linux/tools/perf/util/cpumap.c

622 lines
12 KiB
C
Raw Normal View History

License cleanup: add SPDX GPL-2.0 license identifier to files with no license Many source files in the tree are missing licensing information, which makes it harder for compliance tools to determine the correct license. By default all files without license information are under the default license of the kernel, which is GPL version 2. Update the files which contain no license information with the 'GPL-2.0' SPDX license identifier. The SPDX identifier is a legally binding shorthand, which can be used instead of the full boiler plate text. This patch is based on work done by Thomas Gleixner and Kate Stewart and Philippe Ombredanne. How this work was done: Patches were generated and checked against linux-4.14-rc6 for a subset of the use cases: - file had no licensing information it it. - file was a */uapi/* one with no licensing information in it, - file was a */uapi/* one with existing licensing information, Further patches will be generated in subsequent months to fix up cases where non-standard license headers were used, and references to license had to be inferred by heuristics based on keywords. The analysis to determine which SPDX License Identifier to be applied to a file was done in a spreadsheet of side by side results from of the output of two independent scanners (ScanCode & Windriver) producing SPDX tag:value files created by Philippe Ombredanne. Philippe prepared the base worksheet, and did an initial spot review of a few 1000 files. The 4.13 kernel was the starting point of the analysis with 60,537 files assessed. Kate Stewart did a file by file comparison of the scanner results in the spreadsheet to determine which SPDX license identifier(s) to be applied to the file. She confirmed any determination that was not immediately clear with lawyers working with the Linux Foundation. Criteria used to select files for SPDX license identifier tagging was: - Files considered eligible had to be source code files. - Make and config files were included as candidates if they contained >5 lines of source - File already had some variant of a license header in it (even if <5 lines). All documentation files were explicitly excluded. The following heuristics were used to determine which SPDX license identifiers to apply. - when both scanners couldn't find any license traces, file was considered to have no license information in it, and the top level COPYING file license applied. For non */uapi/* files that summary was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 11139 and resulted in the first patch in this series. If that file was a */uapi/* path one, it was "GPL-2.0 WITH Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 WITH Linux-syscall-note 930 and resulted in the second patch in this series. - if a file had some form of licensing information in it, and was one of the */uapi/* ones, it was denoted with the Linux-syscall-note if any GPL family license was found in the file or had no licensing in it (per prior point). Results summary: SPDX license identifier # files ---------------------------------------------------|------ GPL-2.0 WITH Linux-syscall-note 270 GPL-2.0+ WITH Linux-syscall-note 169 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17 LGPL-2.1+ WITH Linux-syscall-note 15 GPL-1.0+ WITH Linux-syscall-note 14 ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5 LGPL-2.0+ WITH Linux-syscall-note 4 LGPL-2.1 WITH Linux-syscall-note 3 ((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3 ((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1 and that resulted in the third patch in this series. - when the two scanners agreed on the detected license(s), that became the concluded license(s). - when there was disagreement between the two scanners (one detected a license but the other didn't, or they both detected different licenses) a manual inspection of the file occurred. - In most cases a manual inspection of the information in the file resulted in a clear resolution of the license that should apply (and which scanner probably needed to revisit its heuristics). - When it was not immediately clear, the license identifier was confirmed with lawyers working with the Linux Foundation. - If there was any question as to the appropriate license identifier, the file was flagged for further research and to be revisited later in time. In total, over 70 hours of logged manual review