf_env__read_cpu_topology_map(struct perf_env *env);
> >+
> >  #endif /* __PERF_ENV_H */
> >diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
> >index 151b8310ac70..d6437465f70f 100644
> >--- a/tools/perf/util/header.c
> >+++ b/tools/perf/util/header.c
> >@@ -415,8 +415,6 @@ struct cpu_topo {
> >  	u32 thread_sib;
> >  	char **core_siblings;
> >  	char **thread_siblings;
> >-	int *core_id;
> >-	int *phy_pkg_id;
> >  };
> >  static int build_cpu_topo(struct cpu_topo *tp, int cpu)
> >@@ -479,9 +477,6 @@ try_threads:
> >  	}
> >  	ret = 0;
> >  done:
> >-	tp->core_id[cpu] = cpu_map__get_core_id(cpu);
> >-	tp->phy_pkg_id[cpu] = cpu_map__get_socket_id(cpu);
> >-
> >  	if(fp)
> >  		fclose(fp);
> >  	free(buf);
> >@@ -509,7 +504,7 @@ static struct cpu_topo *build_cpu_topology(void)
> >  	struct cpu_topo *tp;
> >  	void *addr;
> >  	u32 nr, i;
> >-	size_t sz, sz_id;
> >+	size_t sz;
> >  	long ncpus;
> >  	int ret = -1;
> >@@ -520,9 +515,8 @@ static struct cpu_topo *build_cpu_topology(void)
> >  	nr = (u32)(ncpus & UINT_MAX);
> >  	sz = nr * sizeof(char *);
> >-	sz_id = nr * sizeof(int);
> >-	addr = calloc(1, sizeof(*tp) + 2 * sz + 2 * sz_id);
> >+	addr = calloc(1, sizeof(*tp) + 2 * sz);
> >  	if (!addr)
> >  		return NULL;
> >@@ -532,10 +526,6 @@ static struct cpu_topo *build_cpu_topology(void)
> >  	tp->core_siblings = addr;
> >  	addr += sz;
> >  	tp->thread_siblings = addr;
> >-	addr += sz;
> >-	tp->core_id = addr;
> >-	addr += sz_id;
> >-	tp->phy_pkg_id = addr;
> >  	for (i = 0; i < nr; i++) {
> >  		ret = build_cpu_topo(tp, i);
> >@@ -554,7 +544,7 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
> >  {
> >  	struct cpu_topo *tp;
> >  	u32 i;
> >-	int ret;
> >+	int ret, j;
> >  	tp = build_cpu_topology();
> >  	if (!tp)
> >@@ -579,11 +569,17 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
> >  			break;
> >  	}
> >-	for (i = 0; i < tp->cpu_nr; i++) {
> >-		ret = do_write(fd, &tp->core_id[i], sizeof(int));
> >+	ret = perf_env__read_cpu_topology_map(&perf_env);
> >+	if (ret < 0)
> >+		goto done;
> >+
> >+	for (j = 0; j < perf_env.nr_cpus_avail; j++) {
> >+		ret = do_write(fd, &perf_env.cpu[j].core_id,
> >+			       sizeof(perf_env.cpu[j].core_id));
> >  		if (ret < 0)
> >  			return ret;
> >-		ret = do_write(fd, &tp->phy_pkg_id[i], sizeof(int));
> >+		ret = do_write(fd, &perf_env.cpu[j].socket_id,
> >+			       sizeof(perf_env.cpu[j].socket_id));
> >  		if (ret < 0)
> >  			return ret;
> >  	}
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222918 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-11 17:40 +0200
Subject	Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7yAF-5N1-17@gated-at.bofh.it>
In reply to	#1222891

Em Fri, Sep 11, 2015 at 11:40:28AM -0300, Arnaldo Carvalho de Melo escreveu:
> So this is a problem before and after my patches, i.e. If I go on and
> do, with what we have in acme/perf/core, i.e. none of the changes I'm
> playing with in perf/env:
> 
>   $ git remote update acme
>   Fetching acme
>   $ git checkout -b tmp acme/perf/core
>   Branch tmp set up to track remote branch perf/core from acme.
>   Switched to a new branch 'tmp'
>   $ git log --oneline | head -5
>   7e150fb33a91 perf tests: Introduce iterator function for tests
>   1765d9b26f84 perf test: Add entry for hists socket filter
>   207bb55e9193 perf hists browser: Zoom in/out for processor socket
>   9a2843a5f421 perf report: Introduce --socket-filter option
>   99851c76436a perf tools: Introduce new sort type "socket" for the processor socket
>   $ rm -rf /tmp/build/perf ; mkdir -p /tmp/build/perf ; make DEBUG=1 -C tools/perf O=/tmp/build/perf install-bin
>   # echo 0 > /sys/devices/system/cpu/cpu2/online
>   $ cat /sys/devices/system/cpu/cpu2/topology/core_id
>   cat: /sys/devices/system/cpu/cpu2/topology/core_id: No such file or directory
>   $ ls -la /sys/devices/system/cpu/cpu2/topology/
>   ls: cannot access /sys/devices/system/cpu/cpu2/topology/: No such file or directory
>   $ perf record usleep 1
>   [ perf record: Woken up 1 times to write data ]
>   [ perf record: Captured and wrote 0.012 MB perf.data (7 samples) ]
>   [acme@felicio linux]$ perf report --header-only -I 
>   # ========
>   # captured on: Fri Sep 11 11:34:18 2015
>   # hostname : felicio.ghostprotocols.net
>   # os release : 4.2.0
>   # perf version : 4.2.g7e150f
>   #  arch : x86_64
>   # nrcpus online : 4
>   # nrcpus avail : 3
>   # cpudesc : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
>   <SNIP>
>   # node0 meminfo  : total = 8085412 kB, free = 5317596 kB
>   # node0 cpu list : 0-1,3
>   <SNIP>
>   $ 

Stephane, Namhyung, the bug report below is about the online/avail
above, they are swapped, full explanation below.
 
