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 22:07:57 +08:00
|
|
|
// SPDX-License-Identifier: GPL-2.0
|
2009-08-12 17:07:25 +08:00
|
|
|
#include "symbol.h"
|
2019-09-28 09:39:00 +08:00
|
|
|
#include <assert.h>
|
2010-03-26 23:11:06 +08:00
|
|
|
#include <errno.h>
|
2011-01-23 06:37:02 +08:00
|
|
|
#include <inttypes.h>
|
2010-03-26 06:58:58 +08:00
|
|
|
#include <limits.h>
|
2009-08-12 17:07:25 +08:00
|
|
|
#include <stdlib.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <stdio.h>
|
2010-04-19 13:32:50 +08:00
|
|
|
#include <unistd.h>
|
2016-09-13 03:47:57 +08:00
|
|
|
#include <uapi/linux/mman.h> /* To get things like MAP_HUGETLB even on older libc headers */
|
2019-08-30 22:11:01 +08:00
|
|
|
#include "dso.h"
|
2010-03-26 06:58:58 +08:00
|
|
|
#include "map.h"
|
2019-08-31 02:09:54 +08:00
|
|
|
#include "map_symbol.h"
|
2012-07-21 07:25:47 +08:00
|
|
|
#include "thread.h"
|
2012-09-11 00:50:19 +08:00
|
|
|
#include "vdso.h"
|
2012-10-28 05:18:28 +08:00
|
|
|
#include "build-id.h"
|
2014-07-15 05:46:47 +08:00
|
|
|
#include "debug.h"
|
2014-07-22 21:17:53 +08:00
|
|
|
#include "machine.h"
|
2013-01-25 09:16:43 +08:00
|
|
|
#include <linux/string.h>
|
2019-07-04 22:32:27 +08:00
|
|
|
#include <linux/zalloc.h>
|
2017-04-18 03:30:49 +08:00
|
|
|
#include "srcline.h"
|
2017-07-06 09:48:08 +08:00
|
|
|
#include "namespaces.h"
|
2016-07-04 20:16:23 +08:00
|
|
|
#include "unwind.h"
|
2018-12-04 08:18:48 +08:00
|
|
|
#include "srccode.h"
|
2019-08-30 03:18:59 +08:00
|
|
|
#include "ui/ui.h"
|
2009-08-12 17:07:25 +08:00
|
|
|
|
2015-05-23 00:45:24 +08:00
|
|
|
static void __maps__insert(struct maps *maps, struct map *map);
|
|
|
|
|
2016-09-06 12:58:27 +08:00
|
|
|
static inline int is_anon_memory(const char *filename, u32 flags)
|
2009-08-12 17:07:25 +08:00
|
|
|
{
|
2016-09-13 03:47:57 +08:00
|
|
|
return flags & MAP_HUGETLB ||
|
2016-09-06 12:58:27 +08:00
|
|
|
!strcmp(filename, "//anon") ||
|
2015-11-26 19:42:32 +08:00
|
|
|
!strncmp(filename, "/dev/zero", sizeof("/dev/zero") - 1) ||
|
|
|
|
!strncmp(filename, "/anon_hugepage", sizeof("/anon_hugepage") - 1);
|
2009-08-12 17:07:25 +08:00
|
|
|
}
|
|
|
|
|
2011-08-24 21:18:34 +08:00
|
|
|
static inline int is_no_dso_memory(const char *filename)
|
|
|
|
{
|
2012-11-07 15:27:11 +08:00
|
|
|
return !strncmp(filename, "[stack", 6) ||
|
2014-08-20 10:31:14 +08:00
|
|
|
!strncmp(filename, "/SYSV",5) ||
|
2011-08-24 21:18:34 +08:00
|
|
|
!strcmp(filename, "[heap]");
|
|
|
|
}
|
|
|
|
|
2014-05-20 17:48:50 +08:00
|
|
|
static inline int is_android_lib(const char *filename)
|
|
|
|
{
|
2020-03-10 03:53:41 +08:00
|
|
|
return strstarts(filename, "/data/app-lib/") ||
|
|
|
|
strstarts(filename, "/system/lib/");
|
2014-05-20 17:48:50 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline bool replace_android_lib(const char *filename, char *newfilename)
|
|
|
|
{
|
|
|
|
const char *libname;
|
|
|
|
char *app_abi;
|
|
|
|
size_t app_abi_length, new_length;
|
|
|
|
size_t lib_length = 0;
|
|
|
|
|
|
|
|
libname = strrchr(filename, '/');
|
|
|
|
if (libname)
|
|
|
|
lib_length = strlen(libname);
|
|
|
|
|
|
|
|
app_abi = getenv("APP_ABI");
|
|
|
|
if (!app_abi)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
app_abi_length = strlen(app_abi);
|
|
|
|
|
2020-03-10 03:53:41 +08:00
|
|
|
if (strstarts(filename, "/data/app-lib/")) {
|
2014-05-20 17:48:50 +08:00
|
|
|
char *apk_path;
|
|
|
|
|
|
|
|
if (!app_abi_length)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
new_length = 7 + app_abi_length + lib_length;
|
|
|
|
|
|
|
|
apk_path = getenv("APK_PATH");
|
|
|
|
if (apk_path) {
|
|
|
|
new_length += strlen(apk_path) + 1;
|
|
|
|
if (new_length > PATH_MAX)
|
|
|
|
return false;
|
|
|
|
snprintf(newfilename, new_length,
|
|
|
|
"%s/libs/%s/%s", apk_path, app_abi, libname);
|
|
|
|
} else {
|
|
|
|
if (new_length > PATH_MAX)
|
|
|
|
return false;
|
|
|
|
snprintf(newfilename, new_length,
|
|
|
|
"libs/%s/%s", app_abi, libname);
|
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2020-03-10 03:53:41 +08:00
|
|
|
if (strstarts(filename, "/system/lib/")) {
|
2014-05-20 17:48:50 +08:00
|
|
|
char *ndk, *app;
|
|
|
|
const char *arch;
|
|
|
|
size_t ndk_length;
|
|
|
|
size_t app_length;
|
|
|
|
|
|
|
|
ndk = getenv("NDK_ROOT");
|
|
|
|
app = getenv("APP_PLATFORM");
|
|
|
|
|
|
|
|
if (!(ndk && app))
|
|
|
|
return false;
|
|
|
|
|
|
|
|
ndk_length = strlen(ndk);
|
|
|
|
app_length = strlen(app);
|
|
|
|
|
|
|
|
if (!(ndk_length && app_length && app_abi_length))
|
|
|
|
return false;
|
|
|
|
|
|
|
|
arch = !strncmp(app_abi, "arm", 3) ? "arm" :
|
|
|
|
!strncmp(app_abi, "mips", 4) ? "mips" :
|
|
|
|
!strncmp(app_abi, "x86", 3) ? "x86" : NULL;
|
|
|
|
|
|
|
|
if (!arch)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
new_length = 27 + ndk_length +
|
|
|
|
app_length + lib_length
|
|
|
|
+ strlen(arch);
|
|
|
|
|
|
|
|
if (new_length > PATH_MAX)
|
|
|
|
return false;
|
|
|
|
snprintf(newfilename, new_length,
|
|
|
|
"%s/platforms/%s/arch-%s/usr/lib/%s",
|
|
|
|
ndk, app, arch, libname);
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2018-04-27 03:52:34 +08:00
|
|
|
void map__init(struct map *map, u64 start, u64 end, u64 pgoff, struct dso *dso)
|
2009-10-31 02:28:23 +08:00
|
|
|
{
|
2013-01-25 08:59:59 +08:00
|
|
|
map->start = start;
|
|
|
|
map->end = end;
|
|
|
|
map->pgoff = pgoff;
|
2014-01-29 22:14:36 +08:00
|
|
|
map->reloc = 0;
|
2015-06-02 22:53:26 +08:00
|
|
|
map->dso = dso__get(dso);
|
2013-01-25 08:59:59 +08:00
|
|
|
map->map_ip = map__map_ip;
|
|
|
|
map->unmap_ip = map__unmap_ip;
|
|
|
|
RB_CLEAR_NODE(&map->rb_node);
|
|
|
|
map->erange_warned = false;
|
2017-02-21 23:34:59 +08:00
|
|
|
refcount_set(&map->refcnt, 1);
|
2009-10-31 02:28:23 +08:00
|
|
|
}
|
|
|
|
|
2014-07-22 21:17:53 +08:00
|
|
|
struct map *map__new(struct machine *machine, u64 start, u64 len,
|
2019-11-19 23:40:29 +08:00
|
|
|
u64 pgoff, struct dso_id *id,
|
|
|
|
u32 prot, u32 flags, char *filename,
|
2018-04-27 03:52:34 +08:00
|
|
|
struct thread *thread)
|
2009-08-12 17:07:25 +08:00
|
|
|
{
|
2013-01-25 08:59:59 +08:00
|
|
|
struct map *map = malloc(sizeof(*map));
|
2017-07-06 09:48:09 +08:00
|
|
|
struct nsinfo *nsi = NULL;
|
|
|
|
struct nsinfo *nnsi;
|
2009-08-12 17:07:25 +08:00
|
|
|
|
2013-01-25 08:59:59 +08:00
|
|
|
if (map != NULL) {
|
2009-08-12 17:07:25 +08:00
|
|
|
char newfilename[PATH_MAX];
|
2009-10-31 02:28:23 +08:00
|
|
|
struct dso *dso;
|
2014-05-20 17:48:50 +08:00
|
|
|
int anon, no_dso, vdso, android;
|
2009-08-12 17:07:25 +08:00
|
|
|
|
2014-05-20 17:48:50 +08:00
|
|
|
android = is_android_lib(filename);
|
2016-09-06 12:58:27 +08:00
|
|
|
anon = is_anon_memory(filename, flags);
|
2012-09-11 00:50:19 +08:00
|
|
|
vdso = is_vdso_map(filename);
|
2011-08-24 21:18:34 +08:00
|
|
|
no_dso = is_no_dso_memory(filename);
|
2014-05-20 03:13:49 +08:00
|
|
|
map->prot = prot;
|
|
|
|
map->flags = flags;
|
2017-07-06 09:48:09 +08:00
|
|
|
nsi = nsinfo__get(thread->nsinfo);
|
2013-08-21 18:10:25 +08:00
|
|
|
|
2018-04-27 03:08:38 +08:00
|
|
|
if ((anon || no_dso) && nsi && (prot & PROT_EXEC)) {
|
2017-07-06 09:48:09 +08:00
|
|
|
snprintf(newfilename, sizeof(newfilename),
|
|
|
|
"/tmp/perf-%d.map", nsi->pid);
|
2009-08-12 17:07:25 +08:00
|
|
|
filename = newfilename;
|
|
|
|
}
|
|
|
|
|
2014-05-20 17:48:50 +08:00
|
|
|
if (android) {
|
|
|
|
if (replace_android_lib(filename, newfilename))
|
|
|
|
filename = newfilename;
|
|
|
|
}
|
|
|
|
|
2012-09-11 00:50:19 +08:00
|
|
|
if (vdso) {
|
2017-07-06 09:48:09 +08:00
|
|
|
/* The vdso maps are always on the host and not the
|
|
|
|
* container. Ensure that we don't use setns to look
|
|
|
|
* them up.