was done on the spreadsheet to determine the SPDX license identifiers to apply to the source files by Kate, Philippe, Thomas and, in some cases, confirmation by lawyers working with the Linux Foundation. Kate also obtained a third independent scan of the 4.13 code base from FOSSology, and compared selected files where the other two scanners disagreed against that SPDX file, to see if there was new insights. The Windriver scanner is based on an older version of FOSSology in part, so they are related. Thomas did random spot checks in about 500 files from the spreadsheets for the uapi headers and agreed with SPDX license identifier in the files he inspected. For the non-uapi files Thomas did random spot checks in about 15000 files. In initial set of patches against 4.14-rc6, 3 files were found to have copy/paste license identifier errors, and have been fixed to reflect the correct identifier. Additionally Philippe spent 10 hours this week doing a detailed manual inspection and review of the 12,461 patched files from the initial patch version early this week with: - a full scancode scan run, collecting the matched texts, detected license ids and scores - reviewing anything where there was a license detected (about 500+ files) to ensure that the applied SPDX license was correct - reviewing anything where there was no detection but the patch license was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied SPDX license was correct This produced a worksheet with 20 files needing minor correction. This worksheet was then exported into 3 different .csv files for the different types of files to be modified. These .csv files were then reviewed by Greg. Thomas wrote a script to parse the csv files and add the proper SPDX tag to the file, in the format that the file expected. This script was further refined by Greg based on the output to detect more types of files automatically and to distinguish between header and source .c files (which need different comment types.) Finally Greg ran the script using the .csv files to generate the patches. Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org> Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-11-01 17:07:57 +03:00
// SPDX-License-Identifier: GPL-2.0
#include <api/fs/fs.h>
perf tools: Fix sparse CPU numbering related bugs At present, the perf subcommands that do system-wide monitoring (perf stat, perf record and perf top) don't work properly unless the online cpus are numbered 0, 1, ..., N-1. These tools ask for the number of online cpus with sysconf(_SC_NPROCESSORS_ONLN) and then try to create events for cpus 0, 1, ..., N-1. This creates problems for systems where the online cpus are numbered sparsely. For example, a POWER6 system in single-threaded mode (i.e. only running 1 hardware thread per core) will have only even-numbered cpus online. This fixes the problem by reading the /sys/devices/system/cpu/online file to find out which cpus are online. The code that does that is in tools/perf/util/cpumap.[ch], and consists of a read_cpu_map() function that sets up a cpumap[] array and returns the number of online cpus. If /sys/devices/system/cpu/online can't be read or can't be parsed successfully, it falls back to using sysconf to ask how many cpus are online and sets up an identity map in cpumap[]. The perf record, perf stat and perf top code then calls read_cpu_map() in the system-wide monitoring case (instead of sysconf) and uses cpumap[] to get the cpu numbers to pass to perf_event_open. Signed-off-by: Paul Mackerras <paulus@samba.org> Cc: Anton Blanchard <anton@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@infradead.org> LKML-Reference: <20100310093609.GA3959@brick.ozlabs.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-03-10 12:36:09 +03:00
#include "cpumap.h"
#include "debug.h"
#include "event.h"