> We can see multiple bugs here, right? online/avail is swapped, and when
> online != avail we simply do not record the cpu topology info at all!
 
> If we get that CPU back online:
> 
>   # echo 1 > /sys/devices/system/cpu/cpu2/online
> 
> Then all works:
> 
>   $ perf record usleep 1
>   [ perf record: Woken up 1 times to write data ]
>   [ perf record: Captured and wrote 0.012 MB perf.data (7 samples) ]
>   $ perf report --header-only -I
>   # ========
>   # captured on: Fri Sep 11 11:37:31 2015
>   # hostname : felicio.ghostprotocols.net
>   # os release : 4.2.0
>   # perf version : 4.2.g7e150f
>   # arch : x86_64
>   # nrcpus online : 4
>   # nrcpus avail : 4
>   # cpudesc : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
>   <SNIP>
>   # sibling cores   : 0-3
>   # sibling threads : 0
>   # sibling threads : 1
>   # sibling threads : 2
>   # sibling threads : 3
>   # CPU 0: Core ID 0, Socket ID 0
>   # CPU 1: Core ID 1, Socket ID 0
>   # CPU 2: Core ID 2, Socket ID 0
>   # CPU 3: Core ID 3, Socket ID 0
>   # node0 meminfo  : total = 8085412 kB, free = 5316992 kB
>   # node0 cpu list : 0-3
> 
> So, again, this was not introduced by this patchkit, but it is good that you did
> these offline tests, so we can fix it!

Stephane, this was introduced in:

  commit fbe96f29ce4b33e0a22219cc7f5996d9157717e3
  Author: Stephane Eranian <eranian@google.com>
  Date:   Fri Sep 30 15:40:40 2011 +0200

    perf tools: Make perf.data more self-descriptive (v8)

------------------

When you write this part:

      - HEADER_NRCPUS: number of online/avail cpus

You do:

static int write_nrcpus(int fd, struct perf_header *h __used,
                       struct perf_evlist *evlist __used)
{
       long nr;
       u32 nrc, nra;
       int ret;

       nr = sysconf(_SC_NPROCESSORS_CONF);
       if (nr < 0)
               return -1;

       nrc = (u32)(nr & UINT_MAX);

       nr = sysconf(_SC_NPROCESSORS_ONLN);
       if (nr < 0)
               return -1;

       nra = (u32)(nr & UINT_MAX);

       ret = do_write(fd, &nrc, sizeof(nrc));
       if (ret < 0)
               return ret;

       return do_write(fd, &nra, sizeof(nra));
}

I.e. write what you called 'nrc' using what is in SC_NRPROCESSORS_CONF, that in
the documentation for glibc reads:

---------------------

http://www.gnu.org/software/libc/manual/html_node/Processor-Resources.html

  sysconf (_SC_NPROCESSORS_CONF)

which returns the number of processors the operating system configured. But it
might be possible for the operating system to disable individual processors and
so the call 

---------------------

Which menas "NR_AVAILABLE", right?

But then you call a variable 'nra' which sounds like you think that what is in
_SC_NPROCESSORS_ONLN is the "available" number of CPUs, which is confused a bit
more by the glibc docs when refering to _SC_NPROCESSORS_ONLN:

---------------------
  sysconf (_SC_NPROCESSORS_ONLN)

returns the number of processors which are currently online (i.e., available). 
---------------------

Then, when printing the number of CPUs encoded in the perf.data file by the
above write_nrcpus() routine, you did:

static void print_nrcpus(struct perf_header *ph, int fd, FILE *fp)
{
       ssize_t ret;
       u32 nr;

       ret = read(fd, &nr, sizeof(nr));
       if (ret != (ssize_t)sizeof(nr))
               nr = -1; /* interpreted as error */

       if (ph->needs_swap)
               nr = bswap_32(nr);

       fprintf(fp, "# nrcpus online : %u\n", nr);

       ret = read(fd, &nr, sizeof(nr));
       if (ret != (ssize_t)sizeof(nr))
               nr = -1; /* interpreted as error */

       if (ph->needs_swap)
               nr = bswap_32(nr);

       fprintf(fp, "# nrcpus avail : %u\n", nr);
}

You inverted it, no?

So, could you please check if the below patch can have your Acked-by?
Namhyung?

Thanks,

- Arnaldo

diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 41814547da15..fce6634aebe2 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -1438,7 +1438,7 @@ static int process_nrcpus(struct perf_file_section *section __maybe_unused,
 	if (ph->needs_swap)
 		nr = bswap_32(nr);
 
-	ph->env.nr_cpus_online = nr;
+	ph->env.nr_cpus_avail = nr;
 
 	ret = readn(fd, &nr, sizeof(nr));
 	if (ret != sizeof(nr))
@@ -1447,7 +1447,7 @@ static int process_nrcpus(struct perf_file_section *section __maybe_unused,
 	if (ph->needs_swap)
 		nr = bswap_32(nr);
 
-	ph->env.nr_cpus_avail = nr;
+	ph->env.nr_cpus_online = nr;
 	return 0;
 }
 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222939 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	Namhyung Kim <namhyung@kernel.org>
Date	2015-09-11 18:20 +0200
Subject	Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7zdo-6M4-13@gated-at.bofh.it>
In reply to	#1222918