|
|
|
|
*/
|
|
|
|
nnsi = nsinfo__copy(nsi);
|
|
|
|
if (nnsi) {
|
|
|
|
nsinfo__put(nsi);
|
|
|
|
nnsi->need_setns = false;
|
|
|
|
nsi = nnsi;
|
|
|
|
}
|
2012-09-11 00:50:19 +08:00
|
|
|
pgoff = 0;
|
perf machine: Fix up vdso methods names
To make it consistent with the other dso lifetime routines.
For instance:
struct dso *vdso__new(struct machine *machine, const char *short_name,
const char *long_name)
Becomes:
struct dso *machine__addnew_vdso(struct machine *machine, const
char *short_name, const char *long_name)
Because:
1) There is no 'struct vdso' for us to have vdso__ prefixed routines.
2) Because it will not really just create a new instance of 'struct
dso', it'll call dso__new() but it will also insert it into the
DSO's list/rbtree, and we have a method name for that: 'addnew',
just like we have dsos__addnew().
3) So it is really a 'struct machine' operation, it is the first
argument, etc.
This way the place where this is used gets consistent:
if (vdso) {
pgoff = 0;
- dso = vdso__dso_findnew(machine, thread);
+ dso = machine__findnew_vdso(machine, thread);
} else
dso = machine__findnew_dso(machine, filename);
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/n/tip-r3w3tvh8exm9xfz3p4tz9qbz@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-05-29 22:54:08 +08:00
|
|
|
dso = machine__findnew_vdso(machine, thread);
|
2012-09-11 00:50:19 +08:00
|
|
|
} else
|
perf dso: Move dso_id from 'struct map' to 'struct dso'
And take it into account when looking up DSOs when we have the dso_id
fields obtained from somewhere, like from PERF_RECORD_MMAP2 records.
Instances of struct map pointing to the same DSO pathname but with
anything in dso_id different are in fact different DSOs, so better have
different 'struct dso' instances to reflect that. At some point we may
want to get copies of the contents of the different objects if we want
to do correct annotation or other analysis.
With this we get 'struct map' 24 bytes leaner:
$ pahole -C map ~/bin/perf
struct map {
union {
struct rb_node rb_node __attribute__((__aligned__(8))); /* 0 24 */
struct list_head node; /* 0 16 */
} __attribute__((__aligned__(8))); /* 0 24 */
u64 start; /* 24 8 */
u64 end; /* 32 8 */
_Bool erange_warned:1; /* 40: 0 1 */
_Bool priv:1; /* 40: 1 1 */
/* XXX 6 bits hole, try to pack */
/* XXX 3 bytes hole, try to pack */
u32 prot; /* 44 4 */
u64 pgoff; /* 48 8 */
u64 reloc; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
u64 (*map_ip)(struct map *, u64); /* 64 8 */
u64 (*unmap_ip)(struct map *, u64); /* 72 8 */
struct dso * dso; /* 80 8 */
refcount_t refcnt; /* 88 4 */
u32 flags; /* 92 4 */
/* size: 96, cachelines: 2, members: 13 */
/* sum members: 92, holes: 1, sum holes: 3 */
/* sum bitfield members: 2 bits, bit holes: 1, sum bit holes: 6 bits */
/* forced alignments: 1 */
/* last cacheline: 32 bytes */
} __attribute__((__aligned__(8)));
$
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-g4hxxmraplo7wfjmk384mfsb@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-20 05:44:22 +08:00
|
|
|
dso = machine__findnew_dso_id(machine, filename, id);
|
2012-09-11 00:50:19 +08:00
|
|
|
|
2009-10-31 02:28:23 +08:00
|
|
|
if (dso == NULL)
|
2009-08-12 17:07:25 +08:00
|
|
|
goto out_delete;
|
|
|
|
|
2018-04-27 03:52:34 +08:00
|
|
|
map__init(map, start, start + len, pgoff, dso);
|
2009-10-31 02:28:23 +08:00
|
|
|
|
2011-08-24 21:18:34 +08:00
|
|
|
if (anon || no_dso) {
|
2013-01-25 08:59:59 +08:00
|
|
|
map->map_ip = map->unmap_ip = identity__map_ip;
|
2011-08-24 21:18:34 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Set memory without DSO as loaded. All map__find_*
|
|
|
|
* functions still return NULL, and we avoid the
|
|
|
|
* unnecessary map__load warning.
|
|
|
|
*/
|
2018-04-27 03:08:38 +08:00
|
|
|
if (!(prot & PROT_EXEC))
|
2018-04-27 03:52:34 +08:00
|
|
|
dso__set_loaded(dso);
|
2010-02-04 02:52:02 +08:00
|
|
|
}
|
2017-07-06 09:48:09 +08:00
|
|
|
dso->nsinfo = nsi;
|
2015-06-02 22:53:26 +08:00
|
|
|
dso__put(dso);
|
2009-08-12 17:07:25 +08:00
|
|
|
}
|
2013-01-25 08:59:59 +08:00
|
|
|
return map;
|
2009-08-12 17:07:25 +08:00
|
|
|
out_delete:
|
2017-07-06 09:48:09 +08:00
|
|
|
nsinfo__put(nsi);
|
2013-01-25 08:59:59 +08:00
|
|
|
free(map);
|
2009-08-12 17:07:25 +08:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2012-08-06 12:41:20 +08:00
|
|
|
/*
|
|
|
|
* Constructor variant for modules (where we know from /proc/modules where
|
|
|
|
* they are loaded) and for vmlinux, where only after we load all the
|
|
|
|
* symbols we'll know where it starts and ends.
|
|
|
|
*/
|
2018-04-27 03:52:34 +08:00
|
|
|
struct map *map__new2(u64 start, struct dso *dso)
|
2012-08-06 12:41:20 +08:00
|
|
|
{
|
|
|
|
struct map *map = calloc(1, (sizeof(*map) +
|
|
|
|
(dso->kernel ? sizeof(struct kmap) : 0)));
|
|
|
|
if (map != NULL) {
|
|
|
|
/*
|
|
|
|
* ->end will be filled after we load all the symbols
|
|
|
|
*/
|
2018-04-27 03:52:34 +08:00
|
|
|
map__init(map, start, 0, 0, dso);
|
2012-08-06 12:41:20 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return map;
|
|
|
|
}
|
|
|
|
|
2015-07-23 22:06:16 +08:00
|
|
|
bool __map__is_kernel(const struct map *map)
|
|
|
|
{
|
2019-11-05 03:31:33 +08:00
|
|
|
if (!map->dso->kernel)
|
|
|
|
return false;
|
|
|
|
return machine__kernel_map(map__kmaps((struct map *)map)->machine) == map;
|
2015-07-23 22:06:16 +08:00
|
|
|
}
|
|
|
|
|
2018-05-22 18:54:35 +08:00
|
|
|
bool __map__is_extra_kernel_map(const struct map *map)
|
|
|
|
{
|
|
|
|
struct kmap *kmap = __map__kmap((struct map *)map);
|
|
|
|
|
|
|
|
return kmap && kmap->name[0];
|
|
|
|
}
|
|
|
|
|
2019-04-17 00:01:22 +08:00
|
|
|
bool __map__is_bpf_prog(const struct map *map)
|
|
|
|
{
|
|
|
|
const char *name;
|
|
|
|
|
|
|
|
if (map->dso->binary_type == DSO_BINARY_TYPE__BPF_PROG_INFO)
|
|
|
|
return true;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If PERF_RECORD_BPF_EVENT is not included, the dso will not have
|
|
|
|
* type of DSO_BINARY_TYPE__BPF_PROG_INFO. In such cases, we can
|
|
|
|
* guess the type based on name.
|
|
|
|
*/
|
|
|
|
name = map->dso->short_name;
|
|
|
|
return name && (strstr(name, "bpf_prog_") == name);
|
|
|
|
}
|
|
|
|
|
2020-08-27 05:30:17 +08:00
|
|
|
bool __map__is_bpf_image(const struct map *map)
|
|
|
|
{
|
|
|
|
const char *name;
|
|
|
|
|
|
|
|
if (map->dso->binary_type == DSO_BINARY_TYPE__BPF_IMAGE)
|
|
|
|
return true;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If PERF_RECORD_KSYMBOL is not included, the dso will not have
|
|
|
|
* type of DSO_BINARY_TYPE__BPF_IMAGE. In such cases, we can
|
|
|
|
* guess the type based on name.