perf tools: Fix sparse CPU numbering related bugs At present, the perf subcommands that do system-wide monitoring (perf stat, perf record and perf top) don't work properly unless the online cpus are numbered 0, 1, ..., N-1. These tools ask for the number of online cpus with sysconf(_SC_NPROCESSORS_ONLN) and then try to create events for cpus 0, 1, ..., N-1. This creates problems for systems where the online cpus are numbered sparsely. For example, a POWER6 system in single-threaded mode (i.e. only running 1 hardware thread per core) will have only even-numbered cpus online. This fixes the problem by reading the /sys/devices/system/cpu/online file to find out which cpus are online. The code that does that is in tools/perf/util/cpumap.[ch], and consists of a read_cpu_map() function that sets up a cpumap[] array and returns the number of online cpus. If /sys/devices/system/cpu/online can't be read or can't be parsed successfully, it falls back to using sysconf to ask how many cpus are online and sets up an identity map in cpumap[]. The perf record, perf stat and perf top code then calls read_cpu_map() in the system-wide monitoring case (instead of sysconf) and uses cpumap[] to get the cpu numbers to pass to perf_event_open. Signed-off-by: Paul Mackerras <paulus@samba.org> Cc: Anton Blanchard <anton@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@infradead.org> LKML-Reference: <20100310093609.GA3959@brick.ozlabs.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-03-10 12:36:09 +03:00
#include <assert.h>
#include <dirent.h>
perf tools: Fix sparse CPU numbering related bugs At present, the perf subcommands that do system-wide monitoring (perf stat, perf record and perf top) don't work properly unless the online cpus are numbered 0, 1, ..., N-1. These tools ask for the number of online cpus with sysconf(_SC_NPROCESSORS_ONLN) and then try to create events for cpus 0, 1, ..., N-1. This creates problems for systems where the online cpus are numbered sparsely. For example, a POWER6 system in single-threaded mode (i.e. only running 1 hardware thread per core) will have only even-numbered cpus online. This fixes the problem by reading the /sys/devices/system/cpu/online file to find out which cpus are online. The code that does that is in tools/perf/util/cpumap.[ch], and consists of a read_cpu_map() function that sets up a cpumap[] array and returns the number of online cpus. If /sys/devices/system/cpu/online can't be read or can't be parsed successfully, it falls back to using sysconf to ask how many cpus are online and sets up an identity map in cpumap[]. The perf record, perf stat and perf top code then calls read_cpu_map() in the system-wide monitoring case (instead of sysconf) and uses cpumap[] to get the cpu numbers to pass to perf_event_open. Signed-off-by: Paul Mackerras <paulus@samba.org> Cc: Anton Blanchard <anton@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@infradead.org> LKML-Reference: <20100310093609.GA3959@brick.ozlabs.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-03-10 12:36:09 +03:00
#include <stdio.h>
#include <stdlib.h>
#include <linux/bitmap.h>
#include "asm/bug.h"
perf tools: Fix sparse CPU numbering related bugs At present, the perf subcommands that do system-wide monitoring (perf stat, perf record and perf top) don't work properly unless the online cpus are numbered 0, 1, ..., N-1. These tools ask for the number of online cpus with sysconf(_SC_NPROCESSORS_ONLN) and then try to create events for cpus 0, 1, ..., N-1. This creates problems for systems where the online cpus are numbered sparsely. For example, a POWER6 system in single-threaded mode (i.e. only running 1 hardware thread per core) will have only even-numbered cpus online. This fixes the problem by reading the /sys/devices/system/cpu/online file to find out which cpus are online. The code that does that is in tools/perf/util/cpumap.[ch], and consists of a read_cpu_map() function that sets up a cpumap[] array and returns the number of online cpus. If /sys/devices/system/cpu/online can't be read or can't be parsed successfully, it falls back to using sysconf to ask how many cpus are online and sets up an identity map in cpumap[]. The perf record, perf stat and perf top code then calls read_cpu_map() in the system-wide monitoring case (instead of sysconf) and uses cpumap[] to get the cpu numbers to pass to perf_event_open. Signed-off-by: Paul Mackerras <paulus@samba.org> Cc: Anton Blanchard <anton@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@infradead.org> LKML-Reference: <20100310093609.GA3959@brick.ozlabs.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-03-10 12:36:09 +03:00
#include <linux/ctype.h>
#include <linux/zalloc.h>
static int max_cpu_num;
static int max_present_cpu_num;
static int max_node_num;
/**
* The numa node X as read from /sys/devices/system/node/nodeX indexed by the
* CPU number.
*/
static int *cpunode_map;
static struct perf_cpu_map *cpu_map__from_entries(struct cpu_map_entries *cpus)
{
struct perf_cpu_map *map;
map = perf_cpu_map__empty_new(cpus->nr);
if (map) {