Hi Arnaldo,

On Fri, Sep 11, 2015 at 12:33:23PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Fri, Sep 11, 2015 at 11:40:28AM -0300, Arnaldo Carvalho de Melo escreveu:
> > So this is a problem before and after my patches, i.e. If I go on and
> > do, with what we have in acme/perf/core, i.e. none of the changes I'm
> > playing with in perf/env:
> > 
> >   $ git remote update acme
> >   Fetching acme
> >   $ git checkout -b tmp acme/perf/core
> >   Branch tmp set up to track remote branch perf/core from acme.
> >   Switched to a new branch 'tmp'
> >   $ git log --oneline | head -5
> >   7e150fb33a91 perf tests: Introduce iterator function for tests
> >   1765d9b26f84 perf test: Add entry for hists socket filter
> >   207bb55e9193 perf hists browser: Zoom in/out for processor socket
> >   9a2843a5f421 perf report: Introduce --socket-filter option
> >   99851c76436a perf tools: Introduce new sort type "socket" for the processor socket
> >   $ rm -rf /tmp/build/perf ; mkdir -p /tmp/build/perf ; make DEBUG=1 -C tools/perf O=/tmp/build/perf install-bin
> >   # echo 0 > /sys/devices/system/cpu/cpu2/online
> >   $ cat /sys/devices/system/cpu/cpu2/topology/core_id
> >   cat: /sys/devices/system/cpu/cpu2/topology/core_id: No such file or directory
> >   $ ls -la /sys/devices/system/cpu/cpu2/topology/
> >   ls: cannot access /sys/devices/system/cpu/cpu2/topology/: No such file or directory
> >   $ perf record usleep 1
> >   [ perf record: Woken up 1 times to write data ]
> >   [ perf record: Captured and wrote 0.012 MB perf.data (7 samples) ]
> >   [acme@felicio linux]$ perf report --header-only -I 
> >   # ========
> >   # captured on: Fri Sep 11 11:34:18 2015
> >   # hostname : felicio.ghostprotocols.net
> >   # os release : 4.2.0
> >   # perf version : 4.2.g7e150f
> >   #  arch : x86_64
> >   # nrcpus online : 4
> >   # nrcpus avail : 3
> >   # cpudesc : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
> >   <SNIP>
> >   # node0 meminfo  : total = 8085412 kB, free = 5317596 kB
> >   # node0 cpu list : 0-1,3
> >   <SNIP>
> >   $ 
> 
> Stephane, Namhyung, the bug report below is about the online/avail
> above, they are swapped, full explanation below.
>  
> > We can see multiple bugs here, right? online/avail is swapped, and when
> > online != avail we simply do not record the cpu topology info at all!
>  
> > If we get that CPU back online:
> > 
> >   # echo 1 > /sys/devices/system/cpu/cpu2/online
> > 
> > Then all works:
> > 
> >   $ perf record usleep 1
> >   [ perf record: Woken up 1 times to write data ]
> >   [ perf record: Captured and wrote 0.012 MB perf.data (7 samples) ]
> >   $ perf report --header-only -I
> >   # ========
> >   # captured on: Fri Sep 11 11:37:31 2015
> >   # hostname : felicio.ghostprotocols.net
> >   # os release : 4.2.0
> >   # perf version : 4.2.g7e150f
> >   # arch : x86_64
> >   # nrcpus online : 4
> >   # nrcpus avail : 4
> >   # cpudesc : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
> >   <SNIP>
> >   # sibling cores   : 0-3
> >   # sibling threads : 0
> >   # sibling threads : 1
> >   # sibling threads : 2
> >   # sibling threads : 3
> >   # CPU 0: Core ID 0, Socket ID 0
> >   # CPU 1: Core ID 1, Socket ID 0
> >   # CPU 2: Core ID 2, Socket ID 0
> >   # CPU 3: Core ID 3, Socket ID 0
> >   # node0 meminfo  : total = 8085412 kB, free = 5316992 kB
> >   # node0 cpu list : 0-3
> > 
> > So, again, this was not introduced by this patchkit, but it is good that you did
> > these offline tests, so we can fix it!
> 
> Stephane, this was introduced in:
> 
>   commit fbe96f29ce4b33e0a22219cc7f5996d9157717e3
>   Author: Stephane Eranian <eranian@google.com>
>   Date:   Fri Sep 30 15:40:40 2011 +0200
> 
>     perf tools: Make perf.data more self-descriptive (v8)
> 
> ------------------
> 
> When you write this part:
> 
>       - HEADER_NRCPUS: number of online/avail cpus
> 
> You do:
> 
> static int write_nrcpus(int fd, struct perf_header *h __used,
>                        struct perf_evlist *evlist __used)
> {
>        long nr;
>        u32 nrc, nra;
>        int ret;
> 
>        nr = sysconf(_SC_NPROCESSORS_CONF);
>        if (nr < 0)
>                return -1;
> 
>        nrc = (u32)(nr & UINT_MAX);
> 
>        nr = sysconf(_SC_NPROCESSORS_ONLN);
>        if (nr < 0)
>                return -1;
> 
>        nra = (u32)(nr & UINT_MAX);
> 
>        ret = do_write(fd, &nrc, sizeof(nrc));
>        if (ret < 0)
>                return ret;
> 
>        return do_write(fd, &nra, sizeof(nra));
> }
> 
> I.e. write what you called 'nrc' using what is in SC_NRPROCESSORS_CONF, that in
> the documentation for glibc reads:
> 
> ---------------------
> 
> http://www.gnu.org/software/libc/manual/html_node/Processor-Resources.html
> 
>   sysconf (_SC_NPROCESSORS_CONF)
> 
> which returns the number of processors the operating system configured. But it
> might be possible for the operating system to disable individual processors and
> so the call 
> 
> ---------------------
> 
> Which menas "NR_AVAILABLE", right?
> 
> But then you call a variable 'nra' which sounds like you think that what is in
> _SC_NPROCESSORS_ONLN is the "available" number of CPUs, which is confused a bit
> more by the glibc docs when refering to _SC_NPROCESSORS_ONLN:
> 
> ---------------------
>   sysconf (_SC_NPROCESSORS_ONLN)
> 
> returns the number of processors which are currently online (i.e., available). 
> ---------------------
> 
> Then, when printing the number of CPUs encoded in the perf.data file by the
> above write_nrcpus() routine, you did:
> 
> static void print_nrcpus(struct perf_header *ph, int fd, FILE *fp)
> {
>        ssize_t ret;
>        u32 nr;
> 
>        ret = read(fd, &nr, sizeof(nr));
>        if (ret != (ssize_t)sizeof(nr))
>                nr = -1; /* interpreted as error */
> 
>        if (ph->needs_swap)
>                nr = bswap_32(nr);
> 
>        fprintf(fp, "# nrcpus online : %u\n", nr);
> 
>        ret = read(fd, &nr, sizeof(nr));
>        if (ret != (ssize_t)sizeof(nr))
>                nr = -1; /* interpreted as error */
> 
>        if (ph->needs_swap)
>                nr = bswap_32(nr);
> 
>        fprintf(fp, "# nrcpus avail : %u\n", nr);
> }
> 
> You inverted it, no?
> 
> So, could you please check if the below patch can have your Acked-by?
> Namhyung?