|
|
|
|
*/
|
|
|
|
name = map->dso->short_name;
|
|
|
|
return name && is_bpf_image(name);
|
|
|
|
}
|
|
|
|
|
2020-05-12 20:19:19 +08:00
|
|
|
bool __map__is_ool(const struct map *map)
|
|
|
|
{
|
|
|
|
return map->dso && map->dso->binary_type == DSO_BINARY_TYPE__OOL;
|
|
|
|
}
|
|
|
|
|
2018-04-24 04:13:49 +08:00
|
|
|
bool map__has_symbols(const struct map *map)
|
|
|
|
{
|
2018-04-27 03:52:34 +08:00
|
|
|
return dso__has_symbols(map->dso);
|
2018-04-24 04:13:49 +08:00
|
|
|
}
|
|
|
|
|
2015-06-02 22:53:26 +08:00
|
|
|
static void map__exit(struct map *map)
|
2009-11-21 06:51:27 +08:00
|
|
|
{
|
2019-10-26 02:14:50 +08:00
|
|
|
BUG_ON(refcount_read(&map->refcnt) != 0);
|
2015-06-02 22:53:26 +08:00
|
|
|
dso__zput(map->dso);
|
|
|
|
}
|
|
|
|
|
|
|
|
void map__delete(struct map *map)
|
|
|
|
{
|
|
|
|
map__exit(map);
|
2013-01-25 08:59:59 +08:00
|
|
|
free(map);
|
2009-11-21 06:51:27 +08:00
|
|
|
}
|
|
|
|
|
2015-05-26 03:59:56 +08:00
|
|
|
void map__put(struct map *map)
|
|
|
|
{
|
2017-02-21 23:34:59 +08:00
|
|
|
if (map && refcount_dec_and_test(&map->refcnt))
|
2015-05-26 03:59:56 +08:00
|
|
|
map__delete(map);
|
|
|
|
}
|
|
|
|
|
2013-01-25 08:59:59 +08:00
|
|
|
void map__fixup_start(struct map *map)
|
2009-11-21 06:51:27 +08:00
|
|
|
{
|
2018-12-07 03:18:17 +08:00
|
|
|
struct rb_root_cached *symbols = &map->dso->symbols;
|
|
|
|
struct rb_node *nd = rb_first_cached(symbols);
|
2009-11-21 06:51:27 +08:00
|
|
|
if (nd != NULL) {
|
|
|
|
struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
|
2013-01-25 08:59:59 +08:00
|
|
|
map->start = sym->start;
|
2009-11-21 06:51:27 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-01-25 08:59:59 +08:00
|
|
|
void map__fixup_end(struct map *map)
|
2009-11-21 06:51:27 +08:00
|
|
|
{
|
2018-12-07 03:18:17 +08:00
|
|
|
struct rb_root_cached *symbols = &map->dso->symbols;
|
|
|
|
struct rb_node *nd = rb_last(&symbols->rb_root);
|
2009-11-21 06:51:27 +08:00
|
|
|
if (nd != NULL) {
|
|
|
|
struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
|
2013-01-25 08:59:59 +08:00
|
|
|
map->end = sym->end;
|
2009-11-21 06:51:27 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-10-31 02:28:25 +08:00
|
|
|
#define DSO__DELETED "(deleted)"
|
|
|
|
|
2016-09-02 06:25:52 +08:00
|
|
|
int map__load(struct map *map)
|
2009-10-29 07:51:21 +08:00
|
|
|
{
|
2013-01-25 08:59:59 +08:00
|
|
|
const char *name = map->dso->long_name;
|
2009-12-15 23:32:33 +08:00
|
|
|
int nr;
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
|
2018-04-27 03:52:34 +08:00
|
|
|
if (dso__loaded(map->dso))
|
2009-12-15 23:32:33 +08:00
|
|
|
return 0;
|
|
|
|
|
2016-09-02 06:25:52 +08:00
|
|
|
nr = dso__load(map->dso, map);
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
if (nr < 0) {
|
2013-01-25 08:59:59 +08:00
|
|
|
if (map->dso->has_build_id) {
|
2016-05-11 21:51:59 +08:00
|
|
|
char sbuild_id[SBUILD_ID_SIZE];
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
|
2013-01-25 08:59:59 +08:00
|
|
|
build_id__sprintf(map->dso->build_id,
|
|
|
|
sizeof(map->dso->build_id),
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
sbuild_id);
|
2018-09-04 21:43:07 +08:00
|
|
|
pr_debug("%s with build id %s not found", name, sbuild_id);
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
} else
|
2018-09-04 21:43:07 +08:00
|
|
|
pr_debug("Failed to open %s", name);
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
|
2018-09-04 21:43:07 +08:00
|
|
|
pr_debug(", continuing without symbols\n");
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
return -1;
|
|
|
|
} else if (nr == 0) {
|
2013-09-30 18:07:11 +08:00
|
|
|
#ifdef HAVE_LIBELF_SUPPORT
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
const size_t len = strlen(name);
|
|
|
|
const size_t real_len = len - sizeof(DSO__DELETED);
|
|
|
|
|
|
|
|
if (len > sizeof(DSO__DELETED) &&
|
|
|
|
strcmp(name + real_len + 1, DSO__DELETED) == 0) {
|
2018-09-04 21:43:07 +08:00
|
|
|
pr_debug("%.*s was updated (is prelink enabled?). "
|
2011-10-19 08:44:45 +08:00
|
|
|
"Restart the long running apps that use it!\n",
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
(int)real_len, name);
|
|
|
|
} else {
|
2018-09-04 21:43:07 +08:00
|
|
|
pr_debug("no symbols found in %s, maybe install a debug package?\n", name);
|
2009-10-29 07:51:21 +08:00
|
|
|
}
|
2012-08-06 12:41:21 +08:00
|
|
|
#endif
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
return -1;
|
2009-10-29 07:51:21 +08:00
|
|
|
}
|
|
|
|
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-09-02 06:25:52 +08:00
|
|
|
struct symbol *map__find_symbol(struct map *map, u64 addr)
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
{
|
2016-09-02 06:25:52 +08:00
|
|
|
if (map__load(map) < 0)
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
return NULL;
|
|
|
|
|
2018-04-27 03:52:34 +08:00
|
|
|
return dso__find_symbol(map->dso, addr);
|
2009-10-29 07:51:21 +08:00
|
|
|
}
|
|
|
|
|
2016-09-02 06:25:52 +08:00
|
|
|
struct symbol *map__find_symbol_by_name(struct map *map, const char *name)
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
{
|
2016-09-02 06:25:52 +08:00
|
|
|
if (map__load(map) < 0)
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
return NULL;
|
|
|
|
|
2018-04-27 03:52:34 +08:00
|
|
|
if (!dso__sorted_by_name(map->dso))
|
|
|
|
dso__sort_by_name(map->dso);
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
|
2018-04-27 03:52:34 +08:00
|
|
|
return dso__find_symbol_by_name(map->dso, name);
|
perf symbols: Allow lookups by symbol name too
Configurable via symbol_conf.sort_by_name, so that the cost of an
extra rb_node on all 'struct symbol' instances is not paid by tools
that only want to decode addresses.
How to use it:
symbol_conf.sort_by_name = true;
symbol_init(&symbol_conf);
struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]");
if (map == NULL) {
pr_err("couldn't find map!\n");
kernel_maps__fprintf(stdout);
} else {
struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL);
if (sym == NULL)
pr_err("couldn't find symbol %s!\n", sym_filter);
else
pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end);
}
Looking over the vmlinux/kallsyms is common enough that I'll add a
variable to the upcoming struct perf_session to avoid the need to
use map_groups__find_by_name to get the main vmlinux/kallsyms map.
The above example looks on the 'variable' symtab, but it is just
like that for the functions one.
Also the sort operation is done when we first use
map__find_symbol_by_name, in a lazy way.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-12 04:50:22 +08:00
|
|
|
}
|
|
|
|
|
2015-08-19 02:19:50 +08:00
|
|
|
struct map *map__clone(struct map *from)
|
2009-08-12 17:07:25 +08:00
|
|
|
{
|
2020-02-10 22:32:17 +08:00
|
|
|
size_t size = sizeof(struct map);
|
|
|
|
struct map *map;
|
2015-08-19 02:19:50 +08:00
|
|
|
|
2020-02-10 22:32:17 +08:00
|
|
|
if (from->dso && from->dso->kernel)
|
|
|
|
size += sizeof(struct kmap);
|
|
|
|
|
|
|
|
map = memdup(from, size);
|
2015-08-19 02:19:50 +08:00
|
|
|
if (map != NULL) {
|
2017-02-21 23:34:59 +08:00
|
|
|
refcount_set(&map->refcnt, 1);
|
2015-08-19 02:19:50 +08:00
|
|
|
RB_CLEAR_NODE(&map->rb_node);
|
|
|
|
dso__get(map->dso);
|
|
|
|
}
|
|
|
|
|
|
|
|
return map;
|
2009-08-12 17:07:25 +08:00
|
|
|
}
|
|
|
|
|
2013-01-25 08:59:59 +08:00
|
|
|
size_t map__fprintf(struct map *map, FILE *fp)
|
2009-08-12 17:07:25 +08:00
|
|
|
{
|
2011-01-23 06:37:02 +08:00
|
|
|
return fprintf(fp, " %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s\n",
|
2013-01-25 08:59:59 +08:00
|
|
|
map->start, map->end, map->pgoff, map->dso->name);
|
2009-08-12 17:07:25 +08:00
|
|
|
}
|
perf annotate: Fix it for non-prelinked *.so
The problem was we were incorrectly calculating objdump
addresses for sym->start and sym->end, look:
For simple ET_DYN type DSO (*.so) with one function, objdump -dS
output is something like this:
000004ac <my_strlen>:
int my_strlen(const char *s)
4ac: 55 push %ebp
4ad: 89 e5 mov %esp,%ebp
4af: 83 ec 10 sub $0x10,%esp
{
i.e. we have relative-to-dso-mapping IPs (=RIP) there.
For ET_EXEC type and probably for prelinked libs as well (sorry
can't test - I don't use prelink) objdump outputs absolute IPs,
e.g.
08048604 <zz_strlen>:
extern "C"
int zz_strlen(const char *s)
8048604: 55 push %ebp
8048605: 89 e5 mov %esp,%ebp
8048607: 83 ec 10 sub $0x10,%esp
{
So, if sym->start is always relative to dso mapping(*), we'll
have to unmap it for ET_EXEC like cases, and leave as is for
ET_DYN cases.