unsigned i;
for (i = 0; i < cpus->nr; i++) {
/*
* Special treatment for -1, which is not real cpu number,
* and we need to use (int) -1 to initialize map[i],
* otherwise it would become 65535.
*/
if (cpus->cpu[i] == (u16) -1)
map->map[i] = -1;
else
map->map[i] = (int) cpus->cpu[i];
}
}
return map;
}
static struct perf_cpu_map *cpu_map__from_mask(struct perf_record_record_cpu_map *mask)
{
struct perf_cpu_map *map;
int nr, nbits = mask->nr * mask->long_size * BITS_PER_BYTE;
nr = bitmap_weight(mask->mask, nbits);
map = perf_cpu_map__empty_new(nr);
if (map) {
int cpu, i = 0;
for_each_set_bit(cpu, mask->mask, nbits)
map->map[i++] = cpu;
}
return map;
}
struct perf_cpu_map *cpu_map__new_data(struct perf_record_cpu_map_data *data)
{
if (data->type == PERF_CPU_MAP__CPUS)
return cpu_map__from_entries((struct cpu_map_entries *)data->data);
else
return cpu_map__from_mask((struct perf_record_record_cpu_map *)data->data);
}
size_t cpu_map__fprintf(struct perf_cpu_map *map, FILE *fp)
{
#define BUFSIZE 1024
char buf[BUFSIZE];
cpu_map__snprint(map, buf, sizeof(buf));
return fprintf(fp, "%s\n", buf);
#undef BUFSIZE
}
struct perf_cpu_map *perf_cpu_map__empty_new(int nr)
{
struct perf_cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int) * nr);
if (cpus != NULL) {
int i;
cpus->nr = nr;
for (i = 0; i < nr; i++)
cpus->map[i] = -1;
refcount_set(&cpus->refcnt, 1);
}
return cpus;
}
struct cpu_aggr_map *cpu_aggr_map__empty_new(int nr)
{
struct cpu_aggr_map *cpus = malloc(sizeof(*cpus) + sizeof(struct aggr_cpu_id) * nr);
if (cpus != NULL) {
int i;
cpus->nr = nr;
for (i = 0; i < nr; i++)
cpus->map[i] = aggr_cpu_id__empty();
refcount_set(&cpus->refcnt, 1);
}
return cpus;
}
static int cpu__get_topology_int(int cpu, const char *name, int *value)
{
char path[PATH_MAX];
snprintf(path, PATH_MAX,
"devices/system/cpu/cpu%d/topology/%s", cpu, name);
return sysfs__read_int(path, value);
}
int cpu__get_socket_id(int cpu)
{
int value, ret = cpu__get_topology_int(cpu, "physical_package_id", &value);
return ret ?: value;
}
struct aggr_cpu_id aggr_cpu_id__socket(int cpu, void *data __maybe_unused)
{
struct aggr_cpu_id id = aggr_cpu_id__empty();
id.socket = cpu__get_socket_id(cpu);
return id;
}
static int aggr_cpu_id__cmp(const void *a_pointer, const void *b_pointer)
{
struct aggr_cpu_id *a = (struct aggr_cpu_id *)a_pointer;
struct aggr_cpu_id *b = (struct aggr_cpu_id *)b_pointer;
if (a->node != b->node)
return a->node - b->node;
else if (a->socket != b->socket)
perf stat aggregation: Add separate socket member Add socket as a separate member so that it doesn't have to be packed into the int value. When the socket ID was larger than 8 bits the output appeared corrupted or incomplete. For example, here on ThunderX2 'perf stat' reports a socket of -1 and an invalid die number: ./perf stat -a --per-die The socket id number is too big. Performance counter stats for 'system wide': S-1-D255 128 687.99 msec cpu-clock # 57.240 CPUs utilized ... S36-D0 128 842.34 msec cpu-clock # 70.081 CPUs utilized ... And with --per-core there is an entry with an invalid core ID: ./perf stat record -a --per-core The socket id number is too big. Performance counter stats for 'system wide': S-1-D255-C65535 128 671.04 msec cpu-clock # 54.112 CPUs utilized ... S36-D0-C0 4 28.27 msec cpu-clock # 2.279 CPUs utilized ... This fixes the "Session topology" self test on ThunderX2. After this fix the output contains the correct socket and die IDs and no longer prints a warning about the size of the socket ID: ./perf stat --per-die -a Performance counter stats for 'system wide': S36-D0 128 169,869.39 msec cpu-clock # 127.501 CPUs utilized ... S3612-D0 128 169,733.05 msec cpu-clock # 127.398 CPUs utilized Signed-off-by: James Clark <james.clark@arm.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Jiri Olsa <jolsa@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: John Garry <john.garry@huawei.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Richter <tmricht@linux.ibm.com> Link: https://lore.kernel.org/r/20201126141328.6509-10-james.clark@arm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2020-11-26 17:13:25 +03:00
return a->socket - b->socket;
else if (a->die != b->die)
return a->die - b->die;
else if (a->core != b->core)
return a->core - b->core;
else
return a->thread - b->thread;
}
struct cpu_aggr_map *cpu_aggr_map__new(const struct perf_cpu_map *cpus,
aggr_cpu_id_get_t get_id,
void *data)
{
int cpu, idx;