Looks good to me.

Acked-by: Namhyung Kim <namhyung@kernel.org>

Thanks,
Namhyung


> 
> Thanks,
> 
> - Arnaldo
> 
> diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
> index 41814547da15..fce6634aebe2 100644
> --- a/tools/perf/util/header.c
> +++ b/tools/perf/util/header.c
> @@ -1438,7 +1438,7 @@ static int process_nrcpus(struct perf_file_section *section __maybe_unused,
>  	if (ph->needs_swap)
>  		nr = bswap_32(nr);
>  
> -	ph->env.nr_cpus_online = nr;
> +	ph->env.nr_cpus_avail = nr;
>  
>  	ret = readn(fd, &nr, sizeof(nr));
>  	if (ret != sizeof(nr))
> @@ -1447,7 +1447,7 @@ static int process_nrcpus(struct perf_file_section *section __maybe_unused,
>  	if (ph->needs_swap)
>  		nr = bswap_32(nr);
>  
> -	ph->env.nr_cpus_avail = nr;
> +	ph->env.nr_cpus_online = nr;
>  	return 0;
>  }
>  
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222964 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-11 18:40 +0200
Subject	Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q7zwK-78E-33@gated-at.bofh.it>
In reply to	#1222939

Em Sat, Sep 12, 2015 at 01:14:02AM +0900, Namhyung Kim escreveu:
> > You inverted it, no?

> > So, could you please check if the below patch can have your Acked-by?
> > Namhyung?
 
> Looks good to me.
 
> Acked-by: Namhyung Kim <namhyung@kernel.org>

Thanks, added it to the patch, after lunch I should have another patch
for another bug introduced in the same patch, i.e. if one CPU is
offlined, we simply refuse to collect the topology information.

To fix it, I think, we need to insert "(offline)" where one expects to
find the thread and core siblings info.

- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1223962 — [tip:perf/urgent] perf header: Fixup reading of HEADER_NRCPUS feature

From	tip-bot for Arnaldo Carvalho de Melo <tipbot@zytor.com>
Date	2015-09-14 11:20 +0200
Subject	[tip:perf/urgent] perf header: Fixup reading of HEADER_NRCPUS feature
Message-ID	<q8y5A-1Tq-5@gated-at.bofh.it>
In reply to	#1222918

Commit-ID:  caa470475d9b59eeff093ae650800d34612c4379
Gitweb:     http://git.kernel.org/tip/caa470475d9b59eeff093ae650800d34612c4379
Author:     Arnaldo Carvalho de Melo <acme@redhat.com>
AuthorDate: Fri, 11 Sep 2015 12:36:12 -0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Sun, 13 Sep 2015 11:41:34 -0300

perf header: Fixup reading of HEADER_NRCPUS feature

The original patch introducing this header wrote the number of CPUs available
and online in one order and then swapped those values when reading, fix it.

Before:

  # perf record usleep 1
  # perf report --header-only | grep 'nrcpus \(online\|avail\)'
  # nrcpus online : 4
  # nrcpus avail : 4
  # echo 0 > /sys/devices/system/cpu/cpu2/online
  # perf record usleep 1
  # perf report --header-only | grep 'nrcpus \(online\|avail\)'
  # nrcpus online : 4
  # nrcpus avail : 3
  # echo 0 > /sys/devices/system/cpu/cpu1/online
  # perf record usleep 1
  # perf report --header-only | grep 'nrcpus \(online\|avail\)'
  # nrcpus online : 4
  # nrcpus avail : 2

After the fix, bringing back the CPUs online:

  # perf report --header-only | grep 'nrcpus \(online\|avail\)'
  # nrcpus online : 2
  # nrcpus avail : 4
  # echo 1 > /sys/devices/system/cpu/cpu2/online
  # perf record usleep 1
  # perf report --header-only | grep 'nrcpus \(online\|avail\)'
  # nrcpus online : 3
  # nrcpus avail : 4
  # echo 1 > /sys/devices/system/cpu/cpu1/online
  # perf record usleep 1
  # perf report --header-only | grep 'nrcpus \(online\|avail\)'
  # nrcpus online : 4
  # nrcpus avail : 4

Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: fbe96f29ce4b ("perf tools: Make perf.data more self-descriptive (v8)")
Link: http://lkml.kernel.org/r/20150911153323.GP23511@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/header.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 4181454..fce6634 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -1438,7 +1438,7 @@ static int process_nrcpus(struct perf_file_section *section __maybe_unused,
 	if (ph->needs_swap)
 		nr = bswap_32(nr);
 
-	ph->env.nr_cpus_online = nr;
+	ph->env.nr_cpus_avail = nr;
 
 	ret = readn(fd, &nr, sizeof(nr));
 	if (ret != sizeof(nr))
@@ -1447,7 +1447,7 @@ static int process_nrcpus(struct perf_file_section *section __maybe_unused,
 	if (ph->needs_swap)
 		nr = bswap_32(nr);
 
-	ph->env.nr_cpus_avail = nr;
+	ph->env.nr_cpus_online = nr;
 	return 0;
 }
 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1224697 — [tip:perf/core] perf env: Introduce read_cpu_topology_map() method

From	tip-bot for Arnaldo Carvalho de Melo <tipbot@zytor.com>
Date	2015-09-15 09:10 +0200
Subject	[tip:perf/core] perf env: Introduce read_cpu_topology_map() method
Message-ID	<q8Sxk-635-27@gated-at.bofh.it>
In reply to	#1221667

Commit-ID:  aa36ddd7afbb0a3db216c1391e28cd6d80ed1706
Gitweb:     http://git.kernel.org/tip/aa36ddd7afbb0a3db216c1391e28cd6d80ed1706
Author:     Arnaldo Carvalho de Melo <acme@redhat.com>
AuthorDate: Wed, 9 Sep 2015 10:37:01 -0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 14 Sep 2015 12:50:28 -0300

perf env: Introduce read_cpu_topology_map() method

Out of the code to write the cpu topology map in the perf.data file
header.