(*) and it is - we've explicitely made it relative. Look for
adjust_symbols handling in dso__load_sym()
Previously we were always unmapping sym->start and for ET_DYN
dsos resulting addresses were wrong, and so objdump output was
empty.
The end result was that perf annotate output for symbols from
non-prelinked *.so had always 0.00% percents only, which is
wrong.
To fix it, let's introduce a helper for converting rip to
objdump address, and also let's document what map_ip() and
unmap_ip() do -- I had to study sources for several hours to
understand it.
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <1265223128-11786-8-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-04 02:52:07 +08:00
|
|
|
|
2012-01-30 12:42:57 +08:00
|
|
|
size_t map__fprintf_dsoname(struct map *map, FILE *fp)
|
|
|
|
{
|
2019-05-08 21:20:05 +08:00
|
|
|
char buf[symbol_conf.pad_output_len_dso + 1];
|
2012-08-27 15:38:26 +08:00
|
|
|
const char *dsoname = "[unknown]";
|
2012-01-30 12:42:57 +08:00
|
|
|
|
2017-02-14 04:11:03 +08:00
|
|
|
if (map && map->dso) {
|
2012-01-30 12:43:20 +08:00
|
|
|
if (symbol_conf.show_kernel_path && map->dso->long_name)
|
|
|
|
dsoname = map->dso->long_name;
|
2017-02-14 04:11:03 +08:00
|
|
|
else
|
2012-01-30 12:43:20 +08:00
|
|
|
dsoname = map->dso->name;
|
2012-08-27 15:38:26 +08:00
|
|
|
}
|
2012-01-30 12:42:57 +08:00
|
|
|
|
2019-05-08 21:20:05 +08:00
|
|
|
if (symbol_conf.pad_output_len_dso) {
|
|
|
|
scnprintf_pad(buf, symbol_conf.pad_output_len_dso, "%s", dsoname);
|
|
|
|
dsoname = buf;
|
|
|
|
}
|
|
|
|
|
2012-01-30 12:42:57 +08:00
|
|
|
return fprintf(fp, "%s", dsoname);
|
|
|
|
}
|
|
|
|
|
2018-05-28 22:05:20 +08:00
|
|
|
char *map__srcline(struct map *map, u64 addr, struct symbol *sym)
|
|
|
|
{
|
|
|
|
if (map == NULL)
|
|
|
|
return SRCLINE_UNKNOWN;
|
|
|
|
return get_srcline(map->dso, map__rip_2objdump(map, addr), sym, true, true, addr);
|
|
|
|
}
|
|
|
|
|
2013-12-06 15:42:57 +08:00
|
|
|
int map__fprintf_srcline(struct map *map, u64 addr, const char *prefix,
|
|
|
|
FILE *fp)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
if (map && map->dso) {
|
2018-05-28 22:05:20 +08:00
|
|
|
char *srcline = map__srcline(map, addr, NULL);
|
2020-02-24 03:34:49 +08:00
|
|
|
if (strncmp(srcline, SRCLINE_UNKNOWN, strlen(SRCLINE_UNKNOWN)) != 0)
|
2013-12-06 15:42:57 +08:00
|
|
|
ret = fprintf(fp, "%s%s", prefix, srcline);
|
|
|
|
free_srcline(srcline);
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2018-12-04 08:18:48 +08:00
|
|
|
void srccode_state_free(struct srccode_state *state)
|
|
|
|
{
|
|
|
|
zfree(&state->srcfile);
|
|
|
|
state->line = 0;
|
|
|
|
}
|
|
|
|
|
2013-10-14 18:43:44 +08:00
|
|
|
/**
|
|
|
|
* map__rip_2objdump - convert symbol start address to objdump address.
|
|
|
|
* @map: memory map
|
|
|
|
* @rip: symbol start address
|
|
|
|
*
|
perf annotate: Fix it for non-prelinked *.so
The problem was we were incorrectly calculating objdump
addresses for sym->start and sym->end, look:
For simple ET_DYN type DSO (*.so) with one function, objdump -dS
output is something like this:
000004ac <my_strlen>:
int my_strlen(const char *s)
4ac: 55 push %ebp
4ad: 89 e5 mov %esp,%ebp
4af: 83 ec 10 sub $0x10,%esp
{
i.e. we have relative-to-dso-mapping IPs (=RIP) there.
For ET_EXEC type and probably for prelinked libs as well (sorry
can't test - I don't use prelink) objdump outputs absolute IPs,
e.g.
08048604 <zz_strlen>:
extern "C"
int zz_strlen(const char *s)
8048604: 55 push %ebp
8048605: 89 e5 mov %esp,%ebp
8048607: 83 ec 10 sub $0x10,%esp
{
So, if sym->start is always relative to dso mapping(*), we'll
have to unmap it for ET_EXEC like cases, and leave as is for
ET_DYN cases.
(*) and it is - we've explicitely made it relative. Look for
adjust_symbols handling in dso__load_sym()
Previously we were always unmapping sym->start and for ET_DYN
dsos resulting addresses were wrong, and so objdump output was
empty.
The end result was that perf annotate output for symbols from
non-prelinked *.so had always 0.00% percents only, which is
wrong.
To fix it, let's introduce a helper for converting rip to
objdump address, and also let's document what map_ip() and
unmap_ip() do -- I had to study sources for several hours to
understand it.
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <1265223128-11786-8-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-04 02:52:07 +08:00
|
|
|
* objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
|
2013-08-07 19:38:50 +08:00
|
|
|
* map->dso->adjust_symbols==1 for ET_EXEC-like cases except ET_REL which is
|
|
|
|
* relative to section start.
|
2013-10-14 18:43:44 +08:00
|
|
|
*
|
|
|
|
* Return: Address suitable for passing to "objdump --start-address="
|
perf annotate: Fix it for non-prelinked *.so
The problem was we were incorrectly calculating objdump
addresses for sym->start and sym->end, look:
For simple ET_DYN type DSO (*.so) with one function, objdump -dS
output is something like this:
000004ac <my_strlen>:
int my_strlen(const char *s)
4ac: 55 push %ebp
4ad: 89 e5 mov %esp,%ebp
4af: 83 ec 10 sub $0x10,%esp
{
i.e. we have relative-to-dso-mapping IPs (=RIP) there.
For ET_EXEC type and probably for prelinked libs as well (sorry
can't test - I don't use prelink) objdump outputs absolute IPs,
e.g.
08048604 <zz_strlen>:
extern "C"
int zz_strlen(const char *s)
8048604: 55 push %ebp
8048605: 89 e5 mov %esp,%ebp
8048607: 83 ec 10 sub $0x10,%esp
{
So, if sym->start is always relative to dso mapping(*), we'll
have to unmap it for ET_EXEC like cases, and leave as is for
ET_DYN cases.
(*) and it is - we've explicitely made it relative. Look for
adjust_symbols handling in dso__load_sym()
Previously we were always unmapping sym->start and for ET_DYN
dsos resulting addresses were wrong, and so objdump output was
empty.
The end result was that perf annotate output for symbols from
non-prelinked *.so had always 0.00% percents only, which is
wrong.
To fix it, let's introduce a helper for converting rip to
objdump address, and also let's document what map_ip() and
unmap_ip() do -- I had to study sources for several hours to
understand it.
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <1265223128-11786-8-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-04 02:52:07 +08:00
|
|
|
*/
|
|
|
|
u64 map__rip_2objdump(struct map *map, u64 rip)
|
|
|
|
{
|
2018-06-05 15:30:00 +08:00
|
|
|
struct kmap *kmap = __map__kmap(map);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* vmlinux does not have program headers for PTI entry trampolines and
|
|
|
|
* kcore may not either. However the trampoline object code is on the
|
|
|
|
* main kernel map, so just use that instead.
|
|
|
|
*/
|
|
|
|
if (kmap && is_entry_trampoline(kmap->name) && kmap->kmaps && kmap->kmaps->machine) {
|
|
|
|
struct map *kernel_map = machine__kernel_map(kmap->kmaps->machine);
|
|
|
|
|
|
|
|
if (kernel_map)
|
|
|
|
map = kernel_map;
|
|
|
|
}
|
|
|
|
|
2013-08-07 19:38:50 +08:00
|
|
|
if (!map->dso->adjust_symbols)
|
|
|
|
return rip;
|
|
|
|
|
|
|
|
if (map->dso->rel)
|
|
|
|
return rip - map->pgoff;
|
|
|
|
|
2016-04-07 18:24:30 +08:00
|
|
|
/*
|
|
|
|
* kernel modules also have DSO_TYPE_USER in dso->kernel,
|
|
|
|
* but all kernel modules are ET_REL, so won't get here.
|
|
|
|
*/
|
2020-08-08 20:21:54 +08:00
|
|
|
if (map->dso->kernel == DSO_SPACE__USER)
|
2016-04-07 18:24:30 +08:00
|
|
|
return rip + map->dso->text_offset;
|
|
|
|
|
2014-01-29 22:14:36 +08:00
|
|
|
return map->unmap_ip(map, rip) - map->reloc;
|
perf annotate: Fix it for non-prelinked *.so
The problem was we were incorrectly calculating objdump
addresses for sym->start and sym->end, look:
For simple ET_DYN type DSO (*.so) with one function, objdump -dS
output is something like this:
000004ac <my_strlen>:
int my_strlen(const char *s)
4ac: 55 push %ebp
4ad: 89 e5 mov %esp,%ebp
4af: 83 ec 10 sub $0x10,%esp
{
i.e. we have relative-to-dso-mapping IPs (=RIP) there.
For ET_EXEC type and probably for prelinked libs as well (sorry
can't test - I don't use prelink) objdump outputs absolute IPs,
e.g.
08048604 <zz_strlen>:
extern "C"
int zz_strlen(const char *s)
8048604: 55 push %ebp
8048605: 89 e5 mov %esp,%ebp
8048607: 83 ec 10 sub $0x10,%esp
{
So, if sym->start is always relative to dso mapping(*), we'll
have to unmap it for ET_EXEC like cases, and leave as is for
ET_DYN cases.