struct cpu_aggr_map *c = cpu_aggr_map__empty_new(cpus->nr);
if (!c)
return NULL;
/* Reset size as it may only be partially filled */
c->nr = 0;
perf_cpu_map__for_each_cpu(cpu, idx, cpus) {
bool duplicate = false;
struct aggr_cpu_id cpu_id = get_id(cpu, data);
for (int j = 0; j < c->nr; j++) {
if (aggr_cpu_id__equal(&cpu_id, &c->map[j])) {
duplicate = true;
break;
}
}
if (!duplicate) {
c->map[c->nr] = cpu_id;
c->nr++;
}
}
/* Trim. */
if (c->nr != cpus->nr) {
struct cpu_aggr_map *trimmed_c =
realloc(c,
sizeof(struct cpu_aggr_map) + sizeof(struct aggr_cpu_id) * c->nr);
if (trimmed_c)
c = trimmed_c;
}
/* ensure we process id in increasing order */
qsort(c->map, c->nr, sizeof(struct aggr_cpu_id), aggr_cpu_id__cmp);
return c;
}
int cpu__get_die_id(int cpu)
{
int value, ret = cpu__get_topology_int(cpu, "die_id", &value);
return ret ?: value;
}
struct aggr_cpu_id aggr_cpu_id__die(int cpu, void *data)
{
struct aggr_cpu_id id;
int die;
die = cpu__get_die_id(cpu);
/* There is no die_id on legacy system. */
perf stat aggregation: Add separate socket member Add socket as a separate member so that it doesn't have to be packed into the int value. When the socket ID was larger than 8 bits the output appeared corrupted or incomplete. For example, here on ThunderX2 'perf stat' reports a socket of -1 and an invalid die number: ./perf stat -a --per-die The socket id number is too big. Performance counter stats for 'system wide': S-1-D255 128 687.99 msec cpu-clock # 57.240 CPUs utilized ... S36-D0 128 842.34 msec cpu-clock # 70.081 CPUs utilized ... And with --per-core there is an entry with an invalid core ID: ./perf stat record -a --per-core The socket id number is too big. Performance counter stats for 'system wide': S-1-D255-C65535 128 671.04 msec cpu-clock # 54.112 CPUs utilized ... S36-D0-C0 4 28.27 msec cpu-clock # 2.279 CPUs utilized ... This fixes the "Session topology" self test on ThunderX2. After this fix the output contains the correct socket and die IDs and no longer prints a warning about the size of the socket ID: ./perf stat --per-die -a Performance counter stats for 'system wide': S36-D0 128 169,869.39 msec cpu-clock # 127.501 CPUs utilized ... S3612-D0 128 169,733.05 msec cpu-clock # 127.398 CPUs utilized Signed-off-by: James Clark <james.clark@arm.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Jiri Olsa <jolsa@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: John Garry <john.garry@huawei.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Richter <tmricht@linux.ibm.com> Link: https://lore.kernel.org/r/20201126141328.6509-10-james.clark@arm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2020-11-26 17:13:25 +03:00
if (die == -1)
die = 0;
/*
perf stat aggregation: Add separate socket member Add socket as a separate member so that it doesn't have to be packed into the int value. When the socket ID was larger than 8 bits the output appeared corrupted or incomplete. For example, here on ThunderX2 'perf stat' reports a socket of -1 and an invalid die number: ./perf stat -a --per-die The socket id number is too big. Performance counter stats for 'system wide': S-1-D255 128 687.99 msec cpu-clock # 57.240 CPUs utilized ... S36-D0 128 842.34 msec cpu-clock # 70.081 CPUs utilized ... And with --per-core there is an entry with an invalid core ID: ./perf stat record -a --per-core The socket id number is too big. Performance counter stats for 'system wide': S-1-D255-C65535 128 671.04 msec cpu-clock # 54.112 CPUs utilized ... S36-D0-C0 4 28.27 msec cpu-clock # 2.279 CPUs utilized ... This fixes the "Session topology" self test on ThunderX2. After this fix the output contains the correct socket and die IDs and no longer prints a warning about the size of the socket ID: ./perf stat --per-die -a Performance counter stats for 'system wide': S36-D0 128 169,869.39 msec cpu-clock # 127.501 CPUs utilized ... S3612-D0 128 169,733.05 msec cpu-clock # 127.398 CPUs utilized Signed-off-by: James Clark <james.clark@arm.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Jiri Olsa <jolsa@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: John Garry <john.garry@huawei.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Richter <tmricht@linux.ibm.com> Link: https://lore.kernel.org/r/20201126141328.6509-10-james.clark@arm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2020-11-26 17:13:25 +03:00
* die_id is relative to socket, so start
* with the socket ID and then add die to
* make a unique ID.