Now if one needs the CPU topology map for the running machine, one needs
to call perf_env__read_cpu_topology_map(perf_env) and the info will be
stored in perf_env.cpu.

For now we're using a global perf_env variable, that will have its
contents freed after we run a builtin.

v2: Check perf_env__read_cpu_topology_map() return in
    write_cpu_topology() (Kan Liang)

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1441828225-667-5-git-send-email-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/perf.c        |  2 ++
 tools/perf/util/env.c    | 28 ++++++++++++++++++++++++++++
 tools/perf/util/env.h    |  2 ++
 tools/perf/util/header.c | 28 ++++++++++++----------------
 4 files changed, 44 insertions(+), 16 deletions(-)

diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index f2fc019..1fded92 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -8,6 +8,7 @@
  */
 #include "builtin.h"
 
+#include "util/env.h"
 #include "util/exec_cmd.h"
 #include "util/cache.h"
 #include "util/quote.h"
@@ -369,6 +370,7 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
 
 	status = p->fn(argc, argv, prefix);
 	exit_browser(status);
+	perf_env__exit(&perf_env);
 
 	if (status)
 		return status & 0xff;
diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c
index ca1e33a..6af4f7c 100644
--- a/tools/perf/util/env.c
+++ b/tools/perf/util/env.c
@@ -1,3 +1,4 @@
+#include "cpumap.h"
 #include "env.h"
 #include "util.h"
 
@@ -56,3 +57,30 @@ out_free:
 out_enomem:
 	return -ENOMEM;
 }
+
+int perf_env__read_cpu_topology_map(struct perf_env *env)
+{
+	int cpu, nr_cpus;
+
+	if (env->cpu != NULL)
+		return 0;
+
+	if (env->nr_cpus_avail == 0)
+		env->nr_cpus_avail = sysconf(_SC_NPROCESSORS_CONF);
+
+	nr_cpus = env->nr_cpus_avail;
+	if (nr_cpus == -1)
+		return -EINVAL;
+
+	env->cpu = calloc(nr_cpus, sizeof(env->cpu[0]));
+	if (env->cpu == NULL)
+		return -ENOMEM;
+
+	for (cpu = 0; cpu < nr_cpus; ++cpu) {
+		env->cpu[cpu].core_id	= cpu_map__get_core_id(cpu);
+		env->cpu[cpu].socket_id	= cpu_map__get_socket_id(cpu);
+	}
+
+	env->nr_cpus_avail = nr_cpus;
+	return 0;
+}
diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
index d0d1a96..0132b95 100644
--- a/tools/perf/util/env.h
+++ b/tools/perf/util/env.h
@@ -39,4 +39,6 @@ void perf_env__exit(struct perf_env *env);
 
 int perf_env__set_cmdline(struct perf_env *env, int argc, const char *argv[]);
 
+int perf_env__read_cpu_topology_map(struct perf_env *env);
+
 #endif /* __PERF_ENV_H */
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index f307b17..46ec6c5 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -415,8 +415,6 @@ struct cpu_topo {
 	u32 thread_sib;
 	char **core_siblings;
 	char **thread_siblings;
-	int *core_id;
-	int *phy_pkg_id;
 };
 
 static int build_cpu_topo(struct cpu_topo *tp, int cpu)
@@ -479,9 +477,6 @@ try_threads:
 	}
 	ret = 0;
 done:
-	tp->core_id[cpu] = cpu_map__get_core_id(cpu);
-	tp->phy_pkg_id[cpu] = cpu_map__get_socket_id(cpu);
-
 	if(fp)
 		fclose(fp);
 	free(buf);
@@ -509,7 +504,7 @@ static struct cpu_topo *build_cpu_topology(void)
 	struct cpu_topo *tp;
 	void *addr;
 	u32 nr, i;
-	size_t sz, sz_id;
+	size_t sz;
 	long ncpus;
 	int ret = -1;
 
@@ -520,9 +515,8 @@ static struct cpu_topo *build_cpu_topology(void)
 	nr = (u32)(ncpus & UINT_MAX);
 
 	sz = nr * sizeof(char *);
-	sz_id = nr * sizeof(int);
 
-	addr = calloc(1, sizeof(*tp) + 2 * sz + 2 * sz_id);
+	addr = calloc(1, sizeof(*tp) + 2 * sz);
 	if (!addr)
 		return NULL;
 
@@ -532,10 +526,6 @@ static struct cpu_topo *build_cpu_topology(void)
 	tp->core_siblings = addr;
 	addr += sz;
 	tp->thread_siblings = addr;
-	addr += sz;
-	tp->core_id = addr;
-	addr += sz_id;
-	tp->phy_pkg_id = addr;
 
 	for (i = 0; i < nr; i++) {
 		ret = build_cpu_topo(tp, i);
@@ -554,7 +544,7 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
 {
 	struct cpu_topo *tp;
 	u32 i;
-	int ret;
+	int ret, j;
 
 	tp = build_cpu_topology();
 	if (!tp)
@@ -579,11 +569,17 @@ static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
 			break;
 	}
 