(*) and it is - we've explicitely made it relative. Look for
adjust_symbols handling in dso__load_sym()
Previously we were always unmapping sym->start and for ET_DYN
dsos resulting addresses were wrong, and so objdump output was
empty.
The end result was that perf annotate output for symbols from
non-prelinked *.so had always 0.00% percents only, which is
wrong.
To fix it, let's introduce a helper for converting rip to
objdump address, and also let's document what map_ip() and
unmap_ip() do -- I had to study sources for several hours to
understand it.
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <1265223128-11786-8-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-04 02:52:07 +08:00
|
|
|
}
|
perf top: Fix annotate for userspace
First, for programs and prelinked libraries, annotate code was
fooled by objdump output IPs (src->eip in the code) being
wrongly converted to absolute IPs. In such case there were no
conversion needed, but in
src->eip = strtoull(src->line, NULL, 16);
src->eip = map->unmap_ip(map, src->eip); // = eip + map->start - map->pgoff
we were reading absolute address from objdump (e.g. 8048604) and
then almost doubling it, because eip & map->start are
approximately close for small programs.
Needless to say, that later, in record_precise_ip() there was no
matching with real runtime IPs.
And second, like with `perf annotate` the problem with
non-prelinked *.so was that we were doing rip -> objdump address
conversion wrong.
Also, because unlike `perf annotate`, `perf top` code does
annotation based on absolute IPs for performance reasons(*), new
helper for mapping objdump addresse to IP is introduced.
(*) we get samples info in absolute IPs, and since we do lots of
hit-testing on absolute IPs at runtime in record_precise_ip(), it's
better to convert objdump addresses to IPs once and do no conversion
at runtime.
I also had to fix how objdump output is parsed (with hardcoded
8/16 characters format, which was inappropriate for ET_DYN dsos
with small addresses like '4ac')
Also note, that not all objdump output lines has associtated
IPs, e.g. look at source lines here:
000004ac <my_strlen>:
extern "C"
int my_strlen(const char *s)
4ac: 55 push %ebp
4ad: 89 e5 mov %esp,%ebp
4af: 83 ec 10 sub $0x10,%esp
{
int len = 0;
4b2: c7 45 fc 00 00 00 00 movl $0x0,-0x4(%ebp)
4b9: eb 08 jmp 4c3 <my_strlen+0x17>
while (*s) {
++len;
4bb: 83 45 fc 01 addl $0x1,-0x4(%ebp)
++s;
4bf: 83 45 08 01 addl $0x1,0x8(%ebp)
So we mark them with eip=0, and ignore such lines in annotate
lookup code.
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
[ Note: one hunk of this patch was applied by Mike in 57d8188 ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <1265550376-12665-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-07 21:46:15 +08:00
|
|
|
|
2013-10-14 18:43:44 +08:00
|
|
|
/**
|
|
|
|
* map__objdump_2mem - convert objdump address to a memory address.
|
|
|
|
* @map: memory map
|
|
|
|
* @ip: objdump address
|
|
|
|
*
|
|
|
|
* Closely related to map__rip_2objdump(), this function takes an address from
|
|
|
|
* objdump and converts it to a memory address. Note this assumes that @map
|
|
|
|
* contains the address. To be sure the result is valid, check it forwards
|
|
|
|
* e.g. map__rip_2objdump(map->map_ip(map, map__objdump_2mem(map, ip))) == ip
|
|
|
|
*
|
|
|
|
* Return: Memory address.
|
|
|
|
*/
|
|
|
|
u64 map__objdump_2mem(struct map *map, u64 ip)
|
|
|
|
{
|
|
|
|
if (!map->dso->adjust_symbols)
|
|
|
|
return map->unmap_ip(map, ip);
|
|
|
|
|
|
|
|
if (map->dso->rel)
|
|
|
|
return map->unmap_ip(map, ip + map->pgoff);
|
|
|
|
|
2016-04-07 18:24:30 +08:00
|
|
|
/*
|
|
|
|
* kernel modules also have DSO_TYPE_USER in dso->kernel,
|
|
|
|
* but all kernel modules are ET_REL, so won't get here.
|
|
|
|
*/
|
2020-08-08 20:21:54 +08:00
|
|
|
if (map->dso->kernel == DSO_SPACE__USER)
|
2016-04-07 18:24:30 +08:00
|
|
|
return map->unmap_ip(map, ip - map->dso->text_offset);
|
|
|
|
|
2014-01-29 22:14:36 +08:00
|
|
|
return ip + map->reloc;
|
2013-10-14 18:43:44 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
void maps__init(struct maps *maps, struct machine *machine)
|
2015-05-22 23:58:53 +08:00
|
|
|
{
|
2019-11-26 09:21:28 +08:00
|
|
|
maps->entries = RB_ROOT;
|
|
|
|
init_rwsem(&maps->lock);
|
|
|
|
maps->machine = machine;
|
|
|
|
maps->last_search_by_name = NULL;
|
|
|
|
maps->nr_maps = 0;
|
|
|
|
maps->maps_by_name = NULL;
|
|
|
|
refcount_set(&maps->refcnt, 1);
|
2010-03-26 23:11:06 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
static void __maps__free_maps_by_name(struct maps *maps)
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
{
|
|
|
|
/*
|
|
|
|
* Free everything to try to do it from the rbtree in the next search
|
|
|
|
*/
|
2019-11-26 09:21:28 +08:00
|
|
|
zfree(&maps->maps_by_name);
|
|
|
|
maps->nr_maps_allocated = 0;
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
void maps__insert(struct maps *maps, struct map *map)
|
2019-01-27 20:44:29 +08:00
|
|
|
{
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
down_write(&maps->lock);
|
|
|
|
__maps__insert(maps, map);
|
2019-11-26 09:21:28 +08:00
|
|
|
++maps->nr_maps;
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
|
2020-02-10 22:32:18 +08:00
|
|
|
if (map->dso && map->dso->kernel) {
|
|
|
|
struct kmap *kmap = map__kmap(map);
|
|
|
|
|
|
|
|
if (kmap)
|
|
|
|
kmap->kmaps = maps;
|
|
|
|
else
|
|
|
|
pr_err("Internal error: kernel dso with non kernel map\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
/*
|
|
|
|
* If we already performed some search by name, then we need to add the just
|
|
|
|
* inserted map and resort.
|
|
|
|
*/
|
2019-11-26 09:21:28 +08:00
|
|
|
if (maps->maps_by_name) {
|
|
|
|
if (maps->nr_maps > maps->nr_maps_allocated) {
|
|
|
|
int nr_allocate = maps->nr_maps * 2;
|
|
|
|
struct map **maps_by_name = realloc(maps->maps_by_name, nr_allocate * sizeof(map));
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
|
|
|
|
if (maps_by_name == NULL) {
|
2019-11-26 08:58:33 +08:00
|
|
|
__maps__free_maps_by_name(maps);
|
2020-01-20 22:15:54 +08:00
|
|
|
up_write(&maps->lock);
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
maps->maps_by_name = maps_by_name;
|
|
|
|
maps->nr_maps_allocated = nr_allocate;
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
}
|
2019-11-26 09:21:28 +08:00
|
|
|
maps->maps_by_name[maps->nr_maps - 1] = map;
|
2019-11-26 08:58:33 +08:00
|
|
|
__maps__sort_by_name(maps);
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
}
|
|
|
|
up_write(&maps->lock);
|
2019-01-27 20:44:29 +08:00
|
|
|
}
|
|
|
|
|
2019-11-25 21:42:38 +08:00
|
|
|
static void __maps__remove(struct maps *maps, struct map *map)
|
|
|
|
{
|
|
|
|
rb_erase_init(&map->rb_node, &maps->entries);
|
|
|
|
map__put(map);
|
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
void maps__remove(struct maps *maps, struct map *map)
|
perf map_groups: Add a front end cache for map lookups by name
Lets see if it helps:
First look at the probeable lines for the function that does lookups by
name in a map_groups struct:
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
12 maps__for_each_entry(maps, map)
13 if (strcmp(map->dso->short_name, name) == 0) {
14 mg->last_search_by_name = map;
15 goto out_unlock;
}
18 map = NULL;
out_unlock:
21 up_read(&maps->lock);
22 return map;
23 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
#
Now add a probe to the place where we reuse the last search:
# perf probe -x ~/bin/perf map_groups__find_by_name:8
Added new event:
probe_perf:map_groups__find_by_name (on map_groups__find_by_name:8 in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:map_groups__find_by_name -aR sleep 1
#
Now lets do a system wide 'perf stat' counting those events:
# perf stat -e probe_perf:*
Leave it running and lets do a 'perf top', then, after a while, stop the
'perf stat':
# perf stat -e probe_perf:*
^C
Performance counter stats for 'system wide':
3,603 probe_perf:map_groups__find_by_name
44.565253139 seconds time elapsed
#
yeah, good to have.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-tcz37g3nxv3tvxw3q90vga3p@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-14 03:33:33 +08:00
|
|
|
{
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
down_write(&maps->lock);
|
2019-11-26 09:21:28 +08:00
|
|
|
if (maps->last_search_by_name == map)
|
|
|
|
maps->last_search_by_name = NULL;
|
perf map_groups: Add a front end cache for map lookups by name
Lets see if it helps:
First look at the probeable lines for the function that does lookups by
name in a map_groups struct:
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
12 maps__for_each_entry(maps, map)
13 if (strcmp(map->dso->short_name, name) == 0) {
14 mg->last_search_by_name = map;
15 goto out_unlock;
}
18 map = NULL;
out_unlock:
21 up_read(&maps->lock);
22 return map;
23 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
#
Now add a probe to the place where we reuse the last search:
# perf probe -x ~/bin/perf map_groups__find_by_name:8
Added new event:
probe_perf:map_groups__find_by_name (on map_groups__find_by_name:8 in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:map_groups__find_by_name -aR sleep 1
#
Now lets do a system wide 'perf stat' counting those events:
# perf stat -e probe_perf:*
Leave it running and lets do a 'perf top', then, after a while, stop the
'perf stat':
# perf stat -e probe_perf:*
^C
Performance counter stats for 'system wide':
3,603 probe_perf:map_groups__find_by_name