*/
id = aggr_cpu_id__socket(cpu, data);
if (aggr_cpu_id__is_empty(&id))
perf stat aggregation: Add separate socket member Add socket as a separate member so that it doesn't have to be packed into the int value. When the socket ID was larger than 8 bits the output appeared corrupted or incomplete. For example, here on ThunderX2 'perf stat' reports a socket of -1 and an invalid die number: ./perf stat -a --per-die The socket id number is too big. Performance counter stats for 'system wide': S-1-D255 128 687.99 msec cpu-clock # 57.240 CPUs utilized ... S36-D0 128 842.34 msec cpu-clock # 70.081 CPUs utilized ... And with --per-core there is an entry with an invalid core ID: ./perf stat record -a --per-core The socket id number is too big. Performance counter stats for 'system wide': S-1-D255-C65535 128 671.04 msec cpu-clock # 54.112 CPUs utilized ... S36-D0-C0 4 28.27 msec cpu-clock # 2.279 CPUs utilized ... This fixes the "Session topology" self test on ThunderX2. After this fix the output contains the correct socket and die IDs and no longer prints a warning about the size of the socket ID: ./perf stat --per-die -a Performance counter stats for 'system wide': S36-D0 128 169,869.39 msec cpu-clock # 127.501 CPUs utilized ... S3612-D0 128 169,733.05 msec cpu-clock # 127.398 CPUs utilized Signed-off-by: James Clark <james.clark@arm.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Jiri Olsa <jolsa@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: John Garry <john.garry@huawei.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Richter <tmricht@linux.ibm.com> Link: https://lore.kernel.org/r/20201126141328.6509-10-james.clark@arm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2020-11-26 17:13:25 +03:00
return id;
id.die = die;
return id;
}
int cpu__get_core_id(int cpu)
{
int value, ret = cpu__get_topology_int(cpu, "core_id", &value);
return ret ?: value;
}
struct aggr_cpu_id aggr_cpu_id__core(int cpu, void *data)
{
struct aggr_cpu_id id;
int core = cpu__get_core_id(cpu);
/* aggr_cpu_id__die returns a struct with socket and die set. */
id = aggr_cpu_id__die(cpu, data);
if (aggr_cpu_id__is_empty(&id))
return id;
/*
* core_id is relative to socket and die, we need a global id.
* So we combine the result from cpu_map__get_die with the core id
*/
id.core = core;
return id;
}
struct aggr_cpu_id aggr_cpu_id__cpu(int cpu, void *data)
{
struct aggr_cpu_id id;
/* aggr_cpu_id__core returns a struct with socket, die and core set. */
id = aggr_cpu_id__core(cpu, data);
if (aggr_cpu_id__is_empty(&id))
return id;
id.cpu = cpu;
return id;
}
struct aggr_cpu_id aggr_cpu_id__node(int cpu, void *data __maybe_unused)
perf stat: Add --per-node agregation support Adding new --per-node option to aggregate counts per NUMA nodes for system-wide mode measurements. You can specify --per-node in live mode: # perf stat -a -I 1000 -e cycles --per-node # time node cpus counts unit events 1.000542550 N0 20 6,202,097 cycles 1.000542550 N1 20 639,559 cycles 2.002040063 N0 20 7,412,495 cycles 2.002040063 N1 20 2,185,577 cycles 3.003451699 N0 20 6,508,917 cycles 3.003451699 N1 20 765,607 cycles ... Or in the record/report stat session: # perf stat record -a -I 1000 -e cycles # time counts unit events 1.000536937 10,008,468 cycles 2.002090152 9,578,539 cycles 3.003625233 7,647,869 cycles 4.005135036 7,032,086 cycles ^C 4.340902364 3,923,893 cycles # perf stat report --per-node # time node cpus counts unit events 1.000536937 N0 20 9,355,086 cycles 1.000536937 N1 20 653,382 cycles 2.002090152 N0 20 7,712,838 cycles 2.002090152 N1 20 1,865,701 cycles 3.003625233 N0 20 6,604,441 cycles 3.003625233 N1 20 1,043,428 cycles 4.005135036 N0 20 6,350,522 cycles 4.005135036 N1 20 681,564 cycles 4.340902364 N0 20 3,403,188 cycles 4.340902364 N1 20 520,705 cycles Signed-off-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexey Budankov <alexey.budankov@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Joe Mario <jmario@redhat.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Michael Petlan <mpetlan@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/20190904073415.723-4-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-08-28 11:17:43 +03:00