-	for (i = 0; i < tp->cpu_nr; i++) {
-		ret = do_write(fd, &tp->core_id[i], sizeof(int));
+	ret = perf_env__read_cpu_topology_map(&perf_env);
+	if (ret < 0)
+		goto done;
+
+	for (j = 0; j < perf_env.nr_cpus_avail; j++) {
+		ret = do_write(fd, &perf_env.cpu[j].core_id,
+			       sizeof(perf_env.cpu[j].core_id));
 		if (ret < 0)
 			return ret;
-		ret = do_write(fd, &tp->phy_pkg_id[i], sizeof(int));
+		ret = do_write(fd, &perf_env.cpu[j].socket_id,
+			       sizeof(perf_env.cpu[j].socket_id));
 		if (ret < 0)
 			return ret;
 	}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221668 — [PATCH 07/13] perf hists browser: Fixup the "cpu" column width calculation

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-09 22:00 +0200
Subject	[PATCH 07/13] perf hists browser: Fixup the "cpu" column width calculation
Message-ID	<q6THc-4vo-13@gated-at.bofh.it>
In reply to	#1221665

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

Since we were not setting it to at least 3 chars ('CPU'), it was being
reset to zero when recalculating the columns width when refreshing the
screen, in 'perf top'. Fix it.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-iqcdnkkqm6sew06x01fbijmy@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/hist.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index 67b48616ab31..b3567a25f0c4 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -146,6 +146,7 @@ void hists__calc_col_len(struct hists *hists, struct hist_entry *h)
 		hists__set_unres_dso_col_len(hists, HISTC_MEM_DADDR_DSO);
 	}
 
+	hists__new_col_len(hists, HISTC_CPU, 3);
 	hists__new_col_len(hists, HISTC_SOCKET, 6);
 	hists__new_col_len(hists, HISTC_MEM_LOCKED, 6);
 	hists__new_col_len(hists, HISTC_MEM_TLB, 22);
-- 
2.1.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222709 — Re: [PATCH 07/13] perf hists browser: Fixup the "cpu" column width calculation

From	"Wangnan (F)" <wangnan0@huawei.com>
Date	2015-09-11 13:00 +0200
Subject	Re: [PATCH 07/13] perf hists browser: Fixup the "cpu" column width calculation
Message-ID	<q7udH-7KK-5@gated-at.bofh.it>
In reply to	#1221668


On 2015/9/10 3:50, Arnaldo Carvalho de Melo wrote:
> From: Arnaldo Carvalho de Melo <acme@redhat.com>
>
> Since we were not setting it to at least 3 chars ('CPU'), it was being
> reset to zero when recalculating the columns width when refreshing the
> screen, in 'perf top'. Fix it.

Tested-by: Wang Nan <wangnan0@huawei.com>

> Cc: Adrian Hunter <adrian.hunter@intel.com>
> Cc: Borislav Petkov <bp@suse.de>
> Cc: David Ahern <dsahern@gmail.com>
> Cc: Frederic Weisbecker <fweisbec@gmail.com>
> Cc: Jiri Olsa <jolsa@redhat.com>
> Cc: Kan Liang <kan.liang@intel.com>
> Cc: Namhyung Kim <namhyung@kernel.org>
> Cc: Stephane Eranian <eranian@google.com>
> Cc: Wang Nan <wangnan0@huawei.com>
> Link: http://lkml.kernel.org/n/tip-iqcdnkkqm6sew06x01fbijmy@git.kernel.org
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> ---
>   tools/perf/util/hist.c | 1 +
>   1 file changed, 1 insertion(+)
>
> diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
> index 67b48616ab31..b3567a25f0c4 100644
> --- a/tools/perf/util/hist.c
> +++ b/tools/perf/util/hist.c
> @@ -146,6 +146,7 @@ void hists__calc_col_len(struct hists *hists, struct hist_entry *h)
>   		hists__set_unres_dso_col_len(hists, HISTC_MEM_DADDR_DSO);
>   	}
>   
> +	hists__new_col_len(hists, HISTC_CPU, 3);
>   	hists__new_col_len(hists, HISTC_SOCKET, 6);
>   	hists__new_col_len(hists, HISTC_MEM_LOCKED, 6);
>   	hists__new_col_len(hists, HISTC_MEM_TLB, 22);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221670 — [PATCH 12/13] perf evsel: Remove forward declaration of 'struct perf_evlist'

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-09 22:00 +0200
Subject	[PATCH 12/13] perf evsel: Remove forward declaration of 'struct perf_evlist'
Message-ID	<q6THc-4vo-17@gated-at.bofh.it>
In reply to	#1221665

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

We have no use for it in evsel.h.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-um03yjrgyi3bj1hzqiqs4dsu@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evsel.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 93ac6b128149..306cfef044e3 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -125,7 +125,6 @@ union u64_swap {
 struct cpu_map;
 struct target;
 struct thread_map;
-struct perf_evlist;
 struct record_opts;
 
 static inline struct cpu_map *perf_evsel__cpus(struct perf_evsel *evsel)
-- 
2.1.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221671 — [PATCH 02/13] perf env: Rename some leftovers from rename to perf_env

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-09 22:00 +0200
Subject	[PATCH 02/13] perf env: Rename some leftovers from rename to perf_env
Message-ID	<q6THc-4vo-23@gated-at.bofh.it>
In reply to	#1221665

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

In ce80d3bef9ff ("perf tools: Rename perf_session_env to perf_env") we
forgot to rename a few functions to the "perf_env" prefix, do it now.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-b3ui3z6ock89z1814pu2er98@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/arch/common.c       | 10 ++++------
 tools/perf/arch/common.h       |  4 ++--
 tools/perf/builtin-annotate.c  |  2 +-
 tools/perf/builtin-top.c       |  2 +-
 tools/perf/ui/browsers/hists.c |  2 +-
 5 files changed, 9 insertions(+), 11 deletions(-)

diff --git a/tools/perf/arch/common.c b/tools/perf/arch/common.c
index b00dfd92ea73..e83c8ce24303 100644
--- a/tools/perf/arch/common.c
+++ b/tools/perf/arch/common.c
@@ -128,9 +128,8 @@ static const char *normalize_arch(char *arch)
 	return arch;
 }
 