44.565253139 seconds time elapsed
#
yeah, good to have.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-tcz37g3nxv3tvxw3q90vga3p@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-14 03:33:33 +08:00
|
|
|
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
__maps__remove(maps, map);
|
2019-11-26 09:21:28 +08:00
|
|
|
--maps->nr_maps;
|
|
|
|
if (maps->maps_by_name)
|
2019-11-26 08:58:33 +08:00
|
|
|
__maps__free_maps_by_name(maps);
|
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-17 22:38:13 +08:00
|
|
|
up_write(&maps->lock);
|
perf map_groups: Add a front end cache for map lookups by name
Lets see if it helps:
First look at the probeable lines for the function that does lookups by
name in a map_groups struct:
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
12 maps__for_each_entry(maps, map)
13 if (strcmp(map->dso->short_name, name) == 0) {
14 mg->last_search_by_name = map;
15 goto out_unlock;
}
18 map = NULL;
out_unlock:
21 up_read(&maps->lock);
22 return map;
23 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
#
Now add a probe to the place where we reuse the last search:
# perf probe -x ~/bin/perf map_groups__find_by_name:8
Added new event:
probe_perf:map_groups__find_by_name (on map_groups__find_by_name:8 in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:map_groups__find_by_name -aR sleep 1
#
Now lets do a system wide 'perf stat' counting those events:
# perf stat -e probe_perf:*
Leave it running and lets do a 'perf top', then, after a while, stop the
'perf stat':
# perf stat -e probe_perf:*
^C
Performance counter stats for 'system wide':
3,603 probe_perf:map_groups__find_by_name
44.565253139 seconds time elapsed
#
yeah, good to have.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-tcz37g3nxv3tvxw3q90vga3p@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-14 03:33:33 +08:00
|
|
|
}
|
|
|
|
|
2015-05-23 00:45:24 +08:00
|
|
|
static void __maps__purge(struct maps *maps)
|
2010-07-31 05:28:42 +08:00
|
|
|
{
|
2019-10-28 22:31:38 +08:00
|
|
|
struct map *pos, *next;
|
2010-07-31 05:28:42 +08:00
|
|
|
|
2019-10-28 22:31:38 +08:00
|
|
|
maps__for_each_entry_safe(maps, pos, next) {
|
|
|
|
rb_erase_init(&pos->rb_node, &maps->entries);
|
2015-05-26 03:59:56 +08:00
|
|
|
map__put(pos);
|
2019-03-16 16:05:51 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
void maps__exit(struct maps *maps)
|
2015-05-22 23:58:53 +08:00
|
|
|
{
|
2017-04-05 00:15:04 +08:00
|
|
|
down_write(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
__maps__purge(maps);
|
2017-04-05 00:15:04 +08:00
|
|
|
up_write(&maps->lock);
|
2015-05-22 23:58:53 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
bool maps__empty(struct maps *maps)
|
2010-07-31 05:28:42 +08:00
|
|
|
{
|
2019-11-26 08:58:33 +08:00
|
|
|
return !maps__first(maps);
|
2010-07-31 05:28:42 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
struct maps *maps__new(struct machine *machine)
|
2014-07-16 16:07:13 +08:00
|
|
|
{
|
2019-11-26 09:21:28 +08:00
|
|
|
struct maps *maps = zalloc(sizeof(*maps));
|
2014-03-22 04:57:01 +08:00
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
if (maps != NULL)
|
2019-11-26 08:58:33 +08:00
|
|
|
maps__init(maps, machine);
|
2014-03-22 04:57:01 +08:00
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
return maps;
|
2014-03-22 04:57:01 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
void maps__delete(struct maps *maps)
|
2014-03-22 04:57:01 +08:00
|
|
|
{
|
2019-11-26 09:21:28 +08:00
|
|
|
maps__exit(maps);
|
|
|
|
unwind__finish_access(maps);
|
|
|
|
free(maps);
|
2014-03-22 04:57:01 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
void maps__put(struct maps *maps)
|
2014-03-26 02:26:44 +08:00
|
|
|
{
|
2019-11-26 09:21:28 +08:00
|
|
|
if (maps && refcount_dec_and_test(&maps->refcnt))
|
|
|
|
maps__delete(maps);
|
2014-03-26 02:26:44 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
struct symbol *maps__find_symbol(struct maps *maps, u64 addr, struct map **mapp)
|
2010-03-26 06:58:58 +08:00
|
|
|
{
|
2019-11-26 09:21:28 +08:00
|
|
|
struct map *map = maps__find(maps, addr);
|
2010-03-26 06:58:58 +08:00
|
|
|
|
2014-01-23 10:29:50 +08:00
|
|
|
/* Ensure map is loaded before using map->map_ip */
|
2016-09-02 06:25:52 +08:00
|
|
|
if (map != NULL && map__load(map) >= 0) {
|
2010-03-26 23:30:40 +08:00
|
|
|
if (mapp != NULL)
|
|
|
|
*mapp = map;
|
2016-09-02 06:25:52 +08:00
|
|
|
return map__find_symbol(map, map->map_ip(map, addr));
|
2010-03-26 23:30:40 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2018-09-07 16:51:16 +08:00
|
|
|
static bool map__contains_symbol(struct map *map, struct symbol *sym)
|
|
|
|
{
|
|
|
|
u64 ip = map->unmap_ip(map, sym->start);
|
|
|
|
|
|
|
|
return ip >= map->start && ip < map->end;
|
|
|
|
}
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
struct symbol *maps__find_symbol_by_name(struct maps *maps, const char *name, struct map **mapp)
|
2010-03-26 23:30:40 +08:00
|
|
|
{
|
2015-05-23 00:45:24 +08:00
|
|
|
struct symbol *sym;
|
2019-10-28 22:31:38 +08:00
|
|
|
struct map *pos;
|
2010-03-26 23:30:40 +08:00
|
|
|
|
2017-04-05 00:15:04 +08:00
|
|
|
down_read(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
|
2019-10-28 22:31:38 +08:00
|
|
|
maps__for_each_entry(maps, pos) {
|
2016-09-02 06:25:52 +08:00
|
|
|
sym = map__find_symbol_by_name(pos, name);
|
2010-03-26 23:30:40 +08:00
|
|
|
|
|
|
|
if (sym == NULL)
|
|
|
|
continue;
|
2018-09-07 16:51:16 +08:00
|
|
|
if (!map__contains_symbol(pos, sym)) {
|
|
|
|
sym = NULL;
|
|
|
|
continue;
|
|
|
|
}
|
2010-03-26 23:30:40 +08:00
|
|
|
if (mapp != NULL)
|
|
|
|
*mapp = pos;
|
2015-05-23 00:45:24 +08:00
|
|
|
goto out;
|
2010-03-26 23:30:40 +08:00
|
|
|
}
|
2010-03-26 06:58:58 +08:00
|
|
|
|
2015-05-23 00:45:24 +08:00
|
|
|
sym = NULL;
|
|
|
|
out:
|
2017-04-05 00:15:04 +08:00
|
|
|
up_read(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
return sym;
|
2010-03-26 06:58:58 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
int maps__find_ams(struct maps *maps, struct addr_map_symbol *ams)
|
2013-10-14 18:43:38 +08:00
|
|
|
{
|
2019-11-05 02:57:38 +08:00
|
|
|
if (ams->addr < ams->ms.map->start || ams->addr >= ams->ms.map->end) {
|
2019-11-26 09:21:28 +08:00
|
|
|
if (maps == NULL)
|
2013-10-14 18:43:38 +08:00
|
|
|
return -1;
|
2019-11-26 09:21:28 +08:00
|
|
|
ams->ms.map = maps__find(maps, ams->addr);
|
2019-11-05 02:57:38 +08:00
|
|
|
if (ams->ms.map == NULL)
|
2013-10-14 18:43:38 +08:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2019-11-05 02:57:38 +08:00
|
|
|
ams->al_addr = ams->ms.map->map_ip(ams->ms.map, ams->addr);
|
|
|
|
ams->ms.sym = map__find_symbol(ams->ms.map, ams->al_addr);
|
2013-10-14 18:43:38 +08:00
|
|
|
|
2019-11-05 02:57:38 +08:00
|
|
|
return ams->ms.sym ? 0 : -1;
|
2013-10-14 18:43:38 +08:00
|
|
|
}
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
size_t maps__fprintf(struct maps *maps, FILE *fp)
|
2010-03-26 23:11:06 +08:00
|
|
|
{
|
2015-05-23 00:45:24 +08:00
|
|
|
size_t printed = 0;
|
2019-10-28 22:31:38 +08:00
|
|
|
struct map *pos;
|
2010-03-26 23:11:06 +08:00
|
|
|
|
2017-04-05 00:15:04 +08:00
|
|
|
down_read(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
|
2019-10-28 22:31:38 +08:00
|
|
|
maps__for_each_entry(maps, pos) {
|
2010-03-26 23:11:06 +08:00
|
|
|
printed += fprintf(fp, "Map:");
|
|
|
|
printed += map__fprintf(pos, fp);
|
|
|
|
if (verbose > 2) {
|
2018-04-27 03:52:34 +08:00
|
|
|
printed += dso__fprintf(pos->dso, fp);
|
2010-03-26 23:11:06 +08:00
|
|
|
printed += fprintf(fp, "--\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-04-05 00:15:04 +08:00
|
|
|
up_read(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
|
2010-03-26 23:11:06 +08:00
|
|
|
return printed;
|
|
|
|
}
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
int maps__fixup_overlappings(struct maps *maps, struct map *map, FILE *fp)
|
2010-03-26 23:11:06 +08:00
|
|
|
{
|
2015-05-23 00:45:24 +08:00
|
|
|
struct rb_root *root;
|
2018-08-07 22:24:54 +08:00
|
|
|
struct rb_node *next, *first;
|
perf tools: Don't keep unreferenced maps when unmaps are detected
For a file with:
[root@emilia linux-2.6-tip]# perf report -D -fi allmodconfig-j32.perf.data | grep events:
TOTAL events: 36933
MMAP events: 9056
LOST events: 0
COMM events: 1702
EXIT events: 1887
THROTTLE events: 8
UNTHROTTLE events: 8
FORK events: 1894
READ events: 0
SAMPLE events: 22378
ATTR events: 0
EVENT_TYPE events: 0
TRACING_DATA events: 0
BUILD_ID events: 0
[root@emilia linux-2.6-tip]#
Testing with valgrind and making perf_session__delete() a nop, so that
we can notice how many maps were actually deleted due to not having any
samples on it:
==== HEAP SUMMARY:
Before:
==10339== in use at exit: 8,909,997 bytes in 68,690 blocks
==10339== total heap usage: 78,696 allocs, 10,007 frees, 11,925,853 bytes allocated
After:
==10506== in use at exit: 8,902,605 bytes in 68,606 blocks
==10506== total heap usage: 78,696 allocs, 10,091 frees, 11,925,853 bytes allocated
I.e. just 84 detected unmaps with no hits out of 9056 for this workload,
not much, but in some other long running workload this may save more
bytes.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2010-08-03 06:45:23 +08:00
|
|
|
int err = 0;
|
2010-03-26 23:11:06 +08:00
|
|
|
|
2017-04-05 00:15:04 +08:00
|
|
|
down_write(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
|
|
|
|
root = &maps->entries;
|
|
|
|
|
2018-08-07 22:24:54 +08:00
|
|
|
/*
|
|
|
|
* Find first map where end > map->start.