{
struct aggr_cpu_id id = aggr_cpu_id__empty();
id.node = cpu__get_node(cpu);
return id;
}
/* setup simple routines to easily access node numbers given a cpu number */
static int get_max_num(char *path, int *max)
{
size_t num;
char *buf;
int err = 0;
if (filename__read_str(path, &buf, &num))
return -1;
buf[num] = '\0';
/* start on the right, to find highest node num */
while (--num) {
if ((buf[num] == ',') || (buf[num] == '-')) {
num++;
break;
}
}
if (sscanf(&buf[num], "%d", max) < 1) {
err = -1;
goto out;
}
/* convert from 0-based to 1-based */
(*max)++;
out:
free(buf);
return err;
}
/* Determine highest possible cpu in the system for sparse allocation */
static void set_max_cpu_num(void)
{
const char *mnt;
char path[PATH_MAX];
int ret = -1;
/* set up default */
max_cpu_num = 4096;
max_present_cpu_num = 4096;
mnt = sysfs__mountpoint();
if (!mnt)
goto out;
/* get the highest possible cpu number for a sparse allocation */
ret = snprintf(path, PATH_MAX, "%s/devices/system/cpu/possible", mnt);
if (ret >= PATH_MAX) {
pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
goto out;
}
ret = get_max_num(path, &max_cpu_num);
if (ret)
goto out;
/* get the highest present cpu number for a sparse allocation */
ret = snprintf(path, PATH_MAX, "%s/devices/system/cpu/present", mnt);
if (ret >= PATH_MAX) {
pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
goto out;
}
ret = get_max_num(path, &max_present_cpu_num);
out:
if (ret)
pr_err("Failed to read max cpus, using default of %d\n", max_cpu_num);
}
/* Determine highest possible node in the system for sparse allocation */
static void set_max_node_num(void)
{
const char *mnt;
char path[PATH_MAX];
int ret = -1;
/* set up default */
max_node_num = 8;
mnt = sysfs__mountpoint();
if (!mnt)
goto out;
/* get the highest possible cpu number for a sparse allocation */
ret = snprintf(path, PATH_MAX, "%s/devices/system/node/possible", mnt);
if (ret >= PATH_MAX) {
pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
goto out;
}
ret = get_max_num(path, &max_node_num);
out:
if (ret)
pr_err("Failed to read max nodes, using default of %d\n", max_node_num);
}
int cpu__max_node(void)
{
if (unlikely(!max_node_num))
set_max_node_num();
return max_node_num;
}
int cpu__max_cpu(void)
{
if (unlikely(!max_cpu_num))
set_max_cpu_num();
return max_cpu_num;
}
int cpu__max_present_cpu(void)
{
if (unlikely(!max_present_cpu_num))
set_max_cpu_num();
return max_present_cpu_num;
}
int cpu__get_node(int cpu)
{
if (unlikely(cpunode_map == NULL)) {
pr_debug("cpu_map not initialized\n");
return -1;
}
return cpunode_map[cpu];
}
static int init_cpunode_map(void)
{
int i;
set_max_cpu_num();
set_max_node_num();
cpunode_map = calloc(max_cpu_num, sizeof(int));
if (!cpunode_map) {
pr_err("%s: calloc failed\n", __func__);
return -1;
}
for (i = 0; i < max_cpu_num; i++)
cpunode_map[i] = -1;
return 0;
}
int cpu__setup_cpunode_map(void)
{
struct dirent *dent1, *dent2;
DIR *dir1, *dir2;
unsigned int cpu, mem;
char buf[PATH_MAX];
char path[PATH_MAX];
const char *mnt;
int n;
/* initialize globals */
if (init_cpunode_map())
return -1;
mnt = sysfs__mountpoint();
if (!mnt)
return 0;
n = snprintf(path, PATH_MAX, "%s/devices/system/node", mnt);
if (n >= PATH_MAX) {
pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
return -1;
}
dir1 = opendir(path);
if (!dir1)
return 0;
/* walk tree and setup map */
while ((dent1 = readdir(dir1)) != NULL) {
if (dent1->d_type != DT_DIR || sscanf(dent1->d_name, "node%u", &mem) < 1)
continue;
n = snprintf(buf, PATH_MAX, "%s/%s", path, dent1->d_name);
if (n >= PATH_MAX) {
pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
continue;
}
dir2 = opendir(buf);
if (!dir2)
continue;
while ((dent2 = readdir(dir2)) != NULL) {
if (dent2->d_type != DT_LNK || sscanf(dent2->d_name, "cpu%u", &cpu) < 1)
continue;
cpunode_map[cpu] = mem;
}
closedir(dir2);
}