-static int perf_session_env__lookup_binutils_path(struct perf_env *env,
-						  const char *name,
-						  const char **path)
+static int perf_env__lookup_binutils_path(struct perf_env *env,
+					  const char *name, const char **path)
 {
 	int idx;
 	const char *arch, *cross_env;
@@ -206,7 +205,7 @@ out_error:
 	return -1;
 }
 
-int perf_session_env__lookup_objdump(struct perf_env *env)
+int perf_env__lookup_objdump(struct perf_env *env)
 {
 	/*
 	 * For live mode, env->arch will be NULL and we can use
@@ -215,6 +214,5 @@ int perf_session_env__lookup_objdump(struct perf_env *env)
 	if (env->arch == NULL)
 		return 0;
 
-	return perf_session_env__lookup_binutils_path(env, "objdump",
-						      &objdump_path);
+	return perf_env__lookup_binutils_path(env, "objdump", &objdump_path);
 }
diff --git a/tools/perf/arch/common.h b/tools/perf/arch/common.h
index 20176df69fc8..7529cfb143ce 100644
--- a/tools/perf/arch/common.h
+++ b/tools/perf/arch/common.h
@@ -1,10 +1,10 @@
 #ifndef ARCH_PERF_COMMON_H
 #define ARCH_PERF_COMMON_H
 
-#include "../util/session.h"
+#include "../util/env.h"
 
 extern const char *objdump_path;
 
-int perf_session_env__lookup_objdump(struct perf_env *env);
+int perf_env__lookup_objdump(struct perf_env *env);
 
 #endif /* ARCH_PERF_COMMON_H */
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 8edc205ff9a7..2bf9b3fd9e61 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -211,7 +211,7 @@ static int __cmd_annotate(struct perf_annotate *ann)
 	}
 
 	if (!objdump_path) {
-		ret = perf_session_env__lookup_objdump(&session->header.env);
+		ret = perf_env__lookup_objdump(&session->header.env);
 		if (ret)
 			goto out;
 	}
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 8c465c83aabf..e5ca6848f01d 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -952,7 +952,7 @@ static int __cmd_top(struct perf_top *top)
 	machines__set_symbol_filter(&top->session->machines, symbol_filter);
 
 	if (!objdump_path) {
-		ret = perf_session_env__lookup_objdump(&top->session->header.env);
+		ret = perf_env__lookup_objdump(&top->session->header.env);
 		if (ret)
 			goto out_delete;
 	}
diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
index e4fd40f72b4a..fea29fbb9d47 100644
--- a/tools/perf/ui/browsers/hists.c
+++ b/tools/perf/ui/browsers/hists.c
@@ -1442,7 +1442,7 @@ do_annotate(struct hist_browser *browser, struct popup_action *act)
 	struct hist_entry *he;
 	int err;
 
-	if (!objdump_path && perf_session_env__lookup_objdump(browser->env))
+	if (!objdump_path && perf_env__lookup_objdump(browser->env))
 		return 0;
 
 	notes = symbol__annotation(act->ms.sym);
-- 
2.1.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221674 — [PATCH 01/13] perf env: Move perf_env out of header.h and session.c into separate object

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-09 22:00 +0200
Subject	[PATCH 01/13] perf env: Move perf_env out of header.h and session.c into separate object
Message-ID	<q6THd-4vo-29@gated-at.bofh.it>
In reply to	#1221665

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

Since it can be used separately from 'perf_session' and 'perf_header',
move it to separate include file and object, next csets will try to move
a perf_env__init() routine.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-ff2rw99tsn670y1b6gxbwdsi@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/Build     |  1 +
 tools/perf/util/env.c     | 19 +++++++++++++++++++
 tools/perf/util/env.h     | 37 +++++++++++++++++++++++++++++++++++++
 tools/perf/util/header.h  | 33 +--------------------------------
 tools/perf/util/session.c | 20 +-------------------
 5 files changed, 59 insertions(+), 51 deletions(-)
 create mode 100644 tools/perf/util/env.c
 create mode 100644 tools/perf/util/env.h

diff --git a/tools/perf/util/Build b/tools/perf/util/Build
index 349bc96ca1fe..4bc7a9ab45b1 100644
--- a/tools/perf/util/Build
+++ b/tools/perf/util/Build
@@ -5,6 +5,7 @@ libperf-y += build-id.o
 libperf-y += config.o
 libperf-y += ctype.o
 libperf-y += db-export.o
+libperf-y += env.o
 libperf-y += environment.o
 libperf-y += event.o
 libperf-y += evlist.o
diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c
new file mode 100644
index 000000000000..0b3e1b2e5263
--- /dev/null
+++ b/tools/perf/util/env.c
@@ -0,0 +1,19 @@
+#include "env.h"
+#include "util.h"
+
+void perf_env__exit(struct perf_env *env)
+{
+	zfree(&env->hostname);
+	zfree(&env->os_release);
+	zfree(&env->version);
+	zfree(&env->arch);
+	zfree(&env->cpu_desc);
+	zfree(&env->cpuid);
+	zfree(&env->cmdline);
+	zfree(&env->cmdline_argv);
+	zfree(&env->sibling_cores);
+	zfree(&env->sibling_threads);
+	zfree(&env->numa_nodes);
+	zfree(&env->pmu_mappings);
+	zfree(&env->cpu);
+}
diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h
new file mode 100644
index 000000000000..b1370516d99a
--- /dev/null
+++ b/tools/perf/util/env.h
@@ -0,0 +1,37 @@
+#ifndef __PERF_ENV_H
+#define __PERF_ENV_H
+
+struct cpu_topology_map {
+	int	socket_id;
+	int	core_id;
+};
+
+struct perf_env {
+	char			*hostname;
+	char			*os_release;
+	char			*version;
+	char			*arch;
+	int			nr_cpus_online;
+	int			nr_cpus_avail;
+	char			*cpu_desc;
+	char			*cpuid;
+	unsigned long long	total_mem;
+
+	int			nr_cmdline;
+	int			nr_sibling_cores;
+	int			nr_sibling_threads;
+	int			nr_numa_nodes;
+	int			nr_pmu_mappings;
+	int			nr_groups;
+	char			*cmdline;
+	const char		**cmdline_argv;
+	char			*sibling_cores;
+	char			*sibling_threads;
+	char			*numa_nodes;
+	char			*pmu_mappings;
+	struct cpu_topology_map	*cpu;
+};
+
+void perf_env__exit(struct perf_env *env);
+
+#endif /* __PERF_ENV_H */
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
index 975d803f46c8..05f27cb6b7e3 100644
--- a/tools/perf/util/header.h
+++ b/tools/perf/util/header.h
@@ -7,7 +7,7 @@
 #include <linux/bitmap.h>
 #include <linux/types.h>
 #include "event.h"
-
+#include "env.h"
 