|
|
|
|
* Same as find_vma() in kernel.
|
|
|
|
*/
|
|
|
|
next = root->rb_node;
|
|
|
|
first = NULL;
|
|
|
|
while (next) {
|
|
|
|
struct map *pos = rb_entry(next, struct map, rb_node);
|
|
|
|
|
|
|
|
if (pos->end > map->start) {
|
|
|
|
first = next;
|
|
|
|
if (pos->start <= map->start)
|
|
|
|
break;
|
|
|
|
next = next->rb_left;
|
|
|
|
} else
|
|
|
|
next = next->rb_right;
|
|
|
|
}
|
|
|
|
|
|
|
|
next = first;
|
2010-03-26 23:11:06 +08:00
|
|
|
while (next) {
|
|
|
|
struct map *pos = rb_entry(next, struct map, rb_node);
|
|
|
|
next = rb_next(&pos->rb_node);
|
|
|
|
|
2018-08-07 22:24:54 +08:00
|
|
|
/*
|
|
|
|
* Stop if current map starts after map->end.
|
|
|
|
* Maps are ordered by start: next will not overlap for sure.
|
|
|
|
*/
|
|
|
|
if (pos->start >= map->end)
|
|
|
|
break;
|
2010-03-26 23:11:06 +08:00
|
|
|
|
|
|
|
if (verbose >= 2) {
|
2016-10-13 05:48:23 +08:00
|
|
|
|
|
|
|
if (use_browser) {
|
2018-09-04 21:43:07 +08:00
|
|
|
pr_debug("overlapping maps in %s (disable tui for more info)\n",
|
2016-10-13 05:48:23 +08:00
|
|
|
map->dso->name);
|
|
|
|
} else {
|
|
|
|
fputs("overlapping maps:\n", fp);
|
|
|
|
map__fprintf(map, fp);
|
|
|
|
map__fprintf(pos, fp);
|
|
|
|
}
|
2010-03-26 23:11:06 +08:00
|
|
|
}
|
|
|
|
|
2015-05-26 02:30:09 +08:00
|
|
|
rb_erase_init(&pos->rb_node, root);
|
2010-03-26 23:11:06 +08:00
|
|
|
/*
|
|
|
|
* Now check if we need to create new maps for areas not
|
|
|
|
* overlapped by the new map:
|
|
|
|
*/
|
|
|
|
if (map->start > pos->start) {
|
|
|
|
struct map *before = map__clone(pos);
|
|
|
|
|
perf tools: Don't keep unreferenced maps when unmaps are detected
For a file with:
[root@emilia linux-2.6-tip]# perf report -D -fi allmodconfig-j32.perf.data | grep events:
TOTAL events: 36933
MMAP events: 9056
LOST events: 0
COMM events: 1702
EXIT events: 1887
THROTTLE events: 8
UNTHROTTLE events: 8
FORK events: 1894
READ events: 0
SAMPLE events: 22378
ATTR events: 0
EVENT_TYPE events: 0
TRACING_DATA events: 0
BUILD_ID events: 0
[root@emilia linux-2.6-tip]#
Testing with valgrind and making perf_session__delete() a nop, so that
we can notice how many maps were actually deleted due to not having any
samples on it:
==== HEAP SUMMARY:
Before:
==10339== in use at exit: 8,909,997 bytes in 68,690 blocks
==10339== total heap usage: 78,696 allocs, 10,007 frees, 11,925,853 bytes allocated
After:
==10506== in use at exit: 8,902,605 bytes in 68,606 blocks
==10506== total heap usage: 78,696 allocs, 10,091 frees, 11,925,853 bytes allocated
I.e. just 84 detected unmaps with no hits out of 9056 for this workload,
not much, but in some other long running workload this may save more
bytes.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2010-08-03 06:45:23 +08:00
|
|
|
if (before == NULL) {
|
|
|
|
err = -ENOMEM;
|
2015-05-26 03:59:56 +08:00
|
|
|
goto put_map;
|
perf tools: Don't keep unreferenced maps when unmaps are detected
For a file with:
[root@emilia linux-2.6-tip]# perf report -D -fi allmodconfig-j32.perf.data | grep events:
TOTAL events: 36933
MMAP events: 9056
LOST events: 0
COMM events: 1702
EXIT events: 1887
THROTTLE events: 8
UNTHROTTLE events: 8
FORK events: 1894
READ events: 0
SAMPLE events: 22378
ATTR events: 0
EVENT_TYPE events: 0
TRACING_DATA events: 0
BUILD_ID events: 0
[root@emilia linux-2.6-tip]#
Testing with valgrind and making perf_session__delete() a nop, so that
we can notice how many maps were actually deleted due to not having any
samples on it:
==== HEAP SUMMARY:
Before:
==10339== in use at exit: 8,909,997 bytes in 68,690 blocks
==10339== total heap usage: 78,696 allocs, 10,007 frees, 11,925,853 bytes allocated
After:
==10506== in use at exit: 8,902,605 bytes in 68,606 blocks
==10506== total heap usage: 78,696 allocs, 10,091 frees, 11,925,853 bytes allocated
I.e. just 84 detected unmaps with no hits out of 9056 for this workload,
not much, but in some other long running workload this may save more
bytes.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2010-08-03 06:45:23 +08:00
|
|
|
}
|
2010-03-26 23:11:06 +08:00
|
|
|
|
2014-10-06 16:35:32 +08:00
|
|
|
before->end = map->start;
|
2019-11-26 08:58:33 +08:00
|
|
|
__maps__insert(maps, before);
|
2016-10-13 05:48:23 +08:00
|
|
|
if (verbose >= 2 && !use_browser)
|
2010-03-26 23:11:06 +08:00
|
|
|
map__fprintf(before, fp);
|
2015-12-09 10:11:31 +08:00
|
|
|
map__put(before);
|
2010-03-26 23:11:06 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
if (map->end < pos->end) {
|
|
|
|
struct map *after = map__clone(pos);
|
|
|
|
|
perf tools: Don't keep unreferenced maps when unmaps are detected
For a file with:
[root@emilia linux-2.6-tip]# perf report -D -fi allmodconfig-j32.perf.data | grep events:
TOTAL events: 36933
MMAP events: 9056
LOST events: 0
COMM events: 1702
EXIT events: 1887
THROTTLE events: 8
UNTHROTTLE events: 8
FORK events: 1894
READ events: 0
SAMPLE events: 22378
ATTR events: 0
EVENT_TYPE events: 0
TRACING_DATA events: 0
BUILD_ID events: 0
[root@emilia linux-2.6-tip]#
Testing with valgrind and making perf_session__delete() a nop, so that
we can notice how many maps were actually deleted due to not having any
samples on it:
==== HEAP SUMMARY:
Before:
==10339== in use at exit: 8,909,997 bytes in 68,690 blocks
==10339== total heap usage: 78,696 allocs, 10,007 frees, 11,925,853 bytes allocated
After:
==10506== in use at exit: 8,902,605 bytes in 68,606 blocks
==10506== total heap usage: 78,696 allocs, 10,091 frees, 11,925,853 bytes allocated
I.e. just 84 detected unmaps with no hits out of 9056 for this workload,
not much, but in some other long running workload this may save more
bytes.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2010-08-03 06:45:23 +08:00
|
|
|
if (after == NULL) {
|
|
|
|
err = -ENOMEM;
|
2015-05-26 03:59:56 +08:00
|
|
|
goto put_map;
|
perf tools: Don't keep unreferenced maps when unmaps are detected
For a file with:
[root@emilia linux-2.6-tip]# perf report -D -fi allmodconfig-j32.perf.data | grep events:
TOTAL events: 36933
MMAP events: 9056
LOST events: 0
COMM events: 1702
EXIT events: 1887
THROTTLE events: 8
UNTHROTTLE events: 8
FORK events: 1894
READ events: 0
SAMPLE events: 22378
ATTR events: 0
EVENT_TYPE events: 0
TRACING_DATA events: 0
BUILD_ID events: 0
[root@emilia linux-2.6-tip]#
Testing with valgrind and making perf_session__delete() a nop, so that
we can notice how many maps were actually deleted due to not having any
samples on it:
==== HEAP SUMMARY:
Before:
==10339== in use at exit: 8,909,997 bytes in 68,690 blocks
==10339== total heap usage: 78,696 allocs, 10,007 frees, 11,925,853 bytes allocated
After:
==10506== in use at exit: 8,902,605 bytes in 68,606 blocks
==10506== total heap usage: 78,696 allocs, 10,091 frees, 11,925,853 bytes allocated
I.e. just 84 detected unmaps with no hits out of 9056 for this workload,
not much, but in some other long running workload this may save more
bytes.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2010-08-03 06:45:23 +08:00
|
|
|
}
|
2010-03-26 23:11:06 +08:00
|
|
|
|
2014-10-06 16:35:32 +08:00
|
|
|
after->start = map->end;
|
2019-09-28 09:39:00 +08:00
|
|
|
after->pgoff += map->end - pos->start;
|
|
|
|
assert(pos->map_ip(pos, map->end) == after->map_ip(after, map->end));
|
2019-11-26 08:58:33 +08:00
|
|
|
__maps__insert(maps, after);
|
2016-10-13 05:48:23 +08:00
|
|
|
if (verbose >= 2 && !use_browser)
|
2010-03-26 23:11:06 +08:00
|
|
|
map__fprintf(after, fp);
|
2015-12-09 10:11:31 +08:00
|
|
|
map__put(after);
|
2010-03-26 23:11:06 +08:00
|
|
|
}
|
2015-05-26 03:59:56 +08:00
|
|
|
put_map:
|
2015-06-16 10:29:51 +08:00
|
|
|
map__put(pos);
|
perf tools: Don't keep unreferenced maps when unmaps are detected
For a file with:
[root@emilia linux-2.6-tip]# perf report -D -fi allmodconfig-j32.perf.data | grep events:
TOTAL events: 36933
MMAP events: 9056
LOST events: 0
COMM events: 1702
EXIT events: 1887
THROTTLE events: 8
UNTHROTTLE events: 8
FORK events: 1894
READ events: 0
SAMPLE events: 22378
ATTR events: 0
EVENT_TYPE events: 0
TRACING_DATA events: 0
BUILD_ID events: 0
[root@emilia linux-2.6-tip]#
Testing with valgrind and making perf_session__delete() a nop, so that
we can notice how many maps were actually deleted due to not having any
samples on it:
==== HEAP SUMMARY:
Before:
==10339== in use at exit: 8,909,997 bytes in 68,690 blocks
==10339== total heap usage: 78,696 allocs, 10,007 frees, 11,925,853 bytes allocated
After:
==10506== in use at exit: 8,902,605 bytes in 68,606 blocks
==10506== total heap usage: 78,696 allocs, 10,091 frees, 11,925,853 bytes allocated
I.e. just 84 detected unmaps with no hits out of 9056 for this workload,
not much, but in some other long running workload this may save more
bytes.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2010-08-03 06:45:23 +08:00
|
|
|
|
|
|
|
if (err)
|
2015-05-23 00:45:24 +08:00
|
|
|
goto out;
|
2010-03-26 23:11:06 +08:00
|
|
|
}
|
|
|
|
|
2015-05-23 00:45:24 +08:00
|
|
|
err = 0;
|
|
|
|
out:
|
2017-04-05 00:15:04 +08:00
|
|
|
up_write(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2010-03-26 23:11:06 +08:00
|
|
|
/*
|
|
|
|
* XXX This should not really _copy_ te maps, but refcount them.