closedir(dir1);
return 0;
}
size_t cpu_map__snprint(struct perf_cpu_map *map, char *buf, size_t size)
{
int i, cpu, start = -1;
bool first = true;
size_t ret = 0;
#define COMMA first ? "" : ","
for (i = 0; i < map->nr + 1; i++) {
bool last = i == map->nr;
cpu = last ? INT_MAX : map->map[i];
if (start == -1) {
start = i;
if (last) {
ret += snprintf(buf + ret, size - ret,
"%s%d", COMMA,
map->map[i]);
}
} else if (((i - start) != (cpu - map->map[start])) || last) {
int end = i - 1;
if (start == end) {
ret += snprintf(buf + ret, size - ret,
"%s%d", COMMA,
map->map[start]);
} else {
ret += snprintf(buf + ret, size - ret,
"%s%d-%d", COMMA,
map->map[start], map->map[end]);
}
first = false;
start = i;
}
}
#undef COMMA
pr_debug2("cpumask list: %s\n", buf);
return ret;
}
static char hex_char(unsigned char val)
{
if (val < 10)
return val + '0';
if (val < 16)
return val - 10 + 'a';
return '?';
}
size_t cpu_map__snprint_mask(struct perf_cpu_map *map, char *buf, size_t size)
{
int i, cpu;
char *ptr = buf;
unsigned char *bitmap;
int last_cpu = perf_cpu_map__cpu(map, map->nr - 1);
if (buf == NULL)
return 0;
bitmap = zalloc(last_cpu / 8 + 1);
if (bitmap == NULL) {
buf[0] = '\0';
return 0;
}
for (i = 0; i < map->nr; i++) {
cpu = perf_cpu_map__cpu(map, i);
bitmap[cpu / 8] |= 1 << (cpu % 8);
}
for (cpu = last_cpu / 4 * 4; cpu >= 0; cpu -= 4) {
unsigned char bits = bitmap[cpu / 8];
if (cpu % 8)
bits >>= 4;
else
bits &= 0xf;
*ptr++ = hex_char(bits);
if ((cpu % 32) == 0 && cpu > 0)
*ptr++ = ',';
}
*ptr = '\0';
free(bitmap);
buf[size - 1] = '\0';
return ptr - buf;
}
const struct perf_cpu_map *cpu_map__online(void) /* thread unsafe */
{
static const struct perf_cpu_map *online = NULL;
if (!online)
online = perf_cpu_map__new(NULL); /* from /sys/devices/system/cpu/online */
return online;
}
bool aggr_cpu_id__equal(const struct aggr_cpu_id *a, const struct aggr_cpu_id *b)
{
return a->thread == b->thread &&
a->node == b->node &&
a->socket == b->socket &&
a->die == b->die &&
a->core == b->core &&
a->cpu == b->cpu;
}
bool aggr_cpu_id__is_empty(const struct aggr_cpu_id *a)
{
return a->thread == -1 &&
a->node == -1 &&
a->socket == -1 &&
a->die == -1 &&
a->core == -1 &&
a->cpu == -1;
}
struct aggr_cpu_id aggr_cpu_id__empty(void)
{
struct aggr_cpu_id ret = {
.thread = -1,
perf stat aggregation: Add separate socket member Add socket as a separate member so that it doesn't have to be packed into the int value. When the socket ID was larger than 8 bits the output appeared corrupted or incomplete. For example, here on ThunderX2 'perf stat' reports a socket of -1 and an invalid die number: ./perf stat -a --per-die The socket id number is too big. Performance counter stats for 'system wide': S-1-D255 128 687.99 msec cpu-clock # 57.240 CPUs utilized ... S36-D0 128 842.34 msec cpu-clock # 70.081 CPUs utilized ... And with --per-core there is an entry with an invalid core ID: ./perf stat record -a --per-core The socket id number is too big. Performance counter stats for 'system wide': S-1-D255-C65535 128 671.04 msec cpu-clock # 54.112 CPUs utilized ... S36-D0-C0 4 28.27 msec cpu-clock # 2.279 CPUs utilized ... This fixes the "Session topology" self test on ThunderX2. After this fix the output contains the correct socket and die IDs and no longer prints a warning about the size of the socket ID: ./perf stat --per-die -a Performance counter stats for 'system wide': S36-D0 128 169,869.39 msec cpu-clock # 127.501 CPUs utilized ... S3612-D0 128 169,733.05 msec cpu-clock # 127.398 CPUs utilized Signed-off-by: James Clark <james.clark@arm.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Jiri Olsa <jolsa@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: John Garry <john.garry@huawei.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Richter <tmricht@linux.ibm.com> Link: https://lore.kernel.org/r/20201126141328.6509-10-james.clark@arm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2020-11-26 17:13:25 +03:00
.node = -1,
.socket = -1,
.die = -1,
.core = -1,
.cpu = -1
};
return ret;
}