 enum {
 	HEADER_RESERVED		= 0,	/* always cleared */
@@ -66,37 +66,6 @@ struct perf_header;
 int perf_file_header__read(struct perf_file_header *header,
 			   struct perf_header *ph, int fd);
 
-struct cpu_topology_map {
-	int	socket_id;
-	int	core_id;
-};
-
-struct perf_env {
-	char			*hostname;
-	char			*os_release;
-	char			*version;
-	char			*arch;
-	int			nr_cpus_online;
-	int			nr_cpus_avail;
-	char			*cpu_desc;
-	char			*cpuid;
-	unsigned long long	total_mem;
-
-	int			nr_cmdline;
-	int			nr_sibling_cores;
-	int			nr_sibling_threads;
-	int			nr_numa_nodes;
-	int			nr_pmu_mappings;
-	int			nr_groups;
-	char			*cmdline;
-	const char		**cmdline_argv;
-	char			*sibling_cores;
-	char			*sibling_threads;
-	char			*numa_nodes;
-	char			*pmu_mappings;
-	struct cpu_topology_map	*cpu;
-};
-
 struct perf_header {
 	enum perf_header_version	version;
 	bool				needs_swap;
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 23fed17307ff..728cb115fbb8 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -170,31 +170,13 @@ static void perf_session__delete_threads(struct perf_session *session)
 	machine__delete_threads(&session->machines.host);
 }
 
-static void perf_session_env__exit(struct perf_env *env)
-{
-	zfree(&env->hostname);
-	zfree(&env->os_release);
-	zfree(&env->version);
-	zfree(&env->arch);
-	zfree(&env->cpu_desc);
-	zfree(&env->cpuid);
-
-	zfree(&env->cmdline);
-	zfree(&env->cmdline_argv);
-	zfree(&env->sibling_cores);
-	zfree(&env->sibling_threads);
-	zfree(&env->numa_nodes);
-	zfree(&env->pmu_mappings);
-	zfree(&env->cpu);
-}
-
 void perf_session__delete(struct perf_session *session)
 {
 	auxtrace__free(session);
 	auxtrace_index__free(&session->auxtrace_index);
 	perf_session__destroy_kernel_maps(session);
 	perf_session__delete_threads(session);
-	perf_session_env__exit(&session->header.env);
+	perf_env__exit(&session->header.env);
 	machines__exit(&session->machines);
 	if (session->file)
 		perf_data_file__close(session->file);
-- 
2.1.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221677 — [PATCH 13/13] Revert "perf evlist: Add backpointer for perf_env to evlist"

From	Arnaldo Carvalho de Melo <acme@kernel.org>
Date	2015-09-09 22:00 +0200
Subject	[PATCH 13/13] Revert "perf evlist: Add backpointer for perf_env to evlist"
Message-ID	<q6THd-4vo-39@gated-at.bofh.it>
In reply to	#1221665

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

This reverts commit 2c07144dfce366e21465cc7b0ada9f0b6dc7b7ed.

We don't need it, machine->env provides what is needed.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.h | 1 -
 tools/perf/util/header.c | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index b39a6198f4ac..436e358300b1 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -56,7 +56,6 @@ struct perf_evlist {
 	struct cpu_map	  *cpus;
 	struct perf_evsel *selected;
 	struct events_stats stats;
-	struct perf_env	*env;
 };
 
 struct perf_evsel_str_handler {
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 085bbc35c186..64adbff36e04 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -2558,8 +2558,8 @@ int perf_session__read_header(struct perf_session *session)
 	if (session->evlist == NULL)
 		return -ENOMEM;
 
-	session->evlist->env = &header->env;
 	session->machines.host.env = &header->env;
+
 	if (perf_data_file__is_pipe(file))
 		return perf_header__read_pipe(session);
 
-- 
2.1.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

Page 1 of 2 [1] 2 Next page →

csiph-web

[RFC 00/13] perf_env/CPU socket reorg/fixes

Contents

#1221665 — [RFC 00/13] perf_env/CPU socket reorg/fixes

#1221667 — [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1221749 — RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222183 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222379 — RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222380 — RE: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222381 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222699 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222891 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222918 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222939 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1222964 — Re: [PATCH 04/13] perf env: Introduce read_cpu_topology_map() method

#1223962 — [tip:perf/urgent] perf header: Fixup reading of HEADER_NRCPUS feature

#1224697 — [tip:perf/core] perf env: Introduce read_cpu_topology_map() method

#1221668 — [PATCH 07/13] perf hists browser: Fixup the "cpu" column width calculation

#1222709 — Re: [PATCH 07/13] perf hists browser: Fixup the "cpu" column width calculation

#1221670 — [PATCH 12/13] perf evsel: Remove forward declaration of 'struct perf_evlist'

#1221671 — [PATCH 02/13] perf env: Rename some leftovers from rename to perf_env

#1221674 — [PATCH 01/13] perf env: Move perf_env out of header.h and session.c into separate object

#1221677 — [PATCH 13/13] Revert "perf evlist: Add backpointer for perf_env to evlist"