|
|
|
|
*/
|
2019-11-26 08:58:33 +08:00
|
|
|
int maps__clone(struct thread *thread, struct maps *parent)
|
2010-03-26 23:11:06 +08:00
|
|
|
{
|
2019-11-26 09:21:28 +08:00
|
|
|
struct maps *maps = thread->maps;
|
2015-05-23 00:45:24 +08:00
|
|
|
int err = -ENOMEM;
|
2015-05-22 22:52:22 +08:00
|
|
|
struct map *map;
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
down_read(&parent->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
maps__for_each_entry(parent, map) {
|
2010-03-26 23:11:06 +08:00
|
|
|
struct map *new = map__clone(map);
|
|
|
|
if (new == NULL)
|
2015-05-23 00:45:24 +08:00
|
|
|
goto out_unlock;
|
2016-07-04 20:16:23 +08:00
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
err = unwind__prepare_access(maps, new, NULL);
|
2016-07-04 20:16:23 +08:00
|
|
|
if (err)
|
|
|
|
goto out_unlock;
|
|
|
|
|
2019-11-26 09:21:28 +08:00
|
|
|
maps__insert(maps, new);
|
2015-12-09 10:11:20 +08:00
|
|
|
map__put(new);
|
2010-03-26 23:11:06 +08:00
|
|
|
}
|
2015-05-23 00:45:24 +08:00
|
|
|
|
|
|
|
err = 0;
|
|
|
|
out_unlock:
|
2019-11-26 08:58:33 +08:00
|
|
|
up_read(&parent->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
return err;
|
2010-03-26 23:11:06 +08:00
|
|
|
}
|
|
|
|
|
2015-05-23 00:45:24 +08:00
|
|
|
static void __maps__insert(struct maps *maps, struct map *map)
|
2010-03-26 06:58:58 +08:00
|
|
|
{
|
2015-05-22 23:58:53 +08:00
|
|
|
struct rb_node **p = &maps->entries.rb_node;
|
2010-03-26 06:58:58 +08:00
|
|
|
struct rb_node *parent = NULL;
|
|
|
|
const u64 ip = map->start;
|
|
|
|
struct map *m;
|
|
|
|
|
|
|
|
while (*p != NULL) {
|
|
|
|
parent = *p;
|
|
|
|
m = rb_entry(parent, struct map, rb_node);
|
|
|
|
if (ip < m->start)
|
|
|
|
p = &(*p)->rb_left;
|
|
|
|
else
|
|
|
|
p = &(*p)->rb_right;
|
|
|
|
}
|
|
|
|
|
|
|
|
rb_link_node(&map->rb_node, parent, p);
|
2015-05-22 23:58:53 +08:00
|
|
|
rb_insert_color(&map->rb_node, &maps->entries);
|
2015-05-26 03:59:56 +08:00
|
|
|
map__get(map);
|
2010-03-26 06:58:58 +08:00
|
|
|
}
|
|
|
|
|
2015-05-22 23:58:53 +08:00
|
|
|
struct map *maps__find(struct maps *maps, u64 ip)
|
2010-03-26 06:58:58 +08:00
|
|
|
{
|
2018-11-23 18:42:39 +08:00
|
|
|
struct rb_node *p;
|
2010-03-26 06:58:58 +08:00
|
|
|
struct map *m;
|
|
|
|
|
2017-04-05 00:15:04 +08:00
|
|
|
down_read(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
|
2018-11-23 18:42:39 +08:00
|
|
|
p = maps->entries.rb_node;
|
|
|
|
while (p != NULL) {
|
|
|
|
m = rb_entry(p, struct map, rb_node);
|
2010-03-26 06:58:58 +08:00
|
|
|
if (ip < m->start)
|
2018-11-23 18:42:39 +08:00
|
|
|
p = p->rb_left;
|
2014-10-15 03:05:38 +08:00
|
|
|
else if (ip >= m->end)
|
2018-11-23 18:42:39 +08:00
|
|
|
p = p->rb_right;
|
2010-03-26 06:58:58 +08:00
|
|
|
else
|
2015-05-23 00:45:24 +08:00
|
|
|
goto out;
|
2010-03-26 06:58:58 +08:00
|
|
|
}
|
|
|
|
|
2015-05-23 00:45:24 +08:00
|
|
|
m = NULL;
|
|
|
|
out:
|
2017-04-05 00:15:04 +08:00
|
|
|
up_read(&maps->lock);
|
2015-05-23 00:45:24 +08:00
|
|
|
return m;
|
2010-03-26 06:58:58 +08:00
|
|
|
}
|
2013-08-07 19:38:51 +08:00
|
|
|
|
2015-05-22 23:58:53 +08:00
|
|
|
struct map *maps__first(struct maps *maps)
|
2013-08-07 19:38:51 +08:00
|
|
|
{
|
2015-05-22 23:58:53 +08:00
|
|
|
struct rb_node *first = rb_first(&maps->entries);
|
2013-08-07 19:38:51 +08:00
|
|
|
|
|
|
|
if (first)
|
|
|
|
return rb_entry(first, struct map, rb_node);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2019-10-28 22:50:12 +08:00
|
|
|
static struct map *__map__next(struct map *map)
|
2013-08-07 19:38:51 +08:00
|
|
|
{
|
|
|
|
struct rb_node *next = rb_next(&map->rb_node);
|
|
|
|
|
|
|
|
if (next)
|
|
|
|
return rb_entry(next, struct map, rb_node);
|
|
|
|
return NULL;
|
|
|
|
}
|
2015-04-07 16:22:45 +08:00
|
|
|
|
2019-10-28 22:50:12 +08:00
|
|
|
struct map *map__next(struct map *map)
|
|
|
|
{
|
|
|
|
return map ? __map__next(map) : NULL;
|
|
|
|
}
|
|
|
|
|
2018-05-22 18:54:35 +08:00
|
|
|
struct kmap *__map__kmap(struct map *map)
|
2015-04-07 16:22:45 +08:00
|
|
|
{
|
2018-05-22 18:54:35 +08:00
|
|
|
if (!map->dso || !map->dso->kernel)
|
2015-04-07 16:22:45 +08:00
|
|
|
return NULL;
|
|
|
|
return (struct kmap *)(map + 1);
|
|
|
|
}
|
|
|
|
|
2018-05-22 18:54:35 +08:00
|
|
|
struct kmap *map__kmap(struct map *map)
|
|
|
|
{
|
|
|
|
struct kmap *kmap = __map__kmap(map);
|
|
|
|
|
|
|
|
if (!kmap)
|
|
|
|
pr_err("Internal error: map__kmap with a non-kernel map\n");
|
|
|
|
return kmap;
|
|
|
|
}
|
|
|
|
|
2019-11-26 08:58:33 +08:00
|
|
|
struct maps *map__kmaps(struct map *map)
|
2015-04-07 16:22:45 +08:00
|
|
|
{
|
|
|
|
struct kmap *kmap = map__kmap(map);
|
|
|
|
|
|
|
|
if (!kmap || !kmap->kmaps) {
|
|
|
|
pr_err("Internal error: map__kmaps with a non-kernel map\n");
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
return kmap->kmaps;
|
|
|
|
}